HEBPETOLOGY OF JAPAN. 295' 



plainly a small triangular, or rather semilunar, scute behind the eye 

 fitting into the blunt angle between the anterior temporals. A 

 similar scute is also discernible in the young specimen No. 34047. 



Thus far the genus has only been recorded from southern China and 

 Japan. Its nearest relatives seem to be found in Ceylon and the 

 Malay islands. However, specimens are as yet rare and it has been 

 but recently discovered, so that it is not safe to draw any conclusions 

 from its distribution and relationship as known at present. 



Boulenger recognizes three species of tliis curious genus, viz, A. 

 rufescens"' and A. hraconnieri^ , from southern China, and A. spinalis, 

 from ''Japan (?)." Of these, the first and the last were only known 

 from one specimen each at the time the first volume of his catalogue 

 was published (1893), while of A. hraconnieT'i he had four specimens. 

 A. rufescens, with its very long internasal suture and partly tricarinate 

 scales'^ niay be regarded as unquestionably distinct. Additional 

 specimens from Japan, however, show so much deviation from the 

 type of A. spinalis and so much variability in the direction of A. 

 hraconnieri that it may become difficult to maintain the latter as a 

 distinct species. The characters which are supposed to distinguish 

 the two species break down one by one. A. hraconnieri was charac- 

 terized as having 23 to 25 scale rows as against 21 in the type of A. 

 spinalis, but five additional specimens from Japan seem to indicate 

 that 23 is the normal number in the latter. Furthermore, in A. hra- 

 connieri the suture between the internasals is said to be shorter than 

 that between the prefrontals, while in the type of A. spinalis these 

 sutures are described as of equal length.^ This is also approximately 

 the case in our specimen from Kiusiu (U.S.N.M. No. 30731), but 

 in the specimen in the Science College Museum, in Tokyo, also from 

 Kiusiu, the interprefrontal suture is much longer, the one between 

 the internasals being only 0.3 the length of the one between the pre- 

 frontals. In the two specimens from Mount Fuji (U.S.N.M. Nos. 

 34046-7) the relations are 0.4 and 0.6. Finally, A. hraconnieri has 



aBouLENGER, Anil. Mag. Nat. Hist. (6), II, 1888, p. 43, (type-locality, Hongkong); 



Cat. Sn. Brit. Mus., I, 1893, p. 308, pi. xx, fig. 2; III, 1896, p. 612. 

 b Ophielaps braconnieriSAvv age. Bull. Soc. Philom. Paris, (7), 1, 1877, p. 109 (type- 

 locality, eastern Kiangsi, China). — Achalinus hraconnieri Boulenger, Cat. Sn. 

 Brit. Mus., I, 1893, p. 309; III, 1896, p. 613. 

 c The ventrals of the type are given as 136, subcaudals 82, and Boulenger (Cat. 

 Snakes Brit. Mus., Ill, 1896, p. 612) in another male specimen from Hongkong coimts 

 137 ventrals and 76 caudals. This is greatly at variance with the number of scutes 

 given by Doctor Wall for four Hongkong specimens, namely, 150+56, 158+58, 158+61, 

 and 154+58, respectively. In view of the apparent narrow limits of variation in this 

 series as well as in A. spinalis, I am at a loss how to explain the above discrepancy. 



d Peters says: "Die Praefrontalia sind nicht lilnger als die Internasalia. "" In the 

 figure (fig. la), however, the prefrontal suture is considerably longer than the inter- 

 nasal suture. Which is correct? 



