Manchester Memoirs, Vol. li. (1907), No. 15. 17 



gradually all over the group. To show next that this 

 theory is in harmony with what we know about elastic 

 straining and non-elastic straining, we will take a group 

 of molecules in two portions, separately mounted on two 

 glass plates, which can be caused to slide past one 

 another. 



Notice what happens when I make one portion of the 

 " crystal " slip past another portion. Up to a certain 

 point there is nothing in the nature of permanent strain. 

 If you watch the molecules which are adjacent to the line 

 of slip, you will see that they are striving, as it were, to 

 hold together. That corresponds to elastic straining. In 

 any strain causing distortion, you can imagine slipping 

 to take place to an infinitesimal extent on any number of 

 planes on the crystal, and on any one of these planes you 

 simply have a slight displacement of the neighbouring 

 molecules, a slight turning of the neighbouring polar 

 axes, but without breaking bond. Suppose we cause 

 the strain to be more violent, the bonds will be broken, 

 and the molecules will swing round into new positions. 

 There will be dissipation of energy, and there will be no 

 recovery ; in other words there will be permanent set. 

 In the model you see this dissipation of energy, when 

 one of the plates is caused to slide so far past the other 

 as to break the molecular bonds. The molecules swing 

 round into new positions, and the energy which they 

 thereby dissipate is lost by its being spread throughout 

 the substance of the piece, and probably spent in gene- 

 rating internal electric currents. That is a non-elastic 

 strain as I conceive it. 



Notice too the effect of periodic straining in opposite 

 directions many times over, which we imitate in the model 

 by making one of the glass plates slide a little way 

 backwards and forwards repeatedly past the other. This 



