38 The American Geologist. July - 1903 - 
then under the lake and around the southern edge of the Mich- 
igan coal-field returning to its starting point near Sarnia. The 
smaller branch which emerges from below lake Erie, at San- 
dusky, and borders the Cincinnati axis, has been already suffi- 
ciently described in writing of its outcrops at Prout's- Station. 
The line of outcrop which doubtfully appears on the west- 
ern side of the Cincinnati arch has not revealed any distinctive 
Hamilton features and the long continuation of the same to the 
northwest almost to the shores of the polar sea merely indicates 
Devonian conditions continuing unchanged throughout the per- 
iod, and the existence of a Devonian ocean throughout the 
whole area, in which no breaks either stratigraphic or palseon- 
tologic can be detected. 
Changes of Level. — As already stated, the nearly uniform 
thickness of the Corniferous limestone indicates a stable con- 
dition of the sea bottom over its whole area and the absence of 
any considerable disturbance along its shores. The incursion 
of mud and sand over part of the same area during the succeed- 
ing periods is equally conclusive proof of instability. 
From the facts already given we can at least roughly out- 
line the more important changes of level that occurred in the 
Appalachian area at the close of the Corniferous period and 
before the setting-in of the Marcellus or Corniferous-Hamilton. 
There is no ground for predicating any rise or fall within the 
Ohio area. The outcrop of the upper division of the Cornifer- 
ous limestone (of the Ohio survey), including the thin shale 
which alone represents the Marcellus in the state, coincides 
with the lower buff beds around the line of their joint outcrop 
in Ohio, Indiana and Kentucky, so that the shore-line of that 
group remains unaltered. 
But eastward important changes occurred. On the reap- 
pearance of the Corniferous-Hamilton as the Marcellus on the 
other side of the Appalachian coal-field we find its shore far 
overlapping the edge of the Corniferous limestone, there invis- 
ible, and the Black shales lying on the Oriskany sandstone with 
nothing intervening. We must hence infer that an area approxi- 
mately equal to that which received the Oriskany sandstone and 
which was raised so as to be above water — probably only to a 
slight extent in the Corniferous period, again sank to receive 
the Marcellus shale. This depression included all middle Penm 
