and Rocks under Pressure. 253 



out the crystal. This specimen was exposed to 3000, 

 5000, 7000, 10,000 and 12,000 kg. for 10 minutes each. At 

 12,000 there was appreciable flaking-off of the inner sur- 

 face, similar to that of quartz at lower pressures. But 

 there was no extensive erosion, so that tourmaline 

 appears as a much stronger crystal than quartz. At 

 7000 and 10,000 minute longitudinal splinters were sep- 

 arated from the inner surface, but this ceased at 12,000. 

 Very minute cracks were formed in the polished faces at 

 the ends of some of the striae. At 12,000 cracks at an 

 angular separation of 120° had started, like those 

 formed in quartz at 6000, but they were small, and there 

 was none of the funnel-shaped appearance of the quartz. 

 It was a surprise that there was no connection between 

 the new cracks which appeared and the original flaws. 

 There was also a system of cracks unlike those of quartz, 

 cracks in planes perpendicular to the axis, reaching in 

 some cases to the outer surface. One such crack was sit- 

 uated diametrically below the end of the hole. No phe- 

 nomena of flow could be detected. 



The behavior of this cylinder, as well as that of the 

 quartz cylinder, was peculiar when compared with ordi- 

 nary substances. Iron and copper, for example, when in 

 the form of cylinders like this, 3 will also stand a much 

 higher stress than indicated by ordinary compression 

 tests, but they do it by a process of accommodation. 

 Under stress they flow viscously until the grains have 

 become properly arranged to stand a higher stress. But 

 the quartz and tourmaline cylinders show no flow, and 

 the only analogy to accommodation is flaking-off. I found 

 it most difficult to see how flaking-off of the inner surface 

 could result in enhanced resistance to rupture, and could 

 think of no explanation except that the effect of time had 

 been neglected. The same cylinders of tourmaline were 

 therefore exposed again to 12,000 kg., this time for two 

 hours. If tourmaline were like an ordinary metal, no 

 effect would be produced by the second application of 

 the maximum stress ; there would be no further cracking 

 or flaking-off, because the elastic limit had been raised 

 by the previous application of pressure. But the effect 

 was the exact opposite. Under the longer duration of 

 stress the cylinder failed completely, being packed tightly 

 full with an impalpable sand. The region of erosion of 



3 P. W. Bridgman, Proe. Amer. Acad., lac cit. ; Phys. Kev., 34, 1-24, 1912. 



