ECOLOGY AND BIOLOGY OF THE PACIFIC WALRUS 



245 



Although ice apparently played a major part in determining the availability of 

 the animals to the St. Lawrence Islanders in 1940-68, adequate harvests prob- 

 ably would have been taken in all years, had the walrus population been larger. 

 In that period, the Eskimos of the island numbered only about 550 to 600 

 people — about one-third as many as in the early 19th century. Nonetheless, even 

 with modern firearms and other advantageous equipment, they were unable to 

 obtain enough walruses in some years to meet their minimum requirements. If 

 remedial food supplies had not been provided by Federal and State governments, 

 the islanders probably would have been afflicted again by starvation and death, 

 at least in 1954-55, as they were in 1878-79. For those and other reasons (Fay 

 1957; Geller 1957), I believe that the Pacific walrus population was at a lower 

 level in the 1940's and 1950's than it had been at any time within the past 

 300 years. 



Aerial Census 



The best estimates of the size of the Pacific walrus population have been 

 derived from sample counts during aerial surveys by Soviet and American biol- 

 ogists (Table 40). In each instance the total size of the population was estimated 

 by extrapolation from the numbers counted in sampled areas to the total area of 

 occupation. The first such survey was conducted in the western Chukchi and 

 eastern East Siberian seas in August-September 1958 by P. G. Nikulin (in 

 Fedoseev 1962), who counted about 20,000 animals on the coastal hauling 

 grounds and estimated that at least 20,000 more were at sea. A more intensive 

 survey by Fedoseev (1962) of the same area and of the northern Gulf of Anadyr 

 in September-October 1960 yielded an estimate of 46,000 animals in Soviet 

 waters; at least 4,000 more were assumed to have been in waters off Alaska. 

 Earlier, Kenyon (1960a) and co-workers conducted two aerial surveys of the 

 population in its wintering area in the Bering Sea from February to April 1960 

 and derived estimates from each of about 70,000 to 100,000 animals. A similar 

 estimate was obtained from a third survey in March 1961 by K. W. Kenyon 

 (unpublished data). In September-October 1964, Gol'tsev (1968) estimated 

 47,000 to 71,000 walruses, mainly on the coastal hauling grounds of Chukotka, 

 and in April 1968, K. W. Kenyon (unpublished data) obtained an estimate of 

 73,000 to 110,000 from a fourth aerial survey over the Bering Sea pack ice. From 

 aerial surveys in Soviet waters in September-October 1970, Gol'tsev (1972) 

 counted 62,000 animals on the hauling grounds and estimated the total popu- 

 lation at 101,000. A fifth survey by Kenyon (1972) of the population in the 

 Bering Sea pack ice in April 1972 yielded a conservative estimate of 85,000 to 

 162,000 walruses. Finally, a Soviet-American cooperative survey in Sep- 

 tember-October 1975 in the Chukchi and western Bering Sea gave an estimate of 

 140,000 to 200,000 animals (Estes and Gilbert 1978; J. A. Estes and V. N. 

 Gol'tsev, unpublished data). 



None of these estimates should be regarded as actual counts. Each is unreliable 

 to some degree because of problems encountered in obtaining and interpreting 

 the data on which it was based. Ray and Wartzok (1976), Chapman et al. 

 (1977), and Estes and Gilbert (1978) have reviewed most of those problems in 

 detail. 



The Soviet estimates, in each instance, were lower than those from the 



