A Quantitative Analysis of Homotherium serum Tooth Marks

Abstract

Humans have coexisted with large predators, such as sabertooth cats, for the entirety of their evolutionary history. Throughout their own evolutionary history, sabertooth cats have played a significant role in their ecosystems as carnivores specialized in consuming the flesh of their prey while being incapable of accessing the majority of within bone nutrients. These behaviors likely created scavengable resources, such as residual flesh and bone marrow, for mesopredators, including hominins. However, because sabertooth species are extinct and lack direct living descendants to serve as behavioral proxies, their impact on the fossil and archeological records remains underexplored. This study examines bone surface modification (BSM) from the fossil assemblage at Friesenhahn Cave, Texas, which has been interpreted as a ~19 kya Homotherium serum den. This assemblage is particularly valuable because it has yielded no evidence of human or other carnivore interaction with the skeletal material, providing a rare opportunity to examine sabertooth tooth marks with minimal risk of mistaken identification. To analyze these marks, silicone molds of fossil BSM were created and scanned using a Sensofar S Neox white light confocal optical profilometer to generate high-resolution 3D surface models. Quantitative micromorphological measurements were extracted from these models and compared to a large database of experimentally produced BSM, including marks created by extant carnivores, stone-tool butchery, percussion, and ungulate trampling. The primary goals of this analysis are to characterize the micromorphology of H. serum tooth marks and to evaluate the degree to which sabertooth-generated marks can be distinguished from those produced by other agents using quantitative methods. Results show that H. serum tooth marks can be reliability classified at the agent level, with minimal confusion between carnivore-generated marks and other modification types and can be distinguished from those of extant carnivores when included as a modeled class. This work contributes to ongoing effort to expand actualistic and middle-range frameworks by testing whether securely attributed fossil carnivore tooth marks can be incorporated into comparative reference datasets traditionally reserved for experimentally produced marks. The results have implications for identifying extinct carnivore feeding traces in the archeological record and for assessing the extent to which contextual evidence is required when evaluating carcass-processing agents. More broadly, this study informs interpretations of sabertooth feeding behavior, ecology, and potential interactions with hominins in Pleistocene ecosystems.