HERPETOLOdY OF POKTO RICO. 



687 



be traced, the iiicshcs of Avhich anteriorly coincide with the outline of 

 the scalcis, but becoming- more and more discordant posteriorly; rostral 

 and anterior nasal brown above, marg-ined with wldtish, underneath 

 whitish; a xcvy abrupt whitish spot occupy ing the anal region and the 

 under side of the tail. 



Dimens!o)is. 



mm. 



Tip of snout to vent 197 



Vent to tip ( )f tai 1 3 



Diameter of 1 » kIv 5 



Variidloii. — Phe individual variation appears to be similar in char- 

 acter and extent to that indicated for Tuplilopx Imnhrkudis. Three of 

 our specimens have 18 scale rows, the fourth has 20. 



JIabifat. — Thus far the only definite locality whence we have 

 received this species is Lares, in the western part of Porto Kico. In 

 this place Mr. A, B. Baker obtained three specimens. A fourth was 

 sent from Porto Rico ])y Mr. Charles E. Adams, but without indica- 

 tion of the exact locality where he collected it. 



lieiiKd'l's. — This interesting novelty resembles in color closely the 

 specimens which Boulenger has described from Dominica under the 

 name of Ty2)1(^ <>l>^ }>^ nt ii<'(-l>^>(fl^i^- The latter differ, however, by hav- 

 ing- 24 scale rows around the body and by a much wider rostral which 

 is said to be one-third the widtli of the head. Dumeril and Bibron's 

 TypJdops j>JatycrphaJ>is^ wdiich is said to have come from Martini({ue, 

 collected by Plee, nuist l)e a very diti'erent one from the Dominica 

 species. It is described as having 20 scale rows and it might conse- 

 quently be suspected to be the same as our Porto Rican species, but 

 apart from the radically diti'erent coloration, the true T. ]>l(tfii<;'plinhiii 

 is descril)edas liaving the head more depressed than T. luiiihf/caJis and 

 the posterior supralabials much lower, while, as we have seen, two of 

 the })est characters of our new T. ro-stelJufus are the lack of depres- 

 sion in the profile of th(» head and the unusual height of the posterior 

 supralabials. 



Boulenger's T. j>I<ft//cc/>/HfIt/s" is apparently different from that of 

 Dumeril and Bibron and may ]>e appropriately named TypJilop^ doinhi- 

 Icana^ from the island of Dominica, of which it is a native. 



Ij'iat <jf sj)c<''uiinii^ of Tijj>li/oj>s n/shi/atu.s. 



«Cat. Sn. Brit. Mu!<., I, 1893, p. 30. 



