matter of fact, the direction in which the axis pointed should not have been absolutely constant. For a dozen or even for fifty journeys round the lamp no difference of direction need be given to it, but after

FIG. 30.-Owing to precession the celestial poles revolve round the poles
of the ecliptic once in 25,870 years.

the globe has made about 6500 journeys the axis should point to a spectator on the same row as the former one, but situated at one quarter of the distance round the circus; after 13,000 journeys it should point to a spectator exactly opposite the original one,

and in about 26,000 journeys it should have the same direction as at the commencement of the cycle. All the individuals on a particular row would thus become pole stars in the course of 26,000 revolutions of the globe, presuming that they had the patience to watch the performance long enough. This is exactly analogous to what takes place in the case of the earth. The axis of rotation remains almost constant in direction for a few years, but in a cycle of 26,000 years each extremity describes a circle, thus producing the change of pole stars to which reference has been made.

In our illustration, the axis of the globe pointed to a particular row during the whole cycle of change; in other words, its inclination to the plane of revolution was kept constant. Suppose now that the arena were filled with water up to the level of the centre of the lamp. When the axis of the globe pointed to a particular spectator, the equator would cut the water-level at two points, and the line connecting these points. would have a particular direction. As the axis turned round, this line would have its direction altered, and would be brought back to the original direction after 26,000 revolutions of the globe. The inclination of the equator to the water-level represents the inclination of the earth's equator to the ecliptic. This is subject to a variation so small that we can leave it out of consideration. But as the equator has its direction changed, the line of intersection between its plane and the ecliptic plane is also moved round. On this account the points at the extremities of the line the equinoctial points-travel round the ecliptic in the same time that the celestial poles gyrate round the poles of the ecliptic.

It has been shown that the earth is bulged out at the equator. The attraction of the moon and sun upon this mass of superabundant matter causes the precession of the equinoxes. If the earth were a perfect sphere, the axis would remain pointing in one direction throughout time; as it is not, the moon and sun try to pull the equatorial protuberance into the level of the ecliptic. And were the earth not a spinning ball its equator would shortly settle down in the ecliptic plane. But a rotating body has a kind of will of its own; hence the earth obstinately refuses to obey the levelling dictates of the greater and lesser lights, and only responds to their actions by majestically changing its direction of rotation.

When the movements of the earth are understood, the apparent motions of all the heavenly bodies are easily followed. An observer on the sun would see the planets moving round the sky in one direction at different rates. We on the earth do not see the workings of our system from a fixed point, but as from a travelling observatory. It is for this reason that the planets appears to travel round the ecliptic in loops. Take the motion of Jupiter as an example. Our big brother performs a revolution round the sun in about twelve years. Hence, while the earth goes completely round its orbit Jupiter only traverses one-twelfth of his path. Suppose the planet to be seen almost in a line with a particular star at any time. Three months later, the earth will have travelled along one-quarter of its annual journey, and Jupiter therefore appears projected to the side of the reference star. There is an apparent stationary point, and then the planet. moves back to the star until he is in conjunction.

The movement of the earth to the opposite side of its orbit causes the planet to appear on the opposite side of the star. Another stationary point occurs when the earth is moving towards or away from the planet, instead of moving across the direction in which he is seen, and then the apparent movement is towards the original position. But the reference star is not reached. While the earth has been going round the sun, and causing Jupiter apparently to swing to and fro, the planet itself has really moved a short distance eastwards; hence the position in which he is found after a period of this character is to the east of the previous Similar reasoning applies to the motions of any of the planets revolving round the sun in orbits outside that of the earth. (For path of Mars see Figs. 24 and 31.)

one.

In the case of the two "inferior" planets Mercury and Venus, the appearances are different. We have shown that neither of these planets are seen beyond a certain distance from the sun. The cause of this is easily understood when it is remembered that Mercury and Venus revolve round the sun in orbits inside that of the earth. If, for simplicity, we consider the earth not to have an orbital motion, we should see the two inferior planets circling round our luminary, now to the west of him as "morning" stars, now to east as " evening" stars, while they would be sometimes behind. him and sometimes between us and his light. The greatest angular distance between the sun and either Mercury or Venus occurs when these bodies are moving towards or away from the earth, when, in fact, we see them separated from the sun by the radii of their respective orbits. Since the earth is in

FIG. 31.-To explain the apparent path of Mars among the that the motion of the earth from A to C while Mars mo own orbit causes the latter to appear to move in a loop

As soon as the Copernican theory pounded, it was seen that the phenomen Mercury and Venus would be a test For if these planets revolve between the sun, and only their sides are illuminat