Wallace Letters Online

Share this:

Record number: WCP4879

Add to My list
Sent by:
Alfred Russel Wallace
Sent to:
Charles Lyell
17 March 1869

Sent by Alfred Russel Wallace, 9 St Mark?s Crescent, Regent?s Park, NW to Charles Lyell [none given] on 17 March 1869.

Record created:
16 November 2012 by Catchpole, Caroline


No summary available at this time.

Record contains:

  • letter (1)

View item:

LETTER (WCP4879.5280)

A typical letter handwritten by author in English and signed by author.

Held by:
American Philosophical Society
Finding number:
The Darwin-Lyell Collection
Copyright owner:
Copyright of the A. R. Wallace Literary Estate

Physical description

Transcription information



[[1]] 4201

9, St Mark’s Crescent N.W

March 17th. 1869


Dear Sir Charles

As I have many objections that appear to me serious ones against your explanation of the origin of lake-basins, and as I presume from your writing to me so fully on the subject you will be glad to have them, I will at once state what they are.

1. From the small amount of delta in many of the Alpine lakes compared with their actual size and depth, it is evident that the time that has elapsed since the ice left there has gone but a small way towards filling them up. It is evident therefore that even if they had not been preserved by being filled with ice, such lakes might continue to exist for a length of time many times as long as that which had elapsed since the glacial epoch.

[[2]] Now as it has never (I believe) been supposed that the glacial epoch lasted any thing [sic] like so long as the period of time since it has ceased, the preservative power of ice during that comparatively short time will not account for the abundance of lakes within glaciated districts and the sudden and almost total cessation beyond them. Even in tropical countries you must suppose that there have been no unequal elevations, turning valleys into lakes, not only during the whole period since the glacial epoch but during a period two or three times as long, for so long at least a lake like Geneva or Maggiore would take to fill up. The difficulty of the remarkable distribution of lakes does not therefore seem to be lessened by supposing that they were preserved [[3]] by ice, not formed by it.

2. The local distribution of groups of lakes, seems to me equally opposed to the view of their having been formed by unequal elevation, unless we suppose such elevation to be limited to districts of a few miles square at a time. For example round Snowdon within a space of 10 miles square there are three valleys radiating to N.E.[,] N.W. and S. each having lakes. Round Helvellyn in a space of 20 miles square, are a dozen lakes radiating to every point of the Compass.

The same distribution occurs more or less strikingly in Perthshire ans in other parts of Scotland.

3. If upheaval has caused lakes then we should generally them in all [[4]] parallel valleys near each other. Yet there are several valleys parallel and near to Lakes Neufchatel[sic]3 and Constance without lakes.

4. The manner in which lakes of all sizes are thickly and irregularly scattered over comparatively level glaciated countries, such as Finland and northern America seems inexplicable on the theory of upheaval, as we should otherwise expect some similar ones in the great network of river vallies[sic] in the South American plains where the surface is so covered in vegetation that there is hardly any denudation going on to fill them up.

The first and last of these arguments appear to me strongest against the upheaval theory. That there has been upheaval to any enormous extent no one denies; -- that it must often have tended to alter the level of valleys and form [[5]]4 lakes must be admitted; -- that it has rarely succeeded in doing so may be due to the fact that the denudation of river channels is generally more than sufficient to counterbalance the upheaval, which is in the more easy to be understood because denudation is unceasing while upheaval is often intermittent. The Dead Sea is probably a case of a lake produced by upheaval of a barrier or depression of the valley, or both combined. The great lakes of Africa may be due to similar causes.

Now for the objections to the Glacial Theory of lakes.

The essence of this theory appears to me to be, the unequal grinding away of the valley, caused by the accumulation of a great mass and thickness of ice, [[6]] which accumulation will be determined by the meeting together of many glacier streams, and a narrow or contracted valley at some point below. If a valley is tolerably wide and has no very contracted exit, it may carry away an enormous glacier which will grind away down its bed uniformly but will not accumulate at any one part so as to grind that part much deeper than another.

Now[?] at the head of Lago Maggiore 4 large valleys from the high Alps meet together, and their combined glaciers would be contracted into a space disproportionately small. Lower down an equally large valley comes in bringing glaciers from a range of 50 miles of high alps, [[7]]5 and below this the valley is again contracted. The enormous ice stream brought in by this valley would cause the greatest accumulation of ice & it is just here the lake reaches the enormous depth of 2500 feet.

At the head of the lake of Como is a similar meeting of two large and widely divergent valleys, bringing down glaciers from an immense extent of Alps, and the formation of the ground leading to the bifurcation of the lake at Bellagio, is a sufficient indication of the heaping up of the glacier that must have occurred above that point.

The whole range of Alps for full a hundred miles, contributed its glaciers to the Lago Maggiore depression -- Nowhere in the whole range district of the Alps can we find [[8]] such a concentration radiating valleys of about[?] equal length & to one point, as this, and it is here we find the deepest lake. That would be an extraordinary coincidence if it were due to upheaval & depression only. The valley of the Baltea Dora has no such converging extent of glaciers to fill it. It is a very wide & open valley and its tributaries come in at pretty regular intervals instead of converging on to one spot, as at the head of the Maggiore or Como. The Rhone valley above Martigny is very long[,] straight & totally wide, & its tributaries come regularly in one after the other, nowhere converging to a focus. Below Martigny the valley [[9]]6 narrows, but the torrent is rapid & the slope probably great. A little lower the tributary glacier of the Diablerets & the Dent du Midi would pour in nearly opposite each other, & it is just here, about Bex, where the lake began. We must I think look at the Lakes of Geneva & Neufchatel[sic] 7as having been (on this theory) ground out at the same time, when the whole great valley between the Alps & Jura was filled with ice, and when it was heaped up by the configuration of the surface more in these two places than elsewhere, and also moved in these valleys more persistently & rapidly. The glaciers from Mont Blanc would at that time probably flow over the low passes from [[10]] the valley of the Arve direct to[?] the centre of the Lake of Geneva.

The last passage of your letter and some others will bear the interpretation, that you think unequal movements of the ground could not form never have formed

lakes except during the time that the denuding power was checked by the valleys being filled with ice; -- at all other times, running water being able to cut itself a passage as fast as its bed was upheaved. This, it is true, would be an answer to my objection (1.) but it would I think introduce an enormous difficulty, since it would imply that in every part of the world, movements are going on with such constancy and [[11]]8 rapidity, that during the comparatively short time the glacial epoch lasted, lakes were formed by the them[?], literally by thousands, -- sometimes thousands of feet deep, and diverging in all directions in the same district. Movements could hardly be so universal and so rapid as this, without producing lakes at other times and places in considerable quantity, unless indeed denudation is even more rapid than has ever yet been supposed.

The more I look at this question from every point of view, the more I see in favour of the glacier theory of lakes. It accords strictly with every fact of their distribution, and with almost every detail of their individual position. The only argument against it is one of degree & time, since it is admitted that glaciers do grind [[12]] rock to some extent. To any other theory the objections seem to me to be more radical, since no other seems able to explain either the distribution or the position of lakes. On the upheaval theory lakes should occur just as frequently in straight open valleys with no narrowing of outlet, & with no converging valleys, -- as elsewhere, -- but few if any such lakes exist, while all the most remarkable lakes can be proved to be situated where, during the glacial epoch great accumulations of ice must have taken place. The study of good maps of lake-districts, will prove this, and it is necessary to consult them to understand fully the extent to which ice accumulation has taken place [[13]]9

My remarks on glacier movement (Canon Moseley’s[)]10 was written hastily,but if you like to show them to Darwin do so, & if you think it worth while[sic] I will put them into better form, for a letter to the Geological Magazine.11

One more remark on lakes. If they have been caused by upheaval during the glacial epochs, they might not to[sic] be confined to vallies[sic] surrounding high mountains, & more or less narrowed at their exit, but should occur also in the neighbouring lowlands whose open valleys must have been glaciated almost as long but with a thinner sheet of ice. But there are no lakes in the lowlands of Scotland, a[nd] in the North of England [[14]] which must have been under ice quite as long as the valleys around Snowdon. The entire absence of lakes over such extensive regions as these, where all the conditions would seem to have occurred for their production, seems to me a much more difficult thing to account for in the upheaval theory, than is the occurrence if any particular lake or the absence of one in any particular valley, or[?] the glacial theory.

Another region where all the conditions for the production of lakes unusual elevation seem to concur in Russian N[orth]. of 55o Lat. This country must have been every where[sic] glaciated, and its numerous rivers must one should think have been [[15]] converted into numerous large lakes, had any such amount of upheaval occurred during the continuance of the glacial period and would have been necessary to have produced the lakes of Switzerland, N[orth]. Italy[,] Scotland and Finland. but there are no lakes of importance till we get towards Finland and the Arctic Ocean.

I do not lay very much stress on these negative facts, but I think they are at least as strong as any that can be brought against the theory.

Believe me | Yours very faithfully | Alfred R. Wallace [signature]

Sir Charles Lyell, Bart.


1. The page / catalogue number, "420", appears at the top left of the page.

2. The page number has been written beneath the address.

3. This is a misspelling. Alfred Russel Wallace, is probably referring to Lac de Neuchâtel.

4. The page / catalogue number, "420", appears at the top left of this page. At the center there is a circled "2", and at the top right ARW has numbered the page himself as page "5".

5. ARW has numbered the page, indicating that it is page 7, at the top right-hand corner.

6. The page / catalogue number, "420", appears at the top left of this page. At the center there is a circled "3", and at the top right ARW has numbered the page himself as page "9"

7. Alfred Russel Wallace is referring to Lac de Neuchâtel.

8. ARW has numbered the page, indicating that it is page 7, at the top right-hand corner.

9. The page / catalogue number, "420", appears at the top left of this page. At the center there is a circled "4".

10. Canon Henry Moseley (1801 - 1872). First Professor of Natural Philo-sophy at King’s College, London, FRS and Canon of Bristol.

11. Geological Magazine: a peer-reviewed scientific magazine first published in 1864.

Please note that work on this transcript is not yet complete. Users are advised to study electronic image(s) of this document, if available.