616 JOURNAL, BOMBAY NATURAL HISTORY SOCIETY, Vol. XVII. 



deviation towards a new type, the effect is the same, and certain 

 individuals must occur which depart in some way or other from the 

 accepted normal type. This being so, any key however well con- 

 structed will fail to correctly indicate certain individual specimens. 

 In framing keys one endeavours to select characters which are found 

 to be most stable in individuals of the same species, so as to minimise 

 the chances of misleading. 



Tkopidonotus xenuka, spec. nov. 



The habitat of this Tery interesting snake is not recorded. That it 

 belongs to this genus I have little doubt, though it differs from all the 

 other known species (80 in number) European, Asian, African, 

 Australian and American in having the subcaudals entire. The 

 entire anal too is a character only as yet known in one other of the 

 many Asian species, viz., T. pealii. Another very unusual feature 

 about this snake is in connection with the supracaudals. These scales 

 in snakes where the subcaudals are in pairs are arranged in even rows. 

 I know of no exception to this rule. On the other hand in snakes 

 such as the genera Bungarus, Aspidura, Achalinus, etc., where the 

 subcaudals are entire, the supracaudals are arranged in odd rows, and 

 even in those snakes where a combination of entire and divided subcau- 

 dals exist, as Bungarus bungaroides, Naia bungarus, Ancistrodon acutus, 

 etc., I have always found the supracaudals arranged in odd rows 

 corresponding to the entire subcaudals, and even rows corresponding to 

 the divided subcaudals. In this snake however the supracaudals are 

 arranged in even rows though the subcaudals are entire. 



Description — Rostral, in contact with 6 shields, of which the anterior 

 nasals form larger sutures than the internasals. Internasah, a pair ; 

 the suture between them subequal to that between the prajfrontal 

 fellows, subequal to the internaso-prffifrontal suture. Prefrontals* 

 a pair; the suture between them subequal to the prEefronto-frontal 

 suture; in contact with internasal, postnasal, loreal, prseocular, suprao- 

 cular, and frontal. Frontal, touches 6 shields, of which the supraocular 

 sutures are longest and about ^ lunger than those formed with the 

 parietals. Supraocular, length subequal to frontal ; breadth | frontal. 

 Nasals, divided; in contact with 1st and 2nd supralabials. Loreal, rather 

 longer than high. Prceocular, one ; not touching frontal. Eye, pupil 

 uncertain. Postoculars, 3. Temporals, 2 ; the lower touching the 7th 



