100 



SCIENCE. 



[Vol. IV., No. 78. 



the common explanation ; and amateur geol- 

 ogists puzzle themselves about their origin. 

 It looks as though there was something special 

 about the Portage group of rocks, every where 

 abounding, as it does, in glens, gorges, and 

 canons, to which the formation of these re- 

 markable places is due. Why is the Portage, 

 of all the ten or twelve formations in this state, 

 the favorite one for these phenomena, being 

 composed, as it is, of sandstones and shales 

 very similar to those in other localities where 

 nothing of the kind is found ? 



The explanation is simple enough ; namely, 

 that it is owing to the peculiar alternation of 

 thin beds of soft shale and harder sandstone 

 rocks, and the presence of a stream of water 

 running into lower ground, which performs the 

 work of erosion, and the size of which must 

 be adapted to the work ; and upon that, too, 

 the size of the gulf depends. Beginning at the 

 lower end or mouth of the present canon, 

 the action of the waterfall first removes a little 

 of the softer layers of shale, leaving the thin 

 beds of harder sandstone projecting for a time, 

 which, in their turn, are also broken off from 

 want of support, in masses small enough to be 

 entirety carried away b} T the stream, especially 

 during the winter floods ; thus preventing the 

 formation of a slope, and exposing the bottom 

 of the falls to another erosion. Thus, by the 

 recession of the falls, the canon is formed, 

 provided another requisite is afforded ; namely, 

 that the side-walls must maintain their erect 

 position : otherwise a valley, instead of a 

 canon, is formed. 



This raises another question ; namely, wiry 

 do not these precipitous side-walls, composed 

 of apparently soft rocks, slope themselves 

 down, and form well-rounded hills, as they do 



elsewhen 



The answer is, because the harder 



layers of sandstone form what a mason would 

 call ' the binders,' which hold the natural wall 

 in its upright position. The erosion of the 

 canon is done by the stream of water under- 

 cutting, like the work of the coal-miner, and 

 then breaking down, the unsupported ' top- 

 bench ; ' while, in the mean time, the action of 

 the air and frost on the side-walls is so much 

 slower than that of the stream of water, that, 

 while the latter is rapidly cutting back, and 

 making the ravine longer and deeper, the sides 

 remain in their original upright position. As 

 the falls recede, and a thicker sandstone or 

 shale rock occurs, without the proper alter- 

 nation of strata, it will form the permanent 

 upper end of the gulf, which will be a precipice 

 if it is a sandstone, and a slope if it is a thick 

 bed of shale. It thus happens that the Portage 



portion of nature's mason^, and the neces- 

 sary streams of water passing over it, are 

 nicely adapted to the formation of glens and 

 canons. In localities where there are no 

 streams of sufficient size passing over the 

 Portage group to a lower level, as on the sum- 

 mit levels, there are high hills, it is true, but 

 no glens or gorges. The eroding action of the 

 elements being more uniform over the whole 

 surface, and the transporting power of a rapid 

 stream to cany away the falling fragments 

 being wanting, therefore slopes, instead of pre- 

 cipices, are there produced. 



There are also, in the state of New York, 

 limestone glens, as they might be called, which 

 are due to the same cause. At Trenton 

 Falls, on the Utica and Black- River railroad, 

 eighteen miles north of Utica, East-Canada 

 Creek has cut a narrow passage, three miles in 

 length, through the Trenton limestone, the 

 formation being named from this localhrv. It 

 is a canon with vertical walls a hundred feet 

 high, in which are the celebrated and very 

 beautiful falls. The cause of the erosion is, 

 that the limestone rock is in thin la} T ers, of 

 from six to ten inches thick, separated quite 

 regularly by thin layers of shale of about the 

 same thickness. It is owing to this regular 

 mixture of hard and soft rock, in alternate 

 courses, that the stream has been able to wear 

 awa} T the rock by undermining the shale into 

 a succession of cascades ; and, what is equally 

 important in forming a canon, the stream, a 

 wild torrent from the Adirondack forests, is 

 large enough to cany away the fragments of 

 the overlying limestone as fast as it gives wslj. 

 After cutting its channel back to a village 

 called Prospect, a thicker and harder layer 

 of gray limestone is encountered, which has 

 stopped the recession of the falls, the stream 

 being unequal to its destruction; and that is 

 the end of the ravine. 



The gulf of the Genesee River from Roches- 

 ter to Lake Ontario was caused in the same 

 waj T : for although the rocks are much thicker 

 and stronger than any of those above referred 

 to, 3'et the river is a correspondingly larger 

 stream, and was able to cut through the alter- 

 nating beds of Medina, Clinton, and Niagara 

 limestone, shale, and sandstone ; and the flood 

 of water is powerful enough, with the aid of 

 the fall below, to carry awa}^ the material, and 

 prevent the formation of a talus. 



At Niagara Falls and the gorge below, in 

 the same formations as at Rochester, is a 

 repetition of the same operation on a vast 

 scale ; and as the river there is larger than the 

 Genesee, so the canon is also longer, deeper, 



