360 H. H. Howorth—A Great Post- Glacial Flood. 
to show in another paper, if we are permitted to print it; but still an 
exception. 
For this reason we cannot accept Mr. Hull’s tripartite classifi- 
cation, nor yet his name of Middle Gravels and Sands, for the 
superficial beds of sand and gravel covering the Boulder-clay in 
some places. The division in our view should not be tripartite, 
and strictly speaking not even duplex. It is quite true that in certain 
cases we find the Boulder-clay overlaid by gravels and sands, and 
these cases are more frequent than those in which there is a three- 
fold arrangement, but even these are local and exceptional. Every 
tyro knows numerous cases of large areas where the clay occurs. 
without any such covering, and where there is no reason to suspect 
any subsequent denudation; and these cases are matched by similar 
ones where the gravels and sands occur over similarly wide areas 
without the subjacent clays. The connection between the two is 
largely accidental. It is a great temptation, in mapping out these 
most eccentric and difficult beds, to form sequences out of them 
which we may delude ourselves into believing are realities, and not 
merely ingenious and unstable hypotheses; but, such hypotheses are 
necessarily ephemeral. 
If we are satisfied that the clays and gravels so often classed 
together are not to be symmetrically arranged either according to 
a duplex or triplex system, but must be treated as substantive 
integers standing apart from each other, and only occurring together 
by accident,—and if we are satisfied that, whatever may be the case 
with the Boulder-clay, the gravels are indisputably not the direct 
_results of glacial action,—we may finally inquire as some have done 
whether these gravels are not the clays in another form—the pebbles 
constituting them being merely the triturated boulders washed out 
of the clay which preceded the gravels. This contention will not 
bear criticism and is inconsistent with experience. I must content 
myself with one carefully selected example as a test case. Mr. H. T. 
Hardman, in a paper published in the Geotocroan Maeazing, Vol. II. 
2nd Series, pp. 172-3, speaking of his experience on the geological 
survey in the North of Ireland, thus discriminates between the 
contents of the Boulder-clay and those of the gravels so often 
covering it. ‘The Till,” he says, “contains invariably a very large 
per-centage of the local rock, whatever it may happen to be—Lime- 
stone when that rock prevails; Sandstone and shale near Coal- 
measure ground; and a plentiful supply of Basalt, both in large 
blocks and small pebbles, when over or within reasonable reach of 
it, together with igneous and metamorphic rocks from some distance; 
but in no case Chalk or Chalk-flints to the amount of more than 
about three or four per cent., even when in the Chalk country, unless 
where in very close proximity to unprotected Chalk exposures. On 
the other hand, the gravels, as a rule, contain perceptibly less of the 
local rock, plenty of travelled or foreign pebbles, and a great abundance 
of chalk and flints; of the last two so much that the ground is 
often white with them; and this often when at a distance of 100 
feet from, or sometimes underlying them, is Boulder-clay, on which 
