Abstract. The changes of size of iron and nickel at their Curie points are too great to be explained by purely magnetic forces. On Heisenberg's theory of ferromagnetism, magnetization is accompanied by a change in the strength of binding between atoms, and the resulting changes of size of crystals of iron and nickel at their Curie points are of the same order of magnitude as those observed.

The calculations for iron make use of a modification of Heisenberg's original theory, in which the restriction that each atom possesses only one electron concerned in magnetic effects is removed. A consideration of the specific heat discontinuity at the Curie point shows, however, that the theory is far from satisfactory.

The change in size which accompanies magnetization varies continuously from the Curie point to saturation, but does not affect measurements of magnetostriction at ordinary temperatures.

Print publication: Issue 5 (15 August 1930)