BEHAVIOUR OF OVER-STRAINED MATERIALS 479 



limiting value of stress be D z . Then when t has the value given 

 by 



the stress in the body is equal to the elastic limit, and, there 

 being now no tendency towards increasing deformation, no 

 further decrease in stress is necessary to maintain constant 

 strain. In this case the stress would become reduced to the 

 elastic limit also within a finite space of time. It is obvious that 

 the formula must cease to be valid as soon as this point is 



Fig. 3. 



reached, for the stress will neither take a negative value nor a 

 value less than the elastic limit. Possibly it does not hold good 

 even up to that point, for one would expect that the stress 

 would not be reduced to its final value until after an infinite 

 time — i.e. the curve representing stress against time would 

 approach asymptotically to that value. Some light is thrown 

 on this question by the curves obtained by the writer 1 for jellies 

 of various concentrations. They are shown in fig. 3, and 

 represent the variation in the couple occurring with progress of 

 time when the torsion is constant. It will be seen that the 

 curves, instead of being smooth throughout, show breaks at 



1 A. O. Rankine, Phil. Mag., April 1906. 



