222 THE MALAGASY REGION. [CHAP. 



mentioned in an earlier chapter 1 , where it was attempted to show 

 that although these instances of discontinuous distribution might 

 be explained by parallel migration from a common northern 

 centre, yet that the Tertiary mammalian evidence indicated that 

 the American forms had reached their present habitat by way of 

 Madagascar and Africa. It will suffice to add here that giant 

 land-tortoises, which existed in the Mascarenes during the present 

 epoch, are represented by extinct species from the superficial 

 deposits of Madagascar; and that the latter have also yielded 

 remains of gigantic flightless birds (dEpyornis) markedly distinct 

 from any other known type. And here it may be mentioned that 

 the chamseleons (Chamaleontida) present a certain similarity in 

 their distribution to the lemuroids, the Malagasy region including 

 23 out of the 49 species, while nearly all the others are Ethiopian. 

 As a whole, the Malagasy reptiles, with the exception of the 

 snakes, are stated to be more nearly allied to those of the main- 

 land than are either the mammals or the birds ; but the 

 amphibians exhibit more decided traces of Oriental affinities. 



Concentrating our attention mainly on the mammals alone, 

 their distinctness from those of all other parts of 



Relations of .... 



Madagascar to the world are quite sufficient to indicate the right of 



the Mainland. _ _ , r -\ r ~\ 



Madagascar to form the centre of a separate zoologi- 

 cal region. In the survey of the lower Oligocene fauna of Europe 

 it has been shown that both lemuroids and civet-like carnivores 

 were common, one of the latter having been referred to the existing 

 genus Viverra. Hence it is probable that to this fauna we must 

 look for the ancestors of the Malagasy mammals. The only 

 lemuroids closely allied to those of Madagascar are the African 

 galagos, and as the civet-family ( Viverridcz] is better represented 

 in Africa than elsewhere, it may be taken for granted that Mada- 

 gascar received its mammalian fauna from the mainland. Putting 

 aside the hippopotamus and bush-pig, which doubtless arrived 

 later, the Malagasy fauna can, however, have been derived from 

 Africa only at a time anterior to the introduction of the modern 

 types of ungulates into that continent, when it was chiefly popu- 

 lated by lemuroids and civet-like carnivores 2 . The question then 



1 Supra, p. 131. 



' 2 It is of course probable that some of the Oligocene primitive ungulates 



