SEA SNAKES IN THE INDIAN MUSEUM. 279 



are normally two placed behind the 3rd infralabial ; and in most of the other species one 

 single scale is wedged between the 3rd and 4th infralabials. 



(ix) Infralabials, — these number three in one species only, viz., D. jerdonii ; in most 

 others there are four, and in a few Distira five. The 4th is almost always the largest of 

 the series, but in some Distira the 5th is che largest. When the 4th is largest it comes 

 into contact with three scales behind as a rule, more rarely two scales only, as in H. 

 gracilis and H. cantoris. The suture between the first is usually larger than that between 

 the anterior sublinguals in most of the genus Hydrophis, while it is usually smaller in the 

 genus Distira. 



(x) Sublinguals are usually present in two pairs, but frequently absent, as in E. 

 valakadyen, Hydrus platurus, H. ccerulescens and the genus Enhydris. The anterior are 

 almost always in contact with three infralabials and the posterior with the 3rd and 4th ; but 

 in D. jerdonii the anterior touch two infralabials only, and the posterior the 2nd and 3rd. 



(xi) Costals, — the costals have been fully studied by others. I count these in three 

 places ; they vary within certain limits in individuals of the same species, but in many 

 cases are valuable aids to differentiation. The condition of imbrication or juxtaposition 

 of these scales, to which Mr. Boulenger has given considerable attention, is, I find in many 

 cases, difficult to decide to one's own satisfaction, though in many cases it is a valuable 

 character. In most of these snakes the costals are subequal everywhere, but in both 

 species of Enhydris a very noticeable enlargement of the lowest three rows is an im- 

 portant aid to recognition. 



(xii) Ventrals, — in almost all the sea snakes these shields are distinct though small, 

 i.e., not twice the breadth of the adjoining costals. Often they are more or less broken up 

 at intervals. In two snakes they are wholly absent, viz. , Enhydris hardwickii and Hydrus 

 platurus. In one snake they are distinct anteriorly and wanting posteriorly, viz. , Enhydris 

 curtus. In two species they are distinct posteriorly and unusually broadly developed 

 anteriorly, i.e., become three or four times the breadth of the adjacent costals, viz., in 

 D. viperina, while in H. schistosus and in the genus Platurus they are as broad throughout 

 as in the land snakes. 



It may be urged that many of these characters are too inconstant to merit the im- 

 portance I attach to them, but I can very emphatically state that I have found them far 

 more constant than many of the characters quoted by other authors, such for instance as 

 the number of the praeoculars, postoculars, and labials, characters which I consider quite 

 useless. Again their constancy is not one whit inferior to that of the characters relied 

 upon to identify other snakes, all of which show occasional departures from the normal. 

 In the descriptions which follow I quote in brackets all the exceptions I have met with, and 

 this refers to all the specimens I have examined, not only those in the Indian Museum. 



The collection includes 6 genera, and 24 species, ail with one exception (H. elegans) 

 occurring within our Indian limits. 



Three I have described as new, whilst I have eliminated three others previ- 

 ously regarded as distinct, viz., Distira gillcspice (Boulenger), Distira tubcrculata 

 (Anderson), and Distira andamanica (Annandale). The collection contains upwards of 

 150 specimens. 



