CONTEMPORARY ADVANCES IN PHYSICS 225 



and temperature, a gas also possesses weight. The builders of the 

 kinetic theory simply said that the weight is a property of the individual 

 atoms, and that the weight of the gas is the sum of the weights of its 

 atoms. Now evidently this was not an explanation of weight at all. 

 Indeed, by assigning weight to the individual atom, the builders of the 

 theory had foregone all attempts at an explanation. A property which 

 you assign to the atom is a property which you refuse to try to explain 

 in terms of the atom — or so at least it always seemed to our fore- 

 fathers. To assign a quality to an atom used to be taken as a confes- 

 sion of incompetence to explain that quality. I can of course make this 

 clear by proceeding to absurd extremes. If I say that an orange con- 

 sists of soft yellow juicy atoms, or that a marshmallow is made of sweet 

 white sticky atoms, or that a piece of iron is made of hard black shiny 

 conductive atoms, you recognize at once that those are not serious 

 atomic theories: they are just futile and somewhat ridiculous state- 

 ments. If I claim to explain the weight of a piece of iron by saying that 

 it is the sum of the weights of the atoms, I am making a claim which 

 unfortunately may not sound ridiculous, but is really just as futile — 

 unless it acquires value by being linked with some other assertion. 

 But when I say that the magnetism of a piece of iron is due to the 

 magnetism of its atoms and its electrons, the statement is by no means 

 a futile one; it is significant and important. For this there are two 

 main reasons or rather groups of reasons, which I will indicate by the 

 words orientation and atomic structure. (In addition there are remark- 

 able experiments on jets of atoms whereby their magnetic moments 

 are measured directly, but these I reserve for another occasion.) 



First a few words about atomic structure. It is a fact of experience 

 — the experience of one hundred and fifteen years— that a current 

 running around a loop of wire is the equivalent of a magnet. If now 

 somebody asserts first that a piece of iron is magnetic because its atoms 

 are magnets, and then goes right ahead and asserts that the atoms are 

 magnets because they have perpetual currents running around inside 

 them — well, the combination of these two statements is not necessarily 

 futile or trivial. At the very least, it is a sensible attempt to reduce 

 the two kinds of magnetism apparently existing in the world to a single 

 kind, that which is due to moving electricity. This was Ampere's idea 

 a hundred years ago. Now if in addition there is independent evidence 

 that the atom comprises mobile electrical particles, then this idea of 

 Ampere's becomes the assertion that those particles inside the atom are 

 actually revolving. It is well known that modern physics is full of 

 such evidence, of evidence that atoms contain very mobile electrons; 

 and some of my readers may recall that thirty years or so ago there 



