372 Mr. G. A. Boulenger on the Batracldan 



XLV. — Remarks on the Batrachian Genera Cornufer, Tschudi, 

 Platv mantis, Gthr., Simomantis, g. n., and Staurois, Cope. 

 By G. A. Boulenger, F.R.S. 



(Published by permission of the Trustees of the British Museum.) 



In the March number of these 'Annals' I have pointed out 

 some differences in the structure of the terminal discs of the 

 digits in the genus Rana, and endeavoured by their applica- 

 tion to the Papuan and Melanesia!! species to ensure a clearer 

 definition of these frogs and a grouping more in accordance 

 with their natural affinities. In proposing the subgenus 

 DUcodeles for certain Indian and Melanesian frogs, the true 

 relationships of which I had failed to grasp before, I stated 

 that it leads to Cornufer, a genus previously defined as 

 differing from Rana in the free or very feebly webbed toes 

 and the complete union of the outer metatarsals by the 

 thickened integument, the web not penetrating between 

 them. In view of the state of things in the species grouped 

 under Discodeles, the definition is no longer sufficient, and I 

 have had to consider whether Cornufer should be treated as 

 a mere subgenus of Rana, or whether other characters, 

 justifying a generic separation, could be discovered. 



I have come to the conclusion, after examining a large 

 material, that, although unquestionably connected with Disco- 

 deles, these frogs can be sharply separated from Rana by 

 reverting to the view of Gunther (1858), that the species 

 with large digital discs {Cornufer, Tschudi, = Halophihts, 

 Girard) should be kept distinct from those with small discs 

 and practically free toes (Platy mantis, Gthr.). 



The discs of Platymantis do not differ from those of Rana 

 buf uniform is, opisthudon, and guppyi, from which group the 

 genus may be derived, whilst those of Cornufer, though 

 agreeing in shape with those of Rana beddutnii and its Indian 

 allies, present this notable feature, that a transverse groove, 

 extending across the lower surface and corresponding in 

 position with the horizontal limb of the T-shaped terminal 

 phalanx, is continuous with the crescentic or horseshoe- 

 shaped groove between the upper and the lower surface, thus 

 defining a hemispherical area within the disc, a feature which 

 is foreshadowed in the species of Rana (subgenus Hylorana) 

 separated by Cope under the name of Amoiops, in which a 

 more or less distinct transverse ridge or groove also corre- 

 sponds to the very long horizontal limb of the terminal 

 phalanx without actually joining the marginal groove. There 

 is, I feel sure, no direct genetic connection between these 

 frogs and Cornufer, as proved by the osteological characters 



