I>U lU-.I.L SYSTEM TECHNICAL JOURNAL 



Now nuilliply both sides of cqualion (1) by Planck's constant /;; 

 it becomes 



livi = hvti„,-h(,. (2) 



On the lefl-luuul side we lia\e hvi, a qiiantiiy of the dimensions of 

 energ>'. Now there is much reason to believe that when radiant 

 cnerg>' streams out from a substance in the form of radiation of fre- 

 quency V, it emerges often if not alwa>s in parcels or packets or units 

 or quanta, each consisting of an amount of energy equal to hv. Sup- 

 pose thai the radiant energy constituting any line of a scries is emitted 

 in quanta such as these; then whenever an atom performs the act of 

 radiating that line, it loses the amount of energy which stands on the 

 left-hand side of Equation (2). The right-hand side represents 

 the same thing, and is itself the difference between two terms which 

 are spectrum-lerms multiplied by //; these are themselves the values 

 (reckoned from a suitable zero) of the energy of the atom before and 

 after the radiation occurs, they are the energy-values of the atom 

 in the state before radiating and in the state after radiating. The 

 spedritni-terms, when multiplied by Planck's constant h, are translated 

 into the energy-values of the Stationary States of the atom. When 

 expressed in proper units, terms are energies and energies are terms. 

 In the decades during which the spectroscopists were analyzing line- 

 spectra, disentangling line-series — by no means a light labor, for the 

 perspicuity of the series shown in the photographs of the First Part 

 is anything but common — and disengaging terms, they were unknow- 

 ingly recognizing and locating the Stationary States of the atom. 

 Spectrum analysis culminates in the fixation of the Stationar\- -States. 

 This is the greatest of the ideas for whicii liu' world is indc'l)ti'(i to 

 Bohr, and e\entually through him to Planck. 



These .Stationary States constitute one of the great systems of 

 facts, which the atom-model of Rutherford and Bohr is designed to 

 interpret. Let me formulate the demands which thus are made upon 

 this atom-model. It must ha\e features to account for these facts: 



First, that there are such things as Stationary States; 



Second, that in passing over in a "transition" from one stationar>- 

 slate to another of which the energy is less by AU. tlic atom releases 

 the energy AU in radiation of the one frequcncx' Af /;; 



Third, that certain transitions do not occur, or occur under ab- 

 normal circumstances only, or occur less frequently than others; and 



Fourth, that the stationary states of each particular kind of atom 

 have the particular numerical energy-values which they are observed 

 to have. 



