196 ANNALS NEW YORK ACADEMY OF SCIENCES 



name applied by Dumeril and Bibron to the Typhlops from St. Thomas, 

 on the ground that it is logically likely to be conspecific with the Porto 

 Eican form. In its more elongate form and greater number of scales 

 from snout to vent, the description of T. richardii applies fairly well to 

 the series from Porto Rico. Typhlops lumbricalis is reported from all of 

 the Greater Antilles and from a large number of the Lesser Antilles. 

 Being unable to examine specimens from the Lesser Antilles, I am unable 

 to form an opinion as to the status of the form found there. The Linnean 

 description of T. lumbricalis is said by Dumeril and Bibron to be taken 

 from the Jamaican Amphisbcena argenta of Browne. Not having 

 Jamaican specimens for comparison, I have compared the Porto Rican 

 series with the specimens from Cuba in The American Museum of Nat- 

 ural History. 



In fourteen specimens examined in detail, the total length varies from 

 216 to 310 mm.; average, 266 mm. The largest specimen available from 

 Cuba measures 244 mm., the average length of nine specimens is 203 

 mm., and the specimen selected by Barbour for description in the 

 "Herpetology of Cuba" (Barbour and Ramsden, 1919, Mem. Mus. Comp. 

 Zool., Cambridge, XLYII, p. 185) measures only 182 mm. It appears, 

 therefore, that the Porto Rican and Cuban Typhlops are distinctly differ- 

 ent in adult size. The ratio of the body diameter' to total length is only 

 slightly different in the two series, 34 to 44 in the Porto Rican, 27 to 38 

 in Cuban. The number of scales about the body is 22-20-20 in seven 

 specimens, 22-20-18 in seven, in the Porto Rican series; the reduction to 

 20 scale rows occurring only a little anterior to the middle of the body. 

 In the Cuban series the scale formula is 20-20-18 in six, 20-18-18 in 

 three specimens. The number of scales counted on the mid-dorsal line 

 from snout to tail spine ranges from 365 to 415 in the Porto Rican 

 specimens and from 270 to 325 in the Cuban. (The Cuban specimen 

 described by Barbour has been reexamined at my request by Mr. Emmett 

 R. Dunn and is found to have between 320 and 325 scales on the mid- 

 dorsal line.) In spite of the relatively small series of Cuban specimens 

 at my disposal, it seems unlikely that the range of variation of a larger 

 number would be greatly different. 



The Porto Rican specimens are darker in color, and, with a single ex- 

 ception, the tail is marked with a white ring. In the specimen described 

 by Stejneger (1904, p. 685), as well as in the exception here noted, the 

 white ventral color forms a prominent notch on the side of the tail, indi- 

 cating the existence of the tendency to form a ring. No such notch or 

 ring is found in Santo Domingan or Cuban specimens. 



