Jantjary 24, 1913] 



SCIENCE 



131 



light, and the experimental evidence is now 

 fairly strong, especially in view of the re- 

 cent work of Hughes,^'^ that it is directly 

 proportional to v. This is what we should 

 expect from the fact that the energy of an 

 electron ejected by an X-ray is propor- 

 tional to the energy of the cathode particle 

 which produces the X-ray. For the ether 

 disturbance set up by stopping a cathode 

 particle corresponds exactly to the ether 

 disturbance set up by a half swing of a 

 vibrating electron. We may then compare 

 roughly the wave-length of one of the pris- 

 matically resolved components of white 

 light with the wave-length of a Rontgen 

 ray impulse by comparing the half-period 

 of the light with the time of stopping the 

 electi'on. This time can be shown to be 

 inversely proportional to the energy of the 

 electron, i. e., the frequency of the X-ray 

 produced by stopping an electron may be 

 taken as directly proportional to the energy 

 of the cathode-ray particle producing it. 

 If, then, an X-ray ejects an electron with 

 an energy proportional to the energy of 

 the original cathode raj', ultra-violet light 

 should eject an electron with an energy 

 proportional to its frequency. Notice that 

 this result is obtained without the aid of 

 Planck's equation, but rather immediately 

 from the fairly well demonstrated inter- 

 convertibility of X-rays and /3-rays and 

 the assumption that light rays are nothing 

 but soft X-rays. But not only is the ab- 

 sorption of energy by an electron from a 

 light wave proportional to v, its numerical 

 magnitude is approximated at least by mul- 

 tiplying the frequency of the light by 

 Planck's value of h. It is true there is 

 here no accurate agreement as yet ; for part 

 of the energy absorbed by the electron is 

 lost in getting out of the metal, and the 

 exact amount of this loss has not been meas- 

 ured with as much accuracy as we hope 

 "'L. c. 



soon to be able to attain. Nevertheless the 

 agreement is now sufficiently good (within 

 some 70 per cent.)^" to lend some support 

 to the notion that the amount of energy 

 actually absorbed from the light by the 

 escaping electron is hv. 



A still further test of the hypothesis can 

 be made by computing the frequency of 

 X-rays from the observed velocity of emis- 

 sion of corpuscles ejected by them; i. e., 

 from the potential difference between the 

 terminals of the X-ray bulb which pro- 

 duces them. Thus we have 



A. = ^ = eF, 



or if y^ 40,000 volts (this would corre- 

 spond to fairly hard X-rays such as Laue 

 used), 



^_Jif^_ 6.55 X10~"X 3X10" 



4.774 X 10-'" X 



40,000 

 300 



: 3 X 10-s. 



Laue^^ gets from his diffraction patterns 

 wave-lengths ranging from 1.27 to 4.83 X 

 10^°. Walter and Pohl's previous diffrac- 

 tion measurements^' also gave a value of the 

 order lO"''. This is certainly striking agree- 

 ment, and lends some support at least to 

 the attempts to extend Bragg 's assumption 

 of the inter-convertibility of X- and ;8-rays 

 to the inter-convertibility of light rays and 

 /?-rays, or more generally, to the assump- 

 tion that whenever an electron is emitted a 

 quantity of radiant energy hv is absorbed. 

 Furthermore, according to the most re- 

 cent experimental results it doesn't seem 

 to make any difference in what form this 

 energy approaches the atom which is to 

 take it on and reemit it. Thus Wliidding- 

 ton^" and Beatty*" show that characteristic 

 X-rays are excited either by other char- 

 acteristic X-rays which are harder than 



^° Hughes, I. c. 



"'Munch. Ber., 363, 1912. 



=» Sommerfeld, Ann. der Phys., 38, p. 473, 1912. 



^'L. c. 



^"Proc. Boy. Soe., 87, p. 516, 1912. 



