404 Dr. Richardson: Effect of a Luminous Discharge 
shown in the figure. In this way the platinum spiral could 
be kept at any desired distance from the electrodes DD,. It 
was heated by means of an electric current let in through 
the mercury (I, I,) ; its resistance was measured and its 
temperature kept constant by the method previously used in 
experiments on the negative leak from hot platinum*. The 
system CAA,C; was insulated when the discharge passed 
between D and D,, but was maintained at +200 volts in 
measuring the imparted emissibility. 
The experiments were conducted at a pressure of about 
-42 mm., and yielded the following numbers :— 
sibility (pro- 2 100 8 0 0 
portional to) 
Distance of 
spiral from ; 2. 2. “ 
Wise ites ia, te 18 2°5 30 6:2 65 
(cms.). 
Imparted Emis- 
240 14 
Evidently the effect rapidly diminishes as the spiral is 
moved away from the electrodes, and at distances greater 
than 4 ems. practically vanishes. 
It was shown in the previous paper} that when the testing 
wire itself was one of the electrodes, the imparted emissibility 
was considerably greater when it was the cathode than when 
it was the anode. This led to the suspicion that the effect 
was a function of the distance from the cathode only; in 
which case no emissibility would be imparted by a discharge 
to a wire forming the anode when the distance between the 
electrodes was great. On trying this, it was found that with 
the electredes 6°5 cms. apart no emissibility was imparted to 
the anode, whereas the effect on the cathode seemed to be 
much the same as with smaller distances. 
Experiments were next made using the discharge in a 
cylindrical tube so that everything was symmetrical about 
the axis of the cylinder. An arrangement was used by which 
the testing wire could be placed in various positions in the 
path of the discharge. It was found that the imparted 
emissibility fell off asymptotically with the distance from the 
cathode, and had no relation to the distribution of luminosity 
in the tube. | 
Finally, an apparatus was set up in which a screen could 
be inserted at will between the cathode and the testing wire. 
This was found to cut off the effect, showing that it was due 
* Phil. Trans. A, vol. cci. p. 508. 
+ Phil. Mag. [6] vol. vi. p. 80. 
