356 Transactions of the South African Philosojjhical Society. 



related to some found in South Africa (Ameghino, 1906, p. 282) and 

 the absence of these forms in the Argentines and Patagonia seem, 

 besides the dissimilarity in the fresh-water fauna, to indicate a 

 connection of South Africa with Tropical America in Triassic times, 

 which was already somewhat indirect by a further northwards carv- 

 ing out of the South Atlantic, and thus even for Jurassic times 

 Neumayer's map previously referred to is hardly quite correct, as 

 it indicates land between the whole of Africa and South America. 

 However, in Cretaceous times only a bridge could have existed 

 which became narrower and narrower until, in Miocene times, as 

 Ameghino has pointed out, it was only represented by a chain of 

 islands. 



We may, therefore, now take it for granted that, at all events 

 until Eocene times, an interchange of plants between Tropical South 

 America and South Africa, by way of Tropical West Africa, w^as 

 possible, and that thus some curious phytogeographical relations 

 become intelligible. 



There is, as far as I can judge, nothing known about the truly 

 indigenous Flora of St. Helena which could be used as evidence 

 against the view here advocated. St. Helena must have been a link 

 in the chain between South Africa and East Tropical America. The 

 affinities of its Flora are essentially African, and while we find that 

 the presence of the genera Phylica, Pelargonium, Mesembrianthe- 

 mitm, Osteospermumj and Wahlenbergia, which are eminently 

 characteristic of extra-Tropical South Africa (though with the 

 exception of Phylica not of the " Cape Province ") indicate a South 

 African connection. Bentham (1873, p. 563) has pointed out " that 

 those composite denizens [of St. Helena], which bear evidence of the 

 greatest antiquity have their affinities for the most part in South 

 America." 



Unfortunately the Flora of the Mauritius and the Seychelles is of 

 very little use in throwing light on the problem before us, since the 

 native Flora of Mauritius especially, like that of St. Helena, only 

 represents the wreck of its constituents as compared with its 

 composition before the island was occupied by Europeans. The 

 obvious relations of the orchids of these islands to those of South 

 Africa, as well as other similar relations, do not necessarily demand 

 a former land-connection with South Africa. 



I regret that I have been unable to make a comparative study of 

 the Flora of Madagascar, which to a large extent might yield the 

 proof that the views here advocated are in the main correct. 

 Palacky (1905), who has recently dealt with it, has only published a 

 scrappy sketch of his ideas as to the genesis of the Flora of Africa, 



