PRECIOUS STONES.	17
	chemical analysis its composition is found to be very fairly uniform ; also it is seen that the older rocks into which it is intrusive are not pushed aside by the mass, but rather replaced by it—there is no material increase of bulk; in other words, no amount of new matter corresponding to the bulk of the intrusive mass can have been introduced unless an equal bulk had been removed. This would manifestly be unlikely, and at once suggests that the intrusive mass may be the result of some alteration of the older rock masses replaced by it. The difficulty that at once con­fronts one is that the composition is so uniform even in passing through considerable masses of rocks of entirely different composition; but an average analysis of the rocks throughout the extent of the intrusion reveals the fact that the composition of the intrusive sheet chiefly differs from that of the surrounding rocks in a greater proportion of alkali. The writer above referred to suggested the possibility that such alkali might be slowly carried into the rocks by osmosis from the area of great terrestrial activity along the seaward margin of the land. Once such alkaline water is introduced at points where activity is great and tempera­ture high, rocks may be liquefied at a very much lower temperature than that of ordinary dry fusion, so the depth and temperature demanded need not necessarily be very great. Most rock-forming minerals have a fairly definite melting point, but of course different minerals melt at different temperatures; so it is quite possible that conditions might obtain which would suffice to introduce into a rock that was on the whole in a solid state, certain minerals in a state from which they could crystallise. In this way a shale may be converted into Lydian Intone, and the p.s.                                                                          q