Origin of Phenocrysts. — Crosby, T^oy 
sured by their comparatively rapid evolution. Thus we again 
find that the principle of the overtaxing of molecular flow by 
increasing viscosity admits of application especially to the in- 
termediate stages ; and we are able to explain the compara- 
tive infrequency of the porphyritic texture, alike in the coarse- 
grained rocks, which have solidified slowly and the fine- 
grained rocks which have solidified rapidly but without a 
sudden increase in the rate of cooling. According to my ex- 
perience with porphyritic dikes the phenocrysts are, as a rule, 
neither peripheral nor central in the dike, but characterize 
intermediate zones. 
While discussing this problem with a friend (Mr. T. A. 
Watson), as we were walking over the quartz porphyry of the 
Blue hills, he suggested that when, during the growth of the 
phenocrysts, the magma becomes over-cooled and diffusive 
crystallization is imminent, it may be precipitated by an earth- 
quake shock, which would give the sudden increase of pres- 
sure which the accepted theory requires (in the absence of a 
like decrease of temperature or hydration). In consequence 
of the over-cooling or potential solidity of the magma, the 
tensile stresses alternating, during the passage of earthquake 
vibrations, with the compressive stresses would not undo the 
crystallizing work of the latter. Have we not here a satis- 
factory compromise? a common ground where sudden changes 
and over-cooling may cooperate? The potency of earth- 
quakes an an exciting cause of crystallization has not, perhaps, 
received the attention that it deserves. 
At a greater depth in the batholith. where the allotrio- 
morphic crystallization of granite is in progress, the physical 
changes afifecting crystallization are so gradual that the ten- 
dency to solidify does not overtax molecular flow, there is no 
over-cooling, and consequently no potential solidity to be 
shaken into actual solidity by the shock of an earthquake. 
The porphyritic texture has been developed in rocks under 
such widely varying and strongly contrasting conditions as to 
indicate that no single explanation will suffice, and that both 
the old and the new views are required to account for the to- 
tality of the phenomena. Thus the sudden increase in the rate 
of cooling incidental to their efifusion is still, apparently, the 
best explanation of the prophyritic texture in surface flows, 
