ABSTRACT
Sex in the adult skeleton can usually be reliably determined through an assessment of features found on the pelvis and cranium. In the lack of these elements it is necessary to elaborate other methods to establish sex in skeletonised remains recovered in forensic cases. Standards for other bones (e.g. humerus, metacarpals and metatarsals) have already been established for the Greek population. The aim of this study is to determine whether the standards for metacarpals provided from a study on the Athens collection are representative of a modern Cretan population. Using a digital caliper we took 7 measurements on each one of the left and right metacarpal bones of 108 adult individuals from a modern collection from Crete. Totally twenty formulae for left and right bones created from the Athens collection were used to sex the sample of this study. The overall classification accuracy obtained for our sample was very close to the cross-validated accuracy reported by the authors. However, looking at the classification accuracy for males and females, a consistent trend for low classification rates in females was observed. New formulae were developed for the Cretan sample yielding up to 85% classification accuracy. This study clearly indicates that the standards for metacarpals developed from the Athens collection are not appropriate for application in forensic cases for the island of Crete as they do not represent the local population efficiently. This may hold true for other regions of Greece thus great caution should be taken when applying these standards. Obviously more research is needed to confirm these results.
