APPARATUS FOR MEASURING STRAIN AND APPLYING STRESS. 277 



75 inch pounds of torque. The rate of recovery is also shown by fig. 19, corresponding 

 to fig. 17 of the former case. The readings at 750 inch pounds are plotted as ordinates, 

 and the times as abscissae. The difference between these latter curves is very apparent. 

 From the diagrams it is apparent that the bar recovers very rapidly at first like the 

 wrought iron ; but this rate of recovery soon slackens and becomes less and less 

 apparent as the time increases, and unless a very considerable time is given the 

 recovery does not become complete (cp. Table XV., col. X.). 



VII. — The Position of the Yield-point as affected by previous Stress. 



The effect of a previous stress upon the properties of a bar has been explained in 

 Section VI., and it remains to point out that overstrain in one direction has a very 

 considerable influence upon the yield-point or curve, separating the elastic from the 

 plastic stage ; in fact it disappears, but gradually reappears again — the recovery in the 

 case of the iron being practically complete in two days, while for the steel nineteen days 

 effects partial recovery only. In both cases, if sufficient time be given, the yield-curve 

 will assume a definite position above the last position, and this rise is augmented by every 

 overstrain. As an example, we may take that of a wrought-iron bar, having a length 

 under test of 4 - 00 inches ; diameter 0'420 inches ; calibration 1 min. = 12*8 divisions. 



The bar was subjected to stress extending beyond the yield-point, and afterwards 

 left to rest for a minimum period of 1-^ days, when the load was repeated. Eight 

 tests were made, and each time there was a perceptible rise in the yield-curve. The 

 observations were plotted with each curve spaced 1000 divisions from its neighbour. 

 The curves are given in fig. 20, and require no further explanation. 



VIII. — Twist in alternately opposite Directions. 



C 



It has long been a common assumption that the limits of elasticity for a bar sub- 

 jected to torsion lie equally distant from the position of no torque, and this is no 

 doubt true for a specimen not previously strained. 



Apparently the first theoretical discussion of the problem is that by James 

 Thomson,* and in his original paper he makes the further assumption " that the limits 

 of elasticity in a substance which has already been strained beyond its limits of 

 elasticity are equal on the two sides of the shape which it has when in equilibrium 

 without disturbing force." This note, added in October 1877, goes on to say: "A 

 supposition which may be true or may not be true. Experiment is urgently needed 

 to test it, for its truth or falseness is a matter of much importance in the theory of 

 elasticity." 



The paper further points out that these assumptions lead to the important result 

 that if a wire be overstrained, its strength to resist torsion in the original direction is 

 twice that in the other direction. 



* Cambridge and Dublin Mathematical Journal, 1848 ; and article "Elasticity," Enc. Brit. 



