A BRIEF OUTLINE OF THE FA'CTS CONCERNING THE 



COMPOSITION OF THE SNAKE FAUNA OF SOUTH 



AFRICA AND ITS RELATIONSHIP TO THE 



MADAGASCAR FAUNA.* 



By John Hewitt, B.A. 



In the proceedings of the Transvaal Biological Society, February, 

 i9io,t I tried to show that the lizards of South Africa resolve 

 themselves into two sections, ancient and modern, the former 

 including families which are either quite peculiar to the region, 

 or which are well represented in the sub-contiinent, occurring spar- 

 sely or not at all in Northern or West Africa, and the latter 

 composed for the most part of species which are merely the South 

 African representatives of a fauna more strongly developed in the 

 northern parts of Africa ; the older section is closely allied to the 

 lizard fauna of Madagascar, whilst the more recent fauna is not 

 represented in that island. 



Those facts certainly favour the hypothesis that the land 

 masses of Southern Africa (Cape Province of Mr. W. L. Sclater) 

 and Madagascar have been united during a late secondary or early 

 tertiary period, perhaps even forming for a time a large con- 

 tinental island, and afterwards when Madagascar had separated 

 off, the African continent assumed its present shape and South 

 Africa became invaded by a new assembly of lizards which came 

 from the North or West Africa ; there is also evidence that this 

 Ethiopian area was about the same time connected both with 

 South America and with Southern India. In this paper it is 

 intended to show how far the evidence of the snakes will conform 

 to the same theories. 



A cursory examination is sufficient to show that the parallel 

 phenomena are not so obvious amongst the snakes, but neverthe- 

 less the broad facts of the case do not conflict with those relating 

 to the lizard faunas. 



At the present day the families of snakes have on the whole 

 a much wider distribution than the families of lizards, so that 

 there are none peculiar either to Madagascar or to South Africa, 

 and in dealing with the present question it will be necessary to 

 consider the general distribution of the Ophidian families over the 

 earth's surface, which distribution is shown in tabular form as 



* Revised by the author, March l6th, 1911. 



t Ann. Tran. Mus. II. 56. 



