PRESIDENTIAL ADDRESS. GG3 



At first, wLen tLc displacement by sliding is exceedingly sninll, tlie strain is a 

 purely elastic one. The molecules adjacent to the plane of sliding pull one another 

 round a little, but without breaking bonds, and if in this stage the strain is 

 removed, by letting the plate slide back to its original position, there is no dissi- 

 pation of energy. The work done in displacing the molecules is recovered in the 

 return movement. We have here a representation of what happens between each 

 pair of adjoining rows in the elastic straining of a metal. So far the action is 

 within the limit of elasticity ; it leaves no permanent eflect : it is completely 

 reversible. 



But now let the process of straining be carried further. The opposing 

 molecules try to preserve their rows intact, but a stage is reached when their 

 resistance is overcome ; the bonds are broken, and they swing back, unable to exert 

 further opposition to the slip. The limit of elasticity has now been passed. 

 Energy is dissipated ; set has been produced ; the action is now no longer 

 reversible. The model shows well the general disturbance that is set up in 

 molecules adjoining the plane of slip, which we may take to account for the work 

 that is expended in a metal in producing plastic strain. 



Moreover, when the slip on any plane stops and the molecules settle down 

 again, the chances are much against their all taking up the normal orientation 

 which they had before the disturbance. What I have called dissenting groups or 

 unstable coteries are formed as a result of the disturbance. Here and there like 

 poles are found in juxtaposition. A'iewed as a whole, the molecular constitution 

 of the metal in the region adjacent to the plane of slip is now uncertain and 

 patchy. It includes parts whose stability is much less than normal. Individual 

 molecules or small groups in it are very feebly stable ; a touch would make them 

 tumble into positions of greater stability. 



Observe how all this agrees with what we know about the nature of plastic 

 strain through experiments on iron or other metals. Its beginning is charac- 

 teristically jerky. Once the critical force is reached, which is enough to start it, 

 there is a big yield, which will not be stopped even bj' reducing the amount of the 

 straining force. 



Again, we know that there is a slow creeping action that continues after the 

 straining force has done its main work. I ascribe this to the gradual breaking up of 

 the more unstable groups v/hich have been formed during the subsidence of dis- 

 turbance in the earlier stage of the slip. 



Further, we know that overstrained iron is very imperfectly elastic until it has 

 had a long rest, or until it has been raised for a short time to a temperature such 

 as that of boiling water.' This is to be expected when we recognise the presence 

 of unstable individuals or groups resulting from the overstrain. When the 

 elasticity of the overstrained piece is tested by removing and reapplying the load, 

 some of these tumble into new positions, making inversible movements, which 

 dissipate energy and produce hystertsis in the relation of the strain to the stress 

 although the strain is quasi-elastic. At ordinary temperatures these unstable 

 groups are gradually becoming resolved, no doubt through the action of the 

 molecular movements that are associated with heat, and hence the slow progres- 

 sive recovery of perfect, or nearly perfect, elasticity shown by the experiments 

 of Muir. Let the temperature be raised and they disappear much more quickly; 

 in warm surroundings the rest-cure for elastic fatigue does not need to be nearly 

 80 long. 



Rosenhain^ has recently shown that after the slip-bands on the surface of an 

 overstrained specimen have been obliterated by polishing, traces of them will 

 reappear on etching if only a short interval of time is allowed to lapse since 

 the ovei-straining ; but if time is given for complete recovery no traces are 

 found. This is in remarkable agreement with the view now put forward, that the 

 layers contiguous to the surface of slip contain for a time comparatively unstable 



'■ J. Jfuir, « On the Recovery of Iron from Overstrain,' Phil. Trans,, vol. cxciii. A, 

 190C. 



* Journ. Iron and Steel Institute, 1906, 



