322 



Annals of the Carnegie Museum. 



Vol. Ill, pp. 95-98). For the purpose of showing clearly the char- 

 acters which have been used to distinguish the species (not to criticise 

 so excellent a naturalist) the principal features of his diagnoses are 

 presented in parallel columns. 



Leptodeira albofusca 

 Rostral twice as broad as deep. 



Frontal once and a quarter to once and 

 two-thirds as long as broad. 



Loreal as long as deep, or a little longer 

 than deep. 



One or two pre-oculars. 



A small subocular. 



Temporals i +2, or i +3. 



Eight upper labials, fourth and fifth, 

 (rarely third, fourth, and fifth) enter- 

 ing the eye. 



Four to six lower labials in contact with 



the anterior chin-shields. 

 Anterior chin-sh-elds as long as, or a 



little shorter than, the posterior. 

 Scales in 21 or 23 rows. 



Gastrosteges 170-2 11. 



Urosteges 71-95. 



A lateral series of spots. 



Lower parts whitish, frequently with 



fine brown specks. 

 Tropical America. 



Leptodeira annulata 



Rostral once and a half to once and 



two-thirds as broad as deep. 

 Frontal once and a half to once and 



two-thirds as long as broad. 

 Loreal as long as deep, or a little longer 



than deep. 

 One pre-ocular. 

 Rarely a very small subocular. 

 Temporals i +2. 

 Eight (rarely seven) upper labials. 



third, fourth, and fifth (rarely third 



and fourth, or fourth and fifth only) 



entering the eye. 

 Five or six lower labials in contact with 



the anterior chin-shields. 

 Antei'or chin-shields as long as, or a 



little longer than, the posterior. 

 Scales in 19 (exceptionally 17 or 21) 



rows. 

 Gastrosteges 175-196. 

 Urosteges 78-107. 



Lateral spots usually small or absent. 

 Lower parts white. 



Tropical South America. 



As all of the characters given vary enough to cross the narrow boun- 

 dary between the species, only a constant combination of several 

 characters can be made the basis of distinction between the two 

 supposed species. The number of scale-rows, the proportions of the 

 rostral shield, the presence or absence of a subocular, the number of 

 upper labials bordering the orbit, and the number of gastrosteges and 

 urosteges, are the most important characters used in the definition of 

 'the species, and if these occur in constant combinations the two species 

 can be distinguished, otherwise not. 



Several authors have used the number of scale-rows and the pro- 

 portions of the rostral as the chief characters by which to identify the 

 two supposed species. The number of scale-rows certainly appears 



