436 Sir J. J. Thomson on the Origin of 



Now -=l-78xl0 7 , 



m 



e = l-6 xlO" 20 , 

 hence k — 5 x 10 ~ 22 « 2 , 



and the time taken for the energy to fall to 1/e of its initial 

 value is 10 22 /5n 2 . 



Thus, if the vibrations have the frequency 5'lxlO 14 of 

 the D line in the sodium spectrum, this time is equal to 

 7*7 x 10~ 9 seconds, and the length of the train emitted in this 

 time would be about 4 million wave-lengths. If the wave- 

 lengths were 10 ~ 8 cm., about that usually given for the L 

 characteristic "Rb'ntgen radiation from platinum, the time 



would be -7-z — tt^tx? and the train emitted would contain 

 4'o x 10 15 



about 660 wave-lengths. Thus, if the spectral lines arise 

 from the effect of magnetic forces, they may be expected to 

 consist of trains containing a very large number of wave- 

 lengths, the number of wave-lengths being inversely propor- 

 tional to the frequency of the vibrations. 



Transference of Radiant Energy into Potential or 

 Kinetic Energy. 



Let us consider a stream of radiant energy passing through 

 a piece of metal. In consequence of the magnetic fields due 

 to the atoms, there will be in the metal magnetic forces 

 reaching very high values in certain places, and falling to 

 quite low ones in others ; and between very wide limits of 

 the magnetic force it will be possible to find places where the 

 magnetic force has any assigned value, if at all such places 

 an electron could be found, a piece of metal would be a 

 system containing vibrators of every possible period within 

 very wide limits. 



In a metal it seems very probable that the electrons are 

 not all contained within the atoms themselves, but that some 

 of them are detached, helping by the forces they exert to 

 make the atoms cohere and form a solid body. Unless, then, 

 the position of these detached electrons were exactly co- 

 ordinated with the intensity of the magnetic field, the effect 

 would be much the same as if these electrons were dis- 

 tributed at random through the magnetic field ; and we 

 might expect to find some electrons in places where the 

 magnetic force had, within wide limits, any specified value^ 

 A piece of metal would then be a system of the kind we are 



