Page:Early Greek philosophy by John Burnet, 3rd edition, 1920.djvu/124

110 54. Anaximander had regarded the heavenly bodies as wheels of "air" filled with fire which escapes through certain orifices (§ 21), and there is evidence that Pythagoras adopted the same view. We have seen that Anaximander only assumed the existence of three such wheels, and it is extremely probable that Pythagoras identified the intervals between these with the three musical intervals he had discovered, the fourth, the fifth, and the octave. That would be the most natural beginning for the doctrine of the "harmony of the spheres," though the expression would be doubly misleading if applied to any theory we can properly ascribe to Pythagoras himself. The word ἁρμονία does not mean harmony, but octave, and the "spheres" are an anachronism. We are still at the stage when wheels or rings were considered sufficient to account for the heavenly bodies.

The distinction between the diurnal revolution of the heavens from east to west, and the slower revolutions of the sun, moon, and planets from west to east, may also be referred to the early days of the school, and probably to Pythagoras himself. It obviously involves a complete break with the theory of a vortex, and suggests that the heavens are spherical. That, however, was the only way to get out of the difficulties of Anaximander's system. If it is to be taken seriously, we must suppose that the motions of the sun, moon, and planets are composite. On the one