Now there is a means of taking the light which comes from a body shining either with its own or with reflected light, and analyzing it into its component colours. The spectroscope is the instrument by which this is accomplished. I do not propose to describe here the nature of this instrument, or the details of the various methods in which it is employed. I note only that it separates the rays of different colour coming from an object, and lays them side by side for us,--the red rays by themselves, the orange rays by themselves, and so with the yellow, green, blue, indigo, and violet. And not only are the rays of these colours set by themselves, but the red rays are sorted in order, from the deepest brown-red[11] to a tint of red (the lightest) which must almost be called orange; the orange in order, from orange which must almost be called red to a tint (the lightest orange) which must almost be called yellow; the yellow, from an almost orange yellow to a yellow just beginning to be tinged with green; the green, from an almost yellow green (the lightest) to a green which may almost be called blue (the darkest); the blue, from this tint to the beginning of the indigo; the indigo, from this tint to the first rays of the violet; and lastly the violet, through all the tints of this beautiful colour to a blackish-brown violet, where the visible spectrum ends. All these tints are sorted in order by the spectroscope, just as a skilful colourist might range in due sequence a myriad tints of colour. But this is only true of really white light, such light as comes from a glowing mass of metal burning at a white heat. In other cases (even when the light may seem white to the eye) some of the tints are found, when the spectroscope spreads out the colours for us, to be missing. And we know that this may be caused in two ways. Either the source of light never gave out those missing tints; or, the source of light gave them out, but some absorbing medium stopped them on their way before they reached the spectroscope with which we examine them. There may be cases where we cannot tell very easily which of these is the true cause. But sometimes we can, as the instances I have now to deal with will show you.