66 Records of the Indian Museum. [Vol, XX, 



{Rec. hid. Mus.^ XIV, 1918, p. 68) been regarded as incorrectly 

 identified. 



Siamese tadpoles, described by Malcolm Smith, on the correct 

 identification of which there can be no doubt, have the beak 

 broadly edged with black ; upper lip with a long, continuous row of 

 teeth, followed b}' a second, broadly interrupted; lower lip with 

 three continuous rows of teeth, or the innermost narrowly inter- 

 rupted. Body considerably flattened; tail nearly twice as long as 

 the body, 4 times as long as deep, tip bluntly pointed, crests 

 rather low, upper a little deeper than lower, not extending on to 

 the back. Olive above, speckled with blackish. 



Habitat. Southern China, Loo Choo Islands and Formosa to 

 the Malay Peninsula and Archipelago (Sumatra, Borneo, Java, 

 Celebes). 



This species varies much individually, but I am unable to 

 find characters by which to define geographical races. Full-grown 

 males (105 millim. from snout to vent) from Tenasserim have an 

 enormously large head, nearly as long as and broader than the 

 body ; the head is also remarkably large in some adult males from 

 Borneo {R. paradoxa, Mocquard), the Loo Choo Islands {R. namiyei, 

 Stejneger), and Formosa. Such males have also a very broad 

 interorbital region, broader than the upper eyelid, and thus con- 

 trast very strikingly with females from the same localities, in which 

 the head is not larger in proportion than in R. limnocharis, and the 

 interorbital region may be even narrower than the upper eyelid. 

 The toes are usually webbed to the terminal discs, except in the 

 specimens from Kuatun and Formosa, and in som^e from Borneo, in 

 which the last phalanx of the fourth is free, or very narrowly bor- 

 dered, the web being more strongly emarginate ; but other speci- 

 mens are intermediate. The fingers vary in length, being very 

 short in the specimens from Kuatun and Formosa : but here again 

 the difference is bridged over by specimens from other localities, as 

 ma^ be seen from the table of measurements. 



The following ' species ' is doubtfully distinct from R. kithlii 

 and I should have been inclined to regard it as identical but for the 

 disposition ascribed to the vomerine teeth. 



Rana khasiana. 



Pyxicephalus kliasiensis, Anders,, Joiirn. As. Soc. Beng. XL, 187 i, p. 23. 

 Rana khasiana, Bouleng., Cat. Batr. Ecaiid. p. 34 (1882). 



" Body short and thick, legs of moderate length. Head short 

 and broad. Snout short and rounded ; no canthus rostralis ; nos- 

 trils directed upwards and backwards, almost on the upper surface 

 and halfwa}' between the eye and the snout. E3'es rather large and 

 prominent. Occiput much swollen. Tympanum invested by the 

 skin, but faintly visible, small, one half the diameter of the eye. 

 Skin in the groin full, smooth throughout, no trace of tubercles j 

 fingers quite free; three small tubercles on the palmar aspect, the 

 inner one the largest, elongated and simulating the shovel-lik e 



