DEFORMATION OF ROCKS 199 
considered as free to escape, and that the viscosity of water in 
minute crevices plays no part. 
If it be supposed that the rocks above the cavities are solid 
to the surface, and therefore that they are not supported by the 
hydrostatic pressure of a column of water to the surface, the 
problem is reduced by Professor Hoskins’ solution to finding 
the height of a column I square centimeter in area, with a spe- 
cific gravity of 2.7, which weighs two-thirds of 1700 kilograms 
for Conclusion I., and weighs 1700 kilograms for Conclusion 
Ill. This gives for I. about 4200 meters; for III. about 6300 
meters. 
If the more probable supposition be made, that the rocks are 
porous, and therefore that the cavities are supported by the 
hydrostatic pressure of a column of water extending to the sur- 
face, it is necessary to subtract 1 from the specific gravity of the 
rocks, and the effective pressure is 1.7 grams per cubic centi- 
meter. Applying Professor Hoskins’ solution, the question is 
reduced to finding the height of a column I square centimeter 
in area, with a specific gravity of 1.7, which weighs two-thirds 
of 1700 kilograms for Conclusion I., and weighs 1700 kilograms 
for Conclusion III. This gives for I., 6667 meters, and for III., 
10,000 meters. For the very strongest rocks the above numbers 
should perhaps be increased by one-fifth, and this gives a maxi- 
mum of 12,000 meters. 
These conclusions do not apply to rock-inclosed, liquid-filled 
cavities. So far as one can understand the conditions, such 
cavities might exist at an indefinite depth, or at least to a depth 
where the liquid and rock may be miscible in all proportions. 
The above numbers fall within the various lmits—roughly, 
yy 
2 to 8 miles—assigned for the ‘level of no strain,’* or as it 
should perhaps be called more properly the level of no lateral 
stress, and thus make it probable that the lateral stress is less 
than the vertical stress of gravity. Therefore it is probable 
that the conditions upon which Conclusion III. are based 
more nearly represent the truth for the greater part of the earth 
‘Manual of Geology, by JAMES D DANA, 4th ed., 1895, pp. 384, 385. 
