188 Mr Almy, On a Diminution of the Potential Difference, etc. 
charge changes, the potential difference increases with increasing 
cathode area. And it is only with this second sort of discharge 
that the effect of the magnetic field is marked, or uniformly 
consistent. 
The transition from the one type of discharge to the other 
is very largely dependent upon external conditions — in fact it 
may be entirely prevented. If, for instance, the discharge tube 
outside the cathode and clear up to the anode be coated with 
tinfoil, connected to earth (the cathode being also earthed) the 
character of the discharge remains the same independent of the 
cathode area. In this case the discharge takes place entirely 
by stream lines, — usually by a single stream — which pass along 
the side of the tube, accompanied by luminous flashes, often with 
crackling noise as a spark discharge. 
The action of the magnetic field producing a diminution of 
potential difference between the electrodes is inseparably con- 
nected with the transition to the stream -like discharge, the 
action of the magnet is to always produce that type of discharge. 
Furthermore, the tinfoil coating of the tube entirely eliminates the 
effect of the magnetic field, and there is a diminution of the P.-D. 
between the electrodes due to the tinfoil, of the same magnitude 
as that produced by the magnet. 
The suggestion offers itself that the 
effect of enlarging the cathode is to in- 
crease the charge which collects on the 
walls of the tube, and this charge acts as 
a counter e.m.f. to increase the P.-D. ne- 
cessary for discharge in the tube. 
To test this idea two different experi- 
ments were tried. 
A tube, as shown in Fig. 2, was 
constructed. The diameter of the smaller 
tube was about 15 mm., that of the large 
part being about 60 mm. Aluminium 
cathodes of 1 cm. area were at each end, 
and a ring of aluminium wire at the centre 
was the anode. 
The effect of a magnetic field, trans- 
verse and parallel, to the direction of the 
discharge was observed with each end of 
the tube ; then the effect of surrounding 
the tube with tinfoil connected to earth, 
was tried. 
These results seem to indicate that 
the effect produced by the magnetic field is not due to the 
removal of a static field, for at the different ends of the tube 
