Anomodont therapsids were the most diverse and abundant group of terrestrial vertebrate herbivores during the Late Permian and much of the Triassic. Their success has been widely attributed to a complex mastication system that was based on a propalinal sliding of the lower jaw. Traditionally, anomodont phylogeny has been viewed as an essentially linear sequence of taxa, each with more specializations for propaliny, culminating in the radiation of the dicynodont anomodonts. However, recent phylogenetic work has shown that similar specializations for propaliny can be found in members of two distinct anomodont clades, the endemic Russian Venyukovioidea and the unnamed clade consisting of Galeops + Dicynodontia. These specializations have been inferred to be present in the common ancestor of the venyukovioids and dicynodonts, implying a single origin of propaliny in anomodonts. Here a reevaluation of the anomodont feeding system is presented, and the possibility that propaliny evolved twice within the group is specifically addressed. First, I examine current hypotheses regarding anomodont phylogeny, using a modified version of data sets presented by previous authors. Then I consider the evolution of four characters related to propaliny in the context of the resulting phylogeny. The results of this investigation suggest that a dual origin of propaliny within anomodonts is not unlikely, but it depends largely on the interpretation of equivocal or poorly preserved features in some critical taxa, especially Otsheria. Given that the power stroke of mastication in basal synapsids involved the application of a posterodorsally directed force on the lower jaw, the evolution of propaliny in basal synapsids may not have been difficult. Furthermore, the independent evolution of propaliny in several other nonmammalian synapsid clades (e.g., Edaphosauria, Gorgonopsia) suggests that homoplasy may be the rule rather than the exception for this feature. Finally, the possibility of a dual origin of propaliny in anomodonts has important implications for the more general problem of the evolution of herbivory in terrestrial vertebrates. In particular, it suggests that propaliny alone was not enough to guarantee success as a herbivore during the late Paleozoic and early Mesozoic.