Mississippi Drainage System. — Westgate. 249" 
and the commencement of the Cretaceous cycle of which there is 
little record in the Mississippi basin. During this time the drain- 
age was changed from the first condition in which the streams 
flowed down, the slope and across the strike of the beds, to a con- 
dition where they followed the softer members of the Paleozoic 
rocks. 
It is probable that the change in outlet from the Ohio to the 
St. Lawrence was effected at a later date and was one of the re- 
sults of the elevation and tilting of the Cretaceous base-level, which 
will be considered later. After the general surface of eastern 
America had been reduced to a lowland under the action of long- 
continued erosion, a slight tilting of this lowland would be suffi- 
cient to divert a part of the Mississippi drainage northward 
through the lower St. Lawrence. If lake Michigan and lake Erie 
drained southward into the Ohio previous to the glacial period, 
the final completion of the present St. Lawrence drainage was not 
effected until the close of the ice invasion. 
(b) The drainage of the Cincinnati and Tennessee anticlinals. 
In the discussion of the post- Carboniferous drainage two similar 
important structural features of the Mississippi basin were left 
out of consideration. These were the Cincinnati and Tennessee 
anticlinals, as they have been called. If these two regions were 
to have a surface form like that indicated by the geological struct- 
ure, there would be a broad low dome over central Tennessee and 
a similar but much larger elevation over parts of Keutuck}', Ohio 
and Indiana. If such elevations did exist they are ignored by 
the present Ohio and Cumberland rivers which cut directl}' across 
them. 
The most probable explanation of this apparent peculiarity is 
that the anticlinals never existed as surface elevations at any 
time subsequent to the Appalachian elevation. Some consider- 
able part of the arching of the strata took place during Paleozoic 
time, previous to the Appalachian uplift. The direction of the 
axis of disturbance is not parallel to the Appalachian axis of 
disturbance, but is more nearly north and south, suggesting an 
independent disturbance. There are evidences that during Pale- 
ozoic time shallow water or land conditions prevailed along the 
axis of uplift in Tennessee, Kentucky and Ohio. Ripple marks 
occur in the Devonian Black Shale in Ohio. Going south from 
lake Erie to Tennessee the Devonian and Upper Silurian succes- 
