246 



ANNUAL REPORT SMITHSONIAN INSTITUTION, 1921. 



therefore, the deduction of the depth of compensation from the 

 geodetic data may give us a value that is somewhat smaller than 

 the depth at which the material changes from one of resistance 

 to one of yielding. The method by which the depth of compensa- 

 tion is derived is a sort of evening-up process, and, therefore, it is 

 practically certain that the depth of compensation is somewhat above 

 the depth or zone at which the change in character of the material 

 takes place. 



It is possible, and in fact probable, that the depth at which the 

 material changes from one of resistance to one of yielding is not 



Fig. 8. — Stress diagram of two columns in isostatic equilibrium. 



A diagram of stress differences between two blocks of the earth's crust of different 

 elevations indicates that the tendency of the material to move horizontally is in the 

 direction from the elevated block toward the lower one. This is on the assumption that 

 the two blocks are in isostatic equilibrium. 



very different in different parts of the earth. Such a change must 

 be the result of a change in temperature. The pressure naturally 

 is approximately the same at all places at a certain depth below 

 sea level, due to the isostatic equilibrium. It appears very probable 

 that the temperature gradient is approximately the same in differ- 

 ent blocks of the earth's crust. The composition of the material 

 of the crust is approximately the same, or it is safe to assume that 

 this is the case, and, therefore, in no one place would the material 

 at a certain depth be much more resistant or yielding than the ma- 

 terial at the same depth at some other place. It seems, therefore, 

 that so long as the change from material that is resistant to one 



