Light on Electrical Tension in Metals. 335 



two metals, nothing but a thermoelectric current could bo 

 detected. This was analogous to the behaviour of illuminated 

 silver plates, whose electrical resistance increased under the 

 influence of the entire solar beam, while it diminished under 

 the influence of particular-coloured rays. 



The phenomena hci'e brought forward might possibly admit 

 of an explanation analogous to that for thermoelectric cur- 

 rents. If these latter owe their origin to the fact that the 

 dift'erence of electrical tension at the point of contact of two 

 metals is altered by a rise of temperature, then it is quite con- 

 ceivable that the difference of tension may also be altered by 

 the penetration of the luminous rays. In this hypothesis we 

 need not assume that the luminous action at the surface of 

 contact of the metal plates penetrates right through the upper 

 plate nor even influences the lower one ; the mere illumina- 

 tion of the upper one can modify the tensional diflerence be- 

 tween the two plates. 



Now, in all the experiments referred to in this paper, the 

 luminous radiation produced two simultaneous eftects. In 

 the first place, rays were absorbed, and consequently the tem- 

 perature of the absorbing body was raised ; and in this way 

 its electrical behaviour was indirectly affected. But, further, 

 there must have been another kind of action due to the lumi- 

 nous rays, because in these experiments, just as in many of 

 the experiments of other observers, the effects of light and 

 heat were opposed to each other. Now if, in the case of this 

 secondary action, the behaviour of the luminous waves is dif- 

 ferent from what it would be if they were merely absorbed, 

 then we may perhaps suppose that such rays as pass through 

 the illuminated metal, or at any rate penetrate it to a certain 

 depth, are the direct cause of the observed variations in elec- 

 trical conductivity and tension, which changes cannot arise 

 from heating. According to this view, it would seem that 

 the more transparent a metal is to a particular kind of light, 

 the greater would be the effect upon it of that light. Now, 

 inasmuch as in this kind of action the rays part with a por- 

 tion of their vis viva, there is here also absorption, but of a 

 different kind from Vv'hat is usually understood by the term — 

 in which it is generally assumed that the energy is at once 

 converted into heat. 



Now, if this attempt at an explanation of the electrical 

 effect of illumination be admitted, then the currents generated 

 by light Yf ould correspond in many respects to thermoelectric 

 ones. And this relation might be expressed by calling those 

 currents which are generated by light without any chemical 

 action, photoelectric currents. 



