Sir David Brewster was led to his discovery of atmospheric bands by observing that the brownish red vapour of nitrous oxide has the property of absorbing solar light, resolving the spectrum into a series of bright and dark bands, alternating. Professors Daniel and Miller found that bromine, iodine, and chlorous acid do the same, and Sir John Herschel observed a multitude of similar bands in the flame of cyanogen; but Dr. W. A. Miller, who has particularly studied the phenomena of absorption bands, has proved that the colour of a vapour does not necessarily determine the position or even the existence of dark bands. He has shown that some simple substances which do not occasion dark bands produce them abundantly by the absorptive power they acquire when in composition, while lines that are produced by a simple vapour, vanish when it is in combination. Dr. W. A. Miller has proved also that none of the preceding vapours exist in the atmosphere. He computed that if free bromine constituted only one in a thousand million parts of atmospheric air, it would betray its presence by absorptive bands; nevertheless he suspects that there may be some substance in the air that occasions certain unaccountable changes. Possibly ozone, so intimately connected with atmospheric electricity, may produce some unknown effect. The spectra from glowing solids and liquids, such as Drummond’s light, which is incandescent lime, the still more brilliant flame of the electric arc between charcoal points, glowing solid and fused metals, and coal-gas flame, are continuous; the spectra exhibit the seven colours, but they are not crossed by dark rayless lines, because such incandescent substances give off light of all refrangibility. But solids and liquids reduced to glowing vapours, and incandescent gases, only give out rays of certain refrangibilities, which cross their spectra at right angles, as bright lines of various colours and intensities. Each glowing vapour and gas has bright lines on its spectrum peculiar to itself. In order to compare these bright lines with Fraunhofer’s dark lines, solar light is transmitted through one half of the vertical slit in Kirchhoff’s apparatus, and the light of the luminous vapour or gas through the other half. Then by prismatic refraction two spectra are seen in looking through the telescope, the gaseous one immediately below the solar one, and only divided from it by an almost imperceptible dark line. So that the bright lines appear to be continuations of the dark lines if they occupy the same position in the two spectra; if not, the deviation is at once visible. The coincidence or deviation of the bright lines on the spectra of two volatilized substances may be determined by the same method.