. ,. 
Process of Diversion. 117 
without appreciable shifting. That the col at a was reduced 
nearly to its present altitude by erosion during the Tertiary 
cycle appears from a comparison of the amount of cutting which 
has since taken place in the most favored localities along the 
rivers and in the least favored localities on the divides. In 
the former the post-Tertiary erosion has been from 150 to 
300 feet, and 100 feet seems a liberal estimate of the erosion in 
the same period upon the divides; but 100 feet added to the 
present altitude of the col at a would still leave the divide in 
soft shale or limestone. Since the divide at b is the one through 
which the contending streams finally forced a passage, it is not 
unreasonable to suppose that it offered some advantage which 
the divide at @ did not possess. This was doubtless its altitude, 
which was in all probability considerably less than that of the 
divide at a. At the same time the divides e and h, between the 
streams G, H and K, had been similarly reduced, although the 
streams belong to a single drainage basin. On the above hy- 
pothesis it appears that the conditions were quite favorable for 
diversion of drainage, since the heavy conglomerate had been 
removed not alone from the main divide at >, but also from a 
series of connecting channels occupied by the streams G, H 
and K. 
A careful study of the Tertiary peneplain in this region shows 
it to be higher on the eastern than on the western side of Walden 
plateau. In the vicinity of Chattanooga its altitude is nearly 
900 feet, while in Sequatchie valley it is somewhat less than 800 
feet; hence there appears to be a difference of at least 100 feet 
in the altitude of these two neighboring peneplains formed dur- 
ing the same period of baseleveling. A corresponding difference 
in the altitude of the Lafayette gravels was noted above. The 
probable explanation of this difference in altitude is found in 
the fact that the Sequatchie river had during the Tertiary cycle 
a more direct outlet to the sea than the Appalachian river, and 
also was flowing on softer and more homogeneous rocks ; hence 
its valley was more perfectly baseleveled, and indeed it seems 
probable that under the exceptionally favorable conditions there 
prevailing the Sequatchie river may have reduced its gradient 
southward from the Tennessee line almost to zero. If the Appa- 
lachian river on the opposite side of Walden plateau were 100 
feet higher than the Sequatchie it would have a descent of LOO 
feet in about 400 miles, or a fall of 3 inches per mile. Consider- 
