506 PROF. HOWES AND MR. A. M. DAVIES ON THE [DcC. 4, 



of the digits in this animal are more markedly sucker-like than are 

 those of other Amphibians. Each terminates in a well-differentiated 

 cup-shaped extremity ; but notwithstanding this, there is, as before 

 said, no supernumerary phalanx. Looked at at first sight, each of 

 the afore-named expansions, with its circular contour, would appear 

 to differ from that of the typically platydactyle Anura in being 

 uniformly developed around the free border of the phalanx. "When 

 examined in section such is seen not to be the case (Plate XXV. fig. 1 5), 

 for the disk, while truly circular and sucker-like, differs in no impor- 

 tant relationship from that of the Anura. Spelerpes is a crawling 

 animal, inhabiting the walls and recesses of a cave, while the Anura in 

 question are saltatorial and arboreal. We have shown (p. 497) reason 

 for beheving the supernumerary phalanx of the latter to be functional 

 in receiving a direct thrust in saltation, and the discovery of the 

 absence of that structure in Spelerpes is as welcome as it is intelligible, 

 on this view. 



Attention has been drawn (p. 503) to the degraded condition of 

 the syndesmodial pads in Rana temporaria (fig. 16), and such is 

 especially the case with that representing the supernumerary phalanx. 

 The same exceeds the more proximal ones in thickness and bulk, 

 despite its degenerate condition ; and it might therefore appear to 

 represent a degraded vestige of the fully iormed supernumerary 

 phalanx, rather than of the simple syndesmosis. The facts 

 of development do not support this idea, for the terminal pad in 

 this species is, to begin with, a true syndesmosis (indistinguishable 

 from that of the lower forms) which, in its subsequent transformation, 

 exhibits no sign of a phalanx-hke stage. 



The condition of the parts in the Discoglossidcs substantiates, 

 more forcibly than anything hitherto recorded for them, their lowly 

 affinities \ They are, in respect to all structures yet investigated 

 which are of service in unravelling their pedigree, by far the least 

 modified of living Anura. The unmodified condition of the syndes- 

 moses, which, in them, persists for life, is, among the higher forms 

 examined, most nearly realized in Pseudophryne bibronii and Rana 

 arvalis. 



V. GENERAL CONSIDERATIONS. 



The observations herein recorded have an important bearing on 

 questions of general morphology of parts of the appendicular skeleton. 

 On reviewing the literature of the subject, it is impossible to avoid 

 the conclusion that while, of late, too much importance has been 

 attached to sundry gristly fragments in their supposed relationships 

 to variation in structure and to the delusive " archipterygium," there 

 has been a lamentable neglect of the study of those leading facts of 

 development, by which these would-be relationships must, in the end, 

 stand or fall. 



A new interest has been recently awakened in the question by the 

 discoveries of Leboucq (12) and Baur and Gadow (1). The former 



' Cf. this vol. p. 178. 



