skip to content

Darwin Correspondence Project


To the Scotsman   [after 20 September 1847]1

– To the Editor of the Scotsman.—2


As the parallel roads of Glen Roy have lately been again brought into notice by Mr Milne’s able & interesting Memoir,3 & as you have always shown much interest on this & allied subjects, I venture to request a little space, for a few remarks, chiefly for the purpose of calling attention to some points, which appear to me to require further examination. My opinion regarding the great improbability of there ever having existed barriers either of detritus or rock at the mouths of the valleys, so as to form lakes, remains quite unaltered: I still think it clear from the facts stated in my paper in the Philosophical Transactions4 that the whole country emerged slowly from beneath the waters of the sea: I believe that this is admitted by Mr Milne. Formerly I thought that if it could be shown that such barriers had had not existed the shelves must have been formed by the waters of the sea. Since my observations, were published, M. Agassiz & Dr Buckland visited the site, & advanced (Eding. New. Phil. Journ. vol 33. p. 236) a most ingenious theory namely that lake had been formed by great ancient glaciers having blocked up the mouths of the valleys. When this theory was first propounded, I mentally rejected it from my belieff that when the water stood at the middle shelf of Glen Roy, there was no outlet for drainage except over the ice itself; & this being so, it appeared to me highly improbable, that the water could have been banked up to the same exact level for the great length of time necessary for the formation of the shelf. but this prominent objection falls to the ground, since Mr Milnes important discovery of an outlet at the head of Glen Glastig.)5 (Mr Milne is strongly opposed to the ice-lake theory & believes that glaciers could not have descended from Ben Nevis in the course suggested by Agassiz & Buckland. But from the published accounts of ancient glaciers in many parts of the world & from what I have myself seen (London Phil. Magazine Vol XXI. p. 184) in North Wales,6 I can feel no difficulty in admitting that the whole valley of the Spean was formerly filled up with ice: Mr Milne’s curious details on the state of the surface near Loch Treig strikes me as corrobarating the former existence of glaciers. To form lakes at the level of the shelves in Lochaber, a glacier must have blocked up Glen Collarig and been pushed up Glen Roy above the mouth of Glen Glastig: at this period the water of the lake would have stood at the upper shelf & its drainage escaped at the head of Glen Roy, down the Spey. As the climate became more temperate, the glacier may be supposed to have retreated a little towards Ben Nevis & then the water would have escaped by Glen Glastig & stood at the level of the middle shelf. Subsequently the glacier must have retreated greatly, a portion of it (or a distinct glacier) still remaining at the mouth of the Spean, & the lower shelf would have been formed, the drainage flowing through the Pass of Muckull. Glen Gluy7 apparently must have been closed by an independent glacier. The blockage in these cases, may be supposed to have been by the ice itself, as in the famous case in the valley of Bagnes,8 & not by a gigantic terminal moraines. As in North Wales, I could in parts, distinguish iceberg action, from that of the ancient glaciers, so here I suspect, that the scored rocks described by Mr Milne at the heads of several glen, will turn out to have been effected by tide-driven icebergs grating over the surface, during the emergence of the country from beneath the sea.) (The ice-lake theory explains beautifully the obvious difficulty of the non-extension of the shelves over the whole country, & the remarkable coincidence in height of the the three shelves of the Roy & Spean,—of the upper shelf in Glen Gluoy—& of that of Kilfinnin, with outlets, over which the drainage-waters of the successive lake might have escaped. The middle shelf of Glen Roy, extends further than the upper one at the mouth both of this Glen & of Glen Collarig, & both shelves at their terminal points die away in the most insensible manner: this further extension of the middle shelf agrees well with the theory of a retreating glacier, & the very gradual dying away of the extremities of both shelves, which always struck me as most remarkable & opposed to the former existence of any barrier, might be explained, as was suggested to me by Mr Lyell (with whom I have discussed this subject, & at whose instigation I have written these few remarks) by the lake having having been generally frozen near the great icy barrier, littoral action having been thus prevented. This theory explains the number of erratic boulders on the shelves, (with which fact I was much struck,) for charged icebergs might have floated off from the glacier when they entered the lake. Finally ice-lakes explain,, as well as lakes formed by barriers of detritus, all the appearances at the outlets & elsewhere, considered by Mr Milne as incompatible with the marine theory; furthermore it would, perhaps, account for the very considerable bodies of water, which Mr Milne describes, as having formerly escaped at the outlets; for the accumulated rain of the whole year, in the form of snow, would flow away only during the summer.) (Opposed to the ice-lake theory, there appear the following facts (1). The existence on Tomhbran of a shelf described by me,9 between the upper & middle shelf & which Mr Milne discovered in two other parts of Glen Roy; & again of a second shelf in Glen Gluy, extending, according to Mr Milne for several miles on both sides of the valley; for it is believed that at the level of these shelves there are no outlets in the surrounding mountains; & if so, the drainage-streams must have flowed in both cases for a very long period, over the ice itself without deepening their channels, so that the lakes were retained at the same level: I say for a long period, for the shelf on Tomhbran, is as plainly marked as that, for instance, in Glen Spean, & the number of buttresses, composed of rounded pebbles, on all the shelves, plainly shows that a long lapse of time was requisite for their formation. We know in the case of the ice-lake in the valley of Bagnes (Lyell’s Principles of Geology 7th Edit. p. 198) that the water running over the ice did rapidly deepen its channel: to suppose that the onward movement of the glaciers exactly coun- teracted the deepening the drainage channels, appears highly improbable. (2nd) The existence of shelves, one above the other, in the other districts described by Mr Milne, & apparently with no drainage-outlets at their levels except over the ice itself. (3rd) The improbability of a glacier having blocked up the short & very wide valley or rather hollow of Kilfinnin; the Great Glen of Scotland, being low in front of it. (4th ) The apparent existence of a shelf (noticed by Macculloch & not fully examined by myself) at the head of the Spey, beyond the limits of Glen Roy; if this be seen a true shelf, there must have been a blockage at both ends of Glen Roy, which on any lake theory, seems improbable. (5th) The manner in which the hard & solid rocks have been worn at the mouth of L. Treigh at the level of the lowest shelf, & cut off at the head of Lower Glen Roy;—the large accumulations of rounded pebbles in, the buttresses on the shelves, & more especially those round once insular little points of rock (as on Meal Roy) & therefore disconnected with any running stream,—all which facts seem to me to indicate greater destroying power in the waves or currents, than can well be attributed to any lake. (6th) The similarity of the “haughs”10 in Glen Spean to those in other vallies in various countries, which in many cases can be proved to have been of marine origin. And lastly the huge buttresses, projecting from the hill-sides at every successive level beneath the lowest shelf in Glen Spean and built of stratified rounded pebbles, for these afford good evidence that water receded very slowly from this valley, standing apparently for centuries at every successive foot in height, so slowly that I cannot believe the recession was caused by the waters of a lake escaping over a barrier of ice.) (Weighing, as well as I am able, these difficulties on both hands, I am still inclined to abide by the marine theory. I do not think my remarks on the non-extension of the shelves over the entire country, & their coincidence in height with the outlets quite so unsatisfactory as Mr Milne does. With respect to this coincidence in height, I would add one hypothetical remark, namely that it has long appeared to me probable, that a slowly rising surface becomes stationary, not at once, but by the movements becoming excessively slow; & if this be the case, it would greatly favour, the cutting down rocky isthmus between islands, when brought within the tidal action in a rising archipelago, to nearly the level which the land held during the quite stationary periods. On the other hand, the outlets at head of Glen Roy & at Kilfinnin appeared to me, (but perhaps I was mistaken) to be brought up to the exact level of the shelves, by having been silted up. If the marine theory be found to be erroneous, then I believe that the theory of Agassiz & Buckland will prove the true solution, & that Macculloch, Sir Thomas Dick Lauder, & Mr Milne are all so far right that the shelves were formed round the shores of lakes & not of arms of the sea.) (The points which appear to me to require further examination or corroboration (& I hope I shall be excused the presumption in enumerating them), are (1.). The nature of the apparent shelf by Loch Spey & beyond Glen Roy: what is its height relatively to the upper one in Glen Gluoy, & to a mound in Glen Turret, mentioned by Macculloch, above the main upper shelf. (2.) In Glen Glastig, are the rocks simply waterworn, or also scratched or polished as if by a glacier or by drifting icebergs: are such worn surfaces confined to the actual bed of the valley or do they reach some way up the sides? Can they be observed in the bed for the whole 212 vertical feet of fall in this valley? Similar points to be attended to at the heads of Glen Roy, Gluoy & the pass of Muckul. Mr Milne has described knolls of rock, at the outlet of Glen Gluoy, waterworn at their bases & rough above: can this be accounted for by their bases alone having been protected by detritus from disintegration: this appears to me most important. Knolls, such as those described by Mr Milne, low down in the bed of the valleys, which forms the outlet of Glen Glastig, would offer convincing evidence that the valleys had been occupied by river & not by a tidal strait. Can the worn rocks at the entrance of Loch Treig, on a level with the shelf, be accounted for by glacier action: they appeared to me, (at that time totally ignorant regarding glaciers) to be the effect of a strong eddying current. (3rd) Is there any trace of a terminal moraine (though partially obliterated by the supposed ice-lake) or of glacier action, such as perched boulders &c at the mouths of Glen Roy, Collarig, Glen Gluoy & Kilfinnin, & likewise in the valleys, where Mr Milne discovered shelves. (5th) Is there any gap in the mountains near the mouth Glen Gluoy, which might have served as an outlet for the waters, when they stood at the level of Mr Milne’s lower shelf in the valley? The same observation with regard to the intermediate shelf of Tomhbran. (6th) M. Agassiz believes that he saw three shelves on the south side of the Spean in front of Glen Roy: do such exist? I saw lines there, & believed that they were sheep-tracks. 7th The nature & height of the shelves mentioned by me (p. 45) on the authority of Sir David Brewster, low down the valley of the Spey.11 (8th) The height of the shelves in Lochaber above the level of the Sea (for my Barometrical measurements made me doubt the accuracy of those by Macculloch) & of the successive shelves, one above the other)—for the sake of comparison with those described as existing elsewhere by Mr Milne. (9th). The horizontality of the shelves, taken in a line at nearly right angles to the Great Glen of Scotland, on probable axis of elevation of the country: I believe the levels have hitherto been taken exclusively in Glen Roy, in a line approximately parallel to the Great Glen. I may remark, that if the shelves should ever be proved to deviate from the horizontal, it would not demonstrate their marine origin, as the advocates of the lake theories, might argue, that Scotland within late times had been uplifted unequally. On the other hand, if they shd be proved strictly horizontal, the vast plains, which in many parts of the world have been upraised above the level of the sea, to my mind render it not improbable that some districts 20 or 30 miles across should have been uplifted with perfect equability.) Charles Darwin


Dated by the relationship to the letter to David Milne, 20 [September 1847]. This letter to the Scotsman was not published (see letter to Charles Lyell, [11 October 1847]).
Charles Maclaren, who was particularly interested in geology and had earlier corresponded with CD (see Correspondence vol. 2), which is perhaps the reason why CD chose to send his letter to the Scotsman.
Milne 1847b.
‘Observations on the parallel roads of Glen Roy’, Collected papers 1: 87–137.
Glen Glaster (Milne 1847b, p. 4; 1849, p. 398).
‘Notes on the effects produced by the ancient glaciers of Caernarvonshire’, Collected papers 1: 163–71, was originally published in the London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science 21 (1842): 180–8. CD’s reference is to having ascended Moel Tryfan in an unsuccessful attempt to find the shells said to have been discovered at a height of 1192 feet by Joshua Trimmer.
Glen Gluoy.
Charles Lyell had described this Swiss valley in C. Lyell 1847, p. 198. A lake had been formed in it by an avalanche of snow and ice blocking a pass. The melting of the ice-dam in 1818 caused a great inundation.
The mountain of Tombhran. See Collected papers 1: 112.
‘A piece of flat alluvial land by the side of a river, forming part of the floor of the river valley’ (OED).
Collected papers 1: 112.


Comments on article by David Milne ["On the parallel roads of Lochaber" (1847), Trans. R. Soc. Edinburgh 16 (1849): 395–418]. Refers to his paper on Glen Roy [Collected papers 1: 87–137]. Comments on Louis Agassiz’s article ["The glacial theory and its recent progress", Edinburgh New Philos. J. 33 (1842): 217–83]. Cites his own observations on glaciers in N. Wales. Discusses possibility of ice barrier creating lake. Notes objections to theory of an ice barrier. Defends his own theory that the roads are sea-beaches. Suggests questions for further investigation.

Letter details

Letter no.
Darwin, C. R.
Sent from
Source of text
DAR 50: B1–9
Physical description
AdraftS 11pp

Please cite as

Darwin Correspondence Project, “Letter no. 1121,” accessed on 28 July 2016,