HYNOBIIDAE. 461 



above and below; 14 costal grooves ; 5 toes; no dorsal light stripe on 

 body or tail ; light with darker dots. 



Description: "A Hynobius with head depressed, longer than 

 broad; snout rounded, the length equalling interorbital width; 

 palatine series as long as f width of tongue, not broader than long; 

 length of eye \\ times in interorbital width; eye prominent; no labial 

 lobes; body cylindric, scarcely depressed; head length 4| in length 

 of body and head (tail omitted) ; snout to gular fold 3| in distance 

 from gular fold to anus; hind feet laid forward not reaching front feet 

 laid back by more than length of head; fingers and toes well developed; 

 fifth toe \\ in length of fourth; carpal and tarsal tubercles indistinct; 

 tail as long as gular fold to anus; tail keeled above and below, tip 

 acuminate; anus a longitudinal slit, no transverse groove, no raised 

 edges; body smooth, a deep middorsal groove forking on occiput, a 

 longitudinal groove crossing parietal region of head; 14 transverse 

 grooves on the sides reaching belly; paratoids on sides of head; a 

 second gular fold between angles of jaws ; a deep horizontal groove in 

 side of head beginning at gular fold ; body reddish brown above with 

 small brown spots; below immaculate yellowish brown; total length 

 90 mm., head 10, body 40, tail 40." 



Nikolski (1918, f. 47) shows the inner row of vomerine teeth equal- 

 ling the distance from outer edge of naris to median line plus width 

 of naris, and thus like those of H. leechii. PI. IV, f. 4, a photograph of 

 the type, shows it to be not fully transformed, and closely resembling 

 Ranodon sibiricus in habit and markings. 



Remarks: This animal is not very different from 11. leechii save 

 in color. In this character it approaches Ranodon sibiricus rather 

 closely. 



It is particularly unfortunate that no specimens of this species are 

 available because its relations with Ranodon sibiricus are of the great- 

 est interest. The legs are shorter and there are more costal grooves, 

 so that it is not particularly close anatomically, but geographically it 

 is the only Hynobius at present available out of which Ranodon could 

 have been evolved, so that the animal should be examined very care- 

 fully for any similarity. 



Hynobius keyserlingii (Dybowski). 



1853. Salamandra terrestris (?) Zerrener, Erdkunde Gouv. Perm., p. 321. 

 1859. Triton nov. sp., Maack, Putesh. Amur, p. 154. 



