HEEPETOLOGY OF JAPAN. 371 



The collective species Dinodon septentrionale is recorded from the 

 eastern Himalayas or Khasi Hills, in Assam, and the Karen Hills, in 

 Burma, to eastern Cliina and Formosa. 



The exact relation of the specimens from the various localities is 

 not very well known, but there seem to be several color forms, of 

 which the easternmost and the westernmost are the extremes, viz, 

 the typical subspecies D. septentrionale from Assam or Himalaya, and 

 Burma, on the one side, and D. septentrionale ruhstrati, from Formosa, 

 on the other. Specimens from the Chinese coast province of Fokien, 

 according to Boulenger,'^ are "intermediate in the pattern of color- 

 ation" between these two forms. The status of the form which 

 occurs in the moimtains of Kiukiang is still uncertain on account of 

 scantiness of the material. 



It is unfortunate that the full scale formulas of the three Fokien 

 specimens has not been published, as altogether the formulas of only 

 seven specimens have been recorded, and the more so since the latter 

 seem to indicate a possible difference in the number of subcaudals in 

 one of the races. I find, namely, that in the three Formosan speci- 

 mens in which the tail was complete the subcaudals number, respect- 

 ively, 103, 104, and 97, while in the three continental specimens on 

 record they are stated to be 83, 87, and 88. 



Under these circumstances it seems best to maintain, for the present 

 at least, the subspecific distinctness of the Formosan specimens. 



A partial translation of Doctor Fischer's original description of the 

 scutellation and coloration of the Formosan specimens is subjoined, 

 as I have not seen any specimens of this form. 



Description. — South Formosa; Mus. Oldenburg; Ruhstrat, col- 

 lector. Rostral just reaching the upper side of the snout; interna- 

 sals slightly broader than long, longer than the prefrontals; frontal 

 pentagonal, as broad as long, shorter than the interparietal suture; 

 nostril between two nasals, of which the posterior is much higher 

 than the anterior; loreal pentagonal, twice as long as high, its pos- 

 terior angle reaching under the preocular; latter well developed, 

 quadrangular, twice as high as long, resting on the third supralabial 

 and part of the loreal, extending to the upper surface of the head, 

 though not reaching the frontal; two postoculars, the lower resting 

 on fifth and sixth supralabials; temporals 2 + 3, the two anterior in 

 contact v.ith both postoculars; eight supralabials, of which third, 

 fourth, and fifth join the eye; ten infralabials, the fii'st six in con- 

 tact with the cliin-shields ; the posterior chin-shields somewhat nar- 

 rower and but slightly shorter than the anterior; between these and 

 the ventrals two or three rows of elongated throat scales; scales in 

 17 rows, only the six or eight median rows ^ weakly keeled; ventrals 



a Proc. Zool. Soc. London, 1899, p. 165. 



6 Must be " seven or nine," as the number of scale rows is uneven. — L. S. 



