Saturn, unless there were some powerful forces at work counteracting the tremendous attractive influence of the matter composing their tremendous globes. Were Saturn or Jupiter in the same condition as the earth, as well as made (which we may well believe to be the case) of the same materials, we know that the tremendous pressure to which their substance would everywhere be exposed would result in producing a density far greater than that of the earth with her much smaller mass and relatively much smaller attractive forces. But so far is this from being the case, that the density of Jupiter is but one-fourth, that of Saturn little more than one-seventh, the mean density of our earth.

In the case of the giant planets, then, as in that of the still vaster orb of the sun, we find what theory had led us to anticipate. We find that these orbs coming next in order, in point of size, to the sun, present all the appearance of being in the next stage of orb life. Just as he presents all the characteristics which we expect to find in orbs in the first or glowing vaporous stage of existence, so do these orbs present all the characteristics of the second or fiery stage.

Passing over the third or middle stage, in which we know our earth to be, let us seek for an illustration of the fourth stage, or the stage of planetary decrepitude. We have seen reason to believe that our moon, so much less than the earth in size and mass, is probably either in a state of advanced decrepitude, or actually dead. If there is any orb between the earth and the moon in size and mass much less than the earth and much larger than the moon, we might fairly expect to find that orb in the middle state between the earth and the moon, not so old as the moon, but still a great deal older than the earth.

The moon has a mass Mars has a mass equal

The planet Mars is such an orb. equal to the 81st part of the earth; to nearly one-ninth of the earth's, or nearly nine times as great as the moon's. So that, precisely as the sun exceeds

Jupiter in mass in about the same degree that Jupiter exceeds the earth, so Mars exceeds the moon in mass in about the same degree that the earth exceeds Mars. We could not therefore have, on à priori grounds, a more convenient middle term than Mars, for the stage midway between those through which the earth and the moon are respectively passing.

Now, when we make comparison between the earth and Mars in that feature which we have regarded as characteristic of advanced age, what do we find? On the earth the water surface is equal to about three-fourths of the entire surface of the globe; in Mars the surface occupied by water is but about one-half (rather less, according to the observations of some astronomers). Again, when we examine the seas of Mars in respect to their shape, we find precisely the feature which we were led to expect in a planet which had passed so far beyond the stage in which the earth now is that the seas had become greatly reduced in extent. We find seas like the Baltic and the Mediterranean as characteristic in the case of Mars as they are unusual in the case of the earth. Nine of the seas on Mars have this peculiar shape, which Stanislas Meunier, following myself, has described as bottle-necked.

So far as we can judge, in fine, all the features of Mars, like those of the sun and Jupiter, correspond with the anticipations which we should have based on the planet's mass.

Lastly, we turn to the moon. This planet, we have seen, should be in the state which the earth will reach 200 millions of years hence, if our assumptions as to the duration of the various stages of the earth's existence are correct.1

1 It is worth noticing that in reality it matters very little whether these assumptions are correct or not, so far as our ideas respecting the stage of life reached by the several planets are concerned. If we have over-estimated the duration of the various stages of the earth's existence, we have correspondingly under-estimated the rate at which the changes which characterise these various stages proceed, so that if we have in consequence over-estimated the difference of the durations of the stages

As we have said, it might fairly be expected that at that remote date the earth will be dead, in the sense of being unfit any longer to support life. The moon, then, if our theory be correct, might be expected to be a dead world. Now, what do we find? So far as the disappearance of seas and oceans is concerned, the moon certainly presents every sign of death. There are astronomers, indeed, who consider that there are traces of moisture on the moon, and some few observers (who certainly are not astronomers) who consider that they can trace signs of water-surfaces-small pools and so forth-on the moon. But, as a matter of fact, all astronomers of repute agree that the surface of the moon is absolutely arid-dry, desolate, and dead. Again, so far has the moon progressed towards the final stage of planetary existence, even if she have not absolutely reached it, that no trace whatever of an atmosphere can be recognised on her surface. As she passes over the stars, she obliterates them from our view, with a suddenness which shows that either she has no atmosphere or an atmosphere so tenuous as to have no appreciable effect in softening the edge of the lunar disc, so as to diminish in any perceptible degree the suddenness with which stars are obliterated from our view. There is no trace of any twilight on the moon. The shadows of mountains are black and sharply defined, instead of presenting those half-tints which we know to exist in the shadows of our own mountains. When, standing on some mountain summit at sunrise or sunset, we look into the shadowed valleys, we do not look into absolute blackness; and if we ask where the light comes from which partially

of these planets' lives, the changes taking place during such over-estimated period will not be greater than those really due to the difference correctly estimated, for they will take place at a correspondingly slower rate. And vice versa, if we have under-estimated the difference of the durations of the stages of other planets' lives, the changes taking place during such under-estimated periods will not be less than those really due to the differences correctly estimated, for they will take place at a correspondingly quicker rate.

illuminates those valleys, we find our answer at once. We know what we should recognise if we walked into those valleys. We should find light reaching us from the sky. The valleys could not be in black shadow unless the skies were also black. Seeing, then, that the shadows of the lunar mountains are absolutely black, we learn as certainly as we should if we could actually have visited the moon and walked into those shadowed regions, that the sky seen overhead there would be absolutely black, with the stars shining as brilliantly there in broad daylight as they do with us on the clearest and darkest night. Now, our sky is in reality illuminated air-air which, receiving the solar rays, refracts and reflects it towards the earth, so that to say the sky is black as seen from the moon is equivalent to saying that either there is no air, or the air is so thin that it reflects no perceptible portion of the solar rays.

Waterless, then, and airless, the moon must be regarded as dead. She tells the same story as the sun, the giant planets, our earth, and Mars. She is in the stage corresponding to her size.

We have then, in the solar system, examples of all the five stages of a planet's history-the first or glowing vaporous stage, the second or fiery stage, the stage of mid-life, the stage of decrepitude or old age, and finally the stage of death, when no life can any longer be borne on the planet's dry, desolate, and dead surface.

THE SUN AS A PERPETUAL MACHINE.

AMONG the problems which have proved most perplexing to astronomers and physicists, there are few which surpass in difficulty the problem of the conservation of solar energy. The mighty orb of the sun pours forth in each second of time as much heat as would come from the burning of 16,436 millions of millions of tons of the best anthracite coal. Yet of all this tremendous radiation of heat all the planets together receive less than one 230,000,000th part. When we consider this it seems at first view as though there were some degree of truth in the saying that in the universe we find Nature upsetting a gallon to fill a wine-glass.'

In company with this great mystery of seeming waste comes the yet more difficult problem, How to explain the apparent continuance of solar light and heat during millions of years. We know from the results of geological research that the earth has been exposed to the action of the solar rays with their present activity during at least a hundred million years. Yet it is difficult to see how on any hypothesis of the generation of solar heat, or by combining together all possible modes of heat-generation, a supply for more than 20 millions of years in the past and a possible supply for as long a period in the future can be accounted for.

It is well known, of course, to all who are likely to read these lines that Dr. Siemens is the inventor of what is called the regenerative furnace, in which the heat, which in ordinary furnaces goes up the furnace chimney and is wasted, is