Theory of Radiation. 265 



molecular encounters will show their influence* We have 

 assumed that yellow vibrations are more easily produced than 

 green vibrations. While, therefore, the green vibrations are 

 allowed to decrease in intensity, the intensity of the yellow 

 rays is constantly renewed by the encounters. The radiation 

 of a body is a phenomenon of dissipation of energy; and the 

 relative intensity of the different radiations merely show 

 through what channels the energy is more quickly dissipated. 

 The rate at which the energy of translation is transformed 

 into energy of periodic motion (which has, according to 

 Boltzmann, nothing to do with the establishment of final equi- 

 librium in the enclosure of uniform temperature) is the most 

 important quantity in determining the intensity of vibration 

 when a body is losing heat. The mere fact, therefore, that 

 we observe lines of different intensities in the spectra of bodies 

 is not in itself sufficient to invalidate Boltzmann's theoretical 

 conclusion that in an enclosure of uniform temperature the 

 vibrations ought to be equally strong. 



It will be seen from what precedes that I have adopted a 

 view on the origin of the different intensities of spectral lines 

 which is different from that usually given ; but it is, as far 

 as I can see, the only one consistent with dynamical prin- 

 ciples. The statement that the spectrum of a body is not 

 independent of the vibrations of the surrounding bodies may 

 sound improbable; but it will, I think, be found to follow from 

 any dynamical explanation of radiation. But though the 

 mere fact of different intensities of spectral lines need not be 

 conclusive against Boltzmann's conclusions, another argument 

 is much more so. If his reasoning is correct we can, from 

 the measurement of the two specific heats, deduce the number 

 of degrees of freedom of a molecule. This number is very 

 small, while we know, by spectroscopic observation, that it 

 must, in fact, be infinite. Those who have observed the com- 

 plicated spectrum of mercury, for instance, will find it hard 

 to believe that a molecule of mercury has only three degrees 

 of freedom, that is, that it can only move about in space ; 

 that it does not vibrate nor rotate. No suggestions which 

 have yet been made seem to me to get over this serious dif- 

 ficulty; and as the reasoning by means of which the result 

 in question has been obtained does not seem open to serious 

 criticism, some of the fundamental assumptions on which the 

 reasoning is based require to be carefully scrutinized. Now 

 it seems to me that in the course of the whole investigation 

 the vibrations of the molecules themselves, as far as their 

 mutual influence (which must take place in consequence of 

 the transmission of these vibrations through the medium) is 



Phil Mag. S. 5. Vol. 12. No. 75. Oct. 1881. X 



