wien: recent theories of heat and radiation 281 



electron is ^ i'^ = hv. The energy of the secondary electrons 



being known one can calculate the value of v and obtain a wave 

 length for the Rontgen rays of the same order of magnitude as 

 that which results from experiments in diffraction. It also 

 seems probable that the Rontgen rays affect only the electrons 

 and that the whole absorption is therefore caused by collisions of 

 the secondary electrons with the atoms. In the production of 

 Rontgen rays almost the whole energy of the -cathode rays is 

 transformed into heat. Sommerfeld has given a formula unit- 

 ing the theory of Rontgen rays with the theory of quanta. He 

 puts the action integral (principle of least action) equal to the 



constant ^ , the integration being taken over the time of the 



molecular action. It is then possible to calculate from the values 

 of the kinetic and potential energies and the constant h, the time 

 of molecular action which corresponds to the time of radiation. 

 According to this reasoning the value of the wave lengths resulting 

 from the electromagnetic theory should agree with the value 

 derived from the theory of quanta. In fact much smaller values 

 for the wave lengths result from the electromagnetic theory. Som- 

 merfeld explains this by assuming that the Rontgen rays are not 

 monochromatic but consist of two kinds of radiation, one depend- 

 ing on the nature of the anticathode while the other is the radia- 

 tion of the retarded electrons. The latter must be polarized in 

 a plane such that the electric vibrations are parallel to the direc- 

 tion of retardation. One can then calculate the radiation of the 

 retarded electrons from the polarized portion of the Rontgen 

 rays. Using the observations upon the polarization of Rontgen 

 rays one now finds an agreement. 



Sommerfeld applies his theorem also to the electrons expelled 

 by ultraviolet light by supposing that the energy acquired by the 

 electrons is accumulated by resonance, but with this supposition 

 I cannot agree. In this case a very long time must elapse before 

 emission begins. For the explanation of the emission of electrons 

 by light it seems to me rather that M^e must take account of the 

 energy concealed in the atom. 





