First--to what extent has that concretion formed behind the falling water? Has it formed there in greater quantities than it has where the face of the cliff has been open to the air? In greater quantities might have been expected, on account of the greater amount and absolute continuity of the moisture on the rocky face. The 400 feet length of cliff, from which the waters have now been permanently shut off, furnishes the answer. Practically, none of that concrete has ever accumulated in the crevices of the rock on the face of the cliff immediately behind the Falls. The currents of air, and the furious blasts of water which they create, rush constantly away from the under surface of the falling sheet, and continuously against the face of the cliff. These scour and cut away the rock, even as a sand blast would do, though more slowly. They allow no chance for deposits. The strata of the Clinton Formation (which commences at about the level of the water in the Gorge, and of the Niagara shale, which overlie it--the two combined having a depth of about eighty feet) are eaten away the faster. The eighty-feet-deep layer of Niagara Limestone, which overlies the shale, being harder, is eaten away slowly; its lower layers being attacked by the winds and waters from below (as the underlying shale disappears) and also on its face, yielding faster than the upper ones. [Illustration: AMERICAN FALL--NIAGARA. CAVE OF THE WINDS IS BEHIND SMALL FALL.] That this concretion has always formed in the limestone, back from the face of the cliff, behind the falling sheet (where the blasts of wind and water cannot reach nor effect it) is certain. That it forms under the river bed, and back from the face of the gorge on both sides of the river, and wherever the water percolates through the upper layer of rock, is also certain. It is so found in the limestone (but not in the shale) wherever deep excavations have been made near the river in the vicinity of the Falls and wherever tunnels have been driven through the limestone--in the crevices and especially where a pocket or hollow space exists in that formation. This process of eating away the lower rocks, undermining the upper limestone, which, as its support is taken away, tumbles into the Gorge, shows the means by which the Falls gradually recede. It is shown to the best advantage in the Cave of the Winds, which, during the past thirty years, by this wind-and-water-blast process, has been enlarged to four times its former size. Some day the layer of rock at the top of that cave will fall; the edge of the Luna Island Fall will be thus moved back a number of feet; the Cave of the Winds will become merely a narrow space between the outward-curving fall of water and the perpendicular rock; and the wind-and-water-blast will continue its erosive work on that rocky face;--and in the course of years will again produce a distinct cave.