586 JOURNAL, BOMBAY NATURAL HISTORY SOCIETY, Vol. X 



shape of the pupil of the eye. This is of less service than the pre- 

 ceding criteria, but it at once excludes the genus JDryopJds from con«- 

 sideration in the exanriination of the snake assumed. The fourth 

 character adopted is the number and arrangement of upper labial 

 shields, the positions of which are shown in the diagram on page 278 

 of Mr. Boulenger's work. We have already arrived at the conclusion 

 that the snake we are examining must either be Zamenis korros or a 

 Dendropliis. The number of upper labials in Z. korros being 8, and 

 that in Dendropliis picius being 9, the area of search would now be 

 narrowed down to the genus Dendrophis^ and as Mr. Boulenger gives - 

 easily applied kejs based on external characters for the ascertainment 

 of all species within the genera, the specific identity of the snake can, 

 with the aid of his book, be now ascertained with certainty. 



Thorough key hei*e printed does not, therefore, pretend in many 

 cases to do more than direct the observer to the correct genus. It will 

 often indicate the species, but in large genera, such as Silt/dura, 

 SimoteSj Tropidonotus, etc., it would have taken up too much space to 

 give criteria sufficient for the identification of each species, and it 

 would also have been waste of time to do so, as they are easily avail- 

 able in Mr. Boulenger's book, which everyone attempting to use this 

 key should possess. The chief drawback which seems likely to attend 

 the use of the key is the variation which undoubtedly occurs in some 

 of the characters adopted as criteria. The number of upper labial 

 plates, of ventrals, and subcaudals is liable to vary from that adopted 

 in the key, and to this extent identification may be made more difficult. 

 But it is hoped that this source of error will not be sufficiently import- 

 ant to deprive the key of all utility. 



It should be added that where the number of scales round the mid- 

 dle of the body varies from the number of scales round the neck, the 

 former has been adopted. The last two snakes in the key, the Cobra 

 and the Hamadrj^ad, did not lend themselves to the arrangement adopt- 

 ed, and being well known, have been placed by themselves at the 

 end. In the Typhlopidce the scales can hardly be said to be in so 

 many " rows" ; but the same phrase has been retained throughout for 

 convenience. In many of the marine snakes the number of ventrals 

 and subcaudals could not be given, but little difficulty is likely to arise 

 over the identification of this class of snake. In conclusion, it should - 

 be once more stated that the imperfections of the key are fully recog- 

 nized and that it is only presented as a possible assistance to those 

 whose study of snakes is in an elementary stage. 



