104 



where they had gone to hibernate and to transform later into 

 mature insects. In making a search in similar locations in the 

 spring, I found that there remained only a very small per cent, 

 of the number that was known to have entered the ground in 

 the fall, the missing ones having evidently been devoured by 

 some small animal whose burrows traversed the ground beneath 

 the trees in every direction.* Several mouse-traps were set in 

 these burrows and in less than a week the traps caught over 

 twenty short-tailed shrews under a single chestnut-oak tree. 

 Some of the shrews were captured alive and kept in confinement 

 and would feed ravenously on the worms with which I supplied 

 them. One ate seventy-two large chestnut-worms, Balaninus 

 proboscideus, in less than five minutes, while another gorged 

 itself with nearly all the flesh of a full-grown white-footed mouse 

 in the morning and in the afternoon of the same day ate thirty 

 chestnut-worms at one meal. 



A few white-footed mice were also caught in the traps but 

 the larger number of shrews taken, and the avidity with which 

 the captive individuals ate the worms given to them, seemed 

 to justify the conclusion that the shrews had been the principal 

 agents in destroying the worms. In many places, too, the ani- 

 mals in searching for the insects had apparently worked over 

 the soil thoroughly between the burrows, which indicated the 

 operations of shrews rather than of mice. The mice when caged, 

 however, would also eat the worms and no doubt they had aided 

 to some extent in the good work done by the shrews. 



Near the oak where the shrews were caught there stood a 

 large chestnut tree that had produced a heavy crop of nuts 

 the previous season. Beneath this tree I selected a spot at ran- 

 dom and drove four stakes enclosing a plat of ground measuring 

 six feet on either side. The plat was then marked off into 

 blocks of one square-foot each and by removing the surface of 

 the ground carefully with a trowel, one section at a time, a 

 diagram was made of the location of all the burrows found. The 

 figure below shows the distribution of the burrows on the thirty- 



