654 BELI. SYSTEM TECUNICAI. JOVRXAL 



through the train of equations; and indeed this is the manner in 

 whieh the Bohr atom-mode! is usuali\' i^resented, so as lo arri\e 

 finally at the agreement between "theor\" and experiment which is 

 expressed in equation (16), and is a most striking chmax to the whole 

 exposition. By working tiirough the train of equations in the inverse 

 sense, I have considerably mitigated the effect of the climax; and this 

 procedure seems hardly fair to the author of the theory, but it is 

 not without its merits, for it enables us to see the exact role of equation 

 (15) more clearly than the commoner procedure. 



The situation now is this. It is possible to construct, out of a 

 nucleus and an electron, an atom-model possessing stationary states 

 of the energy-values displayed by the hydrogen atom, provided that 

 we assume that the electron may revolve only in circular orbits for 

 which the angular momentum of the atom is an integer multiple of 

 h/2T. There is no known reason why an electron should do a thing 

 like this, there is good reason to suppose that it cannot do anything 

 of the sort, for if it started out to revolve in a circular orbit it would 

 radiate its energy and descend spirally into the nucleus. If ne\er- 

 iheless we assert that the electron does just this sort of thing, we ha\e 

 nothing with which to support the assertion, nothing extrinsic by 

 which to render it plausible; it must stand oti its own merits as an 

 independent principle. 



These merits, had we no data other than the energA-\alucs of 

 stationary states catalogued in equation (G), would probably be 

 regarded as scanty. After all, the agreement between the constant 

 pi and the quantity h/2Tz might be fortuitous. But there are other 

 stationary states of the hydrogen atom, beyond those listed in (G). 

 For instance there are the stationary states which are evoked by a 

 strong electric field acting upon hydrogen, and there are the stationary' 

 states which are called into being by a magnetic field applied to 

 hydrogen, as I related in earlier sections of this article. There is 

 also the fact, that at least one of what 1 ha\'e been calling the stationary 

 slates of hydrogen is not a single stationary state at all; there are 

 two states of which the energy-\alues lie exceedingly close together 

 and to the value — Rli 4, so close that nearly all experiments fail to 

 discriminate them. .And tlure is (lie great multitude of stationary 

 states exhibited by other elements than h\(lrogen; l>in we will not 

 think about these for the time being. 



Now the situation is transformed into this. Considci- all these 

 additional stationary states, exhibited b\ tiie hydrogen aKini under 

 unusual or e\en under usual circumstances. Is it possible to trace, 

 for each one of them, an orbit for the electron, such that while the 



