FUR SEALS AND OTHER LIFE, PRIBILOF ISLANDS, I914. 



Comparison of harems and idle bulls, IQ12-IQ14. 



41 



" From the unpublished records of G. A. Clark. 



' Addition for later count, the first count having been made slightly before the height of the season. 



Comparison of nonbreeding seals. — A general increase in all classes of nonbreeding 

 seals is perhaps the most obvious comparative result of the census. The total increase 

 of all classes from 1913 to 1914 is 25,959, and of this only 1,204 are breeding seals, leav- 

 ing 24,755 as the increment of nonbreeding animals. The classes showing the greatest 

 gains are the 4 and 5 year old males, which have more than quadrupled. In the 

 case of the 5-year-olds, the number is still not far from ideal, but the number of 4-year- 

 olds is unquestionably excessive and disproportionate. It is obviously the result of 

 the limited killings of 191 3 and 191 4. That it provides conditions for an overstock of 

 males in 191 5 and 191 6 is so clear that there is no room for argument. Two and 3 year 

 old males also show a steady increase since 191 2 and are now living in numbers beyond 

 all possible future need for breeding purposes. The subsequent effect of these large 

 numbers of surplus males is more fully discussed elsewhere. A tabular comparison of 

 the estimates of nonbreeding seals is found on another page (p. 37). 



Comparison of cows and pups. — A somewhat unexpected result of the census is the 

 very small increase shown in the number of breeding cows and pups. In 191 3 there 

 were 92,269 pups and in 1914 there were 93,250, an increase of only 981. It is incon- 

 ceivable that such a small increment could have been possible unless the herd was in 

 abnormal condition. Any idea that the figures are wrong may be dismissed immedi- 

 ately, for the testimony of nine men who counted in 1914, as well as that of several 

 others who counted in 191 2 and 1913, is that the method of counting is sound and the 

 results reliable. Abnormal conditions, therefore, must furnish the explanation. These 



