the Photoelectric Effect. 593 



since the term involving 6. which depends on the thermo- 

 electric power of the metals, may be disregarded in the cases 

 considered on account o£ its smallness. For similar reasons 

 the differences of iv may be taken to be the same thing as the 

 differences of w Q . 



The values in column II. were obtained by assuming that 

 the value of w for platinum was equal to that given by the 

 thermionic measurements, viz. 5*34, in equivalent volts. A 

 factor A was then determined so that Av = Ax5'34 where 

 7i = 6-55 X 10" 27 erg sec. and v =l*03 x 10 15 sec." 1 the value 

 of v for platinum given by the experiments. This value of 

 A was then used as a reduction factor to determine the value 

 of w from that of hv for each of the other metals. This 

 method is equivalent to assuming that the value of w is the 

 same for the photoelectric as for the thermionic emission, 

 and that the large value of h given by (20) is due to some 

 unknown error in hv which affects the different metals 

 proportionately. 



The values in column IV. are simply the differences of hv 

 reduced to volts per unit charge. This method of calculation 

 is equivalent to assuming either that w Q is in error or that 

 it is not the same thing for photoelectric as for thermionic 

 emission. All things considered, both II. and IV. give a 

 fair agreement with III., although the last column agrees 

 better than the second. On the whole, the view which puts 

 the error in (20) on w has, so far, the best of the argument. 



The Frequency of Characteristic Rontgen Rays. 



If the formula T m = vh— w continues to apply no matter 

 how high the frequency v becomes, the frequency of charac- 

 teristic Rontgen rays can be written down from known data. 

 There is nothing in any derivation of the formula which 

 would tend to confine its application to any limited range of 

 frequency. On the other hand, its validity is closely con- 

 nected with that of Planck's radiation formula, which has 

 only been tested in the neighbourhood of the visible spectrum. 

 The extension to the case of Rontgen rays can therefore only 

 be regarded as speculative ; as a speculation however, it has 

 some merit, since it is the only line of attack which seems 

 open to us at present. 



The researches of Sadler * and Beattie f have shown that 

 when characteristic Rontgen rays fall on different substances, 

 the energy of the emitted electrons is independent of the 

 substance emitting them, being determined entirely by the 



* Phil. Mag. vol. xix. p. 337 (1910). 

 t Phil. Mag. vol. xx. p7 320 (1910). 



