1068 
Physics. — “On the reversible change of the remanent magnetisation 
with the temperature’. By Dr. G. J. Bras. (Communicated 
by Prof. H. A. Lorentz). 
(Communicated in the meeting of December 18, 1915). 
When the temperature of a ferromagnetic body that possesses 
remanent magnetism rises, the magnetism changes too. Let us suppose 
e.g. that a decrease of the magnetisation corresponds to a rise of 
temperature. If then the magnet is again cooled down to its original 
temperature, the magnetisation increases again, but does not return 
to its original value. A following rise of temperature equal to the 
first one causes again a decrease of the magnetisation, but now a 
smaller one than was caused by the first rise. The next return 
to the original temperature brings the magnetisation only back to 
a value below that after the first cooling. With continued changes 
of the temperature this goes on in the same way; the differences 
between the magnetisation after two subsequent equal changes of 
the temperature in the same sense become always smaller, till finally 
a state is reached in which the changes are reversible, that is to 
say, that by a rise of temperature the magnetisation decreases as 
much as it increases again by the following equal fall of the 
temperature. This reversibility only holds however between the 
used limits of temperature; if these limits are taken wider, the 
cyclic process of heating and cooling must first again be gone through 
a few times, before the reversibility is reached. This reversible 
change we shall investigate here. With a change of temperature 
without more we shall denote the reversible change. 
On this subject there are experiments of AsHworTH ') and DurwarD’). 
Both have investigated different kinds of steel; and cast-iron. The 
first author obtained very different values for the change with the 
temperature, according to the kind of steel used. In most cases the 
coefficient a, defined by the equation 
M= Mi, [1 — a(t, — t,)), 
was found to be positive. In this formula J/;, resp. J, is the 
magnetisation at the temperature 7, resp. ¢,; the highest value found 
for steel was about 0,0014, for cast-iron even 0,0029. For some 
kinds e.g. for piano-strings @ had a negative value even down to 
— 0,0003. Further « proved to be strongly dependent on the thermal 
history (process of hardening) of the material, while also the influence 
1) G. R. Asuworrn. Proc. Roy. Soc. A. 62 p. 210, 1898. 
*\ Durwarp. Sill. Journ. 5 p. 245, 1898, 
