Assessment of sex and ancestry variation of the morphology of the fibula.

The fibula is often not used for anthropological estimations, as its morphological variation is expected to be insufficient to allow accurate estimations. However, in forensic or archaeological contexts, there is the possibility that the fibula is the only bone available for analysis. The present study applied geometric morphometrics to evaluate the potential of using fibular morphology for assessment of sex and ancestry. Three-dimensional datasets of 1609 fibulae of adult South Africans (412 females, 507 males) were digitized. Datasets were submitted to Generalized Procrustes Analysis for rotation and scaling to a common centroid. Mean centroid sizes were compared using parametric testing, and morphological variation was assessed using multivariate analyses. Discriminant Function Analysis coupled with leave-one-out cross-validation testing was used to assess classification accuracy of sex and ancestry based on these shape variations. Fibulae were smaller in females than males, but show insufficient shape variation to distinguish the sexes. Fibulae of Coloured individuals were relatively and absolutely smaller than those of the Black and White ancestry groups, likely due to contributions of small-bodied groups to the genetic composition of this group. Based on shape variation, ancestry estimations of 72.4-77.2% were obtained. Evaluation of variation according to sex and ancestry combined still produced insufficient distinction between the sexes and decreased the accuracy of ancestry classification. These results show that morphological variation of the fibula may not be useful for sex estimation, but provides reasonable accuracies for ancestry estimations, making it particularly useful in cases where only the fibula is available for analysis.