826 DEPARTMENT OF THE INTERIOR 



2 GEORGE V., A. 1912 



that such remains become far more abundant and characteristic in the Tertiary, 

 where they are not infrequently preserved in a very perfect manner. This is 

 eminently true of P. baileyi, Gard., and P. plutonis, Gard., as recorded by 

 Starkie Gardner from the Palady beds of Ireland; or P. macluri, Heer, as 

 recorded by Heer (22;' p. vii.), from the Eocene of Greenland. The same is 

 likewise true of several species which Knowlton records from the Laramie of the 

 Yellowstone National Park (35), and of P. florissanti, Lesq., which Lesquereux 

 described from the Green River group (42: p. 138). Some of these cones show 

 decided relations to -existing species, which is also true of P. Columbiana, but 

 the latter cannot be compared with any of the other fossil cones now known, and 

 it therefore stands wholly by itself. 



The general weight of evidence brought forward by the above analysis, 

 would seem to indicate that while the genus Pinus may extend, into the Creta- 

 ceous, it is essentially a Tertiary type, the chief aspects of which are Eocene, 

 rnd it is to this horizon that P. Columbiana probably belongs. 



The genus Ulmus possesses peculiar significance in the present instance,, 

 not only because there are three well defined new species represented by their 

 wood and one undefined species represented by a fragment of a leaf, but also 

 because the genus as at present known, bears a definite relation to geological 

 age. Ulmophyllum is a well recognized Cretaceous type which is chiefly found 

 in the Potomac Formation, although it is also known to the Upper Cretaceous 

 of Vancouver island (5); but Ulmites and Ulmus are confined to the Tertiary 

 where they range from the Eocene to the Pleistocene, and become identified 

 with existing species. An inspection of present records shows that out of nine- 

 teen Tertiary species, twelve are of Eocene age, while only seven are of Eocene 

 and Miocene age, and that out of these latter only five are strictly Miocene. 

 Erom this we may draw the inference that the genus Ulmus is essentially an 

 Eocene type, and our four species from the Kettle river may also be interpreted 

 in that sense. 



The poorly defined species of Betula from the Kettle river afford very little, 

 if anything, in the way of a reliable basis for age determinations. While the 

 genus Betulites is a well defined Cretaceous one, being especially characteristic 

 of the Dakota group, we nevertheless also find Betula beatriciana, Lesq., in the 

 same horizon (42: p. 36), while B. perantiqua, Dn., occurs in the Upper Cre- 

 taceous of Baynes sound (9), and yet another not specifically defined is met with 

 in the Upper Cretaceous of Vancouver island (8). Knowlton enumerates (37) 

 not less than nineteen, while Ward (56) gives seven Tertiary species out of a 

 total of fifteen. As, furthermore, eighteen out of these twenty-six species are 

 distinctly Eocene, it may be concluded that in the absence of definite evidence 

 to the contrary, any large representation of the genus would give to the flora, 

 facies of a distinctly Eocene character. 



Cyperacites haydenii, Lesq., which occurs in the Kettle River flora, and 

 which was originally described from the Green River group (42: p. 140), serves 

 to definitely indicate the probable age of the flora now under discussion. This- 



