In dealing with the Single Stars, which are, like the Sun, self-luminous, can they be proved, like him, to be definite dense masses? [_His_ density is about that of water.] Or are they, or many of them, luminous masses in a far more diffused state, visually contracted to points through their immense distance? Some of those which we have the best means of examining are one-third, or even less, in mass, than he: and if Sirius, for instance, be in this diffused condition, though that would not of itself prevent his having planets, it would make him so unlike our Sun, as much to break the force of the presumption that he must have planets as he has. Again: As far back as our knowledge of our Sun extends, his has been a permanent condition of brightness: yet many of the fixed stars not only undergo changes, but periodical, and possibly progressive changes:—whence it may be inferred, perhaps, that they are not, generally, in the same permanent condition as our Sun. As to the evidence of their revolution on their axis, this has been inferred from their having periodical recurrences of fainter and brighter lustre; as if revolving orbs with one side darkened by spots. Of these, five only can be at present spoken of by astronomers[103] with precision. Nothing is more probable than that these periodical changes indicate the revolution of these stellar masses on their axis—a universal law, apparently, of all the large compact masses of the Universe, but by no means inferring their being, or having accompanying planets, inhabited. The Sun’s rotation is not shown, intelligibly, connected with its having near it the inhabited Earth. In the mean time, in so far as these stars are periodical, they are proved to be, not like, but _unlike_ our Sun. The only real point of resemblance, then, is that of being self-luminous, in the highest degree ambiguous and inconclusive, and furnishing no argument entitled to be deemed one from analogy. Humboldt deems the force of analogy to tend even in the _opposite_ direction. “After all,” he asks,[104] “is the assumption of satellites [attendant planets] to the fixed stars, so absolutely necessary? If we were to begin from the outer planets, Jupiter, &c., analogy might seem to require that all planets have satellites:—yet this is not so with Mars, Venus, Mercury;” to which may now be added the thirty Planetoids—making a much greater number of bodies that have not, than that have satellites. The assumption, then, that the fixed stars are of exactly the same nature as the Sun, was originally a bold guess; but there has not since been a vestige of any confirmatory fact:—no planet, nor anything fairly indicating the existence of one revolving round a fixed star, has ever hitherto been discerned;—and the subsequent discovery of nebulæ; binary systems; clusters of stars; periodical stars; of varied and accelerating periods of such stars,—all seem to point the other way: leaving, though possibly facts small in amount, the