138 BARBOUR: ZOOGEOGRAPHY. 



is valueless. A glance at Boulenger's account of this form will show that 

 variety A occurs in Java and Borneo; B in Straits Settlements and Borneo; 

 C in Straits Settlements, Sumatra, and Borneo; E in Borneo and Sumatra; 

 F in Borneo and Celebes. Of variety D, only one specimen is in the British 

 museum; and of G, only two; both taken at the same time. These varieties 

 have been given names by Boulenger; and these do doubtless afford useful 

 handles by which to speak of the different specimens, but they do not seem to 

 express anything which is valuable to the zoogeographer or student of ophidian 

 relationships. It is noteworthy that all the specimens at the Buitenzorg mu- 

 seum, as well as three taken at Buitenzorg and now in the collection of the M. 

 C. Z., had the coloration designated A by Boulenger. Inasmuch as this phase 

 also occurs in Labuan, we can not name it as a local race, however. Since the 

 above was ^\Titten, Bryant's series has brought in seven from Buitenzorg, and 

 four from Mt. Salak, all of var. A, the only one occurring in Java. 



Annandale has noted the very large size of the poison-apparatus of D. 

 bivirgatus in the Malay Peninsula; and Gadow, in his Amphibia and reptiles, 

 1901, p. 634, has mentioned it for this species (see fig. 17). Boulenger (loc. cit.) 

 notes the pushing of the heart to a position far posterior of its characteristic 

 position. This apparatus is figured here because none of these accounts give 

 one a definite idea of the enormous development of the glands, and this condi- 

 tion seems to be quite unknown to the average comparative anatomist. 



The curious habit of crawling wath the tail up-turned to show the brilliant 

 red under color is also figured. This has been noted before by Annandale (Fasc. 

 Malay. Zool., 1903, 1, p. 167-168). Flower (Proc. Zool. soc. London, 1899, p. 

 693, pi. 37) has described and figured the same habit for Cylindrophis rufus. In 

 spite of the enormous poison-glands, and the widespread native suspicion directed 

 against all the red-tailed snakes, it seems very unlikely that this reptile can be 

 considered a really dangerous form. 



Amblycephalus carinatus (Wagler). 

 Wagler, Nat. syst. Amphib., 1830, p. 181. Boulenger, Cat. snakes Brit, mus., 1896, 3, p. 445-446. 



Dr. Stejneger writes me that A. carinatus Boie is a complete nomen nudum. 

 Reinwardt had sent specimens to Leyden, using this MS. name, which Boie 

 simply copied. Dr. Stejneger points out that this evidently happened before 

 Boie himself left Holland. The authority for the name must therefore stand as 

 Wagler. He described the species as the type of a new genus, Pareas; and, just 

 as Dr. Stejneger also has written me, credits the name "Dipsas car-inata" to 

 "Reinw. in Mus. Lugd." The type locality, then, is Java. 



