Januaet 4, 1907] 



SCIENCE 



9 



Professor Runge, that 'nature is getting 

 more and more disorderly every day.' 



4. Passing now to the pressure shift, we 

 owe to Dr. Humphreys, one of the discov- 

 erers of the phenomenon, a clear explana- 

 tion in terms of the Saturnian atom. 



He points out first of all that an atom 

 built on the model indicated, with quanti- 

 tative specifications such as these given, by 

 experiment, will exert an enormously pow- 

 erful field at points near its center. And 

 since the convection currents here existing 

 are practically equivalent to Amperian cur- 

 rents in circuits devoid of resistance, it is 

 clear that any currents induced in these 

 atoms by attempting to thrust through 

 them more or fewer lines of force than they 

 now contain will be permanent and, there- 

 fore, unlike the induced currents in the 

 wire circuits of our laboratories. 



But an increase of current in such an 

 atom means an increase of speed in the 

 corpuscles, and this in turn means pre- 

 sumably an increase of frequency. From 

 this it follows that, if by any means two 

 atoms are brought closer together there 

 will in general be a change of frequency, 

 and one may expect this change to be some- 

 times an increase and sometimes a decrease ; 

 that is to say, two atoms which are made 

 to approach by pressure, and which we may 

 imagine as strung on a common axis, will 

 sometimes be rotating in the same sense 

 and sometimes in the opposite sense. The 

 increase and decrease of frequency thus se- 

 cured by pressure will have the effect of 

 widening the line. The plausibility of this 

 argument is muoh enhanced by the enor- 

 mous strength of the magnetic field in the 

 neighborhood of one of the magnetic atoms ; 

 in the case of the iron atom amounting to 

 as much as 150,000 C.G.S. units at a dis- 

 tance of ten radii from the center and, of 

 course, a thousand times greater than this 



at the center. So much for the widening 

 due to pressure. 



But Dr. Humphreys has also very clever- 

 ly suggested that this widening will not be 

 symmetrical about the original position of 

 the line, but about a new position on the 

 red side of the old. For when two atoms 

 happen to be rotating in the same direction 

 they will attract each other and then get 

 'into the stronger portion of each other's 

 magnetic field. '^^ In other words, those 

 effects which result in lengthening the 

 waves will be much more marked than 

 those which shorten the waves. Hence in- 

 crease of pressure will be accompanied by 

 shift towards the red. It would be exceed- 

 ingly interesting to know what difference 

 of structure exists between the radiant 

 sources of lines and of bands which causes 

 this explanation to break down when ap- 

 plied to the latter. 



5. Passing now to the effect of a mag- 

 netic field, an experimental fact which 

 largely established, although it did not sug- 

 gest, Lorentz's conception of the electron 

 vibrating about an attracting center, it is 

 at once evident that the Zeeman phenom- 

 enon must follow as a deduction from the 

 Saturnian atom. 



But Preston and Eunge and Paschen 

 have shown that the normal triplet is by 

 no means an ordinary occurrence; on the 

 other hand the breaking up is very much 

 more complicated, a single line yielding 

 anywhere from 3 to 14 components.^" 



Beautiful as the general agreement be- 

 ween fact and theory here is, one finds it 

 peculiarly difiicult to understand how the 

 central line of the normal triplet — the one 

 due to the component of motion along the 

 lines of force— can be split up at all by 

 the magnetic field. But since, as a matter 

 of fact, this component is split up, it is 



'= Humphreys, Astrop. Jour., 23, 243, 1906. 

 "Riinge & Paschen, Aih. Ber. Akad., February 

 6, 1902. 



