Premium
Who were they really? model‐free and model‐bound dental nonmetric analyses to affirm documented population affiliations of seven S outh A frican “Bantu” samples
Author(s) -
Irish Joel D.
Publication year - 2016
Publication title -
american journal of physical anthropology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.146
H-Index - 119
eISSN - 1096-8644
pISSN - 0002-9483
DOI - 10.1002/ajpa.22928
Subject(s) - population , bantu languages , geographical distance , bioarchaeology , evolutionary biology , zulu , isolation by distance , genetic distance , geography , mantel test , genetic divergence , replicate , demography , genealogy , biology , genetic variation , statistics , archaeology , genetic diversity , genetic structure , mathematics , history , sociology , linguistics , philosophy
Objectives For bioarchaeological biodistance analyses it is common to “assume” that skeletal samples are representative of the populations to which they are attributed. Here, alternatively, samples with “known” attribution in the Raymond A. Dart Collection are assessed regarding their suitability for use in such analyses. Prior curation issues may call their ascribed identities into question. Materials and Methods These 20th century samples ostensibly derive from South African Ndebele, Sotho, Swazi, Tswana, Venda, Xosa, and Zulu populations. First, the mean measure of divergence (MMD) is used to obtain among‐sample dental phenetic distances for comparison with documented population relationships. Second, the Mantel test evaluates fit of the isolation‐by‐distance model between MMD and geographic distances, i.e., among the historic homelands. Third, R ‐matrices and minimum and estimated F st from MMD distances give an indication of genetic micro‐differentiation. Results Output from these model‐free and model‐bound analyses suggest that five and perhaps six samples are representative of their attributed populations—presenting differences along population lines and evidence of more ancient ancestry. Discussion Other than the Swazi and perhaps Nedebele, the among‐sample variation: 1) mirrors documented population history, 2) reveals a moderately positive correlation between phenetic and geographic distances, and 3) although evidencing much homogeneity, provides measures of genetic distance in support of the phenetic distances. Therefore, with the two noted exceptions—perhaps from collection issues, swamping of past genetic structure, or both, most samples appear suitable for bioarchaeological analyses. On this basis, results are offered to supplement published findings concerning the biological relationships of these peoples. Am J Phys Anthropol 159:655–670, 2016. © 2015 Wiley Periodicals, Inc.