42 
Correspondence — J fr. Hugh Miller. 
it on, and aid me to show the fallacy of the analogy by which he 
seeks to undermine some of the grounds I occupy. Every one who 
lias studied streams and rivers knows that below rapids and falls, 
and at other places, tkey scoop pools muck deeper, and also broader, 
tkan the average stream near the place. The little runnel makes a 
rough dimple ; the Highland bum a linn ; the alluvial river leaves 
in its old channels small meres; and in the valley of the great 
Amazon these isolated pools are represented by lakelets or lakes 
some teti miles long, roughly speaking, and thirty or forty feet deep. 
The rule is, that the volume of the stream determines the size and the 
contents by volvme of the pools it malces. Glancing back now to the 
question of “reasonable proportions,” it is evident that this rule 
must not be unreasonably stretclied by a use of blind logic. To say 
that it applies witkin reasonable limits, is correct ; to say that any 
one allowing only the pools of the burn to the large river is illogical 
(as well as wrong) is also correct. But to repudiate the rule because 
it cannot explain lakes proportioned by tkeir size to the hypothetical 
pools of impossible rivers, would be simply futile. 
But I proceed to apply this rule elsewliere. Mr. Judd refuses to 
allow that a glacier grinds in a basin with added force, on the grounds 
that “ we are led to infer ” that streams of water and rivers of ice fall 
under similar laws of motion (p. 525), and in a preceding paragraph 
(p. 524) he says what must involve belief on bis part, in the pro- 
duction by glaciers of basins proportional in superficies to the pools 
of the Highland burn and Mississippi river. The above rule, that 
streams of water make tkeir pools according to their volume, being 
correct, rivers of ice, Mr. Judd will probably admit, sliould do like- 
wise. Now streams very much broader than the Amazon do not, 
and probably could not exist, though I am safe in saying that if they 
did their pools would be lakes. But it is a truism now-a-days that 
glaciers many times wider than the Amazon did and do exist. The 
Humboldt glacier is about 60 miles wide ; ancient glaciers moved 
over plateaux and over-rode waterskeds, and by the analogy claimed 
by Mr. Judd we would be justified — nay, encouraged — in predicating 
as possible lakes limited in breadtk only by the volume of glacier 
and ice-sheet. It is not immoderate tlien to ask for the sprinkling of 
tarns and lakes which the nature of the pre-glacial surfaces favoured. 
Although, as I think, legitimately damaging to Mr. Judd’s position, 
his parallel between ice and water cannot strictly be carried out. 
The cascade of a Highland burn tumbles into a pool less broad some- 
times than deep, and not much longer. To accredit glaciers with 
such powers were to forget an important element of difference — the 
greater rigidity of ice. It is this property — the same that makes 
glaciers kabitually scratch rocks as well as smootking them, thus 
giving them a greater rock-hollowing power — that has made it 
possible for me to argue, what could not be argued of water, that 
the deeper a glacier drives a basin, “ the more fully it feels its 
power and the more easily and rapidly it works.” 
A word now upon the stratigraphical division of the question. 
With deference to Mr. Judd’s autkority, I must say that 1 cannot 
agrec with kirn that the liorizontality of the Assynt mountains is “an 
