^10 



Frits Johanskn. 



floating as is usual with the snout above the water) down below the 

 surface and bend the trunk in an arch above the water; then they 

 straighten out the back, so that the head strikes upwards on the 

 under edge of the ice; the head goes down again and so on. 



From what has been said earlier, the impression has perhaps 

 been obtained, that the walrus prefers the open water to the ice; 

 this only refers however to the solid and extensive fjord and coast 

 ice which is formed in the course of the winter, but not to the 

 drift ice, which in the form of small floes and bergs comes in with 



Fig. 4. Two walrus breaking tiio tliiii ice at Sneiues "Iv 1907 

 (to the left the coast covered with snow). 



the current in the summei'. On the contrary, I have observations 

 (Hvalrosodden August 1906 and Stormkap August 1907), which 

 show that the walrus practically follows with the drift ice, mean- 

 ing therewith that it is mostly in the "clear spaces" (open water) 

 of the drift ice, that they remain for a long time diving up and 

 down and playing, while this is never seen in the large, open parts 

 of the fjord or if the latteris quite free of ice. But this agrees 

 well with the present geographical distribution of the animals on 

 the whole (as high-arctic forms). 



With regard to the swimming of the w^alrus, I have the follow- 

 ing observations to record. If they are lying out in the water, they 



