Ill 



rrom this we may at once eliminate all generic forms and the two species of general 

 distribution. We may also exclude Hypnum because of its doubful character. Castanea 

 castanecefolia and Fagus feronice have already been found to be of Oligocene age. There 

 therefore remain only seven well recognized species which can aid in establishing the strati- 

 graphical relations. 



Betula prisca ? is unknown to any other locality in Canada, but in the United States it 

 is a characteristic feature of the flora of Fort Union, Montana, which is the only locality so far 

 known. In Alaska it is found in the Eocene of Port Graham, and it also occurs in the lower 

 Eocene of Spitzbergen and Saghalien. From this evidence it must be regarded as a well 

 defined Laramie species. 



Acir grossedentatum is unique in its occurrence at Quesnel. It is, however, known to the 

 Miocene of Europe as described by Heer. 



Quercus pseudocastanea is unknown in the United States, but it occurs in the Eocene ol Port 

 Graham, Alaska, in the Atane beds of Greenland and Saghalien. 



Nordenskioldia borealis has been recorded from the Mackenzie river in Canada. Its 

 occurrence in the Dakota group of the United States gives it a decidedly Cretaceous aspect 

 which is not diminished by its presence in any other horizon. 



Platanus aceroides, also from the Tertiary of the Mackenzie river, is found in the United 

 States as a common element of the Fort Union group ; but in one instance Knowlton records 

 it, with hesitation, from Van Home's ranch in the John Day basin, Oregon, a formation 

 which he regards as Upper Miocene. 



Dombeyopsis islandica is not known to the Tertiary of the United States, but it should 

 be noted that the genus as represented there, by four species, is characterized by being 

 exclusively Laramie. 



Fagus antipofi occurs in the Eocene of Port Graham, Alaska, and the Laramie of 

 "Wyoming; but according to Knowlton it is also found in the Miocene of California. It is, 

 however, known to the Lignite Tertiary of Greenland and Saghalien. 



A review of these facts shows the following distribution : — 



Cretaceous and Lignite Tertiary ... - . . i 



Laramie or Lignite Tertiary 3 



Eocene to Miocene, chiefly Laramie 1 



Laramie to Miocene, chiefly Eocene 1 



Upper Eocene to Lower Miocene (Oligocene) 1 



Laramie to Pliocene (Oligocene) 1 



General distribution - - 2 



From this analysis it is clear that the beds at Quesnel must be regarded as of Eocene 

 age, with a strong tendency to Laramie. 



