274 wien: recent theories of heat and radiation 



statistical considerations, and has been successful, but the assump- 

 tion that the emission of radiation can only take place by quanta 

 has not yet been necessary in any direct physical experiment. 

 The success of the statistical treatment of the phenomena of heat, 

 which has led to the theory of quanta, must- be due to some peculi- 

 arity of the atom which has found its simplest expression in 

 Planck's hypothesis of quanta, but it remains to be proved 

 whether this assumption represents the true theory of the actual 

 phenomena. 



We can not say that the atom radiates energy only thru quanta, 

 for if we accelerate an atom of canal radiation, it must send out 

 energy according to the laws of the electromagnetic field. This 

 energy can be calculated and may reach any arbitrary value. 

 Nor is it possible to overcome this difficulty by assuming that 

 the theory of quanta holds only for periodic variations, and an 

 acceleration is not periodic — ^for in an alternating field we can 

 impress such vibrations upon a charged atom that it will emit a 

 radiation which is periodic but has nothing to do with the quan- 

 tum. Again, if we regard heat as the elastic vibration of the 

 atoms, it leads to the conclusion that the theory of quanta must 

 apply there. On the other hand, acoustical vibrations of the 

 same character have nothing to do with the quantum. To apply 

 the theory of quanta to every possible kind of vibration seems, 

 therefore, to lead to impossible consequences. 



Once it seemed necessary to limit the application of the 

 theory to the phenomena of irregular molecular motion, and, 

 indeed, we may call Planck's formula the general expression of this 

 irregular motion. From this point of view the theory of quanta 

 tells us that such irregular molecular motion diminishes with 

 temperature and disappears at absolute zero — ^which is synony- 

 mous with the assertion that entropy vanishes at zero as required 

 by the Nernst theorem. 



Since the theory of quanta gives us only a formula for the statis- 

 tical treatment of the partition of energy, it is extremely difficult 

 to find its true physical meaning. We might, of course, content 

 ourselves with applying this statistical treatment to observed 

 phenomena only, merely exchanging the theorem of equipartition 



