Based on extensive experimental work on primates, two masticatory loading regimes have emerged as the likely determinants of mandibular symphyseal fusion—dorsoventral shear and lateral transverse bending (wishboning) (Ravosa and Hylander, 1994; Hylander et al., 1998, 2000). Recently, however, it has been argued that, rather than functioning to strengthen the symphysis during mastication, fusion serves to stiffen the symphyseal joint so as to facilitate increased transverse jaw movements during occlusion (Lieberman and Crompton, 2000). As part of this transverse stiffness model, it has been suggested that taxa with fused symphyses should also exhibit more horizontally oriented occlusal wear facets. Using a series of univariate and bivariate analyses, we test predictions of these three models in a sample of 44 species of selenodont artiodactyls. Consistent with the wishboning and transverse stiffness models, taxa with fused symphyses (camelids) have more horizontally oriented M2 and M2 occlusal wear facets, anteroposteriorly (AP) elongate symphyses, and relatively wider corpora. Contrary to the dorsoventral shear model, camelids do not have relatively deeper corpora (due to greater parasagittal bending). While taxa with ossified symphyses have relatively larger symphysis cross-sectional areas, this appears to be the byproduct of an increase in AP symphysis length due to greater lateral transverse bending of the mandible. Theoretical consideration of the biomechanics of mastication further suggests that strength, not stiffness, is the critical factor in determining symphyseal ossification. Thus, like anthropoid primates, fusion in selenodont artiodactyls appears to function in resisting increased wishboning stresses arising from an emphasis on transverse occlusal/mandibular movements and loads. J. Morphol. 249:221–241, 2001. © 2001 Wiley-Liss, Inc.