-9- 
a two year period at Kure indicates that several factors must be considered 
in censusing the monk seal. The number of individuals on the beach at any 
one time may be affected by season, weather extremes, and time of day, 
% 
but no known rule can be applied with assurance to any set of circum¬ 
stances. Pups can swim at birth and may go out into shallow water with 
the female within two weeks, and thus could be missed on a beach count. 
The low returns (1) of 56 pups tagged at Kure strongly suggest that pups 
leave the island within two months of parturition and do not return for 
► 
several years. There is a tendency for fewer seals to be present on the 
beach during the day in the warmer months from May through October. In 
general the animals avoid high temperatures, high winds, and driving 
. rain, apparently preferring to be in the water during weather extremes. 
Weather data are available for each of the days on which seals were 
counted at Kure, and there is little apparent correlation, with one ex¬ 
ception, between weather conditions and numbers present on the beach. 
The peaks in December 1964 and January 1965 are related to a severe storm 
in December which obliterated all the sandspits west of Green Island which 
are normally favorite hauling grounds. These sandspits were reformed by 
March 1965, and thus this does not explain the peak in that month. There 
are usually more animals on the beaches at night than in the day. Human 
disturbance at Midway Island may have reduced the herd even further from 
that previously estimated (Kenyon and Rice, 1959; Rice, 1960). Construction 
of the Coast Guard facility at Kure Atoll, with a concomitant increase in 
human disturbance to the resident herd, may have reduced the number of 
animals which haul out on Green Island, but lumping Green Island counts 
