SUMMARY 



635 



by the extensive flora of petrified woods found on Antigua, which has 

 several forms common to Central America and the southeastern United 

 States. 



5. In formations correlated with the Aquitanian and Bnrdigalian of 

 the European section, but generally considered as upper Oligocene by 

 American geologists, the tropical types of plants become fewer on the 

 North American mainland and are largely replaced by temperate forms. 



6. The land emergence of the Miocene appears to have developed land 

 connections in Central America and the Antilles reflected in the floras 

 by a general .spreading southward of the temperate flora of North Amer- 

 ica. This emergence resulted in the connection of the Windward Islands 

 with South America, Cuba with Yucatan, and the Jamaica, Haiti, and 

 Porto Rico axis of elevation with Honduras. To this period may be at- 

 tributed the original colonization of the many North American types that 

 still persist in the Antilles and Central America. The invasion of the 

 latter regions, and that beyond in South America, by the oaks (Quercus), 

 walnuts (Juglans), and many 6ther Holarctic types probably occurred at 

 this time, since some of them are found in the Pliocene of Brazil. The 

 radiation of the agaves may also be more appropriately dated from the 

 Miocene rather than from the Pleistocene, as is done by Trelease,^^ since 

 while there was considerable elevation during the Pleistocene the major 

 tectonic lines were- established during the Pliocene by block faulting, as 

 Vaughan has shown, and the failure of Agave to penetrate to any consid- 

 erable extent into South America was due to the impenetrability of the 

 tropical rain forest and not to geographical barriers. 



7. North and South America were connected during the Pleistocene 

 and there was considerable elevation in the Antillean region, resulting in 

 a northward spread from South America of various elements from the 

 rain forests of the Amazon and Orinoco basins, one or two of which occur 

 in the late Pleistocene of southern Florida. Actual land connections 

 among the Antilles or with Florida are not considered probable. 



Note 



There are a number of reasons why the arguments against a land bridge 

 between North and South America based on the evidence of vertebrate 

 paleontology can not be regarded as conclusive. It is opposed by the evi- 

 dence derived from the study of the distribution of other animal types, 

 as has been already pointed out by several students. 



First, as regards actual changes in level. The general thesis ot the 



" William Treleasc : Mem. Natl. Acad. Sci., vol. 11, 1913. 



