Weather as a Mortality Factor 



POPULATION ESTIMATES 



The mean air temperature for St. Paul 

 Island for the 12-month period, 1 July to 30 

 June, and the total count of dead pups in the 

 following August continue to show a signifi- 

 cant inverse relationship, r = -.757, P<.01 

 (fig. 15). This relationship cannot be ex- 

 plained. 



Prior to the recent discovery that hookworm 

 larvae are transmitted from the female to 

 her pup through nursing within a short time 

 after birth, it was believed that an association 

 existed between air temperature and the sur- 

 vival of hookworm larvae in the soil. Since 

 the pups are apparently infected before larvae 

 can enter the female from the soil and migrate 

 to the milk, the influence, if any, of weather 

 on the severity and frequency of infection must 

 come in the preceding year or years. The 

 association between temperature and hook- 

 worm infection now seems less reasonable, or 

 at least, more involved. Possibly the air 

 temperature- mortality relationship is fortui- 

 tous. It will be necessary to analyze the data 

 more thoroughly and trace the apparent rela- 

 tionship for a longer period before making a 

 conclusion. 



TEMPERATURE CC I 



Figure 15.--Pup mortality-air temperature relationship, 

 St. Paul Island. 



The recovery of tags during the kill of seals 

 on land has provided data for estimating the 

 number of pups born annually beginning in 

 1947. Because estimates using this method 

 are necessarily delayed until tags can be 

 recovered 3 and 4 years after birth of the 

 pups, a program- to sample pups after tagging 

 was begun in 1961 to provide information on 

 the tagged to untagged ratio among pups in 

 the year of birth. Pup population estimates 

 from both sources of data are presented 

 in the following analysis, although the pup 

 sampling program is still in an early experi- 

 mental stage. 



Estimates from Tagged Males 



In 1960 and 1961 the observed tag lost to 

 tagged ratio was higher among animals re- 

 covered on St. Paul Island than it was among 

 those recovered on St. George Island. There- 

 fore, the St. Paul Island ratios were used to 

 correct the St. George Island ratios before 

 making the population estimates. Compara- 

 tive analysis of 1962 tag- recovery data showed 

 that a similar correction for this year was 

 unnecessary. The tag lost to tagged ratios 

 among marked male seals taken in 1960, 1961, 

 and 1962 are presented in table 31. Tables 

 31 and 32 show that the tag lost to tagged 

 ratio increases with age and that there has 

 been an increase in the ratio in each year 

 class since 1958. An increase in tag loss with 

 age is expected; however, reasons for the 

 increase are not clear. Several factors that 

 were involved are (1) efforts to identify 

 animals that have lost their tags have been 

 improved since 1959, (2) the "V" notch check- 

 mark used in 1958 is often quite difficult to 

 identify, (3) the more easily recognized slice 

 type of checkmark was used on pups tagged 

 in 1959 and 1960, and (4) numerous clinching 

 failures among tags used in 1960 probably 

 resulted in a higher rate of tag loss. Other 

 factors may also have contributed to the 

 increase. 



Table 33 shows the population estimates 

 based on tag recoveries for year classes 

 1958 through 1960. The estimates were cal- 

 culated using Petersen's formula and treating 



40 



