HERPETOLOGY OF JAPAN. 



309 



shorter than parietals, the sides strongly convergent backward, the 

 width at the posterior hiteral angle being less than width of supra- 

 ocular; nostril large, between two nasals, the posterior of which is less 

 than half as wide as the anterior; loreal much wider than high; one 

 preocular, not in contact with frontal; a subpreocular; eye large, its 

 diameter equaling half its distance from tip of snout; tw^o postocu- 

 lars; temporals 2 + 3; eight supralabials, seventh largest, fourth and 

 fifth entering eye; five lower labials in contact with anterior chin- 

 shields, which are slightly longer than posterior; 23 rows of scales 

 all strongly keeled except outer two rows; 214 ventrals, angularly 

 bent laterally ; anal divided ; 97 pairs of subcaudals. Color (in alco- 

 hol) olive clay color, pale underneath; back with blackish ill-defined 

 cross markings the black color only involving the edge of the scales 



Figs. 260-268.— Elaphe carinata. § X NAT. size. 266, top of head; 267, side of ukAU; 26S, under- 

 side OF HEAD. No. 5, Sci. Coll. Tokyo. 



and the skin between; a longitudinal blackish band on ventrals near 

 the lateral angle and frequently interrupted; each ventral with the 

 posterior edge irregularly marked with black; subcaudals narrowly 

 edged with black so as to form a black median line ; all the upper and 

 lateral cephalic sutures marked with black. 



Dimensions. 



mm. 



Total length 1, 455 



Snout to vent 1, 155 



Vent to tip of tail 300 



Tip of snout to posterior end of parietals 34 



The young (U.S.N.M. No. 36508) are much paler and underneath 

 uniform white; back with a few scattered black dots which, on the 

 anterior half of the bod}", form short and narrow cross lines; on pos- 

 terior half an obscure, pale brown longitudinal line on each side; 

 upturned ends of ventrals with small black dots, forming an irregular 

 longitudinal series; upper and lower labials with dark brown sutures. 



Variation. — The commonest variation is apparently in the number 

 of temporals, the extremes recorded (for Chinese specimens) being 1+2 

 and 3 + 3. The scale rows are almost always 23, but Wall mentions 



