CONSTITUTION AND TEMPERATURE ON MAGNETIC SUSCEPTIBILITY. 207 



(5) FURTHER DISCUSSION OP THE NATURE OF THE LOCAL MOLECULAR FIELD 

 . IN FERRO-MAGNETIC AND DIAMAGNETIC MEDIA. 



This subject has already been discussed in Part III., pp. 89 and 100, but no 

 interpretation was then given to the magnitudes of the constants N and a! c of the 

 local ferro-magnetic and diamagnetic fields. The values of N given by WEISS and 

 BECK* are 0'38x 10* for iron, and l'27x 10 4 for nickel. For diamagnetic crystalline 

 media a' e is of the order 2 x 10*. 



EwiNG and Lowt have shown that in very strong magnetic fields the relation 

 between induction (B) and applied field (H) may be represented by the equation 



B = H + a constant (l) 



This constant has the value 47rl (where I is the saturation intensity) and is equal 

 to 21,360 in wrought iron, 6470 in nickel and 16,300 in cobalt. 

 In the case of wrought iron 



B = H + 213GO (2) 



Suppose we could apply a field equal to the molecular field, 6'53xl0 6 gauss for 

 iron. The limiting value of the permeability fj. L for this field will be from (2) 



21360 



'6-53x10" 



and the limiting susceptibility .per unit volume 



/A.,-1 21360 



L IT" 4.x 6-53 

 and 



= 



= 0'38xlO. 

 XL 



This is equal to the value of N, the coefficient of the molecular field, as we should 

 expect. 



Similar calculations may be made for nickel and cobalt, the limiting susceptibilities 



being respectively 



XL = 0'81 x 10- 4 for nickel, 



XL = 2'0 x 10~ 4 for cobalt. 



Now we may ask the question, why is it that, in spite of the fact that all the 

 molecules are ordered into a definite space lattice under the influence of the respective 

 molecular fields, the materials still show a finite susceptibility to magnetization ? The 



* ' Journ. de Phys.,' s<5r. iv., vol. 7, p. 249, 1908. 

 t ' Phil. Trans. Roy. Soc.,' A, p. 242, 1889. 



