VARIATIONS OF GARTEE-SNAKES. 73 



water is not deep enough, except in occasional pools, to prohibit a 

 rank growth of grasses and sedges, which grow for the most part in 

 clumps on compact elevations formed of decaying roots and rhizomes. 

 Between these hummocks the substratum is soft, plastic mud, with 

 usually a varying amount of water over the surface. This habitat 

 is a very favorable one for frogs, and all stages of the leopard frog 

 are abundant, from the tadpoles in the pools to the large adults hop- 

 ping about in the grass. The margin of these swamps is also pre- 

 eminently the preferred habitat of radix, and the collector can at 

 any time during the summer months pick up several dozen speci- 

 mens of various sizes in a short time at almost an}' of these places. 

 There can be no question as to at least one feature of their food in 

 this habitat, for it is a common experience while walking about the 

 margins of a slough to hear the hoarse "quark" of a frog in distress, 

 and a brief search will almost invariably reveal a leopard frog in the 

 jaws of one of these snakes. The size of the frog which they can 

 eat with comparative ease is scarcely creditable. Tadpoles are also 

 eaten voraciously, and I have opened the stomachs of snakes which 

 were fairly gorged with them. The snakes swim freely about in the 

 pools and probably capture small fish as well as tadpoles. 



About the margins of the lakes it is also abundant, although gener- 

 ally it does not appear to be as numerous in this habitat as about the 

 sloughs. Abundant as it is in swampy habitats, however, radix is far 

 from being confined to these conditions, and we have often taken indi- 

 viduals on the hills (fig. 24) half or three quarters of a mile from water 

 during dry hot days in July and August. In such places, however, 

 they are found commonly only in the morning or evening or on wet 

 and cloudy days, seeking the protection of holes during the heat of 

 the da}'^. Their food habits on these ridges are necessarily somewhat 

 different than in the sloughs. I have observed specimens in August 

 in the stubble, and in and about the shocks of grain, capturing the 

 small tree frog {OhoropMlus riigritus). These tree frogs are, however, 

 of rather rare occurrence in tliis region, and for its food radix must 

 de})end upon other forms. One of these is that staple article of diet 

 for so many of the plains forms in the fall — the grasshopper. This 

 insect occurs in great abundance in late summer and furnishes food 

 for such a variety of animals as wolves, foxes, badgers, gophers, cranes, 

 grouse, and hawks. I have often seen these snakes coiled up on a 

 shock of wheat with a grasshopper's legs protruding from the corners 

 of its mouth, and there can be little doubt that it forms an important 

 part of the food in the upland habitats. It is very possible that they 

 also rob the nests of the field mice which are often made in the shocks 

 of grain, for young mice are often found in the nests at tliis time 

 which have not as yet gotten their eyes open, and would thus fall an 

 easy prey. During September they have also been seen in newly 



