4 8o SCIENCE PROGRESS 



Aj, a 2 , and a 3 , and that the parts following these points are 

 horizontal, indicating a constant couple. Apparently AxNj, 

 a 2 n 2 , etc., are the respective elastic limits for the specimens. 

 No further decrease of couple occurs at constant torsion, and it 

 seems that in these cases the limits are reached after the finite 

 times on 1( on 2 , and on 3 . These curves were obtained by an 

 automatic process, and the time scale used was not sufficiently 

 large to justify fitting equations of the Boltzmann type ; but 

 they at least appear to point to the fact that the stress may reach 

 its final value after a finite time, instead of after an infinite time 

 as might be readily supposed. 



This account of the effects of over-straining materials is by 

 no means exhaustive. It is only intended to call attention to a 

 subject which has been somewhat neglected. More detailed 

 accounts of individual work can, of course, be found in the 

 actual papers to which reference is given, and if this abstract 

 succeeds in creating sufficient interest to cause them to be read, 

 it will have served its purpose. It would not be well to 

 conclude, however, without mentioning certain effects resembling 

 those of over-strain. It is known that analogies exist 

 between the phenomenon of elastic after-effect and many 

 other physical phenomena, and considerable research has 

 been undertaken with the object of establishing possible con- 

 nections between them ; but much yet remains to be done, for, 

 although the various effects have been shown to possess many 

 similarities, the work has not yet proceeded sufficiently far to 

 definitely establish, or otherwise, a general law. Hesehus, in 

 his paper on " Elastische Nachwirkung" published in 1882, seems 

 to have expected that some relation would be shown to exist, 

 particularly in the cases of residual charge on electrical con- 

 densers, magnetic hysteresis, and the decay of phosphorescence. 

 It is well known that when a condenser has been charged, the 

 discharge is not at once complete upon connecting the coatings, 

 but that a further charge develops on them if they are once more 

 insulated. Or, if connection is permanently established, a small 

 current continues to flow for some time after the initial 

 discharge. This secondary discharge is in many respects 

 analogous to the slow recovery which follows the immediate 

 recovery in an over-strained body when suddenly relieved 

 from external stress. Again, it has already been pointed out 

 that it is possible to obtain curves of elastic hysteresis 



