THE STRENGTH OF THE EARTH'S CRUST 447 



the form postulated falls off rapidly with increasing horizontal 

 distance from the epicenter. Second, the test mass has been taken 

 as contiguous to the zone of compensation above and with that 

 Hmited depth given by the hypothesis of uniform compensation. 

 This gives it greater effect according to the law of inverse squares, 

 but postulates either an indefinitely thin zone of isostatic flow at the 

 bottom of the zone of compensation or a capacity in a thicker zone 

 of weakness to maintain within itself heterogeneities of density 

 similar to those of the hthosphere above and the centrosphere 

 beneath. It is thought by the present writer that a more probable 

 presumption is that the centrospheric heterogeneities which may 

 exist are distinctly deeper than 122 km. and separated from the 

 lithospheric outstanding masses by a thick zone which yields to 

 broad inequalities of pressure either upon it or within it and there- 

 fore is incapable of maintaining notable inequahties of mass in 

 this shell. 



The reasons for this preliminary hypothesis are briefly as fol- 

 lows: The depth of compensation seems to be variable and to extend 

 in some regions to as much as 300 km,, even under the assump- 

 tion of compensation uniformly distributed and complete at the 

 bottom. Under a more natural assumption that isostatic compen- 

 sation gradually disappears, those heterogeneities of density which 

 give isostatic compensation would gradually diminish with depth 

 and this diminution would extend to a considerably greater depth 

 than 122 km. If heterogeneities which act isostatically gradually 

 disappear, the heterogeneities which can be borne in excess should 

 also be expected to diminish. 



As to the nature of the shell immediately below the zone of 

 compensation, Schweydar has recently analyzed mathematically 

 the results of the measurements of earth tides by means of the 

 horizontal pendulum. 



The calculations were designed to test the presence or absence 

 of a viscous zone between an elastic crust and elastic interior. It 

 is concluded that even a magma bed with a viscosity as high as 

 that of sealing wax at house temperatures and a thickness of but 

 100 km. cannot be present. The assumption in best agreement 



