^y Professor Pell. 23 



we liave been considering, would give one bright line correspond- 

 ing to its fundamental note ; and fainter lines would correspond 

 to some of the terms of the second and higher approximations. 



It can hardly, I think, have escaped notice, that if the mean 

 translation velocity of the molecules of an incandescent vapour 

 become so great as to bear a sensible ratio to that of wave pro- 

 pagation, the 'wave length of the light corresponding to any 

 bright line would be aifected in a manner and degree depending 

 upon the direction nf motion relative to that of the light observed. 

 As the temperature is gradually increased, this would have the 

 effect of thickening the bright lines, and finally of converting 

 them into a continuous spectrum. If as the temperature is in- 

 creased, a rupture or change of constitution should take place in 

 the molecules, we might expect a sudden change in the appear- 

 ance of the spectrum. In reference to this subject, I may re- 

 mark, although I express an opinion with much hesitation, know- 

 ing how much there is which has been written upon this subject, 

 which I have not had an opportunity of studying, that I believe 

 that the constitution of the luminiferous ether is such as to render 

 it incapable of propagating waves of less than a certain length. 



I see some hope also of an' explanation of what has always 

 appeared to me one of the greatest difficulties in connection with 

 molecular physics ; that the wave length should be so nearly the 

 same for all kinds of heat. It is not difficult to conceive that the 

 molecules in the sun and elsewhere, whose vibrations are the 

 chief sources of heat, should have been so constituted as to 

 vibrate nearly in the same time. The difficulty is to understand 

 why the molecules of bodies of all kinds and constitutions, bein^ 

 heated and then left to vibrate in their own way, should all 

 vibrate so nearly in unison. But if we hold that the arrangement 

 of atoms into molecules is caused by the prevalent heat and 

 depends upon its wave length, the difficulty disappears. Let us 

 suppose for a moment that the sun should radiate heat of one 

 uniform wave length only, and that the values of on for all sub- 

 stances and combinations were such that — — - were in every case 



24/ *' 



a whole number exactly. All the atoms under the influence of 

 the sun's heat would be arranged into molecules, all having the 

 same fundamental note, and collections of such molecules after 

 being heated, would give back that note alone. ISTo substance 

 having its atoms otherwise arranged could continue to exist, for 

 every ray of heat which it encountered would assist in decompo- 

 sing and rearranging its atoms according to the prevailing code. 

 There would be one uniform stability of molecular constitution, 

 and one uniform colour, fx. appears to be as nearly constant as 



