612 F. ^y. kxoavltox — mesozoic floras of the americas 



that it is hardly possil^le to suppose that all or even any considerable per- 

 centage have been incorrectly determined. Therefore, taking the list at 

 its face value, it might well enough have been made of a collection from 

 the Dakota of Kansas or Nebraska. In the 5,000 miles between Kansas 

 and this locality in Argentina no trace of this flora has been reported. 

 Argentine geologists regard these beds as Cenomanian; but, as Berr}^, 

 Halle, and others have suggested, they are probably not older than 

 Turonian. 



Summary 



With the exception of the Dakota flora from Argentina just mentioned, 

 there is comparatively little demonstrable relationship between the Meso- 

 zoic floras of North and South xlmerica. The total known Triassic flora 

 from South America hardly exceeds 30 species, and of these no more than 

 two or three are specifically identical with Xorth American forms, though 

 it is but fair to state that when the floras of the two continents come to 

 be revised in the light of existing knowledge more species will probably 

 be found to be common to the two. As the matter now stands, a majority 

 of the South American species are regarded as endemic, and many of the 

 others are either questionably identified or referred to Old World species. 



Demonstrable specific relationship between the Jurassic floras of North 

 and South America is, if possible, less satisfactory than in regard to the 

 Triassic floras. The Jurassic floras on the South American Continent 

 are so fragmentary and generally unsatisfactory that it is hardly worth 

 while to attempt comparisons. The only flora of importance is the extra- 

 limital one found on Graham Land. This and such as we know from 

 South America are clearly but integral parts of the great world-ranging 

 Jurassic floras. In fact, remoteness appears to have had little influence 

 on distribution. Witness this Graham Land flora, which finds its closest 

 relationship with that of Yorkshire, England, and important relationship 

 with other parts of Europe as well as India, and, it may be added, it is 

 not greatly different from the well known Jurassic floras of California, 

 Oregon, Alaska, and Siberia. 



The relationship between the Upper Cretaceous Dakota group flora of 

 Argentina and the Dakota flora of Texas, Kansas, and Nebraska is direct 

 and positive. The now dominant group of dicotyledons clearly originated 

 in the north, and in late Lower Cretaceous time had spread south over 

 eastern North America and western Europe. By latest Lower Cretaceous 

 and early Upper Cretaceous time they had spread west in North America, 



