366 LIETJT.-GEN. C. A. MCMAHON ON [May I9OO, 



of quartz 1 upon the other minerals during a later stage in the 

 history of the eruption. This action was probably due to a com- 

 bination of two causes: — firstly, the gradual silicification of the 

 residual magma due to the crystallizing- out of the more basic 

 minerals ; and secondly, the increase of heat due to friction and 

 compression caused by the forcible injection of a viscous and 

 partly crystallized granite into fissures and faults, or between 

 strata, in the form of sills, laccolites, or dykes. The increase of 

 heat increased the solubility of the crystals already crystallized, and 

 enhanced the solvent power of the acid silica in the liquid magma. 



A study of the Gilgit granites in thin slices under the microscope 

 shows that the quartz of these granites, during its corroding stage, 

 ate into the other minerals, especially the felspars and biotite, 

 simultaneously attacking numerous points of weakness, and appeared 

 in these minerals as small rounded, blebs. By gradual increase in 

 size these blebs coalesced and formed lakes, bays, and gulfs in the 

 felspars attacked. This action, combined with a nibbling into the 

 outer edges of the felspars by the quartz, produced in some cases 

 fantastically-shaped skeletons of felspar. In other cases the 

 numerous rounded blebs of quartz in the felspars gave the rock an 

 appearance of granular structure or simulated that of perthite. 

 (See PI. XXIII, figs. 2-5.) 



The granophyric structure, so common in these rocks, is only, it 

 seems to me, an example of the same process in an incipient stage. 2 

 When the process of resolution set in, and the liquid quartz began 

 to eat its way into the crystals, the felspars yielded (according to 

 my theory) almost simultaneously at many separated points of 

 weakness, and the invading molecules of silica were influenced in the 

 direction which they took by the planes of solution and of cleavage 

 within the crystal. The result was that, when the process of re- 

 solution was arrested in its initial stage, the corroding quartz 

 assumed those graphic shapes so commonly seen in granophyre. 



Those who have watched crystallization take place from a solution 

 on a glass slide will have noticed that sometimes crystallization 

 begins almost simultaneously at many points separated one from the 



1 This subject was referred to and illustrations were given in my Pres. 

 Addr. Proc. Geol. Assoc, vol. xiv (1896) p. 287. 



2 Several explanations have been suggested by different authors to account 

 for the granophyre in the rocks that they have described. W ithout challenging 

 the correctness of those explanations as regards the rocks dealt with by those 

 authors, I can only say that I do not think any of them applicable to the 

 Gilgit granites, with the exception of the following : — Roland Irving, in his 

 monograph on the Copper-bearing Rocks of Lake Superior ( U.S. Geol. Surv. 

 vol. v, 1883), describes what we should call granophyric structures in augite- 

 syenite and granitell: — ' In the larger number of sections,' he writes (p. 113), 

 ' the felspar-crystals are charged also with secondary quartz, which occurs 

 either in rows of club-shaped or graphic particles which often follow the 

 cleavage-directions of the crystals, or in very fine lines radiating in fan-shape 

 from a central line.' He speaks of this secondary quartz as ' quartz-saturation,' 

 and states that 'in many cases it is evident ... that the replacing process has 

 gone on from without, inwards ' (p. 114). The author does not tell us, however, 

 how the ' quartz-saturation ' is effected, or whether it is an aqueous, aqueo- 

 igneous, or igneous operation. 



