AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 152:273–286 (2013)
Hunter-Gatherer Variability: Dental Wear in South Australia Judith Littleton,1* Rachel Scott,1 Gina McFarlane,1 and Keryn Walshe2 1 2
Department of Anthropology, University of Auckland, Auckland, New Zealand South Australian Museum KEY WORDS
dental attrition; Australian Aborigines; diet; teeth as tools
ABSTRACT Often it is assumed that huntergatherer dentitions are dominated by heavy attrition. Recent analyses, however, have shown unexpected variability in the pattern of wear between groups. It had been previously noted that wear differed between neighboring groups on the Murray River, Australia. This analysis extends that geographic scope as well as focusing on wear across the dentition, including the premolars. The samples came from coastal and riverine regions of southern Australia. The analysis used records from the Yorke Peninsula, Adelaide Plains (Gillman site), and Euston regions. These were compared with previously published work from the Adelaide Plains and four locations on the Murray River. The results confirm the overall severity of wear but reveal systematic differences between the sam-
ples in terms of the pattern of wear. Heavy wear on the incisors and canines is observed among males from the Euston, Kaurna, Middle A, Murray Mouth, and Yorke Peninsula samples but with marked intra-individual variability. Extensive premolar wear is noted among females from Kaurna and Middle B samples as well as among males and females from Euston. It is argued that these patterns relate to gendered non-masticatory use of teeth and reliance upon bulrush (Typha spp.) and related species for both food and fiber among some groups. We argue that analyzing the degree of variability within samples and across all teeth provides a more nuanced understanding of dental wear among huntergatherers. Am J Phys Anthropol 152:273–286, 2013. VC 2013 Wiley Periodicals, Inc.
It is commonly assumed that hunter-gatherer dentitions are characterized by heavy wear and very little caries (Turner, 1979; Lukacs, 1989). Recent work, however, has shown how such generalizations may obscure variability within and between closely related groups (Keenleyside, 1998; Bernal et al., 2007; Lieverse et al., 2007). Yet hunter-gatherer samples are frequently small and can be lacking in precise provenience, making it sometimes difficult to identify and interpret differences in dental wear and pathology. This is especially the case when some of those same collections are no longer available for study, as is the case in Australia. In this article we use two approaches to the analysis of dental wear to address these issues. First, we focus on the pattern of wear across the dentition (looking at wear relative to the first molar), which circumvents the problem of unaged samples. Second, we pay attention to the degree of variability within samples rather than only focusing upon the average degree of wear for each tooth among sample members. This approach provides insight into the degree of constraint and variation within populations which may vary across time and space. We apply these two approaches to the study of dental wear from a large area of southern Australia to examine inter- and intra-regional variation in dental wear. Many people have been fascinated by dental wear of Aboriginal people in Central and Southeastern Australia. In the 19th century European colonists noted the large, “generally very perfect and beautiful” (Eyre, 1845 II:206) teeth of Aboriginal people and the extreme wear of teeth among old individuals. Angas (1847) included a drawing of different Aboriginal mouths in his collection of drawings from Southern Australia, while newspaper articles also touched on the subject (Taplin, 1879 II:109). Research in the 20th century focused upon the develop-
ment of wear in relation to tooth development and age (Richards and Miller, 1991), non-masticatory use of teeth (Barrett, 1977; Brown and Molnar, 1990) and regional variation in dental wear and its possible relationship to dietary variation (Brown, 1978; Richards, 1984; Smith et al., 1988; Molnar et al., 1989; Elvery et al., 1998). In particular, Richards (1984) compared two coastal populations occupying adjacent territories (Kaurna and the Narrinyeri), and Molnar et al. (1989) noted significant differences in the patterning of dental wear when comparing these two samples with four from the Murray River Valley. While all groups experienced moderate to heavy wear, some river groups had more severe wear than others, while wear on the central incisors and canines was marked in Kaurna but not so heavy in the neighboring coastal group. Molnar et al. (1989) suggested that among the Murray River groups there was a division between groups reliant upon terrestrial resources, particularly seed grinding, and those that relied upon more riverine resources, particularly tubers from the river swamps, mussels and fish. The differences in the rate of anterior tooth wear were attributed to differences in the extra masticatory use of teeth although there did not seem to be any apparent differences in the two tool kits of the coastal groups.
Ó 2013 WILEY PERIODICALS, INC.
*Correspondence to: Judith Littleton, Anthropology, The University of Auckland, Private Mail Bag 92019, Auckland, New Zealand. E-mail:
[email protected] Received 18 December 2012; accepted 23 July 2013 DOI: 10.1002/ajpa.22358 Published online 3 September 2013 in Wiley Online Library (wileyonlinelibrary.com).
274
J. LITTLETON ET AL.
Fig. 1. Location of the samples used in the analysis.
A difficulty of the study, however, was that the samples were unaged, so a potential cause of differences was variation in the age composition of the samples. Furthermore, it is now possible to expand the regional coverage of the analysis since samples from the Yorke Peninsula and a single site at Adelaide (Littleton and Scott, n.d.) as well as a sample from Euston in the central Murray Valley (Brown, 1978) have been independently analyzed. This wider regional coverage has the potential to more clearly identify the differences between groups in both masticatory and extramasticatory wear. However, because each author used a different method of scoring wear or different techniques of analysis, we had to devise a way of achieving comparability. Reanalyzing all samples was not possible since many have now been repatriated and are not accessible. This article therefore is a comparison of previously studied materials from Southeastern Australia to explore the extent and underlying causes of heterogeneity in dental wear between and within Aboriginal hunter-gatherers from Southeastern Australia and the implications of that for dietary reconstruction.
THE AREA The study area encompasses the central to lower Murray River, the Adelaide Plains and the Yorke Peninsula (Fig. 1). While all samples lie primarily on semi-arid plain, they vary in their access to water. The Yorke Peninsula, Adelaide Plains (Kaurna/Gillman) and Murray River mouth (Narrinyeri) samples come from contiguous areas of the southern coastline (Fig. 1). This coastline is backed with open woodland/shrubland. While the Yorke Peninsula is surrounded by rocky platforms, the Murray Mouth and Adelaide Plains areas include estuarine areas allowing for swamps and the exploitation of terrestrial, marine, and riverine resources (Littleton and Scott, n.d.). This is particularly true of the Narrinyeri region, which includes Lake Alexandrina, a major freshwater American Journal of Physical Anthropology
lake in the past (Murray-Darling Basin Commission, 2003). The Murray River samples (termed Middle A, Middle B and Euston) came from the central Murray where the major water source is the river. Prior to water regulation it is estimated that the river fell to low flows around 7% of the time (Murray-Darling Basin Commission, 2003), so droughts could and did occur. Nevertheless, the riverine groups are assumed to have had higher population densities than their non-riverine neighbors (Birdsell, 1953). Each river group, however, had a different range of resources they could rely upon. In the Middle A area, the Murray River lies within a low limestone gorge between 400 and 1600 m wide. This gorge constrains the flood plain and limits the relationship between the river (including river swamps) and terrestrial resources. Around Middle B, which comprises the Chowilla Floodplains and lower Darling River, the flood plain is much wider (up to 10 km wide) and there is greater interdigitation of riverine and terrestrial resources. In the past, the area was subject to natural cyclic flooding between one in every ten and one in every two years. Euston lies at the beginning of the riverine plain, close to the junction of the Murrumbidgee and Murray Rivers and at the intersection of the mallee woodland with chenopod scrubland. This area flooded more regularly in the past due to spring snowmelt and supports some wetland regions (Scott et al., 2003). Therefore, the study region is generally drier from east to west, but all sample locations lie within areas of high rainfall variability and the semi-arid zone. In each region resources are varied, combining coastal and terrestrial for Yorke Peninsula and the Adelaide Plains, coastal/riverine and terrestrial for the Murray mouth, and riverine and terrestrial for the Murray River samples. On the basis of this biogeographic diversity we predict that there should be some differences between the riverine samples (Middle A, Middle B, and Euston) and the mixed coastal/terrestrial groups (Yorke Peninsula,
275
DENTAL WEAR IN SOUTH AUSTRALIA
TABLE 1. Sample composition including the total number of adults and subadults, the number of sexed adults, and the number of adults for whom aged could be assessed
Yorke Gillman Kaurna Narrinyeri Middle A Middle B Euston 1 Euston 2
Sub-adult
Total Adult
Male
Female
21–35 yrs
36–50 yrs
511 yrs
Source
4 7
20 13 107 30 35 42 47 33
7 6 59 13 18 18 27 15
11 4 58 17 17 24 20 18
4 3
8 5
4 1
This study This study Richards (1984) Molnar et al.(1989) Molnar et al.(1989) Molnar et al.(1989) Brown (1978) Molnar et al.(1989)
Adelaide Plains), with Murray mouth samples possibly holding an intermediate position given suggestions of internal differences in diet between coastal and inland Narrinyeri populations (Pate, 1998; Owen, 2004).
MATERIALS AND METHODS The Samples The human remains from Yorke Peninsula are a subset of a 20th century collection of individuals deposited in the South Australian Museum. These remains have been uncovered through erosion, in the course of construction, or during other forms of disturbance. Commonly they were sent by the police to the museum. As a result they come from a range of locations, noted with varying degrees of precision, and their antiquity is unknown. Based on the degree of mineralization and dating of at least three individuals, most (but not necessarily all) probably belong to the late Holocene period, i.e., the last 3,000 years. There is at least one postcontact individual in the sample identified by a circular clay pipe facet. These remains were recorded by us in the South Australian Museum as part of a collaborative project run with the Narungga community. The aim of the project is to record and analyze the remains for the community prior to decisions about repatriation. There are two samples from the Adelaide Plains occupied by the Kaurna people: Gillman and Kaurna. Gillman refers to a series of burials excavated from a mound and the surrounding area in 1970 (Hodges, 1973). Hence this is a group which has a very precise burial location, dated to between 1200 and 500 BP (Walshe et al., 2011). In contrast, the Kaurna sample analyzed by Richards (1984) includes remains from across the Adelaide Plains and is, like the Yorke Peninsula remains, undated. Again, these most probably date to the late Holocene. The Murray mouth sample comes primarily from the site of Swanport on the Murray River rather than the coast. Stable isotope analysis indicates that people buried at this site relied more upon lake than coastal resources (Owen, 2004). Swanport was the site of an excavation by Stirling (1911). While he assumed that the burials were post-contact, more recent radiocarbon dating assigns a time period of c3000 to 500 BP indicating long term accumulation (Pate et al., 2003). Richards and Molnar both analyzed remains from the lower Murray and its mouth (Richards, 1984; Molnar et al., 1989). Further upriver is a sample used by Molnar et al. (1989) labeled Murray River A. This is a group of remains from the Renmark region and like the Yorke and Adelaide Plains samples, is the result of collecting
isolated burials or burials within small groups. These remains are undated and have only general proveniences. Following the convention used by Molnar et al. (1989) and later Webb (1995), they are assumed to be primarily late Holocene in date. The group of remains termed Murray River B by Molnar et al. (1989) comes from the area around the junction of the Darling and Murray Rivers as well as Lake Victoria. Many of these remains were part of the George Murray-Black collection (Sunderland and Ray, 1959), collected by him from isolated burials as well as burial sites during the 1940s and 1950s. These remains have been returned to Aboriginal communities. The final group of remains from Euston are also part of the George Murray-Black collection and were studied by Brown (1978) and by Molnar et al. (1989). These authors, however, used two different subsamples. The composition of these samples is given in Table 1. With the exception of the Gillman and Yorke samples, remains were either not aged beyond the attribution of adult or the determination of age relied upon dental wear itself or cranial suture ageing, which Brown (1978) noted had a high degree of error. This, of course, restricts the type of analyses which can be undertaken in the current study.
METHODS Dental wear has been recorded in different ways in these samples. It is, however, possible to derive comparative data. Gillman and Yorke were scored using a 10 point scale based on the degree of dentine exposure and applied to all teeth. This scoring system was adapted from Scott (Scott, 1979a; Littleton and Frohlich, 1993) and is shown in Table 2. Brown (1978) devised a scheme for Euston based on the progressive pattern of wear from one to nine. His reason for scoring was to assess patterns of aging; hence his scoring system scores each tooth independently but does not apply the same score to the same degree of wear on different teeth types. For example, using Scott’s system, any molar with only pinprick exposure of the dentine would be scored 4 (Table 2). In Brown’s system the comparable degree of wear would be scored as 1 for the first molar, 3 for the second.. . . etc. However, as with Scott, the degree of wear is shown by clear diagrams which allow his records to be recalibrated to the Scott scores. Figure 2 shows an example of this while Table 3 tabulates the comparison. This means that the Euston data from Brown can be analyzed with the Yorke and Gillman material, permitting comparison of both the average degree of wear per tooth and of individual patterns of wear based on the calculation of gradients (explained below). American Journal of Physical Anthropology
276
J. LITTLETON ET AL. TABLE 2. The modification of Scott’s (1979) scoring method used here (based on Littleton and Frohlich 1993, Fig. 2)
Score
Molars
0 1
No data. Unworn: minimal wear facets. Large wear facets but cusps still present, pinprick dentine exposed Cusps becoming obliterated rather than clearly defined. Quadrant flat – no dentine exposure rather than pinpricks. Quadrant flat with dentine exposure on 1=4 of quadrant.
2 3 4 5
Premolars
Canine/Incisor
No data. Unworn: minimal wear facets. Wear facets large but cusps still present.
No data. Unworn: minimal wear facets. Wear facets large but incisal edge still clear.
Cusps rounded rather than clearly defined.
Incisal edge rounded rather than clearly defined.
Cusp pattern flat, pinprick dentine exposure.
Incisal edge flat, pinprick dentine exposures.
Cusp pattern flat, dentine exposed on c. 1=4 of cusp.
Edge flat, small dentine exposure.
6
Dentine exposure more than 1=4 of quadrant.
21 dentine exposures clearly separated.
Dentine exposure on 1=41 of incisal surface.
7
Enamel on only 2 sides of quadrant.
21 dentine exposures
Extensive dentine exposure.
8
Enamel on only one side of quadrant but thick.
Enamel still around edge of tooth, thick.
Enamel around edge of tooth, thick.
9
Enamel very thin.
Enamel very thin.
Enamel very thin.
10
No enamel on quadrant.
No enamel on cusp.
No enamel on surface.
Each quadrant (of the molars) or cusp (of the premolars) is scored separately and averaged back to 10.
The other samples were recorded through quantitative analysis of the amount of dentine exposure, digitally recorded (Richards, 1984; Molnar et al., 1989). Although this limits the nature of direct comparison, it is possible to analyze tooth wear for these samples using gradients of wear. Patterns of wear cannot be assessed for each individual but can be assessed for males and females within each sample collectively. This analysis focuses particularly upon the shape and rates of wear since wear becomes more severe with age, creating a major problem when comparing individuals with different ages at death. The Yorke and Gillman samples are the only samples in which age at death has been recorded. The rate at which wear occurs, however, could be calculated for each individual in the Gillman, Yorke, and Euston samples. This is a measure of how much wear accrues between the eruption of the first and the second molars. It is commonly compared between the molars, particularly the first and second molars, when timing of eruption is regularly estimated to be separated by 6 years and where the masticatory forces upon both teeth are similar (Scott, 1979b; Lev-Tor Chattah and Smith, 2006). Gradients were calculated by dividing the degree of wear on each tooth relative to the wear on the first American Journal of Physical Anthropology
molar of that same arch and converting the result to a percentage (e.g., M2/M1 * 100). A gradient close to 100% would suggest that the first molar had worn very little prior to the eruption of the second molar, i.e., a slow rate of wear, whereas a gradient of 50% suggests extensive wear on the first molar prior to eruption of the second. Statistical comparison was undertaken using the second molar to first molar wear gradients (the posterior gradient), the wear on the first premolar relative to the first molar (the centro-posterior gradient), and the anterior wear gradient (central incisor relative to M1). Statistical analysis was undertaken using IBM SPSS statistics software. Samples were compared using T-tests for independent samples and univariate analysis of variance with the Bonferroni correction for multiple comparisons. Normality of the distributions was tested using Q-Q plots. One possibility is that because of the comparison of samples scored quantitatively with those scored qualitatively the resultant patterns of wear might be different. It was, however, possible to assess this in two ways. First, different segments of the Euston sample were scored by both approaches. Second, the Kaurna and Gillman samples come from the same region so these could also be compared.
277
DENTAL WEAR IN SOUTH AUSTRALIA
TABLE 4. Percentage of maxillae and mandibles by age group whose maximum wear is slight, moderate or heavy
10–15 16–20 21–35 36–50 511 Fig. 2. An example of the rescoring of Brown’s (1978) wear stage 5 to Scott’s method as used here. The Scott’s scores are shown below the appropriate tooth. Full comparison of each score is given in Table 3.
Gillman Yorke Gillman Yorke Gillman Yorke Gillman Yorke Gillman Yorke
Slight wear
Moderate wear
50.0 100.0 100.0 50.0
50.0 50.0 25.0 100.0 26.7 33.3
Heavy wear
75.0 100.0 73.3 100.0 66.7
n 2 2 1 2 4 7 7 15 2 3
Slight wear (Classes 1–4) represent wear to pinprick exposure of dentine, moderate wear is represented by wear grades 5–7 while heavy wear represents teeth with only a marginal rim of enamel or no enamel left on the surface at all.
TABLE 3. The comparison of wear stages used by Brown (1978) and the Scott’s scoring method used here Scott’s Scoring Brown’s Wear Stage 1 2 3 4 5 6 7 8 9
Anterior
Premolars
M1
M2
M3
3 5 5 6 7 7 8 9 10
2 4 5 6 7 7 8 9 10
4 5 5 6 7 8 8 9 10
3 3 4 5 6 7 8 9 10
2 2 2 3 4 5 7 8 9
RESULTS Comparison With Age As expected, teeth with extensive dentine exposure were observed in all of the samples. Based on the maximum degree of wear within the maxillae and mandibles, most young adult dentitions from Gillman had only a thin marginal rim of enamel on their first molar (Table 4). All older mandibles and maxillae from Gillman had teeth at this stage of wear (Fig. 3). Wear was less advanced among the Yorke Peninsula. The ultimate stage of wear (tooth roots functional only) was not observed among the sample. Comparisons by age could not be made for the other samples from the region.
Individual Wear Gradients In accordance with the greater wear by age observed at Gillman, the gradient of wear between the first and second molars is steep for both the maxilla and mandible (Table 5). Wear on the M2 relative to the M1 is between 74 and 77%. Females appear to have steeper gradients than males, suggesting faster wear, but the difference is not statistically significant once P-values are corrected for multiple comparisons. Except for the male maxillary gradient, the separate coefficients of variation for Gillman males and females are much smaller than those for the total sample, suggesting that there may be distinct gender patterns among this group. Individuals from Yorke Peninsula have a similar wear gradient to
Fig. 3. Marked anterior wear on individual from Gillman Mound. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Gillman for the mandible but a lower gradient for the maxilla, and unlike the other two samples, maxillary gradients are less steep than the mandibular gradients. Among Yorke Peninsula females the wear gradient appears steeper than among males, suggesting more rapid wear, but this difference is not statistically significant. The lowest gradient for all samples is recorded among the Euston mandibles where there is little difference between the extent of wear on the first and second molar. This pattern is responsible for the statistically significant difference between the three samples for the mandible (F 5 4.29, P 5 0.018). Maxillary molars from Euston show a similar gradient to Yorke Peninsula. There is no statistically significant difference between males and females from Euston for either the maxilla or mandible. Plotting the average degree of wear by tooth (Table 6) suggests fundamental differences between the groups (Fig. 4) that are captured by calculating the anteroposterior gradient, measured by the wear on the central incisor relative to the first molar, and centro-posterior American Journal of Physical Anthropology
278
J. LITTLETON ET AL. TABLE 5. Antero-posterior gradients and molar gradients in dental wear Maxillary N
Molar Gradient (M2/M1*100) Gillman Yorke Euston Gillman Females Gillman Males Yorke Females Yorke Males Euston Females Euston Males Antero-Posterior (I1/M1*100) Gillman Yorke Euston Gillman Females Gillman Males Yorke Females Yorke Males Euston Females Euston Males Centro-posterior (P1/M1*100) Gillman Yorke Euston Gillman Females Gillman Males Yorke Females Yorke Males Euston Females Euston Males
Mean
Mandibular SD
CV
N
Mean
SD
CV
10 17 46 4 4 6 9 19 27
75.22 90.03 82.74 71.74 84.38 79.50 94.28 84.28 81.65
15.98 33.44 15.41 8.48 18.75 19.04 41.73 15.24 15.73
0.21 0.37 0.19 0.12 0.22 0.24 0.44 0.18 0.19
7 12 47 3 3 4 6 20 27
78.51a 76.25 91.17 75.79a 96.30 65.00 75.83 88.61 93.07
23.37 17.95 16.86 9.55 6.42 14.01 16.25 16.71 17.04
0.30 0.24 0.18 0.13 0.07 0.22 0.21 0.19 0.18
5 8 34 1 4 4 4 14 20
108.33a 76.22 95.10 87.50 113.54 73.27 79.17 92.96 96.61
14.73 14.86 23.03 . 10.42 5.64 21.46 8.37 29.46
0.14 0.19 0.24 0.09 0.08 0.27 0.09 0.30
5 5 39 2 3 3 1 15 24
108.06 83.89 93.03 100.00a 113.43 72.22 77.78 92.71 93.22
7.63 29.29 16.06 0.00 2.89 25.46 – 9.14 19.37
0.07 0.35 0.17 0.00 0.03 0.35
9 14 45 4 3 6 7 19 26
102.01c 73.72 106.27 82.99c 108.33 72.46 73.91 101.87c 109.48a
24.14 19.18 32.07 14.18 7.22 19.76 21.53 21.38 38.15
0.24 0.26 0.30 0.17 0.07 0.27 0.29 0.21 0.35
6 10 47 2 3 4 4 20 27
89.22c 67.24 95.21 73.21b 96.30 62.50 59.17 98.58a 92.72
13.17 19.64 18.38 2.53 6.42 15.96 17.72 19.01 17.85
0.15 0.29 0.19 0.03 0.07 0.26 0.30 0.19 0.19
0.10 0.21
Bolded figures represent statistically significant differences using ANOVA. P < 0.05 b P < 0.01 c P < 0.005. a
gradient measured by the wear of the first premolar relative to the first molar (Table 5). The Gillman sample has heavy anterior wear (Table 5). The incisors are on average worn more extensively than the first molar. This pattern is visible among the males rather than females and is statistically significantly different for the mandible although the sample size is very small. In contrast, the first premolars are, on average, worn marginally less than the first molars, which is an expected pattern given that they erupt later. Females have steeper gradients, suggesting more rapid wear on the first molar than the premolars. Males, however, have heavier central than first molar. These gradients are statistically significantly different. Among Yorke Peninsula individuals, anterior wear is less relative to the posterior dentition (Table 5). This is true also for the premolars compared with the molars. There are no statistically significant differences between the sexes. Wear is centered upon the molars in this group. The Euston sample shows a different pattern (Table 5): wear on the incisors is only marginally less than on the molars. This does not differ between males and females although wear of male maxillary incisors is generally more advanced and certainly much more variable. The first premolars are worn equivalently to the first molars or even slightly more in the case of the maxilla. The overall pattern among Euston individuals is of the teeth forming a single, even plane of wear. American Journal of Physical Anthropology
Individual Variation When the wear gradient for each individual is plotted, it becomes apparent how these average patterns of wear are generated. The Gillman sample (Fig. 5) is small and the heavy anterior wear is observed almost exclusively among the males. Also considering the males, the coefficients of variations are small, indicating a lack of variation between individuals (Table 5). In contrast, the Yorke sample is more variable (as seen in the coefficients of variation), particularly among the males (Fig. 6). Variation among the Euston samples is shown through a boxplot (Fig. 7) rather than plotting each individual. This sample also shows greater variation among the anterior teeth of males relative to females and with heavier wear on average affecting these teeth. What also is striking about Euston is the extent of wear on the premolars with many individuals showing wear on the first and second premolar that exceeds the first molar despite their later eruption (Fig. 7). This pattern of wear on the anterior and central teeth is more variable for males than females as shown by the coefficients of variation (Table 5).
Comparing All Samples in Terms of Average Gradients Because of the differences between the recording methods for different groups, the only way in which the Murray River samples recorded by Molnar et al. (1989)
279
DENTAL WEAR IN SOUTH AUSTRALIA
TABLE 6. Mean degree of wear (on scale from 0 to 10) by tooth within each sample (n in brackets) including a comparison of the 36–50 year age group which is of sufficient size to test statistically Maxilla
Site
M3
M2
M1
P2
P1
C
I2
I1
36–50 Male Female Total
5.00(5) 5.33(6) 3.00(5) 3.73(15)
6.50(6) 6.29(7) 6.43(7) 4.81(21)
8.43(7) 8.00(7) 8.71(7) 6.86(21)
8.38(8) 8.50(6) 7.00(6) 6.06(16)
8.17(6) 8.40(5) 6.80(5) 5.65(17)
8.71(7) 7.83(6) 7.50(6) 6.28(18)
9.33(6) 8.43(7) 8.25(4) 7.43(14)
9.40(5) 9.00(6) 7.00(2) 6.83(12)
36–50 Male Female Total
4.75(12) 4.87(15) 3.43(7) 4.60(25)
6.21(14) 5.25(16) 5.08(12) 5.26(31)
8.20(15) 7.12(17) 6.42(12) 6.19(36)
5.62(13) 4.81(16) 4.33(12) 4.71(31)
5.80(10) 4.92(13) 4.44(9) 4.70(23)
6.00(7) 5.50(10) 4.00(9) 4.75(20)
6.39(7) 5.78(9) 4.29(7) 5.12(17)
6.67(6) 6.57(7) 4.25(8) 4.67(18)
Male Female Total
4.56(27) 4.89(19) 4.70(46)
5.70(27) 6.45(20) 6.07(47)
6.81(27) 7.32(19) 7.02(46)
6.74(27) 7.10(20) 6.89(47)
6.81(26) 7.37(19) 7.04(45)
6.54(26) 6.75(20) 6.63(46)
6.73(22) 6.88(17) 6.79(39)
6.95(19) 7.13(15) 7.03(34)
Site
M3
M2
M1
P2
P1
C
I2
I1
36–50 Female Male Total
5.80(5) 5.80(5) 4.83(6) 4.83(12)
8.14(7) 7.14(7) 7.33(6) 5.26(19)
8.60(5) 6.75(4) 8.00(5) 5.73(15)
8.60(5) 6.80(5) 7.60(5) 5.64(14)
8.67(6) 7.71(7) 7.60(5) 5.89(19)
8.67(6) 7.86(7) 9.00(4) 6.53(17)
9.60(5) 6.60(5) 8.80(5) 6.75(16)
9.50(4) 5.75(4) 8.67(6) 7.36(14)
36–50 Female Male Total
5.63(8) 4.13(8) 5.69(13) 5.00(23)
6.50(8) 5.56(9) 6.08(13) 5.58(26)
9.29(7) 6.63(8) 7.33(12) 6.07(27)
5.40(5) 4.57(7) 4.00(7) 4.19(16)
5.40(5) 4.50(6) 4.17(6) 4.43(14)
5.40(5) 4.56(9) 4.25(8) 4.42(19)
6.67(6) 4.80(10) 6.75(4) 5.53(15)
7.00(2) 4.00(6) 7.00(2) 4.45(11)
Female Males Total
5.21(19) 5.41(27) 5.44(46)
6.80(20) 6.96(27) 6.89(47)
7.25(20) 7.33(27) 7.30(47)
7.40(20) 6.89(27) 7.11(47)
7.05(20) 6.67(27) 6.83(47)
6.60(20) 6.72(25) 6.67(45)
6.72(18) 6.81(25) 6.79(43)
6.67(15) 6.67(24) 6.67(39)
Gillman
Yorke
Euston
Mandible Gillman
Yorke
Euston
and the two samples recorded by Richards (1984) can be compared is by constructing graphs of the gradients of tooth wear relative to the upper or lower first molar. In
this analysis there is no indication of the absolute degree of wear (which needs an age at death for comparison) but of the relative wear within a mouth. Furthermore, unlike the average gradients in Figure 4, these graphs represent the average wear of each tooth relative to the population average of molar wear. As a result the graphs may flatten some of the differences. However, there are two ways of assessing this effect. First, the Kaurna sample recorded by Richards (1984) comes from the same area as Gillman but is undated and with less precise provenience. It can be compared directly to Gillman to assess the impact of averaging. Similarly, Molnar et al. (1989) recorded a sample from Euston that can be compared to Brown’s Euston sample (Brown, 1978), allowing us to identify the difference in the two methods of recording (quantitative versus qualitative) and whether the patterns generated are comparable. In both comparisons molar wear gradients are steeper in the quantitatively recorded teeth, probably reflecting the difference in recording method, but for Gillman and Kaurna the pattern is very similar except that females rather than males in the Kaurna sample show the heavy anterior maxillary wear (Fig. 8). In the Euston sample, the two patterns are similar with the exception of the mandibular premolars which are less worn in Molnar et al.’s sample (1989). Given the individual variation in the Euston sample, this seems to reflect sampling; all other comparisons are very similar.
Comparison of Shape of Wear Across Populations Fig. 4. Average degree of wear across the mouth relative to the first molar (at 100). Left and right quadrants combined.
Combining all samples shows the diversity of tooth wear patterns over space (Fig. 9). The Yorke Peninsula American Journal of Physical Anthropology
280
J. LITTLETON ET AL.
Fig. 5. Individual relative wear scores for the Gillman males and females. All teeth relative to first molar (standardized at 100).
Fig. 6. Individual relative wear scores for the Yorke Peninsula males and females. All teeth relative to first molar (standardized at 100).
sample shows a relatively slow rate of maxillary molar wear as well as a lack of wear on the anterior and central teeth. The pattern of wear follows the sequence of eruption for both males and females. Compared to the American Journal of Physical Anthropology
other samples, this population has one of the lower gradients of molar wear (except for female mandibular molars) suggesting relatively slow wear accumulation over time.
DENTAL WEAR IN SOUTH AUSTRALIA
281
Fig. 7. Relative wear scores for the Euston males and females showing maximum, minimum, average, and one standard deviation. All teeth relative to first molar (standardized at 100).
The Kaurna and Gillman samples as noted above vary in terms of which sex has heavy anterior wear, but both are marked by a steep molar gradient indicative of a rapid rate of wear in the posterior dentition. This rate of attrition is faster than observed in the Yorke Peninsula sample and most similar to the Murray mouth sample. The anterior wear is distinctive. The Murray mouth sample also has a steep molar gradient and heavier anterior maxillary wear among males. The difference between this sample and the Adelaide Plains sample is not just the gendered pattern of anterior wear but the heavier central wear among males and females from the Adelaide Plains. Middle A populations have a moderate molar wear gradient (not as steep as Kaurna or the Murray Mouth) and heavy anterior maxillary wear, but the overall pattern is equal wear across the teeth in a pattern similar to Kaurna. In this sample women appear to have engaged in the heavy chewing upon the anterior and central dentition. The forces are centered on the maxilla. Middle B populations coming from slightly further upriver have a steeper molar gradient than for the Middle A group, coming closer to the Adelaide Plains and Murray mouth samples. Anterior wear is moderate but there is no evidence of the heavy central wear evidenced in the Middle A group. There is a female/male split with females possessing a flatter wear plane suggestive of more rapid wear across the premolars while males have marginally more anterior wear. Euston is distinctive overall because of the relatively flat wear plane, but there is heavy anterior maxillary wear for males. The wear gradients between the molars are less, especially for the mandible. In sum, when examining these populations, three broad factors are at play in the shape of wear: the rate of molar wear, the overall wear plane, and the degree of anterior wear. Within samples these patterns may mani-
fest differently for males and females. No two populations have an identical pattern, but there are some similarities between the Adelaide Plains, Murray Mouth and Middle B populations, which have steep molar gradients. In the Adelaide Plains samples, these rapid gradients keep pace with the formation of secondary dentine. The Yorke sample has the lowest average molar wear gradient, but Euston also has a relatively low gradient. The samples fall into two groups in regard to wear plane: those with a plane matching the eruption sequence (less wear on the later erupting premolars) and those with a flat masticatory plane. This distinctive pattern is most clearly seen among Euston, but is also evident in the Middle A females and both sexes in the Adelaide Plains samples. Finally, anterior wear is distinct and has been shown to be variable between and within males and females for the Adelaide Plains (Kaurna and Gillman) and Euston samples. For the Murray mouth sample, males have heavy anterior wear. Heavy anterior wear is not apparent among either the Yorke Peninsula and Murray River Middle B populations. Anterior wear shows the greatest differences between the sexes in the samples. These patterns are summarized in Table 7.
DISCUSSION Interregional Variation All of these samples feature heavy wear due to toothon-tooth attrition, abrasion, and their consequences. The analysis demonstrates, however, that comparing averages between populations tends to obscure significant differences and that considering all the teeth (rather than just molars) yields valuable information. There are, however, problems with this form of analysis. For example, rate and pattern of wear also need to take into American Journal of Physical Anthropology
282
J. LITTLETON ET AL.
Fig. 8. A comparison of the wear gradients for the Gillman males and females recorded using Scott’s scoring system compared to the broader Kaurna sample recorded by Richards (1984) using digital recording of the amount of dentine exposed.
account the full extent of wear. A low molar gradient for instance suggests less wear accumulating over the sixyear period between eruption of the first and second molars, but very severe wear on the first molar can lead to stresses being transferred to the other teeth. This conundrum is apparent in the Euston sample where there is a low molar gradient but overall wear is severe in the sample. There are two related possibilities. First, whatever generates the distinctive flat chewing plane becomes an important activity or dietary component after all teeth are erupted. Alternatively, the flat chewing plane may be generated by a large bolus which transfers stressors across the teeth more evenly. Unfortunately the age data that could distinguish some of these possibilities are not available for these samples. Of the groups considered here, the Yorke Peninsula sample appears distinctive because of the lack of central wear and very limited anterior wear. In addition, there is marked chipping among some of the individuals from this sample, which seems to be related to pitting as seen in dental microwear (Littleton and Scott, n.d.). This pattern seems to suggest a diet and/or a set of activities
American Journal of Physical Anthropology
which involve less heavy chewing, although there are clearly still abrasives in the diet given the presence of chipping particularly of the molars. Microwear analysis confirms both a high degree of variability between individuals and a high frequency of pits relative to striations (Littleton and Scott, n.d.). Schmidt (2009) also notes heavy pitting associated with less macrowear. The variability between Yorke Peninsula individuals suggests that there is not one single dominant resource or activity. Instead, it seems on the Yorke Peninsula people were probably exploiting a range of resources and not heavily relying on processing vegetable matter that could leave striations upon the teeth. All of the remaining groups have heavy molar wear (with the exception of the Euston mandibular molars mentioned above). There is, however, a division between the groups with a flat plane of wear on the anterior and central teeth (all individuals from Euston, females particularly from Middle A and Adelaide Plains) and those without that distinctive plane (Murray Mouth and Middle B). This division does not correspond to a broad ecological division of coastal versus riverine groups. Molnar et al. (1989) suggested that this difference may be due to the greater diversity of ecological zones in the Middle B region of the Murray or because of the reliance of these groups upon seed grinding as described by Allen (1974). Neither explanation, however, explains the Murray Mouth people, who—based on access to resources— ought to have a similar pattern of wear to people from the Adelaide Plains. The overall comparison points to the possibility of a single resource or set of activities on the teeth creating distinctive patterns of wear. Analysis of dental wear often focuses upon the molars or the anterior teeth, ignoring the premolars. These will wear in response to a large bolus moving freely across the mouth or if they are being used to strip fibers and are subject to strong lateral forces (Hall, 1976; Hinton, 1981; Richards and Brown, 1981; Molnar et al., 1989; Taylor, 1991).
The Role of Single Resources The distinctive flat plane is clearest at Euston and is accompanied by marked striations as well as calculus (Brown, 1978). Such wear suggests intensive use of resources that create a coherent and dominant pattern. There are two possible explanations: heavy chewing and/ or the processing of fibers. In the Gillman sample, microwear analysis revealed the same dominance of striations among both men and women (Littleton and Scott, n.d.). Such a detailed level of data is not available for the Middle A sample, but women are marked by a flatter plane of wear than males. We hypothesize that, contrary to seed grinding, which can result in a cupped pattern of wear on the molars (Smith, 1984; Mahoney, 2007), this distinctive plane reflects the reliance of some groups upon tubers for both food and as a source of fiber. While there are a range of underground storage organs used as staples by Aboriginal people in the less arid parts of Southeastern Australia, bulrush (Typha spp. or cumbungi) was particularly noted in the Murray Darling system (particularly the Murray River), to the extent that on the lower Murray River Angas noted: “The staff of their existence is the bulrush. . . root. it is to them what bread is to the European” (Angas, 1847: 890).
283
DENTAL WEAR IN SOUTH AUSTRALIA
Fig. 9. Comparison of the averaged wear by tooth relative to the first molar for all samples (using the broader Kaurna rather than the Gillman sample).
Typha species live in both fresh and brackish water generally about 1–1.5m deep and are thought to tolerate no more than one season of drying (Gott, 1982). In the regulated Murray River their contemporary distribution has been affected by reduced water flows, but Gott (1999) notes mention of the use of cumbungi for both food and fiber from the regions which would correspond to the
Euston, possibly Middle B, Middle A, Murray Mouth and the Kaurna/Gillman samples. The uncertainty about the Middle B region is that while there are records from the Wentworth region where the Murray and Darling Rivers join, it is not clear how far upstream the practice continued and at what point seed grinding was of more importance. The difference between the Middle B samples and
TABLE 7. Summary of dental wear patterns among the samples Yorke Peninsula Molar Gradient Flat central plane Heavy Anterior wear
Narrinyeri
Low None
Steep Flatter
Steep No
One male only
Heavy – females (males) Yes primarily anterior, possibly central
Males heavy
Sexual dimorphism
No
Other
Chipping, calculus High particularly anterior wear for males but both variable centrally
Variability
Kaurna (Gillman)
Middle A Moderate Flatter (females) Males mod. heavy
Middle B
Euston
Steep No
Moderate Flat
No
Males mod. heavy Yes – anterior dentition and degree of variability Calculus
Yes – anterior
Yes – anterior and central
No
Nil
–
–
–
Low
–
–
–
High particularly central for males and females, high anterior variability for males
American Journal of Physical Anthropology
284
J. LITTLETON ET AL.
those from Middle A and Euston may be due to the mixed reliance upon seeds and tubers in this group as opposed to primary reliance upon tubers. The possible role of typha as a major contributor to the pattern and degree of tooth wear is most evident in the Euston and Adelaide Plains samples. In both regions there is an extensive ethnohistoric literature on the use of typha for both food and fiber. The descriptions of processing also make it clear how it could contribute to masticatory forces. According to Beveridge (1889:19), in the area of Tyntynder (southeast of Euston), the rhizomes were steamed in an earth oven with clay balls as heat retainers and grass moistened to add to the steam. Once the cooking was completed, the cortex was stripped off, and the central rhizome was twisted into a knot and then crammed into the mouth. This was chewed to remove the starch which, according to Eyre, tasted like potato (Eyre, 1845 II:62). The fibers were spat out, packed into bags, and retained for making string and netting (Beveridge, 1889:71). These products were then traded. Typha could introduce two abrasives into the mouth: the phytoliths associated with the plant itself and the fine silt from the swamps. In addition to this, the descriptions of typha preparation all refer to what would be a large bolus within the mouth. Gott (1999) following Krefft (1866: 361) argues that in areas reliant upon the use of nets the fiber would probably be of even greater importance than the food and this could be one reason for gendered patterns of central wear observed among some of these samples. It is surprising, however, that the Murray mouth sample does not show this same pattern of heavy wear since Angas’s comments (1847) refer to the lower Murray where Taplin recorded the Narrinyeri word for typha, described as “root of a flag or bulrush, grows in fresh water, steamed and chewed to make string for nets” (Taplin, 1879: 41). It may be that in the area of Lake Alexandrina it was possible to rely less upon starch sources given a more regular supply of freshwater fish so that typha was primarily used for fiber rather than as a major staple and hence less routinely processed. Owen’s analysis of isotopes also suggests that there is some diversity among the Narrinyeri based on the contrast between coastal and freshwater resources (Owen, 2004), which may mean some specialization within the group. The ethnohistoric accounts from the Adelaide Plains refer to bulrush used for both food and for fiber (Teichelmann, 1857) and also point directly to the relationship between this particular resource and dental wear. Taplin quotes his earlier report: “A peculiar cause of toothache is the chewing of fiber for the purpose of making twine; this wears the teeth down to a level and makes them very tender to bite upon” (Taplin, 1879 II:109). It seems likely, therefore, that this is one instance where the ethnohistoric and osteological evidence (both macro- and micro-wear) come into alignment. However, the contrast between the Adelaide Plains where there is a more even pattern of wear (less pulp exposure and less AMTL) and Euston suggests that for the inhabitants of the Adelaide Plains there may have been less dominance of typha in the overall activity, particularly of women.
Intra-Sample Variability: Sex The analysis of intra-individual variation within the samples (where possible) provides insight into gendered American Journal of Physical Anthropology
patterns of diet and activity which can be further tested through analysis of dental microwear and stable isotope analysis. It has often been assumed in studies of Aboriginal teeth that female teeth were more worn and that this reflects a lower quality diet. This inference is frequently based upon ethnological and ethnohistoric accounts suggesting women ate a lower-quality diet (Campbell, 1939; Barrett et al., 1963). What is more striking, however, is the lack of variation in anterior wear among females and among males. This is particularly clear among the Euston sample as indicated by the coefficients of variation. In this sample the Euston males do not have a flat wear plane, particularly over the anterior teeth, but most women do. The difference is also clear in the maxillary premolars. Gendered differences in both anterior and premolar wear are also suggested in the small Gillman sample. Among the individuals from Gillman no female has heavy anterior or premolar wear; some males do (hence statistically significant differences in the gradients). Often it is assumed that gender differences are to do with a heavier reliance among women upon more plant matter and less meat, but the ethnoarchaeological studies by Meehan (1982) challenge this idea, pointing out that women, while gathering, also used the opportunity to catch and eat small animals. In the samples analyzed here the gendered differences are found in the anterior dentition, where it is frequently assumed that differences in wear are related to extramasticatory activity (Hinton, 1981; Smith, 1984), and in the overall plane rather than the molars, which would be affected by the abrasiveness and texture of the diet. Use of the anterior teeth as a third hand is well documented in hunter-gatherer studies, and the Adelaide Dental School expeditions documented a range of activities involving the use of teeth as tools among Aboriginal people in central Australia at Yuendumu (Campbell and Barrett, 1953; Barrett, 1977). Most of these involve tool manufacture. Further analysis might elucidate more clearly what range of activities are in place among the groups where males are using their front teeth, but what should be emphasized is that in these samples this is a variable activity – many but not all men show evidence of these sort of tasks. Part of that difference may be age-based, but it seems likely that such variability also reflects community organization and some specialization in terms of the greater range of acts engaged in by men relative to those available to women. The comparison between the Gillman and Kaurna samples— where in the former, males have heavy anterior wear compared to females in the latter—suggests the possibility these roles are not fixed over either space or time (or both), although the Gillman sample is very small. The lack of a clear gendered pattern in the Middle B and Yorke Peninsula samples is interesting. It is possible that within these groups there was less reliance upon amassing tools and equipment (e.g., nets, baskets, etc.), although that would seem more likely for the Yorke Peninsula than the Middle B, group which along with other Murray River groups is assumed to have a higher population density and to have exploited resources more intensively (Pardoe, 1995; Webb, 1995). However, while examining variation and constraint around patterns of tooth wear has the potential to provide a more nuanced picture of gender divisions (which can be tested by examining the variation in microwear studies or isotope studies), many of these samples are not ideal for that sort of analysis because they are time-
DENTAL WEAR IN SOUTH AUSTRALIA averaged. Time-averaging is a concept most commonly applied to the archaeological record highlighting the long term history of many locales, which represent a palimpsest of variably present activities over time (Stern, 1994; Lyman, 2003; Bailey, 2007). The term, however, highlights that any skeletal sample represents both deaths over time as well as a congregation of individual life histories of varying length. The dating of samples most frequently relies upon associated archaeological materials and typically, as seen here, samples may comprise long periods of time and, sometimes, diverse space. This is particularly true of analyses involving huntinggathering populations (Webb, 1995; Pfeiffer, 2007). How this affects the interpretation of results and whether the practice obscures variability is a critical question. Unfortunately in this analysis the one sample which is least time- and space-averaged—Gillman—is too small to compare to the others in terms of homogeneity. Timeaveraging, however, needs to be taken seriously if we are going to identify the extent and causes of variation within populations.
CONCLUSIONS It was hypothesized that neighboring Aboriginal groups might differ significantly in terms of social and economic practices. These contiguous groups demonstrate that some of the difference relates to availability of resources, but much also relates to resource choices, specializations and patterns of gendered behavior. Even within a region these practices may be neither temporally nor spatially fixed. Focusing our analyses of hunter-gatherers to examine patterns of intra-sample variability across the entire dentition and upon the specific spatial and temporal context of the human remains may yield a much finer-grained and dynamic model of hunter-gatherer adaptations. There is more work to be done, particularly before remains are repatriated.
ACKNOWLEDGMENTS The authors would like to thank the Kaurna National Cultural Heritage Association and the Narangga community for their permission and collaboration in recording these remains. Thanks are also due to Tim Mackrell and Briar Sefton, who assisted with preparation of the illustrations along with Rebekah Candy, Storm Graham and Sam Hoare, who assisted with recording. The authors are grateful for the numerous editorial and reviewer comments on the manuscript as well as those of Bruce Floyd, Harry Allen and Pauline Herbst.
LITERATURE CITED Allen J. 1974. The Bagundji of the Darling Basin: cereal gatherers in an uncertain environment. World Archaeol 15:309–322. Angas G. 1847. Savage life and scenes. London: Smith, Elder and Co. Bailey G. 2007. Time perspectives, palimpsests and the archaeology of time. J Anthropol Archaeol 26:198–223. Barrett M, Brown T, Macdonald M. 1963. Dental observations on Australian Aborigines: mesiodistal crown diameters of permanent teeth. Aust Dent J 8:150–156. Barrett MJ. 1977. Masticatory and nonmasticatory uses of teeth. In: Wright R, editor. Stone tools as cultural markers. Prehistory and material culture series 12. Canberra: Australian Institute of Aboriginal Studies. p 18–23. Bernal V, Novellino P, Gonzalez PN, Perez SI. 2007. Role of wild plant foods among late Holocene hunter-gatherers from
285
central and north Patagonia (South America): an approach from dental evidence. Am J Phys Anthropol 133:1047–1059. Beveridge P. 1889. The Aborigines of Victoria and Riverina. Melbourne: ML Hutchinson. Birdsell JB. 1953. Some environmental and cultural factors influencing the structuring of Australian Aboriginal populations. Am Nat 87:171–207. Brown P. 1978. The ultrastructure of dental abrasion: its relationship to diet. Unpubd. BA Hons Thesis. Canberra: Australian National University. Brown T, Molnar S. 1990. Interproximal grooving and task activity in Australia. Am J Phys Anthropol 81:545–553. Campbell T.D. 1939. Food, food values and food habits of the Australian Aborigines in relation to their dental conditions. I. Aust Dent J 43:2–15. Campbell T, and Barrett M. 1953. Dental observations on Australian Aborigines—a changing environment and food pattern. Aust Dent J 57:1–6. Elvery MW, Savage NW, Wood WB. 1998. Radiographic study of the Broadbeach Aboriginal dentition. Am J Phys Anthropol 107:211–219. Eyre E. 1845. Journals of expeditions of discovery into Central Australia and overland from King Georges Sound in the years 1840–1: Sent by the colonists of South Australia, with the sanction and support of the government: Including an account of the manners and customs of the Aborigines and the state of their relations with Europeans. London: T and W Boone. Gott B. 1982. Ecology of root use by the Aborigines of Southern Australia. Archaeol Oceania 17:59–67. Gott B. 1999. Cumbungi, Typha species: a staple Aboriginal food in Southern Australia. Aust Aborig Stud 1:33–50. Hall RL. 1976. Functional relationships between dental attrition and the helicoidal plane. Am J Phys Anthropol 45:69–75. Hinton RJ. 1981. Form and patterning of anterior tooth wear among Aboriginal human groups. Am J Phys Anthropol 54: 555–564. Hodges J. 1973. Gillman. A report of the excavations at Gillman, S.A., 1970. Unpublished report for the South Australian Museum, Adelaide. Keenleyside A. 1998. Skeletal evidence of health and disease in pre-contact Alaskan Eskimos and Aleuts. Am J Phys Anthropol 107:51–70. Krefft G. 1866. On the manners and customs of the Aborigines of the Lower Murray and Darling. Trans Phil Soc NSW 1: 357–374. Lev-Tor Chattah N, Smith P. 2006. Variation in occlusal dental wear of two Chalcolithic populations in the southern Levant. Am J Phys Anthropol 130:471–479. Lieverse AR, Link DW, Bazaliiskiy VI, Goriunova OI, Weber AW. 2007. Dental health indicators of hunter-gatherer adaptation and cultural change in Siberia’s Cis-Baikal. Am J Phys Anthropol 134:323–339. Littleton J, Frohlich B. 1993. Fisheaters and farmers: dental pathology in the Arabian Gulf. Am J Phys Anthropol 92:427– 447. Littleton J, Scott R. In review. Dietary variability in Southern Australia? In: Katzenberg MA, Lee-Thorp J, editors. Oxford handbook of dietary analysis. Oxford: Oxford University Press. Submitted mss. Lukacs J. 1989. Dental paleopathology: methods for reconstructing dietary patterns. In: Kennedy K, Iscan MY, editors. Reconstruction of life from the skeleton. New York: Liss. p 261–86. Lyman RL. 2003. The influence of time averaging and space averaging on the application of foraging theory in zooarchaeology. J Archaeol Sci 30:595–610. Mahoney P. 2007. Human dental microwear from Ohalo II (22,500–23,500 Cal BP), southern Levant. Am J Phys Anthropol 132:489–500. Meehan B. 1982. Shell bed to shell midden. Canberra: Australian Institute of Aboriginal Studies. Molnar S, Richards L, McKee J, Molnar I. 1989. Tooth wear in Australian Aboriginal populations from the River Murray Valley. Am J Phys Anthropol 79:185–196.
American Journal of Physical Anthropology
286
J. LITTLETON ET AL.
Murray-Darling Basin Commission. 2003. Native fish strategy for the Murray-Darling Basin 2003–2013. Canberra: MurrayDarling Basin Commission. Owen TD. 2004. “Of more than usual interest”: a bioarchaeological analysis of ancient Aboriginal skeletal material from southeastern South Australia. PhD Thesis, Adelaide: Flinders University. Pardoe C. 1995. Riverine, biological and cultural evolution in southeastern Australia. Antiquity 69:696–714. Pate FD. 1998. Stable carbon and nitrogen isotope evidence for prehistoric hunter-gatherer diet in the lower Murray River basin, South Australia. Archaeol Oceania 33:92–99. Pate FD, Owen TD, Lawson EM. 2003. AMS radiocarbon dating of bone collagen: establishing a chronology for the Swanport Aboriginal burial ground, South Australia. Aust Archaeol 56: 8–11. Pfeiffer S. 2007. The health of foragers: people of the later Stone Age, Southern Africa. In: Cohen MN, Crane-Kramer G, editors. Ancient health: skeletal indicators of agricultural and economic intensification. Gainesville: University Press of Florida. p 223–236. Richards L, Brown T. 1981. Dental attrition and age relationships in Australian Aboriginals. Archaeol Oceania 16:94–97. Richards L, Miller S. 1991. Relationships between age and dental attrition in Australian Aboriginals. Am J Phys Anthropol 84:159–64. Richards LC. 1984. Principal axis analysis of dental attrition data from two Australian Aboriginal populations. Am J Phys Anthropol 65:5–13. Schmidt CW. 2009. On the relationship of dental microwear to dental macrowear. Am J Phys Anthropol 142:67–73. Scott A, Olley JM, Murray-Darling Basin Commission. 2003. Settlement, erosion, and muddy waters: lessons from the past. Canberra, ACT: Murray-Darling Basin Commission.
American Journal of Physical Anthropology
Scott EC. 1979a. Dental wear scoring technique. Am J Phys Anthropol 51:213–217. Scott EC. 1979b. Principal axis analysis of dental attrition data. Am J Phys Anthropol 51:200–212. Smith BH. 1984. Patterns of molar wear in hunter-gatherers and agriculturalists. Am J Phys Anthropol 63:39–56. Smith P, Prokopec M, and Pretty G. 1988. Dentition of a prehistoric population from Roonka Flat, South Australia. Archaeol Oceania 23:31–36. Stern N. 1994. The implications of time-averaging for reconstructing the land-use patterns of early tool-using hominids. J Hum Evol 27:89–105. Stirling E. 1911. Preliminary report on the discovery of native remains at Swanport, River Murray; with an inquiry into the alleged occurrence of a pandemic among the Australian Aboriginals. Trans R Soc South Aust 35:4–50. Sunderland S, Ray L. 1959. A note on the Murray Black collection of Australian Aboriginal skeletons. Proc R Soc Victoria 71:45–50. Taplin G. 1879. The folklore, manners and customs of the South Australia Aborigines. Adelaide: E Spiller. Taylor R. 1991. Anatomy and biology of tooth dislocation and wear in the pre-European Maori and Australian Aborigine. Unpubd PhD Thesis Auckland: University of Auckland. Teichelmann CG. 1857. Dictionary of the Adelaide dialect. Mss. no. 59, Bleek’s catalogue of Sir George Grey’s library dealing with the Australian languages. South Australian Archives. Turner C. 1979. Dental anthropological indications of agriculture among the Jomon people of Central Japan. Am J Phys Anthropol 51:619–636. Walshe K, Littleton J, Graham S. 2011. Analysis of the Gillman Mound, South Australia. A report for the Australian Institute of Aboriginal and Torres Strait Islander Studies, Canberra. Webb S. 1995. The palaeopathology of Aboriginal Australians. Cambridge: Cambridge University Press.