Although several long ranges of mountains resembling those of the earth exist on the moon, the great majority of its elevations assume the crateriform aspect. Sometimes, instead of a crater, we find an immense mountain ring whose form and aspect hardly suggest volcanic action. But everywhere the true craters are in evidence, even on the sea-beds, although they attain their greatest number and size on those parts of the moon—covering sixty per cent of its visible surface—which are distinctly mountainous in character and which constitute its most brilliant portions. Broadly speaking, the southwestern half of the moon is the most mountainous and broken, and the northeastern half the least so. Right down through the center, from pole to pole, runs a wonderful line of craters and crateriform valleys of a magnitude stupendous even for the moon. Another similar line follows the western edge. Three or four “seas” are thrust between these mountainous belts. By the effects of “libration” parts of the opposite hemisphere of the moon which is turned away from the earth are from time to time brought into view, and their aspect indicates that that hemisphere resembles in its surface features the one which faces the earth. There are many things about the craters which seem to give some warrant for the hypothesis which has been particularly urged by Mr G. K. Gilbert, that they were formed by the impact of meteors; but there are also many things which militate against that idea, and, upon the whole, the volcanic theory of their origin is to be preferred. The enormous size of the lunar volcanoes is not so difficult to account for when we remember how slight is the force of lunar gravity as compared with that of the earth. With equal size and density, bodies on the moon weigh only one-sixth as much as on the earth. Impelled by the same force, a projectile that would go ten miles on the earth would go sixty miles on the moon. A lunar giant thirty-five feet tall would weigh no more than an ordinary son of Adam weighs on his greater planet. To shoot a body from the earth so that it would not drop back again, we should have to start it with a velocity of seven miles per second; a mile and a half per second would serve on the moon. It is by no means difficult to believe, then, that a lunar volcano might form a crater ring eight or ten times broader than the greatest to be found on the earth, especially when we reflect that in addition to the relatively slight force of gravity, the materials of the lunar crust are probably lighter than those of our terrestrial rocks. [Illustration: Western part of the Mare Serenitatis]