But that is not all. The accompanying photogravures indicate an actual amount of removal over a part of the area enormously in excess of the rate supposed. Fig. 5 shows a portion of the precipice, a hundred feet high, where the road first comes down to the level of the Clinton limestone, and where, consequently, the whole thickness of the Niagara shale is accessible to examination. Fortunately, Patrick MacNamara, the watchman at this station, was a workman on the road at the time of its construction in 1854, and has been connected with the road ever since, having been at his present post for twelve years. We have therefore his distinct remembrance, as well as the appearance of the bank, to inform us where the face of the original excavation then was. In the picture he is standing at the original face, while the other figure is nearly at the back of the space which has been left empty by the crumbling away of the shale. The horizontal distance is fully twenty feet, and the rocks overhang to that amount for the whole distance exposed in the photograph. All this amount of shale has fallen down in forty-four years, making a rate many times larger than the highest we have taken as the basis of our estimate. Of course, this rate for the crumbling away of the Niagara shale on its fresh exposure is much in excess of the average rate for a long period of time; but it is clear that the rate of erosion at the base of the Niagara limestone at the mouth of the gorge can never have been sufficiently slow to reduce the total average much below the assumed rate of a quarter of an inch a year. To impress the truth of this statement it is only necessary to follow the progress, in imagination, of the crumbling process which has brought the side of the gorge to its present condition. At first the face of the gorge was perpendicular, the plunging water making the gorge as wide at the bottom as at the top. At successive stages the strata of shale on the side would crumble away, as is shown in our photograph, and undermine the strata of hard rock. The large fragments would fall to the bottom, and, being too large to be carried away by the current, would form the talus to which we have already referred, which would grow in height with every successive century. The actual progress of the enlargement would thus be periodic, and not capable of measurement by decades; but after centuries the progress would be clearly marked, and especially whenever there was a falling away of the lower stratum of compact Medina sandstone, which is about two hundred feet below the top, would a new cycle of rapid disintegrations in the superincumbent strata follow.