256 B. WILLIS DISCOIDAL STRUCTURE OF THE LITHOSPHfiRE 



the abscissa and ordinates are equal. At that point the resistance which 

 the granite offers to flowing is equal to the load resting upon it. The 

 granite should, therefore, flow into any very small or temporary opening.* 



There is an experiment on the closing of a small cavity in granite which 

 appears to correspond closely to the preceding postulate (figure 2).^^ In 

 this experiment a cylinder of Westerly granite 0.5 inch in diameter was 

 confined in a steel cylinder and so compressed that it sheared or flowed 

 into a cavity 0.5 inch in diameter. The steel cylinder in this experiment 

 did not yield. The pressure, applied to the end of the granite column 

 over its entire area, was transmitted throughout the column and affected 

 its whole mass. There were two holes, the one vertical along the axis of 

 the column, the other transverse at one side. When subjected to sufficient 

 pressure the granite powdered or sheared off in minute flakes on the 

 periphery of the vertical and horizontal holes. 



Eef erring to the results on Westerly granite, experiments 357 and 358, 

 Table III, of the article referred to, we may quote Adams as follows : 



* Professor Hoskins has commented on this statement as follows : 



"There is needed a strict definition of the word 'strength' as applied to a rock or 

 other material. 



"The resistance which a material offers to flow does not depend on one single quantity 

 or factor, hut must obviously be influenced by the whole stress-condition existing in the 

 material. To specify completely the stress-condition requires three quantities, namely, 

 the values of the normal stresses on the three principal planes at the point under con- 

 sideration, 



"In the experiments of Adams two of the principal stresses were kept equal, while 

 the third was made greater and greater up to the point at which the flow occurred ; any 

 general conclusions drawn from his results are applicable only under similar restric- 

 tions. This fact must be remembered in interpreting the curves OMNP and OK on 

 plate 8. 



"Assuming that the location of the point Q by extrapolation is valid, what is its 

 meaning? Merely this: that if the granite were under a stress-condition such that two 

 of the principal stresses were equal, each having a value equivalent to the weight of a 

 40-mile column of the rock, then the third principal stress would have to have double 

 this value in order that flow should result." 



With this understanding I am agreed. I have phrased what seems to me to be the 

 same concept in the statement contained in the summary as to mechanical state, that 

 "a non-uniform stress equivalent in pressure to the weight of a column of rock 40 miles 

 high will cause movement." 



Professor Hoskins further states his view that "there does not seem to be justifica- 

 tion for the statement that granite should therefore flow into any opening." He argues 

 that "the stress-condition in the neighborhood of an empty cavity must be wholly differ- 

 ent from that on which the curve is based. So long as the cavity remains empty, one 

 of the three principal stresses has to be zero at the wall of the opening." 



If the substance surrounding the hole were the ether of pure mathematics, or possibly 

 if It were water, the above argument woul'd apply ; but solid granite obeys a law of 

 shear, which produces wedges that crowd each other in the movement that represents 

 flow. The stress-conditions around an opening are therefore radically different from 

 those which would exist in a fluid, as appears from the results of Adams' experiment on 

 the pressure required to close a small cavity. See the following paragraphs. 



12 p. D. Adams : An experimental contribution to the question of the depth of the zone 

 of flow in the earth's crust. Chicago Jour, of Geol., vol. xx, 1912. 



