MEASUREMENTS OF 3-YEAR-OLD 

 MALES 



The mean lengths and weights of tagged and 

 untagged 3-year-old males are presented by- 

 round for the years 1962 and 1963 in table 41. 

 The measurements of tagged and untagged 

 males were compared by the approximate 

 method (Snedecor, 1956) using unweighted 

 means. The effect of tagging, of rounds, and 

 of the interaction between rounds and tagging 

 were separated by the analysis. Untagged 

 seals were significantly longer than tagged 

 seals for both years (P<.01). The average 

 difference in length was 1.67 cm. Untagged 

 seals were also heavier than tagged seals in 

 1962 and 1963 (P<.01). In 1963, however, 

 the interaction between effects of tagging 



and of rounds was significant. The inter- 

 action resulted from the aberrant data of round 

 7 when the weight of tagged seals was greater 

 than that of untagged ones. It seems to have 

 no biological importance. The F value for 

 the effect of tagging was much larger (28.71) 

 than that (2.80) for the interaction. There- 

 fore, the significant interaction is ignored 

 and the conclusion reached that there was a 

 real difference between the weights of tagged 

 and untagged seals in 1963. Present equipment 

 does not allow accurate weighing of large 

 numbers of seals on the killing fields, but there 

 is no reason to expect bias in comparing 

 weights of tagged and untagged seals. 



Length classes of tagged 3-year-old male 

 seals sampled from the kill are given by date 

 in appendix table 12. 



Table 4-1. — Mean weights and lengths of tagged and untagged 3-year-old male seals, by round, 



St. Paul Island, 1962-63 



[Numbers in parentheses are the number of seals in each sample] 



1 Data for rounds 2-4 (2-21 July) were omitted because untagged seals only were weighed and 

 measured. 



RELATED STUDIES 



LIVE -PUP COUNTS 



Live pup counts were made 5 August on 

 selected areas of Tolstoi and Kitovi Rookeries. 

 Counts made during 5 years are compared in 

 table 42. 



Counts of live pups on these areas were 

 begun originally to provide a means of meas- 



uring population changes expected to occur 

 as a result of herd reduction. The data ob- 

 tained to date, however, cannot be used to 

 measure population changes because they are 

 inconsistent. In the same year, the number 

 of pups on one area has decreased, while 

 on another the count has increased. This 

 approach can be used to measure population 



31 



