Spectrum analysis would also betray the existence of a lunar atmosphere. The solar rays falling on the moon are reflected from its surface to the earth. If, then, an atmosphere existed, it is plain that the solar rays must first pass through such atmosphere to reach the reflecting surface, and returning from thence, again pass through it on their way to the earth; so that they must in reality pass through virtually twice the thickness of any atmosphere that may cover the moon. And if there be any such atmosphere, the spectrum formed by the moon’s light, that is, by the sun’s light reflected from the moon, would be modified in such a manner as to exhibit absorption-lines different from those found in the spectrum of the direct solar rays, just as the absorption-lines vary according as the sun’s rays have to pass through a thinner or a denser stratum of the terrestrial atmosphere. Guided by this reasoning, Drs. Huggins and Miller made numerous observations upon the spectrum of the moon’s light, which are detailed in the “Philosophical Transactions” for the year 1864; and their result, quoting the words of the report, was “that the spectrum analysis of the light reflected from the moon is wholly negative as to the existence of any considerable lunar atmosphere.” Upon another occasion, Dr. Huggins made an analogous observation of the spectrum of a star at the moment of its occultation, which observation he records in the following words:—“When an observation is made of the spectrum of a star a little before, or at the moment of its occultation by the dark limb of the moon, several phenomena characteristic of the passage of the star’s light through an atmosphere might possibly present themselves to the observer. If a lunar atmosphere exist, which either by the substances of which it is composed, or by the vapours diffused through it, can exert a selective absorption upon the star’s light, this absorption would be indicated to us by the appearance in the spectrum of new dark lines immediately before the star is occulted by the moon.” “If finely divided matter, aqueous or otherwise, were present about the moon, the red rays of the star’s light would be enfeebled in a smaller degree than the rays of higher refrangibilities.” “If there be about the moon an atmosphere free from vapour, and possessing no absorptive power, but of some density, then the spectrum would not be extinguished by the moon’s limb at the same instant throughout its length. The violent and blue rays would lay behind the red rays.” “I carefully observed the disappearance of the spectrum of η Piscium at its occultation of January 4, 1865, for these phenomena; but no signs of a lunar atmosphere were detected.”