The 82nd Annual Meeting of the American Association of Physical Anthropologists (2013)


Understanding the role of diet in shaping the lemuriform mandible: Comparing traditional and geometric morphometric approaches

KAREN L. BAAB1 and JONATHAN M.G. PERRY2.

1Department of Anthropology & Interdepartmental Doctoral Program in Anthropological Sciences, Stony Brook University, 2Center for Functional Anatomy and Evolution, The Johns Hopkins University

Thursday Morning, 301D Add to calendar

As chewing biomechanics are influenced by the underlying geometry of the skull, it stands to reason that functional morphological analysis may benefit from explicit consideration of this geometry, which can be quantified by landmark data. Here, we present biomechanically-informed analyses of two sets of variables measured from lemuriform mandibles: a set of standard linear functional variables and a set of landmarks representing the endpoints of those linear dimensions. We performed principal components analyses (PCA) of both sets of variables, followed by regression of the PC scores on dietary variables. We also compared the linear dimensions and landmark configurations of closely related (and similarly-sized) species that differ in their diets.

The first PC in both cases differentiated more frugivorous lemurs from folivores / seed-predators, and highlighted similar morphological differences between these groups. The correlation between Euclidean distances based on PCs 1-3 was also moderately high. Yet, despite the broad congruence of the two analyses, the ordinations were not identical, and both the functional linear and landmark PCAs were also comparable to PCs computed from sets of randomly generated linear measurements based on these same landmarks. Linear regressions of the PC scores on dietary variables confirmed that both analyses captured a dietary signal, but again highlighted differences in the two analyses. One advantage of incorporating landmark analyses was the ability to visualize the morphological transformations underlying the differences in linear dimensions. Future functional analyses of two dimensional morphological data would benefit from consideration of patterns in corresponding 3D data.

Tweet
comments powered by Disqus