The amount of gas and steam given off during an eruption, and from the surface of lava-flows, varies considerably. Many lavas, such as those of Vesuvius, are so saturated with water-gas that the flowing lava is often quite concealed from view in a dense cloud of steam, which is blown off from hundreds of small vents and spiracles all over the surface of the stream. Other lavas, both acid and basic, well up more quietly, and contain but little steam or gas, or else they part with it in a less explosive manner.

Lastly, as to the temperature of lavas at the moment of issue, there is little doubt that this differs in different cases and places, but its actual amount in any case has not yet been ascertained, owing to the difficulties of observation. Copper wire has been melted in lava which indicates a heat of about 2,200° F. The complete fusion of the basic slags which are formed in an iron furnace occurs at about 2,700° F., and we may therefore assume that the temperature of liquid basic lava is at least as much, while for the fusion of an acid lava a greater heat must be necessary.

Lavas, however, are not always in a state of complete fusion; observations have shown that it often consists of a pasty or semi-fluid mass, in which are floating a number of crystals and crystalline particles of several different minerals. These crystals bear evidence to having been formed under great pressure, and in the presence of superheated steam or gas, and have evidently originated while the lava was in the pipe of the volcano. As the lava cools the pasty mass also crystallizes, and forms a finely crystalline stone enveloping the larger separate and previously-formed crystals.

As all rock is a bad conductor of heat, so, when the crust has once formed, the internal portion of the lava-stream always remains hot for a long period of time. The process of solidification is, therefore, a very gradual one; and a great thickness of lava takes a proportionately long time to cool, and solidify into stone. Natural sections of lavastreams in cliffs or ravines show that the upper and lower layers of the cooled mass are always more or less vesicular and scoriaceous, while the central portion is a hard, compact, and frequently a crystalline stone.

The gradual cooling of the lava causes it to contract, and

leads to the production of cracks, which traverse the whole mass from top to bottom. These are sometimes so regular as to split the mass up into a series of hexagonal prisms or rude columns, similar to those assumed by starch and other substances in passing slowly from a fluid to a solid state. This structure will be more particularly described in Part II.

Slow-moving lavas are frequently congealed before they arrive at the base of the cone, but the more liquid lavas often flow to a distance of many miles from the point of eruption. Some lava-currents from Vesuvius are said to have flowed a mile and a half in fourteen minutes, and others a distance of nearly two miles in the space of three hours. The stream which issued from one of the lateral cones of Etna, and destroyed Catania in 1669, was 14 miles in length, with a width of 5 miles in some places. The floods of lava, however, which issued from Skaptar Yokul in Iceland, in the year 1783, are the most extraordinary on record. One of these lava-streams was 50 miles long, with a width of from 12 to 15 miles; another was 40 miles long, with an occasional width of 7 miles; both had an average depth of 100 feet, increasing here and there in narrow ravines to a depth of 500 or 600 feet.

Formation of Lava-filled Fissures.-The position which volcanoes apparently hold in relation to lines of fissure in the earth's crust has already been mentioned; and the dykes which traverse the cone itself have been spoken of as cracks injected with lava. The actual formation of such fissures has, indeed, been observed; and Sir Charles Lyell gives the following account of one which opened in the plain of St. Lio, near Etna: "A fissure 6 feet broad, and of unknown depth, opened with a loud crash, and ran in a tortuous course to within a mile of the summit of Etna. Its direction was from north to south, and its length 12 miles. It emitted a most vivid light, indicating that the fissure was filled with incandescent lava, probably to the height of an orifice not far from Monte Rossi, which at that time opened and poured out a lavacurrent. When the melted matter in such a rent has

1 66 Principles of Geology," tenth edition, vol. ii. p. 21.

cooled, it must become a solid wall or dyke, intersecting the older rocks of which the mountain is composed." There is every reason to believe that the fissures over which volcanoes stand are similarly filled with molten rock, and are, in fact, the feeders or reservoirs of the volcanic vents. On the cessation of volcanic activity in the district these lavafilled fissures must remain as enormous dykes, extending for long distances, and cutting through all the rocks of which the country is composed. Such dykes are sometimes found extending for distances of 50 or 100 miles across the country; and in Iceland, during the eruption of Skaptar Yokul, in 1783, "lava was emitted consecutively at several points on a linear range of 200 miles," a lava-filled fissure doubtless extending throughout the whole of this distance, and now remaining as a solid dyke.1

Such dykes, as Lyell observes, may be called the roots of the volcano, reaching downwards to the regions of subterranean fire, while the external cone, with its lava-streams and loose ashes, may be likened to the branches and light foliage; and, to complete the simile, we may consider the central pipe or chimney to be the stock or trunk of the volcanic tree through which the lava-sap rises from the deep-seated roots to feed the sub-aerial branches.

Exciting Causes of Volcanic Action.-This is rather an abstruse subject upon which to enter near the beginning of a treatise on Geology, but we shall endeavour to put the results of recent researches as simply and briefly as possible. With respect to the immediate cause of eruption-that is to say, the force which causes the explosions, and raises the lava in the volcano-there is a general agreement of opinion, and this has been so well expressed by Professor Judd that his words are here quoted: 2 That the molten materials which issue from volcanic vents have absorbed enormous quantities of steam and other gases we have the most indisputable evidence. The volume of such gases given off during volcanic outbursts, and while the lava streams are flowing and consolidating, is enormous. . . . It is to the violent escape of these gases from the

1

Scrope, "Volcanoes," p. 52.

2 "Volcanoes," Internat. Scientific Series, second edition, p. 357.

molten rock-masses, as the pressure upon them is relieved, that nearly all the active phenomena of volcanoes must be referred; and it was the recognition of this fact by Spalkanzani, while he was watching the phenomena displayed in the crater of Stromboli, which laid the foundations of the science of Vulcanology."

But other questions here present themselves: at what point and in what manner did the gaseous material gain access to the molten rock, and how did it come to be occluded or confined within it? It has been supposed that the water might be supplied by a downward percolation from the oceans and other surface waters, and that even at great depths a capillary conduction of water would still take place toward the earth's interior; but although there is much evidence to show that surface waters do penetrate to great depths, and that they do actually gain access to the ducts of volcanoes when once these have been established, yet it is impossible to conceive that water could penetrate the solid crust of the earth to a depth of 25 or 30 miles, so as to become the initial cause of volcanic action. Not the minutest crack or interstitial space can exist at a depth where the rocks must be in a plastic condition under the influence of enormous heat and pressure. If we could imagine the existence of cavities at such depths, capillary action might be set up; but this seems to be impossible.'

It is much more likely that the water-substance is dissolved or occluded in a liquid substratum of fused rock, its presence there being accounted for in the same way as the presence of other oxygen compounds, that is to say, its materials existed originally in the form of oxygen and hydrogen gases, and it is only when and where conditions allow of their combination that water-substance is formed. If the conditions of the liquid substratum have allowed of this combination, the water is still a gas dissolved in the liquid, for it is a well-known fact that all liquids can absorb large quantities of gaseous matter, and that the quantity so absorbed (though not the volume) is increased by pres

For a refutation of the theory of capillary penetration and volcanic action suggested by M. Daubrée, see Fisher's " Physics of the Earth's Crust," second edition, p. 143.

D

sure. On the relief of pressure, the expansion of the water-gas would lessen the specific gravity of the liquid and enable it to rise, while if it reached the surface explosive ebullition might take place.

That the escape of such occluded gas can produce volcanic action we have an interesting demonstration in the phenomena exhibited by cooling sulphur during the process of its extraction from soda-residues. The molten sulphur is exposed to a temperature of 262° F. and a pressure of two or three atmospheres, in the presence of steam. Under these circumstances it is found that the sulphur absorbs a considerable amount of steam; and when the mass is allowed to cool and solidify, this is expelled again with great violence. The hardened surface crust is agitated and fissured, miniature cones and lava-streams being formed upon it which have a striking resemblance to the grander phenomena of the same kind which occur on the surface of

the earth.

In the presence of confined steam or gas we have, therefore, a competent cause of volcanic action, if a way was once made for its escape through the crust of the earth. This is probably accomplished by the strains to which the lower part of the earth's crust is subjected, aided by movements in the form of convection currents occurring in the subjacent liquid stratum, and resulting in the production of fissures along certain lines of weakness in the crust; but further consideration of the subject must be deferred till the student has acquired more knowledge of structural geology; and, after all, we can but guess at the exact modus operandi of the forces concerned in producing a volcano.