If we compare the mechanical actions here traced out with those which take place upon a glacier, I think it will be seen that the analogy between the ripples and the veined structure is entirely superficial. How the structure ascribed to the Glacier de Lys is to be explained I do not know, for I have never seen it; but it seems impossible that it could be produced, as ripples are, by a fixed obstacle which "cleaves a descending stream." No one surely will affirm that glacier-ice so closely resembles a fluid as to be capable of transmitting undulations, as water propagates rings round a disturbed point. The difficulty of such a supposition would be augmented by taking into account the motion of the _individual liquid particles_ which go to form a ripple; for the Messrs. Weber have shown that these move in closed curves, describing orbits more or less circular. Can it be supposed that the particles of ice execute a motion of this kind? If so, their orbital motions may be easily calculated, being deducible from the motion of the glacier compounded with the inclination of the veins. If so important a result could be established, all glacier theories would vanish in comparison with it. [Sidenote: POSITION OF RIPPLES NOT THAT OF STRUCTURE.] There is another interesting point involved in the passage above quoted. Professor Forbes considers that the ripple is occasioned by the variation of speed from the side to the centre of the stream, and that its _inclination_ depends on the ratio of the central and lateral velocity. If I am correct in the above analysis, this cannot be the case. The inclination of the ripple depends solely on the ratio of the river's translatory motion to the velocity of its wave-motion. Were the lateral and central velocities alike, a momentary disturbance at the side would produce a _straight_ ripple-mark, whose inclination would be compounded of the two elements just mentioned. If the motion of the water vary from side to centre, the velocity of wave-propagation remaining constant, the inclination of the ripple will also vary, that is to say, we shall have a _curved_ ripple instead of a straight one. This, of course, is the case which we find in Nature, but the curvature of such ripples is totally different from that of the veined structure. Owing to the quicker translatory movement, the ripples, as they approach the centre, tend more to parallelism with the direction of the river; and after having passed the centre, and reached the slower water near the opposite side, their inclination to the axis gradually augments. Thus the ripples from the two sides form a pair of symmetric curves, which cross each other at the centre, and possess the form _a o b_, _c o d_, shown in Fig. 49. A similar pair of curves would be produced by the reflection of these. Knowing the variation of motion from side to centre, any competent mathematician could find the equation of the