effect op land connections 189 



Land Connections of South America 



After the great and isolating submergences of the mid-Cretaceous, the 

 late Cretaceous is a time of uplift, when we had a remarkable set of land 

 connections, North America with Asia and South America, Europe with 

 Africa, and Asia with Australia. The union of western Korth and South 

 America continued into the early Eocene to just about Wasatch time. 

 Then the Central American connection was submerged, and South Amer- 

 ica was isolated until the Upper Miocene at least, the mingling of North 

 and South American faunas taking place in the Pliocene. During this 

 long period of insular conditions South America developed a rich and 

 varied mammalian fauna, one of the most distinctive ever developed on 

 any continent. While South America was isolated. North America and 

 Eurasia (connected from time to time) developed our modern advanced 

 types of mammals. I can find no convincing evidence that during this 

 time South America received any real migration from the outside ; there- 

 fore it is to such types as were already in South America at the beginning 

 of this period that we must look for the progenitors of the marsupials, 

 edentates, litopternas, typotheres, toxodonts, primates, etcetera. 



Comparison of Eocene Faunas of North and South America 



At the beginning of the Eocene, South America was connected with 

 North America and evidently with no other continent. The earliest 

 mammalian fauna in South America is the Notostylops fauna of Eocene 

 age. It includes marsupials, edentates, toxodonts, litopternas, typotheres, 

 and probably primates. To consider these more closely, let us take first 

 the marsupials. 



There are true didelphids in the Notostylops beds and all through to 

 modern times, in South America,, in North America, and in Europe. 

 Another group, generally termed sparassodonts, includes large carnivorous 

 marsupials, ranging in size up to Borhysena, nearly as large as a bear. 

 These are close to Thyalocinus, the Tasmanian wolf, so close that Sinclair 

 put them in the same family. Some have postulated convergence in evo- 

 lution for these forms from the Santa Cruz, Deseado, and Notostylops 

 beds. I can not see that they diverge, as the^i are followed backward. A 

 group of tiny marsupials is today represented by Csenolestes. It has been 

 called a pseudo-diprotodont, supposed to be independently derived from 

 some didelphid ; but the Santa Cruz genera Garzonia and Halmarhippus, 

 the Deseado Pseudo-halmarhippus, and the Notostylops Progarzonia 

 carry the diprotodont characters back to the oldest South American beds. 

 Besides, there are other diprotodonts, such as Palseothenes, Pilchena, and 

 Abderites. It looks to me as if we must recognize that several of the 



