450 J. G. GoodchiJd — Granite Junction in Mull. 



veins ramifying into the surrounding rock, along joints and other 

 divisional planes, is so large that in places it is difficult to say 

 whether the section before one should be regarded as a mass of 

 schists traversed by granite veins or as a mass of granite with an 

 unusually large percentage of included blocks. Some of the best 

 Sections for the study of these phenomena are along the coast near 

 Carraig Mhor and at Torr na Sealga. But even from the tourist 

 steamei', on its way past the south coast of the Eoss of Mull, in. 

 travelling from lona to Oban, the phenomena are distinctly visible 

 without the aid of a field -glass, so large are some of the masses of 

 schist and quartzite included within the granite. 



A close examination of the junction between the granite veins and 

 the invaded schists gives some clue to the way in which this extra- 

 ordinary melange of granite and metamorphic rock has been pro- 

 duced. The granite veins have evidently eaten their way into the 

 surrounding rock, along joints or other planes of weakness, melting 

 the adjuining rock as they advanced, and without wedging any of it 

 apart, as the enormous pressure under which the granite was in- 

 truded might have been expected to do. Scores of examples, 

 examined here and elsewhere, showed that the thickening of the 

 wedges of molten rock, as they advanced into the strata beyond, was 

 accomplished by the gradual removal of the rock along the sides of 

 the wedge, the place of the rock so removed being fallen by au 

 equivalent bulk of granite, different in no respect but that of the 

 size of its constituents from the granite nearer the central part of 

 the chief mass. In many cases the course of the granite tongue can 

 be followed along one joint, narrowing as it extends inward, and 

 then turning off in a different direction along another joint, nearly, 

 or quite, up to an intrusion coming from another part. The detach- 

 ment of the blocks by means of wedges, which have eaten their way 

 in along the joint planes, can be traced through every stage of 

 quarrying up to complete isolation. 



Once surrounded in this way by an envelope of molten rock, the 

 extraction of the block, by the further enlargement of the wedges, 

 was no difficult mattei*. The mass of schist so liberated floating 

 upward into the main stream if the current was setting strongly in 

 that direction ; downward, if so determined by the relative specific 

 gravity of the molten as compared with the unmelted rock ; or 

 remaining suspended if the relative specific gravity, etc. so permitted. 



Probably the whole of the blocks so detached were destined sooner 

 or later to undergo complete solution, and, in that form, to undergo 

 amalgamation with the rest of the molten rock forced up from the 

 reservoirs below. The process of difi"usion in such a case may be 

 likened to what takes place when a fragment of a more-fusible metal 

 is introduced into a molten mass of a less-fusible metal. In the 

 case where the two readily form an alloy the difi"usion is effected 

 completely, without stirring, or any other mechanical aid. This 

 principle of diffusion seems to afford the clue to the uniformity of 

 composition of the great majority of intrusive rocks, even where they 

 invade rocks of very diverse composition. It may be remarked. 



