Abstract

Redescription of the small Triassic dicynodont Kombuisia frerensis Hotton reveals new information about its cranial anatomy. On the basis of the new data, the previously suggested hypothesis of a close relationship of Kombuisia and the Permian genus Kingoria is tested within a phylogenetic framework. For this a total evidence analysis of Permian anomodont relationships was performed by combining existing data matrices into a comprehensive data set that includes basal anomodonts, dicynodonts and a large number of morphological characters. The resulting phylogenetic hypothesis corroborates the sister-taxon relationship of Kombuisia and Kingoria. This is based on a number of synapomorphies, including the narrow intertemporal region that forms a sagittal crest, a reduced mandibular fenestra, the presence of a dorsolateral notch in occipital view of the squamosal, a relatively wide mid-ventral plate of the vomer and a dorsal stapedial process. The general topology of this phylogeny supports the main aspects of recent hypotheses of anomodont relationships, and not only resolves critical nodes at the base of the Dicynodontia that were previously obscured by polytomies, but also introduces new hypotheses of relationships. Furthermore, the phylogenetic position of Kombuisia has implications for the survivorship of the Dicynodontia across the Permian–Triassic boundary. With consideration of ghost lineages there are at least four dicynodont lineages that extend beyond the end-Permian extinction event.