much they hit. Indeed, we will try to make them hit. If several of my young friends will each hold one ball in different directions, and then all let go at exactly the same moment I think we shall get a good crash at the centre when they meet: but if they do not let go at exactly the same moment, then they will miss.[1] Now are you all ready to let go when I say "three"? Don't fling the balls at all, just open your fingers and let them drop. One, two, three!
Some of them hit, you see, but one or two missed: and now you see that after a few turns they seem to avoid one another very cleverly partly because some of them have been knocked a little sideways and do not pass accurately through the centre, and partly because some arrive there at one time and some at another: so that though they are all still oscillating through the centre in a general kind of way, collisions are comparatively rare.
I think the stars in our cluster are moving somewhat in this way—oscillating to and fro through the centre—and I have tried to represent the movements very crudely by means of the kinematograph.
[Exhibition here of the results with kinemato-graph. The opportunity was taken to show also some features of a total solar eclipse; such as fowls going in to roost during the darkness and coming out afterwards.]
I think there are occasional collisions, which perhaps make what we call "new stars": and in
- ↑ A fanciful representation of this little experiment is shown in Fig. 95.
