[Illustration: Fig. 25.--Diagrammatic Section of Volcano.]
82. _Lava._--Any rock which has been erupted from a volcano in a molten state is called _lava_. Some modern lava-streams cover a great extent of surface. One of two streams which issued from the volcano of Skaptur Jokul (Iceland) in 1783 overflowed an area fifty miles in length, with a breadth in places of fifteen; the other was not much less extensive, being forty miles in length, with an occasional breadth of seven. In some places the lava exceeded 500 feet in thickness. Again, in 1855, an eruption in the island of Hawaii sent forth a stream of lava sixty-five miles long, and from one to ten miles wide. The surface of a stream quickly cools and consolidates, and in doing so shrinks, so as to become seamed with cracks, through which the incandescent matter underneath can be seen. As the current flows on, the upper crust separates into rough ragged scoriform blocks, which are rolled over each other and jammed into confused masses. The slags that cake upon the face or front of the stream roll down before it, and thus a kind of rude pavement is formed, upon which the lava advances and is eventually consolidated. Thus, in most cases, a bed of lava is scoriaceous as well below as above. Other kinds of lava are much more ductile and viscous, and coagulate superficially in glossy or wrinkled crusts. When lava has inclosed fragments of aqueous rocks, such as limestone, clay, or sandstone, these are observed to have undergone some alteration. The sandstone is often much hardened, the clay is porcelainised, and the limestone, still retaining its carbonic acid, assumes a crystalline texture. But the aqueous rock upon which lava has cooled does not usually exhibit much change, nor does the alteration, as a rule, extend more than a few feet (often only a few inches) into the rock. A lava-current which entered a lake or the sea, however, has sometimes caught up much of the sediment gathering there, and become so commingled with it, that in some parts it is hard to say whether the resulting rock is more igneous or aqueous. Lava which has been squirted up from below into cracks and crevices, and there consolidated so as to form _dykes_, sometimes, but not often, produces considerable alteration upon the rocks which it intersects. The basaltic structure is believed to be due to the contraction of lava consequent upon its cooling. The axes of the prisms are always perpendicular to the cooling surface or surfaces, and in some cases the columns are wonderfully regular. There are numerous varieties of lava, such as _basalt_, _obsidian_, _pitchstone_, _pearlstone_, _trachyte_, &c.; some are heavy compact rocks, others are light and porous. Many are finely or coarsely crystalline; others have a glassy and resinous or waxy texture. Some shew a flaky or laminated structure; others are concretionary. Most of the lava rocks, however, are granularly crystalline. In many, a vesicular character is observed. These vesicles, being due to the bubbles of vapour that gathered in the molten rock, usually occur in greatest abundance towards the upper surface of a bed of lava. They are also more or less well developed near the bottom of a bed, which, as already explained, is frequently scoriaceous. Occasionally the vesicles are disseminated throughout the entire rock. As a rule, those lavas which are of inferior specific gravity are much more vesicular than the denser and heavier varieties. The vesicles are usually more or less flattened, having been drawn out in the direction in which the lava-current flowed. Sometimes they are filled, or partially filled, with mineral matter introduced at the time of eruption, or subsequently brought in a state of solution and deposited there by water filtering through the rock: this forms what is called _amygdaloidal lava_. In volcanic districts, the rocks are often traversed by more or less vertical dykes or veins of igneous matter. These dykes appear in some cases to have been formed by the filling up of crevices from above--the liquid lava having filtered downwards from an overflowing mass. In most cases, however, the lava has been injected from below, and not unfrequently the 'dykes' seem to have been the feeders from which lava-streams have been supplied--the feeders having now become exposed to the light of day either by some violent eruption which has torn the rocks asunder, or else by the gradual wearing away of the latter by atmospheric and aqueous agencies.