1899] THE HABITS OF THE NORTHERN FUR SEAL 19 



cutter " Rush "), and October 4 to 2 2 on St. George Island, thus 

 missing only the earlier part of the season of 1896. In order to 

 complete my knowledge, and to be able to observe the seals in the earlier 

 part of the breeding-season, I reached Bering's Island in 1897, on 

 June 19, and remained there until August 2, when I sailed for Copper 

 Island, and landed there on the following day. On August 19 1 left 

 Copper Island in an unsuccessful search for seal rookeries on the 

 Kamchatkan coast, and did not again return to the seal islands. Dur- 

 ing my stay on the islands I personally examined and walked over 

 the whole extent of all the rookeries, with the exception only of one 

 or two of the lesser ones on the Commander Islands, which I had 

 to be content to observe through my binoculars, either because they 

 are inaccessible from the land side, or because I had not permission to 

 approach them more nearly. 



Enumeration of Seal Islands. 



The islands whither the seals resort for breeding purposes are now 

 all well known, and it is unlikely that the most diligent search can 

 add to their number. They are the Commander and Pribilof Islands 

 in Bering's Sea, certain of the more Northern Kuril Islands, 1 Eobben 

 Island (in the Sea of Okhotsk), and possibly one or two other small 

 rocks 2 and islets in the same sea. It can hardly be doubted that the 

 presence of the seals on these islands, apparently scattered at random 

 throughout the North Pacific, and their absence from many others 

 equally suitable for their purpose, such as the Aleutians, depends 

 entirely on the former presence or absence thereon of man. The Com- 

 manders and Pribilofs are the only large uninhabited islands in the 

 North Pacific, and there are no rookeries on the Aleutian Islands, 

 which, although affording very suitable conditions, are inhabited through- 

 out their extent. 



What Guides the Seals in their Choice of an Island or Rookery. 



A glance at Sir John Murray's map {Geographical Journal, August 

 1898) to illustrate the annual range of the surface temperature of 

 the ocean will show that the question of temperature has had no very 

 great influence on the choice of the seals of islands on which to bring 

 up their young. Whereas the Pribilof and Commander Islands lie in 

 regions where the surface temperature is cold, and has an annual 

 variation of only 20° F., the corresponding figure for the Northern Kuril 

 Islands is 30°, while the little rookery at Eobben Island lies close to 

 the border line of regions where the annual variation amounts to 35° 

 and 40° F. respectively. Again, whereas in the event of a backward 



1 Shnednoi, Raikoke, and Mushir. 



2 St. Iona and the Shantai Islands. 



JUJ ( ■ A R Y 1 ij 



