252 



NORTH AMERICAN FAUNA 68 



beaches. Other accessible beaches in the study area were searched 

 periodically but less intensively than the favored hauling-out areas. 



Each dead animal was autopsied. Specimens — including skins, 

 skulls, bacula, reproductive tracts, stomachs, and organs appearing 

 pathological — were preserved when practicable. The extent of 

 decomposition determined the amount of information and specimen 

 material that was obtained. 



Because it was important to determine the period of greatest 

 mortality, it was necessary to ascertain the usual progress of 

 decomposition, and to estimate the approximate time of death of 

 animals that had been dead for varying periods. Up to a point, 

 this procedure was useful. The time of death for skeletal remains 

 found in midsummer, however, could be placed only to the previous 

 winter-spring season. During periods of high mortality the 

 beaches were searched weekly. Thus, the week of death was known 

 for many animals and the month of death for others. Data for 

 otters that had been dead for more than 1/2 to 2 months were re- 

 corded, but because of uncertainty most of these records were 

 excluded from those aspects of the study which required more 

 exact data. Each carcass was removed from the beach after 

 needed data were recorded and specimens taken from it. 



Average daily air temperatures in winter infrequently exceed 

 36° F., and spring and fall average daily air temperatures above 

 49° F. are unusual. For this reason, otter carcasses decompose 

 slowly on Amchitka beaches. The study was complicated when 

 scavengers (Bald Eagles and rats (Rattus norvegicus)) mutilated 

 carcasses, sometimes preventing the gathering of certain data. 



Results 



Between July 1955 and April 1963, approximately 15 months 

 were spent on Amchitka. During six study periods, a total of 

 331 dead otters found on the beaches were examined. Because 

 data from certain individuals were incomplete, these are elimi- 

 nated, where appropriate, from the following discussions of dif- 

 ferent aspects of mortality. 



THE TIME OF MORTALITY 



It would be ideal, in a study of sea otter mortality, if data could 

 be gathered during some period of consecutive months for several 

 years. This could not be done, so the data from various seasons 

 in different years were used to present a composite picture of 

 the annual cycle of mortality. The mortality study included parts 



