A second day, in company with Dr. Hooker, completed the examination of this glacier in 1856; after which I parted from my friends, Mr. Huxley intending to rejoin me at the Grimsel. On the morning of the 20th of August I strapped on my knapsack and ascended the green slopes from Grindelwald towards the Great Scheideck. Before reaching the summit I frequently heard the wonderful echoes of the Wetterhorn. Some travellers were in advance of me, and to amuse them an alpine horn was blown. The direct sound was cut off from me by a hill, but the echoes talked down to me from the mountain walls. The sonorous waves arrived after one, two, three, and more reflections, diminishing gradually in intensity, but increasing in softness, as if in its wanderings from crag to crag the sound had undergone a kind of sifting process, leaving all its grossness behind, and returning in delightful flute notes to the ear. Let us investigate this point a little. If two looking-glasses be placed perfectly parallel to each other, with a lighted candle between them, an infinite series of images of the candle will be seen at both sides, the images diminishing in brightness the further they recede. But if the looking-glasses, instead of being parallel, enclose an angle, a limited number of images only will be seen. The smaller the angle which the reflectors make with each other, or, in other words, the nearer they approach parallelism, the greater will be the number of images observed. To find the number of images the following is the rule:--Divide 360, or the number of degrees in a circle, by the number of degrees in the angle enclosed by the two mirrors, the quotient will be _one more_ than the number of images; or, counting the object itself, the quotient is always equal to the number of images plus the object. In Fig. 2 I have given the number and position of the images produced by two mirrors placed at an angle of 45 deg. A B and B C mark the edges of the mirrors, and 0 represents the candle, which, for the sake of simplicity, I have placed midway between them. Fix one point of a pair of compasses at B, and with the distance B 0 sweep a circle:--_all the images will be ranged upon the circumference of this circle_. The number of images found by the foregoing rule is 7, and their positions are marked in the figure by the numbers 1, 2, 3, &c. [Illustration: Fig. 2. Diagram of an angular reflector.] [Sidenote: ECHOES EXPLAINED. 1856.] Suppose the _ear_ to occupy the place of the eye, and that _a sounding body_ occupies the place of the luminous one, we should then have just as many _echoes_ as we had _images_ in the former case. These echoes would diminish in loudness just as the images of the candle diminish in brightness. At each reflection a portion both of sound and light is lost; hence the oftener light is reflected the dimmer it becomes, and the oftener sound is reflected the fainter it is.