Hydrostatic Pressure and Conditions of Rupture. 65 



hold without modification, for it demands that a solid body 

 when subjected to uniform hydrostatic pressure all over 

 should break when the pressure rises too high. It is incon- 

 ceivable in this case how the body can break, and the 

 existence of the heavenly bodies with enormous internal 

 pressures is conclusive proof to the contrary. It is still 

 possible, however, that this criterion should be valid as 

 determining set, and that the body might show volume set 

 if the pressure exceeded a certain critical value. This is a 

 question of some interest, and it might possibly seem 

 reasonable at first sight to expect some such set. Kahlbaum * 

 has published results on this subject showing the very 

 surprising result that after the application of pressures up to 

 10,000 atmos. the density of most metals is increased, due 

 to closing up of small pores, but that beyond 10,000 the 

 density is decreased. Although no systematic investigation 

 of this question has been made in the present work, several 

 incidental determinations have been made up to 25,000 

 or 30,000 atmos., and no trace of a change of density has 

 ever been found except in those cases where the metal was 

 obviously porous. There seems to be no question but that 

 Kahlbaum's results were due to the fact that he used castor 

 oil to transmit pressure. This freezes under a few thousand 

 atmos., a fact that Kahlbaum overlooked, so that beyond 

 this, his pressure was no longer hydrostatic. This is shown 

 most strikingly by Kahlbaum's own statement that the test 

 cubes were often curiously distorted after application of 

 pressure. Kahlbaum saw in this support of his theory that 

 all metals tend to become plastic or even fluid under high 

 hydrostatic pressures. But it can be stated unqualifiedly 

 from the present work that, on the contrary, substances tend 

 to become more rigid under high pressures. The distortion 

 of the metal cubes found by Kahlbaum is to be explained by 

 the fact that the effect of pressure in increasing rigidity is 

 very different for different substances. It has been found 

 in the present work that under 20,000 atmos. paraffin wax 

 may become more rigid than Bessemer steel. 



The first criterion is usually modified, therefore, so as to 

 predict rupture when the tension alone, instead of either 

 tension or compression, exceeds anywhere a critical value. 



The first type of test has its most direct bearing on the 

 maximum tensile stress criterion of rupture. In these tests 

 cylinders were exposed to pressure over the curved surface 



* Kahlbaum, Koth, und Seidler, Z.S. Anorg. Chem. pp. 29-30, 

 254-294 (1902). 



Phil Mag. S. 6. Yol. 24. No. 139. July 1912. F 



