Phylogeny of Multituberculata

Overview

Despite the importance of multituberculates as major components of many Mesozoic and early Tertiary faunas, the evolutionary relationships among these mammals have remained poorly understood. To address this issue, a cladistic analysis of relationships among forty-nine multituberculate taxa was conducted using dental and cranial characters. This analysis resulted in a strict consensus tree with a resolution of 66%. The relatively low level of resolution in this tree seems largely to be a result of including incompletely known taxa in the analysis. Sequential addition analyses (in which taxa were added to the analysis in order of decreasing completeness) indicated that even the most incomplete taxa preserve information important for understanding multituberculate relationships. Sequentially deleting such taxa provided a method for testing the stability of observed clades.

Based on the hypothesis of relationships presented in this study, Plagiaulacoidea is recognized as a paraphyletic group composed of several lineages of primitive multituberculates. Ptilodontoidea appears to form a monophyletic group if Boffius and Liotomus are removed to Taeniolabidoidea. Although support for the group is somewhat ambiguous, Taeniolabidoidea may also be monophyletic, provided that Eobaatar and Monobaatar are removed. Members of Cimolomyidae all appear to belong to various lineages within Taeniolabidoidea. Ptilodontoidea and Taeniolabidoidea together constitute a more inclusive monophyletic group, Cimolodonta McKenna, 1975.

The hypotheses of phylogeny presented in this study provide a framework within which evolution of multituberculate morphology can be studied. Results of the current analyses indicate that while cusp numbers and tooth size often vary in direct relationship to one another, these aspects of tooth morphology were not tightly coupled in any lineage during multituberculate evolution. As suggested by Krause and Carlson (1987), “gigantoprismatic” enamel appears to have evolved only once in multituberculates, within derived “plagiaulacoids” ancestral to Cimolodonta. In contrast, small prismatic enamel apparently evolved three times—once in late Cretaceous ptilodontoids and twice in late Paleocene taeniolabidoids. Fully restricted enamel on the lower incisors, long thought to be limited to taeniolabidoids, seems to have evolved once in “plagiaulacoids” and several times in taeniolabidoids. Partial restriction of the enamel appears to be diagnostic for Taeniolabidoidea.