362 H. H. Howorth—A Great Post-Glacial Flood. 
In describing the drift deposits near Thelwall, Mr. Atkinson, 
referring to a bed of fine white and yellow sand overlying the 
Boulder-clay, says, ‘‘Where the deposit of fine sand exists, it 
is found, on removing it, that a very large number of boulders lie 
under it, scattered over the surface of the clay or marl. From this 
it seems probable that before the deposition of this layer of fine sand, 
the clay had undergone considerable denudation by the action of water 
on its surface; being floated away in the form of mud, while the pebbles 
which were contained in it were left behind. This supposition derives 
support from the fact that numerous depressions exist in the surface of 
the clay, and are filled up by the sand. These small accumulations of 
sand are locally termed sand-pots. They vary much in size, shape, 
and depth. When the soil and sand are removed from a moderately- 
sized area, as in a portion of a brickyard, the surface of the clay 
presents the appearance of having been scooped and channelled by the 
action of the water” (Trans. Manch. Geol. Soe. vol. ii. p. 65). 
In volume iii. of the same work, page 72, I find the following 
sentence from my late friend, who had a vast experience of the drift 
beds, Mr. Binney :—*‘ A great deal of Till had been swept away by 
water before the deposition of thé fine sand which often capped it.” 
Another proof that the sand was deposited by a different movement 
and probably long subsequent to the Till is found in the remark of the 
same observer, that at the “junction of the sand and Till (where they 
follow one another horizontally), we sometimes find the sand intruded 
into the Till in the form of a long wedge” (id. pp. 858 and 359). 
All this goes to show that a space intervened between the depo- 
sition of the Till and that of the gravel, and that their distribution 
was due to different circumstances. 
The fact of the gravels and sands being in many places found 
superimposed upon the clays, does not connect them more closely 
than the superposition of the several sections of the Tertiary beds 
upon one another. It simply points to the gravels and sands being 
later in date, and in our view in many places considerably later in 
date. 
This is a considerable step gained, and in what follows we must 
be considered as separating the gravels and sands entirely from the 
Boulder-clay in regard to their origin and the lessons they teach. 
Now, while I propose thus to separate these gravels and sands 
from the clays with which they sometimes occur, I propose to 
argue, on the other hand, that they are to be correlated with 
other gravels and sands which superficially clothe the surface of 
the older deposits where the Boulder-clay does not occur at all, and 
which have been assigned to very different dates and origins, and to 
show that in all probability they are contemporaneous. That the 
fact of the gravels containing in some cases certain specific contents 
not present in others is perfectly explainable without disturbing this 
arrangement. Whether the gravels occur as they do at Macclesfield 
and in the so-called Middle Sands of Lancashire, with a number of 
fragments of marine shells,—whether as in Eastern England they 
occur with similar marine shells mixed with freshwater shells and 
