DISCONTINUOUS DISTRIBUTION OF MAfMMALS, 687 



We see thus that Madagascar has, of the three groups of 

 land Carnivores, only succeeded in giving us the Cat group. The 

 environment in Madagascar has not either the extent or the 

 variety of that of Australia, and we can reasonably assume that 

 the process of differentiation has been much slower in the former 

 than in the latter. 



Rodents. — For our purpose this is a most interesting group. 

 Their linkage is still dubious, although the fossil Tillodontia 

 possess a complex of characters, some of which are also found 

 in Rodents, whilst others are to be found again in the Ungulata. 

 Huxley (says Beddard) used to insist on certain likenesses shown 

 by such apparently diverse animals as the rabbit and the elephant. 

 It is not said what these characters are, but anyone who has 

 before him the skull of a Springha<as (Pedetes), a dassie 

 (Hyrax), and an antelope will find such similarities. It will be 

 noticed that the palatine and pterygoid are sutured to each other, 

 and to the pre- and basi-sphenoid in such a way that in the 

 group with Creodont ancestry the roof of the mouth is com- 

 pletely separated from the orbit. In the above animals, however, 

 there remains a longitudinal slit between these bones through 

 which the obit of the skull communicates with the mouth cavity. 

 Also, if attention is fixed on a point in front of the articulation of 

 the lower jaw, just where the squamosal meets the alisphenoid. a 

 projection will be noticed in the skull of the antelope, and this 

 projection points straight forwards to a backwardly-directed pro- 

 cess of the palatine, starting from the outer edge of the last 

 grinder. This incipient bar of bone is complete in the skulls 

 of the dassie, of the springhaas, of the vlaktehaas, and is com- 

 pletely absent in Creodont-Eutheria, and also' in the horse. 



The Rodent-like appearance of the dassie seems thus to be 

 more than skin-deep ; the affinities of the Duplicidentate Rodents 

 to Ungulates, indicated by the fossil Tillodontia, seems to be 

 borne out by similarities, apparently not of an adaptive nature, 

 between those Rodents and the archaic Ungulate, Hyrax. 



On the other hand the Simplicidentate Rodents are 

 apparently to be linked to the Creodonts. 



One of the oldest groups of Rodents is said to be the Ham- 

 sters (Sigmodontincc) , which is represented by peculiar forms 

 in all three of our Creodont regions. In Madagascar it is the 

 only group of Rodents. In America it is represented by a num- 

 ber of genera. One particular one (Ichthyojuys) of Peru has a 

 skull said to show likenesses to that of the Australian Hydro niys. 

 Now. when we bear in mind the tendency of small marsupials to 

 lose the marsupium and the inflection of the lower jaw, can we 

 not consider the probability of the Australian mice being mar- 

 supial in origin ? 



Another interesting group of Rodents is the family Octo- 

 dontidae. This family has representatives both in Africa and in 

 America; and Lyddeker ("Royal Natural History" and "Harms- 

 worth Natural History ") does not separate the African and 

 American generi as Beddard ("Cambridge Natural History") 



H 



