On the Spectrum of Water-vapour. 355 



pressure. The modifications which it requires when these 

 conditions are not fulfilled, as well as the simple and natural 

 assumptions upon which the proof of the theorem rests, can 

 only be given in a fuller communication. 



From the above proposition it follows in particular that for 

 [«'] = [«], i> e. if the volume of the element a in the new 

 compound C is the same as in the original substance A, the 

 wave-lengths V of the rays emitted by a! must be equal to the 

 wave-lengths X previously produced by a. The only differ- 

 ence will consist simply in a varying change in the ampli- 

 tudes and consequently in the intensities of the individual 

 rays ; from which it results that the brightness of a whole 

 series of rays is so much diminished that they cease to be 

 visible, whereas others become brighter. Thus the spectra of 

 the compounds formed by the union of hydrogen with chlorine, 

 bromine, and iodine (viz. HC1, HBr, HI) consist only of the 

 spectra of hydrogen on the one hand, and of chlorine, bromine, 

 or iodine on the other, but with characteristic changes of in- 

 tensity of the various rays of their components. 



The limitation to the gaseous condition is necessary because, 

 as in chemistry generally so here, easily observable relation- 

 ships exist only for gases. If the pressure increases, or if the 

 gaseous substances in consequence of falling temperature (or 

 w T ith simultaneous increase of pressure) approach the tempe- 

 rature of liquefaction, the number of impacts between two 

 different molecules will be increased under otherwise similar 

 conditions ; and the number of vibrations of the equivalent 

 smallest particles of the different similarly constituted mole- 

 cules of the same substance are altered in varying fashion by 

 varying impacts. Whilst in the former case equivalent cor- 

 responding smallest particles of different similarly constituted 

 molecules of the same substance emit extraordinarily feeble 

 rays of equal wave-length, which in the spectrum are super- 

 posed, and by addition of their extremely small intensities 

 produce a single ray of perceptible brightness, in the latter 

 case equivalent particles of different like-constituted molecules 

 emit not only rays of normal period, but also rays of periods 

 more or less different from the normal, which arise from the 

 impacts of different kinds. Together with the normal line, 

 we have close to it rays of the latter kind — that is, the original 

 sharp line becomes widened. If the pressure increases and 

 the gases approach their point of liquefaction, such a line 

 becomes still further widened ; and it may happen that the 

 resulting band-like line may extend as far as lines due to 

 heterogeneous particles of the different molecules of the same 

 substance. The lines of the spectrum run together, and there 



