REPTILIA. 335 



Alsophis porto-ricensis Reinhardt & Lutken. 



Reinhardt & LtJTKEN, Vid. Meddel. nat. foren Kjobenh. for 1862, 1863, p. 221. Stejneger, Rept. 

 U. S. nat. mus. for 1902, 1904, p. 700, fig. 170. 



Closely related to A. antillensis. These species parallel in their distribu- 

 tion Leimadophis stahli and L. exiguus, since, like the former, this snake is con- 

 fined to Porto Eico and Mona Island. L. stahli has not been found upon Mona, 

 but doubtless occurs there. 



Alsophis melanichnus Cope. 

 Cope, Proc. Acad. nat. sci., Phila., 1862, p. 76. 



I have identified with this poorly described species a fine large male taken at 

 Santiago de la Vega, San Domingo, by Mr. A. H. Verrill. After reading the 

 following in Stejneger's Herpetology of Porto Rico (Rept. U. S. nat. mus. for 

 1902, 1904, p. 701), I compared it carefully with his description of A. porioricen- 

 sis. Stejneger writes: — 



"The question whether Cope's A. melanichnus, from Haiti, is identical with 

 the Porto Rican species can not be said to be settled yet. The scale formula 

 seems to be the same, but it is not certain that there may not be other distin- 

 guishing features We have such startling proof of the effect of isolation 



working with a variable material to produce separate forms of these snakes 

 that it is quite unwarrantable to lump the various names without incontrover- 

 tible proof." 



Stejneger says the scale formula "seem to be the same" in this species and 

 A. portoricensis. Cope, however, only mentioned the one fact that it had seven- 

 teen scale rows. Cope mentions No. 1,522, M. C. Z., from Jeremie, Haiti, as the 

 type; but the specimen can not be found. Upon comparing our only specimen 

 with Stejneger's description, or with a specimen of A. portoricensis, I find that 

 the Haitian snake agrees in all characters, except in a ^'ery elongate loreal, much 

 lower than high, and almost a rough triangle in outline, the base being anterior. 

 It has two first temporals on each side, but these have the appearance of being 

 the elements derived from an abnormally divided single scale. The ventral 

 scales are 189 in number, and subcaudals 108, the anal is divided. These counts 

 will be seen to vary from the range and average of Stejneger's forty-four speci- 

 mens from Porto Rico, which was 17 rows, 169-183 (average 178) ventrals, sub- 

 caudals 112-129. This combination of characters makes it seem almost certain 

 that the Haitian species is perfectly distinct from, but very closely related to, 

 the Cuban A. angulifer, as well as A. portoricensis. 



