244 NORTH AMERICAN FAUNA 68 



calculation are necessary to give equal representation to all times of the year. 

 A regression line fitted to the cube root of the weights against the mean time 

 in months give a regression coefficient of 



a=2.292 per month 

 or on a daily basis 



a=0.076 



This is an estimate of the specific growth velocity of Huggett and Widdas 

 (1951). In view of the very small samples obtained over all parts of the year 

 except the January to April period, it is subject to considerable sampling error 

 and it seems reasonable to round the estimate to 0.08. While this growth 

 velocity is somewhat below the value calculated for the European river otter, 

 it is quite close to values calculated for the northern fur seal (0.085 has been 

 calculated by Chapman, unpublished manuscript). 



If 1850 g. is used as an average fetal size at birth, the estimated period of 

 fetal development after establishment of the fetus is, 



Nyi850 12.3 , 

 ^.08 =70^=^^^ ^^^^ 

 or approximately 5 months. 



It is also necessary to estimate the length of the gestation period during 

 which the blastocyst is unimplanted. From table 54 we have the numbers and 

 percentage of pregnant animals with an unimplanted blastocyst and a con- 

 ceptus during the four seasons. 



It might be argued that lesser weight should be given to the two shorter 

 periods, January to February and March to April. If the results of these two 

 periods are averaged, and this average is incorporated into an overall un- 

 weighted average, the result is essentially the same. (The average so com- 

 puted would be 61 percent versus the 60 shown in table 54.) 



It thus appears that the period during which the blastocyst is unimplanted 

 is half as long again as the implanted period, or about 7V2 months. The full 

 gestation period will also include the period required for establishment of the 

 embryo, i.e., the period before linear growth is established. Huggett and 

 Widdas (1951) indicate a rough procedure to estimate this but it is not 

 known whether their values apply to animals that exhibit delayed implanta- 

 tion. If about V2 month is allowed for this period, it follows that the esti- 

 mated total length of the period of gestation is about 13 months. 



The period during which the female sea otter is not pregnant (lactating 

 and resting period) can be estimated similarly from the data from table 47. 

 This table shows that the percent nonpregnant in the four periods noted is 

 30, 30, 48, and 73, respectively. In this case the weight to be given to the two 

 first periods does make a difference. The average of the four percentages is 



Table 54. — Frequency of unimplanted and implanted pregnancy, by season 



[Data from table 47] 



Number Number 



animals with animals Percent 



Sample unimplanted with unim- 



period blastocyst conceptus planted 



January, February 35 26 58 



March, April 49 49 50 



May-August 4 9 31 



September-December 6 0 100 



Unweighted average 60 



