192 



NORTH AMERICAN FAUNA 68 



placed the annual rate of increase of a population at 7 percent. 

 In the Commander Islands, however, even this rate appeared some- 

 what high when compared with his **on the spot observations." 

 Lensink (1962), on the basis of his surveys, concluded that the 

 rate of increase of the Andreanof Islands population was 10 to 15 

 percent per year. The more comprehensive data gathered in the 

 same areas after his studies were completed, however, indicate 

 that the total population within the Andreanof group is now in- 

 creasing at a lower rate, probably because of density-caused 

 mortality in the heavily populated areas. 



In a local area of sparse population, the 1959-65 data indicate 

 the population may have increased during that period at an annual 

 rate of at least 10 percent. For example, the estimated population 

 at Seguam, the easternmost island of the sample area, increased 

 from 28 to 47 otters (table 25) in the 6-year period 1959-65 or at 

 a mean annual rate of a little more than 11 percent. This con- 

 clusion is based on the assumption that Seguam Pass is a fairly 

 effective barrier to immigration. 



After a large population is produced, as it was, for example, 

 at Adak (in the Andreanof Islands) in the 1954-62 period (table 

 28), the overall annual increase of the Andreanof population was 

 at the rate of about 4 percent. 



The observed 4 percent annual rate of increase is a generalized 

 figure. At islands where the population exceeded the carrying 

 capacity of the habitat, i. e., more than 10 to 15 otters per square 

 mile of habitat (see table 25), population regression was observed. 

 Stress mortality (starvation), as well as emigration, probably ac- 

 counted for the local population decreases. Evidence to support 

 this view was found on 28 April 1965 during a brief examination 

 of one otter hauling ground at Shagak Bay, Adak. There I found 

 a large juvenile otter dead on the beach. It exhibited all the 

 characteristics of starvation that I observed at Amchitka. Prob- 

 ably the rate of mortality during the winter-spring season of stress 

 at Tanaga, Kanaga, and Adak in the 1962-65 period was consider- 

 ably greater than at the relatively sparsely populated islands east 

 of Kagalaska during this same period. 



Thus, the rate of population increase varies according to popu- 

 lation density and its relation to ecological conditions and may be 

 quite different from the rate of reproduction. Elsewhere it is 

 shown that about 16 young may be born annually per 100 in- 

 dependent animals of both sexes and all ages. 



In general, it is concluded that an isolated population having 

 ample unused habitat may grow through local reproduction (no 



