VARIATIONS OF GARTER-SNAKES. 71 



Fortunately in this form no tloubt attaches to the position of tlie 

 lateral stripe, as it is always distinctly present on the third and fourth 

 rows anteriorly, and <2;(uierally on the third only posteriorly aft(>r the 

 fourth is dropped. The scutellation ma}' be generally described as 

 follows: 21-19-17. 19-21-19-17, or 19-17 dorsal scale rows; 7 (8) 

 supralabials; 142 to 176 ventral plates, and 57 to 87 subcaudals. 

 There is so much variation, however, as we shall see later, that this 

 general statement has but little value. The tail length is quite 

 constantly about .23 to .26 in the males and .20 to .23 in the females. 

 The ground color is some shade of brown, with the usual three stripes, 

 the laterals greenish or bluish, antl the dorsal yellow, frequently 

 inclining to orange. The lateral spots, as in niegalops and marcianus, 

 are in three distinct rows, except when the ground color is so dark as 

 to obscure them. 



Hdhifs and luihitat rdafions. — Radix probably enjoys the distinction 

 of having furnished us with as much or more information concerning 

 its habits than any other form in the genus, and yet it will be very 

 apparent from the following summary that our knowledge is still 

 ver3\ incomplete. Taylor (1892, 324), in writing of its habits in 

 Nebraska, saj^s that radix ''in food habits agrees with specimens of 

 E. sirtalis var. panefdlis of the same size. Earthworms and insect 

 larvae seem to constitute the bulk of their food." Elsewhere the food 

 of large individuals of panctalis is said to be the leopard frog, while 

 specimens "not exceeding two and one-half feet in length almost 

 always contain within their stomachs specimens of the common 

 earthworm." 



In 1882 Dr. Henry Brous, in the American Xaturalist (1882, 564), 

 recorded the following notes on the habits of this snake: 



Several of the sumiiiers I passed upon the plains were preceded by rainy springs, 

 swelling to uiuisual height the small streams which l)ecome inhabited by small tishes. 

 During the drouglit of hot summers the receding waters left the fishes in shallow i)ools 

 within creek beds, an easy prey to their numerous enemies. 



The midday lieat caused numbers of snakes to seek shelter from the sun, and the 

 garter-snake {Eutaenia radix) in parti(!ular chose water at this time. Here the fishes, 

 unable to escape or find deep cool water, were unwilling cotenants with the snakes. 

 The latter are fond of fish, and would devour great numbers of the smaller ones, chasing 

 them from one part of the shallow pool to the other. WTien the fishes were in water 

 too shallow to swim in, or were struggling upon the sand, they would be seized by the 

 snakes, who would feed upon them until unable to contain more. The snakes would 

 follow the fish through the water, diving and remaining submerged some time. I did 

 not observe them swallow air. (See Am. Nat., Jan., 1880.) Snakes evince more than 

 ordinarj- energy and sagacity in capturing fish; half a dozen will congregate within a 

 small pool, all acting in concert. 



Mr. J. L. Wortman, who had charge of a scientific party last year, informs me that 

 while fishing one day he caught numbers of chub {Cyprinidse) and, throwing them on 

 the mud, was surprised to see but few remained. \\Tiile quietly continuing to replace 

 those so singularly missing, he observed a garter-snake seize and swallow one of the 

 fish 6 inches in length. There were two of these snakes reaping the benefit of Mr. 



