It has been proposed that the mammalian facial skeleton is optimized for countering or dissipating masticatory stress. As optimized load-bearing structures by definition exhibit maximum strength with a minimum amount of material, this hypothesis predicts that during chewing and biting there should be relatively high and near uniform amounts of strain throughout the facial skeleton. If levels of strain in certain areas of the facial skeleton are relatively low during these behaviors, this indicates that the amount of bone mass in these areas could be significantly reduced without resulting in the danger of structural failure due to repeated masticatory loads. Furthermore, and by definition, this indicates that these areas are not optimized for countering masticatory stress, and instead their overall morphology and concentration of bone mass has most likely been selected or influenced mainly by factors unrelated to the dissipation or countering of chewing and biting forces. An analysis of in vivo bone strain along the lateral aspect of the zygomatic arch of macaques indicates the clear absence of a high and near uniform strain environment throughout its extent. Instead, there is a steep strain gradient along the zygomatic arch, with the highest strains along its anterior portion, intermediate strains along its middle portion, and the lowest strains along its posterior portion. These data, in combination with earlier published data (Hylander et al., 1991), indicate that levels of functional strains during chewing and biting are highly variable from one region of the face to the next, and therefore it is unlikely that all facial bones are especially designed so as to minimize bone tissue and maximize strength for countering masticatory loads. Thus, the functional significance of the morphology of certain facial bones need not necessarily bear any important or special relationship to routine and habitual cyclical mechanical loads associated with chewing or biting. Furthermore, the presence of these steep strain gradients within the facial skeleton suggests that the amount of bone mass in the low-strain areas may be largely determined by factors unrelated to processes frequently referred to as "functional adaptation," or conversely, that the "optimal strain environment" of bone varies enormously throughout the facial skeleton (cf., Rubin et al., 1994). Based solely on anatomical considerations, it is likely that the zygomatic arch is bent in both the parasagittal and transverse planes and twisted about its long axis. Due to constraints on rosette position, the strain data are incapable of determining if one or more of these loading conditions predominate. Instead, the strain data simply provide limited support for the possible presence of all of these loading regimes. Finally, as the masseter muscle is concentrated along the anterior portion of the zygomatic arch and as the arch has fixed ends, the largest shearing forces and the largest bending and twisting moments are located along its anterior portion. This in turn explains why the largest strains are found along the anterior portion of the zygomatic arch.