Age and Sex Estimates of Subadult Ilia from

Age and Sex Estimates of Subadult Ilia from Charnel House A 22, Bab-edh-Dhra': A Model for Determining Subadult Sex Sarah Zaleski, Dept. of Anthropology and Middle Eastern Cultures, Mississippi State University Results and Discussion Introduction The determination of sex in human skeletal remains is crucial to making conclusions about demography and cultural trends of past populations. The bony pelvis exhibits sexually distinct features and is the most accurate indicator of sex for adult specimens. However, tests of subadult (ages 0 -15) sex indicators have produced varying, often low accuracy rates. Methods involving the ilium, the largest bone of the pelvis, have demonstrated the most consistent results across multiple tests (Vlak et al. 2008). Previous studies show that the accuracy rates of ilial features as subadult sex indicators vary with age. This study therefore recommends an age-adjusted model for determining subadult sex using four ilial features. The skeletal collection used to develop this model is an Early Bronze Age (EBII-III; 3000 -2300 BCE) commingled collection from Charnel House A 22 in Bab edh-Dhra’, Jordan. This collection is now housed at the University of Notre Dame. Background During the period in which A 22 was in use, the surrounding region saw developments in agricultural practices and urbanization. A 22 is Bab edh-Dhra’s largest charnel house, and contained prestige items including gold jewelry. A 22 individuals were possibly of a higher social status with greater access to nonlocal goods (Chesson and Schaub 2007). The fragmented nature of this study’s sample necessitates the analysis of multiple ilial features in order to obtain sex estimations for all ilia. This requirement stimulated the development of a multi -feature model for subadult sex determination. This model differs from previous methods for subadult sex determination which are based on single feature evaluations. This study’s sample is of unknown sex; therefore, the accuracy of the proposed model is unknown. Nevertheless, this model provides a foundation for important potential bioarchaeological and forensic applications. To address the fact that the accuracy rates of subadult sex indicators vary with age, this study integrates age data with the preliminary feature assessments. Age was estimated using ilium width measurements. Rissech and Malgosa (2005) found that because ilium width does not present significant differences between males and females, it has great utility as a subadult age indicator for individuals ages 0 -20 years. Due to the fragmentary nature of the ilia, not all bones have the osteological points from which to take the measurement. The sample was therefore seriated from smallest to largest while accounting for the nature of fragmented bones. Ilium width was measured for the bones that had the appropriate points. Using the seriation, the ilium widths of the remaining bones were estimated. Feature Weighting This model gives weight to features that are the most reliable subadult sex indicators for particular ages. The reliability of the features is based on a consensus of their accuracy rates reported in previous studies (Weaver 19080, Mittler and Sheridan 1992, Schutkowski 1993, Sutter 2003, Cardoso and Saunders 2008, and Vlak et al. 2008). elevation, greater sciatic notch (GSN) angle, GSN depth, and the arch criterion – according to the methods set forth by Weaver (1980) and Schutkowski (1993) and expanded upon by Sutter (2003) and Vlak et al. (2008). Each feature was evaluated separately and independently. The ilia were seriated in order of feature expression for auricular surface elevation and arch criterion to provide for a more objective analysis. Digital pictures of the pelvic and gluteal aspects of the GSN were taken to objectively measure the GSN angle and depth with Image. J computer software. Graphical trends of the GSN angle and depth values established male, female, and indeterminate ranges. A preliminary study describes these feature assessments and initial sex estimations (Zaleski et al. 2010). Arch Criterion Figure 3. Standardized landmarks for GSN Angle (A) and GSN Depth (notch breadth/notch depth) (B). Modified from Vlak et al. (2008) Fig. 2 B and 2 C. A. B. GSN Notch Depth Auricular Surface Figure 4. BD 1408. 30. Ilium width is the distance between the two blue dots. Photograph taken by author. Age was estimated via ilium width with reference to a table of subadult Arikara ilium widths and corresponding age data (Ubelaker 1999). Percent deviations of the A 22 sample’s ilium widths from the Arikara means were calculated to ensure that the A 22 ilium widths fell within the range of the Arikara data (Franciscus and Churchill 2002). A modified t-test showed that the A 22 ilium widths do not differ significantly from the Arikara means (Sokal and Rohlf 1981). Six age brackets were created based on the nature of the A 22 and Arikara data. Feature Evaluations The model proposed in this study is based on the assessment of four features of 61 right subadult ilia – auricular surface Figure 2. Bone ID: BD. 1217. 84 (pelvic aspect) with features designated. Photograph taken by author. Table 4. Sex Distribution Age Estimations Figure 1. Bab-edh-Dhra’ is highlighted. Modified from Chesson (1999), Fig. 1. Methodology Ilium Width Measurements Notch Breadth Table 1. Feature Weighting for Age Brackets Ilial Features Auricular Surface GSN Angle GSN Depth Arch Criterion <2 2 -5 0% 0% 33. 33% 25% 33. 33% 50% 33. 33% 25% Feature Weights by Age Bracket 3. 5 -4. 5<x<6. 5 -7. 5 6 -9 7. 5 -8. 5<x<12. 5 -13. 5 11 -15 0% 33. 33% 25% 25% Each bone in the sample was classified as female, likely male, or indeterminate. The ilia’s four pre-existing feature scores (Zaleski et al. 2010) were transformed. Female scores became scores of 1; male scores became scores of 0; and indeterminate scores became scores of 0. 5. Features that were missing or unscorable due to fragmentation were also given a score of 0. 5, based on the idea that there is a 50 -50 chance that feature would score as male (0) or female (1). For each ilium, each of its feature’s scores was multiplied by the weight given to that feature for the particular age bracket into which that particular ilium fell. Then, the products for each bone were summed. The resulting sum or total score became a percent estimation of how likely a bone was to be female. Score ranges for sex estimations are specified in Table 2. Total Scores and Corresponding Sex Estimations The preliminary research for this study was supported by the NSF-REU (SES 0649088) Summer Research in Biological Anthropology at the University of Notre Dame. This study was supported by Dickinson College Academic Affairs and Research and Development Committee. Thank you to the Mississippi State Graduate School TAGG program and Mississippi State’s Dept. of Anthropology and Middle Eastern Cultures. Many thanks as well to: Robert J. Smith (Emory University), Susan G. Sheridan, Ph. D. (University of Notre Dame), Jaime Ullinger (Ohio State University), Brenda Baker, Ph. D. (Arizona State University), Karen J. Weinstein, Ph. D. (Dickinson College), Ann M. Hill, Ph. D. (Dickinson College), Peter A. Zaleski, Ph. D. (Villanova University), Ryan P. Shears, and Nicholas P. Herrmann (Mississippi State University). Sex Estimation . 75 -1. 5 -. 75 0. 5. 25 -. 5 0 -. 25 Female Likely Female Indeterminate Likely Male Table 3. Example from the 11 -15 Age Bracket Ilium ID Weight BD 1500. 13 Feature Score Auricular Surface GSN Angle GSN Depth Arch Criterion Total Score 0. 2 1 0. 4 0. 5 0. 2 0. 5 Males 0% (0/15) 27. 27% (3/11) 11. 11% (1/9) 16. 67% (1/6) 20. 00% (2/10) 44. 44% (4/9) 15. 38% (2/13) 40. 00% (4/10) 33. 33% (3/9) 85. 71% (6/7) 25. 00% (1/4) 50. 00% (1/2) Likely Males 20. 00% (3/15) 0% (0/9) 20. 00% (2/10) 15. 38% (2/13) 33. 33% (3/9) 0% (0/4) Indeterminate 26. 67% (4/15) 33. 33% (3/9) 10. 00% (1/10) 23. 08% (3/13) 22. 22% (2/9) 50. 00% (2/4) Likely Females 40. 00% (6/15) 22. 22% (2/9) 40. 00% (4/10) 46. 15% (6/13) 11. 11% (1/9) 25. 00% (1/4) Females 13. 33% (2/15) 72. 72% (8/11) 33. 33% (3/9) 83. 33% (5/6) 10. 00% (1/10) 55. 56% (5/9) 0% (0/13) 60. 00% (6/10) 0% (0/9) 14. 29% (1/7) 0% (0/4) 50. 00% (1/2) The model produces a classification of 42. 22% males and 57. 78% females in the A 22 sample. When the study sample is broken down by age brackets, the youngest two brackets – <2 years and 2 -5 years – contain much greater female percentages (72. 72% and 83. 33%, respectively) than male percentages. These results suggest preferential treatment of males over females at weaning age. Certain aspects of the A 22 sample should be considered along with the suggested preferential treatment of males during weaning. It is important to note that the A 22 collection is not representative of the entire Bab edh-Dhra’ population, but rather of the elite with access to nonlocal goods and prestige items (Chesson and Schaub 2007). Additionally, the sample size of 61 right ilia was limited to bones that possessed at least one assessable feature of the four ilial features assessed in this study. One aspect that should be noted when considering the A 22 sample at large is that during late EB III (2300 BCE), Charnel House A 22 and the bones inside burned with the destruction of the town (Chesson and Schaub 2007). Burning at high temperatures or for long durations can result in bone shrinkage, which will affect metric measurements. The temperature and burning environment specification for a bone based on its color is its burn score (Walker et al. 2008). Despite the burning, the analysis of sexual characteristics in the A 22 sample should not be affected greatly (Schmidt, 2009 personal communication). An examination of bone color in this study’s sample suggests that about 30% of the sample may have undergone shrinkage; however, this possibility could not be addressed in age estimations. The ilia that for which ilium width measurements were taken did not exhibit bone colors that suggested shrinkage. Conclusions 20% 40% 20% Sex Estimations Total Score Sample <2 year-old Subsample 2 -5 year-old Subsample 6. 5 -7. 5 > x > 3. 5 -4. 5 Subsample 6 -9 year-old Subsample 12. 5 -13. 5 > x > 7. 5 -8. 5 Subsample 11 -15 year-old Subsample 0. 2 0. 5 1 0. 6 Sex Estimation Likely Female Because the A 22 collection is of unknown sex, future research must begin with the assessment of this study’s model on subadult skeletal collections of known sex. For collections with more complete individuals, age should be estimated via dental eruption, and cleaner age brackets should be created when possible. It is likely that the features within each age bracket should be weighted differently for different population groups. Future research should seek to determine how certain environments and stressors impact the growth and sexual dimorphism of the ilium and its features. The proposed model’s use of multiple features and feature weighting based on age provide a foundation for further development. Ultimately, accurate subadult sex determination contributes to a better understanding of subadults in archaeological contexts. This clearer picture of subadults is crucial to understanding archaeological population groups at large. References Cardoso, H. F. V. and S. R. Saunders 2008 Two arch criteria of the ilium for sex determination of immature skeletal remains: A test of their accuracy and an assessment of intra- and inter-observer error. Forensic Science International 178(1): 24 -29. Chesson, M. S. 1999 Libraries of the dead: Early bronze age charnel houses and social identity at urban Bab edh-Dhra', Jordan. J. of Anthropological Archaeology 18(2): 137 -164. Chesson, M. S. and R. T. Schuab 2007 Life in the earliest walled towns on the Dead Sea Plain: Bab edh-Dhra’ and an-Numayra. In Crossing Jordan: North American contributions to the archaeology of Jordan. T. E. Levy, P. M. Michèle Daviau, R. W. Yonker, and M. Shaer, eds. Pp. 244 -252. London: Equinox Pub. Ltd. Franciscus, R. G. and S. E. Churchill 2002 The costal skeleton of Shanidar 3 and a reappraisal of Neandertal thoracic morphology. J. Hu. Ev. 42: 303 -356. Mittler, D. M. , and S. G. Sheridan 1992 Sex determination in subadults using auricular surface-morphology - a forensic-science perspective. J. For. Sci. 37(4): 1068 -1075. Rissech, C. , and A. Malgosa 2005 Ilium growth study: Applicability in sex and age diagnosis. Forensic Science International 147(2 -3): 165 -174. Schutkowski, H. 1993 Sex determination of infant and juvenile skeletons. 1. morphognostic features. Am. J. Phys. Anth. 90(2): 199 -205. Sokal, R. R. and F. J. Rohlf. 1981 Biometry (2 nd edn). New York: W. H. Freeman & Co. Sutter, R. C. 2003 Nonmetric subadult skeletal sexing traits: I. A blind test of the accuracy of eight previously proposed methods using prehistoric known-sex mummies from northern Chile. Journal of Forensic Sciences 48(5): 927 -935. Ubelaker, D. H. 1999. Human Skeletal Remains: Excavation, Analysis, Interpretation. Washington D. C. : Smithsonian Institution. Vlak, D. , M. Roksandic, and M. A. Schillaci 2008 Greater sciatic notch as a sex indicator in juveniles. Am. J. Phys. Anth. 137(3): 309 -315. Walker, P. L. , Miller K. W. P. , and R. Richman 2008 Time, Temperature, and Oxygen Availability: An Experimental Study of the Effects of Environmental Conditions on the Color and Organic Content of Cremated Bone. In The Analysis of Burned Human Remains. C. W. Schmidt and S. A. Symes, eds. Pp. 129 -135. Amsterdam: Academic Press. Weaver, D. S. 1980 Sex-differences in the llia of a known sex and age sample of fetal and infant skeletons. Am. J. Phys. Anth. 52(2): 191 -195. Zaleski SM, Smith RJ, Ullinger J, and Sheridan SG 2010 Let’s talk about sex: using metric and non-metric features of the subadult ilium to identify males and females. Poster presented at the Seventy-Ninth Annual Meeting of the American Association of Physical Anthropologists, Albuquerque, New Mexico.