248 



NORTH AMERICAN FAUNA 74 



10 



■20 



30 



112 

 1277, 



40 



50 



Equivalent Still Air Temperature (°C) 



Fig. 137. Percentage of walruses 

 sighted on the ice (versus in the 

 water), in relation to the range 

 of equivalent still-air tempera- 

 tures. Equivalent temperatures 

 based on measured air tempera- 

 tures and surface wind velocities 

 ("windchill": Siple and Passel 

 1945) at the time of sighting 

 from icebreaking ships in the 

 Bering Sea in February to April. 

 (From J. J. Burns, J. A. Estes, 

 F. H. Fay, and G. C. Ray, un- 

 published data) 



(mean, one-sixth) of the actual number present in the water. 



Aerial photographic capability seems essential, to minimize human error in 

 visually estimating the numbers of animals in large herds. Accurate counts of 

 groups larger than 10 to 15 animals usually are not feasible from the moving 

 aircraft; small animals are particularly difficult to detect in large groups, even in 

 aerial photos. Walrus groups on ice tend to be larger than those in the water and 

 largest on the coastal hauling grounds (Fay and Ray 1968; Estes and Gilbert 

 1978). Furthermore, they differ seasonally and regionally in size (Table 41) as 

 well as in composition. Males tend to be in smaller groups in winter and early 

 spring than in late spring and summer (Table 42); females tend to form larger 

 herds than males in all seasons and to congregate most in late spring and summer 

 (the "nursery herds" of Burns 1970). 



Composition 



The sex-age composition of the Pacific walrus population was estimated 

 earlier by Freiman (1941), Fedoseev (1962), Burns (1965), Krylov (1968), and 

 Fedoseev and Gol'tsev (1969), each of whom concluded that immature animals 

 of both sexes made up about 30 to 40% and that the sexually mature adults made 

 up about 60 to 70 % of the population. The sex ratio at all ages was assumed to be 

 about 1 male:l female, based on the known ratio at birth. Each of those 

 estimates was based in part also on samples of animals observed at sea, on 

 samples from the catch, and on analogy with other mammalian populations. 

 Each was admittedly imprecise but was remarkably similar to the others, 

 considering that each was based on a different set of field and harvest data. 

 Although this similarity suggests that each estimate was close to the actual com- 

 position, the investigators acknowledged that sampling error, especially in the 

 catch data, could have been large because of selection for certain size classes. 

 Also, by analogy with other polygynous pinnipeds, a 1:1 sex ratio of adults seems 

 highly improbable. 



