1 
THE EAPIDS OF THE MADEIiiA AND THE MAMORE. 69 
viucc of Sao I’liulo, aud on tho Ivuliy ami Tibugy,* in the in’ovincG of 
Parana. They are the effects of the loose stones set rolling by the 
floods in the natural hollows of the cliffs. These pebbles, which are of 
(litfcreut sizes, bore into the gneiss-like stone (of metaniorphic origin) 
rows of deep cyhndric holes, with smootldy-polished vertical sides, and 
of all dimensions, from a few inches Avidth, to 17 or 20 feet in 
diameter and depth. 
Whole banks of the hard material liave been broken off in this way, 
as may be distinctly seen by tlic semicircular incisions of the remaining 
parts. In other places the process is still visibly going on. A narroAv 
division is left, at fii-st, between the holes, whose sides are gradually 
[)crforated at the bottom ; and this operation continues until the torrent- 
propelled stones work the channel deeper aud deeper, aud finally break 
off its outer wall. The whole fall must recode in this way, though 
much less rapidly by this simple erosion than by undermining and break- 
ing off the upper layers, as does the Magaiu, which, it has been calcu- 
lated, recedes at the rate of one foot per anuiim. 
Supposing now that these holes deepen only one fiftieth of an inch 
every yeai‘ — which must be thought rather a high estimate, considering 
the great hardness of the structure and the circumstance that most of 
the kettles arc “worked” only during the thi'ee mouths of the floods, 
— it will be seen that 12,000 years are required to make a hole 20 feet 
deep ; and this will bo the scale of retrogi’ession of the fall. 
As at the Eibeirao, the length of the rock-islands, Arhose upper end is 
now near the edge of the fall, probably represents the original breadth 
of the cliff OA er which the river rushed ; and, as this length is of .3,280 
feet, two miUious of years must have elapsed since the present state 
of slow transformation has begun. 
Though such a calculation cannot 
make any pretension to bo regarded as exact, the data bemg insuf- 
ficient, still it gh'es us an idea of the iWAVcrful changes AVrought in 
nature during immense lapses of time, by the smallest means, even a 
pebble set in motion by the water.f 
At tlio Tibagy, a tributary of tho I’arduapanema and Parana, these “ kettles ” 
arc anxiously semched by the diauioud-washors ; as tho precious stone, which indeed 
is a bettor “borer” than any other, is often found among the pebbles at the bottom. 
The material thero is a motamorphic sandstone, while close to it doleritic cones raise 
their rocky ci'owiis. 
f In a A'ory interesting work which appeared lately, “New Trucks in North 
America,” by W. Bell, tho author gives a detailed description of tho gigantic ravines or 
