out that he could; and his discoveries made wireless telegraphy possible. And so you see a practical discovery which you probably regard as one of the most wonderful of modern times can be traced back step by step to this discovery of Roemer's which seems entirely unpractical!
But we are getting rather away from the satellites themselves, in this talk about their eclipses, and I want us to think for a few minutes what a satellite really is. Why should there be these moons, these smaller bodies revolving round the planets? And indeed why should there be planets revolving round the Sun? For the same sort of answer will do for both these questions. The planets are satellites of the Sun just as our Moon is a satellite of the Earth and Jupiter's eight moons are his satellites; in all these cases we have reason to think that parts of the central body have become detached from it to form satellites. First of all remember that all these bodies are rotating—turning round on their axes. We proved that the Earth was rotating by the pendulum experiment in the first lecture; we can see Jupiter rotating by noticing his red spot; we can see the Sun rotating by means of sunspots, as we shall remark in the next lecture. Now, when a body is rotating, the outermost parts tend to fly off, and if they can be detached they will fly off. When a boatman twirls a wet mop, drops of water fly out in all directions. There are several pretty pieces of apparatus in the stores of this Institution which illustrate this principle. In Fig. 45 a weight A slides on a wire BC. A string DEFG attaches it to a hanging weight G. When everything is at rest the
