218 THE ASIATIC FUR-SEAL ISLANDS. 



THE SEAL ROOKERIES AND THE DRIVEWAYS. 



The location of the rookeries on the American and Eussian seal islands offers one 

 very curious contrast. On the Pribilofs the rookeries are distributed pretty nearly 

 on all sides. On the Commander Islands, however, which trend northwest and south- 

 east, the seals are not found, and apparently never were found on the northeast side 

 of the islands. The explanation is not obvious, for there are plenty of localities 

 admirably adapted for rookeries on the eastern shore of both islands. It may 

 possibly be due to the fact that the migrating seals approach the islands from 

 southern and western sides only, and that because of the greater extent of these 

 islands and their elongated shape there has never been any need for them to extend 

 their territory to the opposite side of the islands, the seal herds being approximately 

 in equilibrium and not increasing in number at the time of the discovery of the 

 rookeries by man. On the Pribilof Islands the conditions at that period were very 

 different, inasmuch as nearly all the available coast line, which is much smaller than 

 that of the Commander Islands, was occupied by the seals, which were probably more 

 than twice as numerous as on the Commanders. 



In comparing the nature of the rookeries of the Pribilof Islands with those of 

 the Commanders there was one point which struck me rather forcibly, viz, the lack 

 of sheltered places on the former where the young seals have to learn to swim. 

 There are no shallow bays behind sheltering rocks, nor places protected by submerged 

 long reefs breaking the force of the swell and surf; no place like the bay behind 

 Sivutchi Kamen on north rookery, Bering Island, or the reefs and rocks at Polujimoye 

 Karabelnoye, and Glinka. 



The north roohery of Bering Island most nearly resembles the Eeef rookery on 

 St. Paul so far as its physical characteristics are concerned, though in reality there is 

 no exact counterpart on the latter island. The chief difference lies in the fact that 

 the main portion of the rookery has a much more extended shore line in proportion to 

 its land line, and that consequently it is more difficult to cut out the seals to be 

 driven without a large number escaping into the sea. The sealion rocks of both 

 rookeries are fairly comparable, as both serve as a reserve retreat for many males, 

 who thus escape slaughter to grow up to become bulls, though from its smaller size 

 and greater accessibility the one on Bering Island answers that purpose to a less 

 extent. 



The Bering Island south rookery is again so small and isolated as to have no 

 parallel in the other islands. Its topographical features, however, are by no meaus 

 peculiar. 



The rookeries on Copper Island, again, both the Karabelni and the Glinlca 

 rookeries, are as different from either the Pribilof Island rookeries and those on 

 Bering Island as they can be. Knowing the rookeries of St. George Island, as I did, 

 only from description, I was led to believe that they would compare with those of 

 Copper Island; but my visit to the former island this summer (1896) has shown me 

 my mistake. True, some of the St. George rookeries are situated at the foot of steep 

 bluffs, but in this they do not differ materially from some of the St. Paul rookeries, 

 while on the other hand the cliffs on Copper Island, which rise above the rookery 

 beach, are many times higher, and the beach itself as a rule narrower, more rugged, 

 and indented by reefs and coves. Moreover, the island, back from the bluffs, bears 



