322 NORTH AMERICAN FAUNA 61, FISH AND WILDLIFE SERVICE 



caches totaled about half a peck. The shifting sands must have 

 covered those openings repeatedly, therefore the mouse undoubt- 

 edly found the spot by a sense of location and scent. 



At Urilia Bay, we found another cache that was made up of 

 bases of stems of undetermined plants and roots. 



Microtus oeconomus popofensis 



This is another slightly differentiated form, occupying the 

 Shumagin Islands. Specimens have been obtained on Popof and 

 Unga Islands. This mouse is, of course, very similar to operarius, 

 but the skull appears to have a more slender rostrum, with a 

 little longer and definitely wider incisive foramen. Judging from 

 specimens at hand, the underparts of popofensis are more tawny 

 than in the specimens from the mainland. 



In 1936, these mice were extremely abundant at one place on 

 Unga Island, near a bird colony. The ground was honeycombed 

 with burrows, and mice were seen running about occasionally. 



They are known in the Shumagins as well as on Sanak Island, 

 as "gophers," while shrews are called "mice." 



Microtus oeconomus amakensis 



Strangely enough, this form, which has the most restricted 

 range, is one of the best defined. The skull differs from all other 

 mice in this species, particularly in the occipital region — the 

 flat occipital surface contrasts with the convex surface in the 

 other forms. In this feature, the skull of amakensis suggests the 

 appearance of skull of M. o. kamtschaticus, though the series of 

 the latter is small and not entirely comparable. Also, the incisive 

 foramen of amakensis is short and blunt, contrasting with the 

 attenuated foramina in other forms. The feet and top of tail are 

 paler than in the other forms. 



It is puzzling that this form, which is confined to small Amak 

 Island located only 14 miles north of the coast of Alaska Penin- 

 sula, is more distinct than the mice on other islands that are 

 equally as far, or farther, from the mainland. It is possible that 

 unfavorable transportation aspects have tended to isolate this 

 island, thus emphasizing a distinct form. 



In 1925, when I visited Amak Island, meadow mice were 

 extremely abundant. Runways were everywhere, in the grass, 

 underground, under driftwood, among old whale bones on the 

 beach, as well as among the lava rocks and moss on the higher 

 portions of the island. When walking over the low ground, we 



