THE MECHANICAL INTERPRETATION OF JOINTS 13 
The agreement is sufficiently close to strongly support Mohr’s 
theory. In addition, however, the figures reveal the striking fact 
that as the material changes, under the action of all-sided pressure, 
from a brittle to a ductile substance, the angle of shearing grows 
progressively less and less acute. Karman’s experiments have, 
therefore, completely verified in the case of one and the same sub- 
stance the inference that the less brittle a substance is the larger 
is its angle of shearing. 
TABLE II 
Effective 
Hoop Pressure | Longitudinal | 6 Observed | 6 Observed | - 
o2=03 Pressure without with Computed 
in Atmospheres o1—02 Reduction* | Reduction 
in Atmospheres 
| 0 gio || oe sa 53, 
235 2100 59 5 5 
Ile re OS ee eee 500 2650 72° 65° 63° 
685 2880 83° 70° Be 
° 690 38° 38° 40° 
SHIMCNUOMOs.o 0005600068 280 2040 70° 60° 63° 
555 2580 82° 73% 70° 
* To find the true angle at which rupture actually took place, it is necessary to reduce the observed 
angle to the value it had when the rock cylinder was deformed under its load. 
We can, however, go even one step farther. If growing circum- 
ferential pressure at right angles to the direction of maximum 
(compressive) stress increases the ductility of a substance, and with 
it the angle of shearing, circumferential tension must decrease it, 
that is, render the substance more brittle. This is completely born 
out by the experiments published in 1911 by W. Mason.t. He 
subjected tubes of mild steel to longitudinal compressive stress 
simultaneously with the application of interior hydrostatic pressure, 
and in one series of experiments, made the tubes undergo longi- 
tudinal tension while applying water pressure externally. The 
angle facing the direction of maximum compressive stress, in the 
absence of circumferential pressure, measured? approximately 100°. 
With growing tension normal to the direction of compression, 
tW. Mason, ‘‘The Liiders’ Lines on Mild Steel,’ Proc. Phys. Soc. of London, 
Wolexexalin(Gor1),) pp. 305-33. 
2 Ibid., Table B, values at bottom of column. 
