lxviii PROCEEDINGS OE THE GEOLOGICAL SOCIETY. [vol. lxXlV, 



conditions far outside their theoretical range of stable existence ; 

 that many of them indeed have no stable existence, but are 

 metastable forms wherever found ; and that chemical reactions, 

 no less than changes of state, ma}'" be indefinitely retarded or 

 suspended because the rate of change is so slow as to be insensible. 

 This has far-reaching consequences of a positive as well as a 

 merely negative kind. Thus it appears from the researches of 

 Bowen and others that the manner of evolution of a series of 

 igneous rock-types from one primitive magma is governed largely 

 by the formation of zoned crystals of such minerals as pyroxenes 

 and felspars, and, if there were continual adjustment of chemical 

 equilibrium throughout the process, differentiation would follow 

 quite other lines. 



Some features of metamorphosed rocks, such as the comparative 

 rarity of zoned crystals, suggest that equilibrium is more promptly 

 attained, or more closely approached, here than in a crystallizing 

 magma. Goldschmidt indeed assumes equilibrium Avhen, by 

 direct application of the phase-rule, he lays down the law that 

 the number of different minerals in a metamorphosed rock . is not 

 greater than the number of components in the system. This 

 rule, however, does not seem to be justified by the facts ; and it 

 may also be remarked that some characteristic minerals of meta- 

 morphism, such as andalusite and enstatite, quoted by Grold- 

 schmidt as illustrations, are themselves merely metastable forms. 

 We shall certainly err if we suppose that the minerals and 

 mineral-associations actually found necessarily represent a stable 

 arrangement, either under the conditions of their formation or 

 under existing conditions. Indeed, it is only in virtue of the 

 non - establishment of equilibrium that many products of the 

 higher grades of meta.morphism now exist to be studied. Even 

 in reactions strictly reversible the lag-effect is much greater 

 with falling than with rising temperature, and thus minerals 

 formed at high temperatures, instead of reverting to other 

 products as the temperature falls, may survive to indicate, as 

 it were, the high- water mark of metamorphism. 



Nor shall we justly appreciate some of the phenomena of 

 metamorphism, especially in its earlier stages, unless we recognize 

 that the rocks are in many cases far from being initially in 

 a state of chemical equilibrium. Those argillaceous sediments 

 which are of the nature of residual clays probably do represent 

 approximately the ultimate degradation of the crystalline rocks 



