TWO:Returning to Epicurus, we have next to consider how he obtained the various motions required to bring his atoms into those infinite combinations of which our world is only the most recent. The conception of matter naturally endowed with capacities for moving in all directions indifferently was unknown to ancient physics, as was also that of mutual attraction and85 repulsion. Democritus supposed that the atoms all gravitated downward through infinite space, but with different velocities, so that the lighter were perpetually overtaken and driven upwards by the heavier, the result of these collisions and pressures being a vortex whence the world as we see it has proceeded.163 While the atomism of Democritus was, as a theory of matter, the greatest contribution ever made to physical science by pure speculation, as a theory of motion it was open to at least three insuperable objections. Passing over the difficulty of a perpetual movement through space in one direction only, there remained the self-contradictory assumption that an infinite number of atoms all moving together in that one direction could find any unoccupied space to fall into.164 Secondly, astronomical discoveries, establishing as they did the sphericity of the earth, had for ever disproved the crude theory that unsupported bodies fall downward in parallel straight lines. Even granting that the astronomers, in the absence of complete empirical verification, could not prove their whole contention, they could at any rate prove enough of it to destroy the notion of parallel descent; for the varying elevation of the pole-star demonstrated the curvature of the earth��s surface so far as it was accessible to observation, thus showing that, within the limits of experience, gravitation acted along convergent lines. Finally, Aristotle had pointed out that the observed differences in the velocity of falling bodies were due to the atmospheric resistance, and that, consequently, they would all move at the same rate in such an absolute vacuum as atomism assumed.165 Of these objections Epicurus ignored the first two, except, apparently, to the extent of refusing to believe in the antipodes. The third he acknowledged, and set himself to evade it by a hypothesis striking at the root of all scientific86 reasoning. The atoms, he tells us, suffer a slight deflection from the line of perpendicular descent, sufficient to bring them into collision with one another; and from this collision proceeds the variety of movement necessary to throw them into all sorts of accidental combinations. Our own free will, says Lucretius, furnishes an example of such a deflection whenever we swerve aside from the direction in which an original impulse is carrying us.166 That the irregularity thus introduced into Nature interfered with the law of universal causation was an additional recommendation of it in the eyes of Epicurus, who, as we have already mentioned, hated the physical necessity of the philosophers even more than he hated the watchful interfering providence of the theologians. But, apparently, neither he nor his disciples saw that in discarding the invariable sequence of phenomena, they annulled, to the same extent, the possibility of human foresight and adaptation of means to ends. There was no reason why the deflection, having once occurred, should not be repeated infinitely often, each time producing effects of incalculable extent. And a further inconsequence of the system is that it afterwards accounts for human choice by a mechanism which has nothing to do with free-will.167