An obvious illustration of these general statements is provided by the history of astronomy. The laws of the motions of the planets, as observed from the earth, were tolerably well known to the Greeks. They had also evolved an explanatory mechanism starting from the premise, which seemed to them to be necessary on metaphysical grounds, that the paths of the planets must necessarily be circles. The earth was the centre of the universe and round this revolved spheres to which the planets were at tached. The retrograde motion of the outer planets was explained by supposing that they were attached to secondary spheres revolv ing about points on the primary spheres which in turn revolved about the earth. This mechanism of cycles and epicycles as an explanation of planetary motion held the field for 18 centuries. Finally the observations of Tycho Brahe provided a test which revealed the falsity of the whole structure. The position of Mars was found to differ from that required by the mechanism of epi cycles by an amount as great as eight minutes of arc. "Out of these eight minutes," said Kepler, "we will construct a new theory that will explain the motions of all the planets." The history of the succeeding century of astronomy need not be recapitulated here. The earth yielded its place as the centre of the universe, and the structure of cycles and epicycles crumbled away. The laws of planetary motion were determined with a precision which for the time appeared to be final. The mechanism underlying these laws was supposed to be a "force" of gravitation. This force was supposed to act between every pair of particles in the universe, its intensity varying directly as the prod uct of the masses of the particles and inversely as the square of the distance separating them—the famous law of Newton.
In science, history repeats itself and, in recent years, the theory of relativity has provided a further instance of the general proc esses we have been considering. Under the Newtonian mechanism every planet would describe a perfect ellipse about the sun as f o cus. and these elliptic orbits would repeat themselves indefinitely except in so far as they were disturbed by the gravitational forces arising from the other planets. But, after allowing for these dis
turbing influences, Leverrier found that the orbit of the planet Mercury was rotating in its own plane at the rate of 43 seconds a century. Various attempts have been made to reconcile this ot, served motion with the Newtonian mechanism. The gravitational forces arising from the known planets were demonstrably unable to produce the motion in question, but it was possible that Mercury's orbit was being disturbed by matter so far unknown to us.
Investigations were made as to the disturbance to be expected from various hypothetical gravitating masses—a planet or a ring of planets between Mercury and the sun, a ring of planets outside the orbit of Mercury, a belt of matter extended in a flattened disc in a plane through the sun's centre, an oblateness greater than that suggested by the shape of the sun's surface, in the arrangement of the internal layers of the sun's mass. In every case the mass required to produce the observed disturbance in the motion of Mercury would have also produced disturbances not observed in the motions of the other planets. The solution of the problem came only with the theory of relativity. Just as Tycho's eight minutes of arc, in the hands of Kepler and Newton, revolutionized mediaeval conceptions of the mechanism of the universe, so Leverrier's 43 seconds of arc, in the hands of Einstein, has revolutionized our 19th-century conceptions, not only of purely astronomical mechanism, but also of the nature of time and space and of the fundamental ideas of science. The history of this revolution is in effect the history of the theory of relativity. It falls naturally into three chapters, a first narrating the building of the earlier physical theory of relativity, a second dealing with the extension of that theory to gravitation, and a third, which is still in process of being written, attempting to include electromagnetism in the physical system presented by the existing theory of relativity.