The moon's surface is rough and rocky, and displays no marks of the "weathering" that would necessarily follow, had it possessed anything of an atmosphere in the past. This makes us rather inclined to doubt that it ever had one at all. Supposing, however, that it did possess an atmosphere in the past, it is interesting to inquire what may have become of it. In the first place it might have gradually disappeared, in consequence of the gases which composed it uniting chemically with the materials of which the lunar body is constructed; or, again, its constituent gases may have escaped into space, in accordance with the principles of that kinetic theory of which we have already spoken. The latter solution seems, indeed, the most reasonable of the two, for the force of gravity at the lunar surface appears too weak to hold down any known gases. This argument seems also to dispose of the question of absence of water; for Dr. George Johnstone Stoney, in a careful investigation of the subject, has shown that the liquid in question, when in the form of vapour, will escape from a planet if its mass is less than _one-fourth_ that of our earth. And the mass of the moon is very much less than this; indeed only the _one-eightieth_, as we have already stated. In consequence of this lack of atmosphere, the condition of things upon the moon will be in marked contrast to what we experience upon the earth. The atmosphere here performs a double service in shielding us from the direct rays of the sun, and in bottling the heat as a glass-house does. On the moon, however, the sun beats down in the day-time with a merciless force; but its rays are reflected away from the surface as quickly as they are received, and so the cold of the lunar night is excessive. It has been calculated that the day temperature on the moon may, indeed, be as high as our boiling-point, while the night temperature may be more than twice as low as the greatest cold known in our arctic regions. That a certain amount of solar heat is reflected to us from the moon is shown by the sharp drop in temperature which certain heat-measuring instruments record when the moon becomes obscured in a lunar eclipse. The solar heat which is thus reflected to us by the moon is, however, on the whole extremely small; more light and heat, indeed, reach us _direct_ from the sun in half a minute than we get by _reflection_ from the moon during the entire course of the year.