MOLECULAR HYPOTHESIS 



195 



meats it is proportional to the displacement, and on removal of 

 the magnetising force there is complete return. With large dis- 

 placements, however, there is something analogous to permanent 



n & 



\ 



\ 



6' 71 



FIG. H2. 



set and therefore not complete return. He did not attempt to 

 account for the permanent set, but simply assumed it. 



Ewing's theory and model. Ewing brought a great 

 simplification into the hypothesis by showing that the mutual 

 action of the molecular magnets would account for the controlling 

 or constraining forces, tending to restore the molecules exactly 

 after small displacements, and for the permanent set, that is, the 

 residual magnetisation, after large displacements. Thus no sort of 

 friction need be assumed, and the hypothesis in its simple form 

 accounts for the facts. To illustrate the ideas he constructed what 

 is known as Ewing s model of a magnet, consisting of a number of 

 compass needles placed together as in Fig. 143 (in plan), each needle 

 representing a molecule. Such a model, if inserted within a 



FIG. 143. 



magnetising solenoid in place of the bar in Fig. 129 shows all the 

 phenomena of magnetic induction and gives a curve of magnetisa- 

 tion repeating all the features of the hysteresis curve in Fig. 131.* 

 To understand how these mutual molecular actions account for 

 the phenomena let us follow Ewing in supposing that we have 

 four needles arranged at the corners of a square. With no external 

 magnetising force and under their own mutual actions only, they 

 will set as in Fig. 144, where the arrow-heads are north-seeking 



* Induction in Iron and other Metals, 3rd ed. p. 351. 



