94 ROYAL SOCIETY OF CANADA 
Again, with the cathode À well earthed, B beingthe other terminal, 
a large negative charge was induced when the earth connection to the 
cylinder was made through the wet string ; but when B was put to 
earth instead of À only a feeble inductive action was obtained. With 
A as anode and B cathode and the wet string inserted as part of the 
earth circuit there was a large positive induction if B was put to earth 
and but little when A was earthed. In both these cases the action was 
much more intense when the column of xylol replaced the wet string. 
At higher pressures such as one millimetre of mercury the action 
again varied with the conditions of the experiment. When A and B 
were the electrodes and neither connected to earth a small positive in- 
duction was observed when the column of xylol was inserted, but none 
when the connection was made by means of the wet string. On the 
other hand when one of these electrodes was well earthed there was a 
very strong positive induction such as that obtained at much lower 
pressures. 
The experiments in this direction were not carried further, as the 
presence of this inductive action was sufficient to show the importance 
of having good earth connections in investigations of this class. 
From these results it appears highly probable that the sudden de- 
flections obtained in the experiments described in section I may be 
traced to this cause, as in that case the earth connections were simply 
made by means of a platinum wire placed in contact with the cylinder. 
5. INFLUENCE OF SMALL OPENINGS IN THE FARADAY CYLINDERS 
INSERTED IN Vacuum TUBES. 
While electric conduction along the surface of the glass appeared 
to explain the effects described in sect*on I, it was also thought possible 
that the effects might be due in a measure to the presence of the small 
opening ¢ in the cylinder D, Fig. I. 
A similar opening existed in the cylinder of the tube shown in 
Fig. IV, but, as in this case the hole could be readily closed by simply 
raising the mercury, this form of tube was well adapted to investigate 
the influence of such openings. 
To investigate this point a series of experiments were made with 
the mercury lowered. The disc d was made of metal thick enough to 
cut off the cathode rays, and a discharge was passed through the tube 
at pressures both high and low, each of the electrodes A, B and D being 
in turn selected as the cathode. 
As a result of the tests no evidence was obtained of the presence of 
electrical charges within the cylinder except in one case. 
