A Revision of the Genus Clematis of the United States. 133 
events a form found in Napaul, C. grata, closely resembles Virginiana.* 
This alone shows the close relationship existing between the two 
species under consideration.. 
_- Having now given the general geographical distribution of the 
species and varieties of the Genus Clematis, let us recapitulate and 
see to what sources we can refer the species. As I have elsewhere 
shown,t we must probably look to the north for the place of origin of 
many of our species of plants, and we will find in the glacial theory 
the principal factor for their dispersion. So, too, we must look back 
into the past, to the Cretaceous, or at least the Tertiary Epoch, for the 
time when they first made their appearance. But here we can receive 
no assistance. We have no data to go by, for though the ancestors of 
some of our trees have been found there, there are no known remains 
of Clematis, or, indeed, of any of the Ranunculacee, from the formations 
of the western United States. Of some of them we may be sure. C. 
alpina, with its variety Ochotensis, has undoubtedly come from Asia 
and the north along the highway of the Rocky Mountains. Probably 
this has also been the case with C. verticillaris, now comparatively 
fare, and only found in northern stations. The C. Virginiana and C. 
ligustictfolia, we may certainly regard as the descendants of one form, 
which lived at the north ; while the former came south to the eastward, 
the latter went south on the west and there developed a little differently 
because of a difference in climate. Probable the C. grata, Wall., of 
India, and the C. vitalba, L., of Europe, are derived from the same 
stock as the Virginiuna and ligusticifolia. ‘The C. Viorna, another 
widely dispersed species, and a marked one, has a near relative in the 
C, Japonica, Thunb., of Japan:{ so that probably these two also have 
descended from.a common parent formerly living at the north. Nuttall 
says his C. lasiantha is allied to C. ordentale of Siberia and other parts 
of Asia. It has been refered to C. Peruviana, and Torrey considers 
C. paucifiora, which is also strictly Californian, a variety of lastantha. 
Here we have six species which we seem justified in referring to an 
origin in northern North America and in Asia. Some of the other 
species, such as C. crispa, and Baldwinii, seem to be southern in their 
affinities, and probably have their nearest relatives in the West Indies 
and South America; while C. Bigelovii, Drummondii, and Douglasii, 
seem to have their closest relatives in the south and west. 
* Loudon, Trees and Shrubs of Great Britain, p. 7. 
+ Geographical Dist. of Indig. Plants common to Europe and the N. E., U.S, In Jour. 
Cin. Soc. Nat. Hist., April, 1881. 
t Japan Exped. under Perry, vol. ii., p. 306. 
