In view of what has been said, it is evident that the absence of beds of subaërially stratified silt, sand, and gravel, between beds of till can not be brought in evidence against separate ice epochs. It would rarely be true that topographic and hydro-*graphic conditions would make possible the construction of plains of sufficient extent to serve as criteria for the purpose here indicated, and few of those formed would escape such a degree of destruction as to leave them demonstrably continuous. There is also the further possibility that such beds exist, even though their continuity be not known. To prove the continuity of a buried bed of stratified and incoherent drift, even if it existed, would be a most difficult task. (6) _Differential Weathering._ If, after covering a given region, the ice retreated, the drift which it left in the area which it previously covered would be subject to oxidation, leaching and disintegration. The depth to which this oxidation, leaching and disintegration would extend, would be dependent upon the length of time during which the drift was exposed, and upon the climate which affected the region during its exposure. The longer the exposure and the warmer the climate, the deeper would the weathering extend. If, subsequently, the ice extended over the same region, it might, in some places, override and bury the old surface without destroying it. The earlier oxidized and leached drift would thus come to be buried by the newer, unoxidized, unleached drift. If, therefore, beneath the newer drift of any given locality there be found a lower drift, the surface of which is oxidized and leached to a considerable depth, the evidence is strong that the lower drift was exposed for a long period of time before the upper drift was deposited upon it. Within certain limits a similar result might be brought about, it is true, if the ice, after having reached a certain maximum stage of advance, were to retreat for a short distance only and there remain for a very long period of time. A subsequent minor advance might bury the oxidized surface of the drift beyond the position of the long ice-halt. Under these conditions, the climate which would have obtained in the area of the drift exposed during the minor retreat would have been cold, and oxidation, leaching, and disintegration would have proceeded slowly. If they reached considerable depths, the time involved must have been very long. If this surface of oxidized and leached and disintegrated drift were found to reach far to the northward beneath the layer of newer and upper drift, it would indicate a great recession of the ice. We maintain that if it were found sufficiently far north of the margin of the overlying drift, and if its depth were sufficiently great, extending well down below any possible accumulation of superglacial till, it might be a positive criterion of so great a recession of the ice, protracted