258 B. WILLIS DISCOIDAL STRUCTURE OF THE LITHOSPHERE 



filled by rock powder. The transverse hole remained unaltered. In both 

 experiments the pressures were maintained 21^ months. There can, there- 

 fore, be no question but that the very great pressure pervaded the entire 

 mass and all the granite was involved in the movement that was shown 

 by the local shearing. In other words, the granite approached a state of 

 flow at a pressure equivalent to 31 miles, and yielded at a pressure equiva- 

 lent to 35 miles of depth below the surface of the earth. 



As has been pointed out by Professor Hoskins, in conference, this result 

 is not exactly comparable with those obtained by compressing the granite 

 in a steel cylinder which, though it yielded, nevertheless continued to 

 resist. The resistance raised the internal friction or flow-strength of the 

 granite. If in experiments 357 and 358 there had been substituted for 

 the small hole a steel jacket offering a resistance equivalent to the pressure 

 of 31 or 35 miles of load, the internal friction of the granite would have 

 been even higher than it was. 



The problem thus raised related to the relative values of the stresses 

 exerted in the case of flow into a small central cylindrical hole, as com- 

 pared with those developed by outward flow against the surrounding steel 

 jacket. To what extent does the centripetal movement toward an open 

 cavity develop increased internal friction, which may be regarded as the 

 equivalent of the confining jacket toward which the movement is centrif- 

 ugal? 



King, discussing Adams' experiments with the steel jacket, says:^^ 



"It follows from Tresca's theory that the rock, when stressed under these 

 ideal conditions, will commence to break down or flow simultaneously through- 

 out its entire volume." 



It is evident from the manner in which both the vertical and transverse 

 holes were filled in experiment 358, cited above, that this condition of 

 simultaneous rock flow existed when the pressure on the end of the speci- 

 men equaled the weight of a column of granite 35 miles high. Eock 

 moving under these conditions would seem to be subject to the familiar 

 law that a constriction of cross-section increases friction. Now the 

 centripetal movement of the rock constituted, in any cross-section vertical 

 to the hole, a flow of 99 areal units toward one unit, the hole having an 

 area of but 1 per cent of the cross-section of the cylinder. It is impossible 

 that this should occur without a very notable increase in friction. 



On the other hand, in analyzing the stresses in the specimen which 

 expands inside a steel jacket. King determined "that the average longi- 



13 L. V. King : On the mathematical theory of the internal friction and limiting 

 strength of roclss. Chicago Jour, of Geol., vol. xxv, 1917, p. 642. 



