314 A single interbedded sheet of crystalline rock seldom occurs by itself without any other volcanic accompaniment. It is usually associated at least with bands of tutl' showing that the emission of lava was not unattended with frag- mentary ejections. In the majority of cases it will be found to form part of a series of interbedded sheets with intercalated tuffs. Vast piles of such consecutive flows, reaching a total depth of several thousand feet, remain to witness the energy of former volcanic vents. II. F RAG;IE.'T.-L.—Tl1e rocks embraced under this term include all the fragmentary volcanic ejections which enter into the composition of the earth’s crust, from the coarsest agglomerate to the finest tuff. They differ in lithological character, according to the nature of the lavas with which they are associated and from which they have been derived. Thus in a region of trachyte-lavas, we have tracl1yte—tufi's, trachyte-breccias ; in one of basalts, we find basalt-breccias, basalt-agglomerates, basalt-tufl's; in one of obsidians, we meet with pumiceous tuffs and breccias. The fragmentary matter has been ejected from volcanic vents, and has fallen partly back into the funnels of discharge, partly over the surrounding area. It is therefore apt to be more or less mingled with ordinary sedimentary detritus. We find it indeed passing insensibly into sandstone, shale, limestone, and other strata. Great differences occur in the texture of fragmental rocks even in the same volcanic districts. They are often coarse and tumultuous at or near the vents, and fine-grained at a distance. Alternations of gravelly peperino—like tuft’ with a very fine-grained “ash” may frequently be observed. Large blocks of lava—form rock, as well as of the strata through which the volcanic explosions_have taken place, occur in the tuffs of most old volcanic districts. It has been already pointed out that agglomerate and tuff are not infrequently to be met with occupying the sites of the vents of eruption. Their most connnon disposi- tion, however, is in beds either alone or associated with interbedded lavas. .Iasses of fine or gravelly tuff several hundreds of feet in thickness, without the intervention of any lava-bed, may be observed in the volcanic districts of the Old Red Sandstone and Carboniferous systems in Scotland. On the other hand, in these same areas thin seams of tuff may be seen interlaminated with ordinary sandstone, shale, or limestone. In the one case we have evidence of long-continued and powerful volcanic action, during which fragmentary materials were showered out and spread over the water—basins to the exclusion of ordinary sediment. In the other we have proof of feeble intermittent volcanic explosions, whereby light showers of dust were discharged, which settled down quietly amidst the sand, mud, or limestone accumulating around at the time. Under these latter circumstances tuffs often became fossiliferous ; they enclose the remains of such plants and animals as might be lying on the lake—bottom or sea-floor over which the showers of volcanic dust fell, and thus they form a con- necting link between aqueous and igneous rocks. VIII.——.fr:r,x.or.rmc ROCKS AS PARTS or TIIE Ancnrrecrvnn or THE E_rr.'rn’s CRUST. In part ii. (ante, p- 235) some account has been given of the composition of certain foliatcd rocks frequently met with in the central portions of mountain chains and else- where, lying beneath geological formations of high antiquity. In part iii., in the discussion of the hypogene causes of change within the earth’s crust, reference was again made to these rocks, and they were alluded to as examples of the effects of subterranean processes altering the original character of large mineral masses (ante, p. 263). They were there cited as melanzorgzlue rocks, but their characters GEOLOGY [1v. STP.L’CTL'I:.-L. as integral parts of the earth's crust were reserved for dis- cussion in the present part of this article. At the outset some caution nmst be employed as to the exact meaning in which the terms “metaniorphism" and " metamorphosed” are employed. In a certain sense it may be said that all or nearly all rocks have been metamor- phosed, since it is exceptional to find any, at least among such as are not in a geological sense of modern date, which do not show, when closely examined, proofs of having been altered by the action of percolating water or other daily acting metamorphic agent. liven a solid crystalline mass which, when viewed on a fresh fracture with a good lens, seems to consist of unchanged crystalline pa1'ticl(-s will usually betray under the microscope umnistaliable evidence of alteration. And this alteration may go on until the whole internal organization of the rock has been readjust-.d, though the external form may still remain such as hardly to indicate the change, or to suggest that any new name should be given to the recomposed rock. Among many igneous rocks, particularly the more basic kinds, as basalts, diorites, olivine rocks, the, metamorphism of this kind may be studied in all its stages. But it is not to alteration of this nature, effected at the surface by meteoric water, that the term metamorphi.~m is properly applied. That word is reserved for the process of subterranean change above treated of (ante, p. 2.38), whereby a more or 18:8 complete transformation has been effected throughout vast mineral masses which, while under- going crystalline rearrangement, have usually suffered simultaneously enormous compression. Gneiss, mica-schist, and the other schistose or foliatcd rocks are typical ex- amples of the results of this metamorphic process. Three antagonistic opinions are at l»1'0>'Cnt entertained regarding the origin of these rocks. Some geologists regard the crystalline schists as plutonic rocks representing the early cooled crust of the earth, and suppose that a similar schistose structure has occasionally been super- induced by plutonic action on later sedimentary formations. Again, by some recent writers the “'ernerian notion of chemical precipitation has been revived, a11d the idea of metamorphism has been discarded. These authors suppose that the schistose rocks, in common with many pyroxenic and hornblendic rocks (diabases, diorites, &c.), as well as masses in which serpentine, talc, chlorite, and epidote are prevailing minerals, have been deposited “for the most part as chemically—forn1ed sediments or precipitates, and that the subsequent changes have been simply molecular, or at most confined in certain cases to reactions between the mingled elements of the sediments, with the elimination of water and carbonic acid.” To support this view, it is necessary tu suppose that the rocks in question were formed during-.1 period of the earth’s history when the ocean had a consider- ably different relative proportion of mineral substances dis- solved in its waters, and consequently that they must be assigned to a very early geological period, anterior indeed to what are usually termed the Palaeozoic ages. And it becomes further needful to discredit the belief that any gneiss or schist can by possibility belong to one of the later stages of the geological record. The more thorough-going advocates of the pristine or “eozoic” date and original chemical deposition of the so-called “metamorphic” rocks do not hesitate to take this step, and endeavour, by ingeni- ous explanations, to show that the majority of geologists have mistaken the geological structure of the districts where these rocks have been supposed to be metamorphosed equi- valents of what elsewhere are Palzeozoic, Secondary, or Tertiary strata.‘ Prevalent opinion supports the third or metamorphic theory, according to which the schistose rocks
1 See Sterry IIunt’s ('Izc7m'cal Essays, p. 233 sq.