even a small planet like the moon, we perceive that there could be no cavities. The most solid materials — steel, adamant, platinum — become plastic under pressures far less than those brought into action by the attractive energy of a planet's mass upon all parts of its interior, except those not far from the surface. Be it noticed that it is not, as some seem to suppose who have written on this subject, the force of gravity at different depths which has to be considered. That diminishes as the centre of the planet is approached. What we have really to consider is the pressure produced by the weight of the superincumbent mass above any given level, and this of course becomes greater and greater as the depth below the surface increases. If the rigidity of the solid substances forming the solid crust of a planet were such that any amount of pressure could be borne without impairing it, then of course the various layers of the crust would form a series of arches, stronger and stronger with approach to the centre, because of the increased compression, and therefore the increased density of their substance. There is no a priori reason, perhaps, why this should not be so. Compression, for example, might increase the rigidity or force-resisting power of the materials of the earth's substance in such sort that mines might be dug to any depth, and horizontal tunnelling carried out from the lowest parts of any mine. But experiment shows that the fact is otherwise. Under great pressures the most solid substances become plastic. Steel behaved like a liquid in Tresca's experiments, affording the most conclusive evidence that at a depth of ten or twelve miles no steel walls, however massive, could defend a cavernous space from the surrounding pressures, which would simply crush in the steel until it formed one solid mass without interstices — at least with no interstices which could be seen if the steel were afterwards brought up from that depth to be cut open and examined. It will be readily understood that at the depth of ten or twelve miles there can be no caverns into which the water of the oceans could be bodily withdrawn. Extending similar considerations to the moon, we perceive that there can be no caverns in the moon's interior, at a greater depth than sixty or seventy, or at utmost one hundred, miles. Now one hundred miles is less than the twentieth part of the moon's diameter, and the entire mass of the moon exceeds the mass of the outermost layer (to a depth of one hundred miles) in about the proportion of four to one. So that even on the assumption that all the external parts of the moon, to the depth of one hundred miles, contracted in such a way as to leave cavernous spaces in the manner conceived by Frankland, there would not be nearly enough space for the lunar oceans, supposing them to bear the same proportion to the moon's mass which our ocean bears to the mass of the earth.
But though cavernous spaces would not form throughout the interior of a planet, room would yet be found, even to the degree conceived by Frankland, for the waters of the planet. The greatest possible pressure to which the most solid rock can be exposed would not fill the capillary spaces which exist throughout the material of the rock, while the pressure on the water at great depths would force it into even minuter than capillary spaces. This has been conclusively shown during experiments entered upon for another purpose, viz., to determine the compressibility of water. For when in 1661 Florentine academicians tried to compress water which had been enclosed within a globular shell of gold, they found that the water under great pressure forced its way through the pores of the gold, and stood on the outside of the globe like dew; and since that time the experiment has been repeated with globes of other metals, a similar result being obtained.
It follows from these considerations, that, as a planet cools, more and more space is formed for the retreat of the planet's seas; and that in all probability in the extreme old age of a planet, when its whole frame to the very centre has been sufficiently cooled, space enough is thus formed to hold all the water which had once adorned the planet's surface.
If we consider the whole history of the moon's cooling, partly as indicated by her actual aspect, partly by the evidence given by the aspect of other planets, and partly as justly inferrible from the laws of physics, we shall find abundant reason for believing that her seas at any rate might thus have been withdrawn. During the earlier stages of a planet's history, considered in the essay entitled "When the Seas were Young," the seas are floating in the form of cloud and vapor above the planet's surface. In the next stage, when the crust is still hot, but not too hot for the waters to rest upon it, the process of cooling must take place more rapidly in the crust of the planet than in the planet's interior. All this time, then, the crust would be con-