138 Prof. J. A. Ewing. On Measurements of small [May 16, 



Difference per 

 Load in tons. Extensometer. half ton. 



2 377 46 



2| 422 45 



3~ 469 47 



3| 517 48 



4" 565 48 



4| 613 48 



5 662 49 



200 , 48 



5 661 

 5| 710 49 



6 760 50 

 6-|- 812 52 



7 866 54 

 7f 920 54 

 S~ 975 



977 After 30 seconds. 

 208 



203 After 2 minutes. 



8 974 



977 After 1 minute. 

 210 



207 After 30 seconds. 

 202 After 3 minutes. 

 200 After 20 minutes. 



As in the example already given, the influence of overstrain is 

 apparent (1) by the presence of creeping or " nachwirkung," and (2) 

 by the progressive growth of the differences as the load increases. The 

 first ton or so produces little more strain than it did before the over- 

 strain took place, but the eighth ton produces a quarter as much 

 again. 



An hour after the yielding had taken place the load was again re- 

 applied in stages of 1 ton as follows, and the extensometer readings 

 already show something of elastic recovery through rest. They are 

 given in the following table, along with readings taken on the follow- 

 ing day and on subsequent days. It is interesting to notice the 

 rather slow progress of the elastic recovery from day to day. It takes 

 place much less rapidly in this comparatively hard metal than in 

 milder steel or in wrought iron. To facilitate comparison the 

 readings taken immediately after the overstrain are repeated here : 



