H. H. Howorth—A Great Post-Glacial Flood. 309 
but when we attempt to deduce from these local instances anything 
like a general law, the evidence breaks down at once. Not only isit 
not applicable to the great continental areas occupied by drift at all, 
-but it is not even applicable to a large portion of the United Kingdom, 
nor even to large portions of Lancashire and Wales. This seems 
most plain, and may be tested by a very superficial examination of 
the problem in many districts. This conclusion is by no means my 
own only, but it has the complete support of more than one ex- 
perienced authority on the so-called Glacial beds. 
Mr. Mellard Reade, who has studied these beds with such care, 
says, “ From the time of Joshua Trimmer until now, attempts have 
“been made to distinguish the various beds of clay, sand, and gravel, 
from each other. Trimmer divided them into ‘Lower Hrratics,’ 
‘Boulder-clay,’ and ‘ Upper Erratics’ (sand and gravel). Professor 
Hull more lately into ‘Lower Boulder-clay,’ ‘Middle Sands and 
Gravels,’ and ‘ Upper Boulder-clay,’ and this latter classification is 
the one now most generally adopted. My own investigations lead 
me to doubt the value of any of these classifications, and, what 
makes it more confusing, there is no real agreement among the sup- 
porters of this classification as to which division special beds belong. 
Thus Mr. De Rance considers all the shells of my list to be derived 
from the Lower Boulder-clay, while Mr. Mackintosh and Mr. Shore 
commit them to the ‘ Upper Boulder-clay.’ As regards the fauna of 
the sand beds, I have shown that really no geological distinction can 
be based on these grounds, while, on the other hand, I have failed to 
trace out any stratigraphical connection that would support any of 
these theories of geological subdivision” (Transactions Geol. Soc. 
Glasgow, vol. vi. p. 265). 
Apart from the coasts of North Wales, says Mr. Mackintosh (an 
observer of vast experience), “I have nowhere seen sand and gravel 
regularly and persistently interpolated between a lower and upper 
Boulder-clay ” (GroLocicat Macazinu, Vol. IX. p. 21). 
Treland used to be quoted as a remarkable proof on a considerable 
scale of the reality of the tripartite division of the Glacial beds; but 
it will be exceedingly rash to quote Ireland, after the very sug- 
gestive and able papers by Mr. Kinahan in Vol. IX. Ist Series, and 
Vol. Il. 2nd Series, of the GzonogicaL Magazine. In one of the 
latter papers he sums up the result of his admirable induction in the 
words: ‘‘ There does not appear to be evidence in Ireland for two 
distinct ages of Glacial drift, separated by an interval represented 
by subaqueous accumulations (sand, gravel, marl, clay, etc.), and the 
reasons for this will be given in this paper” (op. cit. p. 111). 
The fact is, like many other generalizations, the tripartite division 
of the loose surface beds is due to our mistaking a particular acre 
for the whole field, and when we base our theory upon an exceptional 
example,—when we poise our pyramid on its apex,—we naturally 
find it unstable. The superposition of the second Boulder-clay 
upon the gravels and sands in some very local deposits in Lan- 
cashire and North Wales is not a normal arrangement, but an 
abnormal one—an exception of great interest, as we shall endeavour 
