Two subspecies of bison (Bison bison) have been described for North America. Plains bison (Bison bison bison) and wood bison (Bison bison athabascae). These subspecies designations were based on morphology (skull and horn measurements, body proportions and size, as well as hair patterns), but there was no consensus on if those subspecies are actually valid. Especially genetic studies have provide no support for plains bison and wood bison being subspecies.
To be frank, I am very critical of the entire subspecies concept, and so far I was not able to find anything that justifies yet another category between species and populations. It is often defined as a taxonomic division of a species which usually arises as a consequence of geographical isolation within that species. I don't see the difference to population in this case and no objective criterion has been brought forward so far.
Good to know that I am not alone with my rejection of the concept. Here is what colleague Matthew Cronin from the University of Alaska, Fairbanks has to say:
The scientific community has debated the whole subspecies category and many if not most population geneticists like myself recognize the entire category has been used very subjectively, in other words, people have designated subspecies without a lot of rigorous scientific evidence. And that’s been going on since the late 1800s, so there’s a lot of subspecies for which there hasn’t been adequate analysis.
Cronin and colleagues have done a comparative assessment of bison and cattle subspecies by analyzing the microsatellite variation within and between the putative bison subspecies, and compared those levels of genetic divergence with those of cattle breeds and subspecies. No matter if one accepts or rejects the subspecies ranking per se bison subspecies are currently recognized by management agencies, e.g. Wood Bison are currently listed as threatened under the US Endangered Species Act.
The results of the study discount the existence of bison subspecies. The data show no phylogenetic distinction of extant wood bison and plains bison. They actually show less divergence of microsatellite allele frequencies than cattle breeds and subspecies. Furthermore, in Wood Buffalo National Park both were mixed and the researchers couldn't find any wood bison without some degree of plains bison ancestry. Consequently extant wood bison herds should be considered a northwestern subpopulation of the North American bison.
These results have ramifications for policies concerning the management and conservation of bison populations. Both Canada and the United States maintain gene banks to conserve germplasm and tissue for agriculturally useful species including bison, and both countries have initiated collections of bison tissues. The results of this study will be used in further development of germplasm acquisition for bison and suggest that plains bison and wood bison should be considered geographic populations and not subspecies. Regarding management of wild bison populations, they could be managed to maintain phenotypic characteristics of interest, as is done with cattle breeds. That is, wood bison can be managed as a geographic subpopulation to maintain morphological characters (whether heritable or environmentally induced) or other potentially unique traits. However, the lack of subspecies-level differentiation suggests that managers of wild and domestic populations could interbreed wood bison and plains bison to maintain or enhance genetic variation and reduce the potential for genetic defects and reduced fitness resulting from inbreeding.