236 EVOLUTION OF PLANTS 
ward retreat of vegetation before the advancing ice- 
sheet, and as the latter retired again the plants were 
enabled to return northward. ~ 
In Europe, owing to the position of the great moun- 
tain chains, as well as its higher latitude, the whole 
region north of the Alps was subjected to the action of 
the glaciers, and the southward retreat of the plants being 
cut off, very many forms perished, while the same plants 
have survived in the more favored regions of Asia and 
America, in both of which a far larger number of sur- 
vivors of the primordial Tertiary flora occur than in 
Europe. The occurrence of nearly related isolated types 
in widely separated regions can almost always be ex- 
plained as a survival from once widely distributed an- 
cestors. In the case of herbaceous plants, such as 
Podophyllum, Stylophorum, and other peculiar types 
common to eastern Asia and Atlantic North America, 
we can only reason from analogy, but in the case of many 
woody plants, especially trees, e.g. the tulip-tree (Lirio- 
dendron), Torreya, etc., this is abundantly proved by the 
fossil remains. 
Perhaps the most striking instance known of close 
correspondence in the flora of widely separated regions, 
is the one already spoken of, z.e. the great number of 
identical or closely related plants found in the temper- 
ate regions of Pacific Asia, northern China, Mantchuria, 
and Japan —and Atlantic North America. Much of our 
knowledge of these extraordinary similarities we owe to 
the labors of Asa Gray. 
The writer recalls vividly the strangely familiar aspect 
of the vegetation of Japan, especially in the northern 
part, where nearly all of the more noticeable plants 
