108 



ALASKA FISHERIES AND FUR INDUSTRIES IN 1917. 



THE AVERAGE HAREM. 



Too mucli stress can not be placed upon the value of ascertaining 

 the average number of cows per bull on the several rookeries and for 

 the herd as a whole. This has long been recognized as one of the best 

 methods of census calculation. The first pup counts ever made were 

 on small rookeries to determine the average harem there, and this 

 factor was applied to the total number of bulls. Thus the total 

 number of cows was obtained and from it the other classes could be 

 deduced. Naturally the value of the process depended upon the 

 applicability to the herd as a whole of tne conditions on the one or 

 two rookeries on which pups were counted. It has been found 

 through five years of complete pup counts that a few rookeries con- 

 sistently have average harems approaching that of the whole. These 

 rookeries should, of course, be taken as a type when only partial 

 counts can be made. 



This was done in 1917. Those rookeries were taken for counting 

 which seemed from observations on the ground and from former con- 

 ditions to have approximately the averages of the herd. Tlie full 

 details of the reasoning followed in then arriving at the average 

 harems on those rookeries which were not counted are shown else- 

 where. 



It is practically certain that the average number of cows to each 

 bull in 1917 was a minimum of 26.39. And it is just as true now as 

 ever that the number of idle bulls makes this large or small. When 

 the idle bulls are few the average harem is large; when they are many 

 it is small. 



In 1917 the percentage of idle bulls to harem bulls was 55 and the 

 average harem 26.39. It is a safe inference, therefore, that whenever 

 the percentage of idle bulls equals or exceeds this figure the average 

 harem will not be far from the minimum. The importance of this 

 fact will be appreciated in future census work when the herd has 

 developed beyond the possibilities of pup counting. The following 

 table of average harems for six years is very instructive. While the 

 curves of each rookery have a general similarity to that of the total, 

 very few are exact enough to furnish an indication of what the 

 average harem will be in future years. 



The Average Harem in the Years 1912-1917, Inclusive. 



> Derived from estimates 



