We shall suppose all space—if we can comprehend what that means—to have been filled with the ether, and the law of attraction to have been in force previous to the time when our operations are supposed to have commenced. These we may consider to have been the first acts of creation, or to have existed from all eternity. Then, in that part of space occupied by our universe—even though it should extend infinitely beyond the reach of our most powerful telescopes—we shall suppose the work of creation to have begun by filling the whole of that space with what are known as the chemical elements, reduced to their atomic state. We do not want to have molecules or particles of matter, or meteorites or meteors; because they involve the idea of previous manipulation or agglomeration, but matter in its very simplest form, if there is any more simple than the atomic. At this stage the most natural idea is to suppose that the whole of this matter was at rest, without motion of any kind, because we cannot understand how motion could be an object of creation, but can very easily see how it might be of evolution; and because, under the law of attraction, matter had the elements of motion in itself. Under that law it is quite possible for us to comprehend that all the suns of our universe could have been formed just as they are, with all their movements of rotation, revolution in the cases of multiple stars, and translation or what is called proper motion. And it is within the bounds of possibility that future astronomers may be able to show how these movements have been brought about, should it ever be possible for them to find out and define with sufficient accuracy what the translatory, or proper, motions are. Then, as for the temperature of this newly created matter, we have no resource left but to suppose that it must have been that of space, whatever that may have been then, even as we have been obliged to say before.

Once created, the atoms of the cosmic matter would immediately begin to attract each other in all directions, and form themselves into groups. At first thought it might be supposed that these groups, and suns formed from them, ought to have been all of the same size, being formed from the same material under the same conditions, but nature, or evolution, seems never to be disposed to produce the same results in its manipulations of matter, whatever they may be. When the water is drawn off from a pond, and the mud left in the bottom of it allowed to dry in the sun, it breaks up into cakes of very various shapes and sizes. No doubt there are physical causes for this being the case, but, though perhaps not altogether impossible, it would be a hard task to find them out. Much more so would it be with originally created matter, and we have only to accept the fact. Moreover, there can be little doubt but that the universe was formed, evolved, according to some design—not at hap-hazard—and that the cosmic matter was created with the distribution necessary to carry out the plan. That the stars differ from each other in magnitude is the best proof of design; for no one can believe that inert matter could determine into what shapes and sizes it could arrange itself. But we have now nothing more to do with the universe, and will confine our operations to the domains of the sun.

Notwithstanding the vagueness and dimness of the description we have been able to give of the part of space to which our work is now to be confined, we can conceive it to resemble in some degree—not a comparatively flat but—a round starfish, with arms more unequal in length, and irregular in position than the kind we are accustomed to see. In such an allotment of space we can easily conceive that the work of attraction and condensation, of the newly created cosmic matter, in forming itself into a nebula, would be most active in the main body; that in the arms, or projecting peaks as they may be called, it would go on more slowly in the direction towards the centre, the quantity being smaller; and that on account of the greater distance in each from the centre of attraction, and of its being more under the influence of the still existing counter-attraction of the matter in the domains of the sun's neighbours, they might become almost, or rather altogether, detached from the more rapidly contracting main body.

We shall, then, suppose that all this has taken place in our incipient nebula. The centre of attraction would at first be the centre of gravity of the whole region occupied by the cosmic matter, which would be ruled in due measure by the projecting peaks, and the indentations or hollows produced in it by the attractive force of the most powerful neighbours; which hollows would gradually disappear as the process of condensation went on, and the main mass would assume the figure of a nebula of some shape. From this stage we may reasonably conclude that, as it was contracting towards the common centre of gravity of all its parts, it would gradually assume a somewhat globular form, and we may now suppose it to have contracted to, say three times the diameter of the orbit of Neptune. Here, then, we may take into consideration what was the interior construction of the main mass which we may now look upon as a nebula; and we have only two states in which we can conceive it to have been. Either that the whole was condensing to the common centre of gravity, in which case its greatest density would be at the centre; or that it was condensing towards the region of greatest mass, in which case its greatest density would be at that region, and its least density at the exterior of the nebula, and also at, or at some distance from, its centre; that is, that the nebula was hollow and without any cosmic matter at all at its centre.

In the first case we must recognise that, from that period of time at least, the cosmic matter that was at, or even near, the centre of gravity then, must be there still all but inert, and being gradually compressed to a greater and greater degree of density. There would, no doubt, be attraction and collisions going on amongst the particles, with condensation towards the centre and production of heat—as long as the particles retained the gasiform condition—which might be increased in activity by the pressure, or superincumbent weight, of the whole exterior mass, but there would be no tendency in them to move outwards—provided their gravitation was always towards the centre; and any motion amongst them would be of the same kind as the vibration of the particles of air shut up in a cylinder and gradually compressed by a piston forced in upon them, and not allowed to escape owing to the sides of the cylinder exerting upon them a pressure increasing exactly in the same proportion as the pressure on the piston was increased. And if this was the case with the matter at or near the centre, it would be the same with that of the whole mass, with the exception, perhaps, of the outer layer, which might act the part of the piston in the cylinder. There could be no motions among the particles, except those of collisions and of falling down towards the centre. The outward impacts of collisions would be less strong than those inwards, on account of gravitation acting against them, and the general tendency of all matter would be to move towards the centre. Even were we to assume that the whole mass was endowed with a rotary motion, the result would be much the same, that is, increasing stagnation of the matter as it approached to the centre. The areolar law teaches us, however, that the increase of condensation at the centre would increase the rotation there; but in that case we have to acknowledge that this increase of rotation would have to be propagated from near the centre to the circumference, which would be by far the most difficult mode of propagation, and we are forced to think of what would be the rate of rotation at the centre, of a nebulous globe, of some sixteen thousand million miles in diameter, required to produce a rotation at the circumference of even once in four or five hundred years; and from that to think of what must be the speed of rotation at the centre of the sun, at the present day, to produce one rotation at the circumference of twenty-five to twenty-seven days. We should also have to think seriously of how the rotary motion was instituted, and we could only appeal either to simple assumption, or to the impact theory, which, applied to a mass of the dimensions of the one we are dealing with, would require more explanation than the whole formation of the nebula itself.

In the second case, that is, looking upon the nebula as a hollow sphere—when it was of the dimensions we have just supposed it to be—we get rid of all the difficulties, and we may add impossibilities, that we encountered in the first case. In such a formation there could be no particle of matter in a state approaching to inertness, not one that could not work its way, through force of attraction and collisions, from the outer to the inner surface of the hollow shell, or vice versâ, or all through and round it and from pole to pole—if it had poles then; it might increase or decrease in density, according to the density of the particles with which it came into collision, as it moved from one place to another, but it would find no spot where it could stand still or be imprisoned. Even arrived at the region of greatest density, it could change places with its neighbours and move all over that region, if it were condemned to remain with one density once it had acquired it; if not, by acquiring or loosing a little density—i.e. by being compressed or allowed to expand a little—it could work its way outwards or inwards, as we have just said, and be as free as the law of attraction would admit of, and as active as that law would oblige it to be. It must be borne in mind that gravitation would act in two opposite directions depending on whether it was acting on the outside or inside of the region of greatest density. We do not go the length of supposing that it could escape altogether from the nebula were its progress outwards; because, as it approached the border, it would meet with plenty of other particles coming in, which would reduce its velocity and prevent its escape. Besides, the law of attraction would take good care to prevent it from passing over to a neighbour nebula or sun.

It may be argued that in the first case—i.e. condensation to the centre—the particles would have the same facilities for changing place, in so far as moving all round the interior of the nebula, or across it, on their way to quasi stagnation, as their densities and the superincumbent weight concentrated and increased; but there could be no motion outwards because the attraction of gravitation would not permit it; nothing could fall upwards, all must gravitate to the centre. Thus the power of motion in the particles would be limited to very much less than half what they would have in the case of the hollow sphere.

It will not do to argue that the increasing heat at the centre would create an upward current. It might create repulsion and prevent the farther-out particles from so soon reaching their final resting or vibrating place, but it could not create an upward convection current of any magnitude; because the colder particles falling down to replace those rising up—that is, if the warmer ones did rise up—being greater in number because occupying greater space, would soon cool down the centre and put an end to the upward current, that is, if it ever came to be set in motion. The greater weight of the greater number would be sure to keep the lesser number in their prison. If any one should say that those nearest the centre would be the heaviest, let him remember that the heaviest liquid or fluid does not rise to the surface. There could be no furnace at the centre to heat the cold particles as they came down to replace those that had just risen up; and if there was, it would be gradually cooled and extinguished. In fact, the centre region would become colder than that immediately outside of it, and so on until the greatest heat would be at the surface of the nebula. Should it be argued that the vastly greater number of particles in the outer regions would help those at the centre to rise up, we agree; but it would be because the attraction would be greater outwards than inwards, as we have shown all along, and not because the pressure forced them out—against itself. But, it must be added, this means that if there was still a plenum at the centre the particles that had once left the centre could never come back again, nor any others to replace them, and that no convection current could ever be formed for carrying heat or matter from the centre, or its immediate neighbourhood, outwards.

In view of the above comparison of the two cases—added as a complement to what we believe we have demonstrated in a former part of our work—we shall adopt the second as being most in harmony with the laws of attraction, and of nature in general, and shall endeavour to describe in some detail, the construction of the nebula out of the matter belonging to the domains of the sun, as we have marked them out.