Forensic anthropology population data Sex assessment using clavicle measurements: Inter- and intra-population comparisons Miroslav Kra ´lı ´k *, Petra Urbanova ´, Martina Wagenknechtova ´ Department of Anthropology, Faculty of Science, Masaryk University, Kotla ´rˇska ´ 2, 61137 Brno, Czech Republic 1. Introduction The determination of sex from a skeleton is a priority in forensic and archeological cases. Methods for sex assessment are based on the existence of morphoscopic features in the skeleton that manifest differently according to sex or statistical differences in skeletal measurements [1]. The human clavicle is considered relatively resistant to taphonomic factors and is frequently well- preserved both in the medico-legal and bioarchaeological context [2]. Due to its specific pattern of ontogeny and age-related changes of the sternal articular surface during adulthood, the clavicle is widely used in the estimation of age at death [2,3] and for age estimation in living people [4]. However, apart from morphological age changes, the human clavicle also exhibits a large degree of sexual dimorphism. Therefore, it comes into consideration for sexing when the skeleton is incomplete, especially when the pelvis and the skull are not available. Sexual dimorphism is reflected both in the size and shape of the clavicle. According to some studies, sexes differ in the shape of the clavicle – male clavicles are more curved horizontally than those of females [5–9]. Sexual dimorphism in clavicle curvature is already present in childhood [10], suggesting that some aspects of shape dimorphism may have originated at an early stage of ontogeny. However, some authors [8,11] found significant sexual character- istics of curvature only in the medial and not the lateral clavicular angle in adults and, moreover, these differences were not consistent between different samples and differed between body sides [11]. The female clavicle is also less robust [5,7,11,12] and it has less marked muscle attachments [5,9,11,13]. No statistically significant sex differences in clavicular length were found in fetuses, newborns and children of all studied age groups [2,8]. Size dimorphism becomes progressively pronounced during puberty and adolescence [8,14,15] and originates predominantly from differences in growth timing and intensity between males and females. Consequently, in a given adult population, size measurements of clavicle are usually, on Forensic Science International xxx (2013) xxx–xxx A R T I C L E I N F O Article history: Received 29 August 2012 Received in revised form 25 May 2013 Accepted 30 August 2013 Available online xxx Keywords: Clavicle Sex assessment Side asymmetry The University of Athens Human Skeletal Reference Collection Inter-population variations Intra-population variations A B S T R A C T We studied sexual dimorphism of the human clavicle in order to describe size variation and create population-specific discriminant tools for morphometric sex assessment. The studied sample consisted of 200 skeletons of adult individuals obtained from the University of Athens Human Skeletal Reference Collection, Athens, Greece. The specimens were well-documented and represented a modern population from cemeteries in the Athens area. Six dimensions typically used for clavicle measurements were recorded. For sexing clavicles, we used both traditional univariate (limiting, demarking and sectioning points) and multivariate discriminant function analysis. The accuracy of the best five classification equations/functions ranged from 91.62% to 92.55% of correctly assigned specimens. By testing new and previously published sexing functions (Greeks, Polynesians, Guatemalans) on four available population samples (English, Indians from Amritsar, Indians from Varanasi, and data from the present study) we found that, for some combinations of tested and reference samples, the accuracy of the sex assessment may decrease even below the probability given by random sex assignment. Therefore, measurements of the clavicle should not be used for sex assessment of individual cases (both forensic and archeological) whose population origin is unknown. However, significant metric differences were also recorded among three different Greek samples (i.e. within a population). As a consequence, application of a sexing method generated from one Greek sample and applied to another Greek sample led to negligible reduction in the success of sex assessment, despite general similarities in ethnic origin (Greeks), generation structure and presumed social background of the samples. Therefore, we believe that future studies should focus on understanding the nature of the differences among within-population reference samples. ß 2013 Elsevier Ireland Ltd. All rights reserved. * Corresponding author. Tel.: +420 776 070 170; fax: +420 541 211 214. E-mail addresses: mirekkralik@seznam.cz (M. Kra ´lı ´k), urbanova@sci.muni.cz (P. Urbanova ´), martina.wag@email.cz (M. Wagenknechtova ´). G Model FSI-7341; No. of Pages 15 Please cite this article in press as: M. Kra ´lı ´k, et al., Sex assessment using clavicle measurements: Inter- and intra-population comparisons, Forensic Sci. Int. (2013), http://dx.doi.org/10.1016/j.forsciint.2013.08.029 Contents lists available at ScienceDirect Forensic Science International jou r nal h o mep age: w ww.els evier .co m/lo c ate/fo r sc iin t 0379-0738/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.forsciint.2013.08.029