480 Profs. Ayrton and Perry on Incandescent 



was as far as possible removed. The lamps subsequently em- 

 ployed were first one with a M-shaped filament, and another 

 with a simple horseshoe- shaped filament, each possessing 

 little self-induction. 



The light was measured by comparison with a standard 

 candle, the two being placed 130 centimetres apart on the 

 photometer P, and a screen composed of two pieces of paraffin- 

 wax W, with silvered paper between them, was adjusted 

 until the two pieces of wax appeared equally bright, the 

 comparison being first made when the screen was looked 

 at through ruby-red glass and then through signal-green 

 glass. 



The following is a sample of the results obtained, a and b 

 being the distances respectively of the screen from the incan- 

 descent lamp and from the standard candle. 



Table I 





Va 5 . 



Vv* 



Watts. 



Green. 



a. 



b. 



Candles. 



Watts per 

 Candle. 



4 Ferranti 



Gramme 



Ferranti 



Gramme 



1-34 

 1-34 

 1-34 

 1-34 



50 

 50 

 50 

 50 



67 

 67 

 67 

 67 



1077 

 105-8 

 107-9 

 108 



22-3 

 22*2 

 22-1 

 22 



23-33 



23-52 

 23-74 

 2411 



2-872 

 2-848 

 2-823 

 2-779 





VjP. 



Vy* 



Watts. 



Bed. 



a. 



b. 



Candles. 



Watts per 

 Candle. 



Ferranti 



Gramme 



Ferranti 



Gramme 



1-34 

 1-34 

 1-34 

 1-34 



50 

 50 

 50 

 50 



67 

 67 

 67 

 67 



106-5 

 106-4 

 106 

 1059 



23-5 

 23-6 

 24 

 24-1 



20-52 

 20-34 

 19-51 

 1932 



3-265 

 3-295 

 3-435 

 3-469 



It will be observed that in the set of four successive observa- 

 tions given in Table I. for green light, the candle-power of the 

 lamp appears to be steadily increasing, a result probably due to 

 the brightness of the candle slowly diminishing ; but although 

 this would make the absolute determination of the efficiency of 

 the incandescent lamp inexact, it introduces no error in "the 

 determination of the relative efficiencies of the lamp with 

 direct and alternating currents, since every observation with 



