How to build a mammalian super-predator

Abstract

Questions surrounding the biology of large fossil predators that differ markedly from living forms have long intrigued palaeobiologists. Among such taxa few have excited more interest than sabertooth cats, whose distinctive hypertrophied canines are suggestive of killing behaviors and feeding ecologies that may have departed widely from those of extant carnivores. Moreover, considerable variation among sabertooth species is further suggestive of intriguing differences within the group. Behavior and ecology in another large, extinct mammalian carnivore, the Australian marsupial lion ('Thylacoleo carnifex'), has also proven contentious. In this study, we assemble a wide range of cranio-dental and postcranial indices in a dataset including machairodont sabertooths, 'T. carnifex' and an extensive sample of extant taxa in order to examine the palaeobiology of these charismatic fossil carnivores. Results of multivariate analyses point to significant relationships between behavior and overall body proportions in extant mammalian carnivores. Postcranial morphologies of two American dirk-tooth species of sabertooth (Smilodon) depart greatly from those of living felids and group most closely with bears among living placentals. Scimitar-tooth species of Homotherium and Machairodus cluster with modern pantherine cats. The marsupial lion groups with Smilodon. If these latter two phylogenetically disparate clades do represent a specialized, robust ecomorph adapted to predation on large prey, then it is a body plan that might be effectively identified on the basis of a handful of 'bear-like' postcranial features in combination with a more typically 'felid-like' carnassialization of the cheektooth row.