THE COMMON INDIAN SNAKES. 37 



Temporals — Two ; the lower touching three supralabials (usually the 

 5th, 6th and 7th). Supralahials — 8 ; the 3rd, 4th and 5th or 4th, 5th 

 and 6th touching the eye. Either the 3rd or 4th usually divided 

 and touching the eye. Infralabials — 6 ; the 4th, 5th and 6th, or 5th 

 and 6th only, touching the posterior sub-linguals ; the 6th largest and 

 in contact with 3 scales behind. Sttblinguals — Two pairs ; the anterior 

 rather the longer, the posterior completely separated. Costals. — Two 

 headslengiihs behind head 21, midbody 23, two headslengths before 

 vent 17. In the step from 21 to 23, a row appears on each side of 

 the vertebral ; in the reduction from 23 to 21 the two rows next to 

 the vertebral unite ; in the reduction from 21 to 19 (which occurs close 

 to the preceding step, in fact, the absorption of rows in these two steps 

 may be reversed), the 3rd above the ventrals is absorbed ; in the 

 reduction from 19 to 17, the 7th or 8th row above the ventrals is 

 absorbed. Vertebrals not enlarged. No keels. Apical facets present 

 in pairs. 



Ventrals. — 197 to 225. Anal — Divided. Subcaudals — 73 to 92 

 in pairs. 



Beoitition. — (From a single skull in my collection ). Maxillary, 

 13 subequal teeth, succeeded (after a gap that would accommodate 

 one tooth), by two teeth little if any larger than the preceding ones. 

 Palatine 1 1 , decreasing in lengiih from before backwards, the anterior 

 subequal to the maxillary. Pterygoid 15 to 16, decreasing in leng-th 

 from before backwards. Mandibular 18, decreasing in leng-th 

 anteriorl}^, and posteriorly from about the 5th. I think the denti- 

 tion is sufficiently distinctive to dissociate this species from the genus 

 Zamenis in which several species with various dental characters are 

 now grouped. 



I take the view in this paper and for the reasons specified in a 

 footnote that the forms of Zamenis described under the names of (1) 

 ventri7naculatus (Gray), (2) rhodorhacMs (Jan.), (3) ladacensis 

 (Anderson), (4) dor sale (Anderson>y, and (5) cJiesnei (Martin), are 

 all varieties of a single species for which the foremost name must 

 stand having priority. Boulenger in his Catalogue (1893, Yol. 1, 

 pp. 398 and 399) has already united Nos. 1 and 5, under the name 

 ventrimacidatus, and Nos. 2, 3 and 4 under rliodorhachis, but 

 distinguishes between these two*. 



* The descriptions of these two forms given by Boulenger in his Catalogue (1893) 

 are identical with the exception of the ranges of the ventral, and subcaudal shields. 

 I have now examined a large series of both forms from localities as widely separated 

 as Ahnora in the East to Aden in the West. The dentition of all these agrees with 

 that of the type of ventrimaculatus in the British Museum and the type of Anderson's 

 ladacensis in the Indian Museum which I have also examined. I can 

 discover no differences in lepidosis (examining specimens of each most critically 

 side by side) except in the ranges of the ventrals and subcaudals. In one form 

 both are more numerous than in the other. This becomes more noticeable in 

 a comparison of the aggregates of these shields. Specimens of each agree in 

 colouration, and they appear to grow to a similar length. I have examined the 



