VARIATIONS OF GARTER-SNAKES. 87 



tail. It is difficult to (letorinino what basis Brown had for the 

 statement that it is a connect injj^ hnk between 'proxiinus and meg- 

 alops, other than the general relationship expressed by the position 

 of the lateral stripe, for the range of radix and proximus overlap 

 to a considerable extent without affecting the purity of either type. 

 Rather it seems, as I shall try to show later, that radix and prox- 

 imus, although both members of the same division (see table, p. 40), 

 have had an independent origin from a common stock. 



It has already been noted that to the eastward on the prairie 

 peninsula radix exhibits a tendency toward a reduced scutellation, 

 and that in this reduction the scale formula 19-17 is approached. 

 This brings radix very close to another form (hutleri), whose western 

 limit, as far as we know at the present time, is close to the eastern 

 limit of radix, and makes it necessary to examine this form before 

 the affinities of radix in this direction can be discussed. 



BUTLERIM 



Description. — East of the prairie peninsula, in Indiana, Ohio, 

 southern Michigan, and western Pennsylvania, radix is replaced by 

 a form which, like it, is a very distinct and well-marked one. In 

 this form, which has received the name of hutleri, the ground color 

 is dark olive brown, the lateral spots when visible on the scales 

 being small and restricted to the scale rows adjacent to the stripes, 

 along which they tend to form narrow broken black borders. On 

 the skin these spots are very seldom to be distinguished in definite 

 rows, being more often partially fused, leaving the interspaces irreg- 

 ularly arranged. The dorsal stripe is usually on the median and 

 adjacent rows, and is bright yellow anteriorly but quickly fades out 

 to a dull yellow posteriorly. The lateral stripe is broad and con- 

 spicuous, occupying all of the third and most of the second and 

 fourth rows anteriorly, but posteriorly, where the fourth row is 

 dropped, it falls on the second and third rows only, and retains 

 this position to the vent. Anteriorly the color of the lateral stripes 

 is bright yellow; posteriorly they are generally somewhat dulliM-, 

 still retaining, how^ever, their yellow tint. The first row of scales 

 is but little, if any, lighter than above, which serves to accentuate 

 the conspicuousness of the lateral stripes. 



The scutellation may be sinnmarized as follows: Dorsal scale rows 

 19-17, occasionally 17-19-17-15, superior labials 7 or 6, ventrals 132 

 to 154, subcaudals 49 to 68. 



Butler's garter-snake has had a rather peculiar history. Described 

 in 1888 by Cope, on the basis of a single specimen said to have been 



a Thamnophis hutleri (Cope), Proc. U. S. Nat. Mvis., XI, p. 399. Includes E. 

 brachystoma Cope, in part E. sirtclis Brown, in part E. sbtalis obscura Cope, and in 

 part E. sirtalis obscura Morse. 



