ECOLOGY AND BIOLOGY OF THE PACIFIC WALRUS 



193 



those of fur seals and sea lions; the same opinion was expressed by Scheffer 

 (1950). ElHott (1875) apparentiy was confused by this when he found that the 

 herd on Walrus Island, Pribilof Islands, was made up only of males. 

 Furthermore, he was told by the Aleut inhabitants of that area that females 

 never joined that herd. Dall (1902), in a less conservative mood, interpreted this 

 masculine group as "an assembly of peculiar character . . . entirely composed of 

 old males driven away from the herds by the competitive valor of their younger 

 and more active congeners, and forming a sort of old gentlemen's club, existing 

 in torpid dignity away from an atmosphere of irritating disrespect." The more 

 recent work of Bel'kovich and Yablokov (1961), Yablokov and Bel'kovich (1963), 

 Gol'tsev (1968), and Miller (1975a, 1976) has fully confirmed that such herds are 

 composed only of males and that they contain not only adults but subadults and 

 juveniles as well. 



Belopol'skii (1939), Freiman (1941), Nikulin (1941), and Brooks (1954) 

 predicted that small, loosely organized "harems" are formed on the ice in spring, 

 and that these are very transient because of the rapidly changing ice conditions. 

 However, they saw no such groups and produced no other supporting evidence 

 for their predictions, apart from reiteration of the fact of sexual dimorphism and 

 analogy with the Otariidae. Probable cases of coition were reported by 

 Mansfield (1958a) in April, by A. V. Yablokov (personal communication) in 

 "spring," by R. A. Ryder (personal communication) in May, by Brooks (1954) in 

 May, June, and July, by Johansen (1912) in August, by Freuchen (1935) in 

 September, and by the St. Lawrence Islanders (personal communication) in 

 November. However, in none of those reports was any mention made of social 

 organization, other than that a male and female were together. 



Some investigators have predicted that walruses are monogamous (Allen 1880; 

 Nutting 1891; Bertram 1940; Popov 1960a; Krylov 1968), rather than 

 polygynous. The principal basis for that prediction seems to have been that some 

 very small, mixed groups, resembling families do occur, especially where 

 remnant populations have survived, for example in the fjords of East Greenland 

 and about the islets of the Svalbard area (see Pedersen 1962). However, "family 

 groups" containing an adult pair (bull and cow) certainly are the exception 

 rather than the rule in the Bering-Chukchi region. 



Walruses in captivity seem to copulate in all months of the year, perhaps more 

 often in spring than at other times and certainly more often in than out of the 

 water (E. D. Asper, D. H. Brown, and G. C. Ray, personal communication). 

 During the period January to March, when most of the mating by free-living 

 walruses probably takes place, the animals tend to be well out in the pack ice, 

 remote from shore and not readily accessible for observ^ation. Only a small 

 amount of information has been obtained on their behavior during that time 

 (Ray and Watkins 1975; F. H. Fay, G. C. Ray, and A. A. Kibal'chich, unpub- 

 lished data), which may be summarized as follows. 



In most winters, nearly all of the adult females and young are situated in the 

 St. Lawrence and Bristol Bay concentration areas. There, the potentially estrous 

 females (i.e., those not carrying a near-term fetus) congregate in herds, each of 

 which is attended by one or more large adult bulls. When those herds of females 

 and young are at rest (on the ice or in the water) , the bulls station themselves in 

 the water alongside the herd, where they perform visual and acoustical displays. 



