684 DISCONTINUOUS DISTRIBUTION OF MAMMALS. 



that the process of differentiation has not followed exactly 

 similar lines in Australia and America, and we can therefore 

 expect that in Africa the process will also have been charac- 

 teristic. 'But on such a view we must accept that the Creodont 

 stock which populated Australia and America would also have 

 emigrated to Africa; and we must be able to find traces of this 

 emigration. Now Africa has, near its Southern end, an island 

 whose relation to the continent is similar to that of Australia to 

 Asia. Has this island any mammals which may be compared 

 to the Australian marsupials? I venture to affirm that it does 

 possess such mammals in the Lemurs. This opinion I base on 

 the following features : — 



1. The Galagininge, the members of which are generalised 

 in structure, and which is the only group occurring in both 

 Africa and Madagascar, must be considered to be nearest the 

 ancestral Lemur. In this group the palate is imperfectly ossi- 

 fied, as it also is in Insectivores and Marsupials. 



2. In Creodonts the tail was long, powerful, and therefore 

 possibly prehensile. That it was prehensile is indicated by its 

 possession by numerous living forms usually labelled Creodont- 

 like, e.g., Arctitis, Cercolcptis, Aelurus, Ptilocercus, Capromys, 

 etc. Both the Marsupials (phalangers and Cuscuses) and the 

 Lemurs (Galagininae and others) are still so generalised as to 

 retain this prehensile tail. 



3. The marsupials being apparently descended from placental 

 mammals, must originally have had only a rudimentary mar- 

 supium. The marsupials have developed this structure into a 

 pouch, and of all the Eutherian mammals the Lemur catta has 

 the marsupium best developed. 



4. Of the Creodont mammals, i.e., those — namely, the Insec- 

 tivora, Carnivora, and Lemuroidea — descended from the Creo- 

 donts, projecting incisor teeth are found only in Marsupials^ 

 Lemurs, and the New World monkeys. 



5. The forms originated by the two groups are comparable. 

 It is a deeply significant fact that the marsupials have given 

 rise to forms analogous with a very large number of Creodont- 

 like Eutheria, whereas the Ungulates and Primates are totally un" 

 represented. The former group, judging from their high degree 

 of specialisation in very early periods, did not originate from 

 the Creodonts. although it is possible that the two groups are 

 descended from similar ancestors. But if they and Creodonts 

 are ultimately monophyletic, we have to explain how the Creo- 

 donts lost the potentiality for producing the Ungulate- form a 

 second time. The ordinarily used explanation would be that the 

 Creodonts were too much specialised in a direction different to 

 that oi the Ungulates, but such an answer does not atteinpt to 

 explain. 



The case of Primates is more anomalous, since they are 

 u«inlly considered to be descendants of Lemuroid ancestors ; 

 so that if the primitive marsupial was similar to the primitive 

 Lprniir, we should exnect to find both producine Primates. 

 A ^-^ in St this we mav answer that the Lemurs in Madagascar 



