270 B. WILLIS DISCOIDAL STRUCTURE OF THE LITHOSPHERE 



to the postulate of temporary mobility greater importance, even, than it 

 then seemed to have. 



It was suggested in the work referred to that periodic rises of tem- 

 perature, favoring recrystallization and fusion, were a cause of temporary 

 mobility ; and this view I still hold. 



Gilbert in 1913 presented an argument for recognition of a ^^relatively 

 mobile layer separating a less mobile layer above from a nearly immobile 

 nucleus," and qualified the concept as follows : 



"It is not necessary to suppose that the degree of mobility at the horizon of 

 mobility is that of a liquid at the surface. When such mobility is attained by 

 any but the densest rocks, eruption takes place. It is not necessary to think 

 of the degree of mobility as uniform, either from place to place or from time 

 to time. Its place variation will naturally be coordinate with that of rock 

 types, and its time variation coordinate with epochs of elevation and subsi- 

 dence. Neither should the depth of maximum mobility be thought of as uni- 

 form." 2« 



Gilbert did not assign a cause of mobility. 



Barrell in 1915 gave the ideas of his predecessors more definite form, 

 and defined the "asthenosphere," or zone of weakness, as that zone "in 

 which flowage is conceived as taking place with but little expenditure of 

 energy, by a ready recrystallization at the temperature of primary crys- 

 tallization of magmas.^^ His discussion of the subject is exhaustive, and 

 his definition of the zone of mobility is that which most completely satis- 

 fies the conditions of current knowledge of geophysics.^^ 



-8 G. K. Gilbert : Interpretation of anomalies of gravity, U. S. Geol. Survey, Profes- 

 sional Paper 85 C, 1913, p. 35. 



2'^ J. Barrell : The strength of the earth's crust. Chicago Jour, of Geol., vol. xxiii, 

 1915, pp. 425-429. 



Note. — It is apparently essential to a true record that I should correct a misstatement 

 made, no doubt, through misapprehension of my meaning, by Barrell in his review of 

 Research in China (Science, new series, volume 29, 1909, pages 257-260) and repeated 

 in his later articles (.Journal of Geology, volume xxii, 1914, pages 672-683 ; American 

 Journal of Science, volume xlvi, 1918, page 166 ; ibid., volume xlviii, 1919, pages 301- 

 302). He attributes to me an acceptance of "undertow," due to erosion, as an adequate 

 cause of horizontal crustal movements, saying : 



"Willis in 1907 and Hayford in 1911, overlooking Dana's objections, have attempted 

 to make a lateral isostatic undertow the cause of all horizontal movements in the crust, 

 adopting the mechanism of Button." 



In order that his statement should correctly interpret mine, it would be necessary to 

 replace the words "lateral isostatic undertow'' with lateral underpush due to recrystal- 

 lization, energized "by heat and oriented by isostatic stress. 



I quote my original statement, as follows (Research in China, pages 130-131) : 



"The periodicity of diastrophic effects may be attributed to four conditions which, 

 after a prolonged interval, unite in accentuating the difference between stress and resist- 

 ance. The first is molecular rearrangement, by virtue of which rocks recrystallizing 

 under stress take on the mineralogical constitution and particular structure which occu- 

 pies least space. The second is the gradual accumulation of stress through excessively 

 slow but unceasing movement in suboceanic sectors. It is conceived that there is very 



