VOLUME CHANGES IN METAMORPHISM 551 
be very difficult to account for metasomatism without them. It is 
considered possible, however, by some investigators that replace- 
ment may be effected by gaseous phases, not of magmatic origin, 
but simply induced by the heat of contact metamorphism in the 
outer zone. Goldschmidt' thinks so. It seems to me, however, 
that if this be true there would certainly be contraction evidenced 
in texture and possible to detect by field or microscopic methods. 
Metamor phism under stress.—Under unequal pressure, as pointed 
out by Johnston and Niggli,’ the solubility will be greatly increased, 
and its influence on reactions which are accompanied by the evolu- 
tion of a gas is enormous. ‘‘Unequal stress will cause reactions 
between solids accompanied by the development of a gas phase to 
proceed to an extent which would be inappreciable in the absence 
of stress.’’ The conditions in a rock under heavy stress are greatly 
complicated by the lateral movement of material by flowage (as 
shown by experiments on marble, icé, etc.). This action has 
nothing to do with replacement, but it certainly has a tendency to 
compress and thin the beds. No actual contraction of volume 
except by the closing of pores is involved in this operation. 
Under heavy stress it is commonly assumed that all capillary 
spaces will be closed and that circulation of water and escape of 
volatile phases take place with the utmost difficulty. Whether this 
assumption is correct is not known. It is probably not true in case 
of limestone, which seems to be easily penetrated by liquids and 
gases. 
Under and after conditions of stress there is often a strong 
tendency to development, by replacement, of heavy aluminum 
silicates of small molecular volume, such as staurolite, andalusite, 
and cyanite. This has generally been interpreted as meaning a 
reduction in volume, an interpretation which at first glance seems 
reasonable enough. Most calculations of changes of volume 
according to Lepsius’ volume law are based on such changes. 
«“*The relations [referring to hornfels, etc.] are valid independently of the state 
of the solvents, whether these are watery solutions or a melt or whether the minerals 
simply are separated by a space containing their gaseous phases” (Die Kontactmeta- 
mor phose im Kristianiagebiet, 1911, p. 122). 
0p. Cit. Pu OLA. 
