Extent of the Peneplain. 109 
to the Coosa at Rome, and thence to its junction with the western 
branch. 
DIVERSION OF THE APPALACHIAN RIVER. 
It is stated above that the drainage of the Appalachian valley 
was southwestward, from the New-Kanawha basin to the sea 
margin, until the close of the Tertiary baseleveling period. 
Since the date of diversion of this drainage is an extremely im- 
portant point in the history of the region and since the above 
statement is hable to be questioned, the grounds on which it is 
based will be given in some detail. The evidence is derived 
from (1) the perfectly baseleveled divide between the Tennessee 
and Coosa river basins; (2) a comparison of the volume of ma- 
terial eroded from the Appalachian valley with that of the 
Tertiary sediments in central Alabama; and (3) the immaturity 
of the Tennessee gorge through the plateau below Chattanooga. 
Evidence from the Coosa-Tennessee Divide—As already stated, a 
peneplain, extending from the Cumberland plateau on the north- 
west to the Great Smoky mountains on the southeast, stretches 
from the head of the Holston and Clinch rivers to the edge of 
the Tertiary sediments in central Alabama. This peneplain is 
well shown in the photograph of the relief map of this region 
reproduced as plate 4. It is as perfectly developed across the 
Coosa-Tennessee divide as elsewhere, and shows no perceptible 
variation in the two basins except the gradual southward de- 
scent shown in plate 6 and due to subsequent differential eleva- 
tion. It extends across the Appalachian valley from Pigeon 
mountain to the base of the Cohutta mountains, a distance of 
40 miles, interrupted only by the valley ridges of hard sand- 
stone or by low knobs of silicious Knox dolomite. Since the 
peneplain is developed only on soft rocks, it is possible that the 
divides might have been cut down to their present altitudes 
by backward erosion of headwaters while the streams occupied 
their present courses; but while the altitude of the divides is 
not conclusive evidence that the main streams have flowed 
across them, the breadth of the valley upon the divide materially 
strengthens the evidence. By the backward cutting of streams 
at their headwaters a characteristic dendritic, inosculating drain- 
age is developed, and it seems improbable that the divides should 
have been maintained in their present position throughout the 
Tertiary cycle without producing this characteristic surface, 
which is conspicuously absent. 
