258 P. IT. B ridyman— Failure of Cavities in Crystals 



conceivable Hint interstitial spaces as large as 50 or 30« , 

 will persist, but on the other hand, will the powder fuse 

 to a compact mass? The conditions here were exception- 

 ally favorable for getting some idea of the answer to this 

 question since pressure could be raised in the available 

 apparatus to more than three times the disintegrating 

 pressure. If fusing by pressure alone is ever to be 

 expected, there is a good chance to find it here. For it is 

 to be considered that the manner of formation of the 

 sand, by automatic flaking-off of the solid material inside 

 a closed cavity, ensures that the grains shall be clean. 

 This is an important point, and one that is most difficult 

 to obtain under ordinary conditions of experiment. 



In an attempt to produce fusion of the powder, pics- 

 sure was maintained at 11,400 for 39 hours. The crystal 

 was thereby changed in appearance throughout the entire 

 mass to the opaque white of marble, showing that slip 

 had spread throughout the entire substance. At the 

 inside there was a core of fine sand which had not fused 

 together, but could readily be picked out with a wire. 

 The average density of the entire cylinder was decreased 

 somewhat, as had also the external diameter. In the 

 marbleized region, there were no cracks except a few fine 

 fissures near the core itself. Obviously, then, this mate- 

 rial may suffer flow without losing its mechanical 

 coherence, but if the particles have ever been separated 

 too much, as when the grains of sand are formed, even a 

 relatively high pressure will not cause them to fuse 

 together. The core of sand had lost all rhomboidal shape, 

 and was now ellipsoidal with angular ends. Small fis- 

 sures radiated from the ends, in appearance extremely 

 like lines of force about a bar magnet. 



In spite of this experiment, the conviction is hard to 

 escape that it must be possible to weld together the frag- 

 ments of a solid merely by bringing them into contact, 

 provided they are perfectly clean. In this connection 

 Langmuir's 4 recent work on the significance of the role 

 played by adsorbed layers of a gas only one molecule 

 deep is most suggestive. The particles of sand formed 

 from the barite appeared by microscopic analysis to be 

 of the order of 0-001 mm. in diameter. A very rough cal- 

 culation shows that the air originallv in the cavitv was 

 much more than sufficient to cover all the grains of sand 



4 1. Langmuir, J. Amer. Chem. Soc, 38, 1145, 1916. 



