59 



and granite be the constant result of the internat congelation of that fluid , no solid crust 

 could ever have been formed. But assuming the existence of the crust apart from all 

 theories of ils origin , and limiting our view to the great lakes or oceans of fused rock 

 beneath particular portions of the crust or occupying great cavities in a solid globe , which 

 are all that Mr. Lyell will recognise as necessary to explain plutonic and volcanic pheno- 

 mena , it is obvious that , unlil the temperature of the whole lake or ocean were reduced 

 to that of incipient congelation , the upper layer would retain its heated fluid condition. 

 Ifso, the igneous fluid would have tiine to penetrate fissures to great distances in the sarae 

 way as that of trap. But granite veins are generally found to be short and sinuous , a 

 circumstance which contradistinguishes them from dykes of trap, and which seemsonly expli- 

 cable on the assumplion that the injective tendency of the fluid was counteracted by its 

 tendency to congeal at the surface from contact with the rocks above and to remain there in that 

 state in consequence of its expandingor becoming less dense when it assumes the solid form (1). 

 The phenomena which have been discussed in this paper disclose two structures in hypogene 

 rocks, the laminar, and the globular passing into the cuboidal. Now the first may be due, in 

 the case of gneiss, simply to excess of mica or to this in combination with other causes which 

 have been adverted to. The second is clearly the original mode of crystallization, if the 

 above views are correct, or if it be granted , without reference to them, that granite soli- 

 difies from the surface downwards. In the upper layer of the granite fluid , nuclei are 

 formed , which gradually enlarge till their mutual expansion prevents further increase, and 

 the layer , bearing the form of its mould (2) f i s complete. This layer is the upper bed or 

 stratum of the granite, and others are successively formed beneath, each, of course, con- 

 forming to that above it. The spheres not oaly when first formed, but during the long 



(1) The granite veins of Cornwall and Devon seem to demonstrate that a crust must have been formed while 

 the granite beneath relained its fluidity , for it not only sends veins into the slates but is itself traversed by 

 veins of the saine kind of granite. These are also sometimes continued in the slate above. (Db li Bkchb's Report 

 p. 171 , 2). 



From the point of fusion of granite being very high compared with volcanic rocks , and , I presume, as high as 



that of any sedimentary rocks, it follows that so long as the temperature of granite is above that point , it will 



continue reduciug the incumbent rocks at the plane of contact into its own substance , and will only cease to do 



«o when it is on the point of ceasing to be a fluid. Hence , probably , the shape of the veins. They were filled 



with granite when its temperalure was reduced to that of incipient congelation and when the fluid had, conse- 



quently , become thick. They shew , as it were , its last efforts in its fluid state to melt into the rock above it. 



Eren where the granite fluid at a temperature above 2786° entered a straight mechanical fissure or crack in any 



rock, it would immedialely begin to melt the sides. The veins are therefore more pyrogenous than mechanical. 



Trappean fluid, on the other hand, might remain in a fluid state long after its heat was inadequate to melt the 



adjaceut rocks. It might therefore be forced into fissures without altering tbeir previous form. Trappean veins 



might thus be considered as generally mechanical ,— granite , as generally dyrochemical, for the irregular distribulion 



of chemical ingredients in a rock would affect the course of granitic veins in it. From the great difference in tbe 



fusing point of granite and trap, it probably also resulls that the former is never found as an overlying rock, 



whereas the later from ils reteution of fluidity 790° lower in the thermometric scale , admits of beiug impelled 



through fissures and spread over the surface. 



(2) See note in appendix. 



