200 SLUDIES FOR SELODENTLS 
than do those of Conclusion I. However, in mountain-making 
regions the lateral stress may be so great as to comply with the 
conditions of Conclusion I., and therefore cavities may close at 
the minimum depth. 
The maximum result reached by the calculation is probably 
in excess of the truth, for all of the assumptions excepting that 
concerning the free escape of water are those favorable to requir- 
ing a great depth for the closing of cavities. However, it is 
believed that the result is valuable, because it is certainly large 
‘enough, and we may be sure that at depths greater than 12,000 
meters no cavities can exist. Also it is reasonably certain that 
in the weaker rocks cavities are closed much nearer the surface 
than this. As more accurate data for the solution of the prob- 
lem become available, it may be possible to obtain reasonably 
accurate results for the rocks of greatest strength and also for 
weaker rocks of different kinds. For instance, if the crushing 
strength of ice be determined for temperatures at and some- 
what below zero centigrade, the greatest possible depth to which 
crevasses in glaciers extend at such temperatures may be readily 
calculated. 
It is highly probable that the very greatly increased plasticity 
of rocks when saturated with superheated water, due to the rise 
of temperature with depth, would lead to the closing of the 
deeper lying crevices by flowage and welding rather than by 
fracture. If the increase in temperature is 1° C. for 30 meters, 
at a depth of 10,000 meters, the material would have a tempera- 
ture of 333° C. above the average temperature of the level where 
climate produces no influence. To this, in mountain-making 
regions, would have to be added any increase in temperature due 
to dynamic action. Under gravity alone rocks at such a depth 
would be subjected to a vertical pressure of 2550 kilograms per 
square centimeter. As has been seen, the lateral pressure 
might be less than this if the rock was not in mountain-making 
areas and near the level of no lateral stress, or it might be as 
much or more than this if in mountain-making areas and there- 
fore subjected to great lateral pressure. It is probable that at 
