326 



SCIENCE 



[N. S. Vol. XLV. No. 1162 



an absorption rather than an emission of 

 energy, since its egress diminishes rather 

 than increases the nuclear charge. Per- 

 haps this is why beta rays are always ac- 

 companied by gamma rays, while alpha 

 rays are not so accompanied. This is, how- 

 ever, a speculation which does not imme- 

 diately concern U5 here. The important 

 conclusion, for the purposes of our present 

 subject, is that Moseley's facts and unques- 

 tionable mechanics combined with our two 

 assumptions of circular orbits and radia- 

 tion frequencies proportional in different 

 atoms to corresponding orbital frequencies, 

 lead inevitably to the explosive emission of 

 energy in definite quantities accompanying 

 orbital readjustments. And there is noth- 

 ing particularly disturbing or radical about 

 this conclusion either, for we have no basis 

 for knowing anything about how an elec- 

 tron inside an atom emits its radiation. 

 The act of orbital readjustment would be 

 expected to send out ether waves. The 

 only difficulty lies in the conception of 

 stable, non-radiating orbits between which 

 the change occurs, and whether or not we 

 can see how such orbits can exist, the ex- 

 perimental evidence that they do so exist is 

 now very strong, and it is to further evi- 

 dence for their existence, since that is the 

 main point to be established if this theory 

 of atomic structures is to prevail, that I 

 now wish to direct your attention. 



I have already mentioned some facts of 

 magnetism and of light which support the 

 orbital point of view. But the strongest 

 evidence is found in the extraordinarj^ suc- 

 cess of the Bohr atom, which was devised 

 before any of these Moseley relationships, 

 which have forced us to the essential ele- 

 ments of the Bohr theory,^^ had been 

 brought to light. Bohr, however, was 

 guided solely by the known character of 

 the line spectra of hydrogen and helium, 



11 N. Bohr, Pliil. Mag., 26, 1 and- 476 and 8.57, 

 1913. Also 29, 332, 1915, also 30, 394, 1915. 



together with the rapidly growing convic- 

 tion, now dissented from, so far as I know, 

 by no prominent theoretical physicist, that 

 the act of emitting electromagnetic radia- 

 tion by an electronic constituent of an atom 

 must, under some circumstances, though 

 not necessarily under all, be an explosive 

 process. To show what is the character of 

 this evidence, let us consider first what are 

 the essential elements in the Bohr theory, 

 and, second, what have been the accom- 

 plishments of that theory. Bohr's experi- 

 mental starting point is the Balmer series 

 in hydrogen, the frequencies in which are 

 exactly given by 



i/ = ^ - = 





(4) 



n-^ having always, for the lines in the vis- 

 ible region, the value 2, and w, taking in 

 succession the values 3, 4, 5, etc. As pre- 

 viously noted, Paschen had already brought 

 to light a series in the infra-red in which n^ 

 was 3 and ■)i, took the successive values 4, 

 5, 6, etc. Lyman 's discovery, subsequent to 

 the birth of the Bohr atom, of an ultra-vio- 

 let series of hydrogen lines in which 

 «i = 1 and ^2 takes the values 2, 3, 4, etc., 

 is not to be regarded as a success of the 

 Bohr atom, but merely as a proof of the 

 power of the series relationships to predict 

 the location of new spectral lines. To ob- 

 tain an atomic model which will predict 

 these series relationships for the simplest 

 possible case of one single electron revolv- 

 ing around a positive nucleus, Bohr as- 

 sumed : 



A, a series of non-radiating orbits gov- 

 erned by equation (1) . This is the assump- 

 tion of circular orbits governed by the laws 

 which are known to hold inside as well as 

 outside the atom. 



B, radiation taking place only when an 

 electron jumps from one to another of these 

 orbits the amount radiated and its fre- 

 quency being determined by 7i = J.^ — A^, 



