182 



Less is known about the phocines, as their poor fossil record combined with a recent post- 

 early Pliocene and/or Pleistocene adaptive radiation (Ray 1976a) largely hinder any 

 detailed description of the dispersal pattern within the subfamily. One theory advocates 

 two separate dispersal movements (McLaren 1975; Ray 1976a; Repenning et al. 1979). 

 The first movement involves an initial northward migration from the Pusa-like ancestors 

 of the Paratethys Sea into the Arctic basin, followed by an eastward migration to give rise 

 to modern Pusa spp. The land-locked Pusa caspica and Pusa sibirica are thus apparently 

 remnant populations of this initial Paratethyan stock (although the case is not as clear for 

 P. sibirica), rather than migrating in from the Arctic basin during the Pleistocene (see 

 below; also McLaren 1960a; Grigorescu 1976). The second movement involves a 

 westward migration of the remaining phocines from the North Atlantic to their current 

 ranges. Both migrations used the Arctic basin to traverse the North American continent. 

 A second theory (Davies 1958b), although not mentioned per se, apparently holds for a 

 North Atlantic origin for all phocines, with subsequent migrations into the Arctic basin 

 during interglacial periods. Repeated glacial events then divided the different species into 

 their Atlantic and Pacific subspecies, or pushed them away from their original North 

 Atlantic range (as was felt to be the case for Histriophoca, Pusa caspica, and Pusa 

 sibirica). Pagophilus and Histriophoca, in particular, were posited to be sister species 

 separated by the most recent glaciation event (Davies 1958b). 



Altogether, the phylogeny of the phocids presented in this study raises several conflicts 

 with the biogeography of the family as outlined above. These conflicts are present 

 throughout the cladogram. Among outgroup relationships, the potential discrepancy 

 between the increasingly accepted monophyly (likely in a strict cladistic sense) of the 

 pinnipeds, and the separate origins of the otarioids and phocids indicated by the fossil 

 record might be due to the inadequacies of the latter. The most parsimonious solution is 

 for a common origin for all pinnipeds (presumably in the earlier North Pacific site), with 

 our first record of a phocid not being until seven million years later, by which time its 

 ancestors had migrated to the North Atlantic, either northward through the Arctic basin or 

 southward through the Central American Seaway. Surprisingly, to our knowledge, only 

 Costa (1993) has recently suggested such a scenario (and specifically via the southerly 

 route), despite the continual recent allying of the Atlantic phocids with the Pacific 

 desmatophocids (an extinct group of pinnipeds comprised of the genera Allodesmus, 

 Desmatophoca, and Pinnarctidion) within a monophyletic Pinnipedia (Wyss 1987; Berta 

 1991; Wyss & Flynn 1993; Berta & Wyss 1994). Indeed, the possible biogeographical 

 ramifications of a monophyletic Pinnipedia have been virtually ignored (e.g., see Wyss 

 1987:25), possibly in light of the strong counter-arguments provided by Ray (1976a: 396- 

 397). 



Of the two possible routes to the Atlantic, the southerly route is the more probable. A 

 migration through the Arctic basin does not even appear to be feasible as the Bering land 

 bridge generally blocked access to it from about the late Oligocene to the early Miocene 

 (Hopkins 1967). Some northward migration might have occurred given that modern 

 pinnipeds are capable of migrating surprising distances over land (see Scheffer 1967), 

 something likely even more readily accomplished by their less aquatically adapted 



