9. _Probable effects of a retreating ice sheet on drainage._ The generally accepted conclusion that glaciers advanced southward and occupied the Laurentian basin during the Glacial epoch and retreated northward toward the close of that epoch, is sustained by a vast body of evidence. As the ice sheet withdrew it left a superficial deposit frequently one or two hundred feet thick over nearly all of the region it abandoned, and pre-glacial drainage lines were obstructed and mostly obliterated. As long as the slope in front of the ice was southward, the drainage from it found ready means of escape, but when the slope was northward towards the ice front, the drainage was obstructed and lakes were formed. We have good reasons for believing that the topography of the Laurentian region was essentially the same at the close of the Glacial epoch as it is now, but the broader question of continental elevation is less definite. The inequalities of the surface being essentially as we now find them, it would follow that the first lake formed when the ice retreated to the north of the divide running through central Ohio and central New York, would be small and dependent on minor features in the relief of the land, and would discharge southward. As the ice retreated, the lakes would expand and become united one with another and the larger lakes thus formed would still find outlet across the southern rim of the basin. As the glaciers continued to retreat lower and lower, passes would become free of ice and the lakes would be drained at lower levels, old beaches would be abandoned, the lakes would contract, and finally separate lakes would be formed in the lowest depression in the basins of the more ancient water bodies. The shape of the retreating ice front would be determined by topographic conditions and would in turn determine the northern outline of the lakes along its margin. This in brief is one hypothesis that has been proposed to explain the varied history recorded by the shore records, sediments, etc., within the basin. 10. _Communication with the sea._ Another hypothesis which assumes to account for some of the facts observed, is that the continent was depressed at the close of the Glacial epoch sufficiently to allow the sea to have access to the Laurentian basin. This hypothesis is coupled with others which do not recognize a period of Pleistocene glaciation, but, as already suggested, this is a matter that is considered by the great body of American geologists as not being any longer open to profitable discussion. In the study here outlined the question whether the water bodies which formerly occupied the Laurentian basin were lakes or arms of the sea, should not be difficult of direct and positive determination. If fossils can be found within the basin, they might yield definite testimony, but even if they are absent or if their evidence is inconclusive, topography can be appealed to with the expectation of receiving a conclusive decision.