DEPTH OF COMPENSATION 279 



of the crust does not yield horizontally to any marked degree under the 

 stress differences between high columns and low ones until the depth of 

 compensation has been reached. Below the depth of compensation-, the 

 material must yield horizontally to rather low stress differences acting 

 for long periods. It is readily understood that the stress difference is 

 from a column of the isostatic shell under a mountain area toward a 

 column under the adjacent plain until we have reached the lower surface 

 of the isostatic shell. 



It is hardly probable that the temperature and pressure conditions of 

 the plastic material differ much at various places, and therefore we may 

 assume that the depth of compensation does not vary materially from 

 place to place. It is possible that the depth below sealevel of flow is 

 somewhat greater under oceans than under continents, owing to the 

 shorter columns. 



Eegional Distribution of Compensation 



In making the computations involved in the isostatic investigations, it 

 was assumed that every topographic feature was compensated by a defi- 

 ciency or an excess of density in the earth's crust directly beneath it, even 

 to the extent of having the compensation confined to a column that is 

 only a fraction of a square mile in cross-section. No advocate of isostasy 

 would agree that it is possible to have each small column compensated 

 independently of surrounding regions. 



A test was made and reported on in special publication number 40 of 

 the United States Coast and Geodetic Survey, to show whether regional 

 distribution of isostatic compensation is more reasonable than the local 

 compensation. It was found that the gravity anomalies were reduced as 

 effectively by regional distribution of compensation out to a distance of 

 36 miles from the station as by local compensation; but when the com- 

 pensation was distributed to a distance of 100 miles from the station, the 

 gravity anomalies became systematic in their appearance and it was 

 found that there was a definite relation between the elevation of the sta- 

 tion and the gravity anomaly. It is reasonably certain that there is no 

 strictly local distribution of the isostatic compensation, but it is equally 

 improbable that the compensation is distributed for great distances from 

 the topographic features. After consideration of the resistance to vertical 

 movement by a column of the isostatic shell, the speaker believes that the 

 column which may be in isostatic equilibrium, independently of the sur- 

 rounding areas, is of the order of magnitude of one square degree at the 

 equator, or about TO miles square. 



