Chimpanzees, bonobos (Pan), and gorillas (Gorilla) are the only extant knuckle-walking primates. Two hypotheses have been proposed to explain the evolution of knuckle-walking in Pan and Gorilla. The knuckle-walking hypothesis holds that the last common ancestor of Pan, Gorilla, and Homo knuckle-walked. The alternative hypothesis holds that Pan and Gorilla evolved knuckle-walking independently. One unresolved issue is that, despite sharing knuckle-walking as a common locomotor mode, Pan and Gorilla exhibit different carpal morphologies. This morphological variation has been used to support the theory that Pan and Gorilla evolved knuckle-walking independently. To date, however, little work has focused on the effect of body mass on this functional complex. Here, we compare shape changes in the carpals of Pan and Gorilla to those of other quadrupedal mammals with similar body size differences. If morphological trends in the carpus of Pan and Gorilla mirror those of other mammals with similar body mass variation, then carpal variation in African apes may be attributed to differences in body mass, rather than the independent evolution of knuckle-walking. More specifically, we hypothesize that the carpals of high body mass taxa will be relatively proximo-distally shorter and medio-laterally/antero-posteriorly wider than those of low body mass taxa. The capitate, hamate, and triquetrum of high and low body mass taxa from the mammalian families Hominidae, Cercopithecidae, Bovidae, Canidae, and Myrmecophagidae were surface scanned. A weighted spherical harmonic (SPHARM) analysis was used to quantify the shape of each carpal. During the SPHARM analysis, surface models were mapped onto the surface of a sphere, and then decomposed into a weighted sum of spherical harmonic functions. Shape variation of each carpal (within each mammalian family) was summarized using a principal component analysis (PCA) of coefficients from spherical harmonic functions. The log-centroid size of each surface model was added to each PCA to determine the effect of size on shape variation. Shape changes were visualized by comparing surface models of the average specimen from species with the highest and lowest body mass within each family. Species separate by body mass in all PC plots, and comparison of average specimen surface models indicates that across all mammalian families included in the analysis, the capitate, hamate, and triquetrum exhibit similar shape changes between high and low body mass taxa (i.e., the carpals of high body mass taxa are proximo-distally shorter and medio-laterally wider than those of low body mass taxa). These distinctions are likely caused by the need to accommodate the higher forelimb loading associated with an increase in body mass. Furthermore, because these patterns of shape change are seen in the carpals of multiple mammalian families, morphological variation in the carpals of chimpanzees and gorillas is more likely due to differences in body size than the independent evolution of knuckle-walking.