54 



BULLETIN OF THE BUREAU OF FISHERIES. 



no important change in the death rate is indicated. This is shown by the following 

 table : 



Mortality in relation to idle bulls. 



Harems 



Idle bulls 



Total bulls 



Total cows 



Dead cows 



Per cent of dead cows in relation to total num- 

 ber of bulls 



Per cent of dead cows in relation to total num- 

 ber of cows 



Dead bulls 



Per cent of dead bulls in relation to total num- 

 ber of bulls 



Total pups on Kitovi 



Dead pups on Kitovi 



Per cent of dead pups on Kitovi 



1896 



4i932 



7»932 



o 209, 873 



131 



1-7 



.079 

 38 



.380 



6,049 



c 109 

 1.8 



I.3S8 



113 



1. 471 



81,984 



27 



1-7 



•037 

 3 



. 304 



1.97S 



37 



l|403 



10s 



I.S08 



92,269 



30 



1.9 



.033 

 6 



•390 

 l,8ss 



lrS59 



173 



I. 731 



93.3SO 



18 



I-O 



.037 



2, 119 



47 

 2. 3 



o Estimated; see Rept. Fur-Seal Investieations. 1896-97, vol. 1, p. 98. 



fr One-half more than formerly estimated. 



c Including only those which died prior to August 10 to make the figures comparable with those of later years. 



The above table is somewhat unsatisfactory, since pelagic sealing was going on in 

 1896 and the counts were made then by methods not exactly the same as those used 

 in 1912-1914. Moreover, it is not possible to state what percentage of the dead was due 

 to trampling and fighting, so it is necessary to compare those dead from all causes. 

 Notwithstanding that the figures are less extensive than desirable, several points of 

 considerable interest are to be noted. The percentage of dead cows in relation to the 

 total number of cows was greater in 1896 than in the years when idle bulls were few. 

 In all other respects there is nothing in the figures to indicate that the percentage of 

 death from all causes during the breeding season was any greater in 1896 than in the 

 later years. The percentage of dead pups on Kitovi was exactly the same in 1896 as 

 in 1 91 2, and even less than in 1914. Likewise the percentage of dead cows in relation 

 to the total number of bulls was the same in 1896 as in 191 2 and less than in 191 3. 

 In 1 914, with an increase of bulls, there were fewer deaths of bulls and cows than in 

 1912 and 1913. 



The imperfection of the data precludes absolute proof, but so far as any conclusion 

 is indicated by these figures it is that a considerable number of idle bulls does not pro- 

 duce an excessive death rate. If, in spite of these figures, it is held that an increased 

 mortality must necessarily follow even a moderate increase of idle bulls, the most that 

 can be maintained is that it is likely to be relatively small and not of itself sufficient 

 to justify apprehension. Fighting and commotion must always be a part of rookery 

 life, and if the idle bulls have any legitimate function it need not be very great to 

 offset the harm they may do and justify their preservation in reasonable numbers. 



THE IDEAL PROPORTION OF IDLE BULLS. 



What has been said in the preceding paragraphs indicates that while a gteat excess 

 of idle bulls is highly undesirable, an entire lack of them is equally so and for the mainte- 

 nance of a well-balanced herd a definite proportion of them is necessary. This propor- 

 tion should be large enough to insure the distribution of idle bulls on all the rookeries in 

 numbers sufficient to serve all the virgin cows. The number of virgin cows at a given 



