PREGLACIAL DRAINAGE. 187 
the narrowing and thinning eastward of the Medina, Clinton, Niagara, 
and Salina formations. 
The Adirondack waters on the south and west were thus diverted by 
the Mohawk, the Black, and what we may calla ‘‘Rome” river. The Sus- 
quehanna was robbed of much of its territory, and its beheaded affluents 
now contest the ground in a losing struggle with the southern tributaries 
of the Mohawk, for in steep gradients and proximity to tidal water the 
Mohawk has a large advantage over the upper Susquehanna. Some head- 
waters of the Oriskany and Sauquoit seem clearly to have been pirated 
away from their former connection southward. This conquest of terri- 
tory finds an apt parallel in the landward migration of the Appalachian 
divide and the beheading of the rivers of Kentucky and Tennessee by the 
“Appalachian ” river.* 
COL AT LITTLE FALLS 
The existence of a water parting at this point is set forth briefly, but 
with much confidence, by Chamberlin. His language is as follows: 
“Tt was here, as I think, that a preglacial watershed parted streams that, on the 
one hand, flowed eastward to the Hudson, and, on the other, westward to the On- 
tario basin. Neither glacial erosion nor accumulation has been here sufficient to 
obliterate the main features of the preglacial topography. The watershed may be 
quite confidently and exactly located near Little Falls. The cutting of the upper 
and broader rock-gorge of the Mohawk at that point appears to have been the work 
of the earlier Glacial epoch and of interglacial drainage. It certainly antedated 
the close of the Glacial period, for the ice passed through it, rounding and scoring 
its ledges, but appears to have in no great degree modified its form or enlarged its 
capacity.’ + 
The writer believes that the above conclusion is fully sustained by fur- 
ther studies. The Little Falls station of the New York Central railroad 
is not far below the top of the gneiss at that point, and the altitude is 
0/6 feet. Thence the river descends for about a mile and a half to an 
altitude of 322 feet at the fault. Restoring at least 118 feet for postglacial 
gorge-cutting in the gneiss at that point, we find a barrier having an alti- 
tude of 440 feet. We must add to this an unknown amount for glacial 
or other erosion during the long progress of the Glacial period. As will 
appear later, the Little Falls fault determined the position of the divide 
by interposing the most resistant rock-mass between Hudson valley and 
lake Ontario. In addition, we have the important fact that the rock 
bottom of the valley descends, though not uniformly, from Little Falls to 
*C. W. Hayes: Southern Appalachians. Nat. Geog. Monographs, p. 322. Hayes and Campbell : 
Geomorphology of the southern Appalachians. Nat. Geog. Mag., vol. vi, p. 103. 
¢ T. C. Chamberlin: Preliminary paper on the Terminal moraine of the second Glacial epoch, 
3d Am, Rep. U. 8, Geol, Survey, p. 362, 
