336 
MR. J. J. THOMSON ON SOME APPLICATIONS OP 
Now, suppose that no heat is supplied to the body, but that 77 is increased by dr), 
the vs will remain constant, so that 
{“Phi 
(90) 
or if/( 77 ) be small compared with unity 
since in this case we may write 
= hf'(v)®Zv 
thus 
or 
©: 
KM 
4- . . • nearly 
P© 
dr/ 
=U'(v)® 
■ J, y,® by equation (87) 
iiB_ e <a 
<71 *0 <70 
(91) 
(92) 
(93) 
if I be the intensity of magnetisation ; if H be the magnetising force we may write 
this equation, since I = Z:H, as 
P© _ PI 
PH - ”® P©’ 
where the magnetising force is supposed to remain constant in the differential 
coefficient on the right hand side of this equation and the coefficient of magnetic 
induction on the left. 
Thus if the coefficient of magnetic induction increases with the temperature, the 
temperature of a soft iron bar will be lowered by magnetisation. It is difficult to tell 
from the experiments which have been made what the effect of temperature on the 
coefficient of magnetic induction really is; according to Wiedemann, when the 
substance has been repeatedly heated and cooled down again to its initial temperature, 
the coefficient of magnetic induction of soft iron diminishes as the temperature 
increases, while for hard steel it increases with the temperature provided the tempera¬ 
ture does not exceed a certain limit. Thus the temperature of hard steel ought to 
fall when it is magnetised and the temperature of soft iron to rise. As far as is 
known the temperature of all bodies rises on magnetisation, but this may be due to 
the electric currents induced by the sudden starting of the magnetic field; unless the 
heat generated by these currents is allowed for the results tell us nothing as to 
whether the temperature is raised or lowered when the magnetisation is increased. 
