Barriers to Dispersion of River Fishes W277 
of which abound in all waters of northern Asia, Europe, and 
western North America, has nowhere crossed the basin of the 
Mississippi, although one of its species finds no difficulty in passing 
Bering Strait. The trout and whitefish of the Rocky Moun- 
tain region are all species different from those of the Great Lakes 
or the streams of the Alleghany system. To the grayling, the 
trout, the whitefish, the pike, and to arctic and subarctic 
species generally, Bering Strait has evidently proved no serious 
obstacle to diffusion; and it is not unlikely that much of the close 
resemblance of the fresh-water faunze of northern Europe, Asia, 
and North America is due to this fact. To attempt to decide 
from which side the first migration came in regard to each group 
of fishes might be interesting; but without a wider range of facts 
than is now in our possession, most such attempts, based on guess- 
work, would have little value. The interlocking of the fish faunas 
of Asia and North America presents, however, a number of inter- 
esting problems, for migrations in both directions have doubtless 
taken place. 
Causes of Dispersion Still in Operation.—One might go on 
indefinitely with the discussion of special cases, each more or less 
interesting or suggestive in itself, but the general conclusion is in 
allcases the same. The present distribution of fishes is the result 
of the long-continued action of forces still in operation. The 
_ species have entered our waters in many invasions from the Old 
World or from the sea. Each species has been subjected to the 
various influences implied in the term “natural selection,”’ and 
under varying conditions its representatives have undergone 
many different modifications. Each of the six hundred fresh- 
water species we now know in the United States may be con- 
ceived as making every year inroads on territory occupied by 
other species. If these colonies are able to hold their own in 
the struggle for possession, they will multiply in the new condi- 
tions, and the range of the species becomes widened. If the 
surroundings are different, new species or varieties may be formed 
with time; and these new forms may again invade the territory 
of the parent species. Again, colony after colony of species 
Salmo clarki Richardson, throughout the Rocky Mountain range to the Mexican 
boundary and the headwaters of the Kansas, Platte, and Missouri. 
