It was only natural that as advances were made the instrumental equipment should keep pace with them. Spectroscopes were built on a larger scale; more prisms, which meant greater dispersion, were employed to render the measurements of the lines in spectra more accurate. The growth of our knowledge especially necessitated the making of maps of the lines in the solar spectrum, and in the spectra of the chemical elements which had been compared with it on a natural scale. This was done by Angström, who utilized for this purpose the diffraction grating invented by Fraunhofer, and defined the position of all lines in spectra by their “wave lengths,” in ten-millionths of a millimetre or “tenth-metres.” In 1862 Rutherfurd extended Fraunhofer’s work on the stars by a first attempt at classification. Two years later Huggins and Miller produced maps of the spectra of some stars. Donati demonstrated that comets gave radiation spectra, and Huggins did the same for nebulæ. By these observations comets and nebulæ were shown to be spectroscopically different from stars, which at that time were studied by their dark lines only. Chiefly by the labors of Pickering, the energetic head of the Harvard Observatory, science has been enriched during the later years by observations of thousands of stellar spectra, the study of which has brought about the most marvellous advance in our knowledge. These priceless data have enabled us now to classify the stars not only by their brightness, or their color, but by their chemistry. Next to be chronicled is the application of the so-called Doppler-Fizeau principle, which teaches us that when a light source is approaching or receding from us the light waves are crushed together or drawn out, so that the wave length is changed. The amount of change gives us the velocity of approach or recess, so that the rate of movement of stars towards or from the earth, or the up-rush or down-rush of the solar vapors on the sun’s disk can be accurately determined. A further utilization of this principle is found when the stars are so close together that they appear as one if the plane of motion passes near the earth. A line common to the spectra of both stars will appear double twice in each revolution, when the motion to or from the earth, or, as it is termed, “in the line of sight,” is greatest. “Spectroscopic doubles,” as these stars are called, yield up many of their secrets which otherwise would elude us. Their time of revolution, the size of the orbit, and the combined mass can be determined. To return from the stars to the sun. By the device of throwing an image of the sun on the slit of the spectroscope the spectra of solar spots have been studied from 1866 onward, and a little later the brighter portions of the sun’s outer envelopes, revealed till then only during eclipses, were brought within our ken spectroscopically, so that they are now studied every day. CELESTIAL PHOTOGRAPHY