12 BARBOUR — HERPETOLOGICAL NOTES [^v^i^vlF' 



The type of Celestus stenurus is, in my opinion, simply a large 

 adult of the same form. The species should then stand as 



Celestus co status (Cope). 



Panalopus costaius Cope, Proc. Acad. Nat. Sci. Phila., 1861, p. 494. Type, 

 M. C. Z., no. 3606, Jeremie, Haiti, collected by Dr. D. F. Weinland. 



l>iploglossus stenurus Cope, Proc. Acad. Nat. Sci. Phila., 1862, p. 188. Type, 

 M. C. Z., no. 3612, Jeremie, Haiti, collected by Dr. D. F. Weinland. 



Celestus phoxinus Cope, Proc. Acad. Nat. Sci. Phila., 1868, p. 125. Type, 

 M. C. Z., no. 12,457, formerly Phila. Acad., no. 9226, Jeremie, Haiti, col- 

 lected by Dr. D. F. Weinland. 



Celestus weinlandii Cope, Proc. Acad. Nat. Sci. Phila., 1868, p. 125. Type, 

 U. S. Nat. Mus., no. 12,145, Gonaives or Gonaves Island, Haiti, T. 

 Yoimglove collector. 



These types are mostly of different sizes. Thus weinlandii 

 represents, I believe, the very young; phoxinus and costaius 

 were based upon half-grown individuals, almost identical in 

 size, but one normal and the other artificially mutilated (Bar- 

 bour, Mem. M. C. Z., 44, 1914, p. 306) ; while stenurus is a very 

 large adult. Dr. Stejneger and I now both believe that the 

 number of striae on the dorsal scales increases with age, and is of 

 little diagnostic value except when comparing specimens of 

 similar size. Carman and I have both said in print that we 

 agreed with Boulenger that stenurus could not be separated from 

 stnatus ( = occiduus) of Jamaica. It is probably most wise, how- 

 ever, to await still more material before concluding that any of 

 the Jamaican species are really identical with the Haitian. 

 Nevertheless, after seeing a Haitian series, which tends to show 

 a great change between the coloration of young and adults, I 

 am strongly inclined to suspect that Celestus crusculus (Carman) 

 ( = bakeri Boulenger) is the young of Celestus occiduus (Shaw). 

 Of one series, labelled crusculus, all are small, including the type 

 (M. C. Z., no. 6051), and in color distribution and pattern they 

 are very similar to the types of phoxinus and costatus, which are 

 of about the same size; while our specimens labelled occiduus 



