662 
GEOLOGY: S. TABER 
otile veins are sometimes remarkably uniform in width, and always 
have well defined walls that are easily separable from the veins. 
The last objection applies likewise to the hypothesis that chrysotile 
veins represent portions of the serpentine that have crystalHzed in 
situ. Recrystallization is a common phenomenon in rocks, but it is 
never confined to a narrow vein-like zone with sharply defined walls. 
If chrysotile can be formed by such a process, why is the ratio of chryso- 
tile to massive serpentine limited? Why is the entire mass of serpentine 
never recrystallized to form chrysotile? 
Not one of the theories previously cited satisfactorily explains the 
presence of one or several partings in some veins and their absence in 
others. They do not explain the angular inclusions of massive ser- 
pentine that frequently mark the partings in chrysotile veins and are 
also found irregularly distributed through them. They do not explain 
why the fibers are normal to the vein walls in some instances and ob- 
lique in others. They do not explain the presence of sharp bends in 
the fibers, nor veins of cross fiber that grade into slip fiber. In view 
of all these facts it must be concluded that previous theories are in- 
adequate to explain the origin of fibrous veins. 
As noted above, cross-fiber veins with structural features similar to 
those found in rocks have been produced in the laboratory where their 
formation and growth could be observed. The evidence thus obtained 
supplements that furnished by veins of fibrous minerals, and makes 
inevitable the conclusion that cross-fiber veins are formed through a 
process of lateral secretion the growing veins making room for them- 
selves by pushing apart the enclosing walls. 
The force that enables the growing veins to push apart their walls 
is not due to the tendency of a crystalline substance to assume a regular 
polyhedral form, for the columnar or fibrous structure of most minerals 
occurring in cross-fiber veins is not a crystallization property, but is 
caused by the conditions of growth.^ Under similar conditions the 
fibrous structure will develop in substances that crystallize in any of the 
systems of crystallization. In most cases it is not the normal habit, 
and therefore is unstable. The writer believes that the force is due 
chiefly to the expansion in volume which accompanies the separation of 
most solids from solution, for, as yet, he has obtained no pressure effects 
during the crystallization of substances that separate from solution with 
decrease in volume. When a substance separates from solution with 
increase in volume, the pressure developed depends on the magnitude 
of the forces resisting expansion, and may be much greater than the force 
required to crush the substance. It is improbable that pressure alone 
