From the moon's incessant changes we perceive that she is in constant motion, and we now further see that whatever these movements may be, the earth and the moon must at present remain at nearly the same distance apart. If we further add that the path pursued by the moon around the heavens lies nearly in a plane, then we are forced to the conclusion that our satellite must be revolving in a nearly circular path around the earth at the centre. It can, indeed, be shown that the constant distance of the two bodies involves as a necessary condition the revolution of the moon around the earth. The attraction between the moon and the earth tends to bring the two bodies together. The only way by which such a catastrophe can be permanently avoided is by making the satellite move as we actually find it to do. The attraction between the earth and the moon still exists, but its effect is not then shown in bringing the moon in towards the earth. The attraction has now to exert its whole power in restraining the moon in its circular path; were the attraction to cease, the moon would start off in a straight line, and recede never to return.
The fact of the moon's revolution around the earth is easily demonstrated by observations of the stars. The rising and setting of our satellite is, of course, due to the rotation of the earth, and this apparent diurnal movement the moon possesses in common with the sun and with the stars. It will, however, be noticed that the moon is continually changing its place among the stars. Even in the course of a single night the displacement will be conspicuous to a careful observer without the aid of a telescope. The moon completes each revolution around the earth in a period of 27·3 days.
In Fig. 24 we have a view of the relative positions of the earth, the sun, and the moon, but it is to be observed that, for the convenience of illustration, we have been obliged to represent the orbit of the moon on a much larger scale than it ought to be in comparison with the distance of the sun. That half of the moon which is turned towards the sun is brilliantly illuminated, and, according as we see more or less of that brilliant half, we say that the moon is more or less full, the several "phases" being visible in the succession shown by the numbers in Fig. 25. A beginner sometimes finds considerable difficulty in understanding how the light on the full moon at night can have been derived from the sun. "Is not," he will say, "the earth in the way? and must it not intercept the sunlight from every object on the other side of the earth to the sun?" A study of Fig. 24 will explain the difficulty. The plane in which the moon revolves does not coincide with the plane in which the earth revolves around the sun. The line in which the plane of the earth's motion is intersected by that of the moon divides the moon's path into two semicircles. We must imagine the moon's path to be tilted a little, so that the upper semicircle is somewhat above the plane of the paper, and the other semicircle below. It thus follows that when the moon is in the position marked full, under the circumstances shown in the figure, the moon will be just above the line joining the earth and the sun; the sunlight will thus pass over the earth to the moon, and the moon will be illuminated. At new moon, the moon will be under the line joining the earth and the sun.
As the relative positions of the earth and the sun are changing, it happens twice in each revolution that the sun comes into the position of the line of intersection of the two planes. If this occurs at the time of full moon, the earth lies directly between the moon and the sun; the moon is thus plunged into the shadow of the earth, the light from the sun is intercepted, and we say that the moon is eclipsed. The moon sometimes only partially enters the earth's shadow, in which case the eclipse is a partial one. When, on the other hand, the sun is situated on the line of intersection at the time of new moon, the moon lies directly between the earth and the sun, and the dark body of the moon will then cut off the sunlight from the earth, producing a solar eclipse. Usually only a part of the sun is thus obscured, forming the well-known partial eclipse; if, however, the moon pass centrally over the sun, then we must have one or other of two very remarkable kinds of eclipse. Sometimes the moon entirely blots out the sun, and thus is produced the sublime spectacle of a total eclipse, which tells us so much as to the nature of the sun, and to which we have already referred in the last chapter. Even when the moon is placed centrally over the sun, a thin rim of sunlight is occasionally seen round the margin of the moon. We then have what is known as an annular eclipse.
It is remarkable that the moon is sometimes able to hide the sun completely, while on other occasions it fails to do so. It happens that the average apparent size of the moon is nearly equal to the average apparent size of the sun, but, owing to the fluctuations in their distances, the actual apparent sizes of both bodies undergo certain changes. On certain occasions the apparent size of the moon is greater than that of the sun. In this case a central passage produces a total eclipse; but it may also happen that the apparent size of the sun exceeds that of the moon, in which case a central passage can only produce an annular eclipse.
There are hardly any more interesting celestial phenomena than the different descriptions of eclipses. The almanac will always give timely notice of the occurrence, and the more striking features can be observed without a telescope. In an eclipse of the moon (Fig. 26) it is interesting to note the moment when the black shadow is first detected, to watch its gradual encroachment over the bright surface of the moon, to follow it, in case the eclipse is total, until there is only a thin crescent of moonlight left, and to watch the final extinction of that crescent when the whole moon is plunged into the shadow. But now a spectacle of great interest and beauty is often manifested; for though the moon is so hidden behind the earth that not a single direct ray of the sunlight could reach its surface, yet we often find that the moon remains visible, and, indeed, actually glows with a copper-coloured hue bright enough to permit several of the markings on the surface to be discerned.
This illumination of the moon even in the depth of a total eclipse is due to the sunbeams which have just grazed the edge of the earth. In doing so they have become bent by the refraction of the atmosphere, and have thus been turned inwards into the shadow. Such beams have passed through a prodigious thickness of the earth's atmosphere, and in this long journey through hundreds of miles of air they have become tinged with a ruddy or copper-like hue. Nor is this property