158 



Drs. R. Pohl and P. Pringsheim on the 



which is due exclusively to the alkalies, the assertions 1 and 

 4 are true, even in the visible part of the spectrum. We 

 call it the normal effect. All the well-known facts such as 

 the independence of the temperature, the curves of the 

 initial velocity and others refer to this effect, and so it may 

 be explained by the atomistic theory of light according to 

 Planck-Einstein* [18]. 



The results are very different for pure alkali metals or 

 Hg alloys of certain concentrations. Also in this case the 

 normal effect exists for E || and E -L as in fig. 1, but for 

 E || there is a second effect superposed on the normal one, 



Fia:. 2. 



A = 200/; a 300 



Wave-length. 



in a short range of the spectrum ; it grows rapidly from a 

 small value to a very great intensity, and decreases then 

 again with decreasing wave-length to zero, suggesting a 

 resonance phenomenon ; it is the more intense the stronger 

 the component of the electric vector normal to the surface 

 of the metal. The orientation of plane of polarization is of 

 great importance for this effect, which may be called the 

 selective one. Pig' 2 shows the superposition of the two 

 effects [18]. Fig. 3 and Table II. contain the results which 

 we obtained with polarized light on a sodium-potassium 



Cf. E. Ladenburg, Jahrb. d. Bad. u. El. vi. pp. 425-487 (1909). 



