ORIGIN OF GRANITES. 203 



those just enumerated we find the representatives of eruptive rocks 

 like peridotite, phonolite, leucitophyre, and similar rocks, which are 

 so many exceptions in the basic group of Bunsen. As, however, they 

 are represented in the sediments of the earth's crust, their appearance 

 as exotic rocks, consequent upon a softening and extravasation of the 

 more easily liquefiable strata of deeply buried formations, is readily 

 and simply explained." 1 No other interpretation gives so simple and 

 logical an explanation of the variety of lavas which volcanoes emit. 



Texture of Volcanic Rocks. In any endeavour to comprehend 

 their variety, we are always thrown back from considerations of 

 texture and mineral structure to the fundamental facts of chemical 

 composition. The texture of lavas may be nearly paralleled by that 

 of iron-furnace slags ; and we have seen at the Lowther Iron Works 

 at Workington, slags which have run their full length exhibiting a 

 compact earthy fracture and deep blue colour, while other portions 

 of the flow which have descended like the waters of a cataract, have 

 consolidated into stalactitic sheets and films of nearly transparent 

 glass ; while in one portion of the stream, where the slag had overrun 

 a minute spring, the whole mass was elevated into a cone of yellowish 

 white cavernous pumice, with a well-defined cup at the summit. 

 And occasionally, under special conditions, these slags are found per- 

 fectly crystallised, with a development of Humboldtilite and other 

 minerals. This circumstance, no less than the differences seen in 

 extinct volcanoes between the texture and the mineral composition of 

 lava streams, and the parent masses from which they were poured 

 out, seem to render any purely mineralogical classification impossible. 

 And it has been experimentally proved that slight changes in the 

 chemical composition of the mass necessarily result in the develop- 

 ment of different minerals. Wo therefore proceed to examine the 

 origin of granite as a type of the history of igneous rocks ; because if 

 the hypothesis of igneous evolution can be applied to granites and 

 rhyolites, we make lio doubt that its application to other igneous 

 rocks must follow. 



Origin of Granites. It is convenient first to turn our attention 

 to the granites, and the lavas which most nearly correspond to them 

 in composition because they are perhaps the best known. Not only 

 do granites vary greatly in the relative proportions of their mineral 

 elements, but they also exhibit considerable variation in their con- 

 stituent minerals. For although we may use the general formula of 

 quartz, felspar, and mica to describe the rock, yet the felspar or mica 

 may be almost any member, or members, of these families of minerals, 

 and they may be supplemented or partly replaced by minerals which 

 are no essential component of granite, and are local in their develop- 

 ment. And when the chemical composition of granite is examined, 

 the variation is almost as remarkable ; for although we may regard 

 the normal composition as including silica, alumina, peroxide and 

 protoxide of iron, lime, magnesia, soda and potash, and water, yet 



1 See also Clarence King, U. S. Geol. Exploration, Fortieth Parallel, vol. i. 

 p. 112, on the " Genesis of Granite and Crystalline Schists." 



