Oro Ty JOHN JOHNSTON AND PAUL NIGGLI 
these fluids any small fragment of the same kind subject to the hydrostatic 
pressure of the fluid, such a fragment would tend to increase. Even when no 
such fragment is present. ... . the presence of the solid which is subject to 
the distorting stresses will doubtless facilitate the commencement of a solid — 
of hydrostatic stress upon its surface... . . But in the case of a solid of con- 
tinuous crystalline structure, subjected to distorting stresses and in contact 
with solutions satisfying the conditions deduced above . . . . within certain 
limits the relations given must admit of realization, especially when the solu- 
tions are such as can be easily supersaturated.t 
In other words, when a crystal is strained, the solubility on the 
strained face is increased. Consequently material tends to dis- 
solve off the strained faces of a crystal in contact with a saturated 
solution in any solvent, and to be redeposited where the strain is 
‘less.2. The effect of this is that the crystal changes in such a way 
as to diminish the stress upon it—an example of the well-known 
principle that the readjustment of a system following disturbance 
of equilibrium is always such as to minimize the effect of the dis- 
turbing factor. 
In order to illustrate how stress affects crystallization let us 
suppose an isotropic sphere exposed to stress. The stress may be 
resolved into three stresses acting along axes mutually perpendicu- 
lar; the sphere is thereby deformed into an ellipsoid, the axes of 
which in the simplest case coincide with those of the stress-ellipsoid. 
The final effect is the same when the substance is immersed in a 
solvent medium, for the solubility is increased most in a plane 
perpendicular to the greatest stress. 
Let us consider now the growth under stress of new isotropic 
particles. The stress would tend to make the particle ellipsoidal, 
even though it were to make equal growth in all directions. But 
the saturation limit is reached soonest in the plane perpendicular 
to the smallest stress, so that the isotropic particle in growing 
assumes of itself the form of the strain-ellipsoid. From this it 
follows that particles growing in a stressed medium will be flattened 
in a direction perpendicular to the greatest stress and will be paral- 
1 W. Gibbs, loc. cit. 
2 This does not necessarily imply that the material be redeposited on the unstrained 
faces of the original crystals; it may go to form new individuals. In very many cases, 
of course, a reaction or transformation will occur, so that the material will not be rede- 
posited in its original form. 
