ribbon, 10 or 11 feet long-so long, in fact, that only a sinall part of it can be seen in the telescope at once.

288. But to complete the proof that white light is in reality composed of a number of differently coloured lights blended together, it is necessary not only to decompose a beam of white light into its constituents, but to recombine these constituents once more into white light.

This may be done in several ways. Suppose, for instance, that instead of one prism, we use two prisms turned opposite ways, as in Fig. 93. The first of these prisms will separate

the white light into its constituents, and the second will reunite these into white light, so that after the light has passed the combination, it will be a beam similar to that which entered, and pursuing also a parallel path.

But the various colours of the spectrum may be combined so as to form white light without the aid of any optical instrument. For instance, we may take a circular disc, of which the sectors or spaces proceeding from the centre to the circumference are coloured according to the different colours of the spectrum of white light, and in the proper proportion, one being red, another orange, another yellow, a fourth blue, and so on. Now cause this disc to rotate very rapidly, and it will appear to the eye as white, the reason being that, owing to the rotation, the various colours pass so rapidly before the eye that they are blended together, and the impression received is that due to their joint effect; that is to say, the disc will appear white.

LESSON XXXII.-THERMO-PILE.

289. Here it may be asked, how are we to compare together the intensity of light of different colours? and the difficulty of this comparison is magnified if we reflect that there are some rays which are absolutely invisible to the eye, while their heating influence is nevertheless very powerful. Now, how are we to compare together the intensity In order to of a ray of light and of a ray of dark heat? answer this question, let us suppose that the rays of light which we wish to compare are received upon a black screen, which absorbs or stops all kinds of radiant light and heat, and neither reflects back any rays, nor allows any to pass through its substance. What becomes, then, of the energy of these rays, and into what is this converted? We answer, that it is entirely transmuted into absorbed heat; the rays, in fact, are wholly spent in heating the surface upon which they have fallen, and the amount of this heating effect is a true measure of the energy or intensity of these rays, whether they be visible or invisible, coloured or white. If this heating effect be very marked, we may measure it by means of a thermometer, using for the purpose the differential thermometer already described (Art. 171).

290. But it is desirable to have a much more delicate method of measuring heating effect than this. Now this desideratum is supplied by the thermo-pile, the principle of which was first discovered by Seebeck.

In this instrument it is an electric current which produces the result, and we must anticipate so far as to explain that when a circuit composed of two different metals soldered together has one of its junctions heated, an electric current will be produced. We ought likewise to state that a magnetized needle will always, if free to move, place itself at right angles to an electric current. Thus in Fig. 94, let B represent a plate of bismuth, and C C C a plate of copper soldered to the bismuth, while n s is a magnetic needle delicately

swung, and so placed as when at rest to lie along the circuit in the direction of its length. Now heat the junction by a

spirit-lamp, and in virtue of the electric current which the heating gives rise to, the marked or north pole of the needle will be pushed forward as in the figure.

HEAT

291. Now, since our present object is to obtain a very delicate instrument wherewith to measure radiant heat, we must first of all obtain as strong a current as possible; secondly, we must render it as effective as possible in turning a magnetic needle; and thirdly, we must have an arrangement by which the smallest motion of the needle may be rendered visible.

B

G

To produce a strong current we solder together a number of pieces of antimony and bismuth, as in Fig. 95; and if the upper junctions of this arrangement be heated, we have the united effect of the positive currents at all the hot junctions passing through the circuit in the direction of the arrow-heads.

FIG. 95.

But in order to utilize this current we must have a galva nometer in the circuit. The best galvanometer for the purpose is that of Sir W. Thomson. In it we have, as in Fig. 96, a very

small magnet attached to the back of a small circular flat mirror, magnet and mirror being delicately suspended by a very fine thread. This arrangement is surrounded by a number

of circles of the wire which conveys the current; in fact, the wire of Fig. 95 may be supposed to be left loose, so as to be capable of being wrapped many times round the suspension frame of the mirror and needle, taking care to insulate the various folds from each other. Now, if a current pass through this coil, it will tend to make the needle lie at right angles to the plane of the coil; but, on the other hand, the magnetic attraction of the earth will tend to prevent the needle moving; there will thus be a strife between the force of the current and the attraction of the earth, and the result will be that the needle will only move a small distance before it will be stopped.

FIG. 96.

The magnetic force of the earth is, however, overcome by means of a large magnet M (Fig. 97), so placed as to counteract the force of the earth upon the small needle m,

This figure is inserted by the kind permission of the late Mr. C. Becker, of Messrs. Elliot Brothers, London.