1920-21.] Experiments with an Electrified Pith Ball. 155 
(4) The position of the source of ionisation can be shown to have an 
important bearing on the oscillations. Thus a candle held directly 
opposite the ball at a distance too great to be effective may be made 
so at the same distance if held at a lower level than the ball. From 
this we may conclude that the ions formed in the flame are carried 
upwards by the convection currents of air, and entirely disposes of any 
suspicion of projection of ions from the flame. Also, should the ball 
happen to be in any position other than directly between the source of 
ionisation and the centre rod of the jar, the source may he approached 
very close indeed to the ball without oscillations taking place. This is 
in confirmation of the previous deduction arrived at, namely, that the 
highly charged centre rod of the jar exerts a directive influence on 
the ions. 
In addition to the above, the atmospheric condition seems to play an 
important part. This appears to be a reasonable explanation of the 
varying maximum distance at which effects could be produced on different 
days when to all appearance other conditions were identical. 
The conclusion drawn from the whole series of experiments was that 
the nature of the charge given to the jar had no bearing on the results 
obtained, and though there were occasions when this did not seem to be 
the case, subsequent experiments showed that the discrepancies were 
probably due to variation in some of the factors just mentioned. 
In performing the experiments we employed the following method : — 
The jar was charged from a Wimshurst machine and placed on a bench 
with the pith ball standing out towards the ionising source, which was 
placed 30 inches off, with a screen interposed between it and the ball. A 
reading telescope was mounted in such a position as to have the ball in 
the field of view of its graduated scale. The screen was then removed 
and the ball carefully watched for any slight drop. This distance was 
invariably too great to show any effect, but on most occasions, when the 
distance was decreased to 25 inches, a slight drop could be noticed. As 
a rule, at 20 inches the drop of the ball was large enough to he seen easily 
with the naked eye, and occasionally oscillations commenced at that 
distance. This was when the ionising source was a bunsen burner ; but 
as a rule oscillations did not commence until the bunsen was about 18 
inches from the ball, though sometimes, under unfavourable conditions, 
much shorter distances were necessary before effective oscillations could 
be produced. Once oscillations had commenced, it was usually found that 
the source could be moved further oft' and the movement maintained at 
the increased distance. 
