voLcANoEs.] a chief part in volcanic eruptions, how it issues in vast clouds from tl1e crater, and continues to rise copiously from the lava even after the molten rock has travelled for some miles, and l1as assumed a solid surface. The quantity of water which descends into the interior must be enormous. The floor of the sea, the beds of rivers and lakes, are all leaky. Of the annual rain which sinks beneath the surface of the land, we cannot tell what proportion is detained and prevented from rising again in springs. Not only does this subterranean water percolate down cracks and joints, it infiltrates through the very pores of the rocks, and can do so even against the pressure of steam on the further side. Accordingly, there has arisen a very prevalent belief among geologists, that it is to the enormous expansive force of perhaps white—hot water finding access to, and imprisoned in, some of the heated empty spaces at the roots of volcanoes that the explosions of a crater and the subsequent rise of a lava—column are due. It has been supposed that, somewhat like the reservoirs in which the l1ot water and steam accu- mulate under the Icelandic geysers, these volcanic spaces receive a constant influx of water from the surface, which cannot escape by other channels, but is in great part con- verted into vapour or retained in the fluid state at an enor- mously high temperature and under vast pressure. In the course of time, the materials filling up the chimney are unable to withstand the upward expansion of this imprisoned vapour and water, so that, after some premonitory rumb- hugs, the whole Opposing mass is blown out, and the vapour rushes up in the well—known masses of cloud. Meanwhile, the removal of the overlying column relieves some of the pressure from the water—charged lava, which therefore begins to rise in the funnel until it forces its way through some weak part of the cone, or pours over the top of the crater. After a time the vapour is expended, the energy of the volcano ceases, and there comes a variable period of repose, until a renewal of the same phenomena brings on another eruption. ’»y such successive paroxysms it is supposed that the form of the internal reservoirs and tunnels become changed; new spaces for the accumulation of superheated water are formed, whence in time new volcanic vents issue, while the old ones gradually die out. As physical considerations negative the idea of a com- paratively thin crust surmounting a molten interior whence volcanic energy might be derived, geologists have found themselves involved in great perplexity to explain volcanic phenomena for the production of which a source of no great depth would seem to be necessary. They have supposed the existence of pools or lakes of liquid lava lying beneath the crust, and at an inconsiderable depth from the surface. They have sometimes appealed to the influence of the con- traction of the earth’s mass, assuming that the contraction is greater in the outer than in the inner portions, and that the effect of this must be to squeeze out some of the internal molten matter through weak parts of the crust. Cordier, for example, calculated that a contraction of only a single millimetre (about Zrlgtil of an inch) would suffice to force out to the surface lava enough for 500 eruptions, allowing 1 cubic kilometre (about 1300 million cubic yards) for each eruption. The influence of contraction as the grand source of volcanic energy has recently been insisted upon on quite different grounds by Mr Mallet, who has developed the theory that all the present manifestations of hypogene action are due directly to the more rapid contraction of the hotter internal mass of the earth, and the consequent crushing in of the outer cooler shell. He points to the admitted difli- culties in the way of connecting volcanic phenomena with the existence of internal lakes of liquid matter, or of a central ocean of molten rock. Observations made by him, on the effects of the earthquake shocks accompanying the GEOLOGY 253 volcanic eruptions of Vesuvius and of Etna, showed that the focus of disturbance could not be more than a few miles deep,——that, in relation to the general mass of the globe, it was quite superficial, and could not possibly have lain under a crust of 800 miles or upwards in thickness. The occurrence of volcanoes in lines, and especially along some of the great mountain—chains of the planet, is likewise dwelt upon by him as a fact not satisfactorily explicable on any previous hypothesis of volcanic energy. But he contends that all these difliculties disappear when once the simple idea of cooling and contrac- tion is adequately realized. “The secular cooling of the globe,” he remarks, “is always going on, though in a very slowly descending ratio. Contraction is therefore con- stantly providing astore of energy to be expended in crush- ing parts of the crust, and through that providing for the volcanic heat. But the crushing itself does not take place with uniformity; it necessarily acts per saltmn after accu1m1- lated pressure has reached the necessary amount at a given point, where some of the pressed mass, unequally pressed as we must assume it, gives way, and is succeeded perhaps by a time of repose, or by the transfer of the crushing action elsewhere to some weaker point. Hence, though the magazine of volcanic energy is being constantly and steadily replenished by secular cooling, the effects are interniittent.” He offers an experimental proof of the sufficiency of the store of heat produced by this internal crushing to cause all the phenomena of existing volcanoes. The slight com- parative depth of the volcanic foci, their linear arrangement, and their occurrence along lines of dominant elevation become, he contends, intelligible under this hypothesis. For, since the crushing in of the crust may occur at any depth, the volcanic sources may vary in depth indefinitely ; and as the crushing will take place chiefly along lines of weakness in the crust, it is precisely in such lines that crumpled mountain-ridges and volcanic funnels should appear. Moreover, by this explanation it is sought to harmonize the discordant observations regarding the varia- tions in the rate of increase of temperature downward within the earth. In some parts of the crust the crushing must be much greater than in other parts; and since the heat “is directly proportionate to the local tangential pres- sure which produces the crushing and the resistance there- to,” it may vary indefinitely up to actual fusion. So long as the crushed rock remains out of reacl1 of a suflicient access of subterranean water, there would, of course, be no disturbance. But if, through the weaker parts, water enough should descend and be absorbed by the intensely hot crushed mass, it would be raised to a very high tem- perature, and, on sufficient diminution of pressure, would flash into steam and produce the commotion of a volcanic eruption. This ingenious theory requires the operation of sudden and violent movements, or at least that the heat generated by the crushing should be more than can be immediately conducted away through the crust. Were the crushing slow and equable, the heat developed by it would doubtless be tra11quilly dissipated through the crust, the temperature of which might not be sensibly affected in the process, or not to such an extent as to cause any appreciable molecular re- arrangement of the particles of the rock. Moreover, as the action is general throughout the whole mass of the outer shell, there does not seem in the theory any valid reason why volcanic action should be limited to particular lines, and why it should continue so long persistently on these lines. The existence of weak parts of the crust is postu- lated ; but the successive crushings and consequent fusions might be expected to strengthen these weaker parts and to make volcanic conditions less easy.
It cannot indeed be denied that there exists the most