134 University of California Puhlications in Zoology. [Vol. 7 



ing the river at Portage Cove were intersected by numerous 

 winding gulches, large and small, formed by the various streams 

 flowing down from the mountain sides, filled with water at high 

 tide, and nearly empty at low tide. These gulches diminished in 

 size as they left the river until they became mere narrow ditches, 

 sometimes completely hidden by the tall overhanging grass; and 

 here the muskrats had their home. There were not many of them, 

 and Hasselborg, who discovered them, captured probably almost 

 the entire colony at this point. Besides the five secured, one 

 was eaten in the trap, probably by an otter. This seems to be 

 an unusual place for them to winter, as all the swamps freeze 

 solid, and there are no ponds. Old trails show that they have 

 wintered on the surface of the meadows, under the snow, feeding 

 on twigs, roots, grass and shoots. 



The holes they were occupying were in the walls of the 

 gulches, the entrances being under water at high tide (see plate 

 4, figure 1). About these holes and in their trails, were piles 

 of cut grass, and other green stuff. No sign of muskrats was 

 seen in the meadows along the Cliiekamin River, on the mainland 

 directly opposite Portage Cove, in a region of very similar 

 character. 



Synaptomys dalli JNIerriam. Dall Lemming ]Mouse. 



Six specimens secured (nos. 8735-8740), one from the Cliiek- 

 amin River, four from Thomas Bay, and one from Port Snetti- 

 sham. At Thomas Bay the meadows in the river valley were 

 strewn with logs and other drift brought down by the stream 

 at high water, and it was under these that the lemming mice 

 were captured. The single specimen from the Chickamin River 

 was caught under a log in the woods, in a trap set for Evotomys. 

 The one secured at Port Snettisham was taken high up on the 

 mountain side, at the edge of deep snowbanks, in an open marmot 

 meadow. 



These six specimens are quite indistinguishable from a series 

 of S. dalli from the Prince William Sound region, Alaska. 

 Neither in external nor in cranial characters am I able to per- 

 ceive any differences whatever, although it might be supposed 

 that the form inhabiting this region would be S. wrangeli. 



