308 BELL SYSTEM TECHNICAL JOURNAL 



for the frequencies forming tiie spectra of other elements." There is 

 the spectrum of neutral helium, for example, and the spectrum of 

 sodium, and the spectrum of mercury; in each of these there are 

 series of lines, of which the frequencies are clearly best expressed 

 each as the difference between a pair of terms, and these terms should 

 be the energies of the atom before and after radiating. But we have 

 not the shadow of an idea what the corresponding configurations of 

 the atom are; it may be that the outermost electron has certain 

 permissible orbits, but we do not know what these orbits are like nor 

 what common feature they possess. 



Is it then justifiable to write down a Rule such as this: the frequencies 

 of the rays which free atoms emit are such as to confirm the idea that 

 radiant energy of the frequency v is emitted in packets or corpuscles of 

 the amount hv? V'ery few men of science, I imagine, would hesitate 

 to approve this. However one may fluctuate in his feelings about 

 Bohr's model of the atom, there always remains that peculiar relation 

 among the frequencies emitted by the hydrogen atom, which is so 

 nearly copied by analogous relations in the spectra of other elements. 

 When one has once looked at the general formula 



/ hR\ ( hR\ ,_, 



and has once inter|ircted the first term on the right as the energy 

 of an atom before radiating, the second term on the right as the 

 energy of the atom after radiating, and the quantity hv as the amount 

 of the packet of energy radiated, it is very difiicult to admit that this 

 way of thinking will e\er be superseded; particularly when one re- 

 members the auxiliary facts, such as thai fact about the electrons 

 transferring just 4.9 equivalent volts to tiie mercury atoms which 

 they strike, no more and no less. Analyzing the mercury spectrum 

 in the same way as the h>drogen spectrum was anahzed, we find 

 the frequencies expressible as differences between terms; interpreting 

 the terms as energy-values, we find that between the normal state 

 of the mercury atom and the next adjacent state, there is a difference 

 in energy of 4.9 equivalent volts, and between this and the next 

 adjacent state there is a further difference of 1.8 volts. This then 

 is the reason why an electron with less than 4.9 equivalent \()lts of 



" The mathematical experts who have laboured over the theory of the Iieliiiiu 

 atom (two electrons and a nucleus of charge +.^<') seem to have convinced them- 

 selves that the features which distinguish the permitted orbits of the electrons in 

 this atom, whatever they may be, are definitely not the same features as distinguish 

 the permitted orbits of the electron in the hydrogen atom. This cannot be said 

 with certainty for any other atom. 



