1Department of Anthropology, New York University, 2Department of Evolutionary Anthropology, Duke University, 3Center for the Study of Human Origins, Department of Anthropology, New York University
Saturday All day, Clinch Concourse
The vertebral column plays central roles in posture, stability, and locomotion. Its numerical composition is somewhat conserved across phylogenetic groups, which may result from developmental constraints and/or stabilizing selection. Deciphering the role of selection versus constraint in this complex anatomical system is therefore of interest in functional and evolutionary studies.
Compared to other vertebrates, mammals are relatively conserved in pre-caudal vertebral formulae. Hominoids are intra- and inter-specifically quite variable in vertebral counts, and because their vertebral formulae are derived relative to non-hominoid primates and many other mammals, an understanding of the forces that drove their evolutionary history may be aided by a comparative study. Using ARLEQUIN, we analyze a dataset of 5735 mammals from 648 species (435 genera), representing all major divisions of Mammalia. Following Pilbeam, we use a morphological analog of Nei’s genetic identity index to quantify interspecific variation in vertebral formulae by creating a ratio of shared vertebral patterns between two populations to the total amount of variation in both populations.
Results demonstrate a significant case of convergence between giant pandas (Ailuropoda melanoleuca) and hominoids; in turn, giant pandas generate no similarity with other bears. Like hominoids, giant pandas demonstrate reduced trunks and numerically long sacra. Although more detailed analyses on the evolutionary morphology of ursid and hominoid vertebral columns are required, the observation that both groups demonstrate complex manual manipulation during upright feeding postures is intriguing. Whether or not this behavior played a selective role in their extensive convergence is a hypothesis that will require further testing.
This study was supported by National Science Foundation Grant BCS-0925734 and a Beckman Institute for Advanced Science and Technology Cognitive Science/AI Award to SAW.