CONTEMPORARY ADVANCES IN PHYSICS 



2.n 



You may now be expecting me to say that there are many ^ases, and 

 possibly other substances as well, for which experimental curves have 

 been obtained that are comparable with these. I am obliged to disap- 

 point you. You can readily see that in order to get over onto the 

 "curvy" part of these curves, one must work in experimental conditions 

 in which the argument fxHjkT is greater than, or anyhow not very much 

 less than, unity. One thinks, of course, of using the highest accessible 

 field strengths H so as to enhance the numerator of that fraction. 

 This, however, is not sufficient, for it turns out that /u. (the magnetic 

 moment of an atom or a molecule) is so very small that one is obliged to 

 diminish the denominator also by going to the lowest attainable tem- 

 peratures. All the experimental curves of this character have been 

 obtained at temperatures lower than 15° absolute, some at tempera- 

 tures between 1° and 2° absolute. This excludes all the gases. More- 

 over, it has been necessary thus far to choose the atoms with the largest 

 magnetic moments, and these turn out to be, quaintly and inconven- 

 iently enough, the atoms of the rare-earth elements. Probably the 

 best of the experimental curves (Fig. 1) relates to a substance which 

 most people never have heard of, in this or any other connection: it 

 is gadolinium sulphate. There are about a score of such curves ob- 



0.9 



0.7 

 0.6 



0.2 



0.1 



0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 



Fig. 1 — Magnetization of a paramagnetic salt [Gd2(S04)3 ■\- 8H2O] as function of 

 the parameter a; the ordinate is / referred to its saturation-value (deduced bv 

 extrapolation) as unit y . Data from Onnes and Woltjer. The curve is the "classical 

 l.angevin curve (number of permitted orientations, m = « ) which is hardly distin- 

 guishable from the quantum-theory curve for this particular case (« = 15). 



