112 CORRELATION OF GEOLOGICAL FAUNAS. [bull. 210. 
This phase may be recognized in the Sherburne formation of Che- 
nango County. 05 In the same way the Portage of western New York 
should, on this hypothesis, be represented by a black shale laid down 
farther west, such as the Ohio shale, and to the east it actually blends 
into the Ithaca and then into the Oneonta, in accordance with the 
theory. Still higher in the series the Catskill formation of the east- 
ern part of New York is at the same horizon as the fossiliferous 
Chemung of the central part of the State and the Erie shale of the 
sections of western Pennsylvania and Ohio. 
Thus the shifting of faunas furnishes a key by which the chrono- 
logical relations of the formations which hold the fossils may be deter- 
mined with a degree of accuracy not possible on any other basis, and 
reduces to order facts which on the ordinary interpretation are not 
only without apparent order but seem, at least, to be unrelated to 
each other. 
The sequence of the faunas themselves, in each section, furnishes 
a clue to the direction in which the shifting has moved. If, for 
instance, the passage upward is from richly fossiliferous shales into 
black, nearly barren, even-bedded shale, the locality where the sedi- 
ments occur was sinking, and the shore line was becoming more dis- 
tant; and, on the assumption that at the time the general shore lines 
were to the east and north of central New York, the inference is that 
the pushing in of the black Genesee shale over the Hamilton was from 
the southwest. All the facts bear out this conclusion. 
Again, if the succession of beds is from fossiliferous shales into 
red, flaggy, and coarse sandstones, the interpretation is that the region 
was rising. In central New York rising would cause the shore lines to 
encroach upon the sea advancing toward the west. This is the fact 
in the case of the Oneonta sandstone; and all the facts bear out this 
interpretation. 
FAUNAL SHIFTING AND CORRELATION. 
Thus a minute study of the faunules in their relation to the sedi- 
ments and their distribution and succession furnish a means of cor- 
relation far better than continuity of like sediments, a safe method 
when the transgression is parallel to shore lines but fallacious when 
the formation is traced at right angles to the shore line of origin of 
the sediments. It is a surer method of correlation than reliance upon 
identity of fossils alone, for we have ascertained that a prevalent 
fauna retains a general integrity of its specific composition for a time 
of great length, measured by the sedimentation of many hundred 
feet of ordinary shale and sandstone rocks, and through a thickness 
of limestones which may reach several hundred feet. 
The relation of limestones to the other classes of sediments has not 
been indicated in the above analysis. It is more difficult to determine 
a See section at " Nigger Hollow, 11 Prosser, p. 134, XIX C 2. 
