collect on the little table in the middle without wanting to go either up or down. Now Mr. Heath turns the table so that the oil is made to spin; and at once you see it flatten out as the hoops did. Now he turns it a little faster and—yes! there you see a drop of the oil flies off to form a satellite!
And so all we have now to consider is—why a planet should spin quicker. It was easy for us to make the oil spin quicker; all we had to do was to ask Mr. Heath to turn the handle a little quicker. But what turns the handle for a planet? Well, the simple reason is that the planet is continually shrinking with the cold, and as it shrinks it automatically spins quicker. We will illustrate that by experiment. In Fig. 50 two wooden bars, CA, CB, are loosely hinged at C. A string APB, passing between the jaws of a spring clothes-peg at P, holds the bars horizontal; and the whole is hung by a string and can be set spinning round it. To make the experiment more striking, there are two leaden weights at A and B. Now if the spring of the clothes-peg be nipped, the string is released and the ends A and B drop into the vertical position (Fig. 51). This is our way of representing the shrinking of the planet. We might make the bars CA and CB actually shorten themselves, but this is not easy to do, and so we make the weights A and B come near the axis in a different way, by dropping them downwards; the main point is that in the first position they are far out from the axis and in the second they are close to it. In the case of a planet, the shrinking may take millions of years; but to save time, we do it in a fraction of a second. Now I will spin the apparatus
