Another result of the close study of the heavens involved in the making of star catalogues has been the detection of DOUBLE STARS--stars that not only appear to be near together but are really so. Quite a distinct and important department of astronomy has arisen dealing with the continual observation and measurement of these objects. For many double stars are in motion round each other in obedience to the law of gravitation, and their orbits have been computed. {86} Some of these systems contain three or even four members. But in every case the smaller body shines by its own light; we have no instance in these double stars of a sun attended by a planet; in each case it is a sun with a companion sun. The first double star to be observed as such was one of the seven stars of the Plough. It is the middle star in the Plough handle, and has a faint star near it that is visible to any ordinarily good sight. Star catalogues and the work of preparing them have brought out another class--VARIABLE STARS. As the places of stars are not fixed, so neither are their brightnesses, and some change their brightness quickly, even as seen by the naked eye. One of these is called +Algol+, _i.e._ the Demon Star, and is in the constellation Perseus. The ancient Greeks divided all stars visible to the naked eye into six classes, or "+magnitudes+," according to their brightness, the brightest stars being said to be of the first magnitude, those not quite so bright of the second, and so on. Algol is then usually classed as a star of the second magnitude, and for two days and a half it retains its brightness unchanged. Then it begins to fade, and for four and a half hours its brightness declines, until two-thirds of it has gone. No further change takes place for about twenty minutes, after which the light begins to increase again, and in another four and a half hours it is as bright as ever, to go through the same changes again after another interval of two days and a half. Algol is a double star, but, unlike those stars that we know under that name, the companion is dark, but is nearly as large as its sun, and is very close to it, moving round it in a little less than three days. At one point of its orbit it comes between Algol and the Earth, {87} and Algol suffers, from our point of view, a partial eclipse. There are many other cases of variable stars of this kind in which the variation is caused by a dark companion moving round the bright star, and eclipsing it once in each revolution; and the diameters and distances of some of these have been computed, showing that in some cases the two stars are almost in contact. In some instances the companion is a dull but not a dark star; it gives a certain amount of light. When this is the case there is a fall of light twice in the period--once when the fainter star partly eclipses the brighter, once when the brighter star partly eclipses the fainter.