44 



BULLETIN OF THE BUREAU OF FISHERIES. 



are subject to possible unknown factors involved in the operation of pelagic sealing. 

 The comparison of counts is as follows: 



Number of pups and percentages of decrease or increase in igij and in IQ14, as compared with 



the year before. 



•2 From the impublishcd records of G. A. Clark. 



The increase in 191 3 was very general, 17 rookeries showing an increase and only 

 5 a decrease. The increases on different rookeries ranged from i per cent to 34.7 per 

 cent and the total increase was 12.4 per cent. In 1914 there were increases on 12 rook- 

 eries and decreases on 10, with a total increase of only 1.06 per cent. The range of 

 variation is greater than in the previous year, running from a decrease of 28.8 per cent to 

 an increase of 38.1 per cent. Some rookeries which showed an increase in 1913 show a 

 decrease in 191 4 and vice versa. Kitovi, which decreased 6 per cent in 19x3, increased 

 as much as 14.2 per cent in 1914 and nearly the same was true of Lukanin. Polovina, 

 which showed the remarkable increase of 34.7 per cent in 1913, has 3.3 per cent decrease 

 in 1 914, while various other rookeries show similar irregularities. The most consistent 

 large increase is that of Sivutch or Sea Lion Rock, where there was a gain of 25.4 per 

 cent in 1913 and 15.7 per cent in 1914. Ardiguen also increased well in both seasons, 

 but its small size magnifies slight change as expressed in percentages. In the two 

 years since 191 2 most of the rookeries show increase, but Gorbatch, Lagoon, and Morjovi 

 on St. Paul Island, and Zapadni on St. George had fewer pups in 1914 than in 1912. 



On the whole, the comparison of the counts on the various rookeries shows nothing 

 more clearly than that great irregularity prevailed. There is nothing to indicate that 



