454 



SCIENCE. 



[N. S. Vol. V. No. 116. 



"We have for the first time a fairly accu- 

 rate census of breeding seals, based not on 

 estimates, but on an actual count of females 

 and males present on the breeding grounds, 

 checked by comparison with the number of 

 young seals present. This last is an im- 

 portant factor, since a count of the pups 

 shows that only about one-half the females 

 are on the rookery grounds at any one 

 time. While an exact census is, from the 

 nature of things, impossible, it appears that 

 there were in October, 1896, a little less 

 than 360,0(»0 seals left on the Pribilof Is- 

 lands, 123,000 being breeding females, 5,009 

 males with harems, and 2,966 idle bulls, or 

 those which were unable to secure any fe- 

 males through the scarcity of those due to 

 pelagic sealing. 



The breeding habits of the seals were 

 carefully studied and the facts definitely 

 settled that the young ' cows ' make their 

 appearance on the rookeries for the first 

 time at the age of two years. This occurs 

 late in July, after the vast majority of 

 young have been born and the breeding 

 grounds or rockeries have lost the compact 

 appearance they present earlier in the sea- 

 son, and the arrangement of the females in 

 regular fixed harems has been broken up; 

 the young females are not in the harems 

 with the older cows. Mr. Townsend's in- 

 vestigations of the condition of female seals 

 taken at sea were continued and extended 

 and fully corroborate his conclusions that, 

 with rare exceptions, all over two years 

 old taken during August and September 

 are not only pregnant but nursing. The 

 statements that the females taken at sea 

 are largely barren or non-breeding are un- 

 supported by any shadow of evidence, as 

 are the assertions that the fur-seal breeds 

 biennially, all observations proving that 

 seals bear annually and throughout life. 



Some additions have been made to our 

 knowledge of the food of the fur-seal, al- 

 though the data for this have mostly been 



obtained by Mr. Townsend and Mr. Alex- 

 ander in previous years. In Bering Sea 

 the Alaskan Pollack (Pollachiiis clialcogram- 

 mus) forms by far the most important item 

 of the seal's diet, a squid ( Gonatus amcenus) 

 coming next. Salmon are eaten when 

 found, and large numbers of a small, unde- 

 scribed fish, allied to the Surf Smelt {Hypo- 

 mesus olidus). Cod are not touched, at least 

 in only one case out of three hundred, nor 

 any species of bottom-frequenting fish, the 

 seals feeding on such as are found near the 

 surface, while the feeding grounds are in 

 deep water. In regard to the young it is 

 clearly shown that up to the time of leav- 

 ing the islands they subsist entirely on their 

 mother's milk, deriving no nourishment 

 from fish or crustaceans, much less if the 

 hibernicism be allowed, from kelp or sea 

 weed. Thus every seal pup whose mother 

 is killed at sea inevitably perishes from 

 starvation, even the largest gray pups being 

 unable to care for themselves late in Octo- 

 ber. 



An important feature of the work of the 

 commission was the ascertaining of the fact 

 that a large number of seals die when very 

 young and the reason for this loss of life. It 

 has been generally considered that all seal 

 pups which have been found dead of late 

 years have starved to death on account of 

 the loss of their mothers from pelagic seal- 

 ing. On the other hand, the English com- 

 missioners in 1892 took the ground that 

 there was a considerable loss from other 

 causes, particularly drowning and epi- 

 demics. As a result of the investigations 

 of 1896, it is now known that many young 

 seals die before they are a month old and 

 prior to the date on which pelagic sealing 

 begins, through being trampled on by the 

 old seals and mainly by the bulls. This 

 loss does not take place all over the breed- 

 ing grounds, but is almost wholly confined 

 to those sections which are level or free 

 from boulders. In such places there is no 



