August 1, 1899.] 



KNOWLEDGE. 



187 



basaltic rock of Eilauea, in Hawaii, has also a coating of 

 some two inches of black glass. Occasionally, as in the 

 wonderful valley near Hlinik, in north-west Hungary, or 

 at Obsidian CUff, Yellowstone Park, huge rock-masses may 

 be seen, consisting throughout of natural glass. In most 

 cases, however, the inner portion of the lava-flow is dull 

 and partly crystalline, while the centre may reveal no 

 glass, even when seen in section under the microscope. 

 A dense mass of felted crystals has arisen ; the fused rock 

 has passed through its glassy stage while at a high tem- 

 perature ; the heat has been maintained sufficiently long 

 for the molecules of the silicates to creep together ; and a 

 stony lava is the result. The mass has devitrined, as glass- 

 workers say ; and all the stages, as Sir James Hall showed 

 long ago, may be successfully imitated by experiment. 



In the dykes, which are exposed where portions of a 

 volcanic cone have been broken down, matters may have 

 proceeded farther. These wall-like masses of rock repre- 

 sent lava that has cooled in fissures ; and they have 

 probably been exposed at no point to the open air. A 

 glassy film may appear on their edges, where the lava was 

 first chilled by contact with the surrounding rock ; and 

 their thin veins or offshoots may easUy be vitreous 

 throughout. But the central portions of a dyke will in 

 general be more crystalline than any portion of a lava- 

 flow of similar composition. They will, in fact, approxi- 

 mate to the structure of granite and its allies. 



The microscope, indeed, shows that the approximation 

 is a very good one. The minerals that have developed 

 are, with very few exceptions, similar to those that occur 

 in coarsely crystaULne rocks of the same chemical com- 

 position. Their mode of grouping in the rocks of the 

 dykes is again and again paralleled in sections cut from 

 coarsely crystalline materials. The structures seen in 

 lavas are, then, to be found in the outer portions of dykes ; 

 those of granite and its allies occur in the central portions. 

 The scale may be more minute, and that is all. 



Certain types of mineral-aggregation may appear in the 

 intermediate regions of the dykes, or throughout dykes of 

 only moderate thickness, which at first sight seem pecaUar 

 to these masses. These structures are found, however, to 

 be intermediate in character between those of ordinary 

 lavas and the coarser type of igneous rock ; and they are 

 repeated, moreover, in the heart of massive lava-flows. 

 One is forced to correlate these modes of aggregation of 

 the minerals, and the presence or absence of a glassy 

 groundmass, with the conditions of temperature and 

 pressure under which the rock consolidated. 



It may, then, be fairly conceded that very slow cooling 

 underground, under considerable pressure, has produced 

 the crystalline characters of granite. The water that 

 escapes so freely as steam from lava-flows at the surface 

 may be imprisoned in deep-seated masses, and may have 

 its effect on the order of crystallisation of the constituents, 

 and even on the nature of the minerals themselves. Yet, 

 when we examine the ofifshoots of a granite or a gabbro — 

 i.e., the dykes which have arisen along the margin of these 

 masses — we find that the Unks are complete with the dyke- 

 rocks of our present-day volcanoes. We must allow for 

 the weathering of some constituents, and for the superior 

 freshness of rooks formed in our late Cainozoic epochs ; 

 but, with this proviso, the comparison can be carried out 

 down to minute structural details. 



Lavas, again, whether they cool in dykes or flows, 

 frequently contain what are called " porphyritic " crystals. 

 These are larger than those in the general groundmass, 

 and give the rock the character of the " porphyry " of the 

 Greeks and Romans. They float, as it were, in the ground, 

 and are frequently broken and eaten into, showing that 



they have not developed under the conditions that pre- 

 vailed during the flow of the molten mass. They must 

 have separated out in quieter times, and doubtless under 

 greater pressure. May they not be the relics of a stage 

 of consolidation which set in far below the surface ? This 

 stage was interrupted by renewed fluidity being given to 

 the mass, through reheating, or by access of water,* or, 

 perhaps, by mere relief from pressure as the mass rose 

 towards the surface ; and the crystals already formed were 

 carried upward in the lava. Many of them were remelted 

 in the process ; many remained to tell the tale. The latter 

 form the " crystals of the first consolidation," the impor- 

 tance of which has been so folly urged by M. Levy, in his 

 critical studies of the structure of igneous rocks. 



Here we are getting some glimpse of the features of our 

 underlying caldron of the rocks. We may well ask, what 

 would have happened if the slow crystallisation had gone 

 on, if no subsequent movement had occurred ? Surely a 

 completely crystalline mass would have arisen, coarser in 

 all its features than the materials examined by us at the 

 surface. Granite and the lava known as rhyolite have 

 precisely the same chemical composition ; diorite is 

 similarly related to the lava andesite, gabbro to the well- 

 known lava, basalt. Where, then, we have rhyolite, 

 andesite, or basalt at the surface, should not granite, 

 diorite, or gabbro, be respectively forming underground ? 



The transition seems easy to us when put in this form ; 

 but the argument at one time was beset with difficulties. 

 The peculiar Teutonic aptitude for classification rather 

 than comprehension obscured the issue with a flood of 

 papers which were intended to prove the essential dissimi- 

 larity of ancient and modern igneous rocks. Granites and 

 gabbros and such like, our German friends urged, are all 

 of ancient date. Abandoning the theory that such rocks 

 were truly primitive, an arbitrary line was drawn at the 

 close of Mesozoic times, and every igneous material now 

 forming was supposed to differ in some way from those 

 produced in Cretaceous or earlier days. It was vain for 

 such workers as the late Mr. S. AUport.f who ranks as 

 one of our foremost pioneers in microscopic work, to urge 

 that decomposition during long ages was responsible for such 

 mineral and structural differences as could be observed ; or 

 for Mr. F. RutleyJ and others to show that slow secondary 

 devitrlfieation would account for the absence of true glasses 

 among ancient volcanic lavas. The German classifications 

 had been based upon a theory sustained by hand-specimens 

 and microscopic slides. The appeal from an orderly 

 collection to the rude disorder of external Nature was 

 repugnant to the careful mind ; and even the great French 

 authors were drawn into the net, and were enfolded in the 

 meshes of the text-books. 



Prof. Judd, in his papers on the " Secondary Rocks ol 

 Scotland," and on the " Volcano of Schemnitz,"§ laboured 

 hard to prove that highly crystalline igneous masses arise 

 naturally in connexion with volcanoes. We cannot 

 usually trace the actual passage from one type of product 

 to the other, since a long lapse of time is necessary before 

 the caldron itself can be laid bare. In the Western Islea 

 of Scotland, however, the coarse gabbros and well- 

 developed grsmites were proved to be of Cainozoic age ; 

 and the same fact has been admitted with regard to other 

 masses, where denudation has been sufiiciently severe to 



* See Judd, " The Natural History of Lavas," Oeol. Mag., 1888, p. 9. 



t Geol. Mag., April, 1870 ; and Quart. Journ. Geol. Soc, VoU 

 XXX. (1874), p. 565. 



J"Devitrified Eocks from Beddgelert and Snowdon," Quart. Journ, 

 Geol. Soc, 1881, p. 407. 



§ QuaH. Journ. Oeol. Soc, 1874 and 1876. 



