MOLECULAR HYPOTHESIS 199 



which they had under that smaller value when the value of H was 

 rising through it. 



Fig. 154, taken from Ewing (loc. cit. p. 351), shows the magne- 

 tisation curve of a group of twenty-four pivoted magnets when the 

 magnetising force is taken through a complete cycle. 



Dissipation of energy in a hysteresis cycle. A 

 study of the behaviour of the members of a model at once suggests 

 that there is dissipation of energy when a piece of iron or steel is 

 taken through a cycle. For work is done on the magnets in putting 



FIG. 154. 



them into line, and when on the diminution and reversal of the 

 magnetising force a point of instability is reached, the magnets fly 

 round and vibrate about their new position through large ampli- 

 tude. This means that some of the work done appears as energy 

 of vibration. In the molecular magnets we should expect similar 

 behaviour, and the molecular vibration would imply the develop- 

 ment of heat and a rise in temperature of the iron. In fact, it can 

 be shown that the area of the hysteresis loop is proportional to the 

 energy dissipated in a cycle. 



The cycle may be performed in another way than that which 

 we have supposed, viz. by the rotation of the specimen in a con- 

 stant field. This is what takes place in a dynamo armature when 

 the core is a cylinder of iron rotated between strong poles. Besides 

 changing the magnetisation, the rotation will tend to produce 

 electric currents in the iron and to prevent these currents the iron 

 is laminated transversely to the direction in which the currents 

 would flow, so that they are practically destroyed. The work done 

 in a revolution is then due to the hysteresis. The molecular groups 

 as they move round from a or b in Fig. 155 tend to keep in their 

 original line, given to them by the magnetising force, so that at 

 c or d they should be perpendicular to the magnetising force ; but 



