MAGNETIC RESONANCE. II 391 



nuclei — ■ protons — ■ of the molecule compensate one another in some 

 of the molecules, enter into the parallel coupling in others. There are 

 always some of these last in a beam of hydrogen molecules, and they 

 produce the proton resonance of which so much was said in Part I. 

 The atoms produce the electron resonance. 



Look now again at equation (4), and remember that p is Ih/2Tr — and 

 remember that p is to be interpreted as the maximum permitted com- 

 ponent, along the field-direction, of the angular momentum. 



Consider now the experimenter with molecular beams of hydrogen 

 molecules and hydrogen atoms at his disposal. In a magnetic field of field 

 strength H he finds the proton resonance of the former at frequency 

 Vp , and ascertains (m/Z) of the proton by putting his data into equation 

 (4): 



(m//)p = hvp/H (8) 



In the same field he finds the electron resonance of the latter at fre- 

 quency Ve , and ascertains (fji/l) of the electron similarly : 



(M//)e = hVe/H (9) 



Now he has both values; but the accuracy of both is contingent on the 

 accuracy of the measurement of H, and this is not so good as he desires. 

 However he can dispense with the measurement of i7 at a price — the 

 price of getting his value of the magnetic moment of the electron in 

 terms of units other than c.g.s. units. This is not a great sacrifice; Nature 

 does not share our affection for c.g.s. units; there are others which are 

 more suitable to the enterprises of the theorist. 



If we divide (8) into (9) we get rid of both H and h. This means that 

 if the experimenter measures vp and Ve in one and the same applied field, 

 he can evaluate {/jl/I) for the electron in terms of (fx/I) for the proton 

 without bothering about the values of H and h. Since I is the same for 

 both particles, he obtains the ratio of the magnetic moments of electron 

 and proton. The value of this ratio would be precious in itself, even if 

 one had not the faintest idea of the value of either moment in c.g.s. 

 units. It is 658.2288 d= 0.0006. 



It is also feasible to get the value of (m/Z) for the electron in terms of 

 the ''unit" eh/^irmc. This entity is so important that it has a name of 

 its own: it is called "the Bohr magneton." 



There is also a combination of experiments by which (/x//)e may be 

 evaluated in terms of the unit (eh/4:Trmc) . This unit is so important that 

 it has a name of its own: it is called "the Bohr magneton." The reader 

 can easily show for himself that (n/I) in terms of this unit is none other 

 than the quantity g, of which this is a second definition (not identical 

 with that of g in Part I). 



