‘ f 
Parr IL. Sect. ii. §3.] RIVER EROSION. — 377 

_zontally upon soft shales (Fig. 110). The softer strata at the base 
are undermined, and slice after slice is cut off from the cliff over 
which the cataract pours. The parallel walls of this great gorge 
owe their direction and mural character to parallel joints of the 
strata. The lesser or American fall enters by the side of the ravine 
and falls over its lateral wall. The larger or Canadian (Horse-shoe) 
fall occupies the head of the ravine, and owes its form to the 
intersection of two sets of joints. The structure of the gorge being 
the same at both falls, it seems reasonable to infer that as the 
American fall, which appears to be diminishing in volume, has 
cut back only somewhere about 140 
feet from the original face of the ravine, 
this branch of the river has, compara- 
tively speaking, only recently begun to 
work. Goat Island, which now separates 
the two falls, is an outlier of drift resting 
on the limestone. It has been cut off 
from the rest of the ground on the 
right bank of the river by the branch 
which rejoins the main stream by the (- Sipamy geaatieees tS 
American fall. From the position of Fie. 110—Szcrmton aT THE 
the glacial strie it may be concluded Honsr-snoz Fauzs, Nracara. 
that a great part, if not the whole of «4 Means ee 0 ee . 
the ravine, has been excavated since the 3 eile, Mae ‘Shale, 80 
glacial period. There are indications feet; d, Niagara Limestone, 
indeed of a pre-glacial valley by which 165 feet, of which 85 feet are 
the waters of Lake Erie joined those of  Vi!>le at the fall. 
Ontario before the erosion of the present gorge. Bakewell, from 
historical notices and the testimony of old residents, inferred. 
that the rate of recession of the falls is three feet in a year. Lyell, 

















Py ce 
Dr acai 
An ‘ CN (WN 
[a 
anil 
% a, 
































Fig. 111.—PLAN oF THE RAVINE OF NIAGARA AT THE F'ALLs. 
A, American Fall; C, Canadian Fall; W, Whirlpool; G, Goat Island; D, Bank of 
| Drift resting on ice-worn sheets of limestone. 
on no better kind of evidence, concluded that, “ the average of one 
foot a year would be a much more probable conjecture,” and estimated 
