The water constituting our seas, lakes, rivers and oceans; the air constituting our atmosphere; the materials of the earth on which we dwell--clay, rock, mud, granite, metals--all have been analyzed, and their chemical composition accurately determined. It has even been possible to measure the density and weight of our earth, and to calculate its age, from the salinity of its oceans. (Of this more anon.) But when it comes to ascertaining with great accuracy the chemical constitution of distant stars, _that_ seems a feat well-nigh impossible, and unless the process by means of which it is accomplished were explained, it might very well be disbelieved. How, then, can this be accomplished? For our explanation, we must go back to a classical experiment made by Sir Isaac Newton. He proved that white light, when made to pass through a glass prism, is split up into a variety of colors. There are seven primary colors, constituting the visible spectrum. These are red, orange, yellow, green, blue, indigo and violet. We now know that there are both “ultra-violet” and “infra-red” rays, invisible to the eye, above and below the spectrum, but this was not known until long after. The essential fact is that light, when passed through a prism, is split up into its primary colors. THE SPECTROSCOPE The instruments employed were necessarily soon refined, and the modern “spectroscope” resulted,--a piece of apparatus of great delicacy, capable of studying these effects with exactitude. The function of the spectroscope is to receive a sample of light and to separate its different components. In a broad sense, everything that can be seen has a spectrum--flame, blue sky, red hot metal, the sun, the electric spark, etc. We can at once divide these things into two classes, (1) those that are visible because they emit light of their own; (2) those that can be seen only by virtue of their reflecting, diffusing or transmitting light that falls upon them from other sources. The former are called “emission spectra” and the latter “absorption spectra.” Now, when practically any spectrum be examined in this way, it will be seen that certain bands of shadow, or dark lines, cut across the light spectrum, in absorption spectra, these are the things which are studied. Thus, when we observe the spectrum of the sun, or of many of the stars, we find that the spectrum may be described as a continuous spectrum, from which a number of narrow lines are omitted. The lines consequently appear dark on a bright ground. These are called “absorption lines.”