56 On Experiments with the Brush Discharge. 



of shorter and some of longer wave-length than these. The 

 reduction of the silver salts in the sensitive film may be 

 caused by more than one kind of these waves. Before then 

 it is possible to say that the effects are caused by the actinic 

 power of the waves of short wave-length only, it is necessary 

 to separate these waves in some such manner as that employed 

 by Tyndall to separate heat-waves from those of light. 

 Various experiments have been made to do this, but hitherto 

 without success. The investigation is being pursued in this 

 direction. 



The actual amount of light given out by the brush- discharge 

 is not much, and in the reflexion experiments which have 

 been described it was found impossible for the most delicate 

 eye, even after being kept in absolute darkness for a con- 

 siderable time, to distinguish the outline of the object; but the 

 shadow was nevertheless easily produced on the photographic 

 plate. The numerous experiments which have been made 

 show that, so far as one can judge, the effects produced by 

 the brush are far more definite than would be expected, when 

 the very small luminosity of the discharge is remembered. 

 Or, in other words, that the emanation from the point contains 

 a much larger proportion of rays capable of bringing about 

 chemical decomposition than would be supposed when we 

 remember only its luminosity. 



In order that some approximation might be obtained between 

 the light-giving power of the discharge and ordinary light, the 

 following experiment was made. A comparison was first ob- 

 tained between the light from a standard candle and the smallest 

 burner procurable, the gas being burnt under the usual con- 

 ditions for regulating the pressure. This burner was now 

 compared with the light from the brush. The ordinary Bunsen- 

 photometer, as used in gas-testing, was employed, and the 

 light from the machine carefully screened off. In order that 

 a good brush might be obtained an u earth " was placed near 

 the terminal. The actual figures obtained varied with the 

 climatic conditions, but not so much as would be expected, 

 and the following numbers may be taken as an approximation, 

 but, of course, only an approximation, to the relative lumi- 

 nosity of a standard candle and the brush from the machine 

 used. 



Distance of candle from photometer ... 60 inches. 



Distance of light „ „ ... 3' 2 5 inches. 



Therefore, Candle : Light : : 3600 : 10*56. 



Distance of light from photometer . . .56 inches. 



Distance of positive brush from photometer . 4 „ 

 Therefore, Light from positive brush : Candle : : 1: 267,200. 



