220 THE POPULAR SCIENCE MONTHLY.
of the deserted channels are cut to a great depth ; but, whatever their depth, they are inclosed by banks or bluffs that are still dis- tinct and comparatively steep ; thus showing that relatively short periods of time elapsed both during and since their occupation by large rivers. The rock-cutting done by Niagara in post-glacial time seems to be a much greater piece of work than that accom- plished by any of the temporary lake outlets during the closing phases of the Glacial period ; but none of them, as far as I have read, had an opportunity for active work equal to that of Niagara.* They are nearly all comparatively shallow ; but an exception to this rule has been pointed out to me by Mr. Gilbert, to which a paragraph may be devoted.
Emmons, the geologist of the second district of New York, long ago described what he took to be fissures in the Potsdam sandstone of northern New York, but which to modern interpre- tation appear to be gorges or chasms cut by rivers, presumably constrained into that position by drift or ice obstruction. The Ausable chasm is a well-known instance of these " fissures," but one of the examples described by Emmons has no river running through it. It lies close to the Canadian boundary in Clinton County, sixteen miles west of Lake Champlain, and is thus de- scribed in Emmons's report (Geology of the Second District, New York, 1843, pp. 309, 310) : " The fissure or gulf, as it is called, is three hundred feet deep and about sixteen rods wide. Its walls of sandstone or conglomerate are perpendicular at the deepest part. The small lake at the bottom is said to be one hundred and fifty feet deep. The direction of this fracture is north, seventy degrees west, and the rock dips at a small angle from each side of it. . . . At Keeseville and Cadysville large rivers, the Ausable at the for- mer and Saranac at the latter, still occupy these gorges as their channels, and have sufficient force and power to sweep out, es- pecially in the time of high water, all rocks of an ordinary size. At this place there is merely a small rill discharging itself from a small lake of dead water, insufficient in itself to accomplish any perceptible change. To account for the present condition of this rock, we have therefore to go back to a period when some current swept through this gorge with great force and power ; for by no other means could the materials which once filled the space be- tween the present walls of the gulf be removed."
Returning to the general features of the abandoned channels.
��* Russell's Geological Reconnaissance in Central Washington gives an account of a temporary displacement of the Columbia River when its valley was obstructed by ice and its waters ran through the Grand Coul6e : a basin was then excavated beneath a cataract in the course of the river, and the basin now holds a lake, although the river and the cata- ract have disappeared (Bulletin 108, United States Geological Survey, pp. 91, 92).