FUR SEALS AND OTHER LIFE, PRIBILOF ISLANDS, I914. 49 



injur}' so far as the great majority of cases are concerned. Although the bulls some- 

 times seemed conscious of the helpless young ones and appeared to avoid trampling 

 them, innumerable instances were noted in which they passed directly over small pups 

 without harming them. In one case a copulation was observed during which a pup 

 was imprisoned beneath a bull and subjected to continued mauling for nearly 15 

 minutes, after which the little fellow wriggled out apparently none the worse. It is 

 plain, therefore, that considerable commotion is unavoidably connected with the harem 

 system whether idle bulls are present or not ; and although much of it is comparatively 

 harmless, some fatalities are bound to result under the best of circumstances. 



SIGNIFICANCE OF THE INCREASE OF HAREM BULLS. 



The increase of harem bulls from 1,358 in 191 2 to 1,403 in 191 3 and again to 1,559 

 in 1 91 4 shows clearly that the reduced state of male life in 191 2 and preceding years 

 was at most only parti)' due to the killing of males on land. As stated elsewhere, the 

 accessions' of harem bulls in 1914 consisted of animals that escaped the undiminished 

 land kilUngs of 1910 and 191 1 when they were 2 and 3 years old, respectively. ■ Their 

 continued sunaval in larger numbers than formerly is thus due to the fact that their 

 later years were free from the effect of pelagic sealing. To some extent this justifies 

 the behef frequently expressed that with the closest land killing possible a certain 

 number of males would always escape and come to maturity. It can not be said, how- 

 ever, that it justifies land killing now at the former rate, for with pelagic sealing stopped, 

 the increase in the number of cows would be proportionately larger than formerly. 



The increase in harems in 1914 without a corresponding increase in cows shows that 

 the number of harem bulls in immediately preceding years, if not insufficient, was at 

 least abnormally small. That the bulls are fully capable of meeting such emergencies 

 may not be doubted, but it is nevertheless clear that when more bulls are present the 

 average number of cows per bull is immediately reduced. The number of cows in 191 3 

 was almost equal to the number in 1914, but in 1913 they were divided among 1,403 bulls 

 and in 1914 among 1,559 bulls. It is evident, therefore, that had 1,559 bulls been pres- 

 ent in 1913, all or practically all of them would have been able to obtain harems. It 

 is apparent also that a continued increase of bulls will cause the size of the average harem 

 to decline from year to year until it reaches a minimum beyond which no increment of 

 bulls can reduce it. 



THE IDEAL PROPORTION OF HAREM BULLS. 



It may well be doubted whether it would be wise to permit such an increase of bulls 

 as would insure a minimum average harem, for other considerations are involved; but 

 a safe course of procedure in the management of the herd would seem to demand that 

 the average number of cows per bull be kept as near such a minimum as possible without 

 causing injury to the herd. Such a course may be modified as more complete knowledge- 

 is obtained in future years, but at present a prudent poUcy of approximating natural 

 conditions recommends itself. 



To accomphsh a reduction of the average harem to the proportions of former years 

 will require the preser\'ation of a relatively small number of males and large numbers 

 may still be kiUed without the sUghtest apprehension. This is evident when it is con- 

 sidered that an annual increment of only 2,000 bulls would provide for the maintenance 



