204 MR. G. A. BOULENGER ON THE [Mar. 20, 



spots on the feathers of the upper surface rufous-buff instead of 

 whitish yellow. 



Total lengtli. Culmen. Wing. Tail. Tarsus, 

 in. 



N. mandibularis [ ^ ^^- •• |^'5 

 I S imm.. . 18-0 



N. caledoninus, c? ad 19-5 



N, manillensis 21*0 



66. Sterna longipennis. 



Sterna longipennis, Nordm. ; Salvad. t. c. iii. p. 439. 



a. S ad. Rubiana; 4.3.87. Bill black; legs black; feet 

 black. Food, fish. 



March 20, 1888. 

 Henry Seebohm, Esq., F.Z.S., in the Chair. 



The following papers were read : — 



1. Note on the Classificatiou of the Ranidce. 



By G. A. BoULENGER. 



[Eeceived February 10, 1888.] 



When revising the arrangement of the tailless Batrachians for the 

 second edition of the British Museum Catalogue (1882), I had very 

 great difficulty in dealing with the subdivision of the large family 

 Ranida. I had to abandon the Platy- and Oxydactyle character as 

 a generic one, finding that, to say nothing of the complete inter- 

 gradation which destroys its usefulness as an artificial character, it 

 did not, by itself, express affinity, and that in consequence satisfac- 

 tory series could not be formed upon the degree of dilatation or 

 acumination of the digits. As au instance, I may refer to RancB 

 malaharica, galamensis, macrodactyla, and lateralis, which are 

 unquestionably very intimately related to such forms as constitute 

 the platydactyle group Hylorana of certain authors {R. erythraa, 

 temporalis, chalconota, &c.), and yet are more 'oxydactyle' than 

 RancB macrodon, kuJilii, and many others which have by all authors, 

 and very justly, been regarded as true Ranee. I cannot see that the 

 union of all these species, however numerous, under one genus, 

 Rana, presents any serious inconvenience. On the contrary, the 

 systematist should form well-defined genera, regardless of the 



