Origin of the Flora of South Africa. 109 



equals, if it does not exceed, the others in number of species, and 

 certainly occupies a much greater area. It consists largely of out- 

 liers of the tropical African Flora. Now I may here recall the fact 

 that it is assumed that up to eocene times, tropical Africa was con- 

 nected with India through ^Madagascar ; further it is assumed that 

 the whole of the Soudan was covered by an extension of the Mediterr- 

 anean Sea, and that there was probably a more or less direct land- 

 connection with tropical South America. We need not, therefore, be 

 astonished to find that a large portion of the tropical African Flora, 

 although largely composed of elements peculiar to this part of the 

 world, has close affinities with the Indian and Madagascar flora ; 

 we also find, especially in Western Africa, some genera the other 

 species of which are to be found in the corresponding formations of 

 tropical Africa, and not only some genera but even whole families 

 and tribes are developed only in tropical America, and either West 

 Africa, or in the whole of tropical Africa. Thus we can picture to 

 ourselves how South Africa also received in comparatively recent 

 geological periods elements of the tropical American Flora, just 

 as there was an interchange of animals between these two regions 

 (I need only refer to the Edentates and fresh-water fish). The sup- 

 posed former distribution of land and water also throws light on the 

 question why so few European types of plants have formed an integ- 

 ral part of the Flora of South Africa. While South African plants 

 have migrated far north of the Equator, European plants have not 

 advanced correspondingly far to the south, and there is little note- 

 worthy in this fact, if the existing climatic conditions have persisted 

 through previous geological ages, for the South African genera are 

 characteristic of a warmer climate than the North-Temperate forms 

 and are, therefore, so much better suited for an advance into a tropi- 

 cal region than the North-Temperate plants \vould be. Hooker 

 specially points out the South African character of the Sub-Alpine 

 Flora of Kilimanjaro, but Engler mentions that the boreal elements 

 besides the Mediterranean have many species in the microtherm 

 plant-formations of the high mountains of Africa, while these are 

 relativelv poor in plants originating from African types. A fact, 

 which seems to corroborate the views here brought forward, is this, 

 that none of our high mountains in South Africa contain Alpine 

 plants of the Northern Hemisphere. 



