CAT'S CRADLE: PHYLOGENY, FUNCTION, BODY SIZE, AND THE TANGLED WEB OF FELID SYSTEMATICS

Felids are a diverse and ecologically important family, and as such have been the subject of extensive paleobiological study. Any analysis of felid evolution or paleoecology, though, requires a robust taxonomic framework, and the identification and nomenclature of fossil taxa have proven to be frustratingly intractable. This is especially well illustrated by the conflation of the Miocene genera Pseudaelurus, Nimravides, and Machairodus. This taxonomic confusion may be due in large part to the preponderance of postcrania in the cat fossil record and the conserved nature of the felid appendicular skeleton. One possible exception to this rule is the humerus, which has long been used to diagnose higher taxa within Felidae. We tested the diagnostic utility of the distal humerus in order to assess the viability of many fossil felid taxa and occurrences based on postcrania. In order to do so, we gathered photographs and measurements of felid and nimravid specimens from the Miocene through the Recent. We performed a discriminant function analysis of humeral shape in feliforms to determine whether cat species can be reliable identified based only on humeri. The feliform, but non-felid, nimravids could reliably be distinguished from true felids, but within cats there was no clear phylogenetic or ecological signal. The one exception to this trend is the cheetah Acinonyx, which has a highly aberrant cursorial morphology. Instead, body size seems to be the best predictor of humeral morphology, with unrelated large-bodied taxa (e.g., Smilodon and Panthera) clustering together. Similar results were returned even when small cats were excluded. This suggests that extinct taxa based primarily on postcrania should be considered nomina dubia and that identifying and describing previously unpublished specimens including both postcranial and craniodental material are crucial to understanding felid evolution and paleoecology.