estimating the number of seals from a year 

 class alive at age 1. 



Estimates of the Pup Population From the 

 Recovery of Tagged Males . --During the period 

 17-19 July, the tag-lost recoveries on St. Paul 

 Island were recorded incorrectly; therefore, 

 the number of tag-lost for this period was 

 estimated by the tag-lost to tagged ratio ob- 

 served during the remainder of the male kill. 

 Also, the tag-lost to tagged ratio for 2-year-old 

 males was much higher on St. Paul Island than 

 on St. George Island, 0.65 and 0.30, respec- 

 tively (table 8). 



Recent estimates of the pup population are 

 lower than those of the mid- and late-1950's 

 (tables 17, 18, and 19). 



Estimates of the Pup Population From the 

 Recovery of Tagged Females . --Two features 

 should be noted in the tag recoveries from 

 females (table 8). First, the tag-lost to tagged 

 ratio is unrealistically high for 2-year-old 

 females on St. Paul Island--this is probably 

 the result of recording tag loss from 6-year- 

 olds as that from 2-year-olds. Secondly, the 

 tag-lost to tagged ratio for all ages is higher 

 on St. Paul Island than on St. George Island. 



Estimates of the I960 and 1961 year classes 

 based on the recovery of tagged females and 

 of tagged males were similar (tables 17 and 

 20). The high but unreal rate of tag loss re- 

 corded on St. Paul Island for 2 -year- olds 

 resulted in the low estimate for that year 

 class and a high estimate for the 1958 year 

 class. Estimates based on the recoveries from 

 tagged females have generally been more 



erratic than those based on recoveries from 

 tagged males. 



Estimated Number of Yearling Males, 1961 

 Year Class.-- Length was first used as abasis 

 for selecting yearling animals for tagging in 

 1962 (Roppel, Johnson, Bauer, Chapman, and 

 Wilke, 1963). Of those tagged, most were 

 yearlings, some were 2 years old and, for 

 females at least, some were 3 years old. The 

 ages of animals tagged as yearlings are deter- 

 mined from tooth ridge counts when the seals 

 are taken in the kill; thus, the proportion 

 older than 1 at the time of tagging can be 

 estimated. For example, 2 of 46 males tagged 

 as yearlings in 1962 and recovered from the 

 kill in 1963 were older than 1 at the time of 

 tagging. In 1964, 222 males tagged as yearlings 

 were recovered from the kill; the ages of 7 

 had been incorrectly determined at the time 

 of tagging. Thus, 3.4 percent of the males 

 recovered were older than 1 at the time of 

 tagging. On this basis, 600 or 96.6 percent of 

 the 621 males selected and tagged as yearlings 

 in 1962 on the basis of body length were 

 actually yearlings. The estimate of the error 

 (3.4 percent) in determining ages, however, is 

 probably too high as seals older than 1 at the 

 time of tagging have subsequently been avail- 

 able in the kill as 3- and 4-year-olds, while 

 those that were age 1 have to date been avail- 

 able only as 2- and 3-year-olds. The probabil- 

 ity of tagged seals being taken in the kill is 

 greater at ages 3 and 4 than at ages 2 and 3. 

 The error of 3.4 percent, if too high, would 

 inflate the estimates of the number of year- 

 lings. 



Table 17. --Estimates of the pup population based on tag recoveries 

 from males, Pribilof Islands, Alaska, year classes 1959-62 



n = number of males killed from each year class, 1 July through 

 7 August 1964. 



t = number of tags applied to each year class. 



s = number of tagged and tag-lost males recovered from each year 

 class, 1 July through 7 August 1964. 



N = estimated number of pups in each year class at time of tagging; 

 (n+l)(t+l) 



(s + 1) ' 



16 



