40 BULLETIN OF THE BUREAU OF FISHERIES. 



conditions prevailing a larger number would have been killed as 4- year-olds in 1912, the 

 first year of the suspension of commercial killing. This is scarcely probable, so it is 

 plainly evident that the increase was accomplished in spite of land killing and was 

 wholly due to the cessation of pelagic sealing, the toll of which was reduced in 191 1 

 and entirely cut off in 1912 and 1913. Considering both harem bulls and idle bulls, 

 the total stock of breeding males has grown from 1,471 in 1912 to 1,505 in 1913 and to 

 1,741 in 191 4, irrespective of land killing. 



The increased number of harems in 191 4 also shows that the number of young bulls 

 counted in 191 2 and 191 3 did not include all that were living. .The total of harem 

 bulls, idle bulls, and young bulls in the census of 1 9 1 3 was i ,767. These classes furnished 

 the harem bulls and idle bulls for 1914 to the number of 1,731, which makes it necessary 

 to assume a mortality between seasons of only 36, a number much too small, since the 

 evidence is clear that approximately one-sixth of the harem bulls die annually. The 

 difference between 1912 and 1913 is 162, which indicates a smaller discrepancy but still 

 a discrepancy. Both cases illustrate what is evident throughout the study of fur seals, 

 namely, that the numbers of nonbreeding seals are almost invariably underestimated. 



Although the increase in harem bulls is not evenly distributed, the great majority 

 of the rookeries show at least a little increase. The only ones on St. Paul Island failing 

 to do so are Zapadni Reef, Little Polovina, and Vostochni, and of these Zapadni Reef 

 remains unchanged, while the decrease on Little Polovina and Vostochni is insignificant, 

 amounting to only five harems in all. Notwithstanding the great increase on St. Paul 

 Island, there is shown a general decrease on St. George Island, the only rookery having 

 an increase being Staraya Artel, which shows four harems more than in 1913. The 

 total number of harems on St. George in 1914 is 243, as against 261 in 1913 and 281 in 

 1912. The most obvious explanation of this condition is that it is due to the gregarious 

 instinct and the tendency for the larger herd to recruit itself from the smaller one. 

 Whether or not this be a wholly sufficient explanation, it is of interest to note that the 

 same principle does not seem to be effective in all cases when applied to the different 

 rookeries on St. Paul Island. Thus some of the smaller rookeries show large percentages 

 of gain, while Vostochni, the largest of all, is one of the very few showing an actual loss. 

 Moreover, St. George, while having a decrease of bulls, has an increase of cows, so the 

 cause, if there be any definite one, is evidently peculiar. 



The total number of idle bulls in 1914 is 172, and although this is not a large 

 number it should be noted that the rate of increase over 1913 is over 60 per cent. The 

 distribution of idle bulls is somewhat irregular, and although they seem most numerous 

 about large rookeries and massed sections, this is not always the case. For example, 

 East Cliffs, a rookery of 57 harems, had only two idle bulls, while Lagoon, with 8 harems, 

 had the same number, and Zapadni Reef, with 3 harems, had i idle bull. In general, 

 the increase in idle bulls seems to be largest on rookeries having a large increase of 

 harems, as Zapadni and Tolstoi. 



