Origin of Phe?iocrysts. — Crosby. 303 
of sudden changes of temperature, hydration or pressure in the 
contact zone of a batholith, I was led to consider whether a 
cause for the critical point in the crystallization of the magma 
might not be found in its gradually increasing viscosity, its de- 
velopment being thus immediately determined by internal con- 
ditions, and only indirectly or remotely by external conditions. 
The phenocrysts, or idiomorphic quartz and feldspar ,of 
the Blue hills quartz porphyry do not necessarily imply an 
abrupt change in the rate or condition of cooling; for a very 
gradual change would sufifice, or possibly no change at all, if 
we may postulate, as of course we must, a somewhat more 
r^pid rate in the superficial portion of the batholith than at 
greater depths than where granite is formed. /\t tlie be- 
ginning of solidification, when the crystallizing magma has, 
we will say, its maximum liquidity, each growing phenocryst 
of quartz and feldspar is able to derive material from a consid- 
erable volume of magma. But with continued cooling the mag- 
ma becomes more and more viscous and molecular flow more 
restricted, until finally the molecular flow is unable to keep 
pace with the rate of solidification and new centers of crystal- 
lization are established between the phenocrysts, and the 
growth of the phenocrysts is virtually arrested. 
According to this view, the formation of the phenocrysts 
or the porphyritic texture may be quite independent of any 
change in the rate or condition of cooling, being conditioned 
solely by the relation of the rate of solidification (equal rate of 
cooling), which may be absolutely uniform, to the rate of mole- 
cular flow, which varies as the viscosity, the viscosity being 
determined chiefly by the temperature and the proportion of 
water or other mineralizers. The main point is that the change 
from centralized to diffused crystallization, from the formation 
of the phenocrysts to the formation of the groundmass, will 
be abrupt. In the granite this change does not occur, be- 
cause solidification is so slow as not to overtax the molecular 
flow. When the change does occur in the porphyry, the mag- 
ma will be in a condition analogous to that of an over-cooled 
liquid or a supersaturated solution ; and it is easy to see that 
the molecules would add themselves much more rapidly to a 
crystal 5- 10 of a millimeter than to one 5 millimeters distant, 
and that the new centers of crvstallization would be verv near 
