72 EDWARD W. BERRY 



Furthermore if it had originated in the north it should have been 

 a member of that migratory wave of vegetation that swept south- 

 ward along the east coast of North America at about the dawn 

 of the Upper Cretaceous. Our east coast Cretaceous contains a 

 very large flora found in deposits of this age from Marthas 

 Vineyard to Alabama and this flora had a great many elements 

 in common with the Greenland Cretaceous flora. No traces of 

 Cretaceous beeches have, however, been discovered along the 

 Atlantic coast. 



If, on the other hand, Fagus originated in the Asiatic region it 

 •would have had an almost latitudinal pathway with broad and 

 uniform land surfaces both westward to Europe and eastward to 

 western North America, the latter region almost entirely cut off 

 from eastern North America by an epicontinental sea (see map, 

 fig. 2). The presence of petrified beech wood in the Cretaceous 

 of Japan, of Germany and in the Eocene of the Caucasus, as 

 well as leaves in the Cretaceous of Vancouver Island and in the 

 Eocene of Alaska and the early Tertiary of Australia would 

 similarly accord with this hypothesis of its center of radiation 

 better than any other. 



The Eocene records also embrace various North American 

 and European forms which offer no difficulties on such a theory, 

 nor do the numerous succeeding Oligocene and Miocene species 

 of the northern hemisphere. The antipodean records, of which 

 there are a considerable number, seem at first thought to offer 

 greater difficulties. In deposits the age of which is unfortunately 

 somewhat uncertain but which are either Eocene or Oligocene, 

 four species of Fagus are found in southern South America — 

 three in Chile and three in Terra del Fuego and a fifth is found 

 on the edge of the Antarctic continent in Graham Land. They 

 are, in all of these localities, associated with an abundant dis- 

 play of Nothofagus, no less than ten species from Chile, Patagonia, 

 Terra del Fuego and Graham Land having been described by 

 Engelhardt and Dusen. The Tertiary species of Fagus in the 

 Australian region are likewise associated with four or five species 

 of Nothofagus. Another species is found in Tasmania and there 

 are several in the outlying region of New Zealand, one of the 

 latter being common to eastern Australia. 



It would be an attractive hypothesis to consider Fagus as of 



