bo 
MASS DYNAMIC ACTION. 97 
becoming smaller gradually became larger. This increase in coarseness 
of the mineral particles may be followed through all stages to the coarsely 
erystalline schists. 
In the granulated rocks the mineral particles every where show strongly 
the strains of undulatory extinction, but the mineral particles of many of 
the coarsely crystalline schists show no more than slight strain shadows. 
The coarsely crystalline, perfectly schistose rocks, nearly free from 
strain shadows, are always found to be those which have been deeply 
buried and profoundly deformed or adjacent to great intrusive masses, 
or both. It is therefore clear that those rocks represent the most ad- 
vanced stages of metamorphism. 
It is generally agreed that the crystalline schists of this character have 
been recrystallized throughout, and therefore strongly contrast with those 
rocks which have been granulated. However, the granulated and re- 
erystallized rocks are not separated sharply from each other (see pages 
305-312), but, on the contrary, there is every gradation between the two. 
If in the altered sedimentary rocks one passes from a place of granula- 
tion to one of recrystallization, he finds that recrystallization of the matrix 
begins while granulation of the larger particles is still going on. 
The original rock may have varied greatly in the coarseness of its 
constituent particles. In an intermediate stage the matrix may have 
completely recrystallized and the granulation of the coarser particles be 
still incomplete. As a consequence, the mineral particles of the matrix 
are increasing in size at the same time the larger particles are being 
decreased in size. 
At a certain stage the larger grains are granulated into particles which 
average about the same magnitude as those which have crystallized out 
of a fine-grained and perhaps irresolvable matrix, and moreover the 
grains which have formed from the matrix approximate uniformity of 
size. Thus there is a marked tendency toward uniformity in the size of 
the grains of the metamorphosed rocks, and this tendency is ordinarily 
dominant in the crystalline schists so long as mass deformation con- 
tinues. ‘This statement is more nearly accurate in reference to the par- 
ticles of each mineral than to particles of different minerals. This ten- 
deney toward uniformity controls, notwithstanding the principle that 
under ordinary conditions laree mineral particles grow at the expense of 
smaller ones (see pages 274, 275) ; for under mass dynamic conditions a 
large grain, whether original or produced by uneven growth, is especially 
exposed to the mechanical stresses,and therefore is granulated in part or 
putin a state of strain, and thus may be more readily attacked by the 
solutions. Some of the properly oriented smaller particles may them- 
selves grow at the expense of the larger ones or the small ones not prop- 
