Age Estimation From the Auricular Surface

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 119:231–239 (2002)

 Age Estimation From the Auricular Surface of the Ilium:  A Revised Method J.L. Buckbe Buckberry rry* and A.T. Chamberlain  Department of Archaeology and Prehistory, University of Sheffield, Sheffield S1 4ET, UK    KEY WORDS auricular surface; ilium; skeletal age at death; Spitalfields; forensic anthropology; palaeodemography  ABSTRACT  A rev revise ised d meth method od for est estima imatin ting g adu adult lt age at death using the auricular surface of the ilium has been developed. It is based on the existing auricular surface aging method of Lovejoy et al. ([1985] Am. J. Phys.  Anthropol. 68: 68:15–28), 15–28), but the revised technique is easier to apply, and has low levels of inter- and intraobserver error. The new method records age-related stages for different features of the auricular surface, which are then combin com bined ed to prov provide ide a com compos posite ite sco score re fro from m whi which ch an estimate of age at death is obtained. Blind tests of the method were carried out on a known-age skeletal collection from Christ Church, Spitalfields, London. These tests

The estimation of age at death of adult skeletal materi mat erial al is one of the more dif diffic ficult ult tasks underundertaken by phys physical ical anthropologi anthropologists. sts. Mays (1998, p. 50) st stat ated ed th that at “a “att pr pres esen entt th the e la lack ck of a wh whol olly ly satisfactory technique for estimating age at death in adult skeletons from archaeological sites is one of  the mos mostt tho thorny rny pro proble blems ms fac facing ing hu human man ost osteoa eoarrchaeology.” The problem arises from the fact that rates of skeletal remodeling and degeneration, from which wh ich mos mostt met method hodss of adu adult lt age est estima imatio tion n are derive der ived, d, can be hig highly hly var variab iable le bet betwe ween en dif differ ferent ent individuals and populations. The life history of an individual will be an important determinant of the rate of skeletal aging. Factors such as endocrine status, diet, disease, physical activity, and cultural differences will contribute to the range of variation between males and females, old and young, and rich and poor within a site, and also al so be betw twee een n po popu pula lati tion onss fr from om di diffe ffere rent nt si site tess (Brothwell, 1981; Ubelaker, 1989; Schwartz, 1995; Hillson, 1996; Aykroyd et al., 1999). Even with a similar simil ar biolog biological ical backgr background ound and envir environme onmental ntal conditions, two individuals are likely to age at different rates. Pubic symphysis age estimation was first applied to adult males (Todd, 1920), and was subsequently redefined and tested, especially with regard to creating separate separate syste systems ms for females (Broo (Brooks, ks, 1955; Nemeske´ ri et al., 1960; Gilbert and McKern, 1973; Suchey, 1979; Katz and Suchey, 1986; Brooks and Suchey, Such ey, 1990). These methods methods were found to gener gener-ate age distributions for different populations that ©

2002 WILEY-LISS, WILEY-LISS, INC.

showed that the dispersion of age at death for a given morphological morpho logical stage was large, particularly particularly after the first decade of adult life. Statistical analysis showed that the age-related changes in auricular surface are not significantl ca ntly y di diff ffere erent nt fo forr ma males les an and d fe fema males les.. Th The e sc scor ores es fr from om th the e revised method have a slightly higher correlation with age than do the Suchey-Brooks pubic symphysis stages. Considering siderin g the higher survival survival rates of the auricu auricular lar surface compared with the pubic symphysis, this method promises to be useful for biological anthropology and forensic science. Am J Phys Anthropol 119:231–239, 2002.  ©

2002 Wiley-Liss, Inc.

mirrored the age structure of the reference series from which the methods were developed (Bocquet Appel and Masset, 1982), and did not allow for the  variation normal to a skeletal population (Suchey et al., 1986). When the Suchey-Brooks method (Brooks and Suchey, 1990) was developed, this variation was made explicit by the documentation of the dispersion of age at death for each morphological stage of  the pubic symphysis. The auricular auricular surfa surface ce meth method od of age estim estimation ation has not been subject to the same levels of scrutiny as pubic symphysis aging, and it is usually applied in the th e fo form rm or orig igin inal ally ly de deve velo lope ped d by Lo Love vejo joy y et al al.. (1985). Murray and Murray (1991) teste tested d the auricular surface method of Lovejoy et al. (1985) on a sample of skeletons in the Terry Collection, to see how ho w re reli liab able le th the e me meth thod od wa wass fo forr us use e in fo fore rens nsic ic cases. They found that the method was “unbiased regarding race and sex” (Murray and Murray, 1991; Grant sponsor: Universities of Sheffield, York, and Leeds. *Correspondence *Corresp ondence to: Miss Jo Buckberry, Research School of Archaeology and Archaeological Science, University of Sheffield, West Court, 2 Mappin St., Sheffield S1 4DT, UK. E-mail: J.Buckberr J.Buckberry@Sheffield.ac y@Sheffield.ac.uk .uk Received 23 July 2001; accepted 17 April 2002. DOI 10.1002/ajpa.10130 Published online in Wiley InterScience (www.interscience.wiley. (www.interscience.wiley. com).

232

J.L. BUCKBERRY AND A.T. CHAMBERLAIN

p. 1166). However, they also found that the method ricular surface area of the ilium is more durable consistently underestimated the age of older individ- than the fragile pubic symphysis, with higher levels uals, and overestimated the age of younger individ- of survival and recovery, particularly in archaeouals in their sample. They felt that this might have logical populations (Lovejoy et al., 1985; Waldron, been attributable to differences in age structure be- 1987). However, the separate features of the aurictween the Todd Collection, on which the method was ular surface described by Lovejoy et al. (1985), such developed, and the Terry Collection (Murray and as porosity, surface texture, and marginal changes, Murray, 1991). Murray and Murray (1991) concluded appear to develop independently of each other. The that the method was “too unreliable to be used as a age of onset for each stage of different features of the single aging technique for forensic science purposes.” auricular surface appears to vary, and as a conseHowever, they conceded that the method would be quence the 5-year age categories of Lovejoy et al. useful when used in conjunction with other methods (1985) tend to overlap. Early-appearing features of age estimation. still present on the auricular surfaces of older indiThe auricular surface method was also tested on  viduals were described by Lovejoy et al. (1985) as an archaeological sample of known age at death from Belleville, Ontario (Saunders et al., 1992). The “residual.” The fact that this variation can occur system was found to underestimate age at death within a single auricular surface indicates that the especially for the older portion of the sample (over 45 method of Lovejoy et al. (1985) oversimplifies the years), although it predicted the ages of younger changes seen, and that the 5-year intervals in their adults reliably (Saunders et al., 1992). The authors scheme of age estimation may be optimistically narfound that estimates of age at death for many indi- row. This problem contributes to the dif ficulty found  viduals did not fall into the correct modal stages, with applying the method, as it leads to uncertainty, indicating that the method of Lovejoy et al. (1985) and in some cases confusion, in assigning individual may not allow adequately for individual variation in auricular surfaces to a particular age stage. skeletal aging. It was also noted that intraobserver error was high (19.3%) (Saunders et al., 1992), and   A REVISED AURICULAR SURFACE METHOD this may be due to the dif ficulty found in applying OF AGE ESTIMATION the method, especially when classifying “ambiguous In view of these problems, we believe that the specimens . . . that cannot with certainty be assigned to one age category” (Saunders et al., 1992, p. auricular surface age estimation technique would benefit from the same level of reanalysis that has 99). Bedford et al. (1993) also tested the auricular sur- been undertaken for pubic symphyseal age estimaface method on a sample of known-age skeletons tion. A quantitative scoring system, which examines from the Grant Collection at the University of To- each different feature of the auricular surface inderonto. As with other tests of the method, there was a pendently, should not only be easier to apply but will tendency to overestimate ages of younger adults and also accommodate the overlap often seen between to underestimate ages in categories above 60 years different stages. Our revised system has been develold. oped from that of Lovejoy et al. (1985), and utilizes their categories of age-related change seen on the   AURICULAR SURFACE METHOD OF AGE auricular surface. The features used were transESTIMATION OF LOVEJOY ET AL. (1985)  verse organization, surface texture, microporosity, The age-related changes seen on the auricular macroporosity, and changes in morphology of the surface of the ilium were first described by Sashin apex and retroauricular area. Each of the features (1930). He interpreted the changes seen in the ar- seen on the auricular surface was recorded indepenticular cartilage of the joint as osteoarthritic, and dently and assigned a series of numerical scores found that they were “progressive, and increase in corresponding to successive stages of degrees of exextent and intensity with the age of the individual” pression. Although one feature might be obscured by (Sashin, 1930, p. 909). However, the regularity of  another, the use of standardized criteria (defined these changes was not discussed. The age-related below) allowed them to be assessed objectively. In changes of the auricular surface were not exploited as a method of estimating age at death until Lovejoy preliminary analyses, the retroauricular area was found to be a poor estimator of age, and this led to its et al. (1985) developed their technique. Lovejoy et al. (1985) described eight modal age exclusion from the revised method. The surface is described using the same terminolstages (typically encompassing 5 years), into which ogy employed by Lovejoy et al. (1985). Figure 1 each auricular surface could be placed using primary aging features. An age would then be esti- shows the locations used in describing the auricular mated from within this modal stage, using secondary surface. Table 1 gives the descriptions of each locaindicators of age (apical activity and retroauricular tion, as defined in Buikstra and Ubelaker (1994). Each of the features of the auricular surface is deactivity).   Auricular surface age estimation increased the scribed in detail, with a summary description of  range of methods that were available, and the au- each stage observed given in tabular form.

233

REVISED AURICULAR SURFACE AGING METHOD TABLE 3. Scoring system for surface texture Score 1 2 3 4 5

Description 90% or more of surface is finely granular 50–89% of surface is finely granular; replacement of  finely granular bone by coarsely granular bone in some areas; no dense bone is present 50% or more of surface is coarsely granular, but no dense bone is present Dense bone is present, but occupies less than 50% of  surface; this may be just one small nodule of dense bone in very early stages 50% or more of surface is occupied by dense bone

TABLE 4. Scoring system for microporosity Score 1 2 3

Fig. 1. Regions of the ilium used for auricular surface aging (redrawn after Lovejoy et al., 1985). TABLE 1. Definitions used to describe auricular surface (after Buikstra and Ubelaker, 1994) Definition

Description of locations

Apex Superior demiface Inferior demiface Retroauricular area

Portion of auricular surface that articulates with posterior aspect of  arcuate line Portion of auricular area above apex Portion of auricular area below apex Region between auricular surface and posterior inferior iliac spine

TABLE 2. Scoring system for transverse organization Score 1 2 3 4 5

Description 90% or more of surface is transversely organized 50–89% of surface is transversely organized 25–49% of surface is transversely organized Transverse organization is present on less than 25% of surface No transverse organization is present

Transverse organization

Transverse organization refers to the horizontally orientated billows and striae that run from the medial to the lateral margins of the auricular surface (Lovejoy et al., 1985). It was dif ficult to distinguish between billows and striae and to be consistent in the recording of these features. Consequently, this trait is scored in terms of  what proportion of the auricular surface is transversely organized . The proportion is estimated by eye, rather than measured, and is assigned to 1 of 5 easily distinguishable stages defined in terms of the percentage of surface covered. The scoring system for transverse organization is given in Table 2. Surface texture

This feature corresponds to that described as “grain” by Lovejoy et al. (1985). The texture of the

Description No microporosity is present Microporosity is present on one demiface only Microporosity is present on both demifaces

auricular surface is finely grained in early life, and becoming more coarsely granular and densified in older individuals. As with transverse organization, this feature is scored in terms of  what proportion of  the surface is covered by a particular type of texture .   Finely granular bone is defined as having grains predominantly less than 0.5 mm in diameter, and coarsely granular bone consists of grains predominantly over 0.5 mm in diameter. Dense bone refers to surface appearance, rather than the amount of bone present. It is defined as nodules or areas of bone which are compact and smooth, with no surface granularity. The scoring system for surface texture is given in Table 3. Microporosity

In the revised system, microporosity was defined as porosity of the surface (or perforations of subchondral bone), with the pores having a diameter of  less than 1 mm. It is scored according to the presence of microporosity on one or both of the two demifaces of the auricular surface. Microporosity may be localized or spread across large areas. Lovejoy et al. (1985) regarded microporosity as a secondary aging feature. The scoring system for microporosity is given in Table 4. Macroporosity

Macroporosity also perforates the surface of the bone. We define macroporosity as holes greater than 1 mm in diameter. Like microporosity, it may be localized or spread across large areas. Lovejoy et al. (1985) also considered macroporosity a secondary aging feature. Macroporosity is scored according to its presence on one or both of the two demifaces of the auricular surface, in the same manner as for microporosity. The scoring system for macroporosity is given in Table 5. Macroporosity should not be confused with cortical defects, which may be present at any age. Cortical defects are areas where the cortex of the bone is

234

J.L. BUCKBERRY AND A.T. CHAMBERLAIN TABLE 5. Scoring system for macroporosity

Score

TABLE 8. Multiple regression of age against features

Description

1 2 3

No macroporosity is present Macroporosity is present on one demiface only Macroporosity is present on both demifaces

TABLE 6. Scoring system for apical changes Score 1 2 3

Description Apex is sharp and distinct; auricular surface may be slightly raised relative to adjacent bone surface Some lipping is present at apex, but shape of articular margin is still distinct and smooth (shape of outline of surface at apex is a continuous arc) Irregularity occurs in contours of articular surface; shape of apex is no longer a smooth arc

TABLE 7. Spearman correlations between age and feature for auricular surface photographs of Bedford et al. (1989) Feature

rs

Transverse organization Surface texture Microporosity Macroporosity Apical changes

0.942 0.930 0.654 0.883 0.888

P 0.01 0.01 0.01 0.01 0.01

not complete. They are usually smooth-edged and do not cover a significant area of the auricular surface. Macroporosity may also be confused with areas of  postmortem damage, which are usually sharp-edged and irregular, often have paler colored bone around the edges, and may expose underlying trabecular bone that is always unremodelled. Care should be taken in recording this feature, to ensure that no confusion with postmortem damage has arisen.   Apical changes

The apex of the auricular surface can develop small osteophytic growths, or lipping, which when more severe can alter the contour of the surface . This change was regarded as a secondary aging feature by Lovejoy et al. (1985), and was described in terms of “activity.” The scoring system for apical changes is given in Table 6. Preliminary testing

The new scoring system was initially applied to archaeological skeletal material of unknown age at death from the Anglo-Norman site of Blackgate, Newcastle-upon-Tyne (n  56; held at the Department of Archaeology and Prehistory, University of  Shef field), to test for ease of use and to check for reproducibility of results. The scoring system was then applied to photographs of known-age auricular surfaces distributed by Bedford et al. (1989) that accompany the system of Lovejoy et al. (1985), to see how well the individual features of the auricular surface correlated with age (Table 7). Each feature was found to have a high correlation with age, consistent with the fact that the auricular surfaces pho-

Feature (constant) Transverse organization Surface texture Microporosity Macroporosity Apical changes

Unstandardized coef ficients

t

6.500

1.156

4.316 7.105 5.572 4.669 3.282

P

0.249 2.527 0.012 4.115 0.001 3.680 0.001 2.720 0.007 2.675 0.008

tographed were specifically chosen to illustrate the typical appearance seen at a particular age. Blind test on a known-age sample

The new scoring system was then tested by one of  the authors (J.L.B.) on a known-age skeletal population (n  180) from Christ Church, Spitalfields, London (held at the Natural History Museum, London; Molleson and Cox, 1993), to assess how the features related to chronological age. Each auricular surface in the collection was recorded twice, with an interval of 2 weeks between these analyses. Any discrepancies between scores were noted, and if discordant, the surface was studied a third time. The age and sex of the individual was not recorded on the boxes, and care was taken that the observer did not know this information prior to recording. While the sex of strongly dimorphic individuals would be apparent to the observer, it was not felt that this would cause bias, as the recording criteria were rigorous, and applied solely to the auricular surface and not to the surrounding bone. The observer also avoided examining other parts of the skeleton, which may have given subjective clues to age at death. Color photographs were taken of every auricular surface. Once the study was completed, the photographs were seriated for each feature, to check the consistency of the recording. The scores were then entered into a computer database along with the documented age and sex of each individual. RESULTS

  Although Kendall’s coef ficient of concordance indicates a reasonable measurement of agreement among the fi ve features (Wa  0.691, P  0.0005), each feature supplies independent information about age. This is seen in the results of a stepwise multiple regression, which indicates that all fi ve features contribute to the prediction of age (Table 8), and in the partial correlations between features after controlling for the effect of age (Table 9). The partial correlations among the features are low and mostly nonsignificant, confirming that the features provide independent sources of information about age. The combination of transverse organization, surface texture, microporosity, macroporosity, and apical changes was found to give the highest correlation with age (rs  0.609; P  0.01). This “composite score,” calculated as the sum of the scores of the

235

REVISED AURICULAR SURFACE AGING METHOD TABLE 9. Partial correlation coefficients between features controlling for effects of age1 TO TO ST MI MA AP

ST 0.381

0.001

0.070 0.004 0.590

0.065 0.001 0.023

MI

MA

AP

0.136 0.138

0.217 0.316 0.104

0.040 0.169 0.038 0.059

0.164 0.612

0.429

1

Coef ficients above diagonal, p-values below, sample size n  180. TO, transverse organization; ST, surface texture; MI, microporosity; MA, macroporosity; AP, apical changes.

Fig. 2. Plot of composite auricular surface scores against age for Spitalfields sample. Spearman’s correlation coef ficient of score against age is given for males and females.

separate features, was used in all subsequent statistical tests.  A paired t-test was calculated for all individuals with both left and right auricular surfaces present. The results showed no significant difference for the composite score between the left and right sides (t  0.6; df  157; P  0.550). Subsequently, the left side was used for all analyses when both sides were available, and the side available was used for all other individuals. Previous analyses had revealed no significant difference in age-related changes of the auricular surface between males and females (Lovejoy et al., 1985; Murray and Murray, 1991). The ages were plotted against composite scores for males and females as a scatter plot, with a regression line for both data sets (Fig. 2). The regression lines indicate that there is little difference between sexes (males, rs  0.624, P  0.01; females, rs  0.626, P  0.01).   A standard (two-sample, two-tailed) t-test was also carried out to test for significant differences between males and females for each of the score categories. There were no significant differences between ages for males and females ( P  0.05; see Table 10).  A scatter plot of score against age for each feature, shown in Figure 3, reveals wide variation in the age of appearance of the features. A Spearman’s correlation coef ficient was calculated for each feature

TABLE 10. Independent two-tailed t-tests between males and  females for each composite score Composite score 5–6 7–8 9–10 11–12 13–14 15–16 17–19

 Age stage I II III IV V VI VII

t

df 

P

Sample size too small 4 0.757 20 0.083 1.825 0.373 30 0.711 0.005 62 0.996 39 0.888 0.142 1.125 10 0.287 0.331

with age (Table 11). Although the r values are relatively low, all were significant at the 99% confidence level. The scores were combined to produce a composite score, which increased the correlation with age (males, rs  0.624; females, rs  0.626; P  0.01). Several composite scores were found to have similar ranges, distributions, and mean ages. These scores were grouped together to produce seven auricular surface stages for the purposes of age estimation. The age ranges, means, and median ages for these stages are given in Table 12. It should be noted that the mean age for auricular surface stage is affected by the age structure of the reference sample, which in this case is biased toward older ages (Bocquet-Appel and Masset, 1982; Jackes, 1992). To compensate for this, a Bayesian analysis was undertaken, using uniform priors following the method of  Chamberlain (2000) to provide posterior probabilities of age, given the auricular surface stage (Table 13). Reliability testing

The method was tested for intraobserver error by one of the authors (J.L.B.). Six well-preserved auricular surfaces were recorded twice, at an interval of 2 weeks. A paired t-test showed no significant variation ( P  0.05) between the composite scores given on each occasion (t  1.348; df   5; P  0.235). A different group of six auricular surfaces was tested for interobserver error by graduate students in osteology at the University of Shef field and members of the British Association of Biological Anthropology and Osteoarchaeology. Cohen’s weighted kappa measure of agreement for ordinal data indicated a low level of interobserver error (  0.66) (Agresti, 1988). The Spitalfields sample was also recorded using the Suchey-Brooks pubic symphysis method, which was applied using the casts for males and females (Suchey et al., 1988; Brooks and Suchey, 1990). The pubic symphysis was only present for 58.9% of the sample. The rank order correlation between the six Suchey-Brooks stages and age is rs  0.58; P  0.01. This compared favorably with the correlation between the new seven-stage auricular surface aging method and age (rs  0.63; P  0.01). This shows that although the Suchey-Brooks system is thought to be the most reliable skeletal indicator of age (Mays, 1998), the new auricular surface aging

236

J.L. BUCKBERRY AND A.T. CHAMBERLAIN

Fig. 3. Scatter plots of scores against age at death in Spitalfields sample for each of  fi  ve features: transverse organization, surface texture, microporosity, macroporosity, and apical changes. TABLE 11. Spearman’ s correlation with age for each feature and for composite score Feature Transverse organization Surface texture Apical changes Microporosity Macroporosity Composite score

rs 0.433 0.489 0.319 0.408 0.533 0.609

P 0.005 0.005 0.005 0.005 0.005 0.005

method has a higher correlation with age for the Spitalfields sample. DISCUSSION AND CONCLUSIONS

Individual human skeletons show age-related changes which progress at different rates. It would appear that none of the current methods of skeletal age estimation are able to provide age estimates that are both precise (closeness of repeated measurements) and accurate (closeness to true value) (Sokal and Rohlf, 1995, p. 50), and that such precise levels of age estimation are unlikely to be developed in the foreseeable future. While precise age estimates are desirable and will always be sought, it is not feasible to attempt to provide them if they are not a true reflection of the underlying biology. None of the present age-estimation methods are able to provide estimates within narrow age ranges, so consequently age estimates need to be stated either as broad ranges or as probability density functions (e.g., Jackes, 1992; Chamberlain, 2000). Multiple regression shows that the features of the auricular surface age independently of each other, indicating that each feature has to be interpreted independently, rather than being grouped together

with other features into 5-year modal age stages. Our revised auricular surface method of age estimation allows for a more realistic interpretation of the changes. Although the age estimates produced by our method are wider, this revised method is easier to apply and may be more reliable than that of  Lovejoy et al. (1985). The sample size used to develop the method was small (n  180), with younger individuals underrepresented due to the age structure of the Spitalfields Collection. The method outlined here needs to be tested and redefined, using large, multiracial, and known-age modern and, if possible, archaeological populations. This will lead to the redefinition of age ranges and standard deviations. Although Murray and Murray (1991) showed that the method of Love joy et al. (1985) was equally applicable for different populations, this has not yet been demonstrated for our revised method. Such a study is needed to determine if the method is universally applicable across human populations, and will have important implications for the applicability of the method to archaeological samples. It has also been shown that there is no significant difference between sexes and sides. The lack of systematic variation in the rate of aging between the sexes was surprising, but allows the same method and age estimates to be applied to both sexes, allowing age to be assessed independently of sex. The revised method generates scores that have a higher correlation with age than the Suchey-Brooks pubic symphysis system for the Spitalfields sample. The age ranges produced by the revised auricular surface method are similar to those given by Brooks and Suchey (1990), regarded as one of the most

237

REVISED AURICULAR SURFACE AGING METHOD TABLE 12. Age estimates from composite scores and age stages Composite score

Auricular surface stage

No. of specimens

Mean age

Standard deviation

Median age

Range

I II III IV V VI VII

3 6 22 32 64 41 12

17.33 29.33 37.86 51.41 59.94 66.71 72.25

1.53 6.71 13.08 14.47 12.95 11.88 12.73

17 27 37 52 62 66 73

16–19 21–38 16 –65 29 –81 29 –88 39 –91 53 –92

5–6 7–8 9–10 11–12 13–14 15–16 17–19

TABLE 13. Posterior probability of age, given auricular surface stage, assuming uniform prior probability of age Age 15–24 25–34 35–44 45–54 55–64 65–74 75–84 85–94

I

II

0.86 0.14 0.00 0.00 0.00 0.00 0.00 0.00

0.33 0.42 0.21 0.04 0.00 0.00 0.00 0.00

III 0.27 0.18 0.20 0.12 0.07 0.02 0.04 0.09

IV 0.00 0.17 0.07 0.16 0.20 0.14 0.20 0.07

V

VI

0.00 0.03 0.19 0.20 0.18 0.26 0.07 0.06

0.00 0.00 0.02 0.04 0.11 0.10 0.27 0.46

VII 0.00 0.00 0.00 0.27 0.00 0.24 0.49 0.00

important and reliable aging methods used at present. This suggests that wide age estimates are more realistic. Narrow age estimates will always remain desirable, but by their nature do not allow for the variation in age-related changes demonstrated by this study. By using narrower age estimates, many individuals could be aged incorrectly, which may hinder identification in forensics cases. This aging technique is important, because the auricular surface is resistant to decay (Waldron, 1987), especially when compared with the fragile pubic symphysis. It is universally applicable to both

sexes and potentially to different populations (Konigsberg and Frankenberg, 1992), and shows older age-related changes than the pubic symphysis (Lovejoy et al., 1985).  ACKNOWLEDGMENTS

We thank Dr. Louise Humphrey of the Natural History Museum, London, for granting us access to the Christ Church Spitalfields skeletal sample. Comments by Dr. Humphrey and Dr. Pia Nystrom of the Department of Archaeology and Prehistory, University of Shef field, on previous drafts of this paper and the suggestions made by anonymous referees are greatly appreciated. We also thank the osteology students and delegates of the British  Association of Biological Anthropology and Osteoarchaeology for taking part in the interobserver error tests, and Alex Norman for drawing the illustration. J.L.B. is currently funded by a White Rose Studentship from the Universities of Sheffield, York, and Leeds.

 APPENDIX. Age and feature scores for known-age females (n Age Females 16 17 19 23 27 27 29 29 30 34 35 35 37 37 38 39 39 41 43 44 45 45 46 47 47 48 49 49

TO 1 2 1 3 1 2 4 2 3 4 2 2 3 3 2 3 5 2 4 3 2 4 4 4 4 2 4 4

ST 1 1 2 2 1 2 3 3 3 3 2 3 3 2 3 2 3 3 3 3 3 3 3 3 4 3 3 3

MI 1 1 1 2 3 2 3 3 2 3 3 3 3 1 3 2 3 3 3 3 2 3 3 3 3 2 2 3

MA 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 2 1 2 1 1 3 1

AP 2 1 1 2 1 1 2 2 2 2 1 2 1 1 1 1 3 2 2 2 3 2 2 2 2 2 3 2

Age 50 50 50 52 52 52 53 53 53 53 53 55 55 55 56 56 56 57 57 57 58 59 60 61 61 61 62 63 64



94) and males (n

TO 4 4 4 4 4 5 2 4 4 5 2 3 4 4 4 3 4 4 4 2 4 4 4 4 4 3 4 4 3

ST 3 4 4 3 4 3 2 3 2 5 3 3 4 4 4 3 4 3 3 3 4 4 3 3 3 3 3 4 4



86) MI 3 3 3 2 3 3 3 3 3 3 2 1 2 3 3 3 3 3 3 3 3 3 2 3 3 3 2 2 3

MA 1 2 2 2 3 2 1 2 1 3 1 2 1 2 1 1 1 1 1 2 2 2 2 3 2 1 1 2 2

AP 2 1 2 2 2 2 1 2 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 2 2 2

(continued)

238

J.L. BUCKBERRY AND A.T. CHAMBERLAIN  APPENDIX. Age and feature scores for known-age females (n

Age 65 65 65 65 65 67 67 67 68 68 68 68 70 70 71 72 73 73 74 76 76 76 77 77 78 78 79 79 80 80 81 83 85 85 87 87 89 Males 16 18 21 22 25 26 27 32 32 32 34 34 34 34 36 36 37 38 39 39 39 40 44 45

TO

ST

MI

MA

AP

2 4 4 4 4 3 4 4 3 4 5 4 4 3 4 3 4 3 4 4 4 4 5 5 5 4 5 4 4 4 5 4 4 5 4 3 4

3 4 3 3 3 3 3 4 4 3 4 3 3 3 3 4 2 3 4 3 3 3 5 3 5 4 4 4 4 3 4 4 4 4 4 3 4

2 3 3 3 3 3 3 3 3 3 3 3 3 3 2 3 3 3 3 3 3 2 3 2 3 3 3 3 3 3 3 3 3 3 3 3 3

1 2 1 2 2 2 1 2 1 1 3 2 3 2 2 3 2 3 3 2 2 2 3 1 3 3 2 2 3 1 2 3 3 2 3 3 3

2 2 2 2 2 1 1 2 2 1 1 2 1 3 3 2 2 2 2 3 1 2 1 2 3 3 2 2 2 2 2 2 3 2 3 2 2

2 4 2 3 2 2 3 3 4 5 2 3 4 4 3 2 2 2 2 2 4 4 2 5

2 3 3 3 3 1 2 3 3 3 2 3 2 3 2 3 4 2 2 3 3 3 3 3

1 1 1 2 1 3 1 3 3 2 2 2 3 2 3 1 3 2 3 3 3 2 3 3

1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2

3 1 1 1 2 1 2 2 2 2 2 1 2 1 1 2 2 1 2 1 2 1 1 1

Age 46 47 47 48 49 50 51 51 52 52 53 53 54 56 57 58 58 58 58 58 60 60 60 61 61 62 63 63 63 63 63 64 64 64 65 66 66 66 67 67 68 68 69 69 69 70 70 70 71 71 71 72 72 75 76 77 78 79 81 88 91 92



94) and males (n TO 3 3 4 4 5 4 4 4 2 4 4 5 4 5 4 4 4 4 3 2 3 4 4 4 4 4 3 4 3 3 5 5 4 4 4 5 4 4 4 5 4 5 4 5 3 4 4 4 5 4 4 4 4 3 2 4 5 4 3 4 4 4



ST 3 2 3 4 3 3 3 3 3 3 3 4 3 4 4 4 3 3 3 3 4 4 4 4 4 2 3 3 3 4 4 4 4 4 3 3 3 4 3 4 3 5 4 4 4 3 3 3 3 3 4 4 3 3 3 3 4 3 3 3 4 4

86) (Continued) MI 2 3 2 2 2 3 1 3 3 3 3 3 3 3 3 2 3 3 3 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 1 3 3 3 3 3 3

MA 1 1 1 2 1 2 1 1 3 2 1 3 1 1 2 1 1 1 1 1 2 2 3 3 2 1 2 2 2 1 3 2 3 2 1 2 2 1 1 2 3 3 1 3 2 1 2 2 3 1 2 2 2 2 1 1 2 3 1 1 1 3

AP 2 3 2 2 3 3 1 1 2 2 1 2 2 2 2 2 2 2 2 2 1 2 2 2 2 1 2 2 1 3 2 2 3 3 2 2 2 3 2 2 2 2 2 2 2 3 2 2 1 2 2 3 2 2 3 2 2 2 2 2 3 3

 Abbreviations as in Table 9.

LITERATURE CITED   Agresti A. 1988. A model for agreement between ratings on an ordinal scale. Biometrics 44:539 –548.   Aykroyd RG, Lucy D, Pollard AM, Roberts CA. 1999. Nasty, brutish, but not necessarily short: a reconsideration of the statistical methods used to calculate age at death from adult human skeletal and dental age indicators. Am Antiq 64:55 –70.

Bedford ME, Russell KF, Lovejoy CO. 1989. The auricular surface aging technique: 16 color photographs with descriptions. Kent, OH: Kent State University. Bedford ME, Russell KF, Lovejoy CO, Meindl RS, Simpson SW, Stuart-MacAdam PL. 1993. Test of the multifactorial aging method using skeletons with known ages-at-death fron the Grant collection. Am J Phys Anthropol 91:287–297.

REVISED AURICULAR SURFACE AGING METHOD Bocquet-Appel JP, Masset C. 1982. Farewell to palaeodemography. J Hum Evol 11:321 –333. Brooks ST. 1955. Skeletal age at death: the reliability of cranial and pubic age indicators. Am J Phys Anthropol 13:567 –598. Brooks ST, Suchey JM. 1990. Skeletal age determination based on the os pubis: a comparison of the Asca´ di-Nemeske´ ri and Suchey-Brooks methods. Hum Evol 5:227–238. Brothwell D. 1981. Digging up bones. New York: Cornell University Press. Buikstra J, Ubelaker DH. 1994. Standards for data collection from human skeletal remains. Fayetteville, Arkansas: Arkansas Archaeological Survey. Chamberlain AT. 2000. Problems and prospects in palaeodemography. In: Cox M, Mays S, editors. Human osteology in archaeology and forensic medicine. London: Greenwich Medical Media, Ltd. p 101–116. Gilbert BM, McKern TW. 1973. A method for aging the female os pubis. Am J Phys Anthropol 38:31–38. Hillson S. 1996. Dental anthropology. Cambridge: Cambridge University Press. Jackes M. 1992. Palaeodemography: problems and techniques. In: Saunders SR, Katzenberg MA, editors. Skeletal biology of  past peoples: research methods. New York: Wiley Liss, Inc. p 189 –224. Katz D, SucheyJM. 1986. Age determination of the male os pubis.  Am J Phys Anthropol 69:427 – 435. Konigsberg LW, Frankenberg SR. 1992. Estimation of age structure in anthropological demography. Am J Phys Anthropol 89:235–256. Lovejoy CO, Meindl RS, Pryzbeck TR, Mensforth RP. 1985. Chronological metamorphosis of the auricular surface of the ilium: a new method for the determination of adult skeletal age at death. Am J Phys Anthropol 68:15–28. Mays S. 1998. The archaeology of human bones. London: Routledge. Molleson T, Cox M. 1993. The Spital fields project, volume 2: the anthropology. The middling sort. York: Council for British Archaeology.

239

Murray KA, Murray T. 1991. A test of the auricular surface ageing technique. J Forensic Sci 36:1162–1169. Nemeske´ ri J, Harsa´nyi L, Acsa´ di G. 1960. Methoden zur Diagnose des Lebensalters von Skelettfunden. Anthropol Anz 24: 70 –95. Sashin D. 1930. A critical analysis of the anatomy and the pathological changes of the sacro-iliac joints. J Bone Joint Surg 28:891–910. Saunders SR, Fitzgerald C, Rogers T, Dudar C, McKillop H. 1992.   A test of several methods of skeletal age estimation using a documented archaeological sample. Can Soc Forensic Sci J 25:97–118. Schwartz JH. 1995. Skeleton keys. Oxford: Oxford University Press. Sokal RR, Rohlf FJ. 1995. Biometry, 3rd ed. New York: W.H. Freeman and Co. Suchey JM. 1979. Problems in the aging of females using the os pubis. Am J Phys Anthropol 44:263 –270. Suchey JM, Brooks ST, Katz D. 1988. Instructions for use of the Suchey-Brooks system for age determination of the female os pubis. Instructional materials accompanying female pubic symphysial models of the Suchey-Brooks system. Fort Collins, Colorado: France Casting. Suchey JM, Wisely DV, Katz D. 1986. Evaluation of the Todd and McKern-Stewart methods for aging the male os pubis. In: Reichs KJ, editor. Forensic osteology: advances in the identification of human remains. Springfield, IL: Charles C. Thomas. Todd TW. 1920. Age changes in the pubic bone I. The male white pubis. Am J Phys Anthropol 3:285–334. Ubelaker DH. 1989. Human skeletal remains: excavation, analysis, interpretation. Washington, DC: Taraxacum. Waldron T. 1987. The relative survival of the human skeleton: implications for palaeopathology. In: Boddington A, Garland   AN, Janaway RC, editors. Death, decay and reconstruction: approaches to archaeology and forensic science. Manchester: Manchester University Press. p 55– 64.