Shape analysis of the zygoma to assess ancestry and sex variation in modern South Africans

Abstract

Skeletal remains exposed to an outdoor context are prone to post-mortem damage and fragmentation, making skeletal analysis difficult for the anthropologist. Research on ancestry and sex from isolated fragments of the cranium is necessary to improve identification of fragmented remains. The zygoma has proven to be more durable post skeletonization than other cranial bones, making research relevant into variation within the zygoma. Whilst the shape of the zygoma has been studied in a South African population using morphological, metric and geometric morphometric techniques, these studies did not include Indian South Africans. The Indian South African population comprises 2.6% of the total population but make up a larger proportion of the population in certain areas. For example, Indian South Africans comprise 7.4% of the population in Kwa-Zulu Natal and 2.9% in Gauteng. More specifically, Indian South Africans make up to 60% of the population in the suburb of Chatsworth with a further 91% of the population in sub-area of Arena Park, and 80% of the population in the Laudium suburb of Gauteng. Therefore, Indian South Africans must be included in anthropological studies attempting ancestry classifications. The purpose of the study was to assess the shape variation and projection of the zygoma attributable to sexual dimorphism and ancestral variation among South Africans, including Indian South Africans, using a geometric morphometric approach. A sample of 400 three-dimensionally (3D) reconstructed models from head CT scans of black, coloured, white, and Indian South Africans were used with an equal sex and ancestry distribution. Eleven landmarks previously described in the literature were used for the analysis. Each landmark was used to depict the most prominent points on the outline of the zygoma. Additionally, semi-landmarks were placed along the curves of the zygoma. The landmarks and semi-landmarks were tested for observer repeatability and reliability using dispersion analysis and revealed that all landmarks were repeatable. Procrustes ANOVA revealed significant differences among the population groups and between the sexes for all population groups, except between coloured South African males and females. A pairwise post-hoc test revealed that white and Indian South Africans had the most similarities except for males, where coloured and Indian South Africans had the most similarities for landmarks. Three interlandmark distances were created to assess the zygoma’s projection. The ANOVA for the projection of the zygoma revealed significant differences for both sex and ancestry except for white South African males and females and males overall for the zygomaticomaxillary length. The zygomaticomaxillary length (ZML) is defined as the maximum distance between the landmarks zygoorbitale and zygomaxilare. No significant differences were noted for female South Africans for the Superior Zygomatic Length which, is a measure of the maximum length of the superior margin of the zygoma (between porion and zygoorbitale; PorZygool). Further analysis of the zygoma’s projection involved creating angles between the interlandmark distances. The ANOVA for the angles of projection revealed significant differences between sexes and populations, except for white and Indian South African males and females at Angle1 (Angle at the intersection of ZML and PorZygoml) and Angle3 (the angle at the intersection of PorZygool and PorZygoml) and black, coloured and Indian South African males and females at Angle2 (the angle at the intersection of ZML and PorZygool). The large amount of overlap amongst ancestry groups demonstrated substantial group similarities; however, differences were noted at the zygomaxillary, zygomaticotemporal and frontomalar sutures. Overlap was also present between males and females, but on average, males were larger than females. Differences, such as a more inferior placement of the zygoorbitale landmark were noted at the inferior margin of the orbit specifically in females. Differences were also noted at the inferior margin of the orbit across all groups. Discriminant functions were created to assess the classifying ability of the shape of the zygoma. Results revealed low accuracies for ancestry classification for the shape and projection of the zygoma. However, higher accuracies were noted for sex classification for the shape and projection of the zygoma. While results demonstrate shape variation of the zygoma, the classifying ability of the zygoma is precarious at best, and the use of the zygoma in a forensic context may not be an option. However, the differences observed can be taken into consideration during medical procedures such as zygomatic and infraorbital implants. Although landmark placements were reliable and repeatable, further analysis of the zygoma using a semi-automatic surface registration method along with different imaging techniques (MicroCT and CBCT scans) may assist in the data collection procedure and may potentially increase the accuracy of the results. Furthermore, the results of the current study highlight the need for the assessment of the effects of diet, climate, age, edentulism and symmetry on the shape of the zygoma.