we have referred to, with the splendour of the great luminary of our system. And it was seen that the great variety of lustre which is observed among the stars is no evidence of any corresponding variety in their real magnitudes, but may be readily explained by the supposition that the stars are placed at different distances from the Sun. Perhaps astronomers in this respect fell into the opposite error, and were too ready to assume diversity of distance as the sole, or almost the sole, explanation of difference of lustre. "The supposition," says a modern writer, "has been usually adopted; and we accordingly consider the stars to derive their variety of lustre almost entirely from their places in the universe being at various distances from us." Here we may apply the first lesson which is taught us by a consideration of the solar system; and, from the analogy of that system, with the infinite diversity of magnitude presented among its various members, we may learn to expect a corresponding diversity among the components of the great sidereal system.

When the considerations which had been applied to the scattered stars visible either to the naked eye or with the telescope, came to be extended to that vast irregular annulus of nebular light called the galaxy, or Milky Way, the most startling conceptions were suggested of the enormous extent of the sidereal system. This remarkable zone had from the earliest ages engaged the attention of astronomers. Long before Galileo had resolved portions of it into stars, Democritus had maintained just views respecting its structure. Manilius also suggested

"An major densâ stellarum turba coronâ
Contexit flammas, et crasso lumine candet,
Et fulgore nitet collato clarior orbis."

Regarded, however, as a zone of suns, this phenomenon acquired a new and astounding significance. If we could suppose a multitude of suns resembling our own to be so closely compacted together as the component stars of the Milky Way appear to be, there must result an inconceivable splendour in those far distant regions; if, on the other hand, the orbs which seem to lie in such close order, are in reality separated by distances comparable with those which separate the Sun from the nearest fixed star, how inconceivably distant must they lie from us, that such intervals as these should be diminished to evanescence! The last supposition has been the one universally accepted by astronomers. Of the two it clearly accounts best for the observed appearance of the galaxy.

I shall presently have occasion to show the probability, if not the certainty, that neither view represents the relations which actually subsist among the stellar components of the Milky Way.

The phenomena presented by this zone of nebulous light are intimately associated with the remarkable researches of Sir William Herschel among the fixed stars. This eminent astronomer, in whom were presented all the qualifications required to constitute a first-class observer, side by side with that power of systematic reasoning on observed facts which so seldom accompanies the highest observing powers, was early fired by the bold ambition to gauge the depths of our sidereal system. He cast aside the notion, which had been held, almost unquestioned, till his day, that the stars of that system extend on every side to an infinite distance. He saw that the existence of a Milky Way affords evidence that the sidereal system has definite bounds; and he quickly grasped at the only method which is available for the determination of its figure. "This great, inspired, and cautious observer," says Humboldt, "first cast the plumb-line into the depths of heaven, to determine the boundaries and the form of the separate cluster of stars which we inhabit." It has been said of him that "he broke through the barriers of the heavens (cœlorum perrupit claustra)," and we shall presently see in what sense these words have been used. But it must not be forgotten that to his labours and to those of his son are due the ideas we at present hold of what those barriers are. He was at once the Romulus and the Remus of astronomy; he marked out the limits of our system, and he showed how man might boldly venture beyond those limits into the domain of the illimitable.

Herschel's method of gauging the sidereal heavens has been long and deservedly the theme of admiration. The boldness and originality of the conception, and the unwearying perseverance with which the laborious processes involved were carried out by the elder Herschel over the northern hemisphere and by the younger over the southern-are unexampled in the history of observational astronomy. Assuming a certain approach to uniformity in the distribution of the stars, and also (for this is very important) that there is no such law of extinction of light in traversing great distances as would prevent a telescope of great power from penetrating the full depths of the system in every direction, it is clear that a very simple process will serve to indicate the relative distance of the observer from different parts of the system's exterior surface. This method and its results have already been discussed in the pages of

the "Intellectual Observer.' ."* We have seen that the labours of the two Herschels seem to show that the stellar system forms a flattish disc of stars, whose central plane corresponds with that of the Milky Way. In one direction this disc is cloven, and those parts of the heavens which lie opposite the two divisions of the stellar disc are occupied by a double stream of milky light.

We have seen also that Herschel soon recognized a complexity in the structure of the sidereal system, which prevented him from regarding the figure of a cloven disc as any save the roughest representation of the galactic system. He saw that in portions of the Milky Way the stars exhibit a tendency to form themselves into clustering groups, and he saw that a tendency of this sort would be quite sufficient to vitiate, not merely those gauges which were made in the direction of the clustering stars themselves, but also those made in neighbouring regions; for where the stars were clustering together, the star-gaugings would indicate a depth not really presented by the sidereal system in that direction, and the regions in which stars were more sparsely strewn, owing to the influence of neighbouring aggregations, would also give false evidence, but of an opposite kind, respecting the depth of the system.

Other peculiarities militating very strongly against the idea of uniform distribution, and sufficing largely to enhance the complexity of the problem which the Herschels have striven to solve, must be reserved for Part II., in which I shall complete the discussion of accepted theories, leaving for Part III. the presentation of the new views which seem to be suggested by modern discoveries.

* See "Notes on Star-streams," "Intellectual Observer" for August, 1867, Figs. 3 and 4.

(To be Continued.)