Tectonic Evolution of the Tethyan Region ||

Tectonic Evolution of the Tethyan Region NATO ASI Series Advanced Science Institutes Series A Series presenting the results of activities sponsored by the NA TO Science Committee, which aims at the dissemination of advanced scientific and technological knowledge, with a view to strengthening links between scientific communities. The Series is published by an international board of publishers in conjunction with the NATO Scientific Affairs Division A Life Sciences B Physics C Mathematical and Physical Sciences D Behavioural and Social Sciences E Applied Sciences F Computer and Systems Sciences G Ecological Sciences H Cell Biology Plenum Publishing Corporation London and New York Kluwer Academic Publishers Dordrecht, Boston and London Springer-Verlag Berlin, Heidelberg, New York, London, Paris and Tokyo Series C: Mathematical and Physical Sciences - Vol. 259 Tectonic Evolution of the Tethyan Region edited by A.M.C.$engor Department of Geology, Istanbul Technical University, Istanbul, Turkey with assistance from Y. Yllmaz, A. I. Okay, and N. Gorur Kluwer Academic Publishers Dordrecht / Boston / London Published in cooperation with NATO Scientific Affairs Division Proceedings of the NATO Advanced Study Institute on Tectonic Evolution of the Tethyan Region The Faculty of Mines, Istanbul Technical University, Istanbul, Turkey 23 September - 2 October 1985 in honour of Prof. Dr. rer.nat. ihsan Ketin Library of Congress Cataloging in Publication Data NATO Advanced Study InstItute on the "Tectonic Evolution of Tethyan Region" (1986 Istanbul Technical University. Faculty of Mines) Tectonic evolution of the Tethyan Region: proceedings of the NATO Advanced Study Institute on the "Tectonic Evolution of Tethyan Region" held in the Faculty of Mines, Istanbul Technical University, Istanbul, Turkey, 23rd September-2nd October 1986 in honour of Prof. Dr. rer. nat. lhsan Ketin / edited by A.M.C. ~engor with assistance from Y. Yllmaz, A.I. Okay, and N. Gor~r. p. cm. -- (NATO ASI series. Series C, Mathematical and physical sciences; no. 259) Inc 1 udes index. ISBN 0-7923-0067-X 1. Geology--Asia--Congresses. 2. Tethys (Paleography)- -Congresses. 3. Ketin, lhsan--Congresses. I. ~engor, A. M. Celal. II. Ketin, lhsan. Ill. Title. IV. Series. OE289.N38 1986 551.7' . 0095--dc 19 88-31485 ISBN-13: 978-94-010-7509-1 e-ISBN-13: 978-94-009-2253-2 DOl: 10.1007/978-94-009-2253-2 Published by Kluwer Academic Publishers, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. Kluwer Academic Publishers incorpor~te.s the publishing programmes of D. Reidel, Martinus Nijhoff, Dr W. Junk,a(ld MTP Press. Sold and distributed in the U.S.A. and Canada by Kluwer Academic Publishers, 101 Philip Drive, Norwell, MA 02061, U.S.A. In all other countries, sold and distributed by Kluwer Academic Publishers Group, P.O. Box 322, 3300 AH Dordrecht, The Netherlands. All Rights Reserved © 1989 by Kluwer Academic Publishers. Softcover reprint of the reprint of the 1 st edition 1989 No part of the material protected by this copyright notice may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage and retrieval system, without written permission from the copyright owner. TABLE OF CONTENTS Preface List of Participants List of Contributors Professor ihsan Ketin: An Appreciation The Tethyside Orogenic System: An Introduction A.M.C. ~engor One Some Key Features of the Evolution of the Western Alps J. Debelmas The Geometry of Crustal Shortening in the Western Alps ni xi xxvii xxxi 23 R.W.H. Butler 43 Triassic and Jurassic Oceanic/Paraoceanic Belts in the Carpathian-Pannonian Region and its Surroundings M. Kazmer and S. Kovacs 77 Major Events of the Tectono-Sedimentary Evolution of the North Hungarian Paleo-Mesozoic: History of the Northwestern Termination of the Late Paleozoic - Early Mesozoic Tethys S. Kovacs 93 Tectonic Units and Sutures in the Pontides, Northern Turkey A.I. Okay 109 An Example for the Tectonic Evolution of the Arabian Platform Margin (SE Anatolia) During Mesozoic and Some Criticisms of the Previously Suggested Models D. Alt1ner 117 Timing of Opening of the Black Sea: Sedimentological Evidence from the Rhodope-Pontide Fragment N. Gorur 131 Paleomagnetic Study of the Neogene Formations of the Aegean Area C. Kissel, C. Laj, A. Mazaud, A. Poisson, Y. Savascin, K. Simeakis, C. Fraissinet, J.L. Mercier 137 An Approach to the Origin of Young Volcanic Rocks of Western Turkey Y. Y1lmaz 159 Tectonic Evolution of Paleotethys in the Caucasus Sector of the Mediterranean Belt: Basic Problems A.A. Belov 191 VI Palaeomagnetism of Upper Cretaceous Rocks from the Caucasus and its Implications for Tectonics M.L. Bazhenov and V.S. Burtman 217 Tethys Evolution in the Afghanistan-Pamir-Pakistan Region J. Stocklin 241 Tectonogenesis and Evolution of a Segment of the Cimmerides: The Volcano-Sedimentary Triassic of Aghdarband (Kopet-Dagh, North-East Iran) A. Baud and G.M. Stampfli 265 Geology of the Baluchistan (Pakistan) Portion of the Southern Margin of the Tethys Sea G.R.McCormick 277 The Himalayan Orogenic Segment P. Le Fort 289 Crustal Scale Thrusting and Continental Subduction During Himalayan Collision Tectonics on the NW Indian Plate R.W.H. Butler and M.P. Coward 387 The Tectonic Evolution of Qinghai-Tibet Plateau: A Review Chang Chen-Fa, Pan Yu-Sheng and Sun Yi-Ying 415 The Neo-Cimmerian Ophiolite Belt in Afghanistan and Tibet: Comparison and Evolution J. Girardeau, J. Marcoux and C. Montenat 477 The Western End of the Tibetan Plateau A. Baud 505 Thrusting on the Tibetan flateau Within the Last 5 Ma K. Burke and L. Lucas 507 Tectonic Evolution of the Yangtze Tectonic Regime Zhang Qinwen, Qu Jingchuan and Chen Bingwei 513 Mesozoic Suturing in the Huanan Alps and the Tectonic Assembly of South China K.J. Hsu, Sun Shu and Li Jiliang The Shan Plateau and Western Burma: Mesozoic-Cenozoic Plate Boundaries and Correlations with Tibet A.H.G. Mitchell The Palaeo-Tethyan Realm and Indosinian Orogenic System of Southeast Asia C.S. Hutchison The Contribution of Vertebrate Palaeontology to the Geodynamic History of South East Asia E. Buffetaut Convergent-Plate Tectonics Viewed from the Indonesian Region W. Hamilton 551 567 585 645 655 PREFACE The ihsan Ketin NATO Advanced Study Institute on the Tectonic Evolution of the Tethyan Region was conceived in 1982 in Veszprem, Hungary, when three of the organizers (B.C.B., L.H.R. and A.M.C.9.) had come together for a meeting on the tectonics of the Pannonian basin. All three of us had experience in the Tethyan belt and all three of us had been for some time deploring the lack of communication among workers of this immense orogenic belt. Much new work had been completed in such previously little-known areas as Turkey, Iran, Afghanistan, the People's Republic of China, the entire Himalayan region, as well as new work in the European parts of the chain. Also, ironically, parts of the belt had just been closed to field work for political reasons, so it seemed as if the time was right to sit back and consider what had been done so far. Because the Istanbul group had had an interest in the whole of the Tethyan belt and because that ancient city was more centrally locElted with excellent opportunities to see both Palaeo- and Neo-Tethyan rocks in a weekend excursion, we thought that Istanbul was a natural place for such a meeting, not mentioning its own considerable attractions for the would-be contributors. A happy coincidence was that Prof.ihsan Ketin, one of the foremost leaders in the tectonics of the central Tethyan regions and the dean of Turkish geologists, was to retire from active teaching in 1983. We decided to seize the opportunity of gathering to honour Ketin to assemble the Tethyan workers. One of us (B.C.B.) thought that perhaps a NATO ASI was the best medium, but the necessity of involving numerous non-NATO participants and contributors made it imperative that other sources of finance be found. The Istanbul Technical University and its Faculty of Mines immediately declared themselves at our disposal to host the meeting and Dr.Remzi Akkok of Ak§an Engineering and Consulting Co. in Istanbul (a former pupil of Ketin, and a faculty member at the iTO) agreed to organize the logistics in Istanbul and also for the excursion. We also asked Dr.Yticel Yllmaz (then of the University of Istanbul), the noted Tethyan worker and an associate of Ketin whether he would join us as the fourth organizer. Yllmaz enthusiastically accepted and we applied to NATO foran ASI grant, which was swiftly given. To finance the non-NATO workers we were generously supported by the following oil companies: Arco Int., Chevron Overseas Petroleum Inc., Gulf Research and Development Co., Exxon Production and Research Company, Texaco Inc., and Esso Exploration Inc. For the local logistics in Istanbul and for the preparation of the meeting folders we are also grateful to Fruko-Tamek Meyva Sularl Sanayii A.$., istanbul. vii viii The meeting in Istanbul was a success; most of the contributors invi ted had responded .posi ti vely and all of those showed up. The meeting format was an unusual one, with prinicipal contributors designated for 13 segments of the orogenic belts to be supplanted by "commentators". This format worked well h r the meeting, but was a failure for the published proceedings. Four of the principal contributors failed to procude written accounts that considerably disturbed the original plan. Also some of the contributors have provided excellent papers that were more than just "comments" as originally intended. As a result, the original plan had to be given up and a more conventional book with a series of papers of equal standing was accepted. The publication of the proceedings took an unusually long time for a NATO ASI for two principal reasons. One was the slow inflow of manuscripts that had been asked to be delivered at the meeting in September 1985. The last manuscript was received in late 1987 and the book had to be closed in early 1988 owing to pressure of time (and t~1e publisher), despite that one major paper was still pending. The second reason was the requirement that the book had to be author-prepared for copy-printing. This requires high quality typewriters and acceptable English. Many of the contributors seemingly had little experience in preparing author-prepared manuscripts and consequently many had to be retyped in Istanbul, for which some currency exchange problem prevented NATO to supply the funds. Also a large number of the papers had to be thoroughly edited for improving the English, which, in places deteriorated to unintelligibility. Most of the figures had to be reshot for reduction. Having been reduced to a copy-editor and page designer with no funds, I was compelled to ask for help from a few friends to have the papers typed and figures reshot. Done on a charity basis, we simply had to await the gaps in the heavy schedules of the companies who deci.ded to assist us. Yticel Yllmaz, Naci Gortir, Aral Okay and Remzi Akkok all shared bits of the editorial responsibilty and Remzi kindly delegated to us office space in which the book could be assembled away from the daily turbulance of the university. I must thank all of these people and the Ajans-Tek for the numerous photographic reproductions and the Stirat Daktilo Company for typing many of the papers despite their heavy schedules. Yticel Yllmaz did much to expediate the editing while I was busy with other things. The resulting book is a heterogeneous one, both physically and in terms of the contents of the contributions. Only two papers submitted had to be rejected owing to low scientific standards. A few others for which one reviewer recommended rejection were accepted on the basis of two considerations: either the paper represented a common view in a given region and we felt that this had to be known, even if we disaggreed with it for good reasons, or the paper represented a band-wagon approach to a certain problem, for which this approach was not the most appropriate. This last category was given a place so that the reader of this book will know what is being done by a large number of workers. ix The book as a whole is somewhat skewed towards Asia. This was desirable, as Asia is the place from where the flow of information is most irregular. But as a whole we hope it provides a good basis for general Tethyan reading. More important, it is our hope that such Tethyan meetings and Tethyan compendia of papers will become more frequent in the future. This meeting taught us what the problems might be in organizing such endeavours and we now know how to tackle them. But most important of all we know how willing the Tethyan workers are to come together and argue about their problems. That this meeting united people from Europe, Asia, North America·, Austraila, and Africa shows that there really are no barriers that cannot be broken from one end of the Tethysides to the other. This sincere "internationalism" is the spirit of the sponsors of this meeting, it is the spirit of the man in whose honour we all came together, and is the spirit of the founders of the Technical University, whose hospitality we all enjoyed. Before I close this preface, I need to thank Prof.Dr.Kemal Kafall, the energetic Rector Magnificus of the iTtl, and ProLDr .Erdogan Ytizer, the equally enthusiastic Dean of the Faculty of Mines of the iTti. We also thank Kevin Burke who substituted for Clark Burchfiel as Institute Director while Clark was hammering at Tethyside outcrops at Ulu Muztagh! A. Mo C. Sengor for the Organizers: B. C. Burchfiel L. H. Royden A. M. C. Seng~r Y. Yllmaz LIST OF PARTICIPANTS Dr.Remzi Akkok AK$AN A.$., Investigation, Consulting and Engineering Bliylikdere Caddesi 73/11 Mecidiyekoy, Istanbul Telefon: 172 35 04 - 172 35 05 TURKEY Dr.Demir Altlner Ortadogu Teknik Vniversitesi Mlihendislik Fakliltesi Jeoloji Bollimti Ankara, TURKEY 237100/2692 or 2682 Mustafa Aydln Tlirkiye Petrolleri A.$. Arama Grubu Mtidafaa Caddesi 22, P.K.209 Jakanllklar, Ankara TURKEY Ozer Balka$ Tlirkiye Petrolleri A.$. Arama Grubu Mtidafaa Caddesi 22, P.K. 209 Bakanllklar, Ankara TURKEY Muawia Barazangi Institute for the Study of the Continents Snee Hall Cornell University Ithaca, NY 14853 USA (607) 256-6411 Telex: 937478 Aymon Baud Geological Institute Palais de Rumine CH-1005 Lausanne SWITZERLAND xi xii Dr.A.A.Belov Geological Institute Academy of Sciences pzyhewsky 7 109017 Moscow USSR Dr.Ziad R.Beydoun Marathon International Petroleum (G.B.) Ltd. Marathon House, 174 Marylebone Road London, NMI SAT. Tel. 0l-486 0222 Telex 297183 ENGLAND Jean-Pierre Brun Labo Tectoniquew VER Sciences Physique de la Terre Tour 25 ler Et. 2 Pl. Jussieu 75230 Paris, FRANCE 633-07-29 Dr.Eric Buffetaut University of Paris 6 Lab. of Vertebrate Paleontology 4 Place Jussieu 75230 Paris Cedex 05 FRANCE Dr.Jean-Pierre Burg Department of Geology University of Melbourne Parkville, Victoria 3052 AUSTRALIA 3416740 - AA35185 Dr.Kevin Burke Lunar and Planetary Institute 3303 NASA Road Houston, TX 77058 (713) 486- 2180 Robert W.H.Butler Dept. of Geolog. Sciences The University South Road Durham DHI 3LE United Kingdom 0385-64971 (ext 432) Dr.Chang Chen fa Academia Sinica Institute of Geology P.O. Box 634 Beijing, China Tristan M.M. Clube. Department of Geophysics University of Edinburgh Edinburgh, SCOTLAND 031-667-1081 ext. 2944 Telex: 727442 Dana Quentin Coffield Earth Sciences and Resources Inst. University of So. Carolina Columbia, SC 29208 USA (803) 777-6484 Telex 805038 Dr.Millard Fillmore Coffin Bureau of Mineral Resources GPO Box 378 Canberra A.C.T. 2601 AUSTRALIA Dr.Michel Colchen Universite de Poitiers Laboratoire de Geologie Stratigraphique 40 Av. de Recteur Pineau 86022 Poitiers Cedex FRANCE J.Calvin Cooper Dept. of Geology Rice University Houston, TX 77251 USA (713) 525-8101 ext 3337 Michael Peter Coward Dept. of Geology Imperial College London SW7 2BP ENGLAND 01-589-5111 x5504 Telex: 261503 xiii xiv Kenneth M.Creer Department of Geophysics University of Edinburgh James Clerk Maxwell Building Mayfield Road Edinburgh, EH9 3J2 SCOTLAND (031) 667 1081 ext. 2952 Telex: 727442 Unived G. Dr.John Gordon Dennis Department of Geological Sciences California State University Long Beach, CA. 90840 USA (213) 430-6903 (home) (213) 498-4404 (office) Mr.Aksoy Ercan Flrat Universitesi Mtihendislik Faktiltesi Jeoloji Boltimti Ara~tlrma Gorevlisi Elazlg, TURKEY Prof.Dr.Kazlm Ergin iTti Maden Faktiltesi Te~vikiye, istanbul TURKEY Dr.Ayhan Erler Ortadogu Teknik Universitesi Mtihendislik Faktiltesi Jeoloji Boltimti Ankara, TURKEY Phillip Bruce Gans Dept. of Geology, Stanford University Stanford, CA 94305 (415) 497-2537 or 497-1149 Dr.David Gee Geological Survey of Sweden Box 670 751 28 Uppsala SWEDEN Klaus H.A. Gohrbandt Chevron Overseas Petroleum Inc. 6001 Bollinger Canyon Road, San Ramon, California Mail P.O. Box 5046, San Ramon, CA 94583-0946 Tel. (415)842-3708 Telex: ITT 470074 CHEV UI(International) USA Dr.Naci Gortir iTO Maden Faktiltesi Jeoloji Mtihendisligi Boltimti Genel Jeoloji Anabilim Dall Te~vikiye, Istanbul TURKEY Dr.Warren Hamilton U.S.Geological Survey Mail Stop 964 Box 25046 Federal Center Denver, CO 80225 USA Samir S.Hanna Earth Resources Institute Department of Geology University College of Swansea SA2 8PP United Kingdom (0792) 295140 Telex: 48358 ULSWAN G Dr.Mark R.Hempton Shell Dev. Co. P.O.Box 481 Houston, TX 77001 Tel.: (713) 663-2120 Dr.Ken Hsu Geologisches Institut E.T.H.-Zentrum Sonneggstrasse 5 CH-8092 Zurich SWITZERLAND Mary Hubbard MIT 54-1020 Cambridge, MA 02139 USA xv xvi Dr.C.Hutchison University of Malaya Department of Geology Kuala Lumpur 2211 MALAYSIA Cemal Kaplangl Tlirkiye Petrolleri A.$. Arama Grubu Mlidafaa Caddesi 22, P.K. 209 Bakanllklar, Ankara TURKEY Daniel Edmund Karig Department of Geological Sciences Cornell University Ithaca,N.Y. 14853 USA (607) 256-369 Miklos Kazmer Dept. of Paleontology Eotvos University Kun Bela ter 2 Budapest H-1083 HUNGARY Ilyas Erdal Kerey Flrat Vniversitesi Mlihendislik Fakliltesi Jeoloji Mlihendisligi Bollimli Elazlg, Turkey Prof.Dr.lhsan Ketin ITO Maden Fakliltesi Jeoloji Mlihendisligi Bollimli Genel Jeoloji Anabilim Dall Te~vikiye, Istanbul TURKEY Catherine Kissel Centre des Faibles Radioactivites BP no.l Avenue de la Terrasse 91190 Gif s/Yvette FRANCE 3311-(6) 907-78-28 ext 791 691.137 F Carlo E.G. Laj Centre des Faib1es Radioactivites BP No.1 Avenue de 1a Terrasse 91190 Gif s/Yvette FRANCE 3311-(6) 907-78-28 ext 791 691.137 F Patrick LeFort Centre de Recherches Petrographiques et Geochimiques B.P.20 Vanoeuvre-1es-Nancy F-54501 FRANCE (8) 351-2213 Telex: 960431 adnancy William S.Leith Lamont-Doherty Geo1. Observatory Palisades, NY 10964 USA (914) 359-2900 Telex: 710-576-2653 Dr.Jean Marcoux Universite de Paris VII Laboratoire de tectoniquede 1a Terre Sciences physiques de 1a Terre Tour 25-24 10 etagedex 2 Place Jussieu 75230 Paris Cedex 05 FRANCE Mr.Ron Marr Conoco, Inc. P.O. Box 4800 The Woodlands, TX 77380 Tel.: (713) 367-3305 Telex: 775347 George Robert McCormick Department of Geology University of Iowa Iowa City, Iowa 52242 USA (313) 353-4318 or (313)353-4105 Metin Me§hur Ttirkiye Petro11eri A.$. Arama Grubu Mtidafaa Caddesi 22, P.K. 209 Bakan11k1ar, Ankara TURKEY xvii xviii Elizabeth L.Mi11er Dept. of Geology Stanford University Stanford, CA 94305 (415)497-2537 or 497-1149 Andrew H.G.Mitche11 C/o UNDP P.O. Box 7285 ADC M.I.A. Road Pasay City M. Manila Philippines Shankar Mitra Arco Resources Technology 2300 W.P1ano Parkway (PAC 3063) Plano, TX 75075 (214) 422-6219 John Nabe1ek MIT, E34-406 Cambridge, MA 02139 USA Jay Namson 3054 ARCO Resources Technology 2300 Plano Parkway Plano TX 75075 USA Dr.Rudo1f Oberhauser Geological Survey of Austria 23 Rasumofskygasse A-103l Vienna Austria 76 56 74/42 Telex: 13297 GEOBA-A Dr.Aral Okay iTO Maden Fakti1tesi Jeoloji Mtihendisligi Bo1timti Gene1 Jeoloji Anabi1irn Da11 Te~vikiye, Istanbul TURKEY Dr.Necdet Ozgti1 iTO Maden Fakti1tesi Jeoloji Mtihendisligi Boltirnti Gene1 Jeo1oji Anabilirn Dall Te~vikiye, Istanbul TURKEY Laurence Page MIT 54-611 Cambridge, MA 02139 USA Dr.Dimitrios Papaniko1aou Department of Geology University of Athens Panepistimiopo1is Zografou Athens 15771 Greece Prof.Dr.Niko1a Pantic Geo1osko-Pa1eonto1oski Zavod Univerzite Beogradu Kamenicka 6 11000 Beograd JUGOSLOWIA Malcolm G.Parsons Esso Exploration Turkey, S.A. P.O.Box 16 Ankara, TURKEY T.L.Patton Amoco Turkey Petroleum Co. Re§it Ga1ip Caddesi ';'60 Gaziosmanpa§a Ankara, TURKEY Dr.Dogan Perin~ek Ttirkiye Petro11eri A.~. Arama Grubu Mtidafaa Caddesi 22, P.K. 209 Bakan11k1ar, Ankara TURKEY Dr.Zhang Qinwen Chinese Academy of Geological Sciences Baiwanzhuang Road 26 West City, Beijing People's Republic of China Dr.A1ison C.Ries University College of Swansea Singleton Park Swansea SA2 8PP Tel.: 0792-295496 Telex: 48358 ULSWAN G(attn. ERI) UNITED KINGDOM xix xx Dr.John Rodgers Department of Geology and Geophysics Box 6666 Yale University New Haven, CT. 06511 USA (203) 436-0616 Dr.Leigh H.Royden Department of Earth, Atmospheric & Planetary Sciences, MIT, Cambridge, Massachusetts 02139 USA Dr.Branch J.Russell Marathon Research Center P.O. Box 269 Littleton, CO 80160 Tel.: (303) 794-2601 Dr.Mircea Sandulescu Institut de Geologie et Geophysique 1, rue Caransebes Bucarest, ROMANIA Mustafa Sarlbudak iTO Maden Faktiltesi Jeoloji Boltimti Te~vikiye, istanbul TURKEY Elizabeth Schermer MIT 54-1024 Cambridge, MA 02139 USA Michael Paul Searle Department of Geology University of Leicester Leicester LEI 7RH United Kingdom 0533-554455 ext 107 Telex: UNIVLIB LESTER 341198 SUMMER ADDRESS 10 Lagoon Close Lilliput Poole Dorset UNITED KING DOOM Dr.Celal ~engor iTO Maden Fakliltesi Jeoloji Bollimli Te§vikiye, istanbul TURKEY Dr.Sigmund Snelson Shell Development Corporation P.O. Box 481 Houston, TX. 77001 (713) 663-2622 Telex: 76-2248 Dr.Freddie Yiying Sun Institute of Geology Academia Sinica P.o.Box 134 Beijing, China Mr.Ozan Sungurlu Tlirkiye Petrolleri A.~. Arama Grubu Ba§kan1 Mlidafaa Cad. 22, P.K. 209 Bakan11klar, Ankara TURKEY Neptune Srimal Dept. of Geological Sciences University of Rochester Rochester, NY 14627 (716) 275-2409 Dr.J.Stocklin Erbduhlstr 4 8472 Seuzach Zurich, SWITZERLAND Tarquin Teale Dept. of Geology Imperial College Prince Consort Road London SW7 2BP U.K. 01-589 5111 x5537 Telex: 261503 Dr.Cestmir Tomek Geofyzika Brno P.O. Box 62, Brno 61246 CZECHOSLOVAKIA Telephone: 05 57464 Telex: 625 12 xxi xxii James W.Tucker ARCO International Oil and Gas Company 444 South Flower Street Los Angeles, California 90017 Tel.: 213486 2735 Telex: 19-4154 USA Timur Ustaomer ITO Maden Faktiltesi Jeoloji boltimti Te~vikiye, Istanbul TURKEY Lester Craig Ward 10, Walpole Road Colliers Wood London SW19 2B2 GREAT BRITAIN 01-542-9649 Dr.Brian Windley Department of Geology The University Leicester LEI 7RH England (0533) 554455 Ext. 107 Telex: UNIVLIB LESTER 341198 Mr.Barry G.Wood Marathon Int. Petroleum (GB) Ltd. 174 Marylebone Road London, NW15AT ENGLAND 01-486-0222 Telex: 297183 Dr.Yticel Yllmaz Istanbul Dniversitesi Mtihendislik Faktiltesi Jeoloji Boltimti Vezneciler, Istanbul TURKEY Dr.Plnar Yllmaz c/o Dr.Kevin T.Biddle Esso Exploration and Production UK Limited Biwater House Portsmouth Road Esner Surrey KTlO 9SJ ENGLAND Dr.Ken Hsti Geologisches Institut E.T.H.-Zentrum Sonneggstrasse 5 CH-8092 Zurich SWITZERLAND Dr.C.Hutchison University of Malaya Department of Geology Kuala Lumpur 2211 MALAYSIA Li Jiliang Institute of Geology Academia Sinica P.O.Box Beijing PRC Qu Jingchuan The Institute of Geology Chinese Academy of Geological Sciences Baiwanzhuang Road 26 Fuchenmengwai, Beijing PRC Miklos Kazmer Dept. of Paleontology Eotvos University Kun Bela ter 2 Budapest H-l083 HUNGARY Catherine Kissel Centre des Faibles Radioactivites BP no.l Avenue de la Terrasse 91190 Gif s/Yvette FRANCE 3311-(6) 907-78-28 ext 791 691.137 F S.Kovacs Hungarian Geological Institute H-1442 Budapest P.O.Box 106 HUNGARY XXlll xxiv Carlo E.G. Laj Centre des Faib1es Radioactivites BP No.1 Avenue de 1a Terdioactivites BP No.1 Avenue de 1a Terrasse 91190 Gif s/Yvette FRANCE 3311-(6) 907-78-28 ext 791 691.137 F Pa trick LeFort Centre de Recherches Petrographiques et Geochimiques B.P.20 Vanoeuvre-1es-Nancy F-54501 FRANCE (8) 351-2213 Telex: 960431 adnancy L.Lucas Lunar and Planetary Institute 3303 Nasa Road 1 Houstan, TX 77058 and Department of Geosciences University of Houston, University Park, Houston, Texas 77004 USA Dr.Jean Marcoux Universite de Paris VII Laboratoire de tectoniquede 1a Terre Sciences physiques de 1a Terre Tour 25-24 10 etagedex 2 Place Jussieu 75230 Paris Cedex 05 FRANCE George Robert McCormick Department of Geology Universite of Iowa Iowa City, Iowa 52242 USA (313) 353-4318 or (313) 353-4105 Andrew H.G.Mitche11 C/o UNDP P.O. Box 7285 ADC M.I.A. Road Pasay City M. Manila Philippines Dr.C.Montenat Laboratoire de Geologie Institut Geologique Albert-de-Lapparent, 21 rue d'Assas 75270 Paris cedex 06 FRANCE Dr .Aral LOkay iTD Maden Faktiltesi Jeoloji Mtihendisligi Boltimti Genel Jeoloji Anabilim Dall Te~vikiye, Istanbul TURKEY xxv LIST OF CONTRIBUTORS Dr.Demir Altlner Ortadogu Teknik Vniversitesi Mtihendislik Faktiltesi Jeoloji Boltimti Ankara, TURKEY 237100/2692 or 2682 Aymon Baud Geological Institute Palais de Rumine CH-1005 Lausanne SWITZERLAND M.L.Bazhenov Geological Institute, Academy of Sciences of the USSR Pyzhevsky per., 7 109017 Moscow USSR Dr.A.A.Belov Geological Institute Academy of Sciences pzyhewsky 7 109017 Moscow USSR Cheng Bingwei The Institute of Geology Chinese Academy of Geological Sciences Baiwanzhuang Road 26 Fuchengmenwai, Beijing PRC. V.S.Burtman Geological Institute, Academy of Sciences of the USSR Pyzhersky per., 7 109017 Moscow USSR xxvii xxviii Dr.Eric Buffetaut University of Paris 6 Lab. of Vertebrate Paleontology 4 Place Jussieu 75230 Paris Cedex 05 FRANCE Dr.Kevin Burke Lunar and Planetary Institute 3303 NASA Road Houston, TX 77058 (713) 486-2180 Robert W.H.Butler Dept. of Geolog. Sciences The University South Road Durham DHI 3LE United Kingdom 0385-64971 (ext 432) Dr .Chang Chenfa Academia Sinica Institute of Geology P.O. Box 634 Beijing, China Michael Peter Coward Dept. of Geology Imperial College London SW7 2BP ENGLAND 01-589-5111 x5504 Telex: 261503 Prof.Jacques Debelmas Geology Dept. , University, F3803l Grenoble FRANCE Dr.J.Girardeau Laboratoire de Petrologie Physique, Universite Paris VlIet Institut de Physique du Globe de Paris, Place Jussieu, 75230 Paris Cedex OS, FRANCE Dr. Naci Gortir iTti Maden Faktiltesi Jeoloji Mtihendisligi Boltimti Genel Jeoloji Anabilim Dall Te~vikiye, Istanbul TURKEY Dr.Warren Hamilton U.S.Geological Survey Mail Stop 964 Box 25046 Federal Center Denver, CO 80225 USA Dr.Zhang Qinwen Chinese Academy of Geological Sciences Baiwanzhuang Road 26 West City, Beijing People's Republic of China Sun Shu Institute of Geology Academia Sinica P.O.Box Beijing PRC Dr.A.M.Celal $engor iTti Maden Fakultesi Jeoloji BOlumu Te§vikiye, istanbul TURKEY Dr.Freddie Yiying Sun Institute of Geology Academia Sinica P.o.Box 134 Beijing, China Dr.J. Stocklin Erbduhlstr 4 8472 Seuzach Zurich, SWITZERLAND Dr .yucel Yllmaz istanbul tiniversitesi Muhendislik Fakultesi Jeoloji Bolumu Vezneciler, istanbul TURKEY Pan Yu-Sheng Institute of Geology, Academia Sinica P.O.Box 634 Beijing CHINA xxix r-) ~ -_.=::::r'-, :::::0> • PROFESSOR iHSAN KETiN: AN APPRECIATION The organizers and participants of the NATO Advanced Study Insti tute "Tectonic Evolution of the Tethyan Region" wish to dedicate this Institute and its published proceedings to Dr.rer.nat. Ihsan Ketin, emeritus professor of geology in the Istanbul Technical University in grateful recognition of his important contributions to our understanding of the geological structure and evolution of the central part of the Alpine-Himalayan mountain ranges. In doing so we also wish to underline that the influence of Ketin' s work has long overflowed the boundaries of the Alpine-Himalayan system and made a conspicuous impact on theoretical tectonics in general. His discovery in 1948 of the North Anatolian strike-slip fault and with it the 'west-drift' of a rigid Anatolian block with respect to its surroun4ings not only was an important step in the recognition of the widespread occurrence of large strike-slip faults in the world, but it also constitutes one of the earliest papers in which the tectonics of a large area was interpreted in terms of relative horizontal motion of a few internally rigid blocks along narrow zones of displacement. His discovery in 1956 of the late Cretaceous-early Cainozoic age of the central Anatolian crystalline axis disposed of Kober's theory of symmetric orogens with ancient median masses along the axis of symmetry in one of its type localities: Instead, Ketin showed in 1959 that Asia Minor as a whole was an asymmetric south-vergent orogen whose construction lasted through several episodes of mountain-building from the late Palaeozoic to the present. This view formed the main basis for most plate tectonic interpretations of Turkey in the last two decades. Ihsan Ketin was born on 10th of April in the ancient central Anatolian town of Kayseri (Caesarea), located at the foot of the mighty volcano of Erciyes (Mt.Aergus) as a subject of Sultan Mehmed V. in that eventful year of 1914. The family, of which Ketin was the second child, was of modest means. His father, Ali Efendi, was compelled to spend much of the time during which Ketin grew up as a child, fighting for his country: first through World War I and then, until 1922, in Mustafa Kemal's War of Liberation. During this time Ketin came under two powerful influences that eventually determined the course of his life: The first was that of his maternal grandmother Hatice Hanlm, a strong-willed Anatolian woman who instilled in Ketin the desire to do something worthwhile. The second source of influence was mute, but possibly more powerful: the towering Mt.Erciyes awakened in Ketin a love of nature, especially of her mineral kingdom that eventually became the child's life-long occupation. Before Ketin completed the first decade of his life the Ottoman Empire had become history and the new Republic of Turkey had been declared with Mustafa Kemal as its first president. This extraordinary xxxi xxxii man was determined to transform the old Ottoman society into a new Turkish nation and was aware that education was his most effective weapon. He sent hordes of young men to various western European countries to receive a university education with the instruction to come back "to raise Turkey to the highest level of contemporary civilisation." When Ketin boarded the train to go to Berlin in 1932 his heart was filled with the inspiration that radiated from Mustafa Kemal to learn the science of the west and to bring it back to his homeland, where, Ketin hoped, it could take root and flourish. But the Berlin Ketin arrived at was the troubled capital of the Weimar Republic, the artificial child of the Versailles Treaty, which was about to expire in the bloody hands of the architect of the infamous Third Reich, Adolf Hitler. The raging inflation, cancerous unemployment, rampant terrorism and the resulting misery induced the quiet natured Anatolian youth after his first semester in the university to move away from Berlin, where he had been exposed to the ideas of Hans Stille at his lectures. From the Prussian capital Ketin moved to the sphere of influence of another giant of tectonics in Bonn. Hans Cloos, the holder of the chair of geology in Bonn and at the same time the influential editor-in-chief of the Geologische Rundschau became not only Ketin' s teacher and eventual doctoral advisor, but also his close, almost fatherly friend. Between 1935 and 1938 Ketin remained under Closs' tutelage that imparted on him a zeal for careful field observation, especially geologic mapping, and a large reservoir of knowledge along wi th a humanism that contrasted sharply with the prevailing racism of the Nazi Germany, but that found a warm echo in Ketin's upbringing that had taken place in the heartland of the Ottoman Empire, in which numerous ethnic groups had peacefully coexisted for centuries. Ketin ended his studies with a doctoral dissertation on the tectonics and volcanism of the region around Bad Bertrich, which was published in 1940. Following the completion of his formal studies in Germany, Ketin returned to Turkey in the Autumn of 1938 and was appointed assistant professor at the Institute of Geology of the University of Istanbul, where, during World War I, the noted German geomorphologist and structural geologist Walther Penck had been the head of the Institute. When Ketin arrived in Istanbul, he found himself the third Turkish citizen with a Ph D in geology! The first, a certain Anastase Georgiades from Istanbul had obtained his doctorate from Zurich in 1918, but evidently had not returned to Turkey. The second, Dr.Ahmet Can Okay was an immigrant from the Soviet Central Asia and had come to Turkey after he had completed his studies in Germany. Thus Ketin was the first native of Turkey to work in his country with a Ph D in geology. When Ketin joined the Faculty of the Institute of Geology in Istanbul Professor Hamit Nafiz Pamir, the one-time assistant and interpreter to Walther Pecnk was the head of the Institute. A graduate of the Uni versi ty of Geneva, Pamir had had to interrupt his doctoral studies owing to World War I. Since then he had been compelled to spend XXXlll more time organizing the earth sciences in the newly-founded Republic of Turkey than doing research. Therefore, when Ketin returned to his country he found that no research tradition existed in geology. One had to be created and it is perhaps Ketin's greatest achievement that during the course of his professional life his work became in Turkey the cornerstone of a research tradition in geology. Ketin's initial activity in Turkey was split between research and teaching. His first research projects naturally reflected the strong influence of Cloos and Ketin plunged energetically into mapping granites and brittle structures. A year after Ketin' s arrival in Istanbul, a long-dormant zone of earthquakes in northern Turkey, the structure that Ketin was to make popular throughout the world under the designation of the North Anatolian Fault resumed its activity with the disastrous Erzincan quake of 29th December, 1939 that took the lives of more than 30,000 inhabitants. Between 1940 and 1948 Ketin devoted a number of mainly descriptive papers to the earthquakes that progressed westwards from Erzincan. Finally, in 1948 Ketin published his classic paper "tiber die tektonisch-mechanischen Folgerungen aus den grossen anatolischen Erd beben des letzten Dezennium" (On the tectonic-mechanic implications of the great Anatolian earthquakes of the last decade). In this paper he documented that the earthquakes in northern Turkey had all occurred along an east-west fault zone that had the character of a right-lateral strike-slip fault. Ketin noticed that with one exception, all of the recent earthquakes had taken place along this fault zone, while vast areas of the Anatolian highland remained aseismic. Ketin deduced from this that an "Anatolian Block", composed of the aseismic areas was "drifting westwards" with respect to the areas to the north. Ketin also noted that one earthquake had occurred near Kozan near the northeastern corner of the Eastern Mediterranean. This, he speculated, may be the expression of another fault that perhaps delimits the "Anatolian Block" against the Arabian platform. This prediction was vindicated only 23 years later when the Bingeil earthquake of 22nd May, 1971 took place on what was to be called the East Anatolian Fault, the left-lateral conjugate pair of the North Anatolian Fault. Ketin's 1948 paper was the second, after W.Q.Kennedy's 1946 paper on the Great Glen Fault, of a series of papers that led to the recognition of the widespread presence and importance of large, in many places orogen-parallel, strike-slip faults, a recognition for which plate tectonics was to supply the rationale nearly a quarter of a century later. In the meantime Ketin also spent all his summers mapping in diverse parts of his previously only sparsely mapped country. Although he initially had to map on a scale of 1:100.000, his maps were immaculate: I remember going to the field in Bursa with Ketin in 1984, with his 1946 manuscript map in our hands. We were in an ophiolitic melange terrain xxxiv and nearly 40 years ago Ketin had carefully mapped the larger blocks ~ The result of one of these summer1y mapping exercises served as his "Habilitation Thesis" and Ketin was promoted to associate professorship in 1945, three years after he had married a young teacher of geography. Miss Bedia A1plin. In 1953 Ketin moved to the then newly founded Faculty of Mines of the old Istanbul Technical University. Here Ketin continued his studies both on the neo- and palaeotectonics of Turkey. In the interval 1953-1956 he was particularly concerned with testing the hypothesis of Sir Edward Bailey and J.W.McCallien, then of the University of Ankara. Bailey and McCa11ien had discovered an extensive outcrop of an ophiolitic melange to the immediate southeast of Ankara and assumed that it underlay the Klr~ehir Massif, interpreted as a giant klippe of northerly origin. Ketin's mapping showed that this was not the case and the Massif in reality underlay the ophiolites. He showed further that the Massif itself had formed only in the late Cretaceous, contrary to the prevailing view of a much older (Palaeozoic or even Precambrian) age. Ketin thus demonstrated that the northern marginal ranges of Turkey, called Pontides after the Pontus Euxinus (Black Sea), were older than the Klr~ehir Massif, whereas the southern marginal chains, the Taurides (after the Taurus of the classical geographers), in which sedimentary successions reach from the Cambrian to the Eocene (in places even Miocene) without a major angular unconformity, were clearly younger. This implied that Turkey had grown from north to south during much of the Phanerozoic, a recognition that clashed with the then-fashionable two-sided orogen model of Kober and Stille, according to which the Ponti des represented the north-vergent north flank, while the Taurides were the south-vergent south flank of a symmetric Anatolian orogen with the crystalline massifs of Menderes and Klr~ehir formeing the axial Zwischengebirge. When Ketin presented some of his conclusions in 1955 at the "Geotectonics Symposium" held in honour of Stille in Hannover, the old and dogmatic German master told Ketin that he found this story hard to believe. Although Ketin had submitted a manuscript intended for the proceedings of the symposium, his paper was somehow left out of the final Festschrift. He later published different versions in Turkey and in Austria and those papers formed the basis of our modern views of the palaeotectonic evolution of Turkey. In 1959 Ketin published his first palaeotectonic synthesis of Turkey. This paper represents a clear break from the Kober-Stille model and a kind of return to Suess' original view of 1909, that portrayed Turkey as a south-vergent outer arc of his Asiatic structure (Asiatischer Bau). Here Ketin showed that orogenic deformation during the Phanerozoic generally migrated from north to south in Turkey. On the basis of the age of the final orogeny and the palaeogeographic development, Ketin distinguished four major tectono-stratigraphic zones three of which extended west to east along the entire length of the country. Only the fourth, the southernmost unit, was confined to the southeastern extremity of the country, being located on the Arabian Platform. Ketin's zones were the following, from north to south: 1- Ponti des (Palaeozoic and Mesozoic orogenic deformation) 2- Anato1ides (Mesozoic and Cainozoic orogenic deformation) 3- Taurides (early Cainozoic orogenic deformation) 4- Border Folds (late Cainozoic orogenic deformation) xxxv In 1961 and 1966 Ketin further refined this classification, which for many years, until the advent of the theory of plate tectonics, served as the basis for all palaeotectonic studies in Turkey. When ;;engor (1979) and ;;engor and Yl1maz (1981) synthesized the tectonic evolution of Turkey from the viewpoint of plate tectonics, all they had to do was to give dif ferent names to the same units that Ketin had distinguished more than two decades earlier. Thus, the Pontides became the Pontide island arc (to be split into a Rhodope-Pontide arc and a Sakarya arc in 1981), the Anato1ides and the Taurides were united into an Anato1ide/Tauride platform (from which ;;engor et a1., 1982, separated a Klrgehir block as an independent unit), and the Border Folds remained the same (in 1979 Ozan Sungur1u suggested to rename them as the Assyrides to maintain para1e11ism with the other three units' names). Since the publication of these landmark papers Ketin maintained his activity both in pa1aeo- and in neotectonics. His fieldwork largely was the basis for the concept of the East Anatolian Accretionary Complex (perhaps the most fundamental modification introduced into his 1966 classification), for the discovery of the Palaeo-Tethyan suture in Turkey, and for the classification of the neotectonic uinits of Turkey. In addition to his research activity in Turkey, Ketin also stands out as an earth-science teacher and an organizer of the earth sciences in the country. As a teacher he not only instructed myriads of students, but also is the author of the most widely used text-books of physical geology, structural geology and the geology of Turkey in this country. His lecture notes on such diverse topics as the recent developments in the earth sciences and the tectonics of Africa are monuments to conciseness and clarity. Ketin is an enthusiastic field geologist and his enthusiasm is contagious. To this day he delights in introducing students into their first mapping area, in acquanting them "with the language of the rocks" as he is fond of saying, in demonstrating for them how to record their observations in minute detail and showing them how to sketch outcrops and panoramas! Ketin was the one who established the 1.T.U. tradition that every post-graduate geology student has to prepare at least one detailed geological map as a part of his or her thesis work. In addition to his formal teaching duties, Ketin has been also the foremost popularizer of the earth sciences in Turkey. Amidst his multifarious duties he has found time always to write popular articles for the general education of the public. Ketin's organizational skills are best displayed by his ability to form and direct research groups. Today his group is the most active and internationally best-known in this country. As a department head, Ketin has always made sure that even the youngest member of his team became an independent researcher. He has repeatedly stressed throughout his career that he expected his students and associates to improve what had xxxvi been done earlier. More than once he exclaimed: "Don't come to me to tell me that I was right. Come to me if you found that I had been wrong!" Ketin was once the president of the Geological Society of Turkey and twice the Dean of the Faculty of Mines of the I.T.U. For many years he was a panel member of the Turkish National Research Council for Research and Technology. He also represented Turkey on many international scientific committees and was the Turkish contributor of the International Tectonic Map of Europe. Ketin's activity as a scientist, university teacher, and scientific organizer found the highest recognition both in Turkey and abroad. In 1981 he became the first recipient of the Hamit-Nafiz-Pamir Medal of the Geological Society of Turkey. In the same year the Turkish National Research Council for Science and Technology (Ttibitak) gave him the Science Award for the totality of his works, the highest recognition for a scientist in Turkey. Ketin was elected an Honorary Fellow of the Geological Society of London in 1984 and of the Geological Society of America in 1988. Also in 1988 he received the prestigious Gustav-Steinmann-Medai11e of the Geo10gische Vereinigung in the Federal Repbulic of Germany for his "far-sighted geotectonic work, contributions to the geology of Turkey and to international co-operation in the earth sciences". I here speak in the name of the organizers, the contributors, and the participants of the Ihsan Ketin Nato Advanced Study Institute on the Tectonic Evolution of the Tethyan Region in wishing Professor Ketin a long, healthy, and productive life. A.M.C.:?engor THE TETHYSIDE OROGENIC SYSTEM: AN INTRODUCTION A.M.C.:;>engor iTO Maden Faktiltesi, Jeoloji Boliimii, Te§vikiye 80394 istanbul, TURKEY ABSTRACT. The Tethysides are the orogenic belts that grew out of the mostly collisional obliteration of the Tethyan domain. The Tethyan domain includes both the Palaeo-Tethys (the original triangular gap in Pangaea), and the Neo-Tethys (Tethys opened behind the Cimmerian Continent as it rifted away from northern Gondwana-Land and rotated to close Palaeo-Tethys) plus their continental margins of diverse types. The term Tethys was defined as a tectonic equivalent of the Centrall Mediterranean of Neumayr and should not be used as a palaeogeographic entity only. The evolution of the Tethysides discloses large amounts of orogen parallel (or subparallel) strike-slip faulting that significantly disru pted the Tethyan orogenic collage. The concept of "allochthonous terranes" is viewed to be little more than the concept of nappes (since the latter also included strike-slip generated slivers) and found a retrogressive step, if used in place of genetic concepts that have long superceded nappe descriptions in the Alpine System at least. INTRODUCTION Apart from Argand' s monumental La Tectonique de l' Asie there is still no comprehensive account dealing with the entire Tethyan orogenic belt. It is a very difficult mountain system to write a comprehensive synthesis about, because geological knowledge about its various parts is extremely heterogeneous in quantity, in quality, and in terms of the theoretical systems in the framework of which it is gathered and/or the languages in which it is written. This variety makes communication very difficult and, from time to time, political instability renders large parts of the system entirely inaccessible to observation. Such isolation also hinders flow of information in and out of closed areas. If we add to this the great physical difficulties involved in doing field work in considerable portions of the Tethyan belt owing to hostile climatic and physiographic conditions, one could perhaps better appreciate why the Tethyan belt is not better known than it is, des pi te the fact that it stretches from one end of the "Old World" to the other, uniting like a chain two centres of ancient civilisation: Mediterranean Sea and China. Notwithstanding these difficulties, it has been the Tethyan belt that has continuously shaped our theoretical conceptions of mountain-building A. M. C. !'jengor (ed.), Tectonic Evolution o/the Tethyan Region, 1-22. © 1989 by Kluwer Academic Publishers. 2 and indeed earth's evolution since the times of the ancient Greeks. First theories of orogeny were conceived using Alpine examples, and again on the example of the Alps and the Himalaya that these older theories were falsified. It was in the Alpine foreland that the first important overthrust was discovered and debated in the early 1800' (the Lausi tzer Uberschieburg that passes through Dresden in East Germany), and the widespread occurrence of nappes was confirmed first in the Alpine chain, once more including the Himalaya. The first strike-slip fault was recognised in the Alps by Escher and their widespread occurrence was also noticed on the example of the Tethyan chains by Suess. Why did we find out so much so early about such an enormously large and difficult mountain system? I think there are two main reasons. The first is tradition. Human civilisation began developing in and around the Tethyan system and, following Popper(l) we know that critical thinking, i.e. the habit of criticising others' hypotheses originated in the southern Tethyan foreland, in Alexandria. Natural sciences were thus born essentially in the lap of the Tethyan chains and a tradition became established of studying them at least at their western extremity. Although their eastern extremity was looked at from very different angles, it too was studied, and, put to good use: in China, the legend of the Great Yti, and the splendid irrigation system of Dujiangyan, constructed where the Min Jiang leaves the Longmen Shan by the world's first engineering geologist Li Bing around 250 B.C., testify to a profound understanding of the geology and geomorphology of mountains (see, for example, von Richthofen' s acoount of the Great Yti(2)). This tradition affected also the central parts of the chain and the great physician Ibn Sina (Avicenna) of the present Tadjikistan in the USSR wrote a treatise on the origin of mountains, no doubt inspired by the lofty peaks of the Paropamisus (Band-e Turkestan) range and the Pamirs. Following the re-awakening in the Renaissance, the study of mountain ranges by direct observation wath made fashionable in the Tethyan system: Steno in the Apeninnes (XVII century) and later Johagg Jakob Scheuczher and Horace Benedict de Saussure in the Alps(XVIII century). After the birth of modern geology in the XIXth century, two centres of study stood out from among all others in the world, which devoted a huge amount of energy to the study of the earth in general and the Tethyan chains in particular. One of these was Suess' school of global tectonics in Vienna. This school initiated, in the person of its leader, a comparative method of studying world-wide geology. Its members did palaeontology, stratigraphy, palaeogeography, and also tectonics on a truly world-wide basis. Intellectually these people were true children of the Enlightenment and yet in their attitude towards nature, towards out-of-doors activity, they were Romantics. In the second half of the XIXth century they roamed the Alps, the circum-Mediterranean countries studying mainly the Mesozoic and Cainozoic deposits. They corresponded with their colleagues the world over and prepared the first global 3 palaeogeographic maps. These maps led them to the discovery of such novel concepts as eustatic movements of the sea-level. From a study of their maps they also saw that all hopes of arriving at a perfectly geometric law of mountain formation, such as the one Elie de Beaumont derived, were baseless. They realised and taught in their lecture-halls that the only way to understand orogenic belts was to go out and prepare geologic maps, compile stratigraphic colums, analyse magmatic rocks, look at fossils from biogeographic angles and to distill all these data into syntheses of vast scope. Out of this activity emerged, towards the end of the XIXth century, the greatest geological synthesis ever, Das Antlitz der Erde (The Face of the Earth) by Eduard Suess. Thus, in the second half of the XIXth century the capital of the Austro-Hungarian Empire was like a geological bee-hive collecting and working on geological data to generate hypotheses to be tested in the field. The concept of the Tethys, asymmetric structure of orogenic belts, non-catastrophic views of mountain-building were all products of the Vienna school. Even the discovery of nappes, by Bertrand, happened under the direct influence of the Vienna school. Another extremely fruitful centre of Tethyan research was founded in 1851 in Calcutta under the title of the "Geological Survey of India". Initiated to search for coal, this organisation rapidly developed into a first-class research centre in Asiatic geology and has made first-class contributions to the earth sciences throughout its distinguished history until the fifties. Although the Survey continues its activity to this day(3), its reputation as a leading centre of research in the earth sciences has declined. In the last decades of the XIXth century, a historical link was established between Vienna and Calcutta, when the Survey began hiring Suess' students. One of these, C.L.Griesbach eventually rose to the leadership of the Survey. It was during this episode that the Tethyan research really blossomed. Most of the details of the Triassic stratigraphy were worked out and the past history of the Tethys was clarified. The Survey extended its activity from east Iran to Burma including Afghanistan and Tibet, and thus embraced a very large part of the Asiatic Tethyan chains. This activity, when combined with the information from the European part, began to yield a clear picture of the Alpine-Himalayan system. Even the Ottoman Empire had begun to pitch in its share of information, partly through the industry of its own workers such as Dr .Abdullah Bey (an Austrian political refugee from Vienna: ), and partly through the travels of foreigners such as Piotr Tchichatcheff, and Ami Boue. Then came that eventful day of 28th June, 1914 - a day that commenced a deplorable process that not only put an end to European peace and international scientific collaboration, but also to the activities of the Vienna school. Most of its members had already died, such as Neumayr, Mojsisovics, Bittner, and Uhlig and of the remaining, 4 Fuchs had already retired from active research following his wife's demise. Suess himself had passed away on 26th April 1914, a few months before the beginning of the hostilities. The Vienna school did not survive the war. Along with the Empire that had nourished it, it became a part of history. In post-war Europe, reactionary schools of thought, representing a return to Elie de Beaumont's views, dominated thinking both in Vienna and elsewhere, with two exceptions. One of these was a survivor of the Vienna school, Franz Eduard Suess, Suess' geologist son. The other, of greater proportions, was Emile Argand, the true heir to Suess' throne. Argand lived in Switzerland and therefore escaped the war. His training had taken place under the able hand of Maurice Lugeon, heir to Bertrand, and thus under the indirect influence of Vienna. He and Argand had worked out the geometry of the Pennine nappes, and, during the war, Argand had sketched a theory to explain how they might have formed. This theory Argand called "embryonic tectonics" alluding to the dominant role of nappe embryos controlling Mesozoic-Cainozoic Alpine palaeogeography (concept of embryonic folding had originated with Suess' famous booklet "Die Entstehung der Alpen" in 1875). This theory required too much crustal shortening for the prevailing contraction theory, so Argand chose to abandon the theory and opted for the then novel theory of continental drift. Both Argand and F.E.Suess produced their most influential work on the basis of the theory of continental drift. Although the theory itself had not been conceived on Tethyan examples, its intellectual basis, the presence of two fundamentally different crust types, sal (later sial) and sima, was a discovery of the father Suess. --- ---- Argand and F. E. Suess were the only true followers of the Vienna style, although the former never had had direct contact with Vienna. After their deaths, Tethyan geology became almost entirely an exercise of finding more and more nappes and arguing on their place of origin. The intellectual stimulus to solve what Suess had called "die grossere Aufgaben der Geologie" had vanished along with the Viennese giants themsel ves. With Kober and Stille geologizing became little more than the job of an archivsit, cataloging nappes, orogenic phases, and orogens and kratogens on the face of the earth. Although the Alps became a place of pilgrimage for the world's tectonicists, many of them came to admire the classical outcrops of an Arnold Escher, of a Suess, of a Heim, of a Bertrand, of a Mojsisovics, of an Uhlig, of a Schardt, of a Lugeon, of an Argand ... Thus by World War II, the Tethyan geology had alread y lost its leading role in the earth sciences, despite the delightful eulogies written about it such as the one by Sir Edward Bailey(4). With the rise of plate tectonics, the Alpine-Himalayan system became the type example of a collisional orogenic belt and just one of 5 many orogenic belts to be studied and understood in terms of plate tectonics. In the general enthusiasm of acounting for orogenic belts in plate tectonic frames outside the Alpine-Himalayan system, much of what we had learned from them had not been taken into account sufficiently. Conversely, many Tethyan geologists had difficulty switching their language and in the process were left behind. As a result both Tethyan geology and orogenic geology as a whole suffered. One excellent case in point is the "novel" terrane concept that rose in western North America. If one looks at the history of this concept and that of the nappe concept in the Alps, one would be astonished at the amount of ink spilled about the former almost identical to the latter that is now more than a century old and all of the latter's aspects already overtaken in the Alps! One of the most serious dangers threatening the Tethyan geology now is difficulty of communication among its students owing to linguistic, poli tical, and even financial problems. It was to diminish this and perhaps to initiate a tradition of "Tethyan" meetings that the Ihsan Ketin Nato Advanced Study Institute was conceived. As its history is briefly outlined in the Preface, I omit it here. In the following I give a few general definitions concerning the Tethyan orogenic belts as a whole to assist not only the outsider, but also the geologists specializing on only parts of the Tethyan system to be able to conceptualise the entire belt in his or her mind. SOME DEFINITIONS It is no doubt useful to give here first the original definition of the Tethys, as it was spelled out by Suess. He wrote in 1893 "the folded and crumpled deposits of a great ••• ocean which once stretched across part of Eurasia ... stand forth to Heaven in Thibet, Himalaya, and the Alps" (Suess, 1893, p.183; ref.5). The existence of this "ocean" was first conceived by Suess' son-in-law, the great German stratigrapher Melchior Neumayr, when he compiled and synthesized world-wide stratigraphic and palaeobiogeographic data pertaining to Jurassic strata. From the Caribbean to Burma, sandwiched between a number of northern land-masses and a major southern Brasilian-Ethiopian continent, Neumayr (1885, ref. 6) distinguished a Centrales Mittelmeer. When Suess adopted and renamed this ocean, he also attached to it a tectonic connotation, as the mother ocean of the Alpine-Himalayan ranges. In 1895 (ref.7) Suess indicated that the age of the Tethys reached down into the Triassic (as evidenced by the pelagische Trias). In Das Antlitz der Erde, he pointed out that in the eastern Asia, there were places where the Tethys existed already in the late Palaeozoic. Thus, already at its beginning two points concerning the Tethys concept emerged, which later became subjects of debate: One is that it was originally defined as a tectonic concept on the basis of an older palaeogeographic concept. Suess renamed Neumayr's Centrales Mittelmeer, 6 simply because what he was proposing was not conceptually its equivalent (if it were, Suess would not have needed a new term). The second point pertains to the age of the Tethys. Suess made it clear that it was dominantly a Mesozoic phenomenon, which in places began in the late Palaeozoic and in others lasted into the Cainozoic, even into the present day. Suess emphasized that the present Mediterranean was a remnant of the original Tethys (this is now known to be the case only for the Eastern Mediterranean). After Suess two fundamentally different schools of thought dominated tectonics. One, which I elsewhere called the Kober-Stille school(8) adopted a Beaumontian catastrophist outlook on tectonic phenomena with a strictly determinist philosophy. The other school, which I called Wegener-Argandian(8), adopted a uniformitarian approach similar to Suess' and had a non-deterministic philosophy. The Wegener-Argandians quickly went away from the fixist tectonic theories and developed continental drift, whereas the Kober-Stilleans remained anchored in a fixist world. Corresponding with these two schools, two different conceptions of Tethys evolved after Suess. One viewed the Tethys as a late Proterozoic geosyncline (formed after Stille's Algonkian regeneration) that became progressively consolidated during the Caledonian, Hercynian, and the Al pine orogenic eras. Throughout these times it separated a tectonically dead southern world (Gondwana-Land) from a tectonically lively northern world, as Stille expressed it in 1949. The Wegener-Argandians also thought that the Tethys was a geosyncline, but their conception of a geosyncline was vastly different from that of the fixists (see ref.8). Wegener-Argandians adhered somewhat more closely to Suess' original concept in rear ding the Tethys as a late Palaeozoic 1J.hrough Cainozoic phenomenon, although Argand differed from Suess by considering the Mediterranean a new product separate from the Tethys. Fig.l. shows the various conceptions of the Tethys, including Neumayr's. With the onset of plate tectonics, the whole question of the Tethys took on a new countenance. Bullard et a1. (9) reconstructed the continents around the Atlantic and the Indian oceans without disturbing the torsional rigidity of the continents. One consequence of this was the production of a yawning gap in Pangaea that opened eastwards and separated Gondwana-Land from Laurasia (Fig.2). This was promptly proclaimed to be the Tethys (as had been done before Bullard et al. by Carey in 1958 and Wilson in 1963). Problems arose shortly after this victorious (!) reconstruction that predicted the Tethys without looking at the Alpine-Himalayan system itself! Smith(lO) pointed out that no ophiolites older than the Triassic and no continental margins of similar age could be found in the Alpine-Himalayan ranges to corroborate what the reconstructions predicted. It soon became clear that the classical Tethys really was a Mesozoic creation, an ocean that everywhere had opened not before the Permian at the earliest. As in central and western Europe no orogenies o Z' "E A "N : . . . . . . . . _ , _ _ _ ,_ ,_ ,_ ., -, -. ,, ,, ,: ,, :: :o ·: ~: 1 ~ .... C ro ae " w is e/ le n. d cr l Jo ty aJ en u .. n U rd l. /p in aR ig bm Z 01 U ! . . . . . . . - t. + ~ e r w U M o t de r n iir dl id t. . 9t m iiB ig t£ n. w u l ~ Z o n . G rt! Iu e . w i,s c: /u !,n . dt !r ~ u n d· si i.d lic k 9e mi i1 ii g~ n Zo ru . \r~ ~ . tIu sc i£ Iu uu '!J dJ uf Jm uk .If U r.T ur aM it. A N '1 'A R K '1 'l S C H " E R 0 7 ." E A N Fi g. 1A : N eu m ay r's C en tr al M ed it er ra ne an , pr ed ec es so r o f th e T et hy s c o n c e pt . th e pa la eo ge og ra ph ic - . l 00 (.~~ . 1- I 1 8 neoi di sc /'e ;e t/1 ys :I~,ar: sz'sc" I ~ " ¥ I I ~ ~ ' \ , I ! ko ns o/ ld le rte . / \. I I / I : , , i ' d , n i s e I > . ' , _~ _ _ ! ~ Kole 0 r e lh ys ro um e£ .' \ ~_ ,: ~--J-- -__ )_ _ _ l- ; I : , j, I / I / I ' / I _ , F 7 1 os sy nt ls ch ~ \ . . . I , . I I , / , ;. 1 = \ \. .. , .. -"_ ~ __ ~ _ L ,; ,. L_ __ __ __ --- =-= -:£ ;:, ;~' ~_\ L ,1:. _ _ L . . - 80 0 F ig .l B : S ti ll e' s fi x is t v ie w o f th e T et hy s. 9 Fig.1C: Argand's mobilis view of the Tethys. had intervened between the end of the Hercynian events and the onset of Alpine deformation, this new situation led to pessimism concerning the success of plate tectonis and to expanding earth models. In Asia, however, the situation was different. Even in southern Europe, Suess had already indicated the presence of orogenic events of middle Mesozoic age, located in an orogenic system independent of his Alpides, and following Prof.Mrazec's suggestion had called this new system, of which he was able to reconstruct only incomplete fragments, the Cimmerian Mountains, after the Cimmerii the oldest known inhabitants of the NW parts of the Black Sea shore, according to the recitations of Homer. Stocklin announced the discovery of some pre-Jurassic ophiolites in NE Iran,near the town of Meshhed, and pointed out that these could possibly be the remnants of an older Tethys, the one suggested by the reconstructions of Bullard et al. and their successors (e.g.(ll)). In 1978 Ken HsU came to Turkey to attend a meeting of the C.I.E.S.M. in Antalya along the south coast. We had agreed that he would fly to Istanbul and that we would drive down to Antalya while looking at the geology of western Turkey along a N-S cross-section. The first night on the way to Antalya was spent in Bursa, after Ken had seen the Intra-Pontide suture and in anticipation of seeing, on the next day, the lzmir-Ankara suture. In Bursa Ken continuously questioned me whether any of these could be the suture of the older Tethys, what Stocklin had called the Palaeo-Tethys. I laid out the observations to him and he became convinced that they could not be. "Then" he said "you must look Fi g. 2 : An ' o rt ho do x' p la te t ec to n ic r e c o n s tr u ct io n o f th e T et hy s. T hi s m ap is f ro m t he P ha ne ro zo ic C on ti ne nt al M ap s by S m ith , H ur le y, a n d B ri de n (1 98 1, Ca m br id ge U ni ve rs it y Pr es s) o 11 for it north of Turkey, in the Dobrudja, in the Crimea, .where the Cimmerian orogenic events may be the witnesses of its closure.". A year later, while I was preparing some lecture notes on the geology of Turkey in Albany,N.Y., I came across references to pre-Liassic ophiolites, ophiolitic melange, and orogenic deformation in N Turkey, in the Pontide tectonic unit of Ketin(12). I jumped out of my seat: Ken was right! Palaeo-Tethys had been there, its evidence staring at me out of the papers of Ketin, Blumenthal, Fourquin, and Radelli, papers which I had read on several previous occasions never thinking that here was the suture zone of the Palaeo-Tethys. A quick review of the geology of the Dobrudja and the Crimea, along with Stocklin's papers on Iran left no room for doubt that the Cimmerian zone of deformation indeed was a suture zone of mid-Mesozoic age (pre-late Jurassic in Turkey, older in Iran) and that it was located N of the Palaeozoic of Istanbul, i.e. it had no contact with the Neo-Tethyan sutures farther S. Later collaboration with Ketin, Yticel Yllmaz, and Ozan Sungurlu largely clarified the first order organization of this older orogenic bel t and pointed out where the new research effort should be located. I undertook a synthesis of it throughout the Tethyan belt and showed that the closure of Palaeo-Tethys was a complicated affair, involving the generation of a veritable orogenic collage ($engor, 1984, ref.13). At this time new work in Iran (mainly by Berberian and his colleagues), in Afghanistan (mainly by the French geologists), and in China (mainly by Chinese geologists) brought in much new data. A number of Soviet scientists, mainly Belov, Adamia and his colleagues in the Caucasus, the late Viktor Shvolman in the Pamirs, and the late Academician Peyve and his group in the whole of Asia tracked down the remnants of the Palaeo-Tethys. In Bulguria, the painstaking field work of G.A.Chatalov in the Strandja documented the complicated structure of the Cimmeride orogen that involved oceanic rocks i Chatalov's Strandja-type Triassic. A number of these new results are contained in the papers in this book. Fig.3 shows the suture network of the Tethyan chains, whiCh I called the Tethysides, that divide it into innumerable blocks. The new conception of the Tethys and its evolution necessisated the generation of the following nomenclature: Al pine-Himalayan System: This is the morphologic mountain range that stretches from the Pyrenees and the Betic and Riff cordilleras through the Alpine chains of Europe, the Carpathians, Dinarides, Hellenides, the Turkish chains, Iran, Afghanistan, Pakistan, Tibet and Burma into Indochina and Indonesian Archipelago. As seen in Fig.4, this system is accompanied by an apron of lesser mountains to the north (e.g. the Mittelgebirge in German, the Tien Shan in Central Asia, etc.), which also formed as a result of orogeny in the Alpine-Himalayan system. Tethys sensu lato: Includes all Tethyan oceans without connotation of age and thus includes both the Palaeo- and the Neo-Tethys. Tectonically it includes the oceans and their continental margins including the shelves, and I have often recommended that the term Tethys be confined 12 ~~ Alpide suture (ticks on upp2r plate) ~ ~ ~ Alpide sutures of 'sialic oceans' 11111111111111:111 Alpide subduction/accretion complex U I on oce.anlc 5ubstro1tum / Cimmeride suture (ticks on upper plate ~ /polarity unknown) •.• " ...• Conj£ctural Cimmeride suture ~-_-= Cimmeride subduction/accretion complex - -- on oceanic sUbstratum Boundary of major tectonic - ~- subdivisions of Eurasia 0 1.1 __ --1 .. ! Precambrian j consolidation . Areas of Poleozoic outside the Mesozoic-Cenozoic Tethysides :::::: Cimmerian •••••• Continent III : i : Oceanic crust (underwater / onland) ..A......A.. Active subduction (teeth on upp2r plate) Fig.3: Generalized tectonic map showing the major tectonic subdivisions of Eurasia and the suture distribution within the Tethysides. Cimmeride sutures: I- Paleo-Tethyan suture in the Balkan/Carpathian Cimmerides, II- Karakaya, III- Luncavita-Consul, IV- North Turkish, V- Paleo-Tethyan suture in the Caucasus, V- Chorchana-Utslevi zone, VI- Talesh-Mashhad, VII- Waser (Farah-Rud), VIII- Paropamisus-Hindu Kush-North Pamir, VIII- Kopet Dagh, IX- South Ghissar, X- Northern "synclinorium" of western Kuen-Lun, X'- Qiman-Dagh, XI- Altln Dagh, XII- Suelun Hegen Mts., XIII and XIII'- Inner Mongolian, XIV- Suolun-Xilamulun, XV- Da Hingan (G.Khingar), XVI- Tergun Daba Shan-Qinhai Nanshan, XVII- Southern "synclinorium" of western Kuen-Lun, XVIII- Burhan Budai Shan-Anyemaqen Shan, XIX- Lancan Jiang-Litien. XX- Jinsha Jinag, XXI- Litang, XXI'- 13 Luochou "arc-trench belt", XXII- Banggong Co-Nu Jiang, XXII'- Mid-Qangtang, XXIII- Shiquanhe, XXIV- Southwest Karakorum, XXV- Nan-Uttaradit-Sra Kaeo, XXVI- Tamky-Phueson, XXVII- Song Ma (Red River), XXVIII- Song Da (Black River), XXIX- Bentong-Raub, XXX- West Borneo, XXXI- Qin-Ling, XXXII- Longmen Shan-Qionglai Shan, XXXIII- Mid-South China. XXXIV- Korea, XXXV- Helan Shan, XXXVI- Mandalay, XXXVII- Shilka. Alpide sutures: 1- Pyrenean, 2- Betic':', 3- Riff*, 4- High Atlas*, 5- Saharan Atlas*, 6- Kabylian*, 7- Apennine, 8- Alpine, 9- Pieniny Klippen melt, 10- Circum-Moesian, 11- Mures, 12- Srednogorie, 13- Peonias-INtra-Pontide, 14- Almopias-Izmir-Ankara, 15- Pindos-Budva-Blikk, 16- Ilgaz-Erzincan, 17- Inner-Tauride, 18- Antalya, 19- Cyprus, 20- Assyrian, 21- Maden, 22- Sevan-Akera-Qaradagh, 23- Slate-Diabase zone, 24- Zagros, 25- Circum-Central Iranian microcontinent, 26- Oman, 27- Waziristan, 28- Kohistan sutures, 29- Ladakh sutures (northeast: Shyok; southwest: Indus), 30- Indus-Yarlung-ZAngbo, 31- Burma, 32- Mid-Sumatra, 33- Meratus (Asterisks indicate sutures of sialic oceans sensu $engHr & Monod 1980). Tethyside block: a- Moroccan Meseta, b- Oran Meseta, c- Alboran, d- Iberian Meseta, e- African promontory, f- Rhodope-Pontide, g- Sakarya, h- Klr.;;ehir, i- North-west Iran, j- Centarl Iranian, k- Aghdarband arc, 1- Farah, m- Helmand (sensu $engHr 1984), m'- Kohistan arc, n- Western Kuen-Lun Central Meganticlinorium, 0- Qaidam, p- Alxa, q- North China (Sino-Korean) platform, r- North China fold belt, s- Qangtang (possibly divided into s'- East Qangtan ad s", West Qangtan: Chang Cheng-fa, oral communication, 1985), t- Lhasa (possibly divided into t'- Bongthol Tangla, t"- Nagqu, and t"'- Lhasa proper: A. Gansser, oral communication, 1985), t""- Ladakh arc, u- Shaluli Shan arc, v- Chola Shan arc, w- Yangtze, x- Annamia, y- Huanan, z- Songpan Massif. E, M, and S are East Anatolian, Makran, and Songpan-Ganzi accretionary complexes, respectively. 14 40 70 20 20 10 • >6000m E2] 4000- 6000 m . ........ 0 km 1000 2000 - 4000 m. I ! CJ 1000-2000m o s r A B R A BAY OF BENGAL INDIAN OCEAN 80 90 «.. \ 80 50 1'70 20 Fig. 4: THe Al pine-Himalayan system of mountain ranges in the overall topographic framework of Eurasia. 15 to tectonic usage, as it was originally intended. When used in a palaeogeographic or palaeobiogeographic sense, it includes all equatorial seas of + Mesozoic age and frequently generates confusion (cf. ref .13). Palaeo-Tethys: The ocean and its continental margins that formed as a by-product of the assembly of the Pangaea at the end of the late Carboniferous as depicted in Fig.2. It is not an exclusively Palaeozoic ocean, but its largest parts survived into the Jurassic. It simply means "old Tethys". Neo-Tethys: The ocean and its continental margins that formed as a result of the closure of Palaeo-Tethys between the Cimmerian continent and the northern margin of Gondwana-Land. Tethyan domain: Region affected by deformation that resulted from the closure of the Tethys sensu lato. Like Tethys sensu lato itself, it is a tectonic term and should not be used in palaeogeography or palaeobiogeography. Tethysides: This term, from and (shape, form, kind; used also to imply family connection in the sense of being of the same kind), signifies the orogenic system that arose from the destruction of the Tethys ensu lato and covers the entire Tethyan domain. It consists of two major elements: Cimmerides: The orogenic system that resulted from the destruction of the Palaeo-Tethys and her dependencies, such as marginal basins, back-arc basins, etc. It comes from combining and Both the Cimmerides and the Alpides include alpinotype and germanotype areas of deformation. Alpinotype areas correspond with orogenic belt proper, with deformation and metamorphism penetrative on the scale of mm. to hundred metres. By contrast, the germanotype areas correspond with fore- and hinterland deformation fields consisting of blocky, non-penetrative structures, commonly including much strike-slip and not much metamorphism or magmatism. All of these terms were defined in refs.(13_ and(14), to which the reader is referred for further details. EVOLUTION OF THE TETHYS IDES Fig.5 shows a sequence of maps depicting the evolution of the Tethyan domain it its gross outlines as I conceivenit in 1984. It shows the existence of Palaeo-Tethys in late Palaeozoic time, its progressive closure by the rotation of the Cimmerian continent and by the agglomeration of a number of exotic continental fragments to Asia in the extreme east. The rotation of the Cimmerian continent at the same time 16 Fig.SA: Schematic reconstruction showing the palaeo tectonics of the Tethyan domain in the late Permian (Kazanian). Abbreviations for all reconstructions: A- Afghan blocks, An- Annamia, B- Bitlis/Pottirge fragment, BNJ-Banggong Co-Nu Jiang ocean CI- Central Iranian microcontinent, CS- Chola Shan, d- Dnyepr-Donetz aulacogen, F- Farah block, H- Helmand block(sensu ~engor 1984), IBF- Istanbul-Balkan fragment, IR- Iran block, K- Klr~ehir block, L- Lhasa block, LB- Luochou arc, MVL- Mount Victoria Land block, NC- North China block, nc- North Caspian depression, No- Northern branch of Neo-Tethys, p- Pachelma aulacogen, Q- Qangtang block. Qu-Quetta graben, RRF- Red River fault, S- Serindia. Sa- Sakarya continent, SB- Yangtze block, SECB- Huanan block, SG- Songpan-Ganzi system, ShS- Shaluli Shan arc, SIBUMASU- China-Burma-Malaya-Sumatra portion of the Cimmerian continent. So- Southern branch of Neo-Tethys, T- Turkish blocks. Fig.SB: Early Triassic (Induan) paleotectonics of the Tethyan domain. 17 PACIFIC OCEAN Fig.5C: Late Jurassic (Volgian) paleotectonics of the Tethyan domain. PACIFIC OCEAN Fig.5D: Late Cretaceous (Cenomanian) paleotectonics of the Tethyan domain. Fig.5E: Middle Eocene (Lutetian) paleotectonics of the Tethyan domain. 18 o C E A N ------..SANTARCTICA Fig.SF: Late Miocene (Vindobonian) paleotectonics of the Tethyan domain. opened Neo-Tethys in its wake, which, during the Mesozoic and the Cainozoic closed by the disintegration of Gondwana-Land, and the collision of its various pieces with Eurasia. Ongoing research since 1984 has shown that although the gross outlines of this evolutionary scheme may be correct, its details are wrong on the scale of loS km. mainly because of extensive strike-slip motion. Fig.6 and Table I show the main ones of the strike-slip faults that disrupted the Tethyside orogenic collage during and after its construction. The concept of allochthonous terranes, as used in North America, has not been applied to the Tethysides by any of the speakers in this ASI. This is not the place to criticise the terrane concept, but in many parts of the Tethysides, and most notably in the European parts of the system, the identification of "fault-bounded terranes with geological histories differing from one terrane to the other" has been largely accomplished by the thirties and since concepts involving genetic considerations have superceded them. The importance of strike slip faulting has been recognised too on the scale of nappes(e.g. refs.IS and 16) as well as on the scale of mountain ranges (ref.17). The workers in the Tethyan realm would find it hard and perhaps retrogressive to go back to purely descriptive concepts, whose sole contribution would be simple cataloging. That the Tethyan workers have missed the significance of strike-slip in orogenic processes is amply negated by such thoughtful papers as that of TrUmpy(18). EPILOGUE: PROSPECTS Tethyan research has entered an exciting new phase since the advent of the theory of plate tectonics. One of the most important tasks now awaiting Tethyan geologists is the identification of suture zones and A ~:~~ke~s~~n/-h'YPol'hesiZ'dl ~D + +++ .?+++ ++ 1 ~~~ Post-suturing strike-sl I ;1 ............ Pre- syn- & post I I - - - suturing strike-slip b .. - Tethyside sutures o """""--- Subduction zone -"'- Suture zone. -+- -+- + Arc magmatism 0000 Collisional magmatism Strike.-slip re.late.d magmatism ~4 =5 0006 111111 8 19 Fig.6: (A) Map showing pre-, syn-, and postcollisional strike-slip motion along the Tethyside sutures. For numbers and sources, see Table 2. (B, D) Diagrams showing possible complications introduced by strike-slip faulting during and after continental collision. (C) Sketch map of the geology of the Chorchana-Utslevi sheared suture in the Dzirula Massif. (E) Map showing the present-day tectonic settings of pre- and syncollisional strike-slip faults in eastern and southeastern Asia. Key to numbers: 1- subduction zone (teeth on upper plate), 2-young zone of collision, 3- strike-slip fault, 4- normal fault (ticks on hanging wall), 5- spreading centre, 6- arc magmatism, 7- aseismic ridge, 8-oceanic area. Abbreviations: BA- Banda arc, CBR- Central Basin Ridge. MTL- median tectonic line. OT- Okinawa trough, PaT- Palau "Trench", PF- Philippine fault, PKR- Palau-Kyushyu Ridge. 20 Table I: Sense, amount, and timing of strike-slip motion on some Tethy- side suturesa Suture b Sense Amount Time (km) l(XI) Left ? ePale ecX+XI) Left ? ?C-P 3(II) Right 1000 C-?eTr 4(XIV) Left 4000 P 5(XVIII-XXXI) Left 4300 1200 P-eJ 6(25) Right 1000 ITr-IK 7(V' ) ? ? P-ITr 8(8) Left 1500 eJ-eK 9(4-5) Left 10-20 e-mJ Right 10-20 IJ-eK 10(7 south) ? ? eJ 1l(7north) ? ? m-IJ l2(XXII ) Left 500 IJ-eK 13(XIV) Left ? IJ-eK 14(1) Left 150-200 e-IK 15(14) ? ? IK 16(23) Left 350-700 IK-Pal 17(17,21) Right 1000 IK-eE 18 (2) Right 300 IE-1M (3) Left 19 (XXXVI) Right 460 IOI-present 20(27) Left 200 u-mT 21(7 south) (north)Left 50 T (south)Right 22(XXVII) Left 500 OI-M Right A few tens Q 23(XXV) Left 300 OI-M 24(VIII) Right ? end E-?M 25(25) Right ? post eM 26(VII) Right ? end E 27(14) Right 100 M-Pli 28(13,16) Right 100 1M-present 29(X,XI) Left 500-600 M-present 30(30) Right 200 IT 31(24) Right 60- Pli-present 32(20,21) Left 20 Pli-present 33(XXI,XXI') Left ? IT 34(XVIII) Left ? IT 35(XXXI) Left Several hundred post E aAbbreviations: Pale-Paleozoic, C- Carboniferous, Triassic, J- Jurassic, K- Cretaceous, T- Tertiary, Eocene, 01- Oligocene, M- Miocene, Pli- Pliocene, early, m- middle, 1- late bIn the enumeration of sutures, left-hand numerals Figure 5A and right-hand numerals (in parentheses) Figure 1. P- Permian, Pal- Paleocene, Q- Quaternary, refer to those refer to those Tr- E- e- in in 21 the characterization of the blocks they separate, including the establishment of the path of drift before these blocks became a part of the Tethyside collage (and in most cases after they became a part of the Tethyside collage, as a result of the formation of the collage). To this end we need a lot more and a lot more detailed geologic mapping coupled with careful biogeographic analysis, palaeomagnetic sampling, and geochemical and isotopic work. Tethyan geologists must familiarize themselves with the active subduction and transform fault systems better than hitherto, in order to understand the past environments they study (this message is amplified by Warren Hamilton in his paper in this book). One of the most important points that emerges in this book is the importance of field research in all parts of the Tethyan chains. REFERENCES 1. Popper,K.K., 1966. The Open SOCiety and its Enemies, v.1, Princeton University Press, Princeton 361 p. 2. Richthofen,F. Freiherr von,1877. China. Ergebnisse Eigener Reisen und Darauf Gegriindeter Studien, v.1, Dietrich Reimer, Berlin, 758 pp. 3. Anonymous, 1976. One Hundred Twentyfive Years of the Geological Survey of India (1851-1976). A Short History, Dipti Printing and Binding Works, Calcutta 58 pp. 4. Bai1ey,E.B., 1935, Tectonic Essays, Mainly Alpine, Oxford Univ. Press, Oxford, 200 pp. 5. Suess,E., Are great ocean depths permanent? Nat. Sci., v.2, p.180-187. 6. Neumayr,M., 1885. Die geographische Verbreitung der Juraformation, Denkschr. k. Akad. Wiss. Wien, Math.-Naturwiss. C1., v.15, pp.57-114. 7. Suess,E., 1895. Noter sur l'histoire des oceans, C.R.hebd. Acad. Sci. Paris, v.121, pp.1113-1116. 8. $engor,A.M.C., 1982. Eduard Suess' relations to the pre-1950 schools of thought in global tectonics. Geo1. Rundsch., v.71, pp.381-420. 9. Bullard,E.C., Everett,J.E., and Smith,A.G., 1965. The fit of the continents around the Atlantic, Phil. Trans. R. Soc. London, v.A258, pp.41-51. 10. Smith,A.G., 1973. The so-called Tethyan ophiolites: in Tar1ing,D.H., and Runcorn,S.K., edts., Implicationsof Continental Drift to the Earth Sciences, v.2, Academic Press, London, pp.977-986. 11. Smith,A.G., and Ha11am,A., 1970. The fit of the southern continents, Nature, v.225, pp.139-144. 12. Ketin,I., 1966. Tectonic units of Anato1ia (Asia Minor), Bull. Miner. Res. Exp1. Inst. Turkey, v.66, pp.23-34. 13. $engor,A.M.C., 1984. The Cimmeride Orogenic System and the Tectonics of Eurasia, Geo1. Soc. America Spec. Pap. 195, xi+82 pp. 14. $engor,A.M.C., 1985. Die A1piden und die Kimmeriden: Die verdoppe1te Geschichte der Tethys, Geo1. Rundsch., v.74, pp.181-213. 22 15. Argand,E., 1920. P1issements precurseurs et p1issements tardifs des chaines de montagnes. Act. Soc. he1vet. Sci. Nat., Neuchate1, pp.1-27. 16. Schwinner,R., 1951. Die Zentra1zone der Ostalpen, in F.X.Schaffer. edt., Geologie von Osterreich, Franz Deuticke, Wien,-Pp.10S-232. 17. Ketin,I., 1948. Uber die tektonisch-mechanischen Fo1gerungen aus den grossen anato1ischen Erdbeben des 1etzten Dezenniums, Geol. Rundsch., v.36, pp.77-83. 18. Triimpy,R., 1976. Du Pelerin aux Pyrenees. Eclog. Geol. Relvet., v.69, pp.249-264. ONE SOME KEY FEATURES OF THE EVOLUTION OF THE WESTERN ALPS Jacques DEBELMAS Geology Dept. University, F 38031, Grenoble, France. ABSTRACT. Some key features control the modern interpretation of the Western Alps. During the Triassic-Cretaceous distensive phase, two facts are outlined (I) the spreading of the Piemont-Ligurian oceanic realm through a rifting process which is still in discussion, (2) the pattern of the two opposite continental margins in tilted belts or blocks, a few of them being still almost undeformed and showing their primitive NE-SW trend. During the Tertiary compressive phase, the following points are developped (I) the difficulties about the disparition by subduction of the oceanic Tethyan crust, (2) the interpretation of the belt as the stacking up of basement slabs, progressing to more and more external areas. This process seems to be still active. (3) The analysis of the mechanical and mineralogical effects of the successive phases of deformation. RESUME. Un certain nombre de caracteres paleogeographiques et structuraux sont evoques, qui commandent l' interpretation actuelle des Alpes occidentales. Pour la phase de distension Trias-Cretace, on insiste (I) sur l' appari tion de l' ocean liguro-piemontais par un processus de rifting dont les modali tes restent encore conj ecturales, et (2) sur la disposi tion en blocs bascules des deux marges continentales opposees, blocs dont certains ont pratiquement garde leur orientation primitive, NE-SW, qui donne ainsi celIe de l'ocean alpin. Pour la phase compressive Tertiaire-Actuel, on insiste sur les obscuri tes liees a la subduction de la croute oceanique ligure, sur l'interpretation actuelle de la chaine comme un empilement d'ecailles de socle se faisant progressivement de l' interieur vers l' exterieur et probablement toujours actif enfin sur l'analyse des consequences mecaniques et mineralogiques des phases de deformations successives. The Western Alps are a classic area of the Mediterranean belts, and numerous synthetic descriptions have been published during the last years for instance, Debelmas & Kerckhove (1980), Debelmas & al. (1983), Trlimpy (1960, 1980). 23 A. M. C. !'Jengor (ed.), Tectonic Evolution of the Tethyan Region, 23-42. © 1989 by Kluwer Academic Publishers. 24 The main outlines of these papers remain valid and do not need to be repeated. More interesting is to present some features of the Alpine evolution related to the current assumptions of Global Tectonics or to new geophysical data supporting our present interpretation of the belt. Fig.l. Structural scheme of the Western Alps (GR. Grandes Rousses massif) . Grey : External Crystalline massifs ; Wider stippled : Penninic zone Densely stippled Helminthoid flysch nappes and connected slices Vertical hatching: South-Alpine zone, with its crystalline basement. It is however essential to recall briefly the structural terminology of the Western Alps. They show a succession of structural zones (fig. I), each of which is characterized by its own stratigraphic sequence and tectonic style, and appears as a well-delineated 25 paleogeographic domain, separated from those adjacent by probably distensive breaks that evolved later into thrust faults. Thus, paleogeographic domains and structural zones are parallel. Moreover, the Alpine belt, like most collision chains, formed along an ophiolitic suture zone, the remains of a disappeared oceanic realm, the so-called Ligurian-Piemont realm. The two continental margins of this oceanic basin came into contact and, as a consequence of their collision, were fragmented and folded, the resulting structures verging to the northwest in Switzerland, to the west in France, or to the southwest in the Italian Maritime Alps. Table I gives the names of the main paleogeographic and structural units, which are not described further. External zone I n t e r n a I z 0 n e s Dauphine zone I Pe n n i n i I South-Alpine zones c z one s I Helvetic zone Su bb"an~on'ZJB"anconna.s. z. 'Piemont zone,Liguro-Piem.z I Sesla zone Ivrea zone Continental marg In: submarine slope w.th tilted blocks Oceanic area I Conti nental margin helvetic. nappes SUbbnanconn.i Bnanconna.s I SChistes lus~res nappes nappes+slices nappes Iw,.hOU.OPhOO.itesl wolh oph,o.ites / Prealps Simmenappe w scanty S·Alplne Nappe of the a: Breche nappe Helmintho.d :l cover Median Prealps ... Flysch nappe :l cover '" folds 0 (Switzerland) 26 The distensive process began in the Late Triassic, during the so-called "Carnian crisis", which generated a horst-graben dynamic topography with varying subsiding rates. Some blocks subsided and recei ved a sedimentation of "Hauptdolomi te" type, while others were uplifted and eroded. An associated volcanism is known at this time (Middle Triassic to the Upper Carnian) in Italy (Bosellini & al., 1982). The platform later experienced more prominent crises, now well dated at the Hettangian-Sinemurian and Domerian-Toarcian boundaries (Lemoine 1984, 1985). In late Mid-Jurassic or early Late Jurassic, the Ligurian-Piemont oceanic crust started to form by spreading, and the first sediments deposited upon it have been paleontologically dated as Upper Oxfordian-Kimmeridgian (De Wever & Caby, 1981). The pattern of the oceanic realm is still a matter of debate and many sketches are available. 1) The oceanic realm may be interpreted (Bernoulli & Lemoine 1980, Lemoine 1985) as a pull-apart basin, the Ligurian rhombochasm, trending roughly northeast-southwest (see below for evidence). The spreading of this basin implied two transform faults, a northern North-Penninic fault and a southern South-Tethyan fault (fig.2). Fig.2 Palinspastic sketch of the Mesozoic Tethys at the end of the Jurassic (after Lemoine, simplified) 1. North-Penninic transform fault; 2. Gibraltar-Sicily transform fault. On both sides of the oceanic gap, the continental margins (the European margin to the northwest, the South-Alpine margin to the southwest) were thinned and fragmented into blocks tilted along listric faults, producing a horst (or semi-horst) and graben (or semi-graben) topography that controlled the sedimentation rate. This topography of more or less parallel belts has been preserved in the less deformed parts of the Alpine area, i. e. on the edge of the French Central Massif, on the one side, and in the Italian Southern Alps, immediately north of the Po basin, on the other. 27 On the edge of the French Central Massif, these blocks trend northeast-southwest, parallel to the late Variscan faults, the so-called "Cevenol" faults, which controlled the Mesozoic paleogeography as far as they were listric faults between delimited blocks. \ , .... Fig. 3 Palinspastic reconstitution of the Western Alpine area in the Early Cretaceous. Two possibilities. AA. Austroalpine zone ; B. Briangonnais zone ; D. Dauphine zone Helvetic zone; J.F. Judicaria fault; LP. Liguria-Piemont zone North-Pyrenean fault; P. Piemont zone ; PR. Provence platform ; South-Alpine zone; V. Valais zone. ; H. NPF. SA. Little cross lines : paleogeographical features (1. Insubria-Lombardia ridge - 2. Belluno through) Stippled: continental thinned crust ; black : oceanic crust. 28 In the Italian Southern Alps, the Alpine stress has also been too weak to destroy the Mesozoic traces of the same distensive crisis, well documented in the stratigraphical sequences (Castellarin 1982). Their north-south to northeast-southwest trends classically support also the assumption that the oceanic basin trended northeast-southwest. But a problem arises here from the fact that the paleomagnetic data seem to indicate a post-Triassic counterclockwise rotation of 50 0 of these Italian Southern Alps with respect to the European foreland (Vandenberg & al., 1976). Replaced in their primitive position, the above mentionned trends become E-W (fig.3) in a more complex paleogeographical pattern but a pattern which allows a better connection with the "Vardar oceanic realm" to the East (Debelmas et Sandulescu, 1987). In the more strongly folded areas of the Alpine domain proper, the main northeast-southwest trends are more difficult to observe, except in the External Crystalline Massifs of Belledonne, Grandes Rousses and Pelvoux where numerous Jurassic synsedimentary faults have been recognized in recent years (Lemoine & al. 1981), (fig.4). 500 ~ 11.. ______ ---' 10km Calcolres • entroques '" -- _ .... Calcalreo • en""qu". Fig.4 Reconstruction stratigraphic features 1984) . of on the the Jurassic tilted Belledonne-Pelvoux blocks and traverse related (Lemoine, Upper section present state of deformation of the Ornon-Rochail semi-graben. A-Bj. Aalenian-Bajocian ; To. Toarcian ; Do. Domerian Si. Sinemurian ; H. Hettangian; TR. Triassic, Calcaires a entroques : crinoidal limestones. Ca. Carixian This new appraisal of the Alpine paleogeography allows a new interpretation of the so-called "Valais-zone", at the front of the Penninic belt, but restricted to the Swiss Alps and to the northern part of the French-italian Alps. This zone would be a transform zone, along the North-Penninic fault, bounding on the north the Ligurian pull-apart basin (fig.2). Its detailed paleogeography is still rather obscure and currently being studied. 29 2) For other geologists (Dercourt et al., 1985), the oceanic realm is not a pull apart basin, but a California-type gulf, i.e. a relatively narrow trough, the general trend of which would be rather obliquous to the accretionnal axis, this latter being cut by numerous transverse faults (fig.5). Fig.5 Palinspastic sketch of the Mesozoic Tethys at the end of the Jurassic (after Ricou et Dercourt, simplified) LP. Ligurian-Piemont oceanic area. In the N, this narrow Alpine oceanic trough would be connected to the Vardar ocean by a transform zone, distinct from the Valais zone. This latter would be another spreading trough, of less oceanic character, extending in the NW of the Ligurian-Piemont realm, but rapidly closing southwards (see also TrUmpy 1980). The intermediate platform could be the northern prolongation of the Brian90nnais domain. The authors do not describe the continental margins of these more or less oceanic furrows but it is always possible to consider that they are stretched and fragmented into tilted blocks, even if there is here no significant symetry between the two borders of the Tethyan realm. The preceding paleogeographical reconstitutions are both supported by the stratigraphical field data. 30 Generally speaking, the stratigraphical sequences are now well known in all Alpine zones, except in the Piemont and Ligurian~Piemont zones where mono1;:onous, deep-sea clayey-carbonate sediments were deposited thes,e sediments were transformed by the Alpine tectonics and metamorphism into schists, the famous "Schistes lustres", for a long time stratigraphicaly undeciphered. The Schistes lustres have now been compared with the Mesozoic deap-sea deposits of the Atlantic ocean (Graciansky et al. 1979). For instance, the "Lower black shales" (Oxfordian) and the "Upper black shales" (Aptian-Albian) of the Atlantic sediments are now reference levels that have been identified almost everywhere in the Alpine series, even in the Piemont and Ligurian-Piemont sequences. In the Ligurian-Piemont series however, only the Upper black shales have been observed, for these series begin with the Oxfordian, as mentioned above. Thus, the Schistes lustres include in their outer part (Piemont zone of the French geologists) a complete sequence from Lias to Upper Cretaceous, but in their inner part (Ligurian-Piemont zone on the French geologist) only Oxfordian to Upper Cretaceous, thinner and more oceanic. Of course, there is local variability, as the Piemont zone too probably consisted of tilted blocks on which sedimentary sequences of varying thickness and composition were deposi ted. Thus, intercorrelation between the sedimentary sequences is difficult, but work is in progress. B. The compressive phase 1. From the Aptian-Albian and during the Late Cretaceous. The relative north-south motion of Africa toward Europe (Patriat & al., 1982) ended the Tethyan spreading and initiated the closing of the Ligurian rhombochasm. All authors agree that this closure was due to subduction of the European margin south or southeast below the South-Alpine margin. Blue schists, dated at 100-80 Ma (Bocquet & ale 1974, Hunziker 1974), and eclogitic facies in ophiolites (perhaps as old as 130 Ma, Oberhansli & ale 1985) are thought to have been connected with this subduction. The detailed pattern of subduction is not clear, however, and for several reasons : 1. Where exactly was the ophiolitic suture ? Most authors place it just west of the Sesia zone, where strongly squeezed ophiolitic bodies occur (Viu-Locana slices) the Sesia zone would belong to the South-Alpine margin (Table I). But most or all of the Sesia gneisses have been metamorphosed under HP conditions and are considered as also carried down in the sUbduction zone (Dal Piaz & ale 1972). Moreover the southern Sesia zone is separated from the South-Alpine domain by the Canavese slices, which contain possible Mesozoic ophiolites (7) and Schistes lustres (7) showing very low grade (Alpine 7) metamorphism. 31 Hence a few authors (Aubouin & al. 1977 Mattauer ~t al., 1987) considerer the Sesia zone as a part of the European margin, in spite of its close lithological similarities to the Austroalpine Dent Blanche klippe, and consider the Canavese slices as the true ophiolitic suture in spite of their lack of HP metamorphic facies and their equivocal nature. 2. Where is the calc-alkaline magmatism related to the subduction process? Such magmatism is missing and several explanations have been put forward, among which are (i) the small size of the vanished oceanic realm, (ii) the steep dip of the subduction plane, or (iii) the possibili ty of an early intra-oceanic subduction instead of the more commonly assumed subduction marginal to the South-Alpine margin. According to the latter hypothesis, the calc-alkaline magmatic rocks would have disappeared in the collision zone together with the underlying oceanic lithosphere. The explanation of an early intra-oceanic subduction leads to the interpretation of the huge Ligurian-Piemont ophiolitic bodies as simply obducted, with or without any further subduction processes. In both cases, the HP metamorphism would have taken place at the base of this ophiolitic edifice, later carved out by the erosion. 3. At what time did the subduction initiate ? This time is not easily determined, especially as deposition persisted throughout at the surface, without any special disturbance. The HP metamorphism has been dated as from 130 Ma (Early Cretaceous) to 100-80 Ma (Albian to Early Senonian). In the Eastern Alps, field evidence suggests a late Early Cretaceous crisis, when the earliest nappes emplaced at the South-Alpine or Austroalpine front, which is consistent with recent data about Atlantic kinematics (Patriat & al., 1982). In the Western Alps, we have no such tectonic evidences but only synorogenic flysch-type sediments, the earliest of which are Cenomanian to ,Early Senonian, with rare ophiolitic olistolites. The problem remains open. 4. At what depths did the superficial material sink and how did it rise afterwards ? During the Late Cretaceous, the most part of the Ligurian-Piemont material sank down to gain a high grade blue-schist facies. For instance, the Zermatt ophiolites and related units were metamorphosed at 10-12 kb at 100 My (Desmons 1977, Frey and al., 1974), the Monte Rosa complex at as much as 16 kb between 70 and 100 my (Frey and aI., 1976). All these data and others suggest depths up to 60-70 km ! The problem is the further buoyant return upwards of these rocks subducted in the mantle, even if this return is only of tertiary age. A possible explanation (Platt 1986) (fig. 6 ) suggest that the subduction of cold oceanic crust continued beneath these rocks maintening low geothermal gradient. Furthermore, the continued underplating of Lower Penninic units allowed the high-pressure rocks to rise towards the surface at which gravity spreading of the surincombant Austroalpine units induced their thinning through distensive tectonics. 32 B 35m.y. Fig.6 - Tectono-metamorphic evolution of the Central Alps (after Platt, 1987, modified) A. Late Cretaceous, B. Late Eocene. AA. Austroalpine units, L. Lanz9peridotites, MS. Upper Lower, Upper Penninic units, S. Sesia massif, z. (Helvetic) zone. mantle, Ext. Pi, Ps. External 5. What was the superficial sedimentological frame of this subduction process ? Most authors agree that the process was coeval with the deposition of Late Cretaceous flyschs in the Piemont and Ligurian-Piemont areas, but all these flyschs were stripped off the underlying rocks as early as the beginning of the Eocene and glided by gravi ty toward the External zones where they now form the Helminthoid flysch nappes of the Prealps, the Embrunais-Ubaye area and the Italian Maritime Alps (fig. 1). They thus escaped the Eocene metamorphism (see below), while other parts of the Ligurian-Piemont Upper Cretaceous sediments (and the underlying Lower Cretaceous and Upper Jurassic rocks) were transformed by this metamorphism and became a part of the Schistes lustres where they cannot be separated, except in a few places (Lemoine & al. 1984, Dumont & al. 1984) 33 The subduction process was accompanied by the decollement af a great part of the Ligurian-piemont sediments and of some ophiolitic slices and by their emplacement as the earliest nappes of the Alpine evolution, the vergence of which is discussed below. It is possible to consider these first units as a kind of "tectonic accretionary prism", upon which the Upper Cretaceous flysches were unconformably laid down as fore-arc sediments. Because the Helminthoid flysches were stripped off from their underlying rocks, it is impossible (in the present state of the edifice) to observe this unconformable contact. The interpretation is derived from the contrast observed between the strongly tectonized ophioli te-bearing Upper Jurassic-Lower Cretaceous complex, on the one hand, and the much less deformed Upper Cretaceous flysches on the other. Moreover, the Cenomanian black shales at the base of the Helminthoid flysch have been supplied with debris eroded from already folded complexes containing serpentinites, granites, Lower Cretaceous limestones, etc (Grandjacquet & al. 1972). This Albian crisis is probably the beginning of the compression between the Eurasian and African plates, because deformations of that age are classic features in all Mediterranean chains. More externally, the Valais trough closed during the Eocene, this closure being predated by the deposition of the so-called Tarentaise flysch in France, the Pratigau flysch and Niesen flysch in Switzerland. In the present state of our knowledge, no subduction seems to have occured here, but rather the squeezing of a trend floored by thinned continental crust, or by narrow, only locally developed stripes of oceanic crust. 2. After the Late Cretaceous came the collision itself, corresponding to the Alpine folding proper which evolved in several phases. This collision is classical in spite of its complexity and the scarcity of the stratigraphic records. This paper does not aim to describe the present structure of the Western Alps, for such a description has been the topic of numerous papers (see, for instance, Debelmas & al. 1983). According to the theme of this conference, its seems better to emphasize the tectonic evolution of the belt and, for this purpose, we will deal with two problems. The first problem is the outward progression of folding and thrusting, i. e. the superficial expression of the phenomenon through which the basement became a stack of intracontinental thrusts developped one after another between the Eocene and the present time, always in an outward direction (Trlimpy 1975, Tricart 1984, Malavieille & al. 1984, Menard & Thouvenot 1984), (fig.7). a. During the Late Eocene, the collision started. The European margin was split easily into several slabs because it had previously been block-faulted the Mesozoic normal faults (the result of the Jurassic rifting) were reactivated as thrust faults, connected with some flat and deeper crustal decollement planes. The slabs, at least the more internal ones, were emplaced under an important overload, made up by the South-Alpine thrusted margin and probably some related ophiolitic slabs. The resulting bulge induced in the underlying cover and basement slices flow structures and long recumbent folds, the "Penninic folds", arranged both synthetically and antithetically ("retrocharriage"). 34 w ..... a w o 40Km .'----'------', b D DO 2 3 ~ ~ 4 c SA SA SA 20 E E Fig.7 A possible model for the evolution of the Western Alps, showing the progressive outwards accretion of basement flakes, on the Gran Paradiso traverse (after Malavieille & al., 1984). BR. Briangonnais basement DB. Dent Blanche nappe MCE. crystalline massifs MCI. Internal crystalline massifs South-Alpine zone ; Se. Sesia zone. a. Late Cretaceous : end of the Alpine ocean closure Thrust I obduction of oceanic rocks on the European margin External SA. Thrust 2 : individualization of the crustal flakes which will form the internal crystalline massifs (MCI). b. Eocene collision Thrust 2 accentuation of the MCI thrusting toward the W, with Thrust 3 Thrust 4 Thrust 5 Thrust 6 antithetic structures (thrust 5) toward the E. fragmentation of the front of the South-Alpine margin, giving birth to the Dent Blanche-Sesia slice (Se). Austroalpine thrust backthrusting of the Briangonnais zone. individualization of the first External crystalline massifs thrust sheets (MCE). c. Present-day Thrust 6 : thrusting of the MCE slices toward the W 1. Sedimentary cover ; 2. Continental crust ; 3. Ophiolites and their sedimentary cover; 4. Upper mantle. 35 The peridotitic slice of Lanzo was probably emplaced in this context, for geophysical data show that it is still deeply rooted in the upper mantle (fig.8). II II 10 20km BELLEDONNE . _ J_ H ,.----------......... , ~",,' SE , . VANOISE ~~' GRAN ""IIAOISO LANZO I Fig.8 NW-SE cross section through the Western Alps on the Gran Paradiso traverse (compare with fig.7) (compiled from Perrier 1980, Debelmas & al. 1983, Menard & Thouvenot 1984). Black. Sedimentary cover White. Crystalline basement Vertical hatching. Upper mantle ; ZZ. Low velocity zone of present-day seismic waves. A, B. Upper Miocene and present-day Moho offsets (?). The age of this first set of phenomena is given by the radiometric dating of the associated metamorphic minerals (i.e. 38 ~ 2, Late Eocene, Hunziker 1974, Desmons 1977, the "Lepontine" phase of metamorphism). Stacking up these slices and folds produced an uplift of the internal Alpine zones ; the sea (which covered the Brian~onnais domain up to the Middle Eocene) was expelled toward the external zones upon which it transgressed from the Middle Eocene to the end of the Eocene and even into the early Oligocene, and where abundant clastic sediments were deposited, derived from the uplifted more internal parts of the belt. b. In the Late Oligocene and the Early Miocene, a new crisis overprinted the previous structures, but expanded outwards and involved the External crystalline massifs, as shown by (1) radiometric data on those massifs (18 to 15 Ma, overprint ages ranging from 41 to 36 Ma Baggio & al. 1967, Rb/Sr on bioti tes ; Leutwein & al. 1970, K/Ar on adularia and muscovite, Rb/Sr on muscovite, etc) and (2) implication of the Upper Eocene sedimentary cover in the deformation. Moreover, cleavage developed for the first time in part of the External zone, affecting Priabonian sandstone and Oligocene molasse. This cleavage was truncated by an erosion surface classically regarded as pre-Miocene (Tricart, 1984). Again, the European crust was split up and more external areas were involved (fig.8). The basement of the External crystalline massifs was 36 broken up into large slices of granite and gneiss dipping eastward or southward, whereas other crystalline basement units (namely the basement of the Helvetic nappes, lying paleogeographically south of the Aar massif) disappeared into the depths. At this time also, the suture separating the external Alpine zone from the internal zones probably appeared (or was completed), i.e. the so-called "chevauchement pennique frontal", and along it a great part of the more external Penninic zone(the Subbrian90nnais zone) vanished. That zone had probably been a brittle belt in the Alpine framework since the Mesozoic, where numerous normal or strike-slip faults were located on the external side of the Brian90nnais geanticline. It now forms a tectonic scar, marked by allochtonous cover slices without any crystalline basement material. The general splitting up of the east or south part of the external zone and the stripping off of its cover were probably associated with underthrusting of the whole crystalline basement of that zone beneath the internal zones. The resistance offered by the external crust to sinking induced the splitting up of the External crystalline massifs into large slabs thrust over one another. Here, clearly, the main thrust planes were not the normal listric faults that marked off the tilted crustal blocks in Mesozoic time. Recent studies in the French crystalline massifs (Lemoine 1984) demonstrate that the slabs were split and thrust without notable disturbance of their Mesozoic fault-planes (fig.4), along more deep-seated planes in the crust. c. In the Late Miocene and Pliocene, the French Subalpine ranges and the Jura mountains were folded, involving the Miocene sediments laid down in the Alpine area around the reliefs born from the previous tectonic crisis. In this last crisis, the basement of the Subalpine ranges was probably thrust beneath the front of the External crystalline massifs, a process that may still be active because seismic data show that seismic waves slow down at a depth of 11-23 km below the External crystalline massifs and the adjacent more internal area (fig. 7 & 8). Perrier (1980) and Menard & Thouvenot (1984) considerer that this low-velocity channel is a shear plane along which the basement of the External crystalline massifs is being thrust outwards and which leads up to the present-day front of the massifs, to 8-9 km deep, along a line well marked by current seismic activity. The second problem related to the collision is its direction and that of the resulting thrusts. Since the formation of the Alps is linked to the roughly north-south collision of Europe and Africa (Patriat & al. 1982), the first idea is to squeeze the Alpine oceanic basin of fig. 2 and 5 along this same north-south direction, consistently with the east-west direction of the first folds that appeared south of the European margin from the Late Cretaceous to the Middle Eocene, as in Provence and the French southern Subalpine ranges. Their eastern prolongation is unknown or badly known in the Penninic domain. 37 In such a reconstruction, two features must be noted : I. The subduction was more or less oblique to the South-Alpine margin. Offsets of that margin, due to minor transform faults, may have evolved into pinched structures including ophiolitic material, which would be a possible explanation for the Canavese zone, for instance (fig.9C). 2. The most extensive thrusts occured in the Central and Eastern Alps and not in the Western Alps. Moreover, it is not easy to determine the thrust direction. Was it northwards, westwards or both? Field evidences shows that these internal units of the Western Alps exhibit cross-cutting folds and lineations, some being perpendicular to the Alpine arc others parallel to it. In early studies (Caby 1974), and also in more recent ones (Ricou 1984), the transverse features are considered as the oldest, defining an east-west folding pattern overprinting a not yet arcuate Alpine domaine (in good agreement with the relative north-south movement between Europe and Africa). The folds parallel to the Alpine arc are considered as formed later, by a west to northwest motion of the South-Alpine block resulting in the rigid-plastic indentation of the European margin of the South-Alpine (or Apulo-Adriatic) block (Tapponnier 1977). Recent field works show a much more complex pattern, for which very detailed studies are required. The structural history that emerges is rather similar from one area to another (e.g. Central Alps : Savary et al. 1983, Steck 1984, Vanoise: Platt and Lister 1985 a & b). It can be summarized as follows : I. The first recognizable Alpine deformation is characterized by a northwest trending lineation which, in the Southern Vanoise, is associated with a flat-lying foliation. The lineation, which is oblique to transverse to the present axis of the Alps, is considered as probably connected with the major nappe displacement and involved a northwest directed subhorizontal shear. In the Southern Vanoise, this deformation phase clearly postdated the peak of the high pressure metamorphism (the majority of the HP-minerals appear to be mechanically rotated into the new foliation plane) . 2. In a second deformation phase, new schistosities appeared with a new stretching direction linked to a northern vergence of the folds. The corresponding deformation was a dextral simple shear. The associated metamorphic minerals were of greenschist facies in the French Alps, indicating a decrease in pressure and thus uplift. The peak of this new metamorphic phase, dated at 38 Ma (Lepontine phase), (Bocquet & al. 1974), was attained between the second deformation and the next one. 3. In a third phase, backthrusting and backfolding, directed south or southeast, took place under low-grade conditions. This phase was followed by differential uplift along steep faults and by strike-slip displacements. In this tectonic pattern, the general stress field was roughly constant through the evolution and compatible with the overall north-northwest displacement of the South-Alpine block (fig.9). However, this framework does not sufficiently account for the present-day curvature of the Alpine belt. We can only assume that an original curvature was increased in Neogene time, in the southern, part of the Alps, by the northeast-directed Apennine thrusts and the spreading of the Genoa sphenochasm (Debelmas 1986). 38 / \./0-" /n~ A B Fig.9 The collision between Europa and the Adriatic promontory. A. Eocene. For the abreviations, see fig.3 Thick black arrows : dominant compressive stress-field. Thin black arrows : successive motions of the Adriatic block. B. Pliocene : Vertical hatching : folded zone Black : ophiolites Thin arrows : opening of the Genoa sphenochasm Thick arrows : compressive local stress-field. C. A possible explanation for the origin of the Canavese zone. The Prealps and the Embrunais nappes. Thus far, we mainly dealt with the behaviour of the crystalline basement and its immediate cover and did not mention the more superficial features, i.e. the gravity gliding of tectonic units that were emplaced from the Late Eocene to the Neogene. Their relics are the Prealps and the Embrunais-Ubaye nappes, where Upper Cretaceous flysches of either Valais or Ligurian origin, are accompanied by Subbrian 39 A second superficial thrust occured immediately after the Late Eocene-Early Oligocene crisis and generated a new nappe complex upon an erosion surface. In the Embrunais area, this surface can be reconstructed across the eastern Dauphine domain not yet affected by a regional cleavage. Thus, the emplacement predated the Late Oligocene-Early Miocene phase (s). In the Prealps, the erosion surface is at the top of the Subalpine "red molasse" (Middle Oligocene). The emplacement is thus dated at Late Oligocene to Early Miocene. The evolution of the Prealps and the Embrunais nappes ended with the general Neogene folding phase, which also affected the underlying Helvetic or Subalpine units. Acknowledgements:I would like to express my gratitude to J. DESMONS, J. RODGERS and R. TRUMPY for critical comments of the successive drafts of this manuscript and correcting English. 40 REFERENCES Aubouin J., Blanchet R., Labesse B., Wozniak J. (1977). Alpes occidentales et Alpes orientales la zone du Canavese existe-t-elle ? C.R.Som.Soc.Geol.Fr., p. 155-158. Baggio P., Ferrara G., Malaroda R. (1967). Results of some Rb/Sr age determinations of the rocks of the Mt Blanc tunnel. Boll.Soc.Geol.ltal., p. 193-212. Bernoulli D., Lemoine M. (1980). Birth and early evolution of the Tethys: the overall situation. C.R. 26e C.G.I., Paris, Coll.5 Ed. B.R.G.M., p. 168-179. Bocquet J., Delaloye M., Hunziker J., Krummenacher D. (1974). Kr-Ar and Rb-Sr dating of blue amphiboles, micas and associated minerals from the Western Alps. Contrib.Mineral.Petr., 47, p. 7-26. Bosellini A., Castellarin A. (1982). Magmatismo e tettonica nel Trias delle Dolomitic Guida alIa geol.del Sudalp.centro.orient., Guide geol.reg. Soc.Geol.ital. Bologna, p. 189-210. Caby R. (1974). Geodynamic implications of transverse folding in the Western Alps for the Alpine fold belt. Nature, 256, p. 114-117. Caby R. (1973). Les plis transversaux dans les Alpes occidentales implications pour la genese de la chaine alpine. Bull.Soc.Geol.France, (7), XV, p. 624-653. Castellarin A. (1982). Lineamenti ancestrali sudalpini. Guida alIa geologia del Sudalp. Centro-or. Guide geol. reg. Soc. Geol. Ital, Bologna, p. 41-55. Dal Piaz G.V., Hunziker J., Martinotti G. (1972). La zona Sesia-Lanzo e I' evoluzione tettonico-metamorfica dell Alpi nord-occidentale interne. Mem.Soc.Geol.ltal. XI, p. 433-466. Debelmas J. (1986). The Western Alps : new data and hypothesis. In "The Origin of Arcs", Intern.Conf.Urbino, 1986. Elsevier (under press) . Debelmas J., Escher A., Trlimpy R. (1983). Profiles through the Western Alps. Geodyn.series, 10, Am. geoph.Union, p. 83-96. Debelmas J., Kerckhove Cl. (1980). Les Alpes franco-italiennes. Geologie Alpine, Grenoble, t. 56, p. 21-58. Debelmas J., Sandulescu M. (1987). Transformante nord-pennique et problemes de correlation palinspastique entre les Alpes et les Carpathes. Bull.Soc.Geol.Fr. (8), p. 403-408. Dercourt J. et al. (1985). Presentation de 9 cartes paleogeographiques au 1/20 000 000, de I' Atlantique au Pamir pour la periode du Lias a l'Actuel. Bull. Soc. geol.Fr., (8), t.l, p. 623-790. Desmons J. (1977). Mineralogical and petrological investigations of Alpine metamorphism in the Internal French Western Alps. Am.J.Sc., 277, p. 1045-1066. Dumont T., Lemoine M., Tricart P. (1984). Perennite de la sedimentation pelagique du Jurassique super~eur jusque dans les Cretace superieur au des sus de la croute oceanique tethysienne ligure : la serie supra-ophiolitique du lac des Cordes (zone piemontaise des Alpes occidentales au SE de Brian90n). C.R.Ac.Sc.Paris, t. 299, p. 1069-1072. Frey M., Hunziker J.C., Frank W., Bocquet J., Dal Piaz D., Jager E. and Niggli E. (1974). Alpine metamorphism of the Alps. A review. Schweiz. Min. Petro Mitt., 54, p. 121-139. 41 Frey M. Hunziker J.C., O'Neil J., Schwander H. (1976). Equilibrium-desequilibrium relations in the M. Rosa granite. Petrological Rb-Sr and stable isotope data. Contr. Min. petr., v.55, p. 147-179. Graciansky P. Ch. (de), Bourbon M., Chaptal o. (de), Chenet P.Y., Lemoine M. (1979). Genese et evolution comparee de deux marges continentales passives : marge iberique de 1 'Ocean atlantique et marge europeenne de la Tethys dans les Alpes occidentales. Bull.Soc.geol.France, p. 663-674. Grandjacquet Cl., Haccard D., Lorenz Cl. (1972). Essai de tableau synthetique des principaux evenements affectant les domaines alpins et apennins a partir du Trias. Bull.Soc.Geol.France, p.158-162. Hunziker J.C. (1974). Rb-Sr and K-Ar determinations and the Apine tectonic history of the Western Alps. Ann.Soc.Geol.Belg., 94, 116-117. Lemoine M. (1984). La marge occidentale de la Tethys ligure et les Alpes occidentales. In "Les marges continentales en mer et a terre autour de la France"., G. Boilot coord., Masson ed. Paris, p. 159-248. Lemoine M. (1985). Structuration jurassique des Alpes occidentales et palinspastique de la Tethys ligure. Bull. Soc .Geol. France, p. 126-137. Lemoine M., Gidon M., Barfety J .Cl. (1981). Les massifs cristallins externes des Alpes occidentales : d'anciens blocs bascules nes au Lias lors du rifting tethysien. C.R.Ac.Sc.Paris, 292, p. 917-920. Lemoine M., Marthaler M. & al. (1984). Decouverte de foraminiferes planctoniques du Cretace superieur dans les Schistes lustres du Queyras (Alpes occidentales). Consequences paleogeographiques et tectoniques. C.R.Ac.Sc.Paris, 299, p.727-732. Leutwein F., Poty B., Sonet J., Zimmerman J.L. (1970). Age des cavites a cristaux du granite du Mont-Blanc. C.R.Ac.Sc.Paris, 271, p. 156-158. Malavieille J., Lacassin R., Mattauer M. (1984) . Signification tectonique des lineations d' allongement dans les Alpes occidentales. Bull.Soc. Geol.France, p. 895-906. Mattauer M., Malavieille J., Monie P. (1987). Une coupe lithospherique des Alpes occidentales dans l' hypothese ou Sezia n' est pas d'origine sud Alpine. C.R.Acad.Sci., Paris, 304, p. 43-48. Mattauer M., Tapponnier P. (1978). Tectonique des plaques et tectonique intracontinentale dans les Alpes franco-italiennes. C.R.Ac.Sc.Paris, 287, p. 899-902. Menard G., Thouvenot F. (1984). Ecaillage de la lithosphere europeenne sous les Alpes occidentales arguments gravimetriques et sismiques lies aI' anomalie d' Ivree. Bull. Soc .Geol. France, p. 875-884. Oberhansli J . C. , Hunziker G. , Martinotti & Stern WB (1985) . Geochemistry, geochronology and petrology of Monte Mucrone : an example of eoalpine eclogitization of Permian granitoids in the Sesia Lanzo Zone, Western Alps (Italy). Chem.Geol. 52, 165-184. 42 Patriat Ph., Segoufin J., Schlich R., Goslin J., Auzende J.M., Beuzart P., Bonnin J., Olivet J .L. (1982). Les mouvements relatifs de l'Inde, de l'Afrique et de l'Eurasie. Bull.Soc.Geol.France, p. 363-373. Perrier G. (1980). La structure des Alpes occidentales deduite des donnees geophysiques. Ecl.geol.Helv., p. 407-424. Platt J.P. (1986). Dynamics of orogenic wedges and the uplift of high pressure metamorphic rocks. Bull. Am. geol. Soc, v. 97, p. 1037-1053. Platt J. P., Lister G. S. (1985a). Structural history of high pressure metamorphic rocks in the Southern vanoise massif, French Alps, and their relation to the Alpine tectonic events. J.Str.Geol., 7, p. 19-55. Platt J.P., Lister G.S. (1985b). Structural evolution of a nappe complex, southern Vanoise massif, French Penninic Alps. J.Str.Geol., 7, p. 145-160. Ricou L.E. (1984). Les Alpes occidentales : une chaine de decrochement. Bull.Soc.Geol.France, p. 861-974. Savary J., Schneider B. (1983). Deformations superposees dans les Schistes lustres et les Ophiolites du Val d'Herens (Valais). Ecl.Geol.Helv., 76, p. 381-389. Tapponnier P. (1977). Evolution tectonique du systeme alpin en Medi terranee poinc;onnement et ecrasement rigido-plastique. Bull.Soc.Geol.France, p. 437-460. Tricart P. (1984). From passive margin to continental collision a tectonic scenario for the Western Alps. Am.J.Sc., 284, p. 97-120. Triimpy R. (1975). On crustal subduction in the Alps. In Tectonic problem of the Alpine system, Slovak .Acad. Sc. Bratislava, 121-130. Triimpy R. (1980). An outline of the geology of Switzerland. Guide-book, 26e C.G.I. Paris, ed. BRGM, 104 p. Triimpy R. (1986). Die Plattentektonik und die Entstehung der Alpen. Naturf.Ges.Ziirich, Neujahrsblatt auf 1985, 47 p. Vandenberg J. & Wonders A. (1976). Paleomagnetic evidence of large faults displacement around the Po Basin. Tectonophysics, 33, 301-320. Wever P. de, Caby R. (1981). Datation de la base des Schistes lustres post-ophiolitiques par des Radiolaires (Oxfordien superieur-Kimmeridgien moyen) dans les Alpes cottiennes (St Veran, France). C.R.Ac.Sc.Paris, 292, p. 467-472. THE GEOMETRY OF CRUSTAL SHORTENING IN THE WESTERN ALPS Robert W.H.Butler Department of Geological Sciences The University South Road Dur ham DHl 3LE United Kingdom ABSTRACT. The use of structural data to define the convergence history across orogenic belts has been handicapped by the absence of a coherent tectonic model wi thin which it is possible to unravel the relative horizontal displacements, their lateral continuity and compatibility. Such a model can be provided by concepts of linked thrust analysis, this contribution applies this working hypothesis to the Eocene-Pliocene tectonic evolution of the western Alps. Deformation commenced with the obduction of oceanic material and a sheet of previously subducted continental crust onto the Franco- Swiss crust. Locally displacements were transferred onto the base of shelf sediments which were carried out onto the external zones as the Prealpine klippen. The obduction complex was then disrupted by back thrusts and then carried forwards by displacements in the frontal Pennine and external zones. Thrusting terminated in the Jura in latest Miocene times. Throughout the deformation the Franco-Swiss continent was transected by thrusts which followed several mid-crustal detachment levels. A series of balanced sections together with an appraisal of thrust sheet distributions suggest that displacements exceeded 400 km on a WNW-ESE axis in the western Alps at a rate of about 1 cm.yr-l. Substantial volumes of Penninic crust must have been subducted beneath the Po plain requiring wholesale transformation to eclogite facies assemblages to inhibit isostatic compensation. Using this model for western Alpine orogenesis the central Alps are interpreted as an oblique ramp zone within which oblique folds, thrust sheet rotations and complex strain patterns develop. Two dimensional section restoration will not be appropriate in this sector. However, displacements can be transferred onto the base of the Austroalpine sheets in the eastern Alps and, via transfer faults, onto the Apennine thrusts to the south. 1. INTRODUCTION To fully understand the processes involved in forming continental collision belts it is fundamentally important to unravel the relative horizontal displacements which have generated crustal thickening and 43 A. M. c. ~engor (ed.), Tectonic Evolution a/the Tethyan Region, 43-76. © 1989 by Kluwer Academic Publishers. 44 lithospheric subduction with the associated responses of regional metamorphism, flexural basin development and any magmatic activity. One approach is to adopt an external frame of reference by concentrating on patterns of magnetic anomalies and hot spot migration to resolve the spreading histories of adjacent oceanic li thosphere(1). However, while it may be a simple matter to define the relative motions on the margins of these oceans, it is considerably more ambiguous to do so in complex tectonic settings such as the Mediterranean. Problems arise because, even if the external plate configuration is fully defined, the spatial and temporal evolution of plate boundaries within the collision zone itself are often far less obvious. Furthermore, changes in plate boundary location will occur throughout the collision process and convergence may obliterate small areas of oceanic lithosphere which record part of the total plate motion picture. So, while a view from outside the collision zone itself may give an accurate picture of the global plate tectonic setting, it is of limited use in explaining the structural, and hence metamorphic, stratigraphic and igneous, evolution of particular segments of the collision zone. The alternative to using 'extra-orogenic' methods is, as Laubscher(2)recognised, to concentrate on the deformation histories of the collision belts themselves. However, the great majority of accounts, particularly in the Alps, stress the severe structural complexity of collision belts. As England(3) recently pointed out, the problem is not so much the paucity of data but rather its shear volume. Clearly some form of model is required. In the Alps, one school of thought developed(4,S) whereby each regionally correlated deformation event was heto explain the discontinuous nature of deformation at surface within a broad zone of plate convergence. Despi te their theoretical elegance, both of these dynamic models are flawed. If continental deformation was controlled by zones of blocks rotating on vertical axes as envisaged by MacKenzie and Jackson(7) mountain belts would not be regions of crustal thickening and would only develop by essentially thermal processes. High pressure metamorphism of continental upper crust would not occur. Map patterns in collision belts would display lozenge blocks which had rotated but had not been thrust across each other. England's modele 6), which treats the continental lithosphere as a simple vertical rheology, would deform essentially by pure shear processes, the various lithospheric levels would not be inter sliced by thrusts and the deformation would be continuous. Thus neither model satisfies some basic observations made for continental collision belts and hence will not provide either an explanation for,or prediction of patterns of regional metamorphism, flexural basin development and other aspects of the geological evolution of these bel ts. This contribution attempts to outline the rationale behind a model of continental deformation in collision orogenic belts which is primarily geometric and to illustrate its strengths by application to the western Alps. As such the geometric solutions adopted here to solve particular structural problems should not be considered final 45 statements, the aim is to illustrate how thrust tectonic models may be applied to complex orogenic problems like the Alps. Models of thrust tectonics on the crustal scale were in use at the turn of the century(8) and are now being rediscovered by reflection seismologists(9) who use the same models to interpret deep crustal reflectors in ancient mountain belts. So the model to be explored here is that mountain belts are primarily built by the thrusting of continental crust. A great deal of work has been directed at unravelling the structural complexities of thrust belts over the last twenty years(lO,11,12). Many accounts stress the importance of two aspects: The linked three dimensional networks of thrust systems; and the need to construct accurate, balanced cross-sections to test geometric models. The basic philosophy behind balancing is that the volume of deforming materials remains broadly constant so that any geometric model based on a structural interpretation must be capable of restoration to a geologically realistic, undeformed state. To date these restorations have been two dimensional and restricted to individual cross-sections. Therefore, each construction carries the assumption of plane strain; cross-sectional area must be preserved after allowing for erosion. The big advantage for plate tectonic studies is that, by subtracting the length of the final state section from that of the restoration, balanced sections yield estimates of orogenic contraction implicit in the particular structural model adopted at the outset(l3). Full details of the thrust tectonic model are outlined elsewhere(l4), the important feature being that most thrust systems contain important sub-horizontal detachments, often termed 'flats', which transfer the displacements between zones of deformation at diferent vertical levels. The links between flats are termed 'ramps' and are the sites of finite stratal shortening. Thus a thrust system will link patches of crustal shortening, not only in cross-section but also in three dimensions. The other crucial aspect is that thrust arrays generally develop in a systematic sequence from hinterland to foreland(12). At any given time in the development of a thrust system only the frontal thrusts will generally be active; earlier, more internal ones will be frozen. Thus, in plate tectonics terms a thrust belt is a zone of plate boundary adjustment in which the site of relative convergence migrates outwards. This is in marked contrast to MacKenzie and Jackson's model(7) where all their faults operate simultaneously. To unravel the convergence history across orogenic belts not only the magnitude but also the direction of displacements must be determined. This is crucial when making two dimensional restoration since the section line must contain the unique horizontal movement direction. The thrusts which stack continental crust are zones of movement within which the strain will be predominantly simple shear(lS). Therefore, all pre-existing markers together with any structures generated during displacement will be progressively rotated towards the movement direction. Regional patterns of stretching lineations will define displacement trajectories between different crustal segments(16). 46 a f- Fig.l: The consequences of shear zone geometry for the restoration of thrust systems, illustrated on schematic sections. a) final state geometry where displacements are distributed in a shear zone climbing across stratigraphy (i.e. a ramp); b) a line-length restoration of the shear zone and wall rocks, note thinned and extended beds in the restored position of the shear zone (between f and g); c) an area restoration which illustrates the true geometry of beds and shear zone boundaries prior to deformation, note that .the restored length is less than that for (b); d) a line-length restoration of wall rocks alone so that the shear zone is approximated to a discrete plane (hence juxtaposition of f and g), note that this underestimates the total restored width and hence also the magnitude of displacement. The application of thrust geometric ideas, derived from foreland fold and thrust belts, to entire orogenic belts cannot be made without first discussing some of the model's complexities. Firstly, displacement zones need not be extremely narrow but can have a finite width. Within t~ese shear zones folds will develop which are not a direct consequence of thrust geometry(15). Figure 1 illustrates that broad shear zones can be approximated to narrow zones but a restoration of the stratigraphy invol ved in such a structure leads to various estimates of orogenic contraction. Clearly area balancing gives the true answer but a minimum estimate of displacement can be obtained by line-length measurements of the wall rock stratigraphies and by neglecting the extended units within the shear zone. Large scale recumbent folds are kinematically analogous to simple imbricate slices, as noted by Heim(17). Other, initially upright folds can result from inhibited displacements on thrusts and hence approximate to layer-parallel shortening strains. 47 These features can account for most structure displayed on cross-ctions across orogenic belts when allied to strains which accommodate stacking of thrust sheets. It is less clear in three dimensions how the model explains much of the complexity of Alpine tectonics, notably the acurate mature of the belt. However, many foreland thrust belts display lateral variations primarily caused by spatially discontinuous zones of shortening. To maintain compatibility, lateral offsets in thrusts must be linked by tear faults or shear zones running parallel to the movement direction(ll). Figure 2 illustrates some of these features. In their simplest form, thrusts can develop into shoe-box plan forms with dip slip (frontal) ramps linked by strike-slip (lateral) ramps, or local combinations. However, when thrust propagation or displacement is inhibited the lateral and oblique ramp sites will localise oblique fold trends as they approximate transverse simple shear zones(18). Within these zones pre-existing markers will be rotated. It is common for these lateral zones to be stable for substantial periods of thrusting so that oblique shear zones may be piled up, each being rotated by the displacements on lower zones. Kinematic indicators may then show a divergence which results not from a real divergence of displacement but from local rotations within a thrust system which otherwise retains a constant movement direction. Large scale divergence of thrust transport has been invoked for the Alpine arc(8). However, the strain compatibilty problems for simultaneously divergent transport are very great(19) , nappes would have to extent greatly around the arc so that thrust zones themselves would not approximate to simple shear zones. Laubscher & Bernoulli(20) describes sets of transcurrent faults which have accommodated arc parallel extension in western Switzerland. However, these could represent no more than a few kilometers extension, not the much larger (c .100 km) amount required for the divergence of thrust systems with over 100 km displacement directed simultaneously towards the WNW and NNW. Rather, Laubscher's observations(20), together with the scale tectonic coherence of the Alps argues against simultaneous divergence, a single transport direction is most likely for each period of orogenesis. Ricou(21) proposes a dominantly south to north motion of Italy relative to Europe during Eocene and Oligocene times with the western Alps acting as a strike-slip zone. Any thrusting is relegated to have secondary importance and be of Miocene or younger in age. This view has been contested by Malavieille et al.(24) who propose a continuous episode of east to west thrusting from Eocene times on the basis of regional stretching lineation patterns. Here I support this contention, on the basis of thrust geometry on a local(23) and regional (24) scale. The present distribution of basement and cover, large and small scale thrust geometry and local stretching lineation patterns in both the western and central Alps are indicative of relative convergence between Italy and Europe on a ESE-WNW axis from the Eocene to Pliocene times. Thus cross-sections constructed parallel to this transport direction and geographically distant from lateral ramp zones 48 SINIS TRAL SHEAR COUPLE IN HANGING-WALL BLOCK [OUNTER - CLOCKWISE ROTATION OF OBLIQUE FOLDS, LINEATIONS Hr FOLDS fiENERATED PERPENDICULAR TO THRUST TRANSPORT OBLIQUE FOLDS GENERATED WITH urrtE ROTATION Fig.2: Hypothetical block diagram illustrating the relationships between old orientation, three dimensional thrust geometry and displacement. should be restorable. I do not advocate that balanced sections be directly valid for the central Alps since the plane strain assumption will be inappropriate. Regionally, however, and in the western Alps it should be a good first approximation so that the magnitude of the WNW-ESE Eocene-Pliocene convergence between Italy and Europe can be estimated. It also enables some statements to be made on the deep structure and the geometry of lithospheric deformation in the Alpine belt. 2. AN ALPINE FRAMEWORK There is a wealth of general review articles (e.g. refs.25,26,27) on Alpine tectonics which has enabled this article to be written and to which I can do slight justice here. Readers are directed to the few papers referenced for details. A general map is provided as a guide line (Fig.3). This illustrates the basic tectonic divisions of the Alps: a hinterland region now occupied by the Po plain in Italy; an upper, Austroalpine, Nappe complex; an internal Alpine zone composed of interleaved Franco-Swiss material with oceanic or Ligurian units; an external Alpine zone composed of folded and thrust Franco-Swiss basement and cover with local molasse; and a foreland in the north and west. The Austroalpine thrust approximates to the suture between Italian and European crustal segments. The emplacement contact of oceanic material onto the Franco-Swiss crust is the Piemontais thrust although it has been severely modified by subsequent deformation. Penetrative 49 deformation and metamorphism in the Swiss sector are limited to the internal zones which are carried onto the external units by the Frontal Pennine thrust. Klippen of Penninic cover rocks now lie up to 100 km from their main outcrop location(28), tectonically emplaced onto external covermassifs (Fig.3). The external zones themselves are characterised by a general decrease in deformation complexity towards the foreland. Although the basement units of the external and Pennine Alps are comparable, the cover units show marked facies variations. On the basis of much of the Mesozoic succession a basic subdivision can be made between external, Dauphino-Helvetic facies, and Pennine, Brianconnais facies units. Details are provided elsewhere. The deep structure of the Alps has been studied by several geophysical experiments(29,30), involving gravity surveys, magnetics and refraction seismology. Seismic reflection profiles have been acquired in the external thrust belts during hydrocarbon exploration but the planned regional deep seismic lines have, at the time of writing, yet to be shot. All the present data are consistent with the Franco-Swiss crust having a present of 28-30 km on the foreland, the Moho gradually deepening beneath the Alps to a maximum depth of about 50 km. The internal margin of the Alps is marked by dense, high velocity material lying at relatively shallow depths. This is the Ivrea body, a presumed mantle slice that has been emplaced onto the Franco-Swiss crust during Alpine tectonics(31). Its continuation to outrcop as either the Lanzo peridotites or the Ivrea zone is highly unlikely as discussed e1sewhere(32). The crustal thickness beneath the Po plain is probably 35 km (ref. 29), although this is difficult to define since the molasse sediments are very thick near surface. 3. WHOLE CRUSTAL IMBRICATION Based on the surface geology of the central A1ps(25,26) and geophysical data(29((through this sector, a model has built up of Alpine deep structure(33,34). It is illustrated schematically on Fig.4. The main features relate to thrusts having a sigmoidal profile, sub-horizontal at the surface and along the Moho but steeply dipping in the middle crust. It is therefore, essentially a pure shear model where shortening in the upper crust is not greatly removed from shortening in the lower crust. This form of section might be compatible with England's notion(6) of continuum deformation. But is it valid? To answer this question let us examine the implication of the model for a structural restoration (Fig.5). 50 o o o o o Fig.3: Sketch map of the NW Alps with locations of the detailed maps and sections. AAT- Austroalpine thrust, FPT - Frontal Pennine thrust, PT Piemontaise thrust (at structural base of oceanic sheets, not illustratd on the map) SL - Simplon line, MR - Monte Rosa massif, a _ Fig.6, b - Fig.12, c - Fig.13, w-w' - Figs.4 and 5, x-x' - Figs.lO, 11 and 18. Fig.4: An interpretation of Alpine crustal structure, modified from De Jong(33), illustrating the 'whole crustal imbrication' model. section line w-w' on Fig.3, FPT - Frontal Pennine thrust. Jura Fareland b! p SWISS rna/ass£' basin Prt:'-oIpme klippe ~~~~--~~~-~~-~~~~ '- A \ I / ~-/--- - Foreland crustal /- Pennmt:' nappes Insubrtc Po plam B' Helve/,es margmal - ocpamc (deep walerl Pl?nmnp ,over p' 50 ,,,, J I 0 I L - H H 'o ce ne L -_ _ - - - 2.J C Ol lg oc en @ I J Eo ct ?n e o 10 ~ C -= -= -t :- :- ~= -_ ~ ~ re tQ C @ O U S • LA rO U R -D u~ el N o () B as D au ph in e o () c ~ t hr us t ~ n o r m Q I F ig .6 : S im pl if ie d ge ol og ic al m ap o f th e NW e x te rn a l A lp s (b ox a o n F ig .3 , s e e a ls o in se t) w it h th e s e c ti o n li n es o f F ig s. 7 a n d 8 il lu st ra te d . P en ni ne z o n e s ha ve a r u le d o rn a m e n t, AA T - A us tr oa lp in e th ru st , FP T - F ro n ta l P en ni ne t h ru st , UH T - U lt ra h el v et ic t h ru st . [ I Tr /Q 5 - ju ra ss Ic [ T I b as em en t 'J > N 53 4. WESTERN ALPINE THRUST SYSTEMS There is currently much debate over the subsurface profile geometry of thrusts. The geometry outlined above for the whole crustal imbrication model might be termed 'thick-skinned', the thrusts steepen downwards(39). An alternative view might be that the thrusts remain gently dipping beneath the orogenic belt so that the deformation might be regarded as 'thin-skinned'. Although both terms are in current usage neither is particularly useful. At the scale of the lithosphere, the 'whole crustal imbrication' model which detaches at the Moho appears thin-skinned. Conversely, if we are to relate thrust systems to plate motion they must ultimately pass displacements up from the base of the lithosphere so that even the shallowest level of detachment forms part of a whole lithosphere thrust profile. In the following account I will propose that Alpine deformation is controlled by thrusts which have a number of different detachment levels, within cover, basement and li thosphere so that they form a staircase. The crucial point is the degree to which the staircase forms and hence how far various parts of the lithospheric profile are translated. To discuss this problem we will gradually build up a cross-section through the western Alps and so we start in the foreland. 4.1. Geometry of the Alpine Sole Thrust The thrust front in the western Alps presently lies at the eastern margin of the Bresse graben (Figs.3 and 6) where Mesozoic carbonates have been thrust out onto Miocene molasse deposits(38). Behind this front lies the frontal part of a thrust belt, the outlying portions of which have a simple structure constituting the Jura fold belt. Geophysical data(30) confirm the long held belief that basement is not directly involved in these structures which rather have decoupled along Triassic evaporites. A compilation of these data, together with deep bore hole records(30), illustrate tha~ this decoup1ing surface passes down beneath a culmination of basement rocks(24), the external crystalline massifs. These massifs have ~ therefore been carried by the displacements which deformed the Jura and their eastern continuations, the subalpine chains (Fig.6). A restoration of these displacements provides an insight as to the deep structure of the basement massifs and to the geometry of this most outlying movement zone. This is termed the Alpine Sole thrust. Many accounts of Jura folding suggest displacements of about 30 km for the Swiss sector (e. g. Ref .12). Two balanced sections, presented here (Figs.7 and 8) through the southern termination of the Jura, suggest a total displacement of about 35 km in the Mesozoic carbonates of the Jura and frontal subalpine chains. These displacements are considered to pass back down beneath the external Belledonne massif. Figure 9 illustrates how the basement can be restored onto a crustal template using these displacements. Foreland crust upon which the now W NW B U G E Y H IL L S a P IN LI N E A N N fC Y B A S IN ES E [=:Q ] LW R - M ID M IO CE NE ~ U R G O N /A N D N fO [O M IA N B K IM _ O X FO R D IA N ~ B A T H O N IA N D A A L - B A JO C IA N O r R /A S - L I A S o ~ 0 0 0 0 0 o;f :Jt /' '\ ) . SS ;:, 1 >" '· ~ £ £ }? ·§S · L s . ' , )5 :; , , ? .. . $ u ' " :: s; ; . o " Ii lL P IN LI NE K H b S I1 0R T fN IN G (P IN -A ) (P IN - A ') 18 K H F ig .7 : B al an ce d (a ) 8n d re s to re d (b ) c ro s s -s e c ti on c o u pl et fr om th e B re ss e gr ab en to th e A nn ec y ba si n, il lu st ra ti n g t he g eo m et ry a n d s ho rt en in g in t he s o u th er n Ju ra . S ec ti on l in e de pi ct ed o n F ig .6 . V l . ,. (J - tt ) 91 11 11 11 1: l01 1$ :: IlS S ttH )N tt7 8 lN O H - lS M /H i )1 1} A7 }H 'r/ iJ .7 n- 1 Sf lI: lH J. }N IN N ld 7t f1 N O I:I ::l - id :: l n c o j ~ ~ "" ' " '!.6 > }N NO O }7 7} 8 ~ 56 detached Mesozoic carbonates once lay, must pass beneath the basement massifs for 35 km so that the Alpine sole thrust cannot cut steeply down towards the Moho from the leading edge of the massifs. This conclusion is opposed to the 'whole crustal imbrication' model discussed above. The basement massifs therefore constitute a relatively thin sheet at their leading edge, the thickness of which can be defined from their basement cover contact. Thus the external Belledonne massif is just 10 km thick and so, by assuming a pre-existing crustal thickness of 28 km, the Alpine Sole thrust must have a staircase from with a mid-crustal detachment.This corresponds to a zone of low seismic velocity reported by Menard(30,39) who traced the thrust down to an offset of the Moho (Fig.lO). The trajectory between the midcrustal detachment and the Moho is poorly constrained, a simple form is illustrated on Fig.9. d e I r z f! ! I y y y M-M-H-t1-f'1-H-H-t1-L.~H_1 1'1-/1-/'1--/1-1'1-1'1 b 25 km .- --=-- Fig.9: The restoration of the Alpine Sole thrust, illustrated graphi- cally, a) final state section, b) restoration. Based on the surface geometry of the external Belledonne massif (segment x), the Moho depths of Menard(30) and the displacement in the Jura and subalpine chains from Figs.7 and 8. NW J/Jra g foreland basemen' massds Internol zones Fig .10: Menard&Thouvenot' s (39) crustal cross-section through the NH Alps (x-x' on Fig.3) based on gravity and seismic refraction data. Penninic material is densely stippled while the external crust is lightly stippled; AAT - Austroalpine thrust, FPT - Frontal Pennine thrust, AST - Alpine Sole thrust. 57 WNW ESE x' Fig.ll: Balanced (a) and restored (b) crustal cross-section through the western Alps (section x-x' on Fig.3) modified after Butler et al. (24). Ornament and key as in Fig.lO. 4.2. The NW External Alpine Thrust Belt Having considered the geometry of the Alpine Sole thrust beneath the basement massifs it is appropriate now to discuss the structure of the massifs themselves. Figure 8 illustrates that the Belledonne massif is split by a major fault, the Median thrust which is locally marked by Mesozoic cover rocks. The eastern panel of basement rocks, the internal Belledonne massif, displays much greater structural compexity than the simple domal uplift of its outlying counterpart(25). The basement-cover contact has been disrupted so that the internal panel of the Belledonne in fact constitutes a series of very narrow basement slices. Recent work on the northern margin of these slices(23) has confirmed a view held by some Alpine pioneers(25), that they constitute a now deeply eroded imbricate stack of basement and coverrocks. A recently published balanced cross-section(23) through this stack restores the basement-cover contact to a width of at least 77 km, with an implicit shortening of over 65 km. A further 45 km of displacement is required to imbricate and emplace the structurally highest external zone units which constitute the Ultrahelvetic thrust sheets. Compared with traditional accounts of Alpine tectonics these values are rather high but are supported by restorations of the Helvetic sector of the thrust belt along strike(23). Thus, around the Mont Blanc massif (Fig.6), thrust systems can be restored to a width of 140 km from a pin line along the external Belledonne and Aiguilles Rouges massifs. It must be emphasised that these displacement estimates are not at odds with traditional accounts(17,25,26) of Alpine geology, indeed many imply greater values. The same accounts also imply local decoupling of basement and cover beneath the entire Helvetic and Ultrahelvetic system 58 on certain section lines. This situation is similar to that in the Jura, the differences primarily being due to lower displacements on the thrusts in the more outlying setting. The NW external thrust belt can be traced south form the Helvetics, within the basement massifs, as far as Pelvoux(26). Displacements are apparently conserved; a recently restored section through the southern part of the belt has resolved displacements in excess of 150 km (ref.32) . Studies of stretching lineation patterns and thrust geometry in the Mont Blanc sector suggest that the transport direction was towards the WNW (ref.23). This direction has apparently remained constant throughout this episode of thrusting, the precise timing of which can be gained from stratigraphic and radiometric dating. The coverunits in the SW Mont Blanc massif area contain Nummulitic limestones(40) of late Eocene age which have experienced the entire history of thrusting. Upper Miocene molasse sediments are buried beneath thrusts in the subalpine chains near Grenoble(4l). However, the structurally high level Mont Blanc basement sheet has yielded Rb-Sr dates of 13.4+2 Ma and K-Ar values of 18.3 +2 Ma, ages(42) which probably reflect the initiation of uplift on thrusts. It is likely that all the displacements discussed here developed between about 22 Ma and 6 Ma at a constant rate of just under 1 cm.yr-l. The Pelvoux area is shrouded in controversy regarding the timing of deformation and the direction of thrusting(43). Eocene age rocks lie unconformably on basement(44) suggesting a much earlier onset of thrusting. However, the interpretation of field relationships around thrusts is often ambiguous since the basement-cover contact forms an important detachment horizon(43,45) so that overlap can occur by tectonic as well as sedimentary processes. The lower Oligocene sandstones in this area are severely deformed so that a major episode of thrusting, probably contemporaneous with the deformation of the Mont Blanc sector, is well know. The pre-Eocene deformation is unlikely to be important in terms of displacement, certainly there is no evidence for a major phase of thrusting further north in the external Alps(24). The displacement direction in the Pelvoux massif is also difficult, studies by Tricart( 45) have proposed a SW orientation. However, the massif lies at the southern end of the NW external Alpine thrust belt. Displacements are transferred up from beneath the massif to converge with the Frontal Pennine thrust(24). This type of setting would tend to localise thrust propagation problems as indicated on Fig. 2. The SW direction could be more apparent than real, generated by the counter-clockwise rotation of thrust sheets by laterally inhibited, lower fault zones directed towards the WNW. Again there are no indications of substantial arc parallel extension in this sector of the Alps. 59 4.3. Deep Structure Earlier we used the restoration of thrust systems developed in cover rocks of the Jura and subalpine chains to derive a model for the deep structure of the Alpine Sole thrust (Fig.9). The same can be done for the crustal structure of the Alps, this time using the restoration of the NW external Alpine thrust belt(24). Using a restored width of these structures of 140 km, a crustal template can be constructed for the external zones, as depicted on Fig.ll. A problem arises in defining the crustal thickness prior to thrusting. Earlier, for the central Alpine section (Fig.4) we performed an area restoration using the crust still available in the cross-section. However, we can also estimate the crustal thickness from the nature and thickness of cover rocks exposed in the thrust belt. A common assumption in Alpine tectonics is that the mountain belt represents a res tacked continental margin of some initial complexity(35). Most assume that the inner parts of the external Alps were greatly attenuated, as evidenced by thick piles of deep water Liassic shales(46). However, these shales have been heavily imbricated and probably had a thickness never more than 2 km. In the Mont Blanc area they are absent, the Mesozoic succession having a total thickness of just 100 m (ref.47). The presence of limestone units in these attenuated sequences argues against a deep water basin starved of sediment. The strongest arguement against this hypothesis isthe widespread occurrences of encrusting Eocene carbonates lying directly on basement in the Pelvoux area(44), and on the attenuated Mesozoic sequences near Mont Blanc. This strongly suggests that crustal attenuation prior to Alpine thrusting in the external zones was minimal (24). A crustal thickness of 25 km can be adopted with some confidence onto the restored template (Fig.ll). This means that the entire width of present Alpine outcrop must be underlain by crust of external Alpine origin. The Pennine thrust sheets have been carried far from their lower crustal roots. The 'whole crustal imbrication' model is inappropriate for at least the western Alps. This type of detachment model for western Alpine thrusting has been implied recently by Menard & Thouvenot(39) who present a very similar type of crustal cross-section. In this way they are able to explain the gravity and seismic velocity structure. Boyer & Elliott(12) produced a detachment-dominated model which looks similar but which did not consider deep structure. One argument against this approach might be that foreland thrust belt models cannot be applied to lower crustal levels. However ,regardless of deformation mechanism, the need to conserve volume clearly must be as strong as in sedimentary units. Lateral continuity of the NW external Alpine thrust belt and its displacements requires the basic conclusion of wholesale underplating of the Alps by externally derived crust to be valid in three dimensions. The deep structural elements of Moho depth and Ivrea body position are 60 coherent around this sector of the Alps(30), there is no indication of late faulting of sufficient organisation to systematically relocate externally derived crust. 4.4. Internal Alpine thrust systems It is widely assumed that the highly deformed and metamorphosed parts of collision mountain belts have experienced most of the total convergence across the orogen. This belief is entrenched in Alpine literature (e.g. ref.26) so that the further importance of the internal Alps is highlighted by the restorations of external Alpine structure discussed above. The plan now is to continue the attempt at estimating the amount of orogenic contraction in the western sector of the Alpine belt and to apply the same basic concepts of thrust tectonics used in the external zones. Regardless of the apparent complexities of the deformation history in the internal Alps, there is a need for a complete reappraisal following the realisation that the Pennine nappes must be far removed from their lower crustal roots. Boyer & Elliott(12) noted that the Alpine etradition on concentrating on nappe volumes and extensi ve correlation has led to an incomplete definition of thrust system geometry. As shown in the external Alps, the definition of transport direction, lateral displacement continuity and thrust relationships is a basic requirement in any tectonic study of collision processes. These have yet to be resolved in the internal Alps. Thus the following account must be treated as an initial statement, the displacements will be obtained from a single, broad transect across the internal zones, lying ESE of the Mont Blanc massif (Fig.12). The Frontal Pennine thrust separates the Pennine zones from the external Alps. In the Mont Blanc area it carries an imbricated stack of Mesozoic shelf sediments (the Brianconnais facies) with local slices of Carboniferous basement and a thick sequence of Upper Cretaceous to Eocene age sandstones. These frontal units constitute the Tarentaise zone(48), otherwise termed the Valais or Sion-Courmayeur zone(26). A sketch map of part of the Tarentaise (Fig.13) together with a balanced cross-section across it (Fig .14) are presented. As such the structure has been simplified, the thrust geometries commonly include recumbent folds indicating broader shear zone control rather than the discrete faults illustrated on the section. This simplification generates minimum estimates of orogenic shortening across the zone, primarily based on a line-length measurement on the Triassic units in the frontal part (Mya) and on a conglomeratic basal unit to the Upper Cretaceous sandstones in the central sector (Terrasse). Larger thrust sheets lie at the eastern margin of this zone, forming the Petit St. Bernard and Bassa Serra sheets. A total displacement of at least 40 km is implied across this zone. A WNW-ESE displacement axis is defined by sheath folds in evaporites which decorate some thrusts, and by stretching lineations on others. The sandstone units commonly display complex patterns of stretching lineations and folds implying rotational strain histories. .... / / Fig .12: Geological sketch the locations of detailed x-y - Fig.16. map 0 f the NW in terna1 Al ps maps and sections; a - Fig.13, Ausfroa(plnf~ sheets sch,stes (usfrps serpent/fides g gabbros Tarenfolse flysch Pennlflle MesozoIc cover rocks Carboniferous basement crystolf,ne basement 61 illustrating b - Fig .17, However, along thrust shear zones, streching 1inations tend to plunge ESE, consistent with a single bulk transport direction. Tectonically overlying the Tarentaise zone along the French-Italian border lies a thick package of Carboniferous basement, the zone Houi11er (Fig.12). Brianconnais cover units are exposed in the Vanoise(49) where they display evidence for a long, complex and laterally variable deformation history(SO). These cover rocks are tectonically overlain by oceanic rocks of the Piemontaise sheets so that the earlier deformation episodes in the Brianconnais presumably relate to obduction. This occurred during the Eocene since shelf sedimentation continued to this time in the Brianconnais(49). Elsewhere, Piemontaise units lie directly of Brianconnais basement suggesting that the cover was carried out during the later stages of obduction. Interestingly, these relationships lie ESE and far internal of the prealpine klippen of Penninic cover rocks now lying on the subalpine chains and He1vetics (Fig.3). These klippen overthrust, and provide debris for late Eocene conglomerates and sandstones(28). Thus the timing of obduction was contemporaneous with 62 ~ Brton(onnols ~ Tor.nfolse flysch ~ basal flysch E3 Cretaceous !.:-.,:, ·: ... ·:1 Lias g older Fig.13: Simplified geological map of part of the Tarentaise zone (loca- tion a on Fig.12), modified after Antoine(48), illustrating the loca- tions of major thrusts, namely; BBT - Basal Brianconnais thrust, BST - Bassa Serra thrust, FPT - Frontal Pennine thrust, PSBT - Petit Saint Bernard thrust, UHT - Ultrahelvetic thrust. Line of section for Fig.14 is depicted (M-M'). 63 the emplacement of the prealpine klippen which can be interpreted as the locally missing cover to the Pennine thrust sheets. A simple model is illustrated on Fig.lS. WNW ESE break In Mya Imbricates H Pennme Q balanced section o 1 '""--==-d km V = H b restored sections T€>rrasse Imbricates duplex of pre - flysch rocifs required at depth .,., 3' Hya Imbricates f,n,> jJJL I~ " 64 WNW \'rARENJA/HZONII_\ 0 0 •• ~~~~" "\ ~ INITIAL PREALPS [HPLMfH[NT /lIGH PIUSSUR{ 8HlMlNT {IfPL.trlO DIHO 'RIANCON,.,./S SHELF HIGH - PRESSUIU [RUST ESE Fig.IS: Schematic and unbalanced sequential (a-c in time) cross sections through the Aosta valley sector (north of x-x' on Fig.3) of the NW Alps illustrating the progressive thrust-controlled deformation of the internal zones, from about middle Eocene (a) to late oligocene (c) times; AAT - Austroalpine thrust, GP - Gran Paradiso massif. a) illustrates the emplacement of oceanic and Austroalpine sheets across the Franco-Swiss crust with its subducted leading edge with high pressure basement. b) depicts the emplacement of part of the subducted crust onto the Brianconnais shelf concurrent with the emplacement of part of the Brianconnais cover onto the outlying Tarentaise and external zones during late Eocene times. c) illustrates in highly simplified form the disruption of the obduction complex developed in (a) and (b) by hinterland and foreland directed thrusts. The underlying sub-Pennine crust is subducted beneath the Ivrea body with the predicted generation of eclogites. 65 4.5. The Back Thrust Belt Throughout the internal part of the Pennine zones in the western Alps the simple obduction relationships of Piemontaise sheets on Brianconnais cover has been severely disrupted by a suite of ESE-directed, retrocharriage structures (51). These are effectively back thrusts which emplace the Pennine basement and cover into the Piemontaise units, often for great distances. The Gran Paradiso massif is a domal culmination of Penninic basement around which are folded some of the back thrust structures. One of these can be traced for over 15 km in an ESE direction, the appropriate footwall cut-off remaining buried. Movement directions can be defined from stretching lineations in the fault zones. These cluster on a WNW-ESE axis in the Vanoise-Aosta sector. In an attempt to gain a displacement estimate across the back thrust belt, a balanced cross-section has been constructed through the northern Vanoise (Fig.16) . An intriguing initial result of remapping by the author is that the bulk distribution of Brianconnais lithologies can be explained entirely by back thrusting. There is however, abundant evidence for earlier deformation from small scale structures, notably intrafolial folds, mylonitic textures and shear bands, in the cover rocks suggesting penetrative simple shear parallel to bedding. This sector apparently remained in the footwall to the obduction complex and was avoided by the following episodes of WNW-directed ductile imbrication which effected the southern Vanoise. Figure l6b illustrates this stratigraphic coherence prior to back thrusting. The western part of the belt contains a culmination of stacked basement and Triassic quartzites which represent at least 15 km of displacement. Thrust shear zones contain good stretching lineations and sheath folds on a WNW-ESE axis. The other constraint on movement direction comes from the most eastern of these thrusts, the Digitation de l'Iseran (see also Fig.12), where stretching lineations also plunge WNW. The entire belt requires at least 40 km displacement to restore the stratigraphy to its undisrupted form. Further shortening would be required to remove local intraformational folding and thrust imbrication. The back thrusts of the northern Vanoise pass along strike into a belt of basement imbrication on the southern side of the Aosta valley (fig.12) . However, Caby (52) reports that in the Arvier area these structures are overstepped by a high level sheet, itself containing back thrust geometries (Fig .17). This contains a marginal facies of the oceanic sediments, the Combin zone. If the interpretation is valid this sheet would have truncated the Vanoise back thrusts over a width of 25 km, a figure which would then correspond to the minimum displacement on this sheet. The entire belt could therefore represent over 65 km of orogenic contraction following the obduction of oceanic rocks onto the Pennine zones. W NW Oo ", ~ d. 1 0 So ch l' Q i'' (S S/ { LI M ES TO NE S " N O YO U NG EP 8R IM /C O N N AI S U N Ir S b R l!' s/ or .d I# m pf at l' La c d . C h. "r tf , . /' " T sa nl l'I .,n tl hi lls SH O RT EN IN G ON rU ST O R AT IO N IJ '· H , • A fU R TH E R I5 K H IS R EQ UI RE O TO R eS TO RE SH U TS X Y Z ~-'r/ ~,~:~ c=J S(HIST{5 iUHRfS _ OPHIOLITE ~--", ~ BRIANCONNAIS COVER o (ARBONIFEROUS rn XALLINE BASEHENT Q. // + b. 0 -0 + TRUN(ATfO BACKFOWS + 67 5 KH ,--~;...~ ( z ( + Fig .17: The 'overstep' geometry of back thrusts in the Arvier area illustrated by map (a) and section (b), modified from Caby(52), loca- lity (b) on Fig.12. BT - 'Basal Combin' thrust, PT - Piemontaise thrust, AAT - Austroalpine thrust. The schistes lustres of the Combin and Zermatt-Saas zones are marked 'c' and 'z' respectively. A problem remains in rooting these structures. Caby et al. (51) proposed that the back thrusts arose from a reversal in subduction polarity, a model followed by Laubscher & Bernoulli(20). However, restoration of the external zones means that the Pennine zones cannot be underlain by their basement, as envisaged by these workers. The back thrusts must have developed above a detachment situated at high levels in the Pennine crust. The Arvier geometries (Fig .17) contain downward (ESE) facing thrusts, presumably folded by the subsequent uplift of the Mont Blanc massif. The back thrusts probably splay from a Pennine thrust, such as the Basal Brianconnais thrust (Fig.12) at the base of the Zone Houiller. This type of relationship is common in many thrust belts(53) and is illustrated on Fig.15; it does not require a reversal in subduction polarity. 68 4.6. Displacements and Lithospheric Structure Using the balanced sections through the Tarentaise zone (Fig.14) and the northern Vanoise (Fig.16) we can begin to estimate displacements in the internal Alps. The two sections, together with the larger scale back thrust geometry, combine to give at least 100 km WNW-ESE convergence since mid. Eocene times. These figures do not include displacements within the Zone Houiller or on the Basal Brianconnais or Frontal Pennine thrusts. Two further examples of Pennine deformation must also be considered. Firstly, the prealpine klippen were emplaced during late Eocene times, concurrent with the later stages of oceanic obduction in the internal zones. These klippen lie at least 75 km outboard of their original position. However, this distance has been shortened by back thrusting, possibly by 60 km. Thus the prealpine sheets must have travelled at least 100 km and probably nearer 150 km to reach their resting place on Ultrahelvetic and Helvetic units. Imbrication within the klippen themselves would increase these early displacements even further. The second consideration must be the internal Pennine basement massifs of Gran Paradiso, Monte Rosa and Dora Maira (Fig. 3) • These contain high pressure metamorphic assemblages (54), in contrast to the main Pennine units in the Vanoise, which possibly result from pre-Eocene subduction. A window in the Gran Paradiso massif(55) shows that this basement unit has been emplaced onto conglomerates of probable Permian age, similar to the basement of the Brianconnais zones. Furthermore, basement slices south of the Aosta valley display relationships suggesti ve of their lying tectonically above the Brianconnais cover rocks of the Vanoise. The cover slices breach up through into the crystalline basement on back thrusts. The most outlying units of this crystalline sheet are found in the eastern part of the Grand St.Bernard nappe (Fig.12) where Caby et al. (51) report remnant high pressure assemblages. Thus the Brianconnais of the northern Vanoise was overlain locally by a further sheet of crystalline basement, its emplacement possibly relating to the removal of Pennine cover to the prealps (Fig .15) . The present width of outcrop taken by this high pressure sheet is about 40 km but again this distance represents a fraction of its original extent. The back thrusts have reduced it by about 60 km suggesting that the total displacement on the sheet exceeded 100 km. Thus the Gran Paradiso massif lay far to the ESE of the Brianconnais cover rocks prior to thrusting so that it could be subducted during 'Eo-alpine' events while sedimentation could continue in the Vanoise sector. Total shortening achieved times in the internal Alps displacements on the base of (Gran Paradiso-eastern Grand by thrusts from middle Eocene to Miocene is difficult to assess. Much of the the prealps and the crystalline sheet St. Bernard nappe may be equivalent. 69 However, both occurred before back thrusting in the Vanoise(50) and WNW-directed displacements in the Tarentaise zone. These later episodes represent at least 100 km shortening while the earlier thrusts must represent a further 150 km. Total displacements therefore exceed 250 km in the Franco-Italian Pennine Alps in a WNW-ESE direction. These continued out into the external Alps as WNW-directed thrusting as described earlier. The outlying structures represent a further 150 km shortening. Thus, since middle Eocene times, about 400 km of WNW-ESE convergence has occurred across the western Alps. Since thrusting terminated in the late Miocene, the deformation continued for about 40 Myr, a time averaged convergence rate of approximately 1 cm.yr-l. The restoration of external zone thrust systems requires the present width of the Alps to be underlain by crust derived from the external zones so that the Frontal Pennine thrust must lie at relatively high crustal levels. The conclusion implied that the Pennine zones would be separated from their lower crustal roots. To tackle this problem we must first estimate the amount of Pennine crust prior to Alpine thrusting. We can make some estimaters using the stratigraphy and thickness of the Brianconnais units which have been restacked in the Vanoise(48) .. Along the section line described here the Brianconnais zone restores to a width of about 80 km, assuming no shortening in the Zone Houiller. A further 100 km of internal Pennine basement would lie ESE of this and about 50 km of the Tarentaise zone would have lain between the Brianconnais and the external zones. Thus the Pennine units would have had a width in excess of 230 km prior to thrusting. The cross-sectional area available beneath the Pennine zones in Fig .11 is about 1000 km2. Thus to obtain a balance using the material present on the cross-section, the Pennine crustal thickness would originally have been just 4.5 km. Clearly it must have been substantially thicker than this, indeed many accounts of Alpine Palaeogeography(46).talk about the stability of the Brianconnais zone. It is likely then that several thousand km2 of Pennine crustal cross-sectional area have been lost. The location of this 'missing' crust remains a problem. The account of Alpine structure given here has stressed the cohererence of thrusting from the Pennine out into the external zones. There are no obvious punctuations in this system when viewed across the complete transect. Furthermore, the basement lithologies of the Pennine and external zones are very similar. There is no indication in the western Alps that the two domains were separate continental blocks. This assumption may be inappropriate for the eastern parts of the central Alps where a Valais ocean basin has been proposed(26)., separating the two domains. The equivalent unit in the Western Alps is the Tarentaise zone which, although containing good evidence for rifting during Upper Cretaceous times(48), most probably remained coherent. These features suggest that the Pennine and external zone crust must have remained continuous throughout Alpine thrusting. Circumstantial evidence comes from the foreland(56). where flexural basins developed by the lateral redistribution of thrust sheet loads on the lithosphere. For the 70 WNW km 100 LVZ A. Ips 100 km InsubNC - [onavese line ESE Po valley Fig.18: Spectulative lithospheric cross-section through the western Alps (section x-yon Fig.19) illustrating the consequences of the large (c.400 km) displacements deduced in this article. LVZ - low velocity zone, AAT - Austroalpine thrust. foreland to be subjected to flexure during Penninic thrusting there must have been lithospheric and therefore crustal continuity. Thus the have much higher seismic velocities compared with its lower pressure precursor. Thus subducted and ec10gi tic crust may be indistinguishable from mantle on some geophysical grounds. It might be detected by electrical or seismic reflection profiling, the results of planned experiments are awaited with interest. 4.7. The Central Alps as an Oblique Convergence Zone The view from the western Alps is that tectonic evolution between middle Eocene and late Miocene times is controlled by thrusting and, on the lithospheric scale, subduction of continental material which accommodated at least 400 km \"/N\v-ESE convergence. This directly implies that the central Alps, which trend obliquely, experien ce v!N\v-ESE convergence and thrusting rather than the N-S shortening implied by some workers. Steck(59) suggests a similar model, but with variation in transport from nappe forming NVl-SE displacements follmved by E-H dextral shear and associated folding, at least in the Pennine zones. This two-phase model can be reinterpreted in terms of a continuous thrusting model with differential Htnl displacements. At the outset it was proposed that oblique and lateral ramps can experience rotations due to a differential shear couple caused by laterally inhibited thrust motion(Fig.2). In the central Alps this could progressively rotate and fold higher thrusts by displacements on later, lower zones. Thus, while nappe formation could always be fo~ by refolding, it can happen diachronously through a thrust belt. I f '1 1 .f ~/. I / I ? Rotated sheet 5 thrusts L,thospnerre extenSion and generation of ocean baSIn Oligo-Miocene o __ 001 km 100 MUNICH .---. TYRRHENIAN SEA LI thospherJ( /110[(]ne - Pliocene 71 EASTERN ALPS Fig .19: Simplified. tectonic map of the Alpine and Apennine thrust systems developed from middle Eocene to late Miocene times. The primary HNH-ESE displacements in the Alps (400 km) are predicted to transfer onto the Apennines where an additional component is added, coupled to lithospheric extension in the Ligurian and Tyrrhenian seas. The spatial patterns of apparent thrust divergence and folding are based on reasoning depicted on Fig.2. 72 defined by palaeomagnetism (62). There are two ways of explaining this. Firstly, the northern Apennines may have experienced differential movement strains as the thrusts attempts to climb northwards into the Po molasse basin. The systematic rotation of thrust sheets in this environment explains both the documented direction of thrust transport and the apparent orientation of the pole, both have been reoriented. In the central Apennines a W-E transport direction has been proposed which would support this model(63). The second solution is that the Apennines contain an extra component of plate motion compared to the Alps, due to the opening of the Liguran basin by lithospheric extension in Oligo-Miocene times( 63). The rotation pole for this opening would lie at the northern end of the Apennines so that thrust sheets would be rotated greatest in this sector. The central and southern Apennines would experience less rotation since much of the displacement path would cOincide with the direction of Alpine thrusting. Note that both models could have operated simultaneously to rotate the northern Apennine sheets but neither model requires the rotation of the Italian foreland which now lies almost completely buried by Apennine thrust sheets. This crustal block will have converged in a \-IN\-l-ESE direction with the European foreland and represented by the Apennine and southern Alpine units. Thus the greater part of the 400 km of displacement required for western Alpine orogeny can be transferred laterally. All these final statements are highly speculative and serve only as a prompt to obtaining further estimates of displacement directions and magnitudes around the Alpine and western Mediterranean belts. Those values, 400 km of HN\-l-ESE convergence during Oligo-Miocene times, for the western Alps make this task particularly important for they are not predicted by an analysis of the 'extra-orogenic' data sets 1,64 • This suggests that the evolution and location of plate boundaries in the Nediterranean during the latter part of the Tertiary must be reappraised. A preliminary conclusion is that Italy cannot be treated as a simple promontory of the African continent and that another plate boundary existed between Italy and Africa during this time. A further area of lithospheric extension, possibly a small ocean basin, within the Mediterranean system is required to balance the convergent displacements in the Alps. hopefully this will be defined by further structural restorations. Finally, to end on a general point, the scale of displacements being uncovered from the Tethyan collision belts suggests that large scale subduction of continental lithosphere, with parts of its crustal component, can continue long after the subduction of any adjacent oceanic material. In the northern t1editerranean it is likely that most of the total convergence between continental blocks is acheived in this way. Indeed the preceeding subduction of oceanic lithosphere and true continental 'collision' may not be a principle requirement for large 73 This model explains why the earliest, higher level thrust sheets in the Alps also contain the most complex deformation histories. However, further complications within the simple thrust model arise because the Pen nine zones defonred in a metamorphic environment where the thrusts are broad, ductile shear zones. Shear zones can generate complex cyclic folding within their boundaries 15 by local rheological changes which pertubate the displacement pattern. In syn-metamorphic shears the potential for rheological changes(EO) caused by reaction::; is greatly increased so more complex folding histories are expected. If the total displacement rate across a shear zone is altered then this folding will also involve the thrust sheets themselves. Hence the complex deformation histories of ~_ internal parts of orogenic belts are fully predicted by the thrust tectonic models adopted here. A common practice in analysing any thrust belt is to start on parts most ammenable to balanced section construction and then to correlate displacements as an aid to structural interpretation in the more complex sectors. This rationale has been followed here so that displacements in the western Alps can predict structural evolution in the central Alps. In start in complex sectors where more models appear to be valid on a local scale can lead to ambiguities when considering the belt as a whole. Continuing around the belt, the implications of a prolonged WNW-ESE history of displacement in the western Alps is that the Austroalpine sheets also moved in this direction in the eastern Alps. Displacements would be transferred, via the Pennine and external central Alps, onto the Austroalpine thrust. Differential movement shears, folds androtations in the central Pennines may decouple along an upper detachment rather than necessarily continue to the surface. The Austroalpine thrust could have performed this detachment function so that the complex strain zones of the Pennines can remain buried: The Austroalpine sheets merely continued to displace towards the HNH during the Oligo-Miocene times. Clearly this must be tested by further investigations along the Austroalpine front in eastern Switzerland and in the windows of Pennine rocks in the Austrian Alps. 5. DISCUSSION vlhile it may be possible to relate the tectonic evolution of the main Alpine belt during Oligo-Miocene times to an essential plane strain model of lithospheric thrusting, it is less clear how these displacements pass southwards. During this period the Appennines were experiencing a major episode of thrusting directed towards the east. In a plane strain model it is possible to link dimetrically opposed thrusting in terms of a backthrust-forethrust complex, connected by a tear fault. For displacement compatibilty this fault zone would have to lie along the northern margin of the Ligurian basin (Fig.19). A problem arises in applying this simple model to Alpine tectonics because the transport direction in the northern Apennines was towards the NE in the critical period((61). Furthermore, thrust sheets in this northern sector have apparently experienced counter-clockwise rotations 74 scale continental deformation. While this problem may not be relevant for the great eastern Tethyan mountain belts, it will be important if these Cenozoic belts are used as analogous for the less well known, ancient orogenes. A CKNm-lLEDGt1ENTS The ideas presented in this paper were born during Ph.D. research at the University College of Swansea with Rod Graham and Norman Fry, and have reached adolescence during the tenure of a Royal Society (Jaffe Donation) Research Fellowship. I am grateful to colleagues for vigorous discussions of Alpine tectonics, particularly Garry Karner, Steve Hatthews and t1arcus Parish together with two anonymous referees whose comments greatly improved this article REFERENCES 1- Dewey,J.F., Pitman.W.C., Ryan,\v.B.F. & Bonnin,J. 1973. 'Plate tectonics and the evolution of the Alpine system', Bull. Geol. Soc. ~ 84, 3137-3180. 2- Lubscher,H.P., 1971. 'The large scale kinematics of the western Alps and northern Apennines and its palinspastic in)plications'. Am. J. Sci. 271, 193-226. 3- England,P.C., 1986. 'Deformation of continents', Nature 320,399-400. 4- Bearth,P., 1952. 'Geologie und petrographie des 110nte Rosa', ~ Geo1. Karte Schweiz n.f. 96, pp.94. 5- Milnes,A.G., Grel1er,M. 8, t1uller,R., 1981: 'Sequence and style of major post nappe structures, Simplon-Pennine Alps', J. Struct. Geol.3, 411-420. 6- Eng1and,P.C., Houseman,G. & Sonder,C., 1985'. 'Length scales for continental deformation in convergent, divergent and strike-slip environments: analytical and, approximate solutions for a thin viscous sheet model', J. Geopfiys. Res. 90, 3551-3557. 7- MacKenzie ,D. & Jackson,J. 1986. fA block model of distributed deformation by faulting', J. Geol. Soc. London 143, 349-353. 8- Argand,E., 1916. 'Sur l'arc des Alpes occidentales', Eclog. Geol. Helv. 14, 145-191. - 9- ~dell ,D.J., 1984. 'Deformation of the Caledonide lithosphere of northwest Scotland', Tectonophysics 109, 137-145. 10- Bal1y,A.W., Gordy,P.L., & Stewart,G.A., 1966. 'Structure, seismic data, and orogenic evolution of southern Canadian Rocky Mountains', Bull. Can. Petrol. Geol.lLf, 337-381. 11- Dah1strom,C.D.A., 1969. 'Balanced' cross sections', Can. J. Earth Sci. 6, 743-757. 12- Boyer,S.J. & Elliott,D. 1982.'Thrust systems', Bull. Airl.,Assoc. Petrol. Geol. 66, 1196-1230. 75 13':"· Hossack,J .R. 1979. 'The use of balanced cross sections in the calculation of orogenic contraction: a review', J. Geol. Soc. L.ondon 136, 705":71l. 14- Butler,R.\v.H. 1985. 'Thrust tectonics: a personal view', Geol. I~ag. 122, 223-232. 15- Ramsay,J.G. 1980. 'Shear zone geometry: a review', J. Struct. Geol. 2, 83-99. 16- Shackleton,R.t'!. & Ries,A.C. 1984. 'The relation bet,,,een regionally consistent stretching lineations and plate motions', J. Struct. Geol. 6, 111-117. 17- Heim,A. 1921. Geologie der Schweiz 2/1, Tauchnitz, Leipzig. 18- Coward,M.P. & Potts,G.J. 1983: 'Co~plex strain patterns developed at the frontal and lateral tips to shear zones and thrust zones', J. Struct. Geol. 5, 383-399. 19- Goguel,J. 1962. 'L'interpretation de l'arc des Alpes occidentales', Bull Soc. Geol. France 7, 20-33. 20- Laubscher,H.P. & Bernoulli ,D. 1982. 'History and defomation of the Alps', in Hsu,K.J. (ed.) t10untain Building Processes, Academic Press, London, 169-180. 21- Ricou,L.E., 1984. 'Les Alpes occidentales: chaine de decroche~ent', Bu1l.·Soc. Geol. France 26,861-874. 22- Malavieille,J., Lacassin,R. & Mattauer,M. 1984. 'Signification tectonique des lineations d'allonge~ent dans les Alpes oc~identales', Bull. Soc. Geol. France 26, 895-906. 23- Butler,R.H.H.1985. 'The restoration of thrust syste~s and displacement continuity around the t10nt Blanc massif, NH external Alpine thrust belt', J. Struct. Geol. 7, 569-582. 43- Beach,A., 1981. 'Thrust structures in the eastern Dauphinois zone (French Alps) north of the Pelvoux oassif', J. Struct. Geol. 3, 299-308. 44- Barbier, 1948. 'Les zones ultradauphinois et subbrianconnais entre l'Arc et l'Isere', Heo. Servo Carte geol. det. France., pp.291. 45- Tricart,P., 1980. 'TectQnigues.gu@J:,l?osees .. dans les Alpes occidentales, au sud dt! ':leI voux. Evolution structurale d 'une chB.ine de collision', These, Univ. Strasbourg. 46- Debelmas,P., & Kerckhove,C., 1980. 'Les Alpes franco-italiennes', Geol. AJpine 56, 21-58. 47- Landry,P., 1978. 'Donnees nouvelles sur la couverture sedioentaire des massifs externes au sud du }1ont-Blanc', Geol. Alpine 54, 83-112. 48- Antoine,P., 1971. 'La zone des breches de Tarentaise entre Borug St. t·1aurice (Vallee de 1 'Isere) et la frontiere Italo-Suisse',' Geol. Alpine 9, pp.367. -- 49- Ellenberger ,F. , 1958. 'Etude geologique du pays de Vanoise', Hem. expl. Carte Geol.Franc~ pp.56l. 50- Platt,J.P., & Lister,G.S., 1985. 'Structural history of high pressure ~etamorphic rocks in the southern Vanoise ~assif, French Alps, and their relation to Alpine tectonic events', J. Struct. Geol. 7, 19-35. 51- Caby,R., Kienast,J.R., ~ Sa1iot,P., 1978. 'Structure, oetamorphisme et modele d'evolution tectonique des Alpes occidentales', Rev. Geog: Phys. Geo1. Dyn. 20, 307-322. 76 52- Caby,R., 1981. 'Le Mesozoique de la zone du Cocbin en Val d' Aosta (Alpes gr!'iies): imbrication tectoniques entre series issues des domaines pennique, austroalpin et oceanique', Geol. Alpine 57, 5-13. 53- Banks ,C. , & \varburton,J., 1986. 'Passive roof duplex geometry in the frontal structure of the Kirthar and Sulaiman mountain belts, Pakistan', J. Struct.Geo1. 8. 54- Chopin,C., & 110nie,P., 1984. 'A unique magnesiochloritoid bearing, high pressure assemblage from the Monte Rosa, western Alps: petrologic and 40Ar-39Ar radiometric study', Contrib. Min. Pet. 87, 388-398. 55- Vearncombe ,J., 1985. 'The structure of the Gran Paradiso basement massif and its envelope, western Alps', Eclog. Geoi. Hel v. 78, 49-72. 56- Allen,P.A., Homewood,P., 8, \hllial'ls,G.D. 1985. 'Structural and sedimentological evolution of the Holasse Basin of western Switzerland', Abstr. 6tjl. Int. Assoc. Sedimentoi. Reg. Enr. M,tg. Llieda, 15-17. 57- Richardson,S.\v., & England,P.C., crustal ecologite production in Planet. Sci. Lett. 42, 183-190. 58- Ahrens,T.J., & Schubert ,G. , 1975. its geophysical significance', 383--400. 1979. 't1etamorphic consequences of overthrust orogenic zones', ~ 'Gabbro-eclogite reaction rate and Rev. Geophys. Space Phys. 13, 59- 60- Steck,A., 1984. Alpes centrales', Hhite,S.IL, & reaction-enhanced 513-516. 'Structures de deformations tertiaires dans &108. Geoi. Helv. 77, 55-100. Knipe,R.J., 1978. 'Transformation- ductility in rocks', J. Geoi. Soc. London les and 135, 61- Reutter,K.J., & Groscurth,J., 1978. 'The pile of nappes in the northern Apennines, its unravelment and emplacement', in Cloos,H., Roeder,D. & Schmidt,K. (eds.) Alps, Apennines, Hellenides,_ Inter Union Comm. Geodyn. Sci. Rep. 38, Schweizerbart'sche Verlagsbuchhandlung, Stuttgart, 234-243. 62- Klootwijk,C. T. & Van Den Berg,J., 1975. 'The rotation of I~aly: preliminary palaeomagnetic data from the Umbrian sequence, horthem Apennines', Earth Planet. Sen,. Lett,. 25, 263--273. 63- Biju-Duval,B., Letouzey,J., & Montadert,L., 1978. 'Structure and evolution of the Mediterranean basins', in Hsu,K.J. et ai. (eds.) Initial Reports of the Deep Sea Drilling Project 42 (1), US Printing Of fice, Washington, 951-984. 64- Savostin,L.A., Sit>uet,J .C., Zonenshain,L.P., Le Pichon,X. 8·. Roulet,H.J., 1986. 'Kinematic evolution of the Tethys belt from the Atlantic ocean to the Pamirs since the Triassic', Tectonophysics 123, 1-35. TRIASSIC AND JURASSIC OCEANIC/PARAOCEANIC BELTS IN THE CARPATHIAN-PANNONIAN REGION AND ITS SURROUNDINGS M. Kazmer Eotvos University Department of Palaeontology H-1083 Budapest Kun Bela ter 2 Hungary S. Kovacs Hungarian Geological Institute H-1442 Budapest P.O.Box 106 Hungary ABSTRACT. Mafic/ultramafic rocks and associated oceanic/ paraoceanic sediments of two ages occur in the Carpathian -Pannonian region. The older rocks indicate Middle Trias- sic rifting and spreading and occur in the Meliata unit of the Inner Carpathians, in the Transylvanides of the East Carpathians, and in the Vardar zone of the Dinarides and Hellenides. These localities can be united into a Middle Triassic-Late Jurassic Vardar ocean, forming the westernmost embayment of the Triassic Tethys ocean. The younger mafic/ultramafic rocks occur in the Penninic- Valais unit of the Alps, in the Pieniny Klippen Belt of the Carpathians, in the BUkk unit of the innermost West Carpathians, in the Mecsek unit of the Pannonian region, in the Mure~ ophiolite belt of Apuseni Mts., in the Black flysch nappe of the East Carpathians and in the Severin nappe of the South Carpathians. These indicate rifting and/or spreading starting in late Early Jurassic. Most of these localities can be united in a Middle Jurassic- Early Cretaceous "Penninic" oceanic zone. Consequently, in the Carpathian-Pannonian-Dinaride region the Jurassic "Penninic ocean" lay to the north of the Triassic Vardar ocean. Its opening was most likely caused by the inter- action of the opening of the Atlantic and back-arc basin formation connected with Vardar subduction. 1. INTRODUCTION The Carpathian-Pannonian region is dissected by major fault systems /Fig. 1/ into distinct blocks of very dif- ferent history. This area provides the key to trace the continuation of Alpine ophiolitic belts eastward towards the Dinaride-Hellenide system. In the Pannonian basin the 77 A. M. C. ,'jengor (ed,), Tectonic Evolution o/the Tethyan Region, 77-92. © 1989 by Kluwer Academic Publishers. 78 ••• Occur~nce.s 0/ complete ophlo/tle sequences Fig. 1. Tectonic sketch of the pre-Neogene basement of the Pannonian basin. with occurrences of complete ophiolite se- quences. The major tectonic lineaments are indicated. "Pe- lagonian zone" includes the Flamburion, Almopias and Kasto- ria nappes /Papanikolaou, 1984/. DAV = Defereggental-Anter- selva-Valles Lineament; Lub.-Marg. Lubenlk-Margecany Li- neament. 79 Triassic Vardar ocean, which opened to the SE and the Ju- rassic Penninic one, which opened to the Wand was gene- rated by the opening of the Atlantic /Frisch, 1977/ over- lap each other. Their closure in the Late Jurassic lin the Vardar ocean/, and in the Middle-Late Cretaceous lin the "Penninic oceanic belt" sensu lato/ also overlap each other. Complete ophiolite sequences between the eastern- most tectonic window of the Alps /the Koszeg-Rechnitz window/ and the Vardar zone are only known in the Meliati- cum of North Hungary and Southern Slovakia /Reti, in press; Hovorka et al., 1984/ and in the Transylvanides of the East Carpathians /Sandulescu et al., 1981; Sandulescu and Russo-Sandulescu, 1981/. All of these represent the Trias- sic Vardar ocean /Fig. 1/. A large number of deep boreholes have proven the continuation of the nappe structure of the Apuseni Mts. in the basement of the Great Plain in Hungary /Balazs et al., 1985/. Major discontinuities dissect the territory of Hungary into fragments of quite different history and facies affinity. It is now evident, that the heterogenous block structure of the Pannonian basement /FUlop and Dank, 1985/ is due to Late Jurassic to Early Miocene horizontal displacements. By taking into account the sharp facies differences among adjacent blocks and their facies rela- tionships, the original palaeogeographic situation can be reconstructed approximately. In this paper we used the map of Hungarian data /FUlop and Dank, 1985/ in the reconstruction of the Var- dar and Penninic oceanic systems. For the Romanian Carpa- thians the following papers were mostly used: Guidebooks of the 12th CBGA Congress, Bleahu /1976/, Cioflica et al. /1980/, Ianovici et al. /1976/, Sandulescu /1980/, Lupu /1984/; for the West Carpathians: Andrusov et al. /1973/ /with the exception of the South Gemeride areal, Andrusov /1975/, Birkenrnajer /1963, 1985/, Mislk and Sykora /1981/, Un rug /1984/; for the Dinarides: Karamata et al./1980/, Pamic /1983/ among others. 2. MIDDLE TRIASSIC-JURASSIC OCEANS /VARDAR SYSTEM/ /Fig.2/ The Vardar oceanic system formed the northwesternmost em- bayment of the Triassic Tethys ocean /$engor, 1984/. Rift- ing, as evidenced by red, pelagic, Hallstatt-type lime- stones and some volcanic rocks, started in the Late Scy- thian in the Hellenides /Jacobshagen, 1972/. It became gradually younger towards the NW, but not younger than Middle Anisian /Kovacs, 1985/. The ophiolites are distri- buted in two belts /Vardar and Subpelagonian belts/. How- ever, it is now accepted by most authors that all these 80 a~ .... T I I I I I T I ~I I I c) Late Triassic situation - -- -- Locati on of the future ,Jurossic opening ~ I 1/ ( Fig. 2. Review of concepts on the connections of oceanic zones in the Carpathian-Pannonian region. a/ Single Trias- sic ocean, developing into single Jurassic ocean after Late Triassic closure of Dobrudja IHerz and Savu, 1974; Kozur, 1979/. bl Single Tethys ocean: Triassic rifting in the east, Jurassic rifting in the west. Dobrudja is inde- 81 ophiolites fat least in Greece/ originated from the Vardar zone proper, while the western belt constitutes a large ophiolite nappe only /Papanikolaou, 1984/. Sedimentation continued in pelagic, deep-water carbonate and radiolarite facies during the rest of the Triassic. The Transylvanides of the East Carpathians origi- nally belonged to this oceanic embayment /Sandulescu, 1980/ as did the Meliaticum and related deep-water Trias- sic formations in North Hungary and Southern Slovakia /Kovacs, 1985; Reti, in press/ and the Pieniny Exotic Cor- dillera /Kazmer and Kovacs, in preparation/. Their present isolated position is due to later oblique-slip and rota- tional movements of different blocks in the basement of the Pannonian basin. In North Hungary and Southern Slovakia small tecto- nic units bear Triassic deep-water formations and mafic/ ultramafic rocks. These are bordered to the west, north and east along major faults by Central West Carpathian units featuring pre-Alpine crystalline basement and Trias- sic shallow marine cover. To the south lies the BUkk unit, also featuring shelf-type Triassic formations, here of Di- naric character. The existence of a southwestern Triassic seaway connecting the BUkkium with the Dinarides, reviewed by Kovacs /1982/ has been disproven. Drillings along its supposed location between the Balaton and Mid-Hungarian lineaments yielded shallow marine Triassic rocks related rather to the Transdanubian Midmountains in Hungary and the region of the Sava folds in Yugoslavia, both forming the northern border of the postulated seaway, than to the BUkk unit /Brezsnyanszky and Haas, 1985; Kazmer, 1986/. The isolated position of the Triassic deep-water rocks Fig. 2. /Continued/ -pendent of this system /Sandulescu, 1980, 1983, 1984; Lupu, 1984/. c/ The model discussed in this paper: the Triassic Vardar rifting and the Jurassic Penninic rifting are two superimposed steps in Western Tet- hyan evolution. The Vardar ocean has been opened in Middle Triassic, containing the Meliata zone of Northern Hungary and Southern Slovakia and the Vardar zone in Yugoslavia and Greece. The surrounding Carpathian, Alpine and Dinaric units are arranged according to the Norian facies reconst- ruction of Kovacs /1982/. The Eastern Alpine - Carpathian - - Serbo-Macedonian region form a single unit adjoining Europe and forming the northern margin of the Tethys. Dobrudja formed a separate extensional basin with thinned continental crust and mafic volcanism. Ex. R. = Pieniny Exotic Ridge; Trans. = Transylvanidesi Penn. = Penninici Dobr. = Dobrudja. 82 and associated mafic/ultramafic formations is interpreted by large-scale strike-slip faulting connected with the eastward continental escape of the Bakony unit /Trans- danubian Midrnountains/ /Kazmer, 1984/. This Palaeogene displacement removed the North Hungarian-Southern Slova- kian Triassic deep-water formations from their original position at the NW termination of the Vardar zone /Kovacs, 1985/. Palaeogeographic evidences of these displacements are discussed by Kazmer and Kovacs /1985/ and Kovacs /1985/. The North Alpine deep-water Triassic Hallstatt facies belt /Salzberg facies/ /Plochunger, 1976/ formed only an exten- sion of the Vardar ocean on thinned continental crust. Pelagic Triassic rocks, mafic and ultramafic magmatic rocks and glaucophane are known in Albian-Maastrichtian conglomerates in the Pieniny Klippen Belt and south of it. These are derived from the hypothetical Pieniny Exotic Ridge /Mislk et al., 1977; Mislk and Sykora, 1981/. How- ever, because the first sign of their presence are the pebbles in Albian conglomerates, there is no evidence for the pre-Albian history of the Exotic Ridge. Mis{k et al. /1977/ supposed the existence of another Triassic oceanic trench north of the nearshore sediments of the autochtho- nous High Tatric Triassic cover. Due to the shortcomings of this model we propose an alternative one /Kazmer and Kovacs, in preparation/. We suggest that the Pieniny Exotic Ridge was part of the northern margin of the Vardar ocean. Following the Late Jurassic final closing of Vardar, fragmentation and horizontal motions brought it to the neighbourhood of the Pieniny Klippen Belt before Albian time. Closure of the Vardar ocean started in the Early or Middle Jurassic, as witnessed by the olistostromes in the "diabase-chert formation" of the Inner Dinarides and by the 170-160 m.y. radiometric age of the emplaced ophiolite bodies /Karamata and Lovric, 1978/. Collision occurred by the end of the Jurassic /"Eohellenic phase" of Jacobshagen et al., 1976/, and is marked by the appearance of shallow- water carbonates in the Inner Dinarides, Transylvanides and Inner West Carpathians. The northwesternmost manifes- tation of this tectonic activity was the emplacement of Hallstatt nappes over the marginal Dachstein platform in the Northern Limestone Alps /Plochinger, 1976/. The Jurassic subduction of the Vardar ocean s. str. formed several marginal seas along its margins /Fig. 3/. At its westernmost end the BUkk unit of northern Hungary /now displaced by subsequent strike-slip faulting/ contains Jurassic tholeiitic basalts and gabbros indicating the opening of a marginal sea /Balla et al., 1983/. The Mure~ zone was another marginal sea /South Apuseni Mts./; its opening began in the Callovian, forming two marginal seas and an island arc /Lupu, 1983/. This zone occupied the whole southern margin of Tisza unit, from Apuseni Mts. in Romania /Lupu, 1983/ to Vojvodina in Yugoslavia /Canovic and Kemenci, 1975/. 83 The marginal sea of the Sumadija zone opened in Late Jurassic time producing mafic/ultramafic magmatic rocks. Based on stratigraphy and petrography a direct connection to the Mures zone is outlined by Andelkovic and Lupu /1967/. 3. JURASSIC-CRETACEOUS OCEANS /PENNINIC SYSTEM/ /Figs. 3-4/ The opening of the Jurassic-Early Cretaceous Tethys was initiated by the opening of the Atlantic /Frisch, 1977/. The earliest /Middle Jurassic: Trlimpy, 1985/ oceanic zone of the Piemont region extended to the east not farther than the presen-day Rechnitz-Koszeg window /Koller and Pahr, 1980/. The Brianconnais ridge separating it from the Valais zone did not extend even as far as to the Engadin window /Oberhauser, pers. comm., 1985/, so the separation of the two oceanic belts is problematic. The Valais zone is most- ly of paraoceanic character. It is connected to the Outer Carpathians and, farther east, to the Outer Dacides /until the Nis-Trojan unit/, where, with the exception of the Severin nappe, no ultramafic rocks are known. The mafic rocks are usually associated with black shales /schistes lustres facies: Isler and Pantic, 1981/. In the Carpathian-Pannonian region the situation is more problematic. While detailed interpretations exist for individual oceanic or paraoceanic zones, their connections to one another have not been established. Possible connections between the oceans of the Alps and the West Carpathians are given by Birkenmajer /1977/. More recently he introduced an interpretation with three oceans /Birkenmajer, 1985/. Evidence for the northern Silesian ocean is based on the presence of alkali mafic rocks /teschenites/. In the middle ocean, the Magura basin, no mafic rocks are known /except some cineritic tuff inter- calations at Piana Botizei, Romania/. Evidence for the southernmost ocean of the Pieniny Exotic Ridge is based on exotic pebbles; however, the autochthonous position of the source rocks /Misik et al., 1977/ is highly disputable /Kazmer and Kovacs, in preparation/. Deep-water sediments of Middle Jurassic age and younger, including radiolarites, are known in the Pieniny Klippen Belt /Birkenmajer, 1977/. Their similarity with the Mecsek unit of the Pannonian region has been recognized by Birkenmajer /pers. comm., 1983/ and by Kazmer et ale /1984/. No volcanism is known in the Pieniny Klippen Belt but in the Mecsek unit a large volume of alkaline basalts 84 EASTERN ALPS Bakony SOUTHERN ALPS J2-3 paleogeographical situation Fig. 3. Middle to Late Jurassic palaeogeographical situ- ation. The Vardar ocean began to subduct below its northern margin during the Jurassic. Contemporaneously began the Penninic rifting of the Valais-Magura-Pieniny- -Mecsek-Black flysch-Severin-Nis-Trojan zone, producing troughs either with oceanic or with thinned continental crust. The Eastern Alps - Carpathians /including Bihor, Bucovinian, Geta/ and Serbo-Macedonian units of normal continental crust formed a continuous zone between the two oceanic zones are known in Hungary /Juhasz and Vass, 1974/ and in the Soviet Union /Dolenko et al., 1980/. These are interpreted as products of continental rifting /Bilik, 1983/. The vol- canism began not later than Oxfordian time /Fozy et al., 1985/ and culminated in Valanginian time /Bilik, 1974/. The Black flysch nappe in Maramure~ /East Carpa- thians/ contains Tithonian basaltic volcanics interpreted as products of intracontinental rifting /Russo-Sandulescu and Bratosin, 1985/. The E,ast Carpathians contain a Tithonian-Barremian flysch, the Sinaia beds in the Ciuc digitation of the Ceahlau nappe /Sandulescu et al., 1981/; although no mafic rocks are present, this might be consi- dered as direct continuation of the South Carpathian, ophiolite-bearing Sinaia Flysch of the Severin nappe /San- dulescu, 1980/. 85 Paleogeographic situation at the Jurassic/Cretaceous bourdary E. ALPS Bakony S. ALPS Fig. 4. Palaeogeographic situation at the Jurassic/Creta- ceous boundary. The Vardar ocean had been closed during latest Jurassic, forming the marginal basins of BUkk fits dimensions are not known/, Mure~ and Sumadija. The Penni- nic oceanic/paraoceanic zone from Valais to Nis-Trojan has reached considerable extension. The Eastern Alpine - Serbo -Macedonian continent is dissected by strike-slip faults. The Severin nappe contains a complete pre-Late Tithonian ophiolitic suite of ocean-floor origin /Cioflica et al., 1981; Savu, 1985/. A continuation of this unit is found in Yugoslavia: the Kiloma basalts in schistes lustres are of Late Jurassic/?/-Early Cretaceous age TGrubic and Ercegovac, 1983/. A possible continuation of the Severin ocean to the southeast might be the Nis-Trojan trough, which is an ex- tensional basin containing thick Tithonian-Berriasian flysch sediments /HsU et al., 1977; Nachev in Adamia, 1984/. No mafic volcanic rocks of this age are known in Bulgaria. In summary, we think that the oceanic/paraoceanic belt from the Valais zone through the Magura/Pieniny, Me- csek, Black flysch, Ceahlau and Severin zone to the Nis- -Trojan unit was not a continuous ocean. It was rather a chain of extensional basins, partly underlain by oceanic 86 Early CretaceO\s paleogeographic situation (Barremian ?) DINAR ES --=::;::::-.--....-:----:-.....---:~ Fig. 5. Early Cretaceous /Barremian?/ palaeogeographic situation. The former (East Alpine - Serbo-Macedonian con- tinent, dissected by strike-slip faults, has suffered considerable displacements. The promontory of the Pieniny Exotic Ridge /showing Transylvanian, i.e. Vardar charac- ters/ is being introduced between the Pieniny basins and the Central West Carpathians. crust /Valais, Severin/ and partly by thinned continental crust /Magura-Pieniny, Mecsek, Black flysch, Ceahlau, Nis- -Trojan/. These basins were separated and/or connected by transform faults similar to those outlined by Weissert and Bernoulli /1985/ in the Swiss Alps. However, these formed an oceanic/paraoceanic belt /the Penninic ocean s.l./ independent from the Vardar ocean and its marginal seas. These basins are similar in their age of opening: Late Jurassic to Earliest Cretaceous, and their age of closure: Middle to Late Cretaceous. During the beginning of this interval the Vardar ocean was being closed. 4. EMPLACEMENT OF THE PIENINY EXOTIC RIDGE /Figs. 5-6/ This problem is reviewed here only briefly; a detailed discussion will be published elsewhere. The palaeogeographical model outlined in this paper offers an explanation for the origin of Transylvanide /i.e. Vardar/ type exotic pebbles /ophiolitic rocks, Trias- sic deep-water carbonates, glaucophane bearing rocks, etc./ in the Albian and younger flysch deposits of the Pieniny 87 Middle Cretaceous (Albian) paleogeographic situation :I'1AGURA. ·,PIENINY::. o s 4 Fig. 6. Middle Cretaceous /Albian/ palaeogeographic situ- ation. The Mure~ - Sumadija zone and the Penninic zones from Magura to Nis-Trojan have been closed due to NW motion of Apulia and W motion of Moesia. Sedimentation in the East Alpine and Carpathian flysch region continues. The Pieniny Exotic Ridge /Ex. R./ occupied its presen position between the Pieniny sedimentary basins and the central West Carpathians. Klippen Belt and external Central West Carpathian units. By the closure of the Vardar ocean at the end of the Jurassic the East Alpine - Serbo-Macedonian micro- continent was fragmented by transverse faults. These were formed contemporaneously with the formation of the marginal seas. The most important ones bordered the Bihor unit /Fig. 4/ to the east and west, and separated it from the adjacent units. The Bihor unit /Mecsek, Villany and Apu- seni Mts./ suffered considerable counterclockwise rotation during Early Cretaceous time /Marton, 1985/. Its immediate eastern neighbour, containing the remnants of the closed Vardar ocean passed the Bihor unit to the north /Fig. 5/ and began to shed detritus in Albian time into the Pie- niny basins and into the basins of the external Central West Carpathians. This hypothesis /Kazmer and Kovacs, in preparation/ resolves the contradiction of the occurrence of Vardar-type /internal/ detritus in the external zones of the V~est Carpathians. 88 5. CONCLUSIONS 1/ The westernmost termination of the Tethys consisted of: a/ the Vardar zone, which opened in the Middle Triassic, and closed in the Late Jurassic, and its marginal seas: the BUkk, Mures and Sumadija zones, which opened in the Jurassic and closed in the Middle Cretaceous, b/ and the Penninic zone, including the Valais, Magura, Pieniny, Mecsek, Black flysch, Ceahlau, Severin and Nis- -Trojan zones, which opened in the Late Jurassic-Early Cretaceous, and closed in the Middle Cretaceous or later. 2/ The tectonic units from the Eastern Alps through West Carpathians, Bihor,flast Carpathians, South Carpa- thians to Serbo-Macedonian unit formed a continuous unit corresponding to the northern margin of the Vardar ocean until Early-Middle Jurassic time. By Late Jurassic- -Early Cretaceous time Penninic rifting formed a micro- continent or peninsula /this is supported by palaeobio- geographic data of Voros, 1977/. 3/ The breakup of this microcontinent by strike-slip faults during Early Cretaceous finally resulted in the emplacement of the promontory of the internal Pieniny Exotic Ridge in between the external units of the West Carpathians. 6. REFERENCES Adami a , S.A. /ed./ /1984/: Yurskie osadochnie geokompleksi Bolgarii i Gruzii. AN GruzSSR, Inst. Dzhanelidze, Trudy, novaya seriya, vyp. ~~, 98 p., Metsniereba, Tbilisi. Andelkovi6, M.Z., Lupu, M./1967/: 'Die Geologie der Sumadi- ja und Mures Zone. Stratigraphische Gliederung, Fa- zies, Magmatismus , Tektonik' .Carpatho-Balkan Geol. Assoc. VIII. Congress, Re~orts ~, 15-28, Belgrade Andrusov, D./1975/: 'Apercu bref du bati des Carpathes Occi- dent ales '. X. Congress Carpatho-Balkan Geol. Assoc., General proceedin Bilik, 1./1974/: 'Unterkretazische Vulkanite des Mecsek- -Gebirges'. Acta Geol. Acad. Sci. Hung. J:~, 3/4, 315-325, Budapest Bilik, 1./1983/: 'Lower Cretaceous submarine /rift/ vol- canism in South Transdanubia /South Hungary/'. In: Bisztricsany, E., Szeidovitz, Gy. /eds./: Proc. 17th Assembly Eur. Seismol. Comm., Budapest, 1980, pp. 569-576, Akademiai Kiad6, Budapest 89 Birkenmajer, K./1963/: 'Esquisse de la stratigraphie du Mesozoique et du Paleogene dans la Zone des Klippes Pienines en Pologne'. Inst. Geol. Biul. !§~, 207-233, Warszawa --- Birkenmajer, K./1977/: 'Jurassic and Cretaceous lithostrati- graphic units of the Pieniny Klippen Belt, Carpathians, Poland'. Studia Geol. Polon. !2, 1-159, Warszawa Birkenmajer, K./1985/: Main Geotraverse of the Polish Car- pathians /Cracow-Zakopane/. Guide to Excursion ~, XIII. Congress Carpatho-Balkan Geol. Assoc., Cracow, 198 p. Bleahu, M./1976/: 'Structural position of the Apuseni Mountains in the Alpine system'. Rev. Roum. Geol. Geophys. Geogr., Geologie ~Q, 1, 7-19, Bucuresti Brezsnyanszky, K., Haas, J./198~7: 'The new tectonic map of Hungary'. Proceeding reports XIII. Congress Car- patho-Balkan Geol. Assoc. I, 174-177, Cracow canovic, M., Kemenci, R./1975/:-'Le Jurassique et le Cre- tace au fond de la partie du bassin pannonien dans la Voivodine'. Ann. Geol.Penins. Balkanique ~~, 35-59, Beograd Cioflica, G., Lupu, M., Nicolae, I., Vlad, S./1980/: 'Alpine ophiolites of Romania: Tectonic setting, mag- matism and metallogenesis'. Anuarul Inst. Geol. Geofiz. 2~, 79-95, Bucuresti Cioflica, G.,-Savu, H., Nicolae, I., Lupu, M., Vlad, S. /1981/: Alpine Ophiolitic Complexes in South Carpa- thians and South Apuseni Mountains. Carpatho-Balkan Geol. Assoc. XII. Congress, Guide to Exc. ~J, Guide- book series 1§, 80 p., Inst. Geol. Geophys:~ Bucuresti Dolenko, G.N., BOlchevskaya, L.T., Danilovich, L.G. et al. /1980/: Glubinnoe stroenie, razvitie i neftegazonos- nosty Ukrainskih Karpat. Naukova Durnka, Kiev, 148 p. Fozy, I., Lantai, Cs., Schlemmer, K./l985/: 'A Pliens- bachian - Lower Cretaceous profile at Zobakpuszta /Mecsek Mts., Hungary/'. Annales Univ. Sci. Budapest., Sect. ~eol. ~2, 97-115, Budapest Frisch, W./1977/:-T Die Alpen im westmediterranen Orogen -- eine plattentektonische Rekonstruktion'. Mitt. Ges. Geol. Bergbaustud. osterr. ~!, 263-275, Wien FUlop, J., Dank, V./eds./ /1985/: Geological Map of Hun- gary Without Cenozoic Formations. 1 : 500.000. Hun- garian Geological Institute, Budapest 90 Grubic, A., Ercegovac, M./1983/: 'Note on the age of Kiloma metadiabase formation /NE Serbia/'. C.R. Seances Soc. Serbe Geol. pour 12§£, 55-60, Beograd Herz, N., Savu, H./1974T:-'Plate tectonic history of Roma- nia'. Geol. Soc. Amer. Bull. §~, 7, 1429-1440, Boulder Hovorka, D., Jaros, J., Kratochvil~-M., Mock, R./1984/: 'The Mesozoic ophiolites of the Western Carpathians'. Krystalinikum 11, 143-157, Prague Hsli, K.J., Nachev, I:K., Vuchev, V.T./1977/: 'Geologic evo- lutionof Bulgaria in light of plate tectonics'. Tecto- nophysics ~Q, 3/4, 245-256, Amsterdam ----- Ianovici, V., Borco~, M., Bleahu, M., Patrulius, D., Lupu, M., Dimitrescu, R., Savu, H./1976/: Geologia Muntilor Apuseni. Editura Academiei, Bucure?ti, 631 p. Isler, A., Pantie, N./1980/: "'Schistes-lustres Ablage- rungen der Tethys'. Eclogae geol. Helv. Z~, 3, 799-822, Basel Jacobshagen, V./1972/: 'Die Trias der mittleren Ost-Agais und ihre palaogeographische Beziehungen innerhalb der Helleniden'. Z. Deutsch. Geol. Ges. ~~~, 445-454, Hannover Juhasz, A., Vass. G./1974/: 'Mesozoische Ophiolite im Beckenuntergrund des Grosser Ungarischer Tiefebene'. Acta Geol. Acad. Sci. Hung. 1§, 3/4, 349-358, Budapest Karamata, S., Lovric, A./1978/: 'The age of metamorphic rocks of Brezovica and its importance for the explana- tion of ophiolite emplacement'. Bull. Acad. Serbe, Sci. 11, 1-9, Beograd Karamata, S:, Majer, V., Pamic, S./1980/: 'Ophiolites in Yugoslavia'. In: Rocci, G./ed./: Tethyan Ophiolites, Ofioliti 1, Spec. issue, 105-125, Bologna Kazmer, M./1984/: ContinentIa escape of the Bakony-Drauzug unit in the Paleogene'. Altalanos Foldtani Szemle £Q, 55-103, Budapest -- Kazmer, M./1986/: 'Tectonic units of Hungary: Their boun- daries and stratigraphy /A bibliographic guide/'. Annales Univ. Sci. Budapest., Sect. Geol. £§, 45-120, Budapest -- Kazmer, M., Kovacs, S./1985/: 'Permian-Paleogene paleo- geography along the eastern part of the Eeriadriatic Lineament: Evidence for continental escape of the Bakony-Drauzug unit' • Acta Geo!. Hung. £§, 1/2, 69-82, Budapest -- Kazmer, M., Kovacs, S./in preparation/: 'Mesozoic oceanic zones in SE Europe -- a review of concepts' Kazmer M., Kovacs, S., Pero, Cs./1983/: 'Outline of the structure of the East Carpathians'. Altalanos Fold- tani Szemle l§, 3-75, Budapest Kazmer, M., Kovacs~ S., Pero, Cs./1984/: 'Tanulmanyuton a Pienini-szirtovben es a Lengyel-Tatraban'. Altala- nos Foldtani Szemle ~Q, 103-144, Budapest Koller, F., Pahr, A./1980/: 'The Penninic ophiolites on the eastern end of the Alps'. Ofioliti ~, 1, 65-72, Bologna 91 Kovacs, S./1982/: 'Problems of the"Pannonian Median Massif" and the plate tectonic concept. Contributions based on the distribution of Late Paleozoic-Early Mesozoic isopic zones'. Geol. Rundschau 11, 2, 617-639, Stuttgart -- Kovacs, S./1985/: 'North Hungarian facies types: A re- view'. Acta Geol. Hung. ~1, 3/4, 251-264, Budapest Kozur, H./1979/: 'Einige ProbIeme der geologischen Ent- wicklung im slidlichen Teil der Inneren Westkarpaten'. Geol. Paleont. Mitt. Innsbruck 2, 4. 155-170, Innsbruck - Lupu, M./1983/: 'The Mesozoic history of the South Apu- seni Mountains'. Anuarul Inst. Geol. Geofiz. ~Q, 115-124, Bucuresti -- Lupu, M./1984/: 'Problems of the European continental mar- gin in the Transylvanian-Pannonian area'. Anuarul Inst. geol. Geofiz. ~~, 1-10, Bucuresti Maher, M./1981/: 'Plate tectonics and extension of Penni- nicum in West Carpathians'. Mineralia slovaca lJ, 4, 289-306, Bratislava -- Marton, E./1985/: 'Tectonic implications of palaeomagnetic results for the Carpatho-Balkan and adjacent areas'. In: Dixon, J.E., Robertson, A.H.F./eds./: Geological Evolution of the Eastern Mediterranean. Spec. Publ. of the Geological Society 11, 645-654, Blackwell, Oxford Misik, M., Mock, R., SYkora,-g./1977/: 'Der Trias der Klip- penzone der Karpaten'. Geologicky zbornlk £§, 1, 27-69, Bratislava -- Mislk, M., Sykora, M./1981/: 'Der Pieninische exotische Rlicken, rekonstruiert aus Gerollen karbonatischer Gesteine kretazischer Konglomerate der Klippenzone und der Manin Einheit'. Zapadne Karpaty,seria geolo- gia 1, 7-111, Bratislava Pamic, J.71983/: ' Considerations on the boundary between lherzolite and harzburgite subprovinces in the Dina- rides and Northern Hellenides'. Ofioliti§, 1, 153-164, Bologna - Papanikolaou, D./1984/: 'Introduction of the geology of Greece; the pre-Alpine units'. In: IGCP Project No.5, Field Meeting in Greece, Field Guide 1, '1-35, Athens - Plochinger, B./1976/: 'Die Oberalmer Schichten und die Platznahme der Hallstatter Masse in der Zone Hallein -Berchtesgaden'. N. Jb. Geol. Palaeont. Ahh. 1~1, 3, 304-324, Stuttgart --- Reti, Zs./in press/: 'Triassic ophiolite fragments in an evaporite melange, Northern Hungary'. Ofioliti, Special volume of the 1980 Nancy Conference, Bologna 92 Russo-Sandulescu, D., Bratosin, 1./1985/: 'Caracteres et signification du complexe basique de la nappe du Flysch Noir /Mints du Maramure~, Carpathes Orientales/'. Proc. Report XIII. Carpatho-Balkan Geol. Congress, Ad- ditionally received reports, 112-115, Cracow Sandulescu, M./1980/: 'Analyse geotectonique des chaines alpines situees autour de la Mer Noire occidentale'. Anuarul Inst. Geol. Geofiz. ~2' 5-54, Bucure~ti Sandulescu, M./1983/: 'Le probleme-de la marge continentale europeenne dans l'areal Carpatho-Balkanique'. Anua- rul Inst. Geol. Geofiz. ~Q, 199-208, Bucurepti---- Sandulescu, M./1984!: 'Compared Alpine geotectonic models'. Anuarul Inst. Geol. Geofiz. ~!, 343-351, Bucure~ti Sandulescu, M., Krautner, H.G., Balintoni, I., Russo-San- dulescu, D., Micu, M./1981/: The structure of the East carpathians /Moldavia-Maramure~ areal. Carpatho- -Balkan Geol. Assoc. XII. Congress, Guide to Exc. ~1, Guidebook Series ~1, 92 p., Inst. Geol. Geofiz., -- Bucure9ti -- Sandulescu, M., Russo-Sandulescu, D./1981/: 'The ophi- olites from the Rarau and Haghimas synclines. Their structural position, age and geotectonic evolution'. Dari de Seama Inst. Geol. Geofiz. 22, 5, 103-114, Bucure~ti -- Savu, H./1985/: 'Alpine ophiolites in the Severin nappe /Mehedinti Plateau/ and their associated copper and pyrite mineralizations /Romania/'. Proc. report Car- patho-Balkan Geol. Assoc. XIII. Congress, Additionally received reports, 116-119, Cracow Sengor, A.M.C./1984!: The Cirnrneride Orogenic System and the Tectonics of Eurasia. Geol. Soc. Amer. Spec. Paper 1~~, xi + 82 p., Boulder TrUmpy, R./Ig85/: Die Plattentektonik und die Entstehung der Alpen. Natf. Ges. Zlirich, Neujahrsblatt ~~~~, 47 p., Orell FUssli Graph. Betriebe AG., ZUrich Unrug, R./1984/: 'Geodynamic evolution of the ~arpathians'. Ann. Soc. Geol. Polon. 52, 1/4, 39-66, Krakow Voros, A./1977!: 'provinciality of Mediterranean Lower Jurassic brachiopod fauna: causes and plate-tectonic implications'. Palaeogeogr., Palaeoclimatol., Palaeo- ecol. ~1, 1, 1-16, Amsterdam Weissert, H~J., Bernoulli, D./1985/: 'A transform margin in the Mesozoic Tethys: evidence from the Swiss Alps'. Geol. Rundschau Z~, 3, 665-679, Stuttgart MAJOR EVENTS OF THE TECTONO-SEDIMENTARY EVOLUTION OF THE NORTH HUNGARIAN PALEO-MESOZOIC: HISTORY OF THE NORTHWESTERN TERMINATION OF THE LATE PALEOZOIC - EARLY MESOZOIC TETHYS s .. Kovacs Hungarian Geological Institute H-1442 Budapest P.O.Box 106 Hungary ABSTRACT. The North Hungarian region has a peculiar posi- tion in the Alpine orogen: it contains the northwestern- most occurrence of marine Late Paleozoic sediments and Early Mesozoic oceanic assemblages. The oldest recognizable synsedimentary movements in the BUkkium are upper Lower or Middle devonian volcano- tectonic movements that formed olistostromes with basic volcanic matrix and limestone olistoliths. In the Upper Devonian a carbonate platform developed adjacent to a basin that contained basic volcanic rocks. The BUkkide domain did not consolidate during the Variscan tectoge- nesis, but experienced only synsedimentary movements with- out strong folding or metamorphism. During the Sudetic phase the former carbonate platform was disrupted and a .,flysch basin developed / in the Szendro and BUkk Paleo- zOic/. The Leonian phase was followed by the shallowing of the sea. The shallow marine, mostly carbonate sedi- mentation continued from the Upper Carboniferous into the Triassic, but was probably broken by a local uplift in the Lower ,Permian. Middle Triassic rifting led to the formation of a central oceanic basin, flanked on both from the Sand N by wide carbonate shelves. The /Eastern/ bUkk was part of the southern mobile shelf, characterized by signifi- cant volcanism, local uplifts and intraplatform basins. The Rudabanya Triassic /especially the sequence of the B6dva nappe/ and the ~~liata Triassic in Slovakia be- longed to the central, open marine basin. Sedimentation in the latter took place on an oceanic crust; the pre- sence of a complete Middle Triassic ophiolitic suite has recently been proven /Reti, 1984/. The Silice nappe /Aggtelek - Slovak Karst/ was originally part of the northern stable shelf /Carpathian shelf/. The rifting, 93 A. M. C. ~engor (ed.), Tectonic Evolution olthe Tethyan Region, 93-108. © 1989 by Kluwer Academic Publishers. 94 according to the sedimentological record, began in the Middle:Aflisian, when the former uniform carbonate plat- form was disrupted and pelagic sedimentation started. In the central basin deep water sediments, partly below the CCD, were deposited throughout the rest of the Triassic. After the Triassic the outer shelf region of the Blikkium collapsed and "schistes lustres"-type Jurassic sediments were deposited and were associated with an in- complete ophiolitic suite. The Jurassic sediments in the Rudabanya Mts. are similar in many respects, but contain alkaline rhyolitic volcanic rocks. The "geosynclinal stage" was most probably terminated by the uppermost Jurassic tectogenesis marked by sporadic occurrences of shallow water Tithonian sediments. Lower Cretaceous rocks are not known in N Hungary. Senonian Gosau-type conglo- merates in the Blikkium followed the first phases of nappe building. 1. INTRODUCTION The Paleozoic and Mesozoic terrains in North Hungary /Fig. 1/ occupy a unique position within the Alpine-Car- pathian-Dinaric mountain chains: corning from the Helleni- des and Dinarides to the NW, they contain the northwes- ternmost occurrences of continuous marine Late Paleozoic and deep water to oceanic Triassic sequences. Intense in- vestigations carried out in the last few years by the Hungarian Geological Institute Ire-mapping, sedimentolo- gical, petrographical and biostratigraphical studies/ have resulted in major changes in the stratigraphy and structure of the region. The new stratigraphy developed for the Szendro-Uppony Paleozoic and for the Aggtelek- -Rudabanya Triassic is based mostly on conodont-biostra- tigraphic investigations carried out by the present author and on radiolarian studies of the siliceous se- diments, made by H. Kozur and P. de Wever. The area is characterized by a complicated and not yet completely understood nappe structure that has predo- minantly southern vergence. At present, the following major tectonic units are distinguished /Grill et al., 1984 and in preparationj Pelikan, in pressj Kovacs and Pero, 1983b/: A. Aggtelek-Rudabanya Mts. /from the top down/: 1. Silicicum /non-metamorphosed Upper Permian to Ju- rassic/ : - Silice nappe /Triassic of predominantly carbo- nate platform facies, but with Hallstatt facies in the Norian/ - Szolosardo unit /Triassic of predominantly slope Fig. 1. Simplified geological map of the Paleozoic and Mesozoic mountains in North Hungary and adjacent South Slovakian territories. >- 0:: « ::c o 1 I Rudabanya 10 1 D······· ...... 2 I' •• 20 'I 30km 1 1: Paleozoic; 2: Triassic; 3: Jurassic 4: Tertiary + Quaternary 95 96 facies from the Middle Anisian onward/ - Bodva nappe /Tiassic of deep-water facies from the Middle Anisian onward; Jurassic of predo- minantly black shale facies with alkaline rhyo- lite volcanism/ - Komjati unit /evaporitic melange at the base, with obducted oceanic slabs/ 2. Meliaticum /oceanic suite, with radiolarites, ultramafics and glaucophanites; in the Hungarian part present only within the Komjati unit/ 3. Tornaicum /anchi- to epimetamorphosed Triassic of grey basinal facies from the Middle Anisian onward/ B. "Blikkium" /s.1./ 1. Szendro Mts. /epimetamorphosed Plaeozoic/ 2. Uppony Mts. /anchimetamorphosed Paleozoic/ - Uppony unit /carbonate Paleozoic/ - Tapolcsany unit /detrital Paleozoic/ 3. Bukk Mts. - North Bukk anticline /marine Middle Carboni- ferous to Middle Triassic - Kisfennsik /=Little Plateau/ nappe /nonmetamor- phosed Triassic of carbonate platform facies/ - Fennsik /=Plateau/ nappe /anchimetamorphosed Triassic of predominantly carbonate platform facies/ - South Bukk unit /Jurassic black shales, radio- larites and mafic rocks, with isolated occur- rences of Triassic of predominantly carbonate platform facies/ The red radiolarite-pillow lava sequence of Darno Mt. formerly regarded as the westernmost part of the Bukk Mts., is assigned now to the Meliaticum, because its Ladinian age has been proven by radiolarians /De Wever, in press/. The continuous Carboniferous to Lower Triassic se- quences and deep-water Triassic assemblages that occur in northern Hungary are present in a completely isolated position /see Kazmer and Kovacs, this volume, Fig. 1/; their necessary connections to the open sea cannot be ex- plained without microplate movements along the Mid-Hunga- rian tectonic zone /Kovacs, 1982; Kovacs and Pero, 1983a, 1983b; Brezynyanszky and Haas, 1985/. Our better understanding of the stratigraphy enabled us to outline the major events of the tectono-sedimentary development in this area. The rocks were deposited in the small embayment of the vast v-shaped Tethys ocean between Laurasia and Gondwana. They are probably the westernmost deposits of the ancient Tethys ocean. In the present pa- per special emphasize is given to the late Variscan evolution and to the Triassic opening of the ocean. 2. THE VARISCAN EPOCH 2.1. The Devonian rifting The oldest detectable synsedimentary tectonic movements in North Hungary were related to the Devonian rifting, 97 and were accompanied by some tholeiitic volcanic activity. Close relationship with carbonate sediments and the rather small amount of volcanic rocks suggest, that the sedimen- tary basin was underlain by attenuated continental crust /Kovacs andveto-Akos, 1983/. The first manifestation of tectonically controlled sedimentation is recognized in the BASH - KIR IAN SERPUKHOVIAN + UPPER VIS~AN ~/-/-I j.!. / 1- / _ / _ - / _ / _ 1 - 1-. I-I - \--\ - 1- \- \-1-\-1 FRASNIAN / - i.!. - / - / - I - / - / - / I/_/~-'- 1,/-/_/- -/- 1 -I - 1-1-\- ~3 ~ Fig. 2. Tectono-sedimentary evolutionary scheme of the Rakaca unit of Szendro Paleozoic during the Late Devonian- -Middle) (arboniferous. Building of carbonate platforms ceased in the Upper Devonian and only pelagic fissure fil- lings are the witnesses of the sedimentation until the Upper Visean. Due to the Sudetic tectonophase, the former carbonate platform disrupted, and after a short clastic event, patch reefs were built on blocks in higher posi- tion, while over subsided blocks a clayey-sandy flyschoid -type sedimentation took place, with olistostromes and olistothrymmata. 1: Platform carbonates /Lower and Upper Rakaca Marble Formation/; 2: Pelagic limestone fissure fillings in the Lower Rakaca Marble /up to the Upper Vi- sean/; 3: dark basinal limestones /Verebeshegy Limestone Mmeber/ interfingering with the patch reefs of the Upper Rakaca Marble; 4: Clastics of the Szendro Phyllite Forma- tion, with carbonate olistostromes and olistothrymmata. 98 late Lower or Middle Devonian Strazsahegy Formation of the Uppony Mts. This unit contains schalstein-type basic volcanic rocks and is associated with olistostromes with an altered metabasalt lava matrix and Silurian-lowermost Devonian limestone olistoliths /Kovacs, in press c/o In the Frasnian a facies differentiation is recog- nizable in the Szendro Paleozoic: on topographically high- er areas carbonate platforms of the Lower Rakaca Marble Formation were formed, while in the basins conodont bear- ing lime mud with basic tuffitic admixture was deposited /Abod Limestone Formation/. Building of carbonate plat- forms stopped in the Frasnian and there was a long hiatus that lasted until the Upper Visean. During this time in- terval the former carbonate platform formed an underwater plateau, from which bottom currents swept away the lime mud. Pelagic limestone fissure fillings with conodonts from almost all the zones within the hiatus indicate, that sedimentation did occur during that time interval /Fig. 2/. In the deeper basin areas, however, sedimentation of the Abod Limestone with tuffitic admixture continued until the end of Famennian. This formation is present both in the Szendro and Uppony Mts. and in the latter it is accompanied by metabasalts, which probably also is the source for the tuffitic admixture. The outpoured volcanic material was intimately mixed with the unconsolidated lime mud /Kovacs and Veto-Akos, 1983/ and subaqueous volcanic debris flows were also formed. Formations overlying the Abod Limestone are unknown in the Szendro Paleozoic, whereas in the Uppony Paleozoic basin carbonate sedimentation continued until the late Visean, but without associated volcanism. 2.2. The time of Variscan tectogenesis Facies changes are the only manifestations of the Varis- can tectogenesis in the Blikkium s.l. The effect of a Variscan metamorphism, if there was any, was not stronger than that of the Alpine one /Arkai, 1983/, but the sedi- mentary history seems to exclude even significant fold- ing /Kozur and Mock, 1979; Kovacs et al., 1983/. The Szendro Paleozoic rocks contain evidence for the disruption of the Lower Rakaca Marble carbonate plat- form in the late Visean /Fig. 2/, an event that can be assigned to Sudetic tectonism. It was preceded by a short event of clastic sedimentation. After disruption, blocks that stayed high received renewed carbonate platform se- dimentation and small patch reefs were built. These sedi- ments interfingered basinward with dark micritic lime- stones. At the same time in basins developed on subsided blocks, a clayey-sandy turbiditic sedimentation took 99 place, with olistostromes and olistothrymmata deriving mostly from the adjacent patch reefs. This type of sedi- mentation controlled by coeval block-faulting continued till the lower Bashkirian. It was followed by deposition of monotonous black shales with occasional sandy turbidi- tes /Kovacs and Pero, 1983b/. Rocks representing the continuation of the Variscan sedimentary cycle are exposed in the Northern BUkk anti- cline unit. This sequence starts with similar Middle Car- boniferous flyschoid rocks /Szilvasvarad Formation/, as the youngest part of the Szendro Paleozoic, with the same pre-metamorphic mineral composition /Arkai, 1983 and pers. comm./. Because their great thickness and similarity, the flyschoid rocks of the two mountains very likely overlap in age with each other, though no biostratigraphical data are available from them /Kovacs et al., 1983/. On the contrary, the Middle Carboniferous of the Uppony Paleozoic /Upper Visean-Lower Iashkirian is proven biostratigraphically/ is not of flyschoid-type and contain alternating limestones and shales without redepositional features. Fig. 3 shows on a palinspastic reconstruction the correlation of the innermost West Carpathian sector with that of the Eastern and northern Southern Alps /after Schonlaub, in Oberhauser, 1980/ for the Middle Carboni- ferous /post-Sudetic/ time. /Note: assignment of the early Variscan crystalline mountain belt of the Eastern Alps to the "Central West Carpathians" in the Carpathian sector means only its eastward continuation in a general way and not the present-day northern vicinity of the·Geme- ride and BUkkide units; cf. Kovacs and Pero, 1983a,b/. The Middle Carboniferous of the Uppony Paleozoic /Lazberc Formation/ is compared with the Dult Formation of the Graz Paleozoic, which suggests its position on the clastic outer shelf. On the other hand, the Middle Carboniferous flyschoid formations of the Szendro+and BUkk Paleozoic correspond to the Hochwipfel flysch of the Carnic Alps and South Karawanken Alps, so they represent the conti- nuation of the flysch basin. Summarizing the effect of the Variscan tectonophases, the Sudetic phasecaused only facies changes in the BUkkium. The "Erzgebirge" phase left no traces in the analogous sequences of the Carnic Alps and South Karawanken Alps /cf. Schonlaub, 1979 and in Oberhauser, 1980/; similarly, it does not seem to be probable, that it played any sig- nificant role within the flyschoid basin of the BUkkium. /cf. Kovacs et al., 1983/. The Paleozoic of the Carnic +Which itself is only flyschoid, when compared with the typical Alpine flysch of the outer Carpathians. 100 Alps and South Karawanken Alps was folded during the Leonian pahse at the turn of the Lower/Upper Moscovian. In the BUkkium, however, this phase resulted only in facies changes /see below/. It is possible, that the flyschoid basin of the BUkkium simply had been filled up by that time and shallow water sedimentation followed /Kovacs et al., 1983/. -6 l c 9 ~~ j ~ ~~ 6 . u ~ ~~ 0 0 .~ ~ ~ % ~. w w ~ ~~ ztt ~ ~ ~ w~ w w UU ~ ~ 0 m Fig. 3. Correlation of the Middle Carboniferous palin- spastic section of the ~est Carpathians with that of the Austrian Alps /drawing after Schonlaub, in Oberhauser, 1980/. 2.3. Tke late Variscan development The post-tectonic late Paleozoic development can be studied in the Northern BUkk anticline unit. The flyschoid se- quence of the Szilvasvarad Formation was followed by shallow marine fusulinid-coral-phylloid algal limestones, alternating with non-graded shales, sandstones and cong- lomerates /Malyinka Formation, comparable with the Auer- nig Group of the Carnic Alps ans South Karawanken Alps/. No unconformity between the two formations has been dis- covered /Balogh, 1964/. The Malyinka Formation ranges in age from Upper Moscovian to Gzhelian, but Kozur /1984b/ suggested, that it may reach even into the Asselian. 101 The contact between Carboniferous and Permian for- mations is cut by a Cretaceous or Tertiary fault /Peli- kan, pers. comm./. Because a new cycle starts in the Permian, it may indicate, that sed1mentation was broken by local germanotype uplifts. In the lower part of the /?/Lower-Middle Permian detrital-lagoonal Szentlelek Formation sandstones and shales predominate, with traces of acidic volcanic activity, in the middle part evapo- rites, while in the upper part vuggy dolomites /FUlop, in preparation/. This hypersaline environment was trans- formed into a normal shallow marine environment in the Upper . P'.errr.ian , with deposition of black algal limestones /"Bellerophonkalk, Nagyvisnyo Limestone Formation/. In fact, this Permian sedimentation was already the beginning of the Alpine sedimentary cycle. 3. THE ALPINE EPOCH In the BUkk Triassic sedimentation continued with a thick /120 m/, light coloured oolitic limestone sequence /Ge- rennavar Limestone Fm./. However, at the boundary a gap in sedimentation was suggested by Kozur /pers. comm./, which might be a Tethyan-wide event. The major part of the Lower Triassic is constituted by a shallow marine marl-limestone-shale sequence of Werfen facies. Building of carbonate platforms started in the Anisian with the peritidal Hamor Dolomite Fm. In the Aggtelek-Rudabanya domain, the Alpine sedi- mentary cycle started in the Upper Permian with lagoonal evaporitic sedimentation. The environment changed in the Lower Triassic to shallow marine, with the deposition of a sequence of Werfen facies that is detrital below and becomes more calcareous upward. The Lower Anisian Gutenstein Fm. of restricted lagoon facies represents the initial stage of carbonate platform development. The overlying Steinalm Fm. is already a typical open shelf carbonate platform facies. Until this time the facies zones were not differentiated. 3.1. The Triassic rifting Rifting in the Aggtelek-Rudabanya domain started with the disintegration of the uniform Steinalm carbonate platform and the onset of pelagic carbonate sedimentation. it re- sulted in the formation of a central, deep water basin /partly with oceanic crust/, flanked on the north by the outer shelf domain of the Silice nappe /s.s./ and on the south ba that of the BUkk /Fig. 4/. In the Aggtelek- -Rudabanya domain the following facies zones developed, which persisted throughout the rest of the ~iassic S H E L F R F T N G B A S I N S H E L F IC O N T1 N EN - O U T E R O U T R I N N E R 1 FC Rlt lN D B ij k Da rn e H e I i a t a R u d a b a n y a S ii i c a N. M t. S e ri e s (s l.) M ts. (?) (1) 111 . ' . c: I . . . . . . , _ ::: ; . . . . . I , I 1I Ic : _~ ~~ 'C '- = a ::> E 111 + ? + + ? + ? ~ - - ; r 0 9 . _ N 1 . + :: > .t: l ' L ' _ _ + + u. .~ 0>< II N I ~ ~ ~ 1 1 _ 1 6 ~ , . E£ t~ 7 t-H 8 12 B I 7 I+? + 1 ~ Q i5 1l 13 V-. ./'\!a 18 ~ 4 IT 1J § § 1 4 ~ 1 9 ~ E ll 10 I v ,/v .... ! 15 lE .J 20 F ig . 4. S ch em at ic T ri a ss ic p a li n sp a st ic s e c ti o n t h ro u g h th e w e s t C ar p at h ia n s /N ot a ll u n it s a n d fo rm at io n s a r e in d ic at ed !/ 1 : c o n ti n en tl a c r u s t; 2: n e w ly fo rm ed o c e a n ic c r u s t; 3: c r u s t m o bi le t h ro u g h o u t th e L at e P al eo zo ic -M es o zo ic ; 4: U pp er P er m ia n li m es to n e; 5: U pp er P er m ia n e v a p o ri te s; 6: W er fe n F o rm at io n /m ar in e/ ; 7: " B u n ts an d st ei n " /c o n ti n e n ta l/ ; 8: G u te n st ei n F or m a- 13 103 /Grill et al., 1984; Kovacs, in press a/: Aggtelek facies: Building of carbonate platforms continued until the Upper Carnian, with local interruptions at the Anisian/Ladinian boundary interval by pelagic intrashelf basins. This shelf margin domain subsided in the Upper Carnian and then pelagic Hallstatt lime- stones were deposited in the Norian. Szolosard6 facies: It represents the shelf slope environ- ment with pelagic sedimentation from the Middle Anisian onward. Redepositional phenomena /intraforma- tional conglomerates, allodapical limestones/ are very characteristic of this facies zone. B6dva facies: red deep water carbonate facies from the Middle Anisian onward. Characteristic are purple shale intercalations /reflecting breaks in carbonate deposition/ and juvenile pelecypod coquina turbidites in the Uadinian and Carnian. Carbonates locally pass into siliceous shales. In the Norian Hallstatt facies, in contrast to that of the Aggtelek facies, redeposi- tional phenomena are especially common. The bottom was very uneven, controlled by block-faulting, which resulted in a great facies variety. Meliata facies: It represents the deepest, axial part of the basin. It was deposited partly on thin continen- tal crust Ired Ladinian radiolarites immediately above Steinalm Limestones; Meliata fac1es~s~s~/~.and partly on oceanic crust /Tornakapolna facies/. In Hungarian territory, in the Aggtelek-Rudabanya Mts., the oceanic suite is present only as obducted slabs /serpentinites of Iherzolitic origin, gabbros, pillow basalts, red radiolarites/ in an evaporitic melange /the Komjati unit/. Its Triassic age has recently been proven by radiometric age determination and by Ladinian radiolarians /det. H. Kozur/ from inter pil- low radiolarites /Reti, in press/. The Ladinian pillow lava - red radiolarite sequence of Fig. 4. /continued/ -tion; 9: Middle Triassic carbonate platforms /Steinalm+Wetterstein Formations/; 10: Anisian dolomite /Blikk/; 11: Anisian balck shales, marls, cherts; 12: Pelsonian-Illyrian red limestones /Rudabanya-Meliata/; 13: Middle Triassic basinal limestones /Schreyeralm, Reifling, Nadaska, Hallstatt limestones/; 14: Ladinian or Carnian shales; 15: Ladinian-Carnian volcanics; 16: Middle Triassic radiolarites, siliceous shales; 17: Carnian det- rital formations; 18: Upper Triassic carbonate platforms /Tisoves, Dachstein limestone, Main Dolomite+ Plateau limestone in Blikk/; Upper Triassic basinal limestones; 20: Carpathian Keuper. 104 Mt. Darno, formerly regarded as the westernmost part of the BUkk is assigned now also to this facies. Torna facies /building up the lowermost, anchi- and epi- metamorphosed unit, the Tornaicum/: grey pelagic ba- sinal facies from the Middle/Upper Anisian onward. y - y I~ 2~\~ Open sea Self ( In part oceanic basement ) East European platform 3\=-= =-1 Epicontinental sea 4 .... \_~ Drylond Fig. 5. Generalized Middle Triassic /Ladinian/ palins- pastic sketch of the northwestern end of the Tethys, show- ing the position of the North Hungarian units. Abbrevia- tions: R.-M.: Rudabanya-Meliata Triassic; S. Alps: South- ern Alps; Transd. M.: Transdanubian Midmountains; N. L. Alps: Northern Limestone Alps; W. Carp.: west Carpathians; E. Carp.: Last carpathians; S. Carp. South Carpathoans; Balc.: Balcanides; Transylv.: Transylvanides; p.5 Pieniny Exotic Cordillera. In the BUkkide domain rifting did not affect the Hamor dolomite platform, which continued until the end 105 of Anisian. Local emersions at the Anisian/Ladinian boundary produced South Alpine-type conglomerates. Signi- ficant volcanic activity took place in two phases. The first, in the Lower Ladinian, produced intermediate e- ruptions /porphyrites/, partly of ignimbritic type. The second, in the late Ladinian-Carnian, produced basic ef- fusive rocks. Building of carbonate platforms continued throughout the Triassic, and their sediments interfingered w.ith pelagic intrashelf basins /grey, cherty limestones/. The reconstructed paleogeographical position of the North Hungarian Triassic is shown on Fig. 5: the central oceanic basin /Heliata facies/ lay in the northwestern- most termination of a Red sea-type young rifted ocean /the Vardar ocean/, flanked on the NE by the stable Carpathian shelf and on the SW by the labile Outer Dinaric shelf /cf. Kovacs, 1982 and in press, at. 3.2. The Jurassic evolution The strong sedimentary facies differentiation formed by the Middle Triassic rifting was less well developed in the Jurassic, e.g. between the central deep-water basin and its southern shelf. With the predominance of dark shales, both region were in the "schistes lustres" facies realm lin sense of Isler and Pantie, 1981/. The Triassic outer shelf domain of the BUkk collapsed and changed into a deep water environment with deposition of dark shales and black, green and red radiolarites. The Jurassic age is proven by radiolarians /det~: H. Kozur/. The intense synsedimentary tectonic movements initiated slides and sediment gravity flows forming olistothrymmata and oliststromes, which are common i~ the Southern BUkk shale unit. The associated mafic rocks represent an incomplete ophiolite sequence, with the pillow lava and the sheeted dyke complexes, which in- dicate a marginal sea-type rifting /Balla et al., 1983/. Triassic rocks in the Southern BUkk unit occur spora- dically and are of carbonate platform and intrashelf basin facies, similar to the Fennsik /Plateau/ unit, only they are less metamorphosed. Their tectonic posi- tion, however, is not clear. Recently the presence of shallow water Jurassic carbonates /oolitic limestones/ has been proven /Berczi-Makk and Pleikan, 1984/; they probably represent large olistothrymmata that slid into the contemporaneous deep basin from the adjacent region. Jurassic rocks in the central domain are known in theB6dva nappe of the Rudabanya Mts. and in the Meliati- cum lin the Meliata type section in Slovakia/. Bio- 106 stratigraphical evidence for their Jurassic age is also given by H. Kozur's radiolarian investigations /Kozur, 1984a/. They also belong to the "schistes lustres" facies realm, because dark shales are predomi- nant. As a difference against the Southern BUkk unit, marls are also common in the Bodva Jurassic, while radio- larian cherts are subordinate. Quite different are the associated igneous rocks, which are represented here by alkaline rhyolites /det. Cs. Szabo/. Their plate tectonic setting is not yet clear. In connection with this, vol- canic olistostromes are also common here, containing Triassic limestones of basinal facies and alkaline rhyolites. They represent typical seismites generated bysubaqueous volcanic activity. The former northern shelf margin domain /the Silice nappe/ has a North Alpine-Type /Juvavic/ Jurassic, simi- larly to its Triassic. It is exposed only on the Slovakian territory and is composed of spotty marls some red no- dular and crinoidal limestones in the Liassic and Dogger, overlain by radiolarites /Bystricky, 1973, p. 57-59/. Abundant redepositional features already make some tran- sition to the Bodva Jurassic. Younger formations of the Alpine geosynclinal stage are unknown in continuous sequences in North Hungary /nor in the South Slovak territory to the S of the Lubenlk- -Margecany line/. Occurrence of pebbles of Tithonian algal limestones of carbonate platform facies in Senonian conglomerates along the southern margin of Uppony Mts. /Mislk and Sykora, 1980/, among Triassic pebbles derived from the Bodva nappe /Brezznyanszky and Haas, 1984 and Kovacs, unpubl./ indicate a drastic change from deep water into shallow water sedimentation. Such pebbles are also known from younger conglomerates in the Slovakian territory /Mislk and Sykora, 1980/. The sudden appearance of this Late Jurassic shallow water carbonate sedimen- tation is known in the Vardar zone, in the Transylvanides and in the middle sector of the Northern LimestoneAlps and is related to the closure of the Triassic-Jurassic ocean /Kazmer and Kovacs, this volume/. In the Slova- kian territory it is also indicated by frequent occur- rences of blueschist metamorphic rocks, concentrated especially along the Riznava line /Hovorka et al., 1984/. 4. REFERENCES Arkai, P./1983/: 'Very low- and low-grade Alpine regional metamorphism of the Paleozoic and Mesozoic formations of the BUkkium, NE-Hungary'. Acta Geor. Hung. ~g, 1, 83-101, Budapest 107 Balla, Z., Hovorka, D., Kuzmin, M., Vinogradov, V./1983/: 'Mesozoic ophiolites of the BUkk Mountains /North Hungary/'. Ofioliti §, 1, 5-46, Bologna Balogh, K./1964/:'Die geologische Bildungen des BUkk- -Gebirges'. Ann. Inst.:Geol. Hung., 1§, 2, 245-719 Berczi-Makk, A., Pelikan, P./1984/: 'Jurassic formations from the BUkk Mountains'. Ann. Rep. Hung. Geol. Inst. 1~§~, 137-166, Budapest Brezsnyanszky, K., Haas, J./1984/: 'The Nekezseny Conglo- merate Formation of Senonian age: a sedimentological and tectonic study of the stratotype section'. Foldt. Kozl. 111, 1, 81-100, Budapest ----- Brezsnyanszky~ K., Haas, J./1985/: 'The nes tectonic map of Hungary'. Proc. rep. 13th Congr. KBGA, I, 174-177 Bystricky, J./1973/: 'Triassic of the West Carpathians Mts.' Guide tOLBxcursion C, lOth Congr. CBGA, 137 p. De Wever, P./in press/: 'Triassic radiolarians from the Darno area'. Acta Geol. Hung. ~1,3-4, Budapest FUlop, J./in prep./: 'Geology of Hungary'. I. Paleozoic. Grill, J. et al./1984/:'Az Aggtelek-Rudabanyai-hegyseg foldtani felepitese es fejlodestortenete'. /Geologi- cal constitution ind history of evolution of the Agg- telek-Rudabanya Range./ Foldt. Kut. ~1, 4, 49-56 Grill, J. et al./in prep./: 'Geological monograph of the Aggtelek-Rudabanya Mts.' Grill, J., Less, Gy., Szentpetery. 1./1985/: 'Geological and tectonic setting of the Aggtelek-Rudabanya Mts.' /Hungarian part of the Southern Gemerides/. Proc. rep. 13th Congr. KBGA I, 186-187, Cracow ---- Hovorka, D., Jaros., J., Kratochvil, M., Mock, R./1984/: 'The Mesozoic ophiolites of the Western Carpathians'. Krystalinikum 11, 143-157, Prague Isler, A., Pantie, N:- /1980/: ' "Schistes-lustres" Abla- geringen der Tethys'. Eclogae geol. helve Z~, 3, 799-822, Basel Kazmer, M., Kovacs, S. /This volume/: Triassic and Juras- sic oceanic/paraoceanic belts in the Carpathian-Pan- nonian region and its surroundings'. Kovacs, S./1982/: 'Problems of the "Pannonian Median Mas- sif" and the plate tectonic concept. Contributions based on the distribution of Late Plaeozoic - Early Mesozoic isopic zones': Geol. Rundschau 11, 2, 617-640, Stuttgart -- Kovacs, S./1983/: 'The "Tisia" problem and the plate tec- tonic concept'. Proc. 12th Congr. CBGA, An. Inst. )Qeol. Geofiz. ~~, 75-83, Bucuresti Kovacs, S., pero, Cs:- /1983a/: 'Tectonic front of a Dinaric-type Paleozoic in North Hungary'. An. Inst. Geol. Geofiz. ~~, 85-94, Bucuresti Kovacs, S., P~ro, Cs:-/1983b/: 'Report on stratigraphical investigation in the BUkkium /northern Hungary/'. 108 In: Sassi, F., Szederkenyi, T./eds./: IGCP No.5, Newsletter, 2, 58-65, PadovaaBudapest Kovacs, s., Veto-Akos, E./1983/: 'On the age and petro- graphy of basic volcanic rocks in the Uppony Mts., NE Hungary'. Ann. Rep. Hung. Geol. Inst. 12~1, 177-199, Budapest ---- Kovacs, s., Kozur, H., Mock, R./1983/: 'Relations between the Szendro-Uppony and BUkk Paleozoic in the light of the latest micropaleontological investigations'. Ann. Rep. Hung. Geol. Inst. 12~1, 155-175, Budapest Kovacs, s. lin press a/: 'North Hungarian facies types: A review'. Acta Geol. Hung. ~1, 3/4, 251-264 Budapest -- Kovacs, s. lin press b/: 'Olistostromatic formations in Northeastern Hungary'. Veroff. Zentralinst. Physik der Erde, Potsdam Kovacs, S./in press c/: ' Devonian olistostrome with volcanic matrix and limestone olistoliths from Strazsa hill, Uppony Mts., NE HUngary' N. Jb. Geol. Palaeont. Mh. , Stuttgart • Kozur, H./1984a/: 'New radiolarian taxa from the Triassic and Jurassic'. Geol. Paleont. Mitt. Innsbruck 11, 2, 49-88, Innsbruck Kozur~-H./1984b/: 'Stratigraphic evaluation of Upper Carboniferous and \ Errr.ian conodonts, holothurian sclerites and ostracodes'. Acta Geol. Hung. ~1, 1, Budapest -- Kozur, H., Mock, R./1979/: 'Zur Frage der varistischen Orogenese und des Alters der Faltung und Metamorphose im innerwest-karpatischen Raum'. Geologicky zbornik 1Q, 1, 93-97, Bratislava Mis{k~-M., Sykora, M./1980/: 'Jura der Silica-Einheit rekonstruiert von Gerollmaterial uns SUsswasser- kalke'der oberen Kreide im Gemerikum'. Geologicky zbornik 11, 3, 239-261, Bratislava Oberhauser, R~7ed./ /1980/: Der geologische Aufbau 6sterreichs. Springer, Wien, 702 p. Pelikan, P./in prep./: 'Structural outline of the BUkk Mts.' Alt. Foldtani Szemle, Budapest Reti, Zs./in press/: 'Triassic ophiolite fragments in an evaporitemelange, Norhtern Hungary'. Ofioliti, Bologna TECTONIC UNITS AND SlITURES IN THE PONTIDES, NORTHERN TURKEY A. I. OKAY i.T.V. Maden Fakliltesi Jeoloji BCillimli Te~vikiye, istanbul Turkey ABSTRACT. The Pontides is made up of three major tectonic units juxtaposed in Mid- to Late Mesozoic times. The Istranca Massif in the west consists of sandstone, quartzite, shale, limestone and Late Permian granitoid deformed and metamorphosed during the, Late Jurassic. Its contact with the Istanbul Zone further east is covered by the Eocene sediments. The Istanbul Zone is characterised by a well developed, unmetamorphosed and little deformed continuous Paleozoic sedimentary succession extending from Ordovician to the Carboniferous overlain with a major unconformity by latest Permian to lowermost Triassic continental red beds. The Intra-Pontide Suture of Late Triassic-Early Jurassic age separates Istanbul and Sakarya Zones. In marked contrast to the Istanbul Zone, the Sakarya Zone does not have a Paleozoic basement; Karakaya Complex of Triassic age made up of strongly deformed and metamorphosed basic volcanic rocks, limestones and greywackes with limestone olistoliths forms the basement to the undeformed post-Triassic sediments of the Sakarya Zone. The Karakaya Complex probably represents a Triassic magmatic arc/forearc/trench complex and may be part of a Cimmerian "Continent". The Sakarya Zone is separated from the Anatolide-Tauride uni ts by the izmir-Ankara-Erzincan Suture. The izmir-Ankara-Erzincan Ocean must have been in existence during the Triassic as no equivalent of the Karakaya Complex has been found south of the suture. Moreover the Tauride nappes include abundant Triassic pelagic sediments and volcanic rocks representing continental margin deposits of Triassic age. I. INTRODUCTION Geologically Turkey consists of a number of microcontinents or blocks separated by suture zones of various ages. The aim of this paper is to evaluate the evidence for the number, extent and age of some of these blocks and suture zones and suggest a revised tectonic schema. 109 A. M. C. !jengor (ed.), Tectonic Evolution of the Tethyan Region, 109-116. © 1989 by Kluwer Academic Publishers. 110 2. MAJOR FEATURES OF THE BLOCKS AND SUTURES 2.1. Istranca Massif This large area of metamorphic rocks in Thrace (Figs.l and 2) forms part of the Klrklareli Nappe of $engor and others(l). Istranca Massif consists of greywackre, shale, limestone, quartzite and Late Permian granitoid apparently deformed and metamorphosed in greenschist facies during the latest Jurassic. This metamorphic sequence is cut by mid-Cretaceous granitoids and unconformably overlain by Late Cretaceous volcanic rocks 2,3 . The internal structure of the Istranca Massif is very poorly known; especially it might include several tectonic units as yet unrecognised. Istranca Massif is separated from the Istanbul Zone by a small area of undeformed Eocene sediments (Fig.2); the nature of the contact between these two tectonic units is not known(l). However, the lack of regional metamorphism in the Istanbul Zone indicates that they were juxtaposed in post-Late Jurassic times. "EOITERRANEAN Fig.l. Tectonic map of Turkey showing some of the major tectonic zones. Heavy lines indicate major sutures. Dotted line is the Karakaya Suture after $engor and Yllmaz(4). 2.2. Istanbul Zone The Istanbul Zone(S), equivalent to the Istanbul Nappe of $engor(6), is characterised by a well-developed Palaeozoic section extending without any major break form the Ordovician to the Carboniferous. The Palaeozoic section consists entirely of sedimentary rocks representing an Atlantic-type passive continental margin and rests unconformably on a Precambrian metamorphic basement. The Palaeozoic rocks are overlain with a major angular unconformity by the uppermost Permian - lowermost Triassic continental red beds which passes up to a well developed Alpine-type Triassic carbonate facies overlain unconformably by Late Cretaceous-Palaeocene limestones. The term Istanbul Zone is preferred III to that of the Istanbul Nappe as there is no convincing evidence for the complete allochthoneity of the Istanbul Zone. The Istanbul Zone is very distinctive from the neighbouring tectonic units in its stratigraphy, absence of metamorphism and lack of major deformation. It is separated from the Sakarya Zone by the Intra-Pontide Suture. Fig.2. Geotectonic map of the Pontides. 2.3. Intra-Pontide Suture mAllotohd.-TIIIU'.d,Untt, • 'ph"", .... ~GrQn'to'd T N DN'D;,n, I §engor and Yllmaz(4) call the Intra-Pontide Suture as the suture of the northern branch of Neo-Tethys which separated their Rhodope-Pontide Fragment from the Sakarya Continent during the Mesozoic. Here, the Intra-Pontide Suture is used to denote the suture separating the Istanbul and Sakarya zones (Fig.l) or more correctly the Palaeozoic of the Istanbul Zone and the Karakaya Complex of the Salcarya Zone. In stark contrast to the Istanbul Zone, no in situ pre-Triassic rocks are reported in the Sakarya Zone; in fact the Sakarya Zone may not have had a continental basement during the Palaeozoic (see below). Furthermore, no equivalents of the Triassic Karakaya Complex of the Sakarya Zone occur in the Istanbul Zone. All these suggest that Sakarya and Istanbul zones had quite independent and different pre-Jurassic histories. An upper age limit to the juxtaposition of these two zones across the Intra-Pontide Suture is given by the Middle to Upper Jurassic clastics and limestones which lie unconformably over both the Karakaya Complex of the Sakarya Zone and the Istanbul Palaeozoic in the region south of Sinop(l). The Intra-Pontide Suture is the best candidate for the Palaeo-Tethyan suture in Turkey. 112 The western part of the Intra-Pontide Suture might have been reactivated during the Late Cretaceous. In the region of Dokurcun (Fig.2) Palaeozoic rocks of the Istanbul Zone lie tectonically over an ophiolite and ophiolitic mHange of probable Late Cretaceous age(7). Pelagic limestones, serpentinite and blueschist of probable Late Cretaceous age also occur as fault slivers in southern Thrace 8 . However, the striking similarity in the Mesozoic stratigraphy of the Sakarya Continent of !?engor and Yllmaz( 4) and the eastern Pontides indicates that the Cretaceous-Intra-Pontide Ocean did not join into the major Izmir-Ankara-Erzincan Ocean( 4) but rather was a narrow, blind-ended Gulf of California type ocean. 2.4. Sakarya Zone The Sakarya Continent of !?engor and Yllmaz(4) and the eastern Ponti des have a very similar stratigraphy and tectonic development which contrasts with the neighbouring tectonic units. These two areas are therefore considered as a single tectonic unit and named the Sakarya Zone(5,9). The Sakarya Zone is characterised by a variably metamorphosed and strongly deformed Triassic basement called the Karakaya Complex overlain with a major unconformity by Liassic conglomerates and sandstones which passes up to Middle Jurassic-Lower Cretaceous limestones and Upper Cretaceous flysch. The internal structure of the Karakaya Complex is poorly known. However, it is apparently made up of several tectonic units including a thick volcanic section with abundant basic pyroclastics and tuffs intercalated with carbonates, and a greywacke section with Permian and Carboniferous limestone olisto1iths(10 ,11). The metamorphism where it occurs is generally in high-pressure greenschist facise, and sodic amphibole occurs frequently in basic volcanic rocks 5,12 . The rocks are generally steeply dipping and are strongly deformed with isoclinal folds with subvertica1 axial planes. The deformation is locally semi-brittle glvlng a broken formation character to the Karakaya Complex. The Karakaya Complex is also intruded by several pre-Liassic granitoids. !?engor and Yl1maz(4) terminate the Sakarya Continent north of Ankara. However, metamorphic rocks similar to the Karakaya Complex and overlain by Liassic sandstones and conglomerates occur in the Tokat and Agvanis Massifs(1,13,14). The rest of the Mesozoic stratigraphy of the inner eastern Ponti des is also quite similar to that of the Sakarya Continent with the ubiquitous Middle Jurassic Lower Cretaceous limestones and Upper Cretaceous flysch 15 . There is a gradual passage to a Cretaceous-Eocene magmatic arc towards the outer Eastern Pontides presumably constructed on a Karakaya type basement. 113 A deformed and variably metamorphosed dark argillite-sandstone sequence locally with abundant limestone, basic volcanic rock and serpentinite olistoliths outcrops over large areas in the Kargl Massif southwest of Sinop (Fig.2, ~engor and others(l). This Akgol Formation, which forms part of the Ktire Nappe of ~engor and others(l), is apparently of mid-Triassic to Early Jurassic age and is unconformably overlain by undeformed Middle Jurassic conglomerates and sandstones passing upwards to Late Jurassic-Early Cretaceous limestones. Here, Akgol Formation and the Ktire Nappe are regarded as part of the Karakaya Complex, as they are lithologically and structurall1 very similar to it and seem to merge to the Karakaya Complex in the Tokat Massif (Fig.2). Only the deformation and metamorphism may have lasted slightly longer in the Akgol Formation. Karakaya Complex probably represents Triassic magmatic arc-trench deposi ts . The argillite-sandstone sequence with limestone olistoli ths may represent a subduction complex(16) whereas the basic volcanic-pyroclastic-limestone sequence may be magmatic arc-forearc deposits. The absence of in situ pre-Triassic rocks in the Sakarya Zone and the dominance of basic volcanism in the Karakaya Complex indicate that the Karakaya Magmatic Arc was constructed on an oceanic substratum. The deformation and metamorphism of the Karakaya Complex may be related to the collision of the magmatic arc with a continent to the north during the Triassic. The base of the Karakaya Complex is exposed in the Uludag and Kazdag tectonic windows in western Anatolia where high-grade gneisses, marbles, amphibolites and metaperidotites occur tectonically beneath the low-grade metabasites of the Karakaya Complex(17). Micaschists and gneisses of the Gtimti1?hane region in the eastern Ponti des may also be part of this Uludag tectonic unit(18). The age of the Uludag tectonic unit, and the age of the thrusting are not know. However, these occurrences suggest that the Karakaya Complex is allochthonous and is tectonically underlain by the Uludag tectonic unit. ~engor and Yllmaz( 4) interpret the Karakaya Complex as a narrow oceanic marginal basin, which opened and closed during the Triassic, and thus show a Karakaya Suture on their maps (Fig.l). However, it is not possible to delineate such a suture in the field, rather as argued here the whole of the Karakaya Complex represents magmatic arc/trench deposits of Triassic age. The Triassic Karakaya Suture may only be exposed in the Uludag and Kazdag tectonic windows assuming that the thrusting of the Karakaya Complex over the Uludag tectonic unit is Triassic. Rhodope-Pontide Fragment of ~engor and Yllmaz(4) includes in Turkey the Istranca Massif, the istanbul Zone and the eastern Pontides. However the Cretaceous Intra-Pontide Ocean most probably did not extend into the izmir-Ankara-Erzincan Ocean so that the Rhodope-Pontide Fragment has never been a discrete palaeotectonic entity during the geological history. 114 2.5. Izmir-Ankara-Erzincan Suture This suture separating the Sakarya Zone from the Anatolide/Taurides is generally accepted as being the major Tethyan suture in Turkey. Most ophioli tes in the Anatolide/Taurides are believed to have originated from the Izmir-Ankara-Erzincan (IAE) Suture. However, the age of the opening and closing of the ocean along the IAE Suture is not well-established and is quite controversial. Based on sedimentological evidence GorUr and others(19) argue for a Liassic opening of the IAE Ocean. However, the IAE Suture abruptly truncates the metamorphic Karakaya Complex along the whole length of the suture (e. g. in the Eski.;;ehir region, in the Toka t and Agvanis Massifs, Fig. 2), and no Karakaya Complex equivalentsare known south of the suture. For the case of Liassic opening of the IAE Ocean, immense lateral movement across the IAE Suture is required to account for the absence of the Karakaya equivalents south of the suture. Furthermore upper level nappes in the Taurides(20) include abundant Triassic pelagic sediments and basic volcanics suggesting that a continental margin was in existence during the Triassic to the north of the Anatolide/Tauride Platform. Reently discovered Triassic deformation in the Taurides(2l) is cited as evidence that the Karakaya Orogeny has also affected the Anatolide/Tauride Platform 1 . However, such Triassic deformation is generally restricted to the southern and external parts of the Taurides and is probably related to the rifting in the Antalya Unit. There is no evidence of Triassic deformation in the Menderes Massif whose metasedimentary cover records continuous sedimentation during the Paleozoic and Mesozoic. A conformable transition from the Triassic clastics to the Liassic dolomites is also observed in the Lycian Nappes east of the Menderes Massif(22). The question of the age of closure of the IAE Ocean is similarly controversial and unresolved. $engor and Yllmaz( 4) argue for a Late Paleocene - Early Eocene collision of the Anatolide/Tauride Platform and the Pontides. However, based on blueschist evidence Okay(9) prefers a Late Cretaceous age for the final collision. The collision may well have been diachronous starting at Late Cretaceous in northwest Turkey and progressing eastward to the eastern Sakarya Zone where island arc volcanism continues into the Eocene. 3. DISCUSSION As this review illustrates there are still many unresolved major tectonic problems in Turkey such as the timing of the inital rifting and final juxtapositioning of the different terranes, or the extent of the various tectonic zones. These problems will only be resolved by careful detailed fieldwork aimed to answer particular questions. For example, rift-related or continental margin type rock assemblages, which would indicate the timing of the initial opening of an ocean, have not been 115 described along the Intra-Pontide or izmir-Ankara-Erzincan-sutures*. This is partly due to metamorphic overprint but largely due to the lack of recognition of such assemblages. Similarly there is not even a single direct age of the numerous ophiolite bodies thought to have originated from the izmir-Ankara-Erzincan Ocean. Triassic orogenic events, although very important in the Pontides, are very poar1y documented. Very little is known on the internal structure, age of metamorphism, geochemistry of the volcanic rocks and sandstone composition of the Karakaya Complex. With such further data, ideas on the tectonics of Turkey will doub1tess undergo major modifications. REFERENCES 1- Sengor,A.M.C., Yl1maz,Y. and Sungur1u,0. 1984. 'Tectonics of the Mediterranean Cimmerides: nature and evolution of the western termination of Paaeo-Tethys'. The Geological Evolution of the Eastern Mediterranean, ed. J.E.Dixon and A.H.F.Robertson, 117-152. 2- Aydln, Y., 1982. 'Yl1dlZ Daglan (Istranca) Masifi' nin Jeo1ojisi', Thesis. iTU. Maden Fakti1tesi, Istanbul, 107 pp. 3- Aykol,A., 1979. 'Klrk1are1i-Demirkoy Sokulumu'nun petro1oji ve jeo- kimyasl'. Thesis, iTU. Maden Fakti1tesi, 180 pp. 4- Sengor,A.M.C. and Yl1maz,Y. 1981. 'Tethyan evolution of Turkey: a plate tectonic approach'. Tectonophysics, 75, 181-241. 5- Okay,A.I. 1986a. 'High-pressure/1ow-temperature metamorphic rocks of Turkey'. Blueschists and Eclogites, Geo1. Soc. America, Memor 164, Ed. B.W.Evans and E.H.Brown, 333-348. 6- Sengor,A.M.C. 1984. 'The Cimmeride orogenic system and the tecto- nics of Eurasia'. Geo1. Soc. America, Spec. Paper 195, 77 pp. 7- Yl1maz,Y., Goztibo1,A.M. and Ttiystiz,O. 1982. 'Geology of an area in and around the Northern Anatolian Transform Fault Zone Between Bo1u and Akyazl'. Multidisciplinary Approach to Earthquake Prediction, ed. A.M.I§lkara and A.Voge1, 45-65. 8- $enttirk,K. and Okay,A.I., 1984. 'Blueschists discovered east of Saros Bay in Thrace'. Bull. Mineral Research and Exploration Inst. of Turkey, 97/98, 72-75. 9- Okay,A.I. 1984a. 'Distribution and characteristics of the Northwest Turkish blueschists'. The Geological Evolution of the Eastern Mediterranean. ed. J. E. Dixon and A. H. f • Ro bertson, Geo1 . Soc. London * Editors's note: Rift-related and Atlantic-type continental margin rocks have been described from the Ankara-Erzincan suture zone. See ref.19 and the references cited therein. 116 Spec.Pub., 455-466. 10- Bingo1,E., Akyiirek,B. and Korkmazer,B. 1975. 'Geology of the Biga Peninsula and some characteristics of the Karakaya Blocky Series'. Proc. Congo Earth Sci. for the 50th Anniv. Turkish Republic, 71-77. 11- Gautier,Y. 1984. 'Deformations et metamorphismes associes a 1a fermeture Tethysienne en Anato1ie Centrale (region de Sivrihisar, Turquie), • Thesis, Universite de Paris-Sud, Centre d'Orsay, 235 pp. 12- Yl1maz,Y. 1979. 'Sogiit-Bi1ecik bo1gesinde po1imetamorfizma ve bun 1a- nn jeotektonik an1aml'. Tiirkiye Jeo1oji Kurumu Bii1teni, 22, 85- 100. 13- Ozcan,A., Erkan,A., Keskin,A., Ora1,A., Ozer,S., Siimengen,M., and Teke1i,0., 1980. 'Kuzey Anado1u FaYl - Klr~ehir Masifi araSlnln te- mel jeo1oisi'. M.T.A. Report no. 6722 (unpublished). 14- Okay,A.I., 1984b. 'The geology of the Agnavis metamorphic rocks and neighbouring formations'. Bull. Mineral Research and Exploration Inst. of Turkey, 99/100, 16-36. 15- Pe1in,S. 1977. 'A1ucra (Giresun) Giineydogu yoreslnln petrol olanak- 1arl baklmlndan jeo1ojik ince1enmesi'. Thesis, Karadeniz Teknik Uni- versitesi, 103 pp. 16- Teke1i ,0., 1981. 'Subduction complex of pre-jurassic age, northern Anato1ia, Turkey'. Geology, 9, 68-72. 17- Ketin,I., 1985. 'Tiirkiye'nin bindirme1i-nap11 yaplslnda yeni ge1i~­ me1er ve bir ornek: U1udag Masifi'. Ketin Symposium, Tiirk.Jeo1.Kur., Ankara 19-36. 18- Yl1maz,Y., 1972. 'Petrology and structure of the Giimii~hane granite and surrounding rocks, north-eastern Anato1ia'. Ph.D. Thesis, University of London, 260 pp. 19- Goriir,N., Sengor,A.M.C., Akkok,R. and Yl1maz,Y. 1983. 'Pontid1er'de Neotetis'in kuzey ko1unun aC;:llmaslna i1i~kin sedimento1ojik veri- 1er'. Tiirkiye Jeo1oji Kurumu Bii1teni, 26,11-20. 20- Ozgii1,N., 1984. 'Stratigraphy and tectonic evolution of the Central Taurides'. The Geology of the Taurus Belt, ed. O.Tekeli and M.C. Gonciiog1u, 77-90. 21- Monod,O. and Akay,E., 1984. 'Evidence for a Late Triassic - Early Jurassic orogenic event in the Taurides'. The Geological Evolution of the Eastern Mediterranean, ed. J. E.Dixon and A.H.F .Robertson, Geo1.Soc.London.Spec.Pub., 113-122. 22- Okay,A.I. 1986b. 'Deniz1i-Tavas araslndaki bo1genin jeo1ojisi'. Turkish Petroleum Company Report (TPAO), (unpublished). AN EXAMPLE FOR THE TECTONIC EVOLUTION OF THE ARABIAN PLATFORM MARGIN (SE ANATOLIA) DURING MESOZOIC AND SOME CRITICISMS OF THE PREVIOUSLY SUGGESTED MODELS Demir Altlner Middle East Technical University, Department of Geological Engineering Ankara, Turkey ABSTRACT. The stratigraphical and environmental analysis of the Mesozoic platform margin sequences, found as allochthonous units to north and northwest of Hazro region, led to the recognition of two rifting phases in Southeast Anatolia. Middle to Late Triassic rifting, characterized by the sudden development of pelagic deposits including volcanism, abruptly ceased in Late Triassic time. The second rifting phase took place in post-Triassic time and preceded an ocean-opening to the north of the Arabian platform. The study of the platform, platform margin and oceanic sequences lying to the north and south of the Hazro region suggests that the Middle to Late Triassic rifting is an abortive phase of the post-Triassic rifting events. Thus, the data obtained reject the previous models proposing the presence of a Triassic and Liassic ocean floor to the north of the Arabian platform in Southeast Anatolia. INTRODUCTION During the Mesozoic, the Southeast Anatolia represented the northernmost extension of the Arabian platform. With nappes and associated clastic deposits emplaced onto its northern margin it forms a precious laboratory for the study of the tectonic evolution of a platform margin. However, the evolution of the margin of this platform has not yet been clearly understood, because most of the geological research related with petroleum exploration has been concentrated on the autochthonous platform sediments represented by the folded belt and the foreland (Rigo de Righti and Cortesini, 1964). In recent years, the stratigraphy and evolution of the platform margin sequences and related geological events were described by Sungurlu (1974), Perinc;:ek (1979 a,b, 1980, 1981), Perinc;:ek and Ozkaya (1981), Ozkaya (1982 a,b), ~engor and Yllmaz (1981), De1aune-Mayere and others (1983), Fontaine (1981), Michard and others (1984) and Yazgan (1984). Most of these authors discussed the evolution of this margin beginning in the Triassic in terms of classical rifting models and tried to fit their data into the model of the opening of the Neo-Tethyan ocean to the north of the Arabian platform. Particularly ;>engor and Yllmaz 117 A. M. C. $,engor (ed.), Tectonic Evolution o/the Tethyan Region, 117-129. © 1989 by Kluwer Academic Publishers. 118 (1981), ;>engor and others (1984), ;>engor (1984) placed the southern branch of the Neo-Tethyan ocean to the north of this platform. According to these authors, immediately after the Mid-Triassic rifting events, the southern branch of the Neo-Tethyan ocean opened in this area in Late Triassic time and its seafloor began to spread actively during Liassic time. In this study, I will present some geological facts against the models proposed by ;>engor and Yllmaz (1981) by interpreting the data obtained from my studies in a trend extending from the allochthonous sequences in the north and northwest of Hazro to Gercli§ area which is located in the area of the autochthonous sequences of the Arabian platform (Fig. 1). Since I disregard the hypothesis of the Bitlis massif as the continuation of the Arabian platform during the Mesozoic, I do not take into account the models proposed by Michard and others (1984, model 1) and Yazgan (1984) in this study and leave them for a further discussion on the evolution of the Arabian platform. PLATFORM, PLATFORM MARGIN AND OCEANIC SEQUENCES IN THE STUDY AREA Mesozoic platform sequence (Fig. 2). Overlying the fossoliferous upper carbonates of the Gomaniibrik Formation of Djulfian to Dorashamian age paraconformably, the Lower Triassic rocks of the study area are represented by the Uludere Shale in the Hazro region. This formation possibly interfingers with the sandstones and sandy limestones of the Beduh Formation towards the south. The Hazro region is characterized by a non-depositional (?) period from Middle Triassic to Early Cretaceous. However, the sequence south of Hazro, down to the Gercli. area or even farther south towards the Syrian border records a Middle Triassic transgression the deposits of which disconformably overlies the Beduh Formation following a break (?) in sedimentation in the Anisian. The Klireci Formation containing the involutines of Ladino-Carnian affinity evolved rapidly in a pelagic environment (Firmeli Formation) and is dominated mainly by a marl and argillaceous limestone sequence containing thin shelled pelagic pelecypods (Daonella or Halobia). During the latest Triassic (Late Norian ?-Rhaetian), the Arabian platform started to record evaporites and dolomites laid down under supra- to intertidal environmental conditions. The sequence recording this drastic change in sedimentation was later truncated by erosion. No Jurassic and Neocomian sediment seems to be laid down in the limits of the study area. During Late Early Cretaceous, the Arabian platform recorded the deposition of a tremendous transgression (Vail et al., 1977). Because of the rapid subsidence of the platform, the formations of the Mardin Group were successively deposited in the whole .. .. .. .. .. .. .. .. .. .. " - . " An ka ra \ \ s tu dy ar ea -- - - '\'l " , _ , ) . . . . . / . . . . . . . . _ . , o J / tv 'e di te rr an ea n- Se a c M al at ya ' " " - . SY RI A F ig .I : L oc at io n m ap . " US SR . . . . . . . . . _ . _ . , + \ " (~ -\ \ \ N ( . ~ . " '- . I [l ~ ~ D iya rb ak lf . . . . . . . " , / . . . . . . . ' r " ~ . " , . . J . . / .. .. .. .. . IJ . . /" I M~d !D. _._ . ..- / IR AQ /' I - - - . . . . So ut he rn li m it o f th e n a pp es e m pl ac ed in C en oz oi c 1\ 1\ So ut he rn l im it o f th e n a pp es e m pl ac ed in C re ta ce ou s \ \ \ IR AN ~ I ... . . . ( '" ~ , \.. "~ . " - i o 50 km ~ " \0 PL AT FO RM PL AT FO RM MA RG IN OC EA NIC AS SE MB LA GE H ez an Un it KO i;a ll U ni t S N IG ER CU ;; RA MA N HA ZR O f< Dn ak Fo rm at io n AL LO CH TH ON OU S- UN IT S T N W of H az ro ) I- 121 Fig.2: Generalized stratigraphy of the platform, platform margin and oceanic sequences in the study area and their tectonic interpretations. 1. Limestone, 2. High energy limestone, 3. Carbonate turbidite, 4. Dolomi te, 5. Dolomitic limestone, 6. Carbonate breccia with dolomitic cement, 7. Anhydrite, 8. Argillaceous limestone, 9. Sandy limestone, 10. Marl or shale, 11. Paralic shale or marl, 12. Sandstone, 13. Pebbly clastics, 14. Radiolarite, 15. Basaltic flows, 16. Tuffaceous volcanism, 17. Stromatolite or algal mat, 18. Loferite, 19. Pelagic pelecypod, 20. Megalodont, 21. Ammonite, 22. Planktonic foraminifera. Lithostratigraphic units. Platform: U: Uludere Shale, A: Arlllk Formation, Kli: Klireci Formation, Gi: Girmeli Formation, M: Mardin Group, K: Karabogaz Formation, S: SaYlndere Formation, Ks: Kastel Formation, B: Beloka Formation, m: Maymune Member, b: Bada Member, d: Dirik Member, An: Antak Formation, Klr: Klradag Formation, Ga: Garzan Formation, Ge: Germav Formation, Si: Sinan Formation, Platform Margin: Ki: Kilisedag Formation, Hk: Hezankilise Formation, He: Hezan Formation, Ne: Nenyas Formation, V: Verrarecelicoke Formation, K: Kara Marl, Ha: Hacl Formation, Ku: Kuran Formation, Li: Licek Formation, Ka: Karadut Formation. 122 of Southeast Anatolia including the study area. The Cenomanian and earliest (?) Turonian were times of maximum transgression as it was already noted for the Middle East area (Reyment and Bengtson, 1985). The evolution of this transgressive cycle was suddenly interrupted by a ra pid uplift in the region in Turonian time recording the regressive deposits preserved in local areas that escaped from the post-Mardin erosion. After a gap in the sedimentation during the Coniacian and Santonian, the platform was drastically covered by a new transgression in the Campanian. The Kastel basin subsided along the northern margin of the Arabian platform and the Kastel Formation dominated by clastic rocks overlay the pelagic deposits of the Karabogaz and SaYlndere Formations. Towards the south, the Kastel Formation passes laterally into the Beloka Formation. The latter was possibly the product of a transgression coming from the north, related to the subsidence of the Kastel basin. The emplacement of the allochthonous platform margin and the oceanic sequences onto the platform during the Late Campanian-Early Maastrichtian caused a rapid regression in the platform and the Beloka Formation was truncated by an uneven erosion surface after a general uplift in the area. Middle to Late Maastrichtian time was represented by a new transgression and deposition of the post-orogenic sediments. Red clastic rocks of the Antak Formation grade southwards into the paralic deposits of the Klradag Formation which was progressively covered by the transgressive carbonates of the Garzan and the fine clastic rocks of the Germav Formations. The Middle to Upper Maastrichtian deposits were strongly controlled by syn-sedimentary extensional tectonics and at the Mesozoic-Tertiary boundary the region to the south of Hazro was mainly under the pelagic influence of the Germav Formation which also overlapped the carbonates of the Garzan Formation. Mesozoic platform margin sequence (Fig.2). In the study area, the platform margin sequences found as tectonic slices or mass-wasting deposits in the north or northwest of the Hazro region (Hezan, Firdevs, Firdevs Dere, KevrapU 123 sometimes yielded coarse breccia laid down near tidal flat sequenes (Verraracelicoke Formation). Transgressive Kara Marl and the Hacl Formation characterized by the Orbitopsela bearing dolomitic limestones of shallow water origin began to overstep the Triassic sequences in Liassic time. Following this transgression the Dogger saw many important changes in the sedimentation. The lower part of the Kuran Formation characterized by oolitic, oncolitic and intraclastic grainstones and packstones of high energy environment was vertically followed by Bositra bearing pelagic limestones and finally the Rosso Ammonitico facies in MaIm. The red nodular limestones of the Rosso Ammonitico contains calpionellid forms of the Tithonian and the Earliest Berriasian in the upper parts and vertically change into calpionellid and Nannoconus-bearing white argillaceous lime stones and marls of the Licek Formation of Late Berriasian-Hauterivian age. Starting from the Barremian, the platform margin sequence began to receive an important quantity of carbonate detritus influx supplied by the turbidity currents carrying platform material. Orbitolines and prealveolines together with oolites, macrofossil fragments (rudists etc.) and clasts were transported from the platform and deposited together with the pelagic fauna (Globigerinacea) of the Karadut Formation during the Barremian-Cenomanian interval. Above the Cenomanian, the same type of sedimentation alternating with pelagic marls and shales continued from the Turonian to the earliest Campanian. The lower-most Campanian is the youngest level recorded in the platform margin sequence. Mesozoic oceanic assemblage (Fig. 2). The oceanic assemblage represented by the ultramafic rocks, gabbros, basic dykes, volcanic and sedimentary rocks (Koc;:ali Unit) is divided into three formations by Sungurlu (1974) and in Fig.2 only the sedimentary rocks interclated with volcanics are displayed (Konak Formation). The data of this part come from the study of Fontaine (1981) that contains several stratigraphic sections measured in the areas lying to the northwest of Hazro (Firdevs, Pirik, Duburi Tepe, Kerves sections, etc.). The oceanic sequence exhibits an alternation of marls, siliceous red clastic rocks, radiolarites, carbonate turbidites, argillaceous limestones including pelagic fauna and basaltic flows. For the age of this sequence, Fontaine (1981) reports an Hauterivian-Barremian interval. However this age should at least be extended up into the Cenomanian as it was established in the Adlyaman region by Fontaine (1981) and by my own studies. The age of the Konak Formation could also be extended down into the Tithonian and Berriasian by the calpionellid fossil findings (Perinc;:ek, 1979 b) and to the Late Jurassic by the presence of the "facies a filaments" reported in the Koc;:ali Unit by Rigo de Righi and Cortesini 124 (1964). Pinkish wacke stones including pelagic pelecypods (Bositra?) described by Fontaine (1981) from the exposures of the Konak Formation in the Adlyaman region might correspond to the same level reported by Rigo de Righi and Cortesini (1964). TECTONIC INTERPRETATION OF THE DATA The chronostratigraphic and lithostratigraphic studies of the platform, platform margin and oceanic sequences of the study area lead to the following tectonic interpretations: Triassic: Tectonically, the Early Triassic time witnessed an undifferentiated environment both in the platform and possibly also in the platform margin sequences (Fig.2) recording quiet, very shallow water sedimentation (Fig.3 a). During or after the sudden transgression of the carbonates of the Klireci and Kilisedag Formations onto the platform and platform margin sequences respectively, the Middle Triassic paleotopography was rapidly modified by an active block faulting episode affecting the platform and platform margin sequences (Fig.3 b). The presence of tuffaceous volcanism of Pietra verde type, rapid invasion of pelagic fauna (Daonella etc.) and rapid changes in the lithological sequences (Girmeli and Hezankilise Formations) indicate a rift-controlled sedimentation in the area. The Norian was the time of abrupt changes in the sedimentation. The area uplifted as the rifting ceased and the platform and platform margin sequences were covered by supra- to intertidal deposits (Fig.3 c). The rest of the Triassic sedimentation was characterized by regressive deposits. The end of the Triassic Period witnessed the complete emergence of the Triassic sequences of the platform which were subsequently deeply eroded from south to north, towards the Hazro area. During this time the platform margin sequences also emerged completely or partially. Jurassic: Starting from the Lias, the platform margin sequences became completely isolated from the platform that stayed as a highland above sea level during the Jurassic to Neocomian interval. The transgression of the Liassic units (Kara and Hacl Formations) of the platform margin was possibly controlled by the very early stages of a new rifting episode. During the Dogger and MaIm, net changes in the facies represented by oolitic, oncoli tic and intraclastic limestones show the formation of high energy zones controlled by fault-bounded sea floor topography (Fig.3 d). later, these areas submerged completely during the MaIm and the Bositra - bearing pelagic limestones and the Rosso Ammonitico were laid down in the platform margin sequence (Fig.3 e). Also, during the Late Jurassic, the oceanic crust might have formed to the north of the platform margin since the oldest sediments corresponding to the Bositra - bearing pelagic rocks of the platform margin (upper part of the Kuran Formation) are recorded from the Konak Formation of the Ko~ali Unit. Cretaceous: During Late Berriasian-Hauterivian time, the pelagic Vl => o lJ.J W « I- lJ.J 0: w w Vl V) « 0: => w Vl Vl « 0: I- Upper Lower Malm Dogger Liassic Upper Mid dle Lower s _ Haz H 126 facies including few carbonate turbidite beds (Licek Formation, Fig.2) indicate the establishment of first slope environment in the platform margin sequence. Of the sediments deposited on the oceanic crust, the pelagic carbonates yield calpionellid fossils which would confirm the continuation of the sedimentation in the oceanic realm. The accumulation of thick carbonate turbidite deposits of the Karadut Formation starting from the Barremian indicates the formation of a continental slope facing an ocean already opened in the Malm to the north of the Arabian platform (Fig.3 f). During this time, the oceanic assemblage recording fine red clastics, radiolarites, basal tic flows also received carbonate turbidite deposition which spread on the oceanic crust as deep-sea fans. These deposits are the tongues of the Karadut Formation accumulating on the continental slope (Fig.2). During the Barremian to Cenomanian time interval the platform saw an important transgression recording the carbonates of the Mardin Group. As a consequence of the relative subsidence of the platform and the rise in the sea level a transgression coming from the north overlapped onto the areas that stayed a highland since the Triassic (Fig.2,3 f). No Turonian or younger age has been recorded from the oceanic assemblage sequence until now. This fact might have been either because of the lack of observation or the initial imbrication of oceanic lithosphere which would affect the sedimentation regime on the oceanic crust. This event possibly disturbed the continental slope sedimentation. Since the platform margin sequence was not involved in this compressional tectonics yet, no hiatus has been detected in the Karadut Formation owing to its deep environment of deposition that could not have been brought to the surface. However, the relatively shallow platform recorded this event to the south and the carbonates of the Mardin Group emerged rapidly in earliest Turonian probably beause of the elastic arching of the lithosphere as a result of tectonic loading to the north (Fig.2). The subsidence of the Kastel basin in the north of the platform in Campanian corresponds with the time when the sedimentation stopped in the platform margin as a result of emergence. Then the oceanic lithophere and platform margin sequences were carried as tectonic slices or mass-wasting deposits into the Kastel basin in Late Campanian-Early Maastrichtian (Fig.2,3 f) time. Following this, the entire platform, thrusted at its margin, uplifted again in Early Maastrichtian time before the Late Maastrichtian transgression and deposition of the post-emplacement units. CONCLUSION AND CRITICISM OF THE PREVIOUSLY SUGGESTED MODELS The data relevant to the evolution of the platform margin, platform and 127 oceanic sequences in the study area do not indicate the presence of any ocean floor to the north of the Arabian platform during Late Triassic and Liassic times. The Triassic rifting which controls the sedimentation on the platform and platform margin within the limits of the study area appears to be an abortive phase of the post-Liassic ocean opening. Thus, the models proposed by gengor and Yllmaz (1981), gengor and others (1984) and gengor (1984) illustrating an actively spreading branch of Neo-Tethys in the north of the Arabian platform during Triassic and Liassic are not compatible with the data and interpretations in this study. Although the age span of the sedimentary sequence of the oceanic assemblage seems to be based on fragmentary data, especially for the base of the sequence assigned to the MaIm, the overall interpretation including the evolution of the platform and platform margin sequences gives a coherent picture. This picture explains well the rapid rupture and submergence of the platform margin in Late Jurassic and the formation of the oceanic crust to the north of the platform in the MaIm. The post-Liassic evolution of the Arabian platform margin was also correctly explained and interpreted by Fontaine (1981) who studied around the Hazro region. However, he could not give any evidence on the abortive character of the Triassic rifting in Southeast Anatolia. The Triassic ocean opening seems to be confirmed in the Baer-Bassit area (Syria) by Delaune-Mayere and others (1983) and Delaune-Mayere (1984). However the recent discoveries by Garfunkel and Derin (1984) bring evidene on the aborted Late Triassic rifting in the Levant area (Israel). This event was followed by renewed rifting in the Jurassic preceeding an ocean opening in Late Jurassic. The distribution of these diachronous events are significant. It seems as if the margins of the Arabian plate underwent similar processes but at different times. For the case of the study area in Southeast Anatolia, as Monod and Akay (1984) conclude, the Cimmerian orogeny resulting from the deformation of the Paleo-Tethys and its "marginal basin" (Karakaya basin), might have left traces on the margin of the Arabian platform. This orogeny possibly started just after the rifting that supposedly split the Cimmerian continent (gengor and Yllmaz, 1981) from the Gondwana-Land in Middle to Late Triassic time. However, this rifting became unsuccessful or aborted while the "marginal basin" of the Paleo-Tethys was deformed in the northern Turkey and the traces of its deformation were felt further south in the Arabian plate in Late Triassic time. It is this inference that strongly contrasts with the Triassic splitting model of the Cimmerian continent from the Arabian plate by the opening of the southern branch of Neo-Tethys and the counterclockwise rotation of this newly-formed continent to close the Paleo-Tethys in the north. If there was any Cimmerian continent, it should be defined farther to the north and only after the Liassic rifting of the northern branch of Neo-Tethys (Gortir et a1., 1983). In this case, the Anatolide/Tauride platform separated from the Cimmerian continent in the north should have formed the northern continuation of the Arabian platform at least up into Lisasic time. Later it was rifted away from the Gondwanaland leading to the opening of the eastern Mediterranean basin in the MaIm. 128 REFERENCES Delaune-Mayere ,M. (1984). Evolution of a Mesozoic passive continental margin: Baer-Bassit (NW Syria): In: Dixon,J.E. and Robertson,A.H.F. (Edits), Geological Evolution of the Eastern Mediterranean, Geol. Soc. London Spec. Pub. 14, pp.15l-l59. Delaune-Mayere,M., Fontaine,J.-M., and Perinc;:ek,D. (1983). La de la plaque Arabo-Africaine au Mesozoique en Syrie et en du Sud-est: Une comparasion: Cah. O.R.S.T.O.M., Ser. vol.XIII, no.l, pp.3l-4l. bordure Turquie Geol. , Fontaine,J.-M. (1981). La plate-forme arabe et sa marge passive au Mesozoique: l'exemple d'Hazro (S-E de la Turquie): These 3e cycle, Paris XI, 270 p. Garf .. nkel,Z. and Derin, B. (1984). Permian-early Mesozoic tectonism and continental margin formation in Israel and its implications for the history of the Eastern Medi terranean: In: Dixon, J . E. and Robertson,A.H.F. (Edits), Geological Evolution of the Eastern Mediterranean, Geol.Soc. London Spec.Pub. 14, pp.187-20l. Gortir,N., ~engor,A.M.C., Yllmaz,Y. and Akkok,R. (1983). Pontidlerde Neo-Tetis'in Kuzey kolunun aC;:llmaslna ilio!?kin sedimentolojik veriler. Ttirkiye Jeoloji Kurumu Btilteni, Vol.26, pp.11-20. Michard,A., Whitechurch,H., Ricou,L.-E., Montigny,R. and Yazgan,E. (1984). Tauric subduction (Malatya-Elazlg provinces) and its bearing on tectonics of the Tethyan realm in Turkey: In: Dixon,J.E. and Robertson,A.H.F. (Edits.), Geological Evolution of the Eastern Mediterranean, Geol. Soc. London Spec. Pub. 14, pp.36l-373. Monod,O. and Akay,E. (1984). Evidence for a Late Triassic-Early Jurassic orogenic in the Taurides: In: Dixon,J.E., and Robertson,A.H.F. (Edi ts), Geological Evolution of the Eastern Mediterranean, Geol. Soc. London Spec. Pub. 14, pp.113-l22. Ozkaya,1. (1982 a). Origin and tectonic setting of some melange units in Turkey: Journal of Geology, vol.90, pp.269-278. Ozkaya,1. (1982 b). Upper Cretaceous plate rupture and development of leaky transcurrent fault ophiolites in Southeast Turkey: Tectonophysics, vol.88, pp.103-l06. Perinc;:ek,D. (1979 a). Geological investigation of the Qelikhan-Sincik-Koc;:ali area (Adlyaman province): Istanbul Univ. Fen. Fak. Mec. Serie B-44, pp.127-l47. Perinc;:ek,D. (1979 b). The geology of Hazro-Korudag-Qtingtio!?-Maden-Ergani-- Hazar-Elazlg-Malatya area: GEOCOME Guidebook, Excursion "B", pp.3-32. Perinc;:ek,D. (1980). Volcanics of Triassic age in Bitlis Metamorphic rocks: Bull. Geol. Soc. Turkey, vol.23, pp.20l-2ll. Perinc;:ek,D. (1981). Sedimentation on the Arabian shelf under the control of tectonic activity in Taurid belt: 5th Petroleum Geologists Congr. Proc., pp.77-93. Perinc;:ek,D. and Ozkaya,i. (1981). Arabistan levhasl kuzey kenarlnln tektonik evrimi: Yerbilimleri, Bull. Inst. Earth Sciences Hacettepe Univ., vol.8, pp.9l-l0l. 129 Reyment ,R.A. and Bengtson ,P. (1985). Mid-Cretaceous Events: report on results obtained 1974-1983 by IGCP Project No.58. Publications from the Paleontological Institution of the Univeristy of Uppsala. Spec. vo1.5, 132 p. Rigo de Righi,M. and Cortesini,A. (1964). Gravity tectonics in foothills structure belt of the Southeast Turkey: Amer. Assoc. Pet. Geol. Bull., vol.48, pp.19ll-l937. Sungurlu,O. (1974). Geology of the Northern Part of Petroleum District-VII: 2nd Petroleum Geologists Congr. Proc., pp.85-l07. $engor ,A .M.C. (1984). The Cirnrneride Orogenic System and the Tectonics of Eurasia. Geol. Soc. America Spec. Pap. 195, 82 p. $engor,A.M.C. and Yllmaz,Y. (1981). Tethyan evolution of Turkey: a plate tectonic approach: Tectonophysics, vol.75, pp.18l-24l. $engor,A.M.C., Yllmaz,Y. and Sungurlu,O. (1984). Tectonics of the Mediterranean Cirnrnerides: Nature and Evolution of the Western termination of Paleo-Tethys: In: Dixon,J.E., and Robertson,A.H.F. (Edits.), Geologic Evolution of the Eastern Mediterranean, Geol. Soc. London Spec. Pub. 14, pp.77-ll2. Vail,P.R., Mitchum,R.M.J., Jr and Thompson. (1977). Seismic stratigraphy and global changes of sea level. In: Payton,C.E. (edit.), Stratigraphic interpretation of seismic data. Mem. Am. Ass. Petrol. Geol., No.26, pp.83-97. Yazgan,E. (1984). Geodynamic evolution of the Eastern Taurus region: Intern.Symp. on the Geology of the Taurus Belt, p.199-208. TIMING OF OPENING OF THE BLACK SEA: SEDIMENTOLOGICAL EVIDENCE FROM THE RHODOPE-PONTIDE FRAGMENT Naci Gorlir i.T.D.Maden Fakliltesi Jeoloji Bollimli Te~vikiye 80394 Istanbul Turkey ABSTRACT. Sedimentologic evidence from the late Jurassic through the Cretaceous sequences of the Rhodope-Pontide fragment (Northern Turkey) indicate that the present Black Sea basin began opening as an originally ensialic 'back-arc' basin concurrently with the onset of north-dipping subduction of the floor of the northern branch of Neo-tethys. In late Jurassic-Neocomian time both the present Rhodope-Pontide fragment and the northern shores of the Black Sea were parts of a south-facing Atlantic-type continental margin (shelf). In Aptian to Cenomanian time interval the Rhodope-Pontide fragment began rifting from the southern U. S. S. R. coevally with the onset of moderate intensity subduction-related magmatism. In post-Cenomanian time rift-related normal fault ac tivity ceased, thermal subsidence along the northern, rifted margin of the Rhodope-Pontide fragment commenced, and subduction-related magmatism intensified. The BI ack Sea is not a Jurassic structure as maintained by some. It is a late Cretaceous back-arc basin that began opening almost in a pre-arc spreading mode as most present-day extensional arcs. INTRODUCTION One of the major problems of the Mediterranean Alpide chains is the age of the small ocean basins now caught up within the orogen. Of these, the Black Sea has been one of the most intractible, not only because its basement is not accessible to observation, but also because surroundings have been politically divided in a way that has made scientific communication difficult. Although much data have been published on the Soviet side, little was available on the Turkish part, where the most interesting data exist to illuminate the origin of the basin. The purpose of this contribution is to summarize the new observations from the Rhodope-Pontide fragment in northern Turkey that suggest strongly a middle to late Cretaceous rifting for the Black Sea. The synthesis of these data with those from the Soviet side are going to be published elsewhere (Gorlir, in press), to which I refer the reader for details. 131 A. M. C. ijengor (ed.), Tectonic Evolution of the TelhyanRegion, 131-136. © 1989 by Kluwer Academic Publishers. 132 CRETACEOUS GEOLOGY OF THE RHODOPE-PONTIDE FRAGMENT: IMPLICATIONS FOR THE OPENING OF THE BLACK SEA The Rhodope-Pontide fragment is the northernmost Alpide tectonic unit in Turkey and constitutes a part of Ketin's (1966) Pontides. To the north it is bounded by the Srednogorie zone in Bulgaria, the Black Sea, and the Riou depression in Transcaucasia (Fig.l) (;>engor and Yllmaz, 1981). To the south the Intra-Pontide suture and the Ilgaz-Erzincan suture plus the East Anatolian Accretionary complex now delimit it (;>engor and Yllmaz, 1981). Data summarized in Fig.l show that the present Rhodope-Pontide fragment was occupied during the early Cretaceous by an extensive, generally southerly-sloping shallow carbonate shelf constructed along the south-facing Atlantic-type continental margin of Eurasia north of the Neo-Tethys (Fig.la). Geological evidence shows that this situa tion had already been established during the late Jurassic. Two deep, probably oceanic, basins perforated this shelf on both the eastern and western sides of the present Black Sea. The one to the east, the Slate-Diabase zone (Khain, 1975), opened during the Lias and closed during the early Tertiary. No part of this basin survives in the present Black Sea basin. Whatever the Slate-Diabase zone may have been, it is clearly unrelated to the Black Sea proper. The other oceanic basin, the Nish-Trojan trough, was a part of the northern branch of the Neo-Tethys enclosing the Rhodope-Pontide fragment and was diachronously obliterated between the early Cretaceous and the Miocene from south to north (Sandulescu, 1980; Burchfiel, 1980). Although the Slate-Diabase zone trough and the Nish-Trojan trough may have been kinematically related during the late Jurassic-early Cretaceous there is no evidence to suggest that such a connection was achieved through a large pre-Black Sea depression as suggested by Zonenshain and Le Pichon (1986). Following the onset of subduction-related volcanism on the Rhodope-Pontide fragment in the Aptian-Albian (Aydln et al., 1982), disintegration of the carbonate platform of the south-facing Atlan tic-type continental margin 0 f Eurasia started in the Western Pontides and in the Moesian platform (Fig.lb). This block faulting event and subsidence accelerated in the Cenomanian (Fig.lc) and, in a number of places (e.g. the Rhodope-Pontide fragment, Pre-caucasus, Crimea, Moldavia, etc.), caused the formation of large depressions or basins on the subsided blocks with deposition of deeper water sediments (deep-water carbonates, carbonate and terrigenous turbidites), locally intercalated with volcanics. I interpret this Cenomanian disintegration as a rifting event that began tearing the Rhodope-Pontide fragment from the main Eurasian continental block along the juvenile volcanic axis of the south-facing Neo-Tethyan magmatic arc. I see this as the initial rifting and opening of the Black Sea, which underwent a uniform, probably thermally induced, subsidence later in the Senonian. Recent geophysical surveys (e.g. Bulandje, 1976; Sidorenko, 1978) show that the continental crust is block faulted beneath the continental slope and locally dis plays a strong relief beneath the sedimentary cover, 133 which provides support for the rifting hypothesis. These data when combined with regional geologic data from around the Black Sea basin, strongly support the marginal basin opening model above a north-dipping Neo-Tethyan subduction zone as suggested by Adamia et a1. (1974), Hsti et a1. (1977) and Letouzey et a1. (1977). Data presented in this paper tightly constrain the onset of back-arc rifting as Aptian to Cenomanian. From the Cenomanian onwards, normal faulting appeared to have ceased in large parts of the Black Sea margins and was replaced by a uniform subsidence probably as a result of lithospheric cooling (Fig.ld and e). Earlier basin opening and closing events around the Black Sea area were thus unrelated to its evolution, much as the oceanic basin obliterated by the Taiwanese collision is entirely unrelated to Okinawa trough. Da ta reviewed in this paper also seem to support Hsti et a1.' s (1977) opinion that the Senonian opening of the Srednogorie is related to that of the Black Sea and that the two opened as parts of one middle to late Cretaceous marginal basin. Such diachroneity along the strike characterizes many of the present-day marginal basins and is also known from the Okinawa trough (Lefouzey and Kimura, 1985). In conclusion, I wish to underline that the Black Sea basin began rifting during the Aptian-Cenomanian interval. No basin that existed in that area before tha Black Sea opening can be shown to have had any relationship with the Black Sea or with any Neo-Tethyan subduction zone to the south. Al though the data reviewed are compatible with the marginal basin opening model as propounded by Adamia et a1. (1974), Hsti et a1. (1977 and Lefouzey et a1. (1977), the whole story seems to have been more complicated: including other elements as well, especially much strike-slip faulting. That is why I deliberately avoided here any discussion on mechanism and the presentation of palaeotectonic maps, bu concentrated on the timing of the basin formation. - a - >"" . . . . . . . , " . ~ \'- ) I -- w ~ Ci d~ ,,, ,,. .J , . / ilr ~ ~ 2 // // 3 " " 1/4 ~ 5 ':-: '';:= .: 6 - 0- .- .- - d - in op S E A - - - - - - - - - - - - ~ 7 ~ - - - - - - - :" "/0 ( 8 < , . ,,~ 0?: ~-. ; -:; !> 1° \.., 4?-: _:_~ 1·'' '-'. .. \ J ,. F ig .l : C re ta ce ou s pa la eo ge og ra ph ic m ap s o f th e R ho do pe -P on tid e fr ag m en t. a : N eo co m ia n, b: A pt ia n- A lb ia n, c : C en om an ia n, d: C en om an ia n- Se no ni an , e : M aa st ri ch ti an . 1: s ha ll ow -w at er s a n ds to ne , 2: s he lf c a rb on at es c o m po se d m a in ly o f m ic ri ti c, s p ar it ic a n d r e e fa l li m es tb ne s c o n ta in in g o o li te s, ps eu do -o ol it es , in tr ac la st s, a n d bi oc la st s o f c e ph al op od s, bi al ve s, ga st ro po ds , a n d br ac hi op od s. 3: La nd a n d/ or e ro s io na l a ra , 4: La nd w it h s ho re li ne d ep os it io n, 5: D ee p- m ar in e c a rb on at es a n d c a rb on at e a n d te rr ig en ou s tu rb id it es (c on tin en ta l s lo pe /r is e a s s e m ba lg e) , 6: G la uc on it ic s a n ds to ne , c o n gl om er at e, sa n dy a n d c la ye y li m es to ne s, O rb it ol in a be ar in g m a rl s a n d U rg on ia n li m es to ne s w it h ga st ro po ds , 7: W ild f ly sc h (la rg e bl ac k do ts ) a n d tu rb id it es , 8: S ub du ct io n r e la te d v o lc an ic s. W tJ \ 136 REFERENCES Adamia,Sh.A., Gamkre1idze,I.P., Zakariadze,G.S. and Lordkipanidze,M.B., 1974. Adjaro-Tria1etsky progib i problema formirovaniya glubokovodnoi upadiny Chernogo morya. Geotektonika, 1:78-94. Aydln,M., Serdar,S. and $ahintlirk,b., 1972. Orta Karadeniz Bo1gesi Jeo1ojisi ve Petrol Olanaklan. Tlirkiye 6. Petro Kongr., Ankara:63-71. Bu1andji, Y .D. (Editor), 1976. Komp1eksonoye Iss1edovaniye Chernomorsky Vpadini. Akad.Nauk SSR, Moscow, 98 pp. Burchfie1,B.C., 1976. Geology of Romania. Geol.Soc.Am.Spec.Pap., 158, 82 pp. Gorlir,N., in press. Timing of opening of the Black Sea basin, Tectonophysics. Hsli,K.J., Nacev,I.K. and Vuchev,V.T., 1977. Geologic evolution of Bulgaria in the light of plate tectonics. Tectonophysics, 40:245-256. Ketin,i., 1966. Tectonic units of Anatolia (Asia Minor). Bu11.Minera1 Res.Exp1or.Inst. Ankara, 66:23-34. Khain,V., 1975. Structure and main stages in the tectono-magmatic development of the Caucasus: an attempt at geodynamic interpretation. Am.J.Sci. 275-A:131-156. Letouzey,J., Biju-Duva1,B., Dorke1,A., Gonnard,R., Kristchev,K., Montadert ,L. and Sungur1u ,0., 1977. The Black Sea: a marginal basin. Geophysical and geological data. In: B. Biju-Duva1 and L. Montadert (Editors), Structural History of the Mediterranean Basins. Technip, Paris:363-379. Letouzey,J. and Kimura,M., 1985, Okinawa Trough, genesis, structure and evolution of a back-arc basin developed in a continent. Mar.Pet.Geo1. 2:111-130. Sandu1escu,M., 1980. Analyse geotectonique des chains alpines situees autour de 1a Mer Noire occidentale. Annu. Inst. Geol. Geofiz. (Roman), 56:5-54. Sidorenko,A.V., 1978. Karta raz1omov territorii SSR i soprede1nich stran, 1/2,500.000 Akad. Nauka. SSSR, Moscow. $engor,A.M.C. and Yl1maz,Y., 1981. Tethyan evolution of Turkey: a plate tectonic approach. Tectonophysics, 75:181-241. Zonenshain ,L.P. and Le Pichon,X., 1986. Deep basins of the Black Sea and Caspian Sea as remnants of Mesozoic back-arc basins. Tectonophysics, 123:181-211. PALEOMAGNETIC STUDY OF THE NEOGENE FORMATIONS OF THE AEGEAN AREA 112 C. Kissel , C. Laj , A.Mazaud, A. Poisson, 2 Y. Savascin3 , K. Simeakis4 , C. Fraissinet 1, J.1. Mercier 1: Centre des Faibles Radioactivites, Laboratoire Mixte CNRS-(;EA, Domaine du CNRS, 91190 Gif-sur-Yvette, France. 2:Departement des Sciences de la Terre, UA 730 Geophysique et Geodynamique interne, Universite Paris XI, 91405 Orsay, France. 3:Dokuz Ey Iii 1 Uni ve rsi tesi, Mil h. Mi m. Fakiil te si , Jeoloji Bo ltimU , Bornova, Izmir, Turkey 4:IGME, 70 Messoghion Str., Athens, Greece ABSTRACT. New paleomagnetic data have been obtained from middle Miocene to Pliocene sedimentary and volcanic formations in central Aegea, in western Anatolia and in the Antalya basin in southern Anatolia. Together with the previous results obtained in northwestern Greece and with results from northern Aegea reported by other authors they suggest that the lower Miocene arc was originally almost rectilinear and that its curvature has been acquired tectonically most probably by opposed rotational deformations at both terminations. No significant rotation has been measured in the Antalya basin since at least 15 My, confirming that this basin has not been involved in the geodynamic evolution of the Ionian-Lycian arc. Furthermore, the values of inclination obtained from Miocene formations, both sedimentary or volcanic, are systematically shallower than expected for the geocentric dipole field. This suggests that all these regions have undergone a northward drift since middle Miocene. 1. INTRODUCTION The most recent reconstructions of the geodynamical evolution of the Mediterranean assume that the Aegean area is an Apulo-Anatolian microplate accreted to Eurasia during the last stages of the Africa- Eurasia collision (Biju-Duval et al., 1977; Channell et al., 1979; $engor and Yilmaz, 1981; Dercourt et aI., 1984; Vergely, 1984). The Mesozoic relative movements of these two major plates has been described using the marine magnetic anomalies in the Atlantic and Indian oceans (Patriat et al., 1982; Olivet et al., 1982). This method cannot however correctly describe the Tertiary tectonic evolution of the Aegean area, 137 A. M. C. !Jengor (ed.), Tectonic Evolution o/the Tethyan Region, 137-157. © 1989 by Kluwer Academic Publishers. 138 characterized during this period by intense deformation. Geological studies have documented the existence during the Oligo- Miocene period of an arcuate structure (Ionian-Lycian arc) (Mercier et al., 1979) which resulted from a very large compression and consequent shortening between Eocene and middle Miocene (Brunn, 1956; Aubouin, 1959; IGRS-IFP, 1966). The two ends of this arc are clearly marked by the Pindus overthrust and the Ionian Zone imbricated thrusts in northwestern Greece and the front of the Lycian nappes onto the Bey Daglari in southwestern Anatolia (Gutnic et al., 1979) (Fig.l). The present compressive front (Aegean arc) most probably appeared in the lower Pliocene wi th a sudden "jump" from the Miocene arc posi tion (Mercier at al., 1979). Istanbul \ o 100200km Fig.l: Present day configuration of the Aegean domain showing the location of the front of the Miocene arc (Ionian-Lycian arc). The studied regions are indicated. All these models are based on seismic and tectonic data. An important parameter which cannot be attained by these methods is the rotational deformation and its variations along the arc. Paleomagnetism seems to be an appropriate method to study such deformations. Previous paleomagnetic results have shown that northwestern Greece has undergone a large clockwise rotation (50°) which occurred in two separate phases of roughly equal amplitude, one between 24 and 12 My and the second one between 5 My and the present. These data indicate that the original lower Miocene orientation of the Ionian structures was almost E-W (Horner and Freeman, 1983; Kissel et al., 1984, 1985). At the other end of the arc, the presence of windows in the Lycian nappes indicates a progressive increase of the southeastward 139 displacement from the Isparta region to the Aegean coast. This observation suggests an anticlockwise rotation (Poisson, 1984). Published paleomagnetic results from western Anatolia (Kondopoulou and Lauer, 1984) are too scarce for the considered epoch, so that they do not allow any defini tive conclusion. This lack of data is also apparent in central Aegea although some paleomagnetic results have recently been reported from this region (Kondopoulou, 1985). This paper reports on new results from 75 sites in the central part of the Agean area and its eastern and western boundaries (Volos region, Evia, Skyros, Lesbos and the Izmir region) and from southern Anatolia (Antalya region). These data will be analysed together with the ones from northwestern Greece and with the results from northern Aegea and western Anatolia reported by other authors. 2. GEOLOGICAL SETTING AND SAMPLING One of the most important results of the geological and geophysical studies has been the recognition of the widely different regimes which have affected the different regions of this area (Mercier et al., 1979 (and references therein); Le Pichon and Angelier, 1979 (and references therein» • A compressive regime has existed along the two successive margins and is still active today at the two ends of the arc in the northwestern and southeastern regions (Ionian islands and Cyprus) (Mercier et al., 1979; Letouzey and Tremolieres, 1980, Rotstein et al., 1982). Most of the Aegea however (the internal domain as well as Crete and Rhodes) has been submitted to a large scale extensional regime since at least the upper Miocene, only occasionally interrupted by compressive events in the lower Pliocene and lower Pleistocene periods (Mercier et al., 1979; Le Pichon and Angelier, 1979, 1982). This has resulted in large-scale normal faulting grabens and troughs. This area is characterized by widespread calc-alkaline volcanic products of both Tertiary and Quaternary age. Different authors have recognized the existence of two main distinct volcanic phases of Oligo- Miocene and Plio-Quaternary ages separated by a period of quiescence of several million years (Innocenti et al., 1981, Bellon et al., 1979). While the most recent products are found in the southern Aegean sector, older volcanics of Oligo-Miocene age are distributed all over the central and northern Aegean regions and in western Anatolia (Borsi et al., 1972; Bellon et al., 1979; Fytikas et al., 1984). The older volcanic phase is considered to have occurred about 13 My ago, while the firs t Plio-Quaternary products were erupted around 5 My ago along the southern Aegean arc. This time gap between the two cycles and a southward migration of the volcanic front has led different authors to suggest that the two volcanic cycles might be attributed to two distinct geotectonic events (Vilminot and Robert, 1974; Fytikas et al., 1984; Bellon et al., 1979). It has also been noticed that a correlation exists between the timing of the two calc-alkaline volcanic phases and the two distinct phases of rotational deformations observed in northwestern Greece, also suggesting two distinct tectonic events (Kissel et al., 140 1985). Sedimentary formations in the studied area are essentially represented by continental (fluvio-Iacustrine) and brackish limestones and marls which indicate the quasi-definitive emersion of these regions. In western Anatolia, as a result of the main volcanic activity, widespread volcano-sedimentary deposi ts are also present. On the other hand, southern Anatolia was still a marine domain and a thick series of marls and flysch was deposited in the Antalya basin (Keraudren, 1979; Kaya, 1981). For the correct interpretation of the paleomagnetic data it is necessary to restore the formations to their paleo-horizontal posi tion and this might be difficult for the volcanic formations. In some cases interbedded sediments yielded a precise bedding correction, in others a local structural study gave a reasonable estimate of it and finally in others no precise correction for tilting was possible and we had to rely on a subjective choice of "almost horizontal" lava flows based on the morphology of the flow i tse 1f. The new paleomagnetic data presented in this paper have been obtained in the volcanic formations from the islands of central Aegea (Evia, Skyros, Lesbos), western Anatolia and the marine sediments of the Antalya basin (Fig.1). A brief description of these regions is given below: 2.1. Volos region Between the two main phases of volcanic activity (Oligo-Miocene and Plio-Quaternary), small volumes of lava with variable petrogenetic character were emi tted over all central Aegea. In the Volos-Atalanti region they consist of a series of small lava outcrops aligned in a SW- NE direction. Four sites were sampled in the lava flow located near the village of Glifa. This lava shows basaltic-andesitic characteristics and is dated at about 3My (Innocenti et al., 1979). The structural position of this flow and its general aspect indicate that it has not undergone any significant tectonic tilt and we have considered it as almost horizontal. 2.2. Evia. The volcanic complex of lava flows and domes located south of Kymi has been sampled at 6 different si tes. These volcanic products, of calc- alkaline affinity with high K20 content, have been dated at about 13 My (Fytikas et al., 1980). This complex lies on the lacustrine marly limestones of the Kymi basin in which we have sampled at one site (EU 232). The age of these sediments is not very precise but according to Lemei lle (1977), it ranges between Aquitanian and Ponti an • Moreover their bedding plane which is regular allover the region, has been used for restoring the volcanic formations to their paleohorizontal posi tion. 2.3. Skyros. Volcanic products in Skyros are located in the central and southeastern 141 parts of the island but in this last location heavy weathering precludes any paleomagnetic study. We have sampled 5 sites in the calc-alkaline lava near the village of Bares in the central part. The products range between andesites and dacites and are dated at 15 My (Fytikas et al., 1980). They have been intruded into the Mesozoic carbonatic series which appear more or less horizontal in this region. 2.4. Lesbos The Tertiary formations of Lesbos consist of various volcanic series: ignimbrites, lower and upper lava units, dykes and continental sedimentary formations. Thirty one sites were sampled in these different lithological units (Fig.2). The composition of the ignimbrites is rhyodaci tic while the lava flows are classified as lati te-andesi te and dacite. The K/ Ar age of the volcanic formations range from 18 My (lava flows) to 16.2 My (dykes) (Borsi e t a1., 1972). 2.5. Izmir-Bergama region This region is located in western Anatolia between the Pontides to the north and the Menderes massif (Taurides) to the south. It corresponds exactly to the suture zone of the main segment of the north Neo-Tethys. The refore the Neogene formations mask many important deeply rooted ancient faults of supposed NE-SW direction. The present day main structural lines are the result of a general N- S extension (Mc Kenzie, 1978; Le Pichon and Angelier, 1982) which was initiated during lower Miocene (intensive volcanic activity). The graben system faulting occurred mainly during the upper Miocene and lower Pliocene. The morphologically most outstanding structures are the E-W and NNE-SSW graben structures. While the NNE-SSW structures are probably partially inherited from the Neo-Tethys ones the E-W fault system corresponds to a post-Alpine, neotectonic structural direction in this region. The structural framework of the central and southern grabens of the Menderes Massif (mainly E-W) is less complicated than the northern one. In this last region a great number of small blocks and grabens result from the high density of faults following the two main directions, cutting up an heterogeneous substratum. The volcanic activity attained its highes t in tensi ty in this region. The faul ts are thus probably deep and their repeated movements certainly induced crustal thinning. A total of 22 sites were studied in both volcanic, sedimentary and volcano-sedimentary formations (Fig.2). The volcanic products of lower Miocene age (16 to 18 My) (Borsi et al., 1972) show a trend from latite- andesite to dacite and rhyodacite. The basaltic rocks with alkaline affinity (Urla, Karaburun, Fo~a, Ayvalik) correspond to the youngest volcanic activity of the region about 11 My ago. The sampled sedimentary formations are not precisely dated but are supposed to be of middle to upper Miocene age (Kaya, 1981). In spite of its rather loose limits, this age determination is sufficient in the present state of advancement of the paleomagnetic study. 142 " ~ " o ~. ,,' 0 1 Ik-;'~d 2 0 3 bj4 Fig. 2: Geological simplified map of Lesbos and of the Izmir region showing the location of the studied sites. 1: Plio-Quaternary sediments; 2: Neogene volcanic series; 3: plutonic rocks; 4: Pre-Neogene basement. 143 2.6. Antalya basin The Antalya basin occupies an exceptional posi tion in the eastern Mediterranean domain. Indeed it is located upon the suture of the south Tauric branch of the southern Neo-Tethys: the Antalya trough which was closed during Paleocene-Eocene times (Poisson et al., 1983). Moreover it is also situated upon both extremities of the two regional arcuated structures: the eastern end of the Ionian-Lycian arc to the west (formed during the middle Miocene compressive phase, Poisson, 1984), and the western end of the previous Tauric arc to the east (late Eocene, Gutnic et al., 1979). These curved structures resulted from the main Alpine orogenic phases of the Helleno-Tauric belt. MER 30' MEDITERRANEE Fig. 3: Geological environment of the Antalya basin and location of the paleomagnetic sites. l:post-tectonique neogen sequences. 2:ante-tectonic Neogen sequences (Late Oligocene - Lower Messinian). 3:Lycian nappes. 4:Antalya nappes. 5:autochtonous and parautochtonous Tauric carbonates platforms. Compressive phases of deformation: 6:Paleocene - lower Eocene (Antalya nappes); 7: late Eocene (Bey§ehir nappes - Tauric arc); 8:middle Miocene (Lycian nappes - Ionian-Lycian arc); 9:upper Miocene (Aksu thrust). This region is also characterized by a reorientation of the regional stress field which occurred during upper Miocene (after the Alpine ac ti vi ty) as a result of the Red Sea opening and of the northeastward motion of the Arabian plate. During this time E-W shortening occurred in the Antalya basin producing important westward 144 thrust along the Aksu 9ay valley (Aksu thrust, Poisson, 1977; Poisson et al., 1983). It is now clearly established that the Neogene formations of the Antalya basin overlie different old deeply rooted tectonic contacts, and the chronology of the superimposed deformations is well known. One can consider the Antalya basin as a key sector for the understanding of the geological evolution of the entire eastern Medi terranean region. Two sedimentary sequences were recognized in this basin. The lower one ranges in age from late Oligocene to lower Messinian (Poisson et al., 1983). It is constituted by marine molassic facies (marls, sandstones and conglomerates), passing northwards to continental facies (del taic conglomerates). This sequence can be observed in the Manavgat basin and both in the Aksu 'lay and the Kopru Clay basins. These two valleys are separated by a main westward thrust (Aksu thrust) whose movement is of late Miocene age. The upper sequence (post-Aksu thrust) is composed from bottom to top of limestones (Gebiz limestones) of supposed upper Messinian age and Pliocene marls with sandbeds intercalations. These marine facies have lateral continental equivalents (deltaic conglomerates and brackish limestones)(Gutnic et al., 1979). The sedimentary sites (9) were sampled in the two sequences, in the limestones and the marls of Langhian to upper Pliocene age. Moreover, three sites were sampled in the volcanic formations located near Isparta. The lava (IS 04) has been dated by K/Ar method at about 4 My (Lefevre et al., 1983) and the ignimbrites (IS 29-31) are intercalated with Pleistocene sediments (Bering, 1971) (Fig.3). 3. RESULTS 3.1. Generals Standard paleomagnetic techniques were used to study the samples in the laboratory. Both thermal and AF demagnetization were used (up to at least 550·C and 0.08 T respectively). When both methods isolated a stable component of magnetization, its direction was identical for samples from the same core. In many cases, however, AF demagnetizaton did not succeed in decreasing the remanent magnetization to values lower than 10-15% of the natural remanent magnetization (NRM) due to the presence of high coercivity minerals. Thermal treatment gave more satisfactory results. Measurements were done using a Spinner magnetometer for the volcanic samples and a LETI 3-axes cryogenic magnetometer for the sedimentary ones. In almost all cases a single component of magnetization has been isolated after the first steps of magnetization (after 250 or 300·C) and its direction is easily determined within 2-3· of accuracy. Some of the obtained demagnetization diagrams are shown in Fig. 4. As a general rule, the low-field magnetic susceptibility of samples was measured at each step of the thermal demagnetization using a Digico susceptibility bridge both for volcanic and sedimentary samples. No 145 significative change was noticed between room temperature and 550·C indicating that the magnetic minerals were not seriously affected by the thermal treatment. In some cases the value of the susceptibili ty of volcanic rocks exceeded the upper limi t of our instrument and could not be measured. U"W VO 235 0' C Mo= 597 3'" NIN UI;YN IZ 43 08B % Mo=388 AN 35 '0 S~ 221 OIB Mo= 0.57 Mo =407 Fig. 4: Thermal demagnetization diagrams of various samples from si tes of the studied regions. Open cirles: projection on E-W or N-S vertical plane; Full circles: horizontal projec tion. Highest temperatures range from 550·C to 600·C. Mo: NRM intensity (10- 3 Aim). For the sedimentary formations we have measured the anisotropy of magnetic susceptibility using a Digico Anisotropic Magnetometer (CRAD) modified in the laboratory. In almost all cases, this analysis revealed a primary sedimentary magnetic fabric, suggesting that no large syn- or post-sedimentary perturbations have affected the sampled formations. For the sedimentary sites, stepwise acquisition of IRM up to 1.5 T was used to identify the magnetic carriers in the sediments. The IRM saturates at 0.15-0.2 T suggesting that the NRM is mostly carried by magneti te, wi th a contribution of some higher coercivi ty minerals also present in some samples. 3.2. Paleomagnetic results 3.2.1. Volos region - Analysis of the 4 sites reveals a reverse stable component of magnetization. The four sites yield tightly grouped results as can be seen in Table I. These results show that no significant rotation has occurred in this region since 3 My. Moreover an inclination 146 of 59·, agrees closely with that expected at this latitude which confirms that the flows have not undergone any significant tilting. Si tes n D I K 0/.95 VO 235 9 176.0 -61.5 266 2.8 VO 236 6 190.0 -54.0 26 11.3 VO 237 10 166.0 -67.0 120 4.0 VO 238 7 184.5 -54.5 29 9.7 N = 4/4 D = 178· I = -59· K = 85 0/. 95= 7.6· Table I: mean paleomagnetic directions calculated using Fisher's statistic for sites from Volos region. All these sites are dated at about 3 My. The bedding plane has been estimated horizontal. The mean regional direction is also calculated at the end of the Table. 3.2.2. Evia - All the 6 sites studied in the volcanic formations have a single reverse component of magnetization which is isolated in the very early stages of demagnetization (between room temperature and 150· C). This component largely deviates from the N-S direction, indicating a large clockwise rotation of 48· for this region since 13 Ma (Table II). All the data were corrected using the tectonic bedding measured on the underlying sediments. The high scatter of the data from the only sedimentary site (EU 232) shows that these marls are not a faithful recorder of the geomagnetic field. Si tes n Da Ia Db Ib K cX95 EU 220 7 222.0 -28.5 228.0 -36.8 950 1.7 EU 221 7 220.0 -42.0 231.0 -50.5 320 3.0 EU 222 7 223.0 -28.0 229.0 -36.0 63 7.2 EU 223 9 218.0 -32.0 225.0 -41.0 320 2.6 EU 224 10 235.8 -36.0 245.0 -41.0 45 6.5 EU 225 10 199.7 -41.8 206.0 -53.0 201 3.0 EU 232* 10 222.0 -46.0 204.0 -65.0 3 24.0 N = 6/7 D = 228· I = -43.5· K = 53 eX. 95= 7.8· Table II: Mean paleomagnetic direction of each measured sites of Evia. The directions are given before (Da, Ia) and after (Db, Ib) bedding correction. The mean paleomagnetic direction does not take into account si te (EU 232*) which has been rejected because of its ve ry large sca t te r (K=3). 3.2.3. Skyros - The fi ve samp led si te s have been s tudi ed using therma 1 demagnetization. From the results shown in Table III it can be seen that 147 the within-site scatter is small (K always> 20), but that the mean direction of the stable magnetization obtained from site SK 231 is quite significantly different from the others. Contrarily to the other sites which have been sampled in the bottom of the valley, this site is situated on the flank of a hill. A possible explanation of the different direction of magnetization could then be the presence of a local landslide which could be easily be unnoticed because of the rather thick vegeta tion. The mean direction has been calculated using the four sites only. This direction shows that Skyros has undergone about 26° of clockwise rotation since middle Miocene. Si tes n D I K d 95 SK 227 9 31.0 40.0 188 3.4 SK 228 10 20.0 38.0 54 6.0 SK 229 9 32.0 49.0 348 2.8 SK 230 9 22.0 55.0 27 11.0 SK 231* 10 285.0 37.0 735 1.7 N = 4/5 D = 26° I = 45.5° K = 82 d.9;= 7.7° Table III: mean paleomagnetic directions for each si te from Skyros island dated at about 15 My. The site SK 231 is not used for the final calculation (*) because of its very different direction (see text). No indications of tectonic tilting have been found in the field. 3.2.4.Lesbos - Widely different magnetic properties were found for the different sites sampled in Lesbos, as might be expected from their considerably different lithologies. Some of the sampled formations were found not to be suitable for a paleomagnetic study. The sedimentary formations (the lacustrine limestones near the town of My ti lini ) have a NRM lower than 0.03 10-3 Aim and were excluded from further study. In a general way, the ignimbrite flows in Lesbos were not considered as trustworthy material for a paleomagnetic study since consistent directions could not be isolated within a flow. Also in the case of the dykes the results were not as good as expected. Single samples are characterized by perfectly rectilinear demagnetization diagram yielding a very precise paleomagnetic direction, but the wi thin-si te scatter is too high and K < 15 in all cases. The only sites which systematically gave coherent and reliable results were those in the lava flows from either the upper or the lower lava uni ts. All the results are reported in Table IV. Only 17 out of the 27 studied sites can be considered as trustworthy recorders of the geomagnetic field. It can be noticed that all the lava flows of the upper lava unit are of reverse polarity and that from the lower units of normal polarity. This could suggest that both these two units were emitted during rather short eruptive phases. 148 Si tes Ceo!. form. n Da Ia Db Ib K '45 LE 268* 9 rejected LE 273 10 31.4 58.5 19.1 56.0 257 2.7 LE 274 Lower 10 358.6 44.3 353.5 38.7 134 3.8 LE 275 Lava 9 ----- ---- 17.3 64.0 233 3.0 LE 277 Uni t 9 ---- ---- 349.5 45.5 170 3.5 LE 278 9 ----- ---- 10.0 44.5 72 5.0 LE 263 9 ----- ---- 177 .8 -49.0 1333 1.3 LE 264* 8 rejected LE 269 8 ----- ---- 146.0 -63.6 177 3.7 LE 271 Upper 11 ----- ---- 214.0 -20.0 15 10.0 LE 272 Lava 10 ----- ---- 191.7 -44.7 181 3.3 LE 279 Unit 7 ---- ---- 199.5 -20.0 144 4.4 LE 280 7 ----- ---- 191.0 -41.0 46 7.8 LE 256 11 ----- ---- 340.5 55.0 229 2.8 LE 257 10 ----- ---- 350.0 69.0 262 3.0 * 6.0 LE 260 I Ignimbr. 5 ----- ---- 6.0 21.0 101 LE 260 u* 7 ----- ---- 20.0 9.0 107 5.0 LE 260 IU* 7 ----- ---- 10 .2 19.3 203 3.7 LE 260 IV* 9 ----- ---- 6.0 14.0 190 3.3 LE 267"" 11 ----- ---- 221.0 -63.8 3 24.0 LE 270i.o Dykes 6 ----- ---- 251.0 31.5 12 20.2 LE 276 "* 9 ---- ---- 44.8 31.6 2 31.0 LE 249 My ti lini 10 8.0 37.0 11.2 27.0 280 2.6 LE 252 8 0.2 49.0 4.8 40.2 153 4.0 LE 253 basalts 9 13.6 59.5 17.2 49.7 196 4.0 LE 255 limestones 11 ----- ---- 11.3 53.3 704 1.6 LE 258* marls 11 rejec ted N = 17/27 D = 6° I = 49.5° K = 24.7 01.. 95= 6.8° Table IV: mean paleomagnetic direction of each measured sites of Lesbos. *:sites rejected on the basis of undeterminable direction or too large within-site scattering (K and Lauer (1984) were not precise enough to either demonstrate the presence or the absence of such rotation. 149 3.2.5. Izmir-Bergame region - Twenty two sites from this region have been completely studied, sixteen from the lava flows, two from lacustrine limestone and marls and four from tuff formations. Five of these sites did not yield satisfactory results because a stable component of magnetization could not be isolated or because the within- site scatter was too large (K=7). -------- ---------- Si tes ogr. pos. n -------- ---------- IZ 14 8 IZ 11* 10 rejected IZ 15 5 226.0 -84.5 35 10.5 IZ 16 Karaburun 6 43.5 16.3 97 5.8 1Z 17* 6 62.7 -8.2 7 17.6 IZ 08 11 267.0 -34.5 23 8.5 IZ 43 Peninsula 8 234.0 -53.5 450 2.6 (tyel?meI) 7 254.0 -51.9 189 4.4 ( yel?meII) 7 237.0 -51.5 52 8.4 (yelilmeIII ) 5 207.0 -59.5 132 6.7 ------- -------- ------- N = 8/10 D = 50· I = 52· K = 9 r4s= 16· ---------- ---------- ------- ------- -------- ------- IZ 06 12 128.0 -47.0 128.0 -47.0 158 4.2 IZ 07 Izmir 8 144.0 -34.0 141.0 -39.5 249 3.3 IZ 09 12 169.0 -62.5 127.2 -41.0 93 4.2 IZ 10 8 192.0 -55.0 144.0 -47.5 500 2.2 IZ 22 Bergama 11 190.0 -48.0 190.0 -48.0 387 2.4 IZ 24 7 348.0 64.0 56 7.0 1Z 25* 9 207.0 64.8 227.7 -43.0 7 15.3 IZ 26* 4 69.8 47.0 7 17 .5 IZ 28 8 346.0 54.0 349.6 57.0 57 4.0 IZ 44 9 342.0 71.4 828 1.6 IZ 45* Region 8 343.0 4.0 720 4.3 IZ 46 8 319.5 53.8 314.0 53.8 540 2.3 IZ 57* 9 328.0 36.6 325.0 14.7 35 8.0 IZ 58* 9 rejected -;-:-;/~~ -------~-: 3;;~ -----;-~-~;~-------~-:-;;-------;{~::-;~;~---J ~~~~~~~~~[~~~~~~~~~~~[~~~I~=====~[~~====~][~~~~~~~I~~~~~~~]~~~~~I~~~~~ Table V: Mean directions from Western Anatolia. sites rejected for the same reasons than in Lesbos (see Table IV). Two main regions have been distinguished and the mean paleomagnetic direction calculated for each of them shows that they have undergone very different rotational de forma ti ons • 150 In Table V we report all the obtained results. For a better understanding of these data we have distinguished two main groups according to their geographical origin. The first group (IZ 08-14-15-16- 17-43) has been sampled in the Karaburun peninsula which is separated from the mainland by a narrow isthmus cut by a main N-S normal fault. Thus the Karaburun peninsula might be tectonically distinct from the Izmir-Bergama region. Although the bedding plane of the Karaburun sites is not precisely known, except for site IZ 14, the morphological and geological aspects indicate that bedding tilt is never very large. We have thus assumed an horizontal bedding. Of the five sites which give reliable results, four are of lower Miocene age (21-18My) while IZ 14, of basaltic composition, most probably belongs to the alkaline eruptive phase dated at about 10 My. In Table V we have also reported the results obtained by Kondopoulou and Lauer (1984) relative to this same region ( 151 The other 14 sites belong to the main Izmir-Bergama region. In all these cases the bedding plane could be determined on interbedded sediments except for two sites (IZ 44, 45). In contrast with the Karaburun si tes, all the results from these regions indicate an anticlockwise rotation of about 30° (Table V and Fig. 5B). It thus appears that the Karaburun peninsula and the mainland Iz mi r- Be rgama regi on have u nde rg one a di f fe ren t tec tonic e vo lu ti on. The rotation measured on three sites by Kondopoulou and Lauer (1984) is confirmed by these new results. However the new results also show that the geodynamical evolution of this region cannot be considered as representative for the entire western Anatolia. Finally site IZ 54 belongs to the Pontides region north of the Edremit - Balikesir fault. The result does not indicate any major rotation in agreement with its rather young age. 3.2.6. Antalya basin - Eleven sites from the different sedimentary series were completely studied. Samples from AN 36 were found unstable upon thermal demagnetization and no stable component of magnetization could be determined. On the other hand all the other sites, ranging in age from Langhian to lower Pleistocene yield reliable results. The dispersion parameter K calculated using Fisher's statistics for the 10 reliable sites shows a significative increase upon bedding correction (K2/K,=2.5; F(18,18) 5% = 2.2). This indicates that the magnetization has been acquired before the tilting, which is a strong indication of its stability. It can be seen from Table VI that these results indicate that no significative rotation of this basin has occurred since at least the Langhian. Si tes Age n Da Ia Db Ib K 0(95 IS 29 Lower 8 ----- ---- 5.3 59.0 170 3.7 IS 31 Pleis tocene 9 ----- ---- 358.0 58.0 288 2.7 AN 37 Up. Plioc. 9 198.0 -60.0 186.0 -48.0 270 2.8 AN 36* 10 rejected AN 40 Pliocene 10 ---- ---- 358.0 52.0 107 4.0 IS 04 Lower 9 ---- ---- 12 .5 46.6 221 3.0 AN 39 Pliocene 8 192.0 -41.0 182.0 -52.0 19 11.0 IS 32 Se rravallian- 8 332.0 48.0 352.0 43.0 163 4.0 AN 33 -Tortonian 9 3.0 43.0 359.0 45.4 24 9.5 AN 34 Langhian 4 7.4 26.0 11.3 38.4 23 14.0 AN 35 9 355.0 35.0 347.0 59.0 49 7.0 N = 10/11 D = 1 .5° I = 50° K = 85.6 01.. 95= 4.8° Table VI: Mean directions obtained from sites of the Antalya basin before (Da, Ia) and after (Db, Ib) bedding correction. 152 4. DISCUSSION When all the available paleomagnetic results from the Aegean area are analyzed together, one obtains a rather complete picture of Neogene rotational deformations which have occurred in this region. All these results, grouped according to their geographical origin, are shown on the schematic map of Figure. 6. Although the lack of emerged lands in central Aegea does not allow an extensive sampling, it can be seen that the large clockwise rotation (45°-50°) measured in northwestern Greece progressively decreases when moving eastward. Indeed Evia has rotated about 48°, Chalkidiki about 37°, Skyros about 26° and the rotation of Lesbos is only 6°, which is barely significant. The formation sampled in the Volos region is too recent to be used in this geodynamical reconstruction. However they show the validity of the paleomagnetic results, in the sense that the expected declination and inclination values are obtained precisely. In contrast with the coherent evolution of the northwestern part of the Aegean area, the geodynamical evolution of its eastern part seems to be much more complex. This complex behaviour is well documented by the widely different paleomagnetic directions obtained from regions as geographically close as Lesbos, Karaburun and the Izmir-Bergama region. It should be noted however, that the clockwise rotation is limited to the Karaburun peninsula, while reliable sites which indicate an anticlockwise rotation are geographically widely distributed over regions isolated from each other by main faults. Although this fault pattern isolates "microblocks" which might be considered as somewhat independent (~engor et al., 1985) and in spite of a probable recent reactivation of the main faults, the Izmir-Bergama region has a coherent behaviour as indicated by the paleomagnetic data. The scatter of these data about the mean, anticlockwise, direction could result from the relative movements of the different small blocks created by the grabens sys tem. Indeed, the tec tonic ac ti vi ty has pe rsis ted throughout the Miocene, the Pliocene and Quaternary, thus a strike-slip motion of some of the numerous faults during recent periods would probably be most effective in causing these relative movements of the blocks. Although no detailed and extensive study of the Neogene faults in this region has been carried out, motions of this kind have been clearly demonstrated for the Buyuk Menderes faults (Dumont et al., 1981a-b; Angelier et al., 1981). However it should also be borne in mind that in regions where fold axes are not horizontal the simple tilt correction applied here may introduce small spurious anomalies in the value of the declination (MacDonald, 1980). We thus believe that the main tectonic feature of this region is a large anticlockwise rotation. This result is in agreement with the anticlockwise rotation suggested by the geological observations of Poisson (1984) in the Lycian nappes. We are well aware that additional data are nedded before this rotation can be firmly considered as significative for the entire western Anatolia north of the Lycian nappes. N ~ L L J , CD M ' rg l> ' " G) r\ 7' I!l LY " CD [J N ~ ~ , [Sl ' .. N /] }; A L 1 -J , ~ti J' " - ~ 01 \:7 ' N ~L .) Y' CU. !De cl~ ~" . ~'I ~ B LY I nc l. OD [!l ] c CD M ' o LY .. o F ig . 6: M ea n pa le om ag ne ti c d ir ec ti o n s in e a c h r e g io n . T he b ro ke n li n e in d ic at es th e p re se n t- d ay in cl in at io n ( se e te x t) ; A : po st -M io ce ne r o ta ti o n s; B: po st -O li go ce ne r o ta ti o n s; C : r e fe re n ce s (1 : L aj e t a I. , 19 82 ; 2: K is se l e t a I. , 19 85 ; 3: K on do po ul ou , 19 85 ; 4: T hi s p ap er ). D : a ge o f th e s tu di ed fo rm at io ns (M a) . -V I W 154 A palinspastic reconstruction in both northwestern Greece and western Anatolia then shows that the orientation of the main structures was almost E-W in both regions. This implies that the Miocene arc was almost rectilinear at its formation and that the curvature has been acquired tectonically by opposed rotational deformation at both ends. The data are however still too scarce to yield precise ideas about the mechanisms involved in these rotational deformations, such as rotation of small blocks or rigid-plastic deformation (Mc Kenzie, 1972; Tapponnier, 1977; Dewey and ~engor, 1979; Mercier, 1979). The results obtained in the Antalya region indicate that this basin has not been affected by the rotational deformations involved in the geodynamical evolution of the Ionian-Lycian arc. The eastern termination of this arc is thus clearly situated west of the Antalya gulf. The basin has not been affected either by the deformations connected to the more recent Aegean arc, so that it appears that the "jump" of the active front has been spatially larger in this region than in northwestern Greece. Indeed it is only in Cyprus (200 Km south of Antalya) that compressive structures similar, although less active, to those of Ionian islands can be found (Rotstein and Kafka, 1982). The Antalya basin thus appears as a relatively stable point with respect to the evolution of the two successive arcs. A second major point which can be noticed in the results from all the regions is that the measured values of the inclination angles are lower than those expected on the basis of a geocentered dipole field (59°). This is observed in both our own and other authors' results (Figure 6) and in both sedimentary and volcanic formations. Thus these results are a convincing evidence that a significant northward motion of Aegea has occurred during the last 15-20 My. The studied formations were situated about 10° south of their present position. It is significant that a similar northward motion of Africa has been documented in a recent publication by Tauxe et ale (1983). However because the African polar wandering path for the considered epoch differs according to the different authors it is not possible at present to determine whether the two northward dritfts have been identical or not. Acknowledgements - The Director of the Institute of Geology and Mining Research (IGME) at Athens kindly provided the necessary permits. The Earth Sciences group of the Dokuz Eyhil Universitesi has helped us in the field work. Dario Decini and Sylvie Guitton helped with the sampling. The financial support was given by the INAG-ATP Sismogenese, Plis, Failles: Mecanique de la lithosphere. Contribution CFR n0753 References ANGELIER,J., DUMONT,J., KARAMANDERESI,H., POISSON,A., SIMSEK,S. and UYSAL,S. 1981. 'Analysis of fault mechanisms and expansion of Southwest Anatolia since the Late Miocene'. Tectonophysics, 1~, T1-T9. 155 AUBOUIN,J. 1959. 'Contribution a l'etude geologique de la Grece septentrionale, les confins de 1 'Epire et de la Thessalie'. Ann. Geol. Pays Hellen., 1Q, 525p. BELLON,H., JARRIGE,J.J. & SOREL,D. 1979. 'Les activites magmatiques egeennes de l'Oligocene a nos jours, et leurs cadres geodynamiques, donnees nouvelles et synthese'. Rev. Geogr. Phys. Geol. Dyn., ll, 41-55. BERING,D. 1971. 'Lithostratigraphie, tektonishe Entwicklung und Seengeschichte der neogene und quartaren intramontanen Becken der Pisidischen Seenregion (sudanatolien)'. Beih. Geol. Jb., 19,1, 150p. BIJU-DUVAL,B., DERCOURT,J. and LE PICHON, X. 1977. 'From the Tethys Ocean to the Mediterranean Seas: a plate tectonic model of the evolution of the Western Alpine system'. In: Biju-Duval and Montadert Eds. Histoire struct. des Bassins Mediterraneens. Ed. Tecnip Paris, 143- 164. BORSI,S., FERRARA,G., INNOCENTI,F. and MAZZUOLI,R. 1972. 'Geochronology and Petrology of recent Volcanics in the Eastern Aegean Sea (Western Anatolia and Lesbos island)'. Bull. Volc., ~~Yl~1, 473-496. BRUNN,J.H. 1956. 'Etude geologique du Pinde septentrional et de la Macedoine occidentale'. Ann. Geol. Pays Hellen., ~, 258p. CHANNELL,J.E.T., D'ARGENIO,B. & HORVATH,F. 1979. 'Adria, the African Promontory in Mesozoic Medi terranean Palaeogeography'. Earth Sci. Rev., 12, 213-292. DERCOURT~J., RICOU,L.E., LE PICHON,X., BOULIN,J., BlJU-DUVAL.D., LEPURIER,C., GEYSSANT,J., ZONENSHAIN,L.P., KASMIN,V.G., KNIPPER,A.L., SBORSHCHIKOV,I.M. & SOROKHTIN,O.G. 1984. 'Geologic evolution of the Tethys convergence zone from Liassic to Present between the Atlantic and Pamir'. Abstract, 27eme Geol. Int. Congr. Moscou, IX, part 1. DEWEY,J.F. and ~ENGOR,A.M.c:-fcf7~Aegean and surro~;ding regions: complex multiplate and continuum tectonics in a convergent zone'. Geol. Soc. Am. Bull., 90 I, 84-92. DUMONT,J.F., UYSAL~1;.=and SIMSEK,S. 1981a. 'Superposition des jeux sur une faille et succession des evenements neotectoniques. L' exemple d 'Ephese (Turquie)'. C.R. somm. Soc. Geol. Fr., 1, 22-24. DUMONT, J. F., UYSAL, S., SIMSEK, S., KARAMANDERESI,H. '" and LE TOUSEY, J. 1981 b. 'Formation of the grabens in Southwestern Anatolia'. Bull. Min. Research Explor. Inst. Turkey (1979), ~l, 7-18. FYTIKAS,M., GIULIANI,O., INNOCENTI ,F. , MANETTI,P., MAZZUOLI,R., PECCERILLO,A. and VILLARI,L. 1980. 'Neogene volcanism of the Northern and Central Aegean region'. Ann. Geol. Pays Hell., }Q, 106- 129. FYTIKAS,M., INNOCENTI,F., MANETTI,P., MAZZUOLI,R., PECCERILLO,A. & VILLARI.L. 1984. 'Tertiary to Quaternary evolution of volcanism in the Aegean region'. In Eds Dixon and Robertson: The Geological Evolution of the Eastern Mediterranean, Special Publication of J. Geol. Soc. of I:()";don, 17, 687-700. - --- ---- GUTNIC,M.,-MONOD,O., POISSON,A. and DUMONT,J.F. 1979. 'Geologie des Taurides occidentales (Turquie)'. Mem. Soc. Geol. Fr., 137. HORNER,F. and FREEMAN,R. 1983. 'Paleomagnetic evidenC;-from pelagic limestones for clockwise rotation of the Ionian Zone, Western Greece'. Tectonophysics, 2§, 11-27. 156 I.G.R.S-I.F.P 1966. 'Etude geologique de l'Epire (Grece nord- occidentale)', Ed Technip, Paris, 306pp INNOCENTI,F., MANETTI,P., PECCERILLO,A. and POLI,G. 1979. 'Inner arc volcanism in NW Aegean Arc: geochemical and geochronological data'. N. Jb. Mineral. Mh. J.G., 445-458. INNOCENTI ,F. , MANETTI,P., MAZZUOLI,R., PASQUARE,G. & VILLARI,L. 1981. 'Neogene and Quaternary volcanism in the Eastern Mediterranean. Time- space distribution and geotectonic implication'. In: WEZEL,C. (ed.). Sedimentary basins ~ mediterranean margins. Technoprint, 369-385. KERAUDREN,B. 1979. 'Le Plio-Pleistocene marin et oligohalin en Grece: stratigraphie et paleogeographie'. Rev. Geo!. Dyn. Geogr. Phys., H, 17 -28. KAYA,O. 1981. 'Miocene reference section for the coastal parts of West- Anatolia'. Newsl. Stratigr., 1Q, 164-191. KISSEL,C., JAMET,M. and LAJ ,C. i984. 'Palaeomagnetic evidence of Miocene and Pliocene rotational deformations of the Aegean Area'. In: Eds Dixon and Robertson. The Geological Evolution of the Eastern Mediterranean, J. Geol. Soc. L~on, !I, 669-679. ----- KISSEL,C., LAJ ,C. and MULLER,C. 1985. 'Tertiary geodynamical evolution of Northwestern Greece: paleomagnetic results'. Earth Planet. Sci. Let., 1'1:,190-204. KONDOPOUL6u,D. & LAUER,J.P. 1984. 'Palaeomagnetic data on the Northern part and a tentative overview on Neogene rotations in the Aegean zone'. In Eds Dixon and Robertson: The Geological Evolution of the Eastern Mediterranean, J. Geol. Soc.fLondon, !J, 681-686. ----- KONDOPOULOU,D. 1985. 'Paleomagnetic study of the Tertiary intrusives from Chalkidiki (Northern Greece)'. Abstract from EUG III~ Terra Cognita, 5, 265. ----- LAJ ,C., JAMET,M., SOREL,D. & VALENTE,J.P. 1982. 'First lJaleomagnetic resul ts from Mio-Pliocene series of the Hellenic se_dimentary arc'. In: LE PICHON,X. AUGUSTITHIS,S.S. and MASCLE,J. (eds.). Geodynamics of the Hellenic Arc and Trench. Tectonophysics, ~~ ,45-67 • -- --- LEFEVRE,C., BELLON,H. and POISSON,A. 1983. 'Presence de leucititites dans Ie volcanisme pliocene de la region d' Isparta (Taurides occidentales, Turquie)'. C.R. Acad. Sci. Paris, 297, 367-372. LEMEILLE,F. 1977. 'Etudes neotectoniques en Grece ~~~d-orientale (Eubee centrale, Attique, Beotie, Locride) et dans les Sporades du nord (~ile de Skyros)'. These 3eme cycle, Ors ay, 17 3pp. LE PICHON,X. and ANGELIER,J. 1979. 'The Hellenic Arc and Trench system: a key to the neotectonic evolution of the Eastern Mediterranean area'. Tectonophysics, ~~, 1-42. LE PICHON,X. and ANGELIER,J. 1982. 'The Aegean Sea'. Phil. Trans. R. Soc. Lond. A, ~QO, 357-372. LETOUZEY,J. and TREMOLIERES,P. 1980. 'Paleostress fields around the Mediterranean since the Mesozoic derived from microtectonics: comparisons with plate tectonic data'. Mem. B.R.G.M., 115,261-273. MACDONALD,W.D. 1980. 'Net tectonic rotation, apparent te~t~nic rotation and the structural tilt correction in paleomagnetic studies'. J. Geophys. Res. , ~ ~, 3659-3669. -- Mc KENZIE,D. 1972. 'Active tectonics of the Mediterranean Region'. Geophys. J.R. Astr. Soc., )9, 109-185. 157 Mc KENZIE,D. 1978. 'Active tectonics of the Alpine Himalayan belt: the Aegean Sea and surrounding regions'. Geophys. ::G...~Astr. Soc., 12' 217-254. MERCIER,J.L. 1976. 'La neotectonique, ses methodes et ses buts. Un exemple: I' arc egeen (Medi terranee orientale)'. Rev. Geol. Dyn. Geogr. Phys., ¥Y~~J, 323-346. MERCIER,J~L:-1979. 'Signification neotectonique de l'Arc Egeen. Une revue des idees'. Rev. Geol. Dyn. Geogr. Phys., ~1, 5-15. MERCIER,J.L., DELIBASSIS,M., GAUTHIER,A., JARRIGE,J.J., LEMEILLE,F., PHILIP,H., SEBRIER,M. & SOREL,D. 1979. 'La neotectonique de l'Arc Egeen'. Rev. Geol. Dyn. Geogr. Phys., 11, 61-72. OLIVET,J.L., BONNIN,J., BEUZART,P. and AUZENDE,J.M. 1982. 'Cinematique des plaques et paleogeographie: une revue'. Bull. Soc. Geol. Fr. (7), XXIV, 2~2, 875-892. PATRIAT,P-.~ SEGOULIN,J., SCHLICH,R., GOSLIN,J., AUZENDE,J.M., BEUZART, P., BONNIN,J. and OLIVET,J.L. 1982. 'Les mouvements relatifs de l'Inde, de l'Afrique et de l'Eurasie'. Bull. Soc. Geol. Fr. (7), XXIV, l, 363-373. POISSON,A. 1977. 'Recherches geologiques dans les Taurides occidentales (Turquie)'. These d'Etat, Orsay, 795pp. POISSON,A., AKAY,E., CRAVATTE,J., MULLER,C. and UYSAL,S. 1983. 'Donnees nouvelles sur la chronologie de mise en place des nappes d' Antalya au centre de l'angle d'Isparta (Taurides occidentales, Turquie)'. C. R. Acad. Sci. Paris, 296,923-926. -- POISSON,~9~Th~=~xtension of the Ionian trough into Southwestern Turkey'. In Eds Dixon and Robertson: The Geological Evolution of Eastern Mediterranean, J. Geol. Soc. of London, 17, 241-250. ROTSTEIN,Y. and KAFKA,A.L-.-1~'Seismotectonics ;i the Southern boundary of Anatolia, Eastern Mediterranean region: subduction, collision and arc jumping'. ::G... Geophys. Res., ?Z, B9, 7694-7706. $ENGOR,A.M.C. and YILMAZ,Y. 1981. 'Tethian evolution of Turkey: a plate tectonic approach'. Tectonophysics, 12, 181-242. $ENGOR,A.M.C., GORUR,N. and SAROGLU,F.-1985. 'Strike-slip faulting and related basin formation in zones of tectonic escape. Turkey as a case study'. S.E.P.M., Spec. Pub. In press. TAPPONNIER,P. 1977. 'Evolution tectonique du systeme alpin en Medi terranee: poinconnement et ecrasement rigide-plastique'. Bull. Soc. Geol. Fr., XIX-3, 437-460. TAUXE,L.~ESSE:j: & LABRECQUE,J.L. 1983. 'Palaeolatitudes from DSDP Leg 73 sediments cores implications for the apparent polar wander path for Africa during the late Mesozoic and Cenozoic'. Geophys. J. R. Astr. Soc., 73, 315-324. ----- VERGELY ,~P: 1984. 'Tectonique des ophiolites dans les Hellenides internes (de forma tions, me tamorphismes e t phenomenes sedimentai res). Consequences sur l'evolution des regions Tethysiennes occidentales'. These d'Etat. Unpubl. Univers. Orsay, France, 411pp. VIlMINOT,J.C. and ROBERT,U. 1974. 'A propos des relations entre Ie volcanisme et la tectonique en Mer Egee'. C.R.Acad. Sci. Paris, ~]~, 2099-2102. AN APPROACH TO THE ORIGIN OF YOUNG VOLCANIC ROCKS OF WESTERN TURKEY Y.Yllmaz Istanbul Teknik tiniversitesi, Maden Fakultesi, Jeoloji Muh. Bolumu Ma~ka-Istanbul Turkey ABSTRACT. Young volcanism (± Post-Oligocene) in Western Turkey has evolved under a strong tectonic control. North-south shortening in the region lasted until the beginning of the late Miocene (± Tortonian) and was accommodated by an A-type subduction. This zone of A-type subduction reached into sub-crustal depths, and led to liberation of H20, Si02 , K20, LIL and radioactive isotopes. These liquids affected the overlying mantle, depressed its melting point and forced it to melt locally. They generated intermediate hybrid volcanic rocks of diverse composition. Due to the crustal shortening, anatectic granites formed and began rising. Magma formed farther down in the mantle was contaminated as it rose through the lower crust. The hybrid volcanism gradually waned after the late Miocene and was replaced by alkaline basaltic magmas as north-south shortening was replaced by north-south extension. This new volcanism had its source in a partly metasomatized mantle and produced two main types of lavas. Some of the basalts stemming from the metasomatized mantle became enriched in mobile elements much like the hybrid andesites. Those formed from 'normal' mantle gave rise to typical rift-type basalts. These two types of lavas evolved coevally in the same places in Western Turkey. 1. INTRODUCTION The young (i. e. ± Post-Oligocene) magmatic rocks of Western Anatolia (Asia Minor) (Fig.l) and neighbouring regions have been one of the most frequently studied topics by both domestic and foreign researchers. Such studies commenced towards the end of the last century and progressively increased to the present[l to 25]. Most of these studies are geochemical and based on analyses of limited sample populations collected from a small region. Studies based on detailed geological mapping, establishment of the local volcanic stratigraphy leading to the delineation of volcanic episodes and volcanic/sedimentary rock relation- ships and the analysis of volcanism by establishing the petrological character of individual volcanic episodes are surprisingly few. 159 A. M. C. s.engor led.), Tectonic Evolution o/the Tethyan Region, 159-189. © 1989 by Kluwer Academic Publishers. 160 Fig. 1. Distribution of young volcanic and plutonic rocks in Western Anatolia. Numbers indicate the following volcanic and plutonic centres (for the explanation of each centre see table 1) 1: ~anakkale-Bayrami~, 2: Biga, 3: Gonen, 4: Susurluk, 5: Ezine, 6: Gtilplnar, 7: Ayvaclk, 8: Edremit-Burhaniye-Korucu, 9: Ballkesir-Bigadi~, 10: Ayvallk, 11: Ber- gama, 12: Dikili-~andarll, 13: Menemen-Fo~a-izmir, 14: ~e§me-Karaburun, 15: Urla-Cumaovasl-Seferihisar, 16: Soke, 17: Dursunbey-Orhaneli, 18: Tav§anll, 19: Simav, 20: Gordes, 21: Selendi, 22: U§ak, 23: Kula, Gl: Salarya Granite, G2: Eybek Granite, G3: Kozak Granite, G4: Ala~am Granite, G5: Egrigoz Granite, MM=Menderes Massif. 161 Virtually no study in this region so far has considered the evolution of volcanism in its geological habitat by looking at local structural and lithological 'environment'. That is why many of these studies reached conclusions from a limited perspective. This paper was written with the primary purpose of promoting a multidisciplinary approach to Aegean magmatic evolution by proposing a broad-based model to be tested. The writer has synthesized his own data and ideas with the observations of others. In the following sections firstly the tectonic evolution of the area is summarized, then the regional distribution and petrological characteristics of the volcanic rocks are presented. A third section briefly presents the available relevant geochemical data. In the discussion section the origin of the magmatic rocks is discussed in the light of the above-data. A speculative petrological model summarizing the writer's thoughts on the evolution of the western Anatolian magmatism concludes this work. 2. TECTONIC EVOLUTION OF WESTERN ANATOLIA The late Cretaceous marked the beginning of convergent regime at all fronts in Turkey (Fig. 2)[4~ and was particularly charecterised by the emplacement of ophiolite nappes (Fig. 2A). These nappes moved onto extensive carbonate platforms that began subsiding "en masse" with the onset of obduction (Fig. 2.A -B-C). The metamorphism of the Menderes Massif may in part be ascribed to this obduction event (Fig. 2C-D). The regionwide metamorphism of the massif may have initiated as a result of the descent of Tauride-Anatolide platform[49] into progressively hotter regions. Thus, Rb/Sr isotopic data show a spread of ages beginning with 60 m. y • [50] . The Menderes Massif is one of the main tectonic element of Western Anatolia (Fig. 1). It is roughly elliptical in shape with a north- east-southwest trending long axis and is located between the Izmir-Anka- ra suture zone in the north and the Lycian nappes in the south (Fig. 2D-E) [49] . The massif has three major lithological units (Fig. 2E1). It has a gneissic core (Fig. 2E;M1), a schist (Fig. 2E;M2), and fhen a marble envelope (Fig. 2E;Ml ) tha~ overly the core successfully[50]. The late major stage of metamorphic events which is called "main Menderes metamorphism" by $engor et. al. [50] occurred around 35 ± 5 which coincides to the Oligocene (Fig. 2D). Following the obduction of the ophiolitic nappes onto the Tauride-- Anatolide platform during early Eocene, the Pontides collided with the Tauride-Anatolide platform which then began to be internally imbricated during the Eocene. This corresponds to the further burial of the central Anatolian crystalline massifs (The Menderes and Klr~ehir Massifs) under the advancing Lycian nappes[49] (Fig. 2C). . During the late Eocene to early Miocene interval the north-south convergence continued and the Menderes Massif (the western portion of 162 the Anatolides; [49]) was initially uplifted and then unroofed (Fig. 2D). The Oligocene then unconformably covered the Menderes metamorphic rocks. As the tightening continued, imbrication became more intense, and, therefore, increasingly more continental material was underlain by the massif which eased its uplift and caused its reheating (Fig. 2E). Under the over-thickened load of the imbricated piles deep crustal melting occurred (Fig. 2E). This is evidenced by the granitic intrusives that yield the radiometric ages of about 30 ± 5 m. y. [51 ; 46]. Continued uplift under ongoing north-south shortening resulted in some retro- charriage development that is indicated in the Menderes Massif by ductile, shear zones of Miocene age[49]. Rise of the Henderes Massif began about 15 m. y. after its burial (Fig. 2C-D) and, continued until the extensional regime of Western Anatolia at Tortonian (10,5 m.y.). Synchronously with the uplift of the Menderes Massif, south-directed thrust progradation continued in the Lycian Taurus and north-directed retrocharriage in Western Pontides and the Sakarya continent (Fig. 2B-C-E2)[49,52]. As a result of the increasing north-south tightening, the structure in Western Turkey is dominated by composite nappe systems that become progressively younger in the direction of transport (Fig. 2E)). In this regional overview, the Menderes Massif represents a southerry thrust parautochton (Fig. 2E2) and is in a tectonic window on which the higher nappes rest (Fig. 2El ). As the convergence continued the suture zone overturned and the over- riding plate began inserting into the ductile metamorphics as a wedge (Fig. 2E2 , Fig. 9)[49,53]. This was expressed at the surface as a late retrocharriage (Fig. 2E2 , Fig. 9). This wedge possibly initiated a separation between the upper and lower continental crust (Fig. 9) which propagates to the surface in the south as listric thrusts (Fig. 2E ) that involve only the upper part of the crust and eventually root into the hot metamorphic core complex (Fig. E2 , Fig. 9). The active examples of which may be expressed as the low-velocity zones in the cores of orogens as is the case beneath the Central Alps[54]. Synchronously with this, in the upper crust, successive thrust piles were formed e.g. the late Eocene, Oligocene and the post-Burdigalian thrusting episodes in the Lycian nappes (Fig. 2E?1. The lower crust hRS been removed by an A-type subduction (Fig. 2E, -Fig. 9)[49]. Early Hiocene was the time of final emplacement of the Lycian nappes. The Miocene thrusts represent the latest stage of crustal imbrication with a minimum thrust transport of 100 km[55] The average crustal thickness of the over thickened crust as a resul t of the post-Eocene north-south intracontinental shortening, is estimated to have been over 60 km [50] before the compressional regime gave way to extension. The Aegean extensional regime began in the Middle Miocene [56] (Fig. 2F) as a result of the consecutive geological events which was triggered initially in Eastern Anatolia due to the Arabia/Anatolia collision [49,57,58]. Following the collision, the Anatolian plate formed 163 wi th the development of the northern and eastern Anatolian transform faul ts. The Anatolian plate bounded between these two f aul ts began to move from the points of convergence westwards into the Aegean because of its buoyancy, in order to relieve the excessive on-roing compression in Eastern Anatolia. According to Dewey and :;>engor [59 and :;>engor[ 58] , the strong southwesterly bend of the north Anatolian fault zone, in the form of the Grecian shear zone [60] led to east-west compression in Western Turkey. This compression began to be progressively relieved by north-south extension which has governed most of the following tectonic evolution of Western Anatolia since Tortonian. The present crustal thickness of Western Turkey has been calculated to be about 30 to 40 km [53,61 and N.Canltez, personal communication, 1984]. On the other hand the total amount of extension has been estimated [60,62] to be in the order of 50 p.c. and 30 p.c. respectively, since the begining of the extension. Independent of these works, Akkok[63,64] estimated that at least 15 km of the envelope rocks of the Menderes Massif must have been removed from its outer parts, since its uplift (Fig. 2D-E-F). He based his estimate on the PIT regime affecting the core rocks of the metamorphic massif (M3 in Fig. 2El ) during the metamorphism, which crop out presently. All these views support one another to indicate that the crust acquired more than 60 km thickness before the extension. Under the extensional regime about ten east-west trending grabens formed in the Aegean region as the most prominent structural and morphological features (Fig. 2F). There is a great variation in the sizes and length to width ratios of the grabens. Their ages also vary greatly (for a summary of the avilable data on general structural and age relations of grabens see [53,65]. The extension has been accomodated mostly by a limited number of major east-west trending normal faults. Additionally, :;>engor[65] emphasized the importance of the thin-skinned variety of cross faults in taking up some of the extension and creating many structural complications in the whole region over which an irregular subsidence pattern was formed as a result. Occurrene ~f both strike-slip and normal (listric) faults are typical of the tectonic escape regime that have been recognized in Western Anatolia[66]. Under the extensional regime the thickness of the thickened continental lithosphere of the Anatolides (namely the Menderes Massif and the region that surrounds it) decreased greatly especially along the elongate depressions beneath which the lithosphere has ruptured. With the beginning of Tortonian the region which had suddenly been put into an extensional regime and, the thickened and partially melted (lower levels only) crust began stretching (Fig. 2E -E2) and created a horst-graben system within the brittle, thin lithosptere (Fig. 2F) (The present width of most of the Western Anatolian grabens varies between 4 and 20 km). 164 3. NEOGENE MAGMATIC EVOLUTION OF WESTERN ANATOLIA The Neogene geological evolution of Western Anatolia involves a wide- spread magmatism producing intrusive as well as extrusive rocks. The latter is much more extensive than the former (Fig. 1 and Table 1). The spatial and temporal relations of the volcanic rocks in the Turkish part of the Aegean system are shown in Fig. 1,3 and Table 1. The ages of these volcanics are established from stratigraphic data as well as same radiometric datings. It is seen in Fig. 1 that the volcanism started at the Late Oligocene(? )-Lower Miocene and continued upto the historical times with varying intensity from north to south (Table 1, Fig. 3). The volcanism that occurred in Thrace and, along the Bodrum (Dat~a) peninsula will be omitted here because we believe that they have their origin in processes different from that which happened in the region under consideration (such as subduction along the Hellenic trench). For the same reason, older magmatics of the Biga peninsula like those that developed during the Eocene or earlier will also be omitted. Extensive development of the young volcanic activity in Western Turkey began in the early Miocene (Fig. 3, Table 1). Exceptions to this, are possibly one or two doubtful earlier ages determined in the Biga peninsula. Three main types of magmatic rock grou~s can be readily distinguished in the region as suggested previously lIZ]. The first group is an intermediate to felsic variety of volcanic series. They will be referred to as the andesi tic suite in the following paragraphs. The second group is formed predominantly of the felsic varieties. This group will be referred to as the granitic suite. The third group is represented by the locally-developed basic lavas. The basic rocks will be referred to as the basaltic suite. The early manifestations of the volcanism are dominated by intermediate and acid members such as andesites, latites, dacites and rhyolites. Basaltic rocks in this period are totally absent. Rhyolites are much less developed as lavas than in the intermediate varieties, but the pyroclastics are much better developed. The basaltic volcanism is apparently late (Table 1 and Fig. 3) (later than the other two groups which are penecontemporaneous with each other). There may be little overlap between the beginning of the basaltic suite and the cessation of the others (Fig. 3 and Table 1). These conclusions, established through the stratigraphic relations of these rocks in the field are also supported by isotopic age determinations (Table 1). The andesitic suite contains a group of light-coloured and porphyritic rock varieties between which there is an apparent transition 165 ~IAP. s.c .• ~~ •• NB.NT •••••• .. .o •••• J 1, ........ A,. LAla Cretaceous . 0.0. ~JJJJ .... . r ........ .. ~ ...... . ........ .. AI: laic Cretaceous 1A.0.5. ~ ..... W1fl~~ B. Paleocene C. EOtRne O. L.te £oce"a ... OllgoClnl E,o Early MtoceMe F. Prescml Fig. 2. Schematized sequential cross sections showing late Mesozoic- Quarternary tectonic evolution of Wes,tern Anatolia. Discussion is in the text. Abbreviations: TAP=Tauride-An·atolide Platform, NBNT=Northern branch of Neo-Tethys[49 , Fig.6], SK=Sakarya Continent, SC=Subduction Complex (primarily an ophiolitic melange), OO=Ophiolite obduction onto the Taurus carbonate platform, IAOS=Izmir-Ankara Ophiolitic suture zone, MM=Menderes Massif, AZ=Anatectic zone, RM=Regional high grade (amphibolite facies) metamorphism in the Menderes Massif, Ml , M2, M 3=Major lithological envelopes of the Menderes Massif, GD=Grani tic dlapirs, Li=Lithosphere, As=Astenosphere, G=Graben, H=Horst, ATS=A-Type subduction, LN=Lycian nappes, MMB=Miocene molasse basins. Ta bl e 1. y , . ~ - an d 01 - - - - - - - - - _ . - - . _ - . - he ir r o ck . U , _ . an d ra di ar et ri c da te s in Io es t, R ef er en ce t o th e lo ca ti on s T im e/ St ra ti gr ap hy L it ho lo gy c it ed i n F io .l I L at a Pl io ce ne B as al t ~a na kk al e- L at e M io ce ne l(? ) B as al t Ba vr am i~ E ar lv M io ce ne A nd e- ' + -- ". ~' +- -~ ", ," ". ~. + - 2 L at e Pl io ce ne B as al t B ig a L at e M io ce ne (? ) B as al t E ar lv M in ce ne .~ ~- -. . -- ~. -, - rh vn li t.P 3 L at e Pl io ce ne B as al t Go ne n E ar ly P li oc en e( ?) A nd es ite E ar lv -L at e M io ce ne On rl .~ Ho t" H 4 L at e Pl io ce ne B as al t E ar ly P li oc en e( ?) A nd es ite Su su rl uk E ar lv -L at e M io ce no ' n ~ o . H _ h ,F F . n n 1 ~ ~ . + o 5 Pl io ce ne B as al t E zi ne La te l YI io ce ne 6 M id dl e M io ce ne A nd es it e- da d te -r hy ol it e G O lo ln ar E ar lv M io ce ne 7 La te P lio ce ne B as al t A yv ac lk M id dl e M io ce ne R hy ol i t e -i gn im br it e E ar lv M io ce ne A nd es 'te - H o B E ar ly P li oc en e A gg lo m cr a t e- tu f f Ed re m i t -B ur ha ni ye (a rl y JI lio ce ne O ae l t e- th yo ll te (la va -d om e- ig ni m br i te ) K or uc u L at e O liQ oc en e- Ea rlv M io ce ne A nd es i t e- tr ac hv an do .' + e -r i.- ; t o- to ,F F Qu at er na ry B as al t L at e P li oc en e 9 E ar ly P li oc en e( ?) R hy ol it e B al lk es ir - La te P lio ce ne Bi ga di ~ M id dl e M io ce ne A nd es i t e- la ti te -d ac i t e E ar ly l 'L io ce ne O ae l t e- th yo li te -t u ff L at e O li go ce ne -E ar l v M io ce ne A nd es i t e- tr ac hv an de si te -d ac i t e- tu ff A na to l· - - - . - - G eo ch em ic al A ff in it y A A r, A r. A r . A ro A CA A CA r . CA CA CA A CA CA -S H CA R ad io m et ri c Ag e (m .y .) 10 .2 16 17 .1 19 .3 ?1 < 0. 7 13 ,0 0 R ef er en ce l1 ,2 2, 27 ,2 B 12 ,2 4, 48 22 ,2 8 22 ,2 7, 29 22 ,3 0, 31 12 12 ,2 2, 30 14 ,1 7, 32 27 ,2 9, 31 ,3 3 I I I 0- , 0- , R ef p. re nc e to th e lo ca ti on s T im e/ St ra ti gr ap hy c it ed i n F ic .l La te M io ce ne 10 E ar ly I 'li oc en e A yv al lk L at e O lig oc en e- E ar ly M io ce ne 11 la te P li oc en e Be rg am a L at e Pl io ce ne I'I id dl e l'I io ce ne 12 L at e P li oc en e D ik il i - ~a nd a r II L at e fW lio ce ne P1 id dl e Pl io ce ne 13 M en em en -F oc a M id dl e M io ce ne tz m ir E ar lv P lio ce ne 14 M id dl e M io ce ne Ce sm e- K ar ab ur un E ar l v M io ce ne 15 M id dl e M io ce ne Ur la -C um ao va s~ - 5e fe ri hi sa r 16 L at e P lio ce ne St ik a E ar ly P li oc en e- la te M io ce ne P1 io ce ne (?l 17 la te P lio ce ne O ur su nb ey - L at e M io ce ne -M id dl e Pl io ce ne O rh an el i IB P li oc en e T av ,a nl l L at . PI io ce ne (? )- P1 id d1 e Pl io ce ne L ith ol og y G eo ch em ic al A ff in it y B as al t A G ra ni te , r hy ol i t e- da ci te CA An de .. i te -d ac it e rA B as al t A O ac it e- rh yo li te ( l av a- tu fF ) CA A nd es ite (l av a- py ro cl as t) CA B as al t A A nd es i t e( la va -p yr oc la st ) CA A B as al t- ha w ai i t e- rr ug ea ri te SH L at it e la ti U c a n de s· te -r hv nl · t e CA A nd es i t e- la ti ti c a n de si te -d ac i t e- rh yo da ci te CA B as al t- ha w ai i t e- nu ge ar i t e- tr ac hy te - A A lk al ir hv ol i t e B as al t A T ra ch yb as a1 t A A nd es it e- la U ti c a n de s· te -l at i t e- da ci te CA B as al t A A nd es i t e- da ci ts -r hy ol i t e CA B as al t A A nd es ite ( 1 av a- py ro cl as t) CA R ad io m et ri c Ag e (m .y .) 19 .5 ,1 9. 8 20 .3 ,2 0. 8 ?1 _R 16 .7 ,1 7. 3, 17 .6 - 1a 1 0 ? 1 0 ~ 15 .7 19 .0 16 ,5 -1 7 . 0 ?l oS 11 ,5 -1 2 17 ,1 7. 3, 18 .2 19 .2 _7 1. 1 11 .3 ,1 1. 9, 12 .5 6. 99 R ef er en ce 17 ,2 7 , 29 ,3 4 , 35 12 ,2 2, 27 ,2 9, 36 12 ,2 2, 26 ,2 7, 31 ,3 7 12 ,2 0, 21 ,2 4, 38 12 ,3 1, 37 11 ,1 2, 15 ,2 2, 39 ,4 0 27 ,4 1, 42 22 ,2 7 24 I '" - . J R ef er en ce t o th e lo ca ti on s T im e/ St ra ti gr ap hy c it ed i n L it ho lo gy F io .l 19 La te -l 'IJ .d dl e Pl io ce ne Tr ac hy ba sa l t -h aw .i i t e- m ug ea ri te -r hy ol i t e Si m au L at e- M id dl e l'I J.o ce ne A nd es i t e- da ci te -r hy ol i t e 20 L at e l'I J.o ce ne D ac it ... . r hy ol i t e- rh yo da ci te G or de s C ua te m ar y B as al t- ha w ai i t e 21 L at e Pl io ce ne G ra ni te (r hy ol iti c la va -p yr oc la st ) Se le nd i E ar ly P li oc en e I'I J.d dl e l'I J.o ce ne A nd es i t e- da ci te -r hv ol i t e 22 L at e P li oc en e B as al t ~a k La te P lio ce ne A nd es i t e- tr ac hu an de si te -d ac i t e- rh yo da ci te I'I J.d dl e l'I J.o ce ne R hy ol it e ( la va -t uf n Qu a t er n ar y- R ec en t B as al t- ha w ai i t e -t ep hr i t e -t u ff 23 K ul a P li oc en e( ?) A nd es ite 61 Sa la ry a gr an it e Ea rly -l' IJ .o ce ne G ra no di or ite L at e O liQ oc en e 62 Ey be k G ra ni te L at e O lig oc en e G ra na di or i t e- qu ar tz -l" 1o nz on i t e , G ra ni te 63 G ra na di or i t e- m on zo ni te -G r a n i t e K oz ak G ra ni te fY lio ce ne 64 Al ac ;a m G ra ni te E ar ly I 'li oc en e M on zo gr an i t e- m on zo n! te L at e O lio oc en e S en o m o n zo n ra n i t e 65 E~ ri gj jz G ra ni te E ar ly M io ce ne [fI on zo gr an i t e G eo ch em ic al R ad io m et ri c A ff in it y Ag e (m .y .) II eA eA A CA eA 8. 3 A 12 ,2 eA eA 10 .0 00 -1 2. 00 0 20 0/ 30 0. 00 0 Y A 1.1 1 7. 51 1 CA 23 .5 CA 20 .3 35 ,3 3, 30 .5 CA 24 .2 ,2 3. 9 23 .6 24 .0 eA 25 .2 20 .2 1f i. 0 eA 20 .3 eA A (?) 18 .0 1 75 .1 20 eA (A ?) 24 .6 21 .2 R ef er en ce 22 ,2 7, 32 ,4 3 19 ,2 7 22 ,2 7 22 ,2 7, 31 ,4 4 1, 12 ,2 4, 31 ,4 5 46 ,4 7 17 ,4 6, 47 ,4 8 46 ,4 7 46 46 a- - 00 Agi My. Plio. 4 :--! I 7 I , I ~ 9 Ie; I..J 10 I \I I 12 loll"";' I~ zl:;; ~ I'" 14 loll i I~ UI 1& 0 1 17 COOSlol Rigion ~ o " ", Sl~' 01 I L..-...!I- 01 I' I : o I" o I" ~..J : 11m .• KOt. Skz. I: 1- m DIk.Mld. I :r-t-~ 6 tlzm..Kor. . 16 I: :: ... ..!!.!:. ~ 16 I 6 I 6 I I' ....!.J I: I' I 2: 19 6 I 6 ~ t : • -I I >- 20 : I I I ";: 21 • I I o 6 I 6 I I ~ 22 I 6 169 I n Ion d Kulo [J Blks. [!] Em!. [!] I 23 I ~ 24~ __ ~ __ • __ ~C~on~.~E~d~r.~ ________________ -L ________________ -J Fig. 3. Radiometric ages and geochemical affinities of the young volcanic rocks of the Aegean regions. For the data see the references cited in Table 1. Abbreviations (Numbers in bracket indicate the locations cited in Table 1). ~an-Ez=~anakkale-Ezine (1,5-6), Pts:Patmus, Sis-Bdr-K§: Samos "Sisam"-Bodrum-Ka§. Izm-Kar: Izmir - Karaburun(13, 14) , Skz:Khios "Saklz", Dik-Mid:Dikili-lesvos "Midilli" (6,7,10), ~an-Edr:~a­ nakkale-Edremit(1,5,6,7,8), Blks=Ballkesir(9,10), Kula:Kula(23), U§k:U§ak(22), Emt(Emet,Ktitahya). Symbols= 6: Mainl y calc-alkalic (and also shoshoni tic) volcanic suites . • : Alkalic volcanic suites. and continuity. They occur in the field commonly as flow breccias, agglomerates and tuffs. Rhythmic ejection of tefras and lavas were possibly produced by Strombolian type eruptions. The lava flows form block lavas; i.e., lava flows which are covered with angular fragments, and have a hummocky surface. They display some flow laminations due to shear failure of the lava. In the porphyritic rocks, the phenocrysts vary between 25-50 p. c. They occur either as seperate crystals or glomeroporphyric; cumulopurphyric clumps. The grain size of the phenocrysts appears to display steady repetitions suggesting repetitive changes in the crystallization conditions. 170 Among the phenocrysts, plagioclase is the most ubiquitous. They are zoned-crystals and are highly variable in composition due to complex zoning. The composition ranges from An60 to AnZS and AnSO to An20 for basic and acidic varieties of the andesitlc suite respectlvely. Oscillatory zoning is characteristic. Normal-continuous zoning is restricted commonly to the centre and the edge of the crystals. In the andesitic suite pyroxene is commonly represented by augite and enstatite. The latter, although rarely, persists into the dacite. In the more felsic varieties of this group hornblende occurs more often as the mafic mineral. It coexists with augite in the basic andesite. Geological history of the andesitic suite forming widespread stratovolcanoes can be interpreted as recording the complex history of a continuum of magma erupted from a reservoir which were replenished by new magma addition. The evolution from basic andesite to dacite and rhyolite probably occurred as a result of magmatic differentiation. The ejectas usually become more differentiated with time during construction of stratovolcanoes. The granitic suite contains granitic stocks, ring dyke complexes; i.e., radial and concentric dykes, rhyolitic lava domes, obsidian and froth flows, ignimbrite cooling units, pumiceous or vitritic (and ash-fall) tuffs and welded tuffs. They all appear to be associated with each other in space and time, and may be connected to a shallow level granitic intrusion in an association of a caldera collapse or resurgent doming. Such spatial and temporal connections distinguish the rhyolitic rocks of the granitic suite from similar volcanic rocks of the other suites, particularly of the andesitic suite. The granitic and associated rocks that occur around the ~anakkale region (namely Ezine-Ayvaclk and the surroundings), in the northern part of Western Anatolia, is a particularly good site to observe close genetical connections of the above mentioned rock varieties. Almost all of the members of the granitic suite are hydrothermally altered to various degrees. In some cases the whole rock appears to have thoroughly rotted so that it may disintegrate easily. The chief agent of such alteration was probably high-temperature water, related possibly to late "paulopost" magmatic process attacking the feldspars, the chief constituent minerals of this group. The granitic suite members are typically light-coloured rocks as they are almost devoid of mafic minerals. The granites, sensu-stricto, of this suite display commonly granular texture. Porphyritic varieties are also commonly present. Aphyric granitic rocks resemble the micro- granitic rock varieties. The vitritic granitic varieties and the acidic lavas exhibit striking conchoidal fractures. Such irregular fractures give the rock resinous rather than vitreous lustre and result in perlitic structure in obsidian. Eutaxitic texture is very distinctive in the ignimbrite. T ab le 2 . A ve ra ge c he m ic al c o m po si ti on o f yo un g v o lc an ic s u it es i n W es te rn A na to li a - - C al c- A lk al ic a n d Sh os ho ni ti c A ss oc ia ti on s A lk a] ic A ss oc ia ti on s ~ - - N or m al lIi Bh K -C A Sh os ho ni ti c V ol at il e R ic h N or m a] CA A nd es ] t e B as al ti c A nd es it e L at h e D ac it e R hy ol it e Sh os ho ni ti c Sh os ho ni te T ra ch y- H aw ai it e O li vi ne T ep hr it e A nd es it e B as al t ba sa lt ba sa lt - - Si 0 2 58 ,4 0 54 ,7 () 60 ,7 4 59 ,0 0 64 ,0 5 72 ,4 9 50 ,h 4 5~ ,3 2 47 ,3 8 51 ,1 0 45 ,9 0 51 ,8 0 - - - T i0 2 0, 72 0, 72 0, 59 0, 71 0, 55 0, 11 1, 20 0, 95 1, 27 1, 45 2, 05 1, 95 A1 20 3 16 ,8 3 16 ,4 6 ]6 ,2 5 ]6 ,0 5 16 ,0 5 14 ,5 1 ]5 ,] 2 15 ,6 3 17 ,2 5 ]6 ,B O 16 ,9 3 ]4 ,9 0 Fe 20 3 2, 59 4, 39 2, B 5 2, 74 2, 33 0, 75 4, 08 4, 15 4, 55 3, 05 5, 40 4, 68 Fe O 3, 05 2, 56 3, 02 3, 0] 1, 70 0, 35 4, 28 2, 92 4, 26 5, 50 4, 96 4, 22 - M nO 0, 13 0, 14 O ,I l I 0, 09 0, 09 0, 05 0, 14 0, 13 0, 15 0, 13 0, 18 0, 15 Mg O 3, 55 4, 29 3, 22 1 3, 75 2, 05 0, 30 6, 52 0, 24 7, 05 4, 92 B ,5 0 7, 61 Ca O 7, 10 7, 04 5, 26 5, 09 4, 35 1, 30 8, 46 9, 08 B, B4 7, 48 9, 50 8, 6B Na 20 3, 61 3, 51 3, 34 3, 25 3, 40 3, 55 3, 24 3, 05 3, 10 4, 21 4, 45 3, 22 1: 2 0 1, 33 3, 12 3, 10 3, 01 3, 20 3, 05 3, 03 2, 57 2, 85 2, 75 1, 20 1, 20 11 2 0 2, 29 2, 38 2, 25 I 2, B 2 2, 24 2, 88 2, 52 1, 75 2, 72 2, 10 0, 70 1, 03 I - - CO 2 0, 40 0, 80 0, 25 i 0, 50 0, 75 0, 70 0, 68 0, 52 0, 88 0, 68 0, 30 0, 70 T ot al 10 0, 00 10 0, 06 JO O ,9 5 i j( )O ,02 10 0, 76 JO O ,0 3 99 ,9 1 10 0, 01 10 0, 30 10 0, 17 99 ,7 2 10 0, 14 - - - - - - - - - - - - - - - - - - - - - :::i 172 The basal tic suite is much less common when compared with the andesitic suite. The rocks of this suite occur as locally developed patches and preferentially follow some distinc zones of fracture or major fault directions. Although, the members of this suite look alike in the field, with their dark colour and fine-grained texture, two subgroubs may be differentiated among them petrographically. They are; olivine basalt-tephrite or phonolite series in one line, and trachy basalt (kulaite hawaiite-mugaerite (and alkali rhyolite in some rare cases) series in the other line. Between the two series, there is no distinct temporal or spatial difference. The olivine basalt series is composed mainly of plagioclase (An6S_80 ) or nepheline, olivine and clinopyroxene (titanium rich augite) as pnenocrysts. The trachy basalt-mugaerite series contains hornblende as the more common mafic phenocryst rather than clinopyroxene or olivine. The plagioclase (AnSO_7S ) is almost always present. Texture of the basaltic suite varies greatly between glassy and microcrystalline forms. Almost every gradation between these two extreme members is present. In the matrix, intergranular, pilotaxitic, and trachytic textures occur as more common varieties. 4. GEOCHEMISTRY Amongst the young volcanic rock series of '~estern Turkey, geochemically three groups of rocks may be distinguished as calc-alkaline, shoshonitic and alkaline. The early phase of the volcanic activity appears to belong to calc-alkaline and shoshonitic groups. The acidic volcanic rocks of the same phase which are genetically related to the hypabyssal granitic rocks also fall into the calc-alkaline group. The alkaline rocks are predominantly represented by the basaltic lavas and have been formed later than the calc-alkaline volcanic suites. It is seen in Fig. 3 and Table 1 that there is a marked change in the geochemical nature of the volcanism by the time. It corresponds roughly to the Ii tho logical changes from the andesi tic suite to the basaltic suite. The general geochemical features of the main volcanic suites will be summarised in the following paragraphs. The granitic suite shows many similar geochemical features to a crustal-derived post-orogenic, epizonal granite. Therefore it will not be treated here as an independent geochemical group. 4.1. The Andesitic Suite The andesi tic suite shows a wide geochemical variation in which three subgroups may be distinguished depending on their major element contents as normal calc-alkaline, high-K calc-alkaline and shoshonitic (Table 2). 173 Amongst these rocks however, there appears to be no clear and distinct difference in petrographical as well as other geochemical characters. Beside, such groupings are not reflected in the spatial and temporal relations of these rocks. The shoshonites too show an apparent continuity with the calc-alkaline rocks [67]. Almost all of the major elements that these rocks have are overlapping in content. Therefore in the Aegean case calc-alkaline and shoshonitic subgroups appear nothing more than an arbitrary division. Therefore it is best to collect hem all under one continuous rock association and treat them collectively as one large group. The main geochemical characteristics of this group may be summarized as follows; 1- They display a wide variation in Si02 from 55 p.c. in basaltic andesite up to 73 p.c. in the rhyolite. 2- K20/Si02 also or close to (Fig. 4). shows a large variation but most of data fall wi thin high-K calc-alkaline field in Si02 vs K20 diagram 10 RHYCLIT E --/ '/ TRACHYTE ,., \ I ~ __ ) LATITE and SHOSHONITIC M>d / /" HAWAIITE ALKALI BASALT~/ / ' / l /: ; I I / o. /"-/ .. (,! ... / CAC I TE ANCESI TE LOW-K THOLEIITE I LDW-K ANDESITE DACITE I 40 45 50 53 55 56 60 63 65 68 70 75 77 BO OfoSi02 Fig. 4. K ° vs SiO plot for the young volcanic rocks (Redrawn after Sava§~ln[671 ~ ASbreviations: Hi-K.Dacite=high-K.dacite, Hi-K.A: High-K.andesite, HBA=Highly calc-alkalic bazaltic andesite, SL:shoshonitic latite, ST:Shoshonitic trachyte, CR:calc-alkalic rhyolite. Symbols indicate regions of the volcanic rocks (For the locations, and data See Fig. 1 and Table 1); o=izmir-Karaburun, Fo~a, Menemen, e=ganak- kale, Dikili, Lesvos, Ayvallk, Urla, x=Ezine, *=Kula. 174 3- They have high content of alkali elements. Hajority of the inter- mediate rocks of this group have total alkali varying from 5 to 8 p.c. 4- H20 content is high and reaches in some rare cases upto 6 p. c. H20+C02 varies generally between 2.50 and 4.00 p.c. 5- Ti02 content is generally low and is mostly less than 0.72 p.c. Ti group elements such as Zr, Hf, Nb and Ta are also in low concentration. But they show slight increase from moderate to high-K calc-alkaline rocks. 6- They are mostly hypersthene and quartz-normative. 7- Lack of an iron enrichment trend in the group is significant. FeO/MgO ratios are largely variable. However some unexpectedly high MgO and CaO values are rarely seen. 8- LIL elements (Rb, Ba, Sr, Cs) are high. Positive correlation between Si02 and Rb, Ba and increasing alkalinity is noted. Sr increase is higher with respect to the others[67]. 9- Ba values and REE do not show a linear increase from intermediate rocks of calc-alkaline series to the alkaline series. 10- K/Rb ratio (varies between 250-300) compares more favourably with the values of continental crust. K/Rb show a negative slope with increasing alkalinity indicating that Rb increase is more than K20 and Sr, so that K/Rb decreases while Rb/Sr ratio increases. 11- The Th group elements; U, Th and Pb contents are high (higher than those of the island arc associations and are closer to the continental crust values [67]. Th/U ratio approaches to 4. Heavy element concentrations are in an order that, Th-Pb (25-40 ppm»U (5-10 ppm) [27] so that Th/U and Pb/U ratios increase with increasing K. The elements cited above correlate positively with Si02 . Heavy REE, and Y contents increase less with silica and potash. 12- Unexpectedly, the compatible elements in this group, such as Ni (160 ppm), Co (40 ppm), V (150-250 ppm) are slightly higher than those of average values found in common calc-alkaline rocks. 13- Absolute abundances of REE (105-l26)[27J are also higher than those in average calc-alkaline rocks but compare more favourably with high-K calc-alkaline and anorogenic andesitic rocks. The REE pattern of the andesitic suite clearly indicates fractination. This is seen more clearly in the light REE when compared with the heavy REE (Fig .5). From the normal calc-alkaline to the shoshonitic rocks, light REE increase steadily while heavy REE change little[26]. Eu depletion 175 in the REE pattern may be ascribed to the incorporation of this element into the plagioclase lattice. 200 E 0.100 0. ~ 50 '- "0 C 0 .I:. U 10 -------------..-- E 0. 0. -'" u 0 cc La Ce Nd Sm Eu Tb Vb Lu Fig. 5. Chondrite normalized REE pattern of hi!i\h-K calc-alkalic rocks (the Andesitic suite) (After Innocenti et al. [26J, Fig. 6B). 14- Relative to REE, Rb, Ba, Sr, and values Cs show an increase and form a posi ti ve slope while those of Nb, Ti and Zr on the other hand decrease and form a negative trend. 15- La (rock)/La (chondrites) ratios of the rocks of varying silica content in the andesitic suite are displayed in Table 3. The ratio varies between 150 and 400. The ratio vs. SiO is high and thus compares favourably with the values derive~ from anorogenic andesitic rocks. La/Yb, and REE do not show a positive correlation with increasing silica contents. Table 1. Young volcanic and plutonic centres, their rock associations, and radiometric dates in Western Anatolia La rock/ La chondrite Bigadi~(9) 110 Bigadi~(9) 150 Ayvallk(lO) 150 Ayvallk(lO) 190 Gordes(20) 210 (numbers in bracket refer to the locations in Fig.l and Table 1). (Values are taken from Ercan et.al. [27] 68 66 60 52 67 176 16- The 87Sr /86Sr ratio in the andesitic suite varies between 0,703 and 0,706[27]. It is noted that dacite and rhyolite of this group have distinctly lower isotope ratios than the similar rocks of the granitic suite in which such ratios vary between 0,710 and 0,720. 4.2. The Basaltic Suite The basaltic suite as a whole is alkaline in character. Still, two fairly distinc subgroubs may be distinguished in this suite. Of these, one subgroup is volatile and K20-rich when compared with ordinary basaltic rocks. The other subgroup on the other hand is more normal for an average basaltic suite. Such a division corresponds roughly with the olivine basalt-tephrite (or phonolite) series and the trachy-basalt hawaiite-mugaerite series. Most of the rocks within the volatile-normal subgroup have higher Ti than the other subgroup. In this distinction, the low-Ti basaltic rocks show many similar or transitional geochemical features to the shoshonitic and calc-alkalic intermediate rocks. In the volatile rich basaltic rocks, the volatiles reach up to 6 p. c. and, K20 varies between 2 and 3.5 p. c.. Their dependent elements such as, Sr, Ba, Rb and Cs are also higher than those usually found in ordinary basaltic rocks. The main chemical difference between the two basaltic subgroups may also be reflected in their isotope abundances. 87Sr /86Sr ratio of the olivine basalt-phonolite series is markedly lower (ranges between 0,702-0,703) than the rocks of trachy-basalt-hawaiite-mugaerite series (ranges between 0,703-0,705) [27]. It suggests that the two subgroubs may have different histories of generation. The olivine basalt subgroup is olivine-normative. Normative nepheline also occurs commonly in these rocks. The trachy basalt- mugaeri te series on the other hand is hypersthene-normative, and as reflected by normative quartz, tends towards the more acidic end members such as mugaerite, near the silica saturation. The rock series belonging to the basaltic suite are seen to have followed two different crystallization paths; namely the Kennedy trend and Coombs trend of evolution as defined by Miyashiro [68]. In this distinction, the olivine basalts and associated rocks belong to the former, the trachy basalt-mugaerite series to the latter. 5. DISCUSSION The widespread intermediate volcanic rock association, in which andesites predominate together with latite and more felsic varieties such as dacite, is the main product of the initial phase of volcanic activity that lasted to the end of the Miocene. 177 BalLa, Th/La, La/Nb ratios of this suite appear to be high when compared with the mantle derived andesitic magmas (i.e. for a quartz eclogi te or peridotite source rocks, or POAM "plagioclase-oli vine- augite-magnetite" fractionation of a basaltic magma) [69]. It is evident that the magma from which the andesitic suite of rocks formed was apparently enriched in Si02 , K20, K-group of elements, LIL elements, REE, and radiogenic isotopes. This is further supported when the andesitic suite average is plotted on the diagram of CeN/YbN vs CeN (Fig. 6). It plots close to the line indicating an amphlbohte type source rather than peridotitic or eclogitic sources. 251~-; " oq' o 20 10 o . ~:\\ct t\J~\c,.r, OA. -c p.~.-- 55. 1.0 --Fractional crystallization .. /' ...... _.-1-._.""._._.-._,- '" I/J 90 20 I/J 60 80 Ce N Fig. 6. The Andesitic suite average (.) plotted on CeN/YbN vs CeN diagram (In Gill[69], Fig.S.17). The Subcript N means concentrations are normalized to chondri tic values (Ce=O,86S ppm, Yb=O,22 ppm) Numbered tickmarks and segments OA indicate percent fusion where melts are andesitic in major element compositions. The data outlined above suggest collectively a hybrid nature for the andesi tic suite. Most of these characteristics are also common to magmas erupted through a thick continental crust and may arise in a variety of environments and conditions. For this, at least two processes may be envisaged. These are; a) Mixing between basaltic and silicic magmas in the mantle or in lower continental crust. b) Interaction of basaltic magmas with silicic crustal rocks, i.e. assimilation. A brief summary of the possible mechanisms which may lead to the hybrid nature of the andesitic suite, are now briefly outlined. a) Mixing of magmas: There are at least a number of possibilities of mixing of the magmas to produce an andesite in a situation 178 similar to the geological evolution of Western Turkey where the mantle had been enriched in certain elements (e.g. silica, LIL elements, Sr isotopes) prior to the mixing of magmas. (al ) aqueous solution or siliceous magma mixes with the low melting fraction of the mantle (mantle wedge) to yield an intermediate magma. fusion of contaminated mantle: Uppermost mantle beneath thick continents is enriched in LIL elements especially U, Pb, Rb, Sr and light-heavy REE [69]. Consequently partial fusion of such mantle might yield magma with high Sr87 /Sr86 ratio, possibly higher than 0,705. But if such a zone of mantle (peridotite) fuses in the presence of the siliceous fluid, it dissolves the fluid and peridotite derived basalt plus slab derived liquid (an anatectic melt) mix at high pressure to form andesitic or even rhyolitic compositions. This whole thing may be seen as a single stage process. Mixing via crust (assimilation of anatectic melts or metamorphic liquid-water rich- within the lower crust): Binary mixing of basaltic magmas and anatectic melts in the lower crustal environments under high pressure may also yield intermediate compositions. Slab derived fluid or silicate melts are potential agents of extensive metasomatism in the mantle. The eruption of high-Ti alkali basalts after the andesitic intermediate rock varieties but simultaneously with the basic magmas of low equilibration temperatures indicate a fairly heterogeneous source region in the mantle. A significant portion of such heterogeneous mantle volumes is thought to have been modified. Modified mantle may yield andesites and related magmas upon partial melting. (b) Crustal Contamination: Another group of mechanism in the genesis of intermediate compositions involves crustal contamination of mantle derived magma by assimilation of or interaction with crustal rocks. The ideas originally belong to such workers as Daly [70], and Wager and Deer [71]. As in a case of mixing of different magmas mechanisms such as diffusion and isotopic exchange r2] may also permit selective contamination by incompatible elements and radiogenic isotopes. The crustal interaction apparently is selective, affecting isotopic ratios more than element concentrations, because fractionation or assimation alone cannot account for the geochemical budgets of andesites in detail [73] . Following the practise utilised to test a similar case in New Zealand (See Fig. 7) the data derived from the calc-alkaline volcanic rocks of diverse compositions in the Aegean region were plotted in the 179 ratio/ratio diagrams (Fig. 7,8). It is seen in these diagrams that the data, although scattered across a large area appear to be clustered in three crude groups. Of these, one extreme group is located around the area of the rocks of sedimentary origin while the other extreme group overlaps with the basaltic rocks leaving a space for the third group in an intermediate position. These roughly may be interpreted to be the result of binary mixing of basaltic magma and the sialic material. On the Sr 87/Sr 86 vs K/Rb diagram (Fig. 8), the centrally located points crudely define a line which may also be interpreted as a mixture of the basaltic magma and sialic material. 0.708 L- .Y> '" - 0.706 L- !elfl 0.704 500 e6le ~e' B G ,./ 400 61 61 S61 / / e • I· • / / / . /61 61. • ••• 300 200 K / Rb Fig. 7. Plot of the Andesitic suite (Ell) on 87 Sr /86Sr vs K/Rb ratio diagram. For comparison, New Zealand volcanic association indicating the mixing (Ewart and Stipp[74]) are also added. B, S, and I=Average Basalt, Sediment and Ignimbrite values respectively from the New Zealand samples: ~:basalt, o:Andesite from New Zealand. 0.708 ~ 0.706 -L- _ r-lfl '" 0.704 -- -- -- 61 ~--G _ e e III III W 61- - e 5 • 0.4 0.8 1.2 1.6 67Rbl ~r Fig. 8. Comparison plot to Fig. 6 using 86Sr a denominator common to both axes. Straight line and broken line are pesudoisochrons for the New Zealand volcanics and the Andesitic suite respectively. Symbols are the same as in Fig. 7. 180 The possibilities listed above are difficult to test for two major reasons. First, because of the nature of the problems concerned, and secondly due to the regional data presently available. (a) Uncertainties related to the nature of the problems: A method to distinguish different processes leading to the sediment involvement is difficult to establish at the present state of knowledge. It may be due to some direct process e.g. slab recycling (mixing) or crustal contamination of magmas during ascent or to some indirect processes; via mantle metasomatism. The agents of mass transfer to enrich the mantle resulting in suites which converge on andesite with time may be fluid mixing (a silica rich liquid or an anatectic melt) or subsolidus metasomatism or velnlng. Because final products of these processes supposedly bring about similar results and the distinction between mass transfer agents is not easily made in practise. However, the distinction in this case may not be critical for model of petrogenesis. Still few criteria may be used to distinguish between the effects of mantle metasomatism versus fluid mixing. Of these one such criteria involves in addition to petrological evidence, the spatial and temporal limits of the distinctive volcanic phases. Because fluid mixing involves buoyant materials over shorter time intervals whereas metasomatism is a more permanent feature around the region of slab/mantle interaction. Therefore distinctions between the two processes rely more heavily on major geological situations. (b) Problems related to the present data available on the young volcanics of the region: The data at hand are not yet sufficient to make a proper approach to test the possibilities stated above. Most of the previous works about the young volcanic rocks appear to have been sporadic rather than systematic. Most of them are geochemically-oriented works without much stratigraphic or field control. The geological map base and related detailed petrological, petrographical, and mineralogical data are not adequate. The iso- topic approaches have not been problem-oriented. Thus, in the light of the data cited above, interpretations must remain non unique. Therefore to distinguish between the effects of crustal and/or mantle level magma mlxlng versus crustal level assimilation, different models may be offered. All of the models discussed above may have contributed to some degree to the generation of the present situation. In the following paragraphs, in the light of the available data and within the geological evolutionary framework of the region I will offer a new interpratation. 6. MODEL PROPOSED The Aegean sector of Western Turkey, following the continental collision 181 in late Paleocene-Eocene time, began to be progressively squeezed under a north-south compressional regime until the middle Miocene (Tortonian). As a result of this continuing convergence, the continental crust was shortened and thickened to an excessive value (over 60 km). Further convergence is thought to be taken up mostly by continental under- thursting (A-type subduction whose evidence is outlined above and discussed in detail elsewhere ~9,50]. The late stage metamorphism of the Menderes Massif coincides roughly with this period. Anatectic melts were generated in the lower levels of the crust (Fig. 2E) at this time. The granitic intrusions rose into shallow levels (See the radiometric dating on the young granites of the region in Table 1) where they formed stocks, dykes and hypabyssal intrusions. Partly penecontemporaneous with these events the first phase of young volcanic activity began in the region and, widespread intermediate (the andesite suite) volcanic rocks of hybrid nature were formed as a result [75]. These simultaneous events show clearly that the andesitic magma ascended through the potentially wide zone of high grade meta- morphism and anatexis at the lower levels in the crust (Fig. 9). The geological data show that this volcanic activity occurred in a wide area and is remote from any true contemporary subduction zone. ----- 80-=--"-- - ------ -----100 - Fig. 9. Schematic, speculative cross section of Western Anatolia during the later stage of compressional regime (early-middle Miocene) showing geological conditions in the upper mantle and lower crust during generation of magmas of diverse origin, and explaining the genesis of the hybrid magmas. UCC=Upper continental crust; LCC=Lower continental crust; DS=Probable surface of detachment. Vertically ruled area=Metasomatised mantle, Dotted area=Anatectic Zone, Block patch=Where silica-rich liquid or melt released from the down-going lithospheric slab. Line SP l-SP 2=Water-- saturated peridotite (taken from Gill [69] , Fig; 8,2). 182 Fig. 9 is a schematic cross section of the region between the Oligocene and early-middle Miocene, when it was under the later stage of the compressional regime. An A-type subduction zone accomodating the major north-south convergence is seen in the figure. The heavy lines underneath the tightened and thickened continental crust and underthrusting lithosphere represent the stability limits and dehydration reactions of some of the hydrous phases such as amphibole, phlogopite, and serpentine. It is seen from the diagram that along the under-thrusting of the lithosphere, slab dehydration is expected to occur following a few kms of under thrusting, across the lines by hydration reactions possibly within SO to 125 kms. Such dehydration can release copious volumes of water-rich fluid or aqueous solution. Assuming that some carbonates are present within the slab, the fluid released will lie within H20-O-C02 (possibly CL) system. At high pressures and temperatures such fluins are effective solvents. Elements with high volatility or water-solubility such as K, Pb and U will also be dissolved [76 J. The elements as refractory as the REE, partition strongly into the fluid relative to eclogite solids D7J. Consequently, the fluid released during the under-thrusting will be rich in incompatible trace elements, radiogenic Sr, possibly Pb and silica. The geochemical distinctiveness of this event is attested to by the volatile contents and ratios, the relative enrichment of K-group, Th group elements, and the depletion of the Ti group elements with respect to the REE. Even the low geothermal gradients along the major thrust boundary representing the A-type subduction is expected to produce a few percent of such fluid. The underthrust lithosphere, if it contains clay (illite)-rich components provides high concentrations of LIL elements notably Ba and Pb. Partial melts of rock groups containing rocks of such composition will have higher Ba/La ratio and higher Sr87 /Sr86 (and possi bl y Pb 207/Pb 204) as was the case in the andesitic suite of rocks. Normal to the underthrust plane the gradient would be high. The slab derived liquid or siliceous melt may rise into the hotter mantle wedge where it becomes superheated. Greatest effect of the fluid introduced to the mantle is to promote its melting preferably near the zone of underthrusting, because, high water content of the fluid leads to lower temperatures of melting per degree of fusion. Addition of water may cause vapour-saturation, thereby reducing solidus temperatures by up to a few hundred degrees, and increase the percentage of melt present. The liquid mixes with the mantle (peridotite)-derived basalt or simply reacts with peridotite. Mixing dilutes the hydrous, siliceous melt containing less hydrous basalt and produces a wide range of inter- mediate comRositions with appropriate temperature and water content DS,79J. In the mixed melt, Ni and Cr contents would be abnormal because distribution coefficients for olivine and pyroxene will be higher. Unspecified titanifereous refractory phase in the depleted mantle may be the answer of distinctly low concentrations of Ti group 183 elements in the high-K intermediate volcanic rocks of the region. During the ascent this magma affected the continental crust through which it passed and was itself also affected by it. The main influence of the andesitic magma on the crust is thought to be the increase of the thermal gradient and the promotion of crustal anatexis at progressively shallower levels. The sialic crust on the other hand affects the andesitic magma as a density filter and promotes the differentiation by retarding ascent. This contributes to the higher silica mode of rocks as attested to by the great variety of rocks from basic andesite to the rhyolite. When andesi tic magma ascended through it, thick continental crust may have permitted the magma to be selectively contaminated, in particular by radiogenic isotopes. This may be the reason why the isotopic ratios of the andesitic suite scatters so widely (Fig. 7 and 8). The first phase of volcanism gradually died out towards the end of Miocene and was replaced by alkaline basalts. This alkaline volcanism continued into the Quaternary in Western Turkey. We believe that our model also accounts for the conspicious absence of alkaline basaltic volcanism during the early development of the Western Anatolian system during which calc-alkaline volcanism instead was widespread. As a matter of fact the development of the andesitic suite lasted to the end of the compressional regime and even extended to the beginning stage of the extensional regime. The reason for this is possibly that, at the initial phase of the continental stretching the condition that led to the andesite genesis and to crustal contamination prevailed until the mixed(:) melt in the mantle and the remnant anatectic melts in the deeper crust were depleted. After the depletion of the melts no more mixing could take place due to the termination of the underthrusting related phenomena. The remarkable absence of alkaline basalts during the early evolution of the Western Anatolian rift system is thus not because of the absence of the melts, but because these melts were then mixed before they reached the surf ace. The proposed mixing process may also be important in the Tibetian type environment which was also the case during the Plio-Quaternary evolution of Eastern Anatolia [80]. In the Aegean region, during the more advanced stage of rifting in the Pliocene and Quaternary, under the subhorizontal flow regime affecting the subcontinental crust, alkaline basalt melts were liberated from the underlying mantle, possibly because of rifting-induced pressure release (much like beneath active rifts where mantle has played a passive role in rifting; e.g. the volcanism of the upper Rheine Graben). Consequently the extension related magmatism assumed its normal alkaline character. The rift induced alkaline basaltic volcanism also shows wider 184 diversity in geochemical variations. We think that it is related to the heterogeneous source region which has in part been metasomatised prior to the opening of the rift due to the reasons outlined above. The evidence includes eruption of high-Ti alkali basalts simultaneously with the low-Ti alkali basalts, the kulaite of low equilibration temperature in which the amphibole more than clinopyroxene being the stable form of Fe-Mg minerals, the carbonate-rich xenoliths dragged along the volcanics [21]. They imply collectively that at least some significant portion of the mantle (mantle wedge) was strongly and more permanently modified. 7. CONCLUDING SUMMARY In Western Anatolia, the neotectonc episode commence in the middle Miocene with the cessation of the north-south compressional regime which had begun with the complete elimination of the neo-Tethyan ocean floor during the late Creteceous and had continued as a post-collisional convergence causing shortening and thickening of the continental crust to an excessive level. The neotectonic episode began when north-south extensional regime took over the role of the former compressional regime. In the region, the young volcanic activity began during the early Miocene and has continued almost uninterruptedly into the historical times. The early products of the volcanism are commonly rocks of intermediate composition. They are partly co-eval and postdate shallow level granitic intrusives and their equivalent extrusive rocks. Extensive development of the intermediate volcanic suite occurred during the late Miocene. This episode died out at the beginning of Pliocene and was replaced by a new volcanism producing basic lavas and associated volcanic rocks. This late volcanic stage was locally developed during the formation of Western Anatolian rifting (graben) and used the young fracture system to reach the surface. Three petrographical groups can be differentiated among the young volcanic products of Western Turkey. They are; an andesitic suite, a granitic suite, and a basaltic suite. The andesitic suite comprises true andesite, latite, dacite and rare rhyolite. The granitic suite contains shallow level granitic intrusives and their equivalent extrusive products such as rhyolitic domes, ignimbiri tic cooling units, welded tuffs. The basic suite was developed in two independent, but temporally and spatially indistinct groups. These are; a) olivine basalt-phonolite series, and b) Trachy basalt-hawaiite-mugaerite series. The granitic magma with its isotopic signatures was derived from the continental crust. The andesitic suite, on the other hand is hybrid in nature as it was enriched selectively in elements such as Si02 , K20 volatiles, REE, LIL elements and radiogenic isotopes. Of the two suti- groups of the basic suite, the weakly-alkaline trachybasal t-mugaeri te series, shows similarities to the andesite suite as it also shows 185 enrichments in similar elements. The olivine basalt-phonolite suite, on the other hand were crystallized from a typically mantle-derived rift type basaltic magma. The contrasting characters of two co-eval basic magmas are assumed to be due to the mantle he terogenities that formed through the metasomatism that occurred during the continental shortening. ACKNOWLEDGMENTS Dr.A.M.C.gengor contributed to this paper significantly by critical reading of the manuscript. His contribution is heartily acknowledged. Dr.M.Oz~elik did the proof reading, to him I extend my thanks. REFERENCES 1- H.S.Washington, 'The volcanoes of the Kula basin in Lydia'. Ph.D. dissertation, Phil.Fak.Univ.Leipzig. New York, 1893. 2. H. S. Washington, 'On the basalts of Kula'. Amer. J . Sci., 48, 114, 1894. 3. H.S.Washington, 'The composition of Kulaite'. J.Geol., 8, 610, 1910. 4- E .Andra, 'Erupti vgesteine im Nordwestern Kleinasiens' dissertation, Phil.Fak.Univ.Leipzig, 1905. 5. A.Philipson, 'Das Vulkangebiet von Kula in lydien, die Katakekaumene der Alte'. Pet.Geogr.Mitt., 1, 237, 1913. 6. H.Holzer, 'Beyce 54/4 ve Simav 71/2 paftalarlnln jeolojik loveleri', M.T.A. Rep., 2366, 1913. 7. A.Kalafat~log~'Karaburun yarlmadaslnln jeolojisi', Bull. M.T.A., 56, 53, 1961. 8. M.Aslaner, 'Etude geologique et petrographique de la region d'Edremit-Havran (Turquie)', M.T.A. Spec. Publ., 110, 1965. 9. J.D.Becker-Platen, 'Lithostratigraphische Untersuchungen im Kanozoikum Stidwest-Anatoliens (Ttirkei) (Kanozoikum und Braunkohlen der Ttirkei, 2)', Bundesanst. Bodenforch, Geol.Jahr., Berlin, 97, 1, 1970. = 10. J.D.Becker-Platen, C.Besong, W.Harre, H.Kreuzer and 'Kalium-Argon Alter des Afyon vulkanismus (Anatolien) Datierung der Miozan-Pliozangrenze: Datierungsbericht Zwisch.Bundes.Stalt Bodenforch. Hannover, 2/71, 1971. P .Muller, und die 2/71- ' 11. F.Innocenti and R.Mazzuoli, 'Petrology of the izmir-Karaburun volcanic area', Bull.Volc., 36, 83, 1972. 12. J.Borsi, G.Ferrara, F.Innocenti and R.Mazzuoli, 'Geochronology and petrology of recent volcanism in the Eastern Aegean sea (W.Anatolia, Leshos)', Bull. Volc., 36, 473, 1972. 13. J.Keller and L.Villiari, 'Rhyolitic Ignimbrites in the region of Afyon (Central Anatolia)', Bull.Volc., 6, 1, 1972. 14. T.Ongtir, 'Dikili-Bergama jeotermal ara9tlrma sahaslna ili9kin jeolo- ji raporu f , M.T.A. Rep., 5444, 1972. 15. i.O~gen~, 'izmir-Cumaovasl bolgesi perlit 01u9umlarlnln jeo10jisi', TUBITAK V. Bil. Kong. Tebl. Kit., 261, 1975. 186 16. R.Ota and A.Din 187 34. a.Dora and Y.Sava9~ln, 'Alibey-Maden Adalarl (Ayvallk ) BOlgesi Mag- matizmasl' TBTAK VII. Bilim Kongr. Matematik Fen Grubu Bildirileri, 1900. 35. R.D.Krushensky, R.Ak~ay and E.Karaege, 'Geology of an Area East of Edremi t, Biga Peninsula, Northwestern Turkey', U. S. A. Dept. Int. Geol. Surv. Rep. (Int.Rep), T.U-25, 1971. 36. B.AkyUrek and Y.Soysal, 'Biga yarlmadasl gUneyinin (Sava§tepe-Klrk- aga~-Bergama-Ayvallk) temel jeoloji ozellikleri', Bull.M. T .A., 95/96, 1, 1982. 37. M.Fytikas, F. Innocenti , P.Maretti, R.Mazzuoli, A.Pecerillo and L. Villiari, 'Tertian to Quaternary evolution of volcanism In the Aegean region', in: The Geological Evolution of the Mediterranean, Eds. by J.E.Dixon and A.H.F.Robertson, 1984. 38. Y.Sava§~ln and O.O.Dora, 'Fo~a-Menemen yoresi volkanitlerinde pirok- senlerin yaYlllml ve kristalografik degerleri' TUrkiye Jeol.Kur.BUlt., 20, 21, 1977. 39. I.Ozgen~, 'Cumaovasl (izmir) asit volkanitlerde saptanan iki extrUz- yon a§amaSl araslndaki goreli ya§ ili§kisi' TUrkiye Jeol. Kur. BUlt., 21, 31, 1978. 40. M.Akartuna, 'izmir-Torball-Seferihisar-Urla bolgesinin jeo1ojik etU- dU', istanbul Univ.Fac.Sci.Mon.Ser., 18, 1962. 41. V. YUkse1, 'Soke-Germencik bo1gesinin jeo1ojisi ve jeoterma1 ener ji olanak1arl', M.T.A. Rep., 4677, 1971. 42. T.Ercan and E.GUnay, 'Soke-yoresindeki Tersiyer vo1kanizmasl ve bo1- gesel yaYlllml', Jeomorf. Derg. Ankara, 10, 117, 1981. 43. N. Akdeniz and N .Konak, 'Menderes Masifinin Simav dolaYlndaki kaya birim1eri ve metabazik, metau1tramafik kaya1ann konumu', TUrkiye Jeol. Kur. BUlt., 22, 175, 1979. 44. E.Bingo1, 'Muratdagl jeo1ojisi ve ana kaya~ birim1erinin petroloji- si', TUrkiye Jeol. Kur. BU1t., 20, 13, 1977. 45. T.Ercan and O.Oztunall, 'Kula volkanizmaslnln ozel1ik1eri ve i~erdigi 'Base Surge' tabakalanma §ekilleri', TUrkiye Jeol.Kur.BUlt., 25, 117, 1982. 46. E.Bingol, M.Delaloye and G .Ataman, 'Granitic Intrusion in Western Anatolia; a contribution to the geodynamic study of this area', Eclog. Geol. Hevl., 75/2, 437, 1982. 47. G.Ataman, 'P1utonisme --ca1co-a1ka1in d'age alpin en Anato1ia du Nord-Quest', C.R.Acad.Sci. Paris, 280, 2065, 1975. 48. Y.BUrkUt, 'Kuzeybatl Anado1u'da yera1an plUton1arln mukayese1i jene- tik etUdU', dissertation, Istanbul Teknik Univ., 1966. 49. A.M.C.$engor and Y.Yllmaz, 'Tethyan evolution of Turkey: A plate tectonic approach', Tectonophysics, 11, 181, 1981. 50. A.M.C.$engor, M.Satlr and R.Akkok, 'Timing of tectonic events in the Menderes massif, Western Turkey: implications for tectonic evolution and evidence for Pan-African basement in Turkey', Tectonics, 1, 693, 1984. 51. E.Bingol, 'Evolution Geotectonique de l'Anatolie de l'Quest', Bull. Soc. Geol. France, Ser.7/18, 235, 1976. 52. Y.Yllmaz, 'Sakarya kltasl gUney kenarlnln tektonik evrimi', istanbul Yerbi1im1eri, 1/1-2, 33, 1981. 188 53. A.M.C.;>engor, 'Ege'nin neotektonik evnmlnl yoneten etkenler', in. Batl Anadolu'nun gen~ tektonigi ve volkanizmasl paneli, TUrkiye Jeol.Kur.Yay., Ankara, 1982. 54. S.MUller, R.Egloff and J.Asorge, 'Structur des tieferen. Untergrundes entlang der Schweizer Geotraverse', Schweiz. Min. pet. Mitt., 56,685, 1976. 55. M.Gutnic, O.Monod, A.Poisson and F.D.Dumont, 'Geologie des Taurides occidentales (Turquie), , Mem. Soc. Geol. France, N.S., 58, 1979. 56. V.Jacobshagen, St.DUrr, F.Kockel, K.O.Kopp and ~.Kowalczyk, 'Structure and Geodynamic evolution of the Aegean Region', in. Eds. H.Closs, D.Roeder and K.Schmidt, Alps. Apennines, Hellenides, Int.Uni.Comm.Geodyn.Sci.Rep.38, Schweizerbartische Verlag's buchhandlung, Stuttgart, 1978. 57. A.M.C.$engor and W.S.F.Kidd, 'Post-collisional tectonics of the Turkish-Iranian Plateau and a comparison wi th Tibet' , Tectonophysics. 55, 361, 1979. 58. A.M.C.$engor, TJFUrkiye'nin neotektoniginin esaslarl', TUrkiye Jeol.Kur.Konf.Ser .• 2, 1980. 59. J.F.Dewey and A.M.C.$engor, 'Aegean surrounding regions: complex mul tiplate and continuum tectonics in a convergent zone', Geol.Soc.Amer.Bull., 90, pt.I, 84, 1979. 60. A.M.C.$engor, 'The North Anatolian transform fault: its age, offset and tectonic significance', J.Geol.Soc.London, 13, 269, 1979. 61. X.Le Pichon and J.Angelier, 'The Aegean Sea' ,Phi1.Trans.Roy.Soc.- London., A.300, 357, 1981. 62. D.P.Mc K~, 'Active tectonics of the Mediterranean Region', J.R.Astron.Soc. 30, 109, 1972. 63. R.Akkok, 'Menderes Masifinin gnayslarlnda ve 9istlerinde metamorfiz- rna ko9u11arl, Alagehir, Manisa', TUrkiye Jeol.Kur.BUlt., 24, 11, 1981. 64. R.Akkok, 'Structural and metamorphic evolution of the the northern part of the Menderes Massif: New data from the Derbent area and their implication for the tectonics of the massif'. J .Geol., 91, 342, 1983. 65. A .M.C. $engor, 'Cross faul ts and differential stretching of hanging walls in region of low-angle normal faulting: Tectonic implications with examples from Western Turkey', in. Extansional Tectonics, Eds. by. J.F.Dewey and P.L.Hancock, Geol.Soc.Spec.Publ., 28, 575, 1987. 66. K.Burke, A.M.C.$engor, 'Tectonic escape in the evolution of the continental crust', Reflection seismology: The Continental Crust, Geodynamics Series, 14, 41, 1986. 67. Y.Sava9~ln, 'Batl Anadolu Neojen magmatizmaslnln yaplsal ve petro- grafik ogeleri' in. Batl Anadolu'nun gen~ tektonigi ve volkanizmasl paneli, TUrkiye Jeol.Kur. Yay., Ankara, 1982. 68. A .Miyashiro, 'Nature of alkalic volcanic rock series' , Contrib.Mineral Petrol., 66, 91, 1978. 69. J.B.Gill, 'Orogenic Andesites and Plate Tctonics', Springer-Verlag, New York, 1985. 70. R.A.Daily, 'Igneous rocks and the depths of the Earth', Mc Craw-Hill, New York, 1933. 189 71. L.R.Wager and W.A.Deer, 'The Petrology of the Skaergaard intrusion, Kangerdlugssang, East Greenland', Medd. Groenl., 105/4, 1939. 72. R.J .Pankhurst, 'Strontium isotope studies relat~ petrogenesis of the Caledonian basic igneous province of NE Scotland', J.Petrol., 10, 115, 1969. 73. H.P.Jr.Taylor, 'The effects of assimilation of country rocks by magmas on 180 / 160 and 87 Sr /8 6S r sy stematics in igneous rocks', Earth Planet Sci. lett., 47, 243, 1980. 74. A.Ewart and J.J.Stipp, 'Petrogenesis of the volcanic rocks of the Central North Island, New Zealand as indicated by a study of Sr 87 / 86 ratios and Sr, Rb, K, U and Th abundances', Geochim. Cosmochim Acta., 32,699, 1968. 75. Y.Yllmaz and A.M.C.~engor, 'Ege'de kabukevrimi ve neo-magmatizmanln kokeni', TUrkiye Jeol.Kur.Bil.Oz., 1982. 76. C.R.Stern and P.J.Wyllie, 'Phase compositions through crystallization intervals in basalt-andesite-H20 at 30 kb with implications for subduction zone magmas', Am.Min., 63, 641, 1978. 77. B. O. My sen, 'Trace el ement partitioning between garnet peridoti te minerals and water-rich vapor: experimental data from 5 to 30 kbar' , Am.Min., 64, 274, 1979. 78. R.W.Kay, ~leutian magnesian andesites: melts from subducted Pacific Ocean crust', J.Volcanol.Geotherm.Res., 4, 117, 1978. 79. C.W.Burnham, 'Magmas and hydrotermal fluids' in. Geochemistry of hydrotermal ore deposits, Holt, Rinehart and Winston, 1978. 80. Y.Yllmaz, F.~aroglu and Y.GUner, 'Iniation of neovolcanism in the East Anatolia', in. Deep-Seated processes in collisional zone, eds. by M.L.Gupta and C.Froidevaux, Tectononophysics, 34, 177, 1987. TECTONIC EVOLUTION OF PALEOTETHYS IN THE CAUCASUS SECTOR OF THE MEDITERRANEAN BELT: BASIC PROBLEMS A.A.BELOV Geological Institute of the USSR Academy of Sciences Pyzhevsky per., 7 109017 Moscow USRR ABSTRACT. All mobilistic reconstructions indicate the oceanic Paleotethys, but the time of its opening and closing, size, time of existence, position between other paleotectonic elements are considered differently. Prejurassic ophiolite assemblages are known in the area of the Greater Caucasus and Dzirula salient of the Transcaucasus. The available data on occurrence of pre-Alpine ophiolites to the south of the Sevan-Akera and Vedi zones are not convincing. But at the same time they show considerable effect of Hercynian tectono-magmatic processes on the northern margin of Gondwana-Land. Application of paleo biogeographical method for the Caucasus encounter a number of obstacles. Paleomagnetic data on the whole confirm the mobilistic reconstructions, but they are still few. The paper critically considers the available paleoreconstructions, reveals the asymmetric structure of the Mediterranean belt and Paleotethys ocean. Northern part of the Caucasus was its active margin: in the Middle Paleozoic - of West Pacific type and in the Upper Paleozoic - of West Pacific type, and in the Upper Paleozoic - of Andean type, while the southern part of the Caucasus was its passive margin. At the end of Paleozoic - beginning of the Mesozoic this part was separated from Gondwana-Land and transformed into a microcontinent with formation of the Neotethys in its back part. The final closing of the Paleotethys on the Caucasus occurred before the Lias, its suture passes along the southern slope of the Greater Caucasus. INTRODUCTION Among mobilists having identical ideas there are numerous disagreements; so the development of the Caucasus, both in the Mesozoic and Paleozoic, is interpreted differently. Common for all mobilistic reconstructions is the recognition of the oceanic Paleotethys*), but the '~) Under Paleotethys is meant an ocean basin separating Gondwana-Land and Laurasia during the Paleozoic. At the start of its formation the rifting began somewhere at the end of Riphean; the basin 191 A. M. C. !Jengor (ed.), Tectonic Evolution o/the Tethyan Region, 191-216. © 1989 by Kluwer Academic Publishers. 192 time and mechanism of its origin" the extent and duration of its existence, the position between other paleotectonic elements and its time of closing are interpreted differently. All these testify to ambiguity of the interpretations and deficiency of original data. One more complicating factor is that the rocks composing the first, second and partly third layers of the recent oceanic bottom on the one hand, and ophiolite associations, on the other, are far from being identical. In the works published recently on the basis of thorough studies, the authors deny representation of ophiolitic associations as evidence of ancient oceans analogous to the present-day ones(1,2,3), and the ocean basins, as to their sedimentology, are regarded peculiar only to the Meso-Cenozoic stage of the Earth's development(4,S). As shown by S. V .Ruzhentsev(2) a distinct divide of basalts and sediments, i.e. the second and the first ocean layers, a steady substitution of shallow-water deposits by deep-water, up to abyssal ones, higher up the section, spatial and time disconnection of basalts and siliceous deposits, are not characteristic of paleooceans, as opposed to the recent oceans. Paleooceans and present-day oceans differ mainly in composition of sedimentary and volcanic rocks, their relationships in sections and on area, as well as in time of formation of basalts of the second oceanic layer. Mesotethys and Neotethys occupied an intermediate position in the sequence Paleoocean (Paleotethys, in particular) - Recent ocean, and Neotethys differed slightly from the present-day oceans in its formational filling(2). Considering the validity of the above conclusions, it should be said, however, that if we mean under oceans large basins with the oceanic type of crust, such differences between the present and ancient oceans as depth, morphology of the relief, chemical and physical properties of water masses, peculiarities of the inner structure and the corresponding geodynamic environment, presence or absence of certain types of rocks, or their differences, cannot be an obstacle for their correlation. The principle of actualism taken as a basis of our studies opened in Early Paleozoic after appearance of the oceanic crust, and its shortening began at the end of Silurian, being especially intense during two phases: Hercynian-Carboniferous and Indosinian the end of Triassic-Liassic. Along with the consumption of the ocean bottom along the strike of the Mediterranean belt, at the end of Devonian-beginning of Carboniferous continental rifting, and then spreading of the ocean bottom took place in some places. The newly-formed ocean it areas were connected with the remaining ones, forming Paleotethys II, contrary to Paleotethys I, whose greater part had closed in the middle of Carboniferous. Paleotethys I seems to be a continuous corridor between continental masses of north and south, whereas Paleotethys II - as a wedge-shaped westward tapering Panthalassa gulf inside Pangea 6 . 193 of the Earth's geological past, remains valid, as a whole, and the complicating events appear to be related to the irreversible evolution of geological processes in the Earth's history. Proceeding from these theoretical concepts, we shall analyze the development of the Caucasian sector of the Mediterranean belt. Fig.l: Distribution of the Caucasian pre-Alpide complexes. R - Rostov salient of the Ukrainian shield, D- Continuation of the Donbass, SP - Scythian platform. Pre-Alpide zones within the Alpide folded area: B - Bechasin, F - Forerange, M- Main Range, SS - Southern slope, T - Transcaucasian massif, I - Northern part of the Iranian plate. Regions of distribution of pre-Alpide complexes at the surface: 1 - Zangezur, 2 - Daralagez, 3 - Miskhan, 4 - Murguz, 5 - Loku, 6 - Khrami, 7 - Dzirula, 8 - Central Abkhazia, 9 - Svanetia, 10 - Amasya, 11 - Halka river upper course (Bechasin). Faul t zones: PT Pskekish Tyrnyaus, MT Main Thurst, KL Kakhetion-Lechkhum, SZ - Erzincan-Sevan. 194 Ophiolite associations of pre-Jurrassic mountain rocks*) have been known in several regions of the Caucasus (Fig.l): in the Caucasus in Hercynian zones - Bechasin (?), the Forerange and Range, and on the Dzirula salient in the Transcaucasia. A more description of them will be given below. Greater the Main detailed All attempts to prove the presence of pre-Alpine ophiolites in the south in Sevan-Akera, Vedi zones, and in the nearest regions beyond the Soviet boundaries cannot be considered successful. The available data are debatable and unconvincing. Direct data are missing. Thus, Gasanov(7) believes that the Sevan-Akera zone incorporates a gabbro-diabase-effusive complex of the Carboniferous age. His publication gives the only value - 291+3 Ma (after feldspars, K-Ar) and says that in the study there were analyzed 11 samples of gabbro-diabases, 3 samples of diabases and 10 samples of andesitic porphyrites; no analytical data have been presented, and it is not clear what it means: the average from 24 determinations of rocks varying in composition, and other field aspects. The paper also reports of 8 analyses of samples from dykes - pyroxenites and spessartites yielding the values of 322+2 Ma (K-Ar). No analytical data were presented either, the doubts remaining the same. The well-known value of the age of metamorphosed gabbroids of the Levchai massif (583 30 Ma, K-Ar after feldspars from gabbro-pegmatites(8)) could be also considered as a true one for the melanocratic basement of the Mesozoic ophiolite association. Nevertheless, it requires further proof. The detailed structural-petrographic studies of metamorphic rocks inside the serpentinitic melange in the Adjaris area in the Sevan-Akera zone enabled Dobrzhinetskaya and Ez(9) to distinguish among them fragments of rocks of a more ancient ophiolite association than the Mesozoic. Its true age, however, is not known. This association can be much more ancient, i. e. Precambrian, or somewhat more ancient - Early Mesozoic. The other group of researchers(lO) studying xenoliths of metamorphic rocks in gabbroids of the Mesozoic ophiolite complex of the Sevan-Akera zone, failed to recognize such elements among them that could be compared with more ancient Palaeozoic ophiolites. The studies carried out by scientists of Armenia(ll) within the Amasya region, where among Mesozoic ophiolites there are tectonic blocks of eclogitic amphibolites, showed that the latter had undergone two stages of metamorphism: early HT/HP and late LT/LP-HP stages. Relics of the first mineral association within the second one testify to a rapid lifting of the block from the deep zone in the Alpine thrusting. The isotopic dating yielded the following values: 80+5 Ma (after K-Ar) , this corresponding to the age of the second metamorphsim stage, and 330+42 >:,) The age of an effusive-sedimentary series is taken here for the spreading age of ophiolites. 195 Ha (Rb-Sr isochrone). The authors are apt to interpret the last values as the age of the protoliths of metamorphosed rocks. Noting the geochemical similarity of metabasites studied by them to basites of the oceanic islands, they believe that the former are representatitves of the Paleotethys ocanic crust. In this relation it should be noted that, first, under conditions of high pressures and temperatures (to_700o , P-8-9 kbar) of the first metamorphism stage the primary geochemical specificity of the original substratum could hardly remain preserved, and second, the measured system reflects most probably the time of metamorphism. Therefore I support the concepts- suggested by V.A.Agamalyan , and some other researchers that eclogitic (almandine) amphibolites of Amasya are detached masses of deep melanocratic horizons of the Precambrian continental crust of the Gondwana-Land. Even though the future studies will, nevertheless, confirm presence of Paleozoic ophiolites in the Erzincan-Sevan-AKera zone, the corresponding ocean basin can be attributed to Paleotethys II, considering the time of origination (early Carboniferous 7, absolutely no data on older age are available), or to Hesotethys, as to the time of its closing. Noteworthy is by the way, that ;>engor, Yllmaz and Sungurlu(13) support the ideas of Bergougnan and Parrot(14) and Tekeli(15), and present the corkesponding data on a possible opening of the ocean basin Karakaya since the Carboniferous. The reliable data indicate the Jurassic-Neocomian age of ophiolites in the Transcaucasia (Fig.2, X,XII). Very interesting and valuable Rb-Sr isochrones obtained by Agamalyan. et al. (12) for rocks of the crystalline basement of the Tsakhkunyats (Hiskhan) massif and Szomkheto-Karabakh zone. They determined the age of 620 and 300 mln. years respectively, not indicating as it seems to me, an affinity to a block situtated to the north of the Sevan-Akera suture i.e. being a part of the East-European plate. The future Sevan-Akera zone could be the southern boundary of intense influence of Hercynian tectono-magmatic processes on the Gondwana-Land margin. It is possible that the Early Carboniferous dykes recognized by T.Ab.Gasanov(7) as well as radiometric values 304+41 mln.y. and 306+40 mln.y. for ultrabasites and gabbroids of North-West Anatolia(16,17) and others given above, testify to deep tectono-magmatic processes proceeding along this zone at the end of the Hercynian epoch. Some manifestations of magmatism and weak dislocations of this time have been known to the south of the Erzincan-Sevan line (for instance, alkaline monzonitic granites of the Bitlis massif, Rb-Sr 325-351 mln.y. Yllmaz(18). They, together with red formations of the Verrucano type (for instance in the east of Central Taurus(19» testify to influence of Hercynian orogenic processes over the territory situated to the south of the boundary concerned. It is interesting, that along this boundary the K-Ar determinations of plagiogranite coincide almost exactly in the Kesis Dag massif (256 mln.y.) in the region of Erzincan(14), and in the Aparan massif of Armenia (255 mln.y.(20». , ~~ 't '" ~ • r.:J .. lSJ lI: ~Cj 4i' .. ~~~fffiB ~ 15 ... HJl- ~~"~ ~~~~rn DlIIiiI R-l f;il ,. : . ~~~[ij:, ~ .. ill III o? I::: ":.-:: fJOQ PI -. 197 Fig.2: Correlation of material formations of ophiolite belts and major pre-Alpine zones of the Caucasus. I - Precaucasus, II - Bechasin, III - Andryuk-Tokhana, IV-V - Forerange: IV - Allochthon of the island arc complex, V - Ophiolitic allochthon, VI - Main Range, VII - Svanetia, VIII-IX - Transcaucasian massif: VIII - Dzirula salient, IX - Khrami salient, X - Sevano-Akera ophiolite belt, XI - South Transcaucasia, XII - Vedi; 1 - groups of mostly terrestrial terrigenous formations with predominance of sandy-clayey rocks: a) grey, b) red and multi-coloured; 2 - terrestrial and coastal-marine terrigenous with predominance of coarse-clastic rocks; 3 terrestrial bauxite-bearing crusts of weathering; 4 group of shallow-water marine terrigenous and carbonate-terrigenous formations ( 50% carbonates); 5 group of shallow-water terrigenous-carbonate and carbonate formations ( 50% carbonates); 6 - reef limestones; 7 - pelagic limestones of moderate and considerable depths; 8 - flyschoid terrigenous and carbonate-terrigenous deposits of various bathymetry; 9 deep-water flyschoid; a) terrigenous, b) siliceous-terrigenous; 10 mostly deep-water black-shale; 11 deep-water clay-siliceous; 12 marine volcanogenic-sedimentary deposits with mostly acid tuffs 0__ various bathymetry; 13 calcareous-alkaline volcanics of mostly acid composition (rhyolites, dacites); 14 - calc-alcalike volcanics of mostly intermediate composition (andesites, andesite-basalts, in island-arc complexes sometimes together with tholeitic basalts of ~i8her potassium content); 15 volcanics of higher alkali contents of acid and intermediate composition (orthopyres, trachytes); 16 - low-potassium basal ts of oceanic type; 17 - alkaline basaltoids of the potassium series (shoshonitic); 18 group of volcanic formations of the contrasting (bimodal) association; 19 - gabbro-amphibolitic (3D layer of the oceanic crust); 20 - Alpine-type ultramaphics (fragments of the upper mantle); 21 - olistostromes, wild flysch; 22 - tectonic melange; 23 - lateral transitions between synchoronous formations; 24 - coal content in sedimentary formations; 25 unconformities: a) local angular, b) regional structural; 26 - tectonic contact of formations; 27 - metamorphic schists. The Palaeozoic (Precambrian?) ophiolites and other geosynclinal deposits of the Troya zone'~) in North-West Anatolia, including the Kazdag and Uludag massifs, are likely traces of Paleotethys I. Their joining with Paleozoic zones of the Greater Caucasus via the Black Sea is a difficult problem. However, the data on this region cannot characterize the western continuation of Mesozoic ophiolite zones of Asia Hinor and the Lesser Caucasus. In this region of North-West Anatolia the eugeosynclinal piles of flysch, diabases and radiolarites, dated by Brikmann(2l) as the Late Carboniferous-Permian, and referred to by Sh.A.Adamia and LD.Shavishvili(22), after the works by Bingol(25), Radelli(26) and Fourquin(27) should be attributed to Triassic. ,:,) The name is given after A.A.Belov(23); it is situated in the northern part of Sakarya continent after(24). 198 The Triassic system for the entire region under investigation has not been studied properly enough. Much new data on the Triassic have been given in the papers(13,24,59); however, their conclusions give rise to certain doubts(28), the most important of which is ascribing the pre-Ma1m ophiolites in the Central and East Pontian to the Palaeotethys suture. They can prove Mesozoic, and in such a case it would be more logical to regard them as evidence of the early rifting in origination of Neotethys*) i.e. rift structure with a short period of development, like the Karakaya structure in North-West Anato1ia closed at the end of Triassic or the end of Lias. The processes of opening the Neotethys can contribute to explanation of an extremely interesting communication by Solovking(29) of occurrence of a tectonic 1ense of Upper Triassic limestones and calcareous sandstones with peculiar detritus of basic magmatic rocks within the ophiolite zone of the Lesser Caucasus in the Akera river upper course. The pre-Ma1m ophiolites in the region of the Klire town and in some other localities in Central and East Pontian can indicate true Palaeozoic ocean floor. Yl1maz and $engor(30) lower the age of these ophiolites to the pre-Middle Jurassic, as they are crossed and metamorphosed by Dogger syn- and post-tectonic granites. These authors stretch the Palaeotethys suture from the East Pontian northeastward to the Dzirula salient of the Transcaucasian massif where it is cut by a left-latera fault, and find its continuation in the Greater Caucasus. In order to solve the problem of Palaeotethys traces westward of the Caucasus, it is extremely important to obtain a more reliable stratigraphic-palaeontological characteristic of the pre-Upper Jurassic oceanic and marginal complexes. At present all the four arguments given by Yl1maz and gengor(30) speak in favour of the pre-Jurassic age of ophiolites from the Kastamonu region, i.e. their attributing to the pre-Neotethys ocean basins is not undisputable. Moreover, in my view, the close space and time association of the Kastamonu ophiolite and the complexes related to the Karakaya pa1aeoocean basin suggest an idea of their common origin. The same concerns the region of the Bayburt town. The stratigraphy of the complexes of the supposed Palaeotethys ocean affinity in the contiguous region have been por1y studied too. The main reason is that there are no reliable data on indivisible Pa1aeozoic-Triassic-Jurassic sections. Either the lowermost parts of these sections have not been faunistica11y characterized, and, correspondingly, either the Palaeozoic lower part (Akgo1 suite in the Central Pontides), or the Triassic-Jurassic upper part (DFF Strandja) are inferred. The same can be said of the Nucu1ite1 element in Northern Dobrudja). Ultimately, a part of them may belong to Hercynian complexes, and another part - to the early Alpine ones, so their joining up may prove artificial. *) Under the Neotethys is meant an ocean basin formed through rifting from the end of Permian - beginning of Triassic and subsequent spreading of the oceanic bottom, and closed in the Cretaceous and Cainozoic. 199 One of the most curious regions from the view point of the Paleotethys suture position is the Bogrovdag ridge in North-East Iran to the southwest of the town of Resht(3l). Unfortunately, it has not been studied thoroughly enough either. The ophiolites occurring there can by age be attributed either to Precambrian or to Paleozoic. Their exact and comprehensive description is missing. The section lying above of Silurian, Devonian and Permian of the platform type common to Iran abounds in inter layers and piles of basic and median lavas and tuffs, from a few to 650 meters thick. The characteristic of this volcanism is not sufficient for a unique identification of its tectonic environment. Possible presence of true Paleozoic ophiolites and eugeosynclinal piles in this region will require an extra explanation. Their position too far to the south can be related to their displacement along submeridional transform faults, and to the South-Caspian basin, if the latter may be regarded as a Paleotethys relic 23,32,33,34 • Carbonate deposits with radiolarite interbeds of the Julphinian stage in the Northern Alborz mark the northern margin of the Gondwana-Land 35. Their position in the facies affinity of Upper Permian deposits (from south to north) shows that they were deposited in the zone of the Gondwana-Land continental slope towards the open sea basin of Paleotethys situated in place of the present South Caspian depression. These data do not speak, however, in favour of the inherited Neotethys( 22,36) . In this region well-pronounced are Indosinian deformations terminating the evolution of Paleotethys. It should be said that to the south of this region, along the Vedi, Taurus and Zagros sutures, there are no reliable evidence on Paleozoic ophiolites, or any other Paleozoic oceanic sediments either. The sections known there, the southern part of the Soviet Transcaucasus inclusive (Fig.2), belong to the platform type. The use of the paleo biogeographical method for control of mobilistic reconstructions of the Caucasus sector of the Mediterranean belt meets a number of difficulties. The first of these is the absence of faunistically confirmed deposits, of the climatically contrasting Ordovician deposits the second is the sporadic distribution and deficiency of fossils in Silurian and Early Devonian rocks. As to the Late Palaeozoic and Early Mesozoic, the Paleotethys, broad in the east, was narrower here, or completely closed off as a bay, and did not prevent from migration plant and animal assemblages, as shown in many paleogeographical schemes compiled for that time on the basis of paleomagnetic data. The northern margin of Gondwana-Land, West Europe and East European platform were situated in the tropical and subtropical areas; so, it is theoretically not possible to separate them by the paleobiogeographical method in the first approximation. This is confirmed by common Carboniferous floras of North Africa and Middle Europe, finds of the European type Namur-Bashkirian flora on the Khrami salient of the Transcaucasian massif, and by the Carbon-Permian flora in the Zonguldak coal basin of Anatolia, by fauna in the Carboniferous (corals, foraminifers) of the Khrami, Svanetia and Donbas 200 regions(23,37), appearance of representatives of Early Permian fusulinids of the North-Tethys province on the southern margin of Paleotethys in the Alborz(38). In the schemes by Termier and Termier 39 the Late Paleozoic Ocean Tethys on the Caucasian traverse became very narrow, and then continues into Europe as a narrow "couloir tethysien". The available differences in the faunistic assemblages can be explained rather well by local changes of habitat. Therefore, one should not refer to works by Termier and Termier(39) in confirmation of the broad Neotethys inherited from the Paleotethys. If one approaches the analysis of the disposition of Paleozoic faunas and floras eastward of the Caucasus (on the Himalayan sector of the Hediterranean belt, for instance) using the concept of an inherited Neotethys, it will result in unresolvable contradictions(40,4l). As for the Early Hesozoic, the analysis carried out by I.V.Arkhipov, C.~engor and other researchers showed(42,43,59) that there was no large structure of the oceanic type in the western part of Eurasia at that time. Paleobiogeographically, it is confirmed by 80sence of appreciable barriers for dispersion of terrestrial Triassic reptiles and fern flora, and by distribution of mixed Laur8sian-Gondwana-Land floras in the Hediterranean region( 44), on the basis of palynological analysis). The paleomagnetic studies of Paleozoic rocks (Table 1) carried out in the Caucasus and in adjacent regions 45 , confirmed, on the whole, the mobilistic reconstructions done. However, these data do not suffice yet, and the conclusions drawn on the basis of paleomagnetism are not convincing, and require checking. TABLE 1- Paleomagnetic Data on Paleozoic and Triassic Rocks of the Caucasus Locality of the Sections DO JO frno 95 n T Jermanis 327 35 19 P Forerange of the Greater Caucasus 237 -24 -13 12 5 P2 Lesser Caucasus (Daralagez 299 34 18 2 Lesser Caucasus 71 35 19 6 Cl _2 Transcaucasia (Dzirula) 348 24 13 14 1 Transcaucasia (Khrami) 324 23 12 5 1 Lesser Caucasus (Daralagez) 215 39 22 15 5 D2_3 Lesser Caucasus (Dara1agez) 232 21 11 8 5 201 The phenomena of magmatism and metamorphism of the Paleozoic era studied on the background of mobilistic reconstructions 46 , yielded a quite coordinated picture of certain types of magmatic rocks and metamorphism relative to the margins of continents and Paleotethys. It is especially well pronounced in the position of Late Paleozoic orogenic magmatism on the northern (Andian type) margin of Paleotethys in the composition of extended sublatitudinal volcano-plutonic belt 47 . On the basis of the Late Paleozoic acid magmatism of the Transcaucasian massif, both in its intrusive and extrusive form, it is not possible to judge of the associated former subduction zone and its position to the south or north of the massif, because the dating of some manifestations of this magmatism based mostly on the K-Ar ratios is rather approximate. Now we shall deal with pre-Jurassic ophiolites and other Paleozoic complexes of the Greater Caucasus and the Dzirula salient of the Transcaucasus, which, as many authors believe, are the main witnessers of the former existence of the Paleotethys. From the beginning of the Paleozoic and end of Devonian all mobilists assume the existence of a relatively broad ocean, the Paleotethys, on the CAucasian traverse of the Mediterranean belt. One can assume that at first it had a simple structure, i. e. it was a large trough-shaped deepening on the Earth t s surface. The sections of ophiolite nappes of the Forerange characterize the oceanic bottom of this basin. Beginning from the Silurian, the Paleotethys because more complicated due to changes of geodynamic conditions: there appeared island arcs, detached marginal seas situated at the active margin of the East-European plate. The Devonian-Carboniferous piles penetrated by holes in the Preaucasus (Fig.2,1) correspond to sediments of the marginal seas, and ophiolitc clastic material appearing in Silurian deposits of the Forerange, as well as Devonian volcanism, testifying to the first dislocations and complicated structure of the oceanic bottom, and to the appearance of volcanic island arcs. The Silurian-Devonian piles of the platform type, known in the upper course of the MaIka river (Khasaut river, Ullu-Lakhran stream), be speak of existence of some rises with carbonate shallow-water sedimentation, possibly of the Bahama bank type: the facies transitionsto terrigenous shaly piles in the same region, and to the section of the Tokhana zone, may outline the conditions of the continental slope and foot of such a rise (Fig.2, II and III). Since Devonian time the mobilistic reconstructions of the Caucasus begin to differ. Thus, the early Devonian volcanism of the Forerange of the Greater Caucasus (Fig. 2, IV) is asigned by Adamia and Shavishvili 22 to the intraarc type, and the zone of the Forerange for that time isregarded as autochthone and as a rift complicating the Greater Caucasus island arc. The work by Omelchenko, 202 ~~--:!.-~--=!...~~. : __ '1..- > > > > :> > ( ----C -t 1 > > :!: 1 > > 203 Fig.3: Summary palaeo profile through island arc paleozone of Greater Caucasus (vertical and horizontal scales chosen at random). A pile of cherts of Lower-Middle Devonian boundary is taken for levelling surface. FR - present-day width of Forerange zone; LZ, Kn, Kr - Laba - Zaraus, Kendellyar and Kartazhyurt subzones. 1 - basalts, 2 - spilites, 3 - rhyolites, 4 - sub alkali basalts,S - tuffs of various compositions, 6 tuffs of acidic composition, ignimbrites, 7 tuff-conglomerates, 8 basaltic plagioclase porphyries, 9 - andesite-basaltic plagioclase porphyries, 10 - "mixed" lavas, 11 - dacites, 12 - perlites, 13 - andesites, 14 - lapi11i and bomb tuffs, 15 cherts and jaspers, 16 tuff-breccias, 17 conglomerates, 18 - limestones, 19 - claystones and siltstones, 20 - sandstones, 21 - terrigenous flysch. Direction of movements of: 22 - lava material, 23 explosive-clastic material, 24 terrigenous material. 204 A B '0 c \./-1' l~ll~lj I=I~ ~51~~~}11I 1~17 c=J .... ~Y 1~I/"w"t---11.1[Z}.1 205 Fig.4: Comparison of palaeoreconstructions of the Caucasus and surrounding territories for Paleozoic and Early Mesozoic (A - after Sh.Adamia et al.(45) (1982), B after A.A.Belov(23), C after I.P.Gamkrelidze(53) (1982) - with simpilfications and insignificant changes) . 1 rough contours of the present continents, 2 - boundaries of continents and microcontinents, 3 - zone of subdction, 4 - assumed mid-oceanic ridges, 5 - sutures, 6 - island arcs, 7 -intraarc rifts, 8 marginal seas, 9 continental slopes and rises, 10 marginal-continental rifts, 11 - margins of Andean type, 12 - ocean basins, 13 - directions of plate displacement. EE - Europe, AF - Africa, AR - Arabia, AN - Anatolia, I - Iran, NI - North Iran, C - Central Iran, PC - Precaucasia, HC - Greater Caucasus, TC - Trans Caucasia, P - Pontian, IA - Iran and Afghanistan, P-T - Paleotethys, N-T - Neotethys. 206 by the author and Grekov and Grekov 48 contains criticism of these concepts. Baranov 49 and E.V.Khain 50 do not agree with them ei ther. True, volcanic and volcanogenic sedimentary rocks of the Forerange, being in allochthonous occurrence and forming the Zaraus and Kartdzhyurt plates of the Kizilkol nappe, correspond quite well by composition to island-arc calc-alkaline volcanism. In the paleotectonic reconstruction they occupy a zone that is three-four times wider than the present-day outcrops (Fig.3), and form a volcanic ensimatic island arc. The zones of the Main Range adjacent to them, and the Bechasin zone, also consi t of a number of nappes including a part of island arc complexes, not being autochthonus bottoms of an intraarc rift. Baranov and Grekov 49 associating the development in space and time of the Greater Caucasus with the Urals (the same also in 51 7, explain the present-day structure of the Greater Caucasus by collision of East Europe and a hypothetic Makera plate the geodynamic regime of compression in the Early Carboniferous from north-west-south-east to submeridional. Noteworthy is that the existence of the Main Range microcontinent (Makera plate) earlier than the Late Devonian has not been confirmed by any data. The main discussion, however, concerns th elocation of the Transcaucasian massif. Adamia and Shavishvili 22 adjoin it to the northern continent from which it is separated by a small ocean basin. The ophiolite allochthones of the Forerange are considered to originate from this basin (Fig.2, V). True, they differ significantly in their effusive-sedimentary part from the first layer of the recent oceanic crust. According to their reconstruction, the main trunk of Paleotethys Stretched southward from the Transcaucasian massif, and the Sevan-Akera zone is a suture of the ocean existin3 for a long time during the Paleozoic and Mesozoic (Fig.4, A). Below 23 not having found Paleozoic ophiolites to the south of the Greater Caucasus (we shall speak of the Dzirula massif later) but instead continuous Paleozoic-Mesozoic sections in oceanic facies, placed the Transcaucasian massif on the northern margin of Gondwana-Land, and, correspondingly, on the southern margin of the Paleotethys (Fig.4, B). The suture of the latter should be looked for in the Greater Caucasus 23,52 . In our opinion, the peculiarities of ophiolite sections of the Forerange are related to differences of paleooceans from the present ones. Researchers of Georgia consider that the interpretation of the history of the Iranian traverse of the Medi terranean belt 22,45 gi ves way to this point of view. They believe that the Iranian part of Neotethys was formed from the Zagros rift in the process of its expansion during the Late Paleozoic, Early and Middle Triassic, whereas the South-Caspian branch of Paleotethys grew narrow. This concept assumes that the inherited Turkey-Transcaucasian part of Neotethys continued since the Triassic as the Zagros part newly formed (Fig.4, A). 207 An intermediate variant belongs to Gamkrelidze(53) who represented the Palaeotethys as two branches separated by the Transcaucasian microcontinent (or island arc?). In his opinion, the northern part of Palaeotethys was closed as a result of Early Carboniferous compressions. Its traces in the form of obducted ophiolite nappes remained in the Forerange and on the Dzirula massif. Since that time the major part of the Transcaucasian massif has been adjoined to the East European platform (Fig.4, C). The southern part of Palaeotethys began to grow narrow since the Carboniferous or Early Permian, as it was detaching from Gondwana-Land and the northern drift of the Iran-Afghanistan microcontinent, on whose south the Neotethys was formed. The southern branch of Palaeotethys closed, as Gamkrelidze(53).believes, only in the Bathonian phase in Middle Jurassic. Herewith, along the Caucasian isthmus in Middle Jurassic a narrow plate detached from the Iran-Afghanistan microcontinent, drifted northward and adjoined Eurasia. The Lesser Caucasian Neotethys opened in its rear. The suture zone of the Palaeotethyan southern branch is drawn directly to the south of the Khrami salient of the Transcaucasian massif, but exact data on its structure have not been given. The position of the Svanetia zone with a continuous, mostly terrigenous Devonian-Triassic marine section interpreted by all researches as relatively deep-water deposits of the continental slope and rise, depends on placing the Transcaucasian massif on a reconstruc tion. The position of the Svanetian zone on the northern margin of the Transcaucasian massif is not disputed by anybody now, but it is confirmed by establishing a southern source area of clastic material including the sialic components(54,SS(. However, on the reconstruction compiled by Belov(23).the Svanetian zone is at the foot of Gondwana-Land, being separated from the Greater Caucasus by the major space of the Palaeotethys. On the reconstructions by Sh.A. Adamia, I.D. Shavishvili(22), M.B.Abesadze, and others(46).it makes up the southern margin of the little ocean basin between the Greater Caucasian and Pontian-Transcaucasian immature island arcs (Fig.4, A,B). In the reconstruction by Gamkrelidze(53). no particular importance is attached to the Svanetian section of the Dizi series. Metamorphism and tectono-magmatic phenomena of the Late Palaeozoic on the Transcaucasian massif to the south of the supposed suture of Palaeotethys, as Belov(23).considers, can be explained by the fact that tectonic movements in the middle of Carboniferous were so intense that they affected various deep levels in the mantle and crust, and in general displacement of masses from south to north, and the location of the main subduction zone along the northern margin of Palaeotethys I could result in shearing of the crust upper horizons and along its southern, relatively passive margin (Fig.S). The suture zone of Palaeotethys could be drawn through the Dzirula salient where the pre-Jurassic ophiolites and other Palaeozoic rocks have been known, as it was done by French scientists(S6(. During the last years our knowledge of pre-Jurassic complexes of the 5 ffI st -E ur pp eO 'n N {)O fld W fm a- [or td fa! eo tet lr! lsl I'l O !T O l'm - - = c _ _ ~ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ~ ~ \ " ' ~ ~ ~ p " ' i ; i + , + , + P Z \r\~_ r; [ T [ /~7/ '/ 209 Dzirula salient were considerably supplemented(57 ,58), but it is still not satisfactory. The following is the main information: a greater part of the pre-Jurassic basement of the salient is composed of plagiogneisses and crystalline schists (P-C ?) and heavily tectoni zed granites of the Late Palaeozoic. In addition, there occurs the Chiatura pile of quartzy porhpyry conditionally attributed to the Upper Palaeozoic, and a narrow band (2 km) of Chorkhana-Utslevi in which exposed are subvertical tectonic lenses and plates of serpentini tes, mylonites, amphibolites, gabbro, and gabbro-diabases of an unknown age, crystalline schists (P-C ?) and Palaeozoic rocks (Fig.2, VII, Fig.6). Among the latter there are phyllites, metasiltstones and metasandstones with lenses of marbles and quartzites attributed to the Lower Cambrian; actinolitic slates, metabasites and metaporphyrites, basic metatuffs and phyllites are assigned to the Upper Silurian - Devonian; metatuffs of quartzy porhpyries, metaconglomerates, metagritstones and metasandstones - to the Upper Palaeozoic without faunistic evidence. The conglomerates incorporate pebbles of all above mentioned rocks, but serpentini tes. Worthy of notice is absence of normal contacts between all above mentioned piles, and their insignificant thickness (the total visible thickness being not over 700 m). This testifies to their reduction in the process of tectonic displacements. Worthy of note also are the different history of metamorphic transformations, attribution of gabbro and diabases mostly to the differentiates of a calc-alkaline suite(58), and the absence of a uniform section of ophiolite association. All these data show that various complexes of pre-Jurassic rocks of the Dzirula salient most probably belonged to different palaeotectonic zones and joined each other tectonically. Their palaeogeodynamic significance is not obvious, and the age is not always established. Therefore, taking the rocks in the Dzirula salient as a basis, we can suggest several variants of palaeotectonic interpretations: 1) The basement of the Dzirula salient is Hercynian, complicated in the Alpine time, the ophiolites are Middle-Early Palaeozoic(53(; 2) The basement of the Dzirula salient is of the Early Cimmerian age with migration of early Palaeozoic ophiolites and Palaeozoic complexes from south out of the near-frontal part of the Pontian-Transcaucasian island arc(22(. 3) It appears to me that there are all grounds to regard the age of structural formation of the Dzirula salient basement as Early Cimmerian (Indosinian), along with the folding of the Svanetian zone and closure of Palaeotethys. At the same time, there are no data on direction of overthrusting of separate ophiolitic elements and their age, as well as their relation to each other and to faunistically characterized Palaeozoic deposits. It is most likely that their displacement to the present-day position took place with participation of large strike-slip-faults. In this respect, the ideas suggested by $engor(59,60) .for the Caucasian/Anatolian sector of the Alpine zone provide a good perspective. _ so , ~ !O O 2j () . 4- B N U l .OO OTr m. CI- ~'' '~' ,-; 's o • • ''' '~\ I~ :,~ II , :, : : . ' ~, \ . . . ' , '" > 00 : . : : ': ,I, ' :: ' I,' 11; ' '. , '. - ~ +- o 1S 0 . . .. • ~.: . + - + ;(' ' , " _ + I - I l +- / t , ' / , /" - , .. . / ' I . . # ' ' // /" /1 /. "- , I - , , t . +- +- . . . +- + +- ~ +- a +- +- + +- • + +- +- " I c= :J - ' \ 7 ~ 2 ~ 8 . . / + +- +- ~ L J . - - - / 1-- - - IJ IH W~ 14I lh1 ~ ~I O ~ ' ; J J ~ 5 ~ (f I , " + + ~ 6 ~' C +- +- ~ ~f J +- +- F ig .6 : G eo lo gi ca l m ap an d c ro s s s e c ti on s o f th e C ho rc ha na -U ts le vi s he ar in g z o n e o f th e D zi ru la s a li en t (a ft er M .B .A be sa dz e, Sh .A .A da m ia a n d G .K .T si m ak ur id ze ): 1 s e rp en ti ni te s, 2 ga bb ro , ga bb ro -a m ph ib ol it es , a m ph ib ol it es , 3 - ph yl li te s, 4 - m a rb le s, S - qu ar tz -p or ph yr ic v o lc an ic s (C he sh ur a s u it e) , 6 - c r y st al li ne s c hi st s, 7 - m yl on it iz ed a n d c a ta c la st ic gr an it es , 8 - gr an it es , 9 - di or it es , d io ri ti c gn ei ss es , 10 - te ct o n ic b re cc ia , 11 - fa ul t, 12 - po st -P al ae oz oi c ro c ks , 13 - di ps . N o 211 CONCLUSIONS The available materials, including those obtained during the last 3-4 years, show that the history of the tectonic development of the Caucasus is most convincingly and fully revealed through analysis in the light of mobilistic ideas. The asymmetric structure of the Mediterranean belt and existence of the Palaeotethys ocean appear almost generally accepted. The northern part of the Caucasus represented its northern active continental margin: of the West-Pacific type in the Middle Palaeozoic, and of the Andean type in the Late Palaeozoic. The southern part of Caucasia and Iran, and almost entire Anatolia were in the Palaeozoic a relatively passive margin of Gondwana-Land. At the end of Palaeozoic - beginning of Mesozoic this part (according to some concepts the Iranian block only) was transformed into a microcontinent as a result of rifting and formation of the new ocean basin, the Neotethys. There are considerable disagreements among researchers on problems of the position of the Palaeotethys suture and the time of its closing, on the interpretation of the ophiolite age and their conformity with these or other structures with the ocanic crust, on the position and palaeotectonic belonging to the Transcaucasian massif. The width of ocean basins and degree of their conformity with the present-day oceans remains obscure for all mobilit researchers. It follows that it is necessary to go on with both the traditional geological directions and the development of palaeomagnetic and palaeo biogeographical trends. At the same time, a conclusion can be drawn already now of difference between the Earth's image with its large oceans and continents, and the same homologous structures of past geological ages. This, first of all, concerns the Mediterranean belt, of the Tethys area, for which S. Karamata( 61) distinguished a special type of plate tectonics manifestation, when a relatively rapid reconstruction of the dynamic regime took place, as well as a change of extensions for compressions and vice versa, under conditions of a limi ted ocean space with abundance of microcontinents. Besides, there probably existed numerous fissure zones of basalt eruptions (scattered spreading). In this connection the elementary estimations also contradict the concept of the inherited Neotethys, that assumes origination of the ocean, 12000 m wide, with spreading rate 2 cm per year during 300 mln. years. Even a considerable shortening of the northern half at the expense of subduction beneath the East-European plate during the Palaeozoic and Early Mesozoic can by no means reduce its southern half along the passive Gondwana-Land margin which subsequently, in the Late Mesozoic, should be consumed at an improbably great speed. Meanwhile, the acknowledgement of a limited subduction along the southern margin of Palaeotethys(23,24) , i.e. under the Transcaucasian and South Anatolian massifs, in the Late Palaeozoic and Triassic will make the solution of the problem of the Mediterranean palaeooceans space much easier. 212 To finish with, I should like to lay emphasis on my preious conclusions: "There was no analogy between the present oceans and their margins, and conditions existing in the Mediterranean belt in the Late Precambrian and Palaeozoic. One may think that, proceeding from the uneven expansion of the Earth(62), or uneven drift(63), and from some palaeontological and geological considerations, that the pre-Mesozoic ocean basins were more narrow and more complicatedly built relative to their morphology and the crust type, than Neotethys and present secondary oceans"(23). To say more exactly, in the pre-Mesozoic time widely distributed were areas of the present Indonesian and Caribbean types. It appears, that the above mentioned divergencies between the present and ancient oceans, as well as some other disagreements of plate tec tonics can be solved wi thin the framework of the Earth's expansion hypothesis(64) in association with mobilism. " •.. These considerations, as well as the outlined differences between various types of ophiolite associations and corresponding ocean basins, do not change, in principle, the fundamental conclusion saying that "the history of intracontinental Earth folded belts is, in fact, the history of the opening and closing of the oceanic type basins at the expense of convergnece and divergence of continents" (ibidem). REFERENCES 1- Pushcharovsky,Yu.M., Ruzhentsev,S.V. (1983). 'Geological peculiarities of recent and ancient oceans'. Doklady Akad.Nauk SSSR, v.270, No.2, 404-406 (in Russian). 2- Ruzhentsev~S.V. (1984). 'Oceans and paleooceans'. Sedi~entary cover of the bottomn of the world ocean and land. Nauka. Trans.Geol. Inst. Akad.Sci. USSR, v.388, 13-26 (in Russian). 3- Coleman,R.G. (1984). 'The diversity of ophiolites'. Geol en mijnbouw, v.63, n.l, 141-150. 4- Timofeev,P.P., Kholodov,V.N., Khvorova"I.V. (1983). 'Evolution of sedimentation processes on continents and in oceans'. Litologiya i polezn.iskop., No.~, 3-23 (in Rus.). 5- Hein,J.R., Karl,S.M. (1983). 'Comparison between open-ocean and continental margin chert sequences'. In: Siliceous Deposits Pacif.Reg. Amsterdam e.a., 25-43. 6- Belov,A.A., Mossakovsky,A.A., Sokolov,S.D., Shvolman,V.A. 'Late Paleozoic-Early l1esozoic development of the Hediterranean - Central Asian branch of the Tethys (Palinspastic reconstructions)'. In: Problems of the Caucasus geodynamics. t1oscow, Nauka, 21-30 (in Russian) . 7- Gasanov,T.A. (1979). 'On age of ophiolites and independence of the gabbro-diabase complex of the Sevano-Akera zone of the Lesser Caucasus'. Geotektonika, No.5, 86-97 (in Russian). 8- Knipper,A.L. (1975). Oceanic crust in the structure of the Alpine folded areas. t·10scow, Nauka, 193 p. (in Russian). 9- Dobrzhinetskaya,L.F., Ez,V.V. (1982). 'Metamorphic rocks in melange of the ophiolite complex of the Lesser Caucasus (Adzharis area).' Geotektonika, No.1, 70-81 (in Russan). 213 10- Luts,B.G., Knipper,A.L., Dobrzhinetskaya,L.F., Popolitov,E.I. (1980). "Xenoliths of metamorphic rocks in gabbroids of the ophiolite complex of the Lesser Caucasus". Geotektonika, No . .§., 49-61 (in Russian). 11- Meliksetyan,B.M., Bagdasaryan,G.P., Gukayan,R.Kh. (1984). 'Isotopic geochemical and geochronological studies of eclogite-amphibolites associated with ophiolites of the Sevano-Amasya belt (Amasya massif)'. Izvest. Akad.Nauk Armen.SSR, Earth Science, XXXVII, No.1, 3-22 (in Russian). - 12- Agamalyan,V.A., Aslanyan,A.T., Bagdasaryan,G.P., Gukasyan,R.Kh. (1982). 'Areas of the Assyntian and Hercynian consolidation in the crystalline base of structural formation zones on the Armenian SSR territory'. Abstracts of the papers of the International \iJorking Meeting of the participants of IGCP Project 5 in the Caucasus, 15-16 (in Russian). 13- $engor,A.M.C., Yllmaz,Y., Sungurlu,O. (1985). 'Tectonics of the Hediterranean Cirnmerides: nature and evolution of the l,qestern termination of Paleotethys'. In: Geological evolution of Eastern Mediterranean. J.E.Dixon and Robertson,R.H.F. eds. Geol. Soc., London, Spec. Publ., No.17, 66-112. lLf- Bergougnan,H., Parrot,J~F. (1981). 'Le carrefour ophiolitique d'Erzincan, Neo-et Paleotethys'. Cah.O.R.S.T.O.M., ser. Geol., 11, 165-188. -- 15- Tekeli,O. (1981). 'Subduction complex of pre-Jurassic age, Northern Anatolia, Turkey, Geology, v.2, n.l, 68-72. 16- Cogulu,E., Krurnmenacher,D. (1967)? 'PRoblems geochronomeriques dans la partie n.w. de l'Anatolie centrale (Turquie), . Schweiz. mineral. und petrogr.Mitt., v.47, n.l, 825-83l. 17- Dora,O. (1971). 'The Kaz-Dag massif near Edremit: an excursion guide'. In: Geol. and History of Turkey. Tripoli, Libya, 403-411. 18- Yllmaz,O. (1971). Etude petrographique et geochronologique dela region de Casas (Partie meridionale du massif de Bitlis, Turquie), . - THes edoct. Univ. sci. med. Grenoble, 208 p. 19- Baykal,F. (1966). Explanatory text of the Geological map of Turkey, Si vas. Ankara: t-IT A, 106 p. 20- Belov,A.A., Sokolov,S.D. (1973). 'Relics of the Hesozoic oceanic crust among crystalline complexes of the Miskhan massif of Armania'. Sov.geol., No.~, 26-41 (in Russian). 21- Brinkmann,R. (1976). Geology of Turkey. Amsterdam; Elsevier, 150 p. 22- Adamia,Sh.A., Shavishvili,LD. (1979). 'A model of tectonic evolution of the Caucasus Earth crust and contiguous territories (pre-Alpine stage)'. Geotektonika, No.1, 77-84 (in Russian). 23- Belov,A.A. (1981). Tectonic developme~t of the Alpine folded area in Paleozoic. Hoscow, Nauka, 212 p. (in Russian). 24- $engor,A.I1.C. and Yllmaz,Y. (1981). 'Tethyan evolution of Turkey; a plate tectonic approach'. Tectonophysics, vol.75, 181-241. 25- Bingol,E. (1976). 'Evolution geotectonique de l'Anatolie de I'Ouest'. Bull.Soc.Geol.France, Vol.18, n.2, 431-450. 26- Radelli ,L. (1970). 'La nappe de Balya. La zone des plis egeens et l'extension de la zone du Vardar en Turquie occidentale'. Geol.alp., vol. 46, 43-50. 214 27- Fourquin.C. (1975). 'L'Anatolie du Nord-ouest, marge meridional du continen t europeen, histore paleogeographique, tectoniq ue et magmatique durant Ie Secondaire et Ie Tertiaire' . Bull.Soc.Geol.France., Ser.7, vol.17, n.6, 1058-1070. 28- Bergougnan,H., Fourquin,C. -: Discussion; ~engi::ir ,A.M.C., Yllmaz, Y., Ketin,I. Reply (1982). 'Remnants of pre-Late Jurassic ocean in northern Turkey: Fragments of Periman - Triassic Palro-Tethys?' Geol.Soc.Am.Bull., v.93, No.9, 929-936. 29- Solovkin,A.N. (1950):-'On T;iassic of the Akera river upper course (Azerbaijan SSR)'. Dokl.Akad.Nauk.Azerb.SSR, v.Q, No.2, 33-40 p. (In Russian). 30- Yllmaz-,Y., ~engi::ir,A.M.C. (1985). 'Paleo-Tethyan ophiolites northern Turkey: Petrology and tectonic setting'. Submitted Ofioliti. 10(2/3), 485-504. in to 31- Davise,R.G., Jones,C.R., Hamzepour,B., Clark,G.C. (1972). Geology of the Masuleh sheet, 1:100000, Northwest Iran. Geol.Surv.Iran Rept., n.24, 330 p. 32- Mouratov,M.V. (1984). 'The main stages of tectonic development of the near Black Sea area and genetic types of structural elements of the Earth's crusts'. Izv.Akad.Nauk SSSR, ser. geol., No.5, 3-18 (in Russian). - 33- Kornev,V.A.,Lutsuk,E.M., Singurov,A.M. (1962). 'Main features of the Caspian Sea from marine geophysical data'. Sov.Geol., n.Q, 80-89 (in Russian). 34- Sti::icklin,J. (1974). 'Possible ancient continental margins in Iran'. In: Geology of continental margins, Burk,C.A., Drake,C.L. Eds., N.Y.Springer, 873-887. 35- Jenny,J., Sampfly,G. (1978). 'Lithostratigraphic du permien de L'Elburs oriental en Iran'. Eclog. Geol. Hevl. v.7l/3, 551-580. 36- Lordkipanidze,M.B. (1980). Alpine volcanism and geodynamics of the central segment of the Mediterranean folded belt. Tbilisi, Metsniereba, 162 p. (in Russian). 37- Belov,A.A., Reitlinger,E.A. (1966). 'On paleontological characteristic of "lower tuffites" of the Khrami salient of South Georgia'. Doklady Akad.Nauk SSSR, v.170, No.1, 152-154 (in Russian). 38- Leven,E.Ya., Shcherbovich (1978). Fusulinids and stratigraphy of the Asselian stage of Darvaz. Moscow, Nauka, 1983 p. (in Russian). 39- Termier,H., Termier,G. (1974). 'Distribution des faunes marines dans Ie sud Ie da Tethys et sur la bordure septentrionale du Gondwana au cours du paleozoique super. Annal.Soc.Geol.Belg., t.97, n.l, 387-446. 40- Xiao Xuchang, Gao Yanling (1984). 'Tectonic evolution of the Tethys Himalayas of China'. 27 th IGC, Moscow, Colloquium LOS Tectonics of Asia, Nauka, Reports, v.s, 181-189. 41- Chengfa,C., Yuchen,P. (1984). 'Preliminary synthesis of the tectonic evolution of the Qinghai-Xizang (Tibet) Plateau', 27th IGC, Moscow, Colloquim K.Os. Tectonics of Asia, Reports v.s, 190-206. 42- Arkhipov,LV. (1984). 'Early Mesozoic rifting in the structural development of the Alpine folded area'. Izvestia VUZ, ser. geol. i razv., No .1, 3-17 (in Russian). 215 43- $engor,A.M.C. (1979). Mid-Mesozoic closure of Permo-Triassic Tethys and its implications'. Nature, vol.279, No.57l4, 590-593. 44- Visscher,H., Zwan,C.J., van der-,--(1981~'Palyno1ogy of the Circum-Mediterranean Triassic: phytogeographical and paleoclimatological implications'. Geo1.Rundschau, Bd. 70. No.1, 625-634. -- 45- Adamia,Sh.A., the Caucasus Problems of Russian). Asanidze,B.Z., Pechersky,D.M. (1982). 'Geodynamics of (experience in palinspastic reconstructions)'. In: the Caucasus geodynamics, Moscow, Nauka, 13-21 (in 46- Abesadze,M.B., Kekelia,M.A., Mgeliashvili,T.N., Tsimakuridze,G.K., Chkhoutua,T.G., Shavishvi1i,I.D. (1982). 'Pre-Alpine development of the Caucasian acti ve continental pa1eomargin (magmatism and metamorphism)'. In: Problems of the Caucasus geodynamics. Moscow, Nauka, 30-41 (in Russian). 47- Mossakovsky,A.A. (1970). 'On the Upper Paleozoic volcanic belt of Europe and Asia'. Geotektonika, No.4, 65,77, (in Russian). 48- Omelchenko,V.L., Be1ov,A.A., Grekov,I.I. (1984). 'Early-Middle Paleozoic volcanism of the Forerange and paleotectonic zonation of the Greater Caucasus'. Geotektonika, No.5, 61-72 (in Russian). 49- Baranov,G.I., Grekov,I.I. (1982). 'Geodynamic model of the Greater Caucasus'. In: Problems of the Caucasus Geodynamics, Moscow, Nauka, 51-59 (in Russian). 50- Khain,E.V. (1979). 'Ophiolites and nappe structure of the Forerange North Caucasus'. Geotektonika, No.4, 63-80 (in Russian). 51- Samygin,S.G., Khain,E.V. (1985).- 'The Southern Urals and North Caucasus in Paleozoic - experience of comparison'. Geotektonika. No.2, 40-56 (in Russian). 52- Adamia,Sh.A., Shavishvi1i,I.D. (1982). 'On the root zones of Hercynian ophiolitic nappes of the \Vest Caucasus'. Dokl.Akad .Nauk SSSR, v.267, No.6, 1424-1426 (in Russian). 53- Gamkrelidze ,I.P.- (1982). 'Mobilism and problems of the Caucasus tectonics'. In: Problems of the Caucasus geodynamics. Moscow, Nauka, 4-8 (in Russian). 5Lf- Somin,M.L. (1971). Pre-Jurassic base of the Main Range and the Southern slope of the Greater Caucasus. Moscow, Nauka, 246 p. (in Russian). 55- Kutelia,Z.A. (1983). Stratigraphy and conditions of formation of Paleozoic deposits of the Greater Caucasus southern slope (after conodonts). Abstracts of the thesis. Tbilisi, 22 p (in Russian). 56- Bergoungnan,H., Fourquin,C. (1980). 'Un ensemble d'e1ements communs a une magre active alpine des Carpathes meridiona1es a Iran central: Le domaine irano - ba1kanique.' Bull. Soc. Geol. France, ser. 7, vol.22, n.1, 61-83. 57- Abesadze,M~B., P1anderova,E., Tsimakuridze,G. (1980). 'New data on the age of metamorphic schists of the Dzirula salient (Georgia)'. Geo1.pr., spr. 74, 137-143 (in Russian). 58- Gamkrelidze,I.P., Dumbadze,G.D., Kekeliya,M.A., Khma1adze ,I. I. , Khutsishvili,O.D. (1981). 'Ophiolites of the Dziru1a salient and problems of Paleotethys in the Caucasus'. Geotektonika, No.~, 23-33 (in Russian). 216 59- $engor ,A.H.C. (1984). The Cimmeride Orogenic System and the Tectonics of Eurasia. Geol.Soc.Amer.Spec.Pap., 195, 82 p. 60- $engor,A.l'1.C., Yllmaz,Y., Ketin,I. (1980). 'Remnants of pre-Late Jurassic Ocean in Northern Turkey: Fragments of Permian-Triassic Paleo-Tethys?' Bull.Geol.Soc.Am. Pt.I, v.9l, n.lO, 599-609. 61- Karamata,S. (1983). 'Peculiarities and--manifestations of plate tectonics in the areas of the Tethys type'. Geotektonika, No.5, 52-66 (in Russian). - 62- Larin,V.N. (1980). Hypothesis of the initially hydride Earth. Moscow, Nedra, 215 p. (in Russian). 63- Engel,A.E.J., Engel,C.G. (1972). 'Origin of continents'. In: Essays of the present-day geochemistry and analytical chemistry. 11oscow, Nauka, 76-78 p. (in Russian). 6Lf - Stock lin ,J. (1984). 'Orogenesis and evolution of the Tethys in the Far East: evaluation of present-day concepts'. In: Tectonics of Asia, 27 IGC,K., OS, papers, v'2, 53-69 (in Russian). PALAEOMAGNETISM OF UPPER CRETACEOUS ROCKS FROM THE CAUCASUS AND ITS IMPLICATIONS FOR TECTONICS Mikhail,L.Bazhenov, Valentin,S.Burtman Geological Institute, Academy of Sciences of the USSR pzyhevsky per., 7 109017 Moscow USSR ABSTRACT. The Caucasus region was subdivided into two domains according to their Al pine structural patterns. The WNW trends prevail within the northern domain (the Great Caucasus), and fold axes outline a large arc within the southern one (the Lesser Caucasus). Sedimentary rocks of Late Cretaceous age have been sampled in Dagestan near the northern margin of the Great Caucasus, and on both flanks of the Lesser Caucasus arc. Directions of a prefolding, most probably primary, component of natural remanent magnetization have been obtained after treatments in the laboratory. The Upper Cr etaceous limestones from Dageston yielded a mean paleopole at 740 N and lSloE, closely matching the Cretaceous-Palaeogene part of the Eurasian polar wander path. Declinations ranging from 3530 to 370 correspond to changes of fold axis trends within the Lesser Caucasus thus pointing to tectonic origin of the arc. The inclinations for the northern and southern domains differ by 80 +30 • We conclude that the Lesser Caucasus moved about 900+350 kilometers northward with respect to stable Eurasia since Late Cretaceous. 1. INTRODUCTION The Caucasus mountain country occupies the wide neck between the Black and the Caspian Seas. This region can be subdivided into two domains according to their structural patterns which have been forming from Miocene through to Quaternary. The northern domain including the Great Caucasus Range and the Transcaucasian basins is characterized by north-westerly trends of the Alpine folds and faults. The southern one consists of a number of plateau and ranges of the Lesser Caucasus. The Alpine structures here outline the convex-to-north Lesser Caucasus arc (LCA) (Fig.l). This arc is terminated by the East Anatolian fault in the west and can be more or less reliably correlated to structures of North-Western Iran. 217 A. M. C. ~engor (ed.), Tectonic Evolution of the Tethyan Region, 217-239. © 1989 by Kluwer Academic Publishers. 218 -......::: BLACK SEA ::. .... ~.:~p // / 40'N~/ /" --- - ~EA ..... 42'E o 100 200 300l'lM , , Fig.l: The Al pine structural pattern of the Caucasus and North-Eastern Turkey. 1 axes of the Alpine structures, 2 main dextral strike-slips, 3 - main thrusts, 4 - our sampling localities, 5 - sampling localities of other authors, 6 - the area of palaeomagnetic work in the Eastern Pontides. PC - the Ponto-Caspian fault, EA - the East Anatolian fault, GC - the Great Caucasus, LC - the Lesser Caucasus, EP - the Eastern Pontides, S - Sevan Lake. See the text for explanation of other letters. The Ponto-Caspian fault is the boundary between these two domains. It is represented by southward-dipping gentle thrusts and thrust-slips in the central part of the region (Basheleyshvili et a1., 1982). This fault dives under the Black Sea to the west and is overlain by a thick pile of Quaternary sediments to the east, within the Kura depression. Its eastern part is probably a dextral strike-slip. The LCA as a whole is the northermost part of the Arabian Syntaxis and the Ponto-Caspian fault is in turn its northern boundary. In order to estimate the character and scale of horizontal movements within the Caucasian region we undertook palaeomagnetic studies of Upper Cretaceous rocks at three areas. The first one is situated in White (Carbonat~) Dagestan on the northern slope of the Great Caucasus Range. Our aim here was to obtain the reference Late Cretaceous pole as the contemporaneous Eurasian pole is poorly defined (see below). The other two are the Amasia area at the western flank of the LCA near to its apex and a number of localities at its eastern flank (Fig .2). Here we tried to obtain information concerning the origin of this arc as well as some palaeolatitude estimations. A number of Late Cretaceous palaeomagnetic data exists for the Lesser Caucasus (Adamia et al., 1979; Szirunyan, 1981; Khalafov, 1985), but we have difficulties to incorporate them into analysis. Some considerable part of the data set was rejected where only the time-cleaning technique (storage test) was used. In a number of cases 219 fZI'D0~~05 a 20 40KM ...... --'---1 Fig.2: The structural map of the Lesser Caucasus. 1 - Neogene-Quaternary rocks, 2 - older formations, Palaeogene and Mesozoic mainly, 3 - fold axes, 4,5 the same as on Fig.l (NP - sampling locality from (Pechersky, Nugen, 1978». palaeomagnetic data were also rejected where no field tests were applied or these tests were negative. Such a not too rigorous selection diminished the number of acceptable data drastically. Two results only (Pechersky, Nugen, 1978) have been retained. Those are from magmatic and sedimentary rocks of the Ijevan-Tauz Basin (locality NP, Fig .2) and from sedimentary rocks of the Vedi area (locality V, Fig .1). As for Dagestan or other parts of the Great Caucasus there are no palaeomagnetic data in this region. 2. GEOLOGICAL SETTING AND SAMPLING We tried to choose sampling sites within a limited area (locality) at well-exposed gomoclines, at both limbs of one fold preferably. One hand-sample oriented by magnetic compass was taken from each layer. The number of samples and the true thickness studied were chosen so as to average adequately all kinds of palaeomagnetic noise, including secular variations, within a single site (section). Such a way of sampling allowed us to treat each section-mean as an independent palaeomagnetic result. This approach could not be used everywhere mainly becR',se of pour quality of exposures. In this case results from several outcrops were combined and mean directions were computed. 220 The White Dagestan area is an uplifted and deformed marginal zone of the East European platform. The Upper Cretaceous rocks are mainly limestones though some sandstones are present in the Cenomanian and Maastrichtian parts of section. The main folding took place in Late Miocene and Pliocene but angular unconformities of the Palaeocene and Eocene age (Beloussov, 1962) point to older weak deformations here. Palaeomagnetic samples were taken on both limbs of two well-exposed anticlinal folds about 40 kilometers apart (localities Dl and D2, Fig.l). The Cenomanian through to Lower Maastrichtian rocks were studied, the main body of this collection being of the Turonian-Santonian age. The limestones prevail (78 samples) though 15 samples of sandstones were taken too. The rocks are of white to light-grey colour as a rule. Some pinkish limestones were sampled from the Turonian-Santonian strata. The closure of the Mesozoic Tethys at the Lesser Caucasus has resulted in intensive thrusting in the Coniacian. These nappes are overlain by terrigenous Upper Coniacian rocks and limestones and marls of the Santonian-Maastrichtian age (Knipper, 1975; Sokolov, 1977). This neoautochton was folded in Neogene and Quaternary though some angular unconformities were found in older parts of it (Milanovsky and Khain, 1963; Knipper, 1975). Grey and red limestones of the Late Senonian age were sampled at 9 sites in the Amasia area (Fig.2). Their exposures are of poor quality and we had to study all outcrops found. Limestones of the same age and colour were studied at the eastern flank of the LCA (Fig.2). The quality of exposures is much better here and we sampled both limbs of folds at localitites E2, E3 and E5; four long sections were sampled at locality E4. The only exception is locality El, where limestones are "crumpled" into small irregular gentle folds with dips about 15-200 , 350 at most. The attitudes of beds vary from sample to sample and no gomoclines have been found here. 3.METHODS AND PROCEDURES Three or four specimens were cut from each hand-sample in the laboratory. Two specimens per sample were subjected to stepwise thermal cleaning up to 4000 C in a furnace shielded with two )l-metal layers and placed within a three-pair-system of Helmholtz coils. The residual field in the central part of the furnace was estimated to be less than 10 nT. If the within-sample scatter was high the third specimen was added to other two. The mutual orientations of specimens were changed after each heating step in order to detect and evaluate any laboratory-induced magnetizations. All measurements of natural remanent magnetization (NRM) or its components were made with the Chechoslovakian JR-4 spinner magnetometer. As a number of samples proved to be very viscous this magnetometer was placed within large Helmholtz coils with the residual field about 1000 nT. The isothermal remanent magnetization (IRM) acquisition and the IRM thermal demagnetization curves were obtained for some representatitve samples. An astatic magnetometer was used for rock-magnetic measurements. 221 Results of the stepwise thermal cleaning were analyzed with stereonets and the vector subtraction technique sometimes. If laboratory-induced magnetizations were negligible the results after the final heating were always prefered. Otherwise we used one of the previous steps. When calculating mean directions we prefered to use a level of statistics resulting in larger confidence limits (though other variants are presented in tables too). The corrections suggested by Demarest (1983) were taken into account. Due to construction of the lJ-metal shielded furnace collections could be heated only up to 4000 C. Some pilot samples were heated above this temperature in another furnace where cancellation of the ambient field was much poorer. No stable end-points were obtained this way because of spurious laboratory-induced magnetizations. So field tests were the main tools to prove reliability of paleomagnetic data. When possible, the section-means (or group-means, see below) were used for tests (McFadden and Jones, 1981). The concentration parameter ratio (CPR) test (McElhinny, 1964) was applied in "desperate" cases only when it was impossible to use the more rigorous test. From our point of view the CPR test does show correctly which component of NRM prevails in a data set, but it does not prove that this magnetization is the purely prefolding (or postfolding) one. 4. RESULTS 4.1. Dagestan All 15 samples of sandstones were rejected because of high within-sample scatter of palaeomagnetic vectors but we were luckier with limestones. 65 samgles out of 78 yielded very consistent results after heating to 200 C. A part of them was completely stable afterwards while others shifted within 50 _100 during the subsequent heating steps. Four tilt-corrected section-means are very tightly clustered (Table 1). After McFadden and Jones (1981) we found that the observed value, f = 1.860, is much less than the critical one at 95% confidence level, F(6, 122) = 3.705. Therefore, the magnetization of limestones can be considered as a purely prefolding one. Both polarities were found, though the reversed one was only met in the upper Santonian and Campanian rocks. The polarity-means are very nearly antiparalle1. For example, these directions for the section D2(1) are D = 160 , I = 570 and D = 1970 , I = _560 (Fig .3A). This fact points to the univectorial magnetization in these limestones. 222 A I 50· I 70' 0 o· • • ..... • lSI. • • • • • + 00 00 o ooog 180' o· .. ~ ..\ ..... -.. I + I 70' 180· Fig.3: Cleaned tilt-corrected directions of magnetization from Dagestan. A - sample directions (circles) and means for groups of different polarity (double circles) from section D2(1), B - pairs of specimen direction (dots) from each sa,n)le of the strongly magnetized rocks from section Dl(l). Solid (open) symbols denote the downward (upward) - pointing vectors. (The same convention is preserved everywhere). The remanent magnetization intensities in limestones vary widely, their dist ribu tion being clearly bimodal (Fig. 4) and modes differing by an order of magnitude. The strongly magnetized samples, both grey and pinkish ones, were met in the Turonian-Santonian rocks, and the "weak" part of collection is distributed rather uniformly throughout the sec tion. The mean palaeomagnetic directions for these two groups are practically identical (Table 1). It is an interesting fact that the within-sample grouping of palaeomagnetic vectors is almost always much better than the between-sample one. This holds true for both the strongly and weakly magnetized rocks though is more pronounced for the former (Fig.3B). The reason seems to be clear as any measurement errors should affect the clustering of "weak" specimens mainly. There appears to be no doubts as to a prefolding, that is pre-Miocene age of stable magnetization in limestones. Some lines of evidence speak in favour of its much older age. These are: (1) The normally and reversedly magnetized rocks yield antiparallel palaeomagnetic directions. (2) The polarity distribution agrees in general with palaeomagnetic scale (e.g., Lowrie and Alvarez, 1981), for the Cenomanian thorugh to Lower Santonian sediments revealed only 223 the normal polarity, while both the signs were met in younger rocks. (3) The mean directions are identical for both the weakly and strongly magnetized rocks. (4) The relation between the within-sample and between-sample grouping points to layer-to-layer-controlled process of NRM acq uisi tion and this is hardly possible in case of any later remagnetization. (5) The tight grouping of section-means suggests that the stable component of NRM in limestones should be older than any deformations (see above) in the region. We believe that these facts combined prove the primary age of this magnetization. 15 n fa 5 o -7 I I n=43 1 n"23 I I I I I I I 19J (emu.1 -5 Fig.4: The distribution of magnetization in tensi ty afetr TO=4000 for the whole Dagestan collection. 4.2. The Amasia Area (the Western flank of the LCA). Some samples were rejected because of high within-sample scatter but the main part of the collection yielded well-groupped specimen directions after stepwise cleaning up to 4000 • As a rule, an unstable NRM component, of viscous nature probably, was removed below 2500 , the remalnlng magnetization being directionally stable afterwards (Fig.5). A limited number of samples shifted along great circle paths within the whole temperature range used. It took some considerable time to comprehend these data, as at first they appeared to be of "reject-and-forget" type. For example, the tilt-corrected mean vector from site Al has a north-western declination and moderately steep negative inclination, and before tilt-correction it lies within 150 from the modern dipole field direction, but of reversed polarity. As this collection was taken in an erosional window among flat-lying Late Neogene basalts we assumed that the upper Cretaceous limestones had been remagnetized completely during lava eruption. As a result this site was omitted. 224 Njup L-------L-_-+----1'-----L--..L-~2.-~---'3 E/ Hor NRM 2 NRM 3.10- 6 emu Fig.5: Thermal vector demagnetization plots for grey (cricles) and red (triangles) limestones from the Amasia area. Open (solid) symbols denote horizontal (vertical) plane. As the next step we tried to analyse the results from the grey and red limestones separately. Such an approach proved to be successful. The grey rocks yielded almost antipara11el directions of both polarities; their tilt-corrected means are D = 3530 , I = 440 and D = 1750 , I = _450 (Fig.6A). Besides, the grey limestones yielded well-grouped site-means, while the red ones did not (Fig.6B, Table 2). However, we could not use these site-means for a fold test as the bedding attitudes varied considerably within each site (Table 2). We tried to overcome this in the fo11owing way. The bedding attitude at every sampling point was converted into direction of normal to bedding plane. A distribution of a11 normals was drawn (Fig. 7) and then was subdivided into a certain number of approximately isometrical non-overlapping areas. The palaeomagnetic tilt-corrected group-means corresponding to each group of normals were calculated for grey and red samples separately. These two sets of group-means were then tested as usual. Several variants of grouping were tried and fold test was invariably positive for "grey" data (Fig. 7) and negative for "red" ones (not shown for clarity). Of course, there is a certain ambiguity concerning the number of groups and their positions on the stereonet, but nevertheless this procedure seems to be preferable to the CPR test. . ':..- • :I't-•• • • • o o o~O: 0 '0 00 o • ~ ••• • 225 •• Fig.6: Cleaned tilt-corrected directions of magnetization from the Amasia area. A - sample directions (cricles) and means for groups of different polarity (double cricles) for the grey limestones, B - site-mean directions for thle grey (circle) and red (squares) rocks. Such a grouping leads to some loss of information but the same is true for "natural" grouping too. From general considerations the best fold test should deal with unit palaeomagnetic and structural vectors without any grouping, but such a mathematically correct test has not existed so far. So, the positive fold test proves the predeformational age of the stable NRM component in the grey limestones. The only evidence of its primary origin is the antiparallelity of the normal and reversed palaeomagnetic directions. Though the "red" data do not pass the fold test it is interesting to note that some site-means (and group-means too) are in a rather good agreement with the "grey" set (Fig.6B), while others are not. We tentatively conclude that the NRM of red beds is of compo site nature. In other words, the reddening of limestones and their partial or complete (?) remagnetization took some considerable time, including the epoch of the main folding. 4.3. The Eastern Flank of the LCA The whole collection of about 150 samples of grey and red limestones was thermally cleaned up to 4000 • Though very similar in appearance the rocks from each locality yielded rather different results. 226 Fig.7: The distribution of normals to bedding planes for all sampling points at the Amasia area (dots) and palaeomagnetic group-means for the grey limestones (triangles). Dashed lines denote the boundaries of groups (normals from groups 9 and 10 are not shown for clarity). The numbers of groups and group-means correspond to each other. All samples from locality E3 and from one section at locality E2 were too weakly magnetized to be measured reliably. On the contrary, grey limes~~nes from locality E5 bear an unusually strong magnetization, about 5xlO e.m.u. per ccm on average, because of a large amount of tuffaceous material. The sample-means here are very scattered, the concentration parameter being as low as 5, and two section-means after til t-correction differ significantly. All attempts to obtain consistent results failed and this collection was rejected. o· o 000 op o "0' • :1if:. · I •• • • 227 Fig.8: Cleaned directions of magnetization from localities at the eastern flank of the LCA. A - sample directions (circles) and means for groups of different polarity (double cricles) from locality E4, B - sec tion-means with circles of confidence: circles - from locality E4, cross - from locality E2, square and triangle - section-means of prefolding and postfolding components of magnetization from locality El respectively. All directions are corrected for tilt but for the last one, which is given in geographic coordinates. The grey limestones from four sections at locality E4 yielded consistent data. Palaeomagnetic mean vectors of different polarity are almost antipodal, D = 330 , I = 460 and D = 2210 , I = _470 (Fig.8A), and the tilt-corrected section-means clearly pass the fold-test (Fig.8B, Table 3). These results point to prefolding, probably primary, age of magne tiza tion here. A similar palaeomagnetic direc tion was obtained from one section at locality E2 (Fig.8B, Table 3), but this similarity may be fortuitous, as localities E2 and E4 are about 50 kilometres aparts. We can not confirm the reliability of the last result in any other way as the gomocline was studied here and all samples were normally magnetized. Somewhat peculiar data were obtained at locality El. The magnetization intensity here increased about two-wold between 2000 and 2500 and decreased steadily afterwards. A component (A) of reversed polarity and a more stable component (B) of normal polarity were identified by the vector substraction method (Table 3). Neither fold test (McFadden and Jones, 1981) nor the above-described "grouping" 228 modification could be used because the bedding attitudes varied greatly within few meters (see above) and the half of collection was rejected because of large within-sample scatter. So we had to apply the CPR test. The ratio of concentration parameters before and after correction for tilt, Kb/K , is 1.93 for component A and the inverted ratio, K /Kb is 2.59 for co'fuponent B. The critical values of F (24,24) at 90%aanff 95% confidence levels are 1.70 and 1.98 respectively for the first component and F (26,26) = 1.90 at 95% level for the second one. He conclude that component B is of prefolding age and, less surely, that componen t A is of postfolding age. Of course, these conclusions do not mean that both magnetizations are purely univectorial ones. The data from grey limestones only have been discussed so far. According to rock-magnetic measurements magnetite predominates in grey limestones and haematite is the main NRM-carrier in redbeds. The latter sediments gave no consistent results because of either large within-sample scatter or multicomponent magnetization. We have studied a number of sites (not described in this paper) where the same rocks from other parts of the LCA changed their polarity twice while heated from 2000 to 4000 • All these data point to unresolved multicomponent magnetization of redbeds and they were excluded from further analysis. 5. INTERPRETATION AND DISCUSSION 5.1. The Results from Dagestan and the Polar Wander Path of Eurasia The long time-interval studied in Dagestan was subdivided into three smaller ones and the Cenomanian, Turonian-Santonian and Campanian poles were calculated separately (Table 4). We should point out that only the second one is really reliable, as the other two are based on limited number of samples. Nevertheless all three poles are identical sta tisticall y (Fig. 9). There are also shown the Palaeocene - Eocene, EP (Westphal et al., 1986) and the Early Cretaceous, EK (Bazhenov, Shipunov, 1985) mean palaeomagnetic poles for Eurasia, which match perfectly the Dagestan data (Fig. 9, Table 4). The studied part of Dagestan is the margin of the East European platform (see above) and any large-scale horizontal motions are geologically improbable in this region. Palaeomagnetic data confirm that. The tight clustering of poles sohws that Dagestan suffered no local rotations as well. Really, it seems very improbable that the exellent agreement of the Eurasian poles with the Dagestan data is just a coincidence. If these two landmasses were really welded together we can come to conclusion that the Eurasian plate changed neither its latitude nor orientation with respect to modern meridian from Early Cretaceous till Eocene.(Of course, this does not put any constrains to longitudinal component of its motion). 229 50 70· gOoN 90"Er---L-----'-----'----...l...-----'-----. 900W 180· Fig.9: Palaeomagnetic poles for Dagestan (circles), Eurasia (crosses) and North America (squares). C, T and K are the Cenomanian, Turonian-Santonian and Campanian poles respectively for Dagestan (this paper)- EK, EP - the Early Cretaceous (Bazhenov, Shipunov, 1985) and Palaeocene-Eocene (Westphal et al., 1986) mean poles for Eurasia respectively. KE, KM, KL alnd TE - the Early, Middle and Late Cretaceous and Early Tertiary mean poles for North America respectively after rotation to Eurasion coordinates (Westphal et al., 1986). Thin (thick) lines are circles of confidence for Eurasian and North American (Dagestan) ypoles. All data are projected to Northern Hemisphere. It is evident that one must know the accurate kinematics of the Afro-Arabian and Eurasian plates in order to comprehend the Alpine fold belt evolution. The apparent polar wander (APW) paths of these plates are very important for this task. The reliability of the paths depend mainly upon the number and quality of palaeomagnetic data, and both are often unsatisfactory. The Late Cretaceous mean pole for Eurasia is an example. It is based on three independent, rather scattered unit 230 poles and is poorly defined, with A 5=130 (Westphal et a1., 1986). A combined APW-path can be constructe~ according to kinematics of two or more main lithospheric plates (e.g. Van der Voo and French, 1974). In such a way one may hope to fill the gaps on the APW-path of a single plate. But such an approach seems to be valid if and only if the accuracy of kinematics determined without palaeomagnetic data is much better than that of APW-paths. There is presented (Fig .9) the APlv-path for North America transfered to Eurasia according to the most recently clarified kinematics of the Atlantic (Westphal et a1., 1986). It is clear that the difference between the Eurasian and North-American APW-paths is large and statistically significant. The new data from Dagestan do not tend to diminish it. This discrepancy does not depend upon the choice of the APW-path for the North America, as all variants are ,vithin error-limits of mean poles (e.g. Harrison, Lindh, 1982). This fact was known for a number of years (Van der Voo, Zijderveld, 1969) and it was explained usually by poor quality of palaeomagnetic data or long-standing nondipole terms. But this discrepancy seems to be too systematic to be explained by the former hypothesis. The "nondipole" explanation has been disputed recently on the ground that all the Late Cretaceous poles are very tightly clustered throughout the North American platform (Stone, 1985). The same is true all the more so for Eurasia, where sampling localities are very wide-spread, from the Caucasus and the Pechora River to the Transbaikal region and the northeastern part of the Siberian platform. This problem has remained unsolved so far, but the crucial experiment can be carried out rather easily. As the European and North American plate had been near to each other before the opening of the Atlantic, some Cretaceous palaeomagnetic data should be obtained in the westernmost parts of Europe. If new poles (of course, very reliable) agree with the Eurasian APW-path and disagree with the North American one, both the "poor-quality-of-data" and "non-dipole" hypotheses will be ruled out. 5.2. Interpretation of Declination Data We should summarize the above results (Tables 2 and 3) for the Lesser Caucasus together with some structural data. The mean declination is 3540 +50 at the Amasia area and axes of the Alpine folds here strike east-west or deviate slightly to ENE-WSW. North-eastern declinations are characteristic for the eastern flank of the LCA. The most reliable result for locality E4 is D = 350 +50 • The Alpine structures here have slightly varying north-western strikes (Fig.lO). The difference in structural trends between the Amasia area and locality E4 is about 350 and the corresponding declinations differ by 440 +7 0 • There are known no post-Coniacian thrusts of noticeable magnitude within the studied part of the Lesser Caucasus and the observed declinations can not be explained by some systematic rotation of thrust-sheets. All sites and localities, which gave acceptable data are situated within one narrow tectonic zone (the Sevano-Akera zone) and any between-zone rotations would not affect our data. Rocks of 231 the same age and with the similar palaeomagnetic properties were studied at the Amasia area and locality E4 and this vector "fan" cannot be attributed to some disparities in ages of rocks and/or magnetization. So the secondary bending of the Sevano-Akera zone seems to be the only explanation of the observed data. All tectonic zones within the LCA have the approximately the same form and the above conclusion seems to be valid for the LCA as a whole. The result from the Vedi area (Pechersky, Nguen, 1978) supports this interpretation (Fig.lO). As for data from the Ijevan-Tauz basin (ibid.), they show the "banana-type" distribution of site-means with widely, from 3430 to 450 , scattered declinations and inclinations ranging only from 440 to 460 • Such a distribution was explained, after MacDonald (1980), by some local deformations (D.M.Pechersky, personal communication). Naturally, the latter data set could not be usedhere. Fig.lO: The Alpine structural pattern of the Caucasus and North-Eastern Turkey and the generalized palaeomagnetic direcitons. Solid arrows the observed data, dashed arrows the reference directions recalculated from Dagestan to each locality. A, El, E4 - this paper, V - (Pechersky, Nguen, 1978), M - overall mean for the Eastern Pontides (Van der Voo, 1968; Orbay, Bayburdi, 1979). See Fig.l for explanation of other letters.) Moreover, the Lesser Caucasus is the northern part of the Arabian syntaxis, and the secondary origin of thes giant arcuate structure has already been proposed (Argand, 1924; Adamia et al., 1979). This oroclinal model can be checked as it predicts, for example, the nroth-western declinations, about 3300 -3400 , at the western continuation of the Lesser Caucasus into North-Eastern Turkey. But as far as we know, such data have not existed up to now. It is worth pointing out that the LCA itself is outlined mainly by folds and faults of the post-Palaeogene age. Therefore the age of the arc is hardly older than deformations here, and the bending had taken place during Miocene-Pliocene. This dating is valid for the 232 northern part of the orocline only, as nearer to Arabia this process might start earlier. The lack of palaeomagnetic data precludes more reliable conclusions. There are some reliable Late Cretaceous palaeomagnetic data for the Eastern Pon tides (Van der Voo, 1968; Orba6 and Bayburdi, 1979). It was noted that palaeodeclinations about 340 inh this region were near to the African reference direction and deviated from the Eurasian one. We think that the counterclockwise rotation with respect to Eurasia took place during the general bending of structures in this region. As for agreement with the African reference direciton, it may be fortuitous. Such an explanation is disputable too and more data are needed, from younger formations especially, in order to solve this problem. The last question to be discussed is the quantitative estimation of rotations within the LCA. Here and further on we prefer to omit the poorly defined Late Cretaceous mean pole for Eurasia (Westphal et al., 1986). Instead of it the data from Dagestan can be used ass a reference direction. After recalculating the overall mean direction from Dagestan to each locality we obtained that the Amasia area was rotated 270 +70 counterclockwise, locality El was unrotated (6 0 +60 ) and locality E4 was rotated 170 +7 0 clockwise. No data exist which point to rotation of the Lesser Caucasus as a whole. Thus we can conclude tentatively that mass transportation during oroclinal bending was directed approximately along the modern meridian. 5.3. Interpretation of Inclination Data These are presented 18 site- and section-means from the Lesser Caucasus (Fig.11), 11 from this work and 7 from the Ijevan-Tauz basin and Vedi area (Pechersky, Nguen, 1978). The overall mean direction from Dagestan with its circle of confidence is shown too. All the data are recalculated to reference point at 4loN, 460 E. The means from the Lesser Caucasus clearly show the "banana-type" distribution, as a result of tectonic rotations of various kinds. Due to smearness of this distribution the Fischer's (1953) statistics can not be used. We assumed that prior to these movements the means had been distributed Fischer-like. Working on this assumption we calculated the mean inclination and its confidence limits by some slightly different methods (Kono, 1980; McFadden and Reid, 1982). These methods gave confidence limit values ranging from 1.30 to 1.90 while the mean inclination was always the same, I = 450 • Taking in to account the largest value and applying some corrections (Demarest, 1983) we found that the inclinations between Dagestan and the Lesser Caucasus difer by 80 +30 , which is statistically significant. 233 o· Fig.ll: The Late Cretaceous palaeomagnetic results from the Lesser Caucasus and Dagestan. Circles (this paper) and crosses (Pechersky, Nguen, 1978) - the site - or section-means from the Lesser Caucasus, triangle - the overall mean from Dagestan with its circle of confidence. Dashed line and striped band - the Lesser Caucasus mean inclination and its 95-percent confidence interval. This dif ference may step fran a number of ways. The shallowing of inclinations in the rocks from the Lesser Caucasus is one of them. However, some magmatic rocks were studied (Pechersky, Nguen, 1978), where such an effect was never observed. BeSides, it was shown recently (Laj et al., 1982; Bazhenov, 1983) that the shallowing of inclinations is minimal in limestones, if present et all. As the main body of data came from limestones this effect hardly affected the above result. Another possible source of disparity of inclinations is a primary til t of strata. But as soon as a tilt exceeded 20 _3°, unconsolidated carbona ti tic oozes would slide downslope. Besides, such tilts should be uniform within either Dagestan or the Lesser Caucasus. For example, all limestone strata should be primary til ted about 80 northward all over the latter region. It is clear that any non-uniform tilts within any region would result in an enlarged scatter of palaeomagnetic direc tions and not in some regular shift of them. Thus we can discard this effect too. The relative movements between the Great and Leser Caucasus seem to be the only explanation of the observed data. As Dagestan was welded to Eurasia in the Late Cretaceous time, we conclude that the Lesser Caucasus moved 80 +30 (900 km + 350 km) northward with respect to the Eurasian palaeolatitude grid. Though separate, these two landmasses have never been too far apart, say, some thousand kilometers. Any long-standing non-dipole terms of the geomagnetic field could hardly affect the above palaeolatitude estimations at such a "short" distance. 234 Contrary to magnitude little can be said about the age of this motion. The Palaeogene palaeomagnetic data though numerous are rather scattered and are of no help here. The inclination of postfolding component from locality El agrees very well with the prefolding magnetization allover the Lesser Caucasus, but the age of deformations is poorly constrained there. We can hypothesize only that this northward motion most probably preceeded the strong Neogene deformations in the Caucasus region. CONCLUSIONS The results of this study demonstrate that the northern part of the Great Caucasus domain has not suffered any palaeomagnetically detectable horizontal movements with respect to Eurasia since Late Cretaceous. On the contrary, such movements were widespread within the southern domain. The structur eof the Lesser Caucasus were bent as a result of indention of the rigid Arabian plate into more plastic crust of the Alpine fold belt, and the Lesser Caucasus as a whole was translated northward. We do not think that tectonic implications of our data are limited to those just listed. It would be interesting to compare them with various models of tectonic evolution of this region as well as with payaeomagnetic data from Iran and Asia Minor. We have not done it on purpose as these items are worth some special paper. Our main objective was to present new palaeomagnetic results, to discuss their reliability and to give their "self-evident" tectonic interpretation. Acknowledgements. We are very grateful A.L.Knipper and S.D.Sokolov for geological N.Ya.Drovorova for help in the laboratory. REFERENCES to Yu. V. Karyakin, consultations and Adamia Sh.A., Asanidze,B.Z., Gumbashidze,P.A., Nadareishvili,G.Sh., Nguen Tkhi Kim Tkhoa, Pechersky,D.M., 1979. 'Palaeomagnetism of the Upper Cretaceousrocks from Southern Georgia and its geological interpretation.' Izv. Akad. Nauk SSSR, ser. Geo- logicheskaya, N.5, 46-57 (in Russian). Argand,E., 1924. 'La tectonique de l'Asie.' 13th Intern. Geol. Congr. Compte rendus, 169-371. Basheleyshvili,L.B., Burtman,V.S., Gamkrelidze,I.P., 1982. 'On the nature of Suture between Ajaro-Trialet fold zone and Dzirula Massif.' Doklady Akad. Nauk SSSR, 226, 196-198 (in Russian). Bazhenov,M.L., 1983. 'Palaeomagnetism of the Upper Cretaceous and Palaeocene sediments of Kopet-Dagh: a correlation between lithology and palaeomagnetic direcitons." Izv. Akad. Nauk SSSR, ser. Fizika Zemli, N 8, 67-75 (in Russian). 235 Bazhenov,M.L., Shipunov,S.V., 1985. 'Early Cretaceous paleomagnetism of North Eurasia: new results and analysys.' Izv. Akad. Nauk SSSR, ser. Fizika Zemli, N 6, 88-100 (in Russian). Belousov,V.V., 1962. The main problems of tectonics. "Gosgeoltechizdat", Moscow, 608 p. (in Russian). Demarest,H.H. Jr., 1983. 'Error analysis for the determination of tectonic rotation from palaeomagnetic data.' J. Geophys. Res., 217, 295-305. Harrison,C.G.A., Lindh,T., 1982. 'A polar wandering curve for North America during the Mesozoc and Cenozoic.' J .Geophys. Res., 87, 1903-1920. Khalafov,A.A., 1985. 'Magnetic studies of Coniacian-Santonian rocks (from Kasakh ba sin.' Izv. Akad. Nauk AzSSR, ser. Nauk 0 Zemle, N 4, 123-126 (in Russian). Knipper,A.L., 1975. 'The oceanic crust in the structure of the Alpine folded belt'. "Nauka Press", Moscow, 208 p. (in Russian). Kono ,M., 1980. 'Statistics of palaeomagnetic inclination data.' J. Geophys. Res., 85, 3878-3882. Laj,C., Jamet,M., Sorel,D., Valente,J.P., 1982. 'First palaeomagnetic results from Mio-Pliocene series of the Hellenic sedimentary arc.' Tectonophysics, 86, 45-67. Lowrie,W., Alvarez,W., 1981. '100 million years of geomagnetic polarity history.' Geology, 2, 392-397. MacDonald,W.D., 1980. 'Net tectonic rotation, apparent tectonic rotation and the structural tilt correction in palaeomagnetic studies.' J. Geophys. Res., 85, 3659-3669. McElhinny, M. W., 1964. 'Statistical significance of the fold test in palaeomagnetism.' Geophys. J. Roy. astr. Soc., Q, 338-340. McFadden,P.L., Jones,D.L., 1981. 'The fold test in palaeomagnetism.' Geophys. J.R. astron.Soc., 67, 53-58. McFadden,P.L., Reid,A.B.- 1982. Aalysis of palaeomagnetic inclination data.' Geophys. J. Roy. astr. Soc., 69, 307-309. Milanovsky,E.E., Khain,V.E., 1963. Geological structure of the Caucasus. Moscow University Press, Moscow, 358 p. (in Russan). Orbay,N., Bayburdi,A., 1979. 'Palaeomagnetism of dykes and tuffs from the Mesudiye region and rotation of Turkey'. Geophys. J. Roy. Astron. Soc., 69, 437-444. Pechersky,D.M., Ngue;-Tkhi Kim Tkhoa, 1978. 'Palaeomagnetism of volcanic rocks from ophiolites and Late Cretaceous extrusives from Armenia.' Izv. Akad. Nauk SSSR, ser. Fizika Zemli, N 3, 48-63 (in Russian). Sirunian,T.A., 1981. 'Palaeomagnetism of Mesozoic rocks from Armenia.' Publishers: Akad. Nauk Arm. SSR, Erevan, 156 p. (in Russian). Sokolov,S.D., 1977. Olistostromes and ophiolite napes of the Minor Caucasus. "Nauka Press", Moscow, 96 p. (in Russian). Stone,D.B., 1985. "Megaplates, Microplates and Palaeomagnetism.' J. Geomayn. and Geoelec., 37, 147-152. Van der Voo,R., 1968. 'Jurassic, Cretaceous and Eocene pole positions from northeastern Turkey.' Tectonophysics, Q, 251-269. 236 Van der Voo,R., French,R.B., 1974. Atlantic-bordering continents: Earth Sci. Rev., 10, 99-119. 'Apparent polar wandering for the Late Carboniferous to Eocene.' Van der Voo,R., Zijderveld,Z.D.A., 1969. 'Palaeomagnetism in the western Mediterranean area.' Koninkl. Nederlandse Akad. Wetensch. Verh., Gen., 26, 121-138. Westphal,M. ,-Bazhenov,M.L., Lauer,J.P., Pechersky,D.M., Sibuet,J.c., 1986. 'Palaeomagnetic implications of the evolution of the Tethys belt from the Atlantic ocean to Pamirs since Trias.' Tectonophysics, 123, 37-82. 237 Table 1:% Palaeomagnetic data from limestones of Late Cretaceous age from Dagestan (lat 430 N, long 460 E) L(S) th AO dO n /n DO 0 10 K d95 P PIn PH plo Dl(l) 100 25 40 20/17 21 52 103 3,3 0 Dl(2) 150 205 40 24/22 16 57 90 3,2 27 D2(l) 250 230 24 21/18 19 58 91 3,5 33 D2(2) 120 40 30 13/8 22 55 33 8,6 0 Mean (samples) 65 19 56 75 2,0 151 74 36 Mean (sections) 4 20 56 712 2,6 151 74 36 Strongly magneti- 23 18 56 63 3,7 zed samples Weakly magnetized 42 20 56 82 2,4 samples Age G r o u p s* Cenomanian 12 19 53 58 5,3 0 33 Turonian- 43 20 56 75 2,5 23 36 Sanotnian Campanian 10 13 58 154 3,6 80 38 * - statistics on sample level L(S)=the numbers of localities (sctions), th=the true thickness studied, in meters, A,d = the mean azimuth and angle of dip, respectively, n /n=number of samples taken/accepted, D, I = the mean declination and iRclination after thermal cleaning and tilt-correction, K = the concentration parameter, 95=the semi-angle of the cone of 95% confidence, P=the percentage of the reversedly magnetized samples, PIn (oE), PIt (oN) = the longitude and latitude of the North palaeomagnetic pole respectively, pI = the palaeolatitude of the area according to the dipole model, in degrees. 238 Table 2. Palaeomagnetic data from Amasia area (the western flank of the Lesser Caucasus, 1at 41oN, long 44oE) s A2 A3 A5+A6 A7 A8 th 20'~ 25 25 25 115 150-260* 280-20 5-260 150-180 10-45 Mean (samples) Mean (sections) Mean (groups)'~* A2 A3 A4 A5 A6 A8 A9 5 5 350 20 Mean (sections) K Grey limestones 10-60* 5-25 10-30 30-70 10-35 14/12 9/6 3/3 8/8 12/9 38 5 8 355 355 349 354 352 44 38 49 41 43 35 6,8 67 7,0 28 15,3 65 6,6 51 6,5 354 43 46 3,3 353 43 330 3,4 354 43 335 2,7 Red limestones 65 85 4/4 5/0 4/4 6/4 8/8 3/2 6/6 7 24 37 34 4 346 55 341 47 329 40 33 1 12 34 215 4,8 125 6,3 19 16,2 38 8,1 11 17,5 7 20,2 16 16 100 12 o 25 25 25 ':' these parameters are given once for each section as a whole, for grey and red limestones are interbedded. Two numbers for AO and dO denote the limits of their variation within each section. ** the very group means are not presented for lack of space. For explanation of symbols see Table 1. 239 Table 3. Palaeomagnetic data from the eastern flank of the Lesser Caucasus (mean coordinates of this area: lat. 4loN, long. 460 E). Table L(S) El'~ El':"~ th 130 130 n /n o 28/13 28/14 K 37 71 6,4 4,4 P 100 o 26 27 E2(1) E4(1) E4(2) E4(3) E4(4) 50 175 85 10/8 35 41 19 11,5 12 24 75 35 150 30 50 135 40 180 Means for samples loc.E4 (sections) 50 12/10 40 45 55 60 10/6 40 44 39 11 7/7 30 48 69 30 9/8 37 44 39 6,0 9,2 6,4 8,0 31 4 37 45 50 3,6 37 45 460 3,3 * component A (see text) before tilt-correction. ** component B (seex text) after tilt-correction. 40 33 14 25 27 27 For explanation of symbols see Table 1. The parameters AO and dO for gently folded rocks at locality El vary within 00 -3600 and 100 _350 respectively. 4. Palaeomagnetic poles for Dagestan, Eurasia and North America Region Age Plt, oN PIn, °E Dagestan Cenomanian 72 159 Dagestan Turonian-Santonian 74 151 Dagestan Campanian 79 152 Eurasia E. Cretaceous 73 154 Eurasia Palaeocene-Eocene 76 165 N .America* E. Cretaceous 68 208 N .America;~ M. Cretaceous 69 211 N .America* L. Cretaceous 73 212 N .America':' Palaeocene-Eocene 83 159 * These poles were rotated to Eurasia according to the kinematics of the Atlantic (Westphal et al., 1985). A95 6 4 5 4 3 6 7 8 5 PIt, PIn - latitude and longitude of palaeomagnetic poles, o A95 ' circle of confidence at 95-percent confidence level. TETHYS EVOLUTION IN THE AFGHANISTAN-PAMIR-PAKISTAN REGION Jovan Stocklin Geological Consultant Erdblihlstrasse 4 8472 Seuzach (Zurich) Switzerland ABSTRACT. The Afghanistan-Pamir-Pakistan ranges resulted from compression of the Tethyan realm. Continental basement rocks under cover of shallow-marine sediments indicate that the Tethys was largely an epicontinental sea. Early-Middle Paleozoic and Mesozoic oceanic environments are indicated by ophiolitic associations in a Paleo tethyan suture in the north and a Neotethyan suture in the south, which divide the epicontinental realm into three major domains. The Northern Domain represents the south margin of Paleo-Asia affected by Hercynian and Indosinian orogenic processes. Several large blocks constitute the Central Domain, in which Early Cretaceous and Early-Middle Tertiary folding and magmatism played a dominant role. The Southern Domain occupies the Mesozoic shelf of the Indian subcontinent folded and imbricated in the Neogene. In the plate-tectonic concept the fundamental orogenic mechanism is continental collision resulting from successive detachment of the Central and Southern Domains from Gondwanaland and their north-drift across a wide Tethys Ocean; folding started with Late Triassic collision along the Paleotethyan suture, to be followed by Early Tertiary collision along the Neotethyan suture, with a possible intermediate collision event in the Early Cretaceous along a poorly documented additional suture bisecting the Central Domain. The model requires large-scale subduction of Tethyan oceanic crust to compensate spreading in the Neotethyan and Indian Oceans. Subduction is held responsible for the calc-alkaline magmatism of the region. The author's main objection against this model is the lack of geological evidence for the existence of the vast Late Paleozoic Tethys Ocean supposed to have been available for Mesozoic subduction. An excess of crustal expansion in the Indian Ocean over crustal shortening in the orogenic belt, tantamount to an expansion of the Earth, is concluded. 1. INTRODUCTION In the region here reviewed (Fig.l), the Tethyan orogenic belt bends in a tight mountain arc around the northern tip of the Indian continent, 241 A. M. C. !jengor (ed.), Tectonic Evolution of the Tethyan Region, 241-264. © 1989 by Kluwer Academic Publishers. 242 6ULF OF OMAN Fig. 1. Tectonic sketch map of Afghanistan-Pamir-Pakistan region 1 . Northern Domain: Late Paleozoic - Late Triassic folding, Early Cretaceous and Plio-Pleistocene reactivation I· .. : . :1 I~I I x x x I ~ Molasse-filled Neogene foredeeps, intermontane basins Mesozoic-Paleogene intracratonic sedimentary basins 200 MA and older granitea Paleozoic-Triassic sedimentary and volcanic rocks, including zones of pre-Middle Carboniferous deep-sea sediments and ophiolites Central Domain: Early Cretaceous and Tertiary folding I ~Iv I /71'"" Itl~+ I -/\- + 1111111111 l1lllll1J:1 -- Quaternary volcanoes Cretaceous and Tertiary volcanic rocks and granites (including some Cambrian and older ~ranites) Mesozoic shelf and intracratonic basin sediments Waras ophiolites (? Triassic) and Panjao-Roshan- Pshart flyschoid sediments (Triassic-Jurassic) Precambrian continental-type basement and Paleozoic epicontinental sedimentary cover Axial Ophiolite Belt: Paleocene ophiolite-nappe emplacement, Tertiary folding (Upper Cretaceous -) Paleogene flysch, in places under Neo~ene molasse cover "Transhimalayan" ~ranites and diorites, Cretaceous Ophiolites, ? largely Cretaceous Cretaceous ophiolitic melan~es and melange-like volcano-sedimentary associations Kohistan sequence: amphibolites, greenschists, meta- gabbros, etc. (? lar~ely Cretaceous) Mesozoic limestone turbidites and pela~ic sediments, including Cretaceous pillow lavas, diabases, tuffs Southern Domain: Neogene folding Molasse-filled Neogene foredeeps Mesozoic-Paleo~ene shelf sediments Precambrian continental-type basement and Paleozoic sedimentary platform cover Cambrian and older ~ranites (including some Late Tertiary granites in Himalaya) Himalayan nappe fronts 243 244 converging from east and west in the gigantic "Scharung" of the Hindukush-Pamir-Karakorum system. Here, the north-push of India finds its clearest manifestation with extreme arching, crustal compression, metamorphism and granitization. The state of geological knowledge varies greatly. The core of the region is one of the remotest and highest mountain countries in the world, culminating in the Karakorum peaks of 8000 meters altitude; information from this part is fragmentary, and interpretations are accordingly difficult and speculative. Other, more accessible parts like the Helmand Valley in Afghanistan or the Salt Range in Pakistan have been key areas for detailed investigations. A prominent tectonic feature is the transcurrent Chaman-Arghandeh-Panjshir fault, believed to be an onshore extension of the Owen-Murray fracture zone in the Indian Ocean. It bisects the region into two very unequal parts, with the high mountain country in the east and the less prominent ranges and large depressions of Central Afghanistan and Baluchestan in the west. The fault has a large sinistral shear, disrupting the structural continuity and making correlations across the fault difficult. The sedimentary record of the region ranges from Infracambrian to Neogene. A major restructuration of the Tethys realm in Triassic time allows its history to be divided into a Paleotethyan and a Neotethyan Era. 2. THE BASEMENT PROBLEM Outcrops of crystalline basement rocks are widespread. It is worthwhile recalling, though, that their Precambrian age is more often inferred from high metamorphic grade than ascertained from stratigraphic position or radiometric work, and that the criterium of high-grade metamorphism may be misleading. Blaise(2), has demonstrated lateral transitions of fossiliferous Paleozoic sediments into crystalline schists reaching the sillimani te grade in grani tized shear zones of Central Afghanistan. Substantial portions of the meso- and katazonal rocks of the central Karakorum are likely to represent metamorphosed Paleozoic and possibly even Mesozoic sediments(3). Proof of Precambrian age is best in Central Afghanistan, where dated sediments as old as Cambrian rest unconformably on a metamorphic complex, a considerable part of which consists of low-grade metasediments of probably Late Proterozoic age(4). In Hazara of northern Pakistan, the presence of a thick sequence of non-metamorphic, paleontologically dated Cambrian and Infracambrian sediments(5) suggests Precambrian age for metamorphic rocks in the vicinity. Radiometric evidence for ages as old as 2000 MA in gneisses and granites of the western Himalaya is accumulating(6). From these and similar results in Iran, Late Precambrian ("Baikalian") basement consolidation is assumed for most of the region. 245 This statement requires a qualification: Precambrian age of certain crystalline rocks in the West Hindukush - North Pamir foldbelt appears to be wholly conjectural; at least, the author is not aware of confirmation by isotopic work or unequivocal stratigraphic evidence. In fact, Hercynian-Indosinian deformation, metamorphism and granitization have imparted the characteristics of a consolidated basement to large parts of the Paleozoic sequence in this northern area. 3. THE PALEOTETHYS Usage of the term "Paleotethys" in recent publications has been inconsistent. While all authors agree in applying it to a Tethys of Paleozoic age, some(7) extend it to a Mesozoic Tethys as young as Late Jurassic, and many restrict it at least imp1icite1y to "oceanic" realms of the Tethys. In this paper, "Paleotethys" designates the Tethyan realm prior to its restructuration in Triassic time and comprising not only deep oceanic but also, and mainly, extensive epicontinental environments. 3.1. The epicontinental realm In fact, the Paleozoic sediments of most of the region under review (all except pre-Middle Carboniferous sediments in the extreme north) attest to a shelf-type epicontinental regime (Fig.2). Shallow-water carbonates ranging from Infracambrian to Triassic predominate but are frequently interrupted by plainly continental deposits or sedL"entary gaps. The largest gap is found in the southern marginal foldbelt (Salt Range) and in the subsurface of the unfolded Indus foreland (Mu1tan), where Permian tillites rest directly on Cambrian sediments. The latter are underlain by thick Infracambrian salt forming the base of the sedimentary sequence (Fig.2). With increasing distance from the Indian shield the sequences become more comprehensive but show notable differentiations. In Central Afghanistan, nearly complete Paleozoic sections, 3000 m thick and comprising an important portion of sandy-argillaceous material, are found in ancient intracontinental grabens(8) in close vicinity to ancient horsts with a thin and incomplete Paleozoic cover. In northern Pakistan, paleontological evidence, mostly from limestones, has come forth for the presence of all major Paleozoic systems(9), but the evidence is widely scattered and severely limited by metamorphism. Probably the thickest and most comprehensive Paleozoic sequence, comparable to the graben deposits of Central Afghanistan, is present in the "Karakorum Slates", which are largely pre-Permian but otherwise poorly dated. Unconformities suggestive of significant Paleozoic deformation are lacking in all these occurrences, and Paleozoic magmatic rocks are rare. 246 3.2. The oceanic Paleotethys Conditions are quite different to the north of the Herat-Panjshir-Akbaytal lineament, an important fault zone accompanied, in the Ghorband Valley and further northeast, by lenses of ultrabasic and other ophiolitic rocks, which the author(l) chose as south-limit of a "Northern Domain" (Fig. 1). From the West-Hindukush sector of this northern foldbelt, Boulin and Bouyx(lO) described a marked unconformity at the base of the Middle Carboniferous (Namurian B), separating Upper Paleozoic shallow-water deposits from a thick older sequence of argillaceous, siliceous and fine-clastic sediments associated with considerable amounts of basic to acid volcanic material. Precise dating of this older sequence has not been possible, but a broadly Early-Middle Paleozoic age is indicated by the presence of a dated Devonian limestone near the top and of uncontestable though indeterminable organic remains (molluscs, radiolaria) in the lower part. These rocks are folded and metamorphosed to varying degrees and intruded by pre-Namurian granites. Continuing tectonic unrest since the Namurian is indicated by repeated unconformities and volcanic manifestations in the Upper Paleozoic - Triassic section. The strongest deformation, accompanied and followed by granite intrusions, has affected the belt in the Late Triassic (Indosinian folding). The belt as a whole continues from the western Hindukush northeast to the northern Pamir, where diabases, spilites and other volcanic rocks are associated with well-dated Early Carboniferous and poorly dated older sediments (Fig. 2). An oceanic orlgln has been inferred for the pre-Middle Carboniferous rocks from the apparent deep-water facies of the sediments and from their association with the volcanic material, to which the Ghorband ophiolites are thought to be related. Two facts must be stressed, however: 1) The deep-water facies does not extend beyond the Lower Carboniferous; the Upper Paleozoic sediments are predominantly platform and reefal carbonates and shallow-marine to continental clastics. In the Herat area as in large parts of Soviet Central Asia, continental red sandstones and conglomerates of Permo-Scythian age reach a thickness of several kilometers. 2) The Lower-Middle Paleozoic deep-water facies is by no means characteristic of the entire Northern Domain but interrupted by zones of epicontinental type, in particular the uplifted Javai-Beleul axial zone in which thin Permo-Carboniferous sediments rest directly on a highly crystalline basement complex(14). 4. THE NEOTETHYS Rifting in Triassic time split up the vast platform area south of the Herat-Akbaytal lineament into several large blocks. The main rift NORTHERN PAHIR (Barkhalov, 1963) ............... ~.:!:-..:.;;~:~ .. y:~ ~·~Tv·~~.,;~ ·V,,/,/vvvvi./ vv~vv\\V///1 ~p V",,"yll,,"yVV"VVyv 'P vvvvvvvv" U ~~~¥~i~ CENTRAL PAMIR (Sorkhatov, 19.3) a: u 5 km N - KARAKORUM (Desio, 1963,1977) w = .. .. ;.;~ .... .;. ........ I I I SALT RANGE (Ibrahim Shah,1977) Q. ~ ~ l22l ~ ~ ~ ~ tE::EE ~ ~ Irrrl~~~! rr r L. L~ IvVvYVl ~ m Con9lomera~e Sand ,t., .ilt.t. Argillite Shale, phyllite ArSil!. lim .. t. Limes~.) sandy Dolomite Salt / gyp,um VolcanICs, ~Uff5 (rys~alline rocks Plon~ remains Tillites Redbeds sp Spilit" Fig.2. Generalized stratigraphic sections of the Paleozoic-Triassic. 247 E U R A S IA N P R O V IN C E NO RT HtR N DO MA IN NO RT H AF GH AN - TA JIK D EP RE SS IO N (A nd reo v . t 0 1., 19 72 Ku lak ov . t a l., 19 U ) ~ ~ ~ ~ ~~:2 -;B' ~JR ::.~ ~::. ~:.; g.:- .-' p . . , " " " '" a . IIT;:: tt~~'~ ~;~ R FO R LE GE ND S EE F IG . 2 C EN TR AL D O M AI N FA RA HR UD BL OC K (W eip pe rt et a I. , 1 97 0 B la is .'. t 01 .) 19 78 ) ;! ~ r }iirf r~'~~ ~B R H£ LM AN D BL OC K (W eip pe rt . t a I., 19 70 ) ~ ~ ~~ i~ R ~~~ '/1 ]g\ sp " " sp I- ... J W QQ u .J I- ... J o :< : " - C > « x « G 0 N D W A N I A N P R O V I N C E SO U TH ER N D O M A IN KI RT HA R - 50 LE IM AN RA NG ES (Ib ra hi m S ha h, 19 77 ) ~ TR AN S- IN DU S RA NG ES (Ib rah im S ha h, 19 77 ) ~ F ig .3 . G en er al iz ed s tr a ti g ra p h ie s e c ti o n s o f th e M es o zo ic . ~ 00 249 initiated opening of a Neotethyan oceanic trough, recorded in the rock material of the "Axial Ophiolite Belt" (Fig. 1). It separated a "Southern", Gondwanian domain (India) from a "Central" Domain now forming a marginal part of Eurasia. South and north of the Neotethyan trough, in the Gondwanian and Eurasian provinces respectively, the shelf-platform conditions of the Late Paleozoic continued into the Mesozoic but showed marked contrasts between the two sides (Fig. 3). 4.1. The Gondwanian shelf The Gondwanian shelf deposits are widely exposed in the Kirthar-Soleiman-Pontwar-Hazara foldbelt, part of the "Southern Domain" of Fig.l. An increase of Mesozoic shelf subsidence with approach to the oceanic trough is indicated by a change from a thin, lacunary, carbonatic and glauconite-rich sandy sequence (e.g., in the Trans-Indus Ranges) to an apparently continuous limestone/shale alternation rich in argillaceous matter, several thousand meters thick, in the so-called axial zone of the Kirthar-Soleiman Ranges (Fig. 3). Still, in spite of increasing rates of subsidence towards the oceanic trough, the frequent recurrence of reefal carbonates, oolitic and glauconitic beds and of neri tic faunas indicates deposition under relatively shallow waters. Towards the ophiolite belt, agglomerates and pillow lavas appear for the first time interbedded with Upper Cretaceous pelagic sediments. Sedimentation was also strongly controlled synsedimentary block faulting, which according to Auden(lS) may have been continuous since the Early Cretaceous and caused drastic facies and thickness changes across grabens and swells running nearly perpendicularly to the continental margin. Marine sedimentation ceased in the Middle Eocene in the north and in the Early-Middle Miocene in the south. In the Upper Miocene-Pliocene, thick continental-fluviatile molasse deposits (Siwaliks) accumulated in foredeeps in front of the rising orogen. Folding took place only in the Pliocene-Early Pleistocene. The moderate folding in the Kirthar-Soleiman-Potwar belt contrasts with the intense contemporaneous thrusting, metamorphism, granitization and colossal uplifting in the Himalaya further east - an integral but very different element of the continental margin of India. 4.2. The Eurasian Province Much stronger Mesozoic - Early Tertiary tectonic acti vi ty than in the Gondwanian province manifested itself in the vast Eurasian province of the Tethys (the Central and Northern Domains of Fig. 1). The Triassic tensional movements which triggered rifting in the ophiolite belt caused also a fragmentation of the platform in the Central Domain with additional rifting along the Waras ophiolitic line, whereas compressional movements in the Northern Domain culminated in strong Late Triassic folding, accompanied and followed by important granite intrusions. Up-faulting and up-folding related to this tectonic phase 250 initiated the formation of some of the high mountain structures in the north like the West Hindukush - North Pamir foldbelt and the Nouristan Southwest Pamir Block, which remained permanently emerged. Coal-bearing continental to paralic deposits accumulated in large Jurassic depressions in the north, whereas carbonate deposition prevailed through the Jurassic in smaller depressions in the Helmand Block and in the Central Pamir - Karakorum region of the Central Domain. Jurassic subsidence was strongest in the Farahrud Block of Central Afghanistan (Panjao "flysch"). Tectonic unrest characterized the Eurasian province again in the Early Cretaceous. Tight folding and even pressure metamorphism affected the Panjao-Waras trough. Elsewhere, faul ting, tilting, and regional uplifting with regression of the sea took place. Volcanism in the Kandahar area followed these tectonic events in the late Early Cretaceous, and important granite batholiths were emplaced at that time in the Helmand Block, in the central and southeastern Pamir, and in the ~entral Karakorum(20). A widespread transgression with deposition of shallow-water limestones and basinal shales invaded the depressions since Barremian time and extended north onto the Turkmen platform and into the North Afghan - Tajik depression (Fig. 3), reaching, however, some of these northern parts only in the Late Cretaceous. Marine conditions ceased by the end of the Cretaceous in the Central Domain but persisted in the northern depressions to the Eocene, when the Tethys Sea finally retreated from the region. Still, subsidence continued in a number of intermontane and northern foreland basins, which received molasse-type continental and lacustrine deposits in the later Tertiary. While significant folding in the Wakhsh-Transalai ,Kashgar frontal ranges, which surround the Pamir bulge in the north, took place only in the Pliocene - Early Pleistocene, the lack of stratigraphic control makes precise dating of Tertiary events in the Central Domain difficult. Paleocene and Early Oligocene movements were important in neighbouring Central Iran and by inference can be assumed for Central Afghanistan; the extensive volcanic complex south of Herat near the Iranian border may be related to them. Early Tertiary granites are found in the northern granite belt of the Karakorum. Granite intrusions dated radiometricall y as Miocene occur in the Helmand Block. The powerful rhyodacitic to andesitic volcanism of Dasht-e-Nawar, west of Qazni in the Helmand Block, has been dated as Late Pliocene Early Quaternary(2l). Warping of Quaternary lake deposits, active faulting and strong seismicity attest to continuing tectonic activity particularly in the Northern Domain. 4.3. The Axial Ophiolite Belt Sediments believed to have accumulated in continental-slope and deep oceanic environments, together with ophiolites interpreted as remnants of oceanic crust, constitute the Axial Ophiolite Belt (Fig. 1). The Belt extends from the Iranian to the Pakistanian Makran and further northeast 251 through the Las Bela, Zhob and Katawaz areas to south and east of Kabul. A zone of mainly greenschists and amphibole-rich metamorphic rocks, the "Kohistan sequence", extends from east of the Kabul spur through Swat to Gilgit in northern Pakistan and seems to form the link with the ophiolitic Indus suture of the western Himalaya. In the sector between Karachi and Kabul, most characteristic of the Belt are extensive ultrabasic and basic rocks exposed in a number of large, disconnected outcrops near the eastborder of the Belt, in the Las Bela, Muslimbagh, Waziristan and Khost areas. The main rocktypes are more or less serpentinized peridotites with accessory amounts of dunite, pyroxenite, gabbro, pillow lavas and tuffs. Studies in the Muslimbagh area by Gansser 22 and Allemann 23 have shown that these ophiolites overlie with tectonic contacts a highly tectoni zed and lithologically complex sequence, in which deep-water sediments such as various shales, red and green radiolarites and micritic limestones containing planktonic foraminifera alternate wi th resedimented shelf material such as bioclastic and oolitic limestones, limestone turbidites, calcarenites and limestone conglomerates with debris of benthonic fossils. The planktonic foraminifera indicate Early to Late Cretaceous(Maestrichtian) age, the oldest exposed beds probably ranging down to the Jurassic. While in a section southwest of Quetta these sediments were found to be entirely free of volcanic material, abundant doleri tic flows, pillow lavas and tuffs are interbedded further northeast with the Upper Cretaceous sediments, disappearing, however, in the pre-Cenomanian section. Older, Triassic-Jurassic sediments of similar lithology, some 2500 m thick but apparently lacking volcanic admixtures, were reported from the Khost area in the northern extension of this zone(24). The close association of pelagic andre-sedimented shelf material in these Mesozoic sediments is strikingly similar to that described from the Hawasina Group of Oman(25). Like in Oman, these sediments are thrust together with the ophiolites upon the Gondwanian shelf deposits discussed previously. In the Altimour Range south of Kabul, the same rock associations display a more chaotic, melange-like structure, in which large blocks of peridotite and of massive Permo-Triassic limestones are particularly characteristic; the latter are apparently derived from the shelf carbonates that overlie the crystalline basement around Kabul(26). Upper Paleocene or Eocene shallow-water limestones overlie transgressively the ophiolite nappes and pass northwestward into the flysch facies of the Katawaz basin. The Katawaz flysch, like its equi valent in the Makran, grades upwards into molasse-like sandstones and coarser clastics of a shallower environment, ranging into the Neogene. The intensely folded sequence, which contains in its lower part also volcanoclastic material, reaches a thickness of 6-8000 m. The flysch zone is sharply limited in the west by the Chaman Arghandeh fault, where ophiolites and volcanic rocks appear again. In the Kandahar area a clac-alkaline volcanic complex of diabases, spilites, 252 keratophyres and andesites appears between the Chaman fault and the faulted border of the Helmand Block. The volcanics are associated here with upper Jurassic-Lower Cretaceous shales, tuffs and limestones under an unconformable cover of shallow-water limestones that are Aptian in the north and Upper Cretaceous in the south(24). The volcano-sedimentary complex is thought to represent an islandarc environment(27). The continuation of the Axial Ophiolite Belt in northern Pakistan is generally identified with the amphibolitic "Kohistan sequence"(28), although very few typical ophiolites are found in it. Some slices of serpentinized peridotite and pyroxenite occur along the southern boundary fault, the so-called "Main Mantle Thrust", along which the sequence is thrust steeply south onto Indian shield elements. Hornblende-rich gabbros, diorites and gneisses, metamorphosed to the amphibolite and in places to a high-pressure granulite facies, form the lower part. The upper part consists of calc-alkaline volcanic rocks and metasediments attributed to the Cretaceous on the basis of rare Orbitolinas. The sequence, which is invaded by late- and post-kinematic granodiorites, lacks the characteristics of either continental (sialic) or typical oceanic crust and has been interpreted as a crustal section of an island arc, with remnants of mantle material preserved in the basal serpentinite slivers. The thickness may be 15 km or more. 4.4. The Waras Ophiolites Ophiolites have also been reported from the fault zone that separates the Farahrud from the Helmand Block in Central Afghanistan. The main information on them(19) comes from the Waras area near the highly tectonized eastern termination of the Farahrud zone, where the ophiolites and the accompanying Panjao flysch wedge out between the converging Herat and Farahrud faults (Fig. 1). The ophiolites consists of spilites, pillow lavas and tuffs interbedded with flysch-type sediments containing also some radiolarite, and of disrupted bodies of gabbro and serpentinized peridotite. Associated are large olistoliths of massive crystalline and crinoidal limestones of Permo-Triassic habitus. Metamorphism, increasing southward to the garnet-amphibole grade, has affected the whole assemblage. The volcano-sedimentary sequence has yielded a few foraminifera of Middle or Late Triassic age in a nodular limestone said to be "interbedded". The authors had reservations about a generalization of this age. In fact, the occurrence of Upper Triassic Megalodontids in some 01istoliths(29) must throw doubts on the indigenous nature of the Middle-Late Triassic forams, on which the whole age argument is based. The ophiolite-bearing Waras rocks are fault-bounded in the north against the Jurassic Panjao flysch, which is free of volcanic material, but minor extrusive and intrusive products are found associated with the unconformably overlying Cretaceous sediments. The Waraz ophiolite zone has been interpreted as an intermediate ophiolitic suture, situated half-way between the Paleo tethyan suture in the north and the Neotethyan suture in the south and splitting the Central Afghan Block into the Farahrud and Helmand sub-blocks. 253 Karapetov et al. (30) connected the Farahrud zone with the discontinuous Roshan-Pshart zone that intervenes in lenticular form between the Central and South Pamir blocks (Fig. 1). Descriptions of this zone are so contradictory that any interpretation cannot be but utterly speculative. For instance, Karapetov et al.(30) reported a thick volcanosedimentary complex containing Triassic coral limestones to be overlain by thick Jurassic flycsh, whereas Dronov and Abdullah in a more recent report(14) stated that post-Triassic deposits have not been found as yet in this zone. 5. PLATE-TECTONIC INTERPRETATIONS The Afghanistan-Pamir-Pakistan ranges are a classical example of a compressional mountain arc, whose origin has since long been related to the Tethys. For Argand(31), Staub(32) and many of their contemporaries, the arc originated from compression and inversion of a Tethys geosyncline between the converging continental masses of Eurasia and India. In current platetectonic concepts, Eurasia/India convergence is a still more important postulate. In these modern interpretations, the Tethys is no longer a mere geosyncline but a vast ocean, some 6000 km wide, and Tethys compression has taken the form of subduction and consumption of enormous volumes of oceanic crust, climaxing in closure of the Tethys Ocean with collision and compression of the converging continental margins. While collision has thus become the fundamental orogenic mechanism, subduction has been made responsible for the calc-alkaline magmatism of the orogen. In the orogenic evolution of the region here reviewed, several magmatic/tectonic phases are distinguished and have been related to corresponding subduction/collision events. The latter in turn have been explained by successive detachment of several continental fragments from Gondwanaland, their north-drift across the Tethys Ocean, and their accretion to Paleo-Asia. The interpretations by different authors, however, differ in important details. Some examples are discussed in the following. Bordet(27) concludes from paleoclimatic data a first suturing (collision) event in Early Carboniferous time along the Herat-Panjshir-Akbaytal fault. These data suggest a separation of at least several degrees of latitude between the northern and southern lips of the fault in Early-Middle Paleozoic time and their junction since the Carboniferous. This conclusion is supported by the sedimentary record, which, as discussed above, seems to indicate deep oceanic environments for the Early-Middle Paleozoic in parts of the Hindukush-North Pamir belt just north of the fault. Further confirmation appears to be available from biogeographic data(33), which suggest that Central Afghanistan at that time formed the Gondwana margin of the Tethys. The pre-Namurian granites of the Hindukush-North Pamir belt could thus be related to subduction of the Paleo tethyan oceanic realm, and the Hercynian diastrophism of the belt to subsequent collision. Blaise et al. (34) distinguished Devonian collision in the Hind ukush from Visean collision in the Tienshan further north. 254 Tapponnier et al.(24), following Stocklin(l), attach more importance to the 210 MA granites and to the pronounced Indosinian folding of the Hindukush-North Pamir belt and take these as evidence for Triassic subduction and Late Triassic collision along the Herat-Panjshir-Akbaytal suture. In the view of Tapponnier et al., subduction then migrated south to the Waras ("Panjao") oceanic belt which they believe to have opened by north drift of the Farah Block. For Bassoulet et al. (29), the 210 MA granites of the Hindukush indicate, not Triassic subduction terminated by Late Triassic collision, but the beginning of Jurassic subduction ending with Early Cretaceous collision. The Jurassic Panjao flysch "attests to the resorption of the (Waras) oceanic domain" (l.c., p.19l), while Early Cretaceous suturing is thought to be indicated by the important folding event of that age in Central Afghanistan. Subduction is seen as directed to the north in the early stages (Early Jurassic granites of the Hindukush) and to have switched to the south in the later stages (Early Cretaceous granites of the Helmand Block). The main difference between the views of Tapponnier et al. (24) and Bassoulet et al. (29) lies in the interpretation of the Farahrud zone: the former consider it as a continental fragment ("Farah Block") limited by the Herat suture in the north and the Waras suture in the south, whereas the latter consider the whole space between these two ophiolite lines as an oceanic basin, taking the dominant "flysch" facies (Panjao, Waras, Turkman) as principal evidence. Regardless of such differences in interpretation, it is assumed in all models that north drift of northern Gondwana fragments in Central Afghanistan, the Pamirs, the eastern Hindukush and the Karakorum caused closing of older Tethyan oceanic basins in their front and simultaneous opening of an oceanic Neotethys in their rear, the remnants of the latter now being preserved in the Baluchestan-Katawaz-Kabul-Kohistan ophiolitic belt. It is also widely accepted that opening of the Neotethys Ocean began in Triassic time, though Bordet(27) believes that its genesis may be much older, perhaps as old as 400 MA. In the Karachi-Kabul sector the main subduction and collision scar (and, thus, the edge of the Indian continent) is generally identified with the Chaman-Arghandeh fault. Between this fault and the fault-bounded Helmand Block appears the wedge-like volcanic zone of Kandahar, in which Bordet(27) distinguished from NW to SE a foredeep, a volcanic arc, and the main trench bordered by the Chaman fault. The Kandahar volcanics, a typical calc-alkaline association, vary in age from Late Jurassic in the north to Albian and possibly to later Cretaceous in the south. From this it has been inferred that subduction at the Chaman line began in the Late Jurassic - Early Cretaceous, at the time when collision terminated subduction at the Waras line. Much uncertainty remains with regard to the time of closure of the Neotethys Ocean and collision of India with Eurasia. An important tectonic event which could suggest collision was the obduction of the large ophiolite nappes (Muslimbagh, Khost, etc.), which can be reliably 255 dated as Paleocene. However, plate reconstructions based on paleomagnetic data suggest that India at that time was still far away from Eurasia. Also, continuing subduction during Eocene has been inferred from Eocene granite intrusions west of the Chaman fault. Molnar and Tapponnier 35 estimated that intimate contact between the two continents was not established before the Late Eocene. (about 40 MA). This implies that subduction in its final stages took place beneath a thick prism of simultaneously accumulating flysch sediments in the Zhob-Katawaz basin. If, as paleomagnetic data require, India drifted north another 1500-2000 km after Late Eocene collision, this drift must have been accomodated by post-Eocene shortening of continental crust in the orogenic belt north of India. Since shortening by post-Eocene folding and thrusting in the Himalaya is unlikely to have exceeded a few hundred kilometers, Molnar and Tapponnier(35) proposed that accomodation was largely achieved by a strike-slip mechanism, by which large wedge-like crustal blocks were squeezed to the west and east out of the way of the advancing Indian continent. In this way the wedge-shaped Central Afghan Block between the sinistral Chaman and the dextral Herat faults would have been pushed to the southwest, symmetrically to an eastward squeeze of the Tibetan Block. Subduction in East Iran and the Makran would have facilitated the southwest push of Central Afghanistan. Subduction, anyhow, is believed by many to have continued to the present by southward migration to an active subduction zone of the Makran coast, where collision evidently has not yet occurred(36). Metamorphism and the almost total lack of paleontological control in the Kohistan extension of the ophiolitic belt have provided comfortable space for uncontrollable speculations (such as an underthrust of Greater India below the Pamirs), into which the writer will not enter here. 6. INCONSISTENCIES IN THE PLATE-TECTONIC INTERPRETATIONS The author's objections against the plate-tectonic interpretations are based on the insufficiency of geological evidence for some of the fundamental postulates of plate tectonics. His arguments have been presented in some detail in two previous papers(37 ,38) and will but summarily be repeated here, emphasizing the data from the Afghanistan-Pakistan area. 6.1. The hypothetic Late Paleozoic Tethys Ocean Plate reconstructions based on paleomagnetic data, if the data are applied to an Earth of present size, indicate a vast oceanic space between India and Eurasia in pre-drift time(39). This imposing Tethys Ocean, patron saint of this very Ihsan Ketin Nato Advanced Study 256 Institute (Fig.4), is a prerequisite of plate tectonics: if the size of the Earth has not changed (an axiom of plate tectonics), a space about equal to that of the present Indian Ocean must have been permanently occupied by such a Tethys Ocean prior to the break-up of Gondwanaland. As the Indian Ocean opened in not earlier than Triassic time and possibly much later, the Tethys Ocean should have persisted with its full width (about 6000 km) through the Late Paleozoic into the Early Mesozoic. Subsequent narrowing of the Tethys Ocean should have been commensurate in time and scope with the opening of the Indian Ocean. Fig.4. The enigmatic Late Paleozoic Tethys Ocean. The wide Tethys Ocean of Late Paleozoic age is not in agreement wi th the geological data derived from the Tethyan rocks themselves. Bouyx(40) has discussed this problem at length for Afghanistan: the Hindukush belt shows that oceanic conditions in the Paleotethys did not survive the first Hercynian movements in the Early Carboniferous. While a Permian oceanic relict of the Paleotethys can possibly be defended on geological grounds in the Sungpan-Kantse zone of northeastern Tibet (F. Yiying Sun, personal comm.), oceanic rocks datable as Late Carboniferous, Permian or Early Triassic have never been dmonstrated in the Iranian-west-central Asian sector of the Tethys belt. Here, sediments of that age are predominantly platform carbonates and shallow-marine to plainly continental clastics, which northward overstep discordantly the Hercynian structures that incorporate the older oceanic rocks. Some deeper intracontinental graben deposits, void of any volcanic material, are found only in Central Afghanistan. Here, some authors in their desperate search for Late Paleozoic oceanic rocks pointed to the Permian slates and quartzites of the small Turkman basin (west of Kabul) and promptly affixed the suggestive term "flysch" to them; Bouyx( 40) comments that nothing proves an oceanic nature of the Turkman basin. 6.2. Origin of the Neotethys Ocean In the oceanic Neotethys (here defined as the feature documented by the 257 oceanic rocks of the Waras and the main, "Axial", ophiolitic zones), the oldest, datable rocks for which an oceanic origin can be assumed are of Triassic age. Pelagic sediments of that age are rare, but alkaline volcanic rocks occasionally associated with them suggest that initial rifting occurred at that time(4l). The ophiolites of the Waras suture are the only rocks of their kind for which the claim of a Triassic age bears some geological justification, although the evidence, as has been discussed, is not unambiguous. Geology thus indicates that the Neotethys Ocean came into being in not earlier than Early Mesozoic time. Its origin cannot be explained by a migration of the Paleotethys Ocean during the Late Paleozoic-Early Mesozoic as the writer had previously suspected(l). It did not open at the expense of a Permian Ocean. It was virtually a new oceanic feature, born from continental rifting, some 100 million years after the closure of the oceanic Paleotethys. In its early stages the oceanic Neotethys cannot have been more than a narrow trough, or rather a system of narrow troughs which may have opened at different times. Nothing in the rock record suggests that this system of narrow troughs developed into a wide oceanic basin in earlier than Cretaceous time: from Iran to the Himalaya, radiometric work on Neotethyan ophiolites has never given ages older than Cretaceous; by far the largest portion of Neotethyan pelagic sediments is of Cretaceous age, and those ranging from Cenomanian to Paleocene are the only ones showing indisputable synsedimentary relationship with volcanic material of the ophiolite suite(38). 6.3. Subduction It has become routine in plate tectonics to conclude subduction from the presence of calc-alkaline magmatic rocks. But a relation between subduction and magmatism is not as obvious as might appear, if the geological time factor is taken into account. The said magmatism shows a rather clear time relation with distinct folding events of relatively short duration while, long intervals of tectonic quiescence are characterized by magmatic inactivity. In terms of plate tectonics, this should suggest a relation of the magmatism with collision rather than with subduction. Thus, the 200-220 MA granite intrusions of the Hindukush-North Pamir belt coincide fairly closely with the pronounced Late Triassic - Early Jurassic folding of this belt. The next distinct folding event is datable stratigraphically as Early Cretaceous, and it is documented for most of Central Afghanistan and its structural equivalents in the Pamir-Karakorum sector. To this tectonic phase we can relate the Early Cretaceous volcanism of Kandahar and the late Early Cretaceous granites of the Helmand Block and the central Karakorum. During the long Jurassic interval of tectonic quiescence between these two folding phases, no magmatism manifested itself in the region, particularly not in the Farahrud trough where subduction is claimed to have continued during 258 this time. Active subduction off the Makran coast has been construed from the presence of subrecent volcanoes in the distant Makran hinterland, although the extremely weak seismicity of the Makran(42) and much geological and geophysical evidence(36) speak against any active subduction in this area. But even if, in spite of all counter-evidence, the Makran volcanism should be related to active subduction, such a relation cannot hold true for the powerful Quaternary volcanism of Dasht-e-Nawar in Afghanistan, which took place in a sector of the orogenic belt where any possible subduction must have long since been halted by collision. Let us make it clear that a genetic relation of the calcalkaline magmatism with subduction in the Alpine-Himalayan orogenic belt is still a hypothesis which lacks independent geological proof. What is geologically well documented is not subduction but obduction, manifested at the surface by the extensive flat overthrusts of ophiolites such as, in the region under review, the ophiolite nappes of Las Bela, Muslimbagh, Waziristan and Khost. In conspicuous contrast, no indisputable underthursts of ophiolites have been observed in the review area to suggest subduction. Wherever one should expect them, the contacts were found to be vertical or steeply reversed faults. This has been persistently reported, for instance, by Blaise et al. (1978) for the Waras suture, by Lawrence and Yeats(43) for the Chaman ophiolite scar, or by Gansser (22) for the Shanglang La suture line (the "Main Mantle Thrust" of Kohistan), and it is a commonly observed fact in the western and eastern extensions of the ophiolite belts in Iran and the Himalaya. The writer does not want to suggest that subduction has not occurred; subvertical attitude could result from post-subduction steepening. But he wonders why subduction at the gigantic scale inferred by plate tectonics is not a~least as evident in the geological structure as obduction, to which plate tectonics assigns the role of a mere by-product. 6.4. Crustal expansion versus crustal shortening If the Earth has not expanded, crustal expansion by ocean floor spreading must have been compensated by equal amounts of crustal shortening. All estimates of crustal compression in the fold belts north of India fall far short of the approximate 6000 km of shortening required to compensate crustal expansion in the Indian Ocean. Plate tectonics invokes subduction to account for the deficit. A large amount of subduction of Tethyan oceanic crust is, therefore, an indispensable component of every plate-tectonic model for the Iranian-Central Asian orogenic belt. If, however, no wide Permo-Triassic Tethys Ocean was available for subsequent subduction, the enormous amount of spreading in the Indian Ocean cannot have been compensated by Tethyan subduction. Expansion of crust in the Indian Ocean amounted, then, to nothing less than an 259 expansion of the Earth. To this one might object that the Neotethys Ocean could, for instance, have widely opened during the Jurassic to be ready for subduction in the Cretaceous. But this would only mean that the Earth expanded by Tethyan spreading in the Jurassic and did not shrink again, because subsequent subduction would have been equalled by spreading in the Indian Ocean. Subduction at the Tethys margin cannot have accomodated both spreading in the Indian Ocean and the additional spreading in the Tethys itself. In fact, the compensatory spreading/subduction mechanism of plate tectonics makes only sense if the two processes occurred simultaneously. If opening of the Neotethys Ocean began in the Triassic, narrowing of the Paleotethys Ocean should only have started (not ended~) at that time. And if spreading in the Neotethys and Neotethyan subduction took place simultaneously, this alone would have prevented the development of a wide ocean; we can as well operate with a narrow oceanic trough which never produced much more oceanic crust than that which we see today in the ophiolite belts, and which required little, if any, subduction. At any rate, Tethyan subduction can in no way account for compensation of the continued expansion in the Indian Ocean after collision of India with Eurasia in Eocene time. The structure of the Himalaya leaves no doubt that very considerable compression of continental crust was achieved here by folding and thrusting in post-Eocene time. But actual estimates fall again far short of the amount required to accomodate postcollisional north-drift of India, which according to spreading data from the Indian Ocean was in the range of 1500-2000 km. Le Fort(44) considered 600-700 km of shortening in the Himalaya as a maximum, Gansser(4S) estimated about half of this amount (500 km less 200 km of pre-collisional compression). Compared to the Himalaya, post-Eocene crustal shortening farther north, in Tibet, seems to have been insignificant; folding occurred here mainly in the Mesozoic. The strike-slip mechanism proposed by Molnar and Tapponnier( 35) may account for some shortening, but hardly for the amount required: strike-slip along the Chaman fault was perhaps in the order of 200-300 km( 43); and while displacement along the Herat fault may have been considerable till Miocene, post-Miocene deposits are no more clearly offset along it(24). 7. CONCLUSIONS - Early-Middle Paleozoic oceanic basins possibly existed in the north of the Hindukush-Pamir region but closed in not later than Early Carboniferous time. - Geological evidence for the existence of the wide Late Paleozoic 260 Tethys Ocean inferred from paleomagnetic data is lacking. The Late Paleozoic was a period of extensive epicontinental to plainly continental deposition, which in the north overstepped the cratonized former oceanic elements. - Since no Late Paleozoic Tethys Ocean was available for subsequent subduction, crustal expansion in the Indian Ocean cannot have been compensated by Tethyan subduction. - Neotethyan oceanic basins came into being in not earlier than Triassic time. They cannot be explained by southward migration of a Paleotethys Ocean but were entirely new features developing from continental rifting in the Late Paleozoic platform and from Mesozoic generation of new oceanic crust. - Closur of the Neotethyan oceanic basins may have involved subduction of oceanic crust, but such subduction cannot have compensated the cumulative amount of spreading in both the Tethys and the Indian Oceans. - Late Cenozoic compression of continental crust in the fold belts of the region was insufficient to accomodate postcollisional crustal expansion in the Indian Ocean. Geological evidence from the Afghanistan-Pamir-Pakistan region thus converges on one final conclusion: crustal expansion in the Indian Ocean, if the spreading data are correct, must have largely exceeded crustal shortening in the Tethyan orogenic belt. The excess is tantamount to an expansion of the Earth. Owen(46) has demonstrated that the lack of a wide Tethys Ocean in Early Mesozoic time is in perfect agreement with the paleomagnetic data, if these data are applied to an Earth of smaller size. Fig. 5 contrasts the Late Cretaceous Tethys Ocean in Owen's earth expansion model with two conventional Tethys reconstructions applied to an Earth of present size. 261 ~:; '!.......... 9G ItO t= ....... ~ 70 my c Fig.5. Three reconstructions of the Tethys Ocean for Late Cretaceous time. A: Smith and Briden(39) for Santonian (80 MA). Iran-Afghan Block attached to Afro-Arabian plate. B: Powell(47) for Maestrichtian (70 MA). Central Iran and Afghan Blocks form isolated microcontinents north of a Baluchestan-Kohistan ("Chitral") island arc. C: Stocklin(38) based on Owen's(46) Earth expansionmodel for Turonian time (90 MA), with radius of Earth about 90% of current mean value. Central Iran and Afghan Blocks attached to Eurasia. 262 REFERENCES 1- Stock1in,J., 1977. Structural correlation of the Alpine ranges between Iran and Central Asia: M2m. h. ser. Soc. geo1. France, 8, 333-353. 2- B1aise,J., 1968. La structure da 1a region de Maidan, a l'Ouest de Kabou1 (Afghanistan): C.R. Acad. Sc. Paris, 266, 1918-1920. 3- Desio, A., 1963. Review of the geologic "Formations" of the Western Karakorum (Central Asia): Riv.lta1.Pa1eont., LXIX/4, 475-501. 4- Wittekindt,H., 1973. Er1Huterunger zur geologischen Karte von Zentra1- und Slid-Afghanistan 1: 500' 000: Hannover, Bundesanst. f. Bodenforschung. 5- Latif,M.A., 1974, A Cambrian age for the Abbottabad Group of Hazara, Pakistan: Geo1. Bull. Punjab Univ. 10, 1-20. 6- Crawford,A.R., 1981. Isotopic age data for the eastern half of the Alpine-Himalayan belt; p.189-204 in H.K. Gupta and F.M. Delany (Eds.), Zagros-Hindu Kush-Himalaya, Geodynamic Evolution: Geodynamics Ser., 3, Amer. Geophys. Union. 7- Sengor,A.M.C., Yl1maz,Y., and Sungur1u,0., 1985. Tectonics of the Mediterranean Cimmerides: nature and evolution of the western termination of Palaeo-Tethys; p.77-112 in J.E. Dixon and A.H.F. Robertson (Eds.) , The Geological Evolution of the Eastern Mediterranean: Geo1. Soc. London, Spec. Pub1. 17. 8- B1aise,J., Desparmet,R., and Lapparent,A.F. de, 1971. Stratigraphie et structure du Paleozoique de 1a region de Wardak, en Afghanistan: Bull. Soc. Geo1. France, (7), XIII/3-4, 420-429. 9- Ta1ent,J.A. and Mawson,R., 1979. Palaeozoic-Mesozoic biostratigraphy of Pakistan in relation to biogeography and the coalescence of Asia; p.81-102 in Abu1 Farah and K.A.DeJong (Eds.), Geodynamics of Pakistan: Quetta, Geo1.Survey Pakistan. 10- Bou1in,J. and Bouyx,E., 1977. Introduction a 1a geologie de l'Hindou Kouch occidental: Mem. h. ser. Soc.geo1. France, 8, 87-105. 11- Markhatov,B.P., 1963. Tectonics of the Pamir: Leningrad Univ. Press, 243 p. (Russian). 12- Desio,A., 1979. Geologic Evolution of the Karakorum; p.111-124 in Abu1 Farah and K.A.DeJong (Eds.), Geodynamics of Pakistan: Quetta, Geo1. Survey Pakistan. 13- Ibrahim Shah S.M. (Ed.), 1977. Stratigraphy of Pakistan: Mem. Geo1. Survey Pakistan, 11, 138 p. 14- Dronov,V.I. and Abdullah Sh., 1984, Tectonics of the Afghan - South Tajik sector in the Asian part of the Tethys: 27th Int. Geol. Congr., Reports, 5 (Tectonics of Asia), 97-110. 15- Auden,J.B., 1974. Afghanistan West Pakistan; p.235-253 in A.M.Spencer (Ed.), Mesozoic-Cenozoic Orogenic Belts: Geo1. Soc. London Spec. Pub1., 4. 16- Andreev,Yu.,N., Dzha1i1ov,M.R., Mamontov,A.N., and Kortchagin,V.I., 1972. Cretaceous deposits of the southern part of Southwest Tajikistan; p.121-134 in K.V.Babkov (Ed.), Lithology and Stratigraphy of the Oil and Gas Bearing Formations of the Tajik Depression: Dushanbe ("Donish") (Russian). 263 17- Kulakov,V.V., Kolchanov,V.P., and Mikhailov,K.Ya., 1969. On the stratigraphy of the Triassic deposits of northwestern Afghanistan: Izv.vyssh. utch. zav., geol. i razv., 7, 11-16 (in Russian). 18- Weippert,D., Wittekindt, and Wolfart,R., 1970. Zur geologischen Entwikclung von Zentral- und Stidafghanistan: Beih. geo1. Jb. 92, 99 p. 19- Blaise,J., Bordet,P., Carbonnel,J.P. and Montenat,C., 1978. Flyschs et ophiolites dans la region de Panjaw: une suture neocimmerienne en Afghanistan central: Bull. Soc. geol.France, (7), XX/5, 795-798. 20- Le Fort,P., Michard,A., Sonet,J., and Zimmermann,J.L., 1983. Petrography, geochemistry and geochronology of some samples from the Karakorum Axial Batholith (northern Pakistan) ; p. 377-387 in Shams ,F .A. (Ed.), Granites of Himalayas Karakorum and Hindu Kush: Lahore, Panjab Univ., Inst. Geol. 21- Bordet, P. 1975. Les volcans recents du Dacht-e-Nawar (Afghanistan Central): Annales Sc. Univ. Clermont, 53, 86 p. 22- Gansser,A., 1979. Reconnaissance visit to the ophiolites in Baluchestan and the Himalaya; p.11l-l2l in Abul Farah and K.A. Dejong (Eds.), Geodynamics of Pakistan: Quetta, Geol. Survey Pakistan. 23- Allemann,F., 1979. Time of emplacement of the Zhob Valley ophiolites and Bela ophiolites, Baluchestan; p.2l5-242 in A.Frah and K.A. Dejong (Eds.), Geodynamics of Pakistan: Quetta, Geol. Survey Pakistan. 24- Ta pponnier ,P ., Mat tauer, M. , Proust, F ., and Cassaigneau ,C., 1981. Mesozoic ophiolites sutures, and large-scale tectonic movements in Afghanistan: Earth and Planet. Sci. Letters, 52, 355-371. 25- Glennie,K.W., Boeuf,M.G., Hughes Clarke,M.W., Mooddy-Stuart,M., Pilaar,W.F., and Reinhardt,B.M., 1973. Late Cretaceous nappes in Oman Mountains and their geologic evolution: Amer. Ass. Petroleum Geol. Bull., 57, 5-27. 26- Mennessier ,G., 1972. Geologie de la chaine d' Altimour: Rev.Geogr. phys. Geol. dyn., XIV/4, 345-355. 27- Bordet,P., 1980. On the structural evolution of Central and Eastern Afghanistan: Geol. Bull. Peshawar Univ., 13, 5-7. 28- Tahirkheli,R.A.K., Mattauer,M., Proust,F., and Tapponnier,P., 1979. The India-Eurasia suture zone in Northern Pakistan: synthesis and interpretation of recent data at plate scale; p.125-l30 in Abul Farah and K.A.DeJong (Eds.), Geodynamics of Pakistan: Quetta, Geol. Survey Pakistan. 29- Bassoulet, J . P ., Boulin, J ., Colchen ,M., Marcoux, J ., Mascle ,G., and Montenat,C., 1980. L'evolution des domaines tethysiens au pour tour du Bouclier indien du Carbonifere au Cretace: Mem. B.R.G.M., 115, 180-198. 30- Karapetov, S. S., Sonin, 1. 1., and Khain, V . E., 1975. On some of the most important characteristics of the structure and development of the Afghan-Pamir segment of the Alpine folded belt of Eurasia: Vestnik Mosk. Univ., 3(1975), 38-46 (in Russian). 31- Argand,E., 1924. La tectonique de l'Asie: Int. Geol. Congr., XIII, 171-372. 264 32- 33- 34- Staub,R., 1928. Der Bewegungsmechanismus BorntrHger, 270 p. der Erde: Berlin, Termier ,H. and Termier ,G. , 1977. Position, entre Gondwana et Tethys, des provinces afghanes au Carbonifere et au Permien. Essai biogeographique: Mem. h. ser. Soc. geol. France, 8, 309-324. Blaise,J., Bordet,P., Boulin,J., Bouyx,E., and Montenat,C., 1980. La ceinture alpine sud-asiatique en Afghanistan: Mem. B.R.G.M., 115, 139. 35- Molnar,P. and Tapponnier,P., 1975. Cenozoic tectonics of Asia: effects of a continental collision: Science, 189, 419-426. 36- Jacob,K.H. and Quittmeyer,R.C., 1979. The Makran region of Pakistan and Iran: trench-arc system with active plate subduction; p.305-3l7 in Abul Farah and K.A .DeJong (Eds.), Geodynamics of Pakistan: Quetta, Geol. Survey Pakistan. 37- Stocklin,J., 1983. Himalayan orogeny and Earth expansion; p.1l9-l30 in S.W.Carey (Ed.), Expanding Earth Symposium, Syndney, 1981: Univ. Tasmania. 38- Stocklin,J., 1984. Orogeny and Tethys evolution in the Middle East, an appraisal of current concepts: 27th Int. Geol. Congr., Reports, 5 (Tectonics of Asia7- 65-84. 39- Smith,A.G. and Briden,J.C., 1977. Mesozoic Paleo-continental Maps: Cambridge Univ. Press, 63 p. and Cenozoic 40- Bouyx,E., 1981. Remarques sur 1a limite septentrionale du Gondwana durant 1es temps paleozoiques: Bull. Soc. geol. France, (7), XXIII/6, 553, 563. 41- Honegger,K., Dietrich,V., Frank,W., Gansser,A., Thoni,M., and Trommsdorff,V., 1982. Magmatism and metamorphism in the Ladakh Himalayas (the Indus-Tsangpo Suture Zone): Earth and Planet. Sci. Letters, 60, 253-292. 42- Berberian,M., 1981. Active faulting and tectonics of Iran; p.33-69 in H.K. Gupta and F.M .De1any (eds.), Zagros-Hindu Kush-Himalaya, Geodynamic Evolution: Geodynamics Ser., 3, Amer. Geophys. Union. 43- Lawrence,R.D. and Yeats,R.S., 1979. Geological reconnaissance of the Chaman fault in Pakistan; p.351-357 in Abu1 Farah and K.A.DeJong (Eds.), Geodynamics of Pakistan: Quetta, Geol. Survey Pakistan. 44- Le Fort,P., 1975, Himalayas: the collided range. Present knowledge of the continental arc: Am. J. Sci., 275, 1-44. 45- Gansser,A., 1977. The great suture zone between Himalaya and Tibet, a preliminary account; p.18l-191 in Himalaya: Paris, C.N.R.S., Co1loque internat. 268. 46- Owen,H.G., 1983. Atlas of continental displacement: 200 million years to present. A test of the conventional and expanding Earth models: Cambridge Univ. Press, 76 maps, 7 Figs., text. 47- Powe11,C.McA., 1979. A speculative tectonic history of Pakistan and surroundings: some constraints from the Indian Ocean; p.5-24 in Abul Farah and K.A.DeJong (Eds.), Geodynamics of Pakistan: Quetta, Geo1. Survey Pakistan. TECTONOGENESIS & EVOLUTION OF A SEGMENT OF THE CIMMERIDES: THE VOLCANO-SEDIMENTARY TRIASSIC OF AGHDARBAND (KOPET-DAGH, NORTH-EAST IRAN) A.Baud Musee Geologique Palais de Rumine 1005 Laussanne Switzerland G.M.Stlimpfli Shell Winning N.V. P.O.Box 2681 El Horreya Heliopolis, Cairo Egypt ABSTRACT. During the late Palaeozoic, the area of the Kopet-Dagh had been accreted to the Turan plate and partly metamorphosed in greenschist facies, as were the Band-i-Turkestan, the North Hindu-Kush and the North Pamir. During Permian time, parts of these accreted terranes were uplifted, eroded and mainly the southern areas covered by red continental deposits. The Paleeotethys active margin migrated to the South and a new volcano-plutonic arc was emplaced South of the Hercynian collage, just to the North of the new Paleotethys subduction zone. During Triassic time, North of the plutonic arc and in a back arc setting, an arcuate deep subsiding volcano-sedimentary marine belt is recognised from the South Caucasus through the Kopet-Dagh to the North Pamir. A Cimmerian deformed segment of this back arc basin appears now in the erosional window of Aghdarband. From the study of the marine volcaniclastic facies and the Triassic events, a tentative model of the regional geodynamical evolution is presented. 1. INTRODUCTION AND GEOLOGICAL SETTING In the Eastern Kopet-Dagh Range, the erosional window of Aghbarband has been discovered in 1956 and surveyed by several geologists during the last decades. A first general report has been presented by RUTTNER (1984) and a monograph on that area is now being finalised (RUTTNER ed., 1989). The erosional window of Aghdarband shows the pre-Liassic basement of the Kopet-Dagh and is situated about 100 Km ESE of the holy town of Mashhad along the Kashaf Rud river (Fig.l). Structurally, this pre-Liassic basement consists of North vergent 265 A. M. C. !lengor (ed.), Tectonic Evolution of the Tethyan Region, 265-275. © 1989 by Kluwer Academic Publishers. 266 !,.,." \ "~ ; \ ", Iran "! • Mashhad CIMMERIAN o ! 50km , Fig.1: Location of the Aghdarband erosional window. The late Triassi/ early Jurassic Mashhad suture had been obliterated by later intraplate tectonism. thrust slabs of Palaeozoic epimetamorphic metasediments over anchimetamorphic Triassic carbonates and volcaniclastics (RUTTNER, 1984). The Triassic rocks in 3 tectonic slices thrusted northward are folded and bounded to the NE by folded Devonian and early Paleozoic metasediments. The whole area is unconformably covered by late Liassic marine clastic and carbonate rocks (Fig.2). Lithologically, the Triassic Aghdarband Group, 1.2 to 1.5 Km thick, is subdivided into 3 formations and shows a complete sedimentary cycle (Fig.3) from shallow water carbonate rocks (early Triassic) through deep water andesitic to trachytic volcaniclastics (middle to early late Triassic) to continental siltstones and sandstones with coal seams (late Triassic). An analysis of this sedimentary cycle is presented by BAUD et a1. in RUTTNER (1986). Here, we explain and develop the main conclusions of that publication. A geodynamic model is presented, where the Aghdarband area is considered in a much wider geological context. s Ikm "--__ ---'I Precambrian and Paleozoic metasediments - Triassic tectonic slices (I-III) -- Kashaf Rud 1 1 ~ I ____ Paleozoic metasediments (Devonian p.p.) Fig.2: South-North Schematic crose-section through the Aghdarband erosional window, modified and completed from RUTTNER- 1984 & 1986. 2. TRIASSIC GEOLOGICAL EVENTS AND VOLCANO-SEDIMENTARY EVOLUTION 267 N From the petrographical analysis and after comparisons with adjacent areas, ancient and recent models, we can assume that the detritic sedimentation of the Aghdarband Group originated from an active volcanic arc on a continental margin and we are interpreting the general depositional environment as a retro arc or Andean setting. In the Fig.3 we indicate the main characters of each of the formations of the Aghdarband Group and of the underlying red conglomerate. During Triassic time, eight separate geological events have been recorded, delimited on the basis of tectonic movements, or abrupt lithological or environmental changes. At the base of the Aghdarband Group, within the late early Triassic, two events occured. The first was a major one with the superposition of non-metamorphic marine andesitic volcaniclastics on continental red conglomerates. This red conglomerate of late Palaeozoic age is considered as a late Palaeozoic molasse that drained mixed crystalline basement and epimetamorphic acidic rocks. The epimetamorphism is late Palaeozoic or older. The second event was the transgression of the shallow water shelf carbonates (Sefid Kuh Formation) on the volcaniclastics and this event is dated as early Spathian by DONOFRIO in BAUD et al. (1986b). The next two events (third and fourth) occured during early Anisian. The third represents also a major break characterized by a tectonic phase consisting of strong vertical movements initiating the desintegration of the former early Triassic carbonate platform. Part of it was uplifted and eroded producing a thick limestone conglomerate. Other parts of the platform sunk rapidly into considerable water depths where turbidity currents became an important process of sediment transport. This event is dated by Ammonoids as Bi thnynian (KRYSTYN et al. 1986). The same age is considered for the fourth event that corresponded with the replacement of the carbonate sedimentation by volcaniclastic supply probably in response to the effusion of juvenile andesi tic volcanoes on the volcanic arc. During the Middle Triassic, this volcanism evolved to a more acidic composition. 268 u: '" .., < " :::i ] tI.I I- '" < '" ..J :.: u: :§ z "'" c < ,!i ii ::;; 0 Z Il. :l 0 Q( ~ c u: z ~ Vi < z < ,g z Q( Q < N < ..J I c Z :t < iii ~ Z < < u: z -§ < :i: :.: I- .., < ~ Il. '" '" ci. u: ci. c z :g < 'il i .&: Q( ~ tI.I ~ Il. 0' E o o ..., o o N siliciclastic shallow shelf onshore plain: brown weathered siltstone and shale with plant remains _ coal horizon _________________ _ volcanic conglomerate and Iitharenite --7 E o o .,.. 0' o ..., E o o ..., o '" E o o ..., I o o N E o '" N - - - - distal marine deep ramp: calcareous shale tuffaceous lime skeletal packstone. and fine voicano-arenite turbiditic volcano-arenite tuffaceous lime packstone polymict, basinal conglomerate -5 deep ramp. juvenile volcano-Iitharenite and tuffaceous skeletal carbonate monomictic, limy 'Z.. mixed volcano-arenite and --3 - conglomerate.....=z. tuffaceous pelagic carbonate_ shallow water carbonate platform --2 -- and 't' I I' --1 eSI IC vo cano- Itharenite alluvial plain: distal, large braided stream conglomerate, grauwacke and red shale Hercynian metamorphic basement Fig.3: Generalized stratigraphical section of the Permian and Triassic of Aghdarband with interpretation of the environments of depositon (modified from BAUD & a1., 1986, and RUTTNER, 1984). 269 The fifth event occured in the member 2 of the Sina Formation (early Ladinian ?) and is characterized by the deep marine deposition of mixed carbonate/crystalline conglomerate following the erosion of the crystalline basement and its late Palaeozoic sedimentary cover. The rapid erosion is interpreted as the result of a new tectonic phase of uplift (taphrogenesis). The sudden change from the mainly coarse detri tic sedimentation of the member 2 to the fine detritic rocks and carbonate deposits of the member 3 of the Slna Formation forms the sixth event and is dated as late Ladinian (KRYSTYN et a1, 1986). The evolution of the detritic sedimentation was characterized by an increase of felsic non-volcanic grains and the deep marine deposits (deep ramp) evolved to more distal turbidites and pelagic limestones. The seventh event occured in the upper part of the member 3 of the Sina Formation and consisted of a sudden increase of the volcanic and tectonic activities as recorded by massive volcanic sandstone. The filling up and the closure of the marine retro arc basin and the rapid progradation of the continental fluviatile deposits characterize this event. The latest event (eighth) corresponds to orogenic phases of the latest Triassic time recognised as the formation of the Cimrneride orogenic collage (;>ENGOR 1984) that affect all the South Turan Plate. In the Aghdarband erosional window we can observe a general nothward thrusting and folding along WNW axes. The entire Paleozoic to Triassic sedimentary pile is involved in this cover deformation. Some particularities of this orogen and the collisional aspect would be examined in the fo1o1wing section. 3. TENTATIVE PALEOGEOGRAPHIC AND GEODYNAMIC MODEL Figure 4 is shown here to support the following text; it is a tentative model presenting some new ideas still to be confirmed by future field work. It presents an alternative interpretation to the one of ;>ENGOR & HSU (1984). During the Late Palaeozoic, the Turan Plate (Turanian and Karakum Microcontinents of SHEIN 1985) was accreted to the newly formed Eurasian Plate. During that period, due to continuing subduction of the Palaeotethys, the Turan Plate was subjected to important volcanism and pu1otinsm. Most of its Palaeozoic sediments are metamorphosed. Several accretion events took place from the Devonian to the Triassic, marginal basins opened and closed, and the subduction migrated southward several hundred of kilometres, in a way similar to what happened along the West Coast of North America during the same time. The Palaeozoic of the Turan Plate is only known from wells drilled by Soviet exp1orationists. The Siluro-Devonian rocks are metamorphosed up to granulite facies in some areas. 270 co Q) 52 ~ ~ ~ I t:1 )~~ )~ ~ I Q~~DC1ltr.j _CII(').,.LIlIDr- 271 Fig.4: Tentative palinspastic model for the late Triassic (modified from: DERCOURT & a1., 1985, for the western part; STNMPFLI, 1978 and BASSOULET & al., 1980 for the central part; HELMCKE, 1985 & 1986, and SENGOR & HSU, 1985 and SENGOR 1986 for the Eastern part). -1 Eurasia -2 Early late Palaeozoic collage microplates (Tu: Turan, AD: Amu-Daria, Ta: Tarlm, CQ: Chaidam/Qilian, So: Songpan, IN: Indochina) -3 Late Palaeozoic microplates (NT: N-Tibet, SM: Shan-Thai/Malaysia) -4 Cimmerian collage (RH: Rhodope, WP: W-Pontides, EP: E-Pontides, LC: Lesser Caucasus, EL:Alborz, SS:Sanandaj/Sirjan, CI:central Iran, LU:Lut, CA:central Afghanistan, SP:S-Pamir, ST:S-Tibet). -5 Gondwana. -6 Mobile zones on a) late Palaeozoic or b) early Cimmerian sutures affected by Cimmerian/Indo sinian movements (late Triassic/middle Jurassic) and post orogenic volcanism/magmatism. -7 Hercynian front. -8 Palaeo tethyan oceanic crust (SC:S-Caspian, FR:Farah-Rud, TA:Tangula). -9 Back arc basin. -10 Volcanic/magmatic arc, island arc (T: Transcaucasia, K: Kara-Bogaz, M:Mashhad, H:Hindu-Kush, P:N-Pamir, S :Sarawak). -11 Subduction trough/fore arc basin. -12 Passive margin. -13 Oceanic ridge. Gabboric intrusions and local serpentinites have been reported together with basic to intermediate volcanism starting in late Devonian. Flyschoid sequences form the major part of the sedimentary record. Fossiliferous marble is also present (oral communication from Soviet Geologists). A major diastrophism took place between the late Carboniferous and the Permian molassic deposits which cover the whole Turan Plate. This event can be related to the main late Palaeozoic folding in the Ural and surrounding areas. After the late Palaeozoic collage, the subduction migrated southward once more and a new volcanic arc was formed all along the southern border of Eurasia (ST1{MPFLI, 1978) . Associated with that event, extensive metamorphism took place as reported in Band-i-Turkestan, North Hindu-Kush and North Pamir (BAZHENOV et al, 1982). The Kopet-Dagh area is part of the magmatic belt associated with the renewed subduction. The metamorphic Paleozoic of Mashhad with its Devonian ophiolites and radiolarites forms the southernmost extension of it (LAMMERER et a1., 1984). The Hindu-Kush granodiorite (BOULIN, 1981) can be regarded as the eastern extension of the Mashhad complex. During the Permo-Triassic, a marginal basin developed along the new Palaeotethyan active margin. This basin has been regarded by some authors as an intracontinental rift (BASSOULLET et al., 1980, BAZHENOV et al., 1982, SHEIN, 1985), but others recognise it as a back arc basin (BOULIN, 1981, for the Hindu-Kush, KHAIN, 1984, for the Caucasus, present authors for the Kopet-Dagh area). The mode of emplacement in time and space, the geometry and the nature and composition of the marine clastic infilling of this subsiding zone speak in favour of a back arc setting (BAUD et al. 1986). In the back arc sequence of Aghdarband, the first volcanic activity 272 appeared in the late Scythian marine sediments and became preponderant in the early Anisian (event four), it ceased in late Norian. A very similar evolution is reported eastward by SLAVIN (1974) in Band-i-Turkestan and by BOULIN (1972) in the North Hindu-Kush. Between Aghdarband and the Band-i-Turkestan, middle to late TRiassic marine volcaniclastics are also reported by Soviet Geologists in the West Murghab River area (well information). The history of this marginal basin became relatively complex as it evolved in time and its closure was only completed in some areas during Cretaceous time. The closure was induced by the collision of the drifting Gondwanian fragments with Eurasia (Cimmeride orogenic collage of $ENGOR, 1984, or Indosinian orogenesis of STOKLIN, 1977, 1980). These deformations affected the whole back arc area forming the Asian Cimmeride front of $ENGOR (1985). The first area affected by that collage was the Kopet-Dagh on the active margin side, the Eastern Alborz and the North Central Iran on the Gondwanian passive margin. The main events are very well correlatable between the Aghdarband area and the Alborz. Major tectonic inversions affected the passine margin (STAMPFLI, 1978) as well as certain mobile areas within the Irano-Afghan plate (BERBERIAN and KING, 1981). From that time, the subduction of the Palaeotethys under the Eurasian plate ceased and the closure of the marginal basin ended. The magmatic acti vi ty starting at the same time along the South margin of the Iranian plate (BERBERIAN and KING, 1981, DAVOUDZADEH and SCHMIDT, 1984) shows that the subduction once more shifted southward. A major reorientation of the sea-floor spreading occured during the Jurassic (opening of the Neotethys and the Central Atlantic). It is possi ble that szome active margins like South Iran became transform margins at that time. The major transcurrent forces associated with this plate reorganization ended up with the closure of the remnant Paleotethyan Ocean between Central Afghanistan and the Hindu-Kush during mid-Cretaceous time (BOULIN, 1981) and between North-West Iran/Lesser Caucasus and Caucasus in late Cretaceous/Paleocene (BERBERIAN, 1983, KHAIN, 1984). The change in horizontal compressional stress is very well examplified in the whole Alborz area where, after the inversion in late Triassic/Liassic, the inverted basement blocks regained their initial collapsed position and the passive continental margin sedimentation resumed. A continuous carbonate platform progradation took place from Callovo-Oxfordian until the Paleocene all around the South Caspian Basin which thus should be garded as a Palaeotethyan ocanic remnant (STXMPFLI 1978). From this rapid account of the geodynamic evolution of that part of the Cimmeride front, it can be seen that outcrops like those of Aghdarband represent key-points for the understanding of the geology of 273 the whole North Iran. It certainly shows by comparison with similar sequences in Afghanistan and Caucasus that the history of the Cimmerides is po1phased. If the formation of magmatic arcs and associated back arc subsidence can be regarded as relatively synchronous along great portions of an active margin, collision and subsequent closure of the marginal basin is likely to be diachronous. This is shown by many past and present examples, e.g. the collision between Australia and the Indonesian magmatic arc where the geometry of the colliding blocks is the major intervening factor which can also be greatly affected by rapid shifts of the sea-floor spreading (also sohwn in the South Pacific area and related collages in New Zealand). In the case of the Cimmerian front considered here, both factors played an important role. This is further exemplified by the more recent evolution of the Irano-Afghan plate during the A1pide orogeny. Major transcurrent movements took place again in late Cretaceous (South Atlantic opening) after the closure of Pa1aeotethyan oceanic remnants. Intracontinental rifting affected the Irano-Afghan plate at that time and closure of the rifts took place during the Paleogene due again to a main plate motion rearrangement. The final closure of the Neotethys around the Irano-Afghan plate was also diachronous due to the geometry of the continental margins. Continent/continent collision did not even occur all along the suture areas as in Baluchistan, the Neotethyan oceanic remnant there (Arabian Sea) can be a good example for what happened in the South Caspian area during late Triassic and Jurassic periods (the obduction of ophiolites on the Oman passive margin in Late Cretaceous being somewhat more drastic than the inversion affecting the A1borz passive margin in Late Triassic). REFERENCES Bassou1et,J.P., Bou1in,J., Co1chen,M., Marcoux,J., Masc1e,G. & Montenat,C. (1980). L'evo1ution des domaines tethysiens au pour tour du Bouc1ier indien, du Carbonifere au Cretace. In: Co110que "Chaines alpines issues de 1a Tethys", 26eme Congr.geo1.int., Paris 1980. Baud ,A., Brandner ,R. & Donofrio,D.A. (1986). The Sefid Kuh Formation (Triassic Aghdarband Group, Kopet Dagh, Iran). In: RUTTNER ed., Geology of the Aghdarband area (Kopet Dagh, NE Iran). Abh.Geo1.B.-A., Wien, 40, in press. Baud,A., Stampf1i,G. & Steen,D. (1986). The Triassic Aghdarband Group, volcanism and geological evolution. In: Rutnner ed., Geology of the AGhdarbandarea (Kopet Dagh, NE IRan). Abh.Geo1.B.-A., Wien, 40, in press. bazhenov,M.L. & Burtman,V.S. (1982). 'The kinematics of the Pamir arc'. Geotectonics, vo1.16/4, 288-301. 274 Berberian,M. (1983). 'The southern Caspian: a compressional depression floored by a traped modified oceanic crust'. Can.J.Earth Sci., 20, 163-183. Berberian,M. & King,G.C. (1981). 'Towards a paleogeography and tectonic evolution of Iran'. Can.J.Earth Sci., 18, 210-265. Bou1ing,J. (1972). 'L'evolution stratigraphique et structurale de l'Hindu-Kouch central en Afghanistan d'apres la traversale du Salang'. Rev.Geogr.phys.Geol.dyn. (2), XIV/4, 371-382. Boulin,J. (1981). Afghanistan structure, greater India concept and eastern Tethys evolution'. Tectonophysics, 72, 261-287. Davoudzadeh, M. & Schmid t, K. (1984) • ' A review of the Mesozoic paleogeography and paleotectonic evolution of Iran'. N.Jb.Geol.PalMont.Abh., 168, 182-207. Dercourt, J ., Zonenshain, L. P. & 15 authors (1985). cartes paleogeographiques au 1/20.000 .000e l'Atlantique au Pamir pour la peri ode du Bull.Soc.Geol.France, (8), 1/5, 637-652. 'Presentation de 9 s'etendant de la Lias a l'actuel'. Helmcke,D. (1985). 'The Permo-Triassic "Paleotethys" in mainland of SE-Asia and adjacent part of China'. Geol.Runschau, 74/2, 215-228. - (1986). 'Die Al pi den und die Kimmeriden: Die verdoppel te Geschichte der Tethys '-Discussion. Geol.Rundschau 75/2, 495-499. Khain,V.E. (1984). 'The Alpine-Mediterranean fold belt of the USSR'. Episodes, 7/3, 20-30. Krystyn,L. & Tatzreiter,F. (1986). 'Ammonoids from the Triassic Aghdarband Group'. In: RUTTNER ed., Gology of the Aghdarband area (Kopet Dagh, NE-IRan). Abh.Geol.B.-A., Wien, 40, in press. Lammerer ,B., Langhenrich, G. & Manutchehr-Danai ,M. (1984) . 'Geological investigation in the Binalud mountains (NE-Iran), . N.Jb.Geol.PalMont.Abh., 168, 269-277. Ruttner,A.W. (1984). 'The pre-Liassic basement of the eastern Kopet-Dagh Range'. N.Jb.Geol.PalMont.Abh., 168/2-3, 256-268. Ruttner,A.W. (1986). 'Geology of the Aghdarband area (Kopet-Dagh, NE-Iran). Abh.Geol.B.-A., Wien, 40, in press. $engor,A.M.C. (1984). 'The Cimmeride orgenic system and the tectonic of Eurasia'. Geol.Soc.Am., Sp.paper, 195, 82p. $engor,A.M.C. (1985). 'The story of Tethys: how many Wives did Okeanos have?'. Episodes, 8/1, 3-12. $engor,A.M.C. (1986). 'Die Alpiden und die Kimmeriden: die verdoppelte Geschichte der Tethys'-Reply. Geol.Rundschau, 75/2, 501-510. $engor,A.M.C. & Hsu,K.J. (1984). 'The Cimmerides of eastern Asia: History of the eastern End of the Paleotethys'. Mem.Soc.Geol.France, N.S., 147, 138-167. Shein,V.S. (1985). 'A geodynamic model forthe petroliferous Regions of the southern USSR'. Sovietskaya Geeologiya, Moscow, 1985/2, 64-77, in russian. Slavin,V.I. (1974). 'The Triassic Deposits of the Afghan part of the Tethys and their correlation with those in neighbouring regions'. Vestnik Moskovskogo Univ.Geol., 29/2, 22-31, in Russian. Stampfli, G. (1978). 'Etude geologique generale de l' Elburz oriental au Sud de Gonbad-e-Qabus, Iran NE'. These Universite Geneve. 275 Stockling,J. (1977). 'Structural correlation of the Alpine Ranges between Iran and central Asia'. Soc.Geol.France, Mem.h.ser., 8, 333-353. Stocklin,J. (1980). Geology of Nepal and its reginola frame., J.Geol.Soc.London, 137/1, 1-34. GEOLOGY OF THE BALUCHISTAN (PAKISTAN) PORTION OF THE SOUTHERN MARGIN OF THE TETHYS SEA George R. McCormick Professor of Geology The University of Iowa Iowa City, Iowa, USA ABSTRACT. The South Tethyan suture zone in Baluchistan has two distinct tectonic segments. That North-South portion along the Axial belt from Karachi to the Pamirs has primarily a strike-slip motion. Volcanic agglomerates, flows, and breccia in the Parh Formation of Maastrictian age on which ophiolites are obducted are reported to be basaltic. Preliminary microprobe analyses of pyroxenes from agglomerates and flows indicate the volcanic rocks are "within plate alkali basalts". It is proposed that this zone which is a landward extension of the Owen Fracture Zone is an old structural element, perhaps early Mesozoic, and was the site of oceanic islands during the Maastrictian which were perhaps caused by motion of the Indian plate over a "hot spot" as has been suggested for the Ninetyeast Ridge to the East of India. Ophiolites were obducted onto the volcanic deposits on the westward twisting Indian plate in the Paleocene from fractures or perhaps insipient subduction zones near the volcanic islands. The suture zone West of the Axial belt in the Makran is the only remaining active subduction zone of the South Tethys that has not been involved in continent - continent collision. It has been proposed that the northern Makran represents an island arc system on the upper plate of the Afghan block accreted to the Eurasian plate. The Chagai Hills have been interpreted as the andesitic volcanic arc. It is now proposed that the Ras Koh range represents a collision mass of oceanic basaltic islands originally South of the Ras Koh with the outer arc of the island arc sequence and perhaps even with the volcanic arc itself. INTRODUCTION This paper is a preliminary report on observations made on a 6-week reconnaissance of the India-Eurasian suture zone and Makran region of Baluchistan (Pakistan) and on preliminary microprobe data on pyroxenes recently obtained from samples collected during that reconnaissance. This reconnaissance and preliminary analyses were carried out in preparation for a proposed detailed three year investigation of the nature of the south Tethys margin in Baluchistan (Pakistan). A model is 277 A. M. C. ~engor (ed.), Tectonic Evolution of the Tethyan Region, 277-288. © 1989 by Kluwer Academic Publishers. 278 here proposed on the basis of the preliminary work but it is to be understood that it represents thoughts of the first approximation and is subject to much refinement and debate. The nature of the suture line parallel to the trace of the Chaman-(Ornach-Nal) fault zone is distinctly different to that in the Makran region West of the Ornach-Nal fault. For purposes of discussion I will divide Baluchistan into an Eastern section to include the area between the Chaman-(Ornach-Nal) fault zone and the Sulliman-Kirthar ranges to the East and a Western section which will include the Makran region West of the Chaman-(Ornach-Nal) fault. (Fig. 1). Eastern Baluchistan The easternmost region of Baluchistan contains the Kirthar and Sulliman ranges which are composed of Triassic, Jurassic, and Cretaceous sediments which have been folded and thrust onto the Indian plate by its collision with the Eurasian plate. The uppermost Cretaceous (Maastrichtian) on the Westernmost flank of these ranges is the Parh Formation (Bakr, 1963; Kazmi, 1982). The Parh Formation is not continuous along these ranges. The majority of Parh sediments are limestones, however, marls and shales are locally abundant. Bakr (1963), Kazmi (1982), and the Hunting Survey Corp. (1960) have described the Parh formation as containing volcanic ash, flows, tuff, breccia, and agglomerate in addition to the sediments but at a scale too small to be mapped. Kazmi (1979) described a volcanic sequence near Ziarat which he interprets as a part of a remnant island arc. He described the volcanic units as occurring in the uppermost cretaceous unit which he named the Bibai Formation and which immediately overlies the Parh formation. De Jong (1979) has described a series of agglomerates in the Bela region which lie immediatley on top of the Parh formation; he has named these the Porali agglomerates. The Porali agglomerates and Bibai formation are time equivalents and, indeed, they may also have originated in the same tectonic framework. Published data on mineralogy and chemistry of these volcanic units is negligible other than reference by Kazmi (1979) and the Hunting Survey Corp. (1960) that the flows are basaltic and the pyroxene phenocrysts in the flows and agglomerates are titaniferous augite. During April of 1984 I spent several weeks with Mr. Wazir Khan of the Pakistan Geological Survey making a reconnaissance survey of the Parh formation and associated volcanics from near Kach northwards past Chinjun, Bagh, and onto Qila Saifullah. I found excellent sections of pillow lavas, flows, agglomerates, ash, and tuff interbedded in the limestones and marls of the Parh formation. I found xenoliths of Parh limestone in agglomerates and fragments of lava and agglomerates in limestones. There is little doubt that the volcanics were being deposited simultaneously and/or alternately with limestone and marls. Preliminary petrography of the volcanic rocks indicates that they are basaltic and that the augites are titaniferous. The matrix of the agglomerates and flows for the most part is highly altered, however, ... OPHIOLITES ARABIAN SEA I o 50100 I I I km 200 I 279 Fig. 1: Map of Pakistan showing location of major topographic features, structural features and ophiolite massifs in Baluchistan. 280 good unaltered phenocrysts of clinopyroxene are usually present. I performed electron microprobe analyses on 5 pyroxenes from one flow (sample GG1) and 5 pyroxene from an adjacent agglomerate (sample GG3) (table I) both obtained at a large water gap in the Parh formation at Gwanda Gwazi which is located 10 miles southwest of Chinjun which is on the southern flank of the Jang Tor Ghar and Supplai Tor Ghar ophiolite massifs which are immediately southeast of Muslimbagh which is 70 miles northeast of Quetta. These pyroxenes all plot within the field of "within plate alkali basalt" on the MnO-NaZO-Ti02 diagram for pyroxenes of Nisbet and Pearce (1977) (Fig. 2) (table I). • Gwanda Gwazl samples • Raa Koh samples Fig. 2: Clinopyroxene composition and tectonic setting (after Nisbet and Pearce, 1977). Key to fields: A=VAB, B=OFB, C=WPA, D=All, E=VAB+WPT+WPA, F=VAB+WPA, G=WPA. Key to magma types: OFB = ocean floor basalt, VAB = volcanic arc basalt, WPT = within plate tholeiite, WPA = within plate al kal i. 281 The band of volcanic units within the Parh formation seems to widen and narrow indicating centers of volcanism at about 30-40 kms separation. Immediately west of the Parh-volcanic sequences are small to large blocks of ophiolite always in tectonic contact. The major units are the Bela ophiolite that begins near Karachi and extends north to about Kalat and the Zhob Valley ophiolite bodies extending from near Kach north of Quetta northward along the valley to Waziristan (Fig. I). It is important to note that the largest ophiolite masses in the north (Muslimbagh and Waziristan) are present where the fold ranges make a southwest lobe from the Sullimen reentrant near Quetta to Waziristan. (Fig. I). Ahmad (1979) describes slices of metamorphic rocks beneath the north end of the Jang Tor massif and along the west side of the Suplai Tor massif at Muslimbagh. He describes the units as grading from amphibolite grade at the contact with the ophiolite to greenschist grade away from the contact but at a greater depth. Williams (1973) has described a similar sequence in Newfoundland and Lanphere (1981) in Oman. Williams (1973) felt the metamorphic rocks in Newfoundland were greywackes metamorphosed by frictional heat by the obducting ophiolite - this accounted for the higher grade of metamorphism at the contact with the ophiolite and lower grades downsection away from the contact. Ahmed (1979) attributes the metamorphism to either friction or residual heati~ the ophiolite. During late October of 1983 and late April of 1984 I reconned the contact zone around parts of the Jang Tor and Supplai Tor massifs. An excellent exposure of a sequence of metamorphic rocks approximately 1000 feet thick is exposed on the North side of Jang Tor. Hornblendite is at the contact followed by garnet gneiss, chlorite-garnet-schist, marble, garnet-schist, and farthest away is a two-mica schist. The marbles a~d a I found some metamorphosed rock, usually chlorite schist exposed almost everywhere at the contact of the ophiolite massifs. I feel that much more metamorphic rock is associated with the massifs than previously reported. Part of the problem probably is that ophiolite, volcanics, melange, and schists have not been accurately separated. For example, the 1:50,000 geologic map of the Nisai quadrangle shows a large area of ultra-mafic rocks and melange near Bagh on the southeast side of Supplai Tor. Field work revealed that the greater part of this mapped area was volcanic breccia and agglomerate. Additionally a band of chlorite and/or two-mica schist is present above a small melange and below a layered gabbro of the massif. In fact, on my last field day I found that at least 500 feet of the east end of Supplai Tor is schist and gneiss and not gabbro as mapped. It is important to note that I did not observe any andesitic rocks in the volcanic sequences nor have any been reported in the literature although, admittedly, published analytical data is meager to non-existent. Shams (1976) did report calc-alkaline compositions for late dolerite dikes in the area. 282 Western Baluchistan The major structural trends in western Baluchistan (Makran) are all east-west. From north at the Afghanistan border to the south they are the Chagai Hills - Saindak area composed of andesitic volcanics, sediments, and intrussives; the Dalbandin trough filled with Tertiary and Recent sediments; the Ras Koh range composed of andesitic volcanics, basaltic volcanics, intrussives and ophiolites; and the Hamun - i - Maskel depression filled with Tertiary and Recent sediments (Fig. 1). The Chagai Hills and Saindak area have been studied by a number of geologists (Bakr, 1963, 1964; Ahmed, et al, 1972; Hunting Survey Corp., 1960; Arthurton, et al, 1979). The region consists of a sequence of late Cretaceous andesitic agglomerates, breccias, and flows together with limestones, shales, and sandstones. Dioritic to granitic intrussive rocks are present in a number of places and some of the granitic intrussives contain tourmaline (personal communication with Mohamed Munir at the University of Baluchistan). Bakr (1964), Kazmi (1982), and the Hunting Survey maps (1960) report Sinjarani andesitic volcanics as present in the northeast most portion of the Ras Koh range. The remainder of the range is mapped as Bunap on the older maps and as Rakhshani formation on more recent ones. Ophiolitic masses, here called Bunap intrussions, are scattered but for the most part are present in the northern part of the Ras Koh range. Several dioritic intrussives are present in the north central and northwest part of the range (Bakr, 1963). Bakr (1963) in his study of the western Ras Koh describes the Bunap formation (Rakhshani) as being composed of basaltic flows, agglomerates, and breccia together with shaley limestones, shales and sandstones. He mapped peridotite (ophiolite?) on the northernmost part and also mapped dioritic intrussions on the northern side. I made a 6 day reconnaissance of the northern Makran from Nushki to Saindak in early May of 1984 and found two distinct volcanic types in the northeast Ras Koh Range which I interpret to be Sinjarani and Rakhshani. The Sinjarani specimens are all so altered it is difficult to tell whether they are andesitic or basaltic. The Rakhshani units make up the mass of the Ras Koh range; thin sections of 8 specimens collected are all basaltic. I performed electron microprobe analyses on 3 pyroxenes from one breccia (Sample RK-4) and 5 pyroxenes from one flow (Sample RK-6) (table I). All pyroxenes plot in the field of "within plate alkali basalt" on the MnO-Ti02-Na20 diagram of Nisbet and Pearce (1977) (Fig. 2). They do however plot in a different field than do the basalts from Gwanda Gwazi described previously in this paper. A number of andesitic breccia's and flows were observed in the region from Saindak to Rabat and in the Saindak project area. Much field work needs to be done in the Ras Koh in light of new plate tectonic interpretations. It is crucial that all ophiolite blocks be looked at not only as to their chemistry and possible ore potential but as to their tectonic relationship to each other. Bakr (1963) describes schists and hornblendites in the western Ras Koh associated with ultramafics which in 1963 were interpreted as 283 intrussive. I suspect that a restudy would show the ultramafics to be obducted ophiolites and the schist and hornblendites to be metamorphosed wedges under the ophiolites much as at Muslimbagh. Careful mineralogical and chemical study needs to be made of the volcanics in the Ras Koh range to determine if they are all basalt, andesite, or more likely, a mixture. SUGGESTED MODEL FOR THE BALUCHISTAN SUTURE ZONE Eastern Baluchistan Field evidence reveals that basaltic lavas and agglomerates interspersed with Maastrichtian marine sediments are found immediately west of the folded Mesozoic sediments on the western side of the Indian plate (Kazmi, 1979; De Jong, 1979; McCormick, field work 1983-84). Ophiolites have been tectonically obducted onto these basalts and sediments from the west probably in mid to late Pliocene (Ahmed, 1979; DeJong, 1979). Tapponnier, et all (1981) report volcano-clastic material throughout the deposits in the Katawaz basin west of the ophiolite belt. They also report andesitic volcanics in the vicinity of Karandahar, Afghanistan. This portion of the suture is distinctly different from that of northern Pakistan as described by Tahirkeli, et al (1979). If the Indian plate was subducted under the Afghan extension of the Eurasian plate and then collided with it we would expect to find India first colliding with an outer arc composed of flyschoid sediments and ophiolites which would be accreted and obducted onto the margin against the mesozoic sediments. As subduction proceeded the volcanic arc would next be collided with and andesitic volcanics accreted onto the Indian plate westward of the ophiolites. Furthermore, it seems likely that in mid to late Paleocene the Indian plate was not yet close to the position where one would expect to find the outer arc of the island arc system on the subduction zone. I submit that the within plate alkali basaltic volcanics associated with the Parh formation represent oceanic island basalts as described by Searle, et al (1980) for the Haybi volcanics of Oman. Sclater and Fisher (1974) considered both the Ninetyeast Ridge and Chagos-Laccadive Ridge as scars of the northward movement of India after the breakup of Gondwanaland. Curray, et. al (1980) put forth the hypothesis that the Ninetyeast Ridge was formed by passage of the northward moving plate over a "hot spot" perhaps presently located under Kerguelen Island. If this is the case it is reasonable to assume that the Chagos-Laccadive Ridge also was formed by passage of the northward moving plate over a "hot spot" perhaps presently located under the Reunion-Mauritius-Rodriguez islands region. It is not unreasonable to hypothesize that the alkali basalt volcanics associated with the parh formation represent oceanic islands formed on the western side of India as the northward moving plate passed over a "hot spot". The trace of this path is presently seen as the Chagos-Laccadive Ridge, however, the northernmost part (volcanics 284 in the parh formation) has been tectonically wedged in the Indian - Eurasian suture zone. Beginning in early Jurassic small fragments of the Arabian crust broke off and moved north, each colliding with Eurasia and accreting onto it. As this happened the subduction zone would move southward and by Maastrichtian time formed a southward lobe in the suture line. The principle motion along both the Owen Fracture Zone and Ninety East Ridge was strike-slip throughout Mesozoic time. During the Paleocene - Eocene, fractures or even an insipient subduction zone might have formed in the region of the oceanic volcanic islands. When the Indian plate began its northwest twist in the Paleocene, chunks of the ocean floor near the volcanic islands, perhaps including them, were obducted onto the Indian plate from the west. Quite possibly much of the rock was from near conduits or the magma reservoir and was still hot. This would account for the metamorphosed sediments now formed below some of the ophiolite masses. As the westward movement proceeded the Indian plate should have collided with the outer arc as described by Mitchell and Garson (1981). Perhaps some of the smaller ophiolite masses belong to a second collision and some to the obducted floor in the vicinity of the islands. As migration proceeded the Indian plate collided with the volcanic arc represented by reported andesites in Afghanistan. Western Baluchistan (Makran) Beginning in early Jurassic small fragments of the Arabian crust broke off and moved north. By Masstrichtian time they had all been accreted into one mass, the Afghan block (Tapponnier, 1981). By this time the subduction zone had migrated southwards to its approximate position under the Makran of today. The Lut block projects farther south and most likely the subduction zone migrated farther south of the Lut block than the Afghan block either being drapped around the strike-slip faults between the blocks or offset by them. The present day subduction zone might well have a similar bend in it as indicated by the zone of Cenozoic volcanoes from Sultan to Taftan. To the south of the continental blocks was an island arc system on the Eurasian plate as elsewhere along the suture. To the south of the island arc in the region of the Gulf of Oman and perhaps close to Oman was a fracture zone along which were oceanic volcanic islands. The present Chagai Hills and Saindak area are the remnants of the volcanic arc which accounts for the andesitic rocks, dioritic intrussives, and mineralization, particularly of copper and iron. Perhaps magma generated in the subduction zone passed through a thin portion of the overlying Afghan block and thus accounts for some of the more silicic plutons and tourmaline-bearing granites in Chagai. Mitchell and Garson (1981) describe similar situations in Indonesia and Malaysia. The subducting Arabian plate would ultimately carry the oceanic islands to where they would collide with the outer arc and thus create a basaltic island sequence with ophiolites obducted on the southern 285 side much as is observed in Ras Koh today. At the time of this collision or soon after, motion on the subduction zone either slowed more than the spreading rate on the Carlsburg Ridge or other ridges pushing Africa-Arabia northward or stopped altogether. The Arabian plate was forced northward and the motion was countered by raising the Chagai arch and downfolding and/or faulting of the Dalbandin trough. This motion then brought the oceanic island - fore arc collision mass (Ras Koh) close to the andesitic volcanic arc (Chagai Hills). Within the Cenozoic, motion has resumed on the subduction zone with andesitic volcanism occurring at nearly the same geographical region as it did in the Cretaceous and Paleocene. Na 20 Mg O A1 2~ Si 0 2 ca o K 2 0 Ti 0 2 Cr 02 Mn O Fe o S0 J4 Na Mg Al Si Ca K Ti Cr Mn Fe TA BL E I M ic ro pr ob e A na ly se s o f C lin op yr ox en es f ro m G wa nd a Gw az i an d Ro s Ko h Ra ng e 10 K M. W . o f Pa da g GG 1·l GG 1·2 GG 1·3 GG 1·4 GG 1·5 GG 3·1 GG 3·1 GG 3·2 GG 3·3 GG 3·4 GG 3·5 RK 4·1 RK 4·2 RK 4·3 RK 6·1 (c or e) (ri m) 0. 23 0. 50 0. 48 0. 51 0. 27 0. 31 0. 12 0. 24 0. 31 0. 45 0. 33 1. 09 1. 19 1. 29 1. 17 11 .3 5 11 .3 21 1. 49 13 .0 21 1. 25 14 .6 0 15 .2 2 15 .4 2 15 .2 6 15 .1 6 14 .2 7 12 .7 7 12 .5 8 13 .4 9 13 .9 9 9. 09 8. 75 8. 46 6. 46 8. 98 3. 86 2. 98 2. 96 3. 07 3. 76 3. 87 9. 94 9. 27 9. 76 7. 70 44 .6 5 45 .0 9 45 .1 9 47 .7 1 44 .6 6 50 .4 3 51 .0 6 51 .4 4 50 .9 2 50 .5 8 50 .4 6 48 .3 5 49 .0 4 48 .0 7 49 .2 8 23 .4 9 22 .7 9 23 .1 9 23 .1 1 23 .0 6 22 .1 6 22 .0 3 21 .9 7 22 .0 0 22 .0 6 22 .0 6 12 .1 3 12 .5 6 11 .8 8 12 .4 2 0. 06 0. 15 0. 15 0. 16 0. 11 0. 05 0. 06 0. 15 0. 08 0. 13 0. 09 0. 12 0. 08 0. 05 0. 06 3. 01 3. 19 2. 75 2. 30 2. 97 1. 45 1. 05 0. 95 1. 10 1. 34 1. 27 0. 15 0. 11 0. 14 o .n 0. 09 0. 22 0. 11 0. 19 0. 60 0. 54 0. 63 0. 53 0. 47 0. 37 0. 12 0. 10 0. 22 0. 08 0. 18 0. 19 0. 18 0. 20 0. 19 0. 20 0. 35 0. 83 0. 39 0. 51 7. 83 8. 63 7. 95 7. 36 7. 95 7. 64 6. n 6. 67 6. 65 6. 71 7. 47 15 .0 2 15 .5 0 14 .9 7 14 .6 6 99 .8 0 10 0. 86 99 .5 4 10 0. n 99 .3 0 10 1. 30 99 .5 6 10 0. 62 10 0. 12 10 0. 86 10 0. 42 10 0. 03 10 1. 25 10 1. 04 10 0. 49 0. 07 0. 15 0. 14 2. 55 2. 53 2. 59 1. 62 1. 55 1. 51 6. 74 6. 76 6. 83 3. 80 3. 66 3. 76 0. 01 0. 03 0. 03 0. 34 0. 36 0. 31 0. 01 0. 03 0. 01 0. 01 0. 03 0. 99 1. 08 1. 00 0. 15 0. 08 2. 89 2. 54 1. 13 1. 60 7. 10 6. 76 3. 68 3. 74 0. 03 0. 02 0. 26 0. 34 0. 02 0. 01 0. 02 0. 92 1. 01 N "" "r s o f Io ns o n th e B as is o f 24 A to m s o f Ox yg en 0. 09 0. 03 3. 22 3. 37 0. 67 0. 52 7. 42 7. 58 3. 49 3. 50 0. 02 0. 01 0. 16 0. 12 0. 01 0. 06 0. 03 0. 94 0. 83 0. 07 3. 39 0. 51 7. 58 3. 47 0. 03 0. 11 0. 07 0. 02 0. 82 0. 09 3. 37 0. 54 7. 55 3. 49 0. 02 0. 12 0. 06 0. 03 0. 82 0. 13 3. 33 0. 65 7. 45 3. 48 0. 02 0. 15 0. 05 0. 02 0. 83 0. 10 3. 16 0. 68 7. 48 3. 50 0. 02 0. 14 0. 04 0. 03 0. 93 0. 31 2. 84 1. 75 7. 21 1. 94 0. 01 0. 02 0. 01 0. 04 1. 87 0. 34 2. 78 1. 62 7. 26 1. 99 0. 01 0. 01 0. 10 1. 92 0. 37 2. 99 1. 71 7. 15 1. 89 0. 01 0. 02 0. 05 1. 86 0. 34 3. 09 1. 34 7. 30 1. 97 0. 01 0. 08 0. 06 1. 82 RK 6·2 RK 6·3 RK 6·4 RK 6·5 1. 42 13 .7 7 7. 26 48 .2 7 11 .9 1 0. 16 1. 02 0. 64 15 .2 6 10 0. 26 0. 42 3. 09 1. 29 7. 27 1. 92 0. 02 0. 12 0. 08 1. 92 1. 35 1. 41 1. 45 13 .4 4 14 .5 5 13 .4 6 8. 29 7. 75 8. 50 48 .0 0 49 .6 6 47 .6 0 12 .0 4 12 .4 3 12 .1 1 0. 20 0. 08 0. 20 1. 06 0. 82 0. 95 0. 17 0. 20 0. 20 0. 68 0. 64 0. 54 14 .5 4 13 .9 6 15 .3 6 99 .7 6 10 1. 49 10 0. 38 0. 39 3. 00 1. 46 7. 19 1. 93 0. 02 0. 12 0. 02 0. 09 1. 82 0. 40 3. 18 1. 34 7. 28 1. 95 0. 01 0. 09 0. 02 0. 08 1. 71 0. 42 3. 00 1. 50 7. 12 1. 94 0. 02 0. 11 0. 02 0. 07 1. 92 tv 0 0 0 \ 287 REFERENCES Ahmed, Waheeduddin, Khan, Shahid Noor and Schmidt, Robert G., 1972, 'Geology and Copper Mineralization of the Saindak Quadrangle, Chagai District, West Pakistan', U.S.G.S. Prof. Paper 716-A, 21 pp. Ahmad, Zaki and Abbas, S. Gazanfar, 1979, 'The Muslimbagh Ophiolites in Geodynamics of Pakistan', Abul Farah and Kees DeJong ed., pp. 243-249. Arthurton, Russell S., Alam, G. Sarwar, Anisuddin-Ahmed, S., and Iqbal, Saeed, 1979, 'Geological History of the Almreg-Mashki Chah Area, Chagai District, Baluchistan in Geodynamics of Pakistan', Abul Farah and Kees DeJong, ed., Geological Survey of Pakistan, pp. 325-331. Bakr, M. Abu and Jackson, Roy D., 1964, Geological Map of Pakistan, Pakistan Geological Survey. Bakr, M. Abu, 1963, 'Geology of the Western Ras Koh Range, Chagai and Kharan Districts, Quetta and Kalat Divisions, West Pakistan', Records of the Geological Survey of Pakistan, v. X, Part 2-A, pp. 1-28. Curray, J.R., Emmel, F.J. and Moore, D.G., 1980, 'Structure, Tectonics, and Geological History of the Northeastern Indian Ocean' in The Ocean Basins and Margins,Vol. 6. The Indian Ocean, A.E.M. Nairn and F.G. Stehli, ed. Plenum Press, New York. DeJong, Kees A. and Subhani, A.M., 1979, 'Note on the Bela Ophiolites, with Special Reference to the Kanar Area in Geodynamics of Pakistan', Abul Farah and Kees DeJong, ed., Geological Survey of Pakistan, pp. 263-269. Farhoudi, Godratollah and Karig, D.E., 1977, 'Makran of Iran and Pakistan As An Active Arc System', Geology, v. 5, p. 664-668. Hamilton, W., 1979, 'Geotectonics of the Indonesian Region', U.S.G.S. Prof. Paper. Hunting Survey Corp. Ltd., 1960, 'Reconnaissance Geology of Part of West Pakistan'. Colombo Plan Cooperative Project, Canada Government, Toronto. Jacob, Klaus H. and Quittmeyer, Richard L., 1979, 'The Makran Region of Pakistan and Iran: Trench-Arc System with Active Plate Subduction in Geodynamics of Pakistan', Abul Farah and Kees DeJong, ed., Geological Survey of Pakistan, pp. 305-317. 288 Kazmi, Ali Hamza, 1979, 'The Bibai and Gogai Nappes in the Kach-Ziarat Area of Northeast Baluchistan in Geodynamics of Pakistan', Abul Farah and Kees DeJong, ed., Geological Survey of Pakistan, pp. 333-339. Kazmi, Ali H. and Rana, Riaz A., 1982, 'Tectonic Map of Pakistan', Geological Survey of Pakistan. Lanphere, Marvin A., 1981, 'K-ar ages of Metamorphic Rocks at the Base of the Samail Ophiolite, Oman', Jour. Geophys. Res., 86, #84, pp. 2777-2782. Mitchell, A.H.G. and Garson, M.S., 1981, 'Mineral Deposits and Global Tectonic Settings', Academic Press, 405 pp. Nisbet, Evan G. and Pearce, Julian A. 1977, 'Clinopyroxene Composition in Magic Lavas from Different Tectonic Settings', Cont. Mineral. Petrol, 63, pp. 149-160. Sclater, J.G. and Fisher, R.L., 1974, 'Evolution of the east-central Indian Ocean, with emphasis on the tectonic setting of the ninetyeast Ridge', Geol. Soc. of Amer. Bull., v. 85, p. 683-702. Searle, M.P., Lippard, S.J., Smewing, J.D. and Rex, D.C., 1980, 'Volcanic Rocks Beneath the Semail Ophiolite Nappe in the Northern Oman Mountains and Their Significance in the Mesozoic Evolution of Tethys', Jour. Geol. Soc. London, Vol. 137, pp. 589-604. Shams, F.A. and Ahmad, Shafeeq, 1976, 'Petrology of the "Twin Sisters" Soda Dolerite, Southeast of Muslimbagh, Zhob District, Baluchistan, Pakistan', Pakistan Journal of Scientific Research, Vol. 28, pp. 79-84. Tahirkel i, R.A. Khan, et al., 1979, 'The India Eurasia Suture Zone in Northern\Pakistan: Synthesis and Interpretation of Recent Data at Plate Scale in Geodynamics of Pakistan', Abul Farah and kees DeJong, ed., Geological Survey of Pakistan, pp. 125-130. Tapponnier, P., Mattauer, M., Proust, F., and Cassaigneau, C., 1981, 'Mesozoic Ophiolites, Sutures, and Large-Scale Tectonic Movements in Afghanistan', Earth and Planetary Science Letters, 52, pp. 355-371. Williams H.R., and Smyth, W.R., 1973, 'Metamorphic Aureoles Beneath Ophiolite Suites and Alpine Peridotites: Tectonic Implications with West Newfoundland Examples', Am. Jour. Sci., v. 273, pp. 61-71. THE HIMALAYAN OROGENIC SEGMENT Patrick LE FORT Centre de Recherches Petrographiques et Geochimiques B.P. n020 54501 Vandoeuvre-les-Nancy France I. INTRODUCTION "Jusqu'au dernier moment des Siecles, la Matiere sera jeune et exuberante, etincelante et nouvelle pour qui voudra" Hymne de l'Univers Pierre Teilhard de Chardin The year 1964 when the International Geological Congress met for the first time in India, A. Gansser published his "Geology of the Himalayas" that remains more than 20 years after the cornerstone of our geological knowledge of the highest mountain range of the world. Since the Delhi congress, however, a large amount of new information has been gathered from detailed field studies (~maps of Fuchs, 1967; Bordet et ale 1968, 1971; Calkins et ale 1969; Fuchs and Frank, 1970; Hashimoto et ale 1973; Academia Sinica, 1980; Colchen et ale 1980; Stocklin and ----- Bhattarai, 1980, 1982; Polino, 1981; Valdiya, 1981a; Geological Bureau of Xizang, 1982; Burg, 1983; Gansser, 1983; Honegger, 1983; ••• ). Satellite imagery has helped immensely in deciphering the most remote parts of the range as well as its Tibetan s ide and in making correlations (~ Rupke, 1974; Molnar and Tapponnier, 1975, 1978; Gansser, 1977; Ni and York, 1978; Fuchs, 1982; Armijo et ale 1986). New disciplines have been successfully tested, particularly geological and geochemical. New theories and models have enable us to link a variety of observations in a global framework. All this has resulted in the publication of a rapidly increasing number of articles and books on the Himalaya and surrounding regions. Starting from the point where Gansser brought our knowledge to in 1964 I will try to review our new acquisitions and interrogations during these two decades. For this I will follow a chronological order of Himalayan rock formations and events. Classical transverse and longitudinal divisions will be kept as in Gansser (1964) and Le Fort (1975a) (fig.l). 289 A. M. C. !jengor (ed.), Tectonic Evolution a/the Tethyan Region, 289-386. © 1989 by Kluwer Academic Publishers. m ill Tr an sh ,m al ay a ba th O lit h ~O Ph lO hl es ~T ib et an se dl me nl ar ys en es ~N or th Hl ma la ya PI Ul on lC be lt D HI gh H im al ay a le uc og ra M es D T lb e ta n (H lg hH lm al ay a lC flS l8 lh ne Sl ab ~ Le ss er H im al ay a M id la nd s a n d Kr ol b el l D p ,e ls to c e n e b a S ln s ~ H igh p re ss ur e (bl ue sc his t) m e ta m or ph ism M et am or ph ic Is og ra de ~M al nM an ti eT hr US I yM a ,n C e n tr a lT h rU S I ~M al nB OU nd ar YT hr us t l, t. Fi g. 1: M ain st ru ct ur al d iv isi on s of H im ala ya - Tr an s-H im ala ya ( fro m Le F or t, 19 86 ) S ub - Hi ma lay a se rie s ar e no t in di vi du al iz ed . e = Ev er es t, K = K ath ma nd u, Kg = K ar gi l, L = Lh as a, Lh = le h, m = M an as lu, P = Pe sh aw ar , T = T him pu , X = X ig az e. +' IV \0 o Geology of the Tibetan plateau will not be treated as it is the object of another section of this volume. II. THE PRE-COLLISION SITUATION 291 In 1964 it was known that the Himalaya incorporated Precambrian rocks of the Indian shield but exact dating was missing from both areas and equivalence was inferred essentially from lithology. Since 1964, a limited number of isotopic dates have been obtained from Himalayan crystalline formations. The Phanerozoic history of sedimentation that was locally already quite well known has been improved and in some areas completely revised, particularly in the Lesser Himalaya. 11.1. THE VERY OLD TERRAINS (> 1Ga) 292 Table I : Selected Precambrian isotopic ages from the Himalaya (from west to east) formation nature approx. Sr initial reference name longi- method samples age (Ma) ratio tude Hazara slate 72°30' Rb-Sr 3 WR 950±20 (0.700) Crawford & Davies, 1975 Jammu limestone 74°45' U-Pb 4 galena 967 Raha et al., 1978 Kulu quartz i te 77°20' U-Pb uranini te 1232±120 8halla & Gupta, 1979 Chait gneiss 78°20' Rb-Sr 6 WR 1430±150 0.746 Bhanot et al., 1977 Bhitangana augen gneiss 78°40' Rb-Sr 4 WR 2120±60 0.710(±20) Raju et al., 1985 (mail il G.aldam grani te-gneiss 79°40' Rb-Sr 6 WR 1700±70 0.7375(±127) Trivedi et a1., 1984 Askot-Dharamgar augen gneiss 80°00' Rb-Sr 9 WR 1795±30 0.7090(±15) Trivedi et al., 1984 8erinag Ramgarh Almora Askot Tansen Tatopani Kunchha lingtse Kangpar metarhyolite 80°04' Rb-Sr + granite gneiss 4 WR 1840±140 0.7083(±27) frank et al., 1977 porphyry 80°05' Rb-Sr 11 WR 1765±60 0.7235(±46) Trivedi et al., 1984 augen gneiss 80°05' Rb-Sr 7 WR 1820±130 0.7144(±118) Trivedi et al., 1984 gneiss 80°10' Rb-Sr 5 WR 1983±80 0.727(±7) Bhanot et al., 1981 feldspathic schist 80°32' K-Ar biot. 1195±33 Khan & Tater, 1970 carbonaceous shale 83°32' K-Ar detrital 1280 Krummenacher, 1971 (Eocene) muse. gabbro 83°40' K-Ar amph. 819±80 Krummenacher, 1971 pegmati te 84°52' Rb-Sr muse. 1744±84 Deniel, 1985 q-chl.schist 86°01' K-Ar WR 1150 Talalov, 1~n augen gneiss (88°30' ) Rb-Sr WR 1075 (0.700) Paul et al., 1982 deformed granite (91°17') Rb-Sr 4 WR 1012±58 0.762 Sinha Roy & Sen Gupta, - porphyritic peraluminous granites grouped in the "Lesser Himalaya" granitoid belt extending some 1700 km from Pakistan to eastern Nepal (Le Fort et al. 1978, 1980, 1983); 1986 - numerous bodies of augen gneiss from the Higher Himalaya, in particular Formation III of the Tibetan slab in central Nepal and equivalents in eastern Nepal (Le Fort, 1975a,b; Pecher, 1978; Kai, 1981a; Le Fort et al. 1982; Ferrara et al. 1983). From central Nepal to Bhutan they extend for more than 800 km; - protolith source ages from gneisses of the Indus gorge in the Nanga Parbat - Haramosh massif (Zeitler et al. 1986); - masses of porphyritic granite within the North Himalaya or Lhagoi Kangri belt of granitoid that extends through southern Tibet for some 900 km (Academia Sinica, 1980; Debon et al. 1981, 1982, 1984, 1986a) ; 293 - intercalations of tuffaceous volcanic material from the lower Palaeozoic series of Zanskar (Kurgiakh tuffite of basaltic composition and medial to late Cambrian age, Garzanti et al. 1986), and from the top of the Annapurna formation in central Nepal (felsic tuffs of Cambro-Ordovician probable age, Colchen et al. 1986). A similar age has been obtained on a body of K feldspar biotite orthogneiss from Anduo in the northernmost part of the Lhasa block in Tibet (Xu et al. 1985). Contrarily to a view expressed by several authors (~ Srikantia, 1977; Mehta & Rex, 1977; Bhargava, 1980; Jain et al. 1980; Garzanti et al. 1986), this event cannot be equated to an orogenic cycle. The ~gmatism, which generally is associated with very litle to no deformation and regional metamorphism represents a thermal episode that has affected a large portion of the Higher Himalayan realm. It can be found in many of the fragments of Gondwana including Afghanistan, Australia and Antarctica, a 10,000 km megazone of crustal extension and thinning, along which a substantial amount of lower crustal material melted (Le Fort et al. 1986a). 11.3. THE PHANEROZOIC SEDIMENTATION 11.3.1. The Lesser Himalaya fossil finds The Lesser Himalaya has been known for a long time for its general paucity of fossil remains and for the difficulty of assigning reliable stratigraphic ages. A number of microfossils discovered in the epimetamorphic formations of the Lesser Himalaya have recently changed many an inferred age assignment. Most of these findings are still very local and will necessitate confirmation and extension, but it is remarkable that they have been made in the upper lithostratigraphic formations of the Lesser Himalaya and that they belong to the lower Palaeozoic. They include : - stromatolites associated with algae and echinoderms, probably of early Cambrian age in the Dhading dolomite from the Nawakot complex (table II), which has been overthrust by the Kathmandu nappe (Stocklin, 1980); - lower Cambrian (Tommotian) conodonts and other shelly microfauna from an upper Krol phosphorite level of the Musoorie syncline in Garhwal (Azmi and Pancholi, 1983); - upper Cambrian gastropod and brachiopod from Tal formation, below the Singtali limestone, in the Ganga valley (Kumar et al. 1983 in Valdiya 1986); - late Cambrian to early Ordovician (?) microfauna from the lower Tal phosphorite of the same area (Azmi, 1983); - Cambrian paleobasidiospores from magnesite beds of the upper Midlands in eastern Nepal (BruneI et al. 1985). tv Ta bl e II C or re la tio n of th e Le ss er H im ala ya n fo rm at io ns of H im ac ha l (a fte r V al di ya , 19 86 ), Ku ma on ( Va ldi ya , 1 98 1b , 19 86 ), c e n tr al N ep al (P ilc he r, 19 78 ; '£ Co lch en e t a l. 19 80 , 19 86 ; St iic kl in , 19 80 ; Sa ka i, 19 83 ), e a st er n N ep al -S ik ki m ( Ac ha ryy a & Ra y, 19 77 ; B as hy al , 19 80 , 19 84 ), Bh ut an -A ru na ch al (A ch ary ya , 1 97 8; G an ss er , 19 83 a; T ha ku r, 19 86 ). St ra tig ra ph ic a tt ri bu ti on s ar e te nt at iv e. HI MA CH AL KU MA ON CE NT RA L NE PA L £ . NE PA L- SIK KI M B HU T AN -A RU N AC HA L { K as au li Ol -lM i D ha ra m sh ala D ag sh ai Du mr i 1£ Su ba th u Su ba th u Bh a i n sk at i Eo ce ne be d Si rm u r Ta ns en UK -P al £ K ak ar a Ba ns i Am ite J-K M as kh et I~ Lo we r B ijn i C- P (G on dw ana ) Jo gi ra Si sn e Da mu da Da mu da Ba ra ha ks he t r a- Ra n gi t D iu ri 07 Ta l Ta l Si ng ta -R ob an g K ro l K ro l u . M id lan d Gh an p o kh ra -M a l ek h u Kr o 1 (N aw ak ot) In fr a K ro l Ba rp ak -B e n i g ha t B la in i Bl ai n i r - - Ri ba n- D ha di ng Bu xa Bu xa -M ir i I I I G an dr un g- ' a gf o 9 Re ya ng Ja in ti , N ag ht at N ag ht at -B er in ag I Si nc hu L a Ja un sa r " 1. M id lan d U lle ri Li ng ts e (ar gi II o -c al ca re o u s ) I (N aw ak ot) , Ch an dp ur Ch an dp ur . . . . . , Ku nc hh a Sa ng u r i-D a l i n g D al in g- Sh um ar - ~ 'ti Bo m dil a . . . . . B ir et ha nt i ~ ' - I Ba sa nt pu r ! .. ~.l i ; Te jam Ri ph ea n m 1 S ha li (c al ca re o u s ) De ob an - - - ' (~ 1 Ga ) { R au tg ar a Su nd e r n ag a r Da m tha (> 1. 1 Ga ) Ba nj a r- Ra m pu r (ar gi II o- c a l c ar eo u s ) Ch ak ra ta All these fossil finds indicate a general epicontinental marine transgression during the Cambrian on the northern edge of the present Indian shield. The upper Blaini - Krol - Tal group of the Lesser Himalaya would encompass the Precambrian - Cambrian limit (table II). The lower pre-Blaini group, largely unfossiliferous, would have been deposited during Precambrian times. 295 Disseminated amounts of Phanerozoic fossiliferous material have long been known to exist in the Lesser Himalaya. Some of them, such as the Tansen group of south central Nepal, have recently been investigated in detail (Sakai, 1983). Unconformably overlying the pre-Blaini group of the Kali Gandaki, this group consists of some 2400 m of mainly conglomerate, sandstone and shale, with the older unit as old as late Carboniferous to middle to late Eocene and the youngest unit possibly extending into Miocene. 11.3.2. Lesser Himalaya lithostratigraphy Apart from the limited fossiliferous zones, lithostratigraphic correlations prevail in the Lesser Himalaya. The formations have a very wide extent and can usually be traced for hundreds of kilometres; however, lateral variations of lithology, differences in thickness and tectonic complications bring many difficulties and uncertainties. Table II presents a proposed set of correlations for the Lesser Himalaya formations. In Kumaon and Nepal these sedimentary formations are overlain by crystalline nappes of higher metamorphic grade, occupying the core of large-scale synforms (e.g. the Almora, Kathmandu, Mahabharat nappes). Lithostratigraphic correlations are not easy because of the regional metamorphic imprint (biotite - garnet grade mostly). Cambro-Ordovician to Devonian fossiliferous series, however, have been preserved in the core of the Kathmandu syncline, thus requiring that the underlying formations of the Bhimpedi group (Stocklin, 1980) be Precambrian in age and partly equivalent to the lower Midlands group. 11.3.3. Higher Himalaya continuous sedimentation For more than a century, the Higher Himalaya has been known for its extensive fossiliferous series of Palaeozoic to Cenozoic age in contrast with the Lesser Himalaya. This series has been named : - the Tibetan sedimentary series because it lies mostly in the Tibetan cultural area, if not in Tibet itself; - the Tethyan sedimentary series in reference to the eastward prolongation of the Tethys, although it is mainly epicontinental; - the High Himalayan sedimentary series, a more adequate term since it forms the summit of some of the highest mountains such as Everest, Dhaulagiri and Annapurna. Because of the highly dissected relief, the remoteness of large portions of the High Himalaya and the political divisions, the High Himalaya sedimentary series has been studied in five main areas : Kashmir - Zanskar, Spiti, northern Kumaon, north central Nepal and 296 Fig.2: Correlation of the Palaeozoic and Mesozoic series in the five main areas of the High Himalaya (reproduced from Bassoullet et ale 1977). southern Xizang. In all these regions the total measured thickness reaches 11 to 14 km (5 to 9 km of Palaeozoic plus 4 to 8 km of Mesozoic) (fig.2). In most places the sedimentation shows no major break from the Cambrian or Ordovician to the lower Cretaceous or even lower-middle Eocene. A few unconformities and erosional contacts have been noted (particularly at the base of Ordovician, at the base of Devonian and between Carboniferous and Permian); they correspond to variations in subsidence and epirogenic movements, but not to orogenic. 297 The base of the succession is unknown. The Cambrian period is the oldest palaeontologically recognized stage in Kashmir and Spiti (trilobites); in the contiguous Salt range of Pakistan, lower Cambrian sandstone (the 'purple' or 'Khewra sandstone') rests normally on saline series of probable Vendian age and below salt pseudomorph beds of similar lower Cambrian age (Stocklin, 1986). Elsewhere, pre-Ordovician formations are metamorphosed. Ordovician sediments, often beginning with a basal conglomerate have yielded plant remains in Spiti, but through the entire Himalaya rocks of this age are known for a large development of marine carbonates. Silurian rocks generally consist of dark to black shales, silstones and carbonates, but red and green colors predominate in Kumaon. Biozonation is based on numerous fossils (graptolites) particularly in Spiti. Devonian rocks are in stratigraphic continuity with Silurian rocks to the east (Nepal - Xizang), whereas in the west, change in the sedimentation appears at the beginning with plant and fish remains bearing detrital levels; Muth white quartzite of middle Devonian age is a very characteristic marine formation, transgressive over various series and thinning out eastward. Only lower Carboniferous and Namurian ages have so far been documented: with predominantly argillaceous composition, they include Fenestella shales and Syringothyris limestone. An important discontinuity generally separates Permian from Carboniferous rocks, below the agglomeratic slates and tilloids; plant remains (including Gangamopteris and Glossopteris flora) and alkali basalt volcanism (Panjal volcanics) underline these gentle (epeirogenic) movements. The Permian trangression itself has deposited typical black shales with fossiliferous limestone lenses (brachiopods); fusulines are restriced to Ladakh Himalaya; evidence for emersion is locally seen at the top, as in Nepal. Triassic rocks consist predominantly of carbonate sediments with abundant Ammonite faunas; more shaly and more sandy sediments appear towards the middle and/or upper Triassic in Kashmir, Nepal and Tibet. Jurassic sediments show an irregular development of neritic, sometimes dolomitic, limestones (lumachelle, Megalodon) with frequent lacunas (ferruginous oolite); the upper part, including the lowermost Cretaceous, is generally made up of the famous Spiti shale with siliceous and pyritiferous nodules rich in Ammonites ('saligram' in Nepali); it indicates a rapid subsidence of the platform during that period. The Cretaceous deposits often begin with plant-remain-bearing levels (Nepal) or glauconitic sandstone and conglomerate (Spiti-Kumaon) followed by marine green sandstones containing basic volcanic debris (Giumal formation) and by fossiliferous li~stones (Chikkim formation, upper Cretaceous). Eocene deposits, marine and in continuity with Cretaceous ones, are well developed in southern Xizang where sandstones followed by fossiliferous limestones and shales of Tethyan facies (nummulites) are as young as Lutetian in some places. In Zanskar (Baud et al. 1984), Eocene red slates are already continental. 298 This very thick pile of sediment has been deposited in an epicontinental marginal sea during 6some 500 Ma. The average sedimentation rate of some 25m/10 a is actually very variable, reflecting changes in subsidence rate and sea-level fluctuations (Baud et ale 1984). Estimated bottom depths show that the sea was shallow throughout the Palaeozoic era until the upper Jurassic Spiti shales were deposited (ibid.). Lateral variations exist at all scales. Altogether one can say that during the Palaeozoic era, sedimentation was mainly of the detrital type in the western Himalaya and was carbonate-rich in the eastern Himalaya, but during the Mesozoic era, the carbonate sedimentation is dominant in the west (Kashmir), with Kumaon and Nepal having a more varied sedimentation (Bassoullet et ale 1977). Northward also the Mesozoic epicontinental series change into a number of thick turbiditic series that include: - the Triassic - Liassic turbidite of Lamayuru in Ladakh (Gansser, 1977; Frank et ale 1977) and its equivalent in southern Xizang (Mu et a1. 1973)-;-- - the Cretaceous flysch of Nindam in Ladakh (Bassoullet et ale 1978, 1980a, 1983), its equivalent in Kumaon, known for nearly a century (Diener, 1898; von Kraft, 1902; Heim and Gansser, 1939), and southern Xizang (Chang and Cheng, 1973; 't11 et a1. 1973); - the upper Cretaceous wildflysch of southern Xizang (Burg, 1983) possibly equivalent to the Kumaon wildflysch that contains Triassic to lower Jurassic blocks of ammonite bearing pelagic limestone of unknown origin (Gansser, 1964). This northward change of type of deposition from epicontinental to flysch and even pelagic type is only documented for the Mesozoic. During the Palaeozoic, the upper Palaeozoic in particular, similar sedimentary facies, faunae and florae are observed throughout the Himalaya, Transhimalaya, Karakorum and even south Pamir (Bassoullet et a1. 1980b). The palaeogeographic reorganization that took place around~ Palaeozoic-Mesozoic boundary is correlated with the rifting of the Indian shield (Panja1, Abor and other volcanics of Permian age) and the opening of the Neo-Tethys. A number of characteristic events occurred during this general evolution : - numerous episodes of continental sedimentation or influx such as during the lower Ordovician (plant fragments in Spiti), lower Devonian (plant and fish remains in Spiti -Kashmir), Carboniferous (coal lenses in Nepal - Xizang), and lower Cretaceous (plant fragments in Kumaon and Nepal); - periglacial conditions and deposits during Permo-Carboniferous time marked by the widespread occurrence of tillite and ti110idic formations : the well known I agg10meratic slates I of Kashmir and similar formations of Spiti, Kumaon, Nepal, southern Xizang and Bhutan; - limited episodes of basic volcanism including: the (?) Ordovician spilite - keratophyre Baf1iaz volcanics (Shah et a1. 1978), the upper Carboniferous to lower Triassic basalt - andesite - rhyolite volcanics that stretch along the entire range from Kashmir (up to 400 m of Panjal volcanics, ~ Pareek, 1973; Bhat and Zainuddin, 1978) to Arunachal (Abor volcanics of disputed age 299 Jain and Thakur, 1978; Bhat, 1984), Triassic alkaline volcanics associated with the Lamayuru turbidites (Honegger et ale 1982) and lower Cretaceous volcanic debris (Nepal - Kumaon). The latter has often been related to the extensive volcanism that occurred in Kohistan - Transhimalaya during (Jurassic) - Cretaceous (see chapter III). But during middle Cretaceous, the palaeolatitude of the Transhimalaya was about 20 0 N (Pozzi et ale 1982) whereas the Indian northern margin was lying in the southern hemisphere, some 3000 to 4000 km away (~ Klootwijk, 1984; Zhu & Teng, 1984). This implies that two different volcanic settings were active at the same time, very little being known of the southern one. III. THE PRE-HIMALAYAN, MAINLY JURASSIC TO EOCENE EVENTS The Himalayan collision or Himalayan orogeny S.S. has been preceded in Cretaceous-Palaeocene time by a number of geodynamic events that affected the oceanic and continental regions that lay to the north. These events can be related to two main types : subduction and obduction. 111.1. SUBDUCTION 111.1.1. Karakorum The Karakorum range extends parallel with the Himalaya for some 600 km, north of its northwestern segment. This second highest mountain range of the world has a width of 150 km. The geology of it is still poorly known, even though scientific traverses started in the mid-XIXth century. Granitoids have a much larger extent in the Karakorum than in the Himalaya. The backbone of the range is made up of an axial batholith intruding Palaeozoic to Triassic (Desio, 1964) sedimentary series lying on the north side of the range; on its south side the batholith is often in contact with high grade metasediments hypothetically assigned a Permo- Triassic age (Ivanac et ale 1956; Desio and Martina, 1972). Until recently, the limited number of ages measured from samples of the axial batholith ranged from 56 to 5 Ma, with the majority around or younger than 25 Ma. An upper Cainozoic emplacement of the batholith was generally linked to the terminal phase of a northward oceanic subduction (~ Desio, 1976, 1979; Tahirkheli et ale 1979; Jain et ale 1981a). Le Fort et ale (1983) and Debon~l. (1987a) showed that several plutons including the central Hunza granodiorite are actually metagranitoids deformed and metamorphosed at the same time as a phase of high grade metamorphism observed farther south in the surrounding rocks. A lower Discordia intersection on zircons has dated 4the intrusive emplacement of this first stage of magmatism at 95~6 (Le Fort et ale 1983b) and two whole rock Rb-Sr isochrons have given ages of 109 Ma and 117 or 97 Ma (Debon et ale 1987a). 300 A second stage of magmatism occurred during Palaeogene and Eocene time, until at least 43 Ma (middle Eocene). But at that time the oceanic floor to the south has disappeared and Kohistan has collided with Karakorum; this is shown in particular by the similar undeformed light- colZtlred3§ubalkaline plutons found in both domains (Debon et al. 1987a). An Ar- Ar age of 75 Ma for hornblende of undeformed mafic dykes close to the suture led Petterson and Windley (1985) to a similar conclusion. Thus the high-grade regional metamorphism (580 to 640°C and 5 ± 0.5 kbar, Debon et al. 1987a) that has affected the mid-Cretaceous plutons is though~result from the upper Cretaceous-Palaeocene collision of the Kohistan island-arc after subduction of the northern Neo-Tethys oceanic crust along the so-called northern suture zone (fig.3). Another group of granitoids gives very young ages: 9 Ma for the Baltoro (Debon et al. 1986b). These authors suggest that this magmatism, different from the High Himalaya leucogranites (section V.3) may be related to the intracontinental subduction of the Kohistan-Deosai volcanic arc after collision. The cafemic to alumino-cafemic character, the calc-alkaline to subalkaline nature and some rather low Sr isotopic initial ratio of the first stage of magmatism support the inference that magmas were generated from oceanic subduction of oceanic material along an Andean type margin involving some sialic crust (Debon et al. 1987a). This Karakorum zone can be followed to the southwest into the central mountains of Afghanistan through the eastern Hindu Kush (Le Fort et al. 1983b; fig.3). To the east, two models have been suggested (fig.4). They mainly depend on the age aSSigned to the Indus and Shyok suture zones, and on the nature and age assigned to the Karakorum and the Transhimalayan batholiths. If they are different (~ Brookfield, 1981; Sharma, 1983; Srimal, 1986), Shyok and Indus-Tsangpo zones represent two separate sutures along which two different batholiths have been generated; the Shyok suture would be equivalent to the Pangong (Banggong) Tso-Nu Jiang (Qiang) "suture" of central Tibet and the Karakorum would be equivalent to the north Tibetan (Qangtang) block. On the other hand, the similar chemistry and Cretaceous- Palaeogene ages obtained on the Karakorum and Transhimalaya batholiths (Le Fort et al. 1983b, Debon et al. 1987a) show that nothing stands against the prolongation of the Transhimalaya into Karakorum nor against the Shyok being a branch of the Indus-Tsangpo suture (ITS). It is also to be noted that as stated by Pudsey (1986) and Sharma (1987), the Shyok suture and its western prolongation, the northern suture, presents no evidence for the consumption of a major ocean. The eastern prolongation of the Karakorum block becomes the south Tibetan (Lhasa s.l.) blocks (~ Stocklin, 1980; Sengor, 1981, 1984 & 1985; Thakur, 1983; Montenat et al. 1986; Sharma, 1987). The zone where Shyok and western ITS meet is also where the huge Karakorum right-lateral strike-slip fault cuts and bends the Ladakh-Transhimalaya range (fig.4); the relationships have become blurred. However, the western ITS and the Shyok zones have many tectono-stratigraphic and petrologic similarities (Rai, 1983). By assuming a right-lateral movement in the order of 300 km along the Karakorum fault (fig.4), the Pangong-Nu Jiang suture returns to the latitude of the Rushan Pshart neo-Cimmerian suture separating southern from central Pamir (cf. Montenat et al. 1986), and the ~1 V/;!;!!/12 ~3 nHmm~14 I:' ....... j (.:-::.:.::.:.::.:.:.::.~ 301 "/ NO/ A" 400km , Fig.3: Structural divisions and correlations between Himalaya, Karakorum, Pamir and Afghanistan (reproduced from Debon et al, 1987b). 1 = southern margin of central Asia ("northern domain" of Stocklin, 1977),'2 = northern Gondwana Land fragment of central Pamir, 3 = southern Gondwana Land fragment of Karakorum, 4 = ophiolitic and island arc domains, 5 = northern margin of India ("southern domain"). AF Andarab fault, AR Arghandab plutonic belt, BB Band-e Bayan area, Bd Belutchan domain, BW Western Badakhshan, CF Chaman fault, Cm Central Mountains, FA Farah Rod domain, Fd Fayzabad, FK Feroz Koh, HaF Hari Rod fault, HD Helmand plutonic belt, HdF Helmand fault, He Herat, HF Herat fault, HI Himalayan domain, HKE Eastern Hindu Kush, HKW Western Hindu Kush, Is Islamabad, J Jelalabad, Ka Kandahar, Kb Kabul, KF Konar fault, Kh Khash Rod (or Waras) suture zone, KK Karakorum, KL Kun Lun, KO Kohistan, LA Ladakh, MBF Middle Badakhshan fault, MMT Main Mantle Thrust (or southern suture zone), NE Eastern Nuristan, NP Nanga Parbat wedge, NSZ Northern Suture Zone, NW Western Nuristan, PC Central Pamir, PN North Pamir, PSE South-East Pamir, PSW South-West Pamir, RP Rushan Pshart suture zone, SB Spin Boldak plutonic belt, Se Seistan, SF Shiwa fault, SK Safed Khers, Sk Safed Koh, WA Wakhan, WAF Wanch - Akbaital fault. Transhimalayan batholith eoctends in the direction of the Karakorum one. Thus, the simple Transhimalayan arc of southern Xizang could be prolongated westward in the double arc system of Kohistan-Ladakh to the south and Karakorum to the north. 302 o ! ~ Transhimalayan batholith ~ Karakorum axial batholith __ Indus- Tsangpo suture _ Pangong Nu Jiang & Rushan Pshart ·sutures· -... _- ... .... , , \ \ .... \ \ ~ Fig.4: Global structural map of Himalaya, Karakorum and Tibet (after Gansser 1983a and various sources). A right lateral strike-slip movement of some 300 km along the Karakorum fault has been restituted (dotted lines) in order to illustrate the two possible geodynamic settings. In the first solution, the Shyok ophiolitic zone (or northern suture zone NSZ) corresponds to the Pangong (Banggong)-Nu Jiang "suture", the Indus-Tsangpo suture continuing on the southern side of the Ladakh batholith. In the second solution (cf. StBcklin, 1980; SengBr, 1984 & 1985), the Shyok zone (NSZ) is only a branch of the Indus- Tsangpo suture, the Karakorum axial batholith prolongs the Transhimalaya one, and the Pangong (Banggong)-Nu Jiang "suture" continues in the Rushan Pshart one; the Karakorum is part of the Tehtyside south Tibetan (Lhasa) block. [ = Everest, I = Islamabad, K = K2, Kb = Kabul, Kt = Kathmandu, L = Lhasa, NB = Namche Barwa, NP = Nanga Parbat, T = Thimpu. 111.1.2. Kohistan Kohistan is an oval shaped region of some 36000 km2 lying in northwest Pakistan between the northern suture zone and the Indus suture zone, which is also called the Main Mantle Thrust by Tahirkheli et al. (1979). It is separated from the Ladakh region by the Nanga Parbat spur. Access to this region of very rugged mountains, remaining a tribal territory in large part, is very difficult. The construction of the Karakorum highway (KKH) along the Indus gorges and its opening for travel at the beginning of the 1970 's has given the geologist the best observation and has been followed by a large number of publications, bringing into limelight this previously unmapped and almost unknown region (cf. Tahirkheli and Jan, 1979). Large portions of it still remain to be travelled and worked out. 303 According to Tahirkheli (1979), Tahirkheli et ale (1979), Coward ~ ale (1982, 1986), Bard (1983) and Petterson and Windley (1985), Kohistan, in a supposed upward facing sequence, consists of : - the Chilas-Jijal complex, an igneous stratiform complex of layered norites and noritic gabbros with minor diopsidites, chromite bearing dunites, harzburgites and websterites, that have been metamorphosed to medium-pressure granulite grade (Jan and Howie, 1981; Bard, 1983); - an intrusive suite of plutons ranging from tonalites to trondjhemites and from gabbro-diorites and quartz diorites to adamellites. According to Petterson and Windley (1985), they belong to two stages of formations, the first one being gneissified whereas the second remains practically undeformed. The first stage is tholeiitic whereas the second is subalkaline in northern Kohistan at least (Debon et ale 1987a); - a volcanic sequence of tholeiitic basalts succeeded by andesites to rhyolites and associated with minor sediments. In the north, the deformed Chalt volcanics still preserve pillow structures whereas towards the south the Kamila amphibolites may represent the metamorphosed equivalent. To the southwest of Kohistan, the Dir- Utror group of andesitic, dacitic and rhyolitic volcanics, preserved in a syncline, is little deformed and interbedded with Eocene fossiliferous arenaceous limestone, graphitic shale and shale; - the Yasin group (Pudsey, 1986), composed of a lower group of slates, turbidites and limestones, of lower Albian-Aptian age, deposited in small basins, and an upper group made up of fine grained shales and tuffs. All of this sequence has been attributed to an island arc with the lower magma chamber represented by the Chilas-Jijal complex, succeeded by the entire plutonic and volcanic section and topped by remnants of intra- arc basins (Tahirkheli et ale 1979; Coward et ale 1982, 1986; Bard, 1983). The existence of sedimentary rocks of inferred Palaeozoic age (Jan and Mian, 1971) may imply that the volcanic arc was not entirely built on oceanic crust but that it incorporated fragments of continental crust as well. From the few ages now available, the first plutonic stage appears to have been active in lower Cretaceous time according to a 102 ± 12 Ma Rb- Sr isochron age on a northern pluton of metamorphosed and deformed trondjhemiSn bY3getterson and Windley (1985), and possibly a complex 130 to 150 Ma Ar- Ar on a hornblende from a southern metamorphosed quartz- monzodiorite (Reynolds et ale in press). Younger magmatic ages obtained by Rb-Sr whole rock isochron are Eocene, from 58.5 ± 1.6 Ma for the northern adamellite pluton of Gindai (Debon et ale 1987a) to 40 ± 6 Ma for the northern granodiorite pluton of Shirot (Petterson and Windley, 1985). Thus this island arc may have been active from Jurassic to Eocene, but the continuity of the magmatic activity cannot be ascertained from so few absolute dates. The metamorphism and deformation that affected the Chilas-Jijal complex may have been contemporaneous with the blueschist metamorphism to the south of it (see below). 304 111.1.3. Transhimalaya Transhimalaya is but the prolongation of Kohistan east of the Nanga Parbat spur. It continues in southern Xizang for 2600 km (fig.l & 4) and is characterized by a nearly continuous batholith with plutons ranging from gabbro to granite (Debon et al. 1986a). Volcanic rocks are generall y associated with plutonic rocks; they are particularly abundant in the western segment of Transhimalaya known as Ladakh (fig.l & 4), where they form the Dras volcanics. More recent and more varied volcanic rocks are also scattered on top or towards the northern edge of the batholith. The batholith has been studied mainly along two segments : for some 200 km in Ladakh (~ Honegger et al. 1982; Sharma, 1983), and for another 200 km in southern Xizang south of Lhasa (~ Academia Sinica, 1980; TU et al. 1981; Debon et al. 1982, 1986a), where it is also known as the Gangdese (or Kangdese) batholith. Characteristics are similar in both segments. As summarized by Debon et al. (l986a) the batholith is composite, made up of numerous plutonic bodies often continuous with gradational contacts. Postmagmatic foliation is often superimposed on flow structures (cf. Brun, this volume). Compositions range from noritic gabbros to adamellites through quartz-monzonites and granodiorites; contrary to rather common statement true tonalites do not occur. Most rocks are typically metaluminous and subalkaline but not calc-alkaline as mostly considered. It has been known to intrude Mesozoic rocks, since Hayden (1907) described it intruding Jurassic and Cretaceous series in the Kyi-Chu valley (also Chang et al., this volume). Somewhat less attention has been paid to volcanic rocks. In the west, the Dras volcanics consist of a thick pile of volcanics, associated with pyroclastics, volcanoclastic sediments, radiolarian cherts and fossiliferous limestone inclusions; the latter two have yielded Callovian to Tithonian and Albian to Cenomanian fossils, respectively (Honegger et al. 1982). The volcanics themselves are mainly composed of basaltic --- pillow lavas associated with dolerite sills and intercalated with dacitic flows; they belong to an island arc tholeiitic series (Dietrich et al. 1983). The Dras volcanics are intruded by the Transhimalayan plu~n the region of Kargil. Further east, they pass laterally into a flysch (Nindam flysch of Bassoullet et al. 1978). However, volcanics are met along almost the entire lengt~the Transhimalaya; andesitic and daci tic lavas predominate and are found either as volcanic formation s (such as the Lingzizong formation north of Lhasa over 2500 m thick: Wang, 1984; Coulon et al. 1986) or as pebbles and boulders in molasse formations (such as the Kailas conglomerate that transgresses the Transhimalayan batholith: Heim & Gansser, 1939). They are known (Hennig, 1916) to intrude middle Cretaceous limestones containing Orbitolina. The Lingzizong formation has been dated at 97.53~ b40Rb-Sr isochron on the lower volcanics (Wang, 1984), 90 ± 2 Ma by Ar- Ar on hornblende from a dyke cutting the underlying Tak3~a f~omation (Coulon et al. 1986), 50 ± 4 Ma and 51.1 ± 1.1 Ma by Ar- Ar on biotite and sanidine from a 59ach40e of Yangbajain region (ibid.), 59.3 ± 2 Ma and 49.2 ± 1 Ma by Ar- Ar on biotite and plagioclase from the upper unconformable lavas of Lingzhu (Maluski et al. 1982; Coulon et al. 1986) and 48.5 ± 1.5 Ma by K-Ar (Montigny in Westphal et al. 1982).-- 305 A large number of ages have been determined using various methods on samples of plutonic as well as volcanics rocks (fig.5). They mostly range from 120 to 40 Ma (lower Cretaceous to upper Eocene). The oldest age, 180 Ma, (lower Jurassic) comes from the Dras volcanics (Frost et ale 1984) and confirms the fossil findings. From 110 to 38 Ma there ~ obvious quiescence in the magmatic activity at the scale of the entire range, but locally ages generally cluster around one or two values. The youngest ages have been obtained on the Maquiang volcanics, andesites at the base and acid ignimbrites at ~§ to~~ located some 100 km WNW of Lhasa : from 16 to 10 Ma by Rb-Sr and Ar- Ar methods (Coulon et ale 1986). Note that a number of ages, particularly the K-Ar minimal young ages correspond with cooling ages and not with emplacement ages. Rb-Sr initial ratios present a general tendency to increase from west to east (from 0.704 to 0.707) and, when available, from south to north (fig.6). This can be ascribed to an increasing contribution of continental crust toward the east, an island arc to the west (Kohistan) passing progressively to an Andean margin to the east. The Transhimalaya-Kohistan thus corresponds with a composite active margin where subduction continuously operated for 70 Ma at least and may have started as far back as Jurassic (180 Ma). The subalkaline (monzonitic) composition already noted may reflect a strike-slip component opening tension fissures at crustal scale in which subalkaline to alkaline magmas of deep origin may rise (Debon et ale 1986a). The movement would be sinistral here as suggested by Brun (this volume) on other grounds. 111.1.4. Blueschists Very few occurrences of blueschists have been described in the Himalaya : they are limited to several patches south of Kohistan around the Shangla pass area (Shams, 1972; Desio, 1977; Shams et ale 1980), two areas in Ladakh, one around Kargil (Frank et ale 1977) and Sapi pass (Honegger, 1983; Honegger et ale 1985), the other one some 150 km farther east (Virdi et ale 1977) in the vicinity of Puga, and finally a dubious occurrence in southern Xizang (Xiao & Gao, 1984; Burg, 1983, p. 58; Burg et ale 1986). -----The metamorphic assemblage of these mostly volcanosedimentary rocks varies from pumpellyite to glaucophane (or crossite)- lawsonite grade. Jadeite has never been found, but omphacite and phengite are frequent. Piedmontite has been reported in Shangla pass area (Jan and Symes, 1977). The blueschists form ill defined lenses surrounded by rocks that have undergone retrograde metamorphism to the greenschist facies rocks. In Pakistan they are accompanied by serpentine masses sometimes associated with emerald deposits (Jan et ale 1981b). Dating of minerals of the Pakistan blueschists have yielded 84 ± 1.7 Ma (Desio and Shams'4698039by K-Ar on muscovite) and 80 ± 5 Ma (Maluski and Matte, 1984, by Ar- Ar on phengite). Older ages have even been recently obtained on the Ladakh samples: 98 to 96 Ma (Zimmerman, 1985, unpublished, by K-Ar method on whol~Oroc~9and glaucophane) and 103.7 Ma (Maluski, 1985, unpublished, by Ar- Ar on the same glaucophane). 58 .5 ± 1 .6 ( Rb -S r/5 W R/ 0. 70 42 ) D eb on e t al . 87 b 40 ± 6 ( Rb -S rI7 W R/ 0. 70 44 ) P et er so n & W in dl ey 8 5 10 2 ± (R b- Sr /7 W R/ 0. 70 39 P et er so n & W in dl ey 8 5 ~Tr a~S hlm ala yab ath ol' lh EI J O ph io lit es ~T' bet ans ed' men !ar yse "es ~ N or th H im al ay a pl ut on iC b el t D H 'g h H lm al ily a le uc og ra nl !e s [=: =J Tl be la n (H Igh H im al ay a) c rls ta lh ne s la b ~ J >< ·~se r H im al ay a M Id la nd s a n d K ro l b el l c.= J P le 'S lo ce ne b as In s ~ H Ig h pr es su re (b lu( ,sc h.s il m e ta m or ph ,s m M et am or ph ,c . so gr ad e r- -- M ai n M an tle T hr us t ~M a. nC en 'm IT hr us ; , . . , . . . , . . M am B ou nd ar y Th ru st F ig .5 5 4 ± 4 < R b- S r/4 W R /0 .7 04 1) P et er so n & W in dl ey 8 5 34 ± 1 4 (R b- Sr /6 W R/ 0. 70 45 )P et er so n & W in dl ey 8 5 - 5 0 ( Ar -A r/B i-F k) M al us ki & S ch ae ffe r 82 - 4 2 A r- A r/ M u) B ro ok fie ld & R ey no ld s 81 18 9 ± 1 1 (K -A r/H b) F ro st e t a l. 19 84 - 3 9 ( Ar -A r IB i) B ro ok fie ld & R ey no ld s 81 82 ± 6 ( Ar -A r/H b) B ro ok fie ld & R ey no ld s 81 10 3 ± 3 (U -P b/ Zr ) H on eg ge r e t a l 82 10 1 ± 2 ( U- Pb /Z r) Sc ha re r e t a l. 84 a 60 .7 ± 0 .4 ( U- Pb /M on -A II) S ch ar er e t a l. 84 a ~38 ± 2 ( K- Ar /W R) S ha rm a e t a l. 78 b ~? 73 .4 ± 2 .4 ( Rb -S r/4 W R/ 0. 70 45 ) S ch ar er e t a l. 84 a \ 38 .8 ± 4 ( Rb -S r/7 W R/ 0. 70 61 l H on eg ge r e t a l. 82 ~6 .7 ± 3 .0 ( Rb -S r/W R6 /0 .7 07 0) De bo n e t a l. 82 ;:~ :, " ' - , - 6 0 ( U- Pb /Z r) Xu e t a l. 85 55 .8 ± 9 .9 ( Rb -S r 1 7W R /0 . 7 05 7) Ji n & Xu 84 44 .3 ± 1 4. 0 < R b- S r/ 5W R /0 .7 06 7) D eb on e t a l. 82 52 .4 ± 1 1. 8 (R b- Sr /6 W R/ 0. 70 52 ) J in r. S el ec te d is o to p ic a ge s fo r T ra ns hi m al ay a (a nd K oh is ta n) p lu to ni c a n d v o lc an ic r o c ks . '" o 0 \ 307 o.7o3oL----L---L---L---..J......--~5---L---..J......--.L------'---1:':O--- Fig.6: Whole rock Rb-Sr isochrg9 foS6volcanic and showing a general increase in Sri Sr initial references in figure 5. plutonic bodies of the Transhimalaya ratio from west to east. Data from At first, there was a tendency to link the blueschist metamorphism with the obduction and collision of India with Eurasia (~ Tahirkheli et al. 1979; Brookfield and Reynolds, 1981; Bard, 1983), but the Cretaceous ages clearly contradict this hypothesis. The high pressure metamorphism is more likely to be linked to the subduction process and may be correlated to the sudden acceleration of the drift of India toward Eurasia that occurred shortly before 83 Ma according to Patriat et al. (1982) • II 1. 2. OBDUCTION We have been dealing with the Transhimalayan northern margin that was shaped by subduction processes. We now, cross over to the southern margin. 308 111.2.1. Ophiolitic nappes Ophiolite canp1exes, although often dismembered and incanp1ete, are precious remnants of the lost ocean floors, giving unique clues to the oceanic environment. Only few and discontinuous ophiolitic nappes have been preserved in the Himalaya. Excluding Kohistan, where a long and thick section of island-arc oceanic crust is exposed and partially thrust over the Himalayan foreland along a major thrust plane (the Main Mantle Thrust of Tahirkheli et a1. 1979), ophiolites are exposed along less than a fifth of the total length. The ophiolitic remnants can be grouped into three segments (table III, fig.l); Table III Andean type margin fore arc basin ophiolitic nappes radiolarite un it fl ys c h nappes calcareous nappes Indian conti nental margin east-west correlations of the different units of the northern Himalaya-Transhimalaya domain. The actual succession in the field has been modified by the b ackthrusting and othe r phases of defo rmation posterior to the first southward nappe emplacement. west (Ladakh) cente r (Kailas) eas t (south Tibet) Transhimalayan (or Kangdese) batholith Indus series Kargi 1 formation Spongtang oph. Nin dam fl (K) miilanges Lamayuru (T-J) Shillakong series (T-K) = Zanskar nappes Kailas congl. Jungbwa & Amlang La oph. melanges Qiuwu congl. xYgaie-seri"es Xigaze oph. K radiolarites wild flysch (K T-J flysch Kangmar High Himalayan sedimentary series (£-0 to E) - in Ladakh, the Spongtang ophiolite (Frank et al. 1977; Fuchs, 1979), recently studied in detail (Reuber, 1986) forms a 20 km x 309 10 km klippe resting on melanges with both being overthrust onto the High Himalayan sedimentary series, including Palaeocene to Eocene nummulitic limestones, according to Gaetani et al. (1980); - in Kumaon, the famous Kiogar and Amlang La ophiolites (Heim and Gansser, 1939; Gansser, 1964) unfortunately inaccessible now because they ar~ too close to the Sino-Indian border. They expose more than 3500 km of the fresh, uniform Jungbwa peridotite up to 500 m thick and made up mainly of harzburgite (Gansser, 1964); - in southern Tibet, ophiolites may be followed in the Xigaze area for some 200 km east to west. Only recently surveyed (Chang et al. 1977; Bally et al. 1980; Nicolas et al. 1981), they have bee-n---- studied in some detail, and three relatively preserved massifs of this band (Angren, Xigaze s.s. and Dagzhuka) have been mapped at a 1:25,000 scale (Girardeau et al. 1984, 1985a,b). These scattered klippen are quite similar and display somewhat peculiar characteristics. We summarize them from the studies on Xigaze and Spongtang. Of the classical lithological and structural sequence displayed by ophiolite complexes around the world (Coleman, 1977), the Himalayan ophiolites show a reduced but complete section (Bally et al. 1980). The mafic part, comprising basaltic pillow lavas overlyin-g---- dolerite sills and dykes with isotropic gabbro screens, shows very little cumulate gabbro and is remarkably thin « 3 km, Nicolas et al. 1981). The diabase member is made up of a sill complex rather than the usual dyke complex; the ultramafic portion 6 to 10 km thick is made up of harzburgites including abundant Cr diopide-rich layers, even in the upper part, and minor dunites (ibid.). Although the peridotitic unit is tectonically repeated in Spongtang (Reuber, 1986) the main massifs have been preserved from intense deformation, dismembering or alteration (Girardeau et al. 1985b). The peridotites have undergone an intraoceanic plastic flow of variable direction but generally directed to the south (Xigaze, Girardeau et al. 1985b) or the southwest (Spongtang, Reuber, 1986). All mafic rocks are of oceanic tholeiitic nature indicating that they have been generated at an oceanic ridge (Girardeau et al. 1985a). However, the abundant sills and dykes of dolerite proceed from a second magmatic event intruding an already crystallized oceanic crustal sequence (ibid.), probably not very different in time from the first event. Lead isotopic studies (Gapel et al. 1984) show that the Xigaze ultramafics and mafics have very distinct compositions and have been produced by two diferent sources. As they have not been tectonically brought in contact, it is suggested that they proceed from a propagating ridge (ibid.). As stated by Girardeau et al. (1985a), "all the geological, petrological and textural data on the mafic rocks point to very low heat production ••• for the spreading centre in which they have been formed, in good agreement with a slow-spreading ridge origin" as considered already by Nicolas et al. (1981). The palaeo axes of the spreading ridge may be inferred by measuring the direction parallel to the dolerite dykes injected in the accompanying tensional fractures. The direction varies from N60 - 80 0 E for Xigaze 310 (Girardeau et al. 1985b) to N1400E for Spongtang (Reuber, 1986). However, for Xigaze, this does take into account the 85 ± 20° of counterclockwise rotation deduced from palaeomagnetic studies by Pozzi et al. (1984) that would bring the ridge to an orientation of N175°E ± 25~ughly perpendicular to what would be normally guessed. In this respect, it is interesting to note that the oldest magnetic anomalies off the west coast of Australia, the so-called MO to M10 anomalies, ranging from 108 to 122 Ma, have a somewhat similar dLrection with regard to peninsular India (Larson et al. 1979; Johnson et al. 1980). If, as suggested by different authors on different grounds, India was initially welded to Australia in the Gondwana continent (~ Crawford, 1969; Hamilton, 1983; Le Fort et al. 1986a), these anomalies could correspond to the same opening of the Indian Ocean between India, Lhasa block and Australia, during the early Cretaceous some 125 Ma ago. In fact, the age of the Xigaze ophiolite deduced either from the radiolarites interbedded in the pillow lavas (110 Ma: Marcoux in Girardeau et al. 1984) or directly from the U-Pb analysis of the magmatite (120 ± 10 Ma: Gopel et al. 1984) corresponds to the M series of anomalies. According to the palaeomagnetic study of Pozzi et al. (1984) the Xigaze ophiolite was then lying between 10 and 20 0N, close to the southern margin of the Lhasa block. The minimum extent of the southward thrusting of the ophiolite klippen onto the Indian margin reaches 30 km for Xigaze, 40 km for Spongtang and 80 km for Amlang La. The contact of the ophiolite with the sedimentary series is marked by a major tectonic breccia locally containing medium to high temperature metamorphic rocks (Girardeau et al. 1984; Reibel and Reuber, 1982). The Xigaze ophiolite is overlain by~6 to 8 km thick Cretaceous Xigaze flysch; a stratigraphic depositional contact of the flysch over the radiolarites has been described in several places (Marcoux et al. 1981, 1984). 111.2.2. Age of the obduction Bounds on the data of obduction can be constrained by ages of the overlying and underlying rocks. But the problem has been complicated by further movements that occurred during the Himalayan evolution. These movements, successively south and north vergent, are very difficult to trace back, and their importance, specially the amount of displacement involved remains unknown. Keeping in mind these difficulties, one may still consider the few pieces of somewhat conflicting evidences that have been presented. In southern Xizang, on top of the Xigaze formation, the flysch extends from Aptian to upper Cretaceous (Marcoux et al. 1981, 1984) and far to the west, possibly into Paleocene and Eocene (Geological Bureau of Xizang, 1982). Xi gaze ophiolites are thrust over a flysch containing blocks, or wildflysch, of upper Cretaceous (Maastrichtian) to Palaeocene age (Burg, 1983). However, it has been argued that the deformation related to obduction may not have affected the wildflysch and that the obduction therefore occurred just before or during its deposition (Tapponnier et al. 1981; Burg & Chen, 1984). In northern Kumaon, the Jungbwa peridotite lies on upper Cretaceous flysch (Gansser, 1964). In 311 Ladakh, the Spongtang klippe may even have primarily overthrusted the Eocene sediments (Gaetani et al. 1980). The minimum age of the obduction can be constrained by younger formations that overlie the contact and, with more difficulty, by structural analysis. The Liuqu red polygenic conglomerate overlies unconformably the Xigaze ophiolites (fig. n; its age is believed to be Oligo-Miocene (Academia sinica, 1980). The Liuqu transgresses southward thrusts contacts at several places (Burg, 1983, p. 62; fig. n, but it is not affected by the two phases of south vergent deformation (D , penetrative, D2 of abundant chevron folds) described in the optiolite and its sub-ophiolLtic sole (ibid; table IV). The same situation exists in Ladakh (cf. Van Haver, 1984). Altogether, the timing of obduction seems to be bracketed between Eocene and Oligocene. More precisely, one can think that it is responsible for the cessation of sedimentation, particularly marine sedimentation, around middle-upper Eocene, just before collision occurred. This conclusion reminds of the timing determined in Baluchistan (Allemann, 1979) for the emplacement of ophiolites, on top of upper Cretaceous melange but transgressively overlain by middle Eocene sediments. IV. THE COLLISION The next major event is the collision of the two continental masses of Asia (Lhasa block) and India. IV.l. TIMING OF THE COLLISION Scientists of a number of different and independent disciplines have tried to determine the date of the collision; on the whole their results agree. Both the Indian ocean and the continent masses have brought their contribu tions. SSE NNW ITS zone ? L HAS A block flysch (T -J) TransHim granitoids (K-E) Fig.?: Schematic cross section of the Lhasa block and Indus-Tsangpo suture zone (ITS) in southern Tibet. Synthesized after Burg (1983) and Academia Sinica (1980). Age of the formations in parentheses. 312 Table IV : Summary of the main phases of small scale deformations from the suture to the high HiMalaya. Synthesized from various sources, particularly Burg (1983). High-Himalaya High-Himalaya flyschs Liuqu ophiolites Xig8ze unit or series (south) series (north) congl. formation formation o - E 10 - E I - K 01 - Hi T 1 K - E 1 age intraoceanic D 0 decreases southward decreases southward f 1 (N~O & N110) f 1 (N-S) isoclinal D1 f 1 (N90-110) isoclinal recumbent S1 (fracture) to flo. S1 flow S1 flo. tangential 1 inters I If linters l1 mineral ove rturned 1\ stretch (M-S) 11 ~ stretch I/f 1 to the south (N ~O-~O) 60 60 to 75 75 increases northward f 2 (EW) upright to overturned f 2 (N90-120) upright to overturned f 2 (N100-110) chevron f 2(N~0-150) chevron f 2(EW) upright to overturned SO~~h $2 crenulation S2 crenul forming S2 fracture forming S2 fracture vergent 12 crenul, I/f 2 12 crenul, I/f2 to ~~ ---------------------------------------------- kinks + large kinks + D' ? folds + shears open folds 2 • reverse reverse 0 f ~ f (EW) chevron back-thrusting faulting faulting back (riorth)-o~erturned nor th thrusting N10 compression $3 sub-horizontal $3 crenul, subhoriz IV.l.l. From the Indian ocean Recent and extensive surveys of the Indian ocean (~ Schlich, 1975, 1982; Patriat, 1983) have enabled us to reconstruct in some detail the northward motion of India since late Cretaceous (Molnar and Tapponnier, 1975; Patriat and Achache, 1984). The spreading history of the oceans allows us to calculate the drift of India relative to a fixed Eurasia, a movement very similar to the absolute motion of India in the hotspot frame of reference as the Eurasian plate moved very little during the Cainozoic (Morgan, 1983). According to Patriat and Achache, three main phases of drift may be distinguished (table V) with a major change in velocity and direction at the beginning of Eocene around anomaly 23. Molnar and Tapponnier had already related such a drastic slow down of the Indian plate (that they had calculated to occur at the end of Eocene times, cf. table V) to the effect of the collision of India and Eurasia. The new timing of the collision proposed by Patriat and Achache (1984) precedes a somewhat 'erratic' period until steadier and slower movement resumes around the beginning of Oligocene and proceeds with the same characteristics up to the present. Accordingly, collision would have started sometime between 313 anomalies 23 and 22, that is 55 to 52 Ma (La Brecque et al. 1977), that is Ypresian (according to the Cainozoic timescale of Odin et al. 1982). IV.l.2. From India-Eurasia palaeomagnetic data Palaeomagnetic data from peninsular India -- particularly the Himalaya and southern Xizang (cf. Klootwijk, 1984; Klootwijk et al. 1985 for review) define two apparent polar wander paths that intersect near the 55 Ma pole position. The most recent results concerning southern Xizang (on the mid-Cretaceous Takena red beds and the lower Cainozoic Lingzizong volcanics) and northern Himalaya (on the Eocene limestones and sandstones series around Tingri) tend to show that the collision between these two domains occurred around 50 Ma in this eastern portion of the Himalaya (Achache et al. 1984; Besse et al. 1984). These results rely strongly on the age of the Tibetan Cainozoic volcanics and large errors, around 10 Ma, are at tached to them; however geological and other cons iderations have led Besse et al. (1984) to bracket the value to 50 ± 3 Ma. IV.l.3. Other indications Information from the stratigraphical, sedimentological and magmatic records have also been advocated to constrain the timing of collision. The Indus formation in Ladakh (table III) was deposited from Albian to Eocene in a sedimentary basin comparable to the inner fore-arc of Andean type margins (van Haver, 1984). The general regression and, in the southern part of the formation, the abrupt change from a marine to a continental regime at the end of the lower Eocene (Illerdian + Cuisian) Table V magnetic anomaly 13 23 32 phases of northward drifting of India according to (1) Molnar & Tapponnier (1975) and (2) Patriat & Achache (1984). Note the different timing suggested for collision. Anomaly 23 is 55 Ma old according to La 8recque et a1. (1977) and 52 Ma old according to Patriat & Achache (19~ age velocity (i n cm x a -1) (Ma) (1) (2 ) Oligo-Pleistocene ~5 around 5 36 Eocene 10 to 11 314 is thought to be indicative of the onset of collision in this region (ibid.). This mid-Eocene regression is widespread; it covers not only the suture zone but also the outer molasse basins (cf. table IX) as well as numerous basins on peninsular India and may be linked to a general upwarping of the Indian continental crust. Remnants of basic ma~atism have been observed to occur within the Eocene detrital and continental Chulung La formation (Gaetani, pers. comm. 1985), the top Ypresian (53 to 44 Ma) formation of the High Himalayan sedimentary series in Zanskar (Baud et al. 1984). Similarly, the red clastic sand and shale of the Murree formation in the Muzaffarabad region of Pakistan have been shown to be of Illerdian age (lowermost Eocene, base at 53 ± 1 Ma) and to contain heavy minerals such as Cr spinel, Ti magnetite, chromite, etc., that must derive from oceanic crustal material such as those now present in Kohistan and Ladakh (Ottiger, pers. com. 1985). These sudden and contemporaneous arrivals of mafic constituents on the Indian platform are again clear indications that the contact with the northward subduction zone had occurred. Finally, the high intensity of ma~a emplacement in the Transhimalaya during Palaeocene to Eocene (around 60 to 45 Ma) may be attributed to the final stage of convergence when some of the water-rich sediments were dragged along the subduction zone and triggered the magma generation (Debon et al. 1986a). Collision may also have helped ma~as to move up by fragilizing the rim of the Tibet continent. IV.2. OCEANIC CRUST REMNANTS ALONG THE SUTURE As described by Gansser (1977), the suture zone between Himalaya and Transhimalaya is a discontinuous belt of ophiolitic rocks that forms a narrow, mostly vertical and highly tectonized zone containing sheared bands of flysch with lenticular bodies of ultramafic and mafic rocks with associated radiolarites (a 'melange'). These narrow strips of oceanic material pinched between the two continental masses have subsequently been sheared and laminated by back thrusting and strike-slip taking place along the same discontinuity (Burg, 1983). Few oceanic remains are still visible. Where they are not, one can still distinguish a limit between the Tibetan vs. Indian affinities of the different flysch units (Mascle, 1985); howeve;-the suture then becomes almost 'cryptic'. In most places it is dificult to distinguish between the obducted ophiolite nappes and the ophiolitic remnants of the suture. These are often made up of pinched serpentinites and lavas separating two different domains (see Burg, 1983, figs 11, 12 and 15; N. of Saka, W. of Ngamring and S. of Sangsang). These vertical slivers of serpentinite are generally not isolated and it is difficult to give them a unique geodynamic significance because a number of north-dipping thrusts have been later rotated into vertical attitudes by backthrusting on planes dipping south. Thus the suture is often more conspicuous when one looks at the tectonic break than when one tries to trace discontinuous slivers of serpentinites. 315 IV.3. A COMPLEX EVENT Collision of India with Eurasia has no reason to be a unique and synchronous event. As already underlined by Dewey and Burke (1973), the two continental margins were opposing different bulges and embayments. Along the 2500 km front, collision certainly did not occur at once and promontaries were first to impinge on the opposite continental shore. This resulted in local tectonic situations including important strike- slip movements. In a similar line, Burg et ale (1985) suggest that collision was preceeded with the scraping in the subduction zone of oceanic seamounts and pelagic horsts incorporated as exotic blocks in the flysch. V. HIMALAYAN TIMES The period post-dating the collision and suturing is the time of the most important tectonic activity : that of intracontinental subduction by thrusting at plate scale. This major Himalayan orogenic event results in the doubling of the continental crust, in the rapid evolution of pressure, temperature and fluid characteristics, and in the generation of granitic ma~a. For reasons of clarity we will review successively deformation, metamorphism and magmatism although they are inextricably linked and cannot be studied independently. V.I. DEFORMATION Let us again consider the evolution of the deformation in space (from north to south) and in time. V.l.l. In the northern Himalayan domain V.l.l.a. Three main phases of deformation Tectonic studies carried out in three different sectors of the northern domain of the Himalaya: Ladakh to the west (e.g. Bassoullet et ale 1978, 1980a; Honegger, 1983; Sharma and Shah, 1983; Baud et ale 1984; Mascle, 1985), southern Xizang to the east (e.g. Tapponnier~l. 1981; Burg, 1983; Burg and Chen, 1984; Mercier et ale 1984), and northern Kumaon in the middle (~ Gansser, 1964; Sinha, 1980, 1981) have shown that it is composed of a thick pile of nappes (table VI) and that at least three main phases of deformation could be recognized. In Ladakh, a first phase of southward thrusting was responsible for the stacking of nappes on the Indian margin. A second phase of south vergent folding followed probably in the same episode, related to the same compressional system (Bassoullet et ale 1980a); it is marked by chevron to isoclinal folding and minor imbricate thrusting deforming the major phase 1 thrust planes (Baud et ale 1984). During a third phase of backfolding and backthrusting to the north, the initial thrust contacts Ta bl e VI Su mm ary o f th e m ain p ha se s of d ef or m at io n a c ro ss th e In du s- Ts an gp o Su tu re Z on e. to rm at io n w ild f ly sc h fl ys ch co n gl . in fr a o ph io li te o ph io li te as se m bl ag e ag e K T- J O- M ? (J) -K r ad (M a-P al £) 0 2 ph as es 3 ph as es , 1 ph as e 3 ph as es 1 ph as e + E F F 1 i so re cu m be nt ba ck th ru st in g F 1 i so (sh ea th ) in tr ao ce an ic 0 N3 0- 18 0 fa ce S 15 0- 22 0 th ru st in g R r 2 u pr ig ht F 2 c he vr on F 2 c he vr on M u pr ig ht 30 -1 50 A su b i so cl in al T F 3 k in ks , F 3 k in ks F 3 b ac kt hr us t I in c r . do wn 0 S1 s la ty S1 a x ia l N S2 f an S2 f an S2 c re n u l S fa n L 1 s tr et ch ( 0) L; s tr et ch L2 c re n u l 1- 2 (SW ) (N ) 1 (S ) 2 (S ) 3 (N ) M ElA - no M~ e p i t o m es o no M1 MO RP H ( 2) M2 po st M 3 1 = 0 1 ob d 1 o bd uc t IN TE R- ~ - _ th ru st 1 PR ET A- lIO N 2 c o ll is 2 c o ll is 2 c o ll is 3 cf g ra ni t 3 la te 3 la te 3 la te X ig az e F Oi uw u tu r bi di te A c o n gl . U L K -(£ ) T £1 1 ph as e 1 ph as e F u pr ig ht F EW to ov er tu rn S fa n S fa n + (s s) (S ) (N ) (N ) o ~4 0 % o .... 30 % no no 40 to 20 1 3 3 gr an ito Id C- IK se d K -£ T- IK uK 1 sh ea r 3 ph as es 1 NS F 1 i so ,/ S 1 (S ) r 2 u pr ig ht r 3 k in ks S1 a x ia l ph as e S2 f an L1 s tr et ch L2 i nt er s (S ) O 2 .... 50 % no M1 1 An de an 1 10 1 2 " 80 1 (c f o bd uc t) 3 1 3 c o ll is ? < 6 0 lin zi zo ng v o lc Pa lE 1 ph as e (F 3 1) no X X 3 1 w a- , 317 have again been deformed and reworked, the system of nappes being this time overthrust on the Indus detrital series (upper Cretaceous to upper lower Eocene). A last phase of minor folding, once again south vergent, followed by shearing, along horizontal planes has also been documented (Bassoullet et al. 1980a). In southern Xizang three main phases of deformation are also recognized across the suture zone (Burg, 1983). Correlations of deformation patterns are sometimes difficult to establish from one tectonic unit to another because of their generally late juxtaposition and the difference in lithology (fig.7); it is however possible to suggest a general correlation chart for this sector of the northern domain (table vI). It compares well with the one from Ladakh. V.l.l.b. Timing and significance at plate scale The emplacement of nappes during the first phase of deformation clearly relates to the obduction of ophiolites on the Indian continental margin. We have discussed above (III.2.2) the possible lower Eocene age of this obduction, the timing of which does not have to be strictly synchronous along the entire northern margin. The collision itself is then responsible for the second phase whose intensity increases downward and northward. It may be related to the initiation of larger shearing zones in the Indian continental crust. As for the third phase, north vergent, its significance becomes more clear when comparing it to the deformation pattern of the central Himalayan domain (see below). V.l.l.c. Longitudinal patterns in the northern domain The first phase of ductile deformation studied in the Suru valley of Ladakh has shown that the strain ellipoid is of flattening type and has a Al principal axis parallel to the stretching mineral lineation and oriented N800E (Gapais et al. 1984; Gilbert, 1986). The intensity of the deformation is shown by ~importance of the elongation and by the presence of sheath folds parallel to A (Gapais et al. 1984). The amount of flattening has been estimated by Gilbert (1986) using the t' ex method on folds; the ratio A2 /Al of the main strain axes varies from 0.8 to 0.2 with increasing intensity. The dip of the axial plane of the folds similarly decreases from 45° to 0°. The elongation, slightly oblique to the longitudinal trend of the belt corresponds with a direction of displacement and shows that under the general N-S compression, strike- slip motion has been also present. Similar patterns have been observed in the central zone (see below) where their origin also remains hypothe- tical : lateral sliding, lateral movement accomodating regional heterogeneities of the crust, obliquity of Indian-Eurasia convergence ••• In Ladakh, Gilbert (1986) suggests that the E-W longitudinal extension is due to the extrusion and "radial spreading-gliding" of the sedimentary nappes over the crystalline basement at a late stage of the metamorphic evolution. 318 V.1.2. In the central domain: the MCT zone The large scale over thrusting of metamorphic and crystalline units over less metamorphosed sedimentary and volcanic formation has long ago been recognized. Heim and Gansser (1939) gave to this structure, which they partly mapped in Kumaon the name of Main Central Thrust. This structure is equivalent to the now world famous MCT (fig.8). In the 1970's, the detailed work of structural geologists, particularly Pecher (1978), has shown that this thrust was not only a plane but a whole volume of rocks deformed in a ductile way for several kilometres on both sides of the thrust plane, thus introducing the notion of the 'Main Central Thrust zone'. The simultaneous study of structural, metamorphic and petrologic features has given further insight into this large scale intracontinental phenomenon. V.1.2.a. Different structural characteristics of the ductile deformation In the MCT zone, different indicators of the deformation viz. cleavages, folds and lineations have special characteristics that have been particularly studied by Pecher (Pecher, 1977, 1978; Bouchez and Pecher, 1976) and BruneI (BruneI and Andrieux, 1977; BruneI, 1983, 1986) in Nepal. Similar conclusions have been obtained however from the rest of the Himalaya, the Kumaon and Sikkim regions in particular. The main cleavage (S2 according to the nomenclature of Pecher, 1978) is very regularly north-d~pping with an average dip of less than 30° (fig.10). In the Higher Himalaya crystallines - the Tibetan slab -, above the MCT, it is a metamorphic cleavage or gneissic foliation that generates the characteristic monoclinal morphology of the slab. The MGT plane itself seems to be sub-parallel with the cleavage. Below the MCT, the same schistosity is encountered again in the Lesser Himalayan formations. However in the upper part of the Lesser Himalaya and the base of the High Himalaya crystallines, the schistosity S forms contiguous almond shape figures at all scales (thin section, outcrop) (fig.9). These almonds become more and more flattened as one comes closer to the MCT plane. Their shape is symmetrical in longitudinal vertical sections (ca. E-W) and strongly disymmetrical in perpendicular sections (N-S). As shown for the first time by Pecher (1977, 1978), they result from the association of the main cleavage S with the penecontemporaneous plane S' bearing the same metamorphic crystallizations and revealing differential sliding movement along S planes. S' always dips more to the north than S; this constant relation gives the general direction of sliding, the upper layer being thrust towards the south. This shear-cleavage relation (C-S, cf. Berthe et al. 1979) is an important criteria of rotational deformation~ingle shear type in which sliding prevails upon folding. B2 folds on the contrary are very scarce. Coming from the High Himalayan sedimentary series they almost disappear by flattening in the Tibetan Slab (fig.10). Only in the Lesser Himalaya below the MCT, a few syn- to late-folial folds of isoclinal type reappear. Large recumbent folds are known not to exist, as sedimentary structures in the Lesser Himalaya indicate generally normal upward direction of younging. The folds that have been observed show a dispersion of the axis directions in 78 ,, - 30 29 7 8 ~ 1 0 2 0 3 _ M G T 0 4 0 5 79 - 0 6 0 7 D a 0 9 6 § 10 - _ M B T 80 \ 28 "- -+ 81 " ~ 1 1 81 82 +-- 31 83 " + 27 "- f- 83 - 84 - - 1- 30 86 - 8T + f- 34 - 85 2 6" -+ - + 8 T 88 " + + 88 89 . - L 2 00 km + 8 9 90 - 90 91 - 2 9 26 92 fi g. 8: St ru ct ur al m ap of c e n tr al Hi ma lay a fro m Ku ma on t o Bh uta n (fr om P ec he r & Le f or t, 19 86 ). 1 : Hi gh H im ala ya s ed im en tar y se rie s (P ala eo zo ic- M eso zo ic) ; 2 : Hi gh Hi ma lay a c ry st aU in es (" Ti be tan S la b" ); 3 : Hi gh H im ala ya l eu co gr an ite s (M ioc en e); 4 : Le ss er H im ala ya c ry st al lin e na pp es ; 5 : " Ch ai l" na pp e of t he L es se r Hi ma lay a (no t a lw ay s di st in gu is he d) ; 6 to 8 : in te rn al m et a- se di m en tar y be lt of th e Le ss er H im ala ya , 6 : up pe r (ca lca reo us ) gr ou p, 7 : lo we r (p eli tic to ar en ac eo u s) gr ou p (K unc hha t yp e), 8 : Le ss er H im ala ya au ge n gn ei ss es ( Ul ler i ty pe ); 9 : " Le ss er H im ala ya n" g ra ni te s (C am bro -O rdo vic ian ); 10 : ex te rn al s ed im en tar y be lt (K rol b el t) of th e Le ss er H im ala ya ; 11 : Si w ali k m ol as se ( M io- Pli oc en e). S ou rce s in P ec he r & La f or t, 19 86 , fi g. 1. w '- D 320 fig.9: Block diagram in the MeT zone showing the strong stretching lineation and the almond shape figures i.e. phacoidal foliation of the S2-S'2 cleavage (redrawn after Pecher, 1977 & 1978). the S plane with a maximum near the 'a' direction (Pecher, 1978). BruneI (1983) is of the opinion that at least for some of them, the folds were initiated parallel to the 'a' direction because of the constrictive character of the finite strain. Taking into account the mainly flattening type of the finite strain, the existence of curved axis that do not fold S, as well as the synmetamorphic direction of the folds, Pecher has argued that they were early longitudinal folds reoriented by differential sliding in the foliation phase and thus acquiring their horseshoe or sheath shape. But the most conspicuous feature of the MeT zone is the penetrative mineralogical and/or stretching NNE-SSW lineation (fig.9). The synmetamorphic mineralogical lineation Lm lies in the main foliation plane and is marked by the orientation of minerals (kyanite, sillimanite, amphibole, staurolite, quartz, feldspar), their grouping in elongated trails (chlorite) or the shearing and stretching of the pressure shadows (particularly on the northern side of garnets and feldspars). The stretching lineation LS is especially present below the MeT where it is marked by the elongation of detritic pebbles, and quartz and plagioclase grains, the boudinage of quartz lenses, the mullions of more competent layer such as quartzite and augen gneisses, and by the metamorphic striae in schistose layers (Pecher, 1978). Pecher has shown that LS' although parallel in direction to L , is not born by the main cleavage plane S but is parallel with the diagoWal lines of the S-S' almonds. These lineations are extremely well ruled (NIDO to N3DO) although they may locally follow a fluidal pattern around almonds and folds. Regionally they are best expressed near the MeT plane, vanishing away from it, above or below. They have been observed from one end to the other of the Himalaya with this remarkably constant direction, perpendicular to the general cartographic trace of the MGT and thus considered to be the transport direction along the 'a' axis (Bouchez and Pecher, 1976; Pecher, 1977, 1978; Brunel and Andrieux, 1977; Bordet et ale 1981; Brunel, 1986). V.1.2.b. Criteria of shearing movement 321 Pecher (1978) and Brunel (1983, 1986) have paid special attention to the structural criteria resulting from the shear movements in the MCT zone. Based on the observation of small scale asymmetrical patterns, these criteria can be divided into three main categories : - at the scale of the sample or of the thin section the asymmetrical structures concern polycrystalline assemblage and external schistosity pattern (S2); - at the scale of the crystal the structures concern the internal schistosity pattern (S.); - at the scale of one mi~eral species, the structures are preferred orientations and microstructures. The first category has already been partly described when I spoke of the main cleavage and asymmetrical S-S' almonds observed on NNE-SSW (XZ) vertical sections (cf. (a) supra and fig.9). Other remarkable structures include intrafolial drag folds occurring in mica-rich layers (Pecher, 1978, p.278). Again they are best observed on XZ planes where their S or Z shape give the direction of movement, although because of their curved axes they are also visible on YZ sections (ibid.). A last example concerns minerals that remain essentially undeformed and unrecrystallized during the shearing. Their nature depends on the intensity of deformation and metamorphism that has been achieved, but they include quartz (in Lesser Himalaya low metamorphic grade), plagioclase from the volcano- sedimentary formations (up to biotite-garnet grade) and sometimes K- feldspar, detrital tourmaline and even early metamorphic minerals such as garnet and kyanite. These minerals rotate during shearing while pressure shadows develop; the rhombic shape achieved in XZ sections again reveals the sense of movement (Pecher, 1978). In the second category, the internal structure of porphyroblasts is considered using inclusion trails as markers of the successive orientations of crystals. Garnets from the Himalaya provide spectacular examples of such rotated inner schistosity in XZ sections (~ Ishida and Ohta, 1973; Pecher, 1978; Brunel, 1983, 1986; Karanth, 1985) but other minerals such as epidote, chloritoid, staurolite, kyanite, plagioclase may show similar patterns (Pecher, 1978). The sense of movement deduced from these structures is always compatible with a southward shearing in the MCT zone, the rotation reaching up to 360 0 with a variable relative ratio of growth over rotation (ibid.). The concomitant variations of chemistry of the garnets have been used to evaluate the metamorphic PT conditions of the shearing (see bel'ow). The third category concerns the quartz microstructures and preferred orientations. Based on network microstructures in quartz-rich formations (metasandstones and quartzites mainly), Pecher (1977, 1978), Bouchez and Pecher (1976, 1981) and Brunel (1983, 1986) have been able to recognize 322 ~upperl " , Midlands group ~''>., lower Midlands High Hi malayan crystallines ~ Tibetan sed. series ~ Granitic network Fig.10: SSW-NNE section across the Lesser and Higher Himalaya in central Nepal glvlng the general disposition of the Midlands group with the southern flexure F, the High Himalyan crystal lines (Tibetan Slab) and the Tibetan sedimentary series (Cambro-Ordovician to Devonian in this section). The section, roughly passing through Annapurna II summit (7937 m), is synthesized from Mascle & P~cher (1977), P~cher (1978) and Le Fort (1981). The five setreograms (from Picher, 1978) show the general evolution of the deformation along the section: 1 = in the lower part of the Tibetan sedimentary series, the B2 folds are numerous, associated with a strong axial plane cleavage S2 (and an intersection lineation L , not shown on the diagram). 2 = in the Tibetan Slab, Formaiton I, the main tectonic e~ement is the S2 cleavage; few associated B2 folds (cherries) are dispersed in the S2 plane whereas the main lineation is a transverse mineral lineation L2 (dots). 3 = in the upper Midlands beneath the MCT the main cleavage S2 is associated with a strong mineral and stretching lineation L (folds B, not shown, are very few, very small and often sub- parallel to L2). 4 = t~e Midlands ~ave been folded around large 8 late folds, such as the Kunchha-Gorkha anticlinorium, associated more to the south with a S fracture cleavage (dotted contour). 5 = in the southernmost part of the Midlands, large sc~le recumbent folds 82 are associated with smaller folds of variable direction (and S2-S intersection llneation mainly ESE-WNW). 0 Stereogram are Schmidt nets, lower hemisphere. Contours for stereograms 1 to 4 are 2, 5, 10, 15, 20 and 25%; for stereogram 5 it is 1, 2.5 and 5%. Number of measurements: 1 (40 82 a~d 52 S2)' ~ (133 S2)' 3 (111 S2 and 32 L2), 4 (152 So and 67 S3)' 5 (50 B2). No vertical exaggerahon. four to five main microstructural zones corresponding to increasing temperature and plastic deformation. From the bottom upwards they are 323 - a zone of preserved detrital quartz grains extending below some 3 km below the MCT plane (in the lower Midland formation in Nepal), where grains are only slightly deformed, - a zone of relict porphyroclastic microstructures between about 3 and 2 km below the MeT, where the preserved casts of detrital quartz grains are filled with a partially recrystallized mosaic and extend into pressure shadows, - a zone of mosaic microstructure extending for some 2 km below the MeT, where the quartz crystals mostly have a polygonal flattened shape elongated parallel with the stretching lineation, - a zone of ribbon texture, just next to the MeT, where polycrystalline ribbons of quartz elongated parallel to the lineation give to the rock a very typical blastomylonitic appearance, - a zone of exaggerated grain growth microstructures found mainly above the MCT plane in the High Himalaya crystallines, where the large crystal of quartz are interlocked in all three dimensions and display a more or less rectangular grid of subgrains. In these rocks, the statistical orientation of the quartz c-axes (as well as a few a-axes) has been studied in the different zones (Bouchez and Pecher, 1976, 1981; Pecher, 1977, 1978; BruneI, 1980, 1983, 1986; Bordet et al. 1981). In XZ sections, the c-axes are distributed along two unequally populated crossed girdles making a 2 6 angle. The more densely populated girdle may evolve into a single girdle, inclined on the foliation. According to Bouchez and Pecher (1976, 1981) who have studied more than 100 samples, the evolution of preferred orientations is continuous and follows the zoning of microstructure described above. The asymmetry of orientation gives the rotational character of the deformation and enables to predict the sense of shearing in most cases (Bouchez and Pecher, 1981). The 2 6 angle diminishes wih increasing shear but rapidly reaches a minimum value of 30° to 40° and remains constant thereafter. Thus this method does not allow to calculate the strain rates because of the lack of strain markers; only a rough and statistical estimate of shear strain may be obtained using both the microstructure zonation and the population density of the girdles of quartz preferred orientations. Both observations clearly show that the MeT is basically a several kilometres thick shear zone. In the High Himalayan crystallines however, the prevailing high temperature has enhanced the annealing features of the fifth zone. There has been a debate concerning the significance and representate character of the quartz preferred orientation in the Himalaya. For BruneI (1980, 1983, 1986) the quartz c-axis fabrics reflect a late deformation phase with an overall similar thrusting direction to the major phase. Most other tectonic objects (folds, lineations, etc.) are also related to late stage 'cold' thrusting by the same author who has definitely overlooked and confused the relationships between metamorphism and deformation. In fact, as shown by Bouchez and Pecher (1981), when the MeT zone is already too cold to plastically deform, late 324 shear may still induce strong quartz preferred orientations; however the zonal distribution of microstructures parallelizing the thermal one (see also below: metamorphism) strong,ly supports the thermal control of quartz orientation during the main phase of deformation and metamorphism. The fact that fabrics are quite sensitive to successive phases of ductile deformation prevents from retrieving the sequence of deformation phases from them and compels one to study a rather large number of samples in one given area. V.l.2.c. Shear strain estimates Because of the general lack of good strain markers, strain estimates in the MCT zone are very few and very tentative. We have already seen that the obliquity of the quartz c-axis preferred orientations on the main cleavage plane S2 cannot be related directly to the strain intensity and reaches a maximum value at relatively low intensities (Pecher, 1978). Pecher (1978) has used the detrital grains of quartz and feldspars in the metasandstones of the Lesser Himalaya to determine the slope of the strain ellipsoid. Deformation is of flattening type and its intensity increases toward the MCT plane as already shown by the study of the microstructures. For Z arbitrarily chosen as 1, the value of X and Y respectively increase from 1.91 to 4.70 and 1.61 to 3.05. In the Lesser Himalaya of Sikkim, the strain estimates for the main deformation, using detrital quartz grains from the Daling graywacke and feldspars from the Lingtse augen gneiss, have yielded flattening values around 60 % with extension in the direction of the stretching lineation round 150 % (Sinha Roy,1979). Measurements have also been made farther south in the Lesser Himalaya on an oolitic limestone from Tal, Garhwal (Gairola and Saxena, 1981) yielding strain estimates around 24 % on these low grade metamorphic rocks. In northern Pakistan, the Manserah pluton belongs to the Lower Paleozoic 'Lesser Himalayan' belt of cordierite bearing granites extending at the southern edge of the High Himalaya for some 1600 km (Le Fort et al. 1980, 1983a, 1986a; fig. 11). It has been dated by Rb-Sr whole rock isochron at 516 ± 16 Ma (Le Fort et al. 1980). It intrudes metasedimentary formations whose metamorphic grade increases from chlorite to the south to sillimanite in the north (Calkins et al. 1975). Orthogneissification of the granite increases with metamorphic grade. With a homogeneous composition over a large area, this porphyritic granite rich in feldspar augens and microgranular ovoidal inclusions constitutes a particularly fine object for the study of deformation and the evaluation of strain rates. Fernandez (1983) has evaluated both intrusion strain in the non-gneissified Manserah granite and bulk strains in the gneisses. The gneissification bulk strain obtained using deformed microgranular inclusions, quartz-tourmaline nodules and quartz crystals yields AI /A3 values from around 75 in the biotite-garnet grade up to more than 300 in the kyanite-sillimanite grade. Deformation is again of flattening type. + -i'-- _DELli! g!J North Himalaya granites & orthogneisses ...... Formation III type augen gneisses , "lesser Himalaya' granites & orthogneisses Gf!1 Other Lower Paleozoic granites & orthogneisses y-l-..l. Main Mantle Thrust ~ MeT ~ MBT ___ Other faults and thrusts TUflgpO ANDUO .: ... 325 fig.11: Extension of the magmatic formations with an isotopic age around 500 Ma (from Le Fort et a!. 1986a). GM : Gurla Mandata, PESH : Peshawar, PKR : Pokhara, SRI: Srinagar. V.1.2.d. Upward evolution of the stretching lineation We have described the remarkable constancy of the very penetrative stretching lineation from one end of the Himalaya to the other. Through most of the MeT zone, the lineation trajectory is oriented N10° to 40 o E. However, recently, a number of local observations (Ladakh, southern Xizang and central Nepal) on the top of the High Himalayan crystallines have shown that the trajectories were apparently bent towards a more longitudinal direction (Burg, 1983; Gapais et al. 1984; Pecher et al. 1984; Brun et al. 1985; Gilbert, 1986). Upward in the High Himalayan sedimentary rocks, as the metamorphism decreases, this stretching lineation passes into an intersection lineation of b type. Actually, the lineation exhibits continuous-discontinuous recurrent changes in direction as shown in southern Xizang (Brun et al. 1985) and may not be totally synchronous. The stretching lineatio~often marked by fibrolitic aggregates of sillimanite proving that it formed at high temperature. Again, a number of shear criteria show that the mineral- stretching lineation corresponds with a transport direction during ductile shear deformation. However the relative sense of movement is not the same in the three areas where information is availabke : the upper levels moving generally to the west in Ladakh and east in Nepal and southern Xizang. 326 The dynamics of this oblique shearing is not yet well understood. Does it result from an oblique convergence of India vs. Tibet or has it more local causes such as thickness differences in the High Himalayan crystallines or warping accompanying the emplacement of High Himalayan leucogranites ? As stated by Gapais et al. (1984), in addition to the well known north-south thrusting movements, one has now to take into account longitudinal movements of probably regional importance. V.l.2.e. A thick zone of shearing The MCT is thus a several kilometre thick zone of ductile deformation in mid-crustal regime. The thickness is not related to folding; folds such as in Garhwal (Frank et al. 1973; Powell & Conaghan, 1973a) are exceptional because of the intense flattening and stretching. The thrust plane itself corresponds with a thin zone of abnormal superposition of older terrains on younger ones. Where the contrast in lithology is well marked, as in central Nepal between the Palaeozoic schists and dolomitic limestones of the Lesser Himalaya and the metapelites and greywackes of the Higher Himalaya, the MeT plane is easily mapable; where the lithologies are similar, as in Burhi Gandaki valley with quartzites on both sides, the plane cannot be mapped in the field but has to be estimated from the intensity of deformation. Actually this plane is but the geometric locus of the highest strain. Towards the south, the MCT zone reaches into the higher levels of the crust and becomes more brittle. This evolution in space also corresponds with an evolution in time that will be dealt with below, in section V.3. V.l.2.f. How many MCTs ? When Auden (1937) and Heim and Gansser (1939) outlined the Main Central Thrust in Kumaon, they defined it mainly as the basal contact of the crystal lines thrust upon lower grade rocks. For 100 km west of India- Nepal border (Kali valley) they recognized a metamorphic gap less clearly defined more to the west (Anaklanda river). Valdiya (1980) after remapping the entire Lesser Himalaya of Kumaon (1981a) has extensively discussed the position of the MCT shoWing that it lies within the inverted metamorphic sequence (see below) at a higher level than originally proposed. This ductile Main Central Thrust zone as recognized by French (~ Bordet, 1961; Bordet et al. 1972; Le Fort, 1975a; Pecher, 1975, 1977, 1978; Colchen et al. 1980, 1986), Japanese (Hashimoto et al. 1973) and Indian (Valdiya, 1980, 1981a,b) geologists is associated with the main phase of deformation. In a large portion of the Himalaya the MCT lies above a more brittle and later thrust known as Jutogh or Munsiari, etc. thrust (~. Valdiya, 1980), which is responsible for the over thrusting of some of the crystalline outliers of the Lesser Himalaya, especially in the Kumaon Himalaya. Where the MCT has not been reworked by later movements, there is no visible break at its level, all formations are in general acordance and differences relate more to the different lithologies than to metamorphic or deformational jumps (cf. Le Fort, 1975a). The MCT plane thus 327 corresponds with the maximum intensity of rotational deformation and is not always readily recognizable. However the superposition of later brittle movements accompanied by generally limited retrogressive metamorphic reactions have led to confusion and to the multiplication of thrust planes to which similar importance were given. Once again, the thickness of the ductile zone of the MGT is unique in the Himalaya. Recently a structure equivalent to the MGT has been claimed in southern Tibet to the north of the MGT : the Kangmar thrust (Burg, 1983; Burg et al. 1984; Tapponnier et al. 1981). This thrust divides the High Himal~series into two uni~a northern one made up of epi- to anchi- metamorphic Palaeozoic to Mesozoic sequence overthrust over less metamorphic Mesozoic series. It repeats a good part of the Palaeozoic Mesozoic sequence and has been followed over a long distance (Burg, 1983). These authors suggest that at depth it is a MGT equivalent; however from Burg's sections it appears to be more an accompanying thrust at a higher level with none of the high grade deformation and metamorphic character typical of the MGT (cf. section V.1.3). V.1.2.g. Age constraints on the MGT The age of thrusting along the MGT is difficult to constrain as relevant evidence is limited. Stratigraphic constraints include the incorporation and deformation of middle Jurassic in the MGT zone of Kumaon (Powell and Gonaghan, 1973a) and the overthrusting of uppermost Paleocene to lower Eocene nummulite bearing series in western Nepal, Barikot area (Fuchs & Frank, 1970). These Eocene series are overlain by a conglomerate and some 50 m of quartzite sandstone and shale that would represent lower Miocene Dagshai series (ibid.). However, this lithologic correlation has too important conseque~ for the timing of thrusting, to be taken for granted. A rather large number of minimum ages have been obtained on rocks from the MGT zone (cf. Saxena, 1981 for a partial list). They correspond generally to cooling-ages and should thus give an upper limit to the movements along the MGT. However, the dispersion is so large (almost from 3 to 50 Ma for Himalayan ages) that they cannot help to constrain the timing of the MGT. Absolute ages of High Himalayan leucogranites, contemporaneous with the MGT movement that produced them (see below), would also help. Unfortunately these granites proved to be extremely heterogeneous in Sr isotopes (~. Vidal, 1978; Vidal et al. 1982, 1984). It is only recently that a Rb-Sr isochron age of 18.1 Ma has been obtained (Deniel, 1985). U- Pb mineral ages on monazite of the High Himalayan leucogranites are also interesting because of the high blocking temperatures of this mineral, closer to crystallization temperatures of the magma; four ages obtained on two plutons vary from 25 Ma (Deniel, 1985; Deniel et al. 1987) to 14.3 ± 0.6 Ma (Scharer et al. 1986). If the onset is difficult to date, this would give minimum span of 10 Ma for the MGT functioning before the movement shifted to a lower and more southern intracontinental thrust : the Main Boundary Thrust (MBT). Last indication, the contribution of Himalayan metamorphic minerals in the heavy mineral fraction of the Sub-Himalaya molasse (see part VI) 328 at least starts with the lower Siwaliks, some 15 Ma ago (cf. table IX), indicating that the MCT has already been active for some time during middle Miocene. V.1.3. Continuation of the movement V.1.3.a. The "cold nappes" of the Lesser Himalaya In a number of well documented cases, there seems to be an abrupt change in the grade of metamorphism along a plane defined as a thrust in the Lesser Himalaya. This is not taking into account the cases in which there is a change in the feldspathic content of the rock (arkosic or felsic volcanic formation, or reworked granitic basement); in such cases there is no real jump in the metamorphic conditions. Actually, real abnormal superposition of different metamorphic grades do exist; they are very conspicuous in a number of metamorphic and tectonic windows such as along the Sutlej river in Kumaon (Shali window, West,1939), the Rangit river in Sikkim (Rangit window, Acharrya & Ray, 1977), or the Chenab river in Punjab (Kishtwar window, Fuchs,1975) (see also Bhargava, 1982). All examples are several ten of kilometers in length and show rocks around garnet grade, lying on rocks in the sericite grade. Similarly, at the front of the Himalaya, in particular along the southern edge of the border synclinorium of the Mahabharat range and parallel to the MBT, "cold" thrusts occur, showing brittle type of deformation with mylonites and retrograded metamorphic assemblages. Such thrusts have also been followed and mapped for several tens of kilo- meters : ~ Stocklin and Bhattarai (1980) south of Kathmandu. In this case the thrust can be traced all around the Kathmandu nappe until to the north it progressively merges into the MCT zone with the same bulk lithologies and an increasing grade of metamorphism. The explanation becomes obvious : these cold thrusts correspond to the continuation in a more brittle domain of the MCT deformation (fig.12). Such a transition from ductile to brittle accompanies a thrust climbing up section in the crust along the base of a wedge shaped nappe. To this evolution in space we may add an evolution in time. The MCT movement did not stop at once but as erosion and tectonic denudation were operating, the thrust itself became shallower. The transition from ductile to brittle happened as the system was cooling down. Movement was still active on the MCT or transferred to new thrust planes such as the one visible in the Shali and Rangit windows. Ductile movement in the MCT zone is typically synmetamorphic whereas the brittle one is clearly late metamorphic. V.1.3.b. The backthrusting Clearly evidenced in Ladakh, the piling of nappes results from a polyphased tectonics with a first south vergent movement followed by a second north vergent movement partly reactivating the primary thrust contact (Bassoullet et al. 1978 & 1980a; Mascle, 1985). This second phase has been compared to~well known retrocharriage in the Swiss and French Alps and results in the formation of a fan-shaped structure. The 1 2 3 4 5 s Krol belt '.:. ~ '. :,', ', ......... """1.:.:.: .• : ....... . !!H~:~:}~~~~[~1~1~?{~:~1flj;J~;~I~t~~%~i~~,~~t~~y~0~~;~*}~~~~:~~:;;Vi\~~~j;~ S,wal,ks Lesser Himalaya nappes MCT zone 329 N tethyan sediments .0-::- Fig.12: Diagrammatic restored sections for the tectonic evolution of the Lesser Himalaya (reproduced from Pecher in Pecher & Le Fort, 1986). Stages 1 and 2 correspond to the initial thrusting long the MeT developing an inverted metamorphism (shaded area). stages 3 to 5 stacking of the different units due to cold thrust tectonics, after the MGT. Thrusting, along the different planes ¢2 to ¢4 corresponding to the MBT. 330 reason for this backthrusting is not clear; for part of it at least, Gapais et al. (1984) and Gilbert (1986) have suggested that it is due to gliding tectonics to the north. In southern Xizang, backthrusting is reported to occur post-Miocene (Burg et al. 1987). After the nappe emplacement, the second phase consists (Burg & Chen, 1984) of upright folds with a faning cleavage dipping to the north in the south and to the south in the north. Collapse of the thickened crust is suggested to explain the movement to the north. V.1.3.c. The "normal faulting" In southern Xizang, south of the Lhasa-Xigaze area, Burg (1983) after Academia sinica (1980: the Chiatsun thrust), has described and mapped a series of 500 to 600 km long normal faults, parallel to the elongation of the range and approximately located at the limit between the High Himalaya crystallines and the High Himalaya sedimentary series, i.e. at the limit between the infrastructure and the superstructure. A similar but limited structure has been mapped in central Nepal north of the Annapurna range where Devonian to Jurassic rocks are folded in a north vergent fold, the overturned limb of which is thrusted onto amphibolite grade rocks (Bordet et al. 1971, 1975; Colchen et al. 1980). This normal faulting is accompanied by a thick mylonitization that affects the gneisses and leucogranites of the High Himalaya (Burg et al. 1984a). The amount of northward movement taken up by these faults is difficult to estimate. In southern Xizang, the metamorphic gap between the two walls represents several kilometers of thickness missing. Depending on the slope of the thrust (between 15 and 30 e most likely), the down dip movement may amount to anything between 4 and 25 km. In Ladakh (Herren, 1987), the amount of movement along a similar normal fault shear zone, mapped along 35 km E-W, has been estimated around 25 km. The timing of the normal faulting, as in the case of other mountain ranges such as the Alps or the Canadian cordillera (Read & Brown, 1981), is posterior to the thrusting along the MCT and the accompanying metamorphism. It postdates the crystallization of the leucogranites but it is cut by the system of N-S normal faults responsible for the recent Tibetan grabens. Thus it can be bracketed between the mineral ages obtained on the leucogranites, i.e. around 12 to 14 Ma (cf. table 1 in Le Fort et al. 1987), and the age of the oldest deposits in the grabens, i.e. 2 to 5 Ma (see below, Section VI.1). Ten million years are still a long lapse of time, but the normal faulting has to be contemporaneous with the thrusting on the MBT. This means that, around the late Miocene, the Himalaya simultaneously experienced compression in the south and extension in the north. Several tentative explanations have been given. For Burg et al. (1984a) the vertical stress in the superstructure exceeds the horizontal stress and causes the northward gliding. For Burchfiel & Royden (1985), the rotation of the stress axes following the partial loading of the crust with some 5 km of sediments, is an additional factor for the gravitational collapse of the Miocene topographic front between India and 331 Tibet. These authors have overlooked the fact that the Miocene topographic front they suggest, lies north of the present topographic front, which is quite unreasonable. Their model, however, shows how topographic gradients can exert an influence on the tectonic framework. In my opinion, the backthrusting is linked to the thickening of the crust at the back of ongoing thrusts, possibly triggered by the existence of ramps along the MBT and by topographic relief at the surface. The whole system is comparable to the bulldozer wedge model of Davis et al. (1983) in which tectonic wedges in compression deform internally until a critical taper is attained after which it will slide above the brittle- plastic transition. The backthrust is located around the boundary between the infrastructure (crystallines) and the superstructure (sedimentary series) of the peak metamorphism. V.2. METAMORPHISMS We have already seen that the high pressure blueschist metamorphism is much older than the obduction on the Indian continent and consequently than the collision. Following the collision event, metamorphism is apparently limited to the Himalayan domain, south of the suture (Le Fort, 1986). There, two different metamorphic belts can be distinguished: the north Himalayan Tibetan belt and the Main Central Thrust belt (ibid.). V.2.1. The North Himalaya metamorphism The area north of the High range has, for most of it, only been opened to investigation less than a decade ago. The available information essentially comes from the work of the Chinese teams and the Franco- Chinese program in southern Xizang, and from the work of Honegger in Ladakh where this zone narrows. South of the suture in south Xizang, the stack of nappes of the flysch belt has been metamorphosed syntectonically with the first phase of deformation (~ Tapponnier et al. 1981; Burg, 1983; table VI). This epimetamorphic metamorphism of prehnite-pumpellyite to greenschist facies forms a narrow belt that can be followed westward along southern Xizang (the Tsangpo belt of metamorphism of Zhang Qi et al. 1981). It connects with the prehnite-pumpellyite zone of northern Ladakh (Honegger et al. 1982; Honegger, 1983) where it is described as outlasting the nappe-- movement in a late kinematic way for the low grade rocks. Down in the structure, the higher grade ones, mesometamorphic, remain synkinematic with the general pattern of a metamorphic dome. This higher grade of metamorphism is also encountered at the eastern syntaxis of the Himalaya in the flysch of the Namche Barwa region (Chang et al. 1977). In Ladakh, Honegger (1983) describes the following zones of metamorphism : - prehnite-pumpellyite, (actinolite-pumpellyite), actinolite-albite, tschermakitic hornblende-albite, hornblende-oligoclase, (hornblende- andesine), in the metabasic rocks; - chlorite, garnet-biotite-chlorite, garnet-biotite-staurolite, sillimanite, in the pelitic rocks. 332 In southern central Xizang, south of Lhasa-Xigaze, the same succession of metamorphic zones occurs in the Tibetan sedimentary series underneath and to the south of the flysch nappes. The grade of metamorphism increases together with the intensity of the deformation with depth. A few eroded domes, such as Kangmar (Burg, 1983; Burg et ale 1984b) enable us to observe the vertical succession of the metamorphic zones; elsewhere, they can be followed on the map towards the south as deeper stratigraphic and structural levels are encountered. Anchizonal metamorphism may reach the southernmost exposures of Jurassic in Nepal (Dunoyer de Segonzac in Le Fort, 1986) and greenschist affects the Jurassic calcschist of Chandra valley, Himachal Pradesh (Powell and Conaghan, 1973a). Greenschist metamorphism also reaches the Triassic, and lower Palaeozoic rocks are generally in the amphibolite facies. Towards the base of the Tibetan sedimentary series, the amphibolite facies rocks may pass with or without a break into the highly metamorphic and tectonized infrastructure of the Tibetan Slab. According to Burg et ale (1984b) whose view differ from the previous observation by Tapponni;r-et ale (1981), the metamorphic minerals are clearly linked to the first phase of deformation, being syn- (micas essentially) to late-kinematic or early D2 (chloritoid, garnet, staurolite, kyanite). The simplest explanation is that this first phase of deformation is related to the obduction and piling of nappes prior to the collision, and this metamorphism corresponds with the partial thermal reequilibration that accompanies and follows it. This first phase of eoBimalayan metamorphism passes continuously downward and southward into the second one described below. It is most important as it changes completely the thermal structure of the northern edge of the Indian continent. For this reason we need many more detailed studies and data concerning the timing of its evolution. V.2.2. The inverted metamorphism of the Main Central Thrust zone From Kashmir to Arunachal it has been possible to follow a metamorphic belt whose zonal succession of index minerals and P-T range are remarkably similar. Its characteristics include the inverted zonation of less metamorphic terranes overlain by more metamorphic ones (fig.13). The Darjeeling region is the historical place where the inverted metamorphism was early observed (Heim and Gansser, 1939; Ray, 1947). On the map by Bhattacharya and Das (1981) of the Darjeeling hill there is a clear horizontal disposition of the metamorphic isogrades with biotite grade at the bottom, garnet intermediate and staurolite at the top (fig.14). This pattern has been called reverse, or inverted, or divergent metamorphism in various parts of the Himalayan range. It has been observed and mapped west and east of Darjeeling, from chlorite to sillimanite grade: in Bhutan (Gansser, 1983a), in eastern Nepal (Hashimoto et ale 1973), in central Nepal (Colchen et ale 1980) as in Kumaon (e.g. Das & Pande, 1973; Frank et ale 1973; Kumar et ale 1974) and in Ladakh (Honegger, 1983). Figure 13 shows a compilation~hese metamorphic isogrades. Table VII gives a list of the main minerals present in the different zones depending on the lithology. It is also noteworthy that the fluids contained in the fluid inclusions of the 78 3 1 ~ 29 + 7B ~ 1 CT I 2 r.. :I2 l 3 0 4 ~ 5 + 0 6 0 7 0 8 80 + +- 82 + 31 " 84 " + +- 3D " 8 2 ' 27 '+ + 84 " 86 " + 26 "- ' 86 " + 8 8 ' +- - l- 8 8 " 20 0k m ! 90 " + _ _ 2 9' 26 92 90 ' Fi g. 13 : M eta m or ph ic m ap of ce n tr al H im ala ya f rom K um ao n to B hu tan (s am e ar ea as fig ur e 7) (re pro du ced fr om P ec he r in P ec he r & Le F or t, 19 86 ). 1 : Hi gh H im ala ya s ed im en tar y se rie s u su al ly l it tl e or no t m et am or ph os ed , 2 : Hi gh H im ala ya l eu co gr an ite s; 3 si lli m an ite c o rd ie rit e zo ne ; 4 : si lli m an ite + ky an ite z on e; 5 : ky an ite st au ro lit e zo ne ; 6 : bi ot ite + ga rn et z on e; 7 bi ot ite z on e; 8 : ch lo rit e zo ne . So ur ce s in P ec he r & Le Fo rt, 1 98 6, fi g. 1. w w w 334 -bY] garnet [;:::;:;:;:] biotite _ fault _1EOO_ • Fig.14: Metamorphic distribution map around Gangtok (Sikkim) showing the typical inverted zonation of the biotite, garnet and staurolite grades (redrawn from Bhattacharya & Oas, 1981). o abundant quartz lenses show a zonality of composition approximately parallel with the metamorphic zonality, with a maximum CO2 content close to the MCT (Pecher, 1978, 1979 & in press). On the N-S transverse section of figure 15, the isograds clearly appear in their reverse order and the metamorphic zones seem to increas e in thickness. On the EW longitudinal section of the same figure, the slight obliquity of the isogrades on the lithology - schistosity as on the MCT is conspicuous. It is also remarkable that there is a direct correlation between the thickness of each zone and the thickness of the High Himalaya crystalline. The PT conditions of this metamorphism have first been evaluated using more or less calibrated curves of the metamorphic assemblages met in the MCT zone (table VII). The most constraining observations include the absence of regional andalusite, the abundance of kyanite and sillimanite, the occurrence of zoisite with kyanite and quartz, the 335 U Manaslu granite !":":.:::\ .. ::- 336 Table VII the seven zones of progressive and inverted metamorphism in Nepal (after Pecher 1978, Peche r lit Le fort 1986). zone 1 zone 2 zone 3 zone It zone 5 zone 6 zone 7 peli tic (0 always present) Chl, Mu, Ctd B'f;" Chl, Mu, ct d Gr, Bi, Chl, Mu, Ctd St, Ky, Gr, Bi, Mu ± Chl !t, Gr, Bi, Mu Sill, Gr, Bi, Mu ~Sill, Gr, Bi ± Mu, fK carbonated (Cc always present) Phl, £p, Pl Phl, Tr, e:p, Sc, Pl, Sph Di, Hb, Bi, e:p, Sc, Gr, Id? idem no known equivalent Some information about the evolution of the metamorphic conditions is obtained from the zoning of garnets, suggesting that on both side of the MCT (in the Lesser and Higher Himalaya), the starting PT condition is different but gradually converge. In eastern and central Nepal, the spessartine-rich cores of garnets in the Midlands and the pyrope- grossu lar rich cores of garnets in the Tibetan Slab converge towards similar almandine rich rims (Arita, 1983; Le Fort et al. 1986b). So far, in southern Xizang only, relicts of high pressure granulites have been found in the deeper portions of the crystalline slab (Burg et al. 1987), with retrogression of the primary clinopyroxene-garnet assemblage into orthopyroxene-plagioclase-hornblende. In Kali Gandaki valley (central Nepal), a PT estimation on coexisting plagioclase-garnet-biotite-kyanite assemblages shows a normal temperature gradient in the crystallines until about a kilometer above the MCT thrust plane where it starts decreasing, whereas the pressure continues to increase (Le Fort et al. 1986b). These results fit the model of inverted metamorphism by downward transfer of heat from a hot overriding slab. Further east, in Darondi and Burhi valleys where the crystallines are particularly thick, the temperature estimates compared with the pressure estimates remain fairly constant over a wide zone (Hodges et al. 1986). One possible explanation of this is that anatexis and anatectic products being widely dispersed, may have effectively buffered temperatures over large portions of the thick crystal lines (ibid.). ----T-here has been some debate about the importance of the retrogressive metamorphism. Some authors (~ Remy 1972, 1974; Hashimoto et al. 1973; BruneI, 1983, 1986) pretend that the main metamorphism was pre-nappe and that thrusting is accompanied ~ retrogressive assemblages. Others (~ Le Fort, 1975a; Pecher, 1978; Lal et al. 1981; Sinha Roy, 1981; Virdi, 1981; Caby et al. 1983; Pecher & L~t, 1986) maintain that the main metamorphis;-;a; syntectonic or even late tectonic (Das & Pande, 1973; Bhattacharya & Das, 1981), and that retrogression occurred during the late stages of thrusting when the denudation had already been operating and the thrusting was "colder" and less ductile. Actually, the closer to 337 p kb 10 Manaslu 9 4 ________ __________ L-________ __________ -4 T 400 500 600 700 ·c fig.16: Schematic pressure-temperature evolution diagram for the MeT zone in central Nepal, Manaslu (Burhi Gandaki) region (reproduced from Pecher & Le fort, 1986). Evolution starts during eo-Himalayan times (t j ) with Barrovian metamorphism in the High Himalaya crystallines (Tibetan Slab : TS). Thrusting along the MeT zone develops prograde metamorphism in the Lesser Himalaya (Midlands) and cooling at the base of TS until time t 2• During thrusting, erosion and tectonic denudation of the overlying sedimentary series produces quasi-adiabatic heating particularly in the upper part of TS with a maximum around t2• It is the main period of magma generation. Cooling down of the system follows dotted llnes from t2 to t ; it is mainly recorded along cold thrust movements. Aluminium silicate e~uilibrium curves according to Richardson et al. (1967 : RGB) and Holdaway (1971 H). Mineral isogrades and beginning of anatexis curves as in Pecher (1978). the MeT front, the larger the retrogression has developed; on the contrary in the more internal parts, there is a good agreement between the initial temperature of the crysta11ines and the temperature attained by the overthrust Midlands (Pecher, 1978; Pecher & Le Fort, 1986). All the above section about deformation corroborates the syntectonic model. Part of the sillimanite crystallization is also linked to the retromorphic episode. Actually sillimanite has two habits : prismatic and more often fibro1itic. A good part of the fibro1ite occurs as a 338 replacement of kyanite or, in the migmatitic zone, on late shear planes with quartz and muscovite cordierite in the migmatitic zone. This sillimanite is retrogressive in the way that it originates by a drop in pressure and no or little increase in temperature. Although it comes on top of the kyanite, it is different from the inverse metamorphic zonation in the Lesser Himalaya (Pecher & Le Fort, 1986). The study of the fluid inclusions of the numerous quartz lenses interfoliated in the High Himalaya crystallines (Pecher, 1978, 1979, in press; Sauniac & Touret, 1983; France-Lanord et ale in press) have also shown the rather low densities of the (C02 + ~ salt) bearing fluids. This observation points out to the relatively tow pressure conditions of the equilibration of the fluid inclusions associated with the late low pressure - high temperature metamorphic stage (Pecher in press). Thus the main phase of regional metamorphism (t1 on fig.16) is overprinted in the crystallines by a retrogressive episode (t2 ) due to a major drop in temperature at the base and a major drop in pressure at the top of the crystal lines. This evolution continues post-kinematic "statically", as marked by the annealing textures in quartz (Bouchez & Pecher, 1981), around time t3 (fig.16). Deformation becomes restricted to thin and cold shear-zones along which a limited and final greenschist facies retrogression develops (Pecher & Le Fort, 1986). Globally, the rapid drops in P and T during the main thrusting activity can be linked: - for the pressure, to the rapid erosion and/or tectonic denudation (i.e. a gravi tational collapse, cf. fig.12) of the overlying sedimentary series (Caby et ale 1983). Some-of the north-east vergent sillimanite bearing shear planes at the top of the crystallines (Burg, 1983; Pecher et ale 1984; Gapais et ale 1984; Brun et ale 1985; Gilbert, 1986), and the huge northward recumbent folds suc~the Annapurna fold (200 km long x 25 km wide x 6 km high visible) in the sedimentary series (Colchen et ale 1980, 1986) may be markers of this denudation tectonics; ---- for the temperature, to the rapid thrusting of the hot crystallines on the cold Lesser Himalaya formations. V.3. MAGMATISM: THE HIGH HIMALAYA LEUCOGRANITES The High Himalaya granites are the only magmatic products of the Himalayan orogeny and they have very special characteristics. A summary paper on them has recently been published (Le Fort et ale 1987), to which the reader may turn for more detailed informations and references. Only the main characteristics and main points will be repeated here. V.3.1. Localization, petrography and structure The Himalayan leucogranites form, between Nanga Parbat and eastern Bhutan, a dozen or so principal plutons (several te2 to several hundred of square kilometers that total more than 10,000 km and innumerable pods and dykes (fig.1n. They are intrusive in the top of the High Himalaya crystallines - or Tibetan Slab - and in the Tibetan sedimentary series up to Cretaceous. They lie on top or to the north of thick zones of 339 + '00 ! 200 ! 84'E -j- Fig.17: Enlarged geological sketch map of central Himalaya, from Kumaon to Bhutan, and Transhimalaya, showing the distribution of High Himalaya and North-Himalaya granitoids (reproduced from Le Fort et al. 1987). 3 to 13 : plutons of High Himalaya leucogranites listed in ibid; 3 Badrinath; 4 Api; 5 : Mustang (Mugu); 6 : Manaslu; 7 : Shisha Pangma; 8 Nyalam; 9 : Everest-Makalu; 10: Gabug (Yadong); 11: Chung La (Chomolhari); 12 : Gophu La; 13 : Monlakarchung. migmatisation where the crystal lines reach sillimanite to sillimanite- cordierite metamorphic grade. There is a direct correlation between the thickness of the crystal lines above the MCT, the grade of metamorphism and migmatisation attained and the existence of abundant leucogranitic material above. This observation has lead to a model of generation of the leucogranite (Le Fort, 1975a, 1981, 1986). The floor of the plutons is in general concordant with the foliation of the under lying formations whereas the roof has the shape of a large cupola intruding the Tibetan sedimentary series (fig.lS). Contact metamorphism at the floor is limited to the development of wollastonite within a few meters of the contact; at the roof a high grade metamorphic aureole develops for some 50 meters with garnet-staurolite-2 micas or 340 Fig.18: Block diagram of the Manaslu massif showing the High Himalaya sedimentary series intruded by the leucogranite (+), the underlying crystallines above the MeT (dents) and the underthrusted Midlands (broken lines) (reproduced from Le Fort et al. 1987). pyroxene-epidote. The pluton are surrounded by a dense and spectacular network of aplite and pegmatites veins extending longitudinally for a considerable distance. Petrographically, the Himalayan leucogranites are very homogeneous and devoid of igneous (microgranular) inclusions. An average mineralogical composition for the Manaslu includes : 32% of xenomorphic quartz, 37"/. of subautomorphic slightly zoned (An 21 to 2) plagioclase, 21% of perthitic K feldspar, 7% of often euhedral muscovite and 3% of biotite; in addition, minor quantities of fibrolite and/or garnet may be present. Accessory minerals are few and include apatite, zircon, uraninite, monazite. Tourmaline may represent several percents of the rock although it is not usually part of the original assemblage but associated with a pervasive network of dykes, veinlets and pods. The grain size varies from mostly fine to coarse. Structurally, the planar disposition of the micas marks the magmatic foliation. A ductile deformation accompanied by a schistosity is super imposed on it; this deformation also breaks the minerals including the tourmaline. Finally, shear bands may in turn affect the schistosity 341 wi th a stronger dip to the NNW to WNW indicating a "backward" shearing movement. This structural pattern indicates the late tectonic timing of the leucogranites with regard to the main deformation phase DZ of the MCT. V.3.Z. Geochemistry V.3.Z.a. Major elements The High Himalayan leucogranites have a very constant chemical composition reflecting their very steady mineralogical composition. They are mostly leuco adamellites (fig.19) forming an aluminous, very leucoc ratic and quar tz-poor association accor ding to the class ification by Debon & Le Fort (198Z, 1984). The main variation concerns the KZO/NaZO ratio that may be explained by the B and F content and the variable water saturation of the magma (Le Fort, 1986; Le Fort et al. 1987). The minerals themse lves also have a very constant composition from one pluton to the other as well as within one pluton. In particular, no major variation appears over the entire thickness. The whole rock composition is compatible with the minimum melt composition at the range of pressures that can be deduced from the thickness of the plutons and of the Tibetan sedimentary cover. V.3.Z.b. Trace elements Among the numerous trace elements analyzed, some are specially interesting for the comparison that can be done with the gneisses of the root zone. For example, the range of Ba and Sr contents as well as of Ba/Sr ratio in the gneisses overlap the range of the granite (Cuney et al. 1984). - On the contrary, incompatible elements such as Zr, Th and REE show a low to very low content when compared to the Tibetan Slab gneisses as well as to average granitic rocks. Light REE and Th are mainly contained in monazite (ibid.) and, in turn, the abundance of monazite is mainly controlled by~abundance of biotite that contains it as shown by the direct correlation between REE and the parameter B = (Fe + Mg + Ti) (Holz, 1987). This suggests that the low content in these elements is an early characteristic of the magma, acquired either during the melting process or before the emplacement of the granite by fractionation of monazite from saturated magma (Montel, 1986). Zr also has a low solubility in leucogranitic magmas (Watson & Harrison, 1983). It is to be noted that these solubilities may be drastically changed by the addition of such components as CO (diminution), F or B (increase) which are known to occur in large quantities in the High Himalaya granites or surrounding formations. U, mainly contained in uranin~te, is rather abundant, much more than in the gneisses. Large variations£rom one sample to the other do not correlate with other elements jO~ structural pattern. This behaviour could result from the recrystallization of a part of the uraninite from a fluid phase at still high temperature. 342 a -140 gd • / I 84.5._ -- /, / / / .F2 / Q 240 200 160 120 -20 p Fig.19: 284 analyses of the High Himalaya Manaslu leucoadamellite presented as density curves in the nomenclature diagram of Debon & Le Fort (1982, 198~). Q = Si/3 - (K + Na + 2 Ca/3), P = K - (Na + Ca) in gram-atoms x 10 of each element in 100 g of rock; gr : granite, ad : adamellite, gd : granodiorite (reproduced from Le Fort et ale 1987). V.3.2.c. Isotopes The radiogenic isotopic composition of the High Himalaya granites is very special and was unexpected (Vidal, 1978; Vidal et ale 1982, 1984; Ferrara et ale 1983; Scharer, 1984; Deniel, 1985; Denie~al. 1987). These characteristics include an extreme heterogeneity ~~ th S6Sr isotopic compositions within a range of high to very high Sri Sr initial ratio (mainly between 0.74 and 0.78), an equally heterogeneous Nd isotopic composition (E. Nd from -13 to -17), ~tl7 a ~04Y constant Pb isotopic composition wi Eh very high values in Pbl Pb well above the earth evolution curves. These results have several important implications : - because the gneisses and migmatites from the root zone have the same special characteristics these results confirm them as source-rocks; - because of their heterogeneity, whole rocks isochrone dating is usually imposs i bl e. Those that have been published (~ Ka i, 1981a ; Ferrara et al. 1983) result from insufficient sampling, approximation, etc. Only one isochrone age (18.1 ± 0.5 Ma) has so far been obtained (Deniel et al. 1983, 1987) on 11 out of 18 samples from a 150 m long outcrop systematically sampled; - mineral absolute ages are scattered in the Oligocene-Miocene 343 epochs: mainly from 25 Ma by U-Pb on a monazite from the Manaslu (Deniel, 1985) to around 11 Ma by Rb-Sr and K-Ar on micas from Bhutan (Gansser, 1983a). This 14 Ma span reflects the long cooling history of the granites synchronous with the MCT movement; - the isotopic compositions necessitate a long crustal residence time (around 2 Ga). The High Himalaya granites are of extremely pure crustal breed produced in a portion of continental block secluded from mantle influence since Precambrian time at least. Analyzed only on a few plutons, the stable isotopes give uniform compositions (Sheppard in Le Fort 1981; Blattner et al. 1983; Vidal et al. 1984; Ferrara et al. 1985; Fran1S-Lanord et al. 1985 & in --- preparation; Debon et al. 1986a). 6 0 values are high (between 11.0 and 14.0 %0) and simila~those from the source-rocks; 6D values are relatively low (from -90 to -75). The uniform values of 6D in massive pluton and the high temperature indicated by the equilibrium fractionation of muscovite and biotite implies that most of the plutons have not been modified post magmatically by externally derived fluids (France-Lanord in Le Fort et al. 1987; France-Lanord in prep.). Thus, the Sr isotopic heterogeneity has to be magmatic and implies that granite has emplaced as a succession of magma batches that have kept the Sr isotopic ratios of their source. Their high viscosity prevented them from mixing. V.3.3. The North-Himalaya belt Some 60 km to the north of the eastern half of the High Himalaya belt of granite occurs another belt of plutons: the North-Himalaya or Lhagoi Kangri belt CDebon et al. 1981; fig.17 & 11). It extends from the eastern syntaxis to the Gurla Mandata gneissic dome where it seems to end. It is made up of some 16 elliptical bodies aligned around 70 ~ south of the Indus-Tsangpo suture and covering an area of some 4000km (Le Fort, 1986; Debon et al. 1986a; Le Fort et al. 1987). They intrude Palaeozoic~esozoic metasediments where they some- times develop a strong metamorphic aureole with staurolite, garnet and kyanite replaced by andalousite (Burg,1983). The plutons are made up of two different rock types: a more or less gneissic porphyritic granite and a two-mica adamellite. The first type very much resembles the "Lesser Himalaya cordierite granites" and the Formation III augen gneisses from the crystallines (fig.ll), whereas the second, often leucocratic, resem- bles the High Himalaya adamellites. The relative abundance of these two types varies from one pluton to the other; they can be associated or form the entire pluton. Chemical and isotopic studies on this belt support this two-fold grouping and give a lower Palaeozoic age to the first one and a Miocene age to the second (table VIII; Debon et al. 1986a). 344 Table VIII: summary of radiometric ages published on the North-Himalaya belt. WR = whole rock, Bi = biotite, Mu = muscovite. method Rb - Sr 87srt86Sr K - Ar 39Ar _ 40Ar i niti al 485 :I: 6 (6 WR) 0.7186 31.8 (Bi) 17.3 (mica) porphyri tic Wang et a 1. 1981 Zhang et al. 1981 21.7 ( " ) 484:1: 14 (4 WR) 0.7140 14.6 :I: 0.5 (5 Bi) Malu ski 1984 granite Debon et al. 1982 Debon et al. 1982 435 :I: 37 (8 WR) 0.7207 12.8 :I: 1.3 (Bi) type Jin &: Xu 1984 Debon et al. 1985 7.1 (WR-Bi) 0.7889 6.5 (WR-Bi) 0.8669 Debon et al. 1986 two-mi ca 7.1 :I: 1.2 (WR-Bi-Mu) 0.7432 13.3:1: 0.4 (2 Bi-2 Mu) 8.4 :1:5.4 ( " ) 0.7418 10.8 :I: 0.8 (3 WR + 3 Bi) adamellite Debon et a 1. 1986 Debon et al. 1985 type The two types are thus separated in time by some 450 Ma. Their juxtaposition is recent, masses of porphyritic granite being wrapped in and pulled upward by the adamellitic magma. It is suggested that the North Himalaya belt is produced by the same mechanism as the High Himalaya belt (Le Fort, 1986). A thicker zone of migmatisation, a higher temperature and differences in the quantity and quality of the fluids released from the Midlands give rise to a slightly different magma, richer in biotite and anorthite contents, departing from the minimum mel t composition. The melt also is more abundant and instead of emplacing as a succession of small batches, it rises diapirically pushing upward portions of its cap, Formation III augen gneisses, as xenoliths. The North-Himalaya belt is interrupted to the west, approximately as a sharp bend of the frontal part of the Himalaya and a merging of MeT and MBT occur. This may reflect the former configuration of the two plates, the westward decreasing rate of convergence as the pole of rotation lies to the west, as well as a difference in the rheologic behaviour of Tibet after the collision. If Tibet creeps, the deformation by thrusting will be less in the Indian plate and the melting may remain discrete with no big gathering and diapiric movement. On the contrary, if Tibet resists deformation, thrusting and melting will be maximum in the Indian plate. 345 V.3.4. Conclusion The Himalayan orogeny has produced a very small quantity of magma but of a very special and typical type. The fact that it is so recent enables to work out its evolution with a detail hardly ever obtained elsewhere. The fact that it is produced on a continent almost secluded from mantle influence for such a long time increases the differences and optimizes the resolution capacity. In the future, when, after the Himalayan orogeny, the crust will be brought back to normal thickness, the small number of plutons will be replaced by large migmatitic zones. This will contrast with the area north of the suture where Tibet lower crust will probably expose tremendous stretches of plutonic rocks. VI. LATE HIMALAYAN EVOLUTION VI.l. THE LATE OROGENIC BASINS AND GRABENS As the main tectonic activity goes on in the Himalaya, the deposition of detri tal sediments starts on the boundaries of the orogenic zone. A large molassic basin extends progressively more and more to the south while smaller basins start to fill within the range itself (fig.20). They record indirectly the evolution of the neighbouring changing regions where markers are so few; for this reason, their study is of primary importance. V1~.1. The Sub-Himalayan basin By far the largest boundary basin, the Sub-Himalayan basin (the "perisutural basin" of Mascle et al. 1986) includes the Siwaliks hills (the hills of Siva, the Hindu god of destruction) and the Indo Gangetic plain. It is tectonically limited to the north by the MBT and transgresses to the south over the Indian shield (fig.20). It is composed of discontinuous series that encompass most of the Cenozoic era (table IX). However, there was generally a lack of sedimentation during late Eocene and almost the entire Oligocene. The upper Oligocene-lower Miocene transgression of the Murrees or Dharamshala group occurs over a hard ground and locally evaporitic formations, bu t generally without any unconformity. In the apex of the Hazara-Kashmir syntaxis, north of Muzafifarabad, stratigraphic transition with the Murree formation has even been discovered (Bossart & Ottiger, submitted). The major difference is in the environment of the sedimentation: from marine (the nummulite-bearing Subathu) to continental (the Murrees or Dharamshala group) (cf. Srivastava & Casshyap, 1983; Blondeau et al. 1986). Continental environment will remain during the deposition of the overlying Siwalik group and subsequent formations. These freshwater deposits have been grouped by Anderson (1927) in the "Nimadric system". The stratigraphy of these sediments is not easy to establish although the geological study started already during the first half of the XIXth century (Falconer, 1835, 1837). But recent work has greatly 346 o Indian shield Immmm g~~~:~~r~O ~~~ae;se D Quaternary ±400km '--~-'--'-~, -;if. Volcano M B T and similar thrust 300 N Fig.20: General map of the Cainozoic and Quaternary basins of the Himalaya and surrounding regions (after G.D. Johnson, 1976 pers. comm.; Gansser, 1981, 1983a; Burbank & Johnson, 1983; le Fort, 1986). B = Bombay, C = Calcutta, ChF = Chama fault, D = Delhi; DeN = Dacht-e Newar, E = Everest, I = Islamabad, K = Kathmandu, Ka = Kashmir, Kb = Kabul, Ki Karachi, Ko = Kohistan, Ks Kailas, l lhasa, NB = Namche Barwa, NP = Nanga Parbat, P = Peshawar, Pk = Pokhra, T = Thakkhola, US = upper Sutlej. improved our knowledge of the Sub-Himalaya and has had important consequences on our understanding of the late Himalayan evolution. VI.l.l.a. Lithology The main groups of formations have distinct lithological characters that make them easily distinguishable : the Subathus are composed of black shales and minor intercalations of quartzitic limestone; the Murrees- Dharamshala are fine grained shales with minor sandstones and are distinctly red; the Siwaliks are made up of alternating sandstones and mudstones, and they become coarser from bottom to top; the Lei conglomerate contains striated and faceted pebbles. However, to establish Table IX Cenozoic nomenclature of the Sub-Himalaya and neighbouring basins compiled from various sources. Columns for Arunachal and upper Assam valley are only indicative as lithostratigraphy is time transgressive and precise ages are not determined (cf. Das Gupta 1979, Karunakaran & Ranga Rao 1979). Indicative thicknesses in meters. OMa epoch Pleistocene Pliocene 5 5.5 u. 10· Ol c Ol " m. 15 0 ::E 20 f. Chattian 30 40 Peshawar Kashmir Pesh. O~4~ NW (Potwar) Central Him. Arunachal Upper ·~-------'--L-ei~C-g---hD~i~kl~ai~edt~a~1. Assam 1--___ t-=-::::_+-_''''30'''o'''' __ I ___ ? __ Dihing 13001 I. Karewa 113001 - '/ = Hir ur / ;> 151 Murree 7-/"-7 uplift of N. IFTI Subathu Subathu ? u.Siw. m.Siw. I.Siw. _~i~j~r!l__ u. Siwalik Kimin (~1200) ~~~~~~~~~~~. (2300) N~~~~~g Tatrot ---------------- ?------ ? Dhok Pathan (400) m. Siwalik --------- 11500-45001 Nagri (400) Chinji Kamllal I. Siwalik = Nahan 11100-16001 Kasauli til (600) Subansiri «0001 Dafla 138001 Tipam (:so:31001 ?----- Surma 12000) ~--?--­ E Subathu Laki '" ;;; "' Cl Barail 18001 Jaintia 50Ma~------~------~------~------~L-------L-------~~~~L-----~ 347 a detailed lithostratigraphy has proved to be extremely problematic, even in a restricted area, as lateral facies variations and large differences in rate of sedimentation are very common. 2 The Potwar basin (or plateau) lies for 20,OOOkm in northern Pakistan between the Salt range and the Kalachita hills. Murrees and Siwaliks are well exposed in this semi desertic, dissected region of relatively easy access and it has been taken as the type for lower and middle Siwaliks by Pilgrim (1913) on lithological and paleontological grounds (table IX). But detailed mapping in the same region (e.g. Gill, 348 1951a) has shown that the succession is affected by facies variations of regional extension and that all formations boundaries, even the Murree- Siwaliks one, are difficult to agree on and to map (cf. Sahni & Mathur, 1964) • The composition of these formations in major elements and heavy minerals has also been studied in de tai 1 with the hope to find out the origin of the material and to make correlations. For example, a typical sandstone from the lower Siwaliks of Arunachal (Singh et al. 1982) contains: quartz (60%), feldspar (6%), micas (4%), rocks debris such as schist and quartzite (1010), argillaceous material (1510), ferrugeneous and siliceous cement (3%), heavy minerals (2%). Rock debris may also contain gneisses and granites; in Kumaon as in Nepal and Punjab, they mainly appear in the middle and upper Siwaliks (Gansser, 1964; Herail & Mascle, 1980; Abid et a!. 1983). In the latter, Herail and I (unpublished study) have determined that the granitic pebbles do not derive from the High Himalayan granites as previously thought but simply from the "Lesser Himalayan granites" now distant by not more than 10 km. Apparently, all the detri tal material, incorporated in Siwal ik as well as in pre-Siwalik formations (Srivastava & Casshyap, 1983), originated from the Himalaya. The heavy mineral content has been very extensively determined, specially by sedimentologists from the petroleum industry. Raju (1967) for India and Chaudhri & Gill (1981) for Nepal have found that the lower Siwaliks contain mainly epidote and staurolite, the middle Siwaliks kyanite (and staurolite), and the upper Siwaliks hornblende and sillimanite, sometimes in very small quantity only. They represent an increasing grade of metamorphism suggesting the erosion of a normal sequence of metamorphism, or, taking into account the existence of an inverted sequence of metamorphism below the normal one, suggesting a progressive erosion towards northern more metamorphic terrains. In Potwar basin, an abrupt increase of blue-green hornblende proportion of the heavy mineral fraction occurs around 11 Ma and corresponds to a rapid increase in the rate of sediment accumulation (N.M. Johnson et al. 1985). It implies that the source area of the sediments attained a Kohistan arc type of formation possibly under tectonic influence. one may remember that this is the time when backthrusting occurred. VI.l.l.b. Biostratigraphy We have just seen how imprecise and misleading is the lithostratigraphy in the Sub-Himalayan basins. However, these series and specially the Siwaliks have been known for a long time for their very rich fauna, vertebrates in particular. Already from 1906 to 1936, Pilgrim aimed at establishing a comprehensive stratigraphy of the fluvial and continental Siwaliks. Since 1973, systematic paleontologic search has yielded more than 50,000 fossil specimens from the south of the Potwar basin. Vertebrates are most important and include very rich mammalian faunas (~Pilbeam et al. 1977a; Barry et al. 1982, 1985; Tassy, 1983; Thomas, 1983; Beden & Brunet, 1986) and hominoid primate remains (Pilbeam et al. 1977b) including some of the oldest known hominoid forms such as t~ Ramapithecus punjabicus in the Chinji formation, around 11.5 Ma (Pilbeam et al. 1977b). Plant remains and pollens are also abundant but a systematic study of them has not yet been undertaken. 349 A local stratigraphy has been established in Potwar based on this very large collection of fossils and using ages given by equivalents of the European fauna (~ Barry et al. 1982). The problem of possible reworking of fauna in this fluvial environment has been eliminated by the abundance of the systematic collection achieved. The interval zones obtained are more or less equivalent to Pilgrim's divisions (cf. table I) but this stratigraphy is limited to one group in one area of the Sub- Himalaya. Other regions, Nepal for example (West,1984, Beden & Brunet, 1986), as other groups, Subathu for example (Sahni et al. 1981) are less rich in fossils and would not allow the establishment of such a fine stratigraphy along the entire basin. The mammalian faunas, however, are most important for the study of faunal change in southern Asia and its correlation with tectonic, climatic and oceanographic events (Barry et al. 1985). The Siwaliks of the Potwar basin record four major faunal changes (before 20 Ma, between 20 and 16 Ma, at 9.5 Ma and 7.4 Ma) dominantly linked to immigration from eastern Asia, from Europe or Africa. The changes themselves are probably controlled by the rise and disappearance of the physiographic, climatic and ecological barriers (ibid.). VI.l.l.c. Magnetic polarity stratigraphy Provided that the depositon is slow enough, the magnetic particles that are sedimented follow the magnetic field of the earth at that moment. However, secondary magnetization may be introduced during diagenesis. The problem is particularly crucial in the case of red beds such as those of the middle Siwaliks, and it is important to identify the different components of the natural remanent magnetization (NRM). For the Siwaliks red beds of the Potwar plateau, Tauxe et al. (1980) have presented evidence for two phases of hematite, a primary specular phase with a higher blocking temperature and a secondary red pigment phase. They have also shown that the remanence of the primary specular hematite was acquired during or very shortly after deposition whereas the red pigment was produced much later. The measure of the NRM is thus carried out after removal of the red pigment contribution by thermal demagnetization (Opdyke et al. 1982). Following this method, rock magnetic properties of the Siwaliks have been extensively and systematically analysed in the same Potwar region (Opdyke et al. 1982; Tauxe & Opdyke,1982; N.M. Johnson et al. 1985). The local magnetic polarity stratigraphy has been established and compared with the magnetic polarity time-scale (fig.21). In this big task, the team has been helped by the occurrence of three groups of conformable horizons of bentonites and bentonitic mud- stones, rich in euhedral zircons, deriving from volcanic-ash beds (G.D. Johnson et al. 1982). A number of those have been dated by the fission- track method and have provided further chronometric constraints on the magnetic polarity stratigraphy already established (ibid. and fig.21). The origin of these volcanic tuffs is not precisely determined. One could think of the ignimbritic equivalent of the High Himalaya 350 1900 m 18 16 14 12 10 0.8 0.6 04 JALALPUR POLARITY TIMI:-SCALE cnron LITHO srRATI INTERVAL - ZONE 1.5 Ma E{opt/as plamfrons 2.9 HexaprotOdon siva/ensis 5.3 Selenoportax lydcJekeri 7.' 'Hlppanoo s./: 9.5 Fig.21 : Correlation of the magnetic polarity time scale (after La Brecque et al. 1977 and Mankinen & Oarlrymple, 1979) with seven local magnetic polarity sections (from N.M. Johnson et al. 1982 & 1985), the lithostratigraphic units (Wadia, 1975; Gill, 1951a) and the biostratigraphic interval zones of Barry et a1. (1982) in the Potwar region. The European marine biochronologic scale is given after Berggren & Couvering (1974), Barron et al. (1985) and Sen (1986). leucogranites that would have blown out several times. But the bentonitic levels seem to be restricted around the NW syntaxis and to be absent closer to the main granitic bodies. A more likely explanation is that they come from a nearby active volcano during Miocene to Pleistocene. The closest known one is the Dasht-e Newar, SW of Kabul in Afghanistan (fig.20) that has produced enormous quantities of andesitic to rhyodacitic lavas for which the few K-Ar dates available range from 2.9 to 1.4 Ma (Bordet, 1975) and can be equated to the youngest volcanic tuffs horizons comprised in the upper Siwaliks (G.D. Johnson et al. 1982). The present distance from Potwar is of 500 to 600 km, equivalent to the distance of Idaho from Mount St Helens. 351 For the first time in a continental domain, lithostratigraphy, biostratigraphy, magnetic polarity stratigraphy and absolute ages have been put together successfully for a period of time on the order of 20 Ma (fig.21). This detailed stratigraphy enables us to estimate precisely the rates of sedimentation and their evolution. They vary between 0.1 and 0.6 mm/a (N.M. Johnson et ale 1982) with an average around 0.3 mm/a, a maximum during middle Siwaliks and a general slow down or even cessation around 2 Ma (Johnson et ale 1979). Interesting enough, these rates are in the same range as the rate of uplift obtained by fission track method for the source area to the north of the Potwar basin (Zeitler, 1982). The precision of the timing also makes possible to evaluate the time transgressive deposition of one type of sediment and to relate it to the structural evolution of the surrounding terranes. VI.l.l.d. Structure of the Sub-Himalaya The Sub-Himalayan rocks have been rather gently deformed compared with the Lesser and Higher Himalaya. The deformation varies in intensity from almost undeformed terrasses and recent flood sediments to tightly folded and faulted terranes. This evolution has developed both in space and time as the most deformed rocks are those occurring more to the north that are also older. In the Potwar basin of the Punjab Himalaya, Gill (l951b) has in this manner distinguished three major zones of deformation, stretching for some 120 km from south to nor th : - a zone of open folding, affected to a varying degree by fold- faulting, - a zone of open folds and monoclines displaying reversed faults dipping steeply to the north and extending laterally for several ten of km, - a zone of more compressed folds and closely spaced reverse faults. In central Nepal, the anal ysis of the thrust pattern shows a progressive southward shift during the Neogene, from the MBT to MDT (Main Dun Thrusts) to MST (Main Siwalik Thrusts) (Mascle & Herail, 1982; Delcaillau, 1986; Herail et al. 1986; Mascle et al. 1986). Whilst MBT was active during middle Siwalik time, the motion-;;;-transferred to MDT during the late Siwalik time and MST during Pleistocene time, resulting in a piling of thrust slices. This progression has been well recorded and analyzed in the sediments of the NW area of the sub-Himalaya where the rates of sedimentation increase away from the thrusting zone, passing through a maximum that progresses with time towards the south (Burbank & Reynolds, 1984). This progression is discontinuous as pulses of rapid deformation lasting less than 0.5 Ma occur along the thrust. From 3.0 to 1.8 Ma, the locus of thrusting shifted some 50 km to the south, that is some 40 mm/a (ibid.); during roughly the same time, from 3.0 to_0.8 Ma, polymictic conglomerate appears suddenly above the topmost conformable strata and progrades to the south by some 50 km in the so called Jhelum re-entrant, at an average speed of some 24 mm/a (Raynolds et al. 1980; Raynolds & Johnson, 1985). Thus, the sedimentation in one locality forms a cycle during which deposition is followed by uplift and erosion, turning it 352 into a source of the new sediments carried further south (N.M. Johnson ~ al. 1982). The sedimentological analysis also enables us to comprehend to some extent the evolution of the drainage system. For example the Jhelum river, one of the left banks main tributaries of the Indus can be shown to have followed approximately the same course for the last 2.5 Ma (Raynolds et al. 1980). On the contrary, during the early and medial Siwalik time the river system probably flowed from west to east along the Himalayan front (N.M. Johnson et al. 1985) with the possible consequence that the ancestral Indus may have flown into the Ganges and the Bay of Bengal (Raynolds, 1981), a modern version of the "Siwalik river" suggested by Pilgrim (1919). The present drainage system would only date back to the late Siwalik time. The episodes of intense deformation may be very brief as shown in the northern limb of the Soan syncline just south of Rawalpindi (Burbank & Reynolds, 1984). There, the nearly vertical youngest upper Siwalik sediments dated about 2.1 Ma are overlain by sub horizontal Lei conglomerate that contains a volcanic ash level (dated 1.6 ± 0.2 Ma) and that extends down to 1.9 Ma. Thus in approximately 0.2 Ma, more than 3000 m of Siwal ik molasse has been steeply folded and eroded (mean minimum rate of 15 mm/a). The deformation by thrusting is accompanied by (Delcaillau, 1986; Herail et al. 1986 in Nepal): - fr~e cleavage in the vicinity and parallel to the MBT; - a system of conjugated fractures steeply dipping to the NNE. Principal stresses obtained using the graphic method of right dihedrons (Angelier & Mechler, 1977) give most of the time a horizontal NNE-SSW compression; - deformation and fracture of quartz grains and pebbles. The intensity of this deformation is heterogeneous with an overall decrease towards the south. In addition to the NS compression and related deformation, two other movements have been evidenced by the study of the Sub Himalaya: strike slip and rotation. Strike slip movements although secondary compared with the importance of thrusting, has been observed at all scales in the Sub Himalaya region. In particular, the rhombic form of the "duns" (Siwalik inner depressions) correspond to the left lateral regional strike slip movement along the Himalayan foredeep (~ Herail et al. 1986). In NW Sub Himalaya, the paleomagnetic study of the Siwaliks around the Salt range (Opdyke et al. 1982) has corroborated the counter clockwise rotation of 30~0° of the entire region as previously suggested by Crawford (1974). Altogether, the Sub Himalayan basin is structured by the southward progression of a series of thrusts that respond to the continuous slightly oblique convergence. The material which is not underthrust is reworked and adds to the erosion products of the Himalayan range. This structure and evolution are very similar to those of an accretionary prism and may be compared with such older periorogenic basins as the French-Belgian Carboniferous coal basin (Stille, 1919; Herail et al. 1986) • 353 VI. 1. I.e. The Indo-Gangetic plain Extending for some 150 to 350 kin in width to the south of the Siwaliks, the Indo-Gangetic plain is the portion of the Sub-Himalayan basin covered by recent deposits. Its thickness and composition is known from seismic surveys and few drill holes specially made for peroleum and natural gas exploration. The thickness exceeds 7 kin in the deepest areas and the fill is made up of the same Cainozoic deposits already described to the north. The sedimentation shows a number of general variations from north to south including : - reduction in global thickness, - facies changes towards finer grained sediment, - increasingly young transgressive base over the metamorphic basement (cf. Lyon-Caen & Molnar, 1983). The total volume of sedimentt co~tained in the Indo-Gangetic plain can be estimated to some 2.5 x 10 km • At both extremities, the Indo-Gangetic plain continues in the ocean by the very large Indus and GaggesiBrahmapoutrg fa2s. They extend respectively for some 1.5 x 10 km and 3 x 10 km with thickness of up to 15 kIn of almost horizontal deposits dating back to the Oligocene (Mallik, 1978; Curray et al. 1982). VI.l.2. The Indus-Tsangpo episutural basin A narrow residual basin follows discontinuously the Indus-Tsangpo suture for nearly 2000 kin between the Nanga Parbat spur to the west and the Namche Barwa region to the east. The width of the basin seldom exceeds 30 kin presently, but the sediments deposited have been folded and eroded. Access to it has been reopened since 1975 in Ladakh and 1980 in southern Xizang, but the exact extent of these late orogenic sediments is still uncertain. The sediments have been deposited entirely in continental environment. In Ladakh (~ Frank et al. 1977; Baud et al. 1982; Brookfield & Andrews-Speed, 1984; Van Haver, 1984; Mascle et al. 1986), up to 5 kin of molasse made up of alternating red pelites and sandstones to conglomerates transgresses over the marine nummulitic series to the south (up to Cuisian) and over the Transhimalaya batholith to the north. The detritic material has been provided by the Transhimalaya magmatic products as well as by the sedimentary nappes. Mud cracks, birds tracks, rain drops and palaeosoils testify of the frequent emersion of these fluvio-lacustrine basins. Fresh water gastropods and bivalves, vertebrate and plant remains poorly constrain the age of the molasse that probably extends from upper Eocene to Miocene. Given this imprecision, Mascle et al. (1986) evaluate the rapid rate of sedimentation around 0.1 mm/a. --- -- In southern Xizang (Academia sinica, 1980; Burg, 1983), the Liuqu molasse forms a very narrow strip made up of a polygenic red conglomerate containing pebbles of the ophiolitic sequence in the vicinity of which it has been deposited. Its age is considered by Academia sinica (1980) to be Oligo-Miocene. As in Ladakh, it is transgressive over the south vergent structures but it is involved in the backthrusting movements (cf. Table III). 354 As underlined by Mascle et al. (1986) it is difficult to envisage the deposition of these episu~ sediments in a purely compressive regime. Admitting that the discontinuity of the basins is original, these authors suggest that they represent small pull apart basins resulting from right lateral strike-slip along the suture line, a palaeo equivalent of the Hundes or upper Sutlej basin (see below). On the other hand, the link of the known basins with the dense bodies of ultramafic material and their narrow width suggests that they may result from simple isostatic equilibrium. VI.l.3. The Kashmir and Peshawar intermontane basins The same team that has so successfully investigated the Potwar basin has also paid attention to the two basins lying just north on both sides of the syntaxis: the Kashmir and Peshawar basins (fig.20). VI.l.3.a. The Kashmir basin The intermontane basin of Kashmir (130 x 15 km) is filled with more than 1300 m of sediments (base unknown) classically divided into two formations: the lower (Hirpur) and upper (Nagum) Karewa separated by an unconformity (table IX). The importance of lacustrine mudstones in the otherwise fluviatile (deltaic) clastic sediments makes a strong difference from the Siwaliks. Abundant carbon-rich sediments represent swamps and shallow lacustrine deposits. Numerous conglomeratic levels with pebbles derived from the nearby High Himalaya and Pir Panjal ranges punctuate the deposits. Finally, volcanic ash levels similar to those in Potwar have been also encountered. Until recently, the time transgressive nature of the Karewa sediments and the limited paleontological data did not allow reliable correlations. As for the Sub-Himalayan Potwar area, the determination of magnetic-polarity stratigraphy and the fission track dating of the zircons contained in the volcanic tuff levels have enabled us to establish the chronologic and stratigraphic development of the basin (Burbank & Johnson, 1983). These methods go back to 3 Ma. Another 1 Ma at least has to be added; it is the time necessary for the deposition of more than 500 m of the first lacustrine mudstone, fluvial conglomerate and coarse sandstone (ibid.). The more than 4 Ma evolution of the Kashmir basin thus extends over most of the Pliocene and Pleistocene epoch. At the time of the inception, between 5 and 4 Ma, the Siwaliks directly to the SW of Kashmir experienced important changes in lithology (introduction of clasts from the Pir Panjal range) and paleocurrent directions (from eastward to southward) as well as the development of the Jhelum re-entrant. Thus, the creation of the Kashmir basin is thought to be linked to the emergence of the Pir Panjal relief, a direct consequence of the southward thrusting along the MBT (ibid. Burbank, 1983a). Rates of s~diment accumulation range from 0.64 to 0.16 mm/a with an average of 0.32 mm/a which is quite fast for such a largely lacustrine basin. The detailed analysis3 0f these rates indicate periodicities in the order of 100, 40 and 20 x 10 a that may be related to climatic cycles 355 (Burbank & Grant, 1985). Conglomeratic levels occur every 0.4 to 0.5 Ma and probably result from episodic movements along the thrust. The importance of the glacial influence becomes conspicuous during the upper Karewa where reworked moraines and boulders of glacial origin have been observed (e.g. Wadia, 1975); thus the climate evolved from warm and humid during the lower Karewa to sub-temperate during the upper (Fort & Gupta, 1981). As in the Siwaliks of the Potwar basin, volcanic ashes have been dated between 2.5 and 2.1 Ma and are ascribed to the same origin (Burbank & Johnson, 1983). Even the recent Karewa deposits have been affected by tectonic movements, including tilting up to 40° (Wadia, 1975). Important differential movement since the deposition of the lower Karewa has been known for long, but the recent absolute ages determination of detailed sections has enabled us to envisage rates of differential uplift between 4 and 8 mm/a along the southwest margin of the basin (Burbank & Johnson, 1983). VI.1.3.b. The Peshawar basin Symmetrical to the Kashmir basin with regard to the NW syntaxis, and with a similar size (fig.20), the Peshawar basin lies at a much lower altitude (around 300 m) and exposed sections exceeding 50 m in thickness are only found in the southern part of it. The basin, drained by the Kabul, Swat and Indus rivers, is filled by more than 300 m of flood plain mudstone and siltstone with minor sandstone intercalations overlying folded and eroded Eocene limestone (Subathu) or Murree sandstone (Burbank & Tahirkheli, 1985) Fanglomerates prograde northward in the basin. Volcanic ashes are encountered at two levels (ibid.). Similar methods as for the Siwaliks and the Kashmir basins have enabled to restore the stratigraphic and structural evolution of the Peshawar basin (Burbank, 1983a & b; Burbank & Tahirkheli, 1985). The inception of the basin occurred at least 2.8 Ma ago and was linked to the uplifting of a small barrier to the south (the Attock range presently culminating around 1400 m). The uplift of this relief on the back of a thrust induced the northward progradation of the alluvial fans, an accelerated uplift being responsible for the folding and tilting of the older sediments and for the multiple episodes of catastrophic flooding during the last 0.7 Ma. Rates of sediment accumulation are less than for Kashmir (about half at 0.15 mm/a). Volcanic levels have been directly dated at 2.6 and 2.4 Ma and indirectly at 1.8 - 1.6 Ma. A few alkaline rock complexes crop out at the western end of the Peshawar basin where they intrude metamorphosed Palaeozoic sediments along fault zones (Kempe, 1973 & 1983; Kempe & Jan, 1980). These authors suggest that they originated from a Cenozoic rifting event although absolute dating is needed to clarify the matter. To the south of the Peshawar basin and in continuity with it lies the smaller Campbellpore basin filled mainly with lacustrine deposits. Some 90 meters of sediments have been analyzed. They contain one bentonized tuff with a zircon fission track age of 1.6 ± 0.2 Ma and they extend for about 1 Ma (G.D. Johnson et al. 1982). 356 VI.1.3.c. Comparison of Kashmir and Peshawar basins A number of differences have been documented in the comparison of the two basins; among them, let us recall: - a difference in altitude and thickness of fill, - a difference in the time of inception (older for Kashmir) and in the age of the underlying formations (apparently no Murrees in Kashmir) , - post deposition deformation has been more important in the case of Kashmir, - abundant shallow earthquake epicenters in the vicinity of the Peshawar basin (Seeber & Jacob, 1977; Seeber et al. 1981). Explanation of these differences include: - the presence of a salt layer to the west of the syntaxis that facilitates 'decollement and broad scale thrusting (Seeber et al. 1981), - the more hypothetical presence of a segment of buoyant Indian crust that has been rafted below the eastern position of the syntaxis and resulted in the differential uplift of the Kashmir side (Burbank, 1983a). Wha tever the ro Ie of the buoyan t segment, it remains that nor th of the Kashmir basin the Nanga Parbat promontory (the westernmost 8000 m peak of Himalaya) interferes with its transversal pattern of high grade metamorphics where the upl ift rates are the highest in the region (Zeitler et al. 1982). VI.1.4. Other basins A number of other basins give clues to the recent history of the Himalaya. They include (fig.20) the Hundes (upper Sutlej), Thakkhola, Pokhara and Kathmandu basins as well as innumerable valley filling and terrasses. ThZ Hundes or upper Sutlej basin is a huge flat area (200 x 50 km, 8000 km ) with a somewhat rhomboidal shape, lying west of the Kailas at an average elevation of some 4.500 m. It is limited to the south by the high Himalayan range. Little geologic information is known from this historical and archaeological area (cf. Gansser, 1964). Thick sequences of Pleistocene (7) to Sub Recent (7)~ndstones and conglomerates have been tilted and warped recently (Heim & Gansser, 1939). The thickness (above 1000 m) and exact age of the deposits is not known. The parallel NNE-SSW drainage pattern is thought to derive from the bounding strike- slip faults and the entire basin to represent a pull-apart basin active for the last few million years (Ni & Barazangi, 1985). The Thakkhola-Mustang graben (50 x 20 km), lying between 3000 and 4000 m above sea level, is also located on the northern side of the high range within the Tibetan sedimentary series up to lower Cretaceous. To the WNW as to the ESE it is limited by N20° to N40 0 E faults that have been active during the entire evolution of the basin (Fort et al. 1982). At least 850 m thick detrital deposits derive from the surrounding formations including the Mugu-Mustang high Himalaya leucogranite whose pebbles are already found close to the base (ibid., Colchen et al. 1986; fig.22). The fuviatile and lacustrine deposit~ not well dated. Based on correlations with uplift, climatic changes, glaciations, etc. in 357 w Fig.22: East-west section across the southern part of the Thakkhola-Mustang graben (reproduced from Colchen et ale 1986). 1 Paleozoic-Mesozoic substratum, 2 : Tetang formation, 3 : Thakkhola formation, 4 : alluvial deposits. Tibet, Fort et al. (1982) suggest that most of the sedimentation took place during late Pliocene and early Pleistocene. However a study of magnetic polarity (Yoshida et al. 1984) has suggested that the base of the Tetang formation (fig.22) may date back to lower Pliocene or older. In fact, pollens may indicate a PI iocene age for the upper part of the Tetang formation (Colchen et al. 1981). The Thakkhola graben would then be more or less contempora~ with the Kashmir basin. To the north, the graben can be traced into Xizang on satellite imagery. On the contrary, to the south, from field mapping it appears that the fault system does not reach the MCT zone but disappears in the Tibetan slab (cf. Colchen et al. 1986). This contrasting behaviour is important as it shows that ~Tibetan extensional tectonic regime has a southern limit and is not linked to the old "Aravalli trend" (~ Valdiya, 1976) of the Indian shield. Using the sedimentological results of their study, Fort et al. (1982) as Yoshida et al. (1984) suggest that most of the sedi~tion period was dominated by an alternation of semi-arid, then warm and humid climate, glacial environment occurring later only. In consequence, they favour a late uplift of the High range, posterior to the Thakkhola fillings. The Pokhara basin, lying some 900 m above sea level, is limited to the south by the Mahabharat ridge. The more than 200 m thick fillings are dominated by calcareous material from the high range (Yamanaka et al. 1982), probably emplaced by mass transport in a periglacial environment (Fort & Freytet, 1979). The rathes complex evolution of this small basin spans a very short time (a few 10 a: Fort, personal communication, 1985) and seems to be connected to movements of the high range (upl ift and related quakes) and climatic changes, from subtropical to humid and cool with valley glaciation (Fort & Gupta, 1979). 358 The circular Kathmandu basin (30 x 25 km), lying around 1400 m above sea level, is also limited to the south by the Mahabharat range. It is filled with fluvio-lacustrine deposits reaching 500 m in thickness. Vertebrate remains found in the lower part of the sequence (Gupta, 1975) are comparable to lower Pleistocene ones from the Pinjore formation in the Siwaliks (table IX). All the sediments have been deposited during the last normal polarity sequence (Yoshida & Igarashi, 19841. Radiocarbon dating of the last formations only date back to 30 x 10 a (table in ibid.). The sediments would span upper Pliocene to recent epoch (Fort & Gupta, 1979; Yoshida & Igarashi, 1984) and the climate would have been much warmer at the beginning. All these basins have a rather short history compared to the Sub Himalaya. The biggest ones seem to follow a paroxism in tectonic movements around Mio-Pliocene boundary with differential uplift of the high range v. s. the Mahabharat border range. They all seem to be undergoing dissection and erosion in the present stage. VI.2. SURRECTION AND COOLING Recent vertical differential movement in the Himalaya has been observed in various areas and suspected in many others. Documented examples include : - the folded Jalipur molasse (Pleistocene 7) and uplifted terraces of the Indus valley NW of the Nanga Parbat (Gansser, 1983b), - the differential uplift of up to 80 m of the portion of a terrace deposited on Siwalik sandstone compared with the same terrace deposited on Lesser Himalayan formations on the other side of the Balia river valley in Kumaon (Valdiya et al. 1984), - the doming of the Quaternary gravel terraces of Taklakot around the Gurla Mandata (Gansser, 1964), - the uplift on the flank of the Pir Panjal range in Kashmir where the lower Karewa lacustrine sediments have been uplifted of 1400 to 2800 m during the last 0.4 to 0.9 Ma (Burbank & Johnson, 1983). Examples of such movements are numerous but measured cases are still very few and it iS3~nlY40ecently, with new dating techniques such as fission track and Ar- Ar, that real progress has been made. VI.2.1. Direct measurements A few geodatic measurements have been attempted for a limited number of years on the MBT or satellite thrust and faults. Results are scanty and ambiguous. Chugh (1974) and Sinvhal et al. (1973) report values from 0.6 to 6 mm/a with respect to the Ganga basin (fig.23). But, as stated by Molnar (1984), if the main movement is horizontal, the uplift will locally depend on the dip of the thrust-fault, being proportional to the tangent of the dip, that is negligible at shallow angles and very high at moderate values. What is required is repeated levelling on large sections of the Himalaya. The old railways of Simla and Darjeeling or the new roads from 0. 2 (FT , 1 6 M al ~~~ ~~T . 8. 7 M al li~ ~~r e t a l. 82 a Ze itl er 8 5 0. 68 (F T, 5. 4 M al N .H az ar a Ze itl er 8 5 ' 1. 6- 7 (S tra ti, 0. 9 M al Pl r Pa nJ al m ill Tr an sh lm a)a ya bat ~!I~ ~ban k & J oh ns on 8 3 ~ Op hl ob te s ~ Ti be ta n se di m en ta ry s e rie s lli J N or lh H im al ay a pl ut on ic be lt o Hi gh H im al ay a le uc og ra nl le s D T ib et an (H igh H im al ay a) cr lst al bn e sla b ~ ~ Le ss er H im al ay a M id la nd s a n d Kr ol b el t o Pl ei st oc en e ba Si ns I11 III H igh p re ss ur e (bl ue sc his t) m e ta m or ph ism M et am or ph ic Is og ra de ~ M ain M an tle T hr u~ t ~ M ain C en tra l T hr us t ~ M ain B ou nd ar y Th ru st ±. Si nv ha l e t a l. 73 t 10 0 20 0 30 0 40 0 km '+' F ig .2 3: Ra te s o f u pl if t m ea su re d o r c a lc ul at ed in t he H im ala ya b y v a rio us m et ho ds . Th e fi gu re g iv es t he r a te in m m /a; i n pa re nt he si s fo llo w s th e m et ho d (FT fo r fi ss io n tr ac k, K fo r K- Ar , Rb f or R b- Sr , n o th in g fo r di re ct m ea su re m en t), th e m in er al w he n n ec es sa ry (B i = b io ti te , Mu = m u sc o v ite ) an d th e nu m be r o f ye ar s ov er w hic h it h as b ee n av er ag ed . Th e fo llo w in g lin e gi ve s th e na m e of t he l oc al it y or re gi on w he re it h as be en m ea su re d, fo llo w ed b y th e re fe re nc e w ith t he tw o la st d ig its o f th e ye ar . v. > V I ' D 360 the Gangetic plain to Lhasa or along the Indus valley would form excellent sections. VI.2.2. Calculated rates of uplift Absolute dating of different metamorphic or igneous minerals by various methods enable us to estimate cooling rates using specific blocking temperatures of the mineral. These blocking temperatures are not well determined; they even vary with the rate of cooling. However, using the same mineral picked up at different altitudes on a vertical or steep section enables to estimate the time difference in passing through an isotherm. These closure rates enable to evaluate the cooling rate of the rock which in turn, having enough knowledge of the geological evolution of the area and making careful choices (see Zeitler, 1985 for discussion), enable to get average estimates of the erosion and uplift rates. Figure 23 gives the available data for the Himalaya. The longterm uplift rates mostly range from 0.2 to 0.8 mm/a. However they may not be representative of the entire Himalaya as they are concentrated in the NW corner or obtained from the high Himalaya. Data for the lesser Himalaya are missing. Apart from a few cases (Zeitler, 1985) it does not appear that there is a systematic increase or decrease of the uplift rate during Neogene time. Variations appear regionally with the striking example of the Nanga Parbat - Haramosh region that was uplifted and eroded during the Pleistocene by several kilometers (ibid.). Other variations correspond with tectonic discontinuity as sho~either side of the Indus-Tsangpo suture zone (the MMT) in northern Pakistan, by Zeitler et al. (1980), where differential movement apparently ceased some 15 Ma ago. Altogether, these rates of uplift and erosion, if6mai~tained for the last 15 to 20 Ma, would have produced some 3 to 9 x 10 km of detritic material, a figure not disagreeing with the estimated amount of sediments from the Sub-Himalaya and the Indus, Ganges and Tsangpo oceanic fans (from Curray & Moore, 1971) figures for the Bengal deep sea fan, a rate of denudation of 0.7 mm/a could be deduced for the corresponding portion of Himalaya. Finally, let us recall the presence of numerous hot springs mainly scattered along the main thrusts (MCT and MBT) and faults (such as around the Nanga Parbat), at their intersection with deeper valleys (Le Fort & Jest, 1974). They are less abundant than in Tibet but may total up to several hundreds, substantially contributing to the cooling of the Himalayan system. VI.2.3. Himalaya as a mountain range Since several decades there is a debate about when the Himalayan relief came into being, either recently i.e. a few millions years or quite early (Miocene) in the history of the belt. Arguments for a recent origin, the "morphogenetic phase of mountain building" (Gansser, 1983b) include the fast rate of uplift (around 2 mm/a, i.e. 4 km in 2 Ma) that the area north of the high range would 361 have experienced since early Pleistocene, on palaeoecological grounds (~Hsli, 1973 in ibid.; Li et al. 1981; Xu, 1981; Fort et al. 1982). However, the mechanisms of such a rapid and sudden surge are only conjectural. On the contrary, arguments for an early mountain morphology include the importance of clastic sediments derived from the Himalaya in the different molassic basins since early Miocene at least, the presence of metamorphic minerals derived from deep parts of the range since middle Miocene at least, the regular and fast average uplift rate obtained by radiometric methods whatever the period concerned since Miocene, the impossibility to depart strongly from isostasy. This does not mean that the rising of the mountain range is a smooth and regular process. But in our mind, the Himalayan barrier has existed for a long time, providing the erosive detritus, preventing the regular immigration of continental faunas (vertebrates in particular), fostering the monsoon climate and counteracting the tectonic and metamorphic on going processes by a rapid unloading and cooling. VI.2.4. Glaciation Himalaya being the highest range of the earth and fringed by the monsoon system, one would expect a highly glaciated mountain. It is not the case and adjacent areas such as Karakorum where the monsoon is not active, are much more covered and morphologically marked by glaciers than the Himalaya. Actually, most of the present glaciers lie above 4000 m high and very few morainic deposits occur below 3000 m in the Himalaya. The absence of large glacial lakes in the Himalaya is also remarkable. The Himalaya differs from the Alps in that there has been no large extension of the ice cover during the Quaternary. However, the timing of glacier fluctuations in both mountain ranges, as well as in Karakorum, seems to be very similar (Rothlisberger & Geyh, 1985). A glacial peneplenation between 4100 and 5000 m high has been described in Bhutan and ascribed to the "little ice age" (Gansser, 1983a). In the Himalaya, glacial morphology is restricted to the high range and immediate surroundings. The absence of extensive glaciation precludes the use of glacial deposits to date back the existence of a topographical high relief in the Himalaya. VI.3. PRESENT STATE: GEOPHYSICS That the Himalayan building is still going on is not only shown by the recent deformation but also by the recent geophysical observation, mainly seismicity and gravity. Both domains have been thoroughly investigated by P. Molnar and coworkers (Molnar, 1984; Molnar et al. 1977; Molnar & Chen, 1983, 1984; Warsi & Molnar, 1977; Lyon Caen & Molnar, 1983, 1985; Baranowski et al. 1984). For more detailed information, one should consult the~ferences. 362 VI.3.1. Seismicity The Himalayan region is seismically active as known by its inhabitants from the four major earthquakes that occurred during the last century and from the numerous less important tremors that shake the country every year. The precise location of earthquake centers is unfortunately not well constrained as the seismic network of stations within the mountains is very limited and recent, and the geological structure not sufficiently known. Following the construction of the huge Tabela dam on the Indus, a remarkable seismic survey has been done during a decade in NW Pakistan (Seeber et al. 1981), but similar studies in less complicated areas of the rang~ very much needed. For the few earthquakes large enough to have a fairly good location and fault-plane solution, it has been shown that, in the southern part of the Himalaya, they are located around the MBT plane and consistent with it (e.g. Molnar, 1984; Ni & Barazangi, 1984; fig.24). The slip vector plunges to the north with a variable angle, from around 5° in the eastern Himalaya to 20-25° in the western Himalaya (Baranowski et al. 1984), a fact that may be in part related to the differential rate of convergence of India against Eurasia along the arc. Seeber et al. (1981) suggest that in ~ Himalaya, the main active thrust fault, called the "detachment", corresponds to the top of the Indian craton sliding under the sedimentary wedge of the Sub- and Lesser- Himalaya. Farther to the north, under the High Himalaya, as under Tibet, the fault plane solutions show normal faulting (Molnar & Tapponnier, 1975) consistent with east-west extension structures such as the Thakkhola graben. The boundary between the thrust and the extension regime for earthquakes lies around the southern topographic front of the Himalaya where the mountains suddenly rise from the 2000-3000 m Lesser Himalaya to the 7000-8000 m peaks. The absence of thrust fault earthquakes may be due to the transition from brittle to ductile regime around 20 km depth (Seeber et al. 1981), a conclusion obtained also by the mechanical model of Meulebrouck (1983), whereas the normal faulting may be linked to the crustal thickening that brings vertical stress as the main compressive stress (Tapponnier & Molnar, 1976). The existence of vertical maximum and roughly east-west minimum principal stresses is corroborated by microtectonic measurements in the Quaternary of the Tibetan plateau (Tapponnier et al. 1981). To the south also, on the Indian shield, Molnar et al. (1973 & 1977) have emphasized the normal faulting shown by a large 1967 earthquake that would reflect the bending down of the crust before sliding beneath the Himalaya. VI.3.2. Gravity anomalies The concept of isostasy was born in the Himalaya more than a century ago. However, gravity measurements for the Himalaya and surrounding regions remain very preliminary (see Lyon-Caen & Molnar, 1985). All available transverse sections show a negative Bouguer anomaly increasing towards the north from around -40 mGals over the Indo-Gangetic basin, to around - 363 HIMALAYA Fig.24: Fault-plane solutions of some major earthquakes in the Himalaya and neighbouring region (reproduced from Molnar & Chen, 1983; Molnar, 1984). Lower hemisphere projection show quadrants with compressional first motions in black. 400 mGals over the High Himalaya (Verma, 1985; Lyon-Caen & Molnar, 1985). The general interpretation is a thickening of the crust towards the north (~ Qureshy, 1969). However, calculating the anomalies that would result from isostatic compensation of the topography shows that the Indo- Gangetic basin is overcompensated - deficit of mass, and that the Himalayan range is undercompensated - excess of mas s (Lyon-Caen & Mol nar, 1985). A model assuming that the Indian plate is flexed under the load of the Himalaya, using reasonable values for the free parameters such as the flexural rigidity of the plate and the density contrasts, takes care of these departures from isostasy over the Indo-Gangetic basin and the Lesser Himalaya (Lyon-Caen & Molnar, 1983). However to explain the profi les across the High Himalaya, Lyon-Caen and Molnar need to infer a weakening of the Indian plate and a bending moment applied to the end of the plate, the exact origin of which is largely conjectural. Some interesting implications and corroborations of this model are the time transgressive nature of the basal sedimentation in the Indo- Gangetic basin with a rate of 10 to 15 mmla for the last 15 to 20 Ma, the probable existence of molasse sediments below the Lesser Himalaya, the steepening of the Moho beneath the High Himalaya, the more rapid uplift of the High Himalaya and of the front of the Lesser Himalaya (Lyon-Caen & Molnar, 1983, 1985). To the north west, recent gravity mesurements made accross Kohistan probably imply a remaining thickness of the arc around 7 to 9 km only (Malinconico, 1986), in agreement with the rapid uplift obtained by Zeitler (1985). 364 VI.3.3. Rate of convergence in the Himalaya All the above facts imply that convergence has been a long lasting phenomenon and that it is still going on. How much convergence? How much of it is taken up in the Himalaya s.s. ? The global rate of convergence between India and Eurasia can be estimated from the pole of rotation (19.71°N, 38.46°E) and angular velocity (0.698 deg/Ma) of the two plates (Minster & Jordan, 1978). The result varies from 55 mm/a in western Himalaya to 75 mm/a in eastern Himalaya. The movement vector, however, is only perpendicular to the range in Kumaon; elsewhere there is a strike slip component that may explain part of the extension structures of the High Himalaya and that lessens the global convergence rate. The important deformation taking place in Central Asia proves that part of the convergence is accomodated north of the Himalaya (Molnar & Tapponnier, 1975). The respective amount of convergence taken up in the Himalaya and north of it in central Asia have been estimated separately and by independent methods (~ Molnar, 1984; Armijo, 1986; Armijo et ale 1986). In the Himalaya, direct observation is not yet available, but geological estimates of the amount of shortening that has occurred since the collision (~ Gansser, 1966; Le Fort, 1975a) give an average convergence rate comprised between 8 and 25 mmm/a. The southward progradation rate of the Sub-Himalaya basin also gives an estimated rate of 10 to 15 mm/a (Lyon-Caen & Molnar, 1983). Using the seismic moments of earthquakes during the last 80 years gives average values of 10 to 20 mm/a with large uncertainties (Chen & Molnar, 1977; Molnar et ale 1977). In Tibet, the rate of extension deduced from the study~he north- south grabens shows that the amount of convergence taken up by Tibet extension is again in the order of 10 to 20 mm/a (Armijo, 1986). Similar rough estimates of the convergence taken up by strike slip movements in central Asia also give a value around 10 to 15 mm/a (ibid.). Alltogether, the global convergence between India and Eurasia may be divided into three more or less equal parts: Himalayan thrusting, Tibet extension and central Asia extrusion. VII. CONCLUSION During the past two decades, the Himalaya has been the hub of the plate tectonic game. Most authors, but not all (cf. Haller, 1979), agree that the Himalaya is the last of a series of collision due to the progressive breaking up of the Gondwana supercontinent and the accretion of the different detached pieces onto Eurasia after drifting across the Tethys oceans (e.g. Sengor, 1979, 1981, 1985; Colchen, 1981). It is the last but also the major one, India being the largest Gondawana fragment. But apart from this broad framework, almost every possibility has been suggested, varying the number of pieces in presence, the number (0 to 3) and location (MBT to ITS) of oceanic sutures, the direction of dip of Benioff planes (northward or southward), the order of subductions-obudctions- collisions 3happenings, the amplitude of plate thrusting (from 0 to several 10 km), the importance of strike-slip movement, etc. Table General evolution of Himalayan range. 50 I Subathu 10 I Siwalik Ma a I Lei I Ka'"f" ~ ~ ___ -L~_~~~~_~_~~_~~L-____ J-________ ~c back thrust. - - - --- --- m .. o collision nappes ? _ .... >---_~M~C~T!....____ _ _ ___ _ " I:.Ii~p~i~ ~ eo Him. metam Karakorum Trans Himalaya __ ...!M!!..:::B'-T=--__ ~ Himal. metamorphism ___ K.:..:a::.ra!:o~u~ ____ _ N & High Himalay~Y_ ;: m ... .. ;: o " : 366 REfERENCES Abid I.A., Abbasi I.A., Khan M.A •• Shah M.T. 1983 - Petrography and geochemistry of the Siwalik sandstone and its relationship to the Himalayan orogeny. Geol. Bull. Univ. Peshawar, 16, 65-83. Academia Sinica 1980 - A scientific guidebook to south Xizang (Tibet). Symp. on Qinghai- Xizang (Tibet) Plateau, Beijing 1980, I, Beijing: Sci. Press, New York: Gordon. Breach, 104 p. Achache J., Courtillot V •• Zhou Y.X. 1984 - Paleogeographic and tectonic evolution of southern Tibet since middle Cretaceous time: new paleomagnetic data and synthesis. J. Geophys. Res., 89, red, 10311-10339. Acharyya S.K. 1978 - Stratigraphy and tectonic features of the eastern Himalaya. In Tectonic geology of the Himalaya, P.S. Saklani ed., Today. Tomorrow's print •• publ., New Delhi, 243-268. Acharyya S.K •• Ray K.K. 1977 - Geology of the Darjeeling-Sikkim Himalaya. Guide to Excursion n04, fourth Int. Gondwana Symp. India, Calcutta, 25 p. Allemann 1979 - Time of emplacement of the Zhob valley ophiolites and Bela ophiolites, Baluchistan (preliminary report). In Geodynamics of Pakistan, A. farah. K.A. De Jong eds, Geol. Surv. of Pakistan, Quetta, 215-242. Anderson R. van V. 1927 - Tertiary stratigraphy and orogeny of the northern Punjab. Bull. Geol. Soc. Am., 38, 665-720. Angelier J •• Mechler P. 1977 - Sur une methode graphique de recherche des contraintes principales egalement utilisables en tectonique et en sSismologie : la methode des diedres droits. Bull. Soc. geol. fr., (7) XIX, 6, 1309-1318. Ari ta K. 1983 - Origin of the inverted metamorphism of the Lower Himalayas, central Nepal. Tectonophys., 95, 43-60. Armijo R. 1986 - Mecanismes des deformations quaternaires au Tibet: cinematique de la collision Inde-Asie. These Doct. Univ. Paris VII, 177 p. Armijo R., Tapponnier P., Mercier J.L •• Han T.L. 1986 - Quaternary extension in southern Tibet: field observations and tectonic implications. J. Geophys. Res., 91, B 14, 13803-13872. Auden J.B. 1937 - Geological structure of the Garhwal Himalaya. Rec. geol. Surv. India, 71(4), 407-433. Azmi R.J. 1983 - Microfauna and age of the lower Tal phosphorite of Mussoorie syncline, Garhwal lesser Himalaya, India. Himalayan Geol., 11, 373-409. Azmi R.J •• Pancholi V.P. 1983 - Early Cambrian (Tommotian) conodonts and other shelly microfauna from the upper Krol of Mussoorie syncline, Garhwal lesser Himalaya with remarks on the Precambrian-Cambrian boundary. Himalayan Geol., 11, 360-372. Bally A.W., Allen C.R., Geyer R.B., Hamilton W.B., Hopson C.A., Molnar P.H., Oliver J.E., Opdyke N.D., Plafker G •• Wu f.T. 1980 - Notes on the geology of Tibet and adjacent areas; report of the American plate tectonics delegation to the People's Republic of China. U.S. Geol. Surv. open file rep. 80-501, Washington D.C., 100 p. Baranowski J., Armbruster J., Seeber L •• Molnar P. 19B4 - focal depths and fault plane solutions of earthquakes and active tectonics of the Himalaya. J. Geophys. Res., 89, B8, 6918-6928. Bard J.P. 1983 - Metamorphism of an obducted island arc: example of the Kohistan sequence (Pakistan) in the Himalayan collided range. Earth Planet. Sci. Lett., 65, 133-144. Barron J.A., Keller G •• Dunn D.A. 1985 - A multiple microfossil biochronology for the Miocene. In The Miocene ocean, Geol. Soc. Am., 163, 21-36. Barry J.C., Lindsay E.H. & Jacobs L.L. 1982 - A biostratigraphic zonation of the middle and upper Siwaliks of the Potwar plateau of northern Pakistan. Palaeogeogr. Palaeoclimatol. Palaeoecol., 37, 95-130. 367 Barry J.C., Johnson N.M., Raza S.M. & Jacobs L.L. 1985 - Neogene mammalian faunal change in southern Asia: correlations with climatic, tectonic and eustatic events. Geology, 13, 637-640. Bashyal R.P. 1980 - Gondwana type of formation with phosphatic rocks in S.L Nepal. J. geol. Soc. India, 21, 484-491. Bashyal R.P. 1984 - Geological outline of Nepal Himalaya. In Tectonics of Asia, colloq. n05, 27th Int. Geol. Congr., Moscow, rep., 5, 159-180. Bassoullet J.P., Colchen M. & Mouterde R. 1977 - Esquisse paleogeographique et essai sur l'evolution geodynamique de l'Himalaya. Mem. h.ser. Soc. geol. fr., 8, 213-234. Bassoullet J.P., Colchen M., Marcoux J. & Mascle G. 1978 - Une transversale de la zone de l'Indus de Khalsi a Phothaksar, Himalaya du Ladakh. C.R. Acad. Sci., Paris, 286, 0, 563-566. Bassoullet J.P., Colchen M., Juteau T., Marcoux J. & Mascle G. 1980a - L'edifice de nappes du Zanskar (Ladakh, Himalaya). C.R. Acad. Sci., Paris, 290, 0,389-392. Bassoullet J.P., Boulin J., Colchen M., Marcoux J., Mascle G. & Monhnat C. 1980b - l' evolution des domaines tethysiens au pourtour du bouclier indien du Carbonifere au cretace. In Geologie des chaines alpines issues de la Tethys, J. Aubouin, J. Debelmas & M. Latreille eds, Paris, Mem. BRGM, 115, 180-198. Bassoullet J.P., Colchen M., Juteau T., Marcoux J., Mascle G. & Reibel G. 1983 - Geological studies in the Indus suture zone of Ladakh (Himalayas). In Contribution to Himalayan geology, 2, V.J. Gupta ed., Hindust. Publ. Corp., Delhi, 96-124. Bassoullet J.P., Enay R. & Mouterde R. 1986 - La marge nord-himalayenne au Jurassique. In Evolution des domaines orogeniques d'Asie meridionale, P. Le fort, M. Colchen & C. Montenat eds, Sci. de la Terre, Nancy, mem. 47,43-60. Baud A., Arn R., Bugnon P., Crisinel A., Dolivo E., Escher A., Hammerschlag J.G., Marthaler M., Masson M., Steck A. & Tieche J.C. 1982 - Le contact Gondwana - peri- Gondwana dans Ie Zanskar oriental (Ladakh, Himalaya). Bull. Soc. geol. fr., (7) XXIV, 2, 341-361. Baud A., Gaetani M., Garzanti E., fois E., Nicora A. & Tintori A. 1984 - Geological observations in southeastern Zanskar and adjacent Lahul area (northwestern Himalaya). Eclogae Geol. Helv., 77, 1, 171-197. Beden M. & Brunet M. 1986 - faunes de mammiferes et paleobiogeographie des domaines indiens et peri-indiens au Neogene. In Orogenic evolution of southern Asia, P. Le fort, M. Colchen & C. Montenat eds, Sci. de la Terre, Nancy, mem. 47,61-87. Berggren W.A. & Couvering van J.A. 1974 - The late Neogene: biostratigraphy, geochronology and paleoclimatology of the last 15 Ma in marine and continental sequences. Palaeogeogr. Palaeoclimatol. Palaeoecol., 16, 1-216. Berthe D., Choukroune P. & Jeguzo P. 1979 - Orthogneiss, mylonite and non-coaxial deformation of granites: the example of the south Armorican shear zone. J. of Struct. Geo 1., 1, 1, 31-42. Bertrand J.M. & Debon f. 1986 - Evolution tectonique polyphasee de la chaine du Karakorum (Baltoro, Nord Pakistan). C.R. Acad. Sci., Paris, 303, ser.II, 17, 1611-1614. Besse J., Courtillot V., Pozzi J.P., Westphal M. & Zhou V.X. 1984 - Palaeomagnetic estimates of crustal shortening in the Himalayan thrusts and Zangbo suture. Nature, 311, 621-626. Bhalla N.S. & Gupta J.N. 1979 - U-Pb isotopic ages of uraninites from Kulu, Himachal Pradesh and Berinag, Uttar Pradesh. J. Geol. Surv. India, 20, 10, 481-488. 368 Bhanot V.B., Singh V.P., Kansal A.K. & Thakur V.C. 1977 - Early Proterozoic Rb-Sr whole- rock age for central crystalline gneiss of Higher Himalaya, Kumaun. J. Geol. Soc. India, 18, 2, 90-91. Bhanot V.B., Singh V.P., Pandey B.K. & Singh R. 1981 - Rb-Sr isochron age for the gneissic rocks of Askot crystallines, Kumaun Himalaya (U.P.). In Contemporary geoscientific researches in Himalaya, A.K. Sinha ed., Bishen Singh Mahendra Pal Singh, Dehra Dun, 1, 117-119. Bhargava O.N. 1980 - Pre-Tertiary orogenies in the Himalaya: a review of various evidences. Geol. Rundsch., 69, 811-823. Bhargava O.N. 1982 - The tectonic windows of the Lesser Himalaya. Himalayan Geol., 10, 135- 155. Bhat M.I. 1984 - Abor volcanics: further evidence for the birth of the Tethys Ocean in the Himalayan segment. J. Geol. Soc. London, 141, 763-775. Bhat M.I. & Zainuddin S.M. 1978 - Geochemistry of the Panjal traps of Mount Kayol, Lidderwat, Pahalgam, Kashmir. J. Geol. Soc. India, 19, 9, 403-410. Bhattacharya A.R. 1982 - The Lesser Himalayan sediments: Precambrian span. In Geology of Vindhyanchal, K .S. Valdiya, S.B. Bhatia & V.K. Gaur eds, Delhi, Hindust. Publ. Corp., 200-210. Bhattacharyya D.S. & Das K.K. 1983 - Inversion of metamorphic zones in the Lower Himalayas at Gangtok, Sikkim, India. J. of Geol., 91,98-102. 18 Blattner P., Dietrich V. & Gansser A. 1983 - Contrasting 0 enrichment and origins of High Himalayan and Transhimalayan intrusives. Earth Planet. Sci. Lett., 65, 276-286. Blondeau A., Bassoullet J.P., Colchen M., Han T .L., Marcoux J., Mascle G. & van Haver T. 1986 - Disparition des formations marines a l'Eocene inferieur en Himalaya. In Orogenic evolution of southern Asia, P. Le Fort, M. Colchen & C. Montenat eds, Sci. de la Terr e, Nancy, memo 47, 103-111. Bordet P. 1961 - Recherches geologiques dans l'Himalaya du Nepal, region du Makalu. Ed. Cent. Natl. Rech. Sci., Paris, 275 p. Bordet P. 1975 - Les volcans recents du Dacht-e Nawar (Afghanistan central). Ann. sci. Univ. Clermont, n053, fasc. 26, 95 p. Bordet P., Colchen M., Krummenacher D., Le Fort P., Mouterde R. & Remy J.M. 1968 - Esquisse geologique de la Thakkhola (Nepal central). Ed. Cent. Natl. Rech. Sci., Paris, 1/75.000. Bordet P., Colchen M. et Le Fort P. 1971 - Esquisse geologique du Nyi-Shang (Nepal central). Ed. Cent. Nat. Rech. Sci., 1/75.000. Bordet P., Colchen M. & Le Fort P. 1972 - Some features of the geology of the Annapurna range, Nepal Himalaya. Himalayan Geol., 2, 537-563. Bordet P., Colchen M. & Le Fort P. 1975 - Recherches geologiques dans l'Himalaya du Nepal, region du Nyi-Shang. Ed. Cent. Natl. Rech. Sci., Paris, 138 p. Bordet P., Colchen M., Le Fort P. & P~cher A. 1981 - The geodynamic evolution of the Himalaya. Ten years of research in central Nepal Himalaya and some other regions. In Zagros - Hindukush - Himalaya: geodynamic evolution, H.K. Gupta & F .M. Delany eds, Am. Geophys. Union, Geodyn. ser., 3,149-168. Bossart P. & Ottiger R. (preprint) - Rocks of the Murree formation: indicators of a descending foreland basin of late Paleocene to middle Eocene age. Submitted to J. of Geol. Bouchez J.L. & P~cher A. 1981 - Himalayan Main Central Thrust pile and its quartz-rich tectonites in central Nepal. Tectonophys., 78, 23-50. Bouchez J.L. & P~cher A. 1976 - Microstructures and quartz preferred orientations in quartzites of the Annapurna area (central Nepal), in the proximity of the Main Central Thrust. Himalayan Geol., 6, 118-132. 369 Brookfield M.E. 1981 - Metamorphic distributions and events in the Ladakh range, Indus suture zone and Karakorum mountains. In Metamorphic tectonites of the Himalaya, P.S. Saklani ed., Today & Tomorrow's print. & publ., New-Delhi, 1-14. Brookfield M.E. & Reynolds P.H. 1981 - Late Cretaceous emplacement of the Indus suture zone ophiolitic melanges and an Eocene-Oligocene magmatic arc on the northern edge of the Indian plate. Earth and Planet. Sci. Lett., 55, 157-162. Brookfield M.E. & Andrews-Speed C.P. 1984 - Sedimentation in a high-altitude intermontane basin - the Wakka Chu molasse (mid-Tertiary, northwestern India). Bull. Ind. Geol. Assoc., 17, 2, 175-193. Brun J.P., Burg J.P. & Chen G.M. 1985 - Strain trajectories above the Main Central Thrust (Himalaya) in southern Tibet. Nature, 313, 388-390. BruneI M. 1975 - La nappe du Mahabharat, Himalaya du Nepal Central. C.R. Acad. Sci., Paris, 280, 551-554. BruneI M. 1980 - Quartz fabrics in shear zone mylonites: evidence for a major imprint due to late strain increments. Tectonophys., 64, 133-T44. Brunel M. 1983 - Etude pet ro-structurale des chevauchements ductiles en Himalaya (Nepal oriental et Himalaya du Nord-Ouest). These Doct. Etat, Univ. Paris VII, 395 p. Brunel M. 1986 - Ductile thrusting in the Himalayas: shear sense criteria and stretching lineations. Tectonics, 5, 2, 247-265. BruneI M. & Andrieux J. 1977 - Deformations superposees et mecanismes assocles au ch evauch ement central h imalayen "M.C. I.": Nepal oriental. Colloq. in t. 268, Ecologie et geologie de l'Himalaya, Paris, ed. Cent. Natl. Rech. Sci., vol. Sci. de la Terre, 69-83. Brunel M., Chaye d'Albissin M. & Locquin M. 1985 - The Cambrian age of magnesites from E. Nepal as dete rmined through the discovery of paleobas idiospores. J. Geo 1. Soc. India, 26, 4, 255-260. Burbank D.W. 1983a - The chronology of intermontane-basin development in the northwestern Himalaya and the evolution of the northwest syntaxis. Earth Planet. Sci. Lett., 64, 77-92. Burbank D.W. 1983b - Multiples episodes of catastrophic flooding in the Peshawar basin during the past 700,000 years. Geol. Bull. Univ. Peshawar, 16, 43-49. Burbank D.W. & Grant M.J. 1985 - Plio-Pleistocene cyclic sedimentation in the Kashmir basin, Northwestern Himalaya: Milankovitch periodicities and grain-size data. Z. fur Gletscherkunde un Glazialgeologie, Universit~tsverlag Wagner, Innsbruck, Austria, 21, 229-236. Burbank D.W. & Johnson G.D. 1983 - The late Cenozoic chronologic and stratigraphic development of the Kashmir inte rmontane basin, northwestern Himalaya. Palaeogeogr. Palaeoclimatol. Palaeoecol., 43, 205-235. Burbank D.W. & Reynolds R.G.H. 1984 - Sequential late Cenozoic structural disruption of the northern Himalayan foredeep. Nature, 311, 5982, 114-118. Burbank D.W. & Tahirkheli A.K. 1985 - The magnetostratigraphy, fission-track dating, and stratigraphic evolution of the Peshawar intermontane basin, northern Pakistan. Geol. Soc. Am. Bull., 96, 539-552. Burchfield B.C. & Royden L.H. 1985 - North-south extension within the convergent Himalayan region. Geology, 13, 679-682. Burg J.P. 1983 - Tectogenese comparee de deux segments de chaine de collision: - le Sud du Tibet (suture du Tsangpo), - la chaine hercynienne en Europe (sutures du Massif central). These Doct. Etat, Univ. Montpellier, 368 p. Burg J.P. & Chen G.M. 1984 - Tectonics and structural zonation of southern Tibet, China. Nature, 311, 5983, 219-223. 370 Burg J.P., BruneI M., Gapais D., Chen G.M. & liu G.H. 1984a - Deformation of leucogranites of the crystalline Main Central Sheet in southern Tibet (China). J. Struct. Geol., 6, 5, 535-542. Burg J.P., Guiraud M., Chen G.M. & Li G.C. 1984b - Himalayan metamorphism and deformations in the North Himalayan belt (southern Tibet, China). Earth Planet. Sci. lett., 69, 391-400. Burg J.P., Marcoux J. & Cheng G.M. 1985 - Wildflysch and exotic blocks along the Yarlung Zangbo suture zone: age and geodynamic significance. Terra Cogni ta, 5, 2-3, p. 125 (abstr.). Burg J.P., leyreloup A., Girardeau J. & Chen G.M. 1987 - Structure and metamorphism of a tectonically thickened continental crust: the Yalu Tsangpo suture zone (Tibet). Phil. Trans. Royal Soc. london, A, 321, 67-86. Calkins J.A., Offield T.W., Abdullah S.K.M. & Tayyab Ali S. 1975 - Geology of the southern Himalaya in Hazara, Pakistan, and adjacent areas. U.S. Geol. Surv. prof. pap., 716-C, 29 p. Calkins J.A., Offield T.W. & Tayyab Ali S. 1969 - Geology and mineral resources of the southern Hazara district, west Pakistan and part of western Kashmir. U.S. Dep. of the Interior, Geol. Surv., 92 p. Chang C. F. & Cheng H. 1973 - Some tectonic features of the Mount Jo Imo lungma area, southern Tibet, China. Sci entia Sinica, 16, 257-265. Chang C.F., Zheng X.l. & Pan Y.s. 1977 - The geological history, tectonic zonation and origin of uplifting of the Himalaya, Beijing, Inst. of Geol., Acad. Sinica, 17 p. Chaudhri R.S. & Gill G.T.s. 1981 - Heavy mineral assemblage of the Siwalik group of Nepal Himalaya. J. Geol. Soc. India, 22, 5, 220-226. Chen W.P. & Molnar P. 1977 - Seismic moments of major earthquakes and the average rate of slip in central Asia. J. geophys. Res., 82, 2945-2969. Chugh R.S. 1974 - Study of recent crustal movements in India and future programs. Int. Symp. on recent crustal movements, Zurich. Colchen M. 1981 - l'evolution paleogeographique de l'Himalaya et les concepts de Gondwana et de Tethys. In Paleogeographie et biogeographie de l'Himalaya et du sous-continent indien, ed. Cent. Natl. Rech. Sci., 15-21. Colchen M., le Fort P. & Pecher A. 1980 - Carte geologique Annapurna - Manaslu - Ganesh. Himalaya du Nepal (1/200.000). Ed. Cent. Natl. Rech. Sci., Paris. Colchen M., le Fort P. & Pecher A. 1986 - Annapurna - Manaslu - Ganesh Himal. Notice de la carte geologique au 1/200.000i. In Recherches geologiques dans l'Himalaya du NepaL Ed. Cent. Natl. Rech. Sci., Paris, 138 p. Col chen M., Mascle G. & Van Haver T. 1986 - Some aspects of collision tectonics in the Indus suture zone. In Collision tectonics, M.P. Coward & A.C. Ries eds, Geol. Soc. london, spec. publ. 19, 173-184. Coleman R.G. 1977 - Ophiolites: ancient oceanic lithosphere? Springer Verlag, He ide Iber g. Coulon C., Maluski H., Bollinger C & w3§g S~O 1986 - Mesozoic and Cenozoic volcanic rocks from central and southern Tibet: Ar- Ar dating, petrological characteristics and geodynamical significance. Earth & Planet. Sci. lett., 79, 281-302. Coward M.P., Jan M.Q., Rex D., Tarney J., Thirlwall M. & Windley B.F. 1982 - Structural evolution of a crustal section in the western Himalaya. Nature, 295, 5844, 22-24. Coward M.P. & Butler R.W.H. 1985 - Thrust tectonics and the deep structure of the Pakistan Himalaya. Geology, 13, 417-420. Coward M.P., Windley B.F., Broughton R.D., luff I.W., Petterson M.G., Pudsey C.J. & Rex D.C. 1986 -Collision tectonics in NW Himalayas. In Collision tectonics, M.P. Coward & A.C. Ries eds, Geol. Soc. london, spec. publ. 19, 203-219. Crawford A.R. 1969 - India, Ceylan and Pakistan: new age data and comparisons with Australia. Nature, 223, 380-384. 371 Crawford A.R. 1974 - The Salt range, the Kashmir syntaxis and the Pamir arc. Earth Planet. Sci. Lett., 22, 371-379. Crawford A.R. & Davies R.G. 1975 - Age of Pre-Mesozoic formations of the Lesser Himalaya, Hazara district, Northern Pakistan. Geo!. Mag., 112, 5, 509-514. Cuney M., Le Fort P. & Wang Z.X. 1984 - Uranium and thorium geochemistry and mineralogy in the Manaslu leucogranite (Nepal, Himalaya). In Proceedings symposium on Geology -of granites and their metallogenetic relations. oct. 1982, Nanjing Univ., K. Xu & G. Tu eds, Sci. Press, Beijing, 853-873. Curray J.R. & Moore D.G. 1971 - Growth of Bengal deep-sea fan and denudation in the Himalayas. Geo!. Soc. Am. Bull., 82, 563-572. Curray J.R., Emmel F.J., Moore D.G. & Raitt R.W. 1982 - Structure, tectonics and geological history of the norheastern Indian ocean. In The ocean basins and margins, A.E.M. Nairn & F .G. Stehli eds, vol.6, Indian ocean, ch.9, 399-450. Das B.K. & Pande I.C. 1973 - Zones of progressive regional metamorphism in Dudatoli syncline, Garhwal Himalaya. Himalayan Geol., 3, 190-208. Das Gupta A.B. 1979 - Petroleum occurences in the terrain adjoining the northeastern Himalaya. Geol. Surv. India, Miscellenaous pub!, 41, V, 309-321. Davis D.M., Suppe J. & Dahlen F .A. 1983 - Mechanics of fold-and-thrust belts and accretionary wedges. J. Geophys. Res., 88, B2, 1153-1172. Debon F. & Le Fort ~ 1982 - A chemical-mineralogical classification of common plutonic rocks and associations. Trans. R. Soc Edinburgh: Earth Sci., 73, 135-149. Debon F. & Le Fort P. 1984 - Chemical-mineralogical classification of plutonic rocks and associations - examples from southern Asia belts. In Proc. Int. Symp. on geology of granites and their metallogenetic relations, Nanjing Univ. 1982, K.Q. Xu & G.C. Tu eds, Beijing: Sci. Press, 293-311. Debon F., Le Fort P., Dautel D., Sonet J. & Zimmermann J.L. 1987a - Plutonism in western Karakorum and northern Kohistan (Pakistan): a composite Mid-Cretaceous to upper Cenozoic magmatism. Lithos, 20, 19-40. Debon F., Le Fort P., Sheppard S.M.F. & Sonet J. 1986a - The four plutonic belts of the Transhimalaya - Himalaya: a chemical, mineralogical, isotopic and chronological synthesis along a Tibet - Nepal section. J. Petrol., 27(1), 219-250. Debon F., Le Fort P. & Sonet J. 1981 - Granitoid belts west and south of Tibet. About their geochemical trends and Rb-Sr isotopic studies. Proc. Symp. on Qinghai-Xizang (Tibet) plateau, Beijing, 1980, V.1, Beijing: Sci. Press, New York: Gordon & Breach, 395-405. Debon F., Sonet J., Liu G.H., Jin C.W. & Xu R.H. 1982 - Caracteres chimico-mineralogiques majeurs et datations Rb-Sr des trois ceintures plutoniques du Tibet meridional. C.R. Acad. Sci., Paris, 295, 213-218. Debon F., Sonet J., Liu G.H., Jin C.W. & Xu R.H. 1984 - Caracteres chimico-mineralogiques et datations par Rb-Sr des trois ceintures plutoniques du Tibet meridional. In Mission franco-chinoise au Tibet 1980, "Etude geologique et geophysique de la croute terrestre et du manteau superieur du Tibet et de l'Himalaya", J.L. Mercier & G. Li eds, Cent. Natl. Rech. Sci.-Acad. Sci. geol. de Chine redaction, 309-317. Debon F., Zimmermann J.L., Liu G.H., Jin C.W. & Xu R H 1985 - Time relationships between magmatism, tectonics and metamorphism in three plutonic belts in southern Tibet: new K-Ar data. Geol. Rundsch., 74/2, 229-236. Debon F., Zimmermann J.L. & Bertrand J.M. 1986b - Le granite du Baltoro (batholite axial du Karakorum, nord Pakistan): une intrusion sub-alcaline d'age Miocene superieur. C.R. Acad. Sci., Paris, 303, ser.II, 6, 463-468. 372 Debon F., Afzali H., Le Fort P., Sonet J. &: Zimmermann J.L. 1987b - Plutonic rocks and associations in Afghanistan: typology, age and geodynamic setting. Mem. Sci. de la Terre, Nancy, 49, 132 p. Delcaillau B. 1986 - Evolution geomorphostructurale d'un piemont frontal de chaine de collision intracontinentale: les Siwalik de I' Himalaya du Nepal oriental. These Doct. Etat, Univ. Toulouse-le- Mirail, 435 p. Deniel C. 1985 - Apport des isotopes du Sr, du Nd et du Pb a la connaissance de l'age et de l'origine des leucogranites himalayens. Exemple du Manaslu (Himalaya, Nepal). These Doct. Univ. Clermont II, 151 p. Deniel C., Vidal Ph. &: Le Fort P. 1983 - The Manaslu granite (Himalaya, Nepal): new Sr and Nd isotopic data. Ter ra Cogni ta, 3, p. 266. Deniel C., Vidal Ph., Fernandez A., Le Fort P. &: Peucat J.J. 1987 - Isotopic study of the Manaslu granite (Himalaya, Nepal): inferences on the age and source of Himalayan leucogranites. Contrib. Mineral. Petrol. (in press). Desio A. 1964 - Geological tentative map of the western Karakorum. Scale 1/500.000. Inst. of Geol., Univ. Milan. Desio A. 1976 - Some geotectonic problems of the Kashmir Himalaya - Karakorum - Hindu Kush and Pamir area. In Geotettonica delle zone orogeniche del Kashmir Himalaya - Karakorum -Hindu Kush - Pamir. Rend. Accad. Naz. Lincei, Roma, 115-129. Desio A. 1977 - The occurrence of blueschists between the middle Indus and the Swat valleys as an evidence of subduction (North Pakistan). Rend. Accad. Naz. Lincei, 52, 5, 1-9. Desio A. 1979 - Geologic evolution of the Karakorum. In Geodynamics of Pakistan (Farah &: De Jong eds), Geol. Surv. Pakistan, Quetta, 111-124. Desio A. &: Martina E. 1972 - Geology of the upper Hunza valley, Karakorum (W. Pakistan). Bull. Soc. Geol. Italy, 91, 283-314. Desio A. &: Shams F .A. 1980 - The age of the blueschists and the Indus-Kohistan suture line, tftI Pakistan. Rend. Accad. Naz. lincei, ser.VIII, LXVIII, 1, 74-79. Dewey J.F. &: Burke K.C.A. 1973 - Tibetan, Variscan and Precambrian basement reactivation: products of continental collision. J. Geol., 81, 6, 683-692. Diener C. 1898 - Notes on the geological structure of Chitichum region. Mem. Geol. Surv. India, 28, (1), 1-27. Dietrich V. &: Gansser A. 1981 - The leucogranites of the Bhutan Himalaya (crustal anatexis versus mantle melting). Bull. suisse Mineral. petrogr., 61, 177-202. Dietrich V.J., Frank W. &: Honegger K. 1983 - A Jurassic-Cretaceous island arc in the Ladakh-Himalayas. J. Volcanol. Geothermal Res., 18, 405-433. Falconer H. 1832 - Dehradun fossil remains. J. Asiat. Soc. Bengal, 1, p.249. Falcone r H. 1835 - J. Asi at. Soc. Ben gal, 4, p.58. Fernandez A. 1983 - Strain analysis of a typical granite of the Lesser Himalayan cordierite granite belt: the Mansehra pluton, northern Pakistan. In Granites of Himalayas Karakorum and Hindu Kush, F .A. Shams ed., Centenary celebrations, 1882- 1982, Univ. Punjab, Inst. Geol., Lahore, Pakistan, 183-199. Ferrara G., Lombardo B. &: Tonarini S. 1983 - Rb/Sr geochronology of granites and gneisses from the Mount Everest region, Nepal Himalaya. Geol. Rundsch., 72, 119-136. Ferrara G., Lombardo B., Tonarig~ S.8~ Turi B1819~~ - Miocene anatexis in the Higher Himalaya basement nappes: Sri Sr and 0/ 0 evidence from the Gophu La leucogranite of Bhutan. Terra Cogni ta, 5, p. 228. Fort M. 1985 - Phases d'accumulations sedimentaires internes et phases orogeniques au Sud du massif de l'Annapurna: l'exemple du bassin de Pokhara (Nepal). In Montagnes et piemonts, Rev. geogr. des Pyrenees et du Sud-Ouest, Toulouse, trav.I, 25-47. 373 Fort M. I'x Freytet P. 1979 - L'evolution sedimentaire recente du bassin intramontagnard de Pokhara (Himalaya, centre ouest Nepal). C.R. Acad. ScL, Paris, 289, 1195-1198. Fort M., Freytet P. I'x Colchen M. 1982 - Structural and sedimentological evolution of the Thakkhola Mustang graben (Nepal Himalayas). Z. Geomorph. N.f., supppl. Bd.42, 75-98. fort M. I'x Gupta V.J. 1981 - Plio-Pleistocene midland Himalayan basins of Kathmandu, Pokhara and Kashmir. Neogene/Quaternary boundary field conference, India, 1979, Proc., 37-43. France-Lanord C., Sheppard S.M.F. I'x Le Fort P. 1985 - The Manaslu leucogranite (central Nepal Himalaya): hydrogen isotopic data of coexisting muscovite and biotite. Terra Cognita, 5, 228-229 (abstr.). france-Lanord C., Le fort P. I'x Sheppard S.M.F. 1986 - Le leucogranite du Manaslu (Haut Himalaya): compositions isotopiques D/H et origine des fluides magmatiques. 11i Reun. Annu. Sci. de la Terre, Clermont-ferrand, Soc. geol. Fr., p.69 (resume). France-Lanord C., Le Fort P., Pecher A. I'x Sheppard S.M.F. (in press) - fluids of the Himalayan M.C.T. from Central Nepal: carbon and oxygen isotope study of some quartzpods and siliceous marbles. Himalayan Geol. semin., nov.1984, Dehra Dun, Himalayan Geol., 14. Frank W., Hoinkes G., Miller C., Purtscheller f., Richter W. I'x Thoni M. 1973 - Relations between metamorphism and orogeny in a typical section of the indian Himalayas. TMPM Tschermaks Min. Petro Mitt., 20, 303-332. Frank W., Gansser A. I'x Trommsdorff V. 1977 - Geological observations in the Ladakh area (Himalayas). A preliminary report. Bull. suisse mineral. petrogr., 57, 89-113. frost D.L., Sharma K.K. I'x Frost E.G. 1984 - Mesozoic-Cenozoic evolution of the Indus suture zone, Ladakh: K-Ar age data. Int. Assoc. Seismol. and Phys. of the Earth's interior, reg. assem., Hyderabad, India, oct.-nov. 1984, 239-240 (abstr.). fuchs G. 1967 - Geological map of the Dolpo region and the Dhaula Himal, 1/100.000. Osterr. Akad. Wiss., Wien, 113. Fuchs G. 1975 - Contributions to the geology of the north-western Himalayas. Abh. Geol. Bundesanst., 32,59 p. fuchs G. 1979 - On the geology of western Ladakh. Jb. Geol., B.-A., Wien, 122/2, 513-540. fuchs G. 1981 - Outline of the geology of the Himalaya. Mitteil. osterr. Geol. Ges., 74/75, 101-127. Fuchs G. 1982 - Explanations of the geologic-tectonic map of the Himalaya. Geol. Surv. Austria, Wien, 50 p. Fuchs G. I'x Frank W. 1970 - The geology of west Nepal between the rivers Kali Gandaki and Thulo Bheri. Jb. Geo!., B.-A., Wien, 18, 103 p. Gaetani M., Nicora A. I'x Premoli Silva I. 1980 - Uppermost Cretaceous and, Paleocene in the Zanskar range (Ladakh Himalaya). Riv. Ita!. Paleont. Stratigr., Milano" 86, 127-166. Gairola V.K. I'x Saxena R.K. 1981 - Low grade metamorphites and strain study of a part of Garhwal synform. Satpuli, Pauri Garhwal (U.P.). In Metamorphic tectonites of the Himalaya, P.S. Saklani ed., Today and tomorrow's print. I'x pub!" New-Delhi, 43-62. Gansser A. 1964 - Geology of the Himalayas. Intersci. publ., John Wiley I'x sons, London, 289 p. Gansser A. 1966 - The Indian Ocean and the Himalayas. A geological interpretation. Eclogae Geol. Helv., 59, 2, 831-848. Gansser A. 1977 - The great suture zone between Himalaya and Tibet. A preliminary account. Colloq. into 268, Ecologie et geologie de l'Himalaya, Paris, ed. Cent. Natl. Rech. Sci., vol. Sci. de la Terre, 181-191. Gansser A. 1981 - The geodynamic history of the Himalaya. In Zagros - Hindu Kush - Himalaya: geodynamic evolution, H.K. Gupta I'x F .M. Delany eds, Am. Geophys. Union, Washington, Geodyn. ser.3, 111-121. 374 Gansser A. 1983a - Geology of the Bhutan Himalaya. Mem. Soc. Helv. Sci. Nat., Birkhauser Verlag, Bile, 96, 181 p. Gansser A. 1983b - The morphogenetic phase of mountain building. In Mountain building processes, K .J. Hsu· ed., Acad. Press, London, 221-228. Gapais D., Gilbert E. & Pecher A. 1984 - Structures et trajectoires de deformation dans la zone de suture de l'Indus-Tsangpo en Himalaya du Ladakh, region de la Suru. C.R. Acad. Sci., Paris, II, 299, 4, 179-182. Garzanti E., Casnedi R., & Jadoul F. 1986 - Sedimentary evidence of a Cambro-Ordovician orogenic event in the northwestern Himalaya. Sediment. Geol., 48, 237-265. Geological Bureau of Xizang 1982 - Tectonics of Yarlung Zangbo suture zone Xizang (Tibet). Guide to Geological Excursion. Geol. Bur. of Xizang, Geol. Soc. of Xizang, China, 49 p. Gilbert E. 1986 - Evolution structurale d'une chaine de collision: structures et deformations dans Ie Nord de la plaque indienne en Himalaya du Ladakh (cristallin du Haut Himalaya et series tethysiennes). These Doct. Univ. Poitiers, 225 p. Gill W.D. 1951a - The stratigraphy of the Siwalik series in the northern Potwar, Punjab, Pakistan. Quart. J. Geol. Soc. London, CVIl, 4, 375-394. Gill W.D. 1951b - The tect onics of the sub-Himalayan fault zone in the nor ther n Potwa r Region and in the Kangra District of the Punjab. Quart. J. Geol. Soc. London, CVIl, 4, 395-421. Girardeau J., Marcoux J. & Yougong Z. 1984 - Lithologic and tectonic environment of the Xigaze ophiolite (Yarlung Zangbo suture zone, southern Tibet, China), and kinematics of its emplacement. Eclogae geol. Helv., 77, 1, 153-170. Girardeau J., Mercier J.C.C. & Wang X. 1985a - Petrology of the mafic rocks of the Xigaze ophiolite, Tibet: Implications for the genesis of the oceanic lithosphere. Contrib. to Mine ral. & Petro 1., 90, 309-321. Girardeau J., Mercier J.C.C. & Zao Y.G. 1985b - Structure of the Xigaze ophiolite, Yarlung Zangbo suture zone, Southern Tibet, China: genetic implications. Tectonics, vol. 4(3), 267-288. Gapel Ch., Allegre C.J. & Xu R.H. 1984 - Lead isotopiC study of the Xigaze ophiolite (Tibet): the problem of the relationship between magmatites (gabbros, dolerites, lavas) and tectonites (harzburgites). Earth Planet. Sci. Lett., 69, 301-310. Gupta V.J. 1975 - On the stratigraphic position of the Kathmandu valley sediments, Nepal. Geogr. Helv., 30, 27-28. Haller J. 1979 - Himalayan orogenesis in perspective. Foreword to Metamorphic rock sequences of the eastern Himalaya, P.K. Verma ed., K.P. Bagchi & Co., Calcutta, 34 p. Hamilton W. 1983 - Configuration of eastern Gondwanaland. U.S. Geol. Surv. circular 911, 2-3. Hashimoto S., Ohta Y. & Akiba C. eds 1973 - Geology of the Nepal Himalayas. Sapporo (Japan), Saikon Publ. Co., 286 p. Hayden H.H. 1905 - Geology of the province of Tang and U in central Tibet. Rec. Geol. Surv. India, 32, 160-174. Heim A. & Gansser A. 1939-1975 - Central Himalaya. Geological observations of the Swiss expedition 1936. Mem. Soc. Helv. Sci. Nat., ZUrich, 73(1), 245 p. Repr.: Hindustan Publ. Corp., Delhi. Hennig A. 1916 - Zur Petrographie und Geologie von Sudwesttibet. In Hedin S., southern Tibet, 5, Stockholm, 220 p. Herail G. & Mascle G. 1980 - Les Siwaliks du Nepal central: structure et geomorphologie d'un piedmont en cours de deformation. Bull Assoc. Geogr. fr., Paris, 471, 259-267. 375 Herail G. & Mascle G. 1985 - Le piemont himalayen et le front du Mahabharat dans le Nepal central: un piemont frontal de chaine de collision. In Montagnes et piemonts, Rev. Geogr. Pyrenees et Sud-Ouest, Toulouse, trav.I, 9-24. Herail G., Mascle G. & Delcaillau B. 1986 - Les Siwaliks de l'Himalaya du Nepal: un exemple d'evolution geodynamique d'un prisme d'accretion intracontinental. In Evolution des domaines orogeniques d'Asie meridionale, livre jubilaire P. Bordet, P. Le Fort, M. Colchen & C. Montenat eds, Sci. de la Terre, Nancy, mem.47, 155-182. Herren E. 1987 - The Zanskar shear zone: northeast-southwest extension within the Higher Himalayas (Ladakh, India). (Submitted to Geol.) Hodges K.V., Le Fort P. & Pecher A. 1986 - Possible evidence for anatectic buffering of the thermal structures of the Tibetan Slab, central Nepalese Himalaya. Geol. Soc. Am. (ab st r. ). Holdaway M.J. 1971 - Stability of andalusite and the aluminum silicate phase diagram. Am. J. Sci., 271, 97-131. Holz F. 1987 - Etude structurale, metamorphique et geochimique de granitoides hercyniens et de leur encaissant dans la region de Montalegre (Tras-os Montes, Nord Portugal). These Doct. Univ. Nancy I, 220 p. Honegger K .H. 1983 - Strukturen und Metamorphose im Zanskar Kristallin (Ladakh-Kashmir, Indien). Thesis, n07456, ETH, ZUrich, 117 p. Honegger K., Dietrich V., Frank W., Ganser A., Thani M. & Trommsdorff V. 1982 - Magmatism and metamorphism in the Ladakh Himalaya (the Indus-Tsangpo suture zone). Earth Planet. Sci. Lett., 60, 253-292. Honegger K.H., Le Fort P., Mascle G. & Van Haver T. 1985 - Presence de schistes bleus a glaucophane dans la zone de suture de l'Indus au Ladakh. Terra Cognita, 5, p.257. Ishida T. & Ohta Y. 1973 - Ramechhap - Okhaldunga region. In Geology of Nepal Himalayas, Hashimoto S. ed., Sapporo, Japan, Saikon publ. Co., 35-68. Ivanac J.F., Traves D.M. & King D. 1956 - The geology of the North-West portion of the Gilgit agency. Records geol. Surv. of Pakistan, 7, 3-27. Jain A.K. & Thakur V.C. 1978 - Abor volcanics of the Arunachal Himalaya. J. Geol. Soc. India, 19, 8, 335-349. Jain A.K., Goel R.K. & Nair N.G.K. 1980 - Implications of Pre-Mesozoic orogeny in the geological evolution of the Himalaya and Indo-Gangetic plains. Tectonophys., 62, 67- 86. Jan M.O., Asif M., Tahirkheli T. & Kamal M. 1981a - Tectonic subdivision of granitic rocks of north Pakistan. Geol. BUll. Univ. Peshawar, 14, 159-182. Jan M.O. & Howie R.A. 1981 - The mineralogy and geochemistry of the metamorphosed basic and ultrabasic rocks of the Jijal complex, Kohistan, N.W. Pakistan. J. Petrol., 22, 1, 85-126. Jan M.O., Kamal M. & Khan M.l. 1981b - Tectonic control over emerald mineralization in Swat. Geol. Bull. Univ. Peshawar, 14, 101-109. Jan M.O. & Mian 1. 1971 - Preliminary geology and petrography of Swat Kohistan. Geol. Bull. Univ. PeShawar, 6, 1-32. Jan M.O. & Symes R.F. 1977 - Piemontite schists from upper Swat, north-west Pakistan. Mineral. Mag., 41, 537-540. Jin C. & Xu R. 1984 - Les grani toides de la partie centrale de l'Himalaya et du Gangdise au Xizang (Tibet) meridional. In mission franco-chinoise au Tibet 1980, J.L. Mercier & G. Li eds, ed. Cent. Natl. Rech. Sci., Paris, 289-308. Johnson B.D., Powell C. Mc A. & Veevers J.J. 1980 - Early spreading history of the Indian ocean between India and-Australia. Earth Planet. Sci. Lett., 47, 131-143. 376 Johnson G.D., Johnson N.M., Opdyke N.D. 8. Tahirkheli R.A.K. 1979 - Magnetic reversal stratigraphy and sedimentary tectonic history of the upper Siwalik group, eastern Salt range and southwestern Kashmir. In Geodynamics of Pakistan, A. Farah 8. K.A. de Jong eds, Geol. Surv. Pakistan, Quetta, 149-165. Johnson G.D., Zeitler P., Naeser C.W., Johnson N.M., Summers D.M., Frost C.D., Opdyke N.D. 8. Tahirkheli R.A.K. 1982 - The occurrence and fission-track ages of late Neogene and Quaternary volcanic sediments, Siwalik group, northern Pakistan. Palaeogeogr. Palaeoclimatol. Palaeoecol., 37, 63-93. Johnson N.M., Opdyke N.D., Johnson C.D., Lindsay E.H. 8. Tahirkheli R.A.K. 1982 - Magnetic polarity stratigraphy and ages of Siwalik group rocks of the Potwar plateau, Pakistan. Palaeogeogr. Palaeoclimat. Palaeoecol., 37, 17-42. Johnson N.M., Stix J., Tauxe L., Cerveny P.F. 8. Tahirkheli R.A.K. 1985 - Paleomagnetic chronology, fluvial process, and tectonic implications of the Siwalik depostis near Chinji village, Pakistan. J. Geol., 93, 1, 27-40. Kai K. 1981a - Rb-Sr geochronology of the rocks of the Himalayas, eastern Nepal. Part II. The age and the origin of the granite of t~e Higher Himalayas. Mem. Fac. Sci., Kyoto Univ., ser. Geol. 8. Mineral., XLVII, 2, 1119-157. Kai K. 1981b - Rb-Sr ages of the biotite and muscovite of the Himalayas, eastern Nepal j its implication in the uplift history. Geochem. J., 15, 63-68. Karanth R.V. 1985 - New petrographic data on the metapelites of Almora crystallines in Kumaun Himalaya. J. Geol. Soc. India, 26, 435-452. Karu nakaran C. 8. Ranga Rao A. 1979 - Status of exploration for hydrocarbons in the Himalayan region. Contributions to stratigraphy and structure. Geol. Surv. India, miscellenaous publ. 41, V, 1-66. Kempe D.R.C. 1973 - The petrology of the Warsak alkaline granites, Pakistan, and their relationship to other alkaline rocks of the region. Geol. Mag., 110, 385-404. Kempe D.R.C. 1983 - Alkaline granites, syenites, and associated rocks of the Peshawar plain alkaline igneous province, NIl Pakistan. In Granites of Himalayas, Karakorum and Hindu Kush, F .A. Shams ed., Punjab Univ., Lahore, 143-169. Kempe D.R.C. 8. Jan M.Q. 1980 - The Peshawar plain alkaline igneous province, NIl Pakistan. Proc. Int. Comm. Geodyn., Group 6, Peshawar nov.1979, Geol. BUll. Univ. Peshawar, spec. issue, 13, 71-77. Khan R.H. 8. Tater J.M. 1970 - Radiometric dates of some Nepalese rocks. His Majesty's Government of Nepal, Ministry of Industry and Commerce, Nepal Gol. Surv., Kathmandu, 6 p. Klootwijk C.T. 1984 - A review of Indian Phanerozoic palaeomagnetism: implications for the India-Asia collision. Tectonophys., 105, 331-353. Klootwijk C.T., Conaghan P.J. 8. Powell C.Mc A. 1985 - The Himalayan Arc: large-scale continental subduction, oroclinal bending and back-arc spreading. Earth Planet. Sci. Lett., 75, 167-183. Kraft A. von 1902 - Notes on the "Exotic blocks" of Malla Johar in the Bhot Mahals of Kumaon. Mem. Geol. Surv. India, 32, 3, 127-183. Krummenacher D. 1971 - Geochronometrie des roches de l'Himalaya. In Recherches geologiques dans l'Himalaya du Nepal, region de la Thakkhola. Ed. Cent. Natl. Rech. Sci., Paris, VI, 187-202. Krummenacher D., Basett A.M., Kingery F .A. 8. Layne H.F. 1978 - Petrology, metamorphism and K/Ar age determinations in eastern Nepal. In Tectonic geology of the Himalaya, P.S. Saklani ed., Today 8. Tomorrow's print. 8. publ., New-Delhi, 151-166. Kumar G., Prakash G. 8. Singh K.N. 1974 - Geology of the Deoprayag-Dwarahat area, Garhwal, Chamoli and Almora districts, Kumaun Himalaya, uttar Pradesh. Himalayan Geol., 4, 323- 347. 377 Kumar G., Raina B.K., Bhatt D.K. & Jangpangi G.S. 1983 - Lower Cambrian body and trace fossils from the Tal formation, Garhwal synform, uttar Pradesh, India. J. Paleontol. Soc. India, 28, 106-111. Kumar S. 1982 - Vindhyan stromatolites and their stratigraphic testimony. In Geology of Vindhyanchal, K .S. Valdiya, S.B. Bhatia & V.K. Gaur eds, Hindustan publ. Corp., Delhi, 102-112. La Brecque J.L., Kent D.V. & Cande S.C. 1977 - Revised magnetic polarity time scale for late Cretaceous and Cenozoic time. Geology, 5, 330-335. Lal R.K., Mukerji S. & Ackermand D. 1981 - Deformation and Barrovian metamorphism at Takdah, Darjeeling (eastern Himalaya). In Metamorphic tectonites of the Himalaya, P.S. Saklani ed., Today & Tomorrow's print. & publ., New-Delhi, 231-278. Larson R.L., Mutter J.C., Diebold J.B. & Carpenter G.B. 1979 - Cuvier basin: a product of ocean crust formation by early Cretaceous rifting off western Australia. Earth Planet. Sci. Lett., 45, 105-114. Le Fort P. 1975a - Himalaya: the collided range. Present knowledge of the continental arc. Am J. Sci., 275A, 1-44. Le Fort P. 1975b - Les formations cristallophylliennes de la "Dalle du Tibet" en MarsyandL In Recherches geologiques dans l' Himalaya du Nepal, region du Nyi-Shang, Ch.I1, ed. Cent. Natl. Rech. ScL, Paris, 21-47. Le Fort P. 1981 - Manaslu leucogranite: a collision signature of the Himalaya. A model for its genesis and emplacement. J. geophys. Res., 86, B 11, 10545-10568. Le Fort P. 1986 - Metamorphism and magmatism during the Himalayan collision. In Collision tectonics, M.P. Coward & A.C. Ries eds, Geol. Soc. special publ., 19, 159-172. Le Fort P., Cuney M., Deniel C., France-Lanord C., Sheppard S.M.F., Upreti B.N. & Vidal Ph. 1987 - Crustal generation of Himalayan leucograni tes. Tectonophys., 134, n01-3, 39-57. Le Fort P., Debon F., Pecher A., Sonet J. & Vidal P. 1986 - The 500 Ma magmatic event in southern Asia, a thermal episode at Gondwana scale. In Evolution des domaines orogeniques d'Asie meridionale (de la Turquie a l'Indonesie), P. Le Fort, M. Colchen & C. Montenat eds, Sci. de la Terre, Nancy, mem.47, 191-209. Le Fort P., Debon F. & Sonet J. 1983a - The lower Palaeozoic "Lesser Himalayan" granitic belt: emphasis on the Simchar pluton of central Nepal. In Granites of Himalayas Karakorum and Hindu Kush, F .A. Shams ed., Inst. of Geol., Punjab Univ., Lahore, Pakistan, 235-255. Le Fort P., Debon F. & Stebbins J. 1978 - Mise en evidence d'une ceinture de granites a cordierite et enclaves "microgrenues" en Bas-Himalaya (Nepal, Inde, Pakistan~ 6i Reun. Annu. Sci. de la Terre, Orsay, Soc. geol. Fr., 243 (resume). Le Fort P., Debon F. & Sonet J. 1980 - The "Lesser Himalayan" cordierite granite belt. Typology and age of the pluton of Manserah (Pakistan). In Proc. Int. Geodyn. Conf., Peshawar, nov-dec. 1979, Geol. BUll. Univ. Peshawar (Pakistan), spec. issue, 13, 51-61. Le Fort P. & Jest C. 1974 - Les sources thermales du (Nepal. Rev. Mus. de l'Homme, Mus. Natl. Hist. Nat., Objets et Mondes, 14(4), 213-218. Le Fort P., Michard A., Sonet J. & Zimmermann J.L. 1983b - Petrography, geochemistry and geochronology of some samples from the Karakorum axial batholi th (northern Pakistan). In Grani tes of Himalayas Karakorum and Hindu Kush, F .A. Shams ed., Inst. of Geol., Punjab Univ., Lahore, Pakistan, 377-387. Le Fort P., Pecher A. & Upreti B.N. 1986 - A section through the Tibetan Slab in central Nepal (Kali Gandaki valley): mineral chemistry and thermobarometry. In Evolution des domaines orogeniques d'Asie meridionale (de la Turquie a l'Indonesie), P. Le Fort, M. Col chen & C. Montenat eds, Sci. de la Terre, Nancy, mem.47, 211-228. 378 Le Fort P., Pecher A. & Vidal P. 1982 - Les gneiss oeilles de la DaIle du Tibet: un episode magmatique au Paleozoique inferieur en Himalaya du Nepal. 96 Reun. Annu. Sci. de la Terre, Paris, Soc. geol. Fr., p.369 (resume). Li J.J., Li B.Y., Wang F .B., Zhang Q.S., Wen S.X. & Zheng B.X. 1981 - The process of the uplift of the Qinghai-Xizang plateau. In Geological and ecological studies of Qinghai- Xizang plateau, Sci. press, Beijing, 1, 111-118. Lyon-Caen H. & Molnar P. 1983 - Constraints on the structure of the Himalaya from an analysis of gravity anomalies and a flexural model of the lithosphere. J. Geophys. Res., 88, 8171-8191. Lyon-Caen H. & Molnar P. 1985 - Gravity anomalies, flexure of the Indian plate, and the structure, support and evolution of the Himalaya and Ganga basin. Tectonics, 4, 6, 513-538. Malinconico, Jr. L.L. 1986 - The structure of the Kohistan-Arc terrane in Northern Pakistan as inferred from gravity data. Tectonophys., 124, 297-307. Mallik T.K. 1978 - Ganges versus Indus cone - a comparative study. J. Geol. Soc. India, 19, 9, 395-402. Maluski H. & Matte P. 1984 - Ages of 39pin~otectonometamorphic events in the northwestern Himalaya (northern Pakistan) by Ar~9 Ar4~ethod. Tectonics, 3, 1, 1-18. Maluski H., Proust F. & Xiao X.C. 1982 - Ar/ Ar dating of the trans-Himalayan calc- alkaline magmatism of southern Tibet. Nature, 298, 152-154. Mankinen LA. & Dalrymple G.B. 1979 - Revised geomagnetic polarity time scale for the interval 0 - 5 my B.P. J. geophys. res., 84, p.615. Marcoux J., Wever P. de, Nicolas A., Girardeau J., Xiao X.C., Chang C.F., Cao Y.G., Wang N.W., Bassoullet J.P., Colchen M., & Mascle G. 1981 - Preliminary report of depositional sediments on top of volcanic member: Xigaze ophiolite (Yarlung Zangbo suture zone), south Xizang (Tibet). Ofioliti, suppl. vo1.6, 31-32. Marcoux J., Burg J.P., Giradeau J., Chang C.F. & Xiao X.C. 1984 - A key piece of the Yarlung -Tsangpo suture puzzle: the Xigaze group (s.l.). In Himalayan Geology into symp., Chengdu, vol.2 (abstract). Maruo Y. & Kizaki K. 1983 - Thermal structure in the nappes of the eastern Nepal Himalaya. In Granites of Himalayas Karako rum and Hindu-Kush, F .A. Shams ed., Inst. Geol., Punjab Univ., Lahore, Pakistan, 271-286. Mascle G.H. 1985 - L'Himalaya resulte-t-il du telescopage de trois chaines? Bull. Soc. geo1. Fr., (8) I, 3, 289-304. Mascle G. & Herail G. 1982 - Les Siwaliks: Ie prisme d'accretion tectonique associe a la subduction intracontinentale himalayenne. Geol. Alpine, 58, 95-103. Mascle G., Herail G., Van Haver T. & Delcaillau B. 1986 - Structure et evolution des bassins d' episuture et de perisuture lies a la chaine himalayenne. Bull. Cent. Rech. Explor.-Prod. Elf-Aquitaine, 10 (1), 181-203. Mehta P.K. 1980 - Tectonic significance of the young mineral dates and the rates of cooling and uplift in the Himalaya. Tectonophys., 62, 205-217. Mehta P.K. & Rex D.C. 1977 - K-Ar geochronology of the Kulu-Mandi belt, JIIN Himalaya, India. Neues Jahrb. Min., Mh., 343-355. Mercier J.L. et al. 1984 - La collision Inde-Asie cote Tibet. In Mission franco-chinoise au Tibet 1980, J.L. Mercier & G.C. Li eds, ed. Cent. Natl. Rech. Sci., 1-13. Meulebrouck van de J. 1983 - Reconnaissance geophysique des structures crustales de deux segments de chaine de collision: Ie Haut Allier (Massif Central fran~ais) et Ie Sud du Tibet (Himalaya). These Doct. 3e cycle, Univ. Montpellier, 137 p. Minster J.B. & Jordan T.H. 1978 - Present-day plate motion. J. Geophys. Res., 83, B 11, 5331-5354. 379 Misra R.C. & Valdiya K.S. 1961 - The calc-zone of Pithoragarh, with special reference to the occurence of stromatolites and the origin of magnesite. J. Geol. Soc. India, 2, 78-90. Molnar P. 1984 - Structure and tectonics of the Himalaya: constraints and implications of geophysical data. Annu. Rev. Earth Planet Sci., 12, 489-518. Molnar P. & Chen W.P. 1983 - Seismicity and mountain building. In Mountain building processes, K.J. Hs(j ed., Acad. press, London, 41-57. Molnar P. & Chen W.P. 1984 - S-P wave travel time residuals and lateral inhomogeneity in the mantle beneath Tibet and the Himalaya. J. Geophys. Res., 89, B 8, 6911-6917. Molnar P., Chen W.P., Fitch T.J., Tapponnier P., Warsi W.E.K. & Wu F .T. 1977 - Structure and tectonics of the Himalayas: a brief summary of relevant geophysical observations. Coll. into 268, Ecologie et geologie de l'Himalaya, Paris, ed. Cent. Natl. Rech. Sci., vol. Sci. de la Terre, 269-294. Molnar P., Fitch T .J. & Wu F.T. 1973 - Fault plane solutions of shallow earhquakes and contemporary tectonics in Asia. Earth Planet. Sci. Lett., 19, 101-112. Molnar P. & Tapponnier P. 1975 - Cenozoic tectonics of Asia: effects of a continental collision. Science, 189, 419-426. Molnar P. & Tapponnier P. 1978 - Active tectonics of Tibet. J. Geophys. Res., 11, B 83, 5361-5375. Montel J.M. 1986 - Experimental determination of the solubility of Ce-monazite in Si02- AI203-K 20-Na20 melts at 800°C, 2 kbar, under H20 saturated conditions. Geology, 14, 659-662. Montenat Ch., Girardeau J. & Marcoux J. 1986 - La ceinture ophiolitique neo-cimmerienne au Tibet, dans les Pamirs et en Afghanistan. Evolution geodynamique comparative. In Evolution des domaines orogeniques d'Asie meridionale, P. Le Fort, Ch. Montenat & M. Colchen eds, Mem. Sci. de la Terre, Nancy, 47, 229-252. Morgan W.J. 1983 - Hotspot tracks and the early rifting of the Atlantic. Tectonophys., 94, 123-139. Mu A.T., Wen S.H., Wang Y.K., Chang P.K. & Yin C.H. 1973 - Stratigraphy of the mount Jolmo Lungma region in southern Tibet, China. Scientia Sinica, 16, 1,96-111. Ni J. & Barazangi M. 1984 - Seismotectonics of the Himalayan Collision Zone: geometry of the underthrusting Indian plate beneath the Himalaya. J. Geophys. Res., 89, B 2, 1147- 1163. Ni J. & Barazangi M. 1985 - Active tectonics of the western Tethyan Himalaya above the underthrusting Indian plate: the Upper Sutlej River basin as a pull-apart structure. Tectonophys., 112, 277-295. Ni J. & York J.E. 1978 - Late Cenozoic tectonics of the Tibetan plateau. J. Geophys. Res., 83, B 11, 5377-5384. Nicolas A., Girardeau J., Marcoux J., Dupre B., Wang X., Cao Y. Zheng H. & Xiao X. - The Xigaze ophiolite (Tibet): a peculiar oceanic lithosphere. Nature, 294, 414-147. Odin G., Curry D., Gale N.H. & Kennedy W.J. 1982 - The Phanerozoic time scale in 1981. In Nummerical dating in stratigraphy, J. Wiley & sons, II, 957-960. Opdyke N.D., Johnson N.M., Johnson G.D., Lindsay E.H. & Tahirkheli R.A.K. 1982 - Paleomagnetism of the middle Siwalik fo rmations of northern Pakistan and rotation of the Salt range decollement. Palaeogeogr. Palaeoclimatol. Palaeoecol., 37, 1-15. Pant C.C. 1985 - Stromatolites from Middle-Riphean Gangolihat dolomites (Deoban), Berinag- Gangolihat area, Pithoragarh district, Kumaun Lesser Himalaya. J. Geol. Soc. India, 26, 1, 1-9. Pareek H.S. 1973 - Geological setting, petrography and petrochemistry of the Darla trap, and its comparative study with Mandi and Panjal traps. J. Geol. Soc. India, 14, 4, 355-368. 380 Patriat P. 1983 - These Doct., Univ. Paris VI. Patriat P. & Achache J. 1984 - India-Eurasia collision chronology has implications for crustal shortening and driving mechanism of plates. Nature, 311, 615-621. Patriat P., Segoufin J., Schlich R., Goslin J., Auzende J.M., Beuzart P., Bonnin J. & Olivet J.L. 1982 - Les mouvements relatifs de l'Inde, de l'Afrique et de l'Eurasie. Bull. Soc. geol. Fr., (7) XXIV, 2, 363-373. Paul O.K., Chandy K.C., Bhalla J.N., Prasad N. & Sen Gupta N.R. 1982 - Geochronology and geochemistry of Lingtse gneiss, Darjeeling, Sikkim Himalayas. Ind. J. Earth Sci., 9, 11-17. Pecher A. 1975 - The Main Central Thrust of the Nepal Himalaya and the related metamorphism in the Modi-Khola cross-section (Annapurna Range). Himalayan Geol., 5, 115-132. Pecher A. 1977 - Geology of the Nepal Himalaya: deformation and petrography in the Main Central Thrust Zone. ColI. into 268, Ecologie et geologie de l' Himalaya, Paris, ed. Cent. Nat!. Rech. Sci., vol. Sci. de la Terre, 301-318. Pilcher A. 1978 - Deformations et metamorphisme associes a une zone de cisaillement. Exemple du grand chevauchement central himalayen (M.C.l.), transversale des Annapurnas et du Manaslu, Nepal. These Doct. es-Sci., Univ. Grenoble, 354 p. Pecher A. 1979 - Les inclusions fluides des quartz d'exsudation de la zone du M.C.I. himalayen au Nepal central: donnees sur la phase fluide dans une grande zone de cisaillement crustal. Bull. Mineral., 102, 537-554. Pilcher A. (in press) - Fluid inclusion barothermometry in the Higher Himalayan metamorphics: retromorphosis and density re-equilibration. Fluid inclusion seminar, Wadia Inst. Himalayan Geol., Dehra-Dun, India, 1984. Pecher A., Bouchez J.L., Cuney M., Deniel C., France-Lanord C. & Le Fort P. 1984 - Himalaya central: une zone de cisaillement "haute temperature" longitudinale au sommet de la DaIle du Tibet. 10e Reun. Annu. Sci. de la Terre, Bordeaux, Soc. geol. Fr., p.436 (resume). Pecher A. & Le Fort P. 1986 - The metamorphism in central Himalaya, its relation with the thrust tectonic. In Evolution des domaines orogeniques d'Asie meridionale (de la Turquie a l'Indonesie), P. Le Fort, M. Colchen & C. Montenat eds, Sci. de la Terre, Nancy, mem.47, 285-309. Petterson M.G. & Windley B.F. 1985 - Rb-Sr dating of the Kohistan arc-batholith in the Trans-Himalaya of N. Pakistan, and tectonic implications. Earth Planet. Sci. Lett., 74, 45-57. Pilbeam D., Barry J., Meyer G.E., Shah S.M.I., Pickford M.H.L., Bishop W.W., Thomas H. & Jacobs L.L. 1977a - Geology and paleontology of Neogene strata of Pakistan. Nature, 270, 684-689. Pilbeam D., Meyer G.E., Badgley C., Rose M.D., Pickford M.H.L., Behrensmeyer A. & Shah S.M.I. 1977b - New hominoid primates from the Siwaliks of Pakistan and their bearing on hominoid evolution. Nature, 270, 22/29, 689-695. Pilgrim G.E. 1913 - The correlation of the Siwaliks with mammal horizons of Europe. Rec. Geol. Surv. India, 43(4), 264-326. Pilgrim G.E. 1919 - Suggestions concerning the history of drainage of northern India. J. Asiat. Soc. Bengal N.S., 15, 9-15. Polino R. 1981 - Geological map of the upper Imja Khola (Mount Everest region - eastern Nepal). Scale 1:25.000. S.E.L.C.A., Firenze. Powell C. Mc A. 1979 - A speculative history of Pakistan and surroundings: some constaints from the Indian ocean. In Geodynamics of Pakistan, A. Farah & K. de Jong eds, Geol. Surv. Pakistan, Quetta, 5-24. 381 Powell C. Mc A., Conaghan P.J. 1973a - Polyphase deformation in Phanerozoic rocks of the Central Himalayan Gneiss. N.W. India. J. Geol., 81, 2, 127-143. Powell C. Mc A. &. Conaghan P.J. 1973b - Plate tectonics and the Himalayas. Earth Planet. Sci. Lett., 20, 1-12. Pozzi J.P., Westphal M., Zhou Y.X., Xing L.S. &. Chen X.Y. 1982 - Position of the Lhasa block, South Tibet, during the late Cretaceous. Nature, 297, 27 May 1982, 319-321. Pozzi J.P., Westphal M., Girardeau J., Besse J., Zhou Y.X., Chen X.Y. &. Xing L.S. 1984 - Paleomagnetism of the Xigaze ophiolite and flysch (Yarlung Zangbo suture zone, southern Tibet): latitude and direction of spreading. Earth Planet. Sci. Lett., 70, 383-394. Pudsey C.J. 1986 - The northern suture, Pakistan: margin of a Cretaceous island arc. Geol. Mag., 123 (4), 405-423. Qureshy M.N. 1969 - Thickening of a basalt layer as a possible cause for the uplift of the Himalayas: a suggestion based on gravity data. Tectonophys., 7(2), 137-157. Raha P.K., Chandy M.C. &. Balasubrahmanyan 1978 - Geochronology of the Jammu limestone, Udhampur District, Jammu &. Kashmir State, India. J. Geol. Soc. India, 19, 5, 221-223. Rai H. 1983 - Geology of the Nubra valley and its significance on the evolution of the Ladakh Himalaya. In Geology of Indus suture zone of Ladakh, V.C. Thaku r &. K.K. Sharma eds, Wadia Inst. of Himalayan Geol., Dehra Dun (India), 79-97. Raju A.T.R. 1967 - Observations on the petrography of Tertiary clastic sediments of the Himalayan foothills of north India. Bull. Oil Nat. Gas Comm., 4, 5-16. Raju B.N.V., Chabria T., Prasad R.N., Mahadevan T .M. & Bhalla N.S. 1985 - Early Proterozoic Rb-Sr isochron age for central crystalline rocks Bhilangana valley, Garhwal Himalaya. Himalayan Geo 1., 12, 196-205. Ray S. 1947 - Zonal metamorphism in the eastern Himalaya and some aspects of local geology. Quart. J. Geol. Min. Met. Soc. India, 19, 4, 117-140. Raynolds R.G.H 1981 - Did the ancestral Indus flow into the Ganges drainage? Geol. Bull. Univ. Peshawar, 14, 141-150. Raynolds R.G.H. & Johnson G.D. 1985 - Rates of Neogene depositional and deformational processe s, nor th-west Himalayan for edeep mar gin, Paki stan. The Chr onol ogy of the Geological Record, N.J. Snelling ed., Geol. Soc. London, Mem. 10, 297-311. Raynolds R.G.H., Johnson G.D., Johnson N.M. & Opdyke N.D. 1980 - The Siwalik molasse, a sedimentary record of orogeny. Proc. into Comm. Geodyn. Peshawar, 1979, R.A.K. Tahirkheli, M.Q. Jan & M. Majid eds, Geol. Bull. Univ. Peshawar, spec. issue, 13, 47-50. Read P.B. & Brown R.L. 1981 - Columbia river fault zone, southeastern margin of the Shuswap and Monashee complexes, southern British Columbia. Canadian J. Earth Sci., 18, 1127-1145. Reibel G. & Reuber I. 1982 - La klippe ophiolitique de Spongtang-Photaksar (Himalaya de Ladakh); une ophiolite sans cumulats. C.R. Acad. Sci., 294, 557-562. Remy J.M. 1972 - Resultats de l'etude geologique de l'Ouest du Nepal (Himalaya). II. Les series de la nappe du Nepal, de la zone des ecailles et de la zone tibetaine. C.R. Acad. Sci., Paris, 0, 275, 2459-2462. Remy J.M. 1974 - Le metamorphisme et ses divers types dans l'Ouest du Nepal. C.R. Acad. Sci., Paris, 0, 279, 461-464. Reuber I. 1986 - Geometry of accretion and oceanic thrusting of the Spongtang ophiolite, Ladakh-Himalaya. Nature. Reynolds P.H., Brookfield M.E. &. Mcnutt R.H. On press) - The age and nature of Mesozoic- Tertiary magmatism across the Indus Suture Zone in Kashmir and Ladakh (N.W. India and Pakistan). Geol. Rundsch. 382 Richardson S.W., Gilbert M.C. & Bell P.M. 1969 - Experimental determination of kyanite- andalusite and andalusite-sillimanite equilibria; the aluminium silicate triple point. Am. J. Sci., 257, 259-272. Rothlisberger F. & Geyh M.A. 1985 - Glacier variations in Himalayas and Karakorum. Zeitsch. Gletzcherkunde und Glacialgeol., 21, 237-249. Rupke J. 1974 - Stratigraphy and structural evolution of the Kumaon Lesser Himalaya. SedimeM. Geol., 11, 2, spec. issue, 2 parts, 81-265. Sahni A., Bhatia S.B., Hartenbergr J.L., Jaeger J.J., Kumar K., Sudre J. & Vianey-Liaud M. 1981 - Vertebrates from the Subathu formation and comments on the biogeography of Indian subcontinent during early Paleogene. Bul. Soc. geol. Fr., (7) XXIII, 689-695. Sahni M.R. & Mathur L.P. 1964 - Stratigraphy of the Siwalik group. 22nd into geol. Congr., B.C. Roy ed., New-Delhi, India, 1.23. Saini H.S., Waraich R.S., Malhotra N.K., Nagpaul K.K. & Sharma K.K. 1982 - Cooling and uplift rates and dating of thrusts from Kinnaur, Himachal Himalaya. Himalayan Geol., 9, 2, 549-657. Sakai H. 1983 - Geology of the Tansen group of the Lesser Himalaya in Nepal. Mem. Fac. ScL, Kyushu Univ., ser D, Geol., XXV, 1, 27-74. Sauniac S. & Touret J. 1983 - Petrology and fluid inclusions of a quartz-kyani te segregation in the main thrust zone of the Himalaya. Lithos, 16, 35-45. Saxena M.N. 1981 - Chronology of tectonic events of the Himalaya. In Metamorphic tectonites of the Himalaya, P.S. Saklani ed., Today & Tomorrow print. & publ., New Delhi, 303-348. 2 0 Scharer U. 1984 - The effect of initial 3 Th disequilibrium on young U-Pb ages: the Makalu case, Himalaya. Earth Planet. Sci. Lett., 67, 191-204. Scharer U., Hamet J. & Allegre C.J. 1984a - The Transhimalaya (Gangdese) plutonism in the Ladakh region: a U-Pb and Rb-Sr study. Earth Planet. Sci. Lett., 67, 327-339. Scharer U., Xu R.H. & Allegre C.J. 1984b - U-Pb geochronology of Gangdese (Transhimalaya) plutonism in the Lhasa-Xigaze region, Tibet. Earth Planet. Sci. Lett., 69, 311-320. Scharer U., Xu R.H. & Allegre C.J. 1986 - U-(Th)-Pb systematics and age of Himalayan leucogranites, south Tibet. Earth Planet. Sci. Lett., 77, 35-48. Schlich R. 1975 - Structure et age de l'ocean indien occidental. Soc. geol. Fr., Mem. h.ser., 6, 103 p. Schlich R. 1982 - The Indian ocean: aseismic ridges, spreading centers and oceanic basins. In The oceans basins and margins, 6, A.E.M. Nairn & F .G. Stehli eds, Plenum Press, New-York, 51-147. Searle M.P. 1983 - Stratigraphy, structure and evolution of the Tibetan-Tethys zone in Zanskar and the Indus suture zone in the Ladakh Himalaya. Trans. Royal Soc. Edinburgh: Earth Sci., 73, 205-219. Seeber L., Armbruster J.G. & Quittmeyer R.C. 1981 - Seismicity and continental subduction in the Himalayan arc. In Zagros, Hindu-Kush, Himalaya geodynamic evolution, H.K. Gupta & F .M. Delany eds, Geodyn. Ser. vol.3, Am. geophys. Union, Washington, 215-242. Seeber L. & Jacob K.H. 1977 - Microearthquake survey of northern Pakistan: preliminary results and tectonic implications. ColI. into 268, Ecologie et geologie de 1 'Himalaya, Paris, ed. Cent. Natl. Rech. Sci., vol. Sci. de la Terre, 347-360. Sen ~ 1986 - Contribution i la magnetostratigraphie et i la paleontologie des formations continentales neogenes du pourtour mediterraneen. Implications biochronologiques et pa16obiologiques. These Doct. Etat, Univ. Pierre et Marie Curie, Paris. Seng~r A.M.C. 1979 - Mid-Mesozoic closure of Permo-Triassic Tethys and its implications. Nature, 279, 590-593. 383 Senglir A.M.C. 1981 - The evolution of Palaeo-Tethys in the Tibetan segment of the Alpides. In Geological and ecological studies of Qinghai-Xizang plateau, 1, Sci. Press, Beijing, 51-56. Senglir A.M.C. 1984 - The Cimmeride orogenic system and the tectonics of Eurasia. Geol. Soc. America, spec. paper, 195, 82 p. Senglir A.M.C. 1985 - The story of Tethys: how many wives did Okeanos have? Episodes, 8 (1), 3-12. Shah S.K., Sharma T.R. & Gupta K.R. 1978 - Trilobite trace fossils from the Bafliaz formation, western Pir Panjal, and their significance. J. Geol. Soc. India, 19 (6), 273-276. Shams f .A. 1972 - Glaucophane-bearing rocks from near Topsin, Swat. first record from Pakistan. Pakistan J. Sci. Res., 24 (3 & 4), 343-345. Shams f.A., Jones G.C. & Kempe D.R.C. 1980 - Blueschists from Topsin, Swat District, NW Pakistan. Mineral. Mag., 43, 941-942. Sharma K.K. 1983 - Transhimalayan granitoids of Ladakh and Karakorum range, Ladakh Himalaya. In Granites of Himalayas, Karakorum and Hindu-Kush, f .A. Shams ed., Punjab Univ., Lahore, Pakistan, 341-354. Sharma K.K. 1987 - Crustal growth and two-stage India-Eurasia collision in Ladakh. Tectonophys., 134, n01-3, 17-28. Sharma K.K., Saini H.S. & Nagpaul K.K. 1978a - fission track annealing, ages of apatite from Mandi granite and their applications to tectonic problems. Himalayan Geol., 8, p.296. Sharma K.K., Sinha A.K., Bagdasarian G.P. & Gukasain R.Ch. 1978b - Potassium-argon dating of Dras volcanics, Shyok volcanics and Ladakh granite, Ladakh, northwest Himalaya. Himalayan Geol., 8, 288-295. Sharma M.L. & Shah S.K. 1983 - Structure and tectonics of the area between Kargil and Bodhkharbu, western Ladakh, Kashmir Himalaya. In Contributions to Himalayan geology, V.J. Gupta ed., Hindust. publ. Corp., Delhi, 2, 143-157. Sheppard S.M.f. 1986 - Igneous rocks: III. Isotopic case studies of magmatism in Africa, Eurasia and oceanic islands. In Stable isotopes, J.W. Valley, H.P. Taylor & J.R. O'Neil eds, reviews in mineralogy vol.16, Mineral. Soc. Am., 319-371. Sheppard S.M.f., Debon f., Liu G., Jin C. & Xu R. 1983 - Oxygen isotope relationships in three plutonic belts in southern Tibet. Terra Cognita, 3, p.272. Singh T., Srivastava R.A.K. & Kumar S. 1982 - Sedimentology of the Siwalik sandstones of Subansiri district, Arunachal Pradesh. Himalayan Geol., 9, 2, 573-588. Sinha A.K. 1980 - Geology and tectonics of the part of Higher central Himalayan Tethyan zone. Proc. Int. Comm. Geodyn., spec. isue, Geol. Bull. Univ. Peshawar, Pakistan, 13, 29-37. Sinha A.K. 1981 - Geology and tectonics of the Himalayan region of Ladakh, Himachal, Garhwal-Kumaun and Arunachal Pradesh: a review. In Zagros, Hindu-Kush, Himalaya geodynamic evolution, H.K. Gupta & f .M. Delany eds, Geodyn. Ser., vol.3, Am. geophys. Union, Washington, 122-148. Sinha R.S. 1979 - Deformation sequence and strain estimate of the Himalayan nappes; example from Sikkim, eastern Himalaya. In Structural geology of the Himalaya, P.S. Saklani ed., Today & Tomorrow's print. & publ., 285-306. Sinha-Roy S. 1981 - Metamorphic facies and inverted metamorphic sequences of the eastern Himalayan crystalline rocks. In Metamorphic tectonites of the Himalaya, P.S. Saklani ed., Today & Tomorrow's print. & publ., New-Delhi, 279-302. Sinha-Roy S. & Sen Gupta S. 1986 - Precambrian deformed granites and possible basement in the Himalayas. Precambrian Res., 31, 209-235. 384 Sinvhal H., Agrawal P.N., King G.C.P. & Gaur V.K. 1973 - Interpretation of measured movement at a Himalayan (Nahan) Thrust. Geophys. J. R. astr. Soc., 34, 203-210. Srikantia S.V. 1977 - Sedimentary cycles in the Himalaya and their significance on the orogenic evolution of the mountain belt. Colloq. into 268, Ecologie et geologie de 1 'Himalaya, Paris, ed. Cent. Natl. Rech. Sci., vol. Sci. de la Terre, 395-407. Srimal N. 1986 - India-Asia collision: implications from the geology of the eastern Karakoram. Geology, 14, 523-527. Srivastava V.K. & Casshyap S.M. 1983 - Evolution of Pre-Siwalik Tertiary basin of Himachal Himalaya. J. geol. Soc. India, 24, 3, 134-147. Stille ~ 1919 - Die Begriffe Orogenese und Epirogenese. Zeitschr. Dtsch. Geol. Gesell., 71, 164-208. Stlicklin J. 1977 - Structural correlation of the Alpine ranges between Iran and central Asia. Mem. h. ser. Soc. geol. Fr., 8, 333-353. Stlic~' in J. 1980 - Geology of Nepal and its regional frame. J. Geol. Soc., 137, part 1, 1.34. Stlicklin J. 1986 - The Vendian - lower Cambrian salt basins of Iran, Oman and Pakistan: stratigraphy, correlations, paleogeography. In Orogenic evolution of southern Asia (from Turkey to Indonesia), P. Le Fort, M. Colchen & C. Montenat eds, Sci. de la Terre, Nancy, mem.47, 329-345. Stlicklin J. & Bhattarai K.D. 1980 - Geological map of Kathmandu area and central Mahabharat range (1:250.000). H.M. Government of Nepal, Dept. of Min. and Geol. Stlicklin J. & Bhattarai K.D. 1982 - Photogeological map of part of central Nepal, 1:100.000e. Dept. Min. Geol., Kathmandu, Nepal, Tehran Naqsheh Offset Press, Iran. Tahirkheli R.A.K. 1979 - Geology of Kohistan and adjoining Eurasian and Indo-Pakistan continents, Pakistan. Geol. BUll. Univ. Peshawar, 11, 1-30. Tahirkheli R.A.K. & Jan M.O. (eds) 1979 - Geology of Kohistan, Karakoram Himalaya, northern Pakistan. Geol. Bull., Univ. Peshawar, Pakistan, 11, 187 p. Tahirkheli R.A.K., Mattauer M., Proust F. & Tapponnier P. 1979 - The India Eurasia Suture Zone in northern Pakistan: synthesis and interpretation of recent data at plate scale. In Geodynamics of Pakistan, A. Farah & K.A. DeJong eds, Geol. Surv. Pakistan, Ouetta, 125-130. Talalov V.A. 1972 - Geology and ores of Nepal. H.M. Geol. Surv. Nepal, Kathmandu, 4 volumes. Tapponnier P. & Molnar P. 1976 - Slip-line field theory and large scale continental tectonics. Nature, 264, 319-324. Tapponnnier P. et al. 1981 - The Tibetan side of the India-Eurasia collision. Nature, 294, 405-410. Tapponnier P., Peltzer G. & Armijo R. 1986 - On the mechanics of the collision between India and Asia. In Collision tectonics, M.P. Coward & A.C. Ries eds, Geol. Soc. London, spec. publ., 19, 115-157. Tassy P. 1983 - Les elephantoidae miocines du plateau du Potwar, groupe de Siwalik, Pakistan. Ann. Paleont., 69 (2 a 4), 99-136, 235-297, 317-354. Tauxe L., Kent D.V. & Opdyke N.D. 1980 - Magnetic components contributing to the NRM of middle Siwaliks red beds. Earth Planet. Sci. Lett., 47, 279-284. Tauxe L. & Opdyke N.D. 1982 - A time framework based on magnetostratigraphy for the Siwalik sediments of the Khaur area, northern Pakistan. Palaeogeogr. Palaeoclimatol. Palaeoecol., 37, 43-61. Thakur V.C. 1981 - Regional framework and geodynamic evolution of the Indus-Tsangpo suture zone in the Ladakh Himalayas. Trans. Royal Soc. Edinburgh, Earth Sci., 72, 89-97. 385 Thakur V.C. 1983 - Palaeotectonic evolution of Indus-Tsangpo suture zone in Ladakh and southern Tibet. In Geology of Indus suture zone of Ladakh, V.C. Thakur & K.K. Sharma eds, Wadia Inst. of Himalayan geol., Dehra Dun (India), 195-204. Thakur V.C. 1986 - Tectonic zonation and regional framework of eastern Himalaya. In Orogenic evolution of southern Asia (from Turkey to Indonesia), P. Le fort, M. Colchen & C. Montenat eds, Sci. de la Terre, Nancy, mem.47, 347-360. Thomas H. 1983 - Les bovides anH-hipparions des Siwaliks interieurs (plateau du Potwar, Pakistan). Mem. Soc. geol. fr., n. sere n0145, 64 p. Trivedi J.R., Gopalan K. & Valdiya K.S. 1984 - Rb-Sr ages of granitic rocks within the lesser Himalayan nappes, Kumaun, India. Geol. Soc. India, 25, 10, 641-654. Tu G.C., Zhang Y.Q., Zhad Z.H. & Wang Z.G. 1981 - Characteristics and evolution of granitoids of southern Xizang. In Geological and ecological studies of Qinghai-Xizang plateau, Sci. press, Beijing, 1, 353-361. Valdiya K .S. 1969 - Stromatolites of the Lesser Himalayan carbonate formations and the Vindhyans. J. Geol. Soc. India, 10, 1-25. Valdiya K.S. 1980 - The two intracrustal boundary thrusts of the Himalaya. Tectonophys., 66, 323-348. Valdiya K .S. 1981a - Geological map of Kumaun Lesser Himalaya. In Zagros, Hindu Kush, Himalaya Geodynamic Evolution, Geodyn. Ser., vo1.3, Am. Geophys. Union, Washington D.C. Valdiya K .S. 1981b - Tectonics of the central sector of the Himalaya. In Zagros, Hindu Kush, Himalaya Geodynamic Evolution, Geodyn. Ser., vol.3, Am. Geophys. Union, Washington D.C., 87-110. Valdiya K .S. 1986 - Correlation of Lesser Himalayan formations of Nepal and Kumaon. In Orogenic evolution of southern Asia (from Turkey to Indonesia), P. Le fort, M. Colchen & C. Montenat eds, Sci. de la Terre, Nancy, mem.47, 361-383. Valdiya K .S., Joshi D.O., Sanwal R. & Tandon S.K. 1984 - Geomorphologic development across the active Main Boundary Thrust, an example from the Nainital hills in Kumaun Himalaya. J. Geol. Soc. India, 25, 12, 761-774. Van Haver Th. 1984 - Etude stratigraphique, sedimentologique et structurale d'un bassin d'avant arc: exemple du bassin de l'Indus, Ladakh, Himalaya. These 3e cycle, Univ. Grenoble, 204 p. Verma R.K. 1985 - Gravity field, seismicity and tectonics of the Indian peninsula and the Himalayas. D. Reidel Publ. Co., Dordrecht, Boston, Lancaster. Vidal Ph. 1978 - Rb-Sr systematics ~~ grHgite from central Nepal (Manaslu): Significance of the 01 igocene age and high Sr / Sr ratio in Himalayan orogeny: Comment. Geology, 6 (4), p.196. -- Vidal Ph., Bernard-Griffiths J., Cocherie A., Le fort P., Peucat J.J. & Sheppard S.M.f. 1984 - Geochemical comparison between Himalayan and Hercynian leucogranites. Phys. Earth & Planet. inter., spec. issue, 35, 179-190. Vidal Ph., Cocherie A. & Le fort P. 1982 - Geochemical investigations of the origin of the Manaslu leucogranite (Himalaya, Nepal). Geochim. Cosmochim. Acta, 46, 2279-2292. Virdi N.S. 1981 - Chail metamorphics of the Himachal Lesser Himalaya. In Metamorphic tectonites of the Himalaya, P.S. Saklani ed., Today & Tomorrow's print. & publ., New Delhi, 89-100. Virdi N.S., Thakur V.C. & Kumar S. 1977 - Blueschist facies metamorphism from the Indus suture zone of Ladakh and its significance. Himalayan Geol., 7, 479-482. Wadia D.N. 1975 - Geology of India. Tate Mc Graw-Hill publ. Co., New Delhi, fourth ed., 508 p. Wager L.R. 1965 - Injected granite sheets of the Rongbuk valley and the north face of Mt Everest. In D.N. Wadia commem. vol., Min. Geol. and Metall. Inst. India, 358-379. 386 Wang S. 1984 - Caractiristiques de la sirie volcani que de Linzizong dans Ie secteur oriental de l'arc volcanique de Gangdise. In Mission franco-chinoise au Tibet, 1980, J.L. Mercier & G.C. Li ids, id. Cent. Natl. Rech. Sci., Paris, 319-333. Warsi W.E.K. & Molnar P. 1977 - Gravity anomalies and plate tectonics in the Himalaya. CoIl. into 268, Ecologie et giologie de I'Himalaya, Paris, id. Cent. Natl. Rech. Sci., vol. Sci. de la Terre, 463-478. Watson E.B. & Harrison T.M. 1983 - Zr saturation revisited. Temperature and composition effects in a variety of crustal magma types. Earth Planet. Sci. Lett., 64, 295-304. West R.M. 1984 - Siwalik faunas from Nepal. Paleoecological and paleoclimatic implications. In The evolution of the east Asian environment, R.O. Whyte ed., Hong Kong, II, 724-744. West W.D. 1939 - Structure of the Shali window near Simla. Rec. Geol. Surv. India, 74, 1, 133-163. Westphal M., Pozzi J.P. & Zhou Y.X. 1982 - Motion in latitude of the Lhasa block during late Cretaceous - early Tertiary. EOS, 63, n045, p.1093. Xiao X.C. & Gao Y.L. 1984 - Tectonic evolution of the Tethys-Himalayas of China. In Tectonics of Asia, Colloq. n05, 27th into geol. Congr., Moscow, reports, 5, 181-189. Xu R. 1981 - Vegetational changes in the past and the uplift of Qinghai-Xizang plateau. In Geological and ecological studies of Qinghai-Xizang plateau, Sci. Press, Beijing, 1, 139-144 • Xu R.H., Scharer U. & Allegre C.J. 1985 - Magmatism and metamorphism in the Lhasa block (Tibet): a geochronological study. J. Geol., 93, 41-57. Yamanaka H., Yoshida M. & Arita K. 1982 - Terrace landform and Quaternary deposit around Pokhara valley, central Nepal. J. Nepal Geol. Soc., spec. isue, 2, 113-142. Yoshida M. & Igarashi Y. 1984 - Neogene to Quaternary lacustrine sediments in the Kathmandu valley, Nepal. J. Nepal Geol. Soc., spec. issue, 4, 73-100. Yoshida M., Igarashi Y., Arita K., Hayashi D. & Sharma T. 1984 - Magnetostratigraphic and pollen analytic studies of the Takman series, Nepal Himalayas. J. Nepal Geol. Soc., spec. issue, 4, 101-120. Zeitler P.K. 1985 - Cooling history of the NW Himalaya, Pakistan. Tectonics, 4, 1, 127- 151. Zeitler P.K., Tahirkheli R.A.K., Naeser C.W., Johnson N.M. & Lyons J. 1980 - Preliminary fission track ages from the Swat valley, northern Pakistan. Geol. Bull. Univ. Peshawar, 13, 63-65. Zeitler P.K., Tahirkheli R.A.K., Naeser C.W. & Johnson N.M. 1982 - Unroofing history of a suture zone in the Himalaya of Pakistan by means of fission-track annealing ages. Earth Planet. Sci. Lett., 57, 227-240. Zeitler P.K., Sutter J.F., Williams loS., Zartman R.E. & Tahirkheli R.A.K. 1986 - The Nanga Parbat-Haramosh massif, Pakistan: geochronology and cooling history. Geol. Soc. Am., abstr. with program, p.800. Zhang Qi, Li D.l. & Li S.H. 1981 - A pair of metamorphic belts in south Xizang. In Geological and ecological stUdies of Qinghai-Xizang plateau. Sci. Press, Beijing, 1, 443-450. Zhou Z.W. & Teng J.W. 1984 - Preuves paliomagnetiques de la derive vers Ie Nord de fragments de la plaque indienne apres la separation du continent de Gondwana et de leur collision avec la plaque eurasiatique. In Mission franco-chinoise au Tibet 1980, J.L. Mercier & Li G.C. eds, Ed. Cent. Natl. Rech. Sci., 15-20. CRUSTAL SCALE THRUSTING AND CONTINENTAL SUBDUCTION DURING HIMALAYAN COLLISION TECTONICS ON THE NW INDIAN PLATE Robert W.H. Butlerl & M.P. Coward2 lDepartment of Geological Sciences, University of Durham, South Road, Durham DHl 3LE, U.K. Department of Geology, Royal School of Mines, Imperial College, Prince Consort Road, London SW7 2BP. ABSTRACT. Following the subduction of Tethyan oceanic litho- sphere beneath Asia and Kohistan the continued convergence between the upper plate and the Indian continent led to thrust stacking of Indian crust to form the Himalayas. This lasted from Oligocene to Recent times and in an attempt to evaluate displacements, a series of balanced cross-sections have been constructed across the belt. In Pakistan these illustrate that over 600km of relative convergence between India and the Kohistan complex north of the Eocene suture zone has occurred by SSE- directed thrusting. This deformation only involves Indian upper crust at present outcrop levels so that the lower crust and remaining lithosphere must have been subducted beneath Kohistan and Tibet. The northern edge of the Indian lower crust may lie beneath the Pamirs. Similarly large amounts of shortening {several hundred kilometres} are implied by other balanced crustal sections through the central Himalayas and western Pakistan. The continuity of thrust systems around the NW margin of the Indian continent is proposed so that thrusts which stack continental crust step off into oceanic lithosphere in the west. This thrusting mechanism accounts for a substantial fraction of the total, l200-2000km relative convergence between stable India and Asia. Further shortening in the Tibetan region which developed after the Eocene continent-continent collision must be added to the displacements on thrusts which stack Indian lithosphere. Deformation within the entire collision zone approximates more closely to an essentially vertical plane strain model rather than to a process of lateral expulsion of Tibet towards the east. 1. INTRODUCTION Arguably the most spectacular example of crustal shortening developed after the collision between two continental masses is the Himalayas (Fig. 1). Since the suturing in the mid-late 387 A. M. C. $,engor (ed.), Tectonic Evolution of the Tethyan Region, 387-413. © 1989 by Kluwer Academic Publishers. 388 Fig. 1 Sketch map of the Himalayan region, the stacked Indian continental material is stippled and the adjacent foreland basin and other regions of Quaternary continental deposition are depicted by open circle ornament. Major suture zones, including the Indus-Tsangpo (ITSZ) and Anduo sutures (AS) are depicted by solid barbs on the hanging-wall block. Himalayan thrusts are depicted by solid barbs and include the Main Boundary (MBT) and Main Central Thrusts (MCT). The lines of cross-sections presented here are depicted R-R' (Fig. 2), S-S' (Fig. 5), T-T' (Fig. 8) and Q-Q' (Fig. 13). More detailed map areas are boxed, a (Fig. 3), b (Fig. 12). The sketch is compiled and modified from Gansser (28) and Mattauer (6). Eocene, between India and the collage of continental fragments and island arcs (1) which constitute the main mass of Asia, magnetic anomaly data suggest that approximately l200-2000km of NNW-SSE convergence has occurred. This must have been accommodated by substantial subduction of continental lithosphere but opinions differ as to whether this includes crustal material. In a model which essentially derives from Argand (2), Powell (3, 4) proposed that a thick wedge of Indian crust was subducted beneath Asia, the upper portion of which was scraped up to produce the main topographic elements of the Himalayas. This detachment model for crustal shortening may be contrasted with that proposed by Mattauer (5, 6) in which thrusts traverse the whole of the Indian crust. The basal detachment lies near the Moho so that Indian crust is contained to the south of the present outcrop of the suture. Many aspects of this 'crustal duplex' or 'whole crustal imbrication' model (7) are attractive. It explains the markedly heterogeneous deformation at outcrop (8) as being concentrated along thrust-sense shear zones, the mineral/stretching lineations approximating to the transport direction. The general forward, southward propagation of the developing thrust stack is further predicted from the model, a feature which is required to explain the patterns of regional metamorphism and foreland basin sedimentation. It also explains the recently imaged steps in the sub-Himalayan Moho (9) provided the basal detachment lies just within the Indian mantle. However, it is not possible for the crustal stacking geometry to accommodate the total 1200+ km convergence required by the ocean floor magnetic anomaly data (3) because there is insufficient crust south of the suture (Fig. 2). Thus substantial deformation (within the Asian plate) either by crustal thickening (10) or by lateral expulsion (11), is required to operate in conjunction with any whole crustal imbrication model. 389 A number of lines of evidence suggest that the Indian plate deformed with important mid-crustal detachments in the western Himalayas. Seeber and others (12, 13) and Ni and Barazangi (14) provide microseismicity data which are consistent with the detachment of upper from lower crust. The thrusts themselves can be traced back around fold generated re-entrants which suggest low angle trajectories (15) rather than the steeper fault vrofiles required by the whole crustal imbrication model. Furthermore, new work in this western sector in Pakistan suggests that thrust displacements and the inherent crustal shortening values are very large requiring substantial amounts of Indian lower crust to remain at depth (16). Since the crust directly beneath the thrust belt has not been greatly thickened, according to the gravity and topographic data (17) much of the required Indian lower crust must now lie beneath Asian plate rocks. Thus Argand's (2) fugh Himalayas 'LVZ' Tsangpo suture zone south Tibetan block fnd/on Anduo suture zone IlfhoSphere ~--------------- Indian plate thrust /' ~ belt detaches at Moho 100 k.m V : h N Fig. 2. A sketch crustal section through the central Himalayas (line R- R' on Fig. 1) illustrating the 'whole crustal imbrication' model of Himalayan thrustings suggested by Mattauer (6), from which source the diagram was produced. MBT - Main Boundary thrust, MCT - Main Central thrust. 390 implication of crustal subduction is fully supported in the Pakistan sector of the Himalayas. The plan of this contribution is to briefly outline the evidence for this assertion and then to suggest some implications for tectonic evolution within the western Himalayas and the NW margin of the Indian plate. 2. INDIAN PLATE THRUST SYSTEMS IN NORTHERN PAKISTAN 2.1 A cross-section from the Punjab foreland to the Main Mantle thrust. The thrust front in northern Pakistan lies in the Salt Ranges where cover rocks, from Cambrian sandstones through a sequence of Mesozoic and Cenozoic carbonates and shales, have been thrust onto Quaternary river gravels and fanglomerates (18). Farah et al. (19) presented gravity data which imply that the Indian shield continues undeformed beneath the [J'"'''''"', ~J [] Fig. 3. Sketch map of the Himalayan thrust belt in northern Pakistan, based on Coward and Butler (16), Tahirkheli and Burbank (53) and Gansser (28). S-S' - section line of Figs. 4 and 5, MMT - Main Mantle thrust, MBT - Main Boundary thrust, MCT - Main Central thrust, OS - Oghi Shear, NGT - Nathia Gali thrust. 391 thrust front so that basement-cover decoupling has occurred. Cover staking at outcrop must step back far to the north of the thrust front. In the eastern Salt Ranges this decoupling has occurred on a thick Cambrian evaporite formation (20) but further west, and around the adjacent Trans-Indus Range, the thrust carries a hanging-wall of Mesozoic sediments although borehole data indicate that the evaporites remain at depth (20). Yeats et al (18) describe very young deformation features in Quaternary river gravels suggesting that the thrust front is still active. The detachment surface north of the Salt Ranges, beneath the Potwar plateau, is however aseismic since the evaporite formations deform by dominantly ductile processes. This presumably allowed the decoupling zone to propagate freely, accounting for the broadly undeformed nature of the overlying central Potwar plateau and hence the outlying nature of the thrust front from the main Himalayan thrust stack. This major system developed in the Margalla Hills north of Rawalpindi has been carried onto Miocene molasse of the Murree Formation (Rawalpindi Group) along the 'Main Boundary thrust zone'. At outcrop this thrust carries a hanging-wall of Mesozoic-Eocene carbonates with Murree molasse: does it carry basement rocks to shallow depths too? To answer this question we must constrain the subsurface geometry of the Main Boundary thrust and this can be readily achieved by restoring the outlying thrust systems on which it has been carried. Since these outer systems do not involve basement rocks, the basement- cover contact should remain undisrupted beneath the Main Boundary thrust complex for a distance equal to the horizontal shortening on structures further south. We can estimate this value by using balanced cross- section methods (21, 22) which can also be used to test the assumption of no basement involvement in the southern thrust systems. To this end Figure 4 is a cross-section constructed through the thrust and fold belt to the south of Rawalpindi (the southern part of the section line of Fig.5), to the east of the Salt Ranges. Commercial seismic reflection data suggest that the basal detachment developed within the Neogene molasse sediments of the Siwalik group rather than along the basement- cover contact. This change in geometry along the thrust front possibly reflects the distribution of Cambrian evaporites, specifically their absence in the east (23). In this eastern region the deformation front is marked by weakly developed hinterland and foreland directed structures (Fig. 4). Towards Rawalpindi on this section line the older Murree molasse sediments are carried up on thrusts suggesting that the basal detachment cuts up from deeper, older stratigraphic levels in the north. The older shelf carbonates are predicted to be carried by thrusts at depth beneath the northern Potwar region; in the western Potwar region near Kohat (Fig. 3) Eocene rocks are certainly carried on the southern thrusts. Using these geometries the section of Figure 4 has been constructed to minimise displacements so that matching hanging- wall and footwall cut-offs of stratigraphic units against particular thrusts are sited as close as possible while still obtaining a restorable geometry. The restored section has no gaps or overlaps of units so that the deformed state section is geometrically viable (24). The outlying rocks display little penetrative strain, apart from a locally developed weak cleavage in the northern part of the section, so o a. ba la nc ed s e ct io n / / / ha ng ln 9- IIr ol l a n tlc lm e ~ 1 @ ; 1 ~ ~f OC en o< ~' H es oz ol C s r a m p I- -- -s h o rt e n tn g = 3 8 km - - - - - - - - - - i ha ng tn 9- ltI ol l ra m p 0 ' o pp os ed d ip c o m pl ex PI N t, p b. r e s to re d tem p~ '" / ~ . ,,}_- ;.~:: :~":' ::=:: ':-": :~::- ;_::. ~.::: ;:.:; =.;,; :.::; :;.:; ;;,;? ~~,:; ;.:-~ ".:~; .~"~: :;~.: :~ .. ~~ :~': ':l A - - f . ,. .. . o · . " !1 ur re e Fm n ~ I , ¥ . n . - - - H es oz ol cs 10 km F ig . 4. A b al an ce d a n d r e s to re d c r o s s - s e c ti o n c o u pl et c o n s tr u c te d p ar al le l to th e SS E th ru st tr a n sp o rt d ir ec ti o n , be tw ee n th e Is la m ab ad a re a a n d th e fo re la nd e a s t o f th e S al t R an ge s. Th e tw o m a rk er s in t he S iw al ik f or m at io n (u n- or na m en te d) c o rr e s po nd t o t w o pr om in en t r e fl ec to rs s e e n o n c o m m e rc ia l s e is m ic p ro fi le s. T he c r o s s - s e c ti on b al an ce s be ca us e th e p ro fi le r e s to re s w it ho ut h ol es o r o v e rl ap s a n d ha s a s e q u en ti al ly r e tr o de fo rm ab le g eo m et ry . W \0 tv 393 that the line length of the stratigraphy is conserved during deformation and that all the shortening is achieved by discrete dis- placements. By subtracting the deformed from the restored section length the shortening value of 38km is obtained. Note that this is a minimum estimate since some of the offsets of footwall and hanging-wall cut-offs are necessarily arbitrary although the seismic and surface geological data permit few other geometries within the requirement of achieving a restorable section. Thus any basement rocks within the Main Boundary thrust systems must have been carried at least 38km away from their original position and the basement-cover contact must remain near its regional elevation beneath the Margalla hills. The Main Boundary thrust cannot cut steeply back into basement rocks at depth directly beneath its present outcrop trace. The shallow trajectory on the Main Boundary thrust predicted by section restoration methods is supported to the east. Here this thrust has been folded around a large anticlinal structure (the Hazara syntaxis, see Fig. 3) so that its northern continuations are exposed. Yet its footwall always lies in Murree molasse, for a total distance of lOOkm back into the re-entrant, suggesting that the thrust systems in the Margalla and Hazara hills are detached at high levels (15). Models for the development of the Hazara syntaxis and similar structures in the region will be discussed later. Using the high level detachment geometry of the Main Boundary thrust zone discussed above, a regional cross-section has been constructed from the foreland in the Punjab to the Main Mantle thrust which carries the Kohistan island arc complex onto the Indian continent (25). The Kohistan complex itself was attached onto the Asian plate collage during the late Cretaceous, the last unit to do so before the final collision with India. The cross-section across the India plate thrust systems is presented here (Fig. 5; see ref. 16). It has been constructed parallel to the direction of thrust transport which is parallel to the relative plate movement vector between Asia and India (3) as confirmed by mineral/stretching lineations, sheath folds, slickensides, shear fibres and the geometry of thrust systems themselves. There is no evidence for a variation in this direction with time within the thrust system or for any strike-slip movements as predicted by Mattauer (6) further east. Thus all cross-sections drawn parallel to the movement direction should be restorable, provided they do not cross lateral tip zones or sites of markedly variable thrust stacking where the necessary assumption of plane strain could be invalid (26). The thrust systems within the Margalla-Hazara hills are far more complex than those developed further south. In the Rawalpindi area extensive inversion of structure may be related to northward directed backthrusts although these structures appear less significant towards the east. A continuous road section from the town of Murree, on the Main Boundary thrust, runs northward to join the Karakorum Highway so that a complete transect can be followed across the thrust belt. This type of continuity is necessary to avoid miscorrelating thrust sheets with the possible risk of inflating displacement estimates. In - - ::;;, c o s - O gt ", he ar - /'l am /'f an tle th ru st R~ "O lp ln '' '_ Is la m ab ad pl G /P O U Ce o". """ r.~- ~"O" ", [J" "" '"" ,'" ba s~ me '" [o sl er n S ai l {/O f!9 " F ig . 5 . R eg io n al c r o s s - s e c ti o n a c r o s s th e H im al ay an t h ru st b e lt . s e c ti o n l in e 8 -8 ' o n F ig s. 1 a n d 3 . a ft er C ow ar d a n d B u tl er (1 6) . V J '- D ~ a b ~r--4 traillflg ~ft~ .-/ bronch lines sfoclr.d Iratilflg branch 1,(1.5 395 Fig. 6. The development of steep fault profiles during thrusting by the juxtaposition of trailing branch line a - before displacement, b - final geometry; see ref. 34. the Murree area the thrust complex contains steeply dipping repetitions of Murree molasse and Eocene carbonates, often developed on the scale of a few metres. This implies detachment just within the Eocene rocks and the original presence of, but now eroded, stack of Murree and probably Siwalik molasse above. The steep dips can be readily generated by a combination of passive back-rotation within the imbricate stack and back thrusting from beneath (Figs. 5 and 6). Further north the imbricates contain thicker slices of Eocene and then Mesozoic carbonates suggesting that the imbricate thrusts cut up from deeper levels. It is likely that these lower parts of the stratigraphy now form a buried duplex beneath the Main Boundary thrust zone. Further north, Proterozoic rocks are brought up within imbricate slices above the Nathia Gali thrust (Figs. 3 and 5). These Hazara slates form part of the northern Indian crust basement complex; in the Mansehra area they are intruded by granitic bodies which have yielded Rb-Sr isotopic ages of 516 16 Ma (27). In this sense the Nathia Gali thrust is analogous, but not necessarily equivalent, to the Main Central thrust in the Indian and Nepalese sectors of the mountain belt (28, 29). The Hazara slates in the frontal imbricate slices have a cover sequence of Mesozoic through to Eocene limestones similar to those further south in the Margalla hills. The fault dips between imbricate slices are steep and locally southerly yet we do not propose that the thrusts either originated in this orientation or have been steepened by pure shear dominated north-south shortening. Bedding within imbricate slices is sub-parallel to the bounding faults and fossiliferous units within the Mesozoic limestones are only weakly strained. The most likely explanation is that each unit within the thrust stack has been progressively back steepened by the accretion of the next, more southerly imbricate slice (Fig.6). Indeed it is likely that this kind of steepening process is responsible for the majority of presently steep fault dips in orogenic belts. 396 The sub-surface geometry of the imbricate slices north of the Nathia Gali thrust can be predicted using the same reasoning outlined above to constrain the geometry of the Main Boundary thrust. Thus the Nathia Gali thrust cannot cut deeply into basement since the Mesozoic- Eocene-Murree Formation repetitions further south would not restore to lie on basement. Rather this thrust must flatten off at depth, the preferred geometry on Figure 4 has the Nathia Gali and Main Boundary thrusts splaying from a depth of about 20km. This level corresponds to the flexured regional elevation of the Indian plate basement-cover contact and does not imply mid-crustal detachment. Using this logic the repetitions of Hazara slates can be restored although the interpretation illustrated on Figure 5 is not unique but merely that which requires the least amount of orogenic contraction. Less than 10km of stratigraphic section is involved in each imbricate slice implying a mid-crustal detachment on the restored section. However, the present position of the footwall basement-cover cut-off must lie north of the Main Mantle thrust so that all the thrust structures which stack Indian crustal material at outcrop must flatten out at relatively shallow levels. In the Abbottabad area the Hazara slates have a Palaeozoic cover sequence of limestones and quartzites (30) which underlie the Mesozoic carbonates found in outcrops further south. The quartzites display rapid thickness variations. One interpretation might be that the Palaeozoic rocks were repeated by thrusts. However there are few recognizable discordances which would require 'flat-on-flat' imbrication with the hanging-wall and footwall ramps greatly offset on individual thrusts. Such a model implies very great values of orogenic shortening. The alternative adopted here interprets the thickness variations as pre- tectonic which would require syn-sedimentary faulting. One such fault is hypothetically illustrated on the section (Fig. 4). A major variation in structure occurs to the north of the Mansehra thrust. In this region the basement complex contains widespread lower Palaeozoic granitic material (27) and further north it is more gneissose, with strips of metasedimentary material (31). These units are interleaved by isoclinal folds and stacked by thrust-sense ductile shear zones. Major upright fold structures, possibly developed above thrust detachments at depth, deform this stack which, unlike the steeply dipping structures further south, has a recumbent attitude. These geometrics are illustrated on Figure 5 although the complex isoclinal folds are highly simplified here. The cover rocks contain metamorphic mineral assemblages of amphibolite facies (32), in contrast to the middle greenschist facies assemblages near Abbottabad, which suggest significant tectonic overburdens during thrusting. The most likely candidate for this is the Kohistan island arc complex, in the hanging- wall to the Main Mantle thrust. As with the Main Boundary thrust, the Main Mantle thrust can be traced back north of its principal outcrop trace, for over l80km around the Nanga Parbat syntaxial fold. Blueschists and ultramafics of the Kohistan complex also lie in the isolated Dargai klippe (33) in the Swat valley (Fig. 3), some 20km south of the Main Mantle thrust. These gross observations support the notion of substantial displacements on the Main Mantle thrust with the implication that very large amounts of Indian crust, at least a width of 397 250km, were effectively "subducted" beneath the Kohistan complex. The thrust structures which developed on the footwall to the Main Mantle thrust, at least to the North of Mansehra, probably approximate to a duplex (34). 2.2 Amount of crustal shortening in the Indian plate. Apart from testing the internal consistency of a particular structural interpretation, balanced cross-sections also provide estimates of orogenic contraction simply by subtracting the deformed from the restored section lengths (21, 22). Where several interpretations were possible the adopted solution on Figure 5 is that which implies the smallest value of orogenic contraction while still being geometrically restorable. Any values of shortening will therefore be minimum estimates. Figure 5 shows the footwall to the Mansehra thrust restoring to a width of 540km. Displacements on thrusts between this structure and the Main Mantle thrust restore to a further 190km. This second figure is likely to be a gross underestimate since there are widespread ductile strains and isoclinal folds which shorten the basement-cover contact in addition to thrust displacements. Thus, the total restoration of the basement-cover contact, from the pin line on the Punjabi foreland to the Main Mantle thrust, is 540 + 190 = 730km. The present width of the cross-section is just 260km so the shortening inherent in Figure 5 is 730-260 = 470km. 2.3 Deep structure of the Indian plate Most of the thrust structures described from the Main Mantle thrust to the Punjabi foreland only involve cover rocks. Precambrian basement is only included north of the Nathia Gali thrust and in these sheets the Hazara slates are only 6-l0km thick. Similarly the crystalline base- ment sheets north of Mansehra are relatively thin and certainly do not bring up lower crustal rocks to outcrop. It seems most likely then that the Indian plate thrust system in Pakistan developed above an intra- crustal detachment rather than along one at the base of the crust or within the mantle as proposed by Mattauer (1985) for the Central Himalayan region. The basal detachment level in Pakistan can be traced from its deepest level just 10km beneath the basement-cover contact to climb gradually to a basement-cover decoupling surface before climbing into Eocene and Neogene sediments. This kind of gently surface-seeking thrust profile is very common in orogenic belts (34, 35) as oppposed to the relatively steep fault trajectories (>300 ) which step rapidly across the crust (e.g. the Outer Isles 'thrust' imaged by the MOIST and WINCH seismic reflection profiles; ref. 36). Note that this steep kind of fault profile cannot be valid for the Pakistan Himalayas since the large displacements reported here would have brought up lower crust and mantle rocks of the Indian plate. The only such rocks which crop out within this segment of the mountain belt do so as part of the Kohistan complex which lies above the Main Mantle thrust. NN W P am ir s A si an - Tl bp fo n ~~1 SS E 5' K ar ak or um K oh /s ta n c o m pl u ~S' H A IN H IH A L A Y A S -- - FR O NT AL R A .N (j fS -- m t> lo m or ph Jc bo s. t> m £> nJ -c ov pr Im br ic at es Im br ic at es e o yp r- m ol as se Im br ": ol p5 fo rt i't on d ,~ ~ 0 -- -= _ - , < - ; ~ - . c uu s: ;;:g ~~~~ ~~F~ ~~~ - ~ 0 ~'- '~" -C_ V ~ " T E:2 l ., ., .. , ~ In dl on 10 0 km PIT D D Lo w er P al ae oz oI c bo se -m en ! g c r u s t s F ig . 7 . T ru e- sc al e c r u s ta l c r o s s - s e c ti o n f ro m t h e In d ia n f o re la n d to th e P am ir s il lu st ra ti n g t h e de ep s tr u c tu re o f th e H im al ay an t h ru st b e lt d et er m in ed u s in g t h e b al an ce d s e c ti o n o f F ig u re 5 . A - th ru st f ro n t, B - ra m p a c r o s s H az ar a s la te s, C - ra m p a c r o s s c r y st a ll in e b as em en t, D - ra m p a c r o s s In d ia n l ow er c r u s t; a ll t h e r e fe re n ce p o in ts w o u ld r e s to re to li e B ' o n B , C ' o n C a n d D ' o n D . N ot e th e r e g io n al , u pw ar dl y s te p p in g s ta ir c a se t ra je ct or y o f th e th ru st s y st em . M M T - M ai n M an tl e th ru st , KO B - K o h is ta n B a th o li th ( se e r e f. 3 1 ); K KB - K ar ak or um b a th o li th ( se e r e f. 3 1 ); NS - N o rt h er n s u tu re , r e th ru st b y th e la te t h ru st , H S - H un za s h ea r. v. > - 0 00 399 To achieve a balance between the lower and upper parts of the Indian crust there must be an equal length of Indian plate Moho to the overlying, and now greatly shortened, basement-cover contact. Certainly the Hazara slates and northern Indian plate gneisses did not directly overlie mantle rocks prior to Himalayan collision orogeny. There is insufficient space immediately beneath the thrust belt to include this lower crust which must therefore continue far beneath the Kohistan complex. Indeed, if this lower crustal wedge has not been stacked up by Himalayan compression its trailing edge would now lie beneath the Pamirs (Fig.7). Models which imply that Kohistan presently overlies its lithosphere root (33) are not valid. It remains to be seen whether the structure within the Kohistan complex developed by decoupling along a proto Main Mantle thrust or whether, as at present seems most likely, this thrust truncates the Kohistan complex at depth. The Main Mantle thrust must have overlain the northern part of the Indian plate thrust systems to provide a tectonic overburden to generate amphibolite facies metamorphism in cover rocks north of the Mansehra thrust and to emplace the Dargai Klippe (Fig.3). Figures 5 and 7 imply at least 100km displacement on the Main Mantle thrust before the Indian plate became imbricated. Note that this is a minimum estimate (37) since the now eroded hanging-wall sheets may have been re-shortened by breaching Indian plate thrusts and may have continued further to the south as a thinner sheet. Syn-displacement erosion may also have occurred to prevent footwall metamorphism. However, the minumum l50km displacement can be added to shortening within the Indian crust (almost 500km) to provide a total estimate of convergence between the Indian foreland and Kohistan. This value, 600+ km, does not include displace- ments within Kohistan or further north in the Pamirs which may have breached up through into the upper plate. 3. Lithospheric thrust geometry along the Himalayan chain. The main conclusion of the discussion above, that the main western Himalayan belt and Kohistan complex are underlain by a wedge of lower crust continuous with the Indian cratonic foreland, confirms the model of Argand (2) and Powell (3, 4) together with that arising from studies of seismicity (12, 13, 14). This geometry is in contrast to that proposed for the central Himalayas by Mattauer (6) and Burg and Chen (38) amongst others (see Fig.2). The plan now is to relate these different models by proposing a speculative three dimensional geometry of crustal scale thrusting along the mountain belt. An alternative crustal cross-section through the central Himalayas is presented here as Figure 8. Using the Moho geometry imaged by Hirn et al. (9) and surface geology (28, 29), the section balances upper and lower crustal segments. The total shortening of Indian plate material (distance Pin-G' subtracted from distance Pin-G) is 330 km of which 260 km has occurred by displacements on the Main Boundary thrust (MBT) and the frontal imbricates (distance Pin-F' subtracted from distance Pin-F). These values are greater than those implied by Figure 2 (5, 6) since there is more of the Indian crust subducted beneath Tibet. Note T - - - - . . ~.. ::: ::- . "" ~~ "' '' -f ~: !i::'H ;,.:': :i:~) a T' N S lJ fu r. 10 0 k, . J ' - >,.;; ::::: ·:·;: i·~»f .;;;, -;;;; :1f;: J;./; ;;:,, ;,j;; ;.:); '/!j8 !:;:; ,;;>/ C i,., :: ;; ~fi,: :0;,: '·.i: ,, :., .,;: .:.' " -.: .. .. - - - b F ig . 8. A g ro ss c r u s ta l c r o s s - s e c ti o n ( a) a n d it s r e s to re d te m pl at e, (b ) th ro ug h th e C en tr al H im al ay as a lo ng l in e T -T ' o n F ig . 1. Th e ge om et ry o f M oh o (t he M l in e) o ff se ts i s a ft er H im e t a l. (9 ). O nl y ba se m en t a re a ha s be en r e s to re d bu t im pl ie s 26 0 km d is pl ac em en t o n th e M ai n bo un da ry ( MB T) (P in F m in us P in -F ') a n d lo w er t h ru st s. T ot al s ho rt en in g o n th e s e c ti o n , in cl ud in g th e fr o n ta l th ru st s ys te m s, th e M ai n C en tr al t h ru st ( MC T) a n d hg ih er s he ar z o n e s is 3 30 km ( Pi n G m in us P in -G '). 8 401 that the shortening implicit in the cross-section (Fig.8) is not simply the subtraction of the line-length of Moho, imagined by the seismic refraction methods (9), from the present length of section. This would not include displacements along the Moho itself, such as those on part of the Main Central thrust (MCT); rather it is necessary to balance the entire crustal profile. Thus large scale crustal shortening can be accommodated within the central Himalayas, indeed several hundred kilometres are required to explain the crustal thickness. However, as Mattauer (6) recognised, further shortening is required to account for the relative convergence between Asia and India since the Eocene. Lateral expulsion of Asia towards the east, as proposed by Tapponnier et al. (11), was favoured by Mattauer (6) to act in combination with the thr~ting of Indian crust. There are however, considerable problems of strain compatibility, notably in the northern part of the thrust belt, in accommodating the two contrasting deformation mechanisms. Complex strain patterns would result from the interaction of two non-coaxial transport directions (north-south compression and east-west strike slip) yet numerous workers in the central Himalayas have interpreted strain patterns in terms of ideal simple shear (6, 8). The solution adopted by Mattauer (6) is that thrusts which initiated at low angles, are passively back steepened (see Fig. 6) and are then reactivated by strike slip transport. This would be possible if the back steepened thrusts were 'cylindrical', that is laterally continuous without major changes in geometry. However, it is not clear that widespread reactivation would occur within a laterally branching and complex thrust belt such as the Himalayas and is more problematic at depth since the back-steepening process is likely to be a high level phenomenon. Secondary lateral movements may arise from gravitational spreading of the thickened Tibetan crust but these would also have to be of minor importance to avoid the strain compatibility problems. At present we have no solution to this problem but future investigations may uncover greater displacements within the thrust belt, with the implication of more Indian crustal subduction beneath Asia. This would then eliminate the need for substantial lateral expulsion. Moving back to the thrust belt, regional studies of mineral and stretching lineations in thrust mylonites by Mattauer (6) show a systematic divergence in transport direction around the Himalayan arc. Thrust movements are apparently arranged radial to the southwardly convex arc (Fig. 9). This arc is considered to be closely followed at depth by the thickened lower crustal stack depicted on Figure 8. Thus divergent thrust transport at outcrop may be due to a component of gravity spreading (39) away from the thickened mass. This, or any other gravity driven mechanism, can not account for much of the thrust transport since any compressional displacements at the active thrust front must be balanced by extensional flow or faulting in the orogenic interior. Only minor extensional faulting (less than 50 km) has been recognised within the Himalayan thrust belt so that the main cause of thrusting must be lithospheric compression. Divergent thrust movements of whatever cause, create new problems for strain patterns within the thrust sheets themselves (Fig. lOa). Simultaneously diverging movement unavoidably increases the length of 402 Fig. 9. The relationship between predicted plate movement vectors and apparent thrust transport directions (6) in the Himalayan belt. MBT- Main Boundary thrust; MCT - Main Central Thrust; Kohistan complex. The stippled area represents the area underlain by subducted continental crust, the Pamir deep earthquake zone (6, 13) is interpreted to represent actrivity at the trailing edge of the Indian crust (Fig. 7). Section lines T-T' and S-S' relate to Figs. 8 and 7 respectively. the arc between the movement vectors. Thus we would expect along strike extension on the Himalayas if the lineation patterns truly reflect different thrust directions. Burg and Chen (38) describe some minor faulting which causes east-west extension in the central Himalayas but these have not been described for areas further west. An alternative mechanism to produce the divergent lineation patterns around the arc is for deep level thrusting to be laterally inhibited at depth in the west (Fig. 10b,c). This model would require a local pole of rotation in the west, swinging transport directions indicators clockwise away from their original north-south orientation. b. oblique I rofallon of folds ~ paSSIve markers -E7 ~ ;X\ unmhlbded I;c:.. transport -x ~4~t~t~ c. Fig. 10. Radial transport directions. (a) The problems of divergent transport, note then the thrust front and all higher structures would have to extend, A-B to A'-B'. The sequential rotation of early movement direction indicators (narrow arrows) at a lateral tip is depicted on (b) and (c); after Coward and Potts (26). 403 It does require arc-parallel extension provided the eastern end of the thrust system is pinned; the alternatives of wholesale detachment and rotation of the entire Himalayan belt seems unlikely. Inhibited thrusting may provide an explanation for the development of the very large anticlinal fold of the Hazara syntaxis (15) and also the Nanga Parbat structure. However, the detachment level for this structure must be at considerable depth, following the reasoning of Hossack (21). They can have little direct relationship to processes operating on the high level thrust systems of the peripheral Himalayan chain. Fission track dating (40) shows that the Nanga Parbat region cooled within the last one million years and that the rate of uplift is almost 1 cm/yr, considerably faster than surrounding regions. Zeitler (40) suggested that this rapid uplift is due to thermal re-equilibration generating buoyancy by converting dense ecologites to lighter granulite or amphibolite facies. However, the Nanga Parbat uplift is controlled by an active back thrust zone which emplaces Indian plate gneisses onto Indus river gravels deposits on Kohistan complex rocks (Fig. 11). Associated deformation includes gouge zones and slickensides in the gravels and surrounding rocks testifying to the late, high level nature of the fault zone. Indeed the older alluvial deposits are locally inverted in the footwall to this fault zone. Further west along the Indus valley there are greenschist-facies mylonites, which retrogress Dare! 1981 Hamran 1972 I: ,_ ~ ~ zone of pseodotachylde ~te gouge ~. ~ bea, 404 the much earlier amphibolite or granulite facies assemblages within the Kohistan complex (31). Elsewhere there are broad zones of individually small, pseudotachylite-filled fault zones. All these features may reflect deeper, earlier parts of the uplift history in the Indus- Kohistan region. Active faulting governing the uplift of the Nanga Parbat syntaxis is supported by seismicity records collected by Seeber and co-workers (12, 13, 41). They defined a broad zone of midcrustal earthquakes (Fig. 11) beneath the Indus Kohistan which they explained by northward backthrusting. Local backthrusting also exists in the Tarbela area (Fig. 3) within the Indian plate thrust belt. Back thrusts, as well as folding, have been related to the inhibition of forward thrust propagation (42, 43). So this inhibition model (Fig. lOb,c) explains not only the divergent pattern of mineral lineations in the west-central Himalayas, but also the development of back thrusts and folding with associated rapid uplift in the Nanga Parbat area and the high seismic activity beneath Hazara and Indus-Kohistan. 4. Continuity of Indian plate thrust systems. 4.1 Thrusting and the tectonic evolution of western Pakistan. It appears possible to relate the Main Himalayan thrust systems to the northward convergence of India into Asia (2, 3, 6) but it is less obvious how to relate the intensely arcuate fold and thrust systems of western Pakistan (Fig. 12) to this simple plate geometry. Within these three dimensional variations there are two other differences between the western Pakistan and central Himalayan thrust systems. Firstly the deformation front in the Sulaiman and Kirthar ranges is marked by hinterland-directed back thrusts (44). Secondly, the hinterland side of the thrust'belt in west Pakistan is truncated by a late, apparently strike-slip zone composed (Fig. 12) of the Ornach-Nal, Ghazaband and Cham an faults (45). Thus the present site of deformation in this western area lies not on the foreland side of the orogen but in the interior; the thrust belt itself has been bypassed. These differences represent greater problems in tracing thrust continuity from northern into western Pakistan than the intensely arcuate outcrop pattern. Recent investigations by the authors in the Salt and Trans Indus ranges suggest that the arcuate thrust front geometry is due to lateral ramps rather than strike-slip segmentation or lateral compression. The model favoured here predicts that thrusts beneath the Sulaiman and Kirthar ranges climb lateral section and have a tendency to join towards the south and north respectively so that the thrust belt appears narrow in the Quetta re-entrant (Fig. 12). Another difference is thrust belt geology between the northern and western ranges is the presence of a thick 'flysch' basin with local ophiolite slices which can be traced (Fig. 12) from the Karkar (in Pakistan) and Katawaz (in Afghanistan) south to the Makran forearc terrain on the Arabian coast (46, 47). In the Katawaz region the external portions of the 'flysch' basin lies on Eocene limestones of the Indian continental shelf. Similar facies of Eocene rocks can be found 405 0 >0O 'DO ... ...... ........ J i 406 in the Sulaiman (44) and Salt Ranges (20). The flysch deposits are probably of Oligocene age (48) in the external Katawaz but in the interior of the flysch zone they may be as old as middle Cretaceous (46. 48). The most likely explanation for these distributions is that the zone started as a marginal to oceanic basin but. as the over-riding Afghan block loaded the lithosphere. flexural subsidence (49) caused the foundering of the adjacent continental margin. Flysch sandstones could then be deposited on shelf limestones. As the collision process continued the lithospheric response of the Indian plate became more rigid, presumably because the continental interior is more cratonic and hence stronger than the originally attenuated margin. Thus syn-orgenic sedimentation changed from submarine (flysch) to alluvial (molasse) deposition (the Siwaliks) in Miocene times (48). It appears then that shortening of the Katawaz flysch occurred contemporaneously with thrusting on the Indian continental shelf in the Hazara Hills. Thus thrust systems must have stepped off the continental margin. much as they presently do between the Makran forearc and the Kirthar range. Within the development of this thrust belt. continental collision will have occurred at different times along the strike of the belt. It is probable the same type of variation occurred during the early part of thrusting in the central Himalayas. It is therefore important to correlate things like isotopic ages of syn-collision thermo-metamorphic events within an established framework of linked oceanic-continental thrust systems. At present lateral thrust continuity has only been established on a gross scale so that large scale correlations must be viewed with some scepticism. Once the oceanic lithosphere which underlay the inner Katawaz area had been subducted beneath the Afghan block. deformation progressed onto the Indian continent to form the Kirthar and Sulaiman fold and thrust belts. The following account is based on new field studies (44) in the Quetta area (Fig. 12). Banks and Warburton (44) provide a balanced cross-section through the thrust belt which is used here to constrain the deep structure of the Katawaz basin and to suggest a relationship between the thrust belt and the Cham an fault zone (Fig. 12). A regional balanced section is provided here (Fig. 13). further details of the thrust belt structure are provided by Banks and Warburton (44). Figure 13 illustrates the backthrust mountain front structure but more importantly here. it provides a restorable geometry for the main thrust belt. This is developed by detachment on pre-Jurassic rocks above crystalline basement. As in the Margalla-Salt Range section (Fig. 5) in northern Pakistan. no basement is caught up in the imbricate stack. Rather a length equal to the restored length of the cover sediments must remain at depth beneath both the thrust belt and the Katawaz hinterland. Stratigraphy of cover rocks suggest that this Indian crust was of stable shelf origin and thus probably thicker than 25 km prior to collision. The thrust belt implies shortening of almost 200 km. This corresponds to the width of the Karkar-Katawaz basin on the line of section which may therefore be entirely underlain by Indian continental crust at present. The Chaman fault zone lies directly over the trailing edge of the Indian continental crust. A possible explanation for this might be the locking of the crustal-scale thrust SSE a Siwalik cover crust b 407 NNW _~_ contlnenfaf _____ Kofkof - KofO'rllDZ _________ _ Afghanlsfan ____ thrust b.1t baSin hInterland (hanlon Q' eT,-;-z-:, -5;!¥:Z S!'~ ''2';=9 iL 5s ,r"""Z '- H---H---H---H---H--H---H--H---H--H--H--H---H~ Fig. 13. Crustal cross-section (M-line is the Moho) through the western Pakistan thrust belt, along line Q-Q' on Figure 12. Thrust belt geometry is after Banks and Warburton (44) and has been restored onto a crustal template (b). The structure of the Karkar-Katawaz basin, being dominated by faulted out anticlines separated by synclines, is interpreted from accounts by Andrieux and BruneI (46) and Arthaud et al. (47);. ramp at this trailing edge causing faults to propagate up into the hanging-wall to bypass the thrust belt. A corrolary would be that the other oblique-slip zones of the Ornach-Nal and Ghazabad faults, lie over the trailing edge of the Indian continent. A pattern then emerges of a relatively narrow continental thrust belt in the south west, compensated for by a wider oceanic accretion complex, passing north and eastwards into a broader continental thrust belt with only minor oceanic involve- ment. This inference is supported by stratigraphy of the Makran-Karkar- Katawaz sediments (46, 48). This speculative geometry is illustrated on Figure 14. A final problem remains. Why should the trailing edge of the western Pakistan thrust system lock up? This possibly occurred also in north Pakistan where a shear zone (the Hunza shear) near the northern suture in the Karakorum (Fig. 7) has brought up rocks which display rapid, recent cooling paths detected by fission track methods (40). One way to lock up the thrust system would be footwall flexure in advance of a subsurface lithospheric load, supplied by the hanging-wall block. In north Pakistan this load may be provided by local stacking of Indian sub-continental mantle into the crust by late-stage stacking of the subducted lithospheric wedge. In west Pakistan the load could be supplied laterally, since flexural loading will not have a polarity, by 408 MAKRAN I I W PAKISTAN OCEANIC ISUJDUCTfON I N PAKISTAN IN[LUDES THRUSTfN(J IN PAH'RS DISPLACEMENT ON I1MT CENTRAL HIMALAYAS INCLUDES DISPLACEMENT ON HIH e SHORTENING IN rlBf T SHORTENING OF INDIAN [ONTINENT (HIMALAYAN THRUSTING - - :~I~a~ __ ~~±-__ ~~~~ 1500 20DO 2500 500 1000 'N' Fig. 14. Distribution of displacements around the Himalayan collison belt. The section lines of Figs. 5, 8 and 13 are depicted S, T and Q respectively, a - present position of continental margin. the subducted Indian ocean lithosphere which originally underlay the Makran forearc (50). The main steepening downbend of this oceanic lithospheric plate must occur far north of the present accretion front to account for the large separation between the magmatic arc and the "trench". This is supported by the structural and stratigraphic work by Platt and Leggett (51) in the Makran who have invoked substantial underplating rather than simple imbrication. Footwall flexure of the western Indian lithosphere may also explain the development of the mountain front back thrusts documented by Banks and Warburton (44). Any such flexure would tend to reduce the surface slope on the developing thrust stack. Most mechanical models recognise that this slope has a critical influence on thrust belt dynamics (42, 52). Lowering of the surface slope would reduce the propensity of forward motion of the thrust stack, generating back thrusts which would in turn increase the topography near the thrust front to regain the critical taper for forward motion. In extreme cases the model might generate large scale shortening and 'break-back' thrusting within the complex to raise the hinterland. It is not clear to what extent this has occurred in, for example, the Tibetan region, or Kohistan. 4.2 Displacement distribution An arrangement of displacements asociated with the convergence between the stable foreland and hinterland of India and Asia during the collision process can now be considered. Figure 14 is a plot of the displacements on thrust systems at various positions around the active convergence zone. It is constructed from the three balanced cross- sections (Figs. 5, 8 and 13) across the western and central Himalayas and western Pakistan. These have provided various estimates of Indian crustal shortening by Himalayan thrusting. These were considered to be minimum values hence the figures are not directly comparable. However, it is likely that shortening of the Indian crust does vary in magnitude around the collision zone. The actual convergence between India and 409 Asia is likely to have been between l200km and 2000km and not greatly variable along strike. Therefore different mechanisms of crustal shortening must account for the variations in Indian continental thrust displacements and in the ubiquitous short-fall of these displacements with respect to the total convergence. In western Pakistan this may be accounted for by the subduction of oceanic lithosphere, such as presently occurs beneath the Makran forearc, so that continent-continent collision occurred significantly later in this sector of the thrust belt. Despite being minimum values, it is unlikely that shortening of the Indian continent accounts for all the convergence between India and Asia since the initial stages of collision in Eocene times. This short-fall in displacements is illustrated on Fig. 14. The deformation of the Asian (northern) side of the suture has not been discussed in this paper although recent observations by one of us (MPC) on the Tibetan plateau suggests substantial shortening of Tertiary sediments; their presently high elevation also indicates large-scale crustal shortening. Much of this strain would be laterally equivalent to thrust in the Pamirs. The ways in which shortening in the hinterland links through to that in the foreland have yet to be resolved. Neither is it clear why the Himalayan collision zone should comprise a series of thrust systems which operate simultaneously rather than the total convergence be contained within a single belt and, at any particular instant, be restricted to a very small number of active thrusts. The simpler model is probably appropriate for the Alps and Pyrenees. One solution may be that the post collision convergence rate between India and Asia has been high, at about 5 cmyr- l , while most thrust belts have documented displacement rates at up to 1.5 cmyr- l Furthermore, studies of thrust sheet rheology and fault geometry suggest that this lower rate approaches the maximum possible without catastrophic failure of thrust sheets. Clearly then, to generate coherent thrust systems within a fast convergence zone like the Himalayas, several thrust belts must operate simultaneously to accommodate the total relative plate motion. It is interesting that such a slow rate control apparently does not apply to the subduction of oceanic lithosphere. Thus a continental collision zone which is required to develop thrust belts which move at rates of greater than 1.5cmyr- l may be by-passed by the initiation of oceanic subduction at a trailing passive continental margin. At present we can provide no answer to why the Himalayan collision persists with multiple thrust belts rather than be by-passed by the subduction of Indian oceanic lithosphere beneath peninsular India. 5. Discussion It would appear then that it is possible to explain the gross structural evolution of the Indian continental lithosphere, following the consumption of Tethyan oceanic lithosphere and Asia-India collision, in terms of thrust tectonic models. This shortening mechanism, originally proposed by Argand (2) appears to operate regardless of the type of lithosphere entering the convergence zone. Thrusts step off continent into oceanic material although the differing flexural response of these 410 two lithospheric types and the intervening transition zone may cause variations in, for example, stacking sequences. Thus, strike-slip zones must truncate some thrust systems, they are in fact the large scale analogues to tear faults and lateral ramps in foreland thrust belts (34, 43) and are required to maintain strain and displacement compatibility around the orogenic system. To fully interpret tectonic evolution within collision belts it is necessary to trace the links between the various thrust systems, which can only be done by evaluating displacements. It is therefore essential to balance cross-sections. At present the restoration techniques only apply to plane-strain, two dimensional profiles so lateral tip and complex transfer zones must be avoided. Future investigations will use three dimensional techniques such as simultaneous serial restorations, in conjunction with palaeo- magnetic and incremental strain studies. Hopefully such geometric studies will then be able to lead to viable theoretical, mechanical models for lithospheric stacking (subduction) and to a better understanding of palaeogeography. Acknowledgements We have benefitted greatly from discussions on Himalayan geology, in the U.K. and in the 'drier' environment of Pakistan, with Garry Karner, Asif Khan, Rob Knipe, Dave Prior and colleagues at Durham, Leeds and Imperial College. This contribution derives from a co-operative project, funded by NERC (U.K.), between UK and Pakistan geologists at the National Centre of Excellence in Geology at Peshawar University. Logistical support from Professors Tahirkheli and Qasim Jan is gratefully acknowledged as is (RWHB) a research fellowship (Jaffe Donation) from the Royal Society. References 1. Chang, C. & Cheng H.: 1973. 'Some tectonic features of the Mt. Polmo Lungma area, southern Tibet, China'. Sci. Sinica 16, pp. 257-265. 2. Argand, E.: 1924. 'La tectonique de l'asie'. 13th International Geological Congress, Brussels, v. I, pp. 170-372. 3. Powell, C.M.: 1979. 'A speculative tectonic history of Pakistan and surroundings: some constraints from the Indian Ocean'. In: A. Farah & K.A. Dejong (eds.) Geodynamics of Pakistan. Geol. Surv. Pakistan, Quetta, pp. 5-24. 4. Powell, C.M. & Conaghan, P.J.: 1973. 'Plate tectonics and the Himalayas'. Earth Planet. Sci. Lett. 20, pp. 1-12. 5. Mattauer, M.: 1983. 'Subduction de lithosphere continentale, decollement croute-manteau et chevanchements d'echelle crustale dans la chaine de collision himalayenne'. C.R. Acad. Sci. Paris 296, pp. 481-86. 6. Mattauer,M:-;-"1986. 'Intracontinental subduction, crust- mantle decollement and crustal stacking wedge in the Himalaya and other collision belts'. In: M.P. Coward & A.C. Ries (eds.) Collision Tectonics. Spec. PubIs. 411 Geol. Soc. London, 19 pp. 7-50. 7. Soper, N.J. & Barber, A.J. 1982. 'A model for the deep structure of the Moine thrust zone'. J. Geol. Soc. London, 139, pp. 127- 38. ----- 8. Brun, J.P., Burg, J.P. & Ming, C.G.: 1985. 'Strain trajectories above the Main Central thrust (Himalaya) in southern Tibet'. Nature, 313, pp. 388-90. 9. Hirn, A. et al.: 1984. 'Crustal structure and variability of the Himalayan border of Tibet'. Nature, 307, pp. 25-7. 10. Dewey, J.F. & Burke, K.C.A.: 1973. 'Tibetan, Variscan and Precambrian basement reactivation: products of continental collision'. :b.. Geol., 81, pp. 683-92. 11. Tapponnier, P., Peltzer, G., Ledain, Y., Armijo, R. & Cobbold, P.: 1982. 'Propagating extrusion tectonics in Asia: new insights from simple experiments with plasticine'. Geology, 10, pp. 611-16. 12. Seeber, L., & Armbruster, J.: 1979. 'Seismicity of the Hazara arc in northern Pakistan: decollement vs. basement faulting'. In: A. Farah & K.A. DeJong (eds.). Geodynamics of Pakistan, Geol. Surv. Pakistan, Quetta, pp. 131-42. 13. Seeber, L., Armbruster, J. & Quittmeyer, R.C .. : 1981. 'Seismicity and continental subduction in the Himalayan arc'. In: H.K. Gupta & F.M. Delany (eds.) Zagros, Hindu Kish, Himalaya, geodynamic evolution. Geodynamic series lL Am. Geophys.. Un., pp. 215-42. 14. Ni, J. & Barazangi, M.: 1984. 'Seismotectonics of the Himalayan collison zone: geometry of the underthrusting Indian plate beneath the Himalaya'. :b.. Geophys. Res., 89, pp. 1147-163. 15. Coward, M.P.: 1983. 'Thrust tectonics, thin skinned or thick skinned, and the continuation of thrusts to deep in the crust'. J. Struct. Geo1., 5, pp. 113-23. 16. Coward, M.P. &But1er, R.W .. H.: 1985. 'Thrust tectonics and the deep structure of Pakistan Hima1yas'. Geology, 13, pp.417-20. 17. Lyon-Caen, He & Molnar, P.: 1983. 'Constraints on the structure of the Himalaya from an analysis of gravity anomalies and a flexural model of the lithosphere'. J. Geophys. Res., 88, pp. 8171-191. 18. Yeats, R.S., Khan, S.H. & Akhtar, M.: 1984. 'Late deformation of the Salt Range of Pakistan'. Am., 95, pp. 958-66. Quaternary Bull. Geo1. Soc. --- --- 19. Farah, A., Mirza, M.A., Ahmad, M.A. & Butt, M.H.: 1977. 'Gravity field of the buried shield in the Punjab Plain, Pakistan'. Bull. Geo1. Soc. Am., 88, pp. 1147-155. 20. Gee, E.R.: 1980. 'Pakistan geological Salt Range series, 1:50 000 (6 sheets)'. Directorate of Overseas Surveys, U.K., Geo1. Surv. Pakistan. 21. Hossack, J.R.: 1979. 'The use of balanced cross-sections in the calculation of orogenic contraction: a review'. J. geo1. Soc. London, 136, pp. 705-11. 22. Elliott, D. & Johnson, M.R.W.: 1980. 'Structural evolution in the 412 northern part of the Moine thrust belt of NW Scotland'. Trans. R. Soc. Edinburgh, Earth Sci., 71, pp. 69-96. 23. Burbank, D.W.: 1983. 'The ohronology of intermontaine basin development in the NW Himalaya'. Earth Planet. Sci. Lett. 64, pp. 77-92. -- -- --- 24. Elliott, D.: 1983. 'The construction of balanced cross-sections'. J. Struct. Geol., 5, pp. 101. 2S. Tahirkheli, R.A.K. & Jan, M.Q.: 1979. 'Geology of Kohistan, Karakorum and Himalaya, North Pakistan'. Geol. Bull. Univ. Peshawar, 11, pp. 189. 26. Coward, M.P. & Potts, G.J.: 1983. 'Complex strain patterns developed at the frontal and lateral tips to shear zones and thrust zones'. J. Struct. Geol., 5, pp. 383-99. 27. Le Fort, P., Debou, F:& Sonet, J.: 1980. 'The lesser Himalaya cordierite granite belt: typology and the age of the pluton of Mansehra, Pakistan'. Geol. Bull. Univ. Peshawar, 13, pp. Sl- 61. ------ 28. Gansser, A.: 1964. 'Geology of the Himalayas'. John Wiley, London, 289 pp. 29. Trommsdorff, V., Dietrich, V. & Honegger, K.: 1983. 'The Indus suture zone: paleotectonic and igneous evolution in the Ladakh-Himalayas'. In: K.J. Hsu (ed.) Mountain Building Processes, Academic Press, London, pp. 213-19. 30. Calkins, J.A., Oldfield, T.Q., Abdullah, S.K.M. & Tayyab Ali, S.: 1975. 'Geology of the southern Himalaya in Hazara, Pakistan and adjacent areas'. U.S. Geol. Surv. prof. pap., 7l6C, 20pp. 31. Coward, M.P., Jan, M.Q., Rex, D., Tarney, J., Thirlwall, M. & Windley, B.F.: 1982. 'Geotectonic framework of the Himalaya of North Pakistan'. J. Geol. Soc. London, 139. pp. 299-308. 32. Windley, B.F.: 1983. 'Meta;orphism and tectonics of the Himalaya'. J. Geol. Soc. London, 140, pp. 849-66. 33. Bard,~.P.: 19~'Metamorphism of an obducted island arc: example of the Kohistan sequence (Pakistan) in the Himalayan collided range'. Earth Planet. Sci. Lett., 65, pp. 133-44. 34. Boyer, S.E. & Elliott, D.: 1982-:---rThrust systems'. Bull. Am. Assoc. Petrol. Geol., 66, pp. 1196-230. 3S. Price, R.A.: 1981. 'The Cordilleran foreland thrust and fold belt in the southern Canadian Rocky Mountains'. In: K.R. McClay & N.J. Price (eds.) Thrust and ~ Tectonics. Spec. pubIs. Geol. Soc. London, 9, pp. 427-48. 36. Blundell, D.J.: 1984. 'Deformation of the Caledonide lithosphere of northwest Scotland'. Tectonophysics, 109, pp. 137-45. 37. Butler, RoW.H.o: 1984. 'Balanced cross-sections and their implications for the deep structure of the northwest Alps: reply. J. Struct. Geol., 6, pp. 607-12. 38. Burg, J.P. & Chen, G.M.: 1984. 'Tectonics and structural zonation of southern Tibet, China'. Nature, 311, pp. 219-23. 39. Elliott, D.: 1976a. 'The motion of thrust sheets'. J. geophys. Res., 81, pp. 949-63. 40. Zeitler, P.K.: 1985. 'Cooling history of the NW Himalaya, 413 Pakistan'. Tectonics, 4, pp. 127-51. 41. Armbruster, J., Seeber, L. & Jacob, H.: 1978. 'The Northwestern termination of the Himalayan mountain front: active tectonics from microearthquakes'. J. Geophys. Res., 83, pp. 269-82. 42. Elliott, D.: 1981. 'The strength of rocks in thrust sheets'. EOS, 62, pp. 397. 43. Butler, R.W.H.:1982. 'The terminology of structures in thrust belts'. J. Struct. Geol., 4, pp. 239-45. 44. Banks, C.J. & Warburton, J.: 1986. 'Passive roof duplex geometry in the frontal structures of the Kirthar and Sulaiman mountain belts, Pakistan' • .:!!. Struct. Geol., 8, 229-237. 45. Lawrence, R.D., Khan, S.H., Dejong, K.A., Farah, A. & Yeats, R.S.: 1981. 'Thrust and strike slip fault interaction along the Cham an transform zone, Pakistan'. In: K.R. McClay & N.J. Price (eds.), Thrust and ~ tectonics. Spec. publ. Geol. Soc. London, 9, pp. 363- 70. 46. Andrieux, J. & BruneI, M.: 1977. 'L'evolution de chaines occidentales du Pakistan'. Mem. h. ~ Soc. geol. France, 8, pp. 189-207. 47. Arthaud, F., Megard, F. & Seguret, M.: 1977. 'Cadre tectonique de quelques bassins sedimentaires'. Bull. Centres. Rech. Explor. Prod. Elf-Aquitaine, I, pp. 147- 88. 48. Auden, J.B.: 1974. 'Afghanistan-West Pakistan'. In: A.M. Spencer (ed.) Mesozoic-Cenozoic orogenic belts: data for orogenic studies. Spec. publ. Geol. Soc. London, 4, pp. 235-53. 49. Beaumon t, C.: 1981. 'Fore land bas ins'. Geophys..:!!. R. astronom. Soc., 65, pp. 291-329. 50. White R.S.: 1979. 'Deformation of the Makran continental margin'. In: A. Farah and K.A. Dejong (Eds.) Geodynamics of Pakistan. Geol. Surv. Pakistan, Quetta, pp. 295-304. 51. Platt, J.P. & Leggett, J.K.: 1984. 'Slip vectors from the Neogene accretionary prism, Makran Coast, Pakistan (abs.)'. Neotectonics meeting, 5th December 1984, Newsletter Geol. Soc. London, 13(6), pp. 39-40. 52. Davis, D., Suppe~& Dahlen, F.A.: 1983. 'Mechanics of fold and thrust belts and accretionary wedges'. J. Geophys. Res., 88, pp. 1153-172. 53. Burbank, D.W. & Tahirkheli, R.A.K. 1985. 'The magneto- stratigraphy, fission-track dating, and stratigraphic evolution of the Peshawar intermontane basin, northern Pakistan'. Bull. Geol. Soc. Am., 96, pp. 539-52. 54. Desio, A.: 1977. --rcorrelation entre les structures des chaines du nord-est de l'Afghanistan et du nord-ouest du Pakistan'. Mem.!!:. ~ Soc. geol. France, 8, pp. 179-88. THE TECTONIC EVOLUTION OF QINGHAI-TIBET PLATEAU: A REVIEW Chang Chen-Fa Pan Yu-Sheng Sun Yi-Ying Institute of Geology, Academia Sinica P.O. Box 634 Beijing China ABSTRACT. This paper, based mainly on the data of the athors' own work, presents information from different sources and views of various workers, especially those of the recent years. Following are the main points: The Kunlun and Qilian Mts. were a part of Paleo-Asia. The west Kunlun was strongly deformed by the Caledonian and Hercynian orogenies, while the middle and east Kunlun was deformed mainly by the Indosinian orogeny. A stabilized platform existed south of the Kunlun until Devonian or Carbanoiferous and probably fragmented at the end of Carboniferous. In late Permian time the north Tibetan block (including Qangtang and Lhasa blocks) broke off from Gondwanaland and caused the opening of the Neo-Tethys and narrowing of the Paleo-Tethys by its subsequent northerly drift. The closure of the Paleo-Tethys is evidenced by ophiolites and arc products of both late Permian and late Triassic ages. The Palaeo-Tethyan floor was subducted both northwards beneath Laur asia andsouthwards beneath the northern Tibetan block. The Bangong-Nu Jiang mini-ocean opened within the northern Tibetan block probably in late Triassic-early Jurassic time and closed in the late Jurassic. The development of the Neo-Tethys spanned an interval from Permian to early Tertiary and obduction of ophiolites took place in Jurassic to Cretaceous time on the subducted oceanic lithosphere of Neo-Tethys of late Triassic. Post-collisional evolution of the plateau has involved much north-south shortening and the present elevation of the plateau is the result of an end-Pliocene uplift. INTRODUCTION The Qinghai-Tibet plateau is bounded to the north by the Qilian Mts. and Altun Mts., to the south and west by the Himalayas and Karakorum Mts., and to the east by the Longmen Mt. and Xiao Jiang as well as the Red River fracture zone. It is a unique feature on the surface of the earth due to its extraordinary high elevation (average elevation about 5,000 m f,bove sea level) and its enormous areal extent (about 2.5 million km ). The high elevation of Tibet has long been attributed to a thick crust which is believed to be a result of the collision of the Indian subcontinent with Eurasia. Moreover, the precise mechanism of the development of the thick crust and the high plateau has been one of the most hotly debated topics in the earth sciences. 415 A. M. c. ,~engor (ed.), Tectonic Evolution olthe Tethyan Region, 415-476. © 1989 by Kluwer Academic Publishers. 416 The geology of the Qinghai-Tibet plateau, particularly the area south of Amdo and Hengduan Mts. in eastern Tibet has been the subject of many recent studies and become relatively well-known as a result of systematic surveying and mapping by Academia Sinica and Ministry of Geology & Mineral Resources. A considerable amount of information towards understanding of many perplexing problems of the geological evolution of this region is already available. Several selected areas, particularly in the eastern part, are now covered by geological maps of 1:200,000 or larger in scale. The geology of other areas is known also in gross outlines from reconnaissance work. However, for the immediate future, several large gaps still subsist, especially in some of the more remote and inaccessible, uninhabited and high mountain regions in the west. MAJOR TECTONIC DIVISIONS OF THE TIBETAN PLATEAU The Tibetan plateau can be tectonically di veded in to several blocks separated by sutures. Four sutures decreasing in age from north to south have been recognized (Fig.l). They are the southern Kunlun suture, the Litian Lake-Jinsha River suture, the Bangong Lake-Nu Jiang River suture and the Indus-Yariung Zangbo River suture. The blocks separated by them are the Kunlun block, the Songpan-Garze System, the Qangtang block (s.l.) which is composed of the Qangtang block (s.s.) Tanggula-Kaixinlin depression and Qamdo block of different origin, considered a Cimmeride "orogenic collage", the Lhasa block and the Himalayan block. The two southern blocks and the Qangtang block were once a part of Gondwanaland and include an old basement. During the Paleozoic the whole region was a comparatively stable continental platform with epicontinental shelf and lack of magmatism and foldin g. However, some 50 km south of the Indus-Yarlung Zangbo suture zone, along the line from Kangmar on the east to Mala Mt. on the west, a series of dome-shaped intrusions in sedimentary formations of mainly upper Paleozoic to Mesozoic age with well-developed metamorphic aureoles have been found. This magmatic belt is composed of two different groups of granitoids: More or less gneissified porphyritic granite group of Paleozoic and two-mica adamellite group of Cenozoic (Debon, 1984). Recent chronological data have defined two completely different groups of age in this belt. The old one is of lower Paleozoic (558 m.y.: lower Cambrian) from U-Pb isochron on the Kangmar pluton (Xu, 1985), and the age of the Lhagoi-Kangri granite is close to that of the Kangmar granite. Further west, the Mala Mt. granite, despite failure to give a well-defined isochron yielded an age older than 400 m.y. based on preliminary data (Xu, 1985). Further north, south of the Bangong-Nu Jiang suture zone, the metamorphic granodiorite in Amdo area also yields a U-Pb age of 531 m.y. In addition, some 1,600 km along the Himalayas and, in general, in a short distance north of the Main Boundary Fault, Le Fort et al. (1980) described and studied a belt of cordierite granites, which are made up of more than 15 independent granitic bodies with very typical and similar characteristics (Le Fort et al., 1978). These granites give a /. /~ r·" '1J J , / - - - ~ F ig .l : A s c he m at ic t e c to n ic m ap o f Qi ng ha i-X iz an g (T ib et ) pl at ea u. 1. O ph io li te , 2. G ra ni te s r e la te d to s u bd uc ti on , 3. V ol ca ni c ro c ks m o s tl y o f c a lc -a lk al in e c ha ra ct er , 4. M el an ge , 5. B lu es ch is ts , 6. E xo ti c bl oc ks , 7. F ly sc h 8. F ly sc h- li ke d ep os it s, 9. O ce an ic c r u s ta l s u it e, 1 0. M et am or ph ic ro ck s - ba se m en t, 11 . A na te ct ic g ra n it e, 12 . M ai n fa u lt s w it hi n th e s u tu re z o n e , 13 . O ve rt hr us ts , 14 . T hr us ts , 15 . N or m al f au lt s, 16 . S tr ik e- sl ip fa u lt s, 17 . F au lt s, 18 . H ig h- te m pe ra tu re m e ta m or ph ic z o n e , 19 . R ec en t v o lc an ic c e n tr e s. 20 . SK S: So ut he rn m a rg in o f K un lu n s u tu re , L JS : L it ia n -J in sh a s u tu re , BN S: B an go ng -N u Ji an g s u tu re , IZ S: In du s- Y ar lu ng Z an gb o s u tu re . 8 - ~ , ~ 3 [2 ]. b: d' ~ , G- -.:J ' c · r- .- -- ~. J9 ~" ~ ' l l L~ , ±. .. ~" /Y y~ 14 :r}s 12 " ~1 7 L;- ;-~ 18 ~~ 19 "" " - . . J 418 well-defined isochron with ages between 466-511 m.y. This lower Paleozoic age fits well with the ages of both the "porphyritic granite" group of the northern HimalayaIlgfel ~~nd the gneiss of the Amdo area in the Lhasa block. The initial Sri Sr ratios are very high in these Paleozoic plutons (between 0.7086-0.720). The lower Paleozoic magmatism discovered in the above-mentioned areas seemingly also effected other areas of Gondwanaland. It may compare with the similar events in Australo-Antarctica in the Ordovician (Crawford & Campbell, 1973) and late Pan-African event or Ross orogeny in the Antarctic (Elliott, 1975). By its widespread distribution, aluminous character, and high initial Sr ratio, this lower Paleozoic magmatism was developed in continental crust in the realm of Gondwanaland at that time. This widespread magmatic event seems more to be of epeirogenic than of true orogenic origin as no clear evidence of regional metamorphism and severe deformation has yet been found. Hence, some authors (Le Fort et al., 1983) suggested that a large crustal thinning with simultaneous arching of the crust would result in the generation of extensive granitic magma from the heated base of the crust. However, the products of this Paleozoic magmatism was affected by the Himalayan orogeny in various ways depending on their localization in the Tethysides. Based on the age of basement consolidation and the lower Paleozoic events as well as the conspicuous similarity of the Paleozoic sections in the Himalayas and Lhasa block (see below) the conclusion is reached that the Indian platform included also the Lhasa block during the Paleozoic (Chang & Pan, 1981). The evidence of connection of the Qangtang-Qamdo block with the Gondwanaland is less good but, in view of the absence of an ocean between the Lhasa block and the Qangtan-Qamdo block un til Juras sic time and the discovery of some metamorphic rocks under the fossiliferous Ordovician sediments in the vicinity of Longmo Lake in Qangtang area, we believe that they constituted a whole during Palaeozoic time. The system of Alpine-Himalayan orogenic belt is the product of the oblitera tion of the Tethys. The Tethyan domain consisted, during the early and middle Mesozoic, of two oceans separated by a strip of continents called the Cimmerian Continent (~engor, 1984) or the Northern Tibet block in the Tibet plateau (Chang & Pan, 1985), which began its separa tion from the northern and northeastern mar gins of Gondwanaland mainly during Triassic time although rifting in the easternmost part had probably begun earlier. North of this northern Tibet block lay the Paleo-Tethys, the original east-facing, equatorial embayment of Permo-Triassic Pangea. To the south, Neo-Tethys was evolving at the expense of the Paleo-Tethys. In the eastern Tethyan domain, three large and independent continental pieces, namely the North China, the South China and the Indochina blocks took part in Tethyan evolution and effected the division of eastern Tethys into a number of branches (~engor, 1984). 419 The double closure of the two Tethyan oceans from late Triassic to early Tertiary generated a double, largely overprinted orogenic system. That which resulted from the elimination of the Paleo-Tethys is here called the Indosinides and for that which is the product of the disappearance of the Neo-Tethys, the designation of Himalayan fold belt is reserved. In Tibet and China, the increasing width of the continental pieces separating Paleo-Tethys from Neo-Tethys effected a clear spatial distinction of the Indosinides from the Himalayan fold belt. The other group of younger ages of the magmatic belt in southern Tibet is Cenozoic with K-Ar ages around 13 m. y. (Debon et al., 1984) implying intraplate tectonic events. NEO-TETHYAN EVOLUTION IN THE TIBETAN SEGMENT The eastern Tethys from Iran to China contains a number of continental blocks and can be divided as mentioned into the northern Paleo-Tethys and the southern Neo-Tethys. These two Tethyan segments show mutual causality in that closing of the former was possibly due to the opening of the latter. It is common knowledge among geologists that the distribution of the sediments of Gondwanian facies crosses the Yarlung Zangbo suture zone but, their characteristics are somehow different on both sides of the zone. So far, the typical Glossopteris flora is found only in the area south of the Yarlung Zangbo suture (Fig .2). A wealth of early Permian (Artinskian stage) flora has been found in Xiagangjiang of the Gerze district by the geologists from the Geological Bureau of Tibet Autonomous Region. Elements of this flora are ascribed to Phyllotheca sp. and Schizoneura gondwanensis FEIS1M., showing strong Gondwanian affinity (Li, et al., 1984). However, they exist contemporaneously with Pecopteris sp. and ? Plagio zamites sp. and other elements of Cathaysiopteris. The fossils found are characterized by absence of typical elements of both Gondwanian or Cathaysian floras, indicating that the Lhasa block had already reached a far northerly latitude by that time as compared with the Himalayan block. It is worthy to note that the Permo-Carboniferous fauna of the Gandise-Nyainqentanglha and the Qangtang areas shows a similar character of "overlapped" evolution. For example, starting from the Chihsia stage of early Permian, the cryophilic fauna of Gondwanian facies was rapidly replaced by the thermophilic fauna of the Cathaysian Tethys. Mutual transition of cold and warm water facies in a short period of time may be seen in some areas. The existence of two different faunas reported previously in Lhasa area is also a reflection of this overlapped evolution. In view of the fact that the above-mentioned phenomena never have been seen in the Himalayan area south of the Yarlung Zangbo suture zone and the age of Glossopteris in India may continue into the Triassic, it is reasonable to hold that these phenomena cannot be interpreted simply as ,:- ,,, ,= , T a ri m B as in ,,, Y>~ ~ Q ai da m B as in - 1 - i,. -\ '\. ) •• Q ia n g- til llg i: )\~\ \ 'l, .~ 1 /~< 9/ ). f ~o "a .l i c y o f G on dw ct na f lo ra (p L ) IL L w i t ia n- Ji n~ .; ha s u tu re z o n e V Y ar lu ng , cc tn iT ,bo s u tu re z o n e • Tr a" e o f g la "i at io n ~v t3 an go ng -ll j"u J ia ng s u tu re z o n e F ig .2 : M ap sh ow in g th e di st ri bu ti on o f tr ac es o f gl ac ia ti on , fl or a an d fa un a o f th e Pe rm o- C ar bo ni fe ro us . SJ > - S t b, ..- . . . . . " '0 ~ N o 421 the global climatic change from cold to warm, which would lead to melting of the glaciers but instead, they are likely to be a consequence of continental drift and provide strong palaeontological evidence for the idea of "Cimmerian Continent" ($engor, 1981). Hence, the rifting of the Cimmerian Continent from the northern margin of the Gondwanaland commenced as early as late early Permian as indicated by the Panjal Trap of Kashmir and the coeval volcanism of early Permian in the Yarlung Zangbo suture zone. Recent work on the Permian to lower Triassic Panjal Trap by Bhat et al. (1981) and by Honegger et al. (1982) indicates that these extensive basalts belong to both tholeiitic and alkaline series and are very similar to the Tertiary alkalic rocks in rift zones such as Aden-Afar region and the Ethiopian plateau. They indicate active rifting ($engor, 1984). There are also some lines of evidence to indicate that the Neo-Tethys experienced spreading from late early Permian to late Triassic-Lias and the oceanic basin did reach a certain width as indicated by palaeogeographic analysis. Late Anisian Paraceratites trinodosus fauna (Ammonoids) in Lhasa area shows very close relationship to those of the Alps, and yet, it is very different from those of the adjacent Himalayas (Gu et al., 1980), suggesting that the main part of this wide Tethys of Triassic age was embodied in the Yarlung Zangbo suture zone. The discovery of abundant palynomorphs in a level from the latest Triassic (Rhaetian) north of Lhasa suggests that an equatorial or tropical latitude was reached by the Lhasa block as early as late Triassic. The flora indicates very close relationships with those of the equatorial (and north equatorial) province and lacks any significant Indian taxa. Most of the taxa are commonly found in Europe and Caucasus and more recently in Sichuan province (Allegre et al., 1984). The Triassic-Liassic flysch and the fragments of Palaeozoic age in the Zhongba area may represent a continental margin flanking the southern part of the Lhasa block and may be a counterpart of the Himalayan passive margin on the southern side. In fact, the Carnian-Norian flysch represented by the Longjiexue group in the eastern segment of Yarlung Zangbo suture zone includes sediments typical of the continental rise. The sedimentary facies and faunal associations are distinctly different from the coeval sediments (Gabo formation) of the Himalayan passive continental margin on the southern side. Haboia yunnanensis and Halobia ganzensis discovered in the Longjiexue group show the same character as those discovered in western Yunnan and Sichuan provinces while the Palaeozoic "fragments" of Zhongba area show some charcteristics that are even closer to the Gondwanian facies of Gandise-Nyainqentanglha area. That is to say, Neoschwagerina and Iranophyllum of thermophilic forms came out in the early late Permian (Maokouian) sediments. Moreover, Claraia wangi, etc. have been found in the Triassic sediments in the Gyirong area by the geologists of the Geological Bureau of Tibet Autonomous Region (Zhou, et al., 1985), which seems to belong to the typical element characterized by Claria-Bumorphotis association of the northern marginal province of the 422 Llzhu 2.0 I .~ uP.fEH 1 ; .0 0.5 '" J> Hz » ~ ~ » ;;- r--- rv > ;{;,~+Ju+ ~ y Lz ~ ;;- > ~ Hz-,)u+ LJZ 0pL ,·/-uu3 t-H,-=+-Phy Py+'dr ul)U rv> [1Z-'-lJU+ > ~ '-'Z Fig.3: Several selected ophiolitic sections along the Yarlung Zangbo suture zone from west to east (after Zhou Xiang et al., 1985). ST: Sedimentary deposits, RC: Radiolarian cherts, Ja: Jadeite, Phy: Phyllite, PL: Pillow lava, AmB: Amygdaloidal basalt, MaB: Massive basalt, Va: Variolite, BrB: Breccia basalt, AnB: Andesitic basalt, SpiB: Spilitic breccia, Spi: Spilite, Ke: Keratophyre, SD: Sheeted dyke complex, Di: Diabase, G: Gabbro, LG: Layered gabbro, Wr: Wehrite, Py: Pyroxenite, Tr: Troctolite, LDu: Layered dunite, Hz+Du+Lz: Harzburgite, minor dunite and lherzolite, SS: Sheeted sill, Do: Dolerite, IG: Isotropic gabbro, LOG: Layered olivine gabbro, OPy: Olivine pyroxenolite, PyTr: Pyroxene troctolite, FiDu: Foidal dunite, Du: Dunite, Hz+Du: Harzburgite and minor dunite. 423 Palaeo-Tethys (Yin, 1981). Thus, it can be seen that the constituents of the passive continental margin involved in the above-rnentioned flysch-melange associations belonged mostly to a part of continental margin of the northern side of Triassic Tethys with a great width. Therefore, the area between it and the Himalayan passive continental margin would be the main location where the said wide oceanic crust was fully developed. Based on the analysis on the Neo-Tethys in the Triassic, there is reason to envisage that steeply-dipping subduction of this earlier oceanic crust of the Neo-Tethys induced a secondary spreading in an ensimatic marginal basin. Hence, supra-subduction zone (SSZ) ophiolites of mainly Cretaceous age formed. Geochemically, the SSZ ophiolites have the characteristics of island arcs but the structure of the oceanic crust is thought to have formed by sea-floor spreading directly above the subducted oceanic lithosphere (Pearce et al., 1984; Gass, 1981). The island arc volcanics and the ophiolites in Lhaze district are a good example of SSZ ophiolites. Other ophiolites in the Yarlung Zangbo suture of Tibet have MORB affinities. The general trend from boninite to island-arc to MORB compositions appears to show the evolutionary characteristics of a marginal basin (Pearce et al., 1984). The enormous ophiolite belt seen at present along the Yarlung Zangbo is mostly the remnants of oceanic crust of this later period and some beautiful ophiolite sequences are preserved in the zone, which are characterized by the thinness of ophiolite layers as compared with those of major ocean basins. This, in turn, implies that these ophiolites probably represent fragments of oceanic lithosphere produced above the subduction zone in back-arc or marginal sea environments (Fig.3). There are some signs that this earlier subduction of the Neo-Tethys may possibly be already involved in the Gandise island arc. The precise lower limit of the age of calc-alkali volcanic rocks there is middle Jurassic (for example the Sangri group). It is quite evident that it is earlier than the age of the oceanic crust preserved now as ophiolites in the Yarlung Zangbo suture zone. Moreover, the Yeba formation tentatively ascribed to the Triassic is also in fact a sequence of epi-rnetamorphosed clac-alkali volcanic rocks. If this is the case, it would imply that the subduction of the Neo-Tethys commenced even earlier. Ultramafic rocks incorporated in the Triassic flysch are probably related to the subduction of this early period. A comparatively concentrated distribution of these ultramafic rocks on the southern margin of Yarlung Zangbo is tentatively regarded as the location of this early subduction zone. Hence, the young oceanic crust represented by the overthrust ophiolitic melange must come from a small ocean formed by sea-floor spreading directly above a subducted oceanic lithosphere, which is consistent with the conjecture that the Xigatse ophiolite was probably formed in a slow spreading oceanic centre. 424 Using radiolaria, Marcoux et al. (pers.comm., 1984) have determined an age of 110 m.y. for the cherts overlying the pillow lavas. Although some complete ophiolite sequences have been observed, the majority of the ophiolites are tectonically dismembered and ophiolitic massifs are generally bounded by thrusts or strike-slip faults and are intensely deformed. A very late compressive phase, postdating the Eocene-Miocene (?) Liqu conglomerate, in some places led to a northward backthrusting of the ophiolites (Fig.4,S). The tectonics of the Upper Cretaceous melange, the Triassic flysch and the autochthonous Indian plate series south of the Indus-Yarlung Zangbo suture involved four shortening phases with intense isoclinal folding and southward thrusting and symmetrical folding with steeply dipping cleavage and varying vergence within the period between late Cretaceous and Eocene. This complex history of superimposed deformation may be consistent with a southward obduction of the ophiolites on top of India from late Cretaceous to Paleocene (Allegre et al., 1984). N Fig.4: View showing a young thrust west of Dazhuka in the Yarlung Zangbo suture zone. 1. Fresh lherzolite, 2. Fault breccia mainly of ultramafic rocks and a few pebbles of probably Quaternary age; 3. Xigaze group and the drag folds related to thrusting. i )eni~Z:C'J l~,jjUrOLi -, L!l'jLU~~'l'~0 J'lE..JANGE i'lAPI'E .::N PEI1GZEJ ~ABUHUU FEN ;JUHTH ()F LAKE ,'lA.l~A3AHuVAH j. '1'('),1, 'l'rLuSS1:: r:. Gcmdl!::ic -.::onglomerate (f-;o..ccne 1 i. strong 426 and Eurasia (Molnar & Tapponnier, 1975). THE LHASA BLOCK The Lhasa block occupies the area between the Yarlung Zangbo suture on the south and the Bangong-Nu Jiang suture on the north. It was probably detached from Gondwanaland by early Permian time and constitued the northern Tibet block with the Qangtang block during early and middle Mesozoic. Proceeding from south to north following tectonic zones of secondary order can be recognized (Zhou Xiang et al., 1985) (Fig.l.6): 1- The Nyainqentanglha-Bomi central axial belt; 2- The Gandise continental marginal volcanic-magmatic arc; 3- The back-arc basin The Nyainqentanglha-Bomi Central Axial Belt It is represented by a belt of remnant uplifts of old rocks which are transgressively overlain by the Jurassic or Cretaceous sequences. The age of the metamorphic rocks that constitute the basement has been clarified by recent isotopic work (Xu, 1984). The age of the gneiss that the basement material had been transformed into appears to be synchronous with the plutonic activity and that is around 40-50 m.y. age although poorly-defined because of the heterogeneity of the inherited Pb components. Inherited Pb components in zircons yield minimum ages up to 1,250 m.y. in Yangbajain granite indicating that Precambrian basement exists in the area of the Nyainqentanglha Mts. In the vicinity of Goqen of Zayu in the eastern segment of this belt, the fossiliferous Lower Ordovician limestone overlies a series of schists of probably Precambrian age. Some sections of a mostly platform-type sedimentary pile from Ordovician onward have been found in Xainza area not far south of the Siling Lake, in which the features of the lower Paleozoic biostratigraphy show similarities to those of the Himalayas but, occassionally display the affinities of western Yunnan and Yangzi as well as northern China and Europe. The Permo-Carboniferous Gondwanian facies have been also found in southern Qangtang, except that the duration of the glacial period is longer, continuing from Visean to Chihsian (Lower Permian) . Cool-water faunas are represented by Stepanoviella, Costiferina, Stenoscisma and Dielasma, etc. as well as the Glossopteris flora, suggesting that the Lhasa block was situated nea r Gondwanaland proper. In Bomi-Zayu area, the basic stratigraphic characters may well compare with those of Xainza area, except the fact that no Ordovician-Silurian system has been found. Only the Carboniferous-Permian volcanic rocks are comparatively better developed in the former. L ha sa A re u. :" 'in gz lz on g Fm K2 T ak en am :h um ul on g Fm KL . w ei bL :z on g Fm D U od ig ou G r. J 3 ll \a ilo ng ga ng F m T3 C ha qu pu G r. L i.e lo ng go u Fm I P 2 ~ ~o ba do i Fm U ru lo ng F m Pa nd o G r. F lu V la l & g la ci al de po si ts A nd es i t ::e , s a n dy s ha le A nd es i t ::e & l~ ll mb ri c e in te rc al at :: ed \ "i th re d be ds , ll ar ls 1 n te r: al at ed w i t h c la s: ::i ::: r O G ks S ll iG ar en i t e & s la te C oa l- be ar in g Fm l.. im es to ne s Sa nd st on e, lim es t:: on e & s ha le C ry st al ll ne l im es to ne & a n de si te s L im es to ne & 5 il i, co li te s " L im es to ne & V 01 C3 I1 ic ' . r o o ks d" ya in qe n- ta n gl ha G r. L lm es co ne i n te rc al at ed ~ w i. th s la te s be an ng - - - - - - - t- -i l- :; -; :'o ':. j'- ' -=;: ::1'. . ,, ~~ o~ ~: ~! ~' s~~ ~~: one ' - . & li m es to ne w i t h '\. B an da pr od u( :t; us , et ~. G ra ni t:: e, gn ei ss , ho rn bl en de , e tc . X a1 nz a. ... (;o qe n A~ 2. B ue rg a G r. 1 r . . 1 _ 2 ~ Ll :r. .e st on e & s ha le Z he lo ng G r. I K l lJ ax un G r. iJ ug a G r. X -,- al a ?m P , lJ an gm ar l G r_ {u nz hu G r. D C ha gu ol uo z1 F m I 3 D ae rd on g C;i ~. De \O ll.1 ka F m SI IU aj ue G r. Sh al lo w m a ri ne d ep os l t s W1 t h :: :a lc -a lk al in e v o l- c: :a ni ::: i nt :: er ca la cl on s V ar ie ga te d c o a rs e t: :e rr i- s de po si ts . . . 1 t::h a am o u r, \:: o f ba Sl ::: & 1n te nn ed ia te v o lc an ic ro c ks Q ua rtz -s an ds to n es , :: :o a rs e s a n ds to ne s :: :a la te d w i t h s il c s to n e & s ha le , y ie ld in g S te pa no vi el la , e tc . Sa nd st on e in te rb ed de d w 1: h s ha le , li m es to ne l en s (} c hi n be dd ed m a rl s li m es to ne , do lo rm t1 c li m e- s to n e w i t h 00 11 te T hi n be dd ed l lm es to ne s & s ha le s 5h 8. 1e s in te rc al at ed W l t h do lo m i t e , li m es to ne & s ha le s S ha le s & m a rb le s l.J un es to ne & s ha le s Z an gg el Fm D uo nl r 'll D on gg u F.~ , JI lu gg ar - K an gr l Fm G en al on gb d Fm F ig .6 : S ev er al s e le ce td s tr a ti g ra p h ic c o lu m ns in th e L ha sa bl oc k (s ee te x t fo r d ea ti ls ). N ag qu A re a l11 ud s " o n e , s a n ds co n e & s a n dy c o n gl or n e r a te Sa nd s ; ;o ne & m u ds to n e s h al e in ~e r: :: al ac ~' m a rl a n d a n de si ;; e . . :'L l... '1d y- sh a1 e .l n~ er ~a la ce d & ':: 08 .1 be ds s a n ds to ne 1 nt cr .: :a - o n e & s ha le Sa nd y s la te & s ha le - " " N - J 428 The Gandise Continental Marginal Volcanic-Magmatic Arc It is composed of a plutonic and a volcanic belt corresponding with an Andean subduction environment. The Gandise plutonic belt comprises numerous bodies of gabbro, diorite, granodiorite and granite intruded from 95 to 40 m.y. ago with Sr initial ratios from 0.7036-0.7053 implying the presence of the products of a mixture of mantle derived components with continental crust material (Allegre et al., 1984). Farther north, the Yangbajain and Lhasa granitoid belt yields ages around 50 m.y. The Pb-isotopic composition of these rocks is compatible with an origin of partia~ melting of basement material. From the view of geochemistry, the Gandise belt appears to be a subduction-related complex built on the continental crust, while the Yangbajain and Lhasa intrusives are anatectic granitoids. Volcanic edifices in arcs of this continental marginal orogen are large composite volcanoes featuring inter layered flows and pyroclastic deposit s, which can be subdivided into three different associations (Zhou Xiang et al., 1985): 1- The marine volcano-sedimentary series of Jurassic-Cretaceous ~. Smaller patches of volcanic rocks of mainly Lower-Middle Cretaceous and partly Upper Jurassic ages remain wi thin the magmatic arc and are represented by the Sangri group in the Zetang area, which comprises andesite-dacite-keratophyre associations that can be assigned to typical calc-alkali series. They also coexist with a sequence of shallow-sea and littoral carbonate and clastic deposits in which minor mafic volcanic and ultramafic rocks may be seen showing sedimentary features of island-arc or less mature island-arc. 2- The volcanic formation of the continental marginal volcanic arc of late Cretaceous-early Tertiary. Volcanic rocks are very abundant in the southern area of Lhasa block. They are mainly of Lower Tertiary age and the representatives are the Dadou group (El _2) in the western segment and the Linzhou group (K~-E) in the eastern segment, respectively. The volcanics occur as igniri1bri tic tuffs in the middle Cretaceous Takena formation and form most of the 1,500 m thick Linzizong formation of Paleocene to Recent age and with piles of andesitic and rhyolitic ignimbri te flows that can be classified also as calc-alkaline. Minor trachyandesites occur in the upper part of these volcanic sequences that show an increase of alkaline content reflecting the gradual thickening of the crust. The volcanics of Linzizong formation have the same isotopic characteristics as those of the Gandise granitoids implying a similar origin. 3- Continental volcanics of the Neogene. Volcanic rocks of this age are associated with fault-bounded depressed basins. They are mainly rhyolites, tuffs, and volcaniclastics and would have formed in an intra-arc basin in the late stage of the magmatic arc development. Hence, the above-mentioned three volcanic environments appear to 429 be suggestive of the development of this Andean continental marginal orogen into an island arc in the later stages. The molasse doposits accumulated on the southern side of this island arc are a sign of the violent uplift of this range from Eocene to Oligocene. The molasse deposi tion and volcanism in the Neogene were likely to be associated with fault-bounded depressions. The Sedimentary Associations of Back-Arc Marginal Sea of the Gandise Magmatic Arc The sedimentary rocks of Jurassic-Cretaceous age are widespread in the area between Gandise-Nyainqentanglha on the south and Bangong-Nu Jiang belt on the north. Paleozoic formations also crop out as remnants of the Gandise-Nyainqentanglha median massif, which are coveredtransgressively by the Jurassic-Cretaceous sediments indicating that the Mesozoic sequence represents sediments of a continental margin on the northern side of the Gandise-Nyainqentanglha median massif. As the development of the Gandise island-arc proceeded, it gradually acquired the sedimentary features of a back-arc basin, which can be divided into following stages by sedimentary characters. 1- Late Triassic-early Jurassic. The sediments of this stage only occur in the vicinity of Siling Lake and Jiali and are composed mainly of terrigenous coarse sediments with minor thin shallow marine limestones and sub-aerial sediments. The Wuga group distributed in the area northwest of Siling Lake is mainly composed of variegated coarse terrigenous sediments with a large amount of basic and intermediate volcanics (Fig.6). Some faunas of late Triassic have been found in the limestone intercalations in the upper part. These associations probably represent the products of the initial rifting. 2. Middle-late Jurassic Stage. This is the main stage during which shallow marine sediments accumulated on the northern side of the Gandise magmatic are, where thick calc-alkaline volcanic intercalations are locally found. However, based on the analysis of the volume and the distribution of the volcanic rocks as well as the character of the sedimentary facies, the rock series seem not likely to represent an integrated island-arc belt, but only reflects a sedimentary environment of continental marginal belt. 3. Early-Middle Cretaceous stage. This is the sta ge of development of back-arc basin. The sedimentary belt of this stage was broken into two belts due to the influence of back-arc spreading. In the southern belt, extending from Shiquanhe to Nam lake as in the vicinity of Deqen, terrigenous detritus with calc-alkali volcanic rocks were developed. The mid-Cretaceous limestones are not so thick and only locally distributed. In the northern belt, extending from Anglonggangri through Baingoin as far as to upper Nu Jiang River valley, the Lower Cretaceous are characterized by deficiency in volcanic components. The coal series in 430 its upper part yields a Wealden flora. Recently, near Peng Lake close to Bard, new species of silicified woods assigned to genus Protodocarpoxylon (gymnosperms, Cheirolepidiaceae) have been discovered, which are always associated with Albian orbitolinids. This genus is widely represented in the Lower Cretaceous of western Europe, northern Africa ad southern Gondwania areas. That these fossil woods show the presence of growth rings suggesting a paleolatitude not too close to the equator during Albian and Aptian times, is a conclusion that agrees well with paleomagnetic results (Allegre et al., 1984). The mid-Cretaceous limestones widely overlap the underlying older rocks, reflecting an environment of comparatively stable shallow sea and littoral sediments. A considerable amount of ultramafic rock has been found in area between these two belts and whether they represent an independent ophiolite belt remains to be determined. This back-arc marginal sea was regressive and underwent folding and uplifting during late Cretaceous to early Tertiary. Some red molasse deposits locally accumulated in the fault-bounded depressions or intermontane basins. Structural Deformation of the Lhasa Block During the whole of the Paleozoic and most of the Mesozoic, there is no evidence to suggest compressive deformation, but only gentle epeirogenic movement influencing this area, which led to the development of local stratigraphic gaps. In the late Mesozoic, however, compressi ve structures of several generations have been recognized. Among these, the first is by far the most intense, which produced low grade metamorphism and isoclinal folds in the lower Mesozoic rocks. The folds are overturned to the south as indicated by sedimentary structures and bedding/cleavage relations and a north-south stretching lineation. The second generation structures are uplift folds trending E-W and refolding the earlier structures. A well-developed fan-like cleavage is roughly parallel with the main attitude of the axial plane with inconsistent vergence. The second phase is overlain unconformably by the Linzizong volcanics that spanned possibly from Paleocene to Oligocene. The volcanics themselves underwent a third phase of crustal shortening with long wavelength folds and both high and low angle thrust faults with small displacements (Allegre et al., 1984; Burg, 1983). The thrusts that developed around the Tertiary would be the main means of crustal thickening in the Lhasa block. Tertiary sediments are widely thrust by older rocks (Chang & Pan, 1981). Detailed studies of seismic sounding and granitic plutons in the Yangbajain, Gulu and Amdo bel ts also show that they are related to major thrusts yielding ages around 50, 100 and 130 m.y. respectively (Allegre et al., 1984) (Fig.7). Moreover, Quaternary sediments thurst by older rocks are locally discovered. Hence, the notion that the Quaternary tectonics in Tibet plateau is largely concerned with the eastwest extension is clearly an oversimplification. 431 A B c s N ~~ 100 km Fig.7: A. Wide-angle seismic refractions from the Moho along the section from Ngamring to Siling Lake revealing a complex Moho topography. Particularly noteworthy is the 20-km step in the Moho near IZS and the presence of two superimposed Mohos at the latitude of Yanbajain; B. A model for the structure of the lithosphere in Tibet and in the Himalaya. The role of obduction related to sutures (Indus-Yarlung Zangbo suture or IZS, Bangong- Nu Jiang suture or BNS), continental thrusts in the Indian plate (main boundary thrust, main central thrust, Kangmar thrust labelled as CTl to CT3) and in the Lhasa block (CT4 to CT6 corresponding to possible thrusts near Yangbajain, Gulu and south of Amdo) are emphasized. Continental and oceanic crusts are shaded in light and dark grey respectively. Subcrustal lithosphere is ruled. Older sutures and continental thrusts are rotated vertically and reactivated as strike-slip faults (from Allegre, et al., 1984). 432 THE BANGONG-NU JIANG SUTURE ZONE The Bangong-Nu Jiang suture separated the Lhasa block to the south from the Qangtang block to the north. It is located approximately 350 km north of the Yarlung Zangbo suture zone and extend over a length of some 1,500 km from the Bangong Lake in the west to Nu Jiang river valley in the east. A quite similar suture zone has been described west of Tibet in Farah Rud, Afghanistan (Bassoullet et al., 1980) and to the east in Sittang valley, Mytkyina of Burma (Mitchell, 1981). Up till the present, numerous ophiolitic elements and melange have been found along this suture (Fig .8). Although it is tectonically dismembered, all terms of a classical ophiolite sequence have been ascertained. In this suture only the ophiolites from Dongqiao to Gyangco have been studied in detail during the three-year(1980-l982) Chinese-French joint project in Tibet. km 2.0 1.0 0.5 Hz ,....;» » ,-.J Dong Co "'» » '" r->? » rv Hz » ~ ,...,» >- ,.... \~est of Dengqen RC Do+PL, G+Di Hz Fig.8: Several selec ted ophiolitic sections along the Bangong-Nu Jiang suture zone from west to east. The abbreviations are the same as those in Fig.3 (from Zhou Xiang et al., 1985). The ophiolites of this belt are characterized by the development of cumulates of different types. In general, the cumulates have a considerable thickness (about 1,000 m) and good continuation reflecting the presence of a relatively stable magma chamber. Although the replenishment of the magma seems to have been abundant, the spreading rate would be comparatively low, as indicated by the relatively poor 433 development of dikes. Based on the geochemical characteristics and the analysis of lead isotopes in samples of ophiolites from Bangong-Nu Jiang suture zone, it is suggested that the Dongqiao ophiolites resemble ophiolites from the eastern Mediterranean. This is compatible with the suggestion that the Bangong-Nu Jiang ophiolites were formed in a small back-arc or interarc basin. The age of the Bangong-Nu Jiang ophiolites is well-constrained by the foraminifers discovered in the radiolarian cherts associated with the ophiolites, in which Sethocyrtis, Cryptcapsa, Dicolacapsa, Tricolocapsa, Dictyomitra, Lithocampe, Cenellipsis, Rhopalastrum, etc. have been found indiating a Jurassic age (Wang, 1980). A typical ophiolitic sequence has been found in Dongqgiao-Gyangco area, which is composed of a mantle part comprising harzburgites clearly representing depleted residual rocks, and wehrlites, pyroxenites as well as duni tes that can be either cumulates or impregnation rocks, and also a crustal part comprising cumulate-layered gabbros, sheeted dolerites and volcanics. Foliated garet-bearingamphibolites which lie beneath the sheeted dolerites would represent remnants of a metamorphic sole. Preliminary geochemical data indicate that these ophiolites were probably formed in a small ocean in a subduction environment (Girardeau et al., 1984). North of Dongqiao and in Deqen area these ophilites are covered by a transgressive shallow marine de trial formation, the Zigetang formation (Chang & Pan, 1981). Numerous foraminifers and algae indicate an uppormost Jurassic-lowermost Cretaceous age (Fig.9). In the area west of Baingoin the ophiolites are transgressi vely covered by middle Cretaceous limestones. Hence, the obduction of the Dongqiao-Gyangco ophiolite occurred in late Jurassic, which is also consistent with the age of 179 m.y. of the hornblende in the metamorphic aureole beneath the metamorphic sole of the peridotite and the age of 172 m.y. of metamorphism in the Amdo schists. This is the signature of an even earlier emplacement of the ophiolites. The short time interval between the formation of the oceanic crust and its obduction shows that parts of the ocean floor obducted were newly formed and still close to a spreading ridge. Opening of the Bangong-Nu Jiang basin as a continental marginal basin initiated in late Triassic time, which appears to have resulted from intra-continental rifting, because a broad continuity of the Paleozoic shelf sedimentary facies occurs crossing the present Bangong-Nu Jiang suture zone. It was only from the late Triassic onward that differences became apparent between the sediments on either side of the Bangong-Nu Jiang suture. Similarly, Norin (1946) and Ling et al. (1983) described an upper Paleozoic section in western Himalayas, north of the Bangong-Nu Jiang suture, which is closely comparable both lithologically and paleontologically with those in Kashmir and Lhasa as well as Himalayan area to the south. Major differences between these two regions are merely a Mesozoic phenomenon. On the southern side along the J 3 - K i : rpl 2 cpl 1 Z ig et an g Co - ;; .. . U pp er m os t; Ju ra ss ic to lo w er m os t C ce ta ::: :e ol ls s e dL ne nt s S er pe nt in iz ed h ar zb ur gi te :; xs m D un it es F ig .9 : Th e tr an sg re ss iv e u n c o n fo rm ity be tw ee n u lt ra m af ic ro c ks an d th e u pp er m os t Ju ra ss ic -l ow er m os t C re ta ce ou s Z ig et an g fo rm at io n. Fr om bo tto m to to p: A. Se rp en ti ni ze o ha rz bu rg it es a n d du ni te s, B. Re d s il ic if ie d e n c ru s ti ng s, C. Co ar ~ gr ai ne d c o n gl om er at e w it h c hr om it e pe bb le s, Sa nd st on e m em be r: XS M 1- 9, t hi ck -b ed de d te rr ig en ou s s a n ds to ne s (c ro ss -b ed de d la m in at io ns an d pl an t re m a in s) , C al ca re ou s m em be r: XS M 10 -1 8, m a in ly w e ll -b ed de d c a lc ar eo us m u ds to ne s, w a c ks to ne s, s u bo rd in at e te rr ig en ou s in fl ux a n d pa tc h- re ef s (XS M 14 ), D ep os it io na l e n v ir on m en ts : XS M 1- 8 s u b- ae ri al d el ta ic t o X SM 9 -1 8 v e ry s ha ll ow m a ri ne . . . . '. . ;. l lI:s m 48 00 m .. . 435 line from Siling Lake northwestward to Coqen, a series of coarse detrital sediments with gypsum and a considerable amount of basic-intermediate volcanics and pyroclastics of late Triassic age have been discovered recently by the geologists of Geological Bureau of the Tibet Autonomous Region. This series would probably be related to the rifting process and represent the products of initiation of the opening. Structural Deformation of the Bangong-Nu Jiang Suture Zone Several successive tectonic events have been preliminarily identified (Girardeau et al., 1984). The first tectonic event, ~1. This event led to the folding of the Jurassic flysch series folrowed by a large thrusting event Dl _2 corresponding with the southward emplacement of the ophiolites ann thrusting of the Paleozoic metasedimentary rocks onto the Jurassic series. Thrusting was from north to south as demonstrated by the N-S stretching lineation in limestones beneath the ophiolite series and by subsequent shear structures. The second folding event, D2 . This event induced the formation of large symmetrical folds accompanied by vertical fracture cleavage, postdated the deposition of the Aptian-Albian red detrital sediments and associated volcanic rocks, and deformed all the basal sedimentary or thrust contact planes. It was followed by large scale reverse faulting. The obduction of the Dongqiao-Gyangco ophiolite occurred in the late Jurassic. It is, therefore, likely that the Dl tectonic event observed in the middle-upper Jurassic flysch series was due to the obduction of the ophiolite. The obduction probably occurred from north to south as indicated by both the vergence of folds in Jurassic flysch series and the stretching direction. Microtectonic analysis indicates that the ophiolites were overthrust from north to south in a relatively hot state. There is a big nappe from Dongqiao to Deqen and a series of thrust formed after the formation of the nappe. The thrusting of later ages cuts the earlier main thrust resulting in imbricate structures. Thrusting of the later ages occurred after the deposition of Eocene red beds. Hence, the ophiolitic slabs scattered on the area south of the Bangong-Nu Jiang suture zone are all allochthonous masses (Fig.lO). THE QANGTANG - SAN JIANG (THREE RIVER) BLOCK The Qangtan-San Jiang block is situated between the Jinsha river belt on the north and the Bangong-Nu Jiang belt on the south (Fig.l). Based on the stratigraphic and faunal data as well as the structural features such as folding and faulting, the Qangtang-San Jiang block appears to be a composite block. In general, its geology is not yet understood in detail because of its special location between Gondwanaland and Eurasia. The following units from west to east have been preliminarily recognized: :.), SO O ro S, O O G 4, ~O O s LH A SA BL OC I: F ig .1 0: G en er al iz ed s tr u c tu ra l s e c ti on Sa nd st on e, 2. S la te , 3. L im es to ne , 4. P~ ez oi c, 5. C on gl om er at e, 6 . U lt ra m af ic ro c ks , 9. G ra ni te , 10 . P il lo w l av a. - - ¥ - - * ]H K m tlA J ! 30 N G -N U J ~ANG ,;Ui 'U HE l O N E i ' )o ng qL uu ~ l G Jz ~ .\ ~ 4 L S §J 5 ~ 6 [3 :J 7 ~ 8 ~ 9 ~ ll l fr om D on gq ia o to D eq en ; 1. T ec to ni ze d li m es to ne o f la te ro c ks , 7. M el an ge , 8. V ol ca ni c N 11 _ t" ~G TA :\ JG B :"' U= J\ - 1'0 '" 0 - 437 The Qangtang Block The basement of the Qangtang block is largely buried under the Paleozoic-Mesozoic sediments and outcrop only as a few elevated fault-bounded blocks. They are represented by green schists, meta-diabase and meta-andesites, mica-schists and garnet-mica-schists gradually changing upwards into quartz-schists, muscovite-quartz schists, and marbles (Wang et al., 1984). These metamorphic rock series are covered by the Silurian and Devonian sediments in some places. No isotopic age dating has been made so far but, in the farther western extension in central and southeastern Pamirs, which is correlative to Qangtang of Tibet plateau between the Kunluns and the Indus-Yarlung Zangbo suture, the platform-type Paleozoic sediments rest with a pronounced unconformity on the Proterozoic-Archaen series (Kravchenko, 1979) . Therefore, these schists are temporarily placed to the Precambrian . Earlier reports of the presence of a Gondwanian facies in the southeast Karakorum (Horpa-Tso formation-Norin, 1946) have been confirmed recently by the group from Whuan Geological College (Fig.ll) (Liang D. et al., 1983). The Cameng and Zhanjin formations of Upper Carboniferous age and Qudi formation of Lower Permian age found by Liang et al. from Zaggar Co to Daoma district in southern Karakorum are typical marine sediments of Gondwanian facies. The Cameng formation consists of sandstones, slates, pebbly slate, sandy pebbly siltstones and diabase with a thickness of more than 500 m. The pebbles of various kinds are scattered and inharmoniously inserted in the grey-brownish argillaceous and silty matrix. The boulders comprise round, sub-round and sub-angular or even angular clasts mainly of quartzites of various types, slate, limestone, volcanic rocks and some granite and gneiss of greatly varying sizes. Glacial striae, slickenside or pressure hollow can be commonly seen on the surfaces of some pebbles. The pebbly slates and siltstones do not show stratification but always contain sandstones and sandy conglomerates with horizontal or cross bedding. Similar formations are widely distributed in the Indian subcontinent and Tibet proper. They have been thought to be glacio-marine origin and considered as one of the typical characters of the Gondwanian facies. The Zhanjin formation of Upper Carboniferous age consists of sandstones, slates intercalated with basic volcanics showing characteristics of turbidites and with a thickness of more than 3,250 m. Some cool-climate faunas, such as Eurydesma-Mourlonia, Amplexocarinia-Cyathaxonia, Ambikella-Anidanthus fusiformis associations in the sediments of the Gondwanian facies have been discovered. The Qudi formation of Lower Permian age consists of flysch intercalated with relatively rich sandy conglomerates, calcareous sandstones with macro-cross-bedding. They are more than 2,000 m in K ar ak or um e : a 1. , L, OO O '' lJ aa ge no ph yl lu m s p. {a be ln a- Ne os ch wa ge rln a Im op hy ll um T ib ec o ph yl lu m C os ti fe ri na -J uv es an ia L yt vo la sn a P ar ac an in a jl\ on oc ile xo di na -P ar af 'us u- 1i na - - - P am ir in a, R ug os of us u- li n a, P se ud of us ul in a, A iT iP ie xo ::a ri ni a, £ i§ . th ax on i a , E ur yd es m a m o u rl on ia , P m bl ke ll a- fu si fo nn is - - - Qa ng c:a n g A re a (F ra o l , 20 0 !'Y l 2, 00 J m 90 0 m ). B ra ch io _ po ds Ja 'T ld o A re a F ig .1 1: S ev er al s e le ct ed s tr a ti g ra p h ic c o lu m ns in Qa ng ta ng -S an J in ag bl oc k (s ee t e x t fo r d et ai ls ). ~s la nd -a rc ' /0 1 - 'e :n ic r o :: ks D )-P G Ig an to p' :: en s c f . ~ . j>. W 00 439 thickness. The faunas that have been found can be ascribed to the Pamirina-Rugosofusulina, Neospirifer fasiger-Subansiria ranganensis and Oriorassallella-Schizodus associations, which are w.idely distributed in the south Gondwania area. However, fusulinids of Tethyan affinity appear also in the upper part of the section. The Tunlonggongba formation of Lower Permian age consists of detrital rocks in the upper part and limestones in the lower part. Volcanic rocks also can be seen locally. The thickness of this formation is more than 500 m. Fossils that have been found belong to Monodiexodina-Parafusulina, Lytvolasma-Paracaninia-Tachylasma, Polythecalis-Chusenophyllum, Paraderbyia duomaensis-Jipuproductus, Costiferina-Juresania associations and Bellerephon sp., which clearly show the mixture of cool climate forms of the Gondwanian facies and warm climate forms of the Tethyan facies. It is likely that the widespread shallow sea at that time facilitated communication of marine organisms and resulted in provincial obscurity except in a strictly latitudinal sense. The Longge formation of lower Upper Permian is composed mainly of limestones and is more than 400 m in thickness. The fossils found belong to the typical Tethyan forms including Neoschwagerina-Yabeina association and Iranophyllum-Tibetophyllum assoiation. The fusulinidis belong to species widely distributed in the Tethys. The Jipuria group of Upper Permian is composed of conglomerates, sandy conglomerates, calcareous sandstones and sandy limestones in the lower part and dolomitic limestones locally with andesite intercalations in the upper part. The total thickness is more than 500 m. 0 nly some poorly preserved fossils are discovered such as Waagenophyllum sp. The L0wer Triassic consists of marls and has a thickness of more than 100 m. It rests with an unconformity on the Upper Permian and in turn overlain unconformably by the Middle Jurassic. The latter is composed of sandstones and limestones in the lower part and limestones intercalated with conglomerates and intermediate-basic volcanics in the upper part. The total thickness is around 3,000 m. The Cretaceous comes into tectonic contact with the Jurassic and is composed of reef limestones with intercalations of basic volcanics in the upper part and flysch intercalated with volcanics in the lower part. Its total thickness is more than 1,500 m. The Tertiary also is composed of sandstones and conglomerates with intercalations of reef limestones yielding corals and bivalves. To sum-up briefly, the Qangtang, in the narrow sense, is an area with Bangong-Nu Jiang suture on the south and the Hoh XiI Mts. on the north and the Xiyaergong-Mayigangri constituting its hinterland. The Qangtang geologically is a block with a metamorphic crystalline basement consolidated probably in Precambrian time and a platform-type PalffDzoic development since the Silurian. The ~aeozoic sediments here are distinctly different from those of southern Karakorum described above. 440 Li Xing-Xue and Yao Zhao-Qi (1981) described an Upper Permian Cathaysian flora in Shuanghu area, which bears a stronger resemblance to the late Permian flora or the Lungtan-Changhsingian flora of South China rather than to that of the upper Shihhotze formation of north China. In addition, none of these plants give any indication of an affinity to the Angara flora, nor do they show a close relationship with the Glossopteris flora. The Mesozoic is characterized by shallow marine sediments with relatively great thickness. Thick molasse deposits of Tertiary and red sandstones with gypsum of Neogene accumulated and the total thickness is greater than 3,000 m. However, it is not clear whether some sediments of the Gondwanian facies are involved in the metamorphic rock series that we have placed into the basement. It is reported that there is a clue of the existence of pebbly slates in the metamorphic rocks. If this is the case, it would imply that there was apparently stretching and fragmenting during late Carboniferous-early Permian. The trough where thick sedimentary-volcanic sequence accumulated was probably a down-faulted basin of extensional origin. Nevertheless, the continental crust on both sides of the trough was never completely separated by an oceanic trough, as no true ophiolites of late Carboniferous-early Permian are present. Hence, the Qangtang area was likely an active marginal area during late Paleozoic time. The Tanggula-Kaixinling Depression Upper Paleozoic-Tertiary sediments with volcanics of more than 19,000 m in thickness are accumulated here including coal beds and limestone wi th the coral Kweichouphyllum sp. of Lower Carboniferous, sandstones with intercalations of limestones and coal seams of Middle-Upper Carboniferous, intermediate-acidic volcanic rocks and limestones of Lower Permian, limestones interbedded with sandstones and gypsum in diapirs and volcanic rocks of Middle Jurassic and thick molasse deposits with intermediatebasic volcanic rocks of Tertiary times. Two important unconformities are found between the Upper Triassic and Middle Jurassic as well as the Middle Jurassic and the Tertiary. Based on the field observations, the volcanic rocks beneath the limestone of Permian age were formed in an extensional tectonic setting. The andesite in the Triassic can be referred to volcanic rocks of island-arc type and has a considerable strike length. Some volcanics discovered in the Jurassic were also formed in an extensional tectonic regime. Some gypsum in diapirs in the Jurassic seems to be derived from depth. Red beds of various ages are widely overlying the Jurassic series and have involved strong folding and northward thrusting. The deformation of red beds that occurred after the Tertiary was related to the deformation in the Himalayas. 441 The Qamdo Block The Qamdo block is bounded on the east by the Jinsha river arc-basin system and on the west by the Taniantaweng orogenic arc. The oldest sediments that crop out belong to the Lower Ordovician and are present only discontinuously in the Qingnidon-Haiton area, consisting of sandstones, slates, phyllites and limestones with 4,000 m and more in thickness. This sequence appears to have a flysch like character, but also contains some shallow marine limestones. They all are sediments of an epicontinental environment and are slightly metamorphosed. The Ordovician is overlain across a pronounced unconformity by the Devonian. The Middle Devonian to Permian section is represented by a sequence of shallow marine-littoral facies and terrigenous clastic deposits. Relatively thin molasse deposits are developed at the base. The Lower Carboniferous and the Upper Permian contain some coal series and in general, only a small amount of volcanic rock is intercalated. The thickness of this formation is between 1, 000 and 4,000 m. The upper Paleozoic marine and sub-aerial sediments are mostly continuous with few disconformi ties. The faunas which show strong Tethyan affinities are similar to those of Yangzi and south China. The late Permian coal series contains comparatively typical Cathaysiopteris represented by Gigantopteris cf. iza11ei indicating that there were relatively close relations between the Qamdo block and the Yangzi-South China block since late Paleozoic time. The sediments from end-Permian to late Triassic are characterized by frequent unconformities indicating certain crustal mobility. The unconformities between the Lower and Upper Permian and within the Upper Permian may have been the prelude of this mobile stage and several unconformities in the Triassic are basically in harmony with the orogeny that occurred in the Jinsha River arc-basin system along the eastern side. The Lower and Middle Triassic are represented by shallow marine littoral sandstones, mudstones and carbonates, in which a few intermediate-acidic volcanic rocks are intercalated. The thickness of the Triassic sediments reaches nearly 1,000 m. They probably represent sedimentation in a back-arc area bordering the continental side. The widespread Upper Triassic consists of red continental clastics, shallow marine carbonate rocks and marine-continental sandstones and mudstones with coal series from bottom to top. This sequence and fossils found in it are similar to those seen in Jomda and Batangtan areas except that there are no volcanics, which indicates that it was hinterland to a continental marginal volcanic arc on the east. The Jurassic sediments in the Qamdo area principally inherited the basin pattern of the late Triassic, in which the middle and lower parts consist of littoral and marine-continental clastic rocks with carbonates. The contact between the Triassic and the Jurassic is conformable. The Upper Jurassic together with a part of the Cretaceous are locally of a continental origin. The basement rocks of the Qamdo block are buried under a thick Paleozoic-Mesozoic sequence. No sign shows whether the high-grade metamorphic rocks sporadically outcropping in the vicinity of Xiaosuma, 442 Qinghai province represent the crystalline basement (Zhou, et al., 1985). If the basement of the Qamdo block is Precambrian in age, it would connect with the Indochina block, where some Precambrian schists and gneisses also outcrop in the Kontum massif. The Paleozoic, particularly sediments of the upper part and the faunal associations all show the characteristics of Yangzi-South China area. Moreover, there is no sign of any relic of plate activity extending from Qamdo to Indochina at that time, except the strike-slip faults of later stage along the line from northern Lancang river-Lanping-Wuliang Mt.-Puer. PALAEO-TETHYAN EVOLUTION OF TIBET PLATEAU The Palaeo-Tethys was the original triangular oceanic embayment of the Permo-Triassic Pangaea that came into existence as a by-product of the Pangaean assembly. The closure of Paleo-Tethys had very profound effects not only on the tectonics of the Tethyan realm but also on the tectonics of nearly the whole Eurasia. The fore- and hinterland deformations associated with the closure locally reached as far north as Siberia and a broad, continuous band paralleling the Paleo-Tethyan suture zone from the southern tracts of the Russian platform through central Asia to northern China and eastern Siberia was extensively disrupted by normal, strike-slip, and thrust faulting and folding forming a partly interconnected network of discontinuities that bounded innumerable blocks of various sizes ($engor, 1984), which are much like those of the Cenozoic tectonoic regime of Asia related to the collision with India (Molnar & Tapponnier, 1975). In the eastern Tethyan domain, three large, independent continental terrains namely the North China and South China blocks and the Indochina block took part in the Tethyan evolution and effected the division of eastern Tethys into a number of branches (Fig.12). The closure of the Paleo-Tethys occurred from the late Permian to the late Triassic up to the end of the Lias. The subduction of the floor of the Paleo-Tethys, today represented by the ophiolitic zone that rim the Songpan-Garze system, was going on both northward beneath Laurasia and southward beneath Gondwana. This is indicated by the granitic intrusions, andesitic volcanics and melanges as well as flysch of generally late Permian and Late Triassic in age along the respective margins ($engor, 1981). The elimination of Paleo-Tethys generated the broadest Indosinian foldbelt in the world in western China. The relics of the closure of the Paleo-Tethys along the northern margin are represented by the Jinsha river, Yushu-Litang and the southern Kunlun ophiolitic belts. They are indications of the closure of a series of branches of Paleo-Tethys developed on the continental margin attached to the northern continent, the so-called Cathysia, which 443 Fig.12: A schematic tectonic map of the Hengduan Mts. area; 1. Kunlun-Qunling fold belt, 2. Yangzi block, 3. Maqen-Lueyang suture zone, 4. Songpan-Garze system,S. Longmen Mt. ancient conjunction zone, 6. Garze-Litang suture zone, 7. Yidun island-arc belt, 8. Zhongzan massif, 9. Jinsha River-Ailao Mt. suture zone, 10. Jiangda-Shiamo basin, 11. Qamdo Block, 12. Baoshan block, l3. Dengqen-Nu Jiang suture zone, 14. Bomi-Burma block, 15. Indian plate, 16. Ophiolite. 17. I type granite, 18. S type granite. 444 includes the Yangzi block, Sino-korean platform, Tarim and Qaidam as well as Indochina blocks. The Cathysia is a collage that came into existence during late Proterozoic to early Palaeozoic and was separated by the Ural and the Mongolian-Da Hinggart (Greater Khingan) oceans from Laurentia and Angara-Land, while Palaeo-Tethys developed between Cathysia and Gondwana-Land located in the southern hemisphere. The Southern Kunlun Ophiolite Belt Many ultramafic to mafic rocks emplaced into the Permian, Carboniferous and Devonian sedimentary and metasedimentary rock series have been found along the southern Kunlun suture from the southern side of western Kunlun to Anyemaqen Mts. in southeastern Qinghai with the possible exception in the southern margin of the Burhan Budai Mt., where the ophiolites were probably obliterated by strike-slip faulting of laterperiods. East of the Burhan Budai Mt., from Huashixia, Maqen to Maqu, more than 100 ultramafic rock bodies occur within the Permian, Carboniferous and Devonian sediments (Fig.13). In this belt, huge exotic blocks of Permian carbonates occur within Triassic slates and limestones and fossiliferous exotic blocks of Lower Permian within the Upper Permian flysch. The Songpan-Garze system represents and accretionary complex composed of at least two island arcs and a thick fill of Permian-Triassic flysch. East of 92 E meridian a thick succession of mainly early and middle Triassic flysch intervenes between the Laurasia-bound collage (Kunlun and Anyemaqen Mts.) and the Cimmerian continent-bound collage. This intervening flysch fill rapidly widens eastward and forms almost entirely the major content of Songpan-Garze system. Both the Laurasia and the Gondwana-Land-bound collages plus the intervening flysch fill form the Indosinian foldbelt of western China. The Yushu-Litang Ophiolite Belt This belt is the junction zone between the Yidun-Xiangscheng slab and the Bayan Har slab. A nearly uninterrupted ophiolitic melange is distributed along this belt. Moreover, some intermittent ophiolite outcrops extend about a thousand kilometres along the line from Tongtian river through Yushu, Shancha, Maniganggo, Garze, Xinlong,Litang to Muli. The ophiolitic representatitves among them can be seen in the Zimenda of Yushu area in southern Qinghai and in Litang area in Sichuan province. The ophiolitic section seen in Zimenda consists of a sequence from north to south as follows (Fig.14): 1- Volcanic rocks comprising basaltic lava at the base, lenticular siliceous rocks and thin silieous slates in the middle and mainly tuffaceous rocks in the upper part; 2- Serpentinized and locally schistose harzburgites thrust to the north over the volcanic rocks and to the south over the pillow lavas, 3- Pillow lavas; 8 1 B B ' ~ I E"C il 7 ~l \l ~J :i ~ w C C ' F ig .1 3: O ph io li ti c s e c ti on s in G o1 og o f Qi ng ha i pr ov in ce ; 1. S an ds to ne , 2. S la te , 3. L im es to ne , 4. C on gl om er at e, S . P hy ll it e, 6. G ne is s, 7. M et am or ph os ed di ab as e, 8. M et am or ph os ed g ab br o, 9. S il ic eo us r o c ks , 10 . M et am or ph os ed s a n ds to ne a n d v o lc an ic r o c ks , 11 . P il lo w l av a, 1 2. A lt er ed ro c ks , 13 . G ra ni te , 14 . Se rp en ti ne , 15 . A m ph ib ol e- sc hi st , 16 . Qu at er na ry de po si ts , 17 . T hr us ts , 18 . N or m al f au lt s; A -A ' Se ct io n a lo ng t he h ig hw ay f ro m G ad e to M aq en , s c a le 1 :1 00 ,0 00 B -B ' Se ct io n o f D on gq in gg ou ( va lle y) , s c a le 1 :5 0, 00 0 C- C' Se ct io n in Q ing 10 ng c om m un e, X ia da w u, s c a le 1 :1 00 ,0 00 ~ , b~~ _n ;1 IB d ,. ~ H B ' § !I m il" ~ " m il ~" IE JI7 !Z l1X ~ 446 if, 200m 4,000 3,800 3,600 D , "- Gala Ton ti anghe NEil 200m Fig.14: The ophiolite section seen in Gala situated on the bank of the Tongtian river (D-D' in Fig.13). 1. Vo lcanic rocks comprising basal tic lava and len tic ular siliceous rocks at the base, thick siliceous slate in the middle part and mainly tuffaceous rocks in the uper part, 2. Serpentinized and locally schistose harzburgites thrust to the north over the volcanic rock and to the south over pillow lava, 3. Pillow lava, 4. Cherts and siliceous tuffaceous rocks, 5. Heavily weathered gabbro, 6. Tuffs, tuffaceous sandstones and slates with phyllitici and slaty structure. 4- Cherts and siliceous tuffaceous rocks, 5- Strongly weathered gabbro, 6- Tuffs, tuffaceous sandstones and slates with phyllitic and slate structures. All these rocks show low-grade metamorphism of greenschist facies. Although the ophiolite sequence is strongly dismembered, nearly all the important members of an ophiolite suite are present. The age of this ophiolite is considered by Pan Yu-Sheng as Lower Permian Based on regional correlation. Siliceous rocks, diabases, basalts and andesites are scattered in the vicinity of Litang. In general, the ophiolites are chopped up into countless smaller blocks and lenses and are chaotically in terminglend with blocks and imbrications of sedimentary material forming melange complexes. The matrix is composed of Triassic clastics and flycsh. Blocks of Permian, Carboniferous and even Silurian limestones, radiolarian cherts and various volcanic rocks, in which basalts are predominantly manifest either as huge exotic inclusions within or as sheet-like rafts on the matrix. A wealth of microfossils of early Triassic have been found recently in the variegated silicous rocks overlying the basic lavas of the ophiolite sequence, which suggests that these ophiolites were mainly developed during early-middle Triassic although spreading had commenced in late Permian. The pre liminary results of chemical analyses indicate that the basalt from the suite can be assigned to tholeiites of low-potash type and resemble the tholeiites of the Mesozoic-Cenozoic oceans indicating that the basalts in the melange belong to the MORB type. The Daochen island arc and the intermediate-acidic magmatic belt on the west 447 suggests that the closure of the ocean occurred at about the late Triassic and can be related to a westward-dipping subduction as indicated by the famous Shaluli clac-alkali granitoid belt of 195-215 Ma (Fig .12) . Further north, this suture zone is hidden by younger cover. Northward to the west of Yushu, it gradually joins with the Jinsha river belt and is severed obliquely by the latter. Hence, a series of huge fault-bounded depressions are developed along this strike-slip fault in the Hoh Xil Mts. The Jinsha River Suture Belt The Jinsha river fracture zone is an important former plate margin on the southern margin of Yangzi-South China plate and extends in northwestern direction as an anti "s" pattern. The belt has functioned in more recent times as a big right-lateral strike-slip fault extending from Red river in the south through the western side of Jinsha river to the southern side of the Hoh xn Mts. in the west. The belt displayed quite complex activity in the Paleozoic and Mesozoic and can be divided roughly into three segments from north to south (Zhou Xiang et al., 1985). The Northern Segment. It forms the jnuction of the Bayan Har block and the Qangtang-Sanjiang composite block on the southern margin of Hoh XiI Mts., along which sediments of Tertiary fault-bounded basins reflecting late strike-slip activity are developed. The ophiolite that reflects the closure of the Paleo-Tethys is largely hidden by the younger cover. The ultramafic rocks exposed sporadically in Xijinwulan and Dapeng Lake are the remnants of early ophiolites. A series of flyschoid sediments containing a great amount of Triassic limestones and basic volcanic rocks is present on the southern side of the Samaxujiari. Hence, this series is quite different from the coeval Bayan Har group with less volcanic content in HoI XiI Mts. developed on the northern side. It is assumed that the melange distributed in Samaxujiari area represents remnants of an ocean closed in late Triassic. The Middle Segment. The arc-basin terrane extending along the J insha river represents the junction between Yidun-Xiangcheng slab and Qangtang-San Jiang composite block. The right-lateral strike-slip fault bel t of later periods extends roughly along the western side of the arc-basin terrane and forms the new junction between the Yidun-Xiangcheng composite arcs and the Kaixinlin-Qamdo microcontinent. The ophiolite belt, the major indicator of the arc-basin association, is located within the strongly dynamo-metamorphic zone, which is about 40 km wide and can be divided into two belts: The western one belongs to ophiolitic melange, the matrix of which is composed of Gajinxue Shan group of Permian age consisting mainly of flysch with spilite-keratophyre and radiolarian cherts and commonly shows low-grade metamorphism in green schist facies. Some metamorphic minerals of high-intermediate pressure facies also locally exist. In this belt a 448 considerable number of huge exotic blocks of Permo-Carboniferous limestones occur within the matrix and many small ultramafic rock bodies have been found. They may be products of subduction of the marginal sea and were obducted westward on the Qamdo-Kaixinlin microcontinent. The eastern melange belt contains huge exotic blocks of Devonian, Carboniferous and Permian limestones, which occur within Upper Triassic slate and sandstones and show the character of olistostrome and turbidity sedimentation. They belong to the so-called argillo-arenaceous melange related to subduction (Zhang-Zhi-Meng & Jin Meng, 1979). ~,ooo m NEE 4, '7 ".,' 10 4, ilOe ·t,100 E 3,800 + 3,:000 Fig .15: The ophiolite section seen in Baimaxueshan in north-western Yunnan (E-E' in Fig.13). 1. Granite, 2. Meta-sandstone and slate, 3. Serpentinized peridotite, 4. Tuffaceous sandstone and slate and tuffs,S. Gabbro, 6. Diabase, 7. Breccia lava, 8. Pilolw lava, 9. Volcanic rocks with cherts, 10. Lime- stone intercalated with sandstone of probably late Triassic to Jurassic age. Up till now, a relatively continuous ophiolitic sequence has only been found along the road from Deqen to Benzilan in north-western Yunnan, which consists of cherts-pillow lava-basic sheeted dikes-gabbro and ultramafic rocks from bottom to top (Fig .15). On the other hand, many ultramafic rocks are emplaced into basic volcanic rocks as diapirs, which seems also to reflect a marginal basin with initial sea floor spreading. The preliminary results of chemical analyses of the pillow lavas indicate island arc thole rites or SSZ ophiolite types (Pearce, 1984) instead of the MORB type. These results also indicate that only a small ocean of marginal sea-type existed along the middle segment of the Jinsha river belt. It seems also to indicate that only a limited rifting of the Qamdo microcontinent relative to the Yangzi-South China block on the east occurred during late Paleozoic time. Based on the ages of the volcanic rocks, opening of this small ocean was initiated during middle-late Carboniferous and it closed prior to late Triassic. The polarity of the subduction was westward and the corresponding magmatic arc was located along the line from Jianda to Deqen. For instance, on the eastern margin of the Qamdo block the volcanism began already in 449 late Paleozoic but, expanded in scale during middle Triassic to early late Triassic time. The thickness of the volcanic rocks varies greatly from several metres to several thousand metres. The volcanics are mainly represented by intermediate with subordinate acidic ones and the basic rocks are very rare. Moreover, an intermediate-acidic intrusive belt is present between the eastern margin of the Qamdo block and the Jinsha River suture zone. Guse intrusive body in the southeast of Jiangda, Suwalong intrusive body in the south of Batang and Baimaxueshan intrusive body in Deqen area of northwestern Yunnan are, for instance, the representatives of this belt. These intrusives are generally small and discontinuous in space and the rock types of this belt are mainly diorite, quartz-diorite, granodiorite and granite. Some granites are distributed on the southeast of Jiangda and intruded into late Paleozoic sediments and transgressively overlain by the Triassic (Fig.12). Thus, it seems to be an intrusion of late Paleozoic time. The Baimaxueshan intrusive is characterized by an association of granodiorite-granite which is intruded into the Triassic and yields a K/ Ar age of 220 m. y. reflecting the impact of the Indosinian orogeny. After the Triassic all the Jinsha River arc-basin associations were overthrust toward the foreland of the Qamdo microcontinent. The Southern Segment. This segment represents the junction between the Yangzi block and the Indochina block. The suture zone extendes here along Tengtiao River fracture zone, which may link up in the south with the Black River suture zone in Vietnam (Fig.16). The Ailao Mt.-Tengtiao River fracture zone is a major fracture zone along which a zone of dynamo-metamorphism is developed involving phyllonite, mylonite and ultramylonite and varying in width from a few hundreds meters to 3 km. The strata on both sides of the Ailao Mt.-Tengtiao River fracture zone are quite different. On the northern side of the fault, the Ailao Mt. is composed of high-grade metamorphic rocks of Precambrian age. The overlying Ordovician to Triassic shallow-sea sediments resemble the cover sediments of the Yangzi block (Duan Xin-Hua & Zhao Hong, 1981; Zhou Xiang et al., 1985). In the Lanping-Simao fold belt, the Silurian to Anisian formations to the southwest are mainly geosynclinal-type deep-sea flysch and graptolitic shales and shallow-sea andesitic volcano-sedimentary formations unconformably overlain by continental molasse of upper Triassic. In the area between the Ailao Mt. fracture on the east and the Anding fracture on the west, more than 300 ultramafic bodies greatly varying in size and ophiolitic melanges composed of magnesian ultramafic rocks, gabbros, diabase, deep-sea radiolarian cherts, arenaceous slates, siliceous sandstones, and argillaceous limestones together with blueschists have been found (Gao Yan-Lin, 1984). The youngest sediments into which the ultramafic rocks are emplaced are Triassic formations and they are unconformably overlain by the late Triassic red beds. These observations being compatible with the closure time of this branch of the Paleo-Tethys at this time. 450 a _NE59° b o 5 tokm ~--~ 3Jl W 2 rn3 o:J4 p, QDs T, 0 6 QS]7 ~B [T~9 QIO [3Jll 012 ~13 [j;]14 WlS ~16 ! PP~ 117 Yuan ii an~ !vIIl8 iI ~ L'-.J Q\ W I9 ! 020 1 ~.~ ~~~;fi. S-~~y~ 2~::~:",fflJ;\,i ~~~m:~i 4[~=m~t·m-tll*tf Ll~OOm sDttttH Fig.16: A. A geological map of Mojiang-Yuanjiang are; 1. Quaternary, 2. Upper Tertiary, 3. Middle Jurassic, 4. Upper Triassi, 5. Anisian, 451 Taking the Ailao Mt.-Tentiao River fracture as the boundary, there occurs on the northeastern side a high~medium temperature, low-moderate pressure metamorphic belt represented by andalusite-cordierite and kyanite-staurolite-garnet associations. The isotopic age of the granodiori tes forming a thermal axis is around 217-169 Ma and roughly represents the upper limit of the age of this metamorphic belt, while on the southwestern side, a high-moderate pressure, low-median temperature metamorphic belt represented by glaucophane-stilpnomelane association forms a high pressure belt of a paired metamorphic belt. Hence, some authors consider that the Ailao Mt.-Tengtiao River fracture would represent an ancient Benioff zone: the Lanping-Simao fold belt representing the subducted oceanic plate on the southwestern side of the fracture and the Ailao Mt. uplift belt representing the obducted continental plate on the northeastern side. The latter probably constitutes a part of the Yangzi paraplatform (Duan Xin-Hua & Zhao Hong, 1981). THE SOUTHERN DOMAIN OF PALEO-TETHYS AND ITS CLOSURE Up till the present the only relics of the Paleo-Tethys that can be taken as the signature of the northern boundary of Gondwanaland are the ophiolites along the Lancang River suture. Farther south, the Langcang River suture roughly corresponds with the Chieng Rai-Medial Malaya zone (Mitchell, 1979) or the Uttaradit-Luan Prabang mafic and ultramafic belt (Hutchison, 1976). The ophiolite and melange and corresponding calc-alkaline magmatic belt of Permo-Carboniferous age constitutes the eastern boundary of the distribution of Gondwanian sedimentary facies. Mitchell et al. (1979; 1981) held that this belt showed the character of an island arc-continent collision and caused strong thrusting toward the foreland. After the later stage of the collision, an island-arc intermediate-acidic magmatic complex and anatectic granitic association in the zone of foreland thrusting was formed, which extends from Phuket Mt. of Malaya through the Tanintaweng Mt. of Thailand to Lincang area in Yunnan. Up to now, the tectonic character of this belt is not fully understood and the continuation further north is also not very clear due to the complicated structures and younger cover. 6. Permian, 7. Middle Devonian, 8. Lower Devonian, 9. Upper Silurian, 10. Middle Silurian, 11. Lower Silurian, 12. Epi-metamorphic rocks of late Paleozoic age, 13. Ailaoshan group of Precambrian age, 14. Metavol- canic rocks of Proterozoic age, 15. Granite of early Yanshanian stage, 16. Quartz porphyry, 17. Diabase, 18. Gabbro, 19. Ultramafic rocks, 20. Faults (from Duan Xin-Hua & Zhao Hong, 1981), R.F.: Red river fault, A.F.: Ailao Mt. fault, A.D.F.: Anding fault. B. The ophiolite section of Xiangyang Mt. of Xinping; 1. Sandstone, pebbly sandstone, 2. Silt- stone, siliceous slate, cherts intercalated with sandstone, 3. Basalt, 4. Diabase, diabase-gabbro and gabbro,S. Serpentine (from Duan Xin-Hua & Zhao Hong, 1981). 452 The exact location of the Lancang River suture is also not clear owing to the strong thrusting toward the foreland. Many authors place it along the line from Canning to Shuanjiang on the western side of the above-mentioned magmatic complex. Indeed, there are melanges with a great number of exotic blocks of Devonian, Carboniferous and Permian limestones here. However, the Permo-carboniferous basic volcanic rocks (including matrix and blocks) that constitute the main components of melange show clearly the character of a clac-alkaline sequence and are generally accompanied by more carbonates of platform-type. Despite the presence of many ultramafic rocks, outcrops of no complete ophiolite sequence has been found. But, a distinct melange association occurs here and has a linear distribution. Furthermore, the Devonian-early Carboniferous sediments belong to a graptolithic shale-siliceous rock-basic volcanic association that is sandwiched between the carbonate sediments of platform-type on both sides of the suture. It seems that the Canning-Shuangjiang melange belt probably represents a rift basin with finite spreading on the northern margin of Gondwana-Land. Judging from the stratigraphic record and character of the sedimentary facies, the Lancang River melange may correspond with the Phuket group (Devonian-Lower Carboniferous) in Thai-Malayan Peninsula. The latter is considered as sediments in a rift basin formed in the period when Thai-Malaya peninsula was separated from the Gondwana-Land (Ridd, 1971). In western Yunnan, the main part of the above-mentioned huge belt of tectono-magmatic complex is the Lancang River metamorphic belt, in which the age of the paired metamorphic belt, however, seems to coincide in time with the closure age of the Palaeo-Tethys, that is with the Indosinian episode but, the glaucophane-greenschist belt developed in the metamorphic rocks of the Lancang group of middle Proterozoic or Sinian-Cambrian age. The primary rock of the Lancang group is a fill of flysch-like sediments with intermediate-basic volcanic rocks. The basic volcanics (Huimin formation) in the upper part belong to an island-arc tholeiite-calc-alkaline series. Minor intermediate-acidic volcanics in the lower part show obviously the character of volcanic rocks of a mature island-arc. Moreover, the Lincang granite, the thermal axis of the paired metamorphic belt, is a composite and reworked batholith with a complex history and belongs to the "s" type granites. In addition to some 190-250 m.y. K-Ar whole rock ages, which correspond to the evolution of the paired metamorphic belt, some 700, 500 and 300 m.y. Rb-Sr ages have been reported also. Hence, the predecessor of the Lancang River metamorphic belt was probably an ancient island arc with a history of long-term development along the northeastern margin of Gondwana and (Zhou Xiang et al., 1985). The Lancang River ancient island arc extends southward to Burma and Thai-Malay peninsula. Therefore, prior to the formation of Canning-Shuanjiang oceanic basin in the late Paleozoic, all of the Western Yunnan, Shan State and Thai-Malay peninsula must have been an entity. In western Yunnan, this belt would be located to the east of the above-mentioned island arc with west-dipping subduction polarity as shown by the Lincang batholith, which increases in acidity and potassium content from east to west. 453 Hence, the Canning-Shuangjiang ocean and the whole marginal sea along the northern margin of Gondwana-Land in the Late Paleozoic would be a back-arc-foreland basin of the above-mentioned ancient oceanic consumption. The Lancang River paired metamorphic belt and the calc-alkaline volcanic rocks of Permian-Middle Triassic age on the east probably were the products of the eastward subduction of the oceanic floor. Activity of plate margins in later period destroyed much of the evidence of earlier events (Zhou Xiang et al., 1985). Farther northward continuation of the Lancang River suture in Qinghai-Tibet plateau is somewhat conjectural, because the Lincang magmatic complex disappears surprisingly to the north of Lincang. Some workers hold that the Lancang suture links up north-wards with the Bangong-Nu Jiang suture zone. But, the discovery of sediments belonging to Gondwanian facies in the Qangtang area indicates that the Bangong-Nu Jiang suture was not the northern boundary of Gondwana-Land during Permian-Carboniferous times. Moreover, the geology of ophiolites, the geochemical characteristics and the age of the ophiolite of the Bangong-Nu Jiang suture zone differ completely from those of the Lancang River belt. Some workers deduce that the northward continuation of the Lancang River suture is hidden beneath the Tanggula-Kixinling depression, which is by no means an empty conjecture. Sporadic ophiolite outcrops are found in the Lancang River valley and in the Qangtang area. Among the latter, the Bilong Lake-Xiaori basic-ultrabasic rock belt is several kilometers long and several tens of meters wide and is composed mainly of basic volcanics such as basalt, trachybasalt and diabase, etc., and in places contains Permian limestone slabs. To the north of this belt some ul trabasic rocks are scattered (Wang Cheng-Shan et al., 1984). Hence, these sporadic ophiolites seem to represent the remnants of the Palaeo-Tethys that bordered the northern margin of Gondwana-Land in Permo-Carboniferous time. Available evidence seems also to indicate that the Songpan-Garze system representes an accretionary complex composed of several island arcs and a thick fill of Permo-Triassic flysch. A number of belts of grani tic-granodiori tic plutons, intermediate to fel sic volcanic rocks and ophiolites of different ages are recognized within and along the margin of this accretionary complex. The oldest ophiolites and batholith belts here are of late Palaeozoic age and located along the northern and southern margins of the triangular-shaped Songpan-Garze system. The batholith/ophiolite belt on the northern margin constitutes the Anyemaqen Mts. arc system and the one on the southwestern margin forms Canning-Shuanjiang-Hol XiI Mts. arc, which was costructed directly atop the Gondwanaland basement of Qamdo and Qangtang blocks ($engor, 1984). In the Canningn-Shuanjinag arc system, the ophilitic melange accumulation was mainly formed in Permo-Carboniferous but, arc magmatism continued from late Palaeozoic to late Triassic. In Hoh Xil Mts. arc system, the ophiolitic melange accumulation continued through early Triassic, while arc magmatism continued to the middle Triassic with widespread activity and ended in the Jurassic. 454 To the south of the Songpan-Garze accretionary margin, two other island arcs were in existence. One is the Shaluli Mt. island arc and the arc magmatism mainly developed in Triassic with isotopic ages of 198-213 Ma. The other one is the Jinsha River arc and the arc magmatism here continued through the late Palaeozoic to the early Triassic with one K-Ar age of 220 m.y. To sum up, the consumption of the Palaeo-Tethys commenced in late Permian and ended prior to the late Triassic in Qinghai-Tibetan realm. It seems that the Tethys was rapidly changing its complex patterns of islands and seas during the Permo-Triassic period. Thus, the wedge-shaped triangular oceanic embayment would exist in a short period time during late Triassic. The remnants of closure of the late Triassic ocean are probably still preserved in the Yarlung Zangbo suture zone (Zhou Xiang et al., 1984). THE KUNLUN FOLDBELT As in most orogenic belts, the Kunlun foldbelt varies in structure along its strike. Its western continuation extends to North Pamir, where the mountain belt is divided into three latitudinal segments: a northern synclinorium, a central megaanticlinorium and a southern synclinorium (Belyaevsky, 1976). The central megaanticlinorium consists dominantly of medium-high grade and regionally metamorphosed rocks, whose deformation and metamorphism are pre-Devonian and, in places, Precambrian. The northern and southern synclinoria of North Pamir consist of the upper Palaeozoic and in the southern part of the southern synclinorium, of lower Mesozoic ophiolites, upper Palaeozoic clastic sedimentary rocks, subordinate limestones and volcanic rocks as well as many upper Palaeozoic and a few lowest Mesozoic granitic plutons. The eastern continuation of North Pamir is the western Kunlun, which was divided formerly into three latitudinal segments as those in North Pamir (Huang Ji-Qing, 1980). However, the intensive investigations made in recent years in western Kunlun by geologists from the Geological Bureau of Xinjiang have gathered evidence to cast doubt upon the previous subdivision. Based on the new evidence, the western Kunlun is divided into two latitudinal segments: The northern Kunlun is separated by the Kegang-Talun fault from Tarim massif to the north. The Kangxiwa thrust separates the western Kunlun belt from the Indosinian orogen to the south. Much of the northern Kunlun consists dominantly of a thick seq uence of marine intermediate-basic volcanic rocks, siliceous rocks and terrigenous sediments of Sinian-Cambrian age. Ophiolitic associations including ultramafics, gabbro-diabase and plagio-granites were found along with this sequence. The K-Ar age of hornblende of the granite is 517 Ma. Jadeite and minerals of high-pressure quartz-type have been found in metavolcanic rocks showing structures formed during the closure of an ocean basin. These rock associations are unconformably overlain by the fossiliferous Ordovician on the southern side of the Gongga Mt. The widespread molasse of Devonian age was mostlikely a 455 product of an important collision. The central Kunlun belt is probably not a "crystalline axis" or "central megaanticlinorium" as claimed by some authors since it consists mainly of fossiliferous Permian or Permo-Carboniferous sedimentary intermediate volcanic rocks with granite intrusives in clastic rocks. Therefore, the western Kunlun mountain belt is most likely an accretional wedge of an active continental margin attached to the southern margin of the Tarim massif and is composed of accreted materials of different ages. The northern Kunlun belt is an accreted plate margin of early Palaeozoic age, the precursor of which would be a mini-ocean developed along the margin of Tarim massif in Sinian-Cambrian age, while the central Kunlun belt appears to show the character of an island arc or a continental marginal volcanic arc representing an accreted margin of late Palaeozoic age. These features of western Kunlun belt are seen clearly also in the Altun Mts.-Nan Shan Mt. system. Further east in the Burhan Budai Mt. the oldest volcano-sedimentary rocks exposed on the road from the Kunlun pass to Golmud are Ordovician, consisting of limestone with corals, conglomerates and meta-quartz-porhpyry, black greenish slates and metamorphic sandstones from bottom to top. They are in tectonic contact with or overlain by Permian (7)-Triassic conglomerates, which are successively followed by fine pebbly conglomerates, sandstones, marbles, grey-greenish sandstones and sediments of turbidite facies (Fig.17,18). A considerable amount of volcanic rock is recognized within this series. This volcano-sedimentary series finally grades into a carbonate with late Triassic faunas, which is unconformably overlain by the latest Triassic or Jurassic molasse. The latter underwent only mild deformation with long wave-length folds. Thus, the deformation in the Burhan Budai Mt. occurred during the latest Triassic or even after the Triassic and the deformation was progressively more intensive southward. No indication of early Palaeozoic orogeny has yet been discovered in this area. The major shear zones bordering the southern part of the Burhan Budai Mt. were thrust to the south in the earlier stages and then back thrust later resulting in imbricated structures. The Kunlun fault remains active as a left-lateral strike-slip fault, along which a few gabbro bodies located to the west of Kunlun pass have been found resembling the general picture in the southern part of western Kunlun in the west and the Anyemaqen Mt. in the east. The eastern continuation of the Kunlun comes to the Qinling Mts., where strong subsidence continued through Permian to Triassic and was accompanied by widespread flysch deposition. Deformation of greater intensity probably occurred in late Triassic. These Indosinian deformations are contiguous with those in the Songpan-Garze system and eastern Kunlun and are accompanied by extensive granite intrusions in both areas. The granitoid rocks exposed in the Burhan Budai Mt. can be preliminarily divided into three groups (Fig.17). The one in the southern part paralleling the Kunlun fault is represented by strongly deformed two-mica granites and the other one is represented by biotite granites which are characterized by lack of deformation. The third is 456 -"''''''''''.- 2/ El 8 s 2' LLJ13 (,1"'1" I \~ _~- --;:-"--"-- Q2 eg6 =10 CJI4 Vol"", .. ",.,,,,,, W' 57 2" Cills ( _~I-" GJ4 0' =12 ~t6 Fig.17: A schematic geological map along the highway from Budongquan to Golmud. 1. Quaternary deposits, 2. Variegated silty shales, siltstones occasionally intercalated with argillaceous limestone lenses, 3. Grey shales interbedded with sandstones, 4. Meta-volcano-sedimentary series in the upper part and marbles, sandstones, sandy shale in the lower part as well as conglomerate at the base,S. Arkose sandstone, greywace occasionally intercalated \I/'ith impure limestone, 6. Lower Bayan Har group, slate, greywace intercalated with coals, 7. Grey shales interbedded with sandstone, limestone and intermediate acidic volcanic rocks and conglomerate at the base, 8. Limestone, sandstone intercalated with intermediate-basic volcanic rocks in the upper part and limestones intercalated with shale and sandstone in the middle part, siltstone, sandstone interbedded with shale in the lower part and conglomerate at the base, 9. Intermediate-basic volcanics intercalated with grey limestone in the upper part, 10. Grey grits, sandstone, pebbly sandstone, conglomerate, sandy shale occasionally intercalated with thin bioclastic limestones and marls in the upper part and grey, dark grey limestones in the lower part, 11. Metamorphosed greenschist, shale intercalated with limestone and volcanic metasediments, 12. Granite, 13. Two-mica granite, 14. Granodiorite, 15. Diorite, 16. Faults. 457 s A N ... P-T 2 km ~ B ------S T~~ C Fig .18: A. A section showing the contact between Ordovician limestone and conglomerate-marble sequence of probably Permo-Triassic age east of a small hydropower station north of Nachitai; B. A section showing the contact between the Ordovician and Permo-Triassic meta-volcano-sedimentary sequence west of section A; C. A section showing thrusts and folds with northward vergence of Triassic volcano-sedimentary sequence north of Nachitai. 458 a batholith belt well exposed in the northern part and all the plutons give late paleozoic ages ( 240 m.y.). It appears as a monocline with well-developed foliation dipping north and plunging into the Qaidam basin. In the Burhan Budai Mts. volcanic rocks of various ages are also widely developed. The volcanic rocks exposed in the northern part bordering the batholith are mainly acidic. A rhyolite alone reaches 2-3 km in thikness. The Devonian and Carboniferous volcanic rocks are represented by basic-intermediate-acidic associations from bottom to top and finally pass into rhyolites. The Triassic or Permo-Triassic volcanic rocks are mainly composed of basaltic lavas and probably formed in a back arc extensional environment. No acceptable signatures of the existence of Precambrian basement and early Palaeozoic orogeny have yet been found in Burhan Budai Mts. but this implies by no means that they do not exist until more detailed work is done. THE LONGMEN MOUNTAIN Evidence of a late Proterozoic plate boundary is found in the Longmen Mt. and Ailao Mt. (The Compilatory Group of the Tectonic Map of China, 1974 Institute of Geology, AcadEmia Sinica). On both sides of the Longmen Mt .-Ailao Mt. belt, developments of two different sedimentary facies continued until the late Triassic. Along the line from Qinchuan in the north to Yan bian in the south, the Bikou, Tongmuliang, and Yanbian groups of late Proterozoic age are all represented by marine soda-rich volcanic rocks and flysch series. Suxiong and Kaijianqiao formations of lower Sinian age located at the eastern side, namely on the Yangzi block, are continental red beds with typical island arc volcano-sedimentary associations. The corresponding intermediate-felsic intrusions yield a number of isotopic ages around 800-950 m. y., which provided vague traces of an ancient arc-trench system and constitutes the oldest boundary betwen the Yangzi block and the Songpan Ganzi system. This suture zone may connect southward with the Ailao Mt. fracture zone, which is considered by some workers as a former subduction zone of Sinian-Cambrian age along the southwestern margin of the Yangzi block. The isotopic ages of granites around 590-799 m.y. at the southeastern side of the zone in the Yangzi block and the thick molasse deposits in the lower part of Chengjiang sandstone are the corresponding products of the orogeny and both indicate a western active margin of the Yangzi block during the late Proterozoic (Zhou Xiang et al., 1985). The Longmen Mt. was the site of a northwest-facing passive continental margin (peri-platform depression in some of the Chinese authors' terminology) during the Palaeozoic-early Mesozoic period. In the early Palaeozoic, clastic sediments intercalated with volcanic rocks and carbonates with thicknesses of some ten thousand metres were accumulated. Local folding and uplifting took place during the early Palaeozoic. The sedimentary centre migrated to the middle segment during 459 Devonian time, while the Permo-Carboniferous shows a character of platform-type sedimentation. Prior to late Triassic the Longmen Mt. was folded with the consumption of Palaeo-Tethys. Hence, a hidden subduction zone seems to exist beneath the Longmen Mt. owing to the occurrence of an extensive and largely magmatic fold/thrust granitic and granodioritic batholith belt of Jurassic-early Cretaceous age (;>engor, 1984). The Jurassic-Cretaceous continental coal series and red beds were accumulated on the southeastern side of the newly formed fold belt and the Jurassic widely overlies the Triassic and other older beds with unconformity. The Longmen Mt. fold belt extends towards the northeast, in which faults are well-developed. A series of imbricate structures dominantly with northeast-strike are characterized in this belt. In general, the numerous fault planes dip to the west and the klippes thrust toward the southeast (Fig .19). Folds and faults of dominant northeast-strike are arranged in right-handed en echelon fashion which indicates that counter-clockwise rotational deformation occurs in Longmen Mt. However, many northeast-striking faults show clockwise horizontal torsion in recent geological time (Liu Zengqian et al., 1980). Along with folding and uplifting of the Longmen Mt., the foredeep successively migrated toward the Sichuan basin and corresponding thick molasses of Jurassic and Tertiary ages were deposited. As the Longmen Mt. is still uplifting at the present time, its frontal area continues to descend. The tectonic evolution of Sichuan and Shaanxi-Gansu-Ningxia basins during the Mesozoic and Cenozoic eras started with the Indosinian movement. In consequence, large, asymmetric, compressional fault-bounded depressions were developed as the floor of Palaeo-Tethys compressed against the China platform. These depressions were superimposed on the intra-cratonic basins of Palaeozoic age and show a multi-layered structure. On the western margin of the Ordos and in the foredeep of the Longmen Mt., Triassic-Jurassic sediments reach a great thickness, which indicates that the Palaeo-Tethyan evolution had considerable impacts upon them. The north-south extending structure of Helan Mt. may be considered as the aulacogen of the Palaeo-Qilian geosyncline. It is an area of folding of late Triassic age, whereas the foreland area was deformaed as a consequence of orogeny containing only germanotype structures. POST-COLLISIONAL EVOLUTION OF THE TIBETAN PLATEAU The Qinghai-Tibet plateau is an area of strong wholesale uplift. The crustal thickness reaches 50-70 km. In the southern, northern and eastern parts, thick molasse deposits of Pliocene-Pleistocene ages are well-developed (Fig.20,2l). These young molasse deposits have been subjected to widespread folding, thrusting and overthrusting. 460 Tianlaishan ",,000 m 1,500 1,000 JOO a 5 10 b c Fig .19 Sect ions showing the klippe and imbricated structures across the Longmen Mt. A. Klippe from Tiantai Mt. to Bailudong in Longmen Mt. of Pengxian county area; B. Klippe from Zhoujiaping to Lanbandeng in Longmen Mt. (from reports of regional mapping of Geological Bureau of Sichuan province); C. Schematic structural section across the Longmen Mt. . , ~.-: , H' " w ;" 1' ,' t" lK u :; l, p m () ~j On u, " ;.r ,d ' , H 11 P : ,y :;t um (= hl ..; K 1 In c F ig . 2 1: Is od ep th m ap o f M oh o o f C hi na . Th e Q in gh ai -T ib et an p la te au is bo un de d by z o n e s o f s u dd en c ha ng e in c r u s ta l th ic k n es s. T he g re at es t c r u s ta l th ic kn es s is s e e n in th e c e n tr a l p ar t o f th e pl at ea u re a c hi ng 73 k m . +> 0- , N 463 The plateau is also an area characterized by strong negative Bouguer anomalies. The negative anomalies reach a maximum in the inner part and decrease rapidly out ward (Fig.22,23). The high elevation of Tibet has long been attributed to a thick crust that has been believed to be the result of the collision of the Indian subcontinent with Eurasia. However, the precise mechanism of the development of the thick crust have long been one of the most hotly debated topics in the earth sciences. The exceedingly great thickness of the Tibetan crust and the rapid uplift of the plateau may be due to either underthrusting of India beneath Tibet (Argand, 1924; Powell & Conaghan, 1973) or intracrustal shortening and thickening (Dewey & Burke, 1973). India has continued driving into Asia since the Eocene collision at an average rate of about 5 cm/yr, approximately half of that when it was migrating more freely across the early Indian ocean, which implies the northward motion of India has apparently been accompanied by post-collisional shortening between 2,000 and 3,000 km that occurred solely within a continental plate. Major part of this shortening probably has been caused by crustal thickening but a minor part probably has also been the consequence of the lateral movement of crustal material out of the way of the converging masses of India and Eurasian (Molnar & Tapponnier, 1978; Hallam, 1984). Any model for the post-collisional evolution of the Tibetan plateau must consider the following facts, which are: 1- The pre-Quaternary basins are generally trending in E-W direction, 2- The present elevation of Tibet and the Himalayas is the result of uplifting since the end-Pliocene instead of immediately after the collision between the Indian and the Eurasia plates in Eocene, 3- Most of the Recent structures are N-S trending normal faults and E-W trending thrusts. In the Lhasa block, for example, most of the recent structures are N-S striking normal faults NNW and NEE striking strike-slip faults. The latter two types of faults often formed rhombic blocks with long axes of about E-W orientation and such kind of rhombic fault blocks are well-developed along the Bangong Lake-Nu Jiang suture zone and within the Qangtang block. No evidence of N-S trending normal faulting has been found north of Wenquan along the Qinghai-Tibet highway. In contrast, N-S compression is displayed by folds and thrusts in younger sediments. In addition, the most striking structure in the region north of Wenquan is the presence of an enormous E-W trending strike-slip fault belt. Based on the above-mentioned facts post-collisional evolution of the Tibetan plateau can be subdivided into the following three main stages: 1, 0 )\'0 3~0 61 0 K m ' - - ' - 2 0 0 " - T h e g ra v it y C hi na S ea is o p le th (m ga l . I I I D ia o y u Is la n d C h iw ei y u F ig .2 2: A ve ra ge B ou gu er gr av it y a n o m a ly m ap o f C hi na . Th e Qi ng ha i-T ib et an p la te au i s bo un de d by a be lt o f su dd en c ha ng e in B ou gu er gr av i t y a n o m ly . Th e n e ga ti ve v a lu e in th e c e n tr a l pa rt o f th e pl at ea u m ay r e a c h - 56 0 m ga 1 (a ft er R en J i- Sh un e t a 1 ., 19 80 ). .. .,' ::' ,, ' l ,," . j:> . ~ o 60 1 20 1 80 2 40 3 00 K m • ! H im a la y a A I tu n M t. L o p 1 \u rl lf I I ~ I '!L. . lI!! N y a in q en ta n g lh a M t. / . . . . lO X lO A ve ra ge F re e Y ar lu n g Z a n g b o/ ' { A ir a n o m a ly K u n lu n M t. T a n g g u la M t. I : /'-1 \G a n .t o k /'.1 .... .... 1/1 _ . . . . . l . I t!'! it Il - - - - - if '- ~ - tp . . . . . . / II to . . . . ~ IlJ - - / \ ... / 'if i ... . ' i! IlJ 11: \ " IlJ _ _ ~ _ _ _ ~. I I. I . I Q ai da m B a si n i!!! A 35 40 41i' " / , _ . / \ 55 ;;" '" 65 70 75 80 85 - 90 95 10 5 11 0 - - Il 5 12 0 ' " 92 '2 3' 92 '2 2' 88 '4 5' ' - . . . 35 'SO ' 32 '41 ' 27 'SO ' -~ OO - SO O - 40 0 a n o m a I y A v e ra g e B o u Ji tu er lO X 1 0 a n o m I y - 50 0 F ig ,2 3: G ra vi ty s e c ti on f ro m A lt ay M ts . to H im al ay as ( af te r Re n Ji e t a 1 ., 1 98 0) . +> - 0- , U 1 466 The Fir st Stage The Tibetan plateau is a region effected by several phases of collision. The collision that occurred along the Yarlung Zangbo suture zone in the Eocene was only the last collision episode. Every collision formed a fore-land basin, which deformed during later collisions. The deformations are expressed both as folds and as large-scale intracon tinental thrusts. Continental blocks were thickened by such faulting which generated also expansive occurrences of granitoid magma. Good examples these are the leucogranitoid belt in the Himalayas and the granitoid belts in Yangbajain, Baingoin-Gulu and Amdo. Ages of the granitoids of Baingoin-Gulu and Amdo are older than those of the granitoids in the Gandise without exception. Therefore these granitoid belts record the thrusting events occurred before the collision between India and Tibet. Similar thrusting events probably also occured in the Qangtang-Qamdo composite terrain and the Kunlun terrain. Thrusting will surely lead to crustal thickening and crustal thickening in turn would cause isostatic uplifting. Thus, whether the formation of the Tibet plateau is the result of earlier and later uplifting of separate blocks or of uplifting as a whole still remains a subject to be studied. However, it seems true that the collision along the Yarlung Zangbo suture zone between India and Eurasia led to the intensive compressional deformations in the continental blocks north of the Yarlung Zangbo and resulted in shortening the width of the crust from 2,000 km to about 1,000 km creating a crustal thickness of 70-80 km. Post-Eocene folding and thrusting is widespread, although only weak folding is found in the Lingzizong volcanics of Paleogene age seemingly indicating weak deformations after the Eocene. Nevertheless, older formations thrusting southward upon the Paleogene in the Lhasa block has been confirmed long ago (Chang Chen-Fa, Pan Yu-Sheng, 1981). Other post-Eocene thrusting, some with insignificant strike-slip component are found throughout the whole plateau, especially around Baingoin nearby Lunpola basin and in the region south of Erdaogou. Not only the outcropping Paleogene red beds are in tensi vely folded, but even in Neogene lake sediments over a wide area weak folding is found and the Neogene formations seem to be thrust by the Paleogene formations around Erdargou. Indication of young thrusting is found throughout the plateau (Fig .24,25, Fig .4). On the other hand, Neogene lake sediments are already folded in Lunpola basin. All these phenomena indicate that locally N-S shortening seems to occur even in regions where young normal faulting is active in recent geological times. Based on field observations along the Qinghai-Tibet highway, shortening by folding in an outcrop of 30 km in width is estimated to be 30% in the Fenghuoshan region. Similar shortening is also seen in the Paleogene red beds outcropping further north near Wudaolinag and Tuotuohe. Provided these widespread yet isolated outcrops of red beds represent the situation of the whole plateau, at least 40% shortening must have occurred in the Tibetan plaetau since the Eocene. This order of magnitude allows the total crustal thickening of the Tibetan plaetau s _ Su b- H ir rt al ay a - - : - - - - - M id dl e H im al ay a 1 1 I I I _ _ _ _ _ _ H i g h H im a 1 a y a _ _ _ 1 _ _ _ _ T ib et an S la b I I 1 I 1 I 1 E v e re st ( C ko lm ol uI ':l f.. ,a ) T ib et an o r T et hy an Z on e - - - - - - 1- - I I 1 1 1 I Z an gb o S u tu re - - 1 -- -- 1 I I I 1 Z an c; bo M 3 T - ~ _d- -~' ::- --- ~~- ~'" _____ ~ ~~-- ..c. '-=~ _ , \'\ ,\ K an gm ar th ru st D '- ' 1/ - N S ou th T ib et _ TR AN SH lM AL AY A K an gd es e b at h o li th s _ _ - - , ~'" »~- ?' ~ "' -" "~ ~' ,. . . c -: . '. Ii ' - . ~ - - ~ --~~ - ~-"" '- - - - - " ' _ , , , _ _ _ - _ ' \ = o -_ _ _ ~- - ~-. .:: .~- -- ~ ~ ~ ~ --:. , , ' _ _ _ ::~.:::: -_ __ ' - " ' " P a r :: :- ~ _ _ " ' " -......-- ---~~ "-C "'- "" , _ _ _ - . r - - - - z - ~~ ~ _ _ _ _ _ _ ' a u to c ht ho n -.::o- .:-':" --~~"' -__...-.. . . . - - . . :::' - _~ ~ ~ r ' _ _ _ _ __''"'' "~':::: :~--~ ~ -- ~:c~ ~~~ :~~: ~;~~ cC~; ~-~~ --~,-, -,,~ - - ~ , - , , ' " M -2 - J: L 2 - - - ' ' ' ' " , " - ~ S iw al ik m o la ss e . . . :_- :--: 1 X ig at ze m o la ss e ~ N 8 B lo ck y fl y sc h O p h io li te s T ri as si c fl y sc h a n d o c e a n ic r o c ks de fo rm ed c iu r i ng o bd uc t i on W T et hy an s e di m en ta ry r o c ks b3 o E J K an p; m ar ~n ei ss es H im al ay an le u co g ra n it es K y an it e- si ll im an it e- b ea ri n g g n ei ss es o f th e T ib et an s la b (D al le d u T ib et ) r= = I , ~ D D - ~ " - - - ' : ~ F o li at io n o f " L ow er N ap pe s" P ar a u to c ht ho n Z on e o f in te n se 5 1 _ 2 fo li a ti o n r e la te d to in tr ac o n ti n en ta l th ru st in g t h at c o m m e n c e d in m id d Ie E oc en e F ig .2 4: S tr uc tu ra l s e c ti on f ro m S ub -H im al ay as t o G an di se M t. sh ow in g th e in tr ac on ti ne nt al t hr us ti ng s in ce m id dl e E oc en e (a ft er B ru ne I 19 83 ). . j> . ~ - . l 468 s s b s c Fig.25Sectionsshowing young thrusting in the Tibetan plateau: A. Schematic section showing the Quaternary thrust by the Pliocene in Zhongda basin, Namling county; B. Schematic section showing the Quaternary thrust by the Upper Triassic-lower Jurassic in Zige Tang lake area; C. Schematic section showing the Quaternary and Quaternary volcanic rocks thrust by the Triassic in Wuquanhe in northern Qangtang. 469 to be explained by dispersed shortening within the Tibetan crust. Provided there is no post-Eocene shortening in the Paleogene outcrops, the estimate of overall shortening would only be about 12%, inadequate to explain the overall thickening known. Provided large-scaled underthrusting Asia by Indian crust caused the crustel thickening of Tibet, systematically migrating elevation fronts producing southerly-deri ved molasse sediments would be expected. Yet, we see no evidence of such systematic migration and derivation of the Paleogene and Neogene sediments. The Second Stage It began in the Pliocene and continued to the Pleistocene, and is the main stage of uplift of the Qinghai-Tibet plateau. In this stage, the Tibetan crust had already thickened to 70-80 km by intra-continental thrusting and folding. This considerably thickened crust resting on a base with probably ongoing partial melting at depth, spread laterally southeastward to South China sea and westward to Pakistan due to its own weight. In contrast with the strong uplifting of the Tibet plateau, a series of intensive subsidence basins were formed along the southern and northern margins of the Tibetan plateau, i.e. the southern piedmont of the Himalayas and the northern piedmonts of West Kunlun-Altun Mts.-Qilian Mts., where the Pliocene-Quaternary molasse formations of considerable thicksness were deposited. Similar Pliocene molasse is also found in drill cores around Dayi, Qionglai and Emei of Sichuan province (Fig .20). The younger molasse in these foreland basins is generally folded and thrust forming imbrication structures of Recent age showing that active north-south shortening, must be thickening and east-west extension of the plateau still must be going on (Fig.2S). Recent volcanism and geothermal activity are other very conspicuous tectonic features in the Tibetan plateau. The volcanic rocks discovered in Qangtang area show per-alkalic character in geochemistry which are probably the signature of recent rifting events. Finally, a generalized structural section across the Qinghai-Tibet plateau is given showing structural features of different tectonic units (Fig .26). SUMMARY The tectonic evolution of the Qinghai-Tibet plateau can be divided into several major episodes. 1- The Kunlun and Qilian Mts. were a part of Palaeo-Asia. They are characterized by continental crust but remnants of more or less cratonized former(late Precambrian-early Paleozoic) oceanic crust are 470 included that seems to represent a Proto-Tethys*. In the western Kunlun, the rockseries were mainly deformed and consolidated by strong Caledonian and Hercynian orogenies, while in the middle and eastern Kunlun, they were largely deformed in the Indosinian orogeny. 2- Until Devonian or Carboniferous time in the area south of Kunlun, a vast platform developed, on which initially (late Precambrian-early Cambrian in the Himalayas) a clastic sequence, then a uniform carbonate-shale sequence was laid down indicating a relative structural stability. 3- The end of the would be the Devonian) in strongly subsident future Jinsha river Palaeo-Tethys. Carboniferous and the Permian (in some areas it saw a fragmentation of this platform resulting sub-basins associated with volcanism along the fracture zone, which was the incipient 4- In the late Permian, the northern Tibet block broke off from the Gondwana-Land by rifting and caused the opening of Neo-tethys in its back and narrowing of Palaeo-Tethys in its front by subsequent northerly drift. Separation of these continental fragments would have been part of the general break-up of Gondwanaland as indicating by the Permian Panjal Traps in Kashmir and coeval volcanism in the Yarlung Zangbo zone and by the Triassic flysch deposition and marine volcanism in the northern Himalayan belt. 5- The Bangong-Nu Jiang mlnl-OCean opened within the northern Tibet block as a marginal basin in late Triassic-early Jurassic. Opening of this mini-ocean would be related to the south-dipping subduction of Paleo-Tethyan floor. The closure of the basin and the obduction of the ophiolite occurred in late Jurassic as the ophiolite was transgressively covered by the uppermost Jurassic-lowermost Cretaceous subaerial to shallow marine sediments. 6- The closure of the Paleo-Tethys occurred markedly in late Permian and from late Triassic to the end of Lias. That the subduction of the floor of the Paleo-Tethys, today represented by the ophiolite zones bounding the Songpan-Garze system was going on both northward beneath Laurasia and southward beneath the northern Tibet block--a part of Gondwanaland-- is indicated by the granitic and granodioritic intrusives, andesitic volcanic rocks and melanges as well as flysch of generally late Permian and late Triassic ages along the respective margins. * Editor's note: Employment of the term Tethys for pre-Pangaea oceans gives the undesired implication that these oceans constituted integral parts of the Tethys, defined as the ocean between the Angara-Land and Gondwana-Land by Suess, and corresponds with the Pangaean gap. S L e ss e r o n d T ib et o n I Il lm al ay a H im al ay a I I L h as a b lo ck Q an gt an g b lo ck N 50 IO O 15 0 K m ~ 1 " " - ~ 6 [3 11 . . . . . ~ 1 6 ~ 2 1 ~t ~2 6 K?} '~ 2 !4 7 12 ~ I 7 ~ 2 2 ~ 3 < I, ~~ 8 1++ +++ 113 + + + ~1 8 ~ 2 3 ~ 4 1 1 9 ~ 1 4 ~ 1 9 [M ]2 1 g. ~5 E. ~1 1J lb. ili1 5 ~] 20 x x x ~ 2 5 F ig .2 6: G en er al iz ed s tr u c tu ra l s e c ti o n fr om S iw al ik fo re de ep a c r o s s th e Q in gh ai -T ib et an p la et au to th e s o u th er n m a rg in o f th e Qa id am b as in . 1 . S iw al ik m o la ss e; 2. G al nd is e m o la ss e; 3. F ly sc h w it h e x o ti c bl oc ks , 4. O p h io li te ,S . T ri as si c fl y sc h , 6. S ou th er n T et hy an s e r ie s, 7. K an gm ar g n ei ss o se g ra n it e, 8. H ig h H im al ay an le u co g ra n it e, 9. H ig h H im al ay an g n ei ss , 10 . F o li at io n o f " lo w er n a pp es ", 11 . P ar a- au to ch to n , 12 . Z on e o f in te n se fo li a ti o n , 13 . G an di se b at h o li th , 14 . P al ae oc en e la v a a n d lo w er M es oz oi c s e di m en ts , 15 . Ju ra ss ic li m es to n e a n d s h al e, 16 . M id -l at e Ju ra ss ic v o lc an ic a n d s e di m en ta ry s e r ie s, 17 . Ju ra ss ic fl y sc h , 18 . C re ta ce ou s r e d be ds , 19 . M id dl e Ju ra ss ic a r g il la ce o u s a n d c a lc ar eo u s s e di m en ts , 20 . T an gg ul a g ra n it e, 21 . M id -l at e Ju ra ss ic a r g il la ce o u s a n d c a lc ar eo u s s e di m en ts w it h s a lt d ia p ir , 22 . T ri as si c fl y sc h 0 f So ng pa n- G ar ze s ys te m , 23 . S ou th er n K un lu n g ra n it e w it h w e ll -d ev el op ed fo li a ti o n , 24 . P al eo zo ic -l o w er M es oz oi c v o lc an o- se di m en ta ry r o c ks w it h w e ll -d ev el op ed c le av ag e, 25 . D ev on -C ar bo ni fe ro us v o lc an o -c la st ic r o c k s, 26 . K un lu n b at h o li th w it h fo li a ti o n di pp in g to th e n o r th . . ,.. - ' 472 7- The development of the Neo-Tethys would span in time from Permian to early Tertiary. The subduction of the floor of the Neo-Tethys would occur in late Triassic resulting in the formation of a supra-subduction zone ophiolites of Jurassic-early Cretaceous age directly above the subducted oceanic lithosphere. The final subduction of the floor of marginal basin occurred in middle Cretaceous, when a complete trench-arc-basin system was established. The collision between the Indian plate and the Lhasa block probably occurred in pre-middle Eocene, prior to the deposition of the Qiuwu formation (and molasse of equivalent age). The slow-down of the spreading rate in the Indian ocean in early Eocene would have been related to this collision. 8- Post-collisional evolution of the plateau has involved much north-south shortening and strong uiplifting. The present elevation of Tibet and the Himalayas is the result of an end-Pliocene phase of uplifting, which seems still in progress. Recent volcanism and geothermal activity are very conspicuous tectonic features. The volcanic rocks discovered in the Qangtang area show per-alkalic character and may be the signature of recent rifting events. 473 ACKNOWLEDGEMENT The present paper is completed mainly on the basis of our own work in the Tibetan plateau. A study of this topic in a high mountainous region covering a vast area must have benefited from the help, advice and comments of many workers. We are grateful to Dr. Zhou Xiang and his colleagues forkindly referring us their unpublished information. A special note of thanks we owe to Dr .Zhang Qi who provided us much valuable information in western Yunnan. We are also grateful to Dr.Deng Wan-Ming and Dr.Xu Rong-Hua for helpful discussions on the plutonism in southern Tibet and recent volcanism in Qangtang area. The authors are sincerely grateful to all their generous collaborations, instructive discussions and valuable suggestions. REFERENCES Allegre,C.J. et al., 1984. 'Structure and evolution of the Himalaya-Tibet orogenic belt', Nature, v.307'3no.5946, p.17-22. Argand,E., 1924. 'La tectonique de L'Asie', IF Congr. Geol. Int., Valilant-Carmanne, Liege, p.17l-373. Bassoullet ,J. et al., 1980. 'L' evolution des domaines tethysienne au pourtour du Bouclier indier du Carbonifere au Cretace'. Collugue C. 5, 2~ Congr. Geol. Int., Editions du B.R.G.M., p.180-l97. Belyaevsky,N.A., 1976. 'Tectonics of the northern part of the Pamirs-Himalayan syntaxis'. Intern. Collogium on geotectonics of the Kashmir Himalaya, Karakorum-Hindu-Kush-Pamir orogenic belts, Academia Nazionale dei lincei, Roma, p.29-42. Bhat,M.I., et al., 1981. 'Panjal Trap chemistry andthe birth of Tethys', Geol. Magazine, 118(4), p.367-375. Brunel,M., 1983. 'Etude petro-structurale des chevanchements ductiles en Himalaya (Nepal oriental et Himalaya du nordouest)', Ph.D. these, 395 pp., Universite Paris VII. Burg ,J.P., 1983. 'Tectogenese compares de deux segments de chaine de collision', Ph.D. thesis, 361 pp., Montpellier University. Chang Cheng-Fa, Zheng Xi-Lan, 1973. 'Some tectonic features of the Mt. Jolmo Lungma area, southern Tibet, China', Scientia Sinica, 16(2), p.257-265. Chang Cheng-Fa et al., 1978. 'The geological history tectonic zonation of uplifting of the Himalayas', Proc. Geo. for Intern. Exchanges, regional tectonics & geomechanics; separate pap. in English, 1977. Chang Cheng-Fa, Pan Yu-Sheng, 1981. 'A brief discussion on the tectonic evolution of Qinghai-Xizang plateau', Proc. Symp. on Qinghai-Xizang (Tibet)Plateau, v.l, p.1-18. Chen,H.J., Zhou,X., 1982. 'Plate tectonics and oil-gas resources', Lectures on plate tectonics, Geological Publishing House, Beijing, p.293-3l5. Crawford,A.R., Campbell,K.S.W., 1973. 'Large scale horizontal displacement within Australo-Antarctica in the Ordocian', Nature, 241, 11-14. 474 the Debon,F. et al., 1984. 'The four plutonic belts of Transhimalaya-Himalaya: A chemical-mineralogical, isotopic and chronological synthesis along a Tibet-Nepal section (in press). Dewey,J.F., Burke,K., 1973. 'Tibetan, Variscan and pre-Cambrian basement reactivation: products of a continental collision', Jour. Geol., 81, p.683-692. Duan Xin-Hua, Zhao Hong, 1981. 'The Ailaoshan-Tengtiaohe fracture-the subduction zone of an ancient plate', Acta Geologica Sinica, v.55, no.4, p.258-266. Elliot,D.H., 1975. 'Tectonics of Antarctica: A review', Am.J.Sci., 275-A, 45-106. Gansser,A., 1964. The geology of the Himalayas, Wiley, Intrescience, New York, 289 pp. Gansser,A., 1977. 'The great suture zone between Himalaya and Tibet, a preliminary account', Jest ,C., ed., Collogues internationaux du CNRS, no.268, pp.1183-ll9l. Gao Yan-Lin, 1984. 'Blueschist in China', Bull. of Chinese Academy of Geol. Science, v.lO, p.6l-76. Gass,I.G., 1981. 'Ophiolitic products and processes: Cyprus and the Oman', Proc. Symp. on Qinghai-Xizang (Tibet) plateau, v.l, p.557-566. Girardeau,J. et al., 1984. of significance Bangong-Nujiang suture p.27-31. the 'Tectonic environment and geodynamics Neo-Cimmerian Dongqiao ophiolite, zone, Tibet', Nature, v.307, no.5946, Gu Qing-Ge et a1., 1980. 'Discovery of the late Anisian Paraceratites trinodosus fauna (Ammonoidea) from Duilungdeqen, Tibet and its significance', Acta Palaeontologica Sinica, v.19, no.5, p.343-356. Hallam,A., 1984. 'The unlikelihood of an expanding Earth', Geol. Magazine, v.12l, p.653-655. Honegger,K. et a1., 1982. 'Magmatism and metamorphism in the Ladakh Himalayas (The Indus-Tsangpo suture zone)', Earth Planet. Sci. Let. , v.60, p.253-292. Huang,T.K., 1980. 'An outline of the tectonic characteristics of China', Continental Tectonics: Studies in Geophysics, National Academy of Sciences, Washington,D.C., p.84l-866. Hutchison,C.A., 1976. Ophiolite in Southeast Asia', Geol. Soc. Am. Bull., v.86, p.797-806. Kravchenko,K.N., 1979. 'Tectonic evolution of the Tian Shan, Pamir and Karakorum', Farah,A. & DeJong,K.A. eds., Geodynamics of Pakistan, Geol. Surv. Pakistan, Quetta, p.25-40. Le Fort ,P. et al., 1981. 'Lower Paleozoic "LesserHimalayan granitic belt": emphasis on the Simchar pluton of central Nepal', Shams,F.A. ed. Granites of the Himalaya, Karakorum and Hindu Kush, Lohore, Panjab Univ. Li Xing-Xue, Yao Zhao-Qi, 1981. 'Discovery of Cathaysia flora in the Qinghai-Xizang plateau with special reference to its Permian phytogeographical provinces', Proc. Symp. on Qinghai-Xizang (Tibet) plateau, v.l, p.145-l48. 475 Li Xing-Xue et al., 1984. 'Preliminary study on a mixed Permian flora from Xiagangjiang of Gerze district, Xizang and its paleogeographic significance', Abstracts(Supplement) of Intern. Symp. Himalayan Geology, Chengdu, China, 3-4. Liang Ding-Yi et al., 1983. 'Permo-Carboniferous Gondwana-Tethys facies in southern Karakorum,Ali, Xizang (Tibet)', Earth Science Jour. of Wuhan Geological College, No.1, p.9-27. Liu Zeng-Qian et al., 1980. 'The basic characteristics of geology of Qinghai-Xizang (Tibet) plateau'. Bulletin of the Chinese Academy of Geol. Science., v.2, no.l, p.23-46. Mitchell,A.H.F., 1977. 'Tectonic settings for emplacement of southeast Asian tin granites', Geol. Soc. Malaysia Bull., v.9, p.123-l40. Mitchell,A.H.F., 1981. 'Phanerozoic plate boundaries in mainland SE Asia, Himalayas and Tibet', Jour. Geol. Soc. London, v.130, p.109-l22. Molnar,P., Tapponnier,P., 1975. 'Cenozoic tectonics of Asia: effects of a continental collision', Science, v.189, p.4l9-426. Molnar,P., Tapponnier,P., 1978. 'Active tectonics of Tibet', J. Geophys. Res., v.83, p.536l-5375. Noring,E., 1946. 'Geological exploration in western Tibet', Sino-Swedish expedition publication (29), Stockhom, p.47-89. Pearce ,J.A., et al., 1984. 'Characteristics and tectonic significance of supra-subduction zone ophiolites', Kokelaar,B.P. & Howells,M.F., eds. Marginal Basin geology, publ. forthe Geol. Soc. by Blackwell Sic. Pub. Powel,C.M., Conaghan,P.J., 1981. 'Plate tectonics and the Himalayas', Earth Planet Sci. Letter, (20), 1-2. Ren Ji-Shun et al., 1980. The geotectonic and its evolution of China, Science Press, Beijing, (in Chinese) Ridd,M.F., 1971. 'The Phuket group of Peninsular Thailand', Geol. Mag., 108, p.445-446. Rothery,D.A., Drury,S.A., 1984. 'The neotectonics of the Tibetan plateau', Tectonics, v.3, no.l, p.19-26. ~engor,A.M.C., 1981. 'The evolution of Paleo-Tethys in the Tibetan segment of the Alpide', Proc. Symp. on Qinghai-Xizang (Tibet) plateau, v.l, p.5l-56. ~engor,A.M.C., 1984. 'The Cimmeride orogenic system and the tectonics of Eurasia', Special Paper, 195, Geol. Soc. Am. Tectonic map compilatory group, Inst. of Geol., Academia Sinica, 1974. 'A preliminary note on the basic tectonic features and their developments in China', Scientia Geol. Sinica, p.1-17. Wang Cheng-Shan etal., 1984. 'The geological features of Qangtang block and the tectonc pattern of Paleo-Tethys', Intern. Symp. Himalayan Geol., Chengdu, China, Abstracts (Supplement), p.5-9 (in Chinese) Hang Nai-Wen, 1980. 'Qingzang-India paleocontinent and its welding to Cathaysia', Li Guang-Cen & Mercier,J.L. eds. Sino-French Cooperative Investigation in Himalayas, Geol. Publ. House, Beijing, 1984, p.39-58 (in Chinese). 476 Wu Ji-Yuan, 1975. 'On Ailaoshan tectonic belt and plate tectonics', Dizhi Keji, p.3l-39, (in Chinese) Xu Rong-Hua, Jin Cheng-Wei, 1985, 'A geochronological study of the Quku batholith, Xizang', Scientia Geol. Sinica, 4, 414-422. Xu Rong-Hua, 1985. 'A geochronolgical study of the granitoids of central part of Tibet', (in press). Yin Hong-Fu, 1981. 'Paleogeographical and stratigraphical distribution of the lower Triassic Claraia and Eumorphotis (Bi val via) " Acta Geol. Sinica, v.55, no.3, p.16l-l69. Zheng Bing-Hua, Zheng Jian-Dong, 1984. 'Qinghai-Xizang (Tibet) plateau lineaments and their features of activity', Intern. Symp. Himalayan Geol., Cehgndu, China, Abstract, p.52-54. Zhou Xiang et al., 1985, Tectonic and formation map of Tibetan plates, (in press), (in Chinese). THE NEO-CIMMERIAN OPHIOLITE BELT IN AFGHANISTAN AND TIBET: COMPARISON AND EVOLUTION J .GIRARDEAU*, J .MARCOUX** AND C .MONTENAT*>!'* ABSTRACT. From the Farah Rud belt in Afghanistan to the Donqiao ophiolite belt in Tibet, similar magmatic and sedimentary series crop out which were formed or deposited in the same time span. They point to the existence of an EW-trending oceanic domain during Triassic time within a major continental domain separating the two main oceans (Palaeo and Neo-Tethys) which were the result of break-up of Gondwana. This ocean opened in late Palaeozoic time and began to close during the Middle or Late Jurassic. A northward-dipping subduction zone was active in the northern part of this ocean during the Triassic resulting in the formation of the Donqiao-Xainxa ophiolite (Tibet) behind the subduction zone and in the plutonism of the Hindu Kush - Badakhsan complex (Afghanistan) at the southern margin of Eurasia. The ophiolite in Tibet was obducted in late Jurassic time as indicated by ophiolite olistoliths locally occuring in Upper Jurassic limestones (Ilansen olisthostrome Formation) and covered by transgressive shallow marine to continental sediments (Zigetang Formation). During Aptian and Albian times, shallow marine to continental sequences were deposited in Afghanistan as well as in Tibet. A late backthrusting event postdating the deposition of the Albian series in Tibet and of a Palaeogene series in Afghanistan is responsible for the main structures now observed in these regions. * Laboratoire de Petrologie Physique, Universite Paris 7 et Institut de Physique du Globe de Paris, Place Jussieu, 75230 Paris cedex OS, France. ** Laboratoire de tectonique, Universite Paris 7, Place Jussieu, 75230 Paris cedex OS, France. *** Laboratoire de Geologie, Institut Geologique Albert-de-Lapparent, 21 Rue d'Assas, 75270 Paris cedex 06, France. 477 A. M. C. iJengor (ed.), Tectollic Evolution of the Tethyan Region, 477-504. © 1989 by Kluwer Academic Publishers. 478 1. INTRODUCTION Since Palaeozoic time, after the break-up of Gondwana several continental blocks moved from south to north. Their subsequent accretion to Laurasia has led to the formation of the Tibetan plateau 1,2,3,4,5,6 . The Palaeo and Neo-Tethys, which during Palaeozoic to middle Cainozoic times were the two major oceanic domains resulting from the Gondwana break-up, were separated by one or several major continental blocks which may have formed a large continental domain sometimes referred to as the "Cimmerian continent" 7,8,9 . The northern limit of these continental blocks with Laurasia is marked by a major suture zone, recognized from Afganistan (Herat fault) in the west, to Tibet (Hoh Xil suture zone) and Indochina (Dien Bien Phu suture zone) in the east and their southern border is the Eocene suture zone well known in Afghanistan (Waziristan suture zone) and Tibet (Indus-Yarlung Zangbo suture zone 10 . A third suture zone of late Jurassic age has been recognized midway between the two others 1,2 . It can be followed from Afganistan (Farah Rud suture zone) in the west, to Tibet (Bangong-Nu Jiang suture zone) and Burma (Sittang Valley, Myitkyina suture zone) in the east. The ophiolite remnants cropping out all along this suture zone are according to §engor 7,8,9 the remnants of a small ocean basin the formation of which would be related to a southward subduction of Palaeo-Tethys during Palaeozoic to Triassic times; this interpretation, though, is highly debatable. Very few data are as yet available on this suture zone except for Afghanistan and Tibet where we have mapped several sections across or south of the suture zone. In this paper, we will characterize the nature and evolution of this oceanic domain exemplified by ophiolite remnants and by thick sedimentary sequences deposited since Triassic times which crop out everywhere along the suture zone. Our data will also be compared to those of Pamir and Burma. II. BLOCKS AND SUTURE ZONES FROM AFGHANISTAN TO BURMA The continuity between the tectonic blocks and related suture zones is not always clear because most of them have been severly telescoped along thrust faults and displaced along major strike-slip faults during the Eocene India-Eurasia collision 5,11,12,13 . Three major SW-NE suture zones have been recognized in Afghanistan (Fig.l). The Waziristan suture zone in the south, separates the Indian plate from the Central Mountains block (Helmend block). The Farah Rud basin is located between the Central Mountains and the Band-e Bayan block in the north. The southern limit of the basin is bounded by the Helmend fault, north of which abundant ophiolite lenses crop out suggesting that this fault results form reactivation of an ancient 10 - 0 -0 - C en o zo ic s u tu re - + -+ - la te J ur a • • lc - lo w er c re ta c e o u s . . . . . . . . . T rl a .s lc s u tu re P a le o zo ic s u tu re B B : B an d- e B ay an s W P : S ou th w e s t Pa m ir SE P : So ut h e a s t Pa m ir A F : A kb a yt a l F a u lt ~ ~JF ~ ~ !': ~:: "h H EL F : H e/ m en d H F : H ef et H R F : H ar l R ud I(K F : K ar ak or um 7 0 · B O · TA R IM / - 'tt l) / - 't tv c M .e .T . M ai n C en tr al T hr us t M .B .T . M ai n B o u n d ar y M .M .T . M ai n M an tl e S tu dy a re a s R. P. R us ha n P sh ar t S V M Z Si tta ng V al ie y- M yi tk yi na Z on e B O ' Fi g. I. B lo ck s a n d s u tu re z o n e s fr o ill A fg ha ni st an t o B ur ill a . 0 0 ' ,.1f ~ TS AI D AM LU N BL O C K J(I L S U TU R E Z O N E TA N G B LO C K R A T o \? " , § " '15 tIl " 'I ~ 'd> \- ~ \d o - ;. ~ .. '! '" 01 . . . - J 'C ! 480 suture zone. The Herat fault marks the limit between the Band-e Bayan block and the Paropamisus - Indu Kush - North Pamir block in the north. It is also underlain by a few ophiolite massifs. A similar geometry is observed further east in Pamir. However, it appears that two suture zones are intercalated between the Indian plate to the south and the South Pamir block to the north. The southern one, parallel to the Main Mantle Thrust (MMT), is characterized by peridotites overlain by gabbros and calc-alkaline volcanic rocks 14,15 . The northern one is the Shyok suture zone which bounds the South Pamir block in the south. This later block is separated from the Central Pamir block by the Rushan-Pshart ophiolite belt. The Akbaytal fault separates the Central Pamir block in the north from the North Pamir block. This fau] Leatures also a few ophiolite massifs. The extent of the different blocks and related suture zones is well-known in Tibet. In the south, the Yarlung-Zangbo suture zone separates the Indian plate from the Lhasa block, the northern contact of which with the Quantang block is marked by the Bangong-Nu Jiang suture zone. The Hoh Xil suture zone marks the boundary of the Quantang and Kunlun blocks. In Burma, the Nagaland ophiolite belt seems to mark the suture zone between the Indian plate and the Western Burma block. This block is limited in the east by the Sittang-Valley Myitkyina suture zone. The Thai eastern border of Thai Malay Peninsula is the Dien Bien Phu suture zone. III. THE FARAH RUD BELT: CENTRAL AFGHANISTAN The first stratigraphic and structural data on the Farah Rud basin have been published by Slavin and Mirzard 16 . Subsequently, this region has been investigated by many geologists who have characterized its main lithological units 4,17,18,19,20,21,22,23,24,25,26 . The Farah Rud basin (Fig. 2) is located between two continental blocks with Precambrian basement: the Band-e Bayan and the Central Mountains (Helmend) blocks. These blocks are partly covered by Palaeozoic and Mesozoic platform sediments. The Farah Rud basin has a funnel-like shape, openning to the west, and pinching out toward the east along the Gardan Dewal-Maydan range, and cut off by the Chaman fault, west of Kabul. Its northern contact is the steeply north-dipping Band-e-Bayan fault zone and, to the south it extends to the south - south-east dipping Helmend fault zone. Several major lithological units have been distinguished within the Farah Rud basin. These are: 1) an ultramafic series mainly cropping out along the Helmend fault zone, 2) volcanic and sedimentary series part of which is Triassic to Jurassic named the War as series which has 481 undergone a greenschist to amphibolite metamorphism, 3) the Jurassic to Neocomian flysch of Panjaw, 4) Lower Cretaceous red detrital sediments, calc-alkaline volcanic rocks and orbito1inid limestones, Palaeogene red detrital volcanic-rich series (Red Grits), and 5)(Senonian limestones. The total thickness of these different units may exceed 10 km in the western part of the basin(19). The main characteristics and relationships between the different units have been determined mainly from a section along the road from Yakao1ang (Chatu pass) in the north, to Panjaw, Waras and Chariston (Koh-e Nak pass) in the south (Fig.2-3). 1- The Central Mountains Block The Central Mountains block has a Precambrian basement. It is covered by Palaeozoic and Mesozoic continental-margin deposits. This block has been thrust northward across the Farah Rud trough; the contact zone has been retectonized strongly during late sinistral strike-slip fault movements. 2- The Ultramafic Series The ultramafic rocks from the Farah Rud trough occur as small-sized (several hundreds to thousands of metres long) lenses cropping out within the southern part of the Waras flysch series. These lenses could represent olistoliths. The peridotites are generally completely transformed into secondary serpentine minerals. Only a few lenses in the eastern part of the Farah Rud basin (north of Waras, near Mullah Yaqub NW of Behsud and near Mianah, NE of Serkhmiran), have been investigated. Although the petrological character of these ultramafic rocks has not yet been determined precisely, we can present nevertheless their main mineralogical and textural features. Although most of the peridotites are totally transformed into secondary minerals including 1izardite, chrysoti1e, magnetite and chlorite, some samples are remarkably fresh. These peridotites are either harzburgites, 1herzo1ites, or wehr1ites. The harzburgites contain enstatite (about 20%), olivine (about 80%) and a few percent brown spinel. They display mylonitic textures with very elongate and kinked orthopyroxenes. Olivine is strongly recrystallized into a fine-grained matrix of equant neob1asts and only a few highly-strained porphyroc1asts can be recognized. These peridotites have been deformed at high temperatures by the (010) (100) slip system under large deviatoric stresses (up to 800MPa) according to the recrystallized grain size piezometer. These high-temperature - hig-stress deformation structures are generally related to early intra-oceanic thrusting events or to transform-fault motion. The lherzolite contains about 5% clinopyroxene (diopside), 10% ortho.pyroxene (enstatite), 85% olivine and less than 1% spinel. 482 Clinopyroxene is small-sized and forms discontinuous aggregates parallel to the foliation. Spinel has a reddish color indicating a chromium-rich composi tion. The wehrli tes are unstrained and show cumulate textures. The orlgln of the completely serpentinized peridotites cannot be determined exactly. They seem however to derive from harzburgites or dunites according to their secondary mineral assemblage. They do however contain fresh chromite with pyrrhotite, pentlandi te and magnetite in various amounts (J. Bolze's analyses). 3. The upper Triassic to Upper Jurassic Waras Series The study area is mainly underlain by the Waras series. To the top, this series mainly comprises turbidite deposits formed by rhythmic intercalations of pelites and sandstones, displaying dark-green or blue colours. The turbidites are generally siliciclastites showing flutes, grove casts, bioturbation and slump structures. They usually are very fine-grained and they form beds a few tens of centimetres thick. Some limestone micro-breccias, derived from Triassic or Jurassic carbonate - platform depOSits, are locally intercalated within the flysch. This series, almost devoid of volcanic rocks, is more than 1000 m thick. Near the base of the Waras series, igneous rocks are abundant, forming flows or sills a few meters to tens of meters in thickness. The volcanic rocks display a greenish colour and correspond to tuffs, hyaloclastites, spili tes and pillow lavas. They have a basaltic or andesi tic composition(20,27). A few sill-like dolerite bodies are locally intercalated within the flysch, in close association with the volcanic rocks. They show typical doleritic textures with elongate laths of plagioclase partly transformed into secondary albite, and interstitial amphiboles probably resulting from the alteration of primary pyroxenes. They also contain a few opaque minerals. Chlorite-bearing greenschists are often associated with these volcanic rocks. Peridotite and coarse-grained amphibolite lenses occur in this part of the flysch. Limestones olistoliths, a few tens to hundreds of metres (sometimes more than one kilometre) long, are locally abundant, particularly in the Dekhundi area. These limestones are transformed into marbles in which fragments of Megalodontidae have been found(20). These limestones are similar to the Triassic deposits cropping out in the Central Mountains to the south. A few grey cherts with radiolarian remnants are observed throughout the entire Waras series. These cherts from 10 cm thick beds, inter layered with more pelitic horizons. The age of the Waras series is very poorly defined. In the lower part of the series, some limestone olistoliths contain Megalodontidae of late Triassic age. The age of the surrounding sediments has not been determined. Some deposits of Triassic age have been described in the Navzac series 19,20,24 , but the biostratigraphic data are questionable. In the upper part of the Waras series, the limestone micro breccias contain some fossils. From bottom to top, the microbreccias show reworked microfauna associations of Late Triassic age 28, and ~ ~ __ ~ ~ __ ~~ 50 Km D [ [ ] ~ E J 4 F ig .2 : A- L oc at io n o f th e Fa ra h Ru d ba si n (B : B eh du s; H: H er at ; K: K ab ul ; K h: K an da ha r; P: Pa nja w; c ro s s e s : m ai n T ri as si c gr an it oi ds ). B - G eo lo gi ca l s ke tc h m ap o f th e e a s te rn pa rt o f th e Fa ra h Ru d ba si n( 18 ,2 0, 23 ). 1: C en tr al M ou nt ai ns ba se m en t; 2: B an d- e B ay an ba se m en t; 3: Tu rk m an fl ys ch s e ri es ; 4: O ph io li ti c ro c ks (p er id ot ite s a n d fo li at ed a m ph ib ol it es ); 5: W ar as f ly sc h s e ri es ( a: m ai n v o lc an ic o u tc ro ps ); 6: P an jaw f ly sc h s e ri es ; 7: Se no ni an li m es to ne s; 8: K oh -e B ab a O lig oc en e pl ut on ic an d v o lc an ic m a s s if . Th e Re d G ri ts a ll c ro pp in g o u t a lo ng th e m a in fa ul ts c ro s s c u tt in g th e Pa nja w fl ys ch , a re n o t re pr es en te d. B: B eh su d; B a: Ba m ya n; D: D ek hu nd i; G: G ar da n D ew al ; M : M ay da n v a ll ey ; P: Pa nja w; S : Se rk hm ir an ; W : W ar as ; Y: Y ak ol an g. . . . 00 w ss w lJ c lJd K OH -c N AK Pa ss CE NT RA L M OU NT AI NS W AR AS PA NJ AW o FA RA H RU D BE LT ( L ,5 0 Km c n y. ) CH AT U P u s NN E BA ND -c BA YA N F ig .3 : Sc he m at ic s e c ti on a c ro s s th e Fa ra h Ru d ba si n. 1: M et am or ph ic ba se m en t o f th e B an d- e B ay an bl oc k; 2: Re d d et ri ta l s e ri es o f C ha tu pa ss ; a : a n de si ti c fl ow s da te d a t 67 M a w hi ch c ro p o u t n e a r th e ba se o f th e Pa la eo ge ne s e ri es ; 3: U pp er C re ta ce ou s li m es to ne s (T ur on ia ll- Se no ni an ); 4: P el it e an d s a n ds to ne s o f th e Pa nja w fl ys ch s e ri es (th e fl ys ch , th ic kn es s o f w hi ch is up to 10 00 m , ha s u n de rg on e a lo w -g ra de m et am or ph ism m ar ke d by s e r ic it e) ; 4a : T it ho ni c a m m o n it e- be ar in g s e r ie s; 4 b: m e tr e th ic k c o a l- be ar in g p h yl li te s; 4c : s il ic ic la st ic s e ri es w it h a fe w re -d ep os i t ed c a rb on at es ; 5: Se no ni an li m es to ne s o v e rl yi ng u n c o n fo rm ab ly th e Ju ra ss ic fl ys ch s e ri es , th e Lo w er C re ta ce ou s Re d G ri ts a n d th e B an d- e B ay an ba se m en t; 6: R ed G ri ts c o v e re d by A pt ia n n e r it ic l im es to ne s; 7: Lo w -g ra de m e ta m or ph ic W ar as f ly sc h s e ri es ( lo ca lly c o n ta in in g a bu nd an t v o lc an ic r o c ks ); 7a : pi ll ow l av as ; 7b : rh yt hm ic i nt er ca la ti on s o f s c hi st s an d qu ar tz it es (5 00 m ), w it h v o lc an ic tu ff s, hy al oc la st it es , gr ee n s c hi st s, an d c a rb on at e o li st o li th s; 7c : Ta kh ak gr ey r a di ol ar it es (5 0 m ); 7d : r hy th m ic in te rc al at io n s o f q u ar tz it es , s e r is it e- b ea ri n g s c h is ts , c a r bo na te o li st o li th s, s c a r c e h y a lo cl a st it es , a n d li m es to n e m ic ro b re cc ia s c o n ta in in g r e w o r ke d T ri as si c to Ju ra ss ic m a te ri al s; 7e : da rk -b lu e s c h is ts r ic h in s a n ds to ne tu rb id it e a n d s la te be ds (a bo ut 1 00 0 m ); 8: Re d G ri ts (a bo ut 5 00 m ; pr es en ce o f a ba sa l c o n gl om er at e o f re w o rk ed fl ys ch , o v e rl ai n by m o re p el it ic be ds w it h a fe w c o n gl om er at e be ds ); 9: W ar as fl ys ch s e ri es (g re en sc hi st f ac ie s) ; 9a : s c hi st s a n d qu ar tz it es (t ur bi di te s) w ith in te rc al at io n o f c a r bo na te m ic ro b re cc ia s (p re se nc e o f r e w o r ke d Ju ra ss ic c a n a bi ne o o li te s) w it h s e v e r a l te n -m et re -t h ic k v o lc an ic fl ow s an d c a rb on at e o li st o li th s; 9b : s c hi st s in be dd in g s e v e ra l te n s- o f- m et re s lo ng u lt ra m af ic l en se s an d c a rb on at e o li st o li th s (th e fl ys ch is e x tr em el y fo ld ed ; o v e r a bo ut 5 00 m i n th ic kn es s) ; 9c : gr ey a n d gr ee n s e ri ci te -b ea ri ng s c hi st s (a bo ut 1 00 0 m ) ; 10 : Re d G ri ts ; lO a: in te rc al at io ns o f tu ff s, c o n gl om er at es a n d a n de si te b re cc ia s da te d a t 31 M a; lO b: a n de si te a n d m a rb le b ea ri ng c o n gl om er at e o v e rl ai n by p el it es , s il ts to n e a n d sa n ds to ne be ds , a n d by c o n gl om er at es w it h gr ee n s c h is ts , m ic as ch is ts , m a r bl es , a n d gn ei ss es p eb bl es ; 11 : D ar k c r y st a ll in e s c h is ts , m ic as ch is ts , m a r bl es , a n d gn ei ss es p eb bl es ; 11 : D ar k c r y st a ll in e s c h is ts o v e r la in by ba nd ed m a r bl es (t he c a rb on at es a re s im il ar to th e P er m ia n- T ri as si c s e ri es o f th e C en tr al M ou nt ai ns ); 12 : W ar as fl ys ch s e ri es (m eta mo rp his m in t he lo w er a m p h ib ol it e fa ci es ); l2 a: r hy th m ic in te rc al at io n o f b io ti te m ic as ch is ts , fi n e- gr ai n ed gn ei ss es a n d q u a rt zi te s. T h is u n it , s e v e r a l th ou sa nd m e tr es in th ic kn es s, c o n ta in s a bu nd an t u lt ra m af ic , m a fi c (a m ph ib ol ite s) , qU qr tz it e an d c a rb on at e o li st o li th s; l2 b: s m a ll -s iz ed m e ta m or ph ic be dd ed c he rt s (r ad io la ri te s) ; l2 c: sa m e a s 12 a w it h v o lc an ic ro c ks ; 12 d: rh yt hm ic in te rc al at io ns o f m ic as ch is ts an d fi ne -g ra in ed gn ei ss es ; 13 : S er ie s o f th e n o rt he rn m a rg in o f th e C en tr al M ou nt ai ns b lo ck ; l3 a: B eh su d u n m et am or pi 1o se d Pr ec am br ia n s e ri es ; l3 b: m ig m at it es de ri ve d fr om th e B eh su d s e ri es ; 13 c: gr an it e in tr us iv e in th e B eh su d s e ri es (th e gr an it e an d su r r o u n di ng m ig m at it es ha ve be en da te d a t 49 5 M a) ; 13 d: P er m ia n- T ri as si c li m es to n es ly in g u n c o n fo rm ab ly o v e r th e P re ca m br ia n B eh su d s e r ie s; 13 e: B eh su d P re ca m br ia n s e r ie s w it h s e r p en ti n it es f or m in g 10 t o 10 0 m i n tr u si on -l ik e b od ie s; 13 f: fa u lt ed z o n e v -ii th m yl on it ic pe ri do ti te . . j>- 00 . j>- 485 Cladocoropsis mirabilis Felix fragments, indicating a Malm age. Thus, the age of the Waras series may range from the Late Trias to Late Jurassic (Malm). The Waras series has undergone a metamorphic event. Metamorphism grades continuously from greenschist facies in the north (chlorite-sericite-albite-muscovite) to amphibolite facies south of Waras and in the eastern part of the Maydan Range. The structure of the flysch is still recognizable in the highest-grade metamorphic areas where biotite-garnet micaschists are intercalated with fine-grained locally quartz-rich gneisses and a few quartzite lenses derived from cherts. CIS structures, and helicitic garnets are abundant. Some coarse-grained amphibolites, probably derived from gabbros, form lenses within the micaschists. These amphibolites show an hornblende - zoizite - sphene assemblage of the amphibolite facies. The hornblende is locally rimmed by a thin mantle of blue amphiboles identified as crossites, indicating that these rocks have undergone a low temperature intermediate pressure metamorphic event. Furthermore, the limestone lenses are transformed into marbles. All these series are strongly tectoni zed , being crosscut by abundant south-facing reverse faults. They show a strong south-facing schistosi ty, parallel to the axial planes of tight isoclinal folds. South of the study area, the folds clearly show a northward vergence. In the Gardan Dewal-Maydan Range, the series shows north-facing isoclinal folds. The deformations have been related to EW dextral strike-slip fault motion(23,29). 4. The Upper Lias to Neocomian Panjaw Flysch Series This series crops out north of the Waras series from which it is separated by a vertical fault zone (Fig.2b,3). This series is generally not metamorphic and it is characterized by the absence of volcanic rocks and olistoliths. It shows a clear polarity from north to south 30 . In the north, the series is rich in allodapic limestone deposits forming lenses or beds intercalated within pelite and sandstones beds. These are dark corals-and sponges-(Cladocoropsis) bearing limestones, and blocks and grain-flow consisting of bioclastic and oolite sand-bearing limestones which were probably derived from the northern margin of the basin (Jurassic limestone deposits of the Band-e Bayan platform). The Panjaw flysch also contains coal-bearing schists (north of Panjaw) formi~g thin beds intercalated with pelites and sandstones. In the sandstones, some plant and cephalopod fragments have been found. Locally, red phyllite occurs at the top of this series. Towards the south, this series thickens greatly and consists of monotonous siliciclastic turbidite deposits with Helminthoides traces, while allodapic limestone deposits become less and less abundant. 486 Some fossils have been found within the Panjaw flysch. North of the village of Panjaw, it contains Hemispi ticeras steinmanni STAUR indicating a Tithonian age(3l). It also has Cladocoropsis mirabilis FELIX in reefal limestones indicating an late Malm age. In the center of the basin south of Panjaw, limestones interbedded within the flysch have yielded various Calpionellids of Berriasian Valanginian age(l8,30). The continental shelf deposits of the Band-e Bayan north of the Panjaw basin, mainly consist of shallow marine, locally reefal carbonates. The age of these sediments varies from Late Lias to Malm(18,20,32). Thus, the age of the Panjaw flysch may range from Late Lias to Neocomian (Valanginian). The Panjaw flysch has undergone a very low grade metamorphic event (sericite-chlorite). As the War as series, this flysch has been strongly tectonized. It shows tight isoclinal folds with subvertical axial planes and cleavages(19). The folds are slightly overturned to the north. East of Panjaw, the flysch is intruded by the Oligocene rhyolite dikes of the Koh-e Baba. At the contact, the flysch series is deformed and has undergone high-temperature contact metamorphism. 5. The Mesozoic and Cenozoic Molasse Series These detrital series(33) unconformably overlie all the series described previously. In the eastern part of the study area, these detrital rocks crop out in discontinuous and narrow depressions, lying along major late EW strike-slip faluts. In the west, they crop out over much larger areas and the total thickness of the deposits exceeds several thousands of meters. Two types of Red Grits, with distinct ages, have been distinguished: a) The first one comprises mainly red terrigenous deposits, consisting of reworked epimetamorphic material from the Panjaw flysch. This series often contains andesi tic or rhyodaci tic volcanic rocks, overlain by marine bioclastic limestones containing Orbitolinids of Barremian to Albian age(2l). b) The second does not contain any marine deposits: (i t consists of reworked material derived from the metamorphic rocks of the Waras series. Andesite breccias of Oligocene age (31 to 37 Ma; Bellon(34() are abundant within these Red Grits. 6. The Senonian Massive Limestones Locally, the Lower Cretaceous Red Grits, Band-e Bayan basement (Panjaw area) are the Panjaw flysch and the disconformably overlain by 487 well-bedded massive limestones. Some limestones have a micritic texture and contain abundant silex, other have a bioclastic texture and contain rudists. The former has yielded Pithonella ovalis and Globotruncana lapparenti indicating a Senonian age. 7. The Band-e Bayan Block In the north, the Farah Rud trough is overthrust to the south by the Band-e Bayan block. As the Central-Mountains block, this block has a Precambrian basement and is covered by Plaeozoic and Mesozoic platform series. 8. Lateral Extension of the Farah Rud Trough The Farah Rud trough and the Central Mountains block do not extend to the west in Iran. They are truncated by the nearly-NS Hari-Rud fault(25) which represents a major tectonic discontinuity. In the east of the Siahsang fault, the Turkman basin contains a sequence several thousand meters thick of mainly quartz-rich schists. Toward the base, the series contains a few limestone horizons of probable lower to middle Palaeozoic age. The upper parts of the series, named "schistes et quartzites du Haut Helmend"(23), contains a few Permian goniatites(26) but has probably a Carboniferous to Triassic age(17). Toward the top of the series in the northern part of the basin, some limestone microbreccia crop out, containing elements of carboniferous, Permian and Triassic microfaunas(28). These microbreccias overlie some radiolarites and greenschists. IV. THE RUSHAN-PSHART BELT: CENTRAL PAMIR The relationship between Afghanistan (Farah Rud) and Pamir (Rushan Pshart) has been discussed first by Karapetov et a1. (24). The most recent Russian literature(35,36,37,38) gives more details on the stratigraphy, structure and geodynamic evolution of this area. The Rushan Pshart zone is located between the Central Pamir block to the north and the South Pamir block. The Central Pamir block is a continental block equivalent to the Band-e Bayan block to the west. It has a Precambrian basement and is covered by lower Palaeozoic to Palaeogene series(25,37). The Rushan Pshart area contains several distinct strongly tectonized lithostratigraphic units., The lowermost is represented by Carboniferous to Lower Permian terrigenous series. The first appearance of deep-water sediments represented by radiolarites, flysch and allodapic limestones, occurred during late Permian. A rifting event 488 PLATE + + 6 -:-:.:- 7 ~++ D··· + + + •••• Fig.4: General geological sketch map of the Bangong-Nu Jiang suture zone in Northern Tibet (from the 1/1 500 000 scale Chinese geological map, modified), 1: Triassic series; 2: Palaeozoic series; 3: Jurassic series; 4: Cretaceous series (v: volcanic rocks); 5: Continental (Eocene ?) series; 6: Intrusive granites; 7: North Lhasa Block crystalline basement; 8: Ophiolites. Detailed maps of the outlined areas have been published elsewhere(40,42). S - - - - - - - - - - - - - LH A S A B LO C K _ _ _ _ _ _ _ _ _ _ _ _ _ _ ~~ ~T ~~ GN B LO C K A M D O A B 11 • B LO C K _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ c 10 km F ig .5 : Sc he m at ic S N c ro s s -s e c ti on t hr ou gh t he B an go ng N u Ji an g s u tu re z o n e . Fr om e a s t to w e s t: A: Am do a re a ; B: D on qi ao a re a , a n d C: X ai nx a a re a . 1: M id dl e C re ta ce ou s v o lc an ic a n d re d d et ri ta l s e r ie s; 2: Ju ra ss ic li m es to ne a n d s la te s e ri es i n th e Qu an tan g bl oc k; 3 : C ry st al li ne b as em en t o f th e Lh as a bl oc k( th es e ro c ks h av e fo rm ed 53 1 M a a go , u n de rw en t a m e ta m or ph ic e v e n t a t a bo ut 17 0 M a; R on g- H ua Xu e t a l. 4 3 ); 4: ba nd ed gn ei ss es , a m ph ib ol it es a n d m a rb le s (m ); 5: Ju ra ss ic fl ys ch s e ri se ; 6: o ph io li ti c s e ri es (o lis to lit hi c fo rm at io n o f Il an se n: bl oc ks o f pe ri do ti te s a n d la ye re d ga bb ro s a re re w o rk ed i n s a n ds to ne s a n d li m es to ne s o f L at e Ju ra ss ic a ge ; 7: Ju ra ss ic fl ys ch s e ri es (M aIm ); 8: Z ig et an g c o n ti ne nt al to sh al lo w m a ri ne Fo rm at io n (U pp er Ju ra ss ic -L ow er C re ta ce ou s) ; 9: Pa la eo zo ic li m es to ne s (th es e ro c ks ha ve u n de rg on e a lo w -g ra de m e ta m or ph ic e v e n t); 10 : pi ll ow la va s in te rb ed de d w it h th e Ju ra ss ic fl ys ch (M aIm ); 11 : c hr om it e- ri ch ha rz bu rg it es an d du ni te s; 12 : ha rz bu rg it es a n d du ni te s; 13 : Ju ra ss ic fl ys ch s e ri es (A al en ia n) ; 14 : P il lo w la va s an d la va fl ow s w it h a fe w is ot ro pi c ga bb ro s in a s e rp en ti ni te b re cc ia (N alo ng a re a ); 15 : re d pe li te s an d s a n ds to ne s (c on ta in s om e w oo d fo ss il s an d o rb it ol in oi ds o f A lb ia n a ge (4 4) ; 16 : L at e in tr us iv e gr an it e a n d rh yo li te c o v e re d to th e n o rt h by Eo ce ne (? ) se di m en ts ; 17 : Lo w er C re ta ce ou s (p el i te s a n d s a n ds zt on es ) s e r ie s( at th e c o n ta ct w it h th e in tr us iv es , th ey u n de rw en t a hi gh -t em pe ra tu re c o n ta ct m e ta m or ph is m ); 18 : A lb ia n li m es to ne s (r ic h in O rb it ol in oi ds an d m a c ro fo ss il s, G ir ar de au e t a l. (4 2) ; 19 : H ar zb ur gi te s an d du ni te s w it h a fe w c hr om it e de po si ts ; 20 : S er pe nt in it e br ec ci a c o n ta in in g U pp er J ur as si c - Lo w er C re ta ce ou s li m es to ne le ns es (Z ig et an g Fo rm at io n) ; 21 : P al ae oz oi c s e ri es (a a n d b: Pe rm ia n (? ) an d C ar bo ni fe ro us li m es to ne s; c , d a n d e : D ev on ia n, S il ur ia n a n d O rd ov ic ia n p el it es , s ha le s, s il ts to n es a n d s la te s w it h a fe w be dd ed l im es to ne s) . Th es e s e ri es h as y ie ld ed a bu nd an t fo ss il s( 45 ,4 6) . ~ 00 '> D 490 within the continental block at that time is indicated by the occurrence of large transitional basalt flows within the flysch. The same types of sedimentary and volcanic rocks have been deposited during the entire Triassic. These series are about 500 m thick. Abundant magmatism (granodiorite intrusions) occurred in the northern part of the Rushan Pshart basin at the end of the Triassic and during the early Jurassic. A 700 m thick flysch series comprising pelagic cherts, sandstones and carbonate olistoliths (of lower to upper Palaeozoic age), formed during Jurassic, particularly in the eastern Rushan Pshart zone. Terrigenous red and grey detrital sediments of Lower Cretaceous age were deposited disconformably over the series previously described. Granites were emplaced in the southern part of the Rushan Pshart at that time. Remnants of oceanic lithosphere are very scarce and occur as tectonic slices at the southern margin of the Rushan Pshart. All the series were tectoni zed during Cenozoic time due to the India-Eurasia collision. The late intense northward backthrusting makes a palinspastic reconstruction of this area very difficult. V. THE BANGONG-DONQIAO-NU JIANG OPHIOLITE BELT: CENTRAL TIBET The first detailed data concerning the Bangong-Donqiao-Nu Jiang suture zone in Tibet have been published by Chinese geologists(l, 2). More recent data have been obtained by French and Chinese geologists during the 1980 to 1983 joint project in Tibet(6,39,40,4l). The Bangong Nu Jiang suture zone is the boundary between the Lhasa block to the south and the Quantang block to the north (Fig .1). This suture zone extends in a nearly EW direction over several hundreds of kilometres and is marked by isolated massifs of ophiolites. Such massifs are abundant in the Amdo, Donqiao, Gyanco and Xainxa areas (Fig.4) where they have been studied in detail(40,4l). To the south of the Quantang block, several lithological units have been distinguished (Fig.4,5). These are: l)(a crystalline basement, 2)(an ophiolitic series, 3) a Palaeozoic series, 4)(a Jurassic flysch series, 5)(an Upper Jurassic-Lower Cretaceous marine to continental series and 6) (a Mesozoic sedimentary and volcanic series. The relationships between all these series have been largely discussed by Girardeau et.al.(40,42). 1. The Quantang Block Very few data are available on the Quantang block, north of the suture zone (Fig.l,4,5). The Chinese geological map of the area shows that this block is mostly covered by marine Jurassic limestones, overlying Triassic sediments. In the west, these sediments overlie some Palaeozoic rocks disconformably. They are themselves unconformably overlain by a few volcanic and red detrital rocks of Cenozoic age. All these series 491 seems to have undergone a large symmetrical folding event( 6). Plutonic and volcanic rocks seem to be rare within the Quantang block. 2. The Crystalline Basement The crystalline basement crops out south of Amdo (Fig.4,5). It consists of banded gneisses and amphibolites covered in the south by a few marbles. All these metamorphic rocks are crosscut by granites. The basement does not extend northward across a major EW fault (Amdo graben; Armijo et al.(47(). In the south, it is overlain by the Jurassic flysch series, but the contact between the two has not been observed. The gneisses may have crystallized 531Ma ago, according to Rong-Hua Xu et al.(43). They may have undergone a low-grade metamorphic event at about l71Ma, ascribed to the collision of the Lhasa and Quantang blocks. Late intrusive granites have been emplaced between 140 and 121 Ma, probably related to intra, block tectonic events(43). 3. The Ophiolitic Series Although strongly dismembered, a complete ophiolitic sequence has been discovered in the Donqiao-Gyan~o area(40). This ophiolite sequence comprises strongly deformed bedded cherts, and undeformed pillow lavas, sheeted dolerites, isotropic and cumulate gabbros, wherlites, dunites, depleted harzburgites sometimes rich in podiform chromite deposits, and some foliated amphibolites representing the remnants of a metamorphic sole related to an intra-oceanic thrusting event. The ophiolite forms small klippen thrust from north to south over the Jurassic flysch series. Based on lead isotope studies of the mafic rocks(48), trace element data(49), major element, petrological and textural data of the mafic and ultramafic rocks(50) , the Donqiao-Xainxa ophiolite has been interpreted as having formed in a supra-subduction zone environment. The age of this ophiolite may be Triassic according to radiolaria found within strongly deformed bedded cherts associated with the volcanic part of the ophiolite sequence(5l). A 220-220 Ma age has been obtained for the formation of the oph&&li~ and for its primary intra-oceanic thrusting on the basis of Ar / Ar spectra on magmatic and metamorphic amphiboles from isotropic gabbros and foliated amphibolites from the ophiolite pile, respectively(52). 4. The Palaeozoic Series Palaeozoic rocks are best exposed in the Xainxa area (about 250 km south of the suture zone), where they form a 500 m thick sequence, which was thrust to the north over an ultramafic slice(53). In the lower parts, this series mainly consists of pelites, shales, siltstones and slates 492 inter layered with thin limestone beds. Near the top, the sequence is mostly calcareous, consisting of grey bioclastic limestones forming beds several tens of meters in thickness. The age of this very fossiliferous series rich in graptolites, conodonts, brachiopods, ammonoids, tabulate corals and a few trilobites, may vary from Middle Ordovician to Lower Permian(45,46,54). A few Palaeozoic rocks are also observed in the Donqiao area, south of the suture zone. These rocks consist of ten meters thick limestone beds with minor yellow banded quartzites and black shales over less than one hundred meters thick. Whereas in the Xainxa area these series are probably autochtonous, they clearly form small klippen thrust onto the underlying Jurassic flysch, similar to the ophiolites in the Donqiao area. In both areas, these series have undergone a low-grade metamorphic event which, however, is of higher grade near Donqiao. These rocks are deformed into folds slightly overturned to the south, but without significant cleavage. 5. The Jurassic Flysch Series A widespread very monotonous flysch series crops out in the Donqiao-Gyanc;:o area. It consists of rhythmic intercalations of dark shales, black slates, and grey to yellow limestones forming beds a few tens of centimeters in thickness. South of Gyanc;:o, this series contains a few re-sedimented bioclastic limestone lenses, several meters in size, in which the following microfossils have been found: Gutnicella (Lucassela) cayeuxi, Haurania sp and Nauticulina sp, yielding a Middle Jurassic (Aalenian) age. Near Donqiao, thick quartzi tic beds are locally intercalated with the flysch. Locally, it is intruded by thick (more than 10 meters) andesitic sills. A pillow lava unit, about 30 m in thickness, is locally interbedded with the flysch. This flysch is coarse-grained in the Donqiao area, and fines toward the south. These pillow lavas are covered by reddish limestone encrustations and by a few bedded cherts. Fossils (gastropods, Sequania sp) found in this area, indicate that the flysch may be Upper Jurassic (Malm). About 100 km south of Amdo (Ilansen area), the ophiolitic series are reworked wi thin an olistostrome formation thrust over the Jurassic flysch series(40). Blocks of ophiolitic rocks are locally enclosed in sandstones and shallow marine limestones containing gastropods and corals of probable Late Jurassic age. Far to the east of Donqiao, abundant volcanic horizons are intercalated with the flysch, which may have an early Jurassic age, based on ammonites(55). Flysch members south of the suture zone, considered as Triassic by Chinese geologists, on the basis of stratigraphic considerations(55) have never yielded fossils. 493 This flysch constitutes the relative autochton for the ophiolitic series. It has undergone a low-grade metamorphic event. In the Donqiao area, the flysch displays a subvertica1 to south-dipping cleavage. Southward, it locally shows large asymmetrical folds with subhorizonta1 axes and a slightly north-dipping cleavage. These folds have been ascribed to the late southward emplacement of the ophiolit'ic series as also evidenced by deformation structures in the Palaeozoic series(40). 6. The Upper Jurassic - Lower Cretaceous Zigetang Formation Locally, as in the Donqiao area the ophiolite series are unconformably overlain by a transgressive continental to shallow marine formation: the Zigetang Formation(40). The oldest deposits of this formation is a basal conglomerate containing pebbles of ophiolitic material. It is overlain by a detrital member with some limestone intercalations. towards the top, terrigenous influx decreases and the sedimentary deposits are dominantly shallow marine carbonates, with isolated patchreefs containing stromatoporoids and sponges. This formation contains abundant foraminifera and algae yielding a Late Jurassic-Early Cretaceous age(40). This formation has also been found in the Xainxa area, far to the south(42). 7. The Mesozoic Sedimentary and Volcanic Series Red detrital and volcanic series uncomformab1y cover all the units previously described. These series locally begin (Gyan~o area) by volcanic (basalts) flows covered by coarse detrital co~glomerate beds with basalt pebbles. They are overlain by red pe1ites and sandstones and by a thick sequence of basalts with minor andesites and ignimbrites(40,56). The age of the red detrit~6 r03~s in the Gyan~o area is Albian based on orbito1inids(44). Several Ar/ Ar radiometric whole rock ages obtained from volcanic rocks indicate that they were deposited 110 to 675 Ma ago(52). In the Xainxa area (about 250 km to the south), a thick (several hundreds of metres) marine sedimentary formation overlies both the ophiolitic and Palaeozoic series(53). The lower part consists of a thin (less than 300 m) detrital member comprising microcong1omerate and sandstone horizons with a few peli tes. It is sequence mainly composed of grey limestones forming beds several tens of meters in thickness, and by a few pe1ites. This series is rich in fossils including pelecypods, gastropods, algae, and foraminifera. Its age varies from Late Aptian to Early Albian as determined by orbito1inoids(53). A volcanic unit, about 100 m thick, consisting of brownish rhyolites and a few ignimbrites, is interstratified with the Upper Aptian - Lower Albian limestones. These Mesozoic series (continental from the Donqiao-Gyan~o area and marine from the Xainxa area) are not metamorphic. They often show large 494 open symmetrical folds, sometimes gently overturned to the north as in the Xainxa area, with subhorizontal axes and without any cleavage. They often are overlain by overthrust Jurassic flysch or ophiolitic series, as result of late backthrusting events. VI. THE SITTANG VALLEY - MYITKYINA BELT: BURMA The Sittang Valley - Myitkyina zone suture zone separates the Western Burma block in the west from the Thai-Malay peninsula block in the east. Very few data have been published dealing with this area(57,58). According to Mitchell(58), this suture zone may represent the extension of the Bangong Nu Jiang suture zone further north in Tibet. This suture zone is however very little constrained by geological data. Hutchinson(57) reports the presene of ultramafic rocks (harzburgites and large chromite deposits) and mafics lenses (gabbros) which may represent the remnants of a strongly dismembered ophiolite suite. These rocks may be included in a metamorphic assemblage comprising low to high grade (Kyanite-glaucophane) metamorphic schists. In the east, Palaeozoic sediments and metamorphic rocks overlie Precambrian rocks of the Thai Malay Peninsula block. In the west, upper Mesozoic to Cenozoic volcanic and sedimentary series crop out; they may be related to a hypothetical subduction zone located further east. Based on highly speculative arguments, Mitchell (58) has proposed that the closure of the Sittang Valley - Miytkyina suture zone occurred during the mid-Jurassic. All these rocks have been strongly tectonized during Tertiary tectonic events as the Sittang Valley - Myitkyina suture zone has been reactivated by a dextral strike slip fault(59). Approximately 130 km west of this suture zone, Mitchell(58) described ophiolitic remnants( (ul tramafic rocks and basalts) thrusted over an Upper Jurassic flysch series. These ophiolite remnants and the Jurassic flysch are unconformably overlain by Orbitolina-bearing limestones of Albian age. Mitchell(58) considered that these ophiolite remnants correspond to the Eocene Yarlung-Zangbo suture zone which is very debatable as the lithological and structural succession occurring in that area is the same as that observed along the Bangong Nu Jiang suture zone in Tibet. More data are needed to understand the origin, age and significance of these two ophiolitic series which unfortunately crop out in an almost inaccessible area. VII. GEODYNAMIC EVOLUTION OF THE FARAH RUD AND DONQIAO OCEANIC DOMAINS As described above, the rock series deposited in the Farah Rud belt are similar, both in lithology and age, as those in the Donqiao belt as listed in Figure 6. They consist of ophiolitic series, Jurassic flycsh A FC H A N IS IA N I PA H IR II B E I BU RH A 17 ~ . . 1n tl nm e n Jr th et d ba dd .l'l ns tU 'g an t N N N ~ laL er a1 a tr lk &- sh p fa Jl tll 'g ~ ~ ~ t - - - ~ ~ t- " . . < 1" '1 "" - Io :i! :s ltl e n a; r- at lSl l P . . In te rs lW f' lll' t.t wa rd b id t1 l8 1t al le d G r1 ts : rd 65 " ' P I ~ o f ca io -a lk al1 l'lO " " 1" 'tl Sl l 61 61 6' ~ I- - ~ I- - - Ce pa ut 1O 'l o f T UC 'Q" U3 '\..S eru um C re t= o. .s 1. ln es ta es (t C'a 'liQ re5 SlC n) t.C t.C t.C 1m lil l lil l . . , . , _ b a: I< tl1 ru sti rg a -d S )'I I'I 'tri ca i !i ll t - - - I- - I- - fo ld irg . t- ~ . . , _ ba dc -tr us l:1 rg ~ h U l o f re d a ll g re y c 2t ri bc ~ t u n o f re d de tr ib .c s Al b1 a-V - Tr CI "B J! $S la\ o f O rtu to lJr la ~a (w .-E ) NI l cm t.i ne nt al s er ie s . >8 a la g tt e su tu re m -e < n 1 o f B rb l~ - . >8 - te p: :e it ic n Df Ba r", ."" . . te pc m tla 'l o f F e1 G nt s w Ith c aJ .o.. !in a lim os ta "o s fa r to t re ~ cd ub lir e al ka l.l re 'V olC 3'U Sll _ ln tt: nU W l m ag ra tis n (g rln Xf ior ite s) 12 0 - 12 0 ~ 12 0 12 0 to t te S Q. .Ith or ~ f - - - I- - I- - - r r ~ o r a n ti re rt al to t - . . . Ye ta nJ rp us n o f W ara s fly sc h se n e s "" "" "" '" U lr g l ts sc u ttw ar d th ns tu g : dJ d. .ct lC Jl ~ ~ ~ If B fl fl ! ct ;c s. its a .-e r tr e ~ 14 0 to ai rd t .~ s w th ., 111 0 14 0 c pu ol lt ic s er J.e 9 14 0 . . . Q xi I:t ia l o f ~ I- h-= I - qt U ol ite s ? ~ t l a l o f s lh ax :l as tl c r ly sd l ~ o f th e q: :tu u1 lti c se ri es : W l.t to .J t En f"' Ol .ca 'u. sa rm tl nr ta l c o ll is un 1 . . o ep os iti m o f l.J ra ss lC - 8e g1 m lr g o f d ep aU tl al o f W ar ", J J J fly sc h se r ie s A d Pa 1j. .., n ys d' l s er J.e S - Ce pa uU cn o f I:Iu .d< r ly sd 'l se r. le9 19 5 19 5 19 5 19 5 I- - - - _ 'tI rtt r.e rd S lb :1 J:t la1 o f tI'l o a: ea u c ~ _ ln te re 1\ 8 I18 J1 Bt isn (g rr lte s) to th e i- - So ut Iw >t d oc ea Uc t :h n. 6t irg l- i ~th ::& sP" 'er e b ef' I!a :h t re O t:r d- e Ba yc n ro rt h o f f \.s ha "l Ps ha rt z tr e o f tr e q: tu o1 ite I 1. la ss lC blc rl< : S" d m tr us lO l o f re la te d ca lc - I . . flr -r l> d ep au tl '" (0 "" "" "" ,, o f v o l= ra :I< s I ~o ma ti al o r tt e D: :n: ua J qt u. cl lte in a I a. 1k aJ. J.r e k m n- tY J= e re la te d m a gr &b sJ I a d d 'e rt s) in e m te m fU sh al P sh ar t S4 )ta -st .ix Lc tic rr d zc re E nli l'O 'll' e" ll: 2l J Z.' O 2l J 2lO I- -- -- - I- - . . . ()e pa ut i.c n o f fly sc h se ri es W Ith f - f- ljl pe r . . . & gl rr un ; o f a :e a u c a :: cr eu m ? lP d' er ts a u S L. tn an re L rc nu tlc rn a. l b as al ts lP lP Pa le az ul C . . {:e pD sIt lO lof ep 1S 1a llc t er rI ge nn .e I~ )4 , )45 )45 se rl es (C art xru fe1 'O. .f3 to l oe " Pe m u.m ) ~ t l C n o f p lat .fo on s er Ie S - - - - - 11 1 Ce nt ra l ll1 as a bl a: k I~ ~ Il < > er l- & sl t'a i . . . " . , . t= . n l u" ," ", " ' " , " , pJ at fo on lP I Pa leu zu LC OC >p Jsl tl< Jl ov er a ajo ct.n t t 'U ltl fO lta i l na ~r r. I F ig .6 : C o rr el at io n s be tw ee n th e m a in s tr a ti g ra p h ic a n d te c to n ic e v e n ts in A fg ha ni st an , P am ir , T ib et a n d B ur m a. . . . ' 0 I.J l 496 series, and Mesozoic (Cretaceous) to Cenozoic sedimentary and volcanic series overlying the preceeding units disconformably. A similar, but not-well constrained situation is described in Burma. This suggests that in Triassic time, an oceanic domain existed along a EW trending area, located between the Central Mountanis - South Pamir - Lhasa - Western Burma block in the south and the Eurasiatic block in the north. Unfortunately, there are not enough detailed petrological data on the ophiolite remnants, to compare the nature of the ancient oceanic domaign throughout the entire suture zone. The only important point is the existence of chromite deposits within the depleted harzburgites in the Farah Rud as well as in the Donqiao and Burma areas. However, we can discuss the evolution of the Farah Rud - Donqiao oceanic domain by combining the stratigraphic, tectonic and geochronological data from Afhganistan and Tibet. The main events are shown on Figure 7. 1. Initial Stage (Fig.7A) Remnants of the continental domain existing before the opening of Farah Rud-Donqiao oceanic basin are observed in Afghanistan and Tibet. These continental domains show traces of migmatization and magmatism around 500 Ma in Afghanistan(22) and 530Ma in Tibet(43). In Afghanistan, they correspond to the Band-e Bayan block to the north and to the Central Mountains block to the south. Both are formed of similar Precambrian series(19) which suggeststhat, at that time, they belonged to the same continent (Gondwana). During Palaeozoic times, thick platform series were deposited on the Band-e Bayan and Central Mountains blocks in Afghanistan, the Central and South Pamir blocks and also on the central part of the Lhasa block in Tibet whereas pelagic sediments were deposi ted in the deepest part of the subsiding basin. Transitional basalts and deep pelagic sediments including radiolarites in Afghanistan and Pamir are of Permian age which may indicate that a rifting process occurred at that time(36,37). 2. Triassic Events (Fig.7B and C) There is abundant evidence for an oceanic domain with an active north-dipping subduction zone in Triassic time. The presence of Triassic pelagic marine deposits in Pamir(36,37). Metamorphic and tectonic events described north of the Hindu Kush occurred before Visean time(60). It seems therefore more likely that the Triassic Hindu Kush - Badakhshan magmatic belt resulted from the northward subduction of the Farah Rud ocean beneath the Eurasiatic margin as previously suggested(4,28,6l). In the Donqiao region (Tibet), foliated amphibolites which reresent metamorphosed oceanic curst are associated with the ophiolite series(40,50). Their metamorphism occured in Triassic time as determined by radiometric dating( (200,220 Ma; Maluski et a1. (52(). It is likely that this metamorphic event occurred during the initiation of the 497 SOUTH NORTH A D 12 E[~ Fig.7: Schematic sections illustrating the evolution of the Farah Rud-Rushan Pshart-Bangong Nu Jiang ophiolite belt. A: Initial stage (Paleozoic time); 1: magmatism in Afghanistan (500Ma) and Tibet (530Ma); 2: deposition of thick flysch series in Afghanistan (Turkmann series) and Pamir and of marine limestones in Tibet (Xainxa region); 3: transitional basalts and radiolarites interbedded with Permian sediments in Pamir; B: Triassic events; 4: deposition of Lower Triassic flysch series in Pamir (and also probably in Afghanistan and Tibet); 5: formation of the Donqiao-Xainxa ophiolite in a subduction zone enviornment (Tibet); 6: plutonism of the Indu Kush - Badakshan (Afghanistan) and Central Pamir; 7: North Indu Kush rift; 8: formation of the foliated amphibolites during initiation of the subduction zone (Tibet and Afghanistan ?); C: Jurassic events: deposition of ophiolite olistoliths in Upper Jurassic limestones (Tibet); D: Neocomian events: 10: deformation and metamorphism of the flysch series deposited south of the subduction zone (Afghanistan and Tibet); 11: Deposition of the continental to shallow marine Zigetang Formation with reworked ophiolite materials; E: Late events: late backthurstings (post-Albian in Tibet and post-Paleogene in Afghanistan). 498 northward-dipping subduction zone or during a later intra-oceanic thrusting event. The foliated amphi boli tes found together with peridotites in the Waras flysch series in Afghanistan could also represent the remnants of this metamorphic event. 3. Jurassic Events (Fig.7C-D) During the Jurassic, a thick sedimentary flysch series, has been deposited in both the Farah Rud and Donqiao oceanic domains. In Afghanistan, the Waras fysch was probably deposited south of the main subduction zone in order to explain the presence of ophiolitic olistoliths in its lower part whereas the Panjaw flysch was deposited north of this subduction zone at the southern margin of the Band-e Bayan block, as shown by its clear sedimentary polarity (transition from shelf carbonate deposits in the north to pelagic deposits toward the south). In the Ilansen area (Tibet), ophiolite olistoliths were deposited in Upper Jurassic coral bearing limestones(40), which suggests that the obduction was terminated at this time. The Donqiao-Xainxa ophiolite in Tibet was clearly obducted onto the continent by Later Jurassic time, because it has been partly eroded and reworked in upper Jurassic-Early Cretaceous sediments (Fig.7D) as shown by the presence of chromite and well-rounded quartz pebbles in upper Jurassic early Cretaceous conglomeratic horizons in the Donqiao massif(40). Rong-Hua Xu et al.(43) suggested that the collision of the Lhasa and the Quantagn blocks had occurred in Dogger time. The high-grade metamorphic event which has affected the Waras flysch series partly resulted probably from the juxtaposition of several slices of sediments thrust toward the south during the obduction. Accordingly, the Panjaw flysch, deposited to the north of the subduction zone, only shows low - grade metamorphic assemblages. 4. Late Jurassic - Neocomian Events (Fig.7D) There was probably no large oceanic domain in the Donqiao area in Early Cretaceous time, as shown by the very shallow marine to subaerial deltaic sedimentation(40). A late tangential tectonic event must have occurred before the deposition of the Upper Aptian - Lower Albian sedimentary and volcanic series in Tibet, to explain the actual position of the ophiolitic series relative to the Jurassic flysch series (basal truncation of the primary thrust contacts). In the Farah Rud basin, the flysch series were deposited until the Neocomian(30). As in Tibet, a strong deformation event occurred before the Barremian-Aptian series was deposited. 5. Barremian to Albian Events There is evidence that a very shallow marine basin existed in Barremian 499 to Albian time, both in Afghanistan and Tibet. In Afghanistan, it is indicated by the transgression of the "orbitolinid sea" from west to esat. The shallow marine sediments (calcareous detrital sediments) are very thick (several thousands of meters) in the western part of the basin(18,19,20). They are much thinner in the east and consist dominantly of continental deposits. Andesitic volcanic rocks extruded during the same time span. In Tibet· orbitolinid limestones were deposited far to the south of the suture zone during Late Aptian - Early Albian times(42), while red detrital sediments were deposited at the same time further north in the Donqiao-Gyan~o area. In both areas, this sedimentary event was accompanied by a basaltic-rhyolitic and andesitic volcanism which ended near 110 Ma(S2). In both areas, plutonism also began at this time (Pasaband diorites dated at about 100 Ma(18(); and Amdo granites dated at 100Ma; Rong-Hua Xu et al.(43). This magmatism has been related to the northward subduction of the Neo-Tethys at the southern margin of the Lhasa and Central Mountains blocks(6,23,62,63). 6. Late Events (Fig.7E) A late symmetrical folding and reverse faulting event occurred in north Tibet after the deposition of the upper Aptian-lower Albian sedimentary and volcanic series(40,42). It is responsible for the late backthrusting of the Palaeozoic and Jurassic flysch series over the Mesozoic series. This is also the case in Afghanistan, where the Central-Mountains block was thrust to the north and the northwest across the southern part of the Farah Rud basin deposits including the Palaeogene detrital and volcanic series. This late thrusting event is likely responsible for the actual northward vergence of the folds observed in Farah Rud series. It could also be responsible for the development of the "late" intermediate-pressure blue amphiboles (crossites), locally observed in the amphibolites along the northern margin of the Central Mountains block. A red molasse series, associated with andesitic and rhyolitic lavas, were deposited during the Palaeogene in the Farah Rud basin. It was accompanied by plutonism 36Ma ago as shown by the major intrusion of the Koh-e Baba granite into 37Ma old rhyolites (18 ,34). No magmatic activity occurred at this time in the northern part of the Lhasa block; magmatism was restricted only to its southern part(S6,62,63) being clearly related to the northward subduction of the Neo-Tethys lithosphere beneath Eurasia(3,S,6). Due to the India-Eurasia collision(13,S9), large strike-slip faulting occurred in the Eurasia block which postdate the Oligocene. Ductile deformation accompanied by granitic intrusions of Miocene age occurred along these main faults (Helmend and Chaman faults) in Afghanistan(23). In northern Tibet, it is marked by nearly EW strike-slip faultin (Xainxa-Jiali· fault) and by NE-SW normal faultings(47,64). 500 ACKNOWLEDGEMENTS This worked has been partly supported by INAG and CNRS through the French-Chinese joint project in Tibet from 1980 and 1983. The authors greatly thank H.Ave La11emant for reviewing the manuscript and correcting the English. REFERENCES 1- Chang Chengfa, and Cheng Hsilan (1973). Some tectonic features of the Mt. Jo1mo Lungma area, southern Tibet, China. Sci entia Geologica Sinica, 16, 257-265. --- 2- Chang Chengfa, Zheng Xia1ang, Pan Yasin (1977). The geological history, tectonic zonation and origin of uplifting of Himalayas. Inst Geo1 Academia Sinica, Beijing, 1-17. 3- Ba11y,A.W., A11en,C.R., Geyer,R.B., Hami1ton,W.B., Hopson,C.A., Mo1nar,P.H., 01iver,J.E., Opdyke, N.D., P1apfker,G., Wu,F.T. (1980). Notes on the geology of Tibet and adjacent areas: report of the .American plate tectonics delegation to the People f s Republic of China. Open File Report U.S. geol. Surv. IVashington,D.C., 80-501. 4- Bassou11et,J.P., Bou1in,J., Co1chen,M., Marcoux,J., Masc1e,G., Montenat,C. (1980). L'evo1ution des domaines Tethysiens au portour du Bouclier Indien du Carbonifere au cretace. In "26 Cong. Geo1. Intern .• Paris. 1980, Co11. CS et Mem. BRGM 105, 190-198. 5- Tapponnier,P., Mattauer,M., Proust,F., Cassaigneau,C. (1981). Mesozoic ophiolites, sutures and large scale tectonic movements in Afghanistan. Earth Planet. Sci. Letters, 52, 355-371. 6- Allegre ,C.J., Courti11ot, V., Tapponnier ,P., Hirn ,A., Mattauer ,M. , Cou1on,C., Jaeger,J.J., Achache,J., SchHrer,U., Marcoux,J., Burg,J.P., Girardeau,J., Armijo,R., Gariepy,G., Gope1,C., Li Tindong, Xiao Xuchang, Chang Chengfa, Li Guancen, Lin Bayou, Teng J i wen, Wang Nai wen, Chen Guoming, Han Tonglin, Wang Xi bin, Den Wanming, Sheng Huaibin, Zao Yougong, Zhou Ji, Qiu Hongrong, Bao Peishang, Wang Songchan, Wang Bixiang, Zhou Yaoixu, Ronghua Xu. (1984). Structure and evolution of the Himalaya-Tibet orogenic belt. Nature, 307, 17-22. 7- :;>engor ,A .M.C., 1979. Mid-Mesozoic closure of permo-Triassic Tethys and its implication. Nature, 279, 590-593. 8- :;>engor,A.M.C. (1981). The evolution of Palaeo-Tethys in the Tibetan segment of the A1pides: in "Proc. Symp. Quingai-Xizang (Tibet) plateau, Beijing, China, 1, 51-56. 9- :;>engor,A.M.C. (1984). The Cirnrneride orogenic system and the tectonics of Eurasia. Geo1. Soc. of America, Sp. Paper, 195, 1-82. 10- Gansser,A. (1977). The great suture zone between Himalaya and Tibet: a preliminary account. ColI. Intern. 268, "Eco1ogie et Geologie de 1 'Himalaya" , C.N.R.S. (ed), 171-191. - 11- Mo1nar,P., Tapponnier,P. (1975). Cenozoic tectonics of Asiua: effects of continental collision. Science, 189, 419-426. 12- Tapponnier,P., Mo1nar,P. (1977). Active faulting and tectonics in China. J. Geophys. Res., 82, 20, 2905-2930. 501 13- Pe1zer,G., Tapponnier,P., Cobbo1d,P. (1982). Les grands decrochements de l'Est Asiatique: evolution dans 1e temps et comparasion avec un modele experimental. C. R. Acad. Sc. Paris, 294, 1341-1348. ~ 14- Tahirkhe1i Khan,R.A., Mattauer,F., Proust,F., Tapponnier,P. (1979). The India-Eurasia suture zone in northern Pakistan: synthesis and interpretation of recent data at plate scale. In "Geodynamics of Pakistan" Abu1 Farah and Kees De Jong, ed), Geological Survey of Pakistan, Quetta, 125-130. 15- Coward ,M.P. , Jan,M.Q., Rex,D., Tarney,J., Thirwa11,M., Wind1ey,B.F. (1982). Geotectonic framework of the Himalaya of North Pakistan. J. Geo1. Soc. London, 139, 299-308. 16- ~Slavin, V.1., Mirzad ,S.Ch. (1969). Tectonic structures of Afghanistan. SOy. Geo1., 4, 68-81 (in Russian). 17- Sborskchikov,I.M., Dronov,V.I. et a1. (1973). The main features of the tectonics of Af ghanistan. in "Geology and mineral resources of Afghanistan", Dept. Geo1. Surv. Afgh., Kabul, 1-30. 18- Wittekindt,H. (1973). Er1Muterungen zur geologischen Karte von Zentra1 und SUd Afghanistan, 1/500 OOOe. Hannover, 109 pp. 19- Slavin,V.I. (1976). Tectonics of Afghanistan. Nedra, Moscou, 204p (in Russian). 20- Abdullah,S., Chyriov,V.M., Staji1o-A1ekseev,K.F., Dronov,V.I., Gannan,P.J., Rosovskiy,L.N., Kafarskiy,A.Kh., Ma1yarov,E.P. (1977). Mineral resources of Afghanistan. Geo1. Survey Kabul, 1-149. 21- Montenat,C., B1aise,J., Bordet,P., Debon,F., Deutsch,S., Le Fort,P., Sonet ,J. (1981). Metamorphisme et p1utonisme au Paleozoique ancien en domaine gondwan sur 1a marge nord-ouest des Montagnes Centra1es d'Afghanistan. Bull. Soc. Geo1. Fr. 23, 1, 101-110. 22- Montenat,C., Moullade,M., Philip,J. (1983). Le Cretace inferieur a orbito1ines et rudistes d'Afghanistan central. Geo1. Mediter. 23- B1aise,J., Bordet,P., Debon,F., Desparmet,R., Le Fort,P., Mistiaen,B., Montenat,C., Vachard,D. (1982). Geologie de l'Afghanistan central. Cartes geo1ogiques au 1/100 OOOe de Tezak et de Maydan. Doc. et Tray. IGAL, Paris, 5, 56p, 2 cartes h. t. 24- Karapetov,S.S., Sonin,J.J., Khain,V.E. (1975). Some major properties of the structure and development of the Afghan-Pamir segment of the Eurasian Alpine folding belt. Vestn. Mosk. Univ. Geol., 30, 3, 38-46. 25- Stock1in,J. (1977). Structural correlation of Alpine ranges between Iran and Central Asia. In "Livre a 1a memoire de A.F. de Lapparent", Mem. h.s. Soc. Geo1. Fr. 8, 333-353. 26- B1aise,J., Bou1in,J., Bouyx,E., Lys,M., Termier,G., Vachard,D. (1978). Decouverte de Permien dans 1a "serie des schistes et xuartzites du Haut-He1mende dans 1es Monts de Turkman, en Afghanistan central, et incidences paleogeographiques. Bull. Soc. Geo1. Fr., 20, 109-113. 27- Stazhilo-A1ekseev,K.F., Chmyriov et al. (1973). The main features of magmatism in Afghanistan. In "Geology and mineral resources of Af hanistan", Dept. Geo1. Surv. Af h., Kabul, 31-43. 28- Montenat,C., VAchard,D. 1980). Le Trias des Montagnes Centra1es et autres regions d'Afghanistan. Ec10gae Geo1. He1v. 73, 3, 697-725. 502 29- Nicolas,A., Bouchez,J.L., Blaise,J., Poirier,J.P. (1977). Geological aspect of deformation in continental shear zones •. Tectonophysics, 42, 55-73. 30- Montenat,C., Bassoullet,J.P. (1983). Le Jurassique et le N~ocomien d'Afghanistan Central. Stratigraphie et ~volution pal~og~ographique. Eclogae Geol. Helv., 76, 1, 194-241. 31- Mensink,H. (1967). Mariner Jura im westlichen Hindu Kush Afghanistan. Geol. Rdsch., 56, 3, 812-818. 32- Wolfart,R., Wittekindt,H. (1980). 9~_19~v~~~fghan~~~an~ 500 pp. Gebruder Borntraeger (ed.). 33- Griesbach,C.L. (1886). Afghan and Persian field notes. Rec. Geol. Surv. India, 19, Part 1, 48-65. 34- Bellon ,H., Bordet ,P., Montenat ,C. (1979). Histoire magmatique de lm'Afghanistan central: nouvelles donn~es chronom~triques K-Ar. C.R. Acad. Sc. Paris, 289, 1113-1116. ---- 35- Shvol'man,V.A. (1978). Relicts of the Mesotethys in the Pamir. In "Himalayan Geology", 8, 369-378. 36- Pashkov,B.R., Shvol'man,V.A. (1979). Rift margins of Tethys in the Pamirs. Geotectonics, 13, 6, 447-456. 37- Ruzhentsev,S.V., Shvol'man, V.A. (1981). Tectonics and structure of the Pamir metamorphics. In "Metamorphic tectonites of the Himalaya", Saklani (ed), 27-41.-----·----- -_. 38- Khain, V. E. (1984). The Alpine-Mediterranean fold belt of the USSR. Episodes, 7, 3, 20-29. 39- Tang Youking and Wang Fangguo (1984). Primary analysis of tectonic environment in northern Xizang (Tibet) Lake District. In '~l!iJ1l.a~.tl.yan Geology" geological publishing house, Beijing, China, 113 (abstract). 40- Girardeau,J., Marcoux,J., Allegre,C.J. et al. (1984). Tectonic environment and geodynamic significance of the Neo-Cimmerian Donqiao ophiolite, Bangong Nu Jiang suture zone, Tibet, China: Nature, 307, 27-31. 41- Wang Xibin, Bao Peisheng, Zheng Haixiang (1984). A dirupted ophiolite in the Lake area in northern Tibet and its geochemistry. In "Himalayan Geology" geological publishing house, Beijing, China, 140 (abstract). 42- Girardeau,J., Marcoux,J., Fourcade,E., Bassoullet,J.P., Tang Youking (1985). The Xainxa ultramafic rocks central Tibet, china: Tectonic environment and geodynamic environment. _GEOo1-.2zL, 13, 330-333. 43- Rong-Hua,Xu, SckMrer,U., Allegre,C.J. (1985). Magmatism and metamorphism in the Lhasa block (Tibet): a geochronological study. Journ. of Geology, 95, 41-57. 44- Jaeger,J.J., et al. (1982). The distribution of fossils to paleogeography of the Lhasa block (Tibet). EOS, 63, 1093. 45- Lin Baoyu and Qiu Hongrong (1982). The Ordovician system in Xizang (Tibet), in Colloque Franco-Chinois, Tibet Himalaya, Campagne 1982: Monograph, Montpellier, France, 4. 46- Qiu Hongrong. (1984). Les conodontes du Nord et du Sud du Yarlung Zangbo au Xizang (Tibet). In "Mission Franco-Chinoise au Tibet 1980". Li Guancen et J.L.Mercier (ed) CNRS, Paris, 109-132. 47- Armijo,R., Tapponnier,P., Han Tonglin (1984a). faulting in southern Tibet: evidence for right between India and Asia. In "Himalayan Geology", Chengdu, China, 1984 (abstract). 503 Active strike-slip lateral decoupling Intern. Symposium, 48- Gopel,C., Allegre,C.J., Rong Huaxu. (1984). Lead isotopic study of the Xigaze ophiolite (Tibet). The problem of the relationship between magmatites (gabbros, dolerites, lavas) and tectonites (harzburgites). Earth Planet. Sci. Letters, 69, 301-310. 49- Prinzhofer,A., Allegre,C.J., Bao Peisheng, Wang Xibin. (1984). Magmatism in the southern Tibet: trace element constraints. in "Himalayan Geology", Intern. Symp. 1984, Chengdu, China (abstract). 50- Girardeau,J., Mercier,J.C.C., Tang Youking, (1986). Petrology of the Donqiao-Xainxa ophiolite (North Tibet, China): evidences for a formation in a subduction zone environment. Contrib. Mineral. Petrol. (in press). ophiolite its plate Symposium. from the tectonic Chengdu, 51- Wang Fangguo and Tang Youking (1984). The northern Xizang (Tibet) Lake District and significance. In -Himalayan Geology" Inter. 1984, 41 (abstract). 52- Maluski,H., Coulon,C., Songchan,W. (1985). ages of orogenic volcanics from central and Cognita, 5, 2, 279 (abstract). 40Ar /39Ar radiometric southern Tibet. Terra 53- Girardeau, J ., Mercier, J . C . C., Zao Yougong (1985) . Origin of the Xi gaze ophiolite, Yarlung Zangbo suture zone. Southern Tibet. Tectonophysics, 119, 407-433. 54- Lin Boyu and Qiu Hongrong (1984). New development in palo zoic stratigraphy and paleontology ion Xizang (Tibet). In "Himalayan Geology". Intern. Symposium, Chengdou, China, 1984, 4. 55- Wang Naiwen (1982). Development of the Mesozoic formations in the Lakes region, North Tibet and its plate tectonic application. In "Recueuil d' arti£~~_ colJ-()gue I1r-anco-Ch:Lnoi,, ___ sur __ lei ~~lQgj~ de 1 'Himalaya", Guikin, Chine, 3 (abstract). ---- 56- Coulon,C., Wang Songchan (1983). A field study of volcanism from Central and Southern Tibet. Terra Cognita, 3, 265 (abstract). 57- Hutchinson,C.S. (1975). Ophiolites in SE Asia. Geol. Soc. America Bull., 86, 797-806. 58- Mitchell,A.H.F. (1981). Phanerozoic plate boundaries in mainland SE Asia: the Himalaya and Tibet. J. Geol. Soc. London, 138, 109-122. 59- Tapponnier,P., Pelzer,G., Le Dain,A.Y., Armijo,R. (1982). Propagating extrusion tectonics in Asia: new insights from simple experiments with plasticine. Geology, 102-133. 60- Vachard,D. (1980). Tethys et Gondwana au Paleozoique superieur: les donnees afghanes; biostratigraphie, micropaleontologie, paleogeo- graphie. Doc. et Trav. IGAL, 2, 1-463, 35pl. 61- Boulin,J. (1981). Afghanistan structure, greater India concept and Eastern Tethys evolution. Tectonophysics, 72, 261-287. 62- SchMrer,U., Ronghua Xu, Allegre,C.J. (1984a). U-Pb geochronology of Gangdese (Transhimalaya) plutonism in the Lhasa-Xigaze region, Tibet. Earth Planet. Sci. Lett., 69:311-320. 504 63- SchHrer,U., Hamet,J., Alelgre,C.J. (1984b). The Transhimalaya (Gangdise) plutonism in the Ladakh region: a U-Pb and Rb-Sr study. Earth Planet. Sci. Letter~, 67, 327-339. 64- Armijo,R., Tapp~n-ni;~:P".-, Mercier ,J .L., Han Tonglin (1984b). Quaternary extension of the Tibetan plaetau: Field observations and tectonic implication. In "Himalayan Geology", Intern. Symposium. Chengdu, China, 1984 (abstract). THE WESTERN END OF THE TIBETAN PLATEAU Aymon BAUD Musee Geologique Palais de Rumine CH-1005 Lausanne Switzerland ABSTRACT: The Tibetan plateau terminates between the NW-SE oriented, right-lateral strike slip Karakorum fault and the E-W oriented, left-lateral strike slip Altyn-Tagh fault, forming a wedge from wich Tibet is now being extruded eastwards. The Plio-Pleistocene Sutlej basin, the Garpo pull-apart, the Purang and the Shiquanhe grabens are associated with the late Tertiary displacement of the Karakorum fault system. In this western part of the plateau, the S (Lhasa) and N (Qangtang) Tibetan Blocks can be subdivided into four zones (fig. 1): 1- The Ladakh - Kailas zone is bounded to the S by the Indus- Yarlung suture and to the N by the Shiquanhe ophiolitic melange.The latter can be followed along the Karakorum fault, dismembered, and correlated westward with the Shyok melange (fig. 2). 2- The South-East trending Bancong Lake zone consists mainly of early to middle Cretaceous sediments intruded by huge (?)late Creta- ceous granitic bodies. This zone can be correlated with the Saltoro belt appearing W of the Karakoram fault, between the Shyok and Nubra valleys (fig. 2). 3- The East Changchemno zone belong to the Qangtang Block, bounded to the S by the late Jurassic Bancong - Nujiang suture and is separa- ted from the Loqzung zone to the N by the SW-NE oriented, left-la- teral strike slip Changchemno fault. We correlate the Bancong -Nuj- iang suture with the N Saltoro suture in Ladakh and the Northern suture in Pakistan. To the NW, the East Changchemno zone can be followed into the Aghil - Shaksgam sedimentary belt of the Karakorum Tethys. 4- The Loqzung zone consists of a curved NW to E trending belt of folded late Paleozoic to late Cretaceous mainly shallow water sedi- ments. To the N, the Lake Lighten (Longmu Co) cryptic suture separates the Qangtang Block from the Aksai Chin zone belonging to the Western Kun-Lun Block. This last zone is characterized by a thick Cretaceous sedimentary cover lying disconformably on partly metamorphosed and folded Paleozoic black slates (Kilian facies). 505 A. M. C. $,engor (ed.), Tectonic Evolution of the Tethyan Region, 505-506. © 1989 by Kluwer Academic Publishers. 506 o I Junqbw8 Ophiolites : Sut,lei I T~tt~)'~ Himalaya BLOCK 100km Kar.kolum fault Changchemno f autt 5 Kun lun fault Rudole Ilncll$ OPhiol~liC M. lone .. LHASA BLOCK .. Figure 1: Block diagram of the western Tibet region looking west and showing the main tectonic features. HIMALAYAN BLOCK IOYkm Sponqtanq OPhi~l;tel J Indus S.l. Tethy",; I Hi ..... ,.., 'Indus I I 5h),0_, S.L. , Sh~IOk LADAKH BLOCK ku.lcoru ... J iCault , Northern Of I SanOID, S.L. Chon')l Tuh rault I I KARAKORUM BLOCK Figure 2: Block diagram of the Ladakh region (NW Himalaya) adjacent to the western Tibet and showing the main tectonic features. N THRUSTING ON THE TIBETAN PLATEAU WITHIN THE LAST 5 Ma Kevin Burke and Lynette Lucas Lunar and Planetary Institute 3303 NASA Road One Houston, TX 77058 and Department of Geosciences University of Houston, University Park Houston, Texas 77004 ABSTRACT. The Lunpola basin, in the middle of the Tibetan plateau, contains about 4 km of non-marine sediments deposited since early Cenozoic times. This remarkable structure, localized in the Bangong-Nujiang suture zone has much to teach us about how the plateau has developed. In this short note we emphasize the strong evidence of young compression in the basin. Persuasive evidence of young thrusting on the Tibetan plateau has not been described. We suggest that the geology of the Lunpola basin (Zhengyu, 1980; Lee, 1984) provides such evidence. Whether there is late thrusting on the Tibetan plateau has been an important question in Tethyan tectonics for sixty years since Argand (1977) published a sketch cross-section illustrating the collision of India and Asia and indicating that the elevation of the Tibetan plateau could be attributed to uplift in response to the thrusting of Indian continental crust beneath Asia. Holmes (1965) quantified the average 5 km elevation of Tibet as a reasonable isostatic response to such a process, but Dewey and Burke (1973) in drawing another sketch cross-section of the collision zone minimized the under-thrusting of India and depicted the elevation of Tibet as largely the result of the shortening of Asia. Studies mostly by Tapponnier and his co-workers over the last decade (Molnar and Tapponnier, 1975; Mercier et al., 1984; Rothery and Drury, 1984; Tapponnier et al., 1982) have emphasized that although evidence of active extensional faulting and strike-slip faulting is widespread on the surface of the Tibetan plateau both in LANDSAT imagery and on the ground they have seen no evidence of recent compression and teleseismic data include no thrust mechanisms. These observations are perhaps more compatible with operation of the Argand-Holmes mechanism than the Dewey-Burke mechanism. On the other hand, wide angle seismic reflection measurements indicate the presence of thrusts at the base of 507 A. M. C. ~engor (ed.), Tectonic Evolution o/the Tethyan Region, 507-512. © 1989 by Kluwer Academic Publishers. 6 Nt 32 "O O 'N 31 l> 30 'IL _ _ _ _ . L _ _ _ _ _ _ _ _ _ 5:: :~J ... ._= !:: ::: . _ _ _ _ _ _ . L _ _ _ _ _ lJ ' I 08 1 0 0.5 =::J -Oua te rn ar y E . . . . . . . . . N dD N eo ge ne O ,n gq tn gF m ~ En Dp Me og en e~ Ca no zo ic N lu t. oF ", K [J Und iff er en tia te d Cr et ac eo us J 0 Un di ffe re nb a1 ed J ur al lic • Op hi ol ite s p z lI P at eo zo ic 811 SiI u_ I _ _ - Bo un da ry 01 li thO u n its Bo un da ry o f b od ro ck + S yn cl in e + An tlc lln . - - SC rik o f ro m In N Ig O ry ~ T hr us t - ~ S 1r i.l llp fa ul t - - - Fa ul t o f u n ce rt ain ty pe 509 the thick continent beneath the plateau (at the level of the M discontinuity) and several studies by the Chinese and their French collaborators have drawn attention to evidence of shortening on the plateau (Hirn et al., 1984; Girardeau et al., 1984; Chang and Cheng, 1973). Although much of this shortening is related to Mesozoic collisional events, some is linked to Himalayan collision because late Mesozoic strata (e. g. the Takena formation) are involved (Girardeau et al., 1984). The issue is not whether there has been any shortening in Tibet since the Himalayan collision (-40 + 5 Ma), but whether this shortening has ceased and if it has how long ago. A related but much more difficult question is: how much shortening has there been during the Cenozoic on the plateau? We here outline the geology of the Lunpola Basin in which there is substantial folding and thrusting involving Pliocene rocks that appears to have happened within the last 5 Ma. We use published sources and our own interpretation of LANDSAT imagery and metric camera photography and present a new map and cross-section of the basin. Zhi-ming (1981) was first to point out the association between some of the Great Lakes of the Tibetan plateau and compressional basins. Jilin Lake lies in the most prominent of these basins, the Lunpola, which is about 200 km long by 20 km wide (Fig. 1). Tertiary lacustrine and fluviatile deposits about 4 km thick ranging from Paleocene to Pliocene in age occupy the basin and ostracod faunas, spores and pollens have been reported (Zhengyu, 1980; Lee, 1984) (Fig. 2). The latter indicate that the basin lay in wooded terrain with conifers and willows becoming abundant only in the Pliocene. There are no evaporites reported in the section and our impression is of a Tibet much moister and lower than the present dry, elevated plateau. The Lunpola Basin lies in the center of the plateau and (Fig. 1) is localized within the Bangong-Nujiang (B-N) suture zone which marks the place where northern and southern Tibet were joined by continental collision at the end of the Jurassic. A detailed structural study (Girardeau et al., 1984) of an area immediately to the east of the Lunpola Basin makes a distinction between an earlier structural event (Dl) corresponding to Early Cretaceous obduction of the B-N ophiolite and a later event (D2) corresponding to the Himalayan collision. The latter event is characterized by upright and open folding, the development of a fracture cleavage and overthrusting generally toward the north. Open folding and thrusting to the north affect the Cenozoic, including the Pliocene, of the Lunpola Basin (Fig. 3) and we suggest that this D2 phase of folding is the event that is recorded in the Lunpola Basin (Zhengyu, 1980; Lee, 1984). The accumulation of 4 km of sediment in the Lunpola is reminiscent of that in other basins adjacent to thrust belts (e. g. Los Angeles) a particularly close analogue is perhaps with the Mush thrust-bound basin of eastern Turkey in which there are about 3 km of Neogene sediments (Dewey et al., 1986). Sediments in the Lunpola basin are thermally mature (Zhengyu, 1980; Lee, 1984) and a single oil-well has produced (D. A. Skevingtian Britoil, personal communication, March 1985) although the section is relatively thin (-4km). Two possibilities are that this results from the high thermal gradient of Tibet (Francheteau, 1984) or that thrusting 510 C E N E Z T o E I R C T I N METERS P G L a I u N 0 I E C N o E G G N E E F N 0 E M R 1M o A C T E I N 0 E N A ' R g Y c P ~ B A 7 A L , 0 E 0 o C F G ' 0 E ~ A N - M E E A o T _ C I E E 0 g ~ N E I N P E A - L E o C E N E IIIl III I III ~MARL I;,,~? 0"1 TUFF b 511 has increased the depth of burial of the lacustrine source rocks. The Neogene sediments of the Lunpola Basin appear on images to be particularly well exposed in cliff sections between Jilin Lake and Lunpola itself and it would be extremely interesting to know more about the environment of their deposition. For example, the occurrence of vertebrates such as Hipparion might be of both stratigraphic and environmental significance and present evidence of a Pliocene age is not overwhelming (I. G. Sohn, J. W. Neals, and R. M. Forester, personal communication, March-April, 1985). Recent mechanical interpretations of the Tibetan collision (England, 1982; England and McKenzie, 1983) have indicated that the extension on the plateau, the thrusting in the bordering Himalayan and Qaidam areas, as well as the "tectonic escape" of China can be interpreted as a response to the 5 km elevation of the plateau. Lunpolan Neogene floras suggest a much lower elevation for Tibet until only a few Ma ago. A plateau with a lower elevation may not provide an adequate driving force and for this reason alone it would be of great geodynamic interest to know more of the evolution of the Lunpola basin. s N K-TAKENA 5km LJ Fig. 3. Sketch cross-section of the Lunpola basin showing open folding of Tertiary strata and thrusting to the north. No vertical exaggeration. ACKNOWLEDGMENTS Research reported here was conducted at the Lunar and Planetary Institute under NASA Contract NASW-4066 with the Universities Space Research Association and NASA Grant NAG5-585 with the University of Houston. This is LPI Contribution 598. 512 REFERENCES Argand, E. Tectonics of Asia, Fig. 13. Translated and edited by Carozzi, A. V., Hafner Press (1977). Atlas of False Color Images of China 1:500,000, Science Press, Beijing (1983) . Chang, C. F. & Cheng, H. I.,Scientia Sinica, ~, 257-265 (1973). Dewey, J. & Burke, K., Jour. Geol. 81, 683-692 (1973). Dewey, J. F., Hempton, M. R., Kidd, W. S. F., Sarog1u, F., Sengor, ~. M. C., in Collision Tectonics (eds. Coward, M. P. and Ries, A. C.) pp. 3-36, Geological Society Special Publication No. 19 (1986). England, P. in Mountain Building Processes, (ed. Hsu, K. J.) pp. 129-139, Academic Press, London (1982). England, P. & McKenzie, D. Geophys. ~ Astron. Soc., 73, 523-532 (1983). Francheteau, J. et a 1 ., ,~ature, 307, 32-36 (1984). Girardeau, J. et al Nature, 307, 27-31 (1984). Hirn, A. et a1. Nature, 307, 25-27 (1984). Holmes, A., Principles of Physical Geology, Fig. 798, Thomas Nelson and Sons, Edinburgh (1965). Lee, K. Y., ~ ~ ~ ~ Open-File Report, 84-420 (1984). Mercier, J. L., Tapponnier, P., Armijo, R., Tong1in, H., Han Tonglin, Zhou Ji in Mission Franco-Chinoise au Tibet (ed. Mercier, J. L. et Guangcen, Li), pp. 413-422, Centre National de La Recherche Scientifique, Paris (1984). Molnar, P. & Tapponnier, P., Science, 189, 419-426 (1975). Rothery, D. A. & Drury, S. A., Tectonics, ~, 19-26 (1984). Tapponnier, P., Peltzer, G., LeDain, A. Y., & Armijo, R. Geology, 10, 611-616 (1982). Tibetan Bureau of Geology, Geological Map of Tibet 1:1.5M, Tibetan Bureau of Geology, Chengdu (1980). Zhengyu, Xu, Oil! Gas Geology, 1, 153-158 (1980). Zhi-ming, C., Proc. ~ Qinghai, Xizang, Beijing, ~, pp. 1769-1776 Acad. Sinica (1981). TECTONIC EVOLUTION OF THE YANGTZE TECTONIC REGIME Zhang Qinwen Qu Jingchuan Chen Bingwei The Institute of Geology, Chinese Academy of Geological Sciences ABSTRACT. The Yangtze Tectonic Regime, bounded by several great faults, consists of three parts: West Sichuan Fold Belts, Yangtze Platform and South China Fold Belts. The Sichuan Basin Ancient Core covered with thick sedimentary sequences and considered as a oldest part of Yangtze Platform was expanded step by step by accreted wedges or belts composed of trench-arc-basin tectonic system,at it's southeast during the Middle Proterozoic to the Cainozoic. West-Sichuan Fold Belts as a tectonic block were collided with the Yangtze Platform in the late satge of the Proterozoic, when the accretion of Yangtze Island Chain took place. It underwent rifting and separa ting in Middle Proterozoic and recombined maybe by the Middle-Norian Movement. There happened activation in Yangtze Platform and South China Fold Belts to the east of Longmenshan mainly owing to the affect of "Paleopacific Ocean Plate" from the east. I. INTRODUCTION Before Liberation, a few famous Chinese geologists such as ProLLi Shigan and ProLHuang Jiqing (T.K.Huang) et al. had established the outlines of the geology of these regions. The Yangtze Platform or Yangtze Paraplatform etc. had been recognized at this time. But we had known the geology in the West Sichuan region poorly as it had been investigated only by a few geologist such as Prof.Li Chunyu et al. After Liberation, several geological departments and institutes were established and many geologists of these local organs or from geological institutes in Beijing had worked here on stratigraphy, palaeontology, tectonics, mineral deposits etc. and systematic geological mapping was completed at different scales. So that the level of geological investigation has risen greatly. All of these led us to generate many new ideas from the viewpoint of the plate tectonics theory. Our region of interest includes the south of Animaqiang and the north of Jinshajinag in the west, the south of Qinling and the east of 513 A. M. C. !Jengor (ed.), Tectonic Evolution of the Tethyan Region, 513-549. © 1989 by Kluwer Academic Publishers. 514 the Great Danlu Faults in the north, Taiwan and north of Hainan in the east and south. The west part of our region is highly mountainous. The Sichuan Basin is located in the central part of this region. In the south part predominate lower hills and mountains. To the east of our region is the Yellow Sea, East China Sea and South China Sea, which tectonically are marginal seas behind a series of island-arcs. Most of these regions have only continental crust. Oceanic crust occurs only in South China Sea at a large scale. The topographic features correspond with the tectonic units clearly. 2. TECTONIC ELEMENTS (Fig.l). We named the area of the 'Yangtze Platform including the related tectonical belts of a long geological history the Yangtze Tectonic Regime'. Its boundries are the Qinlin Fold Belts in the north, Sanjiang Fold Belts in the west and Hainan Tectonic Regime and Pacific Ocean or its island-arcs in the east. The Yangtze Tectonic Regime can be divided in to 3 parts: West Sichuan (Songpan-Ganzi) Fold belts in NW, Yangtze Platform in the middle and South China Fold Belts in SE; that is, one platform and two fold bel ts. The three parts of the Yangtze Tectonic Regime are genetically related to each other. The West Sichuan Belts as a small and relatively more stable block were combined with the Yangtze Platform probably in the Sinian, and formed the so-called "Great Yangtze Platform". The West Sichuan Platform perhaps was fragmented in middle Permian to early Triassic time. The South China Fo ld Belts are composed of several tectonically accreted belts representing the ancient trench-are-basin systems, which step by step became attached to the main continent forming the Yangtze Platform as accreted wedges or slices from SE during the middle Proterozoic to Cainozoic. From this we consider that the South China Fold Belts are genetically related to the Yangtze Platform as its supplementary tectonic belts. The West Sichuan Fold Belts were broken up into several island-arcs and basins during the middle Permian to early Triassic interval as the Animaqin Daofu-Luhe, Litang and Jinshajiang island-arcs, and Songpan, Yajiang and Daocheng basins were probably interarc basins. The West Sichuan Fold Belts border the Yangtze Platform through the Longmenshan Fold Belts. Several important tectonic units can be recognized in the Yangtze Platform such as the Sichuan Basin ancient core, which may be the oldest tectonic unit in the Yangtze Tectonic Regime, the Fanjinshan arc-trench system, the Jiangnan oldland arc-trench system in the southeastern part and the Yangtze island chain in the west and north. 17 0 29 0 39 0 k m Ha ina n H er cy ;; : F~l d B el t): :T n7 n - IN D O -S IN IA N TE C TO N IC R E G IM E 'I " '- . l - - - " " " " " - - - ~ S o u t h - Ch in a Se a Pl at fo rm F ig .l : T ec to ni c e le m en t m ap o f th e Y an gt ze T ec to ni c R eg im e a n d a dj ac en t re gi m es V ./ U l U l 516 The South China Fold Belts consist of Jianyang-Luoding arc-trench belt with age of early Palaeozoic*, Southeast China Coastal Arch-Trench Belt with age of late Palaeozoic and Taiwan Arc-Trench with age of Cainozoic, and the respective back-arc basins located in the northwest of the each arc-trench system. The Yangtze tectonic regime borders the Hainan Tectonic Regime in the extreme south composed by the Hainan Hercynian Fold Belt and South China Sea Platform through North Hainan Fault. 3. THE TECTONIC EVOLUTION OF THE YANGTZE TECTONIC REGIME BEFORE SINIAN ( 700 Ma) Precambrian rocks predominantly occur in the Yangtze Platform. So we know the Precambrian geological development only or mainly on the basis of data from this region. The west Sichuan Fold Belts exsisted perhaps as a rather stable block during the Precambrian. Based on isotopic data collected up to the present, Archean rocks have not been discovered in the outcrops of this region. The oldest isotopic ages can be detected in the Fanjinshan (1400 Ma) and in the north of Kam-Yunnan Metamorphic Belts (2000 Ma). Also old isotopic ages have been recognized in several regions, but they are not reliable. We can suppose that the Sichuan-Basin ancient core is covered with thick sedimentary sequences probably from the latest Precambrian to the Mesozoic-Cainozoic and constitutes the oldest basement in the Yangtze Platform (Fig.2). This view can be verified by the character of distribution of the old rocks which occur at the surface of the old tectonic belt. They always are around the ancient core. This indicates the presence of an old rigid block which the later tectonic fold belts gradually enveloped. The ancient core had been consolidated possibly at least before the middle Proterozoic or more than 1400 Ma ago. The Fanjinshan arc-trench belt with an age about 1400 Ma is located in the southeast ancient core of the Sichuan-Basin. At present only several separate metamorphic domes with NE-SW long axes can be seen. The other parts of it are mostly deeply buried under the sedimentary sequences of the Palaeozoic. Disrupted ophiolite suites sometimes can be detected in the west parts of the metamorphic domes. The Sichuan-Guizhou back-arc basin is located to the northwest of this arc-trench system. * This age assignment is contradicted by the new observations of Prof. Ken HsU (ETH-ZUrich) and his collaborators mainly from the Institute of Geology of the Academia Sinica in Beijing. See HsU et al. in this volume (Editor's note). -- M et am or ph ic D om e H ai na n lq o 2? 0 30 0 k m IN D O -S IN IA N TE C TO N IC R E G IM E ~ - _ _ I - - S ou th -C hi na S ea P la tf or m ( >6 oo M A) F ig .2 : Sc he m at ic p al ae ot ec to ni c m ap o f th e Pr ec am br ia n - - - - - - - - - \ \ / /" i V .I N A N I/~I H A l ~ . ) I . I il "·_ />~ I ~ __ r =;km V I - J 518 To the southeast, is the Jiangnan oldland arc-trench belt with ages around 800-1200 Ma. As mentioned above, it constitutes with the northwest Jiangnan the oldland arc-trench belt with ages around800-1200 Ma. As mentoned above, it constitutes with the northwest Jiangnan oldland back-arc-basin a rather new trench-arc-basin system. But it had a more complicated history and its east, central and west parts had a different geological development during the late Precambrian. The Yangtze Island Chain, located around the ancient core of the Sichuan Basin to its north and west contains a seres of rather old metamorphic domes such as the south Huaiyang, Huangling, Hannan Pengguan, Baoxing, KalIl-Yunnan, North-Vietnam and a few others. All of these domes possibly had been formed as island-arcs mainly composed of metamorphiC rocks especially geiess and migmatite representing probably products of high-temperature metamorphic belts. We can arrive at some conclusions from tectonic characters in the Precambrian of the Yangtze Platform as follows: (1) The Yangtze Platform of Precambrian age can be subdivided into three parts: the Yangtze island chain, Sichuan-Basin ancient core, and the Southeast Trench-Arc-Basin System. (2) Southeast Trench-Arc-Basin Sysetm subsequently was accereted to the ancient core from the southeast. (3) The Yangtze island chain was accreted to this ancient core from north and west in late stage of the Precambrian probably at 700 Ma. This idea can be verified by the distribution of Dengying Formation representing the latest stage of the Precambrian along the northwest border of the Yangtze Platform (Fig.3). I -.; 0 519 Yangtze Platform in the latest stage of the Precambrian through the Yangtze island chain by accretion and collision to form the so-called Great Yangtze Platform including possibly also the Tarim Block. But up to now the typical ophiolites have not been recognized, only the masses of volcanic rocks under the thick sedimentary deposits of the late stages had been discovered by drilling. They belong to island-arc volcanic rocks. (4) In addition to the Yangtze Platform, the West Sichuan region probably was located as a rather stable block. But we cannot find the rocks with Precambrian age as a result of the extensive post-Permian cover. A few isotopic ages of 500-600 Ma had been detected in the several localities in the metamorphic rocks of rather high grade, particularly in the Jianyang-Louding arc-trench belt in the southeast of Yangtze Platform. They occur possibly owing to the errors in isotopic age analysis or due to parts of the ancient core of island-arcs which detached from the Yangtze Platform. Other than the metamorphic rocks with isotopic age of 200-300 Ma, also several rather young ages can be detected. We thought it 'resulted from the strong reactivation of the Indo-Sinian and Yanshan orogenic movements during the Mesozoic. This situation also can be detected in metamorphic complex of the Jianyang-Luoding Arc-Trench Belt (See the paper by Hsti et al., this volume, for an alternative explanation of these young ages. Ed.). 4. THE TECTONIC EVOLUTION OF YANGTZE TECTONIC REGIME DURING THE SINIAN-EARLY PALAEOZOIC INTERVAL In this stage the Yangtze Platform generally became stable and expanded by accretion of the Jianyang-Louding arc-trench beltfrom the southeast (Fig.4). It is necessary to point out the continuous sedimentation from the Sinian to the Paleozoic in this region. The distribution of facies of the Dengying Formation with a late Sinian age (Fig. 5) is of great significance to explain the tectonic framework in this stage. In most parts of the platform carbonate sediments are distributed and in the border regions silicious calcareous sediment predominated. All of these with small thickness possibly were deposited in a rather shallow-water environment and the rather fine arenaceous sediments, sometimes flysch with deeper-water facies, were deposited beyond the Yangtze Platform and only to the southeast of the Yangtze Platform. It is necessary to show the distribution of the tillites in early Sinian (Fig.6). ~ M ai n H ig h- T em pe ra tu re Z on e lJ.! J.J !J m e M ai n H ig h- P re ss ur e Z on e o P os si bl e Su bd uc tio n Z on e \ \ ~-7 r-- --- --+ /' - - - \ H al n an I 0 10 0 20 0 30 0k m IN D O -S IN IA N TE C TO N IC R E G IM E ' , , , l F ig .4 : T he Sc he m at ic pa la eo te ct on ic m ap o f th e S in ia n -e ar ly P al ae oz oi c in te rv al . r i V .' \J > tv o rl~ 1-e " , O b N or th C hi na O ld la nd . c"" , r:Y~ "'~\ . r- .?J .\\\. e. , U l "' " ~e " .~e~ \'~ ~ Pi ed m on t C la st ic s [ff iJ /o , " 0 / 0 V ol ca n o- M ol as se 0 ,0 .. .. - o C on tm en ta l T il li te s B T d li te s ( Ma rin e) Y an gt ze O !ld :la nd " .. .. .. .. . . /I i" . . . , II 19 0 20 0 39 0 km II F ig .6 . P al ae og ra ph ic m ap sh ow in g th e ti ll it e -s ta g e s 8d im en ts o f th e e a r ly S in ia n /~ ;T ai be i 1\./ . )- -" r i \\~( ! 'I ~ / T ~ 11(-6 r II i ) , N AN '\i i i H A l )~ ~ ! k m V 1 N N 523 We can see from Fig.6 the relationships between the type of tillites and tectonic features. The tillites with continental facies are mostly located on the Yangtze Platform and tillites with marine facies beyond the Yangtze Platform. From these data it seems that the Yangtze Platform had been located possibly in the high latitude at this time. But according to paleo&eomagnetic data the Yangtze Platform was located within the limits of 9 _200 of latitude. During the early Palaeozoic rather thick carbonate and terrigenous arenaceous and argillaceous sediments were deposited on the Yangtze Platform. These very thick argillaceous deposits, sometimes flysch of early Paleozoic age can be recognized in Hunan-Jiangxi back-arc basin.On the Jianyang-Luoding arc-trench belt local metamorphic complexes occur as gneisses and migmatites which probbaly represent rocks formed in island-arc environments under high-temperature. Some green schist formations can be recognized, but always in the west part of the arc-trench belt. It probably was the products of rather high-pressure environment at the time of subduction. The Longmenshan Fold Belt was mobilized due to the irregular subsidence in some regions from Silurian onwards and several subsiding basins, in which mostly thick argilaceous sediments were deposited, were formed in it. Fig.7 is the section showing the above menitoned subduction. Jiangnan Oldland Platform Sedimentary Region Jianyang-Luoding Island- Arc (Active) (High-Thermo- Back- Arc Basin metamorphic Belt) .L..- ZffiB" "VIIiJi! ~ -.-..,..----.- ......... ~.. (+ Ptj- ~!~,) ~ H- +1 \ Fig.7: Idealized section showing the tectono-sedimentary evolution of the SE parts of the Yangtze Tectonic Regime in Sinian-early Palaeozoic time There was subduction of oceanic plate under the Yangtze Platform from the southeast and a back-arc basin, with probably very little oceanic crust and deep-sea sediments, was formed. In the middle and late stages of the early Paleozoic, the Caledonian movement can be detected in most of the Hunan-Jiangxi back-arc basin by the unconformity between the Paleozoic and Devonian sediments especillay near the Jianyang-Luoding arc-trench belt. 524 To the southeast of the east Jiangnan oldland belt an island-arc bel t with a subsiding flysch belt in the west, a thick reef carbonate in the centre and an argillaceous subsiding zone in the east had been detected (Fig.8). Jiangnan Oldland Anticlinal Rising Zone Yuqian-Changhua Flysch Back - Arc Subsiding Belt Donglu - Changshan Reef Rising Belt Changwu Argi llaceous Fore-Arc Subsiding Belt Fig.8: Idealized section showing sedimentation of the Donglu-Changshan island-arc and its bilateral belts at the late Ordovician in the southeast fringe of Jiangnan oldland. The great regression in the middle to late Silurian can be recognized in large parts of the Yangtze platform. The ongoing sedimentation at the boundary of the Silurian and the Devonian can be distinguished only in several localities. So that we could say in the latest stage of Paleozoic a great regression occurred that resulted possibly from the subduction and accretion of trench-are-basin system from the southeast. ~B as al ts of O ce an C ru st ~ Ba sa lt s o f Is la nd -A rc 1)', ,1 B as al ts o f Sh al lo w S ea F ac ie s I~~ ;~ B as al ts o f T er re st ri al F ac ie s F ig .9 : Sc he m at ic p al ae ot ec to ni c m ap o f th e m id dl e P al ae oz oi c a n d th e d is tr ib u ti o n o f c o e v a l b as al ts . U \ N U \ 526 5. THE TECTONIC EVOLUTION OF THE YANGTZE TECTONIC REGIME DURING THE MIDDLE PALAEZOIC The Yangtze Platform was very stable during the late Paleozoic, in particular in its middle stage. During the late Paleozoic the Yangtze tectonic regime expanded by tectonic accretion of the southeast trench-are-basin system in one hand and was reactivated in different degrees in different regions especially in the late stage of late Paleozoic in the other hand. At first, the Southeast Coastal Belt was accreted to the Yangtze Platform from the southeast. It underwent mainly rather high-temperature metamorphism and strong crumpling (Fig.9). In the northwest of our region strong separation occurred possibly from the late Permian. Its details will be elucidated later. In the Longmenshan Fold Belt strong subsidence also subsequently occurred in several regions beginning with the Silurian and ending with the Devonian. In the late or middle stages of the late Palaeozoic, the Southeast Coastal island arc belt with East Zhejiang-Fujian back-arc basin possibly was accreted to the mainland from southeast. The rather high-grade metamorphism that can be recognized in the Southeast Coastal island-arc bel t probably resulted from the subduction of an oceanic plate from the southeast (Fig.lO). Coastal High- thermometamorphic () Belt -SE Hercynian Platform B ck-A B· () J/$§!l//l//!$!$//lI!/!l_.~l\!;~:Vf/I Fig.lO: The tectonic evolution of the coastal belt in the middle Palaeozoic. 6. THE TECTONIC EVOLUTION OF THE YANGTZE TECTONIC REGIME IN THE MESOZIC-CAINOZOIC INTERVAL 1. INTRODUCTION The great change in the tectonic framework occurred during the Mesozoic-Cainozoic interval in our region. Most of East China underwent reactivation possibly as a result of subduction of a palaeooceanic plate 527 from the southeast as a segment of circum-Pacific tectonic belts.But in the west part of China complicated history of mobilization, sea-floor spreading, subduction and collision of different plates took place. One of main characters of the sedimentation on the Yangtze Platform was the predominance of continental facies. The West Sichuan Block separated and fragmented or rifted during the middle Palaeozoic, drifted or spread possibly in no great degree in the early stage of the Triassic and collided in the late Triassic probably in mid-Norian time, as a result of mid-Norian movements. 2. TECTONIC EVOLUTION IN THE EAST PART OF YANGTZE TECTONIC REGIME DURING MESOZOIC-CAINOZOIC We must indicate that the China in its geological recognized in the east geological ridge, called tectonic resistance belt. east part of this region is similar to North development. The similar history can be from Helanshan to Longmenshan-Kam-Yunnan by the authors the Helanshan-Longmenshan The geological history during the Mesozoic-Cainozoic in this region can be divided roughly into four stages, based on the characters of the stress field (Fig.ll). The first stage is the stage of the compres2ive stress field from the Triassic (T) to the medial middle Jurassic (J ). During this stage, the crust of this region underwent compression and was shortened somewhat. The eruption of basaltic lava can be recognized in several locali ties especially in North China mainly in the late Triassic. The general regression on the Yangtze Platform occurred in the middle Triassic and the sea-water has not been present in most of this region from this stage to the present. In addition, some small basins with continental deposits formed possibly owing to local extension. The second stage is the stage of t~ strongly compressive stress fi2ld during the late middle Jurassic (J ) to middle early Cretaceous (K ). During this stage the crust was largely shortened and thickened due to the strong compressive stress possibly under the action of a Paleopacific ocanic crust. Strong eruptions of lava from basic to acidic composition occurred in the basins located at considerable heights and originated as a result of secondary extensional stress in a generally compressive stress field (Fig.12). Besides, abundant intrusions of intermediate-acidic rocks can be detected. Both mostly occurred only in the eastern part of the Yangtze Platform and the South China Fold Belts i.e. in the east of the extinct line of magmatism. During the first and the second stage, eastern East-Cathaysian uplift belt acted as a mega-anticline and the West East-Cathaysian Subsiding Belt as a mega-syncline. Il l~ 0 1 -' - 0. 00 elf -"" ' Ill f-- ' 1- '- • • o o ~ N ro o n I- '-r t n o " " ' p- ;3 :: n o 0. 00 I- '-r t t- hl ll 1- '-0 0 ro ro 0. 00 III I II t- h o r to . ro '" 1r o N < :: re , Ill {:: ° r t 00 1 -'_ . 0 0 1- '_ 0 0 0 ~t- h ° r t Ro ::r ro §f ro III I II 0 00 o o r t :: r: : C J {:: ::r III 1 -'- 1- '-0 N il l g-1 -'- 0 0 00 ~ r t ::r f-- 'ro \D (X l;3 :: Nr o '- - " 00 o N o 1- '- n :-I (") g (") " o , ;- - l :-l = -- >:: s: o 0 < :: > - (t l; :: ) !b :: l 3 ' < 3 ' < : ('t) s= (I) s= :::: :I ~ :::: I OJ ~: :: I ;:. ::1 W es te rn E as t- C at ha ys ia n' I~ • R Is in g B el t " :\1; ~ : A n" I : ::: ;: >: :U l s: ~ ~ ~ ~ ~~ o~ ob 1 ~I ~ § a g ~ ~ . ~ ~ ~ ~ ~ ;:;0 ~ I ('t ) ~ c '" ..c f/l c '" ~ .. c-j ~ Western Depression Belt ~ ~~;:: .. -::::; .:::.'.::;:; :=.,' .. :;: .. :.::;:; .. ~ ~ ~ ~ --- --------- 529 Eastern Rising Belt ~SE - Fig .12: Idealized scheme of mechanism of eruption and intrusion in the east Yangtze Tectonic Regime (late Jurassic-early Cretaceous: the stage of compression). In the third stage, the picture changed again. Some large sedimentary basins appeared mainly on the basement of the Palaeozoic complex in the east as a result of the extension of the crust in this region during the late early Cretaceous to the later late Cretaceous. In this stage, the crust of this region, particularly in the east, thinned and extended. The great continental basins formed as a consequence of the extension of the crust such as the well-known Songliao, Jiangnan and other basins. In the fourth stage, many more large sedimentary basins and rather minor basins developed in the east owing to the extension during the later late Cretaceous to the Miocene. In places some basalt lavas can be detected in the thick terrigenous sediments that formed probably as a result of strong extension. Some of large terrigenous basins were inherited from the last stage, like Jiangnan Basin,North Jiangsu Basin etc., and some were juvenile like the North China Basin which was only separated into small basins in the last stage. The fifth stage is the satge of a composite stress field during the Miocene to the present. But we think that the tensional stress field possibly also predominated and still predominates during this stage. In the large sedimentary basins in east of this region subsidence and sedimentation decreased. Many faults ceased their activity which we know from seismo-stratigraphic profiles and drillings data. During the fourth and fifth stages the continental crust of this region, especially the region near the coastal zone underwent strong extension and many normal faults appeared. Some of these bit progressively deeper into the crust and possibly arrived at the lower crust or even at the upper mantle, and basalt lava could erupt along these deep faults. As mentioned above, we can unite five stages to two main stages: 530 a stage of compressive stress field and a stage of tensional stress field divided by Mid-Wealden Movement which is the most important in tectonic development in East China during the Mesozoic to Cainozoic interval. The large terrigenous basins related to oil and gas developed only after the Mid-Wealden Movement (Fig.13). -------"- S E East Part of Yangtze Regime .-;::;/ ===:::::~~~:=~~§~§.~~/I~JI'~~y+::~~~~~!~ .;::;""'/Pal aeo-Paci fi c Ocean -= / ,(/ - Progressi ve Subduction (Comsupti on of Basement) Fig.13: Ideal scheme of mechanism of eastern basins in the east part of Yangtze Tectonic Regime in later early Cretaceous to early Neogene (the stage of extension) There was a tectonic movement between every two tectonic stages which can be recognized by tectonic unconformity between sedimentary sequences and the different tectonic framework around the satge of the tectonic movement. But most of these tectonic movements did not lead to strong orogenic folding in the sedimentary sequenes and only a few of these had important significance for folding like the Bajocian-Stage Movement and maybe the Anyuan Movement. The sedimentary rocks from the Sinian stage all underwent folding especially in the east part of Yangtze Platform during the second stage of the strong compressive stress field (Fig.14). ~SE East Fault - Subsidmg Belt MarglOai- Sea Belt West SubsIding Belt Fig.14: Mechanism of fault-forming in the tectonic cover in the east of the Yangtze Tectonic Regime in Triassic to early Cretaceous interval. Before this, the sedimentary rocks from the Sinian to the early Triassic had remained horizontal. Some great thrusts, even great ductile 531 shear zones formed under the affect of these compressive stresses, which also formed great exotic tectonic blocks. Strong cover-f olding resulted from the strong shortening of the lower earth crust in addition to the gravitational effect. Besides, it is necessary to pay attention to the strong subsidence in North Guangxi. There flysch deposits with several thousand metres in thickness developed. Owing to strong compressive stress, the continental crust was thickened. The thick continental crust was cut into several blocks and these blocks moved and slid relative to each other. At this time the magma can be generated in the middle and lower parts of continental crust owing to the heating from the increasing of geotherms and the friction along block boundaries, particularly during tectonic movements. This may be the origin of many volcanic rocks and intrusions. Today's tectonic framework in the middle and east parts of our region was established during the latest fourth stage (Fig.1S). There are a series of tectonic belts developed from east to west: Trench Belt (I), Island-Arc Belt (II), Marginal-Sea Belt (III), Coastal Uplift Belt (IV), Littoral Subsidence Belt (V), Daxinganling-central Grangxi Central Uplift Belt (VI), Erdos-Sichuan Western Subsidence Belt (VII) and Longmenshan Resistant Belt (VIII). In addiditon, there is the Qinling Trans-uplift Belt which cut several tectonic belts with a NE-SW trend. Beside Large terrigenous basins such as the North Jiangsu Jianghan, the large marine basins can be recognized such as Guangtong Basins and West Guangxi Basin (Fig.16). The large Sichuan Depression was developed in the early stages of the Mesozoic and has subsided up to the present. But the center of subsidence migrated in from east to west in the Sichuan Depression. 3. THE TECTONIC EVOLUTION IN THE WEST PART OF YANGTZE TECTONIC REGIME DURING THE MESOZOIC-CAINOZOIC INTERVAL In later Permian basalt eruptions developed on the West Sichuan block as well as in the west part of Yangtze Platform (Fig.16). Basalt lavas of continental type were distributied in tectonically rather stable regions such as in the west part of Yangtze Platform (the Emeishan Basalts) by contrast marine eruptions occurred in tectonically active regions such as the West Longmenshan Border Region and in some arc-trench belts in the West Sichuan Fold Belts. We may consider the occurrence of basalt lava in late Permian time as a precurser of the fragmentation of the West Sichuan Block. 532 -_-Ir- ~-7r~ i I The Limit at , 1/1 Extinction I 50:- --of Magmatism. , .-r I ! ______ ~ The Boundary of i ~ Tectonic Belts I ! I 250 ! I ---------_ .. _.-. 120 Fig.1S: Tectonic element map of East China and its adjacent regions after Mid-Wealden Moement (Modfied after Zhang Qinwen & Huang Huaizheng, 1982). I)~ '! ... . I V ol ca ni c B el t (C on tin en ta l) 19 0 29 0 39 0k m F ig .1 6: Sc he m at ic p al ae ot ec to ni c m ap o f th e M es oz oi c- C ai no zo ic i n te rv al . V I v. .> v. .> 534 In the two stratigraphic columns from the West Longmenshan Border (Fig.l7) we find the stable shallow-water deposits in early Permian time, basalt lava with pillow structure of marine facies in late Permian time, the less thick argillaceous deposits of deep-water facies in early Triassic time and very thick flysch deposits in the middle and late Triassic. So we reach the conclusion, that these columns reflect the processes of disintegration of the continental crust of the West Sichuan Block from middle Permian to Triassic. From this we infer that the tectonic activation of this region including the adjacent West Sichuan Block started in the late Permian, and strong subsidence and drift occurred from the earliest Triassic. This explains the uncompensation of deep-water sedimentation during the early Triassic. In the region of West Sichuan, a series of arc-trench systems and interac tic basins alternated with each other such as the Animaqiang arc-trench belt, the Songpan interarc, the Daofu-Luhe arc-trench belt, the Yajiang interarc basin, the Litang arc-trench belt, the Daochen interarc basin and the Jinshajiang arc-trench belt. Among them the Jinshajiang arc-trench belt had a more complicated tectonic history than the rest. It opened as a narrow ocean before the late Permian and the other basins possibly opened later. Several island-arcs or elongated uplift belts concurrently developed in this region. Of these the geological development of the Songpan Basin is very typical in this region. Thick flysch sequences of more than 10 thousand metres developed (Fig. IS) and the volume of the flysch deposits was calculated at about five million cubic kilometers only for the middle to late Triassic. We think that this material of enormous quantity was carried from the northwest from the great Central Asian late Palaeozoic Platform including Northwest China, Kazakhstan, Siberia etc. The Daofu-Luhe arc-trench belt is made up of Permian basalt and limestone and subordinate Triassic limestones with slight metamorphism. In the southwest of this arc-trench system, dark argillaceous rocks representing the environment of an inter-arc basin were deposited. Not far from here a series of limestone breccias and basic lava, possibly representing the volcanic island-arc environment was encountered (Fig .19) . In the Yajiang Basin, thick flysch sediments developed and a metamorphic complex with almost horizontal attitude was detected by us in 1965. The metamorphic complex varies from high grade metamorphism in the lower position, where migmatites occur, along with granite-pegmatities and gneisses through moderate-grade metamorphism in the middle where amphibole-staurolite schists are found to lower-grade metamorphism in the high position where dark andalusite schists occur. The later gradually passes into normal flysch deposits and the very gently dipping layers rapidly steepen and become strongly crumpled. We thought this resulted from subduction of a plate from the northwest. The metamorphic complexes with gentle dip may have belonged to a thermal dome formed by a hot spot under the deep crust, caused by subduction. It developed after sedimentation but before regional folding, so the , " .~ P il lo w s tr uc tu re ~""r X r j Dike z o n e " " I r x r X Ba SI C Y oi ca n lc - _ _ ~) X r X r dla ba~ lc- gab bro /' - " > > _ . _ f7 ' 1, ; " - ," " I 3 > ;U ltr ab as ic Ir o c k ) 40 0r n D -C ~ I f5 , - ; L. I C ar bo na t. te " ' )1 ",, 0/ "'- ' I, r i _ _ , _ _ ]( ~_ I L L,~ogi r \r [ ~ Car..i22n a. tit e I 2 ~\.J ,?-'ii ~ I \ , ( , " , , '" - , J" R. " ,,~ t -- : ' , D. - hI. ~'."" (tx.~aJ In . County g . I ' J B . " " . 0 tx , I f Xl n g C ou nt y_ _ _ _ ~ ;~: ~ . v, I N A N \ ,00 I , , , , , 20 0 30 0k I ' , m F ig .1 7: Sc he m at ic c o lu m na r s e c ti on s sh ow in g s e di m en ta ry s e qu en ce s o f th e P er m o- T ri as si c s ys te m a n d th e o p hi ol it e s u it e in S on gp an R eg io n. u . w u . 536 o Yajiang Fig.18: Palaeotectono-palaeogeographic sketch map of Sikam Formation in the Songpan-Ganzi Geosyncline (Modified after Zhang Qinwen, 1981) C, ) . . . . . . . :: C ,) I· - c - 0 . . C C, ) C, ) . . . . . . . :: I :> " . . . . . . c o . . - N E Y aji an g B ac k- A rc B as in " '- - 0 !: > I n te r- A rc B as in O ::§ So ng pa n Se a T 2- 3 ( J ( ) ( ) ( ) I I , I ( ) :6 .1 6 ~ . ;; :T Z- 3 t T 2 3 () B . Id ea liz ed S ec tio n o f D ao fu -Y aji an g T re nc h- A rc - B as in S ys te m i n T~ Y aji an g I T er m o- M et am or ph ic D om e T er m o- M et am or ph ic D om e . . ;t ?0 ~7 ;~ R~ ~~ d ;r z ~: dZ :7 .~ •• :~ \\ \~ \ o 5 10 15 km I I , I A . D ao fu - Y aji an g T ec to ni c Se ct io n F ig .1 9: D ao fu -Y aji an g tr en ch -a re -b as in sy st em , W es t Si ch ua n, a s it w as in t he m id dl e to l at e T ri as si c. V ol . A rc D ao fu \ I I ~ '1'\ - ~ ~ ~ >1 \>~W - 538 primitive horizontal layers could be preserved. There is a column of a rather complete ophiolite suite recognized in the Litang arc-trench belt (Fig .18). The ophiolite suite from the ultrabasic rocks to the pillow lavas can be seen well in this region. According to the chemical analysis of the basic lavas silicon and potassium values are low and they belong to the tholeiites. Most of the Litang basic lavas fall into the cricle of oceanic ridges (Fig.20). 1.1r-------------------------~~~--------------------------__, Tholeiite of the continental rift ~15 of both the ocean ~17 ~ll of the oceanic ridge Andesite of / island-arc r E ~ 0.1L-____________________________________________________________ ~, -0.2 1.5 Fig .20: Chemical analysis data for all kinds of basalts in the world compared with the basic lava of Litang in West Sichuan (After Jiang Youming) . Here is an idealized section subduction in the Litang Island-Arc Belt. The oceanic crust (narrow) was subdue ted to the Litang Island-Arc from northwest and the ophiolite wedge and melange occurred in the nouthwest of it. The magmatism (Fig.2l) can be recognized in the more southeast (related to plate subduction). Yidun Island-Arc I Litang Ophiolite I Yajiang Sea-Basin I I I I Li tang Subduction Fig.2l: Section showing the Litang subduction in West Sichuan 539 The evalutionary history of the Jinshajiang arc-trench belt with a more complicated tectonic development can be divided into four stages (Fig .22): ., "" ! 540 All of these tectonic belts in West Sichuan ended their main plate tectonic development by the middle Norian. After that the marine deposits in general were absent. Sedimentary deposits underwent strong folding in most of this region probably in the latest Triassic. After a short period of stable tectonic conditions, the remobilization of this region occurred during the Jurassic and the Cretaceous, but the strongest uplift of topography of this region, as a part of the Qinghai-Tibet Plateau occurred from the Middle Pleistocene onwards. During the reactivation (or remobilization) the thrusts and nappes can be recognized not only in the region of fold belts but also on the platform or in its border region as in the Fore-Longmenshan region (Fig.23). Longmenshan 2000 1000 Tangbazi I I Tiantaishan I I ~1200 I I Fig .23: Section showing the nappe structure of Longmenshan in Sichuan (After the Geological Survey Team of Sichuan) We can see from Fig .23 that the rather old Carboniferous and Permian sedimentary rocks were thrust upon the Triassic and Jurassic deposits as the Fore-Longmenshan nappe. Its origin may be related to the gravitational gliding owing to the uplift of the Longmenshan and to the compressive stresses from the northwest. 7. MAGMATISM AND METAMORPHISM 1. THE INTRUSION OF ACIDIC AND INTERMEDIATE MAGMA The intrusion of acidic and intermediate magma at different times is very widely seen in the Yangtze Tectonic Regime (Fig.24). On the basis of their ages, intrusions can be divided in to 7 stages: Late Proterozoic, early Palaeozoic, late Palaeozoic, Triassic, Jurassic, Cretaceous and Palaeogene. Among them, the late Proterozoic, early Palaeozoic ("Caledonian"), Triassic ("Indo-Sinian") and Jurassic intrusives ("Early Yanshan") are the more important. Ta ib ei . ~ ? ' l '{I I II / II / ~ L at e ~ ~ P ro te ro zo ic ~ J ur as sI c ITE Il · . p' . :: Ea rl y ~ ~ I , , :: Z l:: P I ' ;;: K -. ;;. C re ta ce ou s . ' . . . ' a a e o zO iC ~ " - ~ L t II ~I ~ pz P I a e . II IE II I E og en e a a e o zo lc ' ~ Tr ia ss ic 10 0 20 0 30 0k m F ig .2 4: S im pl if ie d m ap s ho w in g th e d is tr ib u ti o n o f th e a c id ic -i n te r- m e di at e in tr u si v e r o c ks . lJ o ~ 542 The Proterozoic magmatism developed on the Yangtze Platform especially in the vicinity of metamorphic domes near its north and west borders. It also was developed on the Jiangnan Oldland. "Caledonian" magmatism mainly was developed in the Caledonian tectonic belt, that is Jianyang-Luoding arc-trench belt and Guangxi-Hunan-Jiangxi back-arc basin fold belt. The development of intrusions is genetically connected with the Caledonian movement, that is, with the accretion of Caledonian island-arc and back-arc basin to mainland by subduction and subsequent collision. Indo-Sinian intrusions developed predominantly in the West Sichuan fold belts and they are related to collision. Early Yanshan intrusions very widely developed in both West Sichuan fold belts and South China fold belts as well as in the east part of the Yangtze Platform. The origin of the intrusions of the Yanshan episode, in our point of view, is different from the above-mentioned mechanisms of earlier episodes. These intrusions may have been generated by intracontinental block-sliding as a result of the compressive stress issuing from a "Paleo-Pacific Plate" to the east of the Yangtze tectonic regime and also from the West Sichuan fold belts through continental collision in the Qinghai-Tibet Plateau. Two important conclusions can be outlined: (1) Magmatism of acidic and intermediate intrusions was generated not only by subduction but also by intracontinental block movement. (2) We can see from Fig.24 the tendency that from Longmenshan-Kam-Yunnan to Taiwan the ages of intrusions became younger from the late Peroterozoic to the Cainozoic. This tendency also can be considered as the expression of the process of enlargin the continent. 2. OPHIOLITES (Fig.25) Up to now we have not yet found a complete and undisrupted suite of ophiolite in our area, though many ophiolite belts of different ages are widely distributed. Among them Litang and Jinshajiang ophiolites initially have been studied. 3. METAMORPHISM (Fig.26) The metamorphism in this region is not very strong in general. The rather high-grade metamorphic complexes can be distinguished in the metamorphic domes with Precambrian ages and in several arc-trench belts of Phanerozoic age. The sedimentary rocks in the West Sichuan fold belts and several back-arc basins of South China fold belts mostly are slightly metmorphic. The sedimentary rocks on the Yangtze Platform did not undergo metamorphism. D O ut cr op o f O ph io lit e l.f f I Pos si bl e O ph io lit e B el t ~ M el an ge 19 0 29 0 3Q Ok m F ig .2 5: D is tr ib u ti o n m ap o f bo th o p h io li te s a n d m e la ng es in Y an gt ze T ee to n ic R eg im e. " I / (! l,_- ---' -C" .,km u . "" '" w [Z ] I nd o- Si ni an m et am or ph ic b el t lZ J H er cy ni an m e ta m or ph ic be lt [Z ] C al ed on ia n m e ta m or ph i c be l t . . ~c t- ~~ ~ ~ P re - C al eq on la n = he av y lin e a r e a ~ m et am or ph l(' b el t M id -h ig h T . o gl au co ph an e s c h is t = o bl iq ue l in e a r e a '1 00 -- 20 0- 30 0k m , I I c - I \",. - 7 /H ai na n , IN D O -S IN IA N TE C TO N IC R E G IM E - - - I ~i 'b eG . ' l 'Y l// r- III \ 1'(" / ~r':' / ~ / , \ N A N ! Vi HA l ) ' ,I i / F ig .2 6: D is tr ib ut io n m ap o f m e ta m or ph ic b el ts i n Y an gt ze T ec to ni c R eg im e u . t 545 From the ages of metamorphism the metamorphic belts can be divided into pre-Caledonian, Caledonian, Hercynian, Indo-Sinian and Himalayan. The metamorphic rocks of the first episode developed in the border regions of Yangtze Platform where high-moderate temperature metamorphism predominated (most of these with isotopic ages around 700-900 Ma). The distinctly elongated metamorphic belts are distributed in southeast part of Yangtze Tectonic Regime, with a NE-SW trend. All of them belong to island-arc belts, so they may mainly represent the products of high-temperature metamorphism as a result of subduction of paleo-plates from the southeast. The metamorphic rocks are gneisses, migmatites and schists of different grades of metamorphism. 8. CONCLUSION 1- The Yangtze Tectonic Regime formed and expanded both by accretion of trench-are-basin systems and collision of minor plates during the Pre-Cambrian to the Cainozoic interval. 2- The oldest core in the Sichuan Basin was covered with very thick sedimentary sequences and had formed probably before 1400 Ma. It expanded stey by step five times by the accretion of trench-are-basin systems during the middle Proterozoic to the Cainozoic interval (Fig.27). -------- SE Sichuan- Basin Ancient Core u ... -< " I "''' .~ ~ ~ ~ Oceanic plate ~ Gneiss-Schist Belt (Possible ~ Ophiolite Belt (Possible High- ~ High-Thermo-Metamorphic Belt) ~ Pressure Metamorphic Belt) Fig.27: Ideal model of the tectonic evolution of the southeast part of the Yangtze Tectonic Regime during Middle Proterozoic to Cainozoic. 3- From the above-mentioned, we could imagine that a paleoocean was present for a very long time to the southeast of the Sichuan ancient core, from at least the middle Proterozoic. 4. The northern part of the Hainan Island tectonically shoulj belong to the Yangtze tectonic regime. Its centre part belongs to the Hainan Hercynian fold belt and its southern part may belong to the 546 northern part of the South China Sea platform. The Hainan Hercynian fold belt probably is a part of the Zhanshan Hercynian fold belt of Viet Nam and the South China Sea platform could have been originally connected with the Kynazhon Mass of Viet Nam. The both may have been severed from each other by a great sinistrial fault. 5- The Yangtze tectonic regime was located in the low latitudes in the Carboniferous and the Permian and was far from the Gondwana Lanad. This is established by the presence of subtropical Cathaysian flora, the absence of tillite and Glossopteris, and palaeogeo- magnetic data. 6. There are several tectonic cycles of Yangtze Platform based on tectono-sedimentary characters in the geological history. Four megacycles can be distinguished: Cycle I stable Platform (Sinian), Cyle II - Active Platform Cycle (early Palaeozoic), Cycle III - Stable Platform Cycle (middle Palaeozoic) and Cycle IV - activation Cycle (Mesozoic-Cainozoic) (Fig.28). 7. According to geological data, the Yangtze tectonic regime may be combined with that of the Sino-Korea Platform from the late Silurian to the Devonian to form a united continent. From the latest stage of the early Permian, the united continent (East China Continent) combined with the northern large Siberian Continent to form a Great Central Asian Hercynian Continent. We could assume the Yangtze Platform can be considered as an extreme northeast coastal region of the Great Tethyan Ocean from the late Permian to the Mesozoic. But this conclusion should be further tested in the future, because also some geological data contradict this idea in favour of the presence of a middle Palaeozoic to the Triassic palaeo--ocean':< ACKNOWLEDGEMENTS We are deeply grateful to the Group of Mapping of the Institute of Gology for drawing beautiful figures and to Tong Guilan and Chen Minghui for typing. We also would like to thank Sun Yiyin, Wu Jisahan and Zhou Huiqin for help in preparing this manuscript. ,~ For these data see esp. :;>engor, 1984, The Cimmeride Orogenic System and the Tectonics of Eurasia. Geol. Soc. America Spec. Pap. 195, xi+82 pp. An extensive field programme conducted jointly by Prof.K.J.Hsu (ETH) and the Geological Institute of the Academia Sinica has been augmenting these observations during the last several years (Ed.). Pl at fo nn Se d. C om pl . ~" . . . r.~ r­ ~I~ : : : ~ I~ 7L I. : : : .1: "' I I: " : r I: :: >1 : . r • • • " It > : II I. II I: I I . II I: II I. I, II ' A ct i v e Pl at fo nn S ed im en ta ry C om pl ex :11 1 I I I I I I I :11 1 I I : I I I I I I I I I J : 11 11 11 I: : I : 11 11 11 : I: : I : 11 1 1 1 1 1 1 : : I : 11 11 11 I: : I I I II . I : I, I, I I I. : I I 1 Pl at fo nn I Se d. C om pl . ~ :I~ :: } :-'1 : I : I~I ~ :: : : 'I~ ::, A ct iv at io n \ Se d. C om pl . .~ 1 N m o l 5!J 191 (")1 ~ I ~ 1 ~ ~ C yc le I C yc le II F ig .2 8: C yc le s o f te ct o n o -s ed im en ta ry c o m pl ex s o n th e Y an gt ze P la tf or m du ri ng S in ia n to t he C ai no zo ic . C yc le I II C yc le N I lJ l . j:. - . ) 548 REFERENCES Atlas of the Palaeogeography of China. (1984). Institute of Geology, ~hinese Academy of Geological Sciences and Wuhan Collauge of Geology (ed.) Publishing House of Geology, Beijing, China. Chen Bingwei. (1983). Some new observation on the tectonic development of Sanjiang. Contribution to the Geology of Qinghai-Xizang (Tibet). Publishing House of Geology, Beijing, China. (In Chinese) Chen Bingwei and Ai Changxing. (1983). Discussion on Indosinian Cycle in the Hengduanshan Region (Transect Mountons). Bulletin of the Chinese Academy of Geological Sciences, N.7, (in Chinese). Chen Bingwei. (1985). Tectonic outline and evolution of the Hengduanshan Region. Scientific paper on Geology for international exchange-prepared for the 27th International Geological Congress (2), Publishing house of Geology, Beijing, China. (in Chinese). Helde,T.W.C., Uyeda,S. and Kroenke,L. (1977). Evolution of the Western Pacific and its margin. ,Tectonophysics. 38 (1-2). Huang,T.K. (1945). On major tectonic forms of China. Geol. Memoirs of National Geological Survey of China. Ser. A(20). ------ Huang,T.K. (1978). An outline of the tectonic characteristics of China. Ecologae Geol. Helv., 71(3). Huang Jiqing and Li Chunyu. (1981). Contribution to the tectonics of China and Adjacent Regions. Publishing House of Geology, Beijing, China. (in Chinese). Huang Jiqing (T.K.Huang). (1984). New Researches on the tectonic characteristics of China. Bulletin of the Chinese Academy of Geological Sciences, No.9. (in Chinese). Huang Huaizheng and Hong Dawei (1985). Geological features of the Jiangnan Region in Early Jurassic to Early period of Early Cretaceous Time and their interpretation in terms of plate tectonics. Scientific Paper on Geology for international exchange-prepared for the 27th International Geological C~_I1~!~SS (2). (in Chinese). Lee,J.S. (1973). Crustal Structure and crustal movement. Scie~tia Sinica. 16(4). (in Chinese). Li chunyu, Wang Quan, Liu Xueya, Tang Yaoqing. (1982). The Tectonic Map of Asia. The Cartographic Publishing House Beijing, China. (in Chinese) . Liu Baotian, Jinag Yaoming, Qui Jingchuan. (1983). The Discover of a palaeoceanic crust strip along the line from Litang to Ganzi in Sichuan and its significance on plate tectonics. Contribution to the Geology of the Qinhai-Xizang (Tibet), Publishing house of Geology, Beijing, China. (in Chinese). Molnar,P. and Tapponnier,P. (1975). Cenozoic tectonics of Asia: Effect of a continental collision. Science, V89, P4l9-425. Qiao Xiufu and Geng Shufang. (1981). On Late Precambrian plate tectonics of South China. Contribution to the tectonics of China and adjacent regions. ed. by Huang,T.K. and Li Chunyu, Publishing House of Geology, Beijing, China. (in Chinese). 549 Qu Jingchuan. (1984). A discussion on characteristics of geological structure in East Xizang (Tibet and West Sichuan and South Qinghai during Permian-Triassic, Disintigration and convergence of the massif itself. Bulletin of the Chinese Academy of Geological Sciences, N 10, Pl17-l28 (in Chinese). Ren Jishun, Jiang Chunfa, Zhang Zhengkun, Qin Deyu (1980). The Geotectonic Evolution of China, Scientific Press, Beijing, China (in Chinese). $engor ,A.M.C. (1984). The Cimmeride Orogenic System, the tectonics of Eurasia, Geol. Sci. of America, Special Paper 195. $engor ,A.M.C. (1985). The sto-ryof Tethys: How many wives did okeanos have? Episodes, V 8, No.1. Tapponnier,P. and Molnar,P. (1977). Active faulting and tectonics in China. Jour. Geophy. 82, P.20905-2930. Zhagn Qinwen (1981). The sedimentary feature of the flysch formation of the Xikang Group in the Indo-Sinian Songpan-Ganze Geosycline and its geotectonic setting. Geol. Rev., V 27, N 5. (in Chinese). Zhang Qinwen, Ai Changxing, Li Taizhao and Yu Xijing. (1983). Evolution of the trench-arc-basin tectonic system in the middle belts of Sanjiang. Contributions to the Geology of the Qinghau-Xizang (Tibet) Plateau (12). Publ. House of Geology, Beijing, China (in Chinese). Zhang Qinwen and Huang Huaizheng. (1982). The evolution of magmato-tectonic activation of the Meso-Cenozoic Era in Eastern China. Acta Geologica Sinica, N 2, P.111-122 (in Chinese). Zhang Qinwen. (1985) .--T~control of plate tectonics over the tectono-magmatometamorphic belts in the Shanjiang Region of Qinghai-Xizang (Tibet) Plateau. Scientific papers on geology for international exchange-prepared for the 27th International Geological Congress (2) .publ. House of Geology, Bei jing, China (in Chinese) . MESOZOIC SUTURING IN THE HUANAN ALPS AND THE TECTONIC ASSEMBLY OF SOUTH CHINA Kenneth J.Hsii Geological Institute Swiss Federal Institute of Technology 8006 Zurich, Switzerland Sun Shu and Li Jiliang Institute of Geology Academia Sinica Beijing, China ABSTRACT. The notion that the whole of South China was a post-Caledonian platform is clearly proven wrong by the evidence of Mesozoic deformation and igneous activity. Reinterpretation of map patterns suggests the presence of a Mesozoic orogenic belt comparable to Southern Appalachians in dimensions and in style of deformation. Field work in 1985 led to the discovery of possible flysch and rigid basement nappes, which have been thrust onto a carbonate platform. A suture zone has been identified. The Huanan Alps owed their origin to the early Mesozoic Indosinian Orogeny, which eliminated a Paleo-Tethyan Seaway, the Xiangganjhe Ocean between the Yangtze and Huanan Terranes. 1. INTRODUCTION The geology of South China has been interpreted as related to Circum-Pacific underthrusting (e.g. Jahn and others, 1976). The Mesozoic and Cenozoic igneous activities of southeastern China have been interpreted as related to Jurassic and Cretaceous subduction of the Pacific Plate along a west-dipping Benioff Zone. In such a scheme the part of South China underlain by the Precambrian and Paleozoic terranes is considered a platform, on the margin of which an Andean Arc has been constructed (e.g., Hamilton, 1979). The notion of a "South China Platform" is erroneous and misleading. Chinese scientists have always been aware of the fact that the "platform sediments" were deformed and intruded by granites during the Mesozoic (e.g., Huang, 1980; Ren, 1984). Huang (1978) chose, therefore, an ambiguous term "paraplatform" to designate this deformed belt. During a visit to China in 1979, Hsii studied geologic maps of China and came up with the idea that the tectonics of South China might be interpreted with reference to a collision type of orogen. A glance at the 1:4,000,000 geologic map of China was sufficient to reveal some resemblance between the tectonic pattern of South China and that of Central and Southern Appalachians. The geologic map of eastern Sichuan 551 A. M. C. ~engor (ed.), Tectonic Evolution olthe Tethyan Region, 551-565. © 1989 by Kluwer Academic Publishers. 552 and northern Guizhou (see Figure 1 for location of provinces) looks very much like the geologic map of Pennsylvania and Maryland. In fact, the overall structural zonation of the two orogenic belts are amazingly similar: The flat-lying Mesozoic strata around Chengdu are equivalent tectonically to those of the Allegheny Plateau, and the long and narrow folds around Zhongqing (Region 1 in Figure 2) find a suitable comparison in the anticlines and synclines of the Valley and Ridge. The folds and thrusts of Guizhou (Region II in Figure 2) are reminiscent of the lower Paleozoic deformation of the Great Valley of the Appalachians. If we use Blue Ridge, the next tectonic element of the Appalachians as a model for comparison, we would have to interpret the "Precambrian" in Figure 3, southeast of the carbonate platform, as having been overthrust onto the Paleozoic. Finally, the metamorphic and plutonic rocks of south-central and southeastern China could be compared to the Appalachian Piedmont. HsU (1981) proposed this speculative, if not outrageous, hypothesis in an invited contribution to the proceedings of the Chinese Academy of Sciences. The idea produced echos (e. g., Zhu, 1983). A cooperative project was, therefore, initiated in 1985 by the Geological Institutes of ETH and of Academia Sinica to test the hypothesis. 100' 110' 100' Fig.1: Geographical Map of China, showing the location of Chinese provinces. 553 South China has been mapped on a scale of 1:200,000 and in places on larger scales. Tectonic interpretations are, however, much condi tioned by "wixist" thinking. We took a field excursion to Guizhou and Hunan Provinces in 1985 to check the possibility of alternative interpretations, and visited several key localities. This article presents the preliminary conclusion after the first year of our field investigations. 2. SEDIMENTARY HISTORY The geologic record of the southern and southeastern provinces of China is significantly different from that of the more interior parts of south-central China. The line separating the two different terranes runs pretty much along the railroad track from Hangzhou in the east to Pingxiang (Jiangxi Province) in the west, in river valleys underlain by late Mesozoic and Cenozoic strata. The southwesterly continuation of this line of separation should be located in between the Guizhou Folded Belt (Region II in Figure 2) and the southeastern granite terrane. We postulate that the line marks the trace of a suture zone, and propose to call it the Xiangganzhe Suture because it stretches across the provinces Xiang, Gan and Zhe, which are abbreviated Chinese expressions for Hunan, Jiangxi, and Zhejiang provinces respectively. The interior region north and west of the suture is designated the Yangtze Terrane. The southeastern terrane is named Huanan (=South China) Terrane. 2.1. Stratigraphic Record of Yangtze Terrane The region is underlain by continental crust and a thick sedimentary sequence ranging from late Precambrian to Cenozoic in age. The terrane is bounded on the north by the Qinling Suture, which marks the site of the Mesozoic (Indosinian) collision of the Yangtze Terrane and the Sinokorean continent (Figure 2). The western boundary is marked by the Kangtien Suture, where the Yangtze Terrane collided with Tibet, and by the Lungmenshan Thrust Belt between the Yangtze and Songpang-Gangtze blocks. The boundary separating the Yangtzte from the Huanan Terranes is the Xiangganzhe Suture. The late Precambrian Sinian System of this terrane has its type locality in the Yangtze Gorge section and includes shallow-water sediments and tillites. Depositional depth was greater in the more southeasterly direction, so that sha1y sediments become more dominant on the southeastern margin of the terrane (Wang and others, 1985). The Paleozoic and earlier Triassic sediments constitute an almost uninterrupted sequence of sedimentation on continental shelves and margins. Carbonate sedimentation predominated, although changes into sha1y facies toward the southeast are common in rocks of several ages. Evapori tes are present in Cambrian and in Triassic formations. The 554 distribution of the Triassic evaporites delimits the disharmonic folding of post-Triassic strata in the Sichuan Basin (Region I in Figure 2), whereas the distribution of Cambrian evaporites coincides more or less with the Guizhou Folded Belt (Region II in Figure 2). The Paleozoic record indicates an early Devonian phase of emergence, when Devonian terrestrial sediments were laid down on older marine sediments. And a late Permian phase of accelerated subsidence occurred so that upper Permian radiolarian chert overlies coal measures. XI'AN Qing/ing· ~~ AlilllilhmT11l111Tl1Illh-\ ~ '\ Vietnam .......,.,.,., Suture zone -+- Anticline -+- Syncline ~ Precambrian cover • Flysch and slates o Paleozoic and Triassic IIIIllllJ Jurassic and younger ~ Precambrian basement • Ophiolite melange W Paleozoic and Precambrian granites ~ Mesozoic granites Fig. 2: Tectonic Sketch Map of South China. South China, bounded by Qinling Suture on the north and by Kangtien Suture on the west, is itself a collage of allochthonous terranes. The Xiangganzhe Suture extends from west of Changsha in Hunan (=Xiang) Province, to south of Nanchang in Jiangxi (=Gan) province, to south of Hangzhou in Zhejiang (=Zhe) Province. The tectonic pattern is comparable to that of the Appalachians. The Sichuan Basin (I) is equivalent to the Allegheny Plateau; the Guizhou Folded Belt (II) and the Jiangnan Folded Belt (IV) are equivalent to Valley and Ridge and the Great Valley provinces; the Banchi Group of flysch and slates is equivalent to the Taconic Klippe; and the granitic terrane of south-eastern China is equivalent to the Appalachian Piedmont. The Tiengui Folded Belt (V) is an east-west trending unit, comparable to the Ouachitas of North America. 555 The Permo-Triassic sediments of the southern Tiengui region, extending from Yunnan to Guangxi include a thick sequence of submarine volcanics, volcaniclastic turbidites, and hemipelagic shales (Hou and Huang, 1984). They are typical of sediments deposited in a back-arc basin, but the arc is now buried under overthrust sheets, and, as we shall discuss in a latter section, we are not certain if the arc is marginal to the Yangtze Terrane or to the Huanan Terrane on the other side of the Xiangganzhe Ocean. The post-orogenic sediments of the Yangtze Terrane, the Indosinian Molasse, include sandstones, coal measure and red beds, ranging in age from late Triassic to Cenozoic. They have all been deposited in continental environments, and are more than 4,000 meters thick in the foreland basin of Sichuan. Volcanic rocks and volcaniclastic deposits are common in the Tiengui area. The most puzzling group of formations of the Yangtze Terrane is the Banchi Group ("Precambrian" of Figure 3, and Flysch Nappes of Figure 2). This sequence of slates and flysch-like interbeds of sandstones and shales has yielded no megafossils. Lenticular bodies of ophiolitic rocks, including pillow lava, gabbro, peridotite, duni te, and serpentinite, are present locally as exotic inclusions. The Banchi rocks, commonly present in the mountains, are surrounded by outcrops of Paleozoic carbonates in hilly regions, and they are capped by Precambrian granite on top of several peaks. Assuming that the unfossiliferous rocks constitute the anticlinal cores of the Guizhou Folded Belt, they were interpreted as the Precambrian basement under the Sinian sediments (Wang and others, 1985). We believe, however, that the Banchi outcrops are erosional remnants of nanpes, which have been thrust from the southeast onto the carbonate platform. The correct interpretation of the significance of the Banchi rocks is the key to the understanding of the tectonics of South China. 2.2. Sedimentary Record of Huanan Terrane In contrast to the coherently folded sequence of the Yangtze Terrane, the pre-orogenic strata of the Huanan Terrane are intruded by granite, and those of the coastal provinces occur mainly as roof pendants in Mesozoic plutons. The terrane can be further subdivided on the basis of their sedimentary and tectonic history into the Hunan Folded Belt and the Coastal Volcanic Belt. 2.2.1. Hunan Folded Belt The Precambrian Sinian strata southeast of the suture in Hunan (province south of the lake, not Huanan which means South China) consist mainly of terrigenous clastics; they may represent the deep-water facies equivalent of the Yangtze Sinian (Anonymous, 1973). The lower Paleozoic 556 t I .CHENGDU D Cretacious / Cenozoic D Paleo~oic/ Mesoloic D Precambrian ~ Volcanics o Yanshanian granites o Indosinian granites [3 Ullrabasics ~Folds Y Faults and Thrust faults __ Edge 01 triassic carbonate plallorm Fig.3: Geologic Sketch Map of South China. Modified from Geological Map of China, 1:4000000, 1976. Note that the Banchi Group of South China was considered Precambrian. includes thick flysch turbidites and graptolite-bearing shales in south-central Hunan, overlain unconformably by middle or upper Devonian terrestrial deposits. The upper Paleozoic formations are mainly platform sediments, carbonates, coal measures, and shales, similar to be mainly carbonate sequence of the Yanatze Terrane. The post-orogenic strata of Hunan are late Triassic and younger continental red beds. 557 2.2.2. Coastal Volcanic Province Precambrian gneisses, amphiobolites, marbles, and slates are present in this region and they constitute the pre-Sinian basement south of the suture zone. The Sinian sediments are mainly terrigenous. Thick sequences of lower Paleozoic flysch turbidites and of graptolite-bearing slates, slightly metamorphosed, occur in Zhejiang, Fujian and Guanmgdong Provinces (Anonymous, 1973), and they have been referred to as Coastal Metamorphics. Devonian quartzite and other coarse clastics, more than 2,000 meters thick locally, overlie older sediments unconformably. The Carboniferous, Permian, and earlier Triassic includes shallow-water carbonates, coal measures, and radiolarian chert, as well as marine sandstones and siltstones. Their facies development is thus similar to the coeval deposits of the Yangtze and Hunan regions. The post-Indosinian sediments are predominately volcanic, including intermediate to acidic volcanic flows and pyroclastics, as well as volcaniclastic sandstones and conglomerates of mainly Jurassic age. The Cretaceous and Cenozoic are red beds. 3. TECTONIC INTERPRETATIONS The classic interpretation of a post-Caledonian South China Platform is obviously incorrect. Ren (1984), in a masterful analysis of the Indosinian Orogeny and its significance in the tectonic evolution of China, has clearly recognized the intense folding and magmatism of South China, and cited evidence for the Mesozoic age of the deformation. Unfortunately, bounded by tradition, he referred to the structures as "folds in platform", and continued to used the term "Yangtze Paraplatform", which serves only to confuse the poorly informed. From a reading of the map patterns, Hsli postulated in 1981 that the deformation of South China resulted from continental collision. He noted the occurrence of ultramafic rocks in eastern Guizhou and western Guangxi and suggested that "we should search for a suture zone in the vicinity of those ultramafic rocks" (p.109). Search we did this summer, and we found ophiolitic melanges above carbonate platform rocks in those regions beneath Precambrian granite. We now entertain the hypothesis that this ophiolite melange is the suture between the two small lithospheric blocks, the Yangtze and the Huanan Terranes. The ophiolites and the Banchi Flysch have been squeezed out of a trench and overthrust northwestward from the Xiangganzhe Suture Zone. A second suture zone is represented by a discontinuous belt of ultranafic outcrops in the metamorphic terranes of the coastal regions and has been compared to the Kings Mountain Belt of the Appalachians (Hsli, 1981, p .109). Li and others (1982, p.l3) described this "deep fault belt extending from the Lishui in Zhejiang Province to Haifeng in Guangdong Province for about 1,100 km, with a width ranging from 20 to 50 km, along which there are sparsely exposed the metamorphosed intermediate and basic volcanics, 558 siliceous rocks, and stratified mafic and ultramafic rocks." They assigned a Paleozoic age to the Lishui-Haifeng Lineament, implying that the Huanan Terrane may have resulted from a pre-Indosinian suturing of t,vo allochthonous terranes along a Caledonian subduction zone. We shall present our preliminary conclusions in the following sections, based primarily upon literature survey and field observations in the Guizhou and Hunan Provinces. 3.1. Deformation of Sedimentary Cover in Yangtze Terrane The Yangtze Terrane is characterized by thin-skinned deformation of its sedimentary cover. The direction of tectonic transport is northwestward, and the intensity of deformation increases gradually to the southeast. Four tectonic and physiographic units have been recognized in the area, namely the Chengdu Plains, the Zheongqing Hills, the Guizhou Folded Belt, and the Banchi (Flysch) Nappes. A fifth, the Tiengui Folded Belt (Region V of Figure 2), constitutes a separate tectonic element in the Yunnan (also known as Tien) and Guangxi (also known as Gi) provinces, with a vergence from south to north. 3.1.1. Decollement deformation under Chengdu Plains and Zhongqing Hills. The Red basin (Region I in Figure 2) is bounded by the Qinling System on the north, the Kangtien System on the west, and by the Guizhou Folded Bel t on the southeast. The Chengdu area is underlain by a foreland basin. Numerous wells have been drilled through the Mesozoic strata. Hsu examined cores of Triassic eva pori tes, and found evidence of strong deformation in salt beds beneath the flat-lying Mesozoic sediments of the basin. The evaporite evidently acted as detachment horizons to facilitate the thin-skinned thrusting of the Mesozoic cover of the Yangtze Terrane. The post-evaporite strata have been thrown into sharp folds in the hilly country around Zhongqing. The narrow and long anticlines, with nearly vertical limbs, have a style of deformation similar to that of the Jura-type of folding. The deformation of the Red Basin took place during late Mesozoic because Jurassic and early Cretaceous deposits have been folded. 3.1. 2. Guizhou Folded Belt. The strata of this folded belt are mainly Paleozoic carbonates, but they grade laterally southeastward to shaly units. The detachment horizon for thin-skin deformation is apparently an evaporite unit of middle to late Cambrian age. Faults are common on the flank of anticlines (see Figure 3). Detailed investigations by geophysical survey and by drilling have revealed that over thrusting predominated. The deformation of the sedimentary cover is tectonically comparable to that of the Valley and Ridge Province of the Appalachians, and is far less severe than that of the Helvetic nappes of the Alps. In the southeast part of this belt adjacent to the Banchi Flysch outcrops, the folded strata are mostly older than the evaporitic decollement horizon. Lenticular bodies of an upper Cambrian carbonate formation occurs as exotic blocks in a tectonic breccia or melange. The middle 559 Cambrian is a dolomite and is invariably badly fractured. The Guizhou Folded Belt extends eastward across Hupeh, northern Jiangxi, southern Anhui and southern Jiangsu Provinces (Region IV of Figure 2), where the eastern end of the Qinging Range (known locally as the Dabie Mountains) is buried under the allochthonous folds and thrust sheets from the southeast. The Tiengui Folded Belt in the south (Region V of Figure 2) has a more or less east-west trend. The mainly Permo-Triassic back-arc rocks seem to have been displaced northward by overthrusting from the south, and they may be allochthonous and thrust onto the Yangtze Terrane. 3.2. Banchi (Flysch) Nappes Northeast of Suture Zone Southeast of the Guizhou Folded Belt, near the postulated suture zone, are rather mountainous terranes underlain by thick sequence of unfossiliferous slates and turbidites. They belong to the Banchi Group and have been considered Precambrian (Figure 3). According to this orthodox interpretation, the unfossiliferous strata of flysch facies constitute the core of anticlinoria. Alternatively those rocks may be trench sediments, equivalent in part at least to the platform carbonates of Yangtze, and they could have been thrust on top of the carbonate platform during the Indosinian Orogeny. We are adopting this second interpretation because of the following observations: a) The Banchi Flysch beds have apparently been compressed during tectonic transport and are characterized by steeply dipping axial-plane cleavage. If the flysch and the Paleozoic beds form autochthonous anticlines and synclines, the cleavage should have been evident in both. In fact, the latter have little or no cleavage, but exhibit evidence of intense bedding-plane shear. b) The middle Cambrian strata are everywhere fractured near the Banchi outcrops, in the fashion of the brittle rocks at the base of the Lewis Overthrust of western North America. This style of deformation cannot be explained if the dolomite is only mildly folded in an autochthonous terrane. On the other hand, the intense fracturing is easily explained by our hypothesis that the dolomite underlies a major thrust plane beneath the Banchi Flysch Nappe. The movement of the nappe acted like a plunger and pushed the Paleozoic strata above the middle/upper Cambrian evaporite northwestward to form the folds of the Guizhou Belt, and, at the same time, pulverized the brittle dolomite under the decollement horizon. c) The contact between the Banchi Flysch and a lower/middle Cambrian limestone is clearly a fault contact on the east side of Fangjingshan, which rises 2,500 m. above the surrounding carbonate terrane in eastern Guizhou. If the Fangjingshan Banchi represents the 560 Precambrian core which had been upthrust in the mode of the Big Horn Range of the Wyoming Rockies, the differential uplift should have been 4000 or 5000 meters, and we should also find post-orogenic basins filled by terrigenous clastics such as the Big Horn and Powder River Basins. In fact, the Banchi Flysch of Fangjingshan sits in the middle of a carbonate terrane in the manner of a huge allochthonous slab, forming a synform between anticlines of platform-carbonate strata. d) The Banchi Flysch is overlain tectonically by an ophiolite melange, which is in turn overlain by Precambrian granite in Fangjingshan and in other mountains of southeastern Guizhou and northwestern Guangxi. The whole package has apparently been thrust from the southeast onto the carbonate platform. The tectonic succession is comparable to that in the Alps, where the Australpine nappe (Silvretta Grani te) is thrust over an ophiolite melange (Arosa Schuppen Zone), a flysch nappe (PrMttigau Flysch) and onto the carbonate platform (Helvetic carbonates). The observed tectonic succession in Fangjingshan cannot be explained by a theory of upthrusting tectonics. Flysch terranes capped by Precambrian granite form a discontinuous belt north and west of the Xiangganzhe Suture (Figure 2). We entertain the working hypothesis that they are all erosional remnants of a huge pile of nappe complex, being comparable to Prealpine klippes of the Alps. Whereas the slates, turbidites and ophiolites should have been squeezed out of a deepsea trench, the granite might have been displaced hundreds of kilometers and had their home in the Huanan Terrane southeast of the suture. As John Rodgers aptly put it, the Fangjingshan rocks sit out there on a folded carbonate platform like the Taconic Klippe of the Appalachians. By the postulate of a Fangjingshan Klippe, we might have started a controversy, which could continue till the next decade, if not the next century. A Sino-Swiss team now is now doing field work in the area to gather evidence to resolve the question if they are autochthonous or allochthonous. 3.3. The Xiangganzhe Suture The Xiangganzhe Suture is recognized by the presence of a discontinuous belt of ultramafic exotics in slates and turbidites. The segment near Tienyang, west of Changsha, has been mapped in detail and thoroughly explored with boreholes drilled by the Exploration Team 107 of the Hunan Geological Survey. We visited the outcrops there this summer and were given the following information: Four large slabs of ophiolites have been found in the Banchi Group near Tienyang in Hunan, just across the provincial border from Guizhou Province. The largest is 27.5 km long, but only 0.41 km thick. The slabs are all dipping steeply, and some have the complete ophiolite sequence of basalt/diabase, gabbro, plagioclase pyroxenite, peridotite, and dunite, with the top of the sequence facing northwest. Boreholes, 561 drilled into the largest in connection with search for chromite deposits, revealed the lenticular nature of the slab, which thins to less than half the outcrop thickness at a depth of 700 m. The geologists of the exploration team interpreted the ophiolites as intrusive bodies into the slates and turbidites, although they noted the absence of intrusive contacts or thermal metamorphism. Where exposed, a fault or a shear zone separates tho ophiolites from the Banchi Flysch, a relation similar to that observed in the Fangjingshan area. The basal t/ diabase at one outcrop had been considered a dike intrusion. We found, however, abundant inclusions of metamorphosed pelagic oozes in the vesicular basalt and recognized its origin as a lava flow extruded onto a deepsea floor. In the suture zone of the more easterly segment near Yueshan, Jiangxi Province, exotic blocks in the Banchi Group include not only ophiolites, but also Permian spilite and radiolarite (Unpublished information, Jiangxi Geological Survey, 1981). Their presence in a melange confirms our postulate of an Indosinian age for the tectonic suturing of the Yangtze and Huanan Teranes. Still farther to the east, the Xianggangzhe Suture is defined by the Jiangshan-Shaoxing Line. The melange is largely covered by Cretaceous and younger sediments, although some ultramafic rocks are present here and there (Xiao and Wang, 1984). The suture has been called a "deep fault", beause it separates terranes of distinctly different geology. 3.4. Huanan Terrane as Overriding Block The Huanan Terrane is noted for its many igneous intrusions and extrusions. The Precambrian basement has a very limited distribution. The Sinian (late Precambrian) and Paleozoic sedimentary cover forms a recognizable folded belt in Hunan Province on both sides of the suture, but occurs mainly as roof pendants in the more easterly provinces. The Institute of Geochemistry, Academia Sinica (1980) published a map showing the distribution of igneous intrusions and their ages. Other than Precambrian granites, which occur as klippes on top of the Yangtze Terrane, the intrusive granites are almost all found southeast of the suture in Huanan. The age varies systematically, and is progressively younger in a southeasterly direction. The Paleozoic granites, commonly referred to as Caledonian in Chinese literature, are present mainly in the Hunan Province. They are intrusi ve into the lower Paleozoic and are overlain unconformably by Devonian clastics and they yield radiometric ages of about 400 m. y. (Anonymous, 1973, p.90). Late Paleozoic granites, yielding radiometric ages of 300 m.y. or somewhat less, have only recently been recognized, mainly in an eastnortheast-trending belt just across the suture 562 southeast of Nanning in Guangxi Province (still shown as Mesozoic Granites in Figure 2, which was modified from a geologic map published in 1976); scattered outcrops of granites from Hunan, Jiangxi and other provinces have also been attributed to Hercynian(Institute of Geochemistry, 1980). The Huanan granites are predominantely Mesozoic, yielding radiometric ages mainly in the range of 210 to 100 m. y. (Jahn and others, 1976). Younger granites, somewhat more alkaline in chemistry,are present in a coastal belt and are as young as 70 m.y. (Xu, 1984; also Yang Koyou, personal communication, 1985). A still younger age of 56 m. y. dates a "quartz porphyrite dike" of an offshore island in Taiwan Strait (Jahn and others, 1976, p.770). The Mesozoic, or the so-called Indosinian and Yanshanian, granites of China have relatively high initial stronsium values of 0.7060 to 0.7159. Pitcher (1982) recognized that the Yanshanian granites of Jurassic and late Cretaceous age belong to the S-type, typical of Hercynian orogenic belts of continental collision. The late Cretaceous and early Paleogene granites of the coastal belt, with their low initial ratios of about 0.706, belong to the I-type, typical of Andean margin. We note with interest that Pitcher recognized another Andinotype belt northwest of the Yanshanian Hercynotype belt, in the area where the granites are considered Indosinian or early Mesozoic in age (cf.Pitcher, 1982 & Institute of Geochemistry, 1980). The coastal volcanics are associated with terrestrial sediments containing Jurassic and Cretaceous plant fossils; they belong apparently to the same period of igneous activities which produced the Yanshanian granites at depth. 4. CONCLUSION The Yangtze and Huanan Terranes constitute an orogenic belt of the collision type. The Yangtze Terrane is underlain by Precambrian continental crust. In early Triassic, the Yangtze was a carbonate platform surrounded by deep oceans, like the Bahama Platform today (Figure 4A), before continental collision on three sides converted the platform into an inland basin. The geologic record indicates an early Mesozoic, or Indosinian, collision of Yangtze and Huanan. The southeastern border of the ancient Yangtze Platform had remained a passive margin from late Precambrian (Sinian) to middle Triassic. Post-collision sediments of late Triassic, Jurassic and early Cretaceous age overlie the deformed and eroded passi ve-margin sediments unconformably, and they have been folded and overthrust. Upper Cretaceous and Cenozoic sediments were laid down in post-orogenic basins of extensional origin. 563 The record of igneous activity suggests that the northwestern border of the Huanan Terrane was an Andinotype margin since mid-Paleozoic, if not earlier (Figure 4A). The Banchi Flysch was probably laid down in a trench on that active margin, on ocean crust which is now preserved only as exotic blocks in ophiolite melanges. Mid-Paleozoic, late Paleozoic, and early Mesozoic granites of Huanan were intruded into this Andinotype of orogenic environment, and are thus I-type as the available data suggested (Pitcher, 1982). The Yanshanian granites were emplaced after the Indosinian collision. Continued underthrusting of crustal wedges beneath the Huanan Terrane, led to thermal regimes conductive to anatexis, and partial melting of crustal Paleozoic oj. ... ... ... ... .. ... + + .. .. ... + .. -+ +++-+++~ f- .... -+ .. -t ... t 1" Triassic b) Late Mesozoic ----'--'--'--'---'-'--'--, , c) Fig.4: Tectonic Evolution of Huanan Alps. a) Paleozoic Xiangganzhe Ocean. b) Indosinian Collision. c) Late Cretaceous and early Tertiary subduction of the Pacific Plate under the Huanan Terrane. 564 materials gave rise to the S-type of granites of Huanan. The young coastal granites of I-type seem to owe their origin to the subduction of the Pacific Plate under Huanan in latest Mesozoic or early Tertiary time (Figure 4C). Pitcher (1982, p.28) notes the characteristic tin and copper mineralizations associated with the Hercynotype and Andinotype of granites respectively; the distribution of mineral deposits in South China gives support to his generalization. The more or less east-west trending Tiengui Folded Belt seems to represent an ancient island-arc assemblage that had been thrust over Guizhou Folded Belt; the relation is reminiscent of that between the Ouachitas and the Southern Appalachians. If the back-arc basin was a marginal basin on the southern edge of the Yangtze Microcontinent, the Permo-Triassic subduction zone should have a nother1y dip, in a direction opposite to that under Huanan. More likely, however, the island-arc had its apex facing north; a south-dipping subduction should thes have been responsible for the back-arc seafloor spreading and for the genesis of the late Paleozoic granites southeast of the suture near Nanning (Figure 2). The traditional view of a Caledonian Orogeny in South China was based upon stratigraphic information: The Devonian is a sandstone or conglomerate, resting unconformably on older marine rocks. The Devonian clastic wedge thins out to the west, indicative of an easterly source from a hypothetical landmass called "Cathaysia" (cf. "Appalachia", Grabau, 1924). The pre-devonian deformation did not greatly disturb the Yangtze Terrane. Both its northern margin bordering the Qinling segment of the Paleo-Tethys and its southeastern margin on the shore of the Xiangganzhe Ocean are underlain by a largely conformable sequence, ranging from late Precambrian to middle Triassic in age, although lower Devonian sediments are in places missing because of local tilting or relative sea-level change. We have no evidence that continental collision took place either in the Qinling Range, or in the Xiangganzhe region during the pre-Devonian deformation. Yet the deformation must have greatly affected the more eastern part of the Huanan Terrane, because a thick Devonian clastic wedge was laid down there on top of folded and eroded lower Paleozoic flysch-like sediments. We tend thus to agree with Li and others (1982), who attributed a Paleozoic age to the Lishui-Haifeng Melange, implying a Caledonian suturing of the two halves of the huanan Terrane. 5. ACKNOWLEDGMENT We are grateful to the many people who were our guides during our 1985 excursion to Guizhou and Hunan, particularly Mr.Wang Yangyeng of Guizhou Geological Survey and Mr.Yang Kouyou of Institute of Geochemistry, Academia Sinica, Guiyang. We thank ProLZhang Guowei of Northwestern University. Xian, for his guidance in an excursion to examine the Indosinian tectonics of the Qinling Range. We are indebted to many 565 colleagues and local government officials for their logistic supports. The field work is supported by a research grant from the Swiss Federal Institute of Technology and one from the Institute of Geology, Academia Sinica. REFERENCES CITED Anonymous, 1973. Geological Atlas of China (in Chinese).Qhinese Academy of Geological Sciences, Ministry of Geology, Beijing, China, 149 pp. Grabau,A.W., 1924. Migration of geosynclines. Geol.Soc.China Bull., v.4, p.207-349. Hamilton,W., 1979. Tectonics of the Indonesian Region. U.S.Geol.Survey.Prof. Paper 1078, 345 pp. Hou,F., and Huang,J., 1984.Research into the Permian and Triassic volcaniclastic turbidite of Nanpan River Seg- a unique turbidite mode without submarine fan. Acta Sedimentologica Sinica. v.2, p.19-32. . - HsU, K.J ., 1981. Thin-skinned plate-tectonic model for collision-type orogenesis.Scientia Sinica. v.24, p.100-llO. Hunag,Chi-ching, 1978. An outline of tectonic characteristics of China. ~clogae geol.Helv., v.71, p.6ll-635. Huang,Chi-ching (editor),1980. Tectonic Evolution of China, -explanation of 1:4,000,000 tectonic map of China (in Chinese), Scientific publisher, Beijing, China, 124 pp. Institute of Geochemistry, 1980. Distribution of Granites and of Fracture Zones in South China. Map published by the Institute, Academia Sinica, Guiyang,China. Jahn,B.M., Chen,P.Y. and Yen,T.P., 1976. Rb-Sr ages of granitic rocks in southeastern China and their tectonic significance. Geol.Soc.America Bull., v.87, 763-776. Li,C., Wang,Q., Liu,X., Tang,Y., 1982. Explanatory notes to the tectonic map of Asia. Chinese Academy of Geological Sciences, Ministry of Geology, Beijing, China, 49 pp. Pi tcher ,W . S ., 1982. Granite type and tectonic environment. in K. J . HsU (editor) Mountain Building Processes, Academic Press, London, p.19-40. Ren,J., 1984. The Indosinian Orogeny and its significance in the tectonic evolution of China. Chinese Acad. Geol. Sci. Bull., no.9, p.43-53. Wang,Y., Xie,J., Chen,Y., Qin,S. and Zhu,S., 1985. Evolution of upper Precambrian volcano-sedimentary sequences in the western part of Jiangnan stratigraphic province, China. Precamb. Research, v .29, p.109-ll9. Xiao,X. and Wang,F., 1984. An introduction to the ophiolite of China. Chinese. Acad. Geol. Sci. Bull., no.9, 19-30. Xu,B:, 1984. Isotopic geochronology of Yenshanian granites in Zhejinag Province. Geochimica for 1984, p.2l7-225. Zhu,Xia, 1983. On the Geodynamic Framework of Chinese Mesozoic and Cenozoic oil- and gas-producing basins. In Zhu Xia (editor) Tectonic Evolution of Chinese Oil- and Gas-bearing Basins. Scientific Publisher, Beijing, China, p.l-10. THE SHAN PLATEAU AND WESTERN BURMA: MESOZOIC-CENOZOIC PLATE BOUNDARIES AND CORRELATIONS WITH TIBET A.H.G.Mitchell ABSTRACT. Four suture zones are probably present in Burma. Regional evidence indicates that the eastern Burma-Chiang Rai-medial Malaya suture resulted from closure of an oceanic basin, ocean I, in the early Triassic. Projection from Tibet suggests that ocean II, north of the Lhasa block, closed in the end-Jurassic, but together with the Lhasa block itself, the suture is largely or entirely buried in Burma. The Mount Victoria Land block rifted from Gondwanaland in the Jurassic and collided with Burma, following northeastward subduction of ocean III, in the early Cretaceous. Triassic flysch and ophiolite overlying this block in western Burma correlate with the Triassic flysch and overlying Zangbo ophiolite in Tibet, suggesting that the Triassic flysch of Tibet was never adjacent to India. In the Eocene, before dextral displacement on the Sagaing Fault, India collided with northern Burma following subduction of ocean IV. This resulted in elevation of the Mogok Belt and Shan Plateau above a southeastward continuation of the Himalayan Main Central Thrust. 1. INTRODUCTION The Shan Plateau ) ,Fig.l) has been interpreted as part of the Western Southeast East Asia or Shan-Thai) ,1), block which collided with an Indochina block to the east in the early Mesozoic(2,3,4,S,6,7,8,9,lO). The Plateau is bordered on the west by the Mogok Belt(ll), beyond which are the Central Lowlands and medial volcanic arc; further west the Indoburman Ranges include a tectonic window of schist beneath ophiolite and Triassic flysch (Fig.2). 567 A. M. C. :jengor (ed.), Tectonic Evolution of the Tethyan Region, 567-583. © 1989 by Kluwer Academic Publishers. w 568 :r: (J) w o a; 20' I.' 10' ..... _.) ( --\-t----.~,.~,-.--.---.-~j 100" CHI N A LAOS THAILAND , '''. I ., \./) ~ ... i 200km 100' 24 22' 20' I.' Fig.L Major structural units, Shan Plateau and western Burma. a, b - locations of cross-sections shown in Figs.6 and SB. ARAKAN COASTAL PLAIN PNDOBURMAN RANGES I CENTRAL VALLEY W. TROUGH I ~~'I E. TROUGH S HAN PLATEAU ~ ~ SCARP G') !D THRUSTS iN rIA;~:: :'l f!l (LATE ME!>OZ) . -; PRE-CAMB-CRET (? U TRIASSIC) ULTRA- .01 ~ ~ GRAN', MrlCS OCEAN '00 500 Fig.2, Schematic geological cross-section, western Burma and margin of Shan Plateau around Lat. 2loN. 4 The aims of this paper are to describe briefly the main features of the geology of the Shan Plateau and western Burma, and to infer the major tectonic events based on the rock relationships in Burma and possible correlations with Tibet. E 569 Southeast Asia has probably rotated clockwise with respect to Tibet since the Palaeocene 12 . Palaeo-geographical orientations given in the text are based on the assumption that the Burma arc faced southwestwards before the Eocene, rather than westwards as at present. 2. DESCRIPTION OF GEOLOGY A. Shan Plateau The Shan Plateau (Fig.l) comprises a thick succession of Precambrian, Palaeozoic and Mesozoic predominantly sedimentary rocks(13,14,15). Similarities in the Palaeozoic sequence to that of western and southwestern Thailand and western Malaya were first noted by Stauffer(16). The stratigraphic column shown in Fig.3 is representatitve of the western part of the Plateau; it may be noted that the Upper Carboniferous to Lower Permian diamictites or pebbly mudstones are confined to the westernmost part of this zone, between the Shan Scarps and Mogok Belt (Fig.4). FORMAnON GROUP INTRUSIVE T'NESS (MEMBER) 1.:.1tt:$1 .7.,:--:_°_:':-; C:NGAL~:~:E ....... " ~~~~~s~ Fig.3. Generalized stratigraphic column, Shan Scarps and adjacent areas(4). 570 The geology of the eastern margin of the Plateau is poorly known. Ultramafic rocks and granites were reported from this area in the early 1970s, and together with the presence of Permian volcanic arc rocks to the east in Thailand, have been used as evidence for a Triassic suture through Burma's eastern salient(2,4), between the Shan-Thai block and Indochina block to the east. SaocunQ • 0 o \) ..... ·.C o ';. 0 II> ... 0\ ~~ 0 o~ i 10 I Fig.4. Geological sketch map, part of Shan Scarps and Mogok Belt with granites at northern end of Western Tin Belt(4,13,14,lS). 571 Along the western margin of the Plateau in the Shan Scarps (Fig.4, SA) a zone of late Mesozoic en echelon north-northwest trending thrusts wi th eastward vergence 17 is cut by high-angle faults. West of the thrust belt pebbly mudstones of Lower Permian and possibly Upper Carboniferous age are intruded by late Mesozoic to early Eocene tin-bearing granites. B. Mogok Belt The metamorphic Mogok Belt of Searle and Haq(ll) lies west and northwest of the Shan Plateau. Northwest of the Plateau the Belt includes rocks of upper amphibolite to granulite facies (Dr .M.S .Garson, oral commun, 1979), which pass southeastwards with decreasing metamorphic grade into schists, late Precambrian turbidites and Lower Palaeozoic sedimentary rocks. Although generally considered to be a continuous sequence, it seems probable that southeast-dipping thrusts or thrust zones are present. w EASTERN TROUGH I MOGOK BELT 1 A MOGOK BELT SHAN-SCARPS, LAT. 21-N w WESTERN SALINGYI TROUGH COMPLEX B SALINGYI, BURMA VOLCANIC ARC LAT. 22°N SHAH SCARPS EASTERN TROUGH WESTERN .. ARGIN SHAN PLATEAU SAGAING FAULT (j) 0 Fig.S. Composite geological cross-sections, A - Shan Scarps and Mogok Belt; references as in Fig.4, location shown in Fig.4; a- biotite schists, b- Upper Proterozoic turbidites, Chaung Magyi Gp, c- metasedimentary rocks, mostly Lower Palaeozoic, d- Lower Palaeozoic sedimentary rocks, e- Upper Carboniferous-Lower Permian diamictites, Mergui Gp, f- Upper Permian to Anisian carbonates, g- Nwalabo Fault Complex, thrust and faulted Permian-Cretaceous sedimentary rocks, h- Jurassic-Cretaceous sedimentary rocks, i- granite, k-alluvium. B. Geological cross-section, Salingyi area, Burma volcanic arc(23); location shown in Fig.l, a- gneiss (projected), b- amphibolite, c- pillow basalt, diabase and gabbro (projected), d- quartz keratophyre, e- mid-Cretaceous hornblende diorite, f- mid-Cretaceous biotite granite, g- hornblende gabbro, h- interbedded tuff and conglomerate with clasts of garnet-biotite schist, quartzite, vein quartz, hornblende gabbro, amphibolite, gneiss, permatite, red and black chert, dacite, i- Lower Oligocene red sandstone, k- alluvium, 1- mostly Upper Tertiary sediments. 572 West of the Shan Scarps and Mergui Group (Fig. 4), the Mogok Belt includes Upper Palaeozoic rocks, metamorphosed to schists and calc-silicates(18); these are locally overlain structurally by unmetamorphosed east-dipping rocks of the Mergui Group. Westwards the metamorphic grade within the Belt increases, and dioritic and granitic plutons intrude gneisses. Along the eastern margin of the Eastern Trough the Mogok Belt is overlain by late Cenozoic sediments, and near Mandalay is truncated by the Sagaing Fault. East of the Mogok Belt, in the northwestern part of the Shan Plateau (Fig.4), biotite schists are exposed as probable tectonic windows beneath late Proterozoic turbidites of the Chaung Magyi Group; the turbidites themselves are overthrust by Lower Palaeozoic rocks. These thrusts are tentatively considered to be related to those bordering the eastern margin of the Mogok Belt. K/ Ar determinations on granite and gneiss samples from the Belt have yielded six ages, mostly on biotite, in the 15 to 30 my range, although some older ages have also been determined(19). The metamorphic age of the Mogok Belt has been considered as Precambrian but most more recent authors have favoured a late Cretaceous or Tertiary age(11,18,20), partly because it was thought that Cretaceous limestones could be traced into the metamorphics. Metchell(2l) suggested an end-Triassic age for a major metamorphic event speculatively related to foreland thrusting following a late Triassic orogeny to the east, but Wolfort et a1.(22) included the "Mogok Gneiss" in the Precambrian. C. Central Lowlands The main topographic feature of the Central Lowlands is the Burma volcanic arc, which includes recently extinct volcanoes and lies between late Cenozoic sediments of the Eastern Trough and the much thicker and in part older Western Trough succession. The oldest dated arc rocks are 100 to 110 my granodiorites and diorites present at Lat 220 N and north of lat 240 20' N Fig.l). In the northern area the plutons, with roof pendants or large xenoliths of gneiss, intrude basaltic andesites, dacites and sedimentary rocks and are overlain by younger volcanic and sedimentary rocks and cut by minor intrusions. At Lat 22 0 N (Fig.5B) minor modifications to the geological map of Barber(23) indicate that gneiss occurs as probable roof pendants in 106+7 my hornblende diorite; the diorite intrudes a large body of amphibolite and is intruded by 103+4 my biotite granite. Undated conglomerates with clasts of schist, gneiss, amphibolite, quartz, dacite, diorite and pegmatite are interbedded with volcanic rocks and overlie small exposures of pillow basalts, gabbro-diabase and diorite. The Western Trough is a broadly synclinal succession of Upper Cretaceous and Cenozoic sedimentary rocks up to 10 km thick. In the east the succession overlies mid-Cretaceous rocks of the volcanic arc, and 573 in the west Maastrichtian conglomerates with andesitic to dacitic cobbles overlie east-dipping Upper Albian rocks of the Indoburman Ranges foothills. Formations in the eastern limb of the syncline cannot easily be correlated with those in the west, and the stratigraphy of the eastern limb is complicated by probable east-directed thrusts. D. Indoburman Ranges Beneath the Maastrichtian conglomerates of the Western Trough the foothills of the Indoburman ranges include an Upper Albian ammoni te-bearing succession of shales and limestones with rare serpentinite sandstone laminae and serpentinite sheets (Fig.6). The upper Albian succession lies with a basal conglomerate on Carnian quartzose turbidites and mudstones(24) to the west. The Carnian rocks, recumbently folded and including widespread broken beds, comprise much of the flysch of Brunnschweiler(25). A dismembered ophiolite and minor amphi bolite, locally present immediately east of the turbidites, lie beneath and in probable stratigraphic contact with the Upper Albian rocks. Hornblend pegmatite dykes in harzburgite have yielded a KjAr age of 158+20 my. Thrust-bounded mud-matrix debris flows or olistostromes of Senonian and probable Campanian age are present locally between the Upper Albian rocks and Maastrichtian succession to the east. w INDOBURMAN RANGES Palaeocene ~ 2kmJi __ -"O::::kCCm_---' l'T R I ASS I C BEL T'l b ! WESTERN TROUGH MAGMATIC ARC E Fig.6. Schematic geological cross-section, Indoburman Ranges(26). Location shown in Fig.l (magmatic arc portion is schematic and refers to Lat.24oN). Schematic cross-section, southern Chin Hills and Western Trough, Burma, surface geology based on Mitchell (4,26) and references therein. (a) Biotite and biotite-graphite schists with intercalated greenstones and marble and in diffuse tectonic contact with (b) quartzose turbidites and carbonaceous mudstones with minor limestone and cherts, recumbently folded, and locally forming 'broken beds', mostly Carnian in age(24); (c) serpentenized harzburgite with hornblende pegmatite veins and areas of pillow lavas, minor massive lava, gabbro and dolerite; (d) carbonaceous ammonite-bearing limestones and black pyritic mudstones and shales containing rare serpentinite sand layers, upper Albian fossils near top and basal conglomerate of Aptian-Albian 574 age with large gastropods, lying unconformably on upper Triassic turbidites and intruded by serpentinite sills with ophicalcite at margins; (e) vein quartz-Triassic sandstone-chert pebble conglomerate, Upper Cretaceous(?Maastrichtian) fauna, lying unconformably on Albian limestone and locally on pegmatite-veined serpentinite sills; (f) olistostromes or melange with blocks of Triassic sandstone, radiolarian chert, basalt, gabbro pebble conglomerate, quartzose conglomerate, serpentinite, ophicalcite, marble and biotite schist, and abundant blocks up to 100 m long of coloured micritic limestone with Senonian (probably Upper Campanian) Globotruncana, in a red and green clay matrix; (g) probable olistostromes comprising mostly structureless, locally tightly folded mudstone with beds and lenses of coloured micritic limestone of Senonian and probably Campanian age, minor conglomerates and thin turbidite sandstones, with blocks of ophicalcite, chert, andesite, and pillow basalt; overlain by siltstones with scattered turbidite units and rare Palaeocene limestones, passing up into (middle?) Eocene cross-bedded sandstones; (h) synclinal fore-arc basin succession of Maastrichtian? and Lower Eocene dacite and diorite boulder-bearing conglomerates and limestones, Middle Eocene to Oligocene turbidites, shales and deltaic sediments, and upper Cenozoic mostly non-marine sediments; (i) magmatic arc with granodiorite batholith yielding 100+5 my K/Ar ages, minor Lower Cretaceous and Upper Cretaceous and extensive Cenozoic volcanic rocks and minor intrusions. Schematic evolution of W Burma. A- eastward subduction of ocean floor; B- entry of a spreading ridge into the trench; C- ophiolite detachment accompanying ridge subduction; D- underthrusting of ophiolite by Triassic continental margin sediments and continental fragment, serpentinization of harzburgite, uplift; E- Albian transgression, perhaps due to uncoupling of subducted continental fragment, followed by rise of serpentinite sheets; F - diapiric rise of melange from deep structural level and its extrusion as debris flows. In the Mt. Victoria area of the eastern Indoburman Ranges biotite and biotite-graphite schists with local greenstones are interpreted as a tectonic window within the Carnian flysch. West of the flysch, broadly folded mudstones and thin sandstones of uncertain age with scattered exotic blocks are overlain by Palaeocene shales and limestones and Eocene sandstones. In the south of the Ranges, on the Arakan coast, broken beds of turbidite sandstones, mudstones and grits with scattered olistostromes mostly of basalt have been described by Brunnschweiler(25); the turbidites yield foraminifera of Eocene age. To the northwest, on Ramree Island, blocks of schist and gneiss have been reported from mud volcanoes overlying Miocene sedimentary rocks. 3. TECTONIC INTERPRETATION AND CORRELATIONS WITH TIBET In the following account the proposed correlations between rock units 575 and tectonic events in Burma and Tibet are revised from those I have suggested previously(4,21,26). A. The Shan-Thai Block-Indochina Block Collision and The Kokoxili Suture In the absence of large-scale geological maps of the eastern part of the Shan Plateau, the evidence for and age of the collision between the Shan-Thai block and Indochina is based on the relative position of three main geological belts to the east and south. The three belts comprise the Permian to early Triassic volcanic arc in eastern Malaya and north central Thailand; the discontinuous zone of serpentinites to the west, passing near Chiang Rai and along the Nan River blueschists(27) in Thailand through medial Malaya; and the continental rocks of the Shan-Thai block comprising Sumatra, western Malaya, western Thailand and eastern Burman, lying west of the serpentinite belt. Immediately west of the serpentinites, the Shan-Thai block is intruded by extensive Middle and Upper Triassic and possibly Lower Jurassic granites. The granites are per-aluminuous, intrude folded and thrust sedimentary rocks, and form the 'Central Tin Belt' of the Indonesian 'Tin Islands', the Main Range Malaya, part of north central Thailand and easternmost Burma and a narrow zone in western Yunnan. Analogy between the Upper Triassic mostly per-aluminuous granites of the Shan-Thai block and the post-collision mid-Tertiary granites of the Himalayas suggests that the collision indicated by the serpentinite belt predated granite emplacement by 10 to 20 my. The Central Tin Belt granites have yielded Rb/Sr ages mostly in the range of 204 to 214 my(7) or approximately Norian, suggesting that collision took place in or before the Anisian, implying the following sequence of events during the Triassic. sw U PERMIAN CARBONATES OVER DIAMtTITES + SHAN-THAI + t t 8LOCKt + A EARLIEST TRIASSIC B LADINIAN OCEAN n OPENED IN EARLY PERMIAN UPPER PERMIAN- ""'" "\""" C END TRIASSIC -EARLY JURASSIC HE INDOCHINA MAGMATIC BLOCK\ FORELAND THRUSTS INDO- CHINA MAIN RA BENTONG EASTERN MALAYSIA LINE MALAYA NCENTRAL THAILAND GRANITES ~ESHANPLATEAU Fig.7. Cartoon illustrating Triassic evolution of the Shan Plateau region. 576 Lste Palaeozoic rifting of the Shan-Thai continer.tal block from Gondwanaland 8 , resulted in generation of Ocean II, with Ocean I to the north of the block (Fig. 7, 8a). The Shan-Thai block moved nor~hwards with consumption of ocean I lithcsphere beneath an arc system extending from eastern Malaya through western Thailand and probably attached to Asia at that time (Fig.7A). Consumption of Ocean I also took place in Yunnan and Tibet, but p:!:'obably involved either southward (9) or both southward and northward subduction. During early Upper Permian to Anisian carbonate sedimentation(13) , the leading edge of the continent entered the trench and began to underthrust an ophiolite and associated arc system to the northeast. In ~he mid to late Triassic a foreland thrust belt, analogous to that of the Lower Himalayas, de7eloped within the continental crust to the southwest (Fig.7B). The deformation, refe 2d to locally as the Indosinian orogeny, resulted in imbrication of lalaeozoic and Permian rocks in the Main Range Malaysia and in northwestern Thailand, and has recently been described by Ko Ko(28) from Bangka Island, where it was pre- to syn-granite emplacement. Generation of the Bangka Island and other late Triassic-early Jurassic granites of the Central Tin Belt can best be explained by anatectic melting of meta-sedimentary rocks of the Shan-Thai block within the foreland thurst belt. In the Main Range foothills of Malaysia the absence of tectonically-emplaced ophiolites and recumbent folds has been used as an argument against a suture zone in the region(29). The narrow width of the medial Malaya belt between the volcanic arc to the east and the belt of granites and sedimentary host rocks to the west is also difficult to explain if a 100 km wide fore-arc system lay west of the volcanic rocks before collision. However, these features have analogies in other late Cenozoic collision belts (e.g. Taiwan), and may reflect syn-collision east-directed back-thrusting; this could have resulted in loss of much of the fore-arc area on the underriding eastern thrust plate, and erosion of nappes of ophiolite and folded flysch from the overriding western plate. Followir.g Bally et al.(30), who indicated that the Red River suture north of the Indochina block was continuous with the Kokoxili (Hoh XiI Shan) suture in Tibet, Mitchell(4) argued that the Eastern Burma-Chiang Rai-medial Malaya suture was continuous with what is now termed the Bangong-Nujiang suture. However, recent evidence(9, 31) indicates that the Kokoxili suture dates from around 200 my, and that the Bangong Co-Nujiang suture south of the Quantang Block is much younger(9,32). This implies that the Kokoxili suture is the continuation of the eastern Burma-Chiang Rai zone (Fig.9), and that the Shan-Thai block is continuous with the Quantang block of Tibet as suggested by $engor(9,33). B. End-Jurassic collision on the Bangon-Nujiang Suture The Quantang Block is separated from the Lhasa Block to the south (Fig.9) by the Bangong Co-Nujiang suture zone with associated tectonic 577 segments of the Donquiao ophiolite(9,3l,34). The ophiolite is overlain unconformably by end-Jurassic clastics and Aptian-Albian red beds(34). It is underlain by Middle and Upper Jurassic sandstones and black shales. The inferred age of collision of the Lhasa block with the Quantang block along the Bangong Co-Nujiang suture is late Jurassic to early Cretaceous(9,3l,32). This collision followed subduction, probably northward(32,34), of an ocean basin which may have opened in the early Permian following deposition of diamictites as discussed by Ridd(8), $engor(9) and Audley-Charles(35). In Burma, rocks equivalent to those of the Bangong Co-Nujiang suture and Lhasa Block have not been identified. They may underlie the Western Trough and Volcanic Arc, and have been underthrust eastwards beneath the Shan Plateau. Speculative collision of a narrow Lhasa Block with Burma, following subduction of ocean II, is shown in Fig.8A,B. C. Early Cretaceous collision in western Burma Stratigraphic relationships indicate that the Indoburman Ranges ophiolite was emplaced onto Carnian turbidites or flysch, and largely eroded, before the Albian. Its emplacement age is hence older than most, and probably all, of the plutonic and volcanic rocks in the Burma volcanic arc. The Carnian turbidites are interpreted as continent-derived sediments deposited west of the ophiolite, and recumbently folded and transformed into "broken beds" during emplacement of the overlying ophiolite nappe. Biotite schists of the Mt.Victoria area, structurally beneath the turbidites, are interpreted as part of a continental fragment, here termed Mount Victoria Land (MVL), which underlay the turbidites and collided with and underthrust western Burma before the Albian. The stratigraphic and structural sequence in' the eastern Indoburman Ranges can be correlated with the Zangbo zone of Tibet described by Allegre et al. (31). In the Zangbo zone pre-110 my ophiolite over lies recumbently folded rocks which include a thick sequence of quartzose Upper Triassic turbidites of similar facies to those in Burma j Upper Cretaceous olistostromes are associated with the Triassic rocks in both zones. Palaeozoic and metamorphic rocks locally present beneath the Zangbo turbidites (Cao Yougong, oral commun.) occupy a position analogous to the MVL schists in Burma. North of and overlying the Zangbo ophiolite, limestones of Aptian-Albian age are overlain by the thick Turonian to Campanian Xigatse Group flysch(36)j the Aptian-Albian rocks show similarities to those overlying the ophiolite of the Indoburman Ranges. The unconformity between Upper Albian limestones and shales and the Carnian turbidites in Burma indicates that the turbidites were highly deformed before the Albian, and in more or less their present position with respect to the volcanic arc. Similarly it can be argued that the 578 INFERRED GOND- OCEANm LHASA BLOCK ? OCEAN :II: SHAN-THAI OCEAN I WANA OPENED IN (LSI OPENED IN BLOCK CSTBI CLOSED TRIASSIC LAND TRIASSIC PERMIAN G .. M.::::...,./ "'- LB? ./ "CZ I I I STB I I 1 ~ IB I IN~~g~~Ny A EARLY JURASSIC INFERRED MT OCEAN m EQUIVALENT OCEAN nr OPENED VICTORIA DONQUIAO IN JURASSIC LAND (MVL) CARNIAN TURes OPHIOLITE ? / I ,t", ~ "-.... MVL /" ~LB? ~ ISTB 17 B END JURASSIC " OCEAN llr OCEAN II[ OPHIOLITE --.......... MVL ./' ..... LB?~ ISTB I I ~ ~. C MID LOWER CRETACEOUS ~ OCEAN :nz: OPHIOLl6[ OVER TURBID TES '\ INOIA\ 'yY ............. '-- MVL ~ LB? ~~ ISTBI o APTIAN BURMA VOLCANIC OCEAN .m: ARC TIN 11/ GRANITE ? ..A. 1\ INDIA \ Y ~'MV~ .'S; LB? ~~)!J l"sT;j E ALBIAN DEBRIS FLOW SHAN OCEAN rr g~:: ~~I'6~ITE 1// SCARPS tJ (JOt:. ./'- :I INDIA -'" 'i..Y ~\.. LB? ~TBftI.Y7 MVL F CAMPANIAN ............... N. INDOBURMAN W MOGOK SHAN UP CRET- RANGES TROUGH 1// BEL T SCARPS EOC~ A ~ 1\ \ \ \ \ \ ".(NOIA \ \ \ \ \ \ ~ STB ~;I' STB / / MVL ~ LB? ,,"X: I STBl77 G OLIGOCENE Fig.8. Cartoon illustrating late Jurassic-Palaeogene evolution of western Burma. Triassic turbidites of the Zangbo zone were deformed and close to their present position relative to the Gandise arc before the late Albian(26). These rock relationships can best be explained by the following sequence of events. End-Jurassic collision of the Lhasa block with the Quantang block, and possibly with the Shan-Thai block in Burma, was preceded by opening of ocean III to the south, possibly beginning in the late Triassic (Fig .8A,B). Perhaps in response to foreland thrusting(37) this basin began to subduct northwards and northeastwards beneath the Lhasa block and Burma (Fig.8B), forming volcanic arc rocks in the Shan Scarps and 579 plutons in the Mogok Belt. South of ocean III the MVL block probably rifted from Gondwanaland, with opening of ocean IV. Audley-Charles(38) has suggested from stratigraphic and floral evidence that this took place in the mid to late Jurassic. Continued subduction of ocean III and detachment of an ophiolite slab (Fig.8C) led to ocean closure. Carnian turbidites were accreted to the overriding plate west of and beneath the ophiolite, and obducted together with the ophiolite onto the MVL continental block in Burma (Fig.8D) and the inferred equivalent pre-Carnian metamorphic and sedimentary rocks in the Zangbo zone. Loss of part of the MVL block by underthrusting was followed by renewed northward and northeastward subduction of ocean IV lithosphere (Fig.8E), with arc volcanism in the Gandise belt on the Lhasa block and in the equivalent Burma Volcanic Arc on the western continental margin of the Shan-Thai block in Burma. Volcanism in the Burma Arc had begun by the early Albian, indicating that collision of the MVL block with Burma took place in or before the Aptian. East-verging thrusts in the Shan Scarps involve the Kalaw Red Beds of post-Bathonian age. The thrusts are tentatively interpreted as Cretaceous back-thrusts related to the inferred early Cretaceous collision with the MVL block to the west (Fig.8E). The two-mica tin-bearing granites which intrude the Mergui Group in the western margin of the Shan Plateau (Fig.4) are of late Mesozoic to Eocene Age, too young to be related to inferred thrusting in the Mogok Belt discussed below. They are speculatively explained by partial melting of crust tectonically thickened in and west of the Shan Scarps thrust zone. D. Upper Cretaceous-Cenozoic Events in Western Burma and Southern Tibet In the Burma arc volcanism has evidently continued, at least intermittently, since the late Lower Cretaceous. Initial volcanism and pluton emplacement were accompanied or followed by an upper Albian marine transgression onto Carnian turbidites, and subsidence of the Western Trough forearc basin. Renwed uplift preceded deposition of Campanian olistostromes, speculatively explained as debris flows derived from extruded mud matrix melange diapirs (Fig. 8F) analogous to mud volcanoes. Thede bris flows and Upper Albian rocks were over lain by conglomerates and a marine sequence in the Maastrichtian. Subsequently a thick Cenozoic succession, predominatly clastc, accumulated in the Western Trough. The Indoburman Ranges werelast elvated above sea level in or after the late Eocene, subsequent erosion exposing the MVL schists beneath Canian flysch and an Upper Cretaceous to Eocene sedimentary succession. The uplift could speculatively be explained by under-thrusting of continental rise and shelf rocks of Greater India during its collision with Asia in the Eocene. 580 Late Cenozoic displacement of at least 400 km 3,19,39 on the Sagaing Fault (Fig.9) has resulted in tectonic juxtaposition of formerly widely separated rock units north of Lat 230 • South of Lat 230 , however, displacement on the Fault was more or less parallel to the regional strike. In Tibet, Upper Cretaceous arc magmatism was accompanied by deposition of the Xigatse flysch, and of late Upper Cretaceous olistostromes of similar age to those in the Indoburman Ranges. Collision with India in the Eocene resulted in north and south-verging folds in the Xigatse Group, and in north-verging back-thrusts also present locally in the equivalent Western Trough in Burma (Fig.2). Correlation of the Triassic flycsh of the Zangbo zone with the Carnian turbidites in Burma implies that collisionof Tibet with India took place south of the Triassic flysch and underlying older rocks, not north of the Triassic flycsh as often supposed(3l). E. Mid-Tertiary Events in the Mogok Belt The latest metamorphism of the Mogok Belt, bordering the Shan Plateau, is here tentatively assigned to the early to mid-Tertiary and related to the India-Asia collision. The relationship of the Mogok Belt to the Shan Plateau resembles broadly that of the Central Gneisses of the Higher Himalayas to the Tethyan Himalayas. Te Tethyan Himalayas were uplifted in the late Cenozoic, and a late Cenozoic age for the uplift of the Shan Plateau is more or less required by the geomorphology (A.R.Crawford, oral commun. 1981), suggesting similarities in the Cenozoic evolution of the Mogok Belt to that of the Himalayan Central Gneisses. It is generally accepted that, following Eocene collision of India wi th Tibet and northern Burma, the Main Central Thrust zone developed in the Indian foreland south of Tibet (Le Fort 1975). I infer that the thrust zone continues southeastwards as the Lohit Thrust through NEFA into northernmost Burma, crossing the Cretaceous and Eocene sutures and pre-Oligocene magmatic arc, and lying on or near the western margin of the Shan-Thai block, beneath what is now the Eastern Trough (Fig.9). Mid to late Tertiary movement on the thrust zone (Fig.8G) resulted in eastward subduction of the Lhasa block equivalent beneath the Shan plateau, and subsequent underthrusting of part of the Lower Cretaceous ophiolite of ocean III. As in the Himalayas, the thrusting was accompanied by metamorphism in the Mogok Belt(19), and resulted in complex tectonic relationships with the overlying mostly unmetamorphosed Shan Plateau succession. Uplift, and consequent molasse sedimentation to the west were less intense in Burma than in the Himalayas, perhaps because crust underthrust beneath the Mogok Belt was in part oceanic rather than entirely continental. 0 , , " " 0 0 0 0 , , '. " 0 0 0 0 0 v \\ 581 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 " , ': ~', .'SLOCK'.','::': .' ... .. ' :.: Fig.9. Possible correlation of suture zones and crustal blocks, Burma and Tibet. West of the Indoburman Ranges and eastward subduction of the Indian Ocean continued to the present. south floor, of the Indian block, part of ocean IV, has 582 4. CONCLUSIONS The Shan-Thai Block is equivalent to the Quantang block of Tibet. The equi valent in Burma of the Tibetan Lhasa block may underlie the Shan Plateau. The Mount Victoria Land block and Carnian turbidites collided with western Burma in the early Cretaceous; a similar event probably took place in Tibet south of the Zangbo ophiolite. Elevation of the Shan Plateau and metamorphism of the Mogok Belt were related to Eocene collision of northern Burma with India; the Mogok Belt, on the margin of the Shan-Thai block, is probably continuous with the Himalayan Central Gneisses on the Indian block. ACKNOWLEDGEMENTS I thank Cela1 §engor, Kevin Burke and Michael Aud1ey-Char1es for discussions on an early version of the manuscript. REFERENCES 1- Bunopas,S. and Vella,P. 1983. Tectonic and geologic evolution of Thailand. In Proceedings of the Workshop on Stratigraphic Correlation of Thailand and Malaysia, Vol.l, Technical papers, 307-322. 2- Mitchel1,A.H.G. 1976. Southeast Asian tin granites: magmatism and mineralization in subduction and collision-related settings. COOP News1., Bangkok, 3, 10-14. ---- 3- Mitche11,A.H.G. 1977. Tectonic settings for emplacement of Southeast Asia tin granites. Bull. Geol. Soc. Malaysia 9, 123-140. 4- Mitche1l,A.H.G. 1981. Phanerozoic plate boundaries in mainland SE Asia, the Himalayas and Tibet. J.Geo1.Soc.Lond. 138, 109-122. 5- Asnachinda,P. 1978. Tin mineralization in the Burmese-Malayan Peninsula - A plate tectonic model. In P.Nutalaya (ed.), Proceedings 3rd Regional Conference on Geology and Mineral Resources of Southeast Asia, 293-299, Bangkok. 6- Hutchison,C.S. 1978. Southeast Asian tin granitoids of contrasting tectonic setting. J.Phys.Earth 26, Supp1. s.211-232. 7- Beckinsa1e,R.D., Suensilpong,S., Nakapadungrat,S. and Walsh,J.N. 1979. Geochronology and geochemistry of granite magmatism in Thailand in relation to a plate tectonic model. J. Geo1. Soc. Lond. 136, 529-538. 8- Ridd,M.F. 1980. Possible Palaeozoic drift of SE Asia and Triassic collision with China. J. Geo1. Soc. Lond. 137, 635-640. 9- Sengor,A.M.C. 1984. The Cimmeride orogenic system and the tectonics of Eurasia. Geo1. Soc. Amer. Spec. Paper 195, XI + 82 p. 10- Burton,C.K. 1985. The tectonic framework of mainland Southeast Asia. In Conference on Appliations of Geology and the National Development. Chula1ongkorn University, Bangkok, 1984, 255-268. 11- Searle,D.H. and Ba Than Haq 1964. The Mogok Belt of Burma and its relationship to the Himalayan orogeny. In Proc. International Geological Congress, 24th, pt.ll, 132-161. 583 12-Tapponnier,P., Peltzer,G., Le Doin,A.Y., Armijo,R. and Cobbold,P. 1982. Propagating extrusion tectonics in Asia: New insights from simple experiments with plasticine. Geology 10, 611-616. 13- Amos,B.J. 1975. Stratigraphy of some of the Upper Palaeozoic and Mesozoic carbonate rocks of the Eastern Highlands, Burma. Newsl. Stratigr. 4, 49-70. 14- Garson,M.S., Amos,B.J. and Mitchell,A.H.G. 1976. The geology of the area around Neyaungga and Ye-ngan, Southern Shan States, Burma. Institute of Geological Sciences, London, Overseas Mem. 2, 70 p. 15- Bender,F. 1983. Geology of Burma. Borntraeger, Berlin. 293 p. 16- Stauffer,P.H. 1974. Malaysia and Southeast Asia in the pattern of continental drift. Bull. Geol. Soc. Malaysia 7, 89-138. 17- Kyaw Win et al. 1985. Geological report on southern Inle Lake area, Southern Shan State, Eastern Highlands Region. Union Burma J. Sci. Tech. (in press). 18- Maung Thein and Soe Win 1969. The metamorphic petrology, structure and mineral resources of the Shantaung-u-Thandawmywet Range, Kyaukse District. Union Burma J. Sci. Tech., 3, 487-514. 19- Mitchell,A.H.G., Tin Hlaing and Zaw Pe, 1978. The Burma orogen. Unpub. rept., Rangoon. 20- Clegg,E.L.G., 1937. Notes on the geology of the second defile of the Irrawaddy River. Rec. Geol. Surv. India 71, (4), 350-359. 21- Mitchell,A.H.G. 1985. Mesozoic and Cenozoic regional tectonics and metallogenesis in mainland SE Asia. In Proc. 5th GEOSEA Conf., Kuala Lumpur, 1984. Bull. Geol. Soc. MalaySia, (in press). 22- Wolfort,R., Myo Win, Saw Boiteau, Myo Wai, Peter UK Cung and Thit Lwin, 1984. Stratigraphy of the Western Shan Massif, Burma. Geol. Jahrb. 57, 92 p. 23- Barber,C.T. 1936. The Tertiary igneous rocks of the Pakokku District and the Salingyi Township of the Lower Chindwin District, Burma, with special reference to the determination of the feldspar by the Federov Method. Mem. Geol. Surv. India 68 (2), 121-292. 24- Myint Lwin Thein 1970. On the occurrence of Daonella facies from the Upper Chindwin area, western Burma. Burma J. Sci. Tech., 3, 277-282. 25- Brunnschweiler,R.O. 1966. On the geology of the Indoburman Ranges. J. Geol. Soc. Aust. 13, 137-194. 26- Mitchell,A.H.G. 1984a. Post-Permian events in the Zangbo 'suture' zone, Tibet. J. Geol. Soc. London, 141, 129-136. 27- Barr,S.M., Macdonald,A.S., Yaowanoiyothin,W. & Panjasawatewong,Y. 1985. Occurrence of blueschist in the Nan River mafic-ultramafic belt, northern Thailand. Geol. Soc. Malaysia Newsletter 11, (2), 47-50. 28- Ko Ko, 1985. Preliminary synthesis of the geology of Bangka Island, Indonesia. In Proc. GEOSEA Conf., Kuala Lumpur. Bull. Geol. Soc. Malaysia, i~ress). 29- Tan,B.K. and Khoo,T.T. 1983. Ultramafic rocks in Peninsular Malaysia and their tectonic implications. In Fourth Regional Conference on Geology, Mineral and Energy Resources of Southeast Asia, 1981, 259-264. Geol. Soc. Phillippines et al., Manila. THE PALAEO-TETHYAN REALM AND INDOSINIAN OROGENIC SYSTEM OF SOUTHEAST ASIA Charles S. Hutchison Department of Geology, University of Malaya 59100 Kuala Lumpur, Malaysia ABSTRACT. Southeast Asia is a composite of relatively stable blocks of Precambrian continental crust overlain in large part by Palaeozoic carbonate-dominated platforms. The blocks are sutured along highly deformed mobile belts characterized by deep water sediments and fragmented ophiolites. After suturing, these mobile belts continued to be the focus of further tectonism because of their inherent weakness relative to the more rigid blocks. A major suture occupies the Song Ma region of North Vietnam, welding the Indosinia and South China blocks. This suture is interpreted as representing a Palaeo-Tethyan oceanic strait which closed in the Tournaisian-Visean to form an East Asian Continent together with the North China Block. This East Asian Continent lay in equatorial latitudes in Permian times and developed Cathaysian Gigantopteris flora. The Song Ma Zone is superimposed by several major Triassic sedimentary troughs with important rhyolitic volcanic association. The West Borneo Basement is interpreted as a detached part of the East Asian Continent. Another major suture extends southwards from Dien Bien Phu through Uttaradit, and the Gulf of Thailand to Raub, Bentong and Malacca in Peninsular Malaysia. All terrains east of this suture have Cathaysian affinities, whereas those to the west are of Gondwana affinity, characterised by one or more of the following features: Permian Glossopteris flora, Carbo-Permian diamictites, and Early Permian cool water faunas. This suture closed in Norian-Rhaetic times, thereby eliminating the Palaeo-Tethys and replacing it by the Indosinian orogenic system. The orogeny resulted in the formation of the large Eurasian landmass, and most formations of Jurassic and Cretaceous age are of continental molasse facies. Southern Sumatra contains Cathaysian flora at Djambi, but northern Sumatra has strong affinities with the Gondwana part of the Malay Peninsula. An Indosinian suture, possibly the southwards continuation 585 A. M. C. fiengor (ed.), Tectonic Evolution of the Tethyan Region, 585-643. © 1989 by Kluwer Academic Publishers. 586 of the Bentong-Raub line, may separate the two, but it is not yet well defined. The Eocene collision of India with Eurasia resulted in the propogation of major wrench faults, locally associated with extensional basins. These Cainozoic tectonics have resulted in the reactivation of the older suture zones. The Song Ma-Red River zone is a presently active right-lateral earthquake belt, and Late Cainozoic alkaline basalts are associated with the other extensional-shear basins. I. INTRODUCTION Southeast Asia, extending from Burma, Yunnan and Indochina in the north, through Thailand and Malaysia into Indonesia in the south represents the ul timate eastern extension of the Palaeo-Tethys(l). The elimination of the Palaeo-Tethys resulted in the formation of the Indosinian orogenic system. The Neo-Tethys Ocean has not been totally eliminated here and partly lives on as the Indian Ocean. Only to the north and east of India has it been converted into suture zones in Tibet and Burma. The region has been tectonically complicated by major Late Cretaceous and Tertiary rotations, wrench fault movements, associated with extensional basins, the on-land extensions of which have not been fully documented. These Cainozoic fault motions have largely moulded the present geological framework and made more difficult the unravelling of the earlier geological history and plate motions (2,3). The Cainozoic events may be largely attributed to escape or extrusion tectonics (4,5) resulting from the collision of the Indian continental block with Eurasia, the timing of which still needs to be refined, but which appears to be Mid-Eocene. The Southeast Asian or Indosinian orogenic system(l) is the least well-known and the gaps in our knowledge are attributed to many reasons, predominantly the equatorial climate leading to deep weathering and lush tropical rainforest, complex political boundaries, and a general lack of access. Nevertheless, its geological framework has emerged, though important essential details are lacking, particularly in the structural aspects. I present in the following pages a broad synthesis which illustrates the great similarities to the western extensions of the Cimmeride orogenic system. The differences are unique to Southeast Asia, where the Palaeo-Tethyan ocean was widest and hence a greater separation between Gondwana-Land and Asia existed. Some general published(I,3,4,6,7) , completion(9). summaries and a have more already been compiled and definitive work is nearing 587 II. ANALYSIS OF THE PHANEROZOIC TECTONIC FRAMEWORK Many attempts have been made to analyse the tectonic evolution of Southeast Asia. The following are considered to be the most significant: Hutchison(lO), Stauffer(2), Acharyya(ll), Gatinsky et al. (12), ;>engor (l3), Ridd (14), Bunopas (15), Gatinsky (16), Mitche11( 17) , Stauffer(3), Gatinsky et al. (6), Metcalfe(18), Stauffer(lS), and Gatinsky and Hutchison(8). Throughout this series of papers, there has developed a growing awareness that Southeast Asia is composed of a number of relatively rigid Precambrian continental blocks, marginally and extensively overlian by Phanerozoic platform strata, and separated by hi~hly deformed mobile belts (Fig. 1), composed of strongly compressed deep water strata deposited upon oceanic crust of the Palaeo-Tethys or its branches. Commonly they are closely associated with ophiolite and m~lange belts marking suture zones. Subduction activity within the mobile belts and along their margins with the continental blocks is documented by magmatic arcs, and their nature and ages are summarized in Figure 1. A plate boundary has long been recognized extending southwards from Yunnan as the Changning-Shuangjiang Suture(20), through Thailand and the Malay Peninsula. Hutchison (21) named this the Uttaradi t-Luang Probang Line in Thailand and Laos, and it extends northwards towards Dien Bien Phu, where it appears to be represented by a major right-lateral strike-slip fault. In Peninsular Malaysia he named it the Bentong-Raub suture zone (Fig. 1). From the tectono-stratigraphic entity distributions it may be concluded that this plate boundary was of major importance until the Late Triassic-Early Jurassic (220 to 200 Ma ago); ;>engor 1 , ;>engor and Hsu(7) and Gatinsky and Hutchison(8) deduced that it represents all that is left of the Palaeo-Tethys Ocean, which was closed by the major Indosinian Orogeny. The continental blocks are composed of two parts: 1) where the Precambrian basement outcrops extensively, such as the Kontum and Shan massifs (Fig. 1), or is covered by extensive platform strata, such as the Yangtze Platform, and 2) as the lateral extensions of the massifs and platforms in the form of Infracambrian to Phanerozoic folded Atlantic-type shelves. The figure also shows some of the oldest radiometric ages obtained, as an indication of the antiquity of the continental infra-structure of the continental blocks. These are of course minimal ages because the older rocks are frequently covered. The oldest rocks of the region are known from India and Indosinia (Kontum Massif). 1. GONDWANA AND CATHAYSIAN AFFINITIES The continental blocks have been numbered in Figure 1 as follows: 588 B PrttJommonHy vo/con" E;JI] PI~ 10 MI~ ~MZ~ mCZ ITO P'lC"mbflon Conl'nenlOI C,y,tol Mon,l, and PlottOff", EHllE f~!;~~~";.~;~~-;:.~~~~~O::~9~~·~lf'~~1 ~~;:~~"on Bo •• mln! [PZ 1 k===j Phon,ro,el< Mo~,I. 8elt, O •• ,loped on 0 589 Blocks of Gondwana affinity: 1 = Lhasa, 2 = Himalaya (not actually an independent block, but a region of crustal thickening resulting from underthrusting and belonging to the Indian subcontinent), 3 = Shillong Massif (may be continuous with the Indian Platform), 4 Indian Platform, 5 = Burma Plate, 9 = Sinoburmalaya (also known as Sibumasu, and Shan-Thai). Blocks of Cathaysian affinity: 6 = Yangtze Platform, 7 = Indosinia, 8 = Phu Hoat microcontinent (within the Trungson foldbelt), 11 = West Borneo Basement, and 12 = Eastmal (a continuous part of Indosinia - the continuity is beneath the downfaulted Tertiary basins of the South China Sea). The combined 7, 11 and 12 has been called 'Manabor' by Metcalfe( 18,22) . Blocks of uncertain affinity: 10 = Sumatra All terrains lying east of the Bentong-Raub and Uttaradit-Luang Prabang lines were predominantly landmasses during the Permian, on which flourished the Lower and Upper Permian Gigantopteris flora(23). Such a flora is thought to have evolved in equatorial latitudes, and McElhinny et al.(24) have confirmed this by palaeomagnetic data on the Permian Emeishan Basalt of the western Yangtze Platform. Cathaysian Gigantopteris flora have been confirmed in Indosinia(12), northern Thailand east of the Uttar ad it-Luang Prabang Suture, eastern Malaya(23), and in the Qantang-Tanggula Block of north Tibet north of the Pangongco-Nu Jiang suture. However this flora seems to have rather mixed characteristics (25). It seems appropriate to include the West Borneo Basement with- in the Cathaysion realm, but no diagnostic Permian flora has been found. However the Late Carnian-Early Norian Krusin flora(26) of the Sadong Formation has a strong similarity with the Norian Tonkin flora of north Vietnam. Also the upper Jurassic Kedadom Formation contains bivalves identical to the Upper Jurassic-Lowest Cretaceous Torinosu fauna of Kyushu, Shikoku and north-east Honshu, Japan(27). It is therefore warranted to assume that the West Borneo Basement was formerly attached to Indosinia-Cathaysia, but now displaced southwards along the faulted continental shelf of Vietnam. Sumatra presents an unresolved problem. Cathaysian-type flora at Djambi in south Sumatra(23) indicates a Cathaysian attachment or proximity, but Carboniferous pebbly mudstones and Triassic formations in north Sumatra indicate a strong similarity with the Phuket, Langkawi and Kedah areas of the western part of the Malay Peninsul~28). However the Djambi flora is Early Permian and not of Gigantopteris lineage(29). This problem may be resolved when more work has been done on central and south Sumatra, and this large island may turn out to be a composite of Cathaysian and Gondwana parts, as suggested by Hamilton(30), who postulated that the Bentong-Raub Line of Peninsular Mayasia may continue through central Sumatra. 590 With the exception of Sumatra, whose affinity is not yet resolved, all terrains lying west of the Bentong-Raub and Uttaradit-Luang Prabang lines are ascribed a Gondwana-Land affinity. They each should possess the following three signatures, but two out of the three are good evidence of a Gondawana-Land affinity: Permian Glossopteris flora, Carbo-Permian diamictites interpreted to be of marine glacial or fluvio-glacial origin, and cool water Carbo-Permian faunas. Even the Gondwanas of India showed a rapid warming in respect of the faunas during the post glaciation Permian, so that not all Permian strata show cool water affinities. The Indian post-glacial Gondwanas indicate that the climate at the start of the Permian was frigid, then gradually became temperate by the time of deposition of the Raniganj coal beds, based on the evidence from invertebrates, megafloras and palynomorphs(3l). These coal beds contain(40)speies of Glossopteris and Gangamopteris. The preferred model is that the terrains rifted off Australia during Carboniferous times, but their proximity to glaciated Gondwana-- Land would have ensured cool waters in which the characteristic fauna lived. The preferred former attachment is adjacent to the northern margins of Australia, contiguous with the Canning and Georgina basins. Burrett and Stait(32) have demonstrated the almost perfect similarity between the Cambro-Ordovician faunas of Peninsular Malaysia and Thailand and those of the aulacogen basins of northern Australia. These basins also have a similar record of cold water Carbo-Permian phenomena including diamictites(33). Many of the Gondwana terrains of SE Asia were composed of rifted platforms which were predominantly beneath sea level, so that they have not provided Glossopteris fossil localities. Therefore an association of diamictites(34,35) and cool water fauna 36 will suffice to prove a Carboniferous Gondwana-Land attachment. It appears from the region that Cathaysian flora is not associated with Carbo-Permian diamictites. Sumatra is the sale Southeast Asian terrain with this apparent association, and it therefore presents an enigma. However mixed Gondwana and Cathaysian flora also exist in Tibeti)7), Turkey and Saudi Arabia(38). 2. METHOD OF ANALYSIS Gatinsky and Hutchison( g) and Hutchison( 9) constructed a series of maps of Southeast Asia for different time intervals throughout the Phanerozoic, beginning with the Silurian, They show the present day geo- graphical distribution of rock formations and structural data which are cri tical for locating the various tectonic elements. The time intervals are somewhat arbitrary but have been chosen because they display important features and significant changes (Figs. 3, 8, 10, 13, 17, 20, 21). The common legend for these figures is given in Figure 2. 591 The formations, rock sequences or assemblages, which have been identified on the maps, are tabulated in Table 1. From their geographical distribution, the positions of the tectono-stratigraphic entities or tectonic elements, may be positioned. They are identified in Table 1 and shown on the sequence of figures 3, 8, 10, 13, 17, 20 and 21 as letter symbols. Letters A through D represent the transition from continent across the shelf to the ocean in an Atlantic-type continental margin. Basins of categories G and H complicate the picture but need to be recognized. Plate margins may then be drawn on the maps along appropriate symmetry axes. Covergent margins are indicated by subduction complexes and associated volcano-plutonic arcs, which may be isotopically dated. Dates are also shown on the maps, together with an indication whether they represent an igneous event, or a tectonic or thermal re-setting. The end product of subduction is usually a collision bel t (L) caused by the welding together of continental blocks along a suture zone. Such an event results in a linear/arcuate fold-belt and the production of potassic S-type granite batholiths. Associated with collisions are depressions such as foredeeps CF) and intermontaine troughs (M), which may receive great thicknesses of marine and continental sedimentary rocks. Table 1: Rock Assemblages and Their Tectonic Settings in SE Asia Tectono-stratigraphic Formation, sequence, or entities in which assemblage they ma~ occur A Be OF TG H5 JK LM PR cont inenta 1 redbeds -x xx -x continental grey deposits -x x- xx -x clastic continental and littoral x -- -x -x xx -x continental bauxite weathering xx nerHic marine < 50J: carbonate x- -x x- -x nerHic marine> 50% carbonate x x- -x reef 1 tmes tone - x- xx -x pe189ic 1 imestone xx x- siliciclastic and carbonate flyschoid - -x x- -x deep water turbidite flysch - -x xx -x flysch with chert x- deep water euxinic mud x- -x -x deep water mudstone and chert x- -x pyroclastic strata with acid tuffs -x -x calc-alkaline volcanics, mainly acid -x -x -x calc-alkaline volcanics, mainly xx Intermediate sub-alkaline volcanics. acid-inter .. x- mediate rap. tnO el1tl~ ~as~ I~ '. ala~ase x_ ~r~.f'l~:si i~~b!~~!~:~!i~l ~~:~!t basal t x- -x x- xx .hoshonite (potassic) b.salt xx -x bimodal volcanic association -x potaSSiC (S) granite (with L1 & F) -x x- subalkaline granitoid x- K-Na granite-granodiorite -x Na-gabbro-granodiori te-plagiograni te -x -x x- IOOn10n1 te-diort te xx x- K-N. alkaline intrusives "- syonitic intrusives x- differentiated mafic-ultramafic intru- xx sives gabbro-amphib~lito (Ophlolite) -x -x -- x- al pi no-type ul tramafites -x -x -- x- sedimentar, melange - -x -x x- -x tectonic m lange -x .... x- salt deposi:ts x -- -x -x linear fo lding x- -- x- 592 Table 1 continued x occurs in this tectonic setting A inner epicontinental basin B continental shelf of platform C continental slope and rise, with contourite deposits D open ocean, sometimes with trubidite fan F continental foredeep T oceanic trench G basins on oceanic or intermediate crust (marginal; fore-arc) H retro-arc and back-arc basins on continental crust S subduction complex or accretionary wedge J ensimatic volcano-plutonic island arc K ensialic volcano-plutonic cordilleran arc L collision orogenic zone M orogenic intermontaine trough P zone of pre-rift thermal reworking of continental crust R continental rifting Extensional tectonics of the continental crust may be recognized in two phases; first the thermal re-working, sometimes accompanied by uplift (P), identified by basaltic volcanism, subalkaline granites, and other features such as those listed in Table 1. The second stage is the development of a continental rift (R) in which continental and later marine strata may accumulate. This later stage is commonly accompanied by alkaline volcanism. A summary tectonic map, based on the sequence of tectono-strati- graphic maps, is given as figure 1, modified from Gatinsky et al.(6). 3. THE INFORMATION SOURCE The literature is too extensive to enumerate, but regional and country reviews have provided an important source of information( 9,39). This style of regional analysis has been used by Gatinsky( 40) and Gatinsky and Hutchison(8), who based their joint interpretations on first-hand knowledge of certain regions of Indonesia, Malaysia, Phillipines, Thailand, Laos, Vietnam, and southern China. Unfortunately many literature descriptions give no indication of the conditions of sedimentation, and many contain misleading inter- pretations. There is confusion over the useage of terms such as flysch and molasse. As understood today, flysch should contain an essential turbidite component but alodapic limestones do not qualify for they are readily sedimented off reef slopes in a continental shelf setting. Many Ii terature examples have turned out after further study to represent shallow water sandstone-shale sequences which include coal beds. It is therefore dangerous to over-interpret the term flysch from the older literature. Thus Mitchell(41) has interpreted the Triassic flysch of the 593 Indo-Burman Ranges as resulting from turbidite sedimentation along the continental slope of Asia, but S.K. Acharyya (pers. comm. 1984) does not recognize Triassic flysch in the Naga Hills. Helmcke(42) also attempts to discredit the interpretation of the Lampang Group (Fig. 18), long held to be flysch, thus: "It can be demonstrated that the Triassic strata of the Lampang area in northern Thailand contain neither pelagic sediments nor real flysch-sequences (Helmcke, in print)". This documentation is eagerly awaited. Molasse is another interpretative term open to abuse. Thus, Helmcke and Lidenberg(43) have used it for what I believe to be shelf-platform successions in Thailand, thereby adding to the confusion that these terms have caused in the literature. In addition to the older literature, and more recent publications on detailed areas, the following are the most general references upon which figures 1,3,8,10,13,17,20 and 21 have been based. They review the older literature: Regional stratigraphy and tectonics:(8,9,40,44,4S,46) Regional palaeontology:(47) Burma: (48,49) Thailand: (15,50) South China and Tibet:(25,26,5J) Malaysia:(50,52,53) I;donesia:(30,54,55) Philippines:(54,55,56) Indochina:(12,40,57,58) . III. TECTONO-STRATIGRAPHIC EVOLUTION No pre-Silurian tectono-stratigraphic map was attempted because of the fragmentary stratigraphic record. Late Proterozoic formations, overlain by ~ow water epicontinental sea deposits occur east of Mandalay in the Shan Highlands. Similar epicontinental sea deposits of Late Cambrian age also occur in Langkawi. The tectono-stratigraphic interpretation for Cambrian2 to Silurianl for the western half of the region is almost identical to figure 3, with the addition of a Cambro-Ordovician rhyolitic are extending from Mawchi (Fig. 4) in the north to Kuala Lumpur in the south. The oldest known rocks of the region form the Indian Shield and the Kontum Massif (Fig. 1), where a complex of high grade metamorphites, including charnockites, is overlain by low grade Proterozoic meta- sediments~2) (Fig. 1). The region of the Yangtze Platform is underlain by Proterozoic formations, which continue upwards as the Late Proterozoic-Cambrian Sinian System(5l). 594 1. SILURIAN2 to DEVONIAN2 (425-280 Ma) The Bentong-Raub and Uttaradit-Luang Prabang lines have been interpreted as the approximate position of the main Palaeo-Tethys ocean, with a Laos-North Vietnam strait of the Palaeo-Tethys occupying the Trungson foldbelt to the south of Hanoi, separating Cathaysia from Indosinia (Fig. 3). The regions to the west of the Paleo-Tethys are assumed to have been attached along their western margin to Gondwana-Land. A FORMATIONS AND GROUPS OF B FM'LTS FORMATIONS ~Mo,nlycont'nentolwllhpredO'1lna,ceof ~(o)GreY,IOJRedbedS /' • Mainly deep water blac~ eUXinlC clay ~~epwoterc\QysondrodIOIOIlles ~ru1fmevOlcaniClaSIICS,malnlywlth(lCld MCOlc-olkal,nevaICOnlCS,mOInIYOCId ~COIC-OlkabneVOICQnlcs.mOlnIYlnlelmedlale ~ ~~~~~Ol~~ r~~~~~I~SI' aCid cnd mter- ~TroP(ThoiellhCbOSOllanddlObase) ~Low-potosslcoceonIClhOleIIIICbaSOIl ~ ff~g~~~~~~lIfub-OlkOlme ohvlne bosolt r=22l Alkaline potassIc basalt (Shoshonite) ~BI-mOdOIVOlcQnICOSSOCIOlion rm ~~hO~~~,e~~~;'te (S-Iype) cnd LI-ond F- ~SubOlkollnegrOnllold ~K-Naqraf1lle-granodlorlle m ~~J~~~:-grOnOdlofite-ploglOgronite ~Moozonlte-dlorlte ~ K-Na alkaline intrUSives ~Potas5lcatkahnemtruslves /iVrenchtaults /Thrustfoult5 /Traf1stormfaults /Non-seporollngfOuItS C BOUNDARIES /OfformollOns Of stfllcturol-formatronal compfe~es ~Dlvergentploleboundafles(Spfeadlngo:(ls) ~rcon~ergentpfoteboundory(Trenches) if!< Collision suture D OTHER SYMBOLS ruSedlmentorymelange ~TectonlEmelange ~Saltdeposltslf1sedlmentorysequence M: lateral tranSition between formations 70 Rodlometrlc age of rock Radiometric age of tectono-magmatic reworklflg of crusl /' PaloeomognetlC vector with poloeo-Iohtude • PalaeozOIC-MesozOIC foufIQ of Europeon- Kozokstan pokleo- Tethys G PalaeOZOic-MesozOic funo of South European ond Mlddle-Eost Tethys @ Mixed fauna (Tethys With Arclrc, PaCifiC and East European) ~ Flora-Iale PalaeozOiC CalhaYSlOfl t' Floro-eody MesolOlC Hanggol ond related ~ Dlfferentloted mafic-ultramafic IntrUSives ~ Linear foldmg IIIID Ophiolite association (Gabbro-amphibolite) ./ Supposed dlrechon of lithosphere motion • AtP"lE-type ultromafltes (U mantle OI'IQln) 9 rGI,llatlOnal-structurol complex Fig. 2. Common legend for figures 3, 8, 10, 13, 17, 20 and 21. In these figures, subcripts subdivide the periods, for example Triassic3 means Late Triassic and Jurassic l means Early Jurassic. 595 L \) 5:: "- u 596 predominantly of nepheline syenite, which intrudes folded Lower Devonian and older strata(57). The terrain to the west of the Palaeo-Tethys has been named Sinoburmalaya 8 and Sibumasu(22). The eastern margin between the Palaeo-Tethys and Sinoburmalaya must have been active sometime in the Lower Paleozoic, for a Cambrian to Ordovician rhyolitic volcanic arc may be traced sporadically from Kuala Lumpur (Dinding Schist), through the Kinta Valley (Sungei Siput Rhyolite Tuff), northwards to the Grik area (Lawin Rhyolite Tuff), to correlate with tuffs in the Mergui Series and the Cambro-ordovician Bawdwin Volcanic Series (Fig. 4) of the Shan States of Burma(IO,45). A summary of the Lower Paleozoic stratigraphy is given in Figures 4 and 5(45) and lithofacies maps of the Lower Paleozoic (Ordovician-Silurian) for mainland Thailand and the Malay Peninsula in Figure 6, after Bunopas(15) and Jones(59). The Bentong-Raub zone contains important Devonian graptolitic shales associated with bedded cherts. (I) N O R TH SH AN ST AT ES ~- r- ;;; , . ~ ~ ~~ ::at: : ~ . . 1 ~ ;~ ~ I S ~ ! ~ . " . . . Q;: "" \ ' \ S .W . Y U N N A N \ ' \ (2 ) ~ ~ t 1~: ~ _ i [_ 6~~ ~Q~ ngb O. . . . . . ~ [ - . lu ~ p'U P" OO 5 $11 It! - li en S U it S 0- 9f - , { P ao - $h on Iu ,S U IU ~ ~ a . ::;; " ~ ~ ~< :> ::: ... . " ~ ~ ~ .. . .. " ~ . " :'! ~ -... ~ ~ Bo se no t I, " b, , " c o lc lS Ko~ -III II U j 5"- ' 0 0 ', -' 0 0 . ""I ~_' $ll Ih _ ~u S ch is t Ch'on 'il~ Sh on rJ .J (J /1 IjI nt lS S Bo st no l s u n G \~~ M AG YI ~~!~ 9r;: ;~u-; .n'l~ h$':: ~::~I : '11 t~bSb trtf, oll:I ~~ 1 B 05 t flO ! ' I d.1 d., 350 m Sa nd st on , lu rb ld il it s (4 ) N O R TH T H A IL A N D M A W C H I (5 ) S R I SA W AT (6 ) M E R G U I (7 ) . ; :r C on l,n uo ul U pw or d' ? /~ ! fi' -- - ;; / . /"0 ..... . . . . . . . . . . c o n fo rm a b le /' / , 0 • • " (V er y 10 . ~ //'.. ..... ..... ..... . , , / '10 00 ;0 9, u" .n o. n) ~ , . , " , , , , / . . . . . . . . IS - • • 'il' . lu tO ll' lo rp hl c) ~ /'iJ t~'I J,*l ul5 . . . . . . . . . . . . , . 1i: 45 0" , :: _ _ _ _ _ /~ /1 '... ... . ." d. . . . / 1 .. u " .. '~T. 1.}_ -- ~ ~ Is "' .. . . . ' . " [ : ' 15 /\ ,\ I . . . . . fI Io ", h, 6 _ } 12 0 . . . . 110 1 kn ow n) ' q _ . / r, ' S el ltS e - r 1 III ii:: .~. 50 01 '11 _ _ _ _ - - 6 ~ , . . . . . . . . j~ OO -1 50 1l l; .. .. ~ • - • -' J'} - _ ~ - S Y-~1 Q Tn llf IC " $o n4 = - (J 0 0 - - - - ~ ~, $" ,,, l 00 A99 10 lf lt rO It ::t: : - r ~ 45 0 0 0 ~ : ' .. ~.~... ...... .... v~~ ; h il ls ~ ' • • ' • 50 0 III. ...... ... Al lie u n tft D .n ~ . . . . . • • • • • . . . . . . . . . . . . . 8 0 u 1 10 1 l" n CIt : . . . . . . . . . . . . . Bo se 110 1 u e n ' - - _ _ _ _ _ ': :. . . . . . . . . . . . . . LA TE ~ , . . . ] So u no l s u n , ~ . . . . . . . E A liL r ~ faoom ' " ~ J 1 E .~ i: o ~ " . ; { E ~~ ,,~ I~ Vi J S ou n ol S "f I :o om ' - - - - - - - - lo ca lil y M ap 1- '. i . 2, .. .. .. 3 . \ ... ' . \ '/ . . ' - t BU RM A i" J" e\ ", 5 ~ 4 . . . . . . . . . . . ~A qq IO me rO 'e § S C h IS !o r ph Jl hl , ~G ne rs s ~C on lJ IO me rO le D S o n d 5 !O n e D S ll h ,o n e ~S ho lt Or m\ ld SI On ' §C h'f l < - < B Sh ol e - so n ds lo ne u qw en u IT.. !.. ;- 1 V olC CI HC fo rm ol lo n ~L "" es 'O nt I :J:;r :;r.j 00 10 '1 11 1, ~C on lo r" ,o bl tc On lo ct ~u nc on fo rr nl ly ~ Fa el U c ho n" , F ig .4 : P re -C ar bo ni fe ro us li th o st ra ti g ra p h ic c o r r e la ti o n c h ar t fo r th e n o rt he rn p ar t o f Si no b ur - m a la ya (4 5) . F or c o m pl et e le ge nd s e e a ls o f ig u re 5 . U l 'C ! - . I (8) P H U KE T - ~ - - Im ..f O bb IY . . . . . 6 . . :.. ! m U (lt on t S OU TH TH AI LA N D (9) ~ - - ' , i B os t n ot s e tn " (M ay In cl ud e , :oa t la te D h a n lo n a t W af ' q Th un k Kh a NQ ok 1 ~ ~ .... I - ~ . . . ~ ~ .", ~ ' - . . . E - - - r - - - ., LA T E ~: g SI LI JR IA N ' " ' " e I .. .! ~L LL _ _ _ "' " ~ 41 (L AT E) ~L o R D oL el A N I (E A ty ) _ 1.:: - - - - - - - - - - " ' " Lo ca lit y M ap V er tic al Ic ol . LA N G KA W I AN D N W M A LA Y A (1 0) KE D AH (I I) KE D AH PE R AK ( 1 2) KI N TA V A LL E Y (1 3) St Ss St ag e Sa nd st on e Up ~e r Lo w er Se rie s Fm Fo rm at io n M : M em be r I I I I I I - - - - - - - - . . . . . . , , r - , m es to ne an d " " U pp e. S, t, l L i " A B S E N T , M em be .s J" '" A B SE N T " I I I I I / - - - , E, 41 '" [& ~ E , / 6 Z _ _ 0 1 0 ~ r - ~ ~~--'-_ co .. .. .. . 6 . ' " 0 :' ~ b - or ~~ . . . . . . f) b4 fzi ; ~ ~I", · 1 : f'; ;1 ; g2 7 m c~4I • ~ :~ ~i J-. !~ m yo e n {.d l 15 0m ' 00 1. 1 Ki m L oo ng No . t b .d . 60 0m Do l C~ m~ I! !. _ Pin k ~ E ~5m II 4I. ~ - " ' ) 5 ~ tu ~o c. ' " Bo se n o t se e n I Bo se n o t se e n ,s ~ c 0+ 1 G nk co S ll ts to ne i _ _ . [ - - - tu 1 599 B w '" D 3 >- DC '" Z 600 MID CARBONIFEROUS TO PERMIAN ( 330 - 26S Mo AGO 1 EAST ASI:''' CONTINENT (INOOSINIA - CATHAVSIA BLOCK) PALAEO-TETHYS VIENTIANE I SOUTHEAST S~~~LS:t~ES TH:'~:~~ SEA TAK OCEAN I TRUNGSON CHINA Slt/OBuRMALAYA 'MASSIF I PAK1 LAY COLLISION HAiOI I ~ l:~~~,,-c, 1- - o !oOO '000 ,~oo ZOOOKM 0 ~ MID DEVONIAN TO MID CARBONIFEROUS (380 -no Mo AGO) PALAEO - TETHYS OCEAN GLACIATED SINO!lURMALAYA I f'HUKfT LAOS -VIETNAM PALAEO -TETHYS aRANCH I INOOSIN!A TC~~~S[OXN CATHAYSIA " o " '" GONOWA"A- LAND (O:-GONOWANA- LAN01 I K[OAH BANGKOK I~ I~ 1 _ BLOCK I PHU HOA T BLOCIC KHORAT I I MICRO rLATE HA~I I _ ,~ II I I I Iii _ :. LATE SILURIAN TO MID DEVONIAN (42' -310 Mo AGO) CATHAYSIA INDOSINIA BLOCK BLOCK i PALAEO - PACIFIC OCEAN PALAEO-TETHYS I LAOS-VIETNAIII EARLY OCEAN PALAEO-TETHYS BRANCH I PALAEOZOIC I GONOWANA- LAND I KONTUM I PHU HOAT HANOI COLLISION BAN~ I ~RAT I MICROtL~ I....... 1 ~ r Btl AmP @ll~{~o ~oo~.. 0 Jo rxlo"",~b J)(l-;&; .Jou ~-:-~ AST~[IIOSP~E~E ITJ cO,"",,,,, L" l A~ISOfVOL(A~O-PLUIO~IC ~ :;CCENOINGIoIAIiTLEOIAPIR Fig. 7. Schematic plate-tectonic cross sections from Phuket on Sinobur- malaya, across Indosinia, into southeastern China, for intervals from Silurian to the Permian. Carboniferous to Early Permian pebbly mudstones, or diamictite, occur extensively in three belts - along the coastal region northwards- from Phuket (The Phuket Group) towards Mandalay. They are known as the Singa Formation in Langkawi, and the Bohorok Formation in northern Sumatra. They are interpreted by Stauffer and Mantajit(34) and Stauffer and Lee(3s) as marine tilloids. The oldest formation in Sumatra is the Bohorok Formation (Fig. 9), of pebbly mudstones(28). Its occurrence as a linear belt along northern axial Sumatra links that island with Sinoburmalaya. The interpretative cross section (Fig. 7) shows Sinoburmalaya beginning to rift away from northern Australia, which lay (15) to 20 S(60), and graben structures filled with marine glacigene sediments. Far away in equatorial latitudes, Indosinia rapidly approached Cathaysia as the oceanic lithosphere of the Laos-Vietnam Palaeo-Tethys branch was consumed by subduction. 0 z 0 0 0 ::;; l> z 0 l> I r l> z 0 0 o o kJ 400Km _0 ' ~KALJ'" DEVONIAN 2- CARBONIFEROUS 1 (TOURNAISIAN) 380-360 Mo. 601 Fig.8: Tectono-Stratigraphic map for the time frame Devonian2 to Carboniferousl (tournaisian) (380 to 360 Ma ago) showing the units in their present geographic positions. See figure 2 for legend. Tectonic settings A through R are explained in Table 1(8). The diamictites shown in Sumatra and along the coastal belt towards Mandalay may range in age up to Early Permian. 602 4 0 N 0 0 Bando Aceh JURASSIC - CRE TACEOUS WOYLA GROUP 1000 PERMIAN - TRIASSIC PEUSANGAN GROUP = Undifferentiated v v v v Volcanic orc OphIOlite ond cover sequence CARBO - PERMIAN TAPANULI GROUP \JAS KM 100 ! 0Asahan A- I 200 ! Kille! ' 0 0 Bohorok ~o Alas 0 0 Formation G L - Geumpong Line KL - Kia LIne T L - Takenqon L me 4 0 N 00 Fig. 9. Simplified map of northern Sumatra showing the distribution of the pre-Tertiary formations(28). 3. CARBONIFEROUSl to PERMIANl (340-265 Ma). Indosinia and Cathaysia became sutured along the Song Ma line in Visean times through extinction of the Early Palaeozoic Laos-Vietnam branch of the Palaeo-Tethys to form the Trungson Foldbelt in which 330 Ma old granites were emplaced (Fig. 10). The embedding of the Phu Boat micro- continent in the fold belt is shown diagrammatically in Figure 7. The existence of major Triassic basins parallel to the Song Ma suture (Fig. 26) zone raises some doubt about the timing of the suturing of Indosinia onto Cathaysia, and this problem will be further discussed later. Cathaysian flora flourished widely throughout south China, Indosinia, Eastern Malaya (Eastmal) and south Sumatra. These terrains, together with the West Borneo Basement, are now called collectively the East Asian Continent. The Kuantan-Sungei LelJ'bing region of Eastmal is pictured as the western margin of the East Asian Continent, and westwards from Kuantan the facies became deeper water towards Sri Jaya. The terrain from Raub to Jengka is pictured as an island arc terrain which closely approached the East Asian Continent as a result of active subduction. This island arc terrain of Central Malaya is characterized by abundant volcanic o 400 km 1..1_ ..... _..1' ~ ~ "-C; ~ KOTA '\ K INABALU CJ 8 604 activity, the predominanriY acidic character of which suggests that the island arc must have had a continental basement (Fig. 16). Figure 11 shows a cross section of the area of Southern Sumatra where the Cathaysian flora was discovered( 61). The Djambi flora was collected from a horizon of the Karing Beds lying above a limestone which contains Sakmarian marine fusulinids. Accordingly the plant beds may indicate an Artinskian (Early Permian) age(23). The Gothanopteris bosschana and Palaeogoniopteris originated in Lonchopteris and Callipteridium respectively. These plants are of "different lineage than Gigantopteridaceae of the Gigantopteris flora. It is most likely that the Djambi flora belongs to the Cathaysian flora, but not neessarily to the Gigantopteris flora, because Djambi flora is older"(29). A B MOII'T" IUIUUA "[Sf , IWllA VAll[Y UDAH/'ULII L .... UW. I THAILAIIO 'UIUUI .O.TH Fig. 11. A: The Djambi nappe of southern Sumatra 61 . B: Correlation of the Carboniferous formations of Sumatra with adjacent parts of Sinobur- malaya The whole Djambi overthrust mass is of 2 km or more of predominantly Permian strata and volcanic rocks, and Van Bemmelen 61 suggested from similar lithologies that the allochthon had its origins in Peninsular Malaysia east of the Main Range, i. e. in the Eastmal 605 Cathaysian province. We must therefore entertain the prospect that the Djambi province is a displaced part of Eastmal, whereas the northern part of Sumatra is an integral part of Sinoburmalaya, following the suggestion of Hamilton(30), although a suture between the two terrains has not been documented in Central Sumatra. ,v "' JURASSIC - CRETACEOUS (140 -801.10) ~P/llAEO- ~ ~~~It< Ct 606 The Carbo-Permian was characterized by important island arc and cordilleran volcanism within and along the margins of the Palaeo-Tethys, especially in northwest Thailand 63 . This subduction-related activity led to the narrowing of the Palaeo-Tethys ocean, and the drift of Sinoburmalaya away from Gondwana-Land (Fig. 12). In contrast to the East Asian Continent, the waters bordering Sinoburmalaya supported a cool-water fauna (36), but conditions rapidly warmed up through Permian times. 4. PERMIAN2 to TRIASSIC3 (255-220 Ma) Continuing narrowing of the Palaeo-Tethys resulted in important cordilleran plutonic arcs on either side of the Ocean in the north and on the eastern side of the Malay Peninsula, giving granites of 255 Ma age (Fig. 13). Another granitoid belt fringed the Pacific along the east coast of Vietnam and China. Important rift structures developed across the suture zone from Indosinia and extended well into south China, named as various depressions on Figure 26. Their stratigraphies are given in Figure 27. Most of Sinoburmalaya was covered by an extensive shallow water platform characteristically free of volcanism and tuffs. A deeper water basin occupied the forearc region between the Bentong-Raub tectonic melange and the Eastmal cordilleran volcano-plutonic arc. Sediments within this basin are all tuffaceous, in strong contrast to those of Sinoburmalaya west of the Main Range. The Raub-Jengka Pass-Gunung Blumut terrain is shown colliding with the East Asian Continent resulting in a collision zone in which an important 220 to 240 Ma granite belt was formed extending northwards through Singapore and Gunung Blumut (Fig. 12). This orogenic collision enlarged the East Asian Continent on its western side. On the eastern side, subduction from the Palaeo-Pacific Ocean gave an active magmatic arc along the Vietnam coast (Fig. 13). The major Palaeo-Tethys was narrowing by subduction beneath the East Asian continental margin. Lithostratigraphic correlation charts for the Upper Palaeozoic to Cainozoic for Sinoburmalaya are given in figures 14 and 15. The dominance of platform carbonates and the total absence of volcanic rocks and tuffs are emphasized. This is in strong contrast to the Cathaysian province of Indosinia-Eastmal to the east. ....... ,~ ... ~-~ [) 0 a o PERMIAN 2 - TRIASSIC 3 (NORIAN) (255-220 Ma) o ! ~. ~/ / if , f 1/ II - T E R T IA R Y L A T E JU R A S S IC " '0 L A T E T R IA S S IC LA TE N O R TH S H AN ST AT ES (t ) W E ST Y U N N AN (2 ) SO U TH SH AN S TA TE (3 ) ~;: ~~: .. - ~U- m\~ ~~~ ~~~ 'r .. .. ·nl~ E A R LY L A T E C AR BO N IF ER O U S EA R LY ? \ - - - - - - ~ . \ D EV O N IA N \ \ \ \ V er tI ca l sc a le \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ 1'000 ... . " BO O . 0 0 4 0 0 20 0 o \ LI M ES TO NE \ \ \ iii o r ~ V NG O (O )'·. -(- I:"' H. U , b _ _ . . . SA LT IJO O- 50 0" '8 \ ' . AO , ~~ ~U ~ST N O R TH T H A IL A N D (4 ) M A E S O T W . T H A IL A N D ( 5 ) TE N A S S E R IM M ER G U I (6 1 lII Q TE I(A OA N TA U N G G R O U P m a y b . , 01 1\ " W ES T C E N TR A L T H A IL A N D (7 ) ~ I a)l neo rr. e!l t'a l.g 'op l'l ~pO $ll lon ,o. b) An A LL O C H TH O N o f P IN N A C LE LI M E ST O N E T E R T IA R Y LA TE A M H E R S T " . " GO ' I JURAS S 'C ': i> j1V RE oS AN DS TO NE 6 11 Q 3 (( '" .~ - SE RI ES r EA RL Y - - - - - _ SO "d ; s~ al e _ _ _ _ r 6 20 0m T R IA S S IC :~MM EAS~ ~~~E ""'' ''': OI . . ;1 ~ Q -2~ O-~ v~m - - - - " '. . . . . r 0 6 P E R M IA N Q.. ..... ..... - - - - - - - - .-~. :.l' " / / Fly $C~ C A R B O N IF E R O U S .:;::: MA =~~ ~A: '/ (5 . _ P.: ~~· :~~ s~o ~. / Co nf or m ab le - - - - - - - I D E V O N IA N I I I I un _ n o .n l Lo co llf y M er , , . . Y U N N A N . ' ) - 2 . BU R . . A ·... . . . . . . :: . . . . . . . . . - 3 . . r;:,~ ":'·. j t .. _ •. (· .... 4 ; ?L A O S I ~ /, . . . . . . . . , / . - . . . . F ig .1 4: Po st -D ev on ia n li th os tr at ig ra ph ic c o rr e la ti on c ha rt fo r th e n o rt he rn pa rt o f Si no bu r- m a la ya (4 5) . Se e fi gu re 1 5 fo r le ge nd . 0- o 00 (1) PHUKET (2) LANGKAWI (3) KEDAH (4) KINTA VALLEY ,----------"\ mT~RL. ..,.~~ ",,;;;", .JIJIIASSIC \ \ \ JURASSIC \ PERIiIAII ~ " % ~ It ....... _. • Purpl, Gr., Rid 810ck Whit, Carbonat.ous UPPER DEVONIAN c/b Crou- beddtd t, Thickness unknown b" Burro.td [D= Continental R,db,d focllS L,t Llmestont ~Norile e ' Ootid by tOUi'sD \ 5010"9 s.ds ItI200",\ \ \ t.:Ul~C.,",~-,,,\ \ \ \ \ CANtY 1i60~) PER.,AN K,," Loon\! No 3 aIds - - - -(l00 In) LATE 1500.) CARBONIFEROUS ~o,~K.:n~ ~$_ ~~L:!- (Conformable) (Conformoble) ~ Schist or phyllite f -=-j Shalt -Iondllo". seQulne, ~ Gn,iss Ic~ cl C!\tft ~ Lunulo", or morblt I-;.~."-;I Volcon;c or "y,oclostic I :" : :.1 Sondston. ~ FoclU chonOI ~ Sholl or Mudston, 0 ConforMable tontoct 100 01 Co""lo""rol' 0 Unconformity o Siltstone D Granite 609 Fig.lS: Post-Devonian lithostratigraphic correlation chart for the southern part of Sinoburmalaya, modified after Hutchison(4S). The lithostratigraphic correlation chart f or the western part of Indosinia (Khorat) and Peninsular Malaysia east of the Bentong-Raub suture zone (Eastmal) is given in figure 16. In contrast to Sinobur- malaya to the west, this Cathaysian province contains important Permo- Triassic rhyolitic volcanic rocks, with subordinate andesites, and rocks older than Carbo-Permian are unknown at outcrop. Farther east in Indosinia proper, the Precambrian basement of Indosinian extensively outcorps in the Kontum Massif (Fig.l). 610 11) KHORAT EARLY -;;r; -:;;'A;S~ ''"' .... , .. '"1 ,,' ! J \ \ \ \ \ 12) PAHANG 13) JOHORE 14) SI~GAPORE \ \ " \ luff \ \ \ \ . \ MID - PERMIA It Fig. 16. Lithostratigraphic correlation chart for the western part of Indosinia (1) and its southwards extensons into Eastmal (2,3,4). Older formations do not outcrop in this region, in contrast to Sinoburmalaya to the west. Legend as in figure 15. 5. TRIASSIC3 to JURASSIC2 (220 to 150 Ma) This interval witnessed the major Indosinian Orogeny, when Sinoburmalaya collided with the East Asian Continent forming a linear foldbelt along the length of the Malay Peninsula extending into Thailand and Yunnan(1,7) (Fig. 17). A foredeep flysch-filled basin (F) (Semantan and Gemas formations), with important rhyolite volcanism, was developed adjacent to the main collision zone. The pre-Rhaetic collision of Sinoburmalaya with the East Asian Continent eliminated the Palaeo-Tethys ocean in continental Southeast Asia creating the major part of Eurasia. Almost immediately after this collisional orogeny, pre-rift structures formed in many localities away from the collision zone, and major strike-slip faulting cut obliquely across the collision foldbelt in a predominantly NNW-SSE direction and also possibly sub-parallel to the 611 suture zone. Impressive post-collision S-type tin granites have been dated predominantly at 220 to 200 Ma( 64). They have their anatectic origins in a 1400 to 1900 Ma old basement which does not outcrop. The post-collision release of pressure through faulting may have been the principal activation of anatexis of the basement. A diagrammatic cross-section is given in figure 12, which also suggests that the West Borneo Basement began rifting from Indosinia at this time. IJ o o o TRIASSIC 3 (NORIAN) - JURASSIC (220-150 Ma) B R L • BENTONG-RAUB UNE Kllometr .. 200 400 600 800 Fig. 17. Tectono-stratigraphic map for the time frame Triassic3 (Norian) to Jurassic3 (220 to 150 Ma ago) showing the units in their present geo- graphic posltions. See figure 2 for legend. Tectonic settings A through R are explained in Table 1. 612 The older sutures, such as the Song Ma-Song Da region of North Vietnam remained zones of weakness and became the foci of wrench faulting combined with basin extension resulting from the Indosinian Orogeny which had its major impact to the west. Localized folding of the Yangtze Paraplatform strata is also attributed to 'thin-skinned tectonics' resulting from the collision. A lithofacies scheme for Thailand during the Late Permian to Triassic times as pictured by Bunopas 15 is given in figure 18A. By contrast, the Early Mesozoic scheme for Peninsular Malaysia is much more asymmetrical (Fig. 18B). Whereas in Thailand there is a volcanic arc west of the suture, in Malaysia the rocks west of the suture are devoid of any signs of volcanism, with the exception of the Sempah Conglomerate and ignimbrite terrain, which I interpret as having been thrust westwards and then downfaulted by the Bukit Tinggi Fault. Both Thailand and Malaysia have Triassic volcanic rocks and tuffs of rhyolitic composition to the east of the suture. Fig. 18. A: Late Permian-Triassic lithofacies map of Thailand( 15). B: Distribution of Mesozoic rocks in Peninsular Malaysia . cc = shelf carbonate zone with shelf clastics, 1 = land, Bab = back arc basin, Va = Lampang volcanic arc, Val = Loei volcanic are, fab = forearc basin, et = emergent arc-trench gap, VI = Nam Pat, V2 = Ko Chang, V3 = Khao Bo Nang Ching volcanics, GWKI = Nam Pat, GWK2 = Pong Nam Ron greywacke flysch belts, Lv = Lorn Sak tuff (M. Triassic 7), LSV = Lamnarai-Saraburi volcanics (Permo-Triassic 7), NV = Nakhon Sawan volcanics, IR = rifted land area 7. 1 Bangkok, 2 = Nakhon Sawan, 3 = Tak, 4 = Chiangmai, 5 = Chiangrai, 6 Kanchanaburi, 7 = Loei, 8 = Phitsanulok, 9 = Lampang, 10 = Kulim, 11 Taiping, 12 = Raub, 13 = Bentong, 14 = Malacca, 15 = Tanjong Murau. 613 The differences between the closure of the Palaeo-Tethys in Thailand and Peninsular Malaysia are emphasized diagrammatically in Figure 19. In Thailand, the narrowing was apparently effected by both eastwards and westwards subduction, rather like the present day Molucca Sea system of Indonesia 30 . This gave the simultaneous volcanic arcs in the Sukhothai and Loeif foldbelts(lS,6S). The two volcanic arcs remained separated and the Nan-Uttaradit subduction complex is separated from both arcs by flysch-filled fore-arc basins - Lampang on dhe west and Nam Pat on the east(lS). DRESENT WEST LATE CARBONIFEROUS PRESENT EAST Non-ul!a'adIT .:lccre!,on~ry prism ToIC~lr Beds Turbidite + SF~~~EJ:I~I ~" .. ) Po.eo- Tethys AUST IN (Glaciated) marine dlomlctite • Plluket Rifting from Gondwana-Land II MIDDLE PERMIAN Sarabur, Rot',,, P"'::4taBp;+H@!i¥lA~":":"' •• #lnOCllve m LATE PERMIAN - EARLY TRIASSIC Ill: MIDDLE - LATE TRIASSIC (THAILAND) Non-Uttcradl! Uplifted subduction comp1u Lompongi Nom Pol VolcaniC Arc',.. ,1>BoSHl t BOSln t Loel Volcorllc Arc SINOBURMALAYAV ==- ~~ ~ --~ !!:.~I£.lent Collision l' Kg~~~S/,~~a~~ (extinction ~f ~ Wedges Poleo- Tethys) :l[ MIDDLE- LATE TRIASSIC (MALAYSIA) Lonchon~ OCH.! volcanism / Fig. 19. East-west schematic sections across northern Thailand showing the formation of diamictites and the elimination of the Palaeo-Tethys by the Indosinian collision orogeny. The bottom two sections attempt to explain the differences in collision geometry as seen in northern Thailand and peninsular Malaysia. However, farther south in Peninsular Malaysia, the collision was more compressive and not bi-lateral (Figure 19). The Kodiang limestone platform on the west of the suture remained characteristically devoid 614 of volcanism, while the sequence east of the Bentong-Raub Suture was characterized by rhyolitic volcanism and the Semantan Basin was filled by tuffaceous flysch. The collision must have allowed Sinoburmalaya to underthrust the Bentong-Raub-Main Range accretionary prism, so that the Main Range S-type granite (200 to 220 Ma) batholith intruded into the subduction complex as well as into the platform sequence on the margin of Sinoburmalaya. Liew and Pag~60 have shown that the granites formed by anatexis of a 1400 to 1900 Ma old basement which does not outcrop. By this model, most of the uplifted accretionary prism must be considered allochthonous, sitting now on underthrust Sinoburmalaya continental crust and its platform cover. The collision in Peninsular Malaysia was therefore more compressive th,m in north Thailand, where subduction-related high pressure metamorphism has been preserved 66 . In Peninsular Malaysia the intrusion of the Main Range Batholith into the suture zone would be expected to have overprinted any older subduction-related metamorphism. 6. JURASSIC2 - CRETACEOUS2 (150-80 Ma) The Eurasian Plate was now predominantly above sea level and only a few epicontinental basins such as the Khorat-Vientiane Basin persisted, becoming saline then continental upwards (Fig. 20). The southern margin was actively convergent, with a volcano-plutonic arc extendin3 'l1ong Sumatra through the Java Sea to the Meratus area of Borneo(67). An important plutonic arc extending northwards from Phuket towards the Shan Highlands may be interpreted to be collision-related as the Burma Plate approached the Eurasion continent (Fig. 12). During Early Yenshanian time (Jurassic), the western belt of East China was dominated by extensive basin subsidence, while the coastal region through Hongkong was occupied by rhyolitic volcanic activity and epizonal plutonism. In the Late Jurassic and Early Cretaceous, the western margins of the basins underwent strong compression from the west( 51). The Yenshanian volcano-plutonic activity may have resulted from a collision event( 7), but some of the Late Yenshanian granites appear to be rift-related. A plutonic arc occupying the coastal belt of SE Vietnam, east of Phnom Penh (Fig. 20), has metamorphosed Jurassic redbeds, formerly interpreted as the Palaeozoic Dalat Series. Subalkaline granitoids occur sporadically in north Vietnam, Peninsular Malaysia, and in West Borneo. They are interpreted as resulting from the pre-rift stages of continental crust reworking. Tholeiitic basaltic dykes, dated at 110 Ma near Kuantan(6~ , witness the early rifting stages in the formation of the offshore Penyu and Malaya Basins, which lie between the Malay Peninsula and the Mekong delta. /' JURASSIC Z -CRETACEOUS Z (150-80 Mal 615 Fig. ZO. Tectono-stratigraphic map for the time frame JurassicZ to CretaceousZ (150 to 80 Ma ago) showing the units in their present geographic positions. See figure Z for legend. Tectonic settings A through R are explained in Table 1. 7. CRETACEOUS2 to NEOGENE (80 to 20 Ma ago) The collision of the Bur~a Plate against Eurasia resulted in the Phuket-Mana&llay foldbelt, associated with S-type granites, dated at 70 to 78 Ma (Fig. lZ). At the time of collision, the Bur~3 Plate lay about 450 km farther south, because there was no Andaman Sea, therefore the actual suture zone partly lies under the Andaman Sea, and is partly now represented by the right-lateral strike-slip Sagaing-Namyin faults, broadly c:oinciding with the Mandalay ophiolite line of Hutchison(Zl). 616 Since its collision, the Burma Plate has been translated northwards along the Sagaing-Namyin right-lateral fault by the length of the present Andaman Sea. A foredeep was formed to the west of the collision zone along the Central Valley of Burma. Fig. 21. Tectono-stratigraphic map for the time frame Cretaceous2 to Neogene (80 to 20 Ma ago) showing the units in their present geographic positions. See figure 2 for legend. Tectonic settings A through Rare explained in Table 1. Opening of the South China Sea Basin caused the rifted microcontinent of Spratley-Reed to approach the West Borneo Basement. The bre3k-up of the shelf of China and Vietnam, and the spreading of the South China Sea is documented in figure 22. Activity along the Lupar Line in Sarawak is indicated by a plutonic arc with dates of 77 Ma. The arc-trench system through Sumatra and Java has been established (67) . 617 Major N.W. to S.E. shear faults have developed. Sometimes movement along them had a sl~ht extensional component giving rise to shear-extensional basins such as the Malaya and Gulf of Bacbo (Fig. 25). Some have been filled by major quartz dykes (Fig. 23). These basins have developed along former zones of weakness such as the Song Ma-Song Da zone of north Vietnam which is now seismically active . ..........-- 3UBOUCTIONIONE ~INACTIVESueoucTIONZONE X> 618 ~ TERTIARY CONTINENTAL BASINS rs-:--l TERTIARY ['" ~ SYENITE AND RELATED ROCKS rv-:-l TERTIARY ~ ALKALINE BASALTS AND REL.ATED ROCKS -cx:IIXD- QUARTZ DYKE -- MAoIOR 'AUL.T - - - _ (EXTRAPOLATED) _____ DOWN 'Aut1f:O SIDE, WHERE KNOWN F?:~1:\~;!M GRANITE OUTCROPS BENEATH STRAITS OF MAL Ace A _ LATE CRETACEOUS GRANITE 175 Mal o [AST JOHORE GRANITE (220-240 Mo) o IoIAIN RANGE GRANITE (200-220 Mal ~ CENTRAL. RANGE GR,ANITES (MINOR) 1200-220 Mo) ~ JURASSIC-CRETACEOUS CONTINENTAL. STRATA ~ FORMING CUESTAS hi MYLONITE Ii TEClOGENIC GLASS ELEVATION IN METRES 104· 00'[ Fig. 23. Outline map of the Straits of Malacca showing the occurrence of Late Cretaceous granite and Tertiary continental basins controlled by northwest-southeast faulting. The Main Range .nay continue southwards from Karimun and Kundur towards the tin-granite island of Bangka. Granite details from Hutchison(70), faults in the Straits from Kudrass and Schlueter(71). The final stages of the Khorat-Vientiane epicontinental basin resulted in redbed and evaporite sedimentation. 8. NEOGENE TO QUATERNARY (20 to 0 Ma ago) The pattern of geological activity is one in which an active convergent plate margin extends from the Indoburman Ranges along the Sunda Arc to Java, then farther east to join up with the Phillipine system (Fig. 24). To the east, South China Sea marginal basin lithosphere is subducting eastwards beneath the Phillipines along the Manila Trench, while that of the West Phillipine Basin subducts to the west along the Philippine Trench 30 . o· 10' INDIAN OCtA N 110' -r-T"'-r Convergent plale margin (thrust lou1l1 TIO(rr-"r Benioff lon contours (depth in km I Tlt\1,--..,.. barbs on down-dip side INDOCHINA 300 600 km ! ! ---~~----'--------~"~O'~---- CHINA SOIIth s.o 80sln Plilltppttle Sec 10' 10 619 Fig. 24. The active plate margins of S~utheast Asia, showing the positions of the trenches, active volcanoes, main transform faults, active marginal seas, and contoured Benioff zones. The seismically active areas in the northwest are related to under- thrusting of India beneath Tibet and Burma. Two extensional fault systems are seismically active; the most active being the north-south QUjiang Fault near Kunming (Fig. 1) in south China. Important seismic activity also occurs along the Red River and Song Ma system of north Vietnam indicating right lateral m~vement (Fig. 25). These two systems together indicate east-west extension coupled with north-south shortening. The 900 km long Red River Fault has been very active in the Pleistocene and Holocene. Rightslip faulting is indicated by drainage pattern offsets of up t~ 6 km along the fault line(72). Movement must also be taking place along the parallel TonIe Sap-Mekong Fault (Fig. 25), with similar shear movement along the axis of the Malaya Basin, resulting in east-west fold axes in the Miocene strata. Slight extension along these predominantly shear faults has 620 given graben-like basins notably in the Gulf of Bacbo (Fig. 26) and the Mekong and Malaya basings (Fig. 25). The fault planes ar,? commonly infilled by thick polyphase quartz dykes, forming prominent geomorpho- logical features, for example near Kuala Lumpur (Fig. 23). The young Age of some of the faults is shown in Thailand where they cut Quaternary alluvium, and are associated with fluorite mineralization. It appears that these major shear faults have split Southeast Asia into three independent sub-plates - the Southeast China sub plate m~ving southeastwards away from the QUjiang fault system, the Indochina sub-plate moving northwestwards away from the South China Sea, the Central Sunda Sub-plate moving southeastwards away from the Andaman Sea (Fig. 25). T~e motion of the Indoc'1i"1a Sub-pl3.te may ~elp account for the compressional seismicity in the Indoburman Ranges, while th,:'! m,)~ion of the Central S'mda Sub-plate helps account for the active spreading (leaky transform) in the Andaman Sea behind it. -- MAJOR INTRA AND INTER PlAT~ SHEARS -- CROSS-PLATE rEN$IONAL FAULTS ~ BACK-ARC SPREADING AXES ~ MAJOR THRUST FAULTS AND TRENCHES MAGNETIC LINEATIONS AND TRANSFORM FAULT ::::;> DIRECTION Of MOTION OF INDIAN-AUSTRALIAN PLATE Fig. 25. Map showing the major Cainozoic faults of Southeast Asia, after -wood( 73) . Further evidence of extensional tectonics is widely found in Indochina and Hainan, where alkaline basalts occupy wide tracts of country, and locally in northern Thailand and Peninsular Malaysia. Some 621 of these basalts intrude through Cainoz~ic strata which fill the TonIe Sap-Mekong Basin. These basalts are famous as a primarj source of sapphire and other gemstones, mined from their regolith. Wood(73) developed t:te hypotCt2sis that the basins of S::>utheast Asia resulted entirely from east-west compression. However it is hard to avoid the c~nclusion that post-Eocene activity on the major shear directions (Fig. 25) has resulted from extrusion or escape tectonics(4) following the collision of India with Eurasia(69). V. RE-ACTIVATION OF SUTURE ZONES Ancient suture zones, whic~ weld stable continental lithospheric blocks, continue to represent zones of crustal weakness. Therefore they are susceptible to re-activation by major strike-slip faulting and accompanying extensional and vertical tectonics if a collisional orogeny occurs somewhere in the region. The extrusion or escape tectonics will be concentrated along these weakness zones. The modelling of Tapponier et al. (4) needs to be modified to take into 'iccoun!: the pre-existing inhomogeneities of the crust before colli.3io"1 of a continental block such as Sinobur~alaya or India. Thus the Qinling suture, bet~een the n~r~h and south :hina Blocks, has been interpreted as a suture formed in Devonian ('Caledonian') times by closure of an ocean. However the suture zone is cut by many east-west strike-slip faults, most of which were active in Teriiary times and some are still active. The zone also contains an important Mesozoic (Indosinian) foldbelt, which ~engor(74) considers the main collisional belt. The Bacbo Foldbelt of North Vietnam is more complex and difficult to analyse satisfactority (Fig. 26). The Vietbac region contains a well documented record of a 'Caledonian' orogeny(57), representing the southwest extensions from the Nanling ranges of southeast China. Simply folded Early Devonian redbeds rest unconformably on strongly deformed and metamorphosed Ordovician-Silurian formations in the Caobaclang uplift close to the Song Chay. The 'Hercynian' Song Ma suture has been superimposed by important Triassic basins, and major Cenozoic strike-slip faulting, much of which is presently seismically active, has complicated the picture (Fig. 26). The Bacbo Foldbelt may be considered as a major Triassic Basin and Range Province of great tectonic complexity. In addition to the Song Ma Suture, which I interpreted as Namurian-Visean(2l), and the Song Da, which I took to be Latest Triassic. Tran et al.(57) now add the little known Tamky-Phuoeson suture which occupies the region between the Trungson Mountains and the Kontum Massif to the southwest (to the south of 'Trungson anticlinorium' on fig. 26). These three sutures have parallel N.W. - S.E. trends. 622 ~ ~:.~G:~"ctu'( 10'11\01'011$1 + O .. on,on ~ ~::!:I:C 'O'lrIOh~;) £"unl'tlo'frIIOIO' () Eo.,hquoku ,., Eo.ly TnGn,c it II" Pho· Pllillonn, Alkohn, 8o101t1 1 C"letc,aul ¢ Poluogtn, Fig. 26. Simplified tectonic map of the Bacbo Foldbelt of North Vietnam, based on Tran et al.(S7) and Gatinsky(46). Fontaine and Workman (12) recognized that there was a major collision in Namurian-Visean times resulting in a folded and uplifted Trungson anticlinorium and that marine sedimenattion continued in basins, which became the venue of strong Indosinian folding. Gatinsky et al. (6) and Gatinsky and Hutchison (8) deduced that the Palaeozoic events were the most important, and proposed the closure of the ocean in Early Carboniferous times, and this view has been followed in this paper. The Permian to Triassic basins which lie between these sutures (Fig. 27) have been interpreted as basins which have subsequently opened as a result of major wrench faulting. They are commonly characterized by rhyolitic volcanism. The deduced polarity of the Ordovician-Silurian subductions are in conflict. Gatinsky At al.(6) proposed a S.W. direction for the Song Ma region beneath the Trungson volcanic arc (Figs. 7, 26). However Tran et al.(S7) suggested N.E. directed subduction from the Tamky-Phuoeson suture to cause the Trungson volcano-plutonic arc. They also suggest a N. E. directed palaeo-subduction zone from the Song Ma dipping beneath the Fan Si Pan. There are not enough data to resolve these differences. TRUNGSON BELT . .. " " J, ~- ~~ ~.If TIilUNGSON RANGE SAM NUA DEPRESSION ~ 4 I . ~ TAYBAC BELT 0 seNGDI DEPRESSION Tull Trol/Ql'ot ""0111 624 assemblages. Polarities of the palaeo-subduction systems may resolved as more refined radiometric dating becomes present only K:Ar dating is available. in future be available; at The Bentong-Raub suture has been interpreted as representing only Late Triassic (Indosinian) closure of the Palaeo-Tethys. Earlier events may have been completely overshadowed by this major event. The Indosinian orogenic closure of the Palaeo-Tethys to suture Sinoburmalaya onto the East Asian continent may be held responsible for strike-slip and extensional re-activation of the Early Palaeozoic sutures of Qin-Ling and Bacbo. In the same way, the collision of India against Eurasia is held responsible for strike-slip re-activation of all the previous sutures. The effects on the Qin-ling and Bacbo are obvious and they are both seismically active, but some yet undetected effects must have been superimposed on the Nan-Uttaradit and the Bentong-Raub suture zones. However Helmcke(42) maintains a minority view that these sutures closed in the Palaeozoic. Re-activation of tectonic faults and following a subsequent is a valid alternative hypothesis of $engor 1 older suture zones as a result of extrusion extensional basins propagating through them, collisional orogeny outside the immediate region, to the "orogenic collage" (= terrain assemblage) VI. PALINSPASTIC RECONSTRUCTIONS It is possible to suggest palinspastic settings of the continental blocks of figure 1, based on the tectono-stratigraphic maps (Figs. 3, 8, 10, 13, 17, 20 and 21). However much more palaeomagnetic research is required to refine the scenario presented here. 1. EARLIEST PALAEOZOIC TIMES There is compelling evidence that all the terrains of Southeast Asia were an integral part of Gondwana-Land and most probably were attached to the northern Australian margin(6,7,30,76). Based on palaeomagnetic studies of the Ordovician, Cambrian and the Late Proterozoic of South China, Lin et al. (77) propose an Early Cambrian reconstruction in which the east side of the Yangtze Platform is positioned adjacent to the Ord and Daly River basins of N.W. Australia. This reconstruction juxtaposes the Cambrian marine basins of Central Australia and South China, and also suggests a satisfactory explanation for the strong affinity of Cambrian trilobites in these two presently distant terrains. The existence of major phosphorites in both regions can also be readily explained. Gypsum and salt pseudomorphs in the Middle and Late Cambrian strata of these two terrains is consistent 625 with their palaeomagnetically determined low lattitudes. This reconstruction places the Late Proterozoic tillite localities in Australia and the Yangtze Platform into a continuous low latitude belt. The pronounced similarity between the Sinian System of south China and the middle and upper Adelaidean System in Australia further supports the hypothesis that the Yangtze Platform was an integral part of the NW Australian part of Gondwana-Land until about Ordovician times(77). The evidence that Sinoburmalaya was also part of Gondwana-Land is also compelling(76). The Cambro-Ordovician faunas of Thailand and N.W. Mayasia have an extremely close similarity to those of Australia, notably to the Bonaparte Gulf, Canning, Amadeus and Georgina basins. These stratigraphical and palaeontological similarities render the palaeomagnetic data of Haile(78) unlikely to be significan~ His results from the weakly magnetic Setul Limestone of Langkawi show an unacceptable spread, and Bunopas(lS) found that the Ordovician Thung Song limestone of nearby Thailand has too weak magnetism and is too unstable to give useful results. Haile(78) proposed that his results indicate that the Langkawi area of Sinoburmalaya lay either 430 N or S of the Ordovician-Silurian equator, but Stauffer( 3) was prepared to admit that a 430 S position would support his earlier thesis(2) of a possible juxtaposition with the Iranian sector of Gondwana-land. Whatever the exact geographical attachment, it is a reasonable assumption that the whole of Southeast Asia was an integral part of Gondwana-Land during the Early Palaeozoic, and the most likely attachment was near northern Australia. Those blocks which were to develop Permian Cathaysian floral affinities must have rifted 3.way from Gondwana-Land during the Ordovician or Silurian. Those blocks that were to develop Permian Gondwana affinities remained attached or lay close to Gondwana-Land in Carbo-Permian times. Such terrains would also be characterized by Carbo-Permian tilloids, whereas the Cathaysian terrains, which lay in equatorial latitudes in Permian times, avoided any further glaciation after their Late Proterozoic glaciation. Audley-Charles .,79i maintains a minority view that all these terrains remained attached to Gondwana-Land till Jurassic times, and the different floral types evolved on coastal and inland zones of the super continent. 2. LATE DEVONIAN TO EARLY CARBONIFEROUS The deduced palaeogeography is shown in Figur e 29. From the palaeomagnetic reconstructions of Scotese et al.(60), the northern margins of Australia lay about 200 S. Gatinsky and Hutchison( 8) showed Sinoburmalaya rifting off Gondwana-Land in early Carboniferous time along the Phuket-Mergui belt. The distinctly cool water fauna described by l'Jaterhouse(36) from the early Permian Phuket Group is convincing evidence that Sinoburmalaya was juxtaposed to glaciated Gondwana-Land 626 and the diamictites of Phuket, Langkawi and Sumatra are likely to be marine ti110ids shed off the rifting margin of this major continent, in which Carbo-Permian ti110ids are well documented(33) . LEGEND 1/1' Linear folding II Precambrian continental / Convergent plate .;r Supposed direction of crust with ancient margin plate motion platforms Phanerozoic shelf on # [J Polaeozoic- Mesozoic Divergenl margin Eastern Malay EM .... " basement Peninsula / - , Ma in Iy acid volconic / Indochino Peninsula ,I I,/~ Transform margin IN -\\y arc # West Borneo ~ Sub-alkaline, Collision zone (Kalimantan) ~" II ...... __ \\ trachyte arc LY Longmuco - Yushu D / Atla~tic- type. , , , Andesite arc , , passive margin NT North Tibet D Tholeiitic basaltic trap / Non-separating fault PH Philippines l l [J / P5C5 Palaeo South Chino Sea , , , Alkaline olivine basalt Wrench fault , , I,: 'I Alfrine - !~pe / Q Kunlun - Qinling ul ramafl e Thrust fault Sumatra [J 627 20 10 o c 10 10 20 10 Fig. 29. Palinspastic map of Southeast Asia for the time period from Late Devonian to Early Triassic. For legend see figure 28(8). 628 Good palaeomagnetic data are needed from Southeast Asia for Carbo-Permian formations to constrain the palinspastic maps. The data of McElhinny et al. (80) for the Carboniferous Singa Formation (the Redbeds they included from the Raub area are likely to be Triassic) suggest a palaeolatitude of 150, which they interpreted to be north. This is likely to be correct for the Raub redbeds in Triassic times, but leads to an impossible situation for the Carboniferous Singa Formation. If their Singa data truly represent Carboniferous remanent magnetism (doubtful in view of the abundant Triassic granites in this part of Malaysia) then magnetic reversal must be presumed, gl vlng a palaeolatitude of 150 south. Such an interpretation is supported by Bunopas et al. (80) who found the Phuket Group rocks at Khaeng Krachan damsite gave a palaeolatitude of 240 south. Later Bunopas(ls) documented a palaeolatitude for Sinoburmalaya of between 10 and 200 south. His reconstruction is given in Figure 30. PALAEOMAGNETIC NORTH (Carboniferous) ~ ? ~ I l.&.J U / 0 ;;e ~ ::t:':r .... u l.&.JZ .... w I ct: 01- l.&.J ~ -.J ~ Cl. o \ PRESENT DAY NORTH o 1000KM. I o ~() o 10 S Fig. 30. Carboniferous reconstruction of Sinoburmalaya against Australia which is consistent with palaeomagnetic data on rocks from Thailand(ls). The position of Sumatra (S on figure 29) is not palaeomagnetically constrained. Gatinsky and Hutchison(8) placed it adjacent to Gondwana-Land in the Early Carboniferous to account for the Bohorok diamictite. It may then have rapidly drifted away towards Cathaysia so 629 as to develop its Early Permian Cathaysian-type flora. However an equivalent of the Bentong-Raub suture may subdivide Sumatra into two blocks(30). Support for a SinoburT§laya proxi~ity to Gondwana-Land comes from Rao(82) who obtained 0 and C values from Lower Permian brachiopods of the Chu ;ing Formation of northwr§t Malaysia , indicating depositional temperatures of 6 to 90 C. Some 0 values are similar to values obtained from subpolar Australian Penlian carbonates and indicate that the Chuping Limestone reacted with cold melt waters. The faunal assemblage of the basal Chuping Limestone is also similar to modern subpolar carbonates(22,82). This information suggests a close proximity of Sinoburmalaya to glaciated Gondwana-Land into Early Permian times. Furthermore, Archibold et al.(83) have shown that the late Lower Permian brachiopods of the Thailand Rathuri Group show close affinities to those of N.W. Australia. The position of Cathaysia and Indosinia are not palaeomagnetically constrained for this time frame. However they missed the Gondwana-Land glaciation. Middle Carboniferous brachiopods of Indosinia show no Australian affinities but have close affinities to China and North America(22) • 3. LATE PERMIAN TO EARLY TRIASSIC The Late Permian position of the East Asian Continent is well-constrained by palaeomagnetism on the Emeishan basalt of the Upper Permian Xuanwei Formation (lat. 29.60 , long. 103.40 ) by Wang et al.(84) and re-confirmed bJ McElhinny et al(24). The calculated palaeolatitudes are respectively 2 and 3.50 from different localities. The low latitudes are consistent with the widespread Cathaysian flora, presumed to have evolved in an equatorial climate(23). Early Permian fusulinid faunas of the Thailand part of Indosinia have Eurasian-Arctic affinities but in the Middle and Upper Permian they are of distinctive Tethyan type. There is therefore a major change in the fusulinids of Indosinia from Lower to Middle Permian(22). The Palaeo-Tethys is shown narrowing as a result of active subduction beneath the eastern margin of Sinoburamalaya. But the residual ocean was still a floral barrier. Glossopteris has been found in the Tibet blocks, but not ye.! in Sinoburmalaya, largely because it was covered by shelf seas during Permian times. Some palaeomagnetic constraints for Sinoburmalaya were provided by Bunopas(15). Permian strata of Peninsular Thailand gave a palaeolatitude of 100 and from Mae Sot gave 19 0 • 630 4. LATE TRIASSIC-EARLY JURASSIC The preferred palaeogeography for this time is given in Figure 31, after Gatinsky and Hutchison(8). The Sinoburmalaya Block collided with the East Asian Continent, resulting in the major Indosinian Orogeny and glvlng rise to the major Main Range 220 to 200 Ma old S-type collision-related granite belt(64). The amalgamation of all these block resulted in the greater Eurasian Plate. gengor(1) favours a Late Triassic assembly of a Cimmeride assemblage and its Jurassic amalgamation onto Eurasia. Gatinsky et al.(6) showed the West Burma Block as being originally a part of Sinoburmalaya which rifted away (Fig. 31A7). This is of course highly specualtive and it could alternatively have been an independent minor continental block. The figure shows the West Borneo Basement beginning to rift away from its attachment to the East Asian Continent. Palaeomagnetic constraints on its position are completely lacking. That the elements of the Eurasion Plate lay predominantly in higher northern latitudes is well supported by palaeomagnetic data. Results from the Triassic and Jurassic(lS,8l,8S,86) indicate that Sinoburmalaya was united with the East Asian Continent to form the greater Eurasian Plate in Late Triassic and therefore continued to act as a single unit into the Late Cretaceous(87). The western margins of the Khorat Basin near Churn Phae gave good palaeomagnetic results from the Upper Triassic (Norian) Hin Lat Formation at the base of the Khorat Group, indicating a palaeolatitude of 250 (86). The Pengerang rhyolite of S.E. Peninsular Malaysia, dated at 238 Ma gave a palaeolatitude of 220 (80). The Lower-Middle Triassic rocks of Songkhla, Peninsular Thailand, gave a palaeolatitude of 20°(15). Lower Jurassic rocks of the Khorat Basin gave consistent palaeolatitudes of 200 by several independent workers(lS,8S). An identical palaeolatitude was obtained for the Upper Jurassic Khorat sequence. The Malay Peninsular palaeolatitude of 17 0 for the Sempah conglomerate and associated rhyolite(80) is consistent with these Khorat readings and indicates that these formations, now known from rhyolite dating to be Late Triassic(64), indicate that the whole region occupied latitudes such as shown in Figure 31. Palaeolatitudes of 14 and 150 S for the Upper Triassic to Lower Jurassic of Central Sumatra(88,89), show that Sumatra was likely to have formed a southern extension of the Sinoburmalaya peninsula as shown in Figure 31. 631 40 30 10 o 500 1000 km ! ! ! 50 y s Fig. 31. Palinspastic map of Southeast Asia for the time period from Late Triassic, when the Eurasian Plate was formed through collision of Sinoburmalaya with the East Asian Continent, to the Early Cretaceous. For legend see Figure 28(8). 632 5. LATE JURASSIC-EARLY CRETACEOUS The Lhasa Block of south Tibet is shown colliding with the Eurasian plate. The collision may be dated as Middle Cretaceous(1,20). The West Borneo Basement Block is shown moving away from its attachment to Indochina. Important convergent activity along the Cathaysian margin resulted in the widespread Jurassic to Cretaceous Yenshanian volcano-plutonic terrain of S.E. China (Fig. 31). Haile and Khoo(90) reported palaeomagnetic results from the Upper Jurassic to Lower Cretaceous Tembeling Formation redbeds of Peninsular Malaysia. Widely separated samples from Maran and Kluang gave identical palaeolatitudes of lIoN, while the locality at Sungei Teka gave a slightly inconsistent 140 • The magnetic declinations varied from 3330 to 3560 • These results collectively indicate that Peninsular Malaysia lay about SO north of its present position. Since Late Cretaceous time the Peninsula has rotated anticlockwise and moved south. Middle to Upper Jurassic measurements b6 Bunopas(15) gave a palaeo- latitude of 21 to 250 from Surat Thani, 27 for the Khorat Basind and 150 for north Thailand. The magnetic declinations are around 30 , so this part of Thailand has rotated clockwise and moved southwards to its present position. 6. LATE CRETACEOUS The Late Cretaceous palaeogeography, constrained by the data of Achache et al. (S 7), is rather similar to that of figure 31B, but with all the blocks moved about 5 to 70 south, and in the same orientation. Achache et al. (S7) deduced a palaeolatitude of 140 30 for the Lhasa block, and a measured declination of 150 indicates that southern Tibet has rotated clockwise by this amount since Late Cretaceous time. The large collection of Mesozoic data on the Khorat Basin of the Indosinia Block by Bunopas(15) has been combined with data from Achache and Courtillot(S6) to show that the Indochina block gradually moved southwards from 220 N in the Early Jurassic to 150 N in the Late Cretaceous. These measurement sare identical to European data, indicating also a slow similar drift over the same period. Therefore Indochina was firmly attached to Eurasia throughout this time. However the Mesozoic data from the Khorat Basin suggest a large clockwise rotation with respect to Eurasia since the Middle Cretaceous6 the declination difference is 190 when computed for Jurassic and 29+16 when computed from Cretaceous poles(S7). Only a major tectonic event such as the collision of India with Eurasia could have produced such a large rotation, as suggested by Tapponier et al.(4), and illustrated in Figure 32. The opening of the South China Sea Basin is seen as a consequence of this relative motion between Indosinia and Eurasia, which 633 must have been accomodated along the Red River Fault Zone (Fl on Figure 32). A f t t PALAEOMAGNETIC NORTH BANDED PLASTICENE I II (t 17\ \u 90° 10 /' / /' c Fig. 32. A: Summary of the plasticene indentation exyperiments of Tapponnier et al.(4) to model the collision of India with Eurasia. B: Summary of the Cretaceous palaeomagnetic data from Hail and Briden(91), updated by data from Tibet and Indochina. C: Upper Cretaceous palaeogeographic reconstruction by Achache et al. (87) consistent with palaeogmanetic data. Dolerite dykes intruded into granite in tILe Kuantan region of Eastmal, dated as Albian (104+10 Ma), gave declination of 3330 with a high palaeolatitude of 230 (68). These results are inconsistent with the data from south Tibet and Indosinia, and would place Eastmal farther north than Hong Kong in Late Cretaceous times (Fig. 32). This is clearly unacceptable. These dyke data can be discarded on the following basis: the region lies closely adjacent to the major offshore extensional Penyu basin. The basin development would have involved rotational faulting and 634 20 10 o F;_g. 33. (Palinspastic map of Southeast Asia for the time period from Early Palaeogene to the present day. The major events are the Eocene collision of India with Eurasia and the opening of the south China Sea basin. For complete legend see figure 28(6). 635 crustal tilting. Haile et al.(68) could make no corrections for such a regional tilt. The regional tilt is proven by the Cretaceous Gagau Group of the type locality 130 km NNW of Kuantan, where the Lotong Sandstone commonly dips within the range 10 to 250 towards the NE or NNE(92). A similar problem applies to the measurements on the West Borneo Basement block by Haile et al.(93). They utalized a suite of basic dykes and granites, leading to the overall conclusion that the rocks are predominantly of late Cretaceous age and have a palaeolatitude of 00 • However the region has rotated anticlockwise by about 500 since the late Cretaceous. As in the Kuantan area, no correc tion could be made for regional tilt. Therefore tha calculated palaeolatitudes for Malaysia are less reliable than those obtained from redbeds in Tibet and Indochina. Nevertheless the Segamat Basalt of Johore, in Peninsular Malaysia, which gave a K:Ar age of 62 Ma(94), gave a declination of 3160 with a calculated palaeolatitude of 170 (80). This is high but not as unacceptible as the Kuantan dykes. The data indicate a 440 anticlockwise rota tion for the Peninsula since the Late Cretaceous, which in no way supports the model of Tapponier et al. (4) which requires a clockwise rotation (Fig.32). The circles of confidence for the palaeomagnetic pole positions for the Kuantan dykes, Segamat basalt, and the igneous rocks of West Borneo, only slightly overlap. Haile et al. (93) therefore concluded that "West Kalimantan and the Malay Peninsula have behaved as a unit since the Middle Cretaceous, have remained at about their present latitudes, but have rotated anticlockwise about 500 since then." Such a conclusion is hardly warranted by the data and shows that a palaeomagnetic challenge still remains in view of the apparent conflict with the indentation model of Tapponier et al. (4). Haile(95) showed that Miocene redbeds in West Borneo have remanent palaeomagnetism indistinguishable from the present day field, suggesting that the rota tions discussed above have been completed by Miocene times. This would also be in conflict with the indentation model. A major orogenic event in the Late Cretaceous was the collision of the Burma Plate (WB on Figure 31) with the Sinoburmalaya margin of Eurasia. This collision is presumed to have been the major cause for the Late Cretaceous tin granite belt which extends northwards from Phuket into Burma. The Burma Plate was about 500 km farther south at the time of collision. Since then it has moved northwards along the right lateral major Sagaing-Namyin Fault to its present position shown in Figure 33. As it moved northwards, the Andaman Sea opened behind it. 636 7. CAINOZOIC ERA In figure 33 the Burma Plate is shown moving northwards along the Sagaing-Namyin Fault to its present position adjacent to the Shan Highlands. The movement of the Burma Plate has not yet been tested by palaeomagnetic research. The Indian Platform moves northwards as oceanic lithosphere subducts ahead of it to form the Gandise Transhimalayan volcano-plutonic arc. Eventually India collided with Eurasia in Eocene times and according to the model of Tapponier et al. 4 the collision propogated major shear faults eastwards and southeastwards (Fig. 33) as the squeezed terrains attempt to escape from the collision zone. The West Borneo Basement has been driven south to equatorial latitudes by the opening of the Palaeo South China Sea and the subsequent opening of the South China Sea Basln. These events pushed many microcontinents southwards from the China shelf towards Borneo. Palaeomagnetic measurements on Lower Tertiary igneous rocks of north Sumatra west of the Semangko Fault(95) indicated palaeolatitudes of from I to 30 south of their gresent position and that the areas have been rotating clockwise up to 30 since the Early Tertiary. A Late Cainozoic clockwise rotation of about 200 of Sumatra relative to Java was earlier suggested by Ninkovic 96 based on the sharp change of curvature of the Sunda Trench at the Sunda Strait. Such a rotation requires the existence of an important shear or dislocation zone lying in the Sunda Strait (Fig. 24). Figure 33 shows that important transform motion began on the Red River, the TonIe Sap-Mekong and the faults of Peninsular Malaysia and the South China Sea well before the collision of India with Eurasia. Slight extension on these predominantly strike-slip faultsled to important graben-like Tertiary basins along the lines of the faults. Such basins are therefore appropriately termed shear basins by Kingston et al.(97). The Early Tertiary convergnece of India with Eurasia must have emphasized movements on the pre-existing faults. I per fer to modify the model of Tapponier et al.(4) to suggest that the escape tectonics made use of pre-existing zones of weakness such as old suture zones, and shear zones which were already in existence before the collision, as suggested by ~engor(l). The Neogene picture (Fig. 33) suggests that Southeast Asia may now be divided into several independent sub-plates. The Southeast China sub-plate is moving eastwards and is detaching from Eurasia along the seismically active N-S trending Qujiang Fault( 72), while strike slip motion along the Red River Fault is right lateral(72). 637 The Indochina sub-plate is thought to be independently moving towards the northwest with respect to the SE China sub-plate, causing present day compressional seismicity in the Indoburman Ranges and extensional tectonics behind it in the South China Sea region. The boundary zone between the Indochina and the central Sunda sub-plate lies along the axis of the TonIe Sap-Mekong Basin, and the extensional tectonics along this predominantly strike-slip fault system may be held responsible for the Pliocene-Pleistocene alkaline basaltic provine which extends along its length from the Mekong Delta northwestwards into north Thailand. The tendency for the Central Sunda sub-plate to move towards the southeast is shown by extension behind it in the Andaman Sea, and important S.E. directed faults in the South China Sea and in the Malay Peninsula. Extension on these predominantly strike-slip faults has r esul ted in im portan t shear basins such as the Malaya basin, and the faul t planes pass through the Malay Peninsula, for example the Kuala Lumpur fault zone (Fig. 23). ACKNOWLEDGEMENTS I wish to thank Mr.S.Srinivass, Mr.Ching Yu Hay, and Mr.Roslin bin Ismail for draughting work, Mr.Jaafar bin Haj. Abdullah for photographic reductions, and Mrs.Zaimah bt. Ahamad Saleh for typing the manuscript. The organizers of the conference are thanked for financial support for my attendance. 638 REFERENCES 1- ~engor,A.M.C., 1984. The Cimmeride Orogenic System and the Tectonics of Eurasia. Geol. Soc. America Special paper, 195, 82 pages. 2- Stauffer,P.H. 1974. Malayas and Southeast Asia in the pattern of continental drift. Geol. Soc. Malaysia Bull. 7, 89-138. 3- Stauffer,P.H. 1983. Unraveling the mozaic of Paleozoic crustal blocks in Southeast Asia. Geol. Rundschau, 72, 1061-1080. 4- Tapponier,P., Peltzer,G., Le Dain,A.Y., Armijo,R. and Cobbold,P. 1982. Propagating extrusion tectonics in Asia, new insights from simple experiments with plasticene. Geology, 10, 611-616. 5- Burke,K. and ~engor,A.M.C. 1986. Tectonic escape in the evolution of the continental crust. Geodynamics Series. 14, 41-51, Washinton,D.C. 6- Gatinsky,Y.G., Hutchison,C.S., Minh,N.N. and Tri,T.V. 1984. Tectonic evolution of Southeast Asia. 27th Internat. Geol. Congress, Colloquim 5, Tectonics of Asia. Reports Vol.5, Nauka, Moscow, 225-240. 7- $engor,A.M.C. and Hsti,K.J. 1984. The Cimmerides of Eastern Asia: History of the eastern end of the Palaeo-Tethys. Soc. Geol. France Mem., 147, 139-167. 8- Gatinsky,Y.G. and Hutchison,C.S. 1986. Cathyasia, Gondwanaland and the Paleotethys in the evolution of continental Southeast Asia. Proceedings of GEOSEA V Congress, Geol. Soc. Malaysia. Bull. 20 (in press). 9- Hutchison,C.S. 1987. Geological Evolution of Southeast Asia Oxford Geological Sciences Series, Oxford University Press, England (in press). 10- Hutchison,C.S. 1973. Tectonic evolution of Sundaland: a Phanerozoic synthesis. Geol. Soc. Malaysia Bull. 6, 61-86. 11- Acharyya,S.K. 1978. Mobile belts of the Burma-Malaya and the Himalaya and their implication on Gondwana and Cathaysia/Laurasia continent configurations. Prinya Nutalaya (ed.) Geology and mineral resources of Southeast Asia, Asian Instit. Technology, Bangkok, 121-127. 12- Fontaine,H. and Workman,D.R. 1978. Review of the geology and mineral resources of Kampuchea, Laos and Vietnam. Prinya Nutalaya (ed.) Geology and mineral resources of Southeast Asia. Asian Instit. of Technology, Bangkok, 538-603. 13- ~engor,A.M.C. 1979. Mid Mesozoic closure of Permo-Triassic Tethys and its implications. Nature, 279, 590-593. 14- Ridd,M.F. 1980. Possible Palaeozoic drift of SE Asia and Triassic collision with China. Journ. Geol. Soc. London, 137, 635-640. 15- Bunopas,S. 1981. Paleographic history of western Thailand and adjacent parts of Southeast Asia: a plate tectonics interpretation. Geo1. Survey paper (special issue) 5, Dept. of Mineral Resources, Bangkok, Thailand. 16- Gatinsky,Y.G. 1981. Geokinematic reconstructions of the Southeastern part of the Asian continent. Izvestia vuzov geol. Lrazv. part 4(61-69) part 5(10-19). 17- Mitchell,A.H.G. 1981. Phanerozoic plate boundaries in mainland S.E. Asia, the Himalayas and Tibet. Journ. Geol. Soc. London, 138, 109-122. 639 18- Metcalfe,I. 1986. Late Palaeozoic palaeogeography of Southeast Asia: some stratigraphical, palaeontological and palaeomagnetic constraints. Geol. Soc. Malaysia Bull. 19, 153-164. 19- Stauffer,P.H.1985. Continental terranes in SoutheastAsia: Pieces of which puzzle? Howell,D.G. (ed.) Tectonostratigraphic terranes of the Circum-Pacific region. Circum-Pacific Council for Energy and Mineral Resources, Earth Science Series, No.1, 529-539. 20- Huang J iging .1984. New researches on the tectonic characteristics of China. 27th InternaL Geol. Congress, Colloguium OS, Reports Vol.5, Nauka, Moscow, 13-28. 21- Hutchison,C.S. 1975. Ophiolite in Southeast Asia. Geol. Soc. America Bull. 86, 797-806. 22- Metcalfe, 1. 1984. Stratigraphy, palaeontology and palaeography of the Carboniferous of Southeast Asia. Mem. Soc. Geol. France, 147. 23- Asama,K. 1984. Gigantopteris flora in China and Southeast Asia. Geology and Palaeontology of Southeast Asia, 25, 311-323. 24- McElhinny,M.W., Embleton,B.J.J., Ma,X.H., Zhang,Z.K. 1981. Fragmentation of Asia in the Permian. Nature, 293, 212-216. 25- Chang Chengfa, and Pan Yusheng, 1984. A preliminary synthesis of the tectonic evolution of the Qinghai-Xizang (Tibet) plateau. Tectonics of Asia Coloquium 5, 27th International Geol. Congress, Report Vol.5, Nauka, Moscow, 190-206. 26- Kon'no,E. 1972. Some Late Triassic plants from the southwestern border of Sarawak, East Malaysia. Geology and Palaeontology of Southeast Asia, 10, 125-178. 27- Tamura,M. and Hon,V. 1977. Upper Jurassic bivalves from the Kedadom Formation of Sarawak, Malaysia. Geol. and Paleontology of S.E. Asia., 18, 33-47. 28- Cameron,N.R., Clarke,M.C.G., Aldiss,D.T., Aspden,J.A. and Djunuddin, A. 1980: The geological evolution of northern Sumatra. Proc. Indonesian Petroleum Assoc. 9th Annual Convention, Jakarta, 149-187. 29- Asama,K. 1976. Gigantopteris flora in Southeast Asia and its phytopalaeogeographic significance. Geology and Palaeontology of Southeast Asia, Univ. Tokyo Press, 17, 191-207. 30- Hamilton,W. 1979. Tectonics of the Indonesian region. U.S. Geol. Survey Professional Paper 1078. 31- Rigby,J.F. and Shah,S.C. 1981. The flora from the Permian nonmarine sequences in India and Australia: a comparison. Cresswe1l,M.M. and Vella,P. (eds.) Gondwana Five, A.A.Balkema, Rotterdam, 39-41. 32- Burrett,C. and Stait,B. 1984. Southeast Asia as a part of Gondwanaland - rifting, drifting and collision. 27th International Geol. Congress, Abstracts Vol.IX Part 1, Nauka, Moscow, page 259. 33- Veevers,J.J. (editor) 1984. Phanerozoic earth history of Australia. Clarendon Press, Oxford. 34- Stauffer,P.H. and Mantajit,J. 1981. Late Paleozoic tilloids of Malaya, Thailand and Burma. Hambrey,M.J. and Harland,iLB. (eds.) Earth's Pre-Pleistocene Glacial Record. Cambridge Univ. Press, 331-337. 35- Stauffer,P.H. and Lee Chai Pengo 1986: Late Palaeozoic glaCial marine facies in Southeast Asia and its implications. Geol. Soc. Malaysia Bull. 20 (in press). 36- Waterhouse,J.B. 1982. An Early Permian cool-water fauna from pebbly mudstones in South Thailand. Geological Magazine, 119, 337-354. 640 37- Li Xingxue, Wi Yiming, and Fu Zaibin, 1984. Preliminary study on a mixed Permian flora from Xiagangjian of Gerze district, Xizang, and its palaeogeographic significance. Intl. Symposium Himalayan Geol., Chengdu, China. Abstracts (supplement), 3-4. 38- gengor,A.M.C. 1985a: Die Alpide und die Kimmeriden: Die Verdoppelte Geschichte der Tethys. Geol. Rudschau, 74/2, 181-213. 39- Wolfenden,E.B. 1976. The information problems of the mining and petroleum geologist working in (or on) South-East Asia: Sources of Geological Information for SE Asia. Geol. Soc. London Misc. Pap. No.9, 1-4. 40- Gatinsky, Y .G. (in press) : The Lateral structural-formational analysis. Nedra, Moscow, (in Russian). 41- Mitchell,A.H.G. 1984. Post-Permian events in the Zangbo 'suture' ,ne, Tibet. Journ. Geol. Soc. London, 141, 129-136. 42- Helmcke,D. 1985. The Permo-Triassic "Paleotethys" in mainland Southeast Asia and adjacent parts of China. Geol. Rundschau, 74/2, 215-228. 43- Helmcke,D. and Lindenberg,H.G. 1983. New data on the "Indosinian" orogeny from central Thailand. Geol. Rundschau, 72, 317-328. 44- Ray,D.K. 1982. Tectonic map of South and East Asia 1:5,000,000 Commission for geological map of the world, Paris. 45- Hutchison,C.S. 1982. Southeast Asia. Nairn,A.E.M. and Stehli,F.G. (eds.) The Ocean Basins and Margins, Vol.6, The Indian Ocean. Plenum Press, New York, 451-512. 46- Gatinsky,Y.G. (compiler) 1983. Geological map of Indo-China Peninsula and adjacent territories 1:2,500,000. All-union Scientific Research Institute of Geology of Foreign Countries, Moscow. 47- Kobayashi,T., Torijama,R., and Hashimoto,W. (eds.) 1984. Geology and Palaeontology of Southeast Asia Vol.25, Univ. Tokyo Press. 48- Bender,F. 1983. Geology of Burma, Gebrlider Borntraeger, Berlin. 49- Mitchell,A.H.G. and McKerrow,W.S. 1975. Analogous evolution of the Burma Orogen and the Scottish Caledonides. Bull. Geol. Soc. America, 86, 305-315. 50- Prinya Nutalaya, 1983. (ed.) 1983. Stratigraphic correlation of Thailand and Malaysia. Vol.l, Technical papers Geol. Soc. Malaysia and Thailand, Kuala Lumpur and Bangkok. 51- Yang Zunyi, Cheng Yuqi, and Wang Hongzhen, 1986. The geology of China. Clarendon Press, Oxford, 303 pages. 52- Gobbett,D.J. and Hutchison,C.S.1973: Geology of the Malay Peninsula: West Malaysia and Singapore, John Wiley-Interscience, New York. 53- Haile,N.S. 1974. Borneo, Spencer,A.M. (ed.): Mesozoic-Cenozoic orogenic belts; data for orogenic studies. Geol. Soc. London Spec. Publ. 4, 333-347. 54- Hayes,D.E. (ed.) 1980. The tectonic and geologic evolution of Southeast Asian seas and islands. Geophysical monograph 23, Amer. Geophysical Union, Washington. 55- Hayes.D.E. (ed.) 1983. The tectonic and geologic evolution of Southeast Asian seas and islands, Part 2. Geophysical Monograph 27, Amer. Geophysical Union, Washington. 641 56- Beraua of Mines, 1981. Geolo~y and mineral resources of the Phillipines, Vol.l, Geology.ureau of Mines and Geo-Sciences, Manila. 57- Tran Van Tri, Tran Kim Thach, and Truong Cam Bao. 1979. Geology of Vietnam (North Part). General department of geology, research Instit. of Geology and mineral resources, Hanoi. 58- Trung,H., Bao,N.Z., My,B.P., Tinh,T.H. 1979. Peculiarities of the intrusive magmatic formation distribution in southern Vietnam. Dia chat va khoang san Vietnam, (in Vietnamese). 59- Jones,C.R. 1973. Lower Palaeozoic. Chapter 3 Gobbett,D.J. and Hutchison,C.S. (eds.) Geology of the Malay Peninsula Wiley-Interscience, New York, 25-60. 60- Scotese,C.R., Bambach,R.K., Barton,C., Van Der Voo,R., Ziegler,A.M. 1979. Paleozoic Base Maps. Journal of Geology, 87,217-277. 61- Van Bemmelen,R.W. 1970. The Geology of Indonesia, Vol.lA General geology of Indonesia and adjacent archipelagoes, 2nd edn., Vol.2 Economic Geology, Vol.lB Portfolio and index. Martinus Nijhoff, The Hague. 62- Metcalfe,I. 1983. Southeast Asia. Carlos Martinez Diaz (ed.) The Carboniferous of the World. I China, Korea, Japan and S. E. Asia:"" I.U.G.S. Publication No.16. Instit. Geol. Y. Minero, Spain, 213-243. 63- MacDonald,A.S. and Barr,S.M. 1978. Tectonic significance of a late carboniferous volcanic arc in northern Thailand. Prinya Nutalaya (ed.) Geology alnd Mineral Resources of Southeast Asia GEOSEA III. Asian Instit. Technology, Bangkok, 151-156. 64- Liew,T.C. and Page,R.W. 1985. U-Pb zircon dating of granitoid plutons from the west coast province of Peninsular Malaysia. Journ Geol. Soc. London, 142, 515-526. ----- 65- Bunopas,S. and Vella,P. 1983. Tectonic and geologic evolution of Thailand. Proc. of workshop on stratigraphic correlation of Thailand and Malaysia, Vol.l: Technical papers. Geol. Soc. Thailand, Bangkok, 307-322. 66- MacDonald,A.S. and Barr,S.M. 1984. The Nan River mafic-ultra-mafic bel t, northern Thailand: geochemistry and tectonic significance. Geol. Soc. Malaysia Bull.17, 209-224. 67- Katili,J.A. 1984. Tectonic framework, resources occurrences and relatedproblems in Southeast Asia. Katili,J.A. and Reinemund,J.A. (eds.) Southeast Asia: Tectonic Framework, earth resources and regional geological programs Internat. Union of Geol. Sciences, Ottawa, Publ. No.13, 18-68. 68- Haile,N.S., Beckinsale,R.D., Chakraborty,K.R., Abdul Hanif Hussein, and Tjahjo Hardjono. 1983. Palaeomagnetism, geochronology and petrology of the dolerite dykes and basaltic lavas from Kuantan, West Malaysia. Geol. Soc. Malaysia Bull., 16, 71-85. 69- Hutchison,C.S. 1985. A re-assessment of the evidence that the S.E. Asian shear basins resulted from the India-Tibet collision. Geol. Soc. Malaysia Petroleum Geol. Seminar '85. Kuala Lumpur Dec. 1985 (abstract). 70- Hutchison,C.S. 1983. Multiple Mesozoic Sn-W-Sb granitoids of Southeast Asia. Roddick,J.A. (ed.) Circum-Pacific Plutonism Terranes. Geol. Soc. America Memoir, 159, 35-60. 642 71- Kudrass,H.R. and Schlueter,H.U. 1983. Report on geophysical nd geological offshore surveys in the Malacca Strait (1976) . Exploration for offshore Placers and construction materials. CCOP/TP 12, Bangkok, 115-154. 72- Allen,C.R., Gillespie,A.R., Han Yuan, Sieh,K.E., Zhang Buchun, Zhu Cheng nan , 1984. Red river and associated faults Yunnan Province, China? Quaternary geology, Slip rates and seismic hazard. Geol. Soc. America Bull., 95, 686-700. 73- Wood,B.G.M., 1985. The mechanics of progressive deformation in crustal plates - a working model for Southeast Asia. Geol. Soc. Malaysia Bull., 18, 55-99. 74- ~engor,A.M.C. 1985b. East Asian tectonic collage. Nature, 318 (No.604l), 16-17. 75- Salun,S.A., Gatinsky,Y.G., Stroganova,S.A. 1975. Experiment in characterization of synclinal suture zones, as in certain structures in East Asia. Internat. Geol. Review 17, 1266-1274. 76- Burrett,C. and Stait,B. 1985. Southeast Asia as a part of an Ordovician Gondwanaland - a paleobiographic test of a tectonic hypothesis. Earth and Planet. Sci. Letters, 75, 184-190. 77- Lin Jin-Lu, Fuller,M., Chang We-You. 1983. Position of South China Block on Cambrian world map (abstract). E.O.S. Transactions, Amer. Geophysical Union, 64(18), page 320. 78- Haile, N. S. 1980. Paleomagnetic evidence from the Ordovician and Sil urian of northwest Peninsular Malaysia. Earth and Planet. Sci. Letters, 48, 233-236. 79- Audley-Charles,M.G. 1983. Reconstruction of Eastern Gondwanaland, Nature, 306, 48-50. 80- McElhinny,M.H., Haile,N.S., Crawford,A.S. 1974. palaeomagnetic evidence shows Malay Peninsula was not part of GondwanalandNature, 252, 641-645. 81- Bunopas,S., Vella,P., Pitakpaivan,K., and Sukroo,J. 1978. Preliminary paleomagnetic results from Thailand sedimentary rocks. Prinya Nu talaya (ed.). Geology and Mineral Resources of Southeast Asia. Asian Instit. of Technology, Bangkok, 25-32. 82- Rao,P.C. 1984. Sedimentology of some Permo-Triassic carbonates of Malaysia. Geol. Soc. Malaysia GEOSEA V, Abstracts of papers Page 28. 83- Archibold,N.W., Pigram,C.J., Ratnam,N., Hakim,S. 1982. Indonesian Permian brachiopod fauna and Gondwana-Southeast Asia relationships. Nature, 296, 566-558. 84- Wang,T., Teng,H., Li,C. and Yeh,S. 1960. Acta Geophysica Sinica, 9, 125-138. 85- Barr,S.M., MacDonald,A.S., and Haile,N.S. 1978. Reconaissance palaeomagnetic measurements on Triassic and Jurassic sedimentary rocks from Thailand. Bull. Geol. Soc. Malaysia, 10, 53-62. 86- Achache ,J. and Courtillot, V. 1985. A preliminary Upper Triassic paleomagnetic pole for the Khorat Plateau (Thailand): consequences for the accretion of Indochina against Eurasia. Earth and Planet. Sci. Letters, 73, 147-157. 87- Achache,J., Courtillot,V. and Besse,J. 1983. Paleomagnetic constraints on the Late Cretaceous and Cenozoic tectonics of Southeast Asia. Earth and Planet. Sci. Letters, 63, 123-136. 643 88- Sasajima,S., Otofuji,Y., Hirooka, Suparka, Suwijanto and Hehuwat,F. 1978. Paleomagnetic studies on Sumatra island on the possibility of Sumatra being part of Gondwanaland. Rock Magn. Paleogeophysics, 5, 104-110. 89- Haile,N.S., 1979a. Palaeomagnetic evidence forrotation and northward drift of Sumatra. Journ. Geol. Soc. London, 136, 541-545. 90- Haile,N.S. and Khoo,H.P. 1980. Paleomagnetic measurements on Upper Jurassic to Lower Cretaceous sedimentary rocks of Peninsular Malaysia. Geol. Soc. Malaysia Bull. 12, 75-78. 91- Haile ,N.S. and Briden ,J.C. 1982. Past and future palaeomagnetic research and tectonic history of east and southeast Asia. Palaeomagnetic Research in Southeast and East Asia , CCOP /TP 13. United Nations Building, Bangkok, 25-46. 92- Rishworth,D.E.H. 1974: The Upper Mesozoic terrigenous Gagau Group of Peninsular Malaysia. Geol. Surv. Malaysia Special paper 1, Ipoh. 93- Haile,N.S., McElhinny,M.W., McDougall,I. 1977. Paleomagnetic data and radiometric ages from the Cretaceous of West Kalimantan (Borneo), and their significance in interpreting regional structure. Journ. Geol. Soc. London, 133, 133-144. 94- Bignell,J.D.1972. The geochronology of Malayan granites. Unpubl. D. Phil. Thesis, Univ. Oxford. 95- Haile,N.S., 1979.b. Rotation of Borneo microplate completed by Miocene, paleomagnetic evidence. Newsletter of the Geol. Soc. Malaysia, 5, 19-22. 96- Ninkovich,D. 1976. Late Cenozoic clockwise rotation of Sumatra. Earth and Planet. Sci. Letters, 29, 269-275. 97- Kingston,D.R., Dishroon,C.P. and Williams,P.A. 1983. Global basin classification. Amer. Assoc. Petrol. Geol. Bull., 67, 2175-2193. THE CONTRIBUTION OF VERTEBRATE PALAEONTOLOGY TO THE GEODYNAMIC HISTORY OF SOUTH EAST ASIA Eric Buffetaut U.A. 720 du C.N.R.S. Laboratoire de Paleontologie des Vertebres Universite Paris VI 4 place Jussieu 75230 Paris Cedex 05 France ABSTRACT. Continental fossil vertebrates are good indicators of former land connections between continental blocks, and their study can provide latest possible ages for collisions between geotectonic units. The vertebrate fauna from the Norian Huai Hin Lat Formation of northeastern Thailand shows close affinities with vertebrate faunas from Laurasia, and indicates that there must have been a continental link between the Indochina microcontinent and Laurasia in late Triassic times. The skull of a dicynodont from near Luang Prabang (Laos) may indicate an earlier (latest Permian or earliest Triassic) connection; unfortunately, the specimen has been lost, so that its precise significance is uncertain. 1. INTRODUCTION According to current theories about the geological evolution of South East Asia, this part of the world consists of microcontinents which have become accreted to the South China block and to each other sometime during the Phanerozoic. There are still many uncertainties about the palaeogeographical origin of these blocks and about the chronology of their drift and eventual collisions. The main 645 A. M. C. ~engor (ed.), Tectonic Evolution o/the Tethyan Region, 645-653. © 1989 by Kluwer Academic Publishers. 646 microcontinents of mainland South East Asia, the Indochina (or Indosinia) block and Bunopas's "Shan- Thai" block (Bunopas, 1981) are supposed by many geologists to have separated from the northern margin of Gondwanaland, and then to have drifted northward before colliding with each other and with South China. However, there is no general agreement about the timing of this geological history; some authors (~engor, 1985; gengor & Hsli, 1985) favour a Triassic date for the collisions, while others (Helmcke, 1985a, 1985b) envisage earlier, Palaeozoic, collisions. Within this geological framework, continental fossil vertebrates from South East Asia can provide important evidence concerning the palaeogeographical relations of the South East Asian blocks at different stages of their evolution. The underlying principle is simple: the systematic affinities of land vertebrates (or, more generally, continental organisms) during a given period reflect the palaeogeographical position of the continental block they inhabited. In the case of a microcontinent separating from Gondwanaland, drifting northward, and finally colliding with Eurasia, three main stages of faunal evolution may be distinguished: - the microcontinent is still part of the northern margin of Gondwanaland, and its continental vertebrate fauna is similar to that of other regions of that supercontinent; - the microcontinent has separated from Gondwanaland, but has not yet come into contact with Eurasia; it has thus become an "island continent", on which evolution in isolation has given rise to an endemic fauna, different from both Gondwanan and Laurasian faunas; - the microcontinent has collided with a part of Eurasia, which has enabled Laurasian continental vertebrates to colonise it. The resulting faunal assemblage may at first be a mixture of endemic and Laurasian elements (plus possible Gondwanan survivors), but experience shows that the organisms from the ~arger land mass rapidly tend to dominate the endemIc forms, and the latter usually suffer severe extinction soon after faunal interchange st~rts. The biological reasons for this are not qUIte c~ear, and ~eed not be discussed here. ThIS the?retIcal picture of the events is of course somethIng of an oversimplification. It shou~d be reme~bered that passive or active dispersal of contlnenta~ anImals across marine barriers may allow the Introduction of allogenous faunal elements into an "island continent" before full terrestrial links have been established with another land mass (however, the amount of faunal interchange which can be caused by such dispersal is usually limited). On the other hand, climatic and geographical barriers within a continuous continental area may prevent free faunal interchange between different regions, at least for some groups of organisms. Moreover, despite the fact that vertebrates often evolve and differentiate rapidly, some groups may not show much geographical differentiation, so that, in the present case, forms from Laurasia and from Gondwanaland may not be easy to distinguish (this is an especially important problem in the case of faunas of late Palaeozoic and early Mesozoic age, because at that time the continents were united to form a Pangaea, which facilitated the dispersal of continental animals). The results of a palaeobio- geographical analysis of fossil vertebrates from an accreted microcontinent are thus unlikely to be as clear-cut as the theoretical model given above would suggest. Nevertheless, it remains that the presence of Laurasian land vertebrates on one of the South East Asian blocks is strong evidence for the existence of a terrestrial connection with the northern continents. In the absence of a continuous fossil record, this kind of evidence only provides a latest possible date for the establishment of such a connection, which may actually have taken place much earlier. Also, it says very little about the nature and location of the connection. How geodynamic models can be constrained by palaeontological data can be illustrated by the case of the fossil vertebrates from the Indochina block, which have become much better known in recent years, following a series of Thai-French expeditions (Buffetaut, 1981, 1982; Buffetaut & Ingavat, 1985). 2. THE LATE TRIASSIC VERTEBRATES OF THAILAND The vertebrate faunas discovered by the above-mentioned Thai-French expeditions in the Mesozoic of the Khorat Plateau (northeastern Thailand) range in age from late Triassic to Cretaceous. The oldest of these faunal assemblages is obviously of prime importance for palaeobiogeographical reconstructions based on the principles outlined above. It has been found in the Huai Hin Lat Formation, which mainly consists of freshwater bituminous shales and limestones. The Huai Hin Lat Formation had been referred to the 647 648 late Triassic on the basis of invertebrate palaeontology, palaeobotany and palynology (see review in Buffetaut & Ingavat, 1982), and this has been fully confirmed by the study of fossil vertebrates, which strongly suggest a Norian age, by comparison with the "classical" late Triassic vertebrate assemblages from Germany (Buffetaut & Ingavat, 1982). The Huai Hin Lat vertebrate fauna includes fishes, amphibians and reptiles. The fishes are represented by actinopterygians and dipnoans. Dipnoan toothplates have been referred by Martin and Ingavat (1982) to Ceratodus cf. szechuanensis. The lungfish Ceratodus szechuanensis had originally been described from Sichuan Province, China. Its occurrence in northeastern Thailand thus suggests that by late Triassic times a continental connection of some kind had been established between the Indochina block and South China. The amphibians are mainly known from an incomplete skull which has been identified as Cyclotosaurus cf. posthumus by Ingavat and Janvier (1981). The stegocephalian Cyclotosaurus posthumus is a species from the Stubensandstein (Norian) of Germany, and the specimen described by Ingavat and Janvier can be considered as a Laurasian faunal element in the Huai Hin Lat vertebrate fauna. The reptiles are represented in the Huai Hin Lat fauna by turtles and phytosaurs. The former are among the few known Triassic turtles (de Broin et al., 1982). On the basis of the available fragmentary material, de Broin (1984) has described a new species of the genus Proganochelys, P. ruchae. The genus Proganochelys has first been described from the late Triassic of Germany, and its occurrence in the Huai Hin Lat Formation may be considered as suggesting links with Laurasia, although de Broin (1984) thinks the evidence is inconclusive. The phytosaurs are represented by a fairly abundant material, only part of which has been described (Buffetaut & Ingavat, 1982). Two forms seem to be present, one of them resembling the North American Rutiodon and the European Belodon (Buffetaut & Ingavat, 1982), the other one referrable to Mystriosuchus, a genus known from the Stubensandstein of Germany (Buffetaut & Ingavat, in preparation). Although phytosaurs were widely distributed, in both Gondwanaland and Laurasia, in the late Triassic, Rutiodon, Belodon and Mystriosuchus have not been reported from Gondwanaland, and the specimens from the Huai Hin Lat Formation do seem to indicate Laurasian affinities. The Huai Hin Lat vertebrates thus show that in the Norian the Indochina block was inhabited by a land and freshwater fauna with essentially Laurasian affinities. There is no evidence among the available fossil material of characteristically Gondwanan or endemic forms. Following the principles outlined above, it can be concluded that in the late Triassic a terrestrial connection of some kind must have existed between the Indochina block and Eurasia. This suggests that accretion to South China had already taken place, which confirms that early conceptions of a late drift and collision of South East Asia (Ridd, 1971) were erroneous, but is in agreement with more recent hypotheses of Triassic collision (Ridd, 1980; Mitchell, 1981; ~engor, 1985). However, the Laurasian affinities of the Huai Hin Lat fauna of course do not preclude a much earlier accretion of the Indochina block, in the Devonian or Carboniferous (Helmcke, 1985; Hutchison, this meeting). The available data on fossil vertebrates do not allow to choose between these hypotheses. Another consequence of the presence of Laurasian vertebrates in the upper Triassic of the Indochina block is that there must have existed continental connections between Indochina, South China, North China, and Russia/Siberia at that time, to allow the colonization of Indochina by forms closely related to European ones. 3. THE DICYNODONT FROM LUANG PRABANG (LAOS) 649 Continental fossil vertebrates from South East Asia older than the Huai Hin Lat fauna are obviously of great potential importance for palaeogeographical reconstructions, as they may provide information concerning the position of the South East Asian blocks before the late Triassic, when collision of Indochina with South China had undoubtedly already taken place. Unfortunately, very few such remains are known. The most interesting specimen is a fragmentary skull of a mammal-like reptile found in the 1890s by a French expedition to Laos. The fossil was collected near Luang Prabang, in northwestern Laos, and first mentioned by Counillon in 1896. It was described as a new species of the genus Dicynodon, D. incisivum, by Repelin in 1923, and then referred to the type species of the genus, D. lacerticeps, by Piveteau in 1938. There are many problems concerning this specimen which make its interpretation in biogeographical terms rather difficult (Buffetaut, 1985). Its 650 geological age is uncertain, with estimates varying from late Permian to middle Triassic. Fromaget (1952) has also suggested that the specimen was reworked from the upper Permian into Norian continental deposits. Moreover, the specimen obviously needs to be restudied as recent revisions (Cluver & Hotton, 1981; Cluver ' & King, 1983) have modified our understanding of dicynodont systematics, and it is clear that an accurate identification of the animal is a prerequisite to any safe palaeobiogeographical interpretation. Unfortunately, the specimen, which belongs to the collections of the University of Marseille, seems to be lost, and the descriptions and rather poor illustrations published by Repelin and by Piveteau do not provide an adequate basis for a reinterpretation. Nevertheless, what is known of dicynodont evolution and biogeography suggests that the Luang Prabang specimen may indicate palaeogeographical connections with Laurasia rather than with Gondwanaland. Dicynodonts appeared sometime during the Permian, with the earliest known form, Eodicynodon oosthuizeni, coming from the lower middle Permian of South Africa (see Cluver & King, 1983), and they soon became extremely widespread, with an almost world-wide distribution (the case of the early Triassic Lystrosaurus is especially remarkable: see Colbert, 1982). However, the occurrence of a presumably late Permian or early Triassic dicynodont on the Indochina block strongly suggests that this microcontinent was not isolated from the larger land masses at the time. It is extremely unlikely that dicynodonts had already appeared when the Indochina block became separated from Gondwanaland (if it was ever part of Gondwanaland), because this separation probably took place well before the Permian. The Luang Prabang dicynodont would rather indicate an immigration from China (where dicynodonts are well represented: see Sigogneau- Russell & Sun, 1981), and thus the existence of an early terrestrial land connection between the Indochina block and the Laurasian regions. If this interpretation is correct, the Indochina block may have been occupied by Laurasian faunal elements well before the appearance of the late Triassic Huai Hin Lat fauna. However, the available evidence is very scanty, and its interpretation is rather uncertain; it can only be hoped that future discoveries in the early Triassic or late Palaeozoic of South East Asia will shed more light on this question. 651 4. CONCLUSION The available evidence concerning the continental vertebrates from the Mesozoic and, possibly, the latest Palaeozoic of South East Asia shows that in the late Triassic the Indochina block was already inhabited by a Laurasian fauna, which in turn shows that a terrestrial connection with the northern land masses must have been realized. The establishment of such a connection may have taken place much earlier, and the Luang Prabang dicynodont may be considered as possible palaeontological evidence for such an early connection. It should be remembered that almost all the Palaeozoic and Mesozoic vertebrates from South East Asia come from the Indochina block. Very little is known about the fossil vertebrates of the Shan-Thai and the East Malaya blocks (the latter possibly being a southern continuation of the Indochina block). Scrivenor (1907, 1931) did report the occurrence of fish scales referrable to the genus Semionotus from Triassic rocks near Kuala Lipis in central Malaya, but these fossils do not bring any useful biogeographical informations. More recently, in 1984, a first locality with Mesozoic vertebrates has been discovered in the southern peninsula of Thailand near the city of Trang (Buffetaut & Ingavat, 1985); the available material is still too scanty to allow accurate stratigraphic and biogeographical conclusions to be drawn, but the remains of fishes, turtles and crocodilians found in 1984 suggest a Jurassic age and resemblances with faunas from the Indochina block. It is hoped that further work in southern Thailand will extend our knowledge of the vertebrate faunas of the Shan-Thai block and shed some light on its past relations with the Indochina block and mainland Asia. 5. REFERENCES BROIN, F. DE (1984). - Proganochelys ruchae n.sp., Chelonien du Trias superieur de ThaYlande. Stud. geol.Salamantic., Salamanca, vol.esp. 1, 87-97. BROIN, F. DE, INGAVAT, R., JANVIER, P. & SATTAYARAK, N. (1982). - Triassic turtle remains from northeastern Thailand. J.Vert.Paleont., Norman, 2, 1, 41-46. BUFFETAUT, E. (1981). - Elements pour une histoire paleobiogeographique du Sud-Est asiatique: l'apport des Vertebres fossiles continentaux. Bull.Soc.geol. France, Paris, (7), 23, 6, 587-593. BUFFETAUT, E. (1982). - Mesozoic vertebrates from Thailand and their palaeobiological significance. 652 Terra cognita, Paris, 2, 1, 27-34. BUFFETAUT, E. (1985). - The palaeobiogeographical significance of the Mesozoic continental vertebrates from South-East Asia. Mem.Soc.geol. France, Paris, 147, BUFFETAUT, E. & INGAVAT, R. (1982). - Phytosaur remains (Reptilia, Thecodontia) from the Upper Triassic of north-eastern Thailand. Geobios, Lyon, 15, 1, 7-17. BUFFETAUT, E. & INGAVAT, R. (1985). - The Mesozoic vertebrates of Thailand. Scient.Am., New York, 253, 2, 80-87. BUNOPAS, s. (1981). - Paleogeographic history of Western Thailand and adjacent parts of South East Asia - A plate tectonics interpretation. Geol.Surv.Paper, Bangkok, 5, 1-810. CLUVER, M.A. & HOTTON, N. (1981). - The genera Dicynodon and Diictodon and their bearing on the classi- fication of the Dicynodontia (Reptilia, Therapsida). Ann.S.Afr.Mus., Cape Town, 83, 6, 99-146. CLUVER, M.A. & KING, G.M. (1983). - A reassessment of the relationships of Permian Dicynodontia (Reptilia, Therapsida) and a new classification of dicynodonts. Ann.S.Afr.Mus., Cape Town, 91, 3, 195-273. COLBERT, E.H. (1982). - The distribution of Lystrosaurus in Pangaea and its implications. In E. Buffetaut, P. Janvier, J.C. Rage & P. Tassy (ed.): Phylogenie et Paleobiogeographie. Livre jubilaire en l'honneur de Robert Hoffstetter, Geobios Mem.spec., Lyon, 6, 375-383. COUNILLON, H. (1896). - Documents pour servir a l'etude geologique des environs de Luang Prabang (Cochin- chine). C.R.Acad.Sc., Paris, 123, 1330-1333. FROMAGET, J. (1952). - Etudes geologiques sur le Nord- Ouest du Tonkin et le Nord du Haut-Laos. 2e et 3e parties. Bull. Serv. geol. Indochine, Hanoi, 29, 6, 1-198. HELMCKE, D. (1985a). - The Permo-Triassic "Paleotethys" in mainland Southeast Asia and adjacent parts of China. Geol.Rundschau, Stuttgart, 74, 2, 215-228. HELMCKE, D. (1985b). - The orogenic evolution (Permian- Triassic) of eastern Thailand. Implications on paleogeographic models for mainland South East Asia. Mem. Soc. geol.France, Paris, 147, INGAVAT, R. & JANVIER, P. (1981). - Cyclotosaurus cf. posthumus Fraas (Capitosauridae, Stereospondyli) from the Huai Hin Lat Formation (Norian, Upper Triassic) of northeastern Thailand, with a note on capitosaurid biogeography. Geobios, Lyon, 14, 6, 711-723. MARTIN, M. & INGAVAT, R. (1982). - First record of an Upper Triassic ceratodontid (Dipnoi, Ceratodon- tiformes) in Thailand and its paleogeographical significance. Geobios, Lyon, 15, 1, 111-114. 653 MITCHELL, A.H.G. (1981). - Phanerozoic plate boundaries in mainland SE Asia, the Himalayas and Tibet. J.Geol.Soc.London, London, 138, 2, 109-122. PIVETEAU, J. (1938). - Un Therapside d'Indochine. Remarques sur la notion de continent de Gondwana. Ann.Paleont., Paris, 27, 4, 139-152. REPELIN, J. (1923). - Sur un fragment de crane de Dicynodon recueilli par H. Counillon dans les environs de Luang Prabang (Haut-Laos). Bull.Serv. geol.lndochine, Hanoi, 12, 2, 5-7. RIDD, M.F. (1971). - South-east Asia as part of Gondwana- land. Nature, London, 234, 531-533. RIDD, M.F. (1980). - Possible Palaeozoic drift of SE Asia and Triassic collision with China. J.Geol.Soc. London, London, 137, 635-640. SCRIVENOR, J.B. (1907). - Federated Malay States. Geologist's report of progress. September, 1903 - January, 1907. Federated Malay States Government Press, Kuala Lumpur, 1-44. SCRIVENOR, J.B. (1931). - The geology of Malaya. Macmillan, London, 1-217. ~ENGOR, A.M.C. (1985). - Die Alpiden und die Kimmeriden: die verdoppelte Geschichte der Tethys. Geol.Rundschau, Stuttgart, 74, 2, 181-213. 9ENGOR, A.M.C. & HSU, K.J. (1985). - The Cimmerides of eastern Asia: history of the eastern end of Palaeo-Tethys. Mem.Soc.geol.France, Paris, 147, SIGOGNEAU-RUSSELL, D. & SUN, A.L. (1981). - A brief review of Chinese synapsids. Geobios, Lyon, 14, 2, 275-279. CONVERGENT -PLATE TECTONICS VIE WED FROM THE INDONESIAN REGION Warren Hamil ton Branch of Geophysics United States Geological Survey Denver, Colorado, 80225, U.S.A. ABSTRACT. Plate boundaries have evolved complexly with time and at anyone time have varied greatly along strike. Seismically defined dips of subducting plates mark positions, not trajectories; slabs sink more steeply than they dip, and overriding plates advance over them. Subduction occurs beneath only one side at a time of an internally rigid plate, and the common regime in an overriding plate, behind a surficial accretionary wedge, is extensional, except where a collision is underway. Back-arc-basin lithosphere is built behind, or by, migrating island arcs, which lengthen and increase their curvatures as they evolve. A collision can involve two active arcs, in which case intervening lithosphere sinks beneath both of them, or an active margin and a passive one. Either type of collision generally is followed by the breaking through of new subduction, beneath the composite mass of light crust, from a new trench oceanward of the aggregate; conversely, a new subduction system commonly is a byproduct of collision. A strip of back-arc- basin crust is often left attached to the aggregate, in front of the new trench, and becomes the basement for a fore-arc basin, the leading edge of which is raised as melange is stuffed under it. Sedimentation in trenches is dominantly longitudinal and can be from distant sources. Accretionary wedges are dynamic, being thickened at both toes and bottoms by tectonic accretion and thinned by gravitational forward flow; melange is largely a product of tectonic imbrication driven by these conflicting processes; olistostromes are minor components of accretionary wedges. High-pressure metamorphic rocks form beneath overriding plates, not within wedges in front of them. Arc magmas incorporate much material from the lithosphere through which they rise and vary correspondingly with the composition of that lithosphere. Continental crust is inflated into geanticlines by magmatic-arc intrusive rocks. Submarine island-arc volcanic rocks are widely spilitized, with enrichment in sodium and depletion in calcium, by hydrothermal reaction with sea water. 655 A. M. C. flengor (ed.), Tectonic Evolution of the Tethyan Region, 655-698. © 1989 by Kluwer Academic Publishers. 656 1. INTRODUCTION The active tectonism and magmatism of onshore and offshore Indonesia and adjacent regions record the complex interactions of the Asian, Pacific, and Indian- Australian lithosphere megaplates and of dozens of lesser plates. This report discusses selected aspects of the evolution of the resulting complexes, and derives from these and other examples generalizations regarding convergent-plate processes. Indonesia and nearby regions provide the most extensive and varied modern examples of convergent-plate processes and products. Comprehension of such examples is essential for actualistic paleotectonic analyses of ancient terrains--but few practitioners of paleotectonic analysis are in fact familiar with modern systems. In my book, Tectonics of the Indonesian region (Hamilton, 1979), the accompanying tectonic map (also published separately: Hamilton, 1978a), and earthquake (Hamilton, 1974a) and sedimentary basins maps (Hamilton, 1974b), I integrated onshore and offshore geologic and geophysical data for Indonesia, southeast Asia, the southern Philippines, western Melanesia, and the adjacent seas into a synthesis of modern plate behavior and of the evolution of plate-tectonic features. (Various earlier reports and maps, 1970-77, presented interpretations repeated in, or superceded by, the monograph and maps noted, and are not cited here.) The descriptive basis for discussions in this essay comes primarily from that monograph, to which the reader is referred for documentation, in the forms both of voluminous data newly reported there and of citations to the findings of others, and for descriptions of many features not discussed here. Unreferenced statements regarding the Indonesian-Melanesian region in this essay are based primarily on that monograph. Many reports published since the completion of the book are cited here; these require modification of details of my synthesis but have in general substantiated it. In this essay, I describe aspects of Indonesian tectonics and petrology, integrate them with analogous systems elsewhere as that seems appropriate, and derive generalizations. Such generalizations are emphasized by underlining wherever they appear. A number of the concepts summarized here have evolved since I com pleted the book (and in part were reported also in my 1985 paper), but this essay is a supplement to the monograph, not a replacement for it. Figure 1 is an index map, and Figures 2, 3, and 4 illustrate many features discussed in the text; the reader unfamiliar with the Indonesian region will find reference to my book and tectonic map to be helpful also. Other useful maps of the Indonesian region include the bathymetric map by Mammerickx et al. (1976), which is more detailed than the one used as the base for my maps~ Cartographic summaries of offshore geophysical data, variously supplementing, duplicating, and incorporating my work, were presented for the same general region by Anderson ~ al. (1978; thermal properties), Hayes and Taylor (1978; earthquakes, and a schematic tectonic map), Hayes et ~. (1978; crustal structure), Mrozowski and Hayes (1978; sediment isopachs), Watts et al. (1978, free-air gravity), and Weissel and Hayes (1978; magnetic anomalies). The diverse subduction systems of the Indonesian region record the interactions between three megaplates and many smaller plates. Relative to internally-stable northwestern Eurasia, the India-Indian Ocean-Australia megaplate is moving approximately northward in this region, whereas the Pacific megaplate is moving approximately west-northwestward. The Asian megaplate is fragmented into dozens of subplates; many small plates intervene between parts 657 of the megaplates; and many small plates have been much deformed internally. Southeast Asia was crowded eastward out of the way of the indenting Indian subcontinent and has since been rotating clockwise over the oceanic Bay of Bengal (Hamilton, 1979; Tapponnier et al., 1982). Within the study region, convergence between Indian and Asian megaplates is now being taken up primarily by the continuous Burma-Andaman-Sunda-Banda subduction system, whereas that between Pacific and Asian megaplates is taken up on many subduction systems that trend mostly northerly into the Philippines and along boundaries farther east. Complex subduction and strike-slip systems bound the megaplates and small plates in the zone of interaction between Indian and Pacific megaplates, along and north of New Guinea and in the tectonic knot in northeastern Indonesia and surrounding areas. 2. PLATE TECTONICS General aspects of plate behavior are relevant to an understanding of plate interactions in the Indonesian region and to the paleotectonic application of concepts derived from that region. Seven very large lithospheric plates, and numerous mid- and small-sized ones (the concept of coherent plates breaks down at the small-scale end), are now all moving relative to all others about the Earth's surface. Corollaries often overlooked are that all plate boundaries-divergent, convergent, strike-slip, and obligue--are also moving, and that most boundaries change greatly in length and shape with time. Although plates tend to be internally rigid and to interact mostly at their boundaries, parts of many plates undergo severe internal shortening, extension, and strike-slip deformation. Many plates grow internally as well as marginally. Velocities of relative motion between adjacent plates presently range up to about 13 cm/yr. If "hot spots" exist in the mantle--I am one of the minority that regards many of the features so designated as products of propagating rifts, and most of the others as imaginary--they too are moving relative to one another. 2.1. Mechanism Absolute velocities of present large plates correlate positively with the lengths of ridges and of trenches along their perimeters, and negatively with the proportion of continental lithosphere within them (Carlson, 1981). It appears from the quantitative relationships between these factors that on average slab-pull is about 2.5x as important as is ridge-slide in moving plates, whereas the drag of thick continental lithos here has an effect about e ual to that of rid e-slide but of opposite sign Carlson, 1981. The great relief on the base of an oceanic- lithosphere plate, between lithospheric mantle and less-dense asthenosphere, is much more important in producing this sliding than is the topographic relief on the top of the plate; hence the term, "ridge slide." (The more common term "ridge- push" incorporates an invalid concept of ridge formation by active spreading; actual ridges form where plates move apart and hot mantle wells into the gap.) Ma ·or late motions thus are controlled b lar e lateral variations in lithos here density and thickness that result primarily from cooling Carlson, 1981; Hager and O'Connell, 1981). Convection in the upper mantle is largely a complex product, not a major cause, of plate motion (Alvarez, 1982). Motions of small plates are 658 90· 100· 110· Figure 1. Index map of the Indonesian region. For bathymetry and more detail, see Hamilton (l978a; or 1979, plate 1) or Mammerickx et al. (1976). primarily byproducts of motions of adjacent large plates ~ • ..s.., Hamilton, 1979). Absolute velocity of lithosphere at low latitudes is in general greater than that at high latitudes, so the Earth's rotation must be a factor in driving forces (Solomon et ~., 1975), perhaps by a gyroscopic feedback mechanism (Hamil ton, 1979). 2.2. Heat and Variations with Time Plate motions are responsible for most of the Earth's heat loss. Of the total heat lost by the Earth, about 60 percent is lost by magmatism at spreading ridges and by the subsequent cooling of new oceanic lithosphere as it moves away from ridges 659 120~ 130· 140· 10· o· 10· (Sclater et al., 1981). Because the rate of heat loss by the Earth has probably decreased with time and because ancient crustal nonmagmatic thermal gradients demonstrable by petrologic thermobarometry were little if any steeper than modern ones, it seems likely that plate motions have become slower on average with time. There may have been fluctuations within this progression representing variations of 10 or 20 percent in the rates of plate generation and consumption (d. Parsons, 1982), and there certainly have been major unidirectional changes in the petrologic evolution of crust and mantle. Nevertheless, plate tectonics appears to have operated, broadly as it has during late Phanerozoic time, at least since the early Proterozoic. Archean crust displays the effects of much more voluminous, and in part higher temperature, magmatism than that of younger time and of expulsion of light, continent-forming elements directly from a llttle- 660 differentiated mantle, and specific processes that then operated are much debated; many of us see the effects as due to more rapid motions of more and smaller plates than those of later time. For an evaluation of Precambrian plate processes, see Windley (1984). 2.3. Subduction The sinking of oceanic lithosphere beneath another plate, either oceanic or continental, is often but mistakenly visualized as consisting of the sliding of oceanic lithosphere down a slot fixed in the mantle; and an overriding plate is commonly assumed to be shortened compressively across the entire width of the magmatic arc and the belt of foreland deformation. Both of these assumptions are disproved by the characteristics of modern convergent-plate systems in which the subducting plate is of normal oceanic lithosphere and the Benioff seismic zone has a moderate to steep inclination, and also by analyses of absolute plate motions. Hinges commonly retreat--roll back--into incoming oceanic plates as overriding plates advance, even though at least most subducting plates are also advancing in "absolute" motion in a total-earth reference frame. Among those who have made these points on various bases are Carlson and Melia (1984), Chase (I978), Dewey (I980), Garfunkel et ale (1986), Hamilton (1979), Malinverno and Ryan (I 986), Molnar and Atwater l1978), and Uyeda and Kanamori (1979). Subducting slabs sink more steeply than the inclinations of Benioff seismic zones, which mark positions, not trajectories, of slabs. As most of the authors just cited have emphasized, the typical regime in an overriding plate above a sinking slab is one of extension, not shortening. Subduction can occur beneath only one side at a time of a rigid plate. Much of the negative buoyancy, mechanical behavior, and seismicity of subducting slabs is due to density-phase changes (Pennington, 1983; Rubie, 1984). 2.4. Arc Migration and Back-arc Spreading Karig ~ • .8.., 1972) early recognized from data in the Philippine Sea region that the Mariana island arc has migrated Pacificward as new back-arc-basin oceanic crust formed behind it. He and many others since ~ • .8.., Taylor and Karner, 1983) have found this to be generally true of island arcs. Oceanic island arcs do not mark the fronts of rigid plates of old lithosphere, but instead mark the fronts of plates of young lithosphere that are widening in the extensional regimes above subducting slabs. Oceanic arcs commonly are not inaugurated by the breaking of subduction through old oceanic crust, but rather break through at boundaries between thin and thick crust and migrate over the plates of thin crust ~ • .8.., Hamilton, 1979; Karig, 1982). In anyone system, periods of back-arc spreading may alternate irregularly with periods of magmatism along a volcanic-arc welt (Crawford et a!., 1981; some authorities disagree). During a spreading phase, the magmatic welt can move forward with the advancing part of the overriding plate, can be abandoned as a remnant arc on the relativel retreatin art or can be s lit longitudinally between them Taylor and Karner, 1983 • An island arc should be viewed as a product of a subducting slab rather than as a fixed feature of an overriding plate. A belt of arc-magmatic rocks forms above that part of a subducting slab whose top is 100 km or so deep, hence migrates to track that contour as the slab falls away. Mechanisms of back-arc spreading are still debated, but it appears to me, as to some others, that oceanic 661 back-are-basin lithosphere forms in at least two ways--by irregular spreading behind an are, and by the rapid migration of a magmatic arc which plates out a variable-thickness sheet of lumpy arc crust rather than forming a welt of thick crust. 2.4.1. Arc festoons. Arcs increase in curvature as they migrate. A migrating arc becomes pinned where it encounters thick crust in the subducting plate that either is nonsubductible or that forms a stiffening girder, and festoons and sharply curving arcs result where migration continues away from such obstructions (McCabe, 1984). Pinning against the Caroline Ridge may explain the Yap-Mariana festoon, and pinning against the Emperor Seamount Ridge may explain the Kamchatka-Aleutian festoon. The edges of the Indian subcontinent control the Pakistan and Assam syntaxes. 2.5. Ophiolites On-land ophiolites are sections of upper oceanic lithosphere that were long assumed to be samples of spreading mid-ocean-ridge materials. It now appears probable that instead most, perhaps all, large sheets of ophiolite incorporated tectonicall into the continents are roducts either of arc ma matism or back-arc spreading Coleman, 1984; Hawkins et al., 1984; Pearce et al., 1984. Much has yet to be learned about the evolution of these complexes, but the mechanisms of irregular spreading and fast-migrating arcs appear capable of explaining many relationships. 3. SUNDA SUBDUCTION SYSTEM The subduction front along Sumatra, Java, Bali, and Lombok consists of concentric, arcuate tectonic features (Figures 2, 3) that typify those along other active margins of continents and mature island arcs. I refer here to this 3000-km sector of the great subduction system that is continuous from Burma around the Banda Arc as the Sunda system. In the south is the trench, and northward from it rises the surface of the accretionary wedge at the front of the overriding plate to a culmination at a fore-arc ridge. Islands rise from the ridge along Sumatra but the ridge is wholly submarine south of Java, Bali, and Lombok. Between the ridge and the magmatic arc is the submarine fore-arc basin. I have previously (as, 1979) used the term "outer-arc ridge" rather than the synonym "fore-arc ridge" because of the possible confusion of the latter term with the classical continental term "foreland", which is on the landward, not the trenchward, side of the Sunda system. I here (Figures 2, 3) yield to the now more common usage. Similarly, the "fore-arc basin" of most recent literature and this essay is the "outer-arc basin" of my previous reports. Note that a "foreland basin" is on the same side of an arc as is a "back-arc basin" but is on the side opposite from a "fore-arc basin". Along the Sunda sector of the great Burma-Banda subduction system, Indian Ocean lithosphere is being subducted, at large to moderate convergence angles, beneath an arc system that changes along strike from continental in Sumatra through transitional in Java to oceanic in Bali and Sumbawa. The Sunda sector of the subduction system has been active only during middle and late Cenozoic time. "' -~ a; E . Q ~ . 663 3.1. Trench The Sunda Trench, like trenches that mark traces of subduction systems along continental margins or mature island arcs elsewhere, has inner and outer "walls" that slope only 7 degrees or so (Figure 2). The trench does not mark an abrupt hinge in the subducting Indian Ocean lithosphere, but rather represents the dihedral angle between a surficial accretionary wedge, at the front of the overriding plate, and oceanic lithosphere depressed by that wedge. An outer rise on the oceanward side of this trench, and other trenches, is an elastic response to that depression. The tectonic hinge, whereat the subducting plate tips downward into the mantle, lies 100-200 km landward from the bathymetric trench. Trench sides can be considerably steeper in young or midocean island arc systems in which little sediment is available for incorporation in an accretionary wedge, and the outer edge of such a trench may approximate the hinge in the subducting plate. 3.1.1. Sedimentation in trenches. Clastic sedimentation in trenches is primar ily in the form of longitudinal turbidites, and the long profiles of trench- floor fills slope smoothly and gently away from the sources of turbidites. Sunda Trench sediments as far southeast as Java came largely from the Ganges and Brahmaputra Rivers, 3000 km away (see also Moore et al., 1982); the supply is now being cut off by collision of the Ninetyeast Ridge with the trench in the Andaman sector. Aleutian Trench turbidites similarly sluice as far as 3000 km from source rivers in south-central Alaska. Source terrains thus need bear little similarity to nearb arts of the overridin late a ainst which trench turbidites are lated in an accretionary wedge. Dickinson 1982 demonstrated this to be true for various fossil accretionary wedges around the Pacific Ocean. (Overlooking of this relationship has led some geologists to postulate vast strike-slip motions within the Cretaceous accretionary wedge of coastal California.) Continental detritus sluiced along trenches, or abyssal-fan materials from continents, can be accreted to oceanic sectors of arc systems. 3.2. Accretionary Wedge Sediments and other materials scraped from subducting Indian Ocean lithosphere accumulate, snowplow fashion, in an accretionary wedge at the front of the overriding Sunda plate. The surface of the wedge is irregular, furrowed by longitudinal ridges and basins defined by imbricate thrust faults (Karig, Moore, Curray, and Lawrence, 1980), but the overall surface slope of the wedge in general decreases exponentially upward to the crest of the fore-arc ridge. Surface slopes of accretionar wed es define broad 1 similar curves of d namic e uilibrium re ardless of the widths and thicknesses of the wed es. See many reflection profiles in Hamilton, 1979. Islands along the ridge-the crest of the accretionary wedge--show rapid uplift by Quaternary coral reefs raised high above sea level, presumably because the wedge is being thickened by underplating. Trench fill can be seen on reflection profiles to be scraped off at the front of a wedge, the shallowest materials being accreted against the toe, the deeper being accreted beneath the toe a little farther back. The top of the subducting plate dips very gently beneath the accretionary wedge, which is a thin, dynamic debris pile only 15 km or so thick 75-150 km from the trench. Reflection profiles showing internal structure of this and analogous wedges elsewhere generally display semiconstant imbrication angles, dipping 30 degrees or so landward, independent of position in 664 the wedge, discordant to the gently-dipping decollement atop the subducting oceanic crust and consolidated strata upon it. 3.2.1. Wedge dynamics. Such features indicate to me that an accretionary wedge is thickened as its base is dragged back but simultaneously is thinned by gravitational spreading. The result is internal imbrication of the wedge and maintenance of a dynamic profile, akin to that of an ice sheet, as the wedge grows both laterally and vertically. Others see wedges as more static features, broadened by imbrication and offscraping at their toes but relatively stable in their landward portions. A prediction implicit in this analysis is that strain is accumulated within a wedge in the plate-convergence direction but is released during intra-wedge earthquakes primarily in the direction orthogonal to the trench. Too few first- motion solutions are available to test this prediction for the arcuate Sunda wedge. The prediction accords with solutions (Perez and Jacob, 1980; they made no such interpretation) from the eastern Gulf of Alaska, where relatively north- northwest convergence is absorbed within a wedge which swings from northwestward to southwestward trends, and where thrust earthquakes within the wedge show slip to be toward the trench rather than parallel to the convergence direction. 3.2.2. Materials of wedge. Reflection profiles across the Sunda accretionary wedge indicate that ratios of imbricated, disrupted, and coherently folded materials vary widely but relate systematically to the convergence rates and directions and to the thickness and character of sedimentar sections be in accreted G. F. Moore, Curray, D. G. Moore, and Karig, 1980. No wells have yet been drilled into the Sunda wedge. Drillholes in other similar sediment-rich modern wedges show them to be typified near their toes by broken formations with scaly-clay matrices, and by highly variable proportions of broken formations and coherently imbricated strata elsewhere. Exposures of the top of the wedge on the islands of the Sunda fore-arc ridge have been studied primarily on Nias (G. F. Moore and Karig, 1980; G. F. Moore, Billman, Hehanussa, and Karig, 1980), where coherent, unmetamorphosed lower Miocene to lower Pliocene strata structurally overlie on the northeast, and elsewhere are imbricated with, polymict melange composed of undated materials. The melange is slightly metamorphosed, extremely sheared and disrupted, and is dominated by terrigenous clastic sediments of deep-water origin but contains abundant fragments of chert and basalt from the upper oceanic crust, and sparse fragments of mafic and ultramafic deeper-seated oceanic rocks. Fossils indicate the coherent strata to have been deposited in water depths that in general decreased with time. Although no depositional contacts of strata on melange were found, Moore and his associates assumed that the Neogene strata were deposited atop the wedge and imbricated into it. I infer (incorporating interpretations quite different from theirs regarding the history of the Sunda system in particular and the relationships between accretionary wedges and fore- arc basins in general) that this explanation may be valid for the younger Neogene strata but that the older Neogene materials were deposited on the crystalline oceanic basement of what was then the outer part of the fore-arc basin, that basement having been since removed by tectonic erosion by the subducting plate and itself subducted. I thus see the fore-arc ridge as having migrated landward with time, relative to the cratonic part of the overriding plate, as its supporting 665 buttress was ground away. (Moore and associates, and Karig, 1982, by contrast inferred seaward migration by growth of stabilized wedge.) I suspect that the formation of the undated polymict melange largely overlaps in time the deposition of the older Neogene strata, and I see the rise to the surface of the relatively deep-seated melange as being due in part to the tectonic erosion of overlying crystalline plate material. Similar inferences seem to me to be required for the analogous Cretaceous systems of coastal California, where relationships are much better known. 3.2.3. Tectonic versus sedimentary melange. Many students of fossil accretionary wedges infer much or most broken formation and melange in accretionary wedges to have formed as thick olistostromes (submarine slumps). Minor slumps may be abundant on the surfaces of wedges, but only one major submarine slump (D. G. Moore et al., 1976) has yet been documented on reflection profiles across the floors of theSunda or other modern trenches. I regard sedimentary-melange interpretations of fossil accretionary wedges as in general implausible. A sedimentary melange that is formed in a trench setting must in any case be imbricated into the wedge and tectonized in the process. Broken formation in accretionary wedges wedges is produced by shear related to subduction, not by downslope sliding. The ratios within exposed wedges of polymict melange, broken formation, and strata coherent at outcrop scale vary greatly. The proportion of soft-sediment to brittle deformation also varies greatly. These variables reflect differences in convergence rates, in amount and character of sedimentary strata being added to the wedges, and in positions within wedges. Consolidated sediments and shreds of the subducting lithosphere plate are scraped off against the bottom of the wedge, or can be carried beneath the overriding plate. 3.2.4. Magnitude of subduction. Oceanic lithosphere disappears beneath overriding plates at rates typically near 50 or 100 km/m.y., and most low density material on subducting plates is destined for tectonic accretion against or beneath overriding plates. Something like 3,000 km of Indian Ocean lithosphere have disappeared beneath Sumatra and Java in the 30 million years that the Sunda subduction system has been operating, so much far-travelled material must be incorporated in the accretionary wedge. Far more subduction is recorded in complexes that incorporate the products of series of such subduction systems. 3.3. Fore-arc Basin Between the fore-arc ridge and the shoreline of the Sunda system is the bathymetric and structural fore-arc basin, 150-200 km wide (Figures 2, 3; Beaudry and Moore, 1981; Hamilton, 1979; Karig, Lawrence, Moore, and Curray, 1980). On the landward side of the basin, undeformed lower Miocene and younger strata lap progressively farther landward onto the basement. On the oceanic side, strata become increasingly deformed toward the fore-arc ridge by landward-directed thrusts and folds and by diapiric rise of shale into folds. Basement generally is not defined by reflection work. Depocenters of successively younger stratal packages are displaced landward. I have seen proprietary reflection profiles on which it appears that deep strata now tilted landward on the oceanic side of the Sunda fore-arc basin were deposited as units prograded seaward in deep water before their outer basement was raised to define the basin in which younger strata were 666 90° 100° 110° 10° . I INDIA PL TE-+ __ -+~~ __ -+~~~-+ __ -+~-+~~~_~ __ ~~ 10° ~--4---~--~----+---4----+--~ 1. ORE- RC B SIN 3. Figure 3. Late Cenozoic tectonic elements of the Indonesian region. Adapted from 1981 edition of Hamilton (l978a) and other sources. Plate boundaries are not complete in the poorly understood region between Sulawesi and New Guinea. deposited. Such a relationship is documented by published data for the fore-arc basins of Peru and Chile (Coulbourn and Moberly, 1977) and Luzon (Lewis and Hayes, 1984). The basement beneath the outer parts of both the Sumatra (Kieckhefer et al., 1980) and Java (Naomi Benaron, written commun., 1982) basins has ve10citiestypical of oceanic, not continental, crust, although the thickness of crust of such velocity in the Sumatra sector is much greater than is typical of oceanic lithosphere. I integrate these features for this and other modern fore-arc basins, the 667 120· 130· 140· features of the fore-arc ridge noted previously, and characteristics of some ancient analogs, including those noted in the next section, to infer that the fill of a fore-arc basin is deposited across the boundary between continental crust and a narrow strip of oceanic upper lithosphere that is attached to the front of the overriding continental plate. The basin is formed primarily by the raising of the thin, oceanic leading edge of the overriding plate as accretionary-wedge melange and packets of sediments are stuffed under it. Depth of the basin is augmented by elastic downflexure behind that raised leading edge. The fore-arc ridge is the crest of the accretionary-wedge debris accumulated in snowplow fashion in front of the leading edge. Debris that overtops the leading edge is imbricated gravitationally landward over the front of the basin. As tectonic erosion trims the leading edge of the overriding plate, the fore-arc ridge migrates landward relative 668 100° 110° KT me ange 0° - i \ 10° JU _f--_-+ ____ +-_--+ ___ +-_--+ __ +-_-+ ___ -+-_-'C'-¥-"nL'-t i!!!n",e-,+,-,t Bul ...!!.!..' "fYJ"'-"-"-4.l!..!JJ"-.- ur de , 0 !,~ errldl l 9 Plat~. I . ~-----"-_-~--,-! ~~--;-r Figure 4-. Selected tectonic elements and boundaries of geologic terrains, mostly pre-Neogene, of the Indonesian region. Adapted from Hamilton (l978a, 1979, and this report). K, Cretaceous; KTp, Cretaceous and Paleogene. to that plate and the fore-arc basin is narrowed. (Karig, 1982, disagrees.) 3.3.1. Other fore-arc basins. Fore-arc basins of similar characteristics are common along subduction-system margins of continents and mature island arcs. The ridge and basin may be displayed in bathymetry as well as structure (as in the modern Sunda system, and in the paleobathymetry and longitudinal deposition of the Lower Cretaceous part of the "Valley Facies" of California), or may appear as a bathymetric shelf underlain by structural ridge and basin (as, modern Chile and southern Alaska, and much of the Upper Cretaceous and Paleogene parts of the 669 120' 130' 140' 10' ~4----+----+---~--~----~----~---~ o· 10· "Valley Facies"). Basal strata in such fore-arc basins are generally pelagic sediments and abyssal-fan strata that predate the inauguration of the basins and of the subduction systems that bound them. Exposed basement of the seaward parts of fore-arc basins consists of oceanic crust ~ • .&., Cretaceous California (Hamilton, 1978b) and Cenozoic Luzon (Bachman et al., 1983», which is at least generally of backarc-basin type. A similar origin of the basement of modern basins, including the Sunda system, accords with geophysical data. I believe that the oceanic basement of such a basin commonly forms by migration of an oceanic island arc which collides with a continent or another arc, and that reversals of subduction polarity consequent on such collisions result in inauguration of the subduction systems that form the basins. 670 Where the leading edge of a continental plate is of continental crust, as was the case in the Klamath Mountains region of California and Oregon during Cretaceous time, fore-arc ridge and basin are not present (Hamilton, 1978b). 3.3.2. Lack of shortening. Sunda and other fore-arc basin fills and the slivers of lithosphere beneath them are not commonly shortened compressively across their width, though they are subjected to tectonic erosion and rumpling at their tips and undersides. Thick and undisturbed basin-filling strata can be seen on reflection profiles across many fore-arc basins in Indonesian and other active subduction systems. This lack of deformation disproves the common assumption ~ • .s.., Hutchinson, 1980) that overriding plates are crumpled against subducting ones. Extreme shear imbricates the surficial accretionary wedge pushed in front of an overriding plate, but that plate itself commonly is not shortened. Slight to severe extension, not shortening, occurs across most modern magmatic arcs, perhaps because the steeply sinking subducting slabs displace underlying mantle downward, resulting in extension of the mantle, asthenosphere, and lithosphere above the slabs. 3.3.3. Relation to arc reversal. Subduction systems typically are inaugurated along continental margins by reversal of subduction polarity after an island arc or a continental mass above a subducting plate collides with a previously stable continental margin. The collision enlarges the continent by adding non-subductible crust to it, but convergence continues between the continent so enlarged and the oceanic lithosphere that lay behind the addition and a new subduction system of opposite sense breaks through from the oceanic side of the enlarged continent. The break commonly occurs not at the boundary between thick and thin crust, but within oceanic lithosphere 100 km or so outboard of that boundary. A strip of oceanic lithosphere thus becomes the thin leading part of the newly defined overriding plate. In the case of a reversal following an island-arc collision, this oceanic strip is the youngest part of the back-arc-basin lithosphere formed by a migrating arc, hence is only slightly older than the collision itself. Such an explanation is best documented for the case of latest-Jurassic California (Hamil ton, 1978b) but accords with data from many other arcs, including Sumatra and some others in the Indonesian region (as discussed subsequently). 3.4. High-pressure Metamorphism The only high-pressure metamorphic rocks yet known within Neogene melange in the Sunda system are blocks of garnet amphibolite found on Nias by Moore and Karig (1980), and blocks of glaucophane schist known from Timor and Leti. High- pressure metamorphic rocks, of blueschist and locally eclogite or garnet- amphibolite facies, are widely known in pre-Neogene Phanerozoic subduction complexes in the Indonesian region and elsewhere about the world. The petrology of such rocks requires that they have been metamorphosed mostly at depths of 25- 45 km at relatively low to moderate temperatures, then returned to shallow depths before equilibration of geothermal gradients to normal values for such depths; this apparently occurs as a return-flow byproduct of subduction ~ • .s.., Wang and Shi, 1984). I infer from the geologic relationships of many occurrences around the world that such metamorphic rocks never form within the accretionary wedge--the wedge between trench, fore-arc ridge, and subducting lithosphere. Rather, the high-pressure metamorphic rocks form only where crustal and 671 supracrustal materials have been subducted beneath the overriding plate (Figure 2). Sediment on subducting plates can partly bypass the accretionary wedge and ride far beneath the overriding plate. This is shown directly where anticlinal windows in southern California broadly expose metamorphosed oceanic sedimentary and crustal rocks (termed Pelona, Orocopia, and Rand Schists) beneath lower continental crust. 3.4.1. Geology beneath the Sunda fore-arc basin. Reasoning again by analogy with Mesozoic California and other deeply eroded ancient systems of accretionary wedges and fore-arc basins, I infer that beneath the sub-basin leading edge of the overriding Sunda plate, melange is now being metamorphosed at blueschist facies, and perhaps eclogite facies, and that beneath this is metamorphosing crust of the subducting oceanic lithosphere. The thick zone with oceanic crustal velocities beneath the basin fill, as defined, and puzzled over, by Kieckhefer et ale (1980), perhaps represents a sandwich of thick arc-type ophiolitic basement to the basin, metasedimentary rocks beneath that, and crust of the subducting Indian Ocean plate still deeper. 3.5. Magmatic Arc Volcanoes are now active in a belt above that part of the inclined Benioff zone of mantle earthquakes of the Sunda system whose top is at a depth near 100 km, or whose midplane is near 130 km (Hamilton, 1974-a, 1978a). This magmatic arc changes along strike from continental in Sumatra to transitional in Java to mature oceanic island arc in Bali and Lombok. 3.5.1. Inception of magmatism. Sunda-system volcanism did not begin until well into early Miocene time in Sumatra. Middle Tertiary volcanic rocks are widespread but poorly dated on land, but the inception and subsequent continuity of major silicic magmatism are defined in sections drilled in the Gulf of Thailand by voluminous volcanigenic clays in middle lower Miocene and younger shales. Volcanism, dated by the paleontologic age of intercalated strata in drill holes, was active by late Oligocene time in offshore southern Java; whether this magmatism records the Sunda system or an oceanic island arc that collided with the pre- Sunda-system continent is unclear, but the continuity of late Oligocene and younger strata across central and western Java and the continental shelves to the north and northwest shows that region to have been a coherent part of Southeast Asia by then. The Paleogene of mainland Sumatra, landward of the collided arc noted subsequently, records pre-arc sedimentation across a low and stable landmass from Southeast Asian cratonic sources. 3.5.2. Magma-inflated geanticline. The volcanoes of the magmatic arc rise above a geanticline, within which are most of the exposed pre-Miocene rocks of Java and Sumatra. Presumably this geanticline is a product of magmatic inflation of pre-existing crust. Continental Sumatra has much the higher geanticline of pre-volcanic rocks, and I infer that there a crustal column was heated by intrusions to near-magmatic temperatures, with formation of voluminous migmatites, before much magma reached the surface to form volcanoes (cf. Hamil ton, 1981). 672 3.5.3. Compositional variations. The compositions of the young volcanic rocks vary systematically with the character of the crust through which their magmas have been erupted. The crust of Sumatra was continental by late Paleozoic time, in the southeast, and Mesozoic time, in the northwest (Figure 4), when silicic, radiogenic granites were formed, and likely was so during the Precambrian, although no rocks of that age have been identified. The modern magmatic-arc rocks atop this continental crust are mostly intermediate to silicic in composition. They approximate rhyodacite (granodiorite) in bulk composition; there is little basalt. Lake Toba caldera (Aldiss and Ghazali, 1984; Knight et al., 1986), produced by collapse accompanying voluminous late Pleistocene silicic ignimbritic eruptions, is the largest caldera known anywhere, and is about the same size and shape as the largest upper-crustal granitic pluton yet mapped, the Late Cretaceous Mt. Whitney pluton of the Sierra Nevada of California. In Java, whe" , the pre-Neogene crust is of near-continental thickness but consists of melanges and mafic to intermediate magmatic rocks, young volcanic rocks are mafic to intermediate-mostly pyroxene andesite and high-alumina basalt, with subordinate dacite--and voluminous silicic rocks are lacking. Similar mafic and intermediate rocks characterize the mature oceanic island arc of Bali and Lombok, where exposed rocks are entirely of Neogene age. Farther east, in the Banda Arc sector discussed subsequently, the volcanic arc is younger and consists mostly of more primitive basalts. Comparable transitions, from evolved and silicic magmatic rocks to more primitive and mafic ones, can be seen wherever about the Pacific continuous magmatic arcs cross from continental to oceanic lithosphere. Indian Ocean lithosphere is being subducted beneath all of the Sunda sector, and presumably the mantle proto-magmas generated by subduction-related processes--melting consequent on dehydration of subducted hydrous rocks?--are similar olivine-rich basaltic melts along the entire length of the sector. The volcanic rocks which reach the surface have been profoundly modified by reactions in and with the crust through which they have passed, and by fractionation in and beneath the crust. Even the primitive rocks farther east record magmas equilibrated at shallow depths: no deep-mantle magmas reach the surface without great modification. Volcanoes landward of the main Sunda magmatic belt show the marked but irregular variations, including increased potassium relative to silicon, that characterize eruptions above deep parts of subducting slabs. The new magmatic material added to crust and mantle in and beneath a volcanic arc must be on average very mafic basalt, so fractionates complementary to, and magmatic rocks whose crystallization provided heat sources for, the relatively silicic and alkalic rocks now in the upper crust must have crystallized at depth. Ultramafic, mafic, and calcic arc-magmatic rocks must be voluminous in the lower crust and upper mantle, crystallized in mineral-assemblage facies appropriate for the operant pressures. The Mohorovicic discontinuity beneath an arc likel is rimaril the shallow limit of cr stallization of rocks rich in olivine roxenes and arnet and not a fossil com ositional boundar Hamilton, 1981 and in press, b. Intermediate and mafic magmas residual from such mantle crystallization rise, because of their lower density, through these ultramafic and ec10gitic rocks and spread out low in the crust. There, they are affected by further fractionation, they provide the heat to melt crustal materials and produce secondary magmas, and they mix with such magms. 673 3.5.4. Composition of submarine rocks. The volcanic rocks of modern subaerial transitional and oceanic island arcs are dominantly calc-alkalic basalts and andesites, and the abundantly documented compositions of such rocks are commonly but improperly used as the basis for comparison with ancient island-arc complexes. The great bulk of island-arc material, including most of that seen in ancient arcs accreted to continents, consists of submarine rocks--and these are variably different in composition from subaerial rocks. The submarine rocks tend to be much altered, and igneous petrologists have given them scant attention in most arcs; Gill (1981) does not even mention them in his monograph on andesites. The best major-element data for submarine rocks in an arc that is still active are from the Aleutian Islands (summary by Hamilton, 1963; I know of no more recent synthesis). There, rocks whose relic phenocryst mineralogy shows them to have crystallized with normal calc-alkalic compositions, like those of subaerial rocks, have been variably altered, typically at temperatures of very low greenschist facies, by reaction with sea water. The most conspicuous change in bulk com osition is an erratic increase in sodium and decrease in calcium and the rocks var throu h a com ositional s ectrum from calc-alkalic like the subaerial rocks to spilitic and keratophyric. These latter terms, which belong to rocks enriched in sodium and depleted in calcium, are often applied mistakenly to rocks of normal calc-alkalic or tholeiitic composition metamorphosed at low grade so that their plagioclase is now albite.) Spreading-ridge basalts, formed in water deeper than 2.5 km, show little such reaction (Alt et al., 1986). Experimental thermodynamic evidence indicates that if changes suchas those observed in the submarine arc rocks are indeed due to hydrothermal reactions between seawater and rock, then low ratios of water to rock (and high concentrations of solutes?) are required (ct. Reed, 1983). The likely explanation for the severe alteration of arc assemblages is that violent hydrothermal systems, involving brines concentrated by boiling, are set up as submarine volcanic rocks cool in water shallower than the 2-km critical depth of water, or are induced around plutons emplaced in shallow-water settings. I showed that an ancient island arc now part of western Idaho had compositional variations, from calc-alkalic to spilitic-keratophyric, quantitatively like those of the submarine Aleutian rocks (Hamilton, 1963). Various investigators ~. li., Roobol et al., 1983) have defined similar spectra in ancient submarine arc assemblages but then have argued mistakenly that the divergence from the compositions of modern subaerial volcanoes indicates alkaline magmatic associations. 3.6. Pre-Neogene Tectonics of Sumatra The modern system of subduction of Indian Ocean lithosphere beneath Sumatra was inaugurated only in middle Tertiary time. Much of the older geology records subduction in quite different tectonic systems. Much of Sumatra has been continental at least since late Paleozoic time and belongs to the same system of late Paleozoic and early Mesozoic sutures and magmatic arcs as does the Malay Peninsula. As this system shares many features with the correlative one of northeastern Australia and on-strike east-central New Guinea, I inferred (Hamilton, 1979) that Sumatra was rifted from New Guinea. Medial New Guinea displays the requisite rifted-margin stratal wedge beginning with the Middle Jurassic, and an analogous wedge can be inferred from meager data to be present in Sumatra. Java, on the other hand, has been constructed 674 entirely by post-Jurassic subduction-related processes of magmatism and tectonic accretion. Much reconnaissance information on the pre-Neogene geology of Sumatra has been released since the completion of my 1979 book, as 1 :250,000 photogeologic maps constrained by field traverses and brief rock descriptions. Particularly useful in the present context are the maps by Bennett et al. (1981), Cameron et al. (1982), and Rock et al. (1983). I interpret these worksto Show that the pre-Late-Jurassic rocks o(t'h-e old continental crust are bounded on the southwest by a broad belt of polymict subduction melange and broken formation of late Mesozoic and(?) Paleogene age (Figures 2, 4). This accretionary-wedge complex includes not only the small areas identified as melange and serpentinite by these authors, but also most of the larger terrains they designated as eastern Woyla Group and as Babahrot and Belok Gadang Formations, the brief descriptions of which indicate the presence of widespread broken formation and polymict melange. (Rock-unit names applied in these reports have minimal lithostratigraphic significance.) This broad accretionary-wedge tract lies within the medial part of far northern Sumatra, where its distribution is complicated by the active right-slip Sumatran fault system, but closer to the southwest coast in central Sumatra; southern Sumatra lacks exposures of pre-Neogene rocks in the relevant coastal belt. To the southwest of the broad belt of probable melange is a bel t of volcanic, volcaniclastic, and sedimentary rocks, of island-arc type, which are of Late Jurassic and Early Cretaceous age where dated paleontologically at a few localities, and which were designated as the western Woyla Group by the mappers. I interpret these relationships to indicate that a northward-migrating oceanic island arc collided with the margin of Sumatra, which had been a trailing edge, on which a stable continental shelf had been developing, since its mid- Jurassic separation from New Guinea, in Paleogene time. Convergence of Sumatra and Indian Ocean continued, and the subduction system that is now active broke through south of the continent as enlarged by the collision, leaving a narrow strip of marginal-sea lithosphere, which had been formed behind the advancing are, as the leading edge of the new upper plate. (Bennett et al., 1981, and Rock et al., 1983, recognized the island-arc character of the southwestern rocks but interpreted them in terms quite different from mine.) 3.7. Pre-Neogene Tectonics of Java The modern subduction system in Java was inaugurated no earlier than late Oligocene time. Exposures of pre-Neogene rocks are limited to small areas, in central and southwest Java, of polymict melange of Late Cretaceous and early Paleogene age and of overlying middle or late Eocene through Oligocene quartzose clastic strata and shallow-water carbonates. Much more information has come from the subsurface of Java and, particularly, the Java Sea shelf. Melange of Cretaceous and early Paleogene age dominates the basement in a broad belt trending northeastward from Java across the Java Sea to southeast Borneo, where it is widely exposed (Figure 4; Hamil ton, 1979). This melange may be paired to widespread granitic and volcanic rocks, which have yielded many Cretaceous K-Ar ages, to the northwest in Borneo and the Java Sea basement. During late Paleogene time, western and central Java and the Java Sea were tectonically and magmatically dormant and were fused to the subcontinent that included most of Sumatra and all of the Malay Peninsula. South China Sea lithosphere was then 675 being subducted southward beneath Borneo. If a northward-migrating arc collided with Java in Paleogene time, as might be expected from the interpretation just made for Sumatra, then it now lies offshore in the subsurface, where the upper Oligocene volcanic rocks south of central Java may belong to such an arc. Eastern Java, Bali, Lombok, Sumbawa, and Flores project east of all identified pre-Neogene complexes and expose only Neogene magmatic-arc and sedimentary rocks. 3.8. Neogene Deformation Popular conjecture, residual from otherwise discredited geosynclinal theory, infers great crustal shortening to be a precursor of what is now recognized as arc magmatism. Such deformation is not recorded in the Sunda system or other modern magmatic arcs. In Java, middle Tertiary strata are openly folded; deformation decreases in intensity away from magmatic centers, about which structures tend to be concentric (d. Djuri, 1975), and gravitational spreading of magmatic chambers and edifices is likely a major cause of deformation. In Sumatra, middle Tertiary, pre-magmatic strata within the modern volcanic belt but distant from local centers are subhorizontal or gently dipping, and display normal faulting. Gravitational spreading related to magmatic crustal thickening can be inferred for Sumatra also. Normal faulting, not compressional deformation, is commonly seen in the old parts of mature oceanic island arcs. 3.9. Batholiths, Shortening, and Extension Batholiths are commonly the upper-crustal manifestations of silicic magmatic arcs, and batholiths must now be forming beneath parts of the Sumatran arc. Such arc systems can be associated with a foreland thrust belt, as in the modern central Andes and in the Cretaceous Sierra Nevada of California; with rapid crustal extension, as in modern North Island, New Zealand; or with neutrality, not conspicuously either compressional or extensional, as in modern Sumatra or the Oligocene Sierra Madre Occidental of Mexico. The mantle component of arc magmatism thickens crust, causing its surface to stand higher, whereas extension above a sinking subducting slab thins crust. Apparently these factors of growth and extension can be combined for a broad range of results. Where extension is rapid relative to addition of mantle magma to the crust, continental crust is thinned, and the magmatic arc can lie near sea level, as in New Zealand. Where magmatism is voluminous and extension slow, as in the central Andes, a very high arc is formed; gravitational spreading of .such magmatically thickened crust may be the major cause of foreland thrusting associated with magmatic arcs. Where the ratio of magmatism to extension results in a continental arc of moderate altitude, as in Sumatra and the Sierra Madre Occidental, neither extension nor thrusting is much in evidence. 3.10. Collisions and Continental Shortening Where collisions of light crustal masses are involved, or where a trench rolls back so rapidly that the overriding plate scrapes on a gently dipping subducted plate, severe shortening and major crustal thrusting can result. The preceding arguments against subduction-related compression apply to subduction of oceanic 676 lithosphere with a steep or moderate dip, beneath either a continental margin or an island arc. 3.11. Strike-slip Faulting An active right-slip fault system trends northwestward the entire length of Sumatra (Sumatra fault, Figure 3), within or near the magmatic arc. Total offset and history are not known. Little or no strike-slip faulting occurs in Java, which trends eastward from the southeast end of Sumatra. The convergence between Indian Ocean and the Sunda landmass is approximately north-south, and the obliquity of this motion in the Sumatra sector is partly resolved by the northwestward slip of a coastal strip. This strip likely is the top of a wedge bounded on the bottom by the northeast-dipping subducting Indian Ocean plate, and on the northeast by the subvertical magmatically-softened zone through which arc magmas rise toward the surface. 4. BANDA ARC The south limb of the Banda Arc continues the Java trend eastward for about 1200 km, and thence curves through a tightening arc northward and back to the west for a short distance. Trench, fore-arc ridge and basin, and volcanic arc are concentric around the tight curve. A well-defined Benioff zone of earthquakes dips northward deep into the mantle from the accretionary wedge of the Sunda Arc and the south limb of the Banda Arc. The seismic zone curves in the east, concentric to the bathymetric features, to define a spoon-shaped zone that plunges gently westward but that can be traced unambiguously only to a little north of the geometric axis of the Banda Arc. The Banda sector illustrates the collision of a migrating, lengthening arc with a continent and the reversal of subduction polarity following such collision. Various conclusions of my 1979 and earlier reports were replicated from marine-geophysical data, including much not incorporated in my work, by Bowin et al. (1980), Jacobson et ale (1981), Johnston and Bowin (1981), Silver, Reed et al-:l(1983), and Von der Borch(1979). (Bowin et al. interpreted the on-land geologyin terms quite different from mine.) Silver et ~. (1985) presented important new information from the Banda Sea. McCaffrey et al. (1985) presented much new analysis of seismicity. Audley-Charles (1986), Audley-Charles et al. (1981), and Barber (1981) continued the slow progression of their earlier papers from stabilist toward plate-tectonic interpretations of the islands but still did not incorporate either marine geophysical data or actualistic analogy with other arcs. 4.1. Trench Whereas the trench in the Sunda sector overlies oceanic lithosphere, the trench in the Banda sector overlies continental crust around the entire curve of the arc. The continuity of the trench and its distinctive tectonic morphology around the arc are shown by scores of reflection profiles, many of which were published in my monograph and in the geophysical references just cited. The shallow bathymetric trench atop continental crust marks the dihedral angle, partly covered by trench turbidites, between shallow-water strata bowed down from the Australia-Arafura- New Guinea continental shelf on the outside, and the toe of the accretionary 677 wedge on the inside. Continental crust is demonstrated by refraction data to extend beneath the accretionary wedge at least to the inner edge of the fore-arc ridge (Bowin et al., 1980; Jacobson et al., 1979). McCaffrey et al. (1985) inferred that the thin leading edge of the continent has been subducted to a depth of 150 km in the Timor sector, and that still deeper subducted oceanic lithosphere is detached and sinking independently. 4.2. Fore-arc Ridge (including Timor) The top of the accretionary wedge is wholly submarine from Java to Flores, but where it stands upon continental crust it forms the large, high island of Timor, lower and smaller islands around the tight eastern curve of the Banda Arc, and large, high Seram on the north limb of the system. Continuity of the wedge around the arc as a thick aggregate of low-density material is indicated by a continuous gravity anomaly (Bowin!:! al., 1980). Descriptions of onshore geology that postdate my monograph include those for part of Timor by Rosidi et al. (1979), for the Tanimbar Islands by Sukardi and Sutrisno (1981) and the Kai Islands by Achdan and Turkandi (1982), for localities in Seram by Audley Charles et al. (1981), and for Buru by Tjokrosapoetro et al. (1981). The wedge consists of polymict melange and broken formations imbricated, with generally arcward dips, with variably coherent strata that include strata from the continental shelf onto which the wedge has been ramped, strata deposited atop the wedge, abyssal pelagic sediments, and slices and fragments of both ophiolitic and continental crystalline rocks. Fore-arc-basin materials may have been imbricated into the wedge after tectonic removal of their overriding- plate basement. Quaternary reefs have been elevated as high as 1000 m above sea level as the top of the wedge has been raised both by thickening of the wedge by accretion and imbrication and by ramping farther onto continental crust. Berry (1981, and unpub. thesis) and Berry and Grady (1981) described metamorphism of sedimentary rocks that decreases from uppermost amphibolite facies (with formation of migmatite) to greenschist facies away from an ophiolite mass at the north edge of central Timor. Potassium-argon ages of h~rnblende show the metamorphism to be of about late middle Miocene age. I infer from the relationships mapped by Berry that the temperature of metamorphism decreased downward beneath the ophiolite sheet, which I regard as the hot leading edge of the onramping island arc. (Berry and Grady inferred vertical or strike-slip tectonics and suggested no heat source.) Farther west on the north coast of Timor, upper Miocene tholeiitic and calc-alkalic basalt are thrust southward onto the wedge (Abbott and Chamalaun, 1981); again, I infer onramping of the advancing arc. Onramping of advancing arcs onto continents is a common feature of collisions, and so is high-temperature metamorphism beneath hot ophiolite sheets of the ramps. Such onramping is in the same geometric sense as is the subduction, and is in the sense opposite to that of the hypothetical and implausible process of "obduction". Dorian et al. (1986) speculated on the mineral-resource potential of several Indonesian islands on the basis of comparisons with several of the United States which they thought might be analogous. They assumed that Timor (which consists of Neogene accretionary-wedge melange) is analogous to the State of Arizona (Proterozoic continental basement, cratonic Paleozoic and lower Mesozoic strata, 678 late Mesozoic and Cenozoic granites and rhyolites), and on the basis of this gross misconception they suggested that Timor has great undiscovered mineral wealth. Perhaps it has--but there could be no more foolish basis for thinking so. 4.3. Fore-arc Basin The fore-arc basin is continuous (except at Sumba) around the Banda Arc. Undeformed basin strata lap onto the fore-arc ridge on the outside of the basin, and grade into volcaniclastic aprons of the magmatic arc on the inside. The bathymetric basin deepens symmetrically along both limbs of the Banda Arc toward the axis of its tight horseshoe curve to define the Weber Deep, the depth of which reaches 7.5 km precisely at that axis. I interpret the basin as formed by the elastic deflection of the thin leading part of the overriding lithosphere as its edge has been ramped up by the stuffing beneath it both of accretionary-wedge melange and of continental crust. This depression is focussed at the deepest sector from three sides. The basin is markedly narrower along northern and eastern Timor than elsewhere around the south limb and eastern curve of the Banda Arc. I infer tectonic erosion of the leading edge of the overriding plate, and the imbrication into the Timor wedge of what were strata deposited on that leading edge. There is no suggestion on reflection profiles of subduction within this or other sectors of the basin; narrowing (or, in the east, deepening) by subduction cannot be proposed. The concentricity of the Banda Arc deteriorates in the Buru-western Seram sector of the north limb, where no fore-arc basin is present inward from the fore- arc ridge. Islands of Pliocene volcanic rocks, which presumably represent the extinct magmatic arc and were erupted through silicic continental rocks (Abbott and Chamalaun, 1981), are separated from the ridge only by narrow straits. Tectonic erosion of the overlying plate may here also be part of the explanation. The large, puzzling island of Sumba rises within the trend of what is otherwise the fore-arc basin, and its almost undeformed Miocene to Quaternary strata are continuous with those of the basin. The small areas of pre-Neogene materials, including late Mesozoic and Paleogene sedimentary, low-grade metamorphic, and igneous rocks remain poorly known. I (Hamil ton, 1979) concluded that the pre-Neogene complex was a minicontinental fragment rifted either from the Java Sea shelf, which is known from well data to consist of similar materials, and in that case to have been carried with the outward-migrating arc, or from northwest Australia, in which case it would have been one of the number of fragments isolated during Mesozoic rifting and have been swept up in the advancing arc. Chamalaun et al. (1982) presented a more detailed summary of Sumba geology and replicated my conclusions. Silver, Reed, et al. (1983) suggested that the island may be part of the fore-arc basin raised by subduction beneath it of one of the crustal fragments in front of Australia. 4.4. Magmatic Arc Although the magmatic arc is continuous around the south limb and eastern curve of the Banda Arc, its history varies systematically with position. The width and volume of the magmatic edifice decrease eastward along the south limb of the arc and correspond to a decreasing age of inception of magmatism, from early Miocene in the west to Pliocene in the east. (Data postdating my monograph include those of Abbott and Chamalaun, 1981, Foden, 1983, Foden and Varne, 679 1980, Suwarna et al., 1981, and Varne, 1985.) Around the sharp eastern curve, the magmatic arc Tsrepresented only by the small, active-volcanic islands atop a narrower and less-continuous ridge; the entire edifice may be Pliocene and Quaternary, and the rocks are little evolved petrologically. Volcanic rocks on the short, irregular north limb of the arc are of Pliocene age, but here tectonic relationships are poorly understood. Volcanoes are active all along the south limb of the magmatic arc and around the eastern curve of the arc, except for a length of about 500 km, to the north and northeast of eastern Timor, and along the short north limb in the Buru-western Seram sector, in both of which activity ended in Pliocene time. 4.5. Arc Reversal Two sectors, each about 500 km long, of the south limb of the Banda Arc are now marked by trenches, the tectonic geometry of which indicates subduction relatively southward, at the north base of the volcanic arc. This polarity is opposite to that of the main Banda system. I (Hamilton, 1978a, 1979) identified the trenches on reflection profiles and argued that arc reversal following collision of arc with continent was displayed. McCaffrey and Nabelek (1984), Silver et al. (1986), and Silver, Reed, et al. (1983) further defined the character and extent of the trenches and accretionary wedge from reflection profiles, sidescan mapping, seismicity, and other data, and supported my general conclusion. The eastern of these new trenches is north of central and eastern Timor, and coincides with that part of the volcanic arc in which magmatism ceased in late Pliocene time. The western of the new trenches lies north of Flores, Sumbawa, and Lombok, where magmatism apparently belonging to the north-dipping subduction system is still active, and also lies north of the anomalous Sumba region noted previously. 4.6. Banda Sea The small Banda Sea is enclosed by the Banda Arc, and the evolution Qf the arc can be understood only in terms of the sea. The sea consists of the oceanic North and South Banda Basins and an intervening zone of submarine ridges. The age of formation of the oceanic crust of the two basins is not yet constrained by drilling. I suggested (Hamilton, 1979) that the basins formed behind a migrating Banda Arc, and hence are of Cenozoic age. Bowin et al. (1980), Pigram and Panggabean (1983), and Silver et al. (1985) by contrast all regarded both basins as trapped bits of Mesozoic lithosphere. Their interpretation appears to me to be plausible for parts of the North Banda Basin but not for the South Banda Basin, for the Banda Arc has lengthened during late Neogene time along the latter basin. Fragments of continental lithosphere form islands and variably submerged platforms around part of the Banda Sea--Buton in the west, Banggai-Sula in the northwest, and Buru-Ambon-western Seram in north-center (Hamilton, 1978a, 1979; Pig ram and Panggabean, 1983; Silver, McCaffrey, et al., 1983). Several shallow, discontinuous submarine ridges trend east-northeast between the Buton fragment and Seram, and the clearly continuous parts of the Banda magmatic arc and fore-arc basin and ridge end at the projection of this zone of ridges. Silver et al. (1985) sampled two of the ridges by dredging and found late Neogene potassic andesites and older low-grade metasedimentary rocks of continental types, including argillite, marble, phyllite, slate, and quartzite. As similar continental 680 rock types are known in the continental fragments around the Banda Sea and also in the Paleozoic basement of northwestern New Guinea, and as the Banggai-Sula minicontinent has Mesozoic strata of New Guinea facies, Silver et al. (1985) concluded, and I agree, that these submarine ridges also are fragments of crust from New Guinea, separated from each other by oblique rifts. 4.7. Interpretation The age of inception of the Banda magmatic arc becomes progressively younger eastward along the arc, from early Miocene to Pliocene or even Quaternary: the arc has lengthened with time. The collision of the arc with the Australia-New Guinea continent also has progressed eastward with time, occurring earlier at Timor than around the axis of the tight curve in the east. Timor has not slid past Australia on strike-slip faults but has remained attached to it since the collision in that sector; Banda Sea lithosphere is now being subducted southward beneath the expanded arc at a new trench, even as sUbduction at the axis of curvature of the arc is relatively westward. Such relationships to me indicate that the crust of the South Banda Basin has formed by spreading behind a rapidly-migrating Banda Arc, or has been plated out by the fast-migrating arc itself. The Banda Sea does not represent an internally rigid plate neatly pre-shaped to slide into the Arafura concavity between Australia and New Guinea; rather, the Banda plate expanded as needed to fill a concavity which likely has itself changed shape as Jurassic oceanic crust attached to the continent sank in front of the Banda plate. This much of the story is analogous to that in my 1979 book; but clearly I erred there in picturing the entire Banda Arc and Banda Sea as a simple paired migrating arc and extensional back-arc basin. The north limb of the arc (Seram and Buru), the North Banda Basin, and the submarine Buton-Seram ridges require much more complex explanations. A viable solution must incorporate rapid northward motion of New Guinea and westward motion of Pacific plates, and probably southward motion of the Sunda system, and must account for the bewildering array of diversely oriented tectonic elements north of the Banda Sea. Simple westward transport of fragments from New Guinea on left-slip faults can not explain the relationships, and a combination of Mesozoic rifting of the fragments followed by Cenozoic strike slip may be indicated. 4.8. Caribbean and Scotia Arcs Many analogs to the Banda Arc can be found in the Caribbean and Scotia regions. These can in my view be explained in terms of eastward-migrating oceanic arcs that collided with the Pacific sides of Central and South America and West Antarctica in late Mesozoic time but that continued to migrate through the oceanic gaps between those landmasses, beaching arc mater ial against north and south sides progressively eastward with time. The initial frontal collisions were followed by reversals of subduction polarity which inaugurated the Andean systems that have operated subsequently along the continental margins. (Regional experts have quite different explanations.) 5. SULAWESI The complex K-shaped island of Sulawesi has eastern arms of Cretaceous(?) and 681 Neogene accretionary-wedge materials, and western arms with a Cretaceous accretionary-wedge basement, a Paleogene platform sequence, and Neogene magmatic-arc rocks. Comprehension of onshore Sulawesi has advanced only modestly since my 1979 synthesis. Several new 1 :250,000 photogeolog1c maps, variably constrained by field reconnaissance, add little that is new at the scale of my treatment. Much important new information has been added offshore. Silver, McCaffrey, and Smith (1983) and SlIver, McCaffrey, et al. (1983) presented, in addition to geologic and geophysical details in east andCentral onshore Sulawesi, much new reflection profiling east and north of Sulawesi. Silver and his associates corroborated my interpretation of clockwise rotation of the East and North Arms relative to the South(west) and Southeast Arms and the Celebes Sea; but they made important corrections to my inferences regarding small-plate boundaries east of Sulawesi. I erred in inferring an inactive trench to be present along the northeast foot of the Buton minicontinent, where a strike-slip fault may be present instead; and the north base of the Banggai-Sulu minicontinent apparently is marked by a north-dipping contact beneath southward-spreading Molucca Sea accretionary-wedge melange. The complex history of Sulawesi spans Cretaceous and Cenozoic time. The South Arm and west-central Sulawesi share a Cretaceous accretionary-wedge complex and overlying, little-deformed Paleogene terrigenous and carbonate platform strata with southeast Borneo, the eastern Java Sea shelf, and much of Java. The East and Southeast Arms are dominated by subduction complexes. Accretion of an eastern belt of nonmetamorphosed to low-grade-metamorphic melange, thick intercalated sheets and broad anticlines of ophiolite, and imbricated strata is dated as of Neogene age, and apparently is paired to the widespread middle Miocene to Pliocene or early Pleistocene magmatic-arc rocks, mostly basalt and andesite, of the South and North Arms. Cretaceous magmatic- arc rocks are widespread in southwest Borneo and in the subsurface of the northwestern Java Sea, and presumably are paired to subduction recorded by the Cretaceous accretionary-wedge complexes exposed in the South Arm of Sulawesi and southeast Borneo and known in the subsurface of the southeastern Java Sea. The western belt of the accretionary terrain of the eastern arms of Sulawesi consists of undated melange, including widespread metamorphic rocks of blueschist, and locally eclogite, facies, and likely is paired to the magmatic-arc rocks just noted and hence of Cretaceous age; dating of the metamorphic rocks is needed. Middle Tertiary rifting of Sulawesi relatively southeastward away from Borneo opened the oceanic North and South Makassar Basins, and at about the same time the eastern and western arms of Sulawesi spread apart. Neogene deformation has included the clockwise rotation of the North and East Arms, bounded by arcuate left-slip faults against the South and Southeast Arms and by the North Sulawesi Trench against the Celebes Sea to the north. No Paleogene magmatic-arc rocks that might have formed paired to eastern Sulawesi accretionary-wedge materials are known anywhere in the relevant region, either exposed or in the subsurface. The lower Eocene through Oligocene of the Java Sea, eastern Borneo, and western Sulawesi is a stable-platform section of carbonate rocks and terrigenous clastic strata. I infer that there is no Paleogene melange in Sulawesi. A very broad accretionary-wedge terrain records Paleogene subduction of South China Sea lithosphere in the direction now southeastward under northwestern Borneo, the opposite side of the Sulawesi- Borneo continental mass: there was subduction beneath only one side of the plate 682 at a time. Magmatic-arc rocks paired to this northwest Borneo wedge may be widespread in little-known interior Borneo. 6. MOLUCCA COLLISION ZONE The Molucca Sea is the site of the southward-progressing collision between the east-facing Sangihe island arc and the west-facing Halmahera arc. The suture zone is fully closed in the north, where it is exposed on land in Mindanao. In the central sector, in the northern Molucca Sea region, the accretionary wedges of the two arcs have been joined by collision and have been thickened to at least 15 km, and the composite surface raised to near sea level, in the medial zone. The overthickened composite wedge has flowed gravitationally across the inward- facing trenches and onto the arcs on both sides so that surficial thrusting of melange has the sense opposite to subduction. Following the collision, arc magmatism ceased in this central sector, and subduction polarity of the Sangihe arc was reversed. In the southern Molucca Sea region, the two accretionary wedges have met in the center, but subduction and arc magmatism are still active in their pre-collision sense. This system of colliding arcs is so important for comprehension of plate behavior that it has been studied extensively, particularly by Eli Silver and his associates, since my early work on it. Important information on this collision system that postdates my monograph, and that documents the conclusions just summarized, was reported by Cardwell et ale (1980), McCaffrey (1982), McCaffrey, Silver, and Raitt (1980), G. F. Moore and Silver (1982), and Silver, McCaffrey, et ale (1983). The melange and its ophiolitic components in the composite accretionary wedge, as exposed on the Talaud Islands, were described by Evans et al. (1983), G. F. Moore et ale (1981), and Sukamto (1980). Weissel (1980) identified sea-floor-spreading magnetic anomalies of the southwestern Celebes Basin, the marginal basin opened behind the Sangihe arc but now being subducted beneath northern Sulawesi, southwestern Mindanao, and the northern Sangihe Arc, as probably Eocene in age. 6.1. Subduction of Molucca Sea Plate The Molucca Sea plate is being subducted simultaneously westward beneath the Sangihe arc and eastward beneath the Halmahera arc. A well-defined Benioff seismic zone dips westward beneath the Sangihe arc to a depth of about 650 km beneath the Celebes Sea, and another zone dips eastward beneath Halmahera to a depth of about 250 km. The active volcanoes of both arcs lie about 100 km above the tops of the respective seismic zones, which merge beneath the Molucca Sea composite melange wedge. This two-sided subduction can not be explained in terms of subduction as a process of injection down fixed slots, and requires that the subducting Molucca Sea plate have fallen away on both sides as overriding plates have advanced over it. 6.2. Halmahera The K-shaped island of Halmahera remains poorly understood. Photogeologic maps constrained by sparse field traverses have been released for all of the island (Apandi and Sudana, 1980; Supriatna, 1980; Yasin, 1980), and the geology has been 683 discussed by Soeria-A tmadja (1981) and Sukamto et ale (1981). Morris et ale (1982) presented new data on the volcanic rocks of theHalmahera arc (andMorrice et al., 1983, did the same for the Sangihe arc). The young arc volcanoes of western Halmahera, associated with the seismic zone dipping eastward from the Molucca Sea, stand atop the North(west) Arm, form small islands rising from the shelf west of central Halmahera, and rise from Bat jan island west of the South(west) Arm. The volcanoes stand on middle and upper Tertiary arc-volcanic, volcaniclastic, and carbonate materials. The Northeast and Southeast arms consist of voluminous, undated ultramafic rocks with subordinate mafic rocks; Cretaceous abyssal- pelagic strata; and lower and middle Tertiary terrigenous, volcanic, and carbonate materials, much of them of shallow-water setting. These eastern materials are imbricated together with dominantly westward dips and are overlain by openly folded Neogene strata. I presume that the eastern arms consist largely of accretionary-wedge materials aggregated in Paleogene time in response to subduction relatively westward of Cretaceous oceanic lithosphere, but there is too little documentation of structural and stratigraphic relationships and of melange characteristics to permit confident interpretation. Likely this accretionary wedge is paired to the older magmatic-arc rocks of the western arms. The South Arm consists of Tertiary volcanic rocks in the west and moderately to steeply dipping Neogene clastic and carbonate rocks, of uncertain tectonic setting, in the east. 6.3. Bat jan The island of Bat jan, west of southern Halmahera, exposes a tract of crystalline rocks that includes aluminous schists and gneisses and hence might represent a small continental fragment, as I suggested in 1979. Meager new data (Sukamto et al., 1981; Yasin, 1980), however, indicate the crystalline rocks to be dominated by mafic schists and gneisses and to include ultramafic rocks, and hence to be more likely of oceanic than continental derivation. 6.4. Southern Complexes Relationships around the southern Molucca Sea region are exceedingly complex and still poorly understood. The accretionary wedge exposed in the Southeast Arm of Halmahera projects southeastward along strike as a broad submarine ridge and emerges as the Paleogene melange of Waigeo, off the northwest tip of New Guinea. The Banggai-Sula minicontinent forms a narrow, west-trending ridge which may be bounded on the north by a left-slip fault that has been in part overtopped by melange spreading from the composite Molucca Sea accretionary wedge. The palinspastic analyses attempted for this region by Cardwell et al. (I980), Hamil ton (I979), G. F. Moore and Silver (I982), and Silver et ale (I 985f are highly speculative. -- 7. AGGREGATION OF THE SOUTHERN PHILIPPINES The Philippine Islands are a collage of variably collided, reversed, oroclinally twisted, and magmatically overprinted components of island arcs--magmatic arcs, accretionary wedges, large and small ophiolitic masses, and sedimentary assemblages. The collided Sangihe and Halmahera arcs and intervening wedge come ashore in southern Mindanao, where suturing was completed during middle 684 Tertiary time (Hawkins et al., 1985), and their products are being overprinted by arc magmatism paired to subduction relatively eastward from the Cotabato Trench, which was inaugurated by arc reversal following the collision. The Benioff zone inclined westward from the Philippine Trench extends only to shallow depth and has no obviously associated arc volcanoes south of Leyte (Cardwell et al., 1983); kinematic relationships of this Philippine Trench system to those of the rest of Mindanao are not yet clear. Farther west, the recently inactivated, northwest-facing Sulu island arc comes ashore as the Zamboanga Peninsula of western Mindanao, and the slow-subduction, southwest-facing Negros arc system apparently is closing southward against the Sulu-Zamboanga arc, colliding with its own earlier projection. Still farther west, the Palawan island arc, beneath which South China Sea lithosphere was subducted during middle Tertiary time, comes ashore in the west-central Philippines, much complicated by a collision with a North Palawan minicontinent rifted from China as the South China Sea was opened. Holloway (1982), McCabe et al. (1982, 1985), and Sarewitz and Karig (1986) presented much new data on the tectonic history and collision of that minicontinent and of other problems of tectonic accretion in the southern Philippines. Six different middle and late Cenozoic subduction systems are thus clearly recorded in the southern Philippines aggregate. As arc-type materials are as old as Cretaceous in the southern Philippines, many additional complexities have yet to be understood. This complex collage likely will at some future time be accreted to mainland Asia. Broad tracts of island-arc materials now parts of orogenic belts within, or along the margins of, continents probably have generally had similarly complex histories of collisions, reversals, and oroclinal deformation. Composite arc terrains likely commonly undergo much offshore accretion. Thus, a broad terrain of composite island arcs and related materials, mostly of Jurassic age, in the Klamath Mountains of California and Oregon was probably largely aggregated offshore, and collided with North America toward the end of Jurassic time (Hamilton, 1978b). Explanations in terms of sequential accretion of simple, narrow arcs probably are less commonly viable. 8. NEW GUINEA AND THE WESTERN MELANESIAN SEAS AND ARCS 8.1. Western New Guinea Much of western Irian Jaya (Indonesian New Guinea) was mapped in field and air- photo reconnaissance by geologists of the Australian Bureau of Mineral Resources, Geology and Geophysics, cooperatively with geologists of the Indonesian Geological Research and Development Centre, from 1978 to 1982. The resulting 1:250,000 maps are available in preliminary form from the Centre. Published summary and interpretive reports include those by Dow and Hartono (1982), Dow and Sukamto (1984), Pieters et al. (1983), Pigram and Panggabean (1984), and Pig ram et~. (1982, 1983). Knowledge of stratigraphy, structure of the foreland thrust belt, and lithologies and ages of the crystalline terrains have been greatly increased. A great left-slip "Sorong fault" has long been assumed to trend westward near the north edge of New Guinea. Sabins (1983) documented active strike-slip faulting with radar imagery showing offset valleys and a shutter ridge along the 685 straight fault trace across north-central Vogelkop (Jazirah Doberai). Total displacement remains unknown. The fault apparently follows in a general way the suture zone, marked by accretionary-wedge melange, between an oceanic island arc (parts of which are exposed in the Tamrau Mountains of northern Vokelkop and in Yapen Island) which advanced southward until it collided with New Guinea in middle Tertiary time. The new work by the Australian-Indonesian group supports my 1979 suggestions that the Paleozoic basement rocks of Vogelkop represent a fragment derived from farther east, and that Geelvink (Sarera) Bay is floored by oceanic crust and is trapped between that minicontinent, the Yapen sector of the northern island arc, and mainland New Guinea. 8.2. Eastern New Guinea Relatively little information has come from onshore Papua New Guinea since completion of my monograph. Brown et al. (1980) replicated my general conclusions regarding the domination of New Guinea tectonics by the middle Tertiary collision of a southward-advancing island arc against a previously rifted continental margin. Connelly (1979) modelled the configuration of the collision complex in the Papuan Peninsula from geophysical data. 8.2.1. Crustal extension. The eastward-widening Woodlark Basin is the product of active sea-floor spreading whereby the continental shelf and submarine ridges east of the Papuan Peninsula are being rotated apart (Hamilton, 1979; Weissel et al., 1982). The apex of the spreading triangle is on the continental shelf in the area of the D'Entrecasteaux Islands and is marked by extensional seismicity (Weissel et al., 1982) and by active bimodal volcanism--alkaline basalt plus trachyte and peralkaline rhyolite-of a type common in extended terrains. The D'Entrecasteaux Islands display gneiss domes, for which recent descriptions by OIlier and Pain (1980, 1981) can be integrated with earlier accounts by Davies (1973) and Davies and Ives (1965) into an extensional-tectonics synthesis based on analogy with the Basin and Range Province of the western United States. Four gneiss domes on the three large islands rise 1000-2500 m above the surrounding lowlands and continental shelf. The diversely oriented elliptical domes are 20-30 km long and are defined by concordant carapaces of gently dipping gneisses that are at least in part mylonitic and that that both cap and underlie gneisses of variable orientations. The pristine topography of several of the domes indicates that they have only recently been uncovered. Dome and flanking gneisses are largely of upper amphibolite facies, but locally of garnet-amphibolite, granulite, and eclogite facies; greenschist-facies schists are common in the southeasternmost dome (in its carapace?); granodiorite is widely present. Large masses of ultramafic rocks are limited to upper-plate positions. Reconnaissance data do not permit confident interpretation, but high-temperature metamorphism beneath an onramping ophiolite sheet, with high-pressure rocks also present within the subduction-related complex, might be inferred. These relatively deep-seated rocks were recently at high temperature, for seven Pliocene K-Ar ages have been determined for the granitic rocks; the complexes may have risen 15 km within the last two million years. OIlier and Pain suggested that the domes are magma-cored diapirs now extruding at the surface, whereas Davies inferred them to be upfolds of gneisses beneath overthrust ultramafic rocks. I have argued (Hamilton, in press, a) that the structural geometry of the Basin and Range Province of the western United States indicates its middle crust 686 to be extended as subhorizontal lenses slide apart along zones of ductile shear. This model may be applicable to the D'Entrecasteaux extension. The gneiss domes may be the upper parts of lenses newly exposed at the surface as initially higher lenses--now the crystalline rocks of the surrounding lowlands--have slid completely off of them. In Basin-and-Range terminology, the domal carapaces are the ductile-fault zones beneath detachment faults, and the domes are "metamorphic core complexes." Extension at depth likely has been accomplished by more pervasive ductile flattening, whereas upper-crustal rocks likely were rotated down on gently dipping normal faults atop the detachment faults. Such upper-crustal rocks have been largely removed by a· combination of erosion and tectonic denudation. If the Basin-Range analogy is indeed applicable, then pre- Quaternary supracrustal rocks in the islands sould be rotated down to moderate dips against detachment faults atop both the domes and the structurally overlying crystalline rocks; this prediction cannot be tested with published data. 8.3. Offshore Western Melanesia A number of reports on the offshore region around eastern New Guinea have been published since completion of my work. Davies et al. (1984) presented a tectonic synthesis very similar to mine for the Solomor1Seii region. Weissel and Watts (1979) interpreted the magnetic anomalies of the Coral Sea in terms of Paleocene and early Eocene sea-floor spreading. Drummond ~ al. (I979) argued (I think mistakenly) against the reality of my buried "Port Moresby Trench", along the south side of the Papuan Peninsula, on the basis of their seismic-refraction data. Weissel et al. (I982) presented new data on the Woodlark spreading system, which extends eastward from the Papuan Peninsula to the Solomon Trench. Bracey (1983) synthesized geophysical data from the Caroline Sea. Fornari et al. (1979) showed that the south part of the west boundary of the Caroline Sea Plate is a zone of southward-increasing sea-floor spreading. Hegarty et al. (1983) added evidence supporting my interpretation of the east boundary of the Caroline Sea Plate in the Mussau Trench sector, while reaching new interpretations regarding the north part of the east boundary. 9. COLLISIONS AND SUBDUCTION Examples such as those discussed in this essay make it obvious that long- continuing, steady-state subduction systems are atypical, that complex sequences of collision, aggregation, reversal, rifting, and internal deformation are the rule, and that aggregates of collided bits can be assembled far from their final resting places. Histories and kinematics can vary dramatically along strike in continuous complexes. Plate convergence commonly is oblique, not orthogonal, to structural trends. Collisions and reversals progress along strike with time, and strike-slip and oroclinal deformation are common. Collisions do not occur between neatly matched shapes; irregular masses meet, and highly variable deformation occurs before they are fully jostled together. Large plates commonly continue to converge after a collision, and the result is the inauguration of a new subduction system outboard of the new aggregate; often this represents a reversal of polarity of subduction as well as a jump in position. Subduction of oceanic lithosphere beneath a continental plate commonly begins as a consequence of a plate collision. Convergence between megaplates 687 continues but the light crust on the subducting plate is too low in density to be subducted, so a new subduction system breaks through outboard of the continental plate as enlarged by the subduction. Such post-collision reversals are now underway in the Timor and Molucca regions; dozens of others are recorded in Circum-Pacific geology. The Solomon-Admiralty arc complex displays two reversals, one of which presently is progressing along strike as the arc slides past a trench-trench-transform triple junction (Hamilton, 1979). Major plate-convergence complexes record convergence at rates on the order of 10 cm/yr, or 100 km/m.y. Although such motion can be distributed complexly across several subduction systems which vary greatly along strike and are linked by diverse boundaries of other types, large motions and great complexity are the common case. 10. THE FUTURE If present gross plate motions continue for another 50 million years or so, the complex continental scraps, composite island arcs, and accretionary wedges of much of the Indonesian-Philippine-northern Melanesian region likely will be squashed between Australia and Asia. The result will be another broad orogenic terrain akin to those we term Tethyan, Hercynian, Caledonian, and Pan-African. lI. TETHYAN TECTONICS Despite the broad acceptance of the reality of plate-tectonic processes, little awareness of the characteristics and behavior of active tectonic systems has gone into many paleotectonic analyses of Tethyan terrains, or for that matter of most orogenic terrains now parts of continents. A notable exception is the analysis of the Tyrrhenian Sea region in terms of a migrating-arc model by Malinverno and Ryan (I 986). lI.I. Paleotectonic Synthesis Comprehension of modern plate interactions provides a framework within which can be seen relationships between correlative magmatic, tectonic, and stratigraphic features across broad, complex belts. Actualistic interpretations of ancient tectonic and magmatic terrains can be tested by examination of their implicit predictions regarding relationships with associated features. I attempted to combine these approaches in an analysis of the Mesozoic tectonics of the western United States (Hamilton, 1978b). Silver and Smith (I983) pointed to various parallels between the modern Indonesian and Mesozoic Cordilleran regions. Dewey (I980, and many other papers) is among other geologists who use modern analogs and dimensions in paleotectonic analysis. Too much other published speculation incorporates many misconceptions and neglects to test hypotheses in terms of such implicitly predicted relationships as the pairing of subduction and arc magmatism. Many plate-tectonic reconstructions postulate long-continuing subduction beneath opposite sides of a moving and internally rigid plate ~ • .8.., Pindell and Barrett, in press). No such reconstruction can be valid. Actualistic arguments aside, no slab could sink beneath a retreating plate, for that would require the slab 688 to scoop sub-asthenosphere mantle upward out of the way. Subduction occurs beneath one side of a plate at a time. (A contrary argument can be made for the present Caribbean, where subduction is relatively eastward beneath Central America yet relatively westward beneath the Antilles; but the Caribbean is not a rigid plate, for it is spreading and deforming internally. Another counter- argument can be based on southern Mindanao, with the short Cotabato, or West Sangihe, Trench on one side and the Philippine Trench on the other; but the Philippine Trench is not in this sector paired to a magmatic arc and its possible Benioff zone is poorly defined, and plate boundaries not yet understood may intervene between the trenches in this exceedingly complex region.) The hypothetical process of "obduction", whereby a great sheet of oceanic lithosphere is split from a subducting slab and shoved high atop the thick crust of an overriding island-arc or continental plate, has been invoked by many writers. "Obduction" defies mechanical analysis and has yet to be proved to have operated anywhere. Examples attributed to it represent instead ramping of advancing arcs onto subducting plates--thrusting in the sense of subduction, not opposite to it. 11.2. Data Sources He who would develop actualistic paleotectonic models for complex terrains will commonly find it necessary to work from the primary geologic descriptions of his study region. Most secondary tectonic literature is so distant from the data bases-it describes concepts rather than rocks and structures-as to be of minimal use in plate-tectonic synthesis. Fundamental assumptions need evaluation. Thus, in the important sector of India, the Himalaya, and Tibet, investigator after investigator continues to base tectonic and geophysical models on the assumption (which is contrary to much of what is known of the actual geology) that the northern extension of India has slid beneath Tibet. In a mobilistic, plate-tectonic interpretation, one must keep track of constantly changing settings and histories of sedimentation and deformation, and the generalizations by prior investigators of other viewpoints usually obscure critical relationships needed to evaluate contrary interpretations. Conventional presentations of the structural history of what was, say, a lengthening, migrating, colliding, and reversing arc system may carry little of the information needed to develop such a history. I think that the Carpathian Arc is the product of just such an evolution, but little of the relevant structural history can be deduced directly from any synthesis yet published for that complex region. I found it essential in Indonesia to work from the primary reports by field geologists in order to discriminate what was known from what was assumed. The identification of accretionary-wedge materials is particularly critical for plate- tectonic interpretation. Useful description to permit such identification in Indonesia came from field reports, even though the writers of those reports had no concepts of melange and subduction and interpreted the materials in other ways, and the identifications were entirely lost in summary papers. (I visited in the field many terrains so identified and confirmed their character.) In my 1979 Indonesian monograph, I attempted to give objective descriptions, and to cite their sources, before going on to my subjective interpretations. I urge others to do likewise. 689 11.3. Regional Kinematics At the symposium for which this paper was written, many of the speakers, all regional experts, presented syntheses of large sectors of the Mediterranean- Tethyan belt in terms of slow, simple, long-continuing motions of lithosphere plates. Velocities inferred were high for the Indian plate but otherwise generally were low, on the order of 1 em/yr. Such slow and simple motions have only local analogs in active convergent-plate systems such as those of the Indonesian- Philippine-Melanesian region, so I infer that the Mediterranean-Tethyan region in fact records the consumption of great areas of oceanic lithosphere at velocities that often were high and that recorded exceedingly complex patterns. 1l.4. Subduction Systems The pairing of sutures and magmatic arcs presents the obvious means of demonstrating the polarity and timing of past subduction of oceanic lithosphere, but this powerful method has been overlooked by many investigators in the Tethyan region. Although blueschist metamorphism and the presence of large ophiolitic masses are generally recognized as evidence for subduction, the much more widespread polymict melanges and broken formations of accretionary wedges are too often treated as in situ stratigraphic units of "olistostromes" or "W ildflysch". Such accretionary-wedge materials contain records of sea-floor materials pushed in front of advancing plates and hence can tell much about the magnitude and history of subduction. Calc-alkalic volcanic rocks are widely recognized, and plutonic rocks less so, as likely formed in magmatic arcs above subducting plates, but little attempt has been made to evaluate possible pairings of sutures and magmatic arcs against dating criteria and against character and dimensions of intervening materials. Analogy with the known behavior of modern arc systems has yet been little used in syntheses of Tethyan tectonics. The concept that plate collisions generally are followed by reversals of subduction polarity, new subduction being directed beneath the plate aggregate from the outside, is commonly overlooked. The frequent presence of strips of oceanic lithosphere at the fronts of overriding plates, continental as well as island-arc, is rarely integrated. Awareness is mostly lacking of the behavior of oceanic island arcs--their increases in length and curvature with time, their complex collisions, aggregations, and reversals. The great variations in history along strike in evolving convergent-plate systems are little considered. The invalid concept of "obduction"--thrusting of slabs of oceanic material onto continents in the sense opposite to that of subduction--is often invoked, as is subduction beneath both sides of a rigid plate. Too much Tethyan geology is still interpreted in terms of theories modified only slightly from obsolete notions of geosynclines and block tectonics. "Nappes" (and, in the Soviet literature, "deep faults") are often invoked to avoid real analysis of abrupt juxtapositions that may have major plate-motion significance. Too little attention has been paid to variations with depth of tectonic and magmatic processes, and hence inconsistent kinematic interpretations are made along strike in obliquely eroded crustal sections. Thus, upper-crustal magmatism may be recognized as of arc-magmatic character, whereas mid-crustal, migmatitic sections in the same belt may be ascribed to "anatexis" of no particular plate-kinematic significance. 690 11.5. Paleobiogeography Evidence for or against plate-motion patterns of much greater complexity and magnitude than commonly visualized for the Mediterranean and Tethyan regions should be sought in paleobiogeographic data from accretionary-wedge materials and displaced continental fragments. The richly tropical Permian "Tethyan" fusulinid and coral fauna (Ross, 1967; Stevens, 1985) should be evaluated in this context. Some 35 Middle Permian fusulinid genera, for example, are largely or entirely restricted to this fauna yet occur widely through the Tethyan, Southeast Asian, and circum-Pacific regions. The circum-Pacific occurrences are as fragments of atolls and reefs within accretionary wedges, formed largely within Jurassic time, in the western Cordillera of the United States and Canada, in the Koryak and Sikhote Alin regions of far-eastern USSR, in Japan, and in New Zealand. (I have examined such occurrences myself in each of the non-Soviet countries.) Upper Carboniferous and Lower Permian fusulinids of similar significance are shared by melanges in Japan and southern Chile (Ozawa and Kanmera, 1984). The tropical assemblages are grossly out of place paleoclimatically in all of these regions and are accreted to continents that were temperate or cold in Permian time; the correlative tropical fusulinid faunas of the interior United States and the Central American and northern Andean regions are quite different. The Middle Permian Tethyan faunas are widespread also in platform limestones on various continental plates of southern China and Southeast Asia, including Sumatra, which were accreted to northern Eurasia after Permian time, and the fauna occurs in southern Tunisia-- paleo-equatorial mainland Africa--south of the Tyrrhenian accretionary-wedge melanges. Otherwise, the fauna occurs within the Tethyan region primarily in accretionary-wedge complexes and displaced crustal fragments in Sicily, the eastern Alps, the Middle East, and Central Asia. The Circum-Pacific distribution of the Permian Tethyan fauna requires the subduction of thousands of kilometers of paleo-Pacific lithosphere beneath the flanking continents, and large plates of such lithosphere may have been subducted within the Tethyan region also. The possibility deserves careful paleobio- geographic analysis of faunas. Vastly greater subduction, perhaps even of paleo- Pacific lithosphere, may be recorded by Tethyan sutures than is commonly assumed. 691 REFERENCES CITED Abbott, M. J., and F. H. Chamalaun, 1981, Geochronology of some Banda Arc volcanics: Ind~l1..esia Geol. Research and Development Centre Spec. Pub. 2, p. 253-268. Achdan, A., and T. Turkandi, 1982, Preliminary geologic map of the Kai (Tayandu and Tual) quadrangles, Maluku: Jndonesia Geol. Research and Development ,Centre. A1diss, D. T., and S. A. Ghazali, 1984, The regional geology and evolution of the Toba volcano-tectonic depression, Indonesia: Jour. Geol. Soc. London, v. 141, p. 487-500. Alt, J. C., Jose Honnorez, Christine Laverne, and Rolf Emmermann, 1986, Hydrothermal alteration of a 1 km section through the upper oceanic crust, Deep Sea Drilling Project Hole 504B--mineralogy, chemistry, and evolution of seawater-basalt interactions: Jour. Geophys. Research, v. 91, p. 10,309- 10,335. Alvarez, Walter, 1982, Geological evidence for the geographical pattern of mantle return flow and the driving mechanism of plate tectonics: Jour. Geophys. Research, v. 87, p. 6697-6710. Anderson, R. N., et al., 1978, A geophysical atlas of the east and southeast Asian seas--Heat flow, thermal conductivity, thermal gradient: Geol. Soc. America Map and Chart Series MC-25. Apandi, T., and D. Sudana, 1980, Geologic map of the Ternate quadrangle, North Maluku: Indonesia Geol. Research and DeveloQ!!1ent Centre, 9 p. + map. Audley-Charles, M. G., 1986, Rates of Neogene and Quaternary tectonic movements in the southern Banda Arc based on micropalaeontology: ~ Geol. Soc. London, v. 143, p. 161-175. __ et al., 1981, Reinterpretation of the geology of Seram--implications for the Banda Arc and northern Australia: Indonesia Geol. Research and Development Centre Spec. Pub. 2, p:2Ti:237..------.-·.---.~-·-··· .. · Bachman, S. B., S. D. Lewis, and W. J. Schweller, 1983, Evolution of a forearc basin, Luzon Central Valley, Philippines: Bull. Amer. Assoc. Petroleum .GeoI., v. 67, p. 1143-1162. Barber, A. J., 1981, Structural interpretations of the island of Timor, eastern Indonesia: Indonesia Geol. Research and Development C.t!ntre Spec. Pub. 2, 183-197. Beaudry, Desiree, and G. F. Moore, 1981, Seismic-stratigraphic framework of the forearc basin off central Sumatra, Sunda Arc: Earth and Planetary Science Letters, v. 54, p. 17-28. Bennett, J. D., et al., 1981, Geologic map of the Banda Aceh quadrangle, North Sumatra: Indonesia Geol. Research and Develo'p!!!ent Centre, 19 p. + map. Berry, R. F., 1981, Petrology of the Hili Manu lherzolite, East Timor: Jour. Geol. Soc. Australia, v. 28, p. 453-469. and A. E. Grady, 1981, Deformation and metamorphism of the Aileu -- Formation, north coast, East Timor and its tectonic significance: Jour. Structural Geol., v. 3, p. 143-167. ~-- Bowin, Carl, et aI., 1980, Arc-continent collision in Banda Sea region: Bull. Amer. Assoc. Petroleum Geo!., v. 64, p. 868-915. Bracey, D. R., 1983, Geophysics and tectonic development of the Caroline Basin: ~.0. Na~al Oce~nograph¥Somf!1an~ Techl1..!~al~Q 692 Cameron, N. R., et al., 1982, The geology of the Tapaktuan quadrangle, Sumatra: Indonesia Geol. Research and Development Centre, 18 p. + map. Cardwell, R. K., B. L. Isacks, and D. E. Karig, 1980, The spatial distribution of earthquakes, focal mechanism solutions, and subducted lithosphere in the Philippine and northeastern Indonesian islands: A.~e.r:d:;~QP~J'.!'i.J,Jl'!iQn Geophys. Mono. 23, p. 1-35. Carlson, R. L., 1981, Boundary forces and plate tectonics: Geophys. Research Letters, v. 8, p. 958-961. and P. J. Melia, 1984-, Subduction hinge migration: Tectonophysics, v. 102, p. -- 399-4-11. Chamalaun, F. H. et al., 1982, Banda Arc tectonics--the significance of the Sumba Island (Indonesia): Amer. Assoc. Petroleum Geologists Mem! 34-, p. 361-375. Chase, C. G., 1978, Extension behind island arcs and motions relative to hot spots: Jour. Geophys. Research, v 83, p. 5385-5387. Coleman, R. G., 1984- The diversity of ophiolites: Geologie en Mij~!>ouV0 v. 63, p. 14-1-150. Connelly, J. B., 1979, Mode of emplacement of the Papuan ultramafic belt: BMR Jour. Australian Geology and Geophysics, v. 4-, p. 57-65. Coulbourn, W. T., and R. Moberly, 1977, Structural evidence of the evolution of fore-arc basins of South America: Canadian Jour. Earth Sciences, v. 14-, p. 102-116. Crawford, A. J., L. Beccaluva, and G. Serri, 1981, Tectono-magmatic evolution of the West Philippine-Mariana region and the origin of boninites: Earth Planetary-Sci. Letters, v. 54-, p. 34-6-356. Davies, H. L., 1973, Fergusson Island, Papua New Guinea: Australia Bureau Mineral Resources, Geol. and Geophys., 1:250,000 Series, Sheet SC/56-5. __ and D. J. Ives, 1965, The geology of Fergusson and Goodenough Islands, Papua: Australia Bureau Mineral Resources, Geo1. and Geophys., Report 82, 65 p. -- -.-J P. A. Symonds, and I. D. Ripper, 1984-, Structure and evolution of the southern Solomon Sea region:. BMR Jour. Australian Geology and Geophysics, v. 9, p. 4-9-68. Dewey, J. F., 1980, Episodicity, sequence and style at convergent plate boundaries: Geol. Assoc. Canada Spec. ~c:t~! 20, p. 553-573. Dickinson, W. R., Compositions of sandstones in circum-Pacific subduction complexes and fore-arc basins: Amer. Assoc. Petroleum Geol. Bull., v. 66, p. 121-137. . Djuri, M., 1975, Geologic map of the Purwokerto and Tegal quadrangles, Indonesia: ~eological Survey of Indonesia. Dorian, J. P., A. L. Clark, and Djumhani, 1986, A geologic and mineral resource assessment of Indonesia: Jour. Southeast Asian Earth Sciences, v. 1, p. 33- 4-4-. . Dow, D. B., and Udi Hartono, 1982, The nature of the crust underlying Cenderawasih (Geelvink) Bay, Irian Jaya: }!ldonesia Geol. Research and .Development Centre Bull. 6, p. 30-36. __ and R. Sukamto, 1984-, Western Irian Jaya--the end product of oblique plate convergence in the late Tertiary: Tectonophysics, v. 106, p. 109-139. Drummond, B. J., C. D. N. Collins, and G. Gibson, 1979, The crustal structure of the Gulf of Papua and northwest Coral Sea: BMR Jour. Australian Geology and Geophysics, v. 4-, p. 34-1-351. 693 Evans, C. A., J. W. Hawkins, and G. F. Moore, 1983, Petrology and geochemistry of ophiolitic and associated volcanic rocks of the Talaud Islands, Molucca Sea collision zone, northeast Indonesia: Amer. Geophys. Union Geodynamics .Ser., v. 11, p. 159-172. Foden, J. D., 1983, The petrology of the calcalkaline lavas of Rindjani volcano, east Sunda Arc--a model for island arc petrogenesis: Jour. Petrology, v. 24, p. 98-130. __ and R. Varne, 1980, The petrology and tectonic setting of Quaternary-Recent volcanic centers of Lombok and Sumbawa, Sunda arc: Chemical Geology, v. 30, p. 201-226 Fornari, D. J., et al., 1979, Petrochemistry of the Sorol and Ayu troughs-- implications for crustal accretion at the northern and western boundaries of the Caroline Plate: Earth and Planetary Sci. Letter~ v. 45, p. 1-15. Garfunkel, Z., C. A. Anderson, and G. Schubert, 1986, Mantle circulation and the lateral migration of subducted slabs: Jour. Geophys. Research, v. 91, p. 7205-7223. Gill, J. B., 1981, 9rgenic andesites and plate tectonics: Secaucus, N.J., Springer- Verlag, 390 p. Hager, B. H., and R. J. O'Connell, 1981, A simple global model of plate dynamics and mantle convection: Jour. Geophys. Research, v. 86, p. 4843-4867. Hegarty, K. A., J. K. Weissel, and D. E. Hayes, 1983, Convergence at the Caroline-Pacific plate boundary-collision and subduction: Amer. Geophys. Union Geophys. Mo~~~ 27, p. 326-348. Hamilton, Warren, 1963, Metamorphism in the Riggins region, western Idaho: U.S. _Geo~~!y.~Y_~~~~X~e~ 436, 95 p. 1974a, Earthquake map of the Indonesian region: U.S. Geol. Survey Map 1- - 875-C, 1 :5,000,000. . 1974b, Map of sedimentary basins of the Indonesian region: U.S. Geol. Survey - J'.i~R 1-875-B, 1:5,000,000. __ 1978a, Tectonic map of the Indonesian region, 1:5,000,000: U.S. Geol. Survey Map 875-D. (Reprinted with revisions, 1981.) 1978b, Mesozoic tectonics of the western United States: Pacific Sec., Soc. -- Econ. Paleontol. Mineral., Paleogeography Symposium 2, p. 33-70. .-- __ 1979, Tectonics of the Indonesian region: U.S. Geol. Survey Prof. Paper 1078, 345 p. (Reprinted, with minor corrections, 1981 and 1985). __ 1981, Crustal evolution by arc magmatism: Philos. Trans. Royal Soc. London,_ v. A-301, p. 279-291. __ 1985, Subduction, magmatic arcs, and foreland deformation: Circum-Pacific Council for Energy and Mineral Resources, Earth Sci. Ser., v. 1, p. 259-262. __ in press, a, Crustal extension in the Basin and Range Province, southwestern United States: in Continental Extensional Tectonics, Geol. Soc. London Spec. Pub. - __ in press, b, Tectonic setting and variations with depth of some Cretaceous and Cenozoic structural and magmatic systems of the western United States: in W. G. Ernst, ed., Metamorphism and crustal evolution of the western United States: Prentice-Hall. Hawkins, J. W., et al., 1984, Evolution of intra-oceanic arc-trench systems: TectonophYSics, v. 102, p. 174-205. __ et al., 1985, Geology of the composite terranes of east and central Mindanao: Circum Pacific Council for Energy and Mineral Resources, _Earth Sci. Ser., v. 1, p. 437-463. 694 Hayes, D. E., and Brian Taylor, 1978, A geophysical atlas of the east and southeast Asian seas--Tectonics: Geol. Soc. America Map and Chart Series MC-25. __ et al., 1978, A geophysical atlas of the east and southeast Asian seas--Crustal structure: Geol. Soc. America Map and Chart Series MC-25. Holloway, N. H., 1982, North Palawan block, Philippines--its relation to Asian mainland and role in evolution of South China Sea: Bull. Amer. Assoc. Petroleum Geol., v. 66, p. 1335-1383. Hutchinson, R. W., 1980, Massive base sulphide deposits as guides to tectonic evolution: Geol. Assoc. Canada Spec. Paper 20, p. 659-694. Jacobson, R. S., et al., 1979, Seismic refraction and reflection studies in the Timor-Aru Trough system and Australian continental shelf: Amer. Assoc. Petroleum Geol. Mem--1.2, p. 209-222. . Johnston, C. R., and C. O. Bowin, 1981, Crustal reactions resulting from the mid- Pliocene to Recent continent-island arc collision in the Timor region: BMR Jour. Australian Geology and Geophysics, v. 6, p. 223-243. Karig, D. E., 1972, Remnant arcs: Geol. Soc. America Bull., v. 83, p. 1057-1068. 1982, Initiation of subduction zones--implications for arc evolution and -- ophiolite development: Geol. Soc. London Spec. Pub. 10, p. 563-576. --.J M. B. Lawrence, G. F. Moore, and J. R. Curray, 1980, Structural framework of the fore-arc basin, NW Sumatra: Jour. Geo!. Soc. London, v. 137, p. 77- 9l. --.J G. F. Moore, J. R. Curray, and M. B. Lawrence, 1980, Morphology and shallow structure of the lower trench slope off Nias Island, Sunda Arc: Amer. Geophys. Union Geophys. Mono. 23, p. 179-208. Kieckhefer, R. M., et al., 1980, Seismic refraction studies of the Sunda Trench and forearc basiniJOur. Geophys. Research, v. 85, p. 863-889. Knight, M. D., G. P. L. Walker, B. B. Ellwood, and J. F. Diehl, 1986, Stratigraphy, paleomagnetism, and magnetic fabric of the Toba tuffs--constraints on the sources and eruptive ages: Jour. Geophys. Research, v. 91, p. 10,355-10,382. Lewis, S. D., and D. E. Hayes, 1984, A geophysical study of the Manila Trench, Luzon, Philippines. 2. Fore arc basin structural and stratigraphic evolution: Jour. Geophys. Research, v. 89, p. 9196-9214. Malinverno, Alberto, and W. B. F. Ryan, 1986, Extension in the Tyrrhenian Sea and shortening in the Apennines as result of arc migration driven by sinking of the lithosphere: Tectonics, v. 5, p. 227-245. Mammerickx, J., et al., 1976, Bathymetry of the east and southeast Asian seas: Geol. Soc. America Map and Chart Series MC-17. McCabe, Robert, 1984, Implications of paleomagnetic data on the collision related bending of island arcs: Tectonics, v. 3, p. 409-428. --.J Jose Almasco, and Wilfred Diegor, 1982, Geologic and paleomagnetic evidence for a possible Miocene collision in western Panay, central Philippines: Geology, v. 10, p. 325-329. --.J Jose Almasco, and G. Yumul, 1985, Terranes of the central Philippines: Circum-Pacific Council for Energy and Mineral Resources, Earth Science Series, v. 1, p. 421-436. McCaffrey, Robert, 1982, Lithospheric deformation within the Molucca Sea arc- arc collision--evidence from shallow and intermediate earthquake activity: Jour. Geophys. Research, v. 87, p. 3663-3678. __ and John Nabelek, 1984, The geometry of back arc thrusting along the eastern Sunda Arc, Indonesia--constraints from earthquake and gravity data: .Jour. Geophys. Research, v. 89, p. 617l-6179. McCaffrey, Robert, et al., 1985, Microearthquake seismicity and fault plane solutions related to arc-continent collision in the eastern Sunda Are, Indonesia: Jour. Geophys. Research, v. 90, p. 4511-4528. 695 ---.J E. A. Silver, and R. W. Raitt, 1980, Crustal structure of the Molucca Sea collision zone, Indonesia: Amer~~ophys. Union Geophys. Mono. 23, p. 161- 178. Molnar, Peter, and Tanya Atwater, 1978, Interarc spreading and Cordilleran tectonics as alternates related to the age of subducted oceanic lithosphere: Earth Planetary Science Letters, v. 41, p. 330-340. Moore, D. G., J. R. Curray, and F. J. Emmel, 1976, Large submarine slide (olistostrome) associated with Sunda Arc subduction zone, northeast Indian Ocean: Marine Geology, v. 21, p. 211-226. Moore, G. F., H. G. Billman, P. E. Hehanussa, and D. E. Karig, 1980, Sedimentology and paleobathymetry of Neogene trench-slope deposits, Nias Island, Indonesia: Jour. Geology, v. 88, p. 161-180. ---.J J. R. Curray, and F. J. Emmel, 1982, Sedimentation in the Sunda Trench and forearc region: Geol. Soc. London Spec. Pub. 10, p. 245-258. ---.J J. R. Curray, D. G. Moore, and D. E. Karig, 1980, Variations in geologic structure along the Sunda fore arc, northeastern Indian Ocean: Amer. Geophys. Union Geophys. Mono. 23, p. 145-160. . ---.J D. Kadarisman, C. A. Evans, and J. W. Hawkins, 1981, Geology of the Talaud Islands, Molucca Sea collision zone, northeast Indonesia: Jour. Structural Geol., v. 3, p. 467-475. __ and D. E. Karig, 1980, Structural geology of Nias Island, Indonesia-- implications for subduction zone tectonics: Amer. Jour. Science, v. 280, p. 193-223. and E. A. Silver, 1982, Collision processes in the northern Molucca Sea: -- Amer. Geophys. Union Geophys. Mono. 27, p. 360-372. Morrice, M. G., et ~., 1983, An introduction to the Sangihe arc--volcanism accompanying arc-arc collision in the Molucca Sea, Indonesia: Jour. Volcanology Geothermal Research, v. 19, p. 135-165. -- Morris, J. D., et al., 1982, The Halmahera island arc, Molucca Sea collision zone, Indonesia--a geochemical survey: Amer. Geophys. Union Geophys. Mono. 27, p. 373-387. Mrozowski, C. L., and D. E. Hayes, 1978, A geophysical atlas of the east and southeast Asian seas--Sediment isopachs: Geol. Soc. America Map and Chart Series MC-25. Ollier, C. D., and C. F. Pain, 1980, Actively rising surficial gneiss domes in Papua New Guinea: Jour. Geol. Soc. Australia, v. 27, p. 33-34. __ and---.J 1981, Active gneiss domes in Papua New Guinea--new tectonic landforms: _Zeit. Ge..()}:r'l()rf>~_~,_N.F., v. 25, p. 133-145. Ozawa, Tomowo, and Kametoshi Kanmera, 1984, Tectonic terranes of late Paleozoic rocks and their accretionary history in the Circum-Pacific region viewed from fusulinacean paleobiogeography: Stanford University Pubs. Geol. Sci., v. 28, p. 158-160. --- Parsons, Barry, 1982, Causes and consequences of the relation between area and age of the ocean floor: Jour. Geophys. Research, v. 87, p. 289-302. Pearce, J. A., S. J. Lippard, and S. Roberts, 1984, Characteristics and tectonic significance of supra-subduction zone ophiolites: Geol. Soc. London Spec. Pub. 16, p. 77-94. 696 Penningto.n, W. D., 1983, Ro.le o.f shallo.w phase changes in the subductio.n o.f o.ceanic crust: Science, v. 220, p. 1045-1047. Pieters, P. E., et al., 1983, The stratigraphy o.f western Irian Jaya: Indo.nesia Geo.l. Research and Develo.eme_~!~C2~n_t~~_B~ll. 8, p. 14-48. Pigram, C. J., et al., 1982, Geo.Io.gical results o.f the 1981 expeditio.n to. the Miso.o.l Archipelago., Irian Jaya: Indo.nesia Geo.l. Research and Develo.pm~~.!Sentre_ Bull. 6, p. 18-29. __ and H. Panggabean, 1983, Age o.f the Banda Sea, eastern Indo.nesia: Nature., v. 301, p. 231-234. __ and --' 1984, Rifting o.f the no.rthern margin o.f the Australian co.ntinent and the o.rigin o.f so.me micro.co.ntinents in eastern Indo.nesia: T ecto.no.physics, v. 107, p. 331-353. --- --' G. P. Ro.binso.n, and S. Lumban To.bing, 1983, Late Caino.zo.ic o.rigin fo.r the Bintuni Basin and adjacent Lengguru fo.ld belt, Irian Jaya: Indo.nesia Geo.l. Research and Develo.pment Cen~~~ Bull. 7, p. 24-36. -- -- ... Pindell, J. L., and S. F. Barrett, in press, Geo.Io.gic evo.lutio.n o.f the Caribbean regio.n--a plate tecto.nic perspective: Geo.l. So.c. America, Decade No.rth __ American Geo.logy_, v. H. Reed, M. H., 1983, Seawater-basalt interactio.n and the o.rigin o.f greensto.nes and o.re depo.sits: Eco.no.mic Geo.l?...8Y, v. 78, p. 466-485. Ro.ck, N. M. S., n al., 1983, The geo.Io.gy o.f the Lubuksikaping quadrangle, Sumatra: Indo.nesia Geo.l. Research and Develo.emen!.~en.!E~J..60 p. + map. Ro.o.bo.l, M. J., N. J. Jackso.n, and D. F. P. Darbyshire, 1983, Late Pro.tero.zo.ic lavas o.f the central Arabian shield--evo.lutio.n o.f an ancient arc system: Jo.ur. Geo.l. So.c. Lo.ndo.n, v. 140, p. 185-202. Ro.sidi, H. r;;CD:; n~., 1979, Geo.Io.gic map o.f the Kupang-A tambua quadrangles, Timo.r: Indo.nesia Geo.l. Research and Devel~~~nt Centre, 17 p. + map. Ro.ss, C. A., 1967, Develo.pment o.f fusulinid (Fo.raminiferida) faunal realms: Jo.ur. Paleo.nto.lo.gy, v. 41, p. 1341-54. --. Rubie, D. C., 1984, The o.livine-spinel transfo.rmatio.n and the rheo.Io.gy o.f subducting litho.sphere: Nature, v. 308, p. 505-508. Sarewitz, D. R., and D. E. Karig, 1986, Pro.cesses o.f allo.chtho.no.us terrane evo.lutio.n, Mindo.ro. Island, Philippines: Tecto.nics., v. 5, p. 525-552. Sclater, J. G., Barry Parso.ns, and Claude Jaupart, 1981, Oceans and co.ntinents-- similarities and differences in the mechanisms o.f heat lo.ss: Jo.ur. Geo.phys. Research, v. 86, p. 11,535-11,552. Silver, E. A., et al., 1985, Evidence fo.r a submerged and displaced co.ntinental bo.rderland, no.rth Banda Sea, Indo.nesia: Geo.Io.gy, v. 13, p. 687-691. et al., 1986, Multibeam study o.f the FIo.res backarc thrust belt, Indo.nesia: --JOUr. Geo.~ys. Research, v. 91, p. 3489-3500. --' R. McCaffrey, Y. Jo.yo.diwiryo., and S. Stevens, 1983, Ophio.lite emplacement by co.llisio.n between the Sula Platfo.rm and the Sulawesi Island Arc, Indo.nesia: Jo.ur. Geo.phys. Research, v. 88, p. 9419-9435. --' R. McCaffrey, and R. B. Smith, 1983, Co.llisio.n, ro.tatio.n, and the initiatio.n o.f subductio.n in the evo.lutio.n o.f Sulawesi, Indo.nesia: Jo.ur. Geo.phys. Resear 697 Silver, E. A., and R. B. Smith, 1983, Comparison of terrane accretion in modern Southeast Asia and the Mesozoic North American Cordillera: Geology, v. 11, p. 198-202. Soeria-A tmadja, R., 1981, Ophiolites in the Halmahera paired belts, east Indonesia: Indonesia Geol. Research and Devel0E.men~ 698 Weissel, J. K., Brian Taylor, and G. D. Karner, 1982, The opening of the Woodlark Basin, subduction of the Woodlark spreading system, and the evolution of northern Melanesia since mid-Pliocene time: Tectonophysics, v. 87, p. 253- 277. and A. B. Watts, 1979, Tectonic evolution of the Coral Sea Basin: Jour. -- Geophys. Research, v. 84, p. 4572-4582. --- Windley, Brian, 1984, The evolving continents: Chichester, John Wiley & Sons, 399 p. Yasin, Aswan, 1980, Geologic map of the Bacan quadrangle, North Maluku: Indonesian Geo!. Research and Development Centre, 9 p. + map. /ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 150 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False /CreateJDFFile false /Description > >> setdistillerparams > setpagedevice