584 Life-histories of Northern Animals 



areas and concluded that these animals completely ploughed 

 the surface of the country, that is, turn it all over, at least once 

 in two years. 



An even better illustration I observed on the South 

 Williams River of Colorado, September i6, 1901. Under the 

 shade of some scattering aspens where we were camped I 

 measured off a space 24 feet square and carefully plotted the 

 Gopher-mounds on it (Fig. 170). The firm outlines are around 

 the hills that were less than a week old, the dotted lines mark 

 hills about a month old, but the whole surface had evidently 

 been turned this summer; the ground was so soft that the horses 

 sank six to ten inches in walking across it. 



This was a fair sample of the flats and hollows throughout 

 the White River region, except where it was rock or very hard 

 clay. 



The Californian example (Fig. 170) shows about 75 

 hills on its area of 57 square yards; at this rate there would 

 be 6,000 hills to the acre. An average hill among these was 

 18 inches across and 4 inches high, as nearly as possible 

 one-third of a bushel, therefore 2,000 bushels per acre were 

 upheaved within two or three months by the Pocket-gophers, 

 enough to cover the whole area evenly to the depth of one 

 inch. 



In the Colorado example (Fig. 171), with 74 hills to 64 

 square yards, the result appears slightly lower, but it should be 

 remembered that in this case the hills over a month old were 

 not counted, as heavy rains had melted them to a general 

 level. Those remaining would, if levelled off, cover the ground 

 to a depth of one-half an inch; this was one month's work for 

 the Gophers. 



In the other cases cited, 4 to 5 bushels were thrown up in 

 one night by a single Gopher. In the second Sierra case the 

 47 hills or 25 bushels thrown up would equal 1,500 bushels 

 per acre, or if levelled, about one-half inch over the entire 

 surface in two days. 



These figures show possibilities rather than general proba- 

 bilities. 



