348 0. Bar us — Inversion in the Wave Mechanism, etc. 



advance in one direction only if for 7we insert the correspond- 

 ing gradient b Y/bx in the direction of advance. 



Mathematically stated, if Y is simple harmonic and positive, 

 b 2 Y/bf will be negative and b* Y/bf positive again. If there- 

 fore clockwise curls are associated with positive fluxes, the 

 magnetic curl around Y conditioned by 6 2 Y/bf must be 

 counterclockwise, whereas the electric curl around 7 or <y' con- 

 ditioned by b 4 Y/bf must be clockwise. 



5. With the wave disposed of, the question nevertheless 

 remains as to what will happen if the original impulses Y y 

 Y' . . . are not simple harmonic but of a character to make 

 6 2 Y/bf positive. If then 6 4 Y/bf is also positive, b*y/bf is 

 positive and increasing and the above diagram is a correct rep- 

 resentation of the case. Energy would be transferred from 

 the original plane PP in both directions v, ?/, not however as 

 a wave but as simple flux in the direction of the energy paths 

 v, v', until the original stress Inbreaks down or follows some 

 other law. It seems to me that the case is not without subtlety 

 and that this nonharmonic rush of electric energy in a direc- 

 tion opposite to the usual energy path cannot be excluded. Its 

 duration would be short since Y, Y' . . . are to remain finite. 



Suppose now that such a rush and breakdown occurs irregu- 

 larly : the irregularity would be accentuated since the energy 

 paths in the rush and the breakdown would in a measure be in 

 opposite directions. Energy so transferred could neither be 

 easily reflected nor refracted and it certainly could not be 

 brought to interference or polarization. In proportion as the 

 rush and breakdown became more and more rythmic or wave- 

 train-like, reflection, etc., would eventually manifest themselves. 

 One is tempted therefore to associate these occurrences with 

 the recent advances in electro-optics, particularly since the 

 deflecting effect of a magnet on the transient electric whirls or 

 their resultants Y, and therefore on the energy path, is not out 

 of the question. During the rush, in other words, the direc- 

 tion of electric current in the whirls or the fluxes J 7 remains 

 unchanged for relatively long periods, whereas in an ordinary 

 wave these directions are regularly and rapidly reversed in the 

 lapse of time. 



Brown University, Providence, R. I. 



