52 Records of the Indian Museuvi. [Vol. XV, 



The differences between these two forms are quite as great as between 

 the typical R. esculenta and the var. lessonae, and, to judge from the 

 rather scanty material at my disposal, there is not the same overlap. 



As regards the distribution, although both forms appear to occur 

 together in some localities (Benares, Malabar, Ceylon), it does not seem 

 to be so generally, and I was assured some years ago by Dr. Henderson 

 that the var. crassa is the only one found near Madras town, where its 

 fossorial habits distinguish it so sharply from the true R. tigrina as to 

 have raised doubts in his mind as to the propriety of uniting both under 

 the same specific name. 



From the following table of measurements it will be seen that the 

 width of the head may considerably exceed its length in both the typical 

 form and the variety. It has been stated that " when the foot is 

 stretched out the margin of the web is slightly convex ^ between the 

 fourth and fifth toes." If R. crassa is to be included in R. tigrina, this 

 statement requires modification, as Peters in his description of Hoylo- 

 hatrachus ceylanicus ascribes to it a rather deeply emarginate web, as is 

 confirmed by a few of the specimens in the British Museum. 



I have another correction to make to Dr. Annandale's definition 

 of R. tigrina. The granular nature of the skin in some specimens may 

 extend to the back of the head, as far as the eyes (Benares, Ceylon). 

 Narrow, interrupted, but well defined glandular folds, 6 to 14 in number, 

 are nearly always present on the back, and their number and regularity 

 constitute a fairly good though not absolutely constant character for 

 distinguishing the typical form and the var. crassa from the other 

 varieties. 



Some specimen of the var. crassa (Benares, Malabar) have large 

 black spots on the gular region. 



There is often a narrow light vertebral streak or fine line, which ma}^ 

 be accompanied by another along the calf, as in the type figured by 

 Daudin ; a broad vertebral band, as in the var. cancrivora, I have never 

 seen. 



Rana rugulosa, Annand. nee Wiegm. 



Wiegmann's figures of R. rugulosa and R. vittigera are excellent 

 and may be relied upon. They demonstrate that these two supposed 

 species, founded on the coloration, are identical, even in a racial sense, 

 and as both show a decidedly pointed snout, the tibia half the length of 

 head and body, and the web between the toes strongly emarginate and 

 not reaching beyond the penultimate phalanx, they answer the defini- 

 tion of R. cancrivora and not that of Annandale's R. rugulosa. 



The name R. burkilli, Annand., should therefore be revived for the 

 form, from Burma, Siam, and China, which differs from R. tigrina, 

 s. sir., in the generally shorter hind limb, the length of the tibia being 

 contained 2^ to 2| times in that of head and body, the heels not or but 

 slightly overlapping, and the tibio-tarsal articulation reaching the 

 shoulder, the tympanum, or the posterior border of the eye. The fourth 



^ No doubt a lapsiis for ' concave'. 



