108 CORRELATION OF GEOLOGICAL FAUNAS. [bull. 210. 
and the fluctuations are not expressed so much by an incursion of 
new species or a disappearance of some of the old species entirely, 
from the fauna, but the differences between the temporary faunules 
of each successive zone are found to consist chiefly in relative abun- 
dance of specimens and in relative size of those which do appear in 
the faunules. 
Other cases have been investigated, and from their study I conclude 
that the ordinary changes which take place in the life inhabitants of 
the seas on passing from one stratum to the next are chiefly differ- 
ences in abundance and vigor of the several species. When it is 
found, on passing from one zone in a section upward to the next, that 
the genera change with each new set of species the inference is at 
once that the change is due to migration. When, therefore, accord- 
ing to the above interpretation, it is observed that the faunas occu- 
pying the formations of the geological scale are not the same in two 
neighboring regions, the interpretation may be one of two: Either we 
have a succession of several faunas which may be contemporaneous, 
but represent different conditions of environment at the same time, 
or we have the modification of a single fauna into numerous local 
faunules — local and temporary — as it has been forced to migrate. 
Interpreting Paleozoic history on this basis, it becomes necessary to 
assume that the faunas must be distinguished geographically as well 
as vertically. 
EFFECT OF SHIFTING OF FAUNAS ON CLASSIFICATION OF GEO- 
LOGICAL FORMATIONS. 
If we trace the sediments of the Devonian for several hundred miles 
in one direction, from the west in Ohio eastward across the States of 
New York and Pennsylvania to their eastern limits, a remarkable 
series of changes is observed in the character of the sediments as a 
whole, which is interpretable by this study of the faunas contained in 
them. The facts developed by the minute analysis of the Devonian 
faunas already presented show that formational equivalency is not in 
accordance with faunal equivalency for the different parts of the region 
examined. In other words, if we attempt to trace a common geolog- 
ical horizon across the country by means of the evidence of forma- 
tional uniformity, we will reach a different conclusion as to equivalent 
formations than if the means of determination be the evidence of 
faunal integrity. 
This fact may be expressed in the case of the Catskill sedimenta- 
tion by saying that the Catskill formation occupies a lower place in 
the geological column in eastern New York and Pennsylvania than it 
does a hundred miles to the northwest. In this statement higher and 
lower are terms the estimation of which is based upon evidence of 
place of marine faunas in the rocks. 
The case of the Oneonta sandstone and its place in the midst of the 
faunas, described in detail on previous pages of this report, is another 
