WORLDS RULED BY COLOURED SUNS.

In the heavens there are stars of many colours; for one star differeth from another in glory. But the colours we see with the unaided eye are far less beautiful and less striking than those which are brought into view by the telescope. And among the coloured stars seen by the telescope there are none more beautiful than the coloured pairs of stars. Amongst these we find the most strongly marked contrasts--such combinations as green and red, orange and blue, yellow and purple; then, again, we sometimes see both the companions of the same colour; and yet again we find combinations where the contrast, though not so striking as in the pairs first mentioned, is nevertheless exceedingly beautiful, as when we have gold and lilac, or white and blue, or white and green stars; and, lastly, we find among the smaller companions of double stars such hues as grey, fawn, ash-coloured, puce, mauve, russet, and olive.

It was long thought that at least the more strongly marked colours, in the case of small companion stars, were due merely to contrast. Thus, if the larger of two

stars were orange, the smaller if really white would look blue, as anyone will perceive who will place on a sheet of dark paper a large orange-coloured wafer and close beside it a small white one. In like manner, if the larger star were red, the smaller would look green; if the larger were yellow, the smaller would look purple; and vice versâ only I may as well remark here that while the larger star of a pair is often red, orange, or yellow, it is never blue, green, or purple-at least, such colours are never strongly marked in any leading star of a pair or in any single star.

But the supposition that the colours seen in double stars are due to contrast has been in several instances completely disposed of, by so arranging matters that one star only of a pair is seen at a time. This can readily be arranged where the stars are not very close, and in a great number of cases it has been found that the small star, seen alone, was really blue or green or purple, as the case might be. The experiment was in one case tried in the case of a very close pair, in a very interesting way. The star in question is the ruddy Antares, called also the Scorpion's Heart. This star has a minute green companion, far too close to the red primary star to be seen. alone by any arrangement of the telescope. But advantage was taken by an eminent observer of the passage of the moon over this star. In a moment or two the moon hid the larger star, leaving the other shining alone, and then it was seen that the small star was unmistakably green.

The colours of the double stars, then, are real, so that if we could pay a visit to one of these pairs we should find coloured suns-red, orange, and yellow ruling suns, and green, purple, or blue minor suns, or, as the case might be, lilac, puce, mauve, russet, or olive suns of the smaller sort. Nor must we think of these smaller suns as really small in themselves. It is only by comparison with the leading orbs of unequal pairs that the lesser is called small. In reality it is probable that many of the lesser suns of these double systems are very much larger than all the planets of the solar system together.

But before proceeding to consider the state of affairs in worlds governed and illuminated by double suns, a point as to the colour of these suns has still to be considered. I have said that the colours are real; but it is to be noticed that there are two ways in which this ⚫ may be explained. The light of a star may be actually coloured; or it may be white, but shine through some coloured transparent substance. We may take for illustration of a coloured light of the former kind the red fire, blue fire, and so on, of fireworks. Here the light is really coloured. As an illustration of coloured light of the second kind we may take red and green railway signals. Here we have lights which in one sense may he said to be really coloured, since their colours are not due to contrast or imagination or any like cause. Yet we know that the light is really white, and only appears red or green according as it shines through red or green

we

glass. Now, it is manifestly a very interesting question to decide whether the colours of the double stars are to be explained in one way or the other. Of course know that the coloured stars are not shining through any substance resembling glass. But since it has been ascertained that the light of every star in the heavens (at least every star yet tested) shines through vapours which must to some degree modify its colour, the question is naturally suggested that in the case of the very marked colours of certain double stars the real cause of the colour is to be sought in the nature of the vaporous envelope.

This has, in effect, been found to be the case in the few instances where it has been possible to try the experiment. It will not be difficult to convey an idea of the general principle on which the enquiry depends. When we examine the light of a star through a series of properly arranged prisms of glass, we get a rainbow-tinted streak of light, as in the case of the sun, only of course very much fainter. Also, precisely as in the case of the sun, the star's rainbow-tinted streak-or spectrum, as it is called-is crossed by a multitude of dark lines, which we know to be due to the presence of a number of vapours in the atmosphere of the star. Here I used the word atmosphere, but the reader must not fancy I mean anything

resembling our own air.

Every one of these stars has an amazingly complex atmosphere of glowing vapours, so intensely hot that such substances as iron, copper, and zinc are not merely

melted, but turned into vapour. Now, all stars are not alike as respects these vaporous envelopes. Some have substances in their atmosphere which others have not. And, again, some have apparently a much greater proportion of some substances than of others. Accordingly the dark lines across their spectra are differently arranged. Some have many dark lines in the red part of the rainbowtinted streak, so as in fact to have a great part of the red light cut off, and to shine therefore with a superabundance of the yellow, green, and blue. Such stars have a greenish light. Others have most of their dark lines in the yellow, and so assume a purplish colour. And others have most of their lines in the blue part of the rainbowtinted streak, and so shine with an orange light. And of course it happens in many instances that the dark lines are spread with tolerable uniformity over the whole length, or the greater part, of the rainbow-tinted streak.

The reader will see at once that the method of observation here indicated supplies the means of answering the question whether the colours of the double stars are inherent or caused by the absorbing action of the vaporous envelopes surrounding these stars. The process has been applied very successfully to a beautiful double star called Albireo, or Beta Cygni (that is, the second star of the Swan). This star is seen, even with a small telescope, to be double, and one of the stars, the brightest, is orange, while the other is of a beautiful blue colour. Now, when Dr. Huggins, the eminent spectroscopist, examined the