ECOLOGY AND BIOLOGY OF THE PACIFIC WALRUS 



61 



hairs occurred per canal, one usually was much larger than the other(s); the 

 small hairs were otherwise exact replicas of the large hairs and not readily 

 classifiable as secondary hairs or underfur. Each small hair arose from a separate 

 follicle but emerged with the larger hair from the same hair canal. The hairs 

 extended 7 to 12 mm above the skin over most of the body but generally were 

 shorter on the face (4 to 8 mm). They were intensely pigmented in the distal 

 third of the shaft in the youngest animals and were virtually colorless in some 

 very old adults. As in the Phocidae, the hairs lacked medullary air spaces, though 

 the shafts were stouter than those of the medullated guard hairs of the Otariidae 

 (Sokolov 1960; Scheffer 1964). The pelage hairs of adult walruses were similar in 

 shape to those of the newborn calves. However, they were stouter and longer, 

 and the blade was more lenticular in cross section (Table 8; Figs. 37 and 38). 



The annual molt in most adults seems to be more gradual than the postnatal 

 molt of the calves, beginning at least 1 to 2 months earlier and continuing well 

 into the autumn; mixed populations of both the new and the old generations of 

 hair are present during most of the molt's course (A. W. Mansfield and J.J. 

 Burns, personal communication) . Although I noticed no outward signs of growth 

 of new pelage hairs in any of the animals examined during March to early June, 

 histological preparations of skin from the top of the head of one female taken on 

 31 March showed hair follicles in anagen. Skin from the crown of three others 

 taken at the same time showed only resting follicles. Another taken on 14 May 

 also had follicles in anagen. 



J. W. Brooks and K. W. Kenyon (personal communication) observed that a 

 few adult males taken in Bering Strait about mid-June already had shed much of 

 their hair. Bailey and Hendee (1926) found new hair "underneath the loose 

 epidermis" in animals taken in the same locality during June and July. Most of 

 the old, cornified layer had been shed from a male that 1 examined at Gambell 

 on 27 August. Mixed populations of old and new hairs (ratio about 1:3) over most 

 of its body were remarkably short (less than 4.5 mm), the old hairs having been 

 broken and the new hairs not completely emerged. The follicles of the new 

 generation of hair were still in late anagen. Mansfield (1958a) obser\'ed that 

 some new hair on adults was up to 12 mm long by August, but that some old hair 

 was retained at least as late as October in walruses of the Canadian Arctic. 



Indications of a more rapid, and perhaps more thorough, molt in animals at 

 the southern edge of the summer range were recognized by J. W. Brooks, K. W. 

 Kenyon, and me on 23 to 27 June at Round Island. There, all but a few of the 

 oldest bulls had shed virtually all of their old hair (Fig. 41), and about one-third 

 of the animals already had a coat of new hair a few millimeters long. The 

 nakedness of males that probably were undergoing a comparable molt was noted 

 in another southern locality, the Pribilof Islands, by Elliott (1875:160) on 5 July. 

 He also observed that "the thick, yellowish-brown cuticle . . . seemed to be 

 scaling off in places." At Round Island in June, shed hairs and flakes from the 

 cornified layer were abundant on the beaches and rocks where the animals had 

 lain. Although the further progress of this molt has not been closely followed, 

 photographs taken in August of animals (presumably the same ones) on the 

 beaches of Round Island (Goro 1960) showed them in a fully developed, though 

 perhaps not fully emergent, new pelage. 



Total depilation, followed by total replacement, seems to be the rule also in 



