The _trappean rocks_ are also of igneous origin. It is evident, from their occurring in the form of dikes, that they have been in a melted state. As they rest upon rocks of a sedimentary origin, they must have been thrown up by volcanic forces. Yet they differ from ordinary lavas. They are not vesicular in their structure, are more crystalline, and there is in no case evidence that they have flowed from craters. If we regard them as the lavas of submarine volcanoes, we shall have conditions which will account for all their peculiarities. At a certain depth the pressure of the water would be sufficient to prevent the formation and escape of vapor, and therefore the lavas thus ejected would not be vesicular. As the rapid cooling of lavas depends, in a great degree, upon the escape of watery vapor, submarine lavas would cool slowly, in consequence of the pressure. The liquidity depending in part upon the retention of the heat, and in part upon the retention of the aqueous vapor, they would consequently remain in a liquid state much longer than the lavas of sub-aërial volcanoes. They would therefore take a more highly crystalline form. All the loose materials thrown out during the eruption would be removed by oceanic currents, and hence no cone would be built up around the orifice of eruption. We may therefore regard the trappean rocks as the lavas of submarine volcanoes. The present volcanoes of this kind are necessarily producing the same kind of rocks, though there will be no other proof that they exist, except the existence of the volcano, till the bed of the sea becomes dry land. The _granitic rocks_ are also the product of igneous causes. Granite is the most abundant of these crystalline rocks; and the others, such as crystalline limestone, are so intimately associated with granite that they must have had the same origin. Granite is everywhere found to send off dikes into the overlying rocks, and must therefore have been in a state of fusion; that is, it must have existed as lava beneath the surface. It is obvious that fluid lava always exists in great quantity beneath areas of energetic volcanic activity. Portions of this lava must in succession take the solid form. Wherever the surface is elevated along a line of fracture, the lava which is accumulated beneath rises above the level of the general reservoir of lava, and will therefore part with its heat more rapidly. On cooling, it becomes the granitic nucleus of the mountain. We ought also to suppose that, by the extremely slow process of the transmission of heat to the surface, the crust of the earth is everywhere increasing in thickness; that is, the upper portion of the great lava mass is solidifying.