This will be shown by an examination of lavas. At the time of their ejection, they are in a fluid or semi-fluid state; but it is not a complete fusion. Even the most fluid lavas contain particles of minerals in a solid state. The liquidity depends upon the fusion of the more fusible portions, and upon the steam of water at a high temperature, which fills the interstices between the solid particles. The porous character of cooled lavas is produced by the steam which filled the cavities previous to solidification. Steam always escapes from the surface of a lava current while it is cooling, and it is always discharged in immense volumes from the orifice of eruption, in connection with the lava, and especially at the close of an eruption.
The geographical position of volcanoes, also, leads to the conclusion that water is essential to their activity. There are five principal lines of volcanic activity. One, commencing at the southern extremity of South America, extends northward along the Andes and Cordilleras to California or Oregon. The second has a north-east and south-west direction, from the Aleutian Islands through the Kurule, Japanese, and Philippine islands, till it meets the third line, lying in a nearly east and west direction, embracing Sumatra, Java, and most of the Pacific volcanic islands. A fourth band commences in the Grecian islands, and extends westward so as to include the volcanoes of Italy and the adjacent islands, and the Azores. The fifth band embraces the volcanic islands of the West Indies, crosses Mexico in about the latitude of the city of Mexico, and extends into the Pacific. There are also some isolated centres of volcanic activity, such as Iceland. These volcanic bands embrace about three hundred volcanoes. It will be seen that they must nearly all be in close proximity to the ocean, or to large seas. About two-thirds of them are on islands. Moreover, the volcanic vents which are wholly submarine are probably very numerous.
This circumstance of the position of volcanoes establishes a presumption that they cannot exist at a distance from some large body of water; and, taking it in connection with the constant presence of aqueous vapor in lava, we are justified in the conclusion that the presence of water is an essential condition of volcanic activity.
Knowing that heat and water exist at the volcanic centres, it is not difficult to form an idea of their mode of operation. The water, diffused through the interstices of the lava, and subjected to a temperature sufficient to melt the lava, would possess an elastic power, which, though never computed, we may well suppose capable of overcoming any resistance which the crust of the earth might present. The repressing force will be the tenacity and weight of the superincumbent strata. Whenever the elasticity is superior to this repressing force, it will manifest itself in the fracture of the strata, and often in the ejection or lava to the surface.
This fracturing of the strata, produced by an uplifting subterranean force, is believed to be the cause of the noise and the vibratory motion which are the chief phenomena of earthquakes. The elastic force may raise lava to the surface, and thus the fracture would become a volcano. But the force may expend itself by the discharge of vapor into the fissure, or by merely filling it with lava. In either case, the only evidence of the existence of the volcanic force would be the noise and the wave-like motion experienced at the surface. The cause of the volcano and earthquake is therefore the same, though the phenomena which characterize them are different.
When the strata are is thus fractured, lava may for a time be discharged along the whole line. By the cooling of lava in the fracture, it would become partially reunited. Still, this would be the line of least resistance. It would therefore be again burst through in certain places, which would long continue to be orifices of discharge, and thus the original fracture would determine a line of volcanic activity.
The repressing force may become greater at an orifice of eruption than at some other point, either by the great accumulation of ejected materials around the opening, or by the dormancy of the volcano long enough for the complete solidification of the lava with which the channel was filled. The least resistance may then be far from any previous vent, when a new orifice of discharge will be opened, and a new volcano make its appearance. It seems probable, also, that volcanoes may become extinct by the reduction of temperature at the volcanic centre, and that new volcanic centres may be formed; but the cause of this change of temperature is not yet well understood. New volcanoes have broken out in the sea, near Iceland, in several instances; others in the volcanic line east of Asia. Graham Island, situated between Sicily and Africa, was formed by an eruption which broke out in the bed of the sea where the soundings were more than one hundred fathoms. The island was at one time two hundred feet above the sea, and three miles in circumference. It was, however, gradually destroyed by the action of the waves, and now remains a dangerous reef, covered by less than two fathoms water. The volcano of Jorullo, in Mexico, was formed in this way. Previous to the formation of the mountain, the region where it now is was a cultivated table-land. During the year 1759 volcanic action commenced and continued, until, at the expiration of twelve months, a cone had been formed having an elevation of sixteen hundred feet above the adjacent plain.
An orifice of eruption is at first but little elevated above the general surface; but, by the accumulation of ejected matter, a cone is at length formed around the vent. The upper portion of a cone always consists of these materials, but there may also be in progress a general elevation of that part of the earth’s crust, and the cone will partake of that general elevation. The cones of the Andes owe their height, in a great measure, to a general movement of elevation; those of Ætna and Vesuvius, in a greater degree, to accumulation of ejected matter.
In either way, the height may become so great that the force necessary to raise a column of lava to the top would be greater than the sides of the cone, weakened as they always are by fractures in all directions, can sustain. Hence, the highest craters of Ætna and South America have long been closed, and the lava escapes through fissures at a lower level, and lateral cones are produced.