OLDER PERIDOTITES OF SCOTLAND. 387 



particularly valuable for the purposes of study ; for there the felspars 

 often show no trace of kaolinization, the augites no trace of uraliti- 

 zation or of amphibolization, and the olivines no trace of serpenti- 

 nization ; and yet these several minerals, as we have pointed out, 

 exhibit in the most striking manner the effects of Schillerization. 



The study of these rocks in the field has clearly demonstrated that 

 the degree of Schillerization of the several minerals can be correlated 

 with the depth from the surface at which the rocks have formerly 

 existed. In the more deeply seated rocks the most perfect Schilleri- 

 zation has taken place, and in those at less depth fewer planes within 

 the crystal have been attacked, all traces of the action disappearing 

 when the rocks have existed near the surface. 



That the action producing Schillerization is a secondary one is proved 

 in several ways. It is perfectly true that enclosures are often formed 

 in crystals during their growth, and that, at high temperatures and 

 under great pressures, abnormal crystal-growths frequently arise. 

 In this way it may be suggested that augite might always crystallize 

 with the diallage- and pseudo-hypersthene-modifications, ferriferous 

 enstatite with the bronzite- or hypersthene-modification, and so on. 

 But against the acceptance of this suggestion several very important 

 considerations may be urged. The contents of the negative crystals 

 are evidently products of decomposition, hydrated oxides like chal- 

 cedony, opal, gothite, and limonite. Purther, as I have already 

 shown, the action of Schillerization can in many cases be seen to be 

 set up. from the surface of the crystals and along the cracks which 

 traverse it. And, lastly, the enclosures are altogether absent from 

 some crystals in deep-seated rocks, which appear to have escaped 

 altogether from the action which has produced this phenomenon. 

 Schillerization is thus proved to be due to local and not to general 

 causes. 



On these grounds, then, I think it is impossible to doubt that 

 what are now crystals of diallage were once common augite, that 

 the bronzite and hypersthene are altered ferriferous enstatites, and 

 that the peculiarities of the deep-seated crystals of labradorite and 

 olivine have been acquired since their original formation. 



Bearing aU these facts in mind, it appears impossible to resist the 

 conclusion that the solvent agents which have produced Schillerization 

 are the water and other fluids which have permeated the rock- 

 masses, and that their solvent action has increased with the pressure, 

 that is to say with the depth from the surface. I need only refer 

 to the classical researches of Daubre'e, Sorby, Guthrie, and others, as 

 placing beyond all doubt the fact of the increase of the actio q of 

 solvents by pressure. 



When we remember the enormous volumes of steam and other gases 

 given off by great volcanoes during their eruption, and further that 

 these eruptions are continued through geological periods of vast 

 duration, bearing in mind too that the evolution of these gases does 

 not terminate with the violent activity of the volcanic vent, but is 

 equally manifested during the " Solfatara stage," and that enormous 

 tracts of volcanic rocks are found altered by surface-emanation of 

 steam, we can well understand how potent must be the influences 



