and liocJcs under Pressure. 263 



Glass. — I have already referred to former experiments 

 in which a sealed glass capillary had been exposed to 

 24,000 kg. without effect, while cavities in copper have 

 been squeezed out of existence by 10,000 kg. Some essen- 

 tial difference between crystalline and non-crystalline 

 materials suggested itself, which it was the partial pur- 

 pose of these experiments to examine. A control speci- 

 men was therefore made of optical glass, exactly like the 

 second specimen of quartz, and it was subjected to the 

 same sort of treatment. It was first exposed to 5100 kg. 

 and pressure immediately released, with no effect what- 

 ever. This confirmed the result with the capillary tube, 

 that stresses much higher than those reached in ordinary 

 compression could be sustained if the material is so 

 arranged as to afford itself mutual support. After this 

 test at 5100 it was a serious question how to proceed. 

 Previous experience had shown that glass is exceedingly 

 likely to receive some sort of internal strain under pres- 

 sure so that on the second application of pressure it is 

 much more likely to rupture than on the first. If a com- 

 plete series of tests were made, as with quartz, there was 

 danger that this effect would obscure the results, whereas 

 if pressure were at once raised to the maximum and 

 rupture found to have been produced, there would be 

 no way of telling at what pressure rupture had occurred. 

 I optimistically chose the second method and lost. 



Pressure was raised to 12,000 kg. and immediately 

 released. Rupture was complete. Failure to reproduce 

 the result found with the capillary was doubtless due to 

 incomplete alignment of the two halves. The results 

 were nevertheless instructive in that the manner of rup- 

 ture was entirely different from that of quartz. The 

 entire mass of glass was filled with haphazard cracks, 

 many of them running through to the outside. Each 

 crack was curved and changed direction many times in a 

 complicated fashion, quite unlike the simple cracks in the 

 crystals. These cracks were more numerous at the inner 

 wall, where they interpenetrated each other so extensively 

 as to produce an apparent erosion of the interior like that 

 of the minerals. But the eroded fragments were large 

 instead of an impalpable powder, and in places the orig- 

 inal polish of the walls of the cavity w r as still intact. It is 

 therefore probable that the mechanism of rupture by 

 flaking off was entirely absent in this specimen of glass. 



