1895.] Strains in the Testing of Materials and Structures. 131 



Differences. 



Here each ton of load corresponds to 2'56 tons per square inch. 

 The mean extension up to 10 tons is 84' 7 per ton, and corresponds 

 to a Young's modulus, E, of 13600 tons per square inch. The 

 extension per ton increases slightly, but perceptibly, with the later 

 loads, although the almost complete absence of set after 12 tons, 

 and even after 14, shows that there was no very clear passing of an 

 elastic limit, even under the greatest strain of 35'8 tons per square 

 inch. There is just a trace of hysteresis in the relation of strain to 

 stress which shows itself in tho removal of the load. 



These two examples, relating as they do to cases where Hooke's 

 law of the proportionality of strain to stress is very approximately 

 true, may be taken to serve as tests of the sensibility and accuracy of 

 the extensometer. 



In the examples which follow, the limits were passed within which 

 Hooke's law applies. In augmenting the stress so as to pass these 

 limits, it very generally happens that one of the first evidences of 

 overstrain is a time effect a creeping up of the extensometer reading 

 while the load is kept constant for a few minutes. This creeping, 

 which is a familiar phenomenon in the measurement of strains, can 

 usually be detected a good way before the yield point is reached. It 

 is associated with failure in proportionality of strain to stress, and 

 also with permanent, or semi-permanent set. In certain conditions 

 of the material as regards previous treatment, this creeping is far 

 more marked than in other conditions. It shows itself most when 

 the piece has, immediately before the test, been subjected to a load 

 sufficient to cause permanent yielding to occur. But if the same 

 piece is allowed to rest for some days, and is then re-tested, the 

 tendency to creeping is found to have disappeared or to be much 



