78 BULLETIN 61, UNITED STATES NATIONAL MUSEUM. 



northern boundary of marcianus, is very imperfectly known. The 

 southernmost locaHties in the plains from which radix specimens are 

 definitely known are Scott and Marion counties, Kansas, so that the 

 ranges of radix and marcianus are at least adjacent if not adjoining 

 or even overlapping. There is no evidence at present to indicate an 

 overlapping, except the single specimen recorded by Cope from Dallas, 

 Texas. Tliis is without question a typical specimen of radix, but the 

 record should be held in reserve until it has been substantiated by 

 further collecting; it suggests, how^ever, that the range of radix may 

 be found to extend farther south in the prairie region of central Texas 

 than on the more arid plains to the west, which would be in accord- 

 ance with our knowledge of the habitat relations of the form. If the 

 respective ranges of marcianus and radix overlap in the debatable terri- 

 tory in Oklahoma and southern Kansas it can not be to any great 

 extent, and we believe that we are safe in considering that the south- 

 ern boundary of radix on the plains approximately conforms to the 

 northern boundary of marcianus. 



The northern termination of the domain of radix is even more imper- 

 fectly know^n than the southern, and it is difficult to even approximate 

 it, for here, as to the south, the plains conditions are evidently con- 

 tinued far beyond the range of the form. The notes and specimens 

 of Coues (1878, 278) indicate that it is abundant on the northern 

 boundary of the United States from the valley of the Red River to 

 the foothills of J.he Rocky Mountains, but that it also occurs farther 

 north is indicated by the specimens in the U. S. National Museum 

 from Regina ana Rush Lake, Assiniboia, British Columbia, which 

 at present seem to be the most northern localities in which it has ever 

 been taken." The extent of its northern occurrence is probably 

 dependent upon its ability to endure the lower temperatures that 

 characterize this region, and since snakes are as a rule a warm climate 

 group, and apparently can not stand extremely cold conditions, we 

 may expect to find the northern limits of the area occupied by radix 

 not far north of Regina. 



As was noted in discussing the habits of radix, even in the more 

 moist parts of its domain, it apparently prefers the wet swamps to the 

 dryer uplands. From this it may be expected that the western bound- 

 ary of its range is not determined b}' the Rockies themselves, but by 

 the high arid table-lands that border them on the east. It is easily 

 conceived how a form of more or less general distribution on the 

 prairies but having a preference for moist habitats might, on the 

 more arid plains to the west, become more closely confined to these 

 conditions, to its exclusion from the plains, and be carried far west- 



o There is a single specimen (No. 9251) in the U. S. National Museum labeled " Lake 

 Winnepeg " collected by Doctor Gunn, which, although a typical specimen of radix, 

 can not be considered here, for the locality is evidently a general one. 



