Coupling genetics and ecological niche modeling to examine evolutionary relationships among subspecies of Ord's kangaroo rat, Dipodomys ordii

Scope and Method of Study: The purpose of this study was to determine the evolutionary relationships among subspecies of Dipodomys ordii which has a broad distribution, ranging from southern Alberta and Saskatchewan, Canada to southern Hidalgo, Mexico and from central Oregon and eastern California to central Kansas and Oklahoma. Currently there are 32 recognized subspecies that have been established based on morphologic and ecologic data. In 1949, Setzer recognized 35 subspecies and arranged these subspecies into six subspecies complexes based on geographical location or morphological characteristics. However, since Setzer's (1949) publication no independent study has been conducted to test this evolutionary hypothesis. By combining ecological niche modeling with genetics, we have found significance of the subspecies complexes partitioned by Setzer. We obtained individuals from 22 and 29 subspecies of the 32 currently recognized subspecies for the genetic and ecologic datasets, respectively.;Findings and Conclusions: The digenomic phylogenetic analyses based on DNA sequence variation at the mitochondrial Cytochrome Oxidase III (COIII) gene and nuclear AFLPs reveal clades that match with three subspecific complexes and percent sequence divergences correlating with those clades. Our study supports the general conclusion by Setzer (1949); however, these data support far fewer than the 32 currently recognized subspecies due to lack of resolution at the subspecies level. The ecological niche modeling showed separation between all of the complexes with little to no overlap, as well as zones of unsuitable habitat. Niche divergence among all subspecies complexes is statistically significant. For D. ordii, the extent of niche divergence among subspecies complexes varies and is a result of geographic isolation and niche adaptation. Ecological divergence appears to be driving genetic divergence. Subspecies within each unique complex are becoming locally adapted to that niche either by geographic barriers or by environmental factors, thereby creating barriers to gene flow resulting in molecular divergence. Although we do not propose any formal taxonomic changes we would suggest that pending further study that incorporates larger samples sizes for each of the currently recognized subspecies and inclusion of subspecies missing from this study, the most parsimonious interpretation of the genetic and modeling data would be to minimally recognize three subspecies. These three subspecies would encompass those subspecies currently placed in Setzer's (1949) Great Plains, Mexican, and Western Desert complexes.