The pelvic girdle is a complex structure with a critical role in locomotion, but efforts to model the mechanical effects of locomotion on its shape remain difficult. Traditional approaches to understanding form and function include univariate adaptive hypothesis-testing derived from mechanical models. Geometric morphometric (GM) methods can yield novel insight into overall three-dimensional shape similarities and differences across groups, although the utility of GM in assessing functional differences has been questioned. This study evaluates the contributions of both univariate and GM approaches to unraveling the trait-function associations between pelvic form and locomotion. Three-dimensional landmarks were collected on a phylogenetically-broad sample of 180 pelves from nine primate taxa. Euclidean interlandmark distances were calculated to facilitate testing of biomechanical hypotheses, and a principal components (PC) analysis was performed on Procrustes coordinates to examine overall shape differences. Both linear dimensions and PC scores were subjected to phylogenetic ANOVA. Many of the null hypotheses relating linear dimensions to locomotor loading were not rejected. Although both analytical approaches suggest that ilium width and robusticity differ among locomotor groups, the GM analysis also suggests that ischiopubic shape differentiates groups. Although GM provides additional quantitative results beyond the univariate analyses, this study highlights the need for new GM methods to more specifically address functional shape differences among species. Until these methods are developed, it would be prudent to accompany tests of directional biomechanical hypotheses with current GM methods for a more nuanced understanding of shape and function. Anat Rec, 298:230-248, 2015. V C 2014 Wiley Periodicals, Inc.Key words: adaptation; geometric morphometrics; pelvis; primates; locomotion; functionThe identification of adaptation to biological function has been a goal of evolutionary biology since the time of Darwin. Within the field of biological anthropology, adaptations to diet and/or locomotion have been examined via hypothesis-testing of craniodental anatomy (Lieberman and Crompton, 2000;Ravosa et al., 2000;Daegling and McGraw, 2001;Spencer, 2003;Vinyard et al., 2003;Terhune, 2013), shoulder anatomy (Taylor, 1997), long bone morphology (Fleagle, 1976;Polk, 2002), and hand and foot anatomy (Orr, 2005;Patel, 2010;Tocheri et al., 2011). However, systematic hypothesistesting of pelvic adaptations has been relatively limited because the mechanical rules that govern pelvic shape are not well understood.Among other requirements, identifying morphological adaptations necessitates a clear association between a morphological trait and its presumed function, as well as evidence that the trait performs the function (e.g., Kay and Cartmill, 1977;Anthony and Kay, 1993). The comparative method is used to distinguish form-function correlations in broad samples of extant taxa (Felsenstein, 1985;Harvey and Pagel, 1991;Ross...