Sex determination in miocene catarrhine primates

Canines of fossil hominoids and primitive catarrhines from several early, middle, and late Miocene sites were analyzed according to the shape indices described in Kelley (1995) and compared to those of males and females of extant great apes. In bivariate plots of the fossil canines utilizing the indices, 90% of the upper canines and 85% of the lower canines fell within or just outside the exclusively male or exclusively female territories delimited by the extant great apes. The remainder fell in the male-female overlap zones. Sex assignments based on these distributions were nearly 100% concordant with classifications according to canine height, suggesting a high degree of accuracy. There were various taxon-specific shifts in bivariate space among fossil genera, reflecting subtle differences in canine shape between taxa within the overall pattern of similarity to extant great apes as a whole. In many cases these shifts are matched by particular extant-ape species and subspecies, while other fossil taxa have no exact analogue for canine shape among the extant great apes. However, the pattern of spatial segregation of canines identified as either male or female at each of the sites largely mirrors that of males and females within the extant-ape sample, indicating that Miocene catarrhines shared with extant great apes a common pattern of shape differences between male and female canines, regardless of taxon-specific morphologies. These observations demonstrate that the canines of fossil catarrhines can be sexed with a high degree of confidence based solely on intrinsic features of shape. This will permit more reliable characterizations of morphological sexual dimorphism among fossil species. It is also argued that canine shape is a more reliable indicator of sex in fossil taxa than are canine/molar size ratios.