- it through the head and ate part of it. I gave him a black hornet, which he too 
_ by this species. He has made the same sound several times. It is something like. 
here, and shows a decided preference for the north side of steep hills. I have not 
Se OP AARON eae ak pik neh 3 
y hy } Sets F 
REPORT OF THE COMMISSIONER 
September 28.—Put a small Hesperomys in his box and he served it in the sam 
way that he did the other one; also gave him a Song Sparrow that [ had killed; he bit” 
and ate greedily, but the tail seemed to bother him, and he evidently got stung on 
the nose, but did not seem to mind itmuch. He is very fond of cream. September _ 
30.—Gave Onychomys a common gray moth, which he ate and seemed to relish. 
October 3.—He ate a piece of another mouse of his own kind, and tried toeat it be- © 
fore it was skinned. He has settled one thing for me; that _a squeaking ery which 
I heard evenings at Brown’s Valley, and once or twice at Devil’s Lake, wasmade 
the ery of a Flying Squirrel. Of 4 skinned, I was able to save but 1 without the loss ate 
of a patch of hair from the belly. Though the weather was cool, they would not ie 
keep more than six hours without the hair loosening over the belly. Most of those ie: 
captured at Brown’s Valley show one or more bare spots. I have noticed the same 
thing in Striped Gophers (S. tridecemlineatus) which had been feeding on grass- as 
hoppers. Probably insect food causes this tendency to early decomposition. The — “ 
excrement of this species is easily known by the remains of insects it contains. At 
Montana.—Tilyou’s Ranch, Dawson County, September and October, 1887: Com- 
mon on prairies and hills; probably lives in the holes of some other mice. 
RED-BACKED MOovusE (Evotomys gapperi). ate 
Minnesota.—lk River, Sherburne County, November, 1887: Common. Brown’s 
Valley, Traverse County, June and July, 1887: Verycommon. They live mostly in 
the banks of the Minnesota River, which is a narrow creek with high, steep, brushy 
banks. I found some in brushy ravines, and one on the prairie. ty 
Dakota.—Fort Sisseton, Marshall County, June, 1887: Caught 2. Seem common —_ 
in the brush land. Pembina, July and August, 1887: Common in the woods. Bot- 
tineau, August, 1887: Common; probably numerous along the creek. Caughtone : 
under a shock of wheat on the prairie. Deadwood, Lawrence County, October, 
1887: Caught one on the mountain top, and saw some tracks in the soft snow. 
MEADOW MOUSE (Arvicola riparius). 
Minnesota.—EKik River, Sherburne County, November 28, 1887: At times very ~~ 
numerous, and again scarce; at present date unusually numerous. The only way 
that I can account for this is that when the water is high in the brooks and on the —~ 
meadows, they are forced to leave their holes, and are destroyed by their natural 
enemies, while in the dry seasons they burrow in the brock banks and meadows. =~ 
This year is very dry. When numerous they destroy much grain if left long in the’ ~~ 
shock, especially corn left standing all winter, sometimes eating it half up. They ~ 
have killed a few apple trees for us by gnawing the bark around the bottom in win- 
ter when the snow was deep. Heron Lake, Jackson County, May 13-22, 1887: 
Caught one. Ortonville, Big Stone County, June 4-14, 1887: Probably common, 
although I found but few. Brown’s Valley, Traverse County, June 21—July 8,  ~ 
1887: Common in all low places. - 
Dakota.—Fort Sisseton, Marshall County, June 15-18, 1887: Common. Flan- 
dreau, Moody County, May 25-31, 1887: Found neara small slough on the flats of 
the Sioux River. Their paths extended in all directions through the grass, and 
were strewn with the grass which they had cut down for food, but I could findno 
holes or nests, and think they live under the dead and fallen grass which is thick in. 
places. Bottineau (on western edge of Turtle Mountain), August 22-30, 1887: Com- - 
mon; not numerous. Fort Buford, September 1-20, 1887: Probably common where 
there are marshes, but there are none near here. Found one dead in the road, and = 
saw one under brush and weeds in a ravine. 
PALLID MEADOW Mousse (Arvicola pallidus). va 
Dakota.—¥ort Buford, September, 1887: The Pallid Arvicola seemstobecommon = 
found them on the south, southeast, or southwest side. The only reason I can sug- 
gest for this distribution is that the twilight (their favorite hour) is longer on the’ 
north side. The hills where I have found them are all steepest on the north side, 
which may have some effect, though there seems to be no difference in the vegeta- 
tion_on different sides. Like other Arvicole they have many holes, and probably - 
live in families or colonies, although I have not caught more than one at a group 
of holes; but from the difficulty in catching them this does not signify anything 
(have caught only 4), Where there is grass or weeds their holes are connected by | 
