sun himself began to be. Yet it is with time-intervals measurable by hundreds of millions of years that we have to deal in considering only our earth's history,-nay, two or three hundred millions of years only carry us back to a period when the earth was in a stage of development long sequent to the gaseous condition we are now considering. That the supply of meteoric and cometic matter not gathered in was then enormously greater than that which still exists within the solar domain, appears to me not a mere fanciful speculation, nor even a theoretical consideration, but as nearly a certainty as anything not admitting of mathematical demonstration can possibly be. That the rate of in-gathering at that time enormously exceeded the present rate, may be regarded as certain. That the increase resulting from such in-gathering during the hundreds of millions of years that it has been in operation since the period when the earth first existed as a gaseous mass, must have resulted in adding a quantity of matter forming no inconsiderable aliquot part of the earth's present mass, seems to me a reasonable inference, although it is certain that the present rate of growth continued even for hundreds of millions of years would not appreciably affect the earth's mass.* And it is a thought worthy of consideration, in selecting between Laplace's theory of contraction and the theory of accretion, that accretion being a process necessarily exhaustive, we are able to

* It is, perhaps, hardly necessary to explain that I refer here not to absolute but to relative increase. The absolute increase of mass would amount to many millions of tons, but the earth would not be increased by the billionth part of her present mass

trace it back through stages of gradually increasing activity without limit until we reach that stage when the whole of the matter now forming our solar system was as yet unformed. Contraction may alternate with expansion, according to the changing condition of a forming system; but accretion is a process which can only act in one direction; and as accretion is certainly going on now, however slowly, we have but to trace back the process to be led inevitably, in my judgment, to regard our system as having its origin in processes of accretion,-though it seems equally clear that each individual orb of the system, if not each subordinate scheme within it, has also undergone a process of contraction from a former nebulous condition.

It

In this early gaseous stage our earth was preparing as it were to become a sun. As yet her gaseous globe probably extended beyond the smaller aggregation out of which the moon was one day to be formed. This may be inferred, I think, from the law of the moon's rotation. is true that a moon independently created, and started on the moon's present course, with a rotation-period nearly equalling its period of revolution, would gradually have acquired a rotation-period exactly equalling the mean period of revolution. But there is no reason in nature why there should have been any such near approach; whereas, if we suppose the moon's gaseous globe to have been originally entangled within the outskirts of the earth's, we see that the peculiar relation in question would have prevailed from the beginning of the moon's existence as a

separate body. The laws of dynamics show us, moreover, that although the conditions under which the moon moved and rotated must have undergone considerable changes since her first formation, yet that since those changes took place very slowly, the rotation of the moon would be gradually modified, pari passu, so that the peculiar relation between the moon's rotation and revolution would continue unimpaired.*

In her next stage, our earth is presented to us as a sun. It may be that at that time the moon was the abode of life, our earth affording the supplies of light and heat necessary for the wants of creatures living on the moon. But whether this were so or not, it may be safely assumed that when the earth's contracting gaseous globe first began to have liquid or solid matter in its constitution, the earth must have been a sun so far as the emission of heat and

light were concerned. I must warn you, however, against an undue regard for analogy which has led some astronomers to say that all the members of the solar system have passed or will pass through exactly similar stages. That our earth once gave out light and heat, as the sun does now, may be admitted as probable; and we may believe that later the earth presented the characteristics which we now recognize in Jupiter; while hereafter

* On the theory of evolution some such view of the origin of the moon's rotation must be adopted, unless the matter be regarded as the result of a strange chance. If we believe, on the contrary, that the arrangement was specially ordained by the Creator, we are left to wonder what useful purpose a relation so peculiar and so artificial can have been intended to subserve.

it may pass through a stage comparable with that through which our moon is now passing. But we must remember that the original quantity of matter in any orb passing through such stages must very importantly modify the actual condition of the orb in each of those stages, as well, of course, as the duration of each stage; and it may even be that no two orbs in the universe were ever in the same, or very nearly the same condition, and that no change. undergone by one has corresponded closely with any change undergone by another.

We know so little respecting the sun's actual condition, that even if we could be assured that in any past stages of her history the earth was nearly in the same state, we should nevertheless remain in almost complete ignorance as to the processes to which the earth's orb was at that time subject. In particular we have no means of forming an opinion as to the manner in which the elementary constituents of the earth's globe were situated when she was in the sun-like stage. We may adopt some general theory of the sun's present condition; for example, we may accept the ingenious reasoning by which Professor Young, of Darmouth, N.H., has supported his theory that the sun is a gigantic bubble; but we should be far from having

*"The eruptions which are all the time” (Anglice, ‘always') "occurring on the sun's surface," says Professor Young, "almost compel the supposition that there is a crust of some kind which restrains the imprisoned gases, and through which they force their way with great violence. This crust may consist of a more or less continuous sheet of rain,-not of water, of course, but of materials whose vapours are shown by means of the spectroscope to exist in the solar atmosphere, and whose condensations and combinations are supposed to furnish the

any exact idea of the processes actually taking place within the solar globe, even if we were absolutely certain that that or some other general theory were the true one.

Assuming that our earth, when in the sun-like stage, was a gaseous mass within a liquid non-permanent shell, we can see that as the process of cooling went on the showers forming the shell would attain a greater and greater depth, the shell thus becoming thicker, the space within the shell becoming less, the whole earth contracting until it became entirely liquid; or rather these changes would progress until no considerable portion of the earth would be gaseous, for doubtless long before this stage was reached large portions of the earth would have become solid. Ast to the position which the solid parts of the earth's globe would assume when the first processes of solidification took place, we must not fall into the mistake of judging from the formation of a crust of ice on freezing water that

solar heat. The continuous outflow of the solar heat is equivalent to the supply that would be developed by the condensation from steam to vapour of a layer about five feet thick over the whole surface of the sun per minute. As this tremendous rain descends, the velocity of the falling drops would be increased by the resistance of the dense gases underneath, the drops would increase until continuous sheets would be formed; and the sheets would unite and form a sort of bottomless ocean, resting upon the compressed vapours beneath and pierced by innumerable ascending jets and bubbles. It would have nearly a constant depth in thickness, because it would re-evaporate at the bottom nearly as fast as it would grow by the descending rains above, though probably the thickness of this sheet would continually increase at some slow rate, and its whole diameter diminish. In other words, the sun, according to this view, is a gigantic bubble, whose walls are gradually thickening and its diameter diminishing at a rate determined by its loss of heat. It differs, however, from ordinary bubbles in the fact that its skin is constantly penetrated by blasts and jets from within."