Influence of carbon and nitrogen sources, relative carbon and nitrogen concentrations, and soil moisture on the growth in nonsterile soil of soilborne fungal antagonists

Influence of carbon and nitrogen sources, relative carbon and nitrogen concentrations, and soil moisture on the growth in nonsterile soil of soilborne fungal antagonists J . P. STACK, C. M . KENERLEY, A N D R. E. PETTIT Department of Plant Pathology and Microbiology, Texas A & M Universih, College Starion, TX, U.S.A. 77843 Received July 20, ! 986 Accepted March 10, 1987 STACK, J. P., KENERLEY, C. M., and PETTIT, R. E. 1987. Influence of carbon and nitrogen sources, relative carbon and nitrogen concentrations, and soil moisture on the growth in nonsterile soil of soilborne fungal antagonists. Can. J. Microbiol. 33: 626-63 1. Three components of dispersal important to the colonization of a soil matrix by fungal parasites of sclerotia were identified and measured: the percentage of the carrier granules from which hyphae extended into soil (PGH), the mean number of hyphae extending into soil from each granule (MNH), and the mean length of the hyphae extending into soil (MLH). Factors that influence dispersal were determined for strains of Gliocladium roseum, Thielavia rerricola, and Trichoderma spp. The source of carbon, the source of nitrogen, and the C:N ratio of the carrier substrate significantly ( p = 0.001) affected all three components of dispersal subsequent to the placement of the carrier granules in nonsterile soil. Increased C:N ratio ( 12: 1 to 80: 1 ) and increased molar concentrations of both carbon and nitrogen sources (0.02 to 0.18 M maltose and 0.006 to 0.024 M arginine) gave increased PGH (1 7 to 82%), MNH (1 to 5 hyphae per granule), and MLH (275 to 782 pm in 24 h) for G . roseum. Similar increases in PGH (80 to 1 OO%), MNH (5 to 10 hyphae per granule), and MLH (983 pm to too long and interwoven to measure after 24 h) were observed for Th. terricola. PGH and MNH were greater at high soil moistures (-0.1 and -0.33 bars matric potential; 1 bar = 100 kPa) than at low soil moisture (- 10 bars). STACK, J. P., KENERLEY, C. M., et PETTIT, R. E. 1987. Influence of carbon and nitrogen sources, relative carbon and nitrogen concentrations, and soil moisture on the growth in nonsterile soil of soilborne fungal antagonists. Can. J. Microbiol. 33 : 626-63 1. Trois composantes de la dispersion sont importantes pour la colonisation de la matrice d'un sol par des champignons parasites des sclirotes. Identifiees et mesurees, ces composantes sont : le pourcentage de granules porteuses de germes fongiques a partir desquelles les hyphes se diveloppent dans le sol (PGH), le nombre moyen d'hyphes qui se diveloppent a partir de chaque granule (MNH) et la longueur moyenne de ces hyphes (MLH). Des souches de Gliocladium roseum, Thielavia rerricola et de Trichoderma spp. ont semi a determiner les facteurs qui influencent la dispersion. Lorsque des granules porteuses de germes furent placies dans un sol non stirilise, la source de carbone, la source d'azote et le ratio C:N du substrat des porteurs ont affect6 les trois composantes de la dispersion de fason significative ( p = 0.00 1 ). Pour le G . roseum, une augmentation du ratio C:N (de 12: 1 a 80: 1) et un accroissement des concentrations molaires des sources de carbone et d'azote (le maltose, de 0,02 a 0,18 M; l'arginine, de 0,006 0,024 M ) ont fait augmenter le PGH de 17 a 82%, le MNH de 1 a 5 hyphes par granule et le MLH de 275 a 782 pM en 24 h. Des augmentations similaires ont ete obsemees avec le Th. rerricola, soit de 80 a 100% pour le PGH, de 5 a 10 hyphes par granule pour le MNH et de 983 pM a des dimensions trop longues d'hyphes entremelis pour itre mesuries aprks 24 h pour le MLH. Dans les sols a humidite ilevie (-0.01 et -0.33 bars du potentiel matriciel; 1 bar = 100 kPa) les PGH et MNH furent plus eleves que dans les sols a faible humidit6 ( - 10 bars). [Traduit par la revue] Introduction propagule. A preliminary report of this work has been published Growth of hyphae through soil is one mechanism of active (Stack et al. lgg5). dispersal for soilborne fungi. There are several reports con- cerning the development and growth of mycelial strands and rhizomorphs through artificial and natural environments (Butler 1984; Trinci 1984; Watkinson 1984). Moisture, nutrition, and C:N ratio were among the factors found to influence growth. Most of this work involved Basidiomycetes. There is little available information about hyphal growth in nonsterile soil by either soilborne plant pathogens or soilborne fungal antagonists of plant pathogens. The nature of this growth and the factors that affect growth must be understood to maximize the performance of biocontrol agents. Dispersal is very important in the epidemiology of diseases of plants. It is likely to be important in the epidemiology of diseases of fungi (e.g., plant pathogens) as well. In the case of Sporidesmium sclerotivorum, active growth in soil was, in part, responsible for succesful control of Sclerotinia minor in the field (Ayers and Adams 1979). The objectives of this research were to determine ( i) the factors that influence the dispersal of soil borne antagonists and (ii) whether, by manipulating these factors, dispersal can be maximized. The intent is to achieve better colonization of the soil and increased probability of contact with the target Materials and methods Strains and media The following strains were isolated from sclerotia of Aspergillus Jlavus Link. ex Fries that had been buried in nonsterile soil from peanut fields of Texas: Gliocladium roseum Bain. ((31-4, Gr6), Thielavia rerricola (Gilman and Abbot) Emmons (Th2, Th3), and Trichoderma sp. (Tri2). Gliocladium roseum Bain. (Gr1620) was obtained from a sclerotium of Phymarotrichum omnivorum L. (Shear) Duggar that had been buried in a nonsterile soil from a cotton field of Texas. These strains were selected because of their frequency of isolation from, and ability to colonize sclerotia of, either A . Jlavus or Ph. omnivorum. Gliocladium roseum (Gr3), obtained from roots of alfalfa in New York State, and the Trichoderma strains (Tham, Tharz), originally obtained from soil, were included for comparative purposes. The Gliocladium and Trichoderma strains were maintained on potato dextrose agar (PDA) (Difco Co., St. Louis, MO) at 25OC. The Thielavia strains were maintained on 20% clarified V-8 medium (Campbells Soup Co.) at 25°C. Soils Two soils were used in this' study. A Houston black clay (HBC) collected from a cotton field near Temple, TX, and a sandy-loam soil (SSL) collected from a peanut field near Stephenville, TX. The SSL Pr~nted In Canada I Imprime au Canada C an . J . M ic ro bi ol . D ow nl oa de d fr om w w w .n rc re se ar ch pr es s. co m b y U ni ve rs ity o f P. E .I . o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. STACK ET AL. 627 had the following characteristics: less than 1% organic matter, pH 7.0 (determined in water), 14 pg phosphorus/g, 84 pg potassium/g, 335 pg magnesiumlg, 560 pg calciumlg, and a predicted low release of nitrogen. The HBC was described previously (Kenerley and Stack 1986). Moisture release curves were generated for both soils using a pressure plate apparatus. In growth potential experiments, soil moisture was adjusted gravimetrically to a percentage moisture that corresponded to a specific matric potential (MP) on the release curve. It is recognized that, with certain soils, rewetting a dry soil to a specific percentage moisture can result in a different MP than that obtained by drying a saturated soil to that same percentage moisture. Since our objective was to determine the effects of large differences in MP (-0.33 vs. - 10 bar; 1 bar = 100 kPa) rather than effects of specific MPs, the possible discrepencies due to hysteresis were not considered. Lignite cultures Lignite was collected from mines near Rockdale,TX. It was ground into small granules and sieved (4-mm openings) to obtain uniform particle size. The granules were impregnated with various nutrient substrates by soaking in flasks (50%, by weight substrate:lignite) and then drying for 24 h at 25OC. When dry, the lignite was again soaked in substrate (50%, by weight) and autoclaved (20 min at 1 21°C). When cool, the lignite was colonized with an antagonist by the addition of either agar blocks from actively growing colonies or suspensions of conidia (approximately lo5- 1 o6 conidia/mL) from sporulating cultures. The flasks of lignite were held at 25 or 35" for 4-7 days to allow thorough colonization of the granules by the antagonist. 'The lignite cultures were then air dried. The substrates used were thin liquid stillage, a by-product of sorghum fermentation (Jones et al. 1984) having a C:N ratio of approximately 20: 1, and a minimal salts medium supplemented with various carbon and nitrogen sources in varying C:N ratios. The total mass of carbon and nitrogen in the substrates were determined. For compounds (e.g., amino acids) containing both elements, the mass of each element was considered in determining the total mass. For example, when the substrate contained both maltose and arginine, the carbon in maltose and the carbon in arginine were both calculated to determine the total carbon mass. The C:N ratios were expressed as the percentage of carbon to the percentage of nitrogen relative to the mass of the total substrate. The ratios considered in this study were 12: 1, 40: 1, and 80: 1, which represent a realistic range of C:N of the organic matter incorporated into agricultural soils, and have been reported to have an influence on the growth of fungi in culture. The C:N ratio was adjusted by increasing or decreasing the amount of carbon or nitrogen sources relative to each other. The minimal salts medium contained the following (per litre): MgS04-7H20, 500 mg; CaC12.2H20, 100 mg; H3B04, 2.86 pg; MnC12.4H20, 1.41 pg; ZnS04-7H20 , 0.22 pg; CuSo4. 5H20 , 0.08 pg; FeKEDTA, 33 mg; thiamine-HC1, 100 pg; and biotin, 10 pg . This medium was prepared in 0.07 M phosphate buffer at pH 5 .5 and 7.0. For experiments on solid media, the buffered minimal salts medium was supplemented with 18 g agar/L medium. The minimal salts medium would not support growth of these fungi without supplemental carbon and nitrogen. Concentrations of biotin and thiamine-HC1 (carbon- and nitrogen-containing compounds) in the minimal salts medium were too low to significantly contribute to the overall C:N. Substrate pH To determine the effect of substrate pH upon the growth of the antagonists, the minimal salts medium containing carbon and nitrogen sources was adjusted (1 N NaOH or 1 N HCl) to pH 5.5 or 7.0. Lignite granules were impregnated with the substrates as described above and colonized by Gr6, Gr1620, or Th2. The C-N combinations tested were maltose-arginine, galactose- arginine, fructose-arginine, fructose- nitrate, glucose-arginine, and thin liquid stillage. For each treatment, 0.5 g lignite was added to 10 mL of nonbuffered sterile, distilled deionized water (pH 5.5), shaken well, and allowed to stand at laboratory temperature. The pH was determined over time by sub- mersing the electrode into the H 2 0 above the lignite. Growth potential experiments Lignite granules colonized by an antagonist were dispersed over the surface of the soil in a plastic petri dish (0.1 g lignitel5-cm dish). A nylon mesh screen (100 p m pore size; Nitex, Tetko Co., Elmsford, NY) was placed on top of the lignite. Soil was added to bury the nylon and lignite to a dept of 0.5-1.0 cm and the soil was firmed. The moisture levels of all soils were preadjusted to correspond to MPs of -0.1, -0.33, - 1 .O, or - 10 bars. The plates were held at 25 or 35°C for 24-72 h at which point the soil was removed to expose the nylon mesh. The nylon was lifted and the lignite granules were observed at 10 X and 40 X magnification. The following observations were made: (i) the proportion of lignite granules from which growth occurred (either hyphal extension or sporulation); (ii) the number of hyphae extending from each granule; and (iii) the length of hyphae extending from the lignite granule into the soil. Initial experiments indicated little variability within and among replicate plates. Therefore, 20 observations per replicate plate with three replicate plates per treatment were used, unless indicated otherwise. Noncolonized substrate-im- pregnated granules were placed in soil as controls to aid distinction of hyphae of the introduced agent from hyphae of soil fungi, which might colonize and grow from the granule after placement in soil. Growth on agar Agar disks (5 mm diameter) from the margin of actively growing PDA cultures of G. roseum (Gr1620) or Th. terricola (Th2) were placed in the center of a 10-cm agar plate containing the minimal salts-agar (pHs 5.0 and 7.0) described above. This medium was supplemented with various carbon and nitrogen compounds in various C:N ratios. The final molarities for the carbon and nitrogen compounds were 0.02-0.05 and 0.01 -0.02 M, respectively. Radial growth was determined at 48-h intervals by measuring the greatest diameter for each of five replicate plates per treatment. Statistical analyses The data were tested for normality with a univariate procedure of SAS (Statistical Analysis Systems, release 1985, SAS Institute, Inc., Cary, NC). Some data sets contained zeros and most data were not normally distributed. It was determined that a log transformation of the nonzero values best approximated a normal probability distribution. This was based on results of the Shapiro-Wilk test. The occurrence of zeros in some data sets precluded a log transformation. To obtain accurate estimates of the true means, the procedure of Aitchison (1955) was utilized. This procedure provides minimum-variance unbiased estimators of the means for log normally distributed data sets containing zeros (Owen and DeRouen 1980; Pennington 1983). An anaylsis of variance was performed (General linear models procedure, SAS) using the Aitchison estimator values. Results EfSect of soil MP on growth of antagonists Lignite granules impregnated with thin liquid stillage were colonized with strains Gr4, Th2, Tri2, and Tham. After 24 h in nonsterile sandy-loam soil at -0.33 or - 1.0 bar MP, hyphal extension into soil at 25 and 35°C occurred from 73 to 100% of the granules colonized by Gr4, Th2, and Tri2 (Table 1). At - 10.0 bars MP, the percentage of granules with extended hyphae (0-18%) was significantly (p = 0.001) less. Even at high soil moisture (- 0.33 bars), Tham grew from fewer granules than the other species. The mean number of hyphae extending into soil from each granule increased with increasing moisture (Table 1). For strains Gr4, Th2, and Tham, there were approximately five times as many hyphae extending from granules at -0.33 than at - 10.0 bars MP in the sandy-loam soil. Strain Tri2 did not grow at - 10 bars. The effect of soil moisture on the length of hyphae extending into the SSL was species dependent. There was no significant moisture effect (p = 0.001) for Th. terricola (Th3) at the three C an . J . M ic ro bi ol . D ow nl oa de d fr om w w w .n rc re se ar ch pr es s. co m b y U ni ve rs ity o f P. E .I . o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. CAN. J . MICROBIOL. VOL. 33, 1987 TABLE 1. The percentage of lignite carrier granules with growth of the biocontrol agent into soil and the mean number of hyphae per carrier granule extending into soil at two temperatures and three MPs after 24 ha % of granules with Mean no. of growth into soil at extending three MPs hyphaelgranule at ( - bars)b three MPs (-bars)" Temperature Agent ("c) 0.33 1.0 10.0 0.33 1.0 10.0 Th. terricola (Th2) 25 3 5 G. roseum (Gr4) 25 3 5 Trichoderma sp. (Tri2) 25 3 5 Tr. hamatum (Tham) 25 3 5 "Lignite granules (1-2 mm diameter) were impregnated with thin liquid stillage, coloni~ed by the indicated species of fungi, allowed toair dry, and then placed in soil. Soil moisture was adjusted gravimetrically. See text for details. bSE of the differences = 8 . 4 . 'SE of the differences = 1.06. TABLE 2. The percentage of lignite carrier granules with growth of the biocontrol agents into soil and the mean number of hyphae per carrier granule extending into soil with two sources of carbon and three C:Nsu Mean no. of % of granules with extending growth into soil at hyphaelgranule at three C : N S ~ three C:Ns" Time Agent Substrate (h) 12: 1 40: 1 80: 1 12: 1 40: 1 80: 1 Th. terricola (Th2) M +A 24 80 100 100 5 10 10 48 90 100 100 8 10 10 M+NO, 24 31 96 100 2 9 10 48 47 100 100 6 10 10 G . roseum (Gr 1620) M +A 24 17 51 82 1 5 5 48 27 38 48 48 6 6 M+NO, 24 25 29 28 2 4 2 48 26 35 38 5 4 5 "Lignite granules were impregnated with minimal salts medium containing maltose and arginine (M+ A ) or maltose and potassium nitrate ( M + N 0 3 ) as the primary sources of carbon and nitrogen at C:Ns of 12: 1 , 40: 1 , and 80: 1 . After colonization by Th2 or Gr1620, the granules were air dried and placed in soil at -0 .33 bars MP and 35°C. bSE of the differences = 5.51. "SE of the differences = 0 .64 . levels of MP tested. However, G . roseum (Gr3), Tr. hamatum (Tharn), and Tr. harzianum (Tharz) had less (p = 0.001) extension into soil at - 10.0 than at -0.33 bars MP. This effect was more pronounced at 25 than at 35°C. Variability with respect to hyphal extension into soil was observed among the three strains of Trichoderma. The Tri-2 strain grew further than the other two strains at all moisture levels tested. The effects of soil moisture and soil temperature upon the three growth parameters of antagonists in the SSL were also observed in the HBC. Substrate pH There was essentially no difference among the pH values recorded at 10 min or 5 h after immersing the lignite granules in water. The pH of lignite granules impregnated with thin liquid stillage was 5.3 whether or not the granules were colonized by Gr6. The granules impregnated with fructose-nitrate and glucose-arginine that were colonized by Th2 had pH 5.7. All other combinations of C-N source and isolate (Th2 or Gr l620), regardless of initial substrate pH (5.5 or 7.0), had pH 6.0-6.2. The efSect of carbon and nitrogen source and C:N ratio on growth in nonsterile soil The source of carbon, the source of nitrogen, and the relative concentrations of carbon and nitrogen affected the growth of the antagonists in a nonsterile soil. The nature of the effect was manifested in three ways: the percentage of carrier granules with growth of the antagonist (Table 2), the number of hyphae growing from individual granules (Table 2), and the length of the hyphae extending into soil from the carrier granule (Table C an . J . M ic ro bi ol . D ow nl oa de d fr om w w w .n rc re se ar ch pr es s. co m b y U ni ve rs ity o f P. E .I . o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. STACK ET AL. 629 TABL from .E 3. The mean length of hyphae extending into soil the carrier granules impregnated with various sources and C:Ns after 24 ha Hyphal length (km) at three C:Ns Agent Substrate 12: 1 40: 1 80: 1 Th, terricola (Th2) M + A 983 734 OG M + N 0 3 166 568 OG G . roseum (Gr 1620) M + A 275 697 782 M + N 0 3 442 434 436 'lignite granules were impregnated with minimal salts medium containing maltose and arginine (M+A) or maltose and potassium nitrate ( M + N 0 3 ) as the primary sources of carbon and nitrogen at C:Ns of 12: 1,40: 1 , and 80: 1. After colonization by Th2 or (31620, the granules were air dried and placed in soil at -0.33 bars MP and 35°C. The SE of the differences = 116. 3). The values in Tables 2 and 3 are the results of a single experiment conducted in the HBC soil. The experiment was repeated twice with comparable results. Similar experiments were conducted in the SSL soil and comparable results were obtained. The magnitude of the values varied among experi- ments, but the nature of the growth response remained consistent (e.g., increased C:N from 12: 1 to 80: 1 resulted in an increased number of hyphae per carrier granule in each experiment). We also observed this effect of C:N on the growth response of Th2 and Gr1620 with other sources of carbon and nitrogen, e.g., galactose-ammonium and galactose-nitrate. In initial experi- ments, we observed that after 48 h in nonsterile soils, HBC and SSL, at 25°C and - 1.0 or -0.33 bars MP, the percentage of granules with hyphal growth and the length of Th. terricola (Th2) hyphae extending into soil were greater with maltose or galactose (primary carbon source) in combination with arginine (primary nitrogen source) than with fructose and arginine. Galactose in combination with arginine gave a greater growth response by Th2 than galactose with KN03. In one experiment with Gr 1620, the growth response from a defined synthetic medium (minimal salts with galactose- ammonium) was much greater than growth from an undefined natural substrate (thin liquid stillage). At 25°C and -0.33 bars MP there were 100 and 13% granules with hyphal extension, and nine and two hyphae per granule for the synthetic and natural substrates, respectively. Similar results were observed at 25°C with - 1 .O bar MP and at 35°C with -0.33 and - 1 .O bars MP. As the ratio of C:N was increased from 12: 1 to 80: 1, the percentage of granules with hyphae extending into the soil increased for both G. roseum and Th. terricola (Table 2). This was true at 24 and 48 h whether the nitrogen source was arginine or nitrate, with the exception of G. roseum at 24 h with nitrate. Thielavia terricola (Th2) grew from a much higher percentage of the carrier granules than G. roseum (Gr1620) under most conditions of substrate. This was statistically significant (p = 0.001). The increase in C:N from 12: 1 to 80: 1 was equivalent to an increase in the molarity of the carbon source maltose from 0.02 to 0.18 M, while the molarities of the nitrogen sources arginine and nitrate remained constant at 0.006 and 0.024 M, respectively. Similar results were observed when the number of hyphae per granule was considered. As the C:N was increased from 12: 1 to 80:1, the number of hyphae extending from each granule increased for both agents (Table 2). For Th2 at 24 h, the number of hyphae at C:N of 80:l was two- (maltose-arginine) and five-fold (maltose-nitrate) higher than at 12: 1. For Grl620 at 24 h, the number of hyphae per granule at the C:N of 80:l was fivefold higher than at the C:N of 12: 1 with maltose-arginine, but there was no increase when the C:N ratio of maltose-nitrate was altered. These differences were significant at p = 0.001. At a C:N of 80:1, there were almost twice as many hyphae extending from the granules colonized with Th. terricola as those colonized with G. roseum. In most experiments, the hyphal growth by Th2 from granules impregnated with maltose-arginine at 80: 1 was too extensive (long and interwoven) to accurately determine hyphal length. When we removed the nylon screen from these plates, a layer of soil would often remain attached to the screen owing to the extensive hyphal network throughout the soil. Similar growth (interwoven hyphae too long to measure) was observed with maltose-nitrate at 80: 1, though less consistently. For Th2 and Gr1620 on maltose-nitrate and Gr1620 on maltose- arginine, the length of hyphal extension into soil increased as the C:N was increased from 12: 1 to 80: 1. For Th2 with maltose-nitrate, the average length increased from 166 mm at the C:N of 12: 1 to hyphae too long to measure at the C:N of 80:l. For Gr1620 with maltose-arginine, there was almost a threefold increase in the average hyphal length. With maltose- nitrate, there was no significant difference in the average length of hyphae at the three C:Ns tested. The effects of substrate, isolate, and C:N ratio, as well as their interactions, were statistically significant (p = 0.00 1). Experiments were conducted to determine the effect upon Gr1620 of decreasing the C:N by holding the concentration of carbon constant (at the level for C:N = 80: 1) and increasing the nitrogen concentration (Table 4). As expected from earlier experiments, increased carbon concentration resulted in a greater number of granules with hyphal growth (sevenfold increase). In addition, Gr 1620 was sporulating from 100% of the granules. Increasing both the carbon and nitrogen con- centrations such that the C:N remained 12: 1 resulted in a comparable increase in the proportion of granules with hyphal growth. Unlike the treatment where only the carbon con- centration was increased, however, Gr1620 did not sporulate on the granules. There were more hyphae per granules and the mean length of hyphae was greater where the nitrogen con- centration was also increased (significant at p = 0.05). Effect of carbon and nitrogen source and the C:N ratio on growth on agar Minimal salts - agar amended with carbon and nitrogen sources in various ratios was seeded with the agents to identify optimal substrates and substrate combinations for use in soil studies. Differences in preference for carbon and nitrogen source were observed on agar for both agents (Table 5). However, the growth response on agar did not correlate well with the growth observed in nonsterile soil (Tables 2 and 3). On agar, galactose was the poorest source of carbon for growth of G. roseum, yet in soil it was one of the best, yielding 100% of the carrier granules with hyphal extension and seven hyphae per carrier granule at -0.33 bars MP and 35°C. Similar results were obtained at 25°C. In contrast to the observations in nonsterile soil (Table 6), G. roseum grew to greater colony diameters than Th. terricola in equal time periods at the three C:Ns tested (Table 6). This was true whether arginine or nitrate was the source of nitrogen. Growth on agar did not predict relative performance in soil of the different antagonist strains nor did it C an . J . M ic ro bi ol . D ow nl oa de d fr om w w w .n rc re se ar ch pr es s. co m b y U ni ve rs ity o f P. E .I . o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. CAN. J . MICROBIOL. VOL. 33, 1987 TABLE 4. The effect of relative carbon and nitrogen concentrations on the growth of G. roseum (Grl620) in nonsterile soil from carrier granules impregnated with different concentrations of maltose and arginine" Molarity X % granules with: No. Mean length C:N Maltose Arginine hyphae sporulation no growth hyphaelgranule of hyphae (pm) "Lignite granules were impregnated with minimal salts medium containing maltose and arginine as the primary sources of carbon and nitrogen at C:Ns of 80: 1 and 12:l. Two concentrations of carbon and nitrogen were used to achieve a C:N of 12: 1. After colonization by G . roseum (Gr1620). the granules were air dried and placed in soil at -0.33 bar MP and 35°C. bNot detennined owing to high proportion of granules with no growth. TABLE 5. Mean colony diameter after 7 days growth on minimal salts - agar supplemented with various sources of carbon (0.02 M) and nitrogen (0.01-0.02 M) Agent Colony diameter (mm)" with various carbon sources Nitrogen source Glucose Sucrose Maltose Fructose Galactose Th. terricola (Th2) Arginine KN03 (NH412S04 Glutamic acid G. roseum (Gr 1620) Arginine KN03 (NH4)2S04 Glutamic acid "Mean of five replicate plates. TABLE 6. Colony diameter after 10 days on a defined medium with maltose and nitrate or arginine at three C:Ns Colony diameter (mm)a at various C:Ns Agent Carbon Nitrogen 80: 1 40: 1 12: 1 G. roseum (Gr 1620) Maltose NO3 55 54 49 Maltose Arginine 57 57 56 Th. terricola (Th2) Maltose NO3 19 18 16 Maltose Arginine 35 34 25 "Mean of five replicate plates per treatment. predict the relative performance in soil of different carbon and nitrogen sources or ratios. Discussion Colonization of a soil matrix by an introduced agent would comprise at least three components of hyphal dispersal: the proportion of the agent-carrier population from which hyphae emerge, the number of hyphae extending into soil from each agent-carrier unit, and the length of hyphal extension into soil from the agent-carrier unit. Also important, but not considered in this study, is the hyphal branching pattern in soil. To adversely affect plant pathogen propagules in soil, an antagonist must first reach the target propagules. In this study, several parasites of sclerotia of A. jlavus and Ph. omnivorum were evaluated for their ability to colonize soil relative to the above criteria. Also, an attempt was made to determine factors that affect the three components of dispersal. The sources of carbon and nitrogen significantly affected the growth of antagonist species in nonsterile soil and on agar. An attempt was made to utilize the growth response on agar as a preliminary screen to identify the optimum sources of carbon and nitrogen for each antagonist and to predict their relative performance in natural soil. However, there was little correlation of growth response on agar to the growth response in soil. Galactose, a good source of carbon for hyphal elongation of G . roseum in soil, was the poorest source for growth on agar. It also was evident that the ability to colonize the substrate-impregnated carrier granules in the preparation stage was not an indication of performance in soil. For certain antagonist-substrate combina- tions, the antagonist sporulated profusely subsequent to addition to soil, but did not grow from the granule into the soil. This was true with isolates of Trichoderma and Gliocladium . Although the substrate was good for biomass production, it was not suitable for achieving the desired growth response (i.e., hyphal extension). The initial concentrations of carbon and nitrogen were consistent with the requirements of fungi for growth and sporulation in culture (Griffin 198 1). Increasing the carbon concentration (0.02 to 0.18 M) such that the C:N ratio of the carrier substrate (maltose-arginine or maltose-nitrate) increased from 12: 1 to 80: 1 resulted in enhanced dispersal in soil, as indicated by a greater percentage of carrier granules from which growth occurred, a greater number of hyphae extending from each carrier granule, and an increased length of the hyphae extending into the soil. By simultaneously increasing the nitrogen source concentration (0.006 to 0.042 M) such that the C:N ratio remained 12: 1 ,.a further increase in the number of hyphae per carrier granule and a greater mean length of hyphae extending into soil resulted. Even after subsequent experiments, it was difficult to determine whether the effects were a result of C an . J . M ic ro bi ol . D ow nl oa de d fr om w w w .n rc re se ar ch pr es s. co m b y U ni ve rs ity o f P. E .I . o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y. STACK ET AL. 63 1 an increase in molar concentration of the carbon or nitrogen source, or a result of a change in the C:N ratio. There are several reports on the effects of the sources and ratios of carbon and nitrogen on the growth rate and habit of fungi (Trinci 1984; Thompson 1984; Watkinson 1984). Most studies were conducted on artificial media; others were con- ducted on natural substrates such as wood. In these studies, the C:N ratio affected the branching pattern and the rate of elongation of hyphae. In trying to encourage a more thorough colonization of soil, these are the precise growth attributes we want to effect. This should lead to a more thorough colonization of the soil and increase the probability of contact with the target propagule. Soil moisture also significantly affected dispersal of the soilborne fungi. By determining the effects of soil moisture on dispersal, we will be in a better position to integrate application of a biocontrol agent with management practices, such as irrigation and (or) to take advantage of natural rainfall. The timing of application of the agent with respect to the environment and the crop will be critical to maximize the agents' performance. There have been few cases of effective control of soilborne plant pathogens by the introduction into soil of antagonists to those pathogens. Many reasons have been suggested to explain this lack of success. Some feel that introducing an alien organism, e.g., antagonist, is predisposed to failure because the introduced organism is inferior to the native soil microflora (Alexander 197 1; Garrett 1956). It has also been stated that reintroduction of indigenous antagonists is predisposed to failure because a give soil will support only a given population (Baker and Cook 1974; Garrett 1956). However, their con- clusions are based on the behavior of microorganisms in natural ecosystems. There is great potential for manipulation of agroecosystems to establish conditions that will allow the introduced biocontrol agent to overcome natural barriers to proliferation. Site modification has been achieved by solarization (Katan 1980), fumigation (Elad et al. 1982), and irrigation. Modification at the microsite level could be accomplished by use of specific carrier substrates that elicit desired agent responses. It has been stated that, before we can develop effective biological control practices, we must first learn as much as possible about the life history of the plant pathogen (Baker and Cook 1974; Cook and Baker 1983; Doupnik 1984'). It can be argued that we must learn equally as much about the antagonist (Ayers and Adams 1981; Cook and Baker 1983; Kenerley and Stack 1986). Similarities between mycoparasites and plant pathogens have been reported with respect to mechanisms of pathogenesis, e.g., prepenetration and penetration activities (Manocha 1981). The analogy could be extended to include other factors which make a pathogen effective, e .g. , production and dispersal of inoculum. This study provides information on dispersal of soilborne fungi and factors that influence dispersal in soil. Acknowledgements This study was supported by the United States Agency for International Development, Department of State, grant no. '~oupnik, B. 1984. Sorghum root and stalk rots, a critical review: Proceedings of the Consultative Group Discussion on Research Needs and Strategies for Control of Sorghum Root and Stalk Rot Diseases. International Crops Research Institute for the Semi-Arid Tropics, Patancheru, A.P. 502 324, India. DAN 4048-G-55-2065-00, and the Texas Agricultural Experi- ment Station. W e thank Curt Cocking, Jeanne Cortese, and Malinda McMurray for technical assistance and Trey Richard- son, Institute of Statistics, Texas A & M University, for assistance with the statistical analyses. AITCHISON, J. 1955. On the distribution of a positive random variable having a discrete probability mass at the origin. J. Am. Stat. Assoc. 50: 901 -908. ALEXANDER, M. 197 1. Microbial ecology. John Wiley & Sons, New York. AYERS, W. A., and ADAMS, P. B. 1979. Mycoparasitism of sclerotia of Sclerotinia and Sclerotium species by ~ioridesmium sclerotivorum. Can. J. Microbiol. 25: 17-23. 198 1. Mycoparasitism and its application to biological control of plant diseases. Beltsville Symp. Agric. Res. 5: 91- 103. BAKER, K. F., and COOK, R. J. 1974. Biological control of plant pathogens. W. H. Freeman and Company, San Francisco, CA. BUTLER, G. M. 1984. Colony ontogeny in basidiomycetes. In The ecology and physiology of the fungal mycelium. Edited by D. H. Jennings and A. D. M. Rayner. Cambridge University Press, Cambridge. pp. 53-71. COOK, R. J . , and BAKER, K. F. 1983. Introduction of antagonists for biological control. In The nature and practice of biological control of plant pathogens. American Phytopathological Society Press, St. Paul, MN. pp. 282-3 11. ELAD, Y., HADER, Y., CHET, I . , and HENIS, Y. 1982. Prevention with Trichoderma harzianum Rifai aggr., of reinfestation by Sclerotium rolfsii Sacc. and Rhizoctonia solani Kuhn of soil fumigated with methyl bromide, and improvement of disease control in tomatoes and peanuts. Crop Prot. 1: 199-21 1. GARRETT, S. D. 1956. Biology of root-infecting fungi. Cambridge University Press, Cambridge. GRIFFIN, D. H. 198 1 . Fungal physiology. John Wiley and Sons, Inc., New York, NY. JONES, R. W., PETTIT, R. E., and TABER, R. A. 1984. Lignite and stillage: Carrier and substrate for application of fungal biocontrol agents to soil. Phytopathology , 74: 1 167- 1 170. KATAN, J. 1980. Solar pasteurization of soils for disease control: status and prospects. Plant Dis. 64: 450-454. KENERLEY, C. M., and STACK, J. P. 1986. Influence of assessment methods on selection of potential fungal antagonists of the sclero- tium-forming fungus Phymatotrichum omnivorum Can. J. Micro- biol. 33: This issue. MANOCHA, M. S. 198 1. Host specificity and mechanism of resistance in a mycoparasitic system. Physiol. Plant Pathol. 18: 257-265. OWEN, W. J., and DEROUEN, T. A. 1980. Estimation of the mean for log normal data containing zeros and left-censored values, with applications to the measurement of worker exposure to air contaminants. Biornetrics, 36: 707-7 19. PENNINGTON, M. 1983. Efficient estimators of abundance for fish and plankton surveys. Biornetrics, 39: 28 1-286. STACK, J. P., KENERLEY, C. M., and PETTIT, R. E. 1985. Growth potential in soil of hyperparasites of sclerotia. Phytopathology, 75: 1 344- 1 345. THOMPSON, W. 1984. Distribution, development and functioning of mycelial cord systems of decomposer basidiomycetes of the deci- duous woodland floor. In The ecology and physiology of the fungal mycelium. Edited by D. H. Jennings and A. D. M. Rayner. Cambridge University Press, Cambridge. pp. 185-2 14. TRINCI, A. P. J. 1984. Regulation of hyphal branching and hyphal orientation. In The ecology and physiology of the fungal mycelium. Edited by D. H. Jennings and A. D. M. Rayner. Cambridge University Press, Cambridge. pp. 23-52. WATKINSON, S. C. 1984. Morphogenesis of the Serpula lacrimans colony in relation to its function in nature. In The ecology and physiology of the fungal mycelium. Edited by D. H . Jennings and A. D. M. Rayner. Cambridge University Press, Cambridge. pp. 165-184. C an . J . M ic ro bi ol . D ow nl oa de d fr om w w w .n rc re se ar ch pr es s. co m b y U ni ve rs ity o f P. E .I . o n 11 /1 3/ 14 Fo r pe rs on al u se o nl y.