pleted its course, or arrived in the lower valley, whilst the other, which was its companion, has advanced only three-fifths

of the distance, or remains perhaps several miles behind. Thus it has been shewn from multiplied measurements of the most precise and accordant kind, that a series of stones or marks being supposed to be laid across a glacier in the line ABCDEFG; they will be found, after a certain time, in the position abcdefg, after other equal intervals at a'b'c'd'éf'g', and at a"b"c"d"e"ƒ"g", by which time it will be seen that the neighbour particles have entirely changed their relative positions, and that the mass can have no pretension to be called rigid, but moulds itself after the manner that a fluid or semifluid body does in like circumstances-the centre advancing fastest, and, for some space in the centre, nearly uniformly, whilst the retardation produced by the friction of the banks is most intense in their neighbourhood, which is conformable to what we know of the movement of viscous fluids. It is, therefore, no hypothesis, but a simple statement of a demonstrated fact, that the manner of movement of the surface of a glacier is not such as is consistent with the continuity of a rigid body, but that it coincides with the manner of motion of a viscous or semifluid body. Whatever may be the difficulty of conceiving the glacier to be a body thus constituted, the fact admits of no doubt;-the effects of forces

1844.]

ICE MAY BE RENT AND REUNITED.

53

applied on a great scale to bodies, are the best and only conclusive proofs of their real constitution, and worth all molecular theories and minute experiments put together.

If a body be really of a pasty consistence, ductile and plastic like lava or tar, such transpositions taking place in the interior of the mass are effected without any injury to the texture or continuity of the substance. With a degree less of plasticity, a violent separation of the parts may take place, but they will, by juxtaposition, soon reunite and take a new set. With a degree more of rigidity, there must be a permanent bruising and rending of the parts, in order that a semirigid body may assimilate in all its movements to a fluid. It must, therefore, be considered as entirely confirmatory and explanatory of the preceding statements of the seeming plasticity of a body so fragile in its elements as pure ice, that the ice of glaciers is found rent in many parts by the forces tending to dislocation, and that, besides, it contains within itself a testimony to the internal partial movements by which its total motion is effected, in the veined structure already alluded to, occasioned by the varying velocity of the adjacent icy strips A a a' a", B b b b′′, etc. This structure is not exactly parallel to the direction of motion of the ice, for reasons which I have elsewhere stated, but which need not now be adverted to. My present object is to shew, that the rigidity of ice, as a physical fact, cannot con'tradict the mathematical evidence of the manner in which glaciers do move, and that the seeming contradiction is reconciled by showing, that the ice bears permanent traces of the violent strain to which it is subjected, and of the actual bruising and disseverment of its parts, producing a phenomenon otherwise impossible to be explained.

I believe that it is during the progress of the glacier thus subjected to a new and peculiar set of forces depending upon gravity, and which remodel its internal constitution, by substituting hard blue ice, in the form of veins, for its previous snowy texture, that the horizontal stratification observed in the higher part of the glacier or névé, is gradually obliterated.

54

SIXTH LETTER ON GLACIERS.

If, as we cannot doubt, the slower motion near its sides be owing to the retardation which friction occasions, there must necessarily be a re bottom in a similar manner, and the surface of move faster than the strata in contact with 1 which it is even supposable, that, in some case entirely frozen. This retardation may, perhap the lateral retardation, because the slope of the the glacier lies is probably more even, generally its breadth is regular. In fact, so great is the the ground-plan of any compound valley-so frec fering ridges or promontories, the bays forme tributary valleys-and so numerous the gorges -that we cannot properly call the lateral re onward motion of a glacier, friction, but rather tion to the exit of a solid body, which render absolutely essential to its progression. Neverth rior slope of the glacier bed being also irregular, great, must cause a retardation in the lower stra must be continually overtaken by the superior appears to me to be so plain and necessary a the combination of facts which we have to con haps the direct proof of it would not repay the would involve, which would be of the most seri we must not expect to find the difference of ve in the superficial strata, even to a considerable know that the retardation is a maximum near bottom, and that, for the same reason, the mo central part of a glacier is nearly uniform, so of all the part of the ice near the surface be nea These considerations suggest the explanation kindly suggested to me by a most competent pressed himself at the same time persuaded of t viscous theory of glaciers. "How comes it, tha of the different parts of a glacier diminishes from

1844.] UPPER STRATA OF ICE MOVE ALMOST UNIFORMLY.

55

by M. Agassiz in the glacier of the Aar, is stated to have remained vertical for a period of many weeks?" In the first place, the fact of the verticality requires confirmation; for it is difficult to understand how, by means of a plummet, a hole 140 feet deep, and only 3 or 4 inches in diameter, could have its verticality tested. Such bores, so far as I have seen them, are more or less twisted, owing to the softness of the material, and the method of working; and it seems beyond all probability, that a hole of such a depth constructed in the ordinary way, should be either mathematically straight or vertical. I apprehend that the verticality alluded to by M. Agassiz, or his coadjutors, is merely that of popular language, indicated by the boring rods standing vertically outwards when plunged into the hole, which, on account of their flexibility, would not be an indication of the verticality of more than the upper twenty or thirty feet of the bore at the most.*

But, even setting aside this important consideration, the principle of the variation of velocity being chiefly confined to the neighbourhood of the sides and bottom, and the comparatively quiescent and passive state of the central and superficial part, seem sufficient to explain the facts within the reasonable limits of error. The depth of 140 feet appears, from M. Agassiz's own observations, not to exceed ONE SIXTH, at most, of the depth of the glacier of the Aar in that part. Now, let ABC, etc., represent points in the vertical section of the glacier; then, from all that we know of the superficial motion of glaciers, or of the parallel case of rivers whose velocity has been ascertained at different depths, the velocities will vary in some such manner as A a, B b, C c, etc.,-the variation being scarcely sensible at first, and very rapid at the bottom, where the velocity may even be zero, if the curve be prolonged

* Since this passage was written, I have had an opportunity of referring to the description of the experiments of Agassiz in the Bibliothèque Universelle; and I find that there is no evidence whatever of the continued verticality of the bore of 140 feet, which existed (to that depth), I believe, but a few days; the observations of continued verticality, such as they are, applied to small bores only, not exceeding 25 or 30 feet, which, of course, greatly increases the force of the reasoning in the text.-Aug. 1844. [See also page 30 of the present volume.]

56

SEVENTH LETTER ON GLACIERS.

to the point h. But, supposing G to be the glacier, it will be seen how insignificant may be

the variation of velocity between A, the surfa sixth of the depth, during the short period of a even months.

VIII. SEVENTH LETTER on GLACIER VEINED STRUCTURE OF THE ICE. Addresse Dr. WHEWELL, Master of Trinity College, O

Mechanical Considerations tending to explain the Forms of th under different Conditions.

*

*

*

*

SALERNO

You object that the shells produced by the ruptu of the ice caused by excessive friction should b the sides and bottom of the trough of the glacier, i inclined from the sides inwards and forwards towa as in fig. 11, and from the bottom upwards and

Fig. 12.

Fig. 11.

fig. 12.

You will find that I have endeavoured