HERPETOLOGY OF JAPAN. 425 



black with narrow whitish rings Avhich are mostly interrupted on the 

 underside, the whole lower neck and a more or less complete longi- 

 tudinal band on the ventrals })eing black; head entirely black with a 

 whitish postocular band and a pale spot on each of the prefrontals; 

 tail black with a few lateral, vertical, white bars or spots. 



Variation. — The extreme specialization of these snakes is accom- 

 panied by a correspondingly great variability. Thus the reduced 

 ventrals are so irregular and so often subdivided to the size of the 

 adjacent scales that their number in a large series shows an excessive 

 range, a feature alluded to above under the preliminary discussion of 

 the species (p. 420). The head shields also show considerable varia- 

 tion, though to a less degree. The supralabials are 7, as a rule, but 8 

 are occasionally found, and 6 very seldom; three of them may enter 

 the eye; two postoculars occur, but one is the normal number; pos- 

 terior chin-shields in contact as a rule, but occasionally separated by 

 a scale. The most notable variation, however, is the occasional, 

 though rare, j^resence of two anterior temporals, because the scale 

 formula of such specimens reads like that of D. cyanocincta, in which 

 two anterior temporals is the rule. The arrangement, however, is 

 different, for while in the former the two anterior temporals are pro- 

 duced by the horizontal division of the larger temporal located above 

 the very small sixth supralabial, in the latter it is caused by the hor- 

 izontal division of the fifth supralabial. In the former case we have 

 the very small sixth supralabial between the much larger fifth and 

 seventh, while in the latter the three are subequal or the sixth inter- 

 mediate in size between the other two. 



In the Pescadores Island specimens the small sixth supralabial on 

 the left side is abnormally fused with the fifth, as shown in fig. 339; 

 but on the other side it is normal. 



Habitat. — The locality whence came the type of this species is 

 given as "Indian Ocean," but as it is one of specimens brought home 

 by Sir E. Belcher, this probably means only that it was obtained 

 during his cruises in the "Samarang" in which case it is just as likely 

 to have been collected in the China Sea or even in the waters about 

 the Riu Kius, where it is evidently a common snake, as I have exam- 

 ined 24 specimens from the southern group (Saki shima), one from 

 Okinawa, and one from the Pescadores in the Formosa Channel. 

 Here also belongs the Il.fasciatus recorded by Boettger from'^Miyako- 

 shima." Finally there is the specimen in the Rijksmuseum in Lei- 

 den (No. 1483). It is labeled "Hydrophis striatus, von Siebold, 

 Japan." However, it is not the specimen figured under that name 

 in Fauna Japonica (Ophid., pi. vii), but most probably the specimen 

 referred to in the text (p. 89) as ' 'harponne pres des iles Lioukiou au 

 27me degr^ de lat. bor." apparently the only specimen collected by 



