POIHONOUS SNAKES OF NORTH AMEKICA. 443 



Variation. — There is some slight variation in the size of the scales 

 covering- the top of the head, which is, in a general way, correlated with 

 the latitndeof the locality, but the character is so exceedingly unstable 

 and apparently unsupported by any other character tliat it would be 

 unprofitable and misleading to adopt any subspecies at the present 

 stage of our knowledge, at least. This instability in the size of the 

 head scales sometimes goes so far as to fuse them into nearly regular 

 shields on the snout and between the eyes, exactly in the same manner 

 as in certain specimens of Crotalus atrox, which have been called Cro- 

 talus sciitulatus, the result being that this so called species or subspecies 

 is a mixture of rt/ro.P and confliientus. Such scutulated specimens have 

 been found both in Arizona and in Montana, the two extremes of its 

 geographical distribution. A specimen from Fort Hayes is recorded by 

 Garman as having the outer edge of the supraocular produced into a 

 horn, as in G. cerastes. 



The color varies also greatly being sometimes duller, sometimes 

 brighter, lighter or darker, depending upon age, season, condition of 

 skin, climate, and the predominating color of the surroundings, but I 

 have seen no differences of such a character or stability as to render 

 the recognition of geographical races possible or profitable. 



Oeographtcal (listribiition. — Broadly speaking, the Prairie Rattlesnake 

 occui^ies the area bounded in the East by the ninety-sixth meridian and 

 the Upper Missouri Valley; by the main divide of the Rocky Moun- 

 tains in the West; by the thirty-third parallel in Texas and the Mexi- 

 can boundary further west in the South; and by the fiftieth parallel in 

 the North. In the Northeast its distribution apx3ears to be limited by 

 the watershed between Missouri and the Red River of the North, 

 according to Dr. Coues (Bull. Geol. Surv. Terr, iv, 1878, p. 207), who 

 cellected numerous specimens along the Canadian border between this 

 watershed and the crest of the Rockies. He also states that it is to be 

 considered fairly common in the region of the Upi^er Missouri and Milk 

 River and some of their northern tributaries; its range thus extending 

 some distance into the British Possessions, where Mr. James M. 

 Macoun informs me that it is most abundant from Medicine Hat, on the 

 Saskatchewan, to the boundary. In the region just south of the above, 

 Dr. J. A. Allen, while attached to the Union Pacific Railroad expedi- 

 tion, found itcommon, esi^ecially in the bad lands of the Little Missouri 

 and along the Yellowstone, outnumbering all the other ophidians 

 together, and on the expedition of 1872 not less than 2,000 were killed 

 (Proc. Boston Soc. Nat. Hist., xvii, 1874, p. 69). In Nebraska, Taylor 

 (Amer. Natural., xxvi, Sept. 1802, p. 752) observes that it was formerly 

 abundant all over the State, but that it is now confined almost wholly 

 to the middle and western part, where it is by no means rare. The dis- 

 tribution in Kansas, Oklahoma, and Indian Territory is very similar, 

 while in northwestern Texas it is recorded from between the main 



