ESTIMATING BODY MASS IN EXTINCT THERIAN MAMMALS USING OCCIPITAL CONDYLE WIDTH

Body mass is a trait of paramount importance to mammalian paleontologists, as it influences numerous parameters of paleoecology and evolutionary history including dietary habits, life history, physiology, and extinction risk. However, estimating body mass has proven difficult in cases where taxa have non-analogous dentitions or postcrania, few or no living relatives, are outside the range of body sizes spanned by living forms, or exhibit proportions unlike any living mammal (e.g., the disproportionately large heads of “creodonts” and other extinct groups). Body mass has traditionally been estimated by regression equations based on craniodental and long bone measurements, though pitfalls with both of these methods have been noted in previous studies. Occipital condyle width (OCW) offers a possible alternative to traditional cranial and postcranial metrics. Due to the occiput’s role in connecting the cranium and postcranium, OCW is expected to correlate with the size of the postcranium and be highly subject to stabilizing selection. I constructed a database of 22 orders and 349 species of mammals using specimens with known live weights to investigate the correlation between OCW and body mass across Mammalia. OCW accurately predicts body mass within Theria with a higher degree of accuracy than many previous craniodental regression equations. Taxa that exhibit high error in body mass estimates typically exhibit morphological apomorphies of the occiput, such as narrow, pulley-like condyles in lagomorphs, allowing them to be identified as unsuitable for OCW-based regression a priori. Phylogenetic signal is low, except for rate shifts related to occiput specialization at the base of several mammalian clades such as lagomorphs. Because the relationship between OCW and body mass is consistent across Theria, it is possible to produce phylogenetically bracketed body mass estimates for wholly extinct groups. Several extinct therian mammals for which body mass estimates have been problematic (creodonts, sparassodonts, giant caviomorph rodents) are used as case studies for potential applications of OCW in body mass estimation. Creodont and sparassodont body mass estimates generally agree with previous postcranial-based ones, though the creodont Megistotherium and the sparassodont Paraborhyaena are estimated as much smaller than previously predicted. Lower body mass estimates (200-400 kg) are supported for the giant caviomorphs Phoberomys and Josephoartigasia.