PEIMATES FEOM MADAGASCAR, 157 



fauna of Madagascar is very much more American than African." The Batrachians and 

 MoUusca again give further evidence tending in the same direction. 



Turning now to the floras of the two regions, we find, according to Baron, that there 

 is a distinct though slight affinity between the plants of Madagascar and those of 

 America. Thus the genus Omvhalea (Euphorbiaceae) includes eight species, of which 

 seven are found in Tropical America and one in Madagascar. Of the genus Pedilanthus 

 (the same order) two species are found in Madagascar, and all the rest (about a dozen) 

 in Tropical America. The well-known Travellers' Tree {Ravenala madagascariensis) 

 finds its representative in Phenacospermum guianense (really a species of Ravenala), 

 which inhabits Northern Brazil and Guiana and is the only other species. No doubt 

 similar instances might be multiplied. 



Once more to quote from Dr. Blanford a concise deduction from the facts bearing on 

 this subject, we may conclude that " the biological evidence of former land-connexion 

 between S. America and Africa is much stronger than that in favour of a belt between 

 Africa, Madagascar, and India, though this latter is supported by geological data. It 

 is probable that the land-barrier across the S. Atlantic, if that was the form of union, 

 lasted to a later geological epoch than that across the Indian Ocean." 



Without attempting to discuss the exact limits at various epochs of this great 

 southern land-mass referred to by different writers under various names, it is sufficient 

 for my present purpose to recognise that such an area existed. 



Now, if we seek to know the nature of the fauna inhabiting this great southern 

 Africo-American continent, it is evident that the only quite satisfactory source of 

 information is the evidence from late Secondary or early Tertiary fossils. But a 

 comparison of allied forms common to two or more isolated regions formerly 

 included within the limits of this continent will furnish valuable inferences as to the 

 probable character of their common ancestry. In Madagascar we have a large area of 

 land which apparently has been almost untouched by lute arrivals from the African 

 continent. 



It is, then, from the study of the existing fauna of Madagascar and its affinities with 

 African and South-American forms on the one hand, and from the examination of the 

 early Tertiary fossils of South America on the other, that we can best form an idea of 

 the character of the forms of animal life existing on this ancient continent. In order 

 to simplify the argument, the only group which we need here consider is the order of 

 the Primates. 



Now in the Santa-Cruz beds, which are probably of Oligocene date, we already 

 have, according to Ameghino, clear evidence of the existence of well-differentiated 

 Simis. These were not merely Ape-like Prosimi^, but four genera of Monkeys closely 

 allied to the extant Cebidse. At the time of the last connexion of Madagascar with 

 Africa (or possibly the eastern part of Africa), which, as we have seen, was probably 

 during middle Tertiary times, the climate of South Africa (including Madagascar) was 



y2 



