treated according to Van der WaaWs Equation. 349 



of an intrinsic pressure, and consequently there is a deve- 

 lopment or! latent heat. 



Where there is a destruction or creation of the magnetic 

 intrinsic field, as there is by heat in the vicinity of the 

 critical temperature, la roe thermal effects are to be expected 

 such as have been found *. 



3. An attempt may be made to determine the ratio (jii) of 

 that part of the intrinsic field which is put in evidence (H t -) 

 to the field (H) which is externally applied. 



If H be very small compared to H l5 it may be numerically 

 neglected and then the ferromagnetic equation becomes 



MH)= R ' T 



where R' is the paramagnetic constant. Also it has been 

 previously shown that the equation to the anhysteretic 

 isothermals, 



H (K) =HiT ' 



is approximately true when R T is a constant")", hence by 

 division we have 



Hy _ TV 



H ~ Rj" 



Now R' is the reciprocal of: Curie's constant and has the 

 value 3*56 for iron and 208 for nickel. An estimated value 

 of R x deduced from some recent experiments, not yet pub- 

 lished, is 0-77 xlO" 6 for iron and 3*27 x 10" 6 for nickel J, 

 when the maximum intensities are 1685 and 510 for these 

 metals respectively, hence, 



XT TJ / 



yi6 7 = ~ = =- =4-6 X 10 6 for iron and 0*4 x 10 6 for nickel, 

 li Hi 



4. Using this ratio the external critical field (H dC ) which 

 corresponds to the critical field (H c ) is H ec =H c /yu, t , and 

 therefore, inserting numerical values, H ec = 0'18 for iron and 

 0*14 for nickel. 



From the value for the critical field the external field 



* Hopkinson, Phil. Trans. Roy. Soc. A. 1889, p. 443 ; Honda, Sci. 

 Reports, Sendai, Japan, vol. ii. no. 2 (1913). 



t Phil. Mag-, xxvii., Feb. 1914. 



} These numbers are derived from curves of 1=6(11) constructed from 

 curves of I=/(T) in which hysteresis was suppressed by heating above 

 the critical temperature, readings being taken as the metal cooled. 



