In a small telescope the star Castor will be found double, the components, one of which is brighter than the other, forming a binary system. The fainter of these was found by Belopolsky, with the spectroscope, to be composed of a system of two stars, one bright and the other either dark or not so bright, revolving around each other in a period of about three days. The brighter component of Castor is also a spectroscopic binary, with a period of about nine days; so that the whole of what we see with the naked eye as Castor, is in reality a remarkable system of four stars in mutual orbital movement. Alpha Centauri--the nearest star to the earth--is a visual binary, the component bodies revolving around each other in a period of about eighty-one years. The extent of this system is about the same as that of Sirius. Viewed from each other, the bodies would shine only like our sun as seen from Neptune. Among the numerous binary stars the orbits of some fifty have been satisfactorily determined. Many double stars, for which this has not yet been done, are, however, believed to be, without doubt, binary. In some cases a parallax has been found; so that we are enabled to estimate in miles the actual extent of such systems, and the masses of the bodies in terms of the sun's mass. Most of the spectroscopic binaries appear to be upon a smaller scale than the telescopic ones. Some are, indeed, comparatively speaking, quite small. For instance, the component stars forming [b] Aurigæ are about eight million miles apart, while in [z] Geminorum, the distance between the bodies is only a little more than a million miles. Spectroscopic binaries are probably very numerous. Professor W.W. Campbell, Director of the Lick Observatory, estimates, for instance, that, out of about every half-a-dozen stars, one is a spectroscopic binary. It is only in the case of binary systems that we can discover the masses of stars at all. These are ascertained from their movements with regard to each other under the influence of their mutual gravitative attractions. In the case of simple stars we have clearly nothing of the kind to judge by; though, if we can obtain a parallax, we may hazard a guess from their brightness.