156 Proceedings of the Royal Society of Edinburgh. [Sess. 
The introduction of a platinum cup containing a sodium salt into the 
bunsen flame did not increase the distance at which oscillations could be 
produced, but seemed to have the opposite effect. 
As already mentioned, a Nernst filament, a candle flame, and a spirit 
flame were all found to be inferior to the bunsen as sources of ionisation ; 
but especial mention is required in the case of the electric arc, which 
was unable to produce oscillations at a greater distance than 6 inches, 
while 2 inches of platinum wire raised to a low white heat by an electric 
current were quite effective at 13 inches. 
An interesting effect was noticed while working with a glass screen 
which was held so near the jar that the pith bail was only a matter of 
1J to 2 inches from the screen. When a source of ionisation (in this case 
the Nernst) was brought up to the side of the screen remote from the 
jar, the ball was visibly attracted toward the screen, and when the Nernst 
was brought nearer still the ball jumped to the screen and stuck there. 
The effect was more marked if in place of the glass screen a thin insulated 
metal screen was used ; but in this latter case the ball, instead of sticking 
to the screen, oscillated between it and the centre rod of the jar. As 
before, this effect was independent of the nature of the charge in the 
jar. If, however, the charged pith ball were detached from the jar and 
suspended from a glass rod at a distance of 2 inches, as before, from the 
screen, the effect was no longer obtainable, and the approach of the Nernst 
to the other side of the screen produced no visible attraction between 
the screen and the ball. 
A possible explanation of these effects is as follows : — The knob of the 
jar being in close proximity to the screen induces on the side next to 
it an opposite and repels to the far side a similar charge. These opposite 
charges in close proximity to one another can have little influence 
on the ball, and the thinner the screen the less will this influence of the 
opposite induced charge be in attracting the ball towards it. If, however, 
the repelled similar charge be removed by any means, then the opposite 
charge is left free to attract the ball. This is what happened when the 
Nernst lamp was brought toward the far side of the screen, for then 
the ionised atmosphere surrounding it rapidly discharged the repelled 
similar charge, leaving the induced opposite charge to attract the pith 
ball. The fact that when the ball alone was present no effect was 
produced seems to show that it was dependent on the proximity of the 
highly charged knob of the Leyden jar. 
When X-rays were used as the source of ionisation the oscillation 
effects, though in the main similar to those produced by flames, etc., differed 
