840 J. BARRELL MEASUREMENTS OP GEOLOGIC TIME 



meiit of the initial rate of rise is probably subject to considerable cor- 

 rection. 



Becker takes next the depth of complete isostatic compensation for 

 a uniformly distributed compensation, 114 kilometers as given by Hay- 

 ford for the most probable value determined by deflections of the vertical. 

 He argues from this that the level where the temperature gradient ap- 

 proaches nearest to the fusion curve of diabase is ll-t kilometers. 



But Hayford later changed this depth to 122 kilometers, and on the 

 basis of measurements of intensity of gravity Bowie has recently reached 

 96 kilometers as the most probable value. Bowie couples this with the 

 statement that he believes future determinations will fall between 80 and 

 180 kilometers.^^ This great range in the depth, as determined by the 

 method of least squares, applied to a large mass of observational data, is 

 owing to the wide range of the local determinations, the imperfection of 

 isostasy, and also the imperfection of the particular hypothesis of dis- 

 tribution. This hypothesis assumes that the density under every topo- 

 graphic feature is so adjusted that at the depth of complete compensa- 

 tion — for example, at 114 kilometers below sealevel — every column con- 

 tains the same mass, irrespective of its height. Furthermore, it assumes 

 that the difference of densities between columns are the same at all depths 

 down, to the level of complete compensation, at which level the differences 

 abruptly cease. 



Such a simple form of hypothesis satisfies the geodetic data, giving the 

 isostatic condition, chiefly between continents and ocean basins, as about 

 nine-tenths complete, but a more natural assumption of variable, regional, 

 irregular, and gradually disappearing compensation satisfies it just as 

 well. Furthermore, a variety of evidence based on the stresses set up in 

 the crust by the loads within it and on it indicates that the shell of great- 

 est strength is less than a hundred kilometers thick, but that the under- 

 lying shell of weakness is very thick, with its center apparently as much 

 as 300 to 500 kilometers in depth. ^* This is the conclusion of the writer, 

 but there are others who hold that the data of isostasy may be so inter- 

 preted as to place the compensation even deeper and who see in the data 

 of isostasy no evidence of weakness. It all goes to show on what an in- 

 secure and artificial foundation Becker's calculations rest. But this is 

 only one of a series of assumptions of equal unreliability which enter into 

 the result. 



83 WiUiam Bowie : Investigations of gravity and isostasy. U. S. Coast and Geodetic 

 Survey, Special Publication No. 40, 1017, p. 112. 



*** J. Barren : The strength of the earth's crust. Journal of Geology, vols. xxii. xxiii, 

 1914, 19ir). 



