H. H. Howorth—A Great Post-Glacial Flood. oor 
by no other agency known to us than that of ice. Granting this to 
the full, however, my contention is that it does not necessitate our 
assigning to ice everything we meet with in recent geology. We 
cannot and must not assign to it work which ice is most clearly 
incompetent to accomplish, nor are we to import fresh transcendental 
laws into physics to supplement what empirical tests tell us are the 
controlling forces of matter, in order to support @ priort theories 
and metaphysical dogmas about Uniformity. 
This has been largely the burden of my previous communica- 
tions. In the present paper I propose to carry my argument 
another step further. The superficial gravels may be divided into 
two well-marked series, according as their contained pebbles are 
rounded or have their angles more or less intact. The latter class 
are more limited in extent than the former. I have already treated 
of them and the lessons they furnish in the paper on the Angular 
Drift. Putting them aside, I now propose to consider the gravels 
containing rounded pebbles and the sand, which so generally accom- 
pany them, and which together form such a conspicuous feature in 
the Drift deposits of Northern Hurope. Discarding for the present 
_ any considerations about the other contents of these gravels, let us limit 
our attention to the pebbles themselves. In the first place, then, these 
pebbles clearly did not originate in the action of ice in any form. 
No observer known to me has argued seriously that the rolled 
pebbles which constitute the gravels in question, could have been 
rolled and rubbed into their present form by glacial action. They 
are, for the most part, unscratched, rounded, and water-worn, and 
quite out of the category of glacial products proper, and are to be 
correlated with the shingle beaches found by the sea, or the gravels 
continually being formed in rapid rivers. This is universally 
accepted even by ultra-glacialists. 
But this by no means exhausts our problem. Much the larger 
portion of the rolled gravels which are so frequently met with in 
Northern Hurope have no traces of marine or freshwater shells in 
them, and therefore according to the canons by which we have tested 
submarine and in fact subaqueous deposits, we cannot treat these 
gravels as they are now found, as long submerged marine beds, or as the 
immediate result of river action. On the other hand, the rolled and 
rounded character of the pebbles makes it clear that they must have 
been triturated and shaped under water. How, then, are we to re- 
concile the position? ‘The reconciliation is by no means at once 
obvious, as may be judged by the various solutions of the difficulty 
that are current, most of them accepting the conditions that rolled 
gravels are as clearly the remains of sea-bottoms as shelly deposits. 
It seems to me the only real solution of the difficulty is to conclude 
that the shaping of the pebbles and their distribution as gravel are 
to be entirely separated: the two facts are not even to be taken as 
elements of one problem. So far as we can see, the pebbles are for 
the most part derivative. This has been long known in regard to 
some of them. For instance, in the gravels south of the Thames, 
where many of the pebbles are rounded, while others have their 
