1885.] Observations on the Muskrat. 1051 
should seek to do this, but in all probability the teeth, from con- 
tinued eating of vegetable food throughout the summer, become 
tender and are unable to cut hard grains of corn with ease. This 
is the case with many domestic animals in autumn when fed on 
corn after some months of pasture life. Muskrats are very fond 
of parsnips, turnips and apples. They frequent apple orchards 
and turnip patches, near their homes, and make use of much of 
the farmer's abundant crop of these articles. When snow, which 
had lain on the ground for some time, melted, I have observed that 
plats of grass near the water’s edge had been eaten bare by these 
animals while they were confined to such diet as they could find 
beneath the ice. Their food is not entirely vegetable; in winter 
and in early spring they subsist, in a great part, upon the flesh of 
river mussels. Many a winter morning have I found a number 
of well cleaned shells of the more delicate mussels upon the ice 
near swift, running water. I have never been able to satisfy my- 
self that this food was used by them at any other time of the 
year. Neither do I believe that this material was originally so 
used. It is very probable that owing to the scarcity of suitable 
vegetable food, they have been forced to include the meat of the 
mussel among their articles of diet; largely on account of its 
abundance near their watery haunts and also on account of the 
ease with which it is obtained. Such change of food has not 
Occurred in this region within historic time, perhaps, but it is evi- 
dent that formerly, when there were few mussels in these rivers, 
not so many of them were eaten. With the conditions favorable 
to their development produced by our canal, mussels multiplied 
very rapidly, and in proportion to their increase in numbers the 
muskrat increased his mussel-eating. Records of this are pre- 
Served in the banks of the canal; alternate deposits of shells, 
cleaned by the muskrat, and of sediment may be seen in many 
localities reaching to the depth of two feet below the present bed 
of the stream. Upon these same piles of bivalve remains the 
muskrat leaves the remains of most of the mussels it eats. I 
have never known the muskrat to eat univalve mollusks. I have 
identified the following shells as forming the principal part of its 
bivalve food in this vicinity: Anodonta plana Lea, A. decora Lea, : 
A. imbecillus Say, Unio luteolus Lam., U. parvus Barnes, Margari- 
tana rugosa Lea, and M. complanata Lea; all common in pro- 
portion to their comparative abundance. In some localities I 
