Comparing individuals buried in flexed and extended positions at the Greek colony of Chersonesos (Crimea) using cranial metric, dental metric, and dental nonmetric traits

Abstract

The Greek colony of Chersonesos was founded in the 5th century BC on the coast of the Crimean Peninsula in the northern Black Sea region. Extensive archeological research has investigated the timing and mode of the Greek land seizure and the responses of the local Taurian tribes, focusing on a large necropolis dating to the earliest period of the colony, somewhere between the 5th and the 4th century BC. Relying on burial traditions as an indicator of the biogeographic origin of the deceased, it was hypothesized that individuals buried in flexed positions were Taurian whereas individuals buried in extended positions were Greek. Here, we test the hypothesis that individuals in flexed and extended burial positions are biologically different by directly analyzing the human skeletal remains. For this, we collected three different types of phenotypic data commonly used for biodistance analysis, namely, (1) cranial measurements, (2) dental measurements, and (3) dental nonmetric traits, recorded for 47 individuals. Using Gower distance coefficients, we combine the three data types in a single analysis and estimate biological relationships among a subset of well‐preserved individuals with documented flexed (n = 8) and extended (n = 13) burial position. The estimated distances show a large amount of overlap between the two groups, with the exception of two individuals that are more divergent. To statistically corroborate this finding, we use distance‐based permutational multivariate analyses of variance (PERMANOVA) and dispersion (PERMDISP). Both analyses reveal no statistically significant differences between the groups, neither in group centroids nor in group dispersions. Our results therefore contradict the idea that burial position was determined by ancestry of the deceased. This has implications for future archeological research at Chersonesos and other Greek colonies of the northern Black Sea region.

Inferring human neutral genetic variation from craniodental phenotypes

“…Although this approach is valuable for handling unpaired samples and accounts for correlations within the four data sets, it does not account for correlations between them. However, in our case, it may still be appropriate since previous research demonstrated that the different data types are largely independent from each other, at least when comparing cranial metrics, dental metrics, and dental nonmetric traits ( 27 ), or cranial nonmetric and dental nonmetric traits ( 98 ). In addition to model-free D 2 distances, we also calculated model-bound P ST distances, which incorporate relative estimates of effective population size ( N e ; SI Appendix, Table S9 ) and average estimates of heritability ( h 2 ; SI Appendix, supplementary text 2 ).…”

“…Cranial metric data collection is performed by defining a set of homologous anatomical landmarks located on the skull and by measuring either linear dimensions between them ( 21 , 22 ) or the relative position of landmark and semilandmark coordinates in two or three dimensions ( 23–26 ). Dental metric data collection is performed in a similar fashion, usually by measuring linear lengths, widths, or diagonal dimensions at the tooth crown or at the cement–enamel junction ( 27 , 28 ) or semilandmark-based crown outlines ( 29 , 30 ). Cranial nonmetric trait data collection is performed by visually scoring minor discontinuous variants, such as extra-sutural ossicles, proliferative ossifications including bridges or spurs, or variation in foramina number and location, for example ( 31–33 ).…”

Missing data in bioarchaeology II: A test of ordinal and continuous data imputation

“…Compared to other subtopics within biological anthropology, bioarchaeologists have made far less use of statistically sophisticated methods for handling missing data and are more likely to rely on deletion methods (see companion article). The areas in which imputation has been used extensively include biodistance analyses and broader investigations of population affinity (Godde & Rangel González, 2022 ; Paul et al, 2013 ; Prevedorou & Stojanowski, 2017 ; Rathmann et al, 2022 ). Noting the limitations of missing data early on, Howells ( 1973 ) proposed three options for handling missing biodistance data: mean imputation, regression, and making an educated guess.…”

Население Раннего Херсонеса Таврического По Результатам Исследования Антропологических Материалов Из Раскопок Г.Д. Белова

“…До настоящего времени данные по этой коллекции представлялись лишь частично в рамках общих сравнительных исследований (Иванов, 2016, с. 14-37;Rathmann et al, 2022) и не были полноценно введены в научный оборот.…”

“…По результатам краниологического анализа доступных материалов (Иванов, 2016) не удалось выявить однозначного соответствия группы населения Херсонеса каким-либо иным группам: по совокупности значений краниометрических признаков полученная выборка в равной степени отличается от синхронных серий из античных колоний других регионов (материковая Греция, Боспор эллинистического и римского времени) и от краниологических серий окружающего варварского населения. Внутри самой выборки также не выявлено значимых различий, которые бы указывали на "смешанный" характер колонизации 1 (Иванов, 2011;Rathmann et al, 2022). На настоящий момент многие аспекты миграций античного времени остаются неясными, в немалой степени из-за недостаточного количества сравнительных материалов.…”