78 Mr. P. W. Bridgman on Breaking Tests under 



yield begins, whereas before yield the mean stress was a 

 tension, because the circumferential tension was greater 

 than the radial compression. The strain, then, after yield, 

 consists of a volume compression at the inner surface, 

 whereas before yield there was a volume extension. Con- 

 sider now what would happen at the inner surface if rupture 

 were to get started there. We are to think of the general 

 action of rupture as one relieving the stresses at any point, 

 for otherwise the rupture would have no excuse for existence. 

 Before plastic yield, then, rupture at the inner surface would 

 result in release of the volume distension, that is, the volume 

 of the steel would become less. As a result, the compressing 

 liquid would have room to become greater in volume, and 

 the pressure would automatically fall. But if the rupture 

 were to start after the plastic yield, that is after volume 

 compression had taken place at the interior surface, release 

 of the pressure would result only in increase of volume of 

 the steel, the compressing liquid would lose volume, and the 

 pressure would as a result be increased instead of decreased. 

 In this case, rupture, instead of affording relief from the 

 existing state of affairs, would only intensify them. That is, 

 rupture would be an explosive affair, during which work 

 was done against the applied forces instead of by them. 

 This is opposed to all experience. This simply means, then, 

 that even if rupture on a small scale should start at the 

 interior surface during the state of plastic yield, it would be 

 unable to spread. If rupture is to start at the centre and 

 travel out it must start before the stage of plastic yield is 

 reached. 



The photograph of the 4 in. Bessemer-steel cylinder, fig. 4, 

 shows this most strikingly. The interior surface is covered 

 with numerous slip bands, some of them very prominent, 

 where the rupture had started but was unable to spread, 

 The pressure on the plastic metal at the inner surface lias 

 been so high that a process similar to cold-welding has gone 

 on. The crystalline grains were probably torn apart so as to 

 slip past each other, but the pressure was great enough so 

 that the separation was never beyond the range of molecular 

 attraction, and true rupture did not occur. The copper 

 cylinder also shows the same slip bands at the interior 

 surface. 



Another interesting question in this connexion is that of 

 the greatest possible raising of the elastic limit by permanent 

 stretching. It is easy bo see that the stress distribution in 

 such a cylinder after the yield-point has been passed ami the 

 pressure released is exactly like that in a gun with shrunk-on 



