Manchester Memoirs, Vol. Hi. (1908), iV<?. 10. 21 



stimulation, — for instance, the fact that excitation by a 

 homogeneous vibration of period close to one of the 

 free periods of the vapour excites the emission of the 

 line of that period, accompanied by a series of lines, 

 equidistant in frequency, ranged on both sides of it. The 

 similarity to the circumstances of G. W. Hill's (and Adams') 

 dynamical analysis of the lunar perturbations, first trans- 

 ferred to problems in other branches of physics by Lord 

 Rayleigh, has been remarked more than once.* On the 

 general view that a molecule is a structure such as a vortex 

 ring or an orbital system of electrons, it has intrinsic cyclic 

 motions of its own, which must, however, be so balanced 

 as to avoid the unlimited draining away of its constitutive 

 energy by radiation ; we should then anticipate that each 

 of these structural cyclic periods would interact with the 

 periodic disturbance exciting a natural vibration, and give 

 rise to the analogues of summation and difference tones, 

 if we may use the acoustic terminology. On this view 

 the common difference of the vibration frequencies, in one 

 of Wood's series, would be the frequency of an intrinsic 

 cyclic motion of the kinetic system which constitutes the 

 molecule. Here again Wood has found that many of the 

 lines which exhibit these phenomena are specially 

 sensitive to magnetic influence. 



Recent investigation seems still to confirm that in a 

 given spectrum the lines that shine out brightly are 

 determined by the temperature, without concomitant 

 chemical change : lines which appear conspicuously at 

 one range of temperatures do not show sensibly at 

 another, where others of the same system show up 

 instead. We are thus required to imagine a way in which 

 the mode of electric or other excitation of the same 

 molecule may be different at different temperatures, just 



*E.g., A. Stephenson, Phil. 3/a^., July, 1907, p. 115. 



