170 Mr Richardson , On an attempt to detect radiation , 
the first. The wire was supported by sealing-wax both where it 
entered and where it left the tube. The other end of the tube 
was closed by a tight india-rubber stopper which could be removed 
to insert the plates. The outside of the tube was painted with 
several layers of black paint so that no light could get through. 
After getting rid of troublesome glows no photographic action 
could be detected. It was therefore concluded that no photo- 
graphically active radiation was shot off at right angles to the 
wire. 
In this apparatus there was no very ready means of discover- 
ing whether the discharge really passed through the part of the 
wire which was tested or not ; although it could be proved to be 
passing through the parts of the wire external to the glass 
apparatus by the insertion of a coil with an electrodeless discharge 
bulb or of a spark gap. This defect was remedied in the form of 
apparatus described below by the insertion of a mercury cup into 
which the excited wire dipped, within the exhausted vessel. The 
wire could be drawn at will out of the mercury and a spark gap 
thus produced. By observing the light from the spark the inten- 
sity of the discharge could therefore be determined. 
It was possible that although the ions did not move off at right 
angles to the wire they might still be produced close to the 
surface and pursue straight paths backwards and forwards parallel 
to the wire under the influence of the electric field. In this case 
the application of a strong enough transverse magnetic field 
would cause them to move away from the wire, so that a photo- 
graphic effect might still be got if the wire were placed in a 
maguetic field. To accomplish this the wire was fixed axially in 
a narrow glass tube, which was placed between the poles of an 
electromagnet. 
The apparatus finally took the form shown in figure 1. The 
discharge wire passed through the top of a T-tube, the top being 
about 30 cms. long and ‘6 cms. diameter. The limb of the T-tube 
was cut off about 1 cm. from the shoulder and inserted into the 
cork e , so that it opened into the inside of a wide test-tube f 
which was cut off about 12 cms. from the open end. Another 
tube g also passed through the cork to the water-pump and mano- 
meter used before. The limb of the T-piece only penetrated 
half-way through the cork which was cut out so that the end of 
the tube was flush with the surface. This enabled the photo- 
graphic plates, which were fixed by soft wax on to the end of the 
support d, to be brought as near the discharge wire as possible. 
The tight india-rubber stopper h allowed of the removal of the 
plates, while the other end of the tube was made air-tight by 
covering the cork etc. with sealing-wax. The central wire was 
free except for being fixed air-tight in a with sealing-wax. A was 
