THE GENESIS OF IGNEOUS ROCKS 315. 



that such magmas are in a liquid state in the ordinary sense 

 of the word. They are, it is true, non-crystalline, but at 

 the same time they must, as recent researches have shown, 

 be as rigid as a solid so far as regards rapidly changing forces 

 like those of earthquake tremors, or even the varying stresses 

 due to the attraction of the moon and sun, though there is 

 reason to believe that they are capable of yielding slowly to 

 continuously operating forces. Under the enormous pressures 

 of the earth's interior the distinctions between the states of 

 solid, liquid and gas would seem to cease to exist, and all 

 discussions as to whether one condition or the other prevails 

 are probably beside the point. To say that the magmas of the 

 earth's interior are in a gaseous state if the temperature is above 

 the critical point is as misleading as to assert that they are in 

 a liquid or solid state, for we have no reason to suppose that 

 at the pressures that must prevail the critical temperature 

 corresponds to any discontinuous change of physical condition. 

 We cannot here follow the author in his interesting de- 

 scriptions of the varying microstructure or texture of igneous 

 rocks and its relation to the phenomena and principles which 

 we have briefly considered. But besides the differences which 

 rocks present in this respect there are important variations in 

 chemical composition which we have every reason to suppose 

 have been developed by a process of segregation, and a number 

 of suggestions have been made as to the means by which they 

 may have been brought about. All are agreed that in many cases 

 at least these variations arise in the course of the crystallisation 

 of the magma in intercrustal reservoirs. The first crystals that 

 form appear on the margin where the magma is cooled by 

 contact with the neighbouring rocks. Here it is mainly the 

 more basic minerals that separate out, leaving the magma 

 poorer in their constituents, which are replaced to some extent 

 by diffusion from other parts of the reservoir. In process of 

 time crystallisation extends towards the interior, the minerals 

 formed becoming gradually less basic, until in the centre of the 

 mass the most acid variety of the rock occurs. The gabbro 

 of Carrock Fell in Cumberland, described by Mr. Harker, is 

 an interesting example. At the actual border it is " a very 

 dense ultrabasic rock, with as much as 27 per cent, of titaniferous 

 iron-ore (silica-percentage 32^)," whereas " in the centre the 

 rock is a quartz-gabbro of relatively acid nature (silica-per- 



