Theory of the Flicker Photometer. 711 



the straight line relation, and the apparently irregular flicker 

 photometer values was left unexplained. It is this connexion 

 which is now developed. 



2. General Outline of Theory. 



The line of thought here pursued is best introduced by 

 the analogy of a common physical case taken from the domain 

 of heat. Consider two incandescent lamp filaments, one of 

 fine diameter, the other coarse, both operated on the same 

 fluctuating E.M.F. Let the frequency of the voltage supply 

 be capable of alteration at will. Then as it is slowly raised 

 from a low value it will be noted that the thick filament 

 ceases to flicker at a much lower frequency than the thin one, 

 in consequence of course of its larger heat capacity. Suppose 

 now that the two lamps are connected to two separate voltage 

 supplies which are exactly opposite in phase. If the two 

 lamps were exactly alike, then a surface equally distant from 



Fin-. 3. 



(Upper diagram.) Two alternating dove-tailed stimuli A and B . 

 At a colour of the photometric field is A n and at /3 it is B . 



(Lower diagram.) The impression as recorded by a transmitting 

 medium. The amplitudes of fluctuation are reduced. At x the colour 

 is a mixture of A and B with A predominating; at /3 a mixture with 

 B predominating. 



the two would be continuously uniformly illuminated. Fig. .'S 

 shows, in the upper diagram, the manner in which the power 

 is supplied to the two lamp filaments, and in the lower 



