160 DISCOVERY REPORTS 



40 per cent (11,506) and in each of the remaining four years at the rates shown in 

 Table XIII. The total number of immature females computed in this manner is 73,024, 

 which includes 13,479 virgin four-year-old animals. 



Under estimate A the total herd is composed as follows : 



Pups, count ... 

 Cows, by inference 

 Bulls, table ... 

 Immature males 

 Immature females 



8o,555 

 8o,555 

 50,895 

 86,471 



73.024 



Total 371,500 



Estimate B. Certain views expressed previously (1934, pp. 310, 314 and 315) require 

 modification or alteration. After somewhat lengthy calculations based on the assump- 

 tion that the annual rate of increase is 8 per cent it was found that this entailed so large 

 an accession of newly adult seal that it seemed extremely unlikely that the death rate of 

 immatures could be so low as to permit it. This death rate, for males, amounted to 

 only 38 per cent of the pups, during the whole of their immature life. 



From what is known, not only of Otaria byronia but also of O. (Enmetopias) jubata, 

 Zalophus calif or nianus and Calorhinus alascanus, it is certain that the death rates of 

 immature seal are very heavy, and it would be unreasonable to suppose that O. byronia 

 departs from the normal state of affairs in the Otariidae (see 1934, p. 309.) 



Additional calculations led to the opinion that a revised estimate of the annual increase 

 at the rate of 2 per cent was in closer accord with the probabilities than the earlier 

 figure of 8 per cent, and 2 per cent has therefore been adopted for the estimate, based on 

 the assumption that the herd is increasing. This permits a death rate of adequate size 

 and also averts the necessity of assuming that the number of breeding cows has increased 

 from something over 49,000 to just under 98,000 in the nine years 1928-37, an enlarge- 

 ment which could scarcely avoid advertising its presence. 



If, then, the herd is increasing at 2 per cent per annum and, as it obviously must, is 

 maintaining the proportions of the different classes, it is easy to calculate the numbers 

 at any time removed from the present. The method of computing the numbers is similar 

 to that which has already been used for estimate A. The loss of one-sixth of the bulls 

 and one-seventh of the cows must be replaced annually, but in this case there must be 

 further accessions large enough to provide an increase of 2 per cent, so that the accessions 

 must number 18-66 and 16-29 P er cent for males and females respectively. The deaths 

 of seals before reaching maturity under this estimate are 63-98 per cent for males, 

 and 64-04 for females. 



Since the census was not confined to one season a correction of 2 per cent per annum 

 has been applied to counts made previous to 1937, with the result that for that year the 

 number of pups appears as 83,879, and by inference the number of parous cows is the 

 same. This number has been made the basis for the other calculations necessary. 



By the use of the same method as for estimate A the number of bulls has been 

 computed, with reference to the 2 per cent annual increase, at 53,277 in January 1937; 

 male immatures number 94,488 and female 73,977- 



