ASTRONOMY
752
interposition of a dark body nearly as large as the star revolving around it ■ but this explanation remained only a surmise until Yogel of Potsdam, by repeated measurements of the motion of Algol in the line of sight, showed that the star is always receding from us before the loss of light, and approaching us afterwards. This was so exactly the result that would be produced by the revolution of such a body as that supposed that no doubt could remain on the subject, and by a legitimate induction the variations of all stars of the Algol type are now assumed to be partial eclipses caused by the interposition of a dark companion moving around them. The number of stars of this class is small, only about twelve being yet known. This may be partly due to the difficulty of detecting variations of the type. Owing to the temporary character of the loss of light, the magnitude of such a star might be observed a great number of times without the variation being detected, since it would be noticed only when the observation happened to be made during the partial eclipse. The following list shows the main particulars relating to stars of this type, their positions, periods, magnitudes, and loss of light when eclipsed :— Variable Stars of the Algol Type. Name. U Cephei /3 Persei X Tauri R Can. Maj S Cancri S Antlia S Velorum S Librse U Coronas R Arse Z Hercules RS Sagittarii Anonymous Anonymous W Delphini
Position. R. A. I Dec. h. m. 0 53 +81-3 3 2 +40-6 3 55 +12-2 7 15 -16-2 8 38 + 19-4 9 28 -28-2 9 29 -44-8 14 56 - 8-1 15 14 +32-0 16 31 -56-8 17 54 +15'1 18 11 -34-1 19 43 +32-5 20 4 +46-0 20 33 ! +17-9
Period. d. h. m. 2 11 49-6 2 20 48-9 3 22 52-2 1 3 15-8 9 11 37-8 0 7 46-8 5 22 24-4 2 7 51'4 3 10 51-2 4 10 12-7 3 23 49-5 2 9 58-6 6 0 8’8 4 13 45-0 4 19 21-2
Magnitude. DiminuUsual Min tion. 79'2 233456-7 89-8 67'3 79-3 5'0 6-2 7'5 8-9 68-0 78-0 6-4 7 *5 10-8 12-9 9-0 119-5 12-
2-1 1-2 0-8 0-8 1-6 0-6 1-5 1-2 11-1 0-9 1-1 21 22-5
Another but analogous type of variation is that of ft Lyras. The characteristic of this type is that in each PL rse Peri°d there are two equal maxima and two type. unequal minima. The variation of light is seemingly continuous. G. W. Meyers {Astrophysical Journal, vii.) has offered a theory of this type of variation which explains the phenomena so precisely that we cannot seriously doubt its correctness. Two unequal bodies, so near together as to be almost in contact, revolve round each other. By their mutual attraction and the centrifugal force they are drawn out into prolate ellipsoids, each rotating in the same period as that of revolution, so that they revolve as a single mass. Each star partially or wholly eclipses the other in each revolution. When the line of centres is at right angles to our line of sight, the two bodies present to us their greatest apparent surface, and therefore send us their maximum of light. As the line becomes oblique from the revolution they are seen more and more foreshortened, and consequently diminish in brilliancy. Were the two bodies of equal surface brilliancy, the apparent magnitude of minimum would be the same whichever was eclipsed. The inequality of the alternate minima indicates a difference in magnitude and brilliancy. It will seem that this type resembles the Algol type in that the variations of light are due to the different aspects under which a pair of stars is seen as they revolve around each other, and especially by one star of the pair eclipsing the other, and not by any actual changes in the bodies themselves. It is now found that the two types merge into each other by insensible gradations. The principal intermediate type is one in which, calling the two stars A and B, A eclipses B and B
eclipses A at each revolution. In such a case, if the orbits were quite circular, it would be impossible by eye-observation alone to distinguish this type from the Algol type, in which the eclipsing body is dark. But if the orbit is eccentric the alternate minima will occur at unequal intervals. Even if such is not the case, the period may be detected by spectroscopic measures of velocity in the line of sight, if only the star is bright enough to admit of its motion being determined in this way.
Systems of Stars, Clusters, and Nebulae. Since 1880 there have been discovered great numbers of double stars, now more properly known as binary systems, which had formerly escaped detection, owing to the difficulty of ascertaining their systems character with telescopes less perfect than those of our time. Few general facts, however, have been brought out by these discoveries. The advance of our knowledge has mainly consisted in the more accurate determination of the elements of the orbits and the times of revolution of those systems whose period of revolution is less than a century. The trend of recent research is towards extending the conception of systems of stars almost indefinitely in two opposite directions. Since, with every increase of telescopic power, closer binary systems of shorter and shorter periods are found in constantly increasing numbers, we cannot set any limit either to the actual minuteness of such objects or 1 to the shortness of 3 the period. Ear within the limit to which 5 telescopic vision can ever extend, such systems are4 now being2 brought 9 2 to light by the spectroscope. These systems appear as a 7 connecting link between certain variable stars on the one 8 hand and the telescopic double stars on the other. Stars 4 9 of the class to which the Algol type of variables belongs 1 will appear to us to vary in brightness only in the wry exceptional cases when the plane of the orbit which one 7 body describes around the other passes so near0 our sun 7 that the one seems to pass over the other, and so causes an eclipse. In all cases, except those in which the line of sight is nearly perpendicular to the plane of the orbit, the revolution of the two bodies around their common centre of gravity will result in a periodic variation of the motion of each in the line of sight. Such a motion may be detected by the spectroscope. If only one of the two bodies is luminous, its motion will be shown by a periodic change in the displacement of its spectral lines. This, so far as yet known, is the usual case. If both are luminous, especially if they do not differ much in brightness, the motion of revolution will be shown by a periodic doubling of the lines. When, were they visible, they would appear to us in conjunction, their spectra are merged into one, which will show nothing unusual; but if one is moving from us and the other towards us, the spectral lines will be displaced in opposite directions, so that all of the lines strong enough to be seen in both spectra will appear double. But while there is thus no well-defined inferior limit to the dimensions of systems of two stars, recent research shows that we cannot set any superior limit to the number of stars which may form a system or to the dimensions of such a system. Considering thdse binary systems whose components are distant from each other, we may, at the first step, meet with a difficulty in deciding whether two stars apparently near each other really form a system, or appear to us together because they happen to be in the same line from us. If their distance is a very few seconds, the presumption would be in favour of a physical connexion. At greater distances the only evidence yet available is found in their relative motion. If the smaller star shares the proper motion of the brighter one, the
