rUE-SEAL CENSUS, PRIBILOF ISLANDS, 101&. Ill 



before they attain the third year. Since an exceedingly small pro- 

 portion is weak physically, it seems evident that the great loss is due 

 to a predatory animal. 



LOSS, 



The loss of cows on the islands due to rookery conditions continues 

 to be large, but a decrease from 1918 is noted. This result would 

 naturally be expected from the increase in the average harem, 

 although small, and the consequent less crowded condition. The 

 dead cows found on five St. Paul rookeries numbered 15. The total 

 alive on the same areas was 12,093. This proportion (-0.001240) 

 applied to the entire herd indicates a total loss of 195 cows. This 

 number is to be compared with 213 for 1918, 129 for 1917, and 39 for 

 1916. With an increase in the average harem to about 40 there 

 should be a material reduction in this loss. 



The practice of killing seals up to August 10 grew up when the herd 

 was smaller and much time could be taken in the segregation of the 

 animals. The natives were thus provided with fresh food 10 days 

 longer. At the present time the practice can hardly be excused on 

 those grounds. The males which would ordinarily be taken in this 

 period can be secured before, at least in large part, by a slight speeding 

 up of the work. 



AGES. 



The maximum age of cows is yet unknown. Animals of the 1902 

 series (branded 1896 to 1902 inclusive) were observed in 1919 to the 

 number of nine. This is a larger number than has been previously 

 recorded for several years. Close watch was kept for them. This is 

 considered a large number in view of the fact that the youngest are 

 at least 17 years old and have passed tlu-ough at least 10 years of 

 pelagic sealing. The number branded of which records are available 

 was 19,587. St. George brandings from 1898 to 1902, inclusive, are 

 not available. 



C. E. Crompton reported seeing a cow on Staraya Artil Rookery 

 August 2, 1919, with the one bar across the back characteristic of the 

 1902 series and, in addition, a longitudinal bar on the left side. No 

 record of such a brand was made on St. Paul Island, so it is probably 

 a St. George seal. 



Cows of the 1912 series were especially in evidence, possibly 

 because branded animals were being searched for. Four were seen in 

 one day on St. George. They have been seen on rookeries occasionally 

 where it is kno\vn they were not born. This lends weight to the 

 theory that the cows wander considerably from place to place, whereas 

 it is known that adult males do not do so to nearly the extent that was 

 formerly supposed. In less than half a dozen cases have males of 

 this series been seen away from parent rookeries since they became 6 

 years old. 



BULLS. 

 HAREM AND IDLE BULLS. 



The count of bulls, harem and idle, made at the height of the breed- 

 ing season furnishes the most reliable data obtainable at present upon 

 which to base the census. From this the average harem can be 

 35286°— 21 31 



