comparative shallowness of the river valleys, which allowed the glacial streams depositing the kames to exhibit such independence of the present waterchannels.

Glacial ice must also be invoked, to explain some of the facts regarding kames. At the time of these deposits the lower stratum of glacial ice still lingered in many localities, and filled many of the depressions. In some places the gravel and pebbles would be deposited in channels worn in the ice by the water, and in other places spread out over masses of ice; thus protecting them from the sun and the warm air. The final result was, that when the various ice-barriers were removed, and the water sought its present lines of outflow, these previous gravel deposits of the glacial rivers were left undisturbed, except where the changed course of the water-flow led across their path and eroded them. In places, where the earthy material had been deposited in ice-channels, a ridge of gravel would be formed when the sides melted away; while in the places where masses of ice had been so covered with sand and gravel that the melting was delayed the earthy material would settle down in a very irregular manner, forming the characteristic reticulated ridges and the frequent kettleholes wherever the material should slide off, and leave an enclosed mass of ice to melt away gradually.

XIV. The Champlain Epoch.

It should also be stated that at the time when the material composing the kames and terraces was being deposited there was a notable depression of land to the north; at any rate, in Northeastern America. The tourist who descends the St. Lawrence and ascends the Saguenay cannot fail to notice at the junction of these rivers extensive deposits of gravel high upon the mountains back of Tadousac. The St. Lawrence is here twenty miles wide; hence it would be absurd to suppose any rise of the river could account for the terraces referred to. They must mark a depression of the land. From similar gravel deposits at Montreal it is inferred that the land was nearly five hundred feet lower at the time of the formation of these terraces than now.

On the shores of Lake Champlain there is evidence that the depression was not far from four hundred feet. In the state museum at Montpelier, Vt., the skeleton of a whale is preserved, which was dug up from a clay deposit on the borders of the lake, one hundred and fifty feet above the level of the ocean. Upon going farther south the extent of this depression continues to diminish, till it has nearly disappeared upon the coast in the latitude of Philadelphia. In the Mississippi valley the depression carried the region four hundred feet below its present level. This period of subsidence

accompanying the close of the glacial period is called by Dana the Champlain epoch. Such a northward and westward depression, even where it did not proceed to the extent of submergence, must have greatly affected the southward flow of water at that time, for by as much as the depression at the north exceeded that at the south, by so much was the slope of all southward flowing streams diminished.

XV. Glacial Floods of the Delaware Valley.

When, now, we return to Trenton, N. J., the presence of human implements in gravels deposited by glacial floods is still more suggestive of Noah than such gravels without any implements. Doubtless the boulder-bearing brick clay of the Delaware valley and the Trenton gravel in which Palaeoliths are found are of glacial origin, though the clay belongs to a much more remote period than the gravel. Both were deposited by the river when swollen by the melting ice toward the close of the glacial period; and the clay belongs to the Champlain epoch. The succession of events in the Delaware valley would seem to be as follows:

During the early part of the glacial period ice continued to accumulate in the upper portion of the valley until it attained a depth of many hundred feet. It is probable that its average depth was not less than fifteen hundred feet. The ice

covered area of the upper Delaware is, however, small when compared with that of the Connecticut. The valley of the Connecticut embraces about twenty thousand square miles, and is all glaciated; while the glaciated portion of the Delaware does not exceed six thousand square miles. But fifteen hundred feet in depth of ice over six thousand square miles make fifteen hundred cubic miles of ice to melt, with which to swell the torrents of the channel below. Still, enormous as the amount seems, it is not sufficient to account for the floods which deposited the Philadelphia brick clay, even though it might explain the Trenton gravel. The depression of the Champlain epoch must be invoked to account for the Philadelphia brick clay. We will give the reasons for this inference more in detail.

The Delaware valley below Trenton is nowhere, at forty feet above its present level, less than five miles in width, and constantly enlarges towards the sea. If the water at Trenton were to rise sufficiently high to deposit the brick clay, namely, one hundred and fifty feet, the slope of the surface would be about two feet per mile to the bay, which is seventy-five miles distant. Now a current of five miles per hour, one hundred and fifty feet deep and one mile wide, would discharge a cubic mile of water every eight hours, or three cubic miles per day. (The mean rate of the Ohio River, with