Ancestral variation in mid-craniofacial morphology in a South African sample

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Ancestry estimation is a critical component of the demographic profile compiled by forensic anthropologists when unknown skeletal remains are discovered. The mid-craniofacial region is most frequently used to estimate ancestry as this region reflects the genetic and morphological ancestry of an individual. The diverse composition of the South African population makes ancestry estimation problematic, and necessitates the development of reliable, population-specific standards. This study sought to characterise variations in mid-craniofacial shape and size between South Africans of European ancestry (EA), African ancestry (AA) and Mixed ancestry (MA). Metric, nonmetric and geometric morphometric assessments were performed on 392 crania from skeletal collections in South Africa. Variations in mid-craniofacial shape and size were assessed in the orbital, nasal, zygomatic and maxillary regions in two-and three-dimensions. Univariate and multivariate statistical analyses were employed to characterise variation and estimate ancestry in AA, MA and EA individuals. Multivariate analyses suggest that tightly integrated ancestral variations in each component of the mid-craniofacial region are associated with functional, regional and developmental proximities of these regions. Specifically, AA individuals exhibited wider and shorter midfacial regions than EA individuals, who exhibited the narrowest orbital, zygomatic and nasal breadths and the longest upper facial, orbital and nasal heights. EA individuals exhibited inferiorly-angled orbits, elongated nasal apertures and anteriorly projecting nasal bridges. Rounder nasal apertures, less anteriorly projecting nasal bridges and more anteriorly projecting maxillary regions were detected in AA individuals. MA individuals exhibited heterogeneity in terms of craniofacial shape and size, and therefore produced the lowest ancestry estimation accuracies. Overall, nasal and maxillary regions were the most ancestrally diverse regions. Antemortem maxillary tooth loss and midfacial trauma were confounding factors in ancestry estimation accuracies. The lowest ancestry estimation accuracies were yielded by two-dimensional metric (27%-60.2%) and nonmetric (57.1%-82.4%) methods. Metric and geometric morphometric assessments yielded the highest repeatability (≥ 95%) indicating that these methods may be more reliable for use in medicolegal contexts. Geometric morphometric shape assessments yielded the highest ancestry estimation accuracies (75-97.9%), suggesting the presence of three dimensional shape variations between ancestry groups. These results suggest that a continuum of ancestral variation, with large areas of overlap, exists across South African populations and emphasises the need to develop multivariate ancestry estimation standards which can estimate ancestry reliably.