ALASKA FISHERY AND FUR-SEAL INDUSTRIES, 1929 



337 



For the present it does not appear to be desirable to increase the 

 percentaf^e used to determine the number born. The final results 

 will be the same if mortality rates are reduced. It has now been 

 definitely determined that the mortality rates applied in 1925 and 

 for several years thereafter (which rates were determined as a result 

 of information secm'cd from branding operations of 1923 and 1924) 

 were too high. 



Observations on a number of rookeries on July 12 and 13 showed 

 at least 25 per cent less harems than w^hen count was actually made. 

 This showed that cows were arriving late. Within the next few days 

 they hauled out in large numbers. 



Distribution of pups on the Pribilof Islands, August 10, 1929, and comparison with 



distribution in 1928 



MORTALITY OF SEALS AT SEA 



In 1925, due to information secured from branding operations, 

 it was necessary to increase the mortality rates of seals for the first 

 three years of their existence. It was advisable to continue using 

 these mortality rates until it could be definitely proved that they 

 were incorrect. It is now felt that it has been distinctly shown that 

 the mortality rates established in 1925 were abnormal. For the 

 present, therefore, it is desirable to use the mortality rates established 

 for males prior to 1925 and to make adjustments accordingly in the 

 number of seals credited to each age class for the first 3 years. This 

 has been done in the computation for this season. The mortality 

 rate for yearling females has been reduced to 40 per cent. 



