Wiixiams.] INTRODUCTION. 7 
ronments present in sections so nearly contiguous to one another as 
to occasion confusion in correlation. 
The case of the Old Red sandstone and the marine Devonian was a 
conspicuous exception to the practice indicated. In this case the 
marine faunas of the Devonian limestone were recognized by Lonsdale 
as holding an intermediate place between the Silurian and Carbonif- 
erous marine faunas; and the Old Red sandstones were known to 
occupy the interval between these two systems ; hence the equivalency 
of a series of marine beds with a series of estuaiy or fresh-water beds 
containing an entirely different fauna was established. But, in gen- 
eral, in the lesser cases, where faunas of the same kind of organisms 
are concerned, it has been the prevailing practice of geologists every- 
where to assume that formations must be classified in a single column. 
Since the correlation and identification of formations has depended 
on their fossil contents, this practice has resulted virtually in the 
assumption that fossil faunas whose identity can not be established 
must be either older or younger than the standard faunas to which 
they are most closely related. 
It was in the belief that this practice was erroneous and was lead- 
ing to false conceptions of geological history that the investigations 
here described were begun. But the difficulties in the way of demon- 
strating the fallacy of the practice were great. Since the fossils are 
the only means by which the identity of two formations found at a 
distance from each other can be established, it seemed like a contra- 
diction to say that two formations with unlike faunas may be identi- 
cal in age. In order to test the question, it was necessary to take a 
region in which, for considerable distance, the structure of the rocks 
was so simple and so little disturbed that the stratigraphical equiva- 
lency of the beds could be traced with a high degree of certainty from 
one end to the other, independently of the fossil contents. Such a 
set of conditions appeared in the Devonian rocks of New York, Penn- 
sylvania, and eastern Ohio. It was proposed to make a series of sec- 
tions cutting through the same general part of the geological column, 
at intervals of about 50 miles, extending eastward as far as the Hud- 
son River Valley and westward as far as the Cuyahoga Valley at 
Cleveland, the first trial section having been made along the meridian 
running through Ithaca, N. Y., in 1881-82. Minute st udy of each sec- 
tion was to be made; the fossils were to be collected from each fos- 
siliferous zone, the position of which was to be carefully noted, and 
the faunules so collected were to be separately analyzed and listed. 
Intermediate traverses were to be made to tie together the sections 
by clearly recognized continuous strata, so that the stratigraphic 
equivalency of the parts of each section could be established with cer- 
tainty. The work was begun privately in Cornell University, bu1 the 
necessity of transgressing Stale lines led to the association of the 
university with the United Slates Geological Survey, by whose official 
