Foetal bison long bones and mortality season estimates at the early Holocene Casper and Horner II sites, North America.

Foetal animal remains present the opportunity to investigate seasonal prey mortality in archaeological contexts. Mortality seasons may be inferred by adding a gestation age, as estimated from foetal skeletal growth curves, to a known conception period. For example, given an October conception period, the remains of a 6‐month‐old foetal animal suggest deposition in April. However, three problems complicate these investigations: (1) Conception periods in animal populations are generally not discrete windows or dates but rather distributions of events though a calendar year; (2) variation in foetal development implies that any given skeletal state corresponds to a range of gestation ages, and (3) evolutionary changes in a taxon's foetal growth may make modern growth curves poor analogues for past populations. We develop a probabilistic method to address the first two problems and partially address the third problem. The method is specific to early Holocene bison in North America (~10 000 to 8000 cal. BP), and we provide example applications: the Horner II Site, with one foetal humerus, and the Casper Site, with five foetal elements. Results for Horner II indicate that the foetal humerus is consistent with the fall or winter mortality season inferred from bison dentitions at the site, whereas results for Casper suggest a spring mortality event or events distinct from most of the site's adult bison specimens. This may reflect real seasonality variability at Casper, although methodological assumptions warrant caution for this interpretation; Future work should focus on testing these assumptions. [ABSTRACT FROM AUTHOR]