BREEDING HABITS OF FIELD MICE. 11 



found them nearly a foot above the ground. Sometimes they are 

 placed under flat stones or logs or under shocks of grain. The struc- 

 tures are so slight that a day's sunshine will dry them out after a 

 storm, and yet they are so compact that the animals pass the coldest 

 weather snugly housed in them under the snow. 



The nests are admirably located with respect to drainage, being so 

 placed that they are not likely to be flooded during excessive rains, 

 and often being in depressions of mounds made by the mice, well 

 above the general surface of the meadows. 



Trails, often of great length and worn smooth by constant use, 

 lead to neighboring feeding grounds. As far as possible they are 

 under shelter of old grass, fallen weeds, leaves, and other material. 

 The trails of some species are almost entirely below the surface of the 

 ground, and short tunnels are common with most species. At inter- 

 vals burrows of varying depth occur, usually but a few inches below 

 the surface. These lead to underground nests, to the roots of food 

 plants, or serve only for refuge from enemies. 



BREEDING HABITS. 



While most surface nests are for shelter only, sometimes the young, 

 especially of swamp species, are produced in them. However, the 

 young of most kinds are born in underground nests and are rarely 

 seen unless uncovered by accident. They are at first hairless and 

 blind. When discovered in the nest the mother vole slips noise- 

 lessly away, sometimes carrying the young attached to her mammae. 



The breeding season includes most months of the year, except mid- 

 winter in cold latitudes and periods of long-continued drought. The 

 number of litters in a .year thus depends on climate, and especially 

 upon the character and length of the winter. In temperate latitudes 

 in normal seasons from four to six litters are produced. Even the 

 same species in different localities differs much in this particular. 

 That the number of young in a litter varies with the different species 

 may be inferred from the variation in the number of teats. Some 

 species produce normally from two to four young, others from four 

 to eight; but the variation in the same species is remarkable, and 

 depends partly upon climate, but probably more on the scarcity or 

 abundance of food. The largest litter recorded by the field natural- 

 ists of the Biological Survey is thirteen, the species being the dwarf 

 vole (M. nanus). The period of gestation is not positively known, 

 but probably is about twenty days. Members of the Biological Sur- 

 vey have recorded finding young of the American species in the 

 nests, or females containing embryos, in every month of the year 

 except January and February. 



European testimony proves conclusively that, like the lemmings, 

 voles at times increase in numbers abnormally, but the causes are 



