H>08.J BATRAOHIAN RHINODERMA DARWINI. 679 



their object was not to enter into the general anatomy of 

 Rhinoderma, but only to describe structures associated with the 

 breeding-pouches, there is naturally some room left for a fuller 

 account of this Engystomatid frog, which I specially compare with 

 its African relative the genus Breviceps, dissected by myself some 

 time since and referred to in a paper communicated to this 

 Society lately*. Rhinoderma is an Engystomatid frog without a 

 narrow mouth. The mouth is not far from being of the ordinary 

 Batrachian capacity. The general aspect also of the species is 

 widely removed from that of Breviceps, with which it would 

 certainly not be associated were external characters alone taken 

 into consideration. Externally, in fact, Rhinoderinia is a typical 

 frog, except indeed for the considerable projection of the upper 

 beyond the lower jaw, and the presence of a narrow projection of 

 the snout region of the head which has nothing to do with the 

 nostrils. This process is not always present, but its presence or 

 absence has nothing to do with sex, for I found the process well 

 developed in one female, and absent in another female, both speci- 

 mens being of about the same size. I have not, however, examined 

 with care a sufficient number of examples to enable me to 

 contribute aiiy material facts with reference to the development 

 ■of this " snout " t. 



§ Visceral Anatomy. 



The viscera show a number of differences from those of 

 Breviceps. 



The liver in Rhinoderma has not the peculiar form and relations 

 to the heart which I have described in Breviceps. It is consti- 

 tuted more on the plan of that of Rana. That is, there is a 

 primary division into light and left lobes of which the right is 

 much the larger and is again divided into two lobes. The whole 

 mass of the liver lies entirely behind the heart, which is not in 

 the least hidden by it in the remarkable fashion which I have 

 noticed in Breviceps. That this is the case is shown by the 

 attachment to the posterior border of the pericardium of a 

 peculiar muscle, which passes from the body-wall under the lobes 

 of the liver without being attached to them and ends on the 

 pericardium X. 



The alimentary tract is proportionately and roughly of the same 

 general appearance and length as that of Rana. There are, how- 

 ever, some differences, especially when the gut is slit up aiid the 

 characters of the lining membrane in different regions compared 

 with each other and with the cori'esiDonding or apparently corres- 

 ponding sections of the gut of Rana. The stomach itself has a 

 tendency to be more spherical in shape than in Rana. The part 

 which ensues and corresponds in its position to the duodenum of 



* P. Z. S. 1908, p. 11. 



+ The skull, showing the cartilaginous hasis of this process, is figured by W. K. 

 Parker (Phil. Traus. 1881, pi. 39, figs. i-iv). 

 X This is described below (on p. 684). 



