Metric Differences in adult second Metacarpal Bones related to Age-at-Death and their Comparison between Recent and Historical Populations
Abstract
Hand bones can reflect external factors during ontogenesis and individual life history. The goal of this study was to examine age-related changes in the 2nd metacarpal in adulthood.
Metacarpal bones of archaeologically excavated adult skeletons from three different medieval burial sites-Cedynia (Poland), Brno-Vídeňská Street (Czech Republic), and Dětkovice-Za zahradama (Czech Republic) - were collected, described, and scanned by means of a flatbed scanner in standardized dorsal and radial positions into two-dimensional images. On them, four measurements (one length and three widths) were taken on each view and subsequently subjected to statistical regression methods in order to quantify their relationship with age at death.
These trends were compared with those in a documented sample of the Athens Human Skeletal Reference Collection of the recent Greek population. In females, no significant relationship between the length of the 2nd metacarpal and age at death (AAD) was observed.
In females, mostly positive relationships between width measurements and AAD were observed, ranging typically between 3 to 7% over 30 years, with maximum of ca. 9% in midshaft width in the right hand in dorsal view. These relationships were more statistically significant for the recent than for the medieval sample which might be attributed to differences in sample size, and the nature of AAD (documented vs. estimated).
In males, relationships between width measurements and AAD (i.e., an increase with age) were also prevalently positive but much lower than in females and mostly not statistically significant. The systematic increase of the width measurements in females and the differences from males of the same samples suggest certain specificities of women's life histories in adulthood that would be worth further investigation in terms of the influence of external factors.
Potential methodological biases due to the cross-sectional nature of the samples and sampling selectivity are further discussed.