FORMER LAND CONNECTIONS. 317 



which increased in the Cretaceovis period, cutting off all direct 

 connection between Africa and Australia; and, further, the 

 Madagascar-Indian section became separated from Africa. In 

 Cretaceous times, however, the Afro- American mass was still 

 existing, although a breach was already developing in its northena. 

 portion. Ultimately, the Houth American and African coasts 

 became completely separated towards ilie end of the Cretaceous. 

 During the Tertiary period, whilst the several sections of Gond- 

 wanaland were moving rapidly apart, the elevated encircling 

 folds, concurrently produced, brought new land connections into 

 existence. The chief of these was one in Miocene times, between 

 South America, Australia and New Zealand. There are some 

 indications of another connection in the region of the line between 

 V,enezuela and Morocco, but there are no geological data to deter- 

 mine whether such connection took the form of an isthmus, or 

 of a chain of islands of which the present day Cape Verde and 

 Canary Islands are persistent relics. In Miocene times also, 

 Northern Africa became linked to Asia and Europe by the" 

 emergence of the Iranian-Himalayan fold ranges, and Madagascar 

 probably had temporally reunion with Africa. 



Now, the geological data are certainly incomplete in respect 

 to the periods which chiefly concern us here, namely, from the 

 Jurassic to the Tertiary, and, as Du Toit says, some of the views 

 expressed may be speculative, but zoologists have the satisfaction 

 of knowing that the main outlines of the theory rest on palaeonto- 

 logical and geological considerations. 



Another able geologist, Prof. E. H. L. Schwarz, has laid 

 stress on the magnitude of the marine invasion during Cretaceous 

 times, even to the extent of almost obliterating the land-surfaces 

 of our continent, although this extreme view does not seem 

 generally acceptable to geologists. Dr. Eogers says that " the 

 area north of latitude 30° has not been under the sea since the 

 remote period when animals and plants possessing stiiictures 

 preservable in sediments began to live." 



A Recent Statement of Evidence submitted in support of an 

 Ancient Land Connection between Africa and South America.* 



The faunas of Africa and South America are on the whole 

 essentially dissimilar, but in a few groups of animals there are 

 some remarkable instances of afhnity. The most striking case 

 is that of the fresh-water fishes specified by Mr. C. T. Eegan^ as 

 follows: — " South America has a very rich and varied fresh-water 

 fish fauna : with the exception of the Osteoglossidae, a generalised 

 and ancient group represented at the present day by a few 

 remnants, it has not a single family in common with either North 

 America or with Australia. On the other hand three South 

 American families, Lepidosirenidae, Characidae [including our 

 Tiger-fish], and Ciehlidae [including the small fish known in the 

 Transvaal as Kurpers], occur also in Africa, and the South 

 American cat-fishes of the family Pimelodidae are clearh^ related 

 to the Bagridae of Africa and India. If South America and 



* See paper (9) of biMiography. 



