LM CH

44

BEFORE entering upon my subject, I must claim your indulgence upon two grounds. A week ago I had not the remotest idea that I was to have the honour of addressing you here this evening; the time which I have had for preparation has, therefore, been exceedingly limited. In the second place, it is my desire (in accordance with the traditions of this Institution) not merely to give a description of the experiments in which I have for the last few months been engaged, but, as far as possible, to reproduce them before you. Now, these experiments are mostly of a very delicate nature. In the quiet of a laboratory-where time is practically unlimited, and where an operation, if it should fail at first, may be repeated an indefinite number of times-success is tolerably certain to be finally obtained; but in exhibiting delicate experiments before an audience, one is working under the most unfavourable conditions, and, in case of failure in the first instance, the attempt cannot generally be repeated. Moreover, the substance with which we are chiefly concerned, selenium, is apparently extremely capricious in its behaviour. This appearance is, of course, really due to our present ignor ance of its properties; but the fact remains that, on account of the great uncertainty of its action, it is a very difficult substance to deal with.

Selenium is a rare chemical element which was discovered in the beginning of the present century. In many of its properties it closely resembles sulphur, and, like sulphur and some other substances, it is capable of existing in more than one form.

The ordinary form is that called vitreous. Selenium in this condition is as absolutely structureless as glass, and in appearance resembles nothing so much as bright black sealing-wax, with, perhaps, somewhat of a metallic lustre; its real colour, however, when seen in thin films, is ruby red. Its melting-point is a little higher than 100° C. In its second modification selenium is crystalline. When in this form its surface is dull, its fracture is metallic (not unlike that of cast-iron), its colour is grey or leaden, and it is quite opaque to light; its melting-point also is considerably higher, being 217° C.

Vitreous selenium, if melted and kept for a certain length of time at a temperature between its own fusing

A Paper read before the Royal Institution of Great Britain, Friday, March 11, 1881.

| point and that of crystalline selenium, will crystallise; and I think I am right in saying, from casual observation, though I have made no experiments to verify the point, that the length of time necessary for crystallisation depends upon the degree of temperature, being proportionately shorter as the temperature approaches 217° C. electricity; it is, indeed, an Vitreous selenium is an exceedingly bad conductor of Crystalline selenium is a moderately good conductor, and almost perfect insulator. it possesses this very remarkable property, which has been utilised in the photophone and other inventions, that it conducts better in the light than in the dark, the change in its resistance to the passage of a current of electricity through it varying, according to Professor W. G. Adams, as the square root of the illuminating power. battery by means of two copper wires. The passage of a Let a galvanometer be connected to the two poles of a tion of the magnetic needle; or, if a little mirror is atcurrent of electricity will at once be denoted by the deflectached to the needle, and a beam of light be reflected from indicate the movement of the needle. Let, now, one of it upon a scale, the movement of the spot of light will the wires be cut, and the two ends be joined together by again move, but its deflection will be very much less than a piece of crystalline selenium. The spot of light will it was before, showing that the resistance of the selenium is very much greater than that of the wire. Moreover, if the piece of selenium be alternately exposed to and screened from a beam of light, the deflection will be greater when it is in the light than when it is in the dark, showing a corresponding variation in its resistance. This remark hibited by Mr. Willoughby Smith in 1873. But the effects able property of selenium was first announced and exproduced by the simple arrangement which I have just described are small, and very delicate instruments are required for their observation.

Since that date several devices have been proposed for exaggerating the effect, but they all depend upon the fact that the amount of the variation increases with the extent of the selenium surface acted upon. It has lately been the fashion to call these arrangements "cells," which, in most It has been suggested that they should be termed cases at all events, seems to be a very inappropriate name. "rheostats," a name which well expresses the purposes for which they are generally used, and is less likely to lead to confusion than the other. In deference to custom, however, I shall call them by the usual name.

The simplest selenium cell which could be devised would be made by placing two short pieces of copper wire parlalel to each other, and very near together, and con