Files
Abstract
The diverse ways in which mammals move through their environments are reflected in adaptations in their appendicular skeletons. One of the most derived postcranial adaptations found in mammals is unguligrady, where only the tips of the toes contact the ground in a resting posture. Unguligrady is often observed in mammals considered to be adapted for efficient long-distance locomotion, and is accompanied by a suite of morphological characters, including moderate to extreme lateral digit loss and elongation of the foot bones. The rise of open habitats over the Cenozoic is thought to have contributed to selection for distally lengthened elements, but the link between the rise of grasslands and locomotor morphology has yet to be explicitly tested, especially in a macroevolutionary context. Chapter two of my dissertation highlights the novelty of the Lateral-Sequence Lateral-Couplet (LSLC) gait, a sequence of footfalls thought to coincide with long limbs, relative to body size. This study shows how, within a broad sample of 244 quadrupedal tetrapods, long-limbed mammals employ unique walking patterns relative to other taxa, demonstrating a need to further investigate the ecology of animals with this gait type. Members of the suborder Ruminantia (Artiodactyla) primarily use the LSLC gait, are unguligrade, and offer a diverse assemblage of ~200 extant species in six families that span several orders of magnitude of body masses and live in ecological extremes. The fossil record of Ruminantia spans 45 million years and is well-represented in North American fossil assemblages. Ruminants therefore are an ideal study system to examine the environmental and functional impacts associated with relatively long limbs and unguligrade locomotion. As it is hypothesized that ungulate limbs lengthened to achieve decreased locomotor costs as a response to increasing grassland habitats over the Cenozoic, I have used a macroevolutionary approach to answer questions about changing locomotor morphology in ruminants and shifts in habitats over time. The third and fourth chapters of my dissertation develop methods for predicting body mass and habitat importance using 65 postcranial measurements across 10 skeletal elements. Using phylogenetically informed bivariate and multivariate analyses, I show in chapter three that a comprehensive approach for estimating body mass from many skeletal elements allows for flexibility when predicting mass from fragmented fossil remains. Chapter four of my dissertation uses a novel Bayesian multilevel modeling approach for predicting habitat importance in living ruminants. Using functional indices, these models show a steady importance of Forest habitats through time with a slow and moderate rise in Savanna habitat importance, while Grassland importance increases dramatically over time. Broadly, my results show that grasslands became an important part of fossil ruminant ecology between the Oligocene and Miocene, consistent with paleobotanical evidence and previous studies showing a shift in limb proportions related to opening habitats. This framework provides a novel way to identify trends in fossil mammal ecology and supports the hypothesis that unguligrade locomotor morphology responds to changes in habitat composition.