HERPETOLOGY OF JAPAN. 471 



The locality whence Captain Belcher obtained the specimens upon 

 which Gray established the new species is not known. Gray suggested 

 'Svest coast of (North?) America?" Bj^ a study of Belcher's itinera- 

 ries both in the Sulphur and in the Samarang I have come to ih^ 

 conclusion that the only likely place where he can have obtained 

 the specimens in question is Ishigaki shima, an island in the southern 

 group of the Riu Kius, where he stayed some time and undertook con- 

 siderable survejnng on shore", the very same island whence come two 

 of the specimens of this species examined by me. In fact I have but 

 little doubt that Ishigaki shima is the true type-locality. 



With regard to the probability of this island also being the type- 

 locality of T. luteus, see remarks farther on under habitat of this pres- 

 ent species (p. 474). 



Thanks to the courtesy of Doctor Boulenger, I have been able to 

 examine Gray's types of C. elegans. They consist of two young speci- 

 mens, No. 47.3.4.62, with the upper edge of snout and supraoculars 

 turned up, and with the temporals slightly but distinctly keeled, 

 and No. 47.3.4.64, an adult male, with temporals scarcely keeled. 

 There is, therefore, no doubt in my mind that Boettger's T. luteus is 

 the same. 



T. elegans is very closely allied to the Formosan T. mucrosguamatus , 

 and while there are various average characters which separate the two, 

 the number of ventrals is the onh^one which seems to offer an abso- 

 lutely reliable fliagnosis. In all the specimens of T. elegans examined 

 by me the supraoculars have the outer edge sharply turned up, but in 

 the only small Formosan specimen before me (Sci. Coll. No. 18) this 

 edge is similarly turned up. Neither the keeling of the temporals, nor 

 the size of the canthal scales offer any stable characteristics. 



In the three specimens of each species now before me — the onh^ ones 

 I can now examine with regard to this character — I find another differ- 

 ence, viz, that in the three Formosan T. mucrosquamatus there are two 

 loreals, one behind the other, followed by a long horizontal upper pre- 

 ocular, while in the three Riu Kiu T. elegans there is only a single loreal 

 between the nasal and the elongate preocular. 



Additional characters may be found in the lower number of supra- 

 labials, the somewhat wider su'f^raoculars, and resultant narrower 

 interorbital space in T. elegans as well as in the lower number of scale 

 rows, characters, however, which intergrade and offer no hard and fast 

 line of separation. The coloration is essentially the same, with a 



"The "examination of Pa-tchung-san " [Ishigaki shima] "occupied us twenty-one 

 days," Belcher, Narrative Voy. Samarang, I, 1848, p. 75. 



See also Doctor Adams's reference to a species of " Trigonocephalus " in the Meiaco- 

 shima group in his "Notes from a Journal of Research into the Natural History of the 

 Countries visited during the Voyage of H. M. S. Samarang under the command of Capt. 

 SirE. Belcher, C. B.." in Belcher's Narrative Voy. Samarang. II. 1848. pp. 305 and 

 306. This may be an alhision, however, to Agkistrodon affinis. 



