In reply to this objection, we may point out that these accumulations of loose materials are of such a nature as to be capable of easy removal by denuding agents, and that as they are formed upon the land they will, if not already washed away by the action of rain, floods, rivers, &c., run great risk of having their materials distributed, when the land sinks beneath the waters of the ocean and the surface is covered by new deposits. With respect to the glassy rocks it must be remembered that the action of water, containing carbonic acid and other substances, in percolating through such masses has a tendency to set up crystalline action, and these glassy rocks easily undergo 'devitrification'; it would therefore be illogical for us to expect glassy rock-masses to retain their vitreous character through long geological periods, during which they have been subjected to the action of water and acid gases. But careful observation has shown that the scoriaceous and vitreous rocks are by no means absent among the igneous materials ejected during earlier periods of the earth's history. Their comparative infrequency is easily accounted for when we remember, in the first place, the ease with which such materials would be removed by denuding forces, and in the second place, the tendency of the action of percolating water to destroy their characteristic features, by filling up their vesicles with crystalline products and by effecting devitrification in their mass. [Sidenote: SIMILARITY OF ANCIENT AND RECENT LAVAS.] If we go back to the very oldest known rock-masses of the globe, those which are found underlying the fossiliferous Cambrian strata, we find abundant evidence that volcanic action took place during the period in which these materials were being accumulated. Thus, in the Wrekin, as Mr. Allport has so well shown, we find clear proofs that before the long-distant period of the Cambrian, there existed volcanoes which ejected scoriæ, lapilli, and volcanic dust, and also gave rise to streams of lava exhibiting the characteristic structures found in glassy rocks. In these rocks, which have undergone a curious alteration or devitrification, we still find all those peculiar structures--the sphærulitic, the perlitic, and the banded--so common in the rhyolites of Hungary, with which rocks the Wrekin lavas, in their chemical composition, precisely agree. Prof. Bonney, too, has shown that the rocks of Charnwood Forest, which are also probably of pre-Cambrian age, contain great quantities of altered volcanic agglomerates, tuffs, and ashes. I have found the sphærulitic, perlitic, and banded structures exhibited by British lavas of the Cambrian, Silurian, Devonian and Carboniferous periods, as well as in those of Tertiary age; and in connection with these different lavas we find vast accumulations, sometimes thousands of feet in thickness, of volcanic agglomerates and tuffs which have undergone great alteration.