58 Records of the Indian Museum. [YOl. XV, 



B. Glandular folds much broken up or absent ; if long, 



fewer in number ; inner metatarsal tubercle ^ to 



I length of inner toe. 



(t. Toes webbed to the tips or at least to the base of the 



last phalanx of the fourth ; tibia 2\ to 2| times 



in length of head and body ; heels not or but 



slightly overlapping. 



Tibio-tarsal articulation reaching the eye or between 



the eye and the nostril ..... var. occipitalis, Gthr. 

 Tibio-tarsal articulation reaching the shoulder, the 



tympanum, or the posterior border of the eye . var. burkilli, Annand. 

 b. Toes incompletely webbed, one or two phalanges 

 of fourth free ; tibia If to 2^ times in length of 

 head and body ; heels strongly overlapping ; 

 tibio-tarsal articulation reaching the eye or 

 between the eye and the nostril .... var. cancrivora, 



Gravh. 



In uniting these different forms under one species, I am simply 

 adhering to the standard adopted in the case of R. esculenta, in which 

 we find the same amount of variation in the shape of the head, in the 

 proportions of the hind limb, in the development of the inner meta- 

 tarsal tubercle and, nearly though not quite, in the extent of the web 

 between the toes ; and as I have not the slightest doubt as to the justi- 

 fication of the course followed in dealing with that highly variable and 

 widely distributed species, of which I have carefully studied a very large 

 material, I feel satisfied that the conclusion adopted in the analogous 

 case of R. tigrina serves best the purposes of exact systematics. It has 

 always been my firm conviction that the multiplication of specific names 

 on differences which break down when put to the test of a large material 

 is not conducive to an advance in our knowledge, whilst the recognition 

 of forms to which subordinate rank is assigned fulfils all requirements 

 and leads to a truer appreciation of the state of things in Nature. 



It is, however, with diffidence and provisionally that I include R. 

 cmicrivora among the varieties of R. tigrina. 



I have not seen examples of van Kampen's R. angustopalmata, 

 from Macassar, but if its tadpole is practically identical with that of 

 R. limnocharis, as he states, may it not be a distinct species ? As to 

 the tadpoles described from Java, is a confusion with R. limnocliaris 

 absolutely out of question ? Dr. van Kampen himself, when alluding 

 to Flower's identification of Siamese tadpoles, regarded it as almost 

 incredible that the Malay frog, so difficult to distinguish from the Burmo- 

 Siamese, should differ to that extent in the larval condition. I there- 

 fore believe the question of the specific rank of R. cancrivora should 

 remain open until Dr. van Kampen adduces further proof of the correct- 

 ness of his identification of the Javan tadpoles. 



I hope I may be pardoned for raising these doubts, in view of the 

 fact that, even in so geographically remote a form as R. occipitalis, the 

 Very striking buccal characters of the tadpole of R. tigrina have 

 remained unchanged. 



If, however, it should be established beyond doubt that R. cancrivora 

 passes through a larval stage so difterent from that of R. tigrina, I would 

 then unhesitatingly endorse Dr. Annandale's conclusion as to the speci- 

 fic distinction. 



