quickly diminishes as we proceed from the mountains, till, in plains not far remote, the strata retain their horizontality, we may say that the trap rocks are most abundant in points and in lines distributed between the granitic axis and the level plains. In some instances trap rocks occupy an extent of country not inferior to the area of granite. The Ochill Hills, the Campsie Hills the Pentland Hills, and others connected with them, form one great trappean country filling the vale of the Forth and Clyde, which is a great natural hollow between the ranges of the Grampians and the Lammermuir mountains, both elevated on axes of granite and syenite. Large breadths of trap rocks appear in Skye, Rum, Eigg, Mull, Arran, and Antrim; but in none of these cases is their appearance connected with ridges of stratified rocks, as granitic masses almost invariably are. Moreover these trap rocks, whether in the shape of dykes or overlying masses, are usually so disposed as to suggest the idea of volcanic action, determined to particular points, and bursting out and overflowing from particular lines, rather than a general expansion beneath immense areas of strata which seems best to agree with granitic elevations.
Trap dykes are frequently manifested along the lines of faults, and these may sometimes be determined in geological age by the circumstances which accompany the disturbed strata.
Keeping in mind these general facts, but disregarding the crude notions which attribute to granite or trap rocks the elevations and fractures which have merely opened to us their subterranean repository, or given them channels to the surface, we shall be able to construct a table of the relative antiquity of igneous rocks, by comparing their distribution with the principal phenomena of convulsion in the crust of the earth. Such a table, however, would be very incomplete if founded upon small geographical areas; as the imperfection of the following sketch, based on the examination of the British islands, will abundantly prove.
