treated according to Van der Waals's Equation. 351 



The thermal energy may be written RT, hence 



2R 

 T 2 



£ISRT = RT, or R' = t2 , 



where R is the gas constant. When R' is calculated in this 

 way for iron, nickel, and cobalt, it is found to be in 

 sufficiently good agreement with experimental values to show 

 that it may be safely used in determining a', and hence in 

 determining the magnitude of the intrinsic field (a'I 2 ) which 

 is thus found to be about 10 7 gausses. 



There is another consideration w r hich may be cited in this 

 connexion. A ferromagnetic substance above the critical 

 temperature behaves as a paramagnetic body, and the inten- 

 sity of magnetization is directly proportional to the field- 

 strength. Let a field now be applied, if possible of 10 7 

 gausses, then the susceptibility being constant the intensity 

 will become as great as the maximum in the ferromagnetic 

 state. If, now, the temperature be lowered below the critical 

 temperature, the field still being in action, no change occurs 

 in the intensity, but the substance has changed from the 

 paramagnetic to the ferromagnetic state, and that without 

 any abrupt transition. According to this argument the two 

 states are continuous and the magnitude of the ferromagnetic 

 field is of the order of 10 7 gausses. 



Again, the progressive increase of the specific heat of a 

 ferromagnetic substance as it is heated up to the critical 

 temperature, when a sudden diminution occurs, is quanti- 

 tatively accounted for with a high degree of accuracy by the 

 gradual loss and final destruction of a field of the order of 10 7 

 gausses*. Thus thermal considerations lead to the conclusion 

 that the intrinsic field is of enormous magnitude. This con- 

 clusion is borne out also by the results of an electrolytic 

 experiment on the electromotive force of magnetization to 

 which I have referred in a former paper f . 



On the other hand, if this intrinsic field is a uniform one 

 and of magnetic origin, then only a small fraction of its lines 

 of force are observable on the outside of the magnetized 

 body, and, moreover, there are difficulties in accounting for 

 the source of so large a field if the molecular specific mag- 

 netism is of the same order as that of the whole mass at 

 its maximum. 



For the present, therefore, it may be left an open question 



* Weiss & Beck, J. de Phys. ser. 4, t. 7. p. 249 (1908). 

 t Ashworth, Phil. Mag. vol. xxx. p. 713 (1915) ; Mem. & Proc. Man- 

 chester Lit. & Phil. Soc. vol. lviii. part ii. (1914 k . 



