Assembling the ruminant tree: combining morphology, molecules, extant taxa, and fossils

A gap exists between paleontological and neontological approaches to ruminant phylogenetics, despite great increases in phylogenetic resolution through molecular work of the last three decades, and a large and growing fossil record. This gap is reflected in differing methodological approaches, with insufficient integration of the large fossil record by molecular studies on the one hand, and insufficient consideration of highly resolved genomic work by paleontological studies on the other. Both paleontological and molecular approaches seek to answer similar broad evolutionary questions, and a synthetic approach is in the interest of all. I demonstrate this by reviewing the development of each field, noting many examples in which paleontological or molecular approaches to ruminant phylogenetics are, on their own, inadequate compared to an approach which considers all sources of data together. In particular, cases such as those of Bison, Capra, and Pelea have shown that integration of genomic and anatomical data presents better resolution of relationships, and I suggest Antilocapra and Moschus may benefit from a similar approach, especially with the integration of fossil taxa into a combined (supermatrix) analysis. I present preliminary results of a new and large (in progress) morphological matrix that is intended to be used for the incorporation of anatomical data and fossil taxa into a combined analysis. The new matrix is much larger than previous morphological matrices assembled for ruminant phylogenetics, meaning it can support a larger number of fossil taxa than was previously possible. Preliminary analysis with 18 taxa recovers a highly supported tree that is mostly compatible with both traditional and molecular phylogenies, although problems of convergence remain, such as between Antilocapra and Bovidae. Finally, I propose standardization of ruminant clade names in order to limit miscommunication between paleontological and neontological workers. I propose phylogenetic definitions based on crown (extant) clades for the names Ruminantia and Pecora, and the use of Pan-Ruminantia and Pan-Pecora to accommodate each respective crown clade plus its stem group.