Contemporary Advances in Physics, XXX — The Theory 



of Magnetism * 



By KARL K. DARROW 



The topic of this article is the explanation of magnetism as ordinarily 

 observed — to wit, the magnetization of pieces of matter of ordinary dimen- 

 sions — by ascribing magnetic moment to the individual molecules, atoms, 

 and electrons of which matter is composed. For paramagnetic bodies it is 

 postulated that the individual atoms are magnets of which the orientation, 

 but not the strength, is altered in the presence of a magnetic field; the theory 

 is so successful as to make it possible to calculate, from magnetization-curves, 

 values for the magnetic moments of these atoms which agree admirably 

 with those deduced from spectroscopic theory and from experiments of other 

 types. For ferromagnetic bodies the same postulate is made, but it is 

 necessary in addition to recognize the existence of huge interatomic forces 

 of which very little is known, so that a large proportion of the science of 

 ferromagnetism still lies beyond the scope of atomic theory. For diamag- 

 netic bodies the phenomena are interpreted in a simple and effective manner, 

 as an immediate corollary of the well-known structure of the atom. 



"IXTAGNETISM is a quality which we attribute to the atom. We 

 ^^ ^ affirm that iron, nickel, gadolinium, gaseous oxygen, and in fact 

 all substances, are magnetic because there is magnetism in their atoms. 

 Indeed we go even deeper, and affirm that the individual electrons and 

 the nuclei within the atoms are magnetic. Nevertheless, the atomic 

 theory of magnetism is a really valuable theory. Perhaps that 

 "nevertheless" sounds out of place; but I assure you that without it 

 there would be a trace of paradox in the statement, which perhaps our 

 grandfathers would have been quicker at discerning than are we. Let 

 me explain my meaning by referring to the atomic theory, or as it is 

 usually called the kinetic theory, of gases. Those who designed this 

 theory succeeded in explaining the pressure, the temperature, and the 

 viscosity of gases, without attributing a single one of those qualities 

 to the atoms. To the atoms they assigned the properties of momen- 

 tum and velocity and kinetic energy; those other qualities which I just 

 named were then interpreted in terms of these, — ^they were interpreted 

 as what we call statistical properties of the great multitude of atoms 

 which constitutes a gas. This was a real explanation of pressure and 

 viscosity and temperature, in the fullest sense of the word "explana- 

 tion" — or anyhow, in the fullest sense of that word which is customary 

 in physics. But along with these properties of pressure and viscosity 



* Expanded from a lecture delivered on January 14, 1936, at the School of En- 

 gineering of Yale University, and still bearing obvious traces of its original form. 

 In preparing it I received invaluable aid from Dr. R. M. Bozorth. 



224 



