to subsist on distilled water; they must therefore be supported by some organic remains. We have tried to feed them with the stalk and sporules of the protococcus without any conclusive result. Inclosed in a vessel surrounded with a cold mixture, they are unable to support a temperature of -18 centigrade degrees. This degree of cold kills them in a few seconds.

These facts ought to be impressed on the memory of those who have occasion to visit glaciers; they shew that there is a peculiar organisation to be studied in such places. A numerous series of microscopic beings belonging to the vegetable and animal kingdom live and prosper in the bosom of the ice, at a height of 2500 metres above the level of the

sea.

On the other hand, it is not less clearly proved, according to the laws which regulate the movements of glaciers, that the entire mass renews itself at the end of a certain number of years, and contains the mineral, vegetable and animal remains which exist on their surface and in their interior; they all reach in succession the terminal talus, and are again found in the waters which flow from the inferior vaults. Hence the origin of their great impurity, and the grey, milky colour, without transparency, which characterises them.-Comptes Rendus de l'Academie des Sciences, t. xxiv. p. 1093.

On the Parallel Roads of Lochaber, with Remarks on the Change of Relative Levels of Sea and Land in Scotland. By DAVID MILNE, Esq., F.R.S.E., Member of the Wernerian Natural History Society; of the Geological Society of London, &c.

There are few questions in geology which have given rise to so many theories, and so much speculation, as the origin of the parallel roads in the valleys of Lochaber.

In the year 1817, the late Dr MacCulloch gave an elaborate description of them, in a paper read before the Geological Society of London. In the year 1818, Sir Thomas Dick Lauder read before the Royal Society of Edinburgh a paper, full of equally interesting details. Both of these observers suggested, in explanation of the

shelves which mark the mountain sides of these valleys, that they had been occupied by lakes, which, by earthquakes or other violent convulsions, had been drained. This theory was generally received, until, in the year 1839, Mr Darwin, so justly celebrated as a geologist, and an accurate observer, published his views, and pronounced the shelves to have been formed by the sea; an opinion which, besides being rested on proofs derived from the locality, he enforced also by his observation of similar appearances in South America.

Mr Darwin's opinion has received the assent of Sir Roderick I. Murchison, Mr Lyell, and Mr Horner, all successively Presidents of the Geological Society, besides other geologists, both at home and abroad, who are justly regarded as authorities in physical science. Relying on the soundness of their views, I confess that when I went to Glen Roy, in the year 1845, it was with a strong conviction that the lake theory was indefensible; a view to which I was the more inclined, from having studied certain marks along different parts of the Scottish coast, on both sides of the island, which satisfied me that the sea had recently stood at a much higher relative level than at present; and that, in its recession, it had formed, all round our coasts, shelves or beach lines, very analogous to those in the Lochaber valleys. I had not been two days in Glen Roy, before I satisfied myself that these views were inapplicable to the shelves in it and its associated valleys. But I was unable during my visit of 1845, to remain long enough to obtain evidence of the manner in which the lakes had been dammed up, and eventually drained. I therefore resolved to defer the farther consideration of the subject, until I could pay a second visit. This I accomplished in September 1846, when I spent a week in the examination.

In the following paper, I shall attempt to explain my reasons for thinking Mr Darwin's theory inadmissible, and to point out the manner in which, as it appears to me, the lakes were drained,not as supposed by Dr MacCulloch and Sir Thomas Dick Lauder, by convulsions of nature, but by the gradual operation of ordinary causes.

Though it is the principal object of this paper to account for the formation of the Lochaber shelves, there are no views regarding them which can be suggested, which have not a more general bearing, and the soundness of which may be tested by evidence supplied from other sources. Former writers, accordingly, and especially Mr Darwin, have felt it to be necessary, after giving their explanation of the parallel roads, to shew, that the principles on which it rests, are, at least, not inconsistent with any established truths in other branches of geology.

I shall not shrink from subjecting the Lake theory, which I have to submit, to a similar ordeal; and the more so, as I feel satisfied that it receives great support from geological considerations now held to be well established.

As the whole details of the parallel roads have been fully de

scribed by former writers, I shall limit myself to points on which I have obtained new information, or with regard to which doubts have been expressed.

1. One of the points of the class last referred to, is the absolute horizontality of the shelves. Mr Darwin, referring to Sir Thomas Dick Lauder's observations on this point (p. 76.), hints at the possibility of errors and omissions in the calculation. M. Bravais, in his paper on the lines of former sea-level in Finmark, suggests, "that an accurate geodetic levelling should be applied in the case of the doubtful lines in Scotland," evidently refering to Glen Roy. Mr Horner, the president of the Geological Society, in his last year's address, observes; "Mr Darwin's explanation of the parallel roads of Glen Roy, that they are ancient sea-beaches, appears to be now generally accepted; and it would be most interesting, if it were ascertained by exact levellings, such as those of M. Bravais, whether they really are parallel." Similar doubts had been expressed by Sir R. I. Murchison, Mr Horner's predecessor, in his anniversary address of 1843; in support of which, he refers to the concurrent opinion of M. de Beaumont and Professor Phillips.

In accordance with the doubts expressed by these authorities, the Geological Section of the British Association, at their last meeting, agreed on an address to Her Majesty's Government, requesting them to cause the parallel roads of Lochaber to be examined by the officers of the Ordnance Survey, to ascertain their supposed horizontality.

I have no doubt that the result of this official survey, if made, will be to establish the absolute horizontality of the shelves. In August 1844, Mr D. Stevenson, at my request, was so obliging as to examine them, and the conclusion at which he arrived, is explained in a letter to me, from which I make the following extracts:-" I have had a number of levels taken, the particulars of which I shall give you afterwards. The result, I think, leaves no doubt as to the perfect horizontality of the roads." The glen is much more extensive, both as regards length and breadth, than I anticipated, and the height of the roads above its bottom is also very considerable, and any thing like a series of cross sections, referred to the same datum, would be a work of very great magnitude; a month, I should say, would not complete it. The whole we have been able to do, therefore, is to test the uniformity of the levels of the different roads, by viewing them with a good instrument from several points, as was done by Sir Thomas Dick Lauder; and, in addition to this, a section was made along the middle road, where it is pretty well defined from Glen Turret downwards, for a distance of nearly 3 miles, and throughout that stretch, the road was found to be perfectly horizontal." "If I had seen that any thing further could be done, I would have left my assistants for a few days longer; they were there a week."

...

These observations of Mr Stevenson, whose professional accuracy is undeniable, confirming, as they so completely do, the result of Sir Thomas Dick Lauder's survey (and he, too, was aided by an engineer,) leave no doubt in my mind as to the horizontality of the roads. It is scarcely necessary to refer to any farther and weaker testimony on the subject. But it may be proper to add, that during the two occasions when I visited Glen Roy, I had a pocket-level with me, which I constantly used; and that on the last visit I was accompanied by Mr R. Chambers of Edinburgh, who had a larger spirit-level, and we never could detect any deviation from horizontality.

2. There is a point of some importance bearing on the theory of the shelves, about which former observers have disputed. MacCulloch found by his barometric observations, that the Glen Gluoy uppermost shelf is 12 feet above the highest in Glen Roy; but he attributed this difference to errors of observation, and his theory in regard to the formation of the shelves proceeds expressly on the assumption that these shelves are precisely on the same level. Sir Thomas Dick Lauder mentions, however, that Mr M'Lean, the engineer who assisted him, made the Glen Gluoy shelf 12 feet above that in Glen Roy, whilst Sir Thomas himself made it 15 feet. According to the observations made by myself and Mr Chambers last September, the difference is much greater. By levelling, we made it 29 feet; by joint barometric and sympiesometer observations, I made it 23 feet.

3. Whilst on the subject of Glen Gluoy, I may mention that I discovered in it a second shelf, which the barometer shewed to be 200 feet, and the sympiesometer 213 feet, below the level of the one before referred to. I detected it first immediately above the mouth of Glen Fintec. It is traceable on both sides of the glen, and for several miles upwards.

4. There is a circumstance of great importance, in the theory of these roads, on which I was so fortunate as to obtain farther information. I allude to the fact, that most of the shelves are coincident with some summit level, so as to admit of the waters flowing over that level as over a lip. Thus the uppermost shelf of Glen Gluoy No. 1, in Sir Thomas Dick Lauder's Memoir, is (as he explains) exactly coincident with the water-shed ridge which divides Glen Gluoy from Glen Roy, so that the waters (whatever they were) which stood at that height and formed the beach No. 1, must have flowed out at the head of Glen Gluoy into Glen Roy. In like manner, the uppermost shelf in Glen Roy, No. 2 in Sir Thomas Dick Lauder's Memoir, is (as he also mentions) exactly coincident with the water-shed ridge which divides Glen Roy from the valley. of the Spey; so that the waters which stood in Glen Roy at No. 2 beach, must have flowed over the head of the Glen into Spey valley.

In like manner, the only shelf which occurs in Glen Spean, No. 4 in Sir Thomas Dick Lauder's Memoir, is exactly coincident with, or rather is a few feet above, the pass of Mukkul at the head of Loch Laggan, through which pass, the waters standing at the level of No. 4 must have flowed eastward into Spey valley. These coincidences, as Mr Darwin admits, are so remarkable, that they must (I use his own words) be intimately connected with the origin of the shelves; although such relation is not absolutely necessary, inasmuch as the middle shelf of Glen Roy is not on a level with any water-shed." (P. 43.)

66

The middle shelf here alluded to is No. 3 in Sir Thomas Dick Lauder's list. The discovery which I made, was its exact coincidence with a water-shed at the head of Glen Glaster, a glen which, though branching up from Glen Roy near the bottom of it, oddly enough does not appear to have been visited, and certainly not to have been described, by any former observer.

Shelves 3 and 4 are the only shelves which enter and run up this glen. Sir Thomas Dick Lauder's map inaccurately represents shelf 2 as marking it on both its sides. Shelf 2 stops, however, on both sides of Glen Roy a little to the eastward of, or above the mouth of Glen Glaster.

In following shelf 3 to the head of this glen, I found that it was there lost in a low mossy flat. A little beyond this flat, and a few feet below the summit-level, an old river-course can be distinctly traced down a slope towards Loch Laggan. It has a rocky bed, over which a great body of water had evidently flowed at some former period. The breadth of the rocky bed is from 30 to 40 feet; the knolls of rock are from 2 to 5 feet high, and amongst them are rounded blocks of stone, such as occur in all great Highland rivers, I traced this rocky channel for about a mile towards Loch Laggan; and I afterwards found the place where it had discharged its waters into Loch Laggan, when that loch stood at shelf 4. It is marked by a huge delta, forming a projecting buttress at the level of that shelf, and bulging far beyond the general side of the Laggan valley.

On examining the rocky knolls attentively in this ancient rivercourse, I found that the smooth faces were all towards Glen Glaster, and the rough faces in the opposite direction, affording proof, if such were needed, that the stream which flowed there had come from Glen Glaster.

A small rivulet trickles now among the rocks, infinitely too feeble to have produced the appearances.

It is now, therefore, established, not only that the whole of the 4 shelves of Lochaber are coincident with water-sheds respectively, but that a great body of water had filled Glen Glaster, and of course Glenroy, the outlet of which was down this ancient rever