_A._ But suppose there be no place empty, (for I will defer the question till anon), how can the earth cast off either the air or anything else? _B._ I shall show you how, and that by a familiar example. If you lay both your hands upon a basin with water in it, how little soever, and move it circularly, and continue that motion for a while; and you shall see the water rise upon the sides, and fly over. By which you may be assured that there is a kind of circulating motion, which would cast off such bodies as are contiguous to the body so moved. _A._ I know very well there is; and it is the same motion which country people use to purge their corn; for the chaff and straws, by casting the grain to the sides of the sieve, will come towards the middle. But I would see the figure. _B._ Here it is. There is a circle pricked out, whose centre is A, and three less circles, whose centres are B, C, D. Let every one of them represent the earth, as it goeth from B to C, and from C to D, always touching the uttermost circle and throwing off the air, as is marked at E and F. And if the world were not full, there would follow by this scattering of the air, a great deal of space left empty. But supposing the world full, there must be a perpetual shifting of the air, one part into the place of another. _A._ But what makes a stone come down, suppose from G? _B._ If the air be thrown up beyond G, it will follow that at the last, if the motion be continued, all the air will be above G, that is, above the stone; which cannot be, till the stone be at the earth. _A._ But why comes it down still with increasing swiftness? _B._ Because as it descends and is already in motion, it receiveth a new impression from the same cause, which is the air, whereof as part mounteth, part also must descend, supposing as we have done the plentitude of the world. For, as you may observe by the figure, the motion of the earth, according to the diameter of the uttermost circle, is progressive; and so the whole motion is compounded of two motions, one circular and the other progressive; and consequently the air ascends and circulates at once. And because the stone descending receiveth a new pressure in every point of its way, the motion thereof must needs be accelerated. _A._ It is true; for it will be accelerated equally in equal times; and the way it makes will increase in a double proportion to the times, as hath heretofore been demonstrated by Galileo. I see the solution now of an experiment, which before did not a little puzzle me. You know that if two plummets hang by two strings of equal length, and you remove them from the perpendicular equally, I mean in equal angles, and then let them go, they will make their turns and returns together and in equal times; and though the arches they describe grow continually less and less, yet the times they spend in the greater arches will still be equal to the time they spend in the lesser.