GENERAL CONCLUSIONS. 47 



distinctly in the Eocene by a number of species of Cissus and Vitis, one 

 of which is recognized in the Lower Miocene of Carbon, and by a hue 

 Ampelopsis scarcely distinct from A. qidnquefolia, in the Upper Miocene of 

 Colorado. The fflianmacece, already in the Cretaceous in one species, be- 

 come predominant in the Eocene of the Territories with Berchemia leaves, 

 which, though described under a proper specific name, cannot be posi- 

 tively distinguished from B. volubilis. Of the following orders in the 

 vegetable series, the Tertiary has especially species of Celastrus, Ceanoihus, 

 and Sapinclns, this last in abundance mostly from the Miocene, with Acer, 

 Negundo, anil Staphylea. The Miocene species of the last genus is hardly 

 separable bom ,S'. trifoliata. The Leguminosem and the Rosacece are little 

 known, and the few forms described are not as yet comparable to those 

 of the present time. The first order has in our present vegetation mostly 

 herbaceous plants. In the second we have a Spircea in the flora of Alaska 

 and another in that of Florissant, Colorado. A Crataegus is also present in 

 the Eocene of Golden. I have described as Hamamelites some Cretaceous 

 leaves considered by Saporta as related to Hamamelis ; we have, however, 

 no leaves in the Tertiary which might by relation of types authorize this 

 reference. But the Araliacece are positively Cretaceous. Species of Aralia 

 described from the Dakota group are reproduced in close conformity 

 of types in the Upper Eocene of Evanston, and especially in the Plio- 

 cene of California. Comparing, for instance, Aralia qidnquepartita of the 

 Cretaceous Flora (PI. XV. Fig. 6), and A. Towneri (PI. IV. Fig. 1) of Dr. F. V. 

 Hayden's Annual Report of 1874, with Figs. 4 and 5, PI. V.. of this memoir, 

 the likeness will certainly appear striking. The fine leaf of A. Saportana, 

 also, with its shorter lobe and fan-like form, is comparable to .1. Wliitnn/i, 

 while the present forms of Aralia with serrate lobes have a more distant 

 affinity to a new species with crenate lobes recently sent from the Creta- 

 ceous of Colorado. This one is quite near to A. furmosa, Heer, of Moletin, 

 perhaps identical with it. 1 The Cornaccw have numerous species of Cormis 

 in the Eocene and two in the Pliocene of California, while Ngssa is by 

 leaves and fruits at Evanston. Viburnum represents the CaprifoUaceoe by a, 

 large number of leaves of different species of the Eocene. Their charac- 

 ters refer them as intermediate to V. dentatum and I', fantanoides, and one 

 of them to V. ellipticum of Oregon. Professor Newberry describes in his 

 Ancient Floras two species from the Fort Union group. We have none as 



1 //< il< ra, also, the well-known Ivy introduced from Europe, is of Cretaceous origin on ihi> continent. 



