CHAPTER XV 



MOLECULAR HYPOTHESIS OF THE 

 CONSTITUTION OF MAGNETS 



Molecular hypothesis Ewing's theory and model Dissipation of 

 energy in a hysteresis cycle Temperature and magnetisation 

 Attempts to explain the constitution of molecular magnets. 



Molecular hypothesis of the constitution of magnets. 



The fundamental phenomena of magnetisation receive an explana- 

 tion on the hypothesis that the molecules of iron and steel are 

 themselves small permanent magnets, capable of being turned 

 round their centres. This hypothoi^ \\a> lir>t propounded in 

 definite form by Weber.* It has .since been developed by 

 G. Wiedemann, Hughes, Maxwell, and above all by Ewing.f 



In the eighteenth century Aepinus supposed that magnetism 

 consisted of two fluids with opposite properties, uniformly mixed 

 in an unmagnetised bar, and that one ua^ pulled to one end and 

 the other pushed to the other end under a magnetising force. 

 Coulomb and others supposed that the separation \\ent on in each 

 molecule separately, and this was the first step in the molecular 

 hypothesis. Poisson developed it in this form and in\< I iS 



mathematical theory. But Weber gave it a new form, now univer- 

 sally adopted, in supposing that each molecule of iron or -ted, or at 

 least each molecule which contributes to the magnetic condition, is 

 itself a permanent magnet, with definite polarity, and capable 

 rotation about its centre. No attempt was made to account for 

 the magnetisation of the molecules, so that the hypothesis attempts 

 to give an account of the structure of a magnet as built up of smaller 

 magnets in a particular way, and does not in any way explain 

 magnetism. 



According to Weber, the axes of these molecular magnets in 

 an unmagnetised bar are pointing equally in all directions, as 

 represented in Fig. 139, so that in any space small compared with 

 the whole magnet, but large compared with that occupied by a single 

 molecule, an equal fraction of the whole number of molecules will 

 point in every given direction, and so neutralise each other's external 



* Pogg, Ann. Ixxxvii. (1852), p. 145. 



t Ewing's paper describing his magnetic model is in Proc. X.S., xlviii. 

 (1890), p. 342. He gives an account of the molecular hypothesis in chap. xi. of 

 Magnetic Induction in Iron and other Metals. 



192 



