ALLEN: mammalia: otariid^. 123 



Islands, and the Commander Islands, have separate feeding as well as 

 breeding grounds, and do not mingle even during migrations ; and also 

 that the animals composing these separate herds are so far differentiated 

 in external and other characters as to be considered worthy of recognition 

 in nomenclature under separate names. Thus instead of the northern Fur 

 Seals being all referred as formerly to a single species, they are considered 

 as separable into three closely related species, under the names respec- 

 tively, of (i) Callotaria itrsiua (Linn.) of Bering Island, the real Sea Bear 

 {Ursus mariiiits) of Steller ; (2) Callotayia alascana (Jordan & Clark) of 

 the Pribilof Islands ; and (3) Callotaria curilensis (Jordan & Clark).' The 

 differences, so far as known, separating these forms are sufficient to render 

 even their peltries distinguishable. This being the case it is more than 

 probable that the southern Fur Seal herds that congregate at different 

 breeding resorts would also show similar differentiation, if material repre- 

 senting them could be brought together in sufficient amount for satisfac- 

 tory comparison. It does not, for example, seem probable that the Fur 

 Seals of the Galapagos Archipelago can be strictly the same as those 

 breeding on the Falkland Islands,^ or even those of Juan Fernandez and 

 Mas a Fuera Islands ; or that the herds that formerly resorted in immense 

 numbers to the Shetland, South Georgian and other Antarctic Islands 

 south of Cape Horn may not have presented minor differences from those 

 whose breeding resorts were in much lower latitudes. In the absence of 

 satisfactory material, however, for such comparison the Fur Seals of 

 southern South America and adjacent islands will be treated in the present 

 connection as belonging to two species, under the names Ayctocephalus 

 aitstralis (Zimm.), of which the type locality is the Falkland Islands, and 

 Arctocephalus philippii (Peters), of which the type locality is Juan Fer- 

 nandez Island. Both species are represented in the material at hand by 

 fifteen skulls, of which four are referable to A. aitstralis and eleven to A. 

 pJiilippii ; ten of the latter are from the Galapagos Archipelago, while 

 one purports to have come from "Straits of Magellan." 



The two South American species may be readily recognized by their 

 marked cranial differences, the skull of A. australis having the rostral 



'Jordan and Clark, Report Fur Seal Investigation, I, 1898, p. 45, and III, 1899, p. 3. 



^ Since this article was prepared, early in 1902, the Galapagos Fur Seal has been separated as 

 Arctocephalus galapagoensis by Heller (Proc. Cal. Acad. Sci., 3d Ser., Z06I., Ill, 1904, p. 245), 

 but on very unsatisfactory evidence of its distinctness. Sec further, p. 1 34. 



