The color of this system of gray plains is far from being uniform. In general appearance it is of a bluish gray, but when observed attentively, large areas appear tinted with a dusky olive-green, while others are slightly tinged with yellow. Some patches appear brownish, and even purplish. A remarkable example of the first case is seen on the surface, which encloses within a large parallelogram the two conspicuous craters, Aristarchus and Herodotus. This surface evidently belongs to a different system from that of the Oceanus Procellarum surrounding it, as, besides its color, which totally differs from that of the gray formation, its surface shows the rugged structure of the volcanic formation. When the Moon is full, some very curious white, luminous streaks are seen radiating from different centres, which, for the most part, are important craters, occupied by interior mountains. The great crater Tycho is the centre of the most imposing of the systems of white streaks. Some of the diverging rays of this great centre extend to a distance equal to one-quarter of the Moon's circumference, or about 1,700 miles. The true nature of these luminous streaks is unknown, but it seems certain that they have their origin in the crater from which they diverge. They do not form any relief on the surface, and are seen going up over the mountains and steep walls of the crater, as well as down the ravines and on the floors of craters. The Moon seems to be deprived of an atmosphere; or, if it has any, it must be so excessively rare that its density is less than of the density of the Earth's atmosphere, since delicate tests afforded by the occultation of stars have failed to reveal its presence. Although no atmosphere of any consequence exists on the Moon, yet phenomena which I have observed seem to indicate the occasional presence there of vapors of some sort. On several occasions, I have seen a purplish light over some parts of the Moon, which prevented well-known objects being as distinctly seen as they were at other times, causing them to appear as if seen through a fog. One of the most striking of these observations was made on January 4th, 1873, on the crater Kant and its vicinity, which then appeared as if seen through luminous purplish vapors. On one occasion, the great crater Godin, which was entirely involved in the shadow of its western wall, appeared illuminated in its interior by a faint purplish light, which enabled me to recognize the structure of this interior. The phenomenon could not be attributed in this case to reflection, since the Sun, then just rising on the western wall of the crater, had not yet grazed the eastern wall, which was invisible. It is not impossible that a very rare atmosphere composed of such vapors exists in the lower parts of the Moon.