322 FOUMEK LAND COXXECTIOXS. 



(1918). The authoress discredited the Tertiary land-routes 

 between America and Africa, but seemed persuaded to beheve in 

 " a limited laud-mass, not trans-oceanic, lying to the north and 

 east of northern South America which supplied rock debris for 

 building u[) the palaeozooics of Brazil, and the oldest rocks of 

 Irinidad." Dr. Maury's special contribution to the evidence was 

 derived from a careful study of the marine shells of Eocene rocks 

 in Trinidad. She found there various very characteristic North 

 American lower Eocene rocks at Pernambuco. The authoress 

 laid great stress on the very close kinship of the North and South 

 American Tertiary life ; and on the other hand minimised the 

 importance of the resemblances between the South American 

 fossils and those of contemporaneous times of the Old World, as 

 reported by many other writers. Her view rests on insufficient 

 foundation, so far as land vertebrates are concerned ; for the 

 facts of mammalian palaeontology point to a long separation of 

 North and South America during Tertiary times, North America 

 exchanging at various times with Eurasia, but contributing 

 nothing to the fauna of South America until the latter portion 

 of the Tertiary period. This paper, through a misreading of the 

 text in Osborn's " Age of Mannnals " (p. 80), states that " Dr. 

 Osborn in 1910 abandoned as a matter of imperfect record the 

 theory of an Antarctic land-connection, even between South 

 America and Australia. . . . He now believes that the greater 

 part of the animals and plants of the southern continent are of 

 northern origin, and that the evidence for Antarctic connections 

 is probabh' explainable through distribution from the north. . . . 

 It was a delight to find that these conclusions of Dr. Osborn, 

 reached from a study of the vertebrates, should so harmonize with 

 my own based on the invertebrates! " But. unhappily, a perusal 

 of the "Age of Mammals" shows that the distinguished American 

 palaeontologist had no such views. The views attributed to him 

 are (or were) those of Dr. W. D. Matthew. Osborn himself quite 

 plainly tells us that " South America appears to have had late 

 Cretaceous or early Eocene connections through Antarctica with 

 Australia " (p. 78). 



(E.) After all this, we read without surprise the views of 

 Messrs. Nichols and Griscom" in their recent paper on the 

 "Freshwater Fishes of the Congo." They conclude that the 

 Characinids and Silurids entered Africa and South America 

 independently from the Northern Hemisphere where they 

 originated, despite the peculiar present-day distribution, and 

 despite the fact that the authors believe in Gondwanaland as an 

 Antarctic continent where certain more ancient groups of fishes 

 took their origin (Dipnoi and Polypterids). This conclusion is 

 based only on indirect evidence, chiefly the fact that the 

 Cyprinidae, which are allies of the above two families, present 

 clear indications of northern origin. Ihe Siluridae, Characinidae 

 and Cyprinidae are held to represent three consecutive waves of 

 invasion from the north, the last wave failing to reach South 

 America and Australia. As for the Cichlidae, they are held to 

 have arisen independently from some ti'opical marine, Acantho- 



