an appreciable extent. The substance of a mass of iron forty or fifty-feet high, would be the same in structure at the bottom as at the top of the mass; for the strength of the metal would resist any change which the weight of the mass would (otherwise) tend to produce. But if there were a cubical mountain of iron twenty miles high, the lower part would be absolutely plastic under the pressure to which it would be subjected. It would behave in all respects as a fluid, insomuch that if (for convenience of illustration) we suppose it inclosed within walls made of some imaginary (and impossible) substance which would yield to no pressure, then, if a portion of the wall were removed near the base of the iron mountain, the iron would flow out like water from a hole near the bottom of a cask. The iron would continue to run out in this way, until the mass was reduced several miles in height. In Jupiter's case a mountain of iron of much less height would be similarly plastic in its lower parts, simply because of the much greater attractive power of Jupiter's mass. Thus we see that the conception of a hollow interior, or of any hollow spaces throughout the planet's globe, is altogether inconsistent with what is known of the constitution of even the strongest materials.
How, then, are we to explain the relatively small mean density of Jupiter's globe? On the supposition that his atmosphere is less than fourteen miles deep, we cannot do so; for there is nothing hypothetical in the above considerations respecting a solid globe as large as Jupiter's, excepting always the assumption that the globe is not formed of substances unlike any with which we are familiar. Even this assumption, though it is one which few would care to maintain in the present position of our knowledge, amounts after all to an admission of the chief point which I am endeavoring to maintain: it is one way—but a very fanciful way—of inferring that Jupiter is utterly dissimilar to the earth. Rejecting it, as we safely may, we find the small density of Jupiter not merely unexplained, but manifestly inexplicable.
All our reasoning has been based on the assumption that the atmosphere of Jupiter exists at a temperature not greatly differing from that of our own atmosphere. If we assume instead an exceedingly high temperature, abandoning of course the supposition that Jupiter is an inhabited world, we no longer find any circumstances which are self-contradictory or incredible.
To begin with, we may on such an assumption find at once a parallel to Jupiter's case in that of the sun., For the sun is an orb attracting his atmospheric envelope and the material of his own solid or liquid surface (if he has any) far more mightily than Jupiter has been known to do. All the difficulties considered in the case of Jupiter would be enormously enhanced in the case of the sun, if we forgot the fact that the sun's globe is at an intense heat from surface to cen-
- The effect of pressure in rendering iron and other metals plastic has been experimentally determined. Cast-steel has been made to flow almost like water, under pressure.