276 wien: recent theories of heat and radiation 



fraction of the total energy, how is it possible for the electric con- 

 ductivity to depend in anj^ considerable degree upon temperature? 

 The assumption that the energy can be distributed only in 

 multiples of hv can, it seems to me, be combined with Planck's 

 new theor}^ only on the hypothesis that the absorption of energy 

 is continuous, but that the absorbed energy exists at first only in 

 the form of energy of electrons. The vibrations of the atoms are 

 identical with the heat energy, and disappear completely at abso- 

 lute zero, but the energy of the electrons remains arid amounts 



hv 

 in the mean to ^. On this assumption, the theory of specific 



heat remains unchanged. 



There is one further difficulty with vibrations of the infra-red 



rays which are assumed in the theory of dispersion to excite elastic 



vibrations of the molecules. These vibrations would also be 



hv 

 heat vibrations, but there would remain an amount ^ not depend- 



ing on temperature. Perhaps in this case also the motion of the 



h 



electrons is primary. Then ^ would again represent the energy 



of the electrons. 



The assumption that the electrons have a motion independent 

 of temperature seems not to be a new hypothesis. It is founded 

 on the theory of quanta, because the emission can only take place 

 if the energy reaches hv. Now we know from Zeemann's phe- 

 nomenon that the radiating particles are moving electrons, and 

 therefore that the electrons must move before the radiation can 

 begin. The theory of diamagnetism also requires the hypothesis 

 of moving electrons independent of temperature. Debye's theory 

 of specific heat is founded directlj^ on the formula of partition of 

 energy. It requires only the assumption that the heat energy is 

 identical with the vibrations of the atoms, and that the forces 

 are those which are assumed in the common theory of elasticity. 



A serious difficulty arises in connection with the heat conduc- 

 tivity which should be determined by the elastic waves in the 

 solid. The theory of elasticity is used for the determination of 

 the number of free vibrations, and is thus connected with the 



