CHAPTER XI

SPECTRUM ANALYSIS

If white light (that of the sun, for instance) be passed through a glass prism, namely, a piece of glass of triangular shape, it will issue from it in rainbow-tinted colours. It is a common experience with any of us to notice this when the sunlight shines through cut-glass, as in the pendant of a chandelier, or in the stopper of a wine-decanter.

The same effect may be produced when light passes through water. The Rainbow, which we all know so well, is merely the result of the sunlight passing through drops of falling rain.

White light is composed of rays of various colours. Red, orange, yellow, green, blue, indigo, and violet, taken all together, go, in fact, to make up that effect which we call white.

It is in the course of the _refraction_, or bending of a beam of light, when it passes in certain conditions through a transparent and denser medium, such as glass or water, that the constituent rays are sorted out and spread in a row according to their various colours. This production of colour takes place usually near the edges of a lens; and, as will be recollected, proved very obnoxious to the users of the old form of refracting telescope.

It is, indeed, a strange irony of fate that this very same production of colour, which so hindered astronomy in the past, should have aided it in recent years to a remarkable degree. If sunlight, for instance, be admitted through a narrow slit before it falls upon a glass prism, it will issue from the latter in the form of a band of variegated colour, each colour blending insensibly with the next. The colours arrange themselves always in the order which we have mentioned. This seeming band is, in reality, an array of countless coloured images of the original slit ranged side by side; the colour of each image being the slightest possible shade different from that next to it. This strip of colour when produced by sunlight is called the "Solar Spectrum" (see Fig. 9, p. 123). A similar strip, or _spectrum_, will be produced by any other light; but the appearance of the strip, with regard to preponderance of particular colours, will depend upon the character of that light. Electric light and gas light yield spectra not unlike that of sunlight; but that of gas is less rich in blue and violet than that of the sun.

The Spectroscope, an instrument devised for the examination of spectra, is, in its simplest form, composed of a small tube with a narrow slit and prism at one end, and an eye-piece at the other. If we drop ordinary table salt into the flame of a gas light, the flame becomes strongly yellow. If, then, we observe this yellow flame with the spectroscope, we find that its spectrum consists almost entirely of two bright yellow transverse lines. Chemically considered ordinary table salt is sodium chloride; that is to say, a compound of the metal sodium and the gas chlorine. Now if other compounds of sodium be experimented with in the same manner, it will soon be found that these two yellow lines are characteristic of sodium when turned into vapour by great heat. In the same manner it can be ascertained that every element, when heated to a condition of vapour, gives as its spectrum a set of lines peculiar to itself. Thus the spectroscope enables us to find out the composition of substances when they are reduced to vapour in the laboratory.

In order to increase the power of a spectroscope, it is necessary to add to the number of prisms. Each extra prism has the effect of lengthening the coloured strip still more, so that lines, which at first appeared to be single merely through being crowded together, are eventually drawn apart and become separately distinguishable.

On this principle it has gradually been determined that the sun is composed of elements similar to those which go to make up our earth. Further, the composition of the stars can be ascertained in the same manner; and we find them formed on a like pattern, though with certain elements in greater or less proportion as the case may be. It is in consequence of our thus definitely ascertaining that the stars are self-luminous, and of a sun-like character, that we are enabled to speak of them as _suns_, or to call the sun a _star_.

In endeavouring to discover the elements of which the planets and satellites of our system are composed, we, however, find ourselves baffled, for the simple reason that these bodies emit no real light of their own. The light which reaches us from them, being merely reflected sunlight, gives only the ordinary solar spectrum when examined with the spectroscope. But in certain cases we find that the solar spectrum thus viewed shows traces of being weakened, or rather of suffering absorption; and it is concluded that this may be due to the sunlight having had to pass through an atmosphere on its way to and from the surface of the planet from which it is reflected to us.

Since the sun is found to be composed of elements similar to those which go to make up our earth, we need not be disheartened at this failure of the spectroscope to inform us of the composition of the planets and satellites. We are justified, indeed, in assuming that more or less the same constituents run through our solar system; and that the elements of which these bodies are composed are similar to those which are found upon our earth and in the sun.

The spectroscope supplies us with even more information. It tells us, indeed, whether the sun-like body which we are observing is moving away from us or towards us. A certain slight shifting of the lines towards the red or violet end of the spectrum respectively, is found to follow such movement. This method of observation is known by the name of _Doppler's Method_,[9] and by it we are enabled to confirm the evidence which the sunspots give us of the rotation of the sun; for we find thus that one edge of that body is continually approaching us, and the other edge is continually receding from us. Also, we can ascertain in the same manner that certain of the stars are moving towards us, and certain of them away from us.

[9] The idea, initiated by Christian Doppler at Prague in 1842, was originally applied to sound. The approach or recession of a source from which sound is coming is invariably accompanied by alterations of pitch, as the reader has no doubt noticed when a whistling railway-engine has approached him or receded from him. It is to Sir William Huggins, however, that we are indebted for the application of the principle to spectroscopy. This he gave experimental proof of in the year 1868.