district, on the contrary, the decomposed crust has been removed, and the hard unaltered rock laid bare.

It has been shown that the decomposition of the rocks has been caused by the slow percolation of rain-water containing a little carbonic acid. It follows that the surface rocks had been exposed for long ages to this influence, and that the time that those of the glaciated districts have been subjected to the same action, since the Glacial period, is exceedingly short in comparison.

Another mark by which the glaciated may be distinguished from the non-glaciated country is by the occurrence of a deposit long known in Scotland under the name of "till," and on the Continent as "grund morane" and "moraine profonde." It is generally a stiff clay, full of small angular fragments of the rocks over which the ice has passed, and sometimes with large angular and subangular stones. It always more or less reflects the characteristics of the strata immediately in the neighbourhood and in the direction from which the ice has come. Thus in the vicinity of Toronto I noticed, as Mr. G. J. Hinde had before remarked,* that the till is packed with small fragments of black Utica shale and blue Trenton limestone-strata that the ice had passed over in its passage from the eastward. Along with these were larger fragments and slabs of the underlying Hudson River group, and a few rounded boulders of gneiss that were far travelled. Around New York many patches of till are left on the glaciated rocks. I visited Marion, near the city of New Jersey, by the advice of Prof. Cook, and found very fine sections of the glacial beds. The till is there not a stiff clay, but a rather sandy deposit, of a dark reddish brown colour, packed with the angular débris of the red triassic sandstones that form a large portion of the bed-rocks of eastern New Jersey. A few of the contained stones may have been brought from a distance, but the great bulk of them, as well as the sandy matrix, are of local origin.

The deposition of the till probably took place during the melting back of the ice-sheet. Dr. Dana has shown that the ice was very thick over New England, and that the pressure at its base would be so great as to force the plastic mass into the crevices of the rocks below, so as to tear off fragments from them, which, with any loose material it met with in its progress, would be gathered up and borne along in the lower portion of the ice-sheet. Prof. Joseph Le Conte,

* Canadian Journal, April, 1877, p. 8.

+"On the Glacial and Champlain Eras in New England." Amer. Journ. Sci., March, 1873.

in his study of the deposits left by the ancient glaciers of the Sierra Nevada,* has arrived at a similar conclusion. Prof. Hall, in his "Geology of New York," has given a large picture of a natural section on the shore of Lake Erie, in Chautauque county, where we may see as it were the till in course of formation. In some parts the top strata are only a little separated, with clay forced in between them; in others they stand on edge, or are broken up and the fragments scattered throughout the till. I have seen similar instances on the shores of Lake Ontario. In our own country there are also many examples of the same action; one of the finest, that I have seen, being in the beautiful section of the glacial beds exhibited in the cliffs at the mouth of the Tyne, in Northumberland. In this way the lower portion of the ice appears to have become charged with stones and finer materials which were left on the surface when the great glacier melted back. The till is therefore restricted to the area that the land-ice covered, and is as much a memorial of its former presence as the scratched and rounded rocks on which it generally lies.

By the glaciated rock-surfaces up to the line I have already mentioned and their absence beyond, by the outspread of the till limited by the same boundary, and by the disappearance of the decomposed surface-rocks up to but not beyond that margin, we know that the land-ice did not reach, in the valley of the Delaware, farther than the neighbourhood of Belvidere, which is 50 miles to the north of Trenton. What relations, then, do the beds of drift that I have described at Trenton bear to the ice-sheet? To answer this question we must take into consideration what we know is taking place at the terminations of existing glaciers. Great streams of water run from underneath them, bearing along fragments of rocks that have been melted out of the glacier or fallen through crevices to the subglacial rivers. These are rounded by attrition, and spread out in sheets of pebbles often extending for miles below the terminations of the glaciers. From the enormous glacier that once filled the Delaware valley as far as Belvidere, and which was itself only a southern prolongation of the still greater northern ice-sheet, there must have been shed a vast amount of similar material. The large rounded drift that lies at the base of the quaternary beds at Trenton (No. 3 in Figs. 1, 2, 3) is almost undoubtedly a similar

deposit. It contains a mixture of all the various rocks over which the glacier of the Delaware would pass; those from the hardest formations being most abundant, as they would best survive the severe abrasions to which they had been subjected.

No implements are yet reported from this bed, so that, putting aside the disputed implement from the morainic accumulation at Butzville, we have no evidence at present that man frequented the valley of the Delaware before or during the greatest extension of the glacier of that valley. The deposits from which Dr. Abbott has obtained the implements lie above this great unstratified bed of rounded drift, and we should have great difficulty in fixing the approximate age of the implement-bearing strata if it were not for the fortunate occurrence of the sandy clay with large. boulders (No. 1 in Figs. 1 and 3) clearly superimposed on the latter. We may now turn our attention to the consideration of this surface-bed and the relation it bore to the Glacial period.

In my discussion of the glacial and post-glacial phenomena bearing on the date of the excavation of the gorge at Niagara, published in this Journal,* I have described the occurrence of large boulders of crystalline rocks lying above all the other glacial beds. In the till which lies next the glaciated bed-rocks the stones are all of local origin; in the surface deposit they are all from the distant north.

Prof. James Hall, so long ago as in 1843, had fully recognised the importance of the occurrence of these far-transported blocks that lie scattered over the surface, and had noted the difference in the mode of their occurrence and in their composition from the rocks included in the lower glacial beds. He shows that the glacial beds belong to two periods: one, the lower, which contains mostly local rocks; the other, the upper, containing far-transported crystalline rocks. He says that on the broad northern slope. towards Lake Ontario, where hills are distant, there are numerous and extensive fields of boulders resting upon the surface, or but partially imbedded in the soil, and holding such a position that it is evident that they are of subsequent origin to the great body of detritus; and again, on the western prairies, long lines of boulders are to be observed stretching away for miles beyond the reach of vision, as if once forming a line of coast or deposited along some channel

or course of a current, though the general surface indicates no influence upon this portion beyond what is common to the whole. Prof. Hall considered that there was no explanation of the transport of these great blocks, excepting on the supposition that the whole surface was covered with water, over which they were floated on icebergs, "Had they been transported," he says, "by a powerful current over the bottom (which cannot be supposed from the inequalities of the surface) all the older drift would have been. removed at the same time, and instead of finding them as we do now, mostly upon the surface, they would have been imbedded indiscriminately in the superficial detritus, and there would have been no means of recognising the products of different periods.'

Dr. Newberry, in his "Surface Geology of Ohio," has fully described the distribution of large boulders over the surface of that State. Even in Southern Ohio they are in some parts very numerous. He says that the large unscratched boulders are generally found on the surface, and that in the great series of excavations, which have been made in the construction of the railways and canals, they have been rarely met with below it. They are often seen resting on the fine stratified clays which form the upper part of the drift. And he observes that "it seems impossible that they should have been brought to such positions by glaciers or currents of water, as either of these agents would have torn up the underlying clays. We also learn, from their relative position, that these boulders were deposited at a later period than the most recent stratified beds of the drift series, and that they were floated to their present resting-places. In short, no argument is required to convince anyone who will glance at the facts that these boulders, and probably the gravel and sand with which they are sometimes accompanied, were floated on icebergs from the north shore of the great fresh-water lake which once filled the lake basin, and that as these icebergs melted, or when they stranded, their loads were discharged on the top of all the drift deposits which had been laid down in the preceding epochs of the Quaternary age."+

On the eastern side of the Appalachians, Prof. Hall has noticed the occurrence of these boulders in the valley of the Hudson, and says that he has searched in vain, near Albany and Troy, for a boulder or pebble of granite, or of

any rock older than the Potsdam sandstone in the deposits below the clay; while in a period subsequent to the deposition of the clays and sands, boulders of granite are by no means rare.*

In the southern part of the State of New York and in New Jersey they are not uncommon. At Marion, at the section of the till to which I have already referred, the top bed is a light-coloured sandy clay, similar to that at Trenton. Lying on and sometimes imbedded in this are large boulders, scattered over the surface. The sandy clay rests directly on the till, and is about 3 feet thick. Both here and at Trenton these great boulders were much larger than any I saw in the underlying till or drift. At Trenton they are often seen in the formation of new streets on the outskirts of the town. Some of them are 7 or 8 feet across, and most require blasting before they can be removed. I learnt from Prof. Smock that these blocks are distributed over much of the State, and he spoke of particular boulders occurring at a considerable altitude. I do not know, however, how high they occur, but probably this interesting question will be worked out by the Geological Survey of New Jersey, as well as the distances which they must have travelled from their parent rocks.

Nor does the Delaware form the southern limit of the far-transported boulders. They appear to bear the same relation to the drift-beds in Virginia, for Mr. Wallace, in his account of the discovery of stone-implements near Richmond, speaks of boulders in the surface-soil, and of large blocks (8 feet by 12) resting on the gravel.

It is obvious, as Prof. Hall and Dr. Newberry have pointed out, that these great blocks of stone must have been carried to their present position by floating ice. Any flood of water sufficient to move them would certainly wash away the sandy loam in and on which they rest, and such a mode of transport would not account for their position scattered here and there over the great undulating plain that extends from Trenton to the sea; nor could they have been left by the great ice-sheet, as they are found far beyond the limits to which it reached. Sometimes we hear the distribution of the upper glacial beds ascribed to a second Glacial period, when the ice again covered the land. But ice could not have moved thus for hundreds of miles over beds of gravel and sand without disarranging them, and nowhere in America has any sign been noticed of a second advance of