Williams] GEOLOGICAL EXPEESSION OF FAUNAL MIGRATION. 39 
succession of formations often differs from classification based upon 
the succession of species, and the paleontologist is often found 
practically differing from the stratigrapher in his interpretation of 
the correlation of the rocks in any particular region. 
Although the matter of shifting of faunas has been, in a general 
way, involved in what is called geographical distribution, I am not 
aware that, in this country, it was deliberately announced as a fact 
until about 1883 or 1884, when such announcement became neces- 
sary in order to explain certain facts in the geology of New York 
State which I then had under investigation. The most conspicuous 
case which came under my notice was reported in Bulletin 41 of the 
United States Geological Survey, On the Fossil Faunas of the Upper 
Devonian — the Genesee Section, New York. The investigations 
which led to the publishing of that report were carried on for the 
direct purpose of ascertaining what kind of modification actually 
occurred in the same formation when it was minutely and compara- 
tively studied for a few hundred miles across the field of its distri- 
bution. • The Upper Devonian was taken because of its possession 
of several successive faunas, the lack of disturbance of the strata, 
and the wide region over which its outcrops could be studied with- 
out any doubt as to their stratigraphical correlations. The investi- 
gation showed unmistakably that the constituent faunas which make 
up the sequence of any particular section had shifted back and forth 
over the region. It was ascertained, for instance, that the place of 
the fauna belonging to the Ithaca group corresponded stratigraph- 
ically to the lower part of the Portage formation of the western part 
of the State; whereas to the east the Hamilton faunas crept up with 
some of their speeies into the same stratigraphical zone; while still 
farther east the same horizon, geologically speaking, was filled by 
sediments of the Oneonta group, which seem to be equivalent, in every 
respect but position, to portions of the typical Catskill formation. 
Again, in 1897 a study of the faunas of the southern Appalachian 
province, in the southernmost point of Virginia, brought to light the 
fact that actual traces of the Carboniferous fauna were found in a 
position in the sequence which, a little to the north, was found to be 
dominated by Chemung species. a Such facts can be explained at 
present only by supposing that there was a shifting of the faunas 
geographically within the common basin in which the}^ lived. 
The theory of the migration of faunas, then, assumes to be true the 
proposition that two faunas, one of which generally succeeds the 
other, may be actually contemporaneous in their life periods, at least 
during the end of one and the beginning of the other. By the theory 
of shifting of species and migration of faunas it is easy to understand 
how a fauna which immediately succeeds any other particular fauna 
of a given region (if the faunas be actually different, or if one be 
«SeeOn the Southern Devonian formations: Am. Jour. Sci., 4th scries, Vol. Ill, ]S<)7, pp. 393-403. 
