234 The American Geologist. October, 1902. 
work to adjust the minerals." Such a conception apphed to ice 
necessarily precludes the possihility of obtainin^c;- a measure ot 
the straininjT^, because the strain is continuously obliterated by 
solution and recrystallization. 
Yet it is easily shown that mechanical stresses in tlie ice 
mass are constantly o])erative. and a condition of strain it 
seems must inevitably exist. Again it has been shown that in 
general, in the process of recrystallization under condition of 
saturation, there is always a growth of large crystals 'at the 
expense of the smaller ones. This fact is due, as pointed out 
by Ostwald, * to the phenomenon of surface tension which 
exists on the boundary surfaces between solids and liquids. 
These surfaces are reduced in size by the tension, with the re- 
sulting enlargement of individual crystals, the process being 
augmented, as is well known, by pressure and increase in tem- 
perature.. Now for a given volume of the substance it is eas- 
ily seen that the surfaces of the crystals are inversely as their 
diameters, whence it appears that an increase in the size of the 
crystals, through the reduction of surface tension hberates 
energy as heat. It is difftcult to estimate the amount of heat 
thus set free in the case of ice, but it would seem that this heat, 
be it great or small in amount, w'ill aid the pressure in increas- 
ing further the rapidity of crystal growth. 
The particular ]:)oint of interest here is the possible appli- 
cation of these principles in explanation not only of the 
movement of the ice mass, but also of the well known fact 
that crvstals of glacier ice increase in size in passing from the 
neve line toward the terminus of the glacier. If this assump- 
tion concerning solution and recrystallization be true of 
the zone of plasticity, the premises relating to the movement in 
the zone of fracture do not preclude its application there also, 
to a greater or less extent. Indeed it is probable that an in- 
crease in size of crystals takes place where the forces of rup- 
ture, solar energy, and regelation. are in active operation. 
It is also logical to suppose that partial recovery from a 
state of strain where the deforming pressure is relieved, means 
partial recrystallization, in which case the bounding surfaces 
might in consecjuence be irregular and cunaxl and the optic 
axes arranged at random, as instanced by Deeley and Fletcher, 
already quoted. 
♦Foundations of Analytical Chemistry. "W. Ostwald. 
