222 THE ASIATIC FUR-SEAL ISLANDS. 



because it is sometimes very diiBcult on the narrow beaches to drive past the breeding 

 grounds without a few females becoming mixed in. On Bering Island, however, it is 

 necessary to drive off the entire rookery iu order to pick out the bachelors from among 

 the females and pups. This is no novel feature due to the decrease of the seals. The 

 same thing took place in 1882 and 1883, during the palmiest days of the seal business, 

 when the condition of the rookeries was at its best, and while the seals were still 

 increasing. Even at that time numbers of females and pups were mixed up in the drives. 

 It is quite likely that the proportionate number of females was much less in those days, 

 when several times as many bachelors were killed as nowadays, but the driving of 

 females on Bering Island was then as much of a regular feature as it is now. The 

 reason for this necessity lies in the peculiar shape of the Eeef rookery. Situated as it 

 is on a long spit with only a narrow base, the breeding seals occupying the water line 

 and the bachelors hauled out behind them, it is impossible to cut the latter off from 

 the water without also cutting off the breeding females. Formerly the rough sorting 

 of the sexes was accomplished easily at the first round-up, and as the drive, divided 

 into small companies, progressed toward the killing grounds, there being no particular 

 anxiety about letting a few bachelors escape with the females. Nowadays, however, 

 when the latter have become scarce it is essential that not a single one be lost, and 

 consequently the females have to go to the killing ground, where a more careful culling 

 can be done in the pods taken out for slaughter. 



NUMBER OF SEALS ON THE ISLANDS. 



On the Pribilof Islands various attempts have been made during the last twenty- 

 five years to estimate, calculate, or count the number of seals in the rookeries. The 

 vast multitudes hauling out in the seventies and early eighties were evidently so 

 overwhelming that no one seriously suggested a census based upon an actual count. 

 Instead, various methods of estimating and calculating were indulged in, which to my 

 mind were nothing more than guesswork of the worst kind in disguise. My reasons 

 for thinking so I have given elsewhere (Russ. Fur-Seal Isds., p. 105), though with the 

 cautious proviso that while I had found such methods utterly inapplicable on the 

 Commander Islands I would express no opinion as to their adequacy on the Pribilofs. 

 My experience on the latter in 1896 has convinced me, however, that the objections I 

 had raised were equally pertinent on the American as on the Russian side. In 1895, 

 however, the number of seals on St. Paul and St. George had become so decimated 

 and scattered that it appeared to Mr. True and Mr. Townsend to be feasible to count 

 them, at least on some of the rookeries, and during 1896 a census on a similar basis, 

 but more extensive and with the aid of a number of assistants, was successfully 

 undertaken by Dr. Jordan and his associates, as reported by him. I am satisfied that 

 the total arrived at is a very close approximation to the actual number of seals hauled 

 out on the Pribilof breeding grounds. By partaking in this work myself I gained 

 enough experience to form a well-founded opinion concerning the feasibility of a 

 similar census of the Commander Islands, and I must adhere to my former declaration 

 that it can not be done, the nature of the rookeries is so different. 



Take first the north rookery on Bering Island. There is no point on shore from 

 which even an approximate count could be made. From the very fact that the bulls 

 located away from the Reef can not be roused, and because of the level nature of the 

 ground, it is not even practicable to get an approximate count of the bulls. Nor could 

 a better result be obtained from a boat, as the surrounding water is too shallow and 



