star movements. They see one lot going to the centre of the system and another lot coming away: and they have argued that since the first lot seem to be converging to a point their movements are really parallel. They might be parallel just as Harry's real name might be Henry; but they also may be actually moving towards a point as they seem to be, just as Harry's name may be really Harry outright. And when there is a very good reason why they should be moving to a point, because all the stars belonging to the system are all being pulled towards the centre, as we know, then it seems better to take that as the explanation rather than go out of our way to treat the movements as parallel. The same may be said of the outgoing stream: their arrows seem to diverge from a point, and they might be merely a perspective appearance of movements that were really parallel; but it might also be true that the stars are moving outwards from the centre (like the pendulum that passes through the centre because it has got way on), and when there is a good reason for the latter alternative it seems sensible to adopt it.
In mentioning this great fact, that we can divide the stars into two lots, one going one way and one going the other, I have omitted several very important things so as to keep the argument simple. One important thing is who first found it out. It was first announced by Professor Kapteyn of Groningen in 1904, not yet ten years ago: and it reflects the very greatest credit upon him that he made the discovery. Many other people have confirmed it since, and the evidence is so clear that
