pressure overcomes the influence of heat, and keeps the rocks solid, in spite of increased temperature.

Whether solid or fluid, we may safely assume that the interior of the earth has a very high temperature of its own, and if this is so, it must have been continually losing heat by conduction or convection from the interior to the surface, and this heat must have been dissipated into space. Consequently we are led back to the conclusion, which was foreshadowed on p. 10, that there was a time when the whole mass of the earth was in a molten or incandescent state, and that it has gradually cooled down to its present condition.

Assuming this to be the case, the condition of the earth's interior at the present time must depend upon the manner in which it has cooled, and Mr. Hopkins has shown that the earth's mass must now be in one of three different states; these are,

1

1. A solid crust with a fluid interior.

2. A solid crust and a solid nucleus with a fluid interstratum.

3. Solid throughout, with or without fluid spaces or cavities.

Hopkins concluded that the balance of probability was in favour of the last (No. 3). He admitted, however, that after the solidification of the nucleus there must have been a time when a solid external crust was formed resting on an imperfectly fluid mass beneath; but he thought that it must have grown to a thickness of not less than 800 or 1,000 miles at the present time, and that it had probably united itself to the internal nucleus.

He bases his opinion on the ground that the movements of precession and nutation could not be as they are, unless the crust possessed such a minimum thickness. Sir W. Thomson, however, has recently shown that these movements would still be experienced in the case of a thin crust covering a fluid substratum if the inner surface of the crust in contact with the fluid was not perfectly spherical. If this surface was only slightly irregular, the fluid and the

1 Phil. Trans. Roy. Soc., 1839.

2 Brit. Assoc. Rep., p. 48, note.

crust "would have sensibly the same precessional motion as if the whole constituted one rigid body."

Sir W. Thomson has nevertheless expressed his concurrence with Mr. Hopkins' conclusion that the earth is either solid or possesses so thick a crust as to give it practical solidity. He considers that the tide-producing power of the sun and moon is so great that if the thickness of the crust was less than 2,000 miles, it would yield to the strain, and would be subject to periodic deformations similar to oceanic tides. He says that, "the solid crust would yield so freely to the deforming influence of the sun and moon, that it would simply carry the waters of the ocean up and down with it, and there would be no sensible tidal rise and fall of water relatively to land." He concludes that the mass of the earth "is on the whole more rigid than a continuous solid globe of glass of the same diameter."

2

Prof. G. H. Darwin has also investigated the tidal deformation of viscous spheroids, and finds that Sir W. Thomson's results must be modified. In his first investigation (1879) he came to the conclusion that the effective rigidity of the earth was great, but not so great as that of glass or steel; and he adds that "no very considerable portion of the interior of the earth can even distantly approach the fluid condition." "

Still more recently Prof. Darwin has returned to the problem, which is one of extreme complexity, and obtains a result which he expresses as follows: "The present result shows that it is not possible to attain any estimate of the earth's rigidity in this way; but as the tides of the long period are distinctly sensible, we may accept the investigation in the "Natural Philosophy" as generally confirmatory of Thomson's view as to the effective rigidity of the earth's whole mass."

Lastly, Mr. Fisher points out that the most recent observations throw some doubt on the actual existence of the

1 Brit. Assoc. Rep., 1876. Sectional Address, p. 5.

2 Loc. cit., p. 7.

Phil. Trans., 1879, Part I.

"Nature," Jan. 20th, 1887.

8 66 Physics of the Earth's Crust," 2nd ed., p. 41, and Appendix,

p. 34.

small fortnightly tide, on the theoretical existence of which the calculations of Thomson and Darwin are based. The whole matter is, in fact, fraught with so many difficulties and uncertainties, that geologists need not consider the results obtained as in any degree approaching to mathematical demonstrations.

Mr. Fisher, moreover, meets the whole case of the astronomers by the following hypothesis: he assumes that there is a liquid substratum, suggesting that it consists not of melted rock only, but of an intimate mixture of fused rock and gas, the gas being, in fact, dissolved in the liquid magma. This gas may be pure hydrogen, or it may be some compound of hydrogen, such as water, maintained in the state of a gas by the great heat of the interior.

This supposition of a liquid substratum containing dissolved gases is supported by much geological evidence. Steam and gases of various kinds are given off in large quantities during volcanic eruptions. The observations of Siemens on Vesuvius in 1878, and of Fouqué on the eruptions of Santorin (1810), have shown that hydrogen is one of the most abundant gases in such eruptions. Siemens concluded that vast quantities of hydrogen gas, or of combustible compounds of hydrogen, must exist in the earth's interior, and that portions escape through the volcanic outlets. It is also supposed to be the rise of these superheated gases which keeps the lava inside a volcano hot and liquid.

Mr. Fisher therefore assumes the existence of a liquid substratum which is saturated with gases, and is therefore expansible; and he endeavours to ascertain how such a magma would behave in respect to the formation of tides. The materials for a proper examination of this problem do not yet exist one of the desiderata being experiments on the capacity of molten rock (such as slag) for dissolving gases; but if the case is similar to the solubility of certain gases in water, and with certain other conditions which are not improbable, Mr. Fisher shows that there would not necessarily be any tide in the substratum which would affect the level of its surface, or cause any rise or fall of the overlying crust.

Mr. Fisher's argument if sound is of great importance to

C

geologists, who have always felt great difficulty in accounting for various geological phenomena on the supposition of the earth being a solid and rigid globe; they have always been inclined to believe in the liquidity of some portion of the earth's interior, for it is only on this assumption that they are able to explain the occurrence of volcanoes, earthquakes, and other earth movements, or the fact that volcanic products are similar all over the world, as if derived from a common source.1

Assuming the existence of a continuous liquid substratum, Mr. Fisher adduces physical arguments for supposing that the thickness of the overlying crust is not everywhere the same; that its average thickness at the sea-level is about 25 miles; that it is thicker than this under mountain chains, and thinner under the oceans. To this question we shall have occasion to recur in the sequel.

In this chapter we have briefly discussed the views which have been held by the best modern authorities respecting the structure and condition of the earth's interior. We have seen that the most probable conclusions are— 1. That the interior is very hot.

2. That a portion of it is liquid, and forms a continuous layer between the solid crust and the central mass, whether that be solid or plastic (see fig. 1, p. 8).

3. That the solid crust is comparatively thin.

4. That the liquid layer is saturated with dissolved gases.

It is obvious that under such conditions the earth's crust is not likely to be perfectly rigid and unyielding, and that if any causes exist which tend to alter the equilibrium of the crust, movements of various kinds are likely to be produced. The crust, in fact, would give evidence of decided instability.

As a matter of fact we know that such movements do take place all over the globe. Volcanoes pour forth vast quantities of molten matter; earthquakes shake large

1 Another mathematician, Professor Harkness, has also admitted the force of the arguments against the solidity of the earth, and the possibility of a thin crust resting in hydrostatic equilibrium on a denser substratum. ("On the Solar Parallax," Washington, 1891.)

tracts of ground, and often produce permanent changes in the relative level of land and sea; and lastly, there is evidence to prove that slow and gradual movements of the crust have been in progress over still larger areas, the land being elevated in one place and depressed in another. Finally it will appear that every part of the land has been below the sea at some period of past time, and is now dry land only because it has been elevated by the movements above mentioned.

It is important that the geological student should thus early realize the fact of the instability of the earth's crust, proofs of which will be given in the next four chapters, for he will find evidence of the strains, pressures, and movements to which the rocks have been subjected in every branch of geological study.