Figure 6.-- Pup shearing, St. Paul 

 Island, 1963. 



Little Polovina from pup sampling was 6,500, 

 89.9 percent of the counted total. 



Annual counts of the live pups on Kitovi 

 amphitheater have been made since 1959. In 

 1963, the count was 1,106 and the average 

 estimate from pup sampling 1,200, a differ- 

 ence of less than 10 percent. 



The estimate of the pup population at the 

 time of sampling on St. Paul Island was in- 

 creased from 229,900 to 255,700 to com- 

 pensate for the difference between the count 

 and the sampling estimate on Little Polovina. 

 An estimate of 316,000 for both islands was 

 obtained by dividing 255,700 by 0.81, the 

 average percentage share of harem bulls found 

 on St. Paul Island. No corrections were made 

 for pups that died before shearing, possibly 

 20,000 or more. 



The count of live pups on Little Polovina, 

 the harem-bull count, and the dead-pup count 

 can be used to make two additional estimates 

 of the pup population. 



Assuming that the number of pups per harem 

 bull is a constant, an estimate of the total 

 of pups can be calculated from the 



number 

 ratio: 



Where: 

 n = 



hb = 



HB = 



N 



(1) 



number of pups counted on Little 



Polovina 

 number of harem bulls on Little 



Polovina 

 total number of harem bulls on all 



St. Paul Island rookeries 

 total pup population on St. Paul Island 



The estimated pup population for St. Paul 

 Island from these data is 237,000. 



Assuming that the mortality rate of pups 

 on land is constant for all rookeries for any 

 one year, an estimate of the pup population 

 can be calculated from the following equation: 



.0362 aN 



(2) 



Where: 



a 



n 



0.0362 



N 



land mortality rate for a year 

 counted pups on Little Polovina 

 the average proportion of total 

 land mortality occurring on 

 Little Polovina based on dead- 

 pup counts from 1953 through 

 1963 

 total pup population on St. Paul 

 Island 



Equation (2) becomes: n = .0362N, from which 

 N is estimated to be 200,000. 



The validity of the two assumptions neces- 

 sary to make the estimates from equations 

 (1) and (2) is not known. 



Discussion of Estimates 



A difference in the mortality rate of tagged 

 and untagged animals is thought to be the 

 only important, unmeasured factor that may 

 affect pup population estimates based on tag 

 recovery data. Differences in weight of tagged 

 and untagged pups and 3-year-old males have 



27 



