Way towards the Nubecula. The former argument presents no difficulty. It is, indeed, rather a confirmation of my views that they afford an easy explanation of what had been held to be a scarcely explicable phenomenon. That the processes of aggregation in portions of space not falling within the galactic annulus should, in certain regions, lead to the exhibition of forms seen within that region, can hardly be considered very wonderful. But, in connection with the second argument, there is a circumstance which deserves to be carefully attended to. Herschel dwells forcibly on the exceeding barrenness of the regions which surround the Nubeculæ. 'The access to the Nubecula Minor on all sides is through a desert,' he says, in one place; and among his notes on this district we find such expressions as a miserably poor and barren region,' 'a region of utter barrenness,' and so on. Now, this peculiarity, so far from confirming Herschel's opinion that the Nubeculæ are disconnected with the sidereal system, is directly opposed to it. One can understand the phenomenon, if one looks on the Nubeculæ as aggregations formed within regions of space belonging to the sidereal system-one would almost expect that the neighbourhood of such regions should be deficient in splendour-drained of stars, so to speak. But, if the Nubeculæ were really distinct systems far beyond the sidereal system, there could be no reason for expecting that their neighbourhood should be more barren than other portions of the sky-still less that it should be oppressively barren. May we not go farther, and say that there is no way of accounting for so remarkable a phenomenon, save on some such hypothesis as I have presented?

But this is not all. It has been well remarked by Sir John Herschel, that the Nubeculæ are so nearly circular as to render the assumption that they are otherwise than globular in figure utterly improbable. It follows, therefore, that the

farthest part of either globe is not much farther off proportionately than the nearest part. Hence the Nubeculæ show us that stars of the seventh and eighth magnitude and irresolvable nebulæ may coexist within limits of distance not differing more in proportion than as nine to ten.' Surely this circumstance is of greater force than Sir John Herschel seems to assume. He says that it must inspire some degree of caution in accepting as certain' the views ordinarily held respecting stars and nebula. To me the fact that stars and irresolvable nebulæ appear intermixed in the Nubeculæ seems to afford decisive evidence of the justice of the views which I have been induced to accept on other grounds. In the face of such evidence, the old theories respecting the universe seem to become wholly untenable.

From the Student for February, March, and April 1869.

WHAT FILLS THE STAR-DEPTHS?

FOR more than two centuries and a half astronomers have studied the depths of heaven with the telescope, piercing farther and yet farther into wondrous abysms of space, gathering clearer and yet clearer information as to the structure of celestial objects, and accumulating an untold wealth of knowledge respecting the habitudes of the great system whereof our sun is a constituent orb. During all this process of research the great end and aim of astronomers has been to extend the range of their instrumental appliances, in order to analyse more scrutinisingly the features of each portion of the celestial depths. Now and then it has occurred to some among their number to endeavour to combine the results which have been gathered together with so much pains; but these attempts have been almost lost sight of amidst the continual accumulation of fresh facts. The efforts made to arrange and systematise our knowledge have been altogether out of proportion with its extent.

And very strangely, when any attempts are made to educe from the labours of observers their proper significance, to reap the harvest which is already ripe, or rather to grind the corn which is already in our garners, the cry is raised that such attempts are fit only for the theorists, that they argue a want of appreciation of the labours of observers, and that we have more to hope from fresh observations than from any process of mere reasoning. Surprising, indeed! that

those who say 'Let us use the observations already made' should be accused of undervaluing observation; and that those who can find no value or significance in past observations should call so eagerly for fresh ones!

I make these remarks because I am about to exhibit certain views respecting the habitudes of interstellar space, which have been formed from the study of the past labours of astronomers. I am fully sensible of the fact that to many I should seem better worthy of a hearing if I nightly timed my watch by the stars, if I had spent a few years of labour in attempting to divide well-known double stars with inadequate telescopic power, or if I had in some other equally convincing manner exhibited my title to be regarded as a member of the now large array of amateur telescopists who work so hard and effect so little and suppose themselves to be practical astronomers. Let me not be misunderstood, however. It is only because I wish to see amateur telescopists engaged on more useful researches, because wish to see them devote a little more consideration than they do now to the thought of advancing astronomy, that I speak slightingly of the modes in which at present they are for the most part wasting time. We want all their help, and more, to advance the interest of our well-loved science; all their telescopic appliances are too few for the work astronomers would like to see them doing.

In studying the heavens, we have always this great difficulty, that we are looking at objects which lie in reality at very different distances, but which appear to lie on the concave surface of a vast spherical enclosure. It seems almost hopeless to attempt by any processes of observation to obtain reliable estimates of the distances of all, save a very few, of the fixed stars. It is not going too far to say that we are tolerably certain of the distance of only one star in the heavens-the star Alpha Centauri. This being the case, and

the heavens spangled with millions of objects at altogether unknown distances, we must look carefully round us for evidence of another kind than that derived from actual measurement, we must look for signs of association, for definite laws of aggregation—if any such exist--and, if possible, we must apply that mode of inquiry from analogy which Sir William, Herschel found in many instances so effective,

And here, as I have mentioned the name of this great astronomer, to whom we owe the first systematic survey of the heavens, and the first attempt to reduce the results of ob servation into law and order, I wish, with extreme diffidence, to point to what I cannot but consider an error of judgment in his selection of the principles which were to guide his survey of the heavens. It appears to me, that it would have been in all respects better had his first processes of stellar observation been directed to gauge the probability that this or that law of distribution prevails in the heavens, rather than to the application of a system of star-gauging, which, if founded on a mistaken assumption was necessarily but a waste of labour. It would have been a misfortune if the unequalled observing qualities of either the elder or the younger Herschel had been misapplied for a single hour; but the possibility that the labours of both these astronomers should have been devoted year after year to a process which (if my views are just) was practically useless, is painful indeed to reflect upon. It is true that the labours of the Herschels have been so numerous and so widely extended, that even the recognition of their star-gaugings as of little real utility would leave the great mass of useful results credited to them almost unaffected; but it would remain none the less a misfortune that labours, which in the case of other men would have worthily filled a lifetime, should have been misdirected.

And yet, when one considers the matter apart from preconceived notions, how inconceivably small the chance