Comments: Phoca largha has been treated as conspecific with P. vitulina by some authors; recent accounts generally have regarded vitulina and largha as separate species (e.g., Reeves et al. 1992; Wozencraft, in Wilson and Reeder 1993, 2005; Rice 1998, Baker et al. 2003). This is supported by limited mtDNA data (Mouchaty et al. 1995).

Two subspecies have been recognized in the North Pacific (richardsi in east, stejnegeri in west [stejnegeri sometimes has been placed in Phoca largha]); populations on the Kuril Islands formerly were regarded as a subspecies, P. v. kurilensis. However, Reeves et al. (1992) stated that variation in North Pacific harbor seals may be regarded more realistically as a single trans-Pacific cline. Additionally, mtDNA and nuclear DNA data (Lamont et al. 1996, Burg et al. 1999) indicate the existence of at least three populations of harbor seals in the Pacific, as follows: (1) Japan, Russia, Alaska, and northern British Columbia, (2) southern British Columbia and Puget Sound, and (3) outer coasts of Washington, Oregon, and California. The data do not support the existence of the two nominal subspecies in the Pacific Ocean.

A phylogenetic analysis using mtDNA data revealed extensive macrogeographic subdivision among a subset of grouped localities that represent centers of harbor seal abundance along a distributional continuum from Japan to Alaska (Westlake and O'Corry-Crowe 2002). "Heterogeneity was influenced by population size and correlated with geographic distance, suggesting that dispersal occurs primarily among neighboring subpopulations. The two currently recognized subspecies of harbor seal in the Pacific, P. v. richardii of North America and P. v. stejnegeri of Asia, do not represent phylogenetically discrete mtDNA assemblages. The greatest differentiation detected was along the Commander-Aleutian Island chain, the region of the presumed subspecies boundary and a likely contact zone for expanding refugial populations of a number of marine mammal species after retreat of ice sheets. Differentiation between the Kodiak Archipelago and Prince William Sound, and between Bristol Bay and the Pribilof Islands, indicates that current management stocks are inappropriate and highlights the need for a detailed analysis of population and stock structure in Alaska."

North Atlantic populations are represented by subspecies concolor (west) and vitulina (east, including Iceland); nominal subspecies mellonae, from the Seal Lakes region of Quebec's Ungava Peninsula, was based on a small sample size and has not gained wide support (Reeves et al. 1992; but see Smith 1997).

Taxonomic separation of Atlantic and Pacific harbor seals is based on disjunct distributions rather than on morphological evidence (Reeves et al. 1992). In summary, current subspecies taxonomy should be regarded as of questionable significance, though regional studies of morphology and reproductive biology have suggested minimal genetic interchange among local populations (Bigg 1969, 1981, Shaughnessy and Fay 1977, Calambokidis et al. 1978, Pitcher and Calkins 1979, Kelly 1981, Burns and Gol'tsev 1984, Burns et al. 1984).

A cladistic analysis of mtDNA data yielded three clades among northern seals: Phoca-Pusa-Halichoerus, Cystophora-Pagophilus, and Erignathus (Perry et al. 1995). Each clade may be regarded as a tribe of the subfamily Phocinae. The magnitude of the differences among Phoca, Pusa, and Halichoerus was on the same order as that between species and subspecies within the genus Odocoileus. Because Cystophora is the closest relative of Pagophilus, the latter cannot be regarded as congeneric with Phoca; the differences between the two are great enough to justify placing them in separate genera (Perry et al. 1995).