1888.] CLASSIFICATION OF THK RANH)/E. 205 



number of species they may embrace ; by so doing he firstly facili- 

 tates identification, for the "student has a right to expect, when using 

 a synoptic work, to get at the name of the genus before that of the 

 species ; and secondly, he more correctly expresses the continuity 

 and breaks in the series of forms as exist in Nature at the pre-eut 

 period. This mode of treatment has therefore both a practical and 

 a philosophical bearing. 



My arrangement has not met with general acceptance. As for 

 myself, I have not lost sight of the question during the six years 

 that have elapsed since the publication of my classification, and I 

 have, on different occasions, taken up the matter again in the hope 

 of finding characters upon which to subdivide the genus Rana, 

 but without success ; and I am now more than ever convinced that 

 it is a natural association. This conviction has been confirmed 

 by a discovery published by Peters (Reise n. Mossamb. iii. 18S2;, 

 shortly after the issue of the Brit'sh Museum Catalogue. He found 

 that the digits of most of his Pohjpedatiiue differ from those of the 

 Ranince by the presence of a small additional phalanx between the 

 uliiinate and what is normally the penultimate; the number of 

 phalanges being 3, 3, 4, 4 in the fore limb, and 3, 3, 4, 5, 4 in 

 the hind limb, instead of 2, 2, 3, 3 and 2, 2, 3, 4, 3. After testing 

 the constancy of this character, I fully endorse Peters's view as to 

 its taxonomic importance ; it affords a far better character for separa- 

 ting Rhacophonis from Rana than does the presence of a web 

 between the fingers. And I find, with satisfaction, that all the 

 species referred by me, from autoptic examination, to the genus 

 Rana have the normal phalanges, irrespective of the presence or 

 absence or size of the digital expansions. Two species which were 

 formerly unknown to me, but of which specimens are now in the 

 Museum, y\z.Hyla bucrgeri,Schleg., and Theloderma leprosum,Tsch. 

 { = Polypedates leprosus,Gt\\r.), must, in spite of their free fingers, be 

 re.^erred to Rhacophorus. I also find that Cassina, though oxydac- 

 tyle, and therefore placed by Peters in his RanirKe, has the additional 

 phalanx like its close ally Hylambates ; and that the genus Ivalus, 

 as hitherto defined, is unnatural, the species opisthorhodus, Gthr., 

 silvaticus, Blgr., fusciis, Blgr., saxicola, Jerd., and doubtless also 

 sarasinorum, F. MiilL, all from Southern India and Ceylon, 

 standing in the same relation to Rana as the typical Ixali to 

 Rhacophorus. For these species, characterized by the normal num- 

 ber of phalanges, I propose the generic name Micrixalus. Con- 

 sidering the importance of the character discovered by Peters, I 

 would suggest the division of the Ranidce into two groups, that 

 which is characterized by the additional phalanx embracing the 

 following genera : — 



Cassina, Gir., Hylambates, A. Dum., Rappia, Gthr., Meyalixa- 

 lus, Gthr., Rhacophorus, Kuhl, Chiromantis, Ptrs., Lvalus, Tsch., 

 and Nyctbcalus, Blgr. 



The following figures show that the character upon which these 

 two groups are based is readily ascertainable. 



Proc. Zool. Soc— 1888. No. XV. 15 



