Page images

we may suppose our rod, though formed of a solid material, to be indefinitely elongated without being broken asunder, precisely as if it were perfectly plastic.

Now, there is only one solid substance known for which this continued alternate breaking and reconstruction of continuity and structure are possible; and there is only one condition under which this alternate process is possible with that particular substance. The substance is ice; the condition is, that its temperature must be 32° (Fahr.); and the process of reconstruction is that of which we denote the result by regelation. We consider our imaginary beam analogous to the ice of the glacier. The latter is broken by extension, or its structure may be broken down by compression, but the continuity and structure rise again, restored by regelation.

This explanation completely reconciles the pliability of the glacial mass with the obvious brittle and unyielding character of a hard specimen of glacial ice, by means of an experiment entirely independent of all glacial accumulations of ice, or of the phenomena attending them, and proving the existence of a peculiar and distinctive property of ice, on which the whole explanation rests. The Viscous Theory only explains the difficulty by an appeal to the phenomena which constitute the difficulty itself.

Most of our readers will be aware that there has been of late considerable discussion respecting the priority of the recognition of that pliability of glacial masses of which we have been speaking. M. Rendu, the late Bishop of Annecy, wrote an essay on the · Théorie des Glaciers de la Savoie,' which was printed in Vol. X. of the · Mémoires de la Société Royale Académique de Savoie, 1841. Principal Forbes has made not unfrequent references to this essay, but it still remained till recently almost unknown to the glacialists of this country. It was not so much from any incompleteness, we conceive, in these references, as from the fact of most of the quotations being insulated from each other, that they entirely failed to convey to the reader any adequate idea of the essay itself; and it was not till the publication of Dr. Tyndall's “Glaciers of the Alps' that it became at all appreciated in this country. More copious extracts from it than had before appeared are given in this work, and what is, perhaps, equally important, they are given in more continuous order,* and not in that insulated form in which they had previously appeared. In this essay it is clearly seen that the author had formed a distinct conception of the unequable motions of different parts of a glacier; of its accumulation in particular loca

[ocr errors][merged small]

lities, and its attenuation in others; and, in fact, of all the principal phenomena which indicate a certain pliability in the general glacial mass, sufficient to enable it to mould itself to the local and temporary conditions to which it may be subjected, and to flow on in a manner analogous to that of a river. In Chap. VIII. he remarks :

Il y a entre le Glacier des Bois et un fleuve une ressemblance tellement complète qu'il est impossible de trouver dans celui-ci une circonstance qui ne soit pas dans l'autre. Dans les courants d'eau la vitesse n'est pas uniforme dans toute la largeur ni dans toute la profondeur; le frottement du fond, celui des bords, l'action des obstacles font varier cette vitesse, qui n'est entière que vers le milieu de la surface.

Again, the author says (Chap. X.):

"Je l'ai dit, les glaciers d'écoulement sont des fleuves d'eau solide ; tous les phénomènes des fleuves s'y retracent avec une fidélité qui suffirait pour faire soupçonner leur usage : ils s'élargissent ou so rétrécissent selon la nature des bords.'

Few persons, we imagine, after reading these simple quotations, will doubt the priority of M. Rendu in the recognition of the fact that the motion of a glacier was analogous to that of a river. But it may be said, and something of the kind has been asserted, that this recognition was little more than a vague idea in his mind, which probably never assumed a form sufficiently definite to make it worthy of notice. If it had been so, and he had speculated no farther without seeing the formidable difficulty which had to be encountered in any attempt to reconcile the rigidity of ice with the pliability of a glacier, it might perhaps have been justly said that he had made only an accidental and faltering step in our knowledge of glacial movements. But let us take another quotation from his memoir. He says (p. 84, Vol. X.):

• Il y a une foule de faits qui sembleraient faire croire que la substance des glaciers jouit d'une espèce de ductilité qui lui permet de se modéler sur la localité qu'elle occupe, de s'amincir, de se renfler, de se rétrécir, de s'étendre, comme le ferait une pâte molle. Cependant, quand on agit sur un morceau de glace, qu'on le frappe, on lui trouve une rigidité, qui est en opposition directe avec les apparences dont nous venons de parler. Peut-être que les expériences faites sur de plus grandes masses donneraient d'autres résultats.'

This quotation shows that he saw the difficulty before him, looked it, as it were, full in the face; felt that the scientific weapons of that day were insufficient to vanquish it; obeyed the call of sound philosophy, and stopped. Dr. Tyndall has well


observed that he stopped where many have since stopped also, till more effective means were discovered of overcoming the difficulty in question.

And here, also, we feel ourselves called upon to observe how difficult it is, by means of a few extracts, to produce the conviction derived from the perusal of the whole memoir here spoken of, as to the caution and modesty of the author's philosophical character. Some of his views are such as have not been sanctioned by advancing science, but they are always put forth, when doubtful, with that care and reserve which, we think, appertains to the highest philosophy, and which assuredly; in the case before us, increases our confidence in the author's clearness of view on points of greater certainty. No mere extracts, however favourably chosen, could have given us the same conviction of the strength of M. Rendu's claim to priority in the case we have been discussing, as the entire perusal of his memoir. Scientific justice calls, we think, for the recognition of the Bishop's claim to the clear perception of the pliability of a glacier, while Principal Forbes appears to have had a stronger conviction of its importance. If the latter had subsequently established his Viscous Theory, he might well have afforded to M. Rendu the inferior merit of recognising this mere pliability of the aggregate glacier ; but those who cannot admit that the term viscosity was ever intended to denote a property of ice equivalent to that clearly expressed by regelation, will scarcely regard the Principal as having laid the real foundations of a true theory of glacial motion in the Viscous Theory.

It has already been explained that the mass of a glacier will be subject to certain internal tensions and pressures due to the more rapid motion of its axial portions. The weight of the mass, the form and inclination of the glacial valley, and particular local causes, may also exert a great influence on these internal forces. When the sides of the valley are parallel, or when they are widely divergent, there are certain results, obtained by the mathematical solution of the mechanical problem thus offered to us, which are directly applicable to the actual cases of glaciers. In those cases, also, in which we are concerned with inore irregular valleys, producing more irregular external forces, though, from the want of sufficient data, we may not be able to calculate the amount of the effects produced, we can often ascertain their nature and character, which is usually all that can be practically useful. .

Let us first suppose the glacial valley to be elongated and of the simplest form, with parallel sides and a uniform inclination ; and suppose, also, the upper and lower, surfaces of the glacier to i. Vol. 114.-No. 227.


move with the same velocity. This will not be exactly true, but will lead to no sensible error so long as we restrict ourselves to those upper portions of the mass to which alone we can commonly penetrate. Now we have already seen that, in consequence of the more rapid axial motion, the mass will be extended in some directions and compressed in others, and it has been distinctly proved by accurate investigation, * that in the case before us the internal tension at any point of the mass will be the greatest in a direction pointing towards the upper extremity of the glacier and outwards towards its nearest side, and inclined to the axis at an angle of 45°. To explain the nature of the problem to which we would here direct the attention of our readers, as well as certain of the results deducible from it, we shall borrow a very simple eluci

dation of it from Dr. Tyndall.t In the annexed figure, a b represents the axis of a regular trough-shaped glacial valley, like that above described. The glacier may be represented by a quantity of any plastic or viscous substance partially filling the trough when placed in a horizontal position. Conceive three equal circles to be stamped on its surface near a; then if the end a of the trough be slightly elevated, and the opposite end at 6 be open, the viscous substance contained in it will flow in the direction a b, and it is found that the circle stamped on the axis retains its circular form, while the two lateral circles are transformed into the ovals represented in the figure, the longest axis of each oval

being inclined at an angle of 45° to the axis a b of the trough, while their shorter axes are perpendicular respectively to the longer ones. This manifestly proves that the longer axis of each oval is a line of maximum extension compared with any other line through the centre of the oval, the shorter axis being in like manner a line of maximum compression. In other words, supposing the mass to have cohesive power, the longer axis of each oval must be in a direction along which the tension at the centre of the oval will be greater than in any other direction through that centre; and, likewise, the shorter axis of each oval must be that in which the pressure

[ocr errors]

will be the greatest. At any point on the axis a b there will be neither tension nor pressure resulting from the motion of the mass. These conclusions are in exact accordance with those arrived at long ago by an exact mechanical solution of the problem.

We are hence enabled to explain the formation of the large crevasses, and their general positions; for it is manifest that when the greatest tension becomes greater than the cohesive power, a crevasse must be formed perpendicular to the direction of that greatest tension ; i.e. it must be formed along the minor axis of each oval in the case elucidated by the figure. Consequently, the crevasses in the two marginal portions of the glacier respectively will converge towards each other as they proceed towards its higher end (a). In the cases of converging valleys, the more general solution of the problem shows that the lateral crevasses will always converge towards each other, as just described ; but will make angles greater than 45o with the axis a b. If on the contrary the valley rapidly diverge, the crevasses will diverge as the lines of motion of each part of the mass diverge with the valley itself. The lower extremity of the Rhone glacier presents a most striking example of these diverging crevasses,

It should be remarked that the directions of the crevasses above determined, are those in which they will be originally formed. They remain open for a certain time, and then close up, and the ice on opposite sides of them is regeled into a continuous mass. During this time the more rapid central motion constantly tends to bring them, in parallel-sided or convergent valleys, more nearly to perpendicularity with the axis a b. Still they are observed to lie within the angular limits above stated, with few, or, perhaps, no exceptions. The exact positions in which large fissures will be formed may doubtless depend materially, in many cases, on local conditions ; but this will not usually prevent a dominant general cause from impressing a dominant general character on the resulting phenomena. We have seen, too, that glacial ice appears to have no greater tendency to cleave in one direction than another, so that the directions of the crevasses must be determined by external causes, and not by the internal structure of the glacier.

We have already spoken of the curious phenomena of the veined structure in glacial ice (p. 94). It appears to be closely associated with the directions of greatest pressure above explained. Wherever it exists in the same locality with crevasses, the directions of the latter are stated to approximate very generally to perpendicularity with the superficial curves of structure. This law is usually observable in the marginal portions of glaciers, in


[ocr errors]
« PreviousContinue »