1917] on Recent Developments of Molecular Physics 



elements, 

 " sound ' 

 energy 



We now see that in beat motion at low temperatures the 

 " is one in which all harmonics do not sound with equal 

 the higher harmonics get nothing, like the share of 

 energy allotted to them by the theorem of equiparfcition, and the 

 energy tends to concentrate in the vibrations of lowest frequency. 



The formula just given has the very special significance that it 

 also expresses the partition of energy amongst the different vibrations 

 in the s^^ectrum of a normal black body, as given by the well-known 

 and now generally-accepted formula of Planck. Indeed, it is 



I'D 



r/0 = -l -2 -3 -4 -5 -6 -7 -8 -9 VO M 1-2 >-3 14 



Fig. 3. 

 Atomic Heats at Low Temperatures (+ Aluminium, o Copper, x Silver). 



probable that for dynamical reasons the partition of energy in the 

 black body spectrum must form an indicator of the partition of 

 energy in the black body itself, out of which the spectrum originates. 

 Planck has proved that this is necessarily the case ; but his argument 

 is weakened, and to some extent invalidated, by the circumstance 

 that the proof is based on the Newtonian laws of motion which we 

 are now discarding. 



In general, however, we may say that in every case examined the 

 partition of energy in vibratory motions of all kinds must be supposed 

 to be that given bv the above formula, When x is small^ 



a2 



