districts in the gravel-period as there is now. I have introduced illustrations of these Welsh gravels, because it would be impossible to form any clear view of the general features of the gravel-period without comparing the sections of the deposits found on the upper part of rivers flowing off high land with those formed along the lower parts of rivers near the sea in the Quaternary period. The general law of deposition will be the same; and it is this law I wish to investigate by the aid of the measured sections. Those sections are chosen specially for description where the rock or clay and the incumbent gravel can be observed together. This gives an opportunity of observing the points or lines on which denudation ceases and deposition commences.
In the London basin, the series of Quaternary strata generally commences with a coarse gravel, a few feet thick, eating into the chalk or London clay in which it lies, and mixing up the clay or chalk with the gravel.
The coarse gravel often passes into coarse sand, and is evidence then that the movement of the water at that point was not sufficiently rapid to transport gravel, or that there was no gravel present to be deposited at that moment.
I have not observed any finely bedded clay or loam in contact with the surface of the chalk or London clay in the London basin; and this appears as if the movement of water over the surface of the clay and chalk had been generally rapid at the commencement of the Thames deposits. No doubt there are many cases where the first deposit has been of fine materials, but I have not been able to meet with them.
Where concavities in the chalk and London clay exist in the neighbourhood of the Thames, and in its side-streams, and are so situated as to be favourable for quiet deposition, the lowest bed of gravel or coarse sand sometimes contains fossils derived from the Eocene beds, and rolled Mammalian remains of a later period. It is succeeded by a series of laminated clays and false-bedded sands, from 20 to 30 feet in thickness, with water-worn materials, and containing fluviatile and land shells, tranquilly deposited with Mammalian remains (only occasionally rolled). No natural sections exist of this class of deposits; but excavations for brick-earth have during the last 30 years opened out several of these fluviatile beds.
Non-fossiliferous gravels are often clearly contemporaneous with these fluviatile brick-earths. If on the same horizon, they are sometimes interstratified with the fossiliferous beds; but the non-fossiliferous gravels reach to much higher levels, and are deposited at much steeper angles, and contain materials not perfectly washed or water-worn. They include masses of Thanet sands, of plastic clay, of London clay, and of the fossiliferous bands of the Woolwich series, buried in great unrolled masses in the gravel, as well as water-worn masses of Druid Sandstone, derived from strata very near the spot of deposition. These materials were evidently washed in by heavy floods, not confined to the valleys, but passing over the whole surface of the land, tearing up the ground and carrying
