EINAR LÖNNBERG, AN ATOM Y OF THE RUMNIANTS. 3 



the left side a deep groove into which the processus urethralis 

 fits in. This groove is continued backwards on tlie lower side 

 of the penis by two folds which surround the urethra. The 

 left of these folds is more strongly developed than the right 

 which is thin and less conspicuous. The processus urethralis 

 is comparatively thick and short. It cnrves up from the 

 lower side of the penis on the left side of the cushion from 

 which it almost immediately detaches itself, although this 

 is not conspicuous, because it is sunk into the deep groove 

 already mentioned. The processus urethralis of this species 

 can evidently not be spöken of as a »filiform appendage» as 

 it does not project but IV2 or 2 mm. beyond the tip of the 

 penis and its diameter a t the end is nearly as great. 



Compared with the condition prevailing in the genus 

 Cephalophus it must thus be said that the processus urethralis 

 has become greatly reduced in length in Sylvicapra. But on 



the other hand, the direction of the 



process and the less strong develop- /^ZT^^ ' ''^•'-' 



ment of the terminal cushion appear ^'vll^-^-^ 



to be less modified or less specialised ^-g o. End of penis 



Structures which might be almost di- of Sylvicapra grimmia. 



rectly deduced from the more primi- 

 tive conditions exhibited in such a stage of evolution as is 

 represented by some species of Tragulus.^ It appears to be 

 almost more difficult to assume that a penis shaped like 

 that of Sylvicapra should ever have passed through a stage 

 resembling for instance that of Cephalophus natalensis. If 

 such a view should be correct it would follow that Sylvicapra 

 grimmia should be more distant from Cephalophus than ge- 

 nerally is assumed. But on the other hand such an assump- 

 tion of primiti veness in Sylvicapra is contradicted by the 

 analogous conditions found in Cohus and Redunca as will be 

 described below. In consequence of this I think that such 

 a hypothese cannot be maintained, but even if it is abol- 

 ished as leading too far, the addition of such a morpho- 

 logical characteristic as the one just mentioned gives greatly 

 increased strength to the others which distinguish Sylvicapra 

 and Cephalophus, and it appears to be necessary to hold 

 these two genera apart. 



^ Conf. Lönnberg (5). Pl. II fig. 20. 



