Be 
J 
“5 
ng 
fk 
: 
i 
eg 
a 
W. M. Davis—Gorges and Waterfalls. ESS 
south lakes of the western arm of New York must frequently 
or always have a similar origin; notably the gorge below 
Taghannock Falls on the west side of Cayuga lake, and Wat- 
ins Glen at the head of Seneca lake. ‘Trenton Falls and 
Ausable Chasm, farther east, most likely come under the same 
class, but so far as the author knows there are no descriptions 
of them sufficiently detailed to make the matter sure. Local 
studies of these interesting spots would lead to valuable results. 
The creek that joins the Mohawk at Canajoharie flows from an - 
open, flat, drift-covered country into a deep, narrow, rocky 
gorge. Schoharie Creek has throughout the greater part of its 
course opened a terraced valley in the drift-filling of its old 
channel ; but a few miles below Gilboa it abruptly enters and 
as suddenly leaves a rocky chasm, a few hundred yards long 
and fifty or more feet deep. Here it has evidently lost its 
former line of flow, and cut its new bed in the rocky slope of 
its old trough. Rondout Creek does the same about two miles 
above its junction with the Hudson. 
Several of Professor Newberry’s writings* call attention to 
the old buried channels in Ohio and the numerous gorges 
-Tesulting from turning old streams into new courses. Rocky 
River, for example, which enters Lake Hrie seven miles west of 
Cleveland, follows an open valley strewn with drift in the 
upper part of its course, but when two miles above its mouth 
It enters a narrow rave with rocky walls and floor and fol- 
a Superior, and of the Ohio at Louisville, are thought by - 
ae Proce edings Bost. Soc. Nat. Hist., ix, 1862-63, 42; Annals N. Y. Lyceum 
“at. Hist., ix, 1870, 1; Amer. Naturalist, iv, 1870-71, 193; Geology of Ohio, i, 
met Na ii, 1874, 12 
eol. Ohio, i 
I. Ohio, i, 174, t Id., i, 511.  § Id., i, 201, ete. 
