the Photoelectric Effect. 581 



Thus when T has its maximum value, ^7^ has a finite value 



which is negative, since F 2 '(0) is negative. The value of 



dN 



™- when T = may be obtained by solving the equation 



F 2 (r) = P/T for r and substituting in (5). This value of ^ 



is readily seen to be positive and finite, so that it is clear that 

 the type of theory which supposes the maximum energy of 

 the emitted electrons to be determined by the energy they 

 receive under the influence of the light less a constant amount 

 of work necessary for them to escape from the material is in 

 satisfactory general agreement with the experimental results. 

 We shall not consider the numerical analysis of curves 

 like those in fig. 3 in the present paper, but shall now turn 

 to the experimental data which have been given by the other 

 metals investigated. 



Experiments with Different Metals. 



Curves similar to those in fig. 2 were obtained by the same 

 method for strips S of copper, bismuth, tin, zinc, aluminium, 

 and magnesium, as well as platinum, using various wave- 

 lengths of light. The curves for aluminium are represented 

 by the full lines in fig. 2 (PL XIII.). The results with the 

 other metals are shown in figs. 4, 5, 6, 7, and 8. 



In each of these cases the curves for different wave-lengths 

 reach saturation at a common point. The dotted ordinate 

 in each figure represents the actual position of the current 

 axis according to the reading of the voltmeter V, and the 

 point where the dotted line meets the volt-axis is the experi- 

 mental position of zero volts. But when this position of 

 zero volts is shifted so as to correct for the contact difference 

 of potential between the silver bulb and the metal strip used, 

 the current-axis is shifted so as to pass exactly through the 

 saturation-point where the curves coincide. For instance, 

 the contact-difference of potential between silver and zinc 

 is just 1 volt. When this is corrected for by shifting the 

 position of zero volts to the right a distance corresponding 

 to 1 volt, it is seen that this point, which really corresponds 

 to zero field acting on the electrons, exactly coincides with 

 the position where the curves reach their common maximum. 

 This correspondence between the contact-difference of poten- 

 tial and the shift necessary to place the ordinate of zero volts 

 at the saturation-point is exact, so far as we have been able 

 to determine, in every case except that of magnesium, fig. 4. 

 But doubtless even here the discrepancy is only apparent. 



Phil. Mag. S. 6. Vol. 24. No. 142. Oct. 1912. " 2 Q 



