Tonization produced in Gases by Réntgen Rays. 89 
it was possible to do so, for solder joints were of no use, as 
some of the temperatures reached in the experiments were 
above the melting-point of solder. On the end B there was 
brazed a heavy brass flange, as shown at FF. PP was a 
thick aluminium plate 6 mms. in thickness, made so as to fit 
flat upon the flange. The central part of this plate was 
recessed to a depth of 5 mms., as shown in the diagram, so 
as to allow the rays to pass easily into the cylinder. To 
make the junction between this plate and the flange air-tight 
a lead wire, drawn down to a thickness of between one and 
two millimetres, was laid flat upon the flange, having one 
end neatly overlapping the other, and then the plate PP was 
placed flat upon this wire and tightly screwed down to the 
ange, by means of strong iron screws, until the lead wire 
became flattened down to halfits thickness orless. This joint 
proved very satisfactory, and was capable of withstanding 
any of the temperatures employed in the experiments. 
E and E’ were two parallel brass plates between which 
the rate of leak was measured, the rays passing between 
them after entering at the end B. The plate EH was sur- 
rounded by a guard-ring connected to earth. These plates 
were supported by the heavy brass rods H and H’ which 
passed out through the large insulating ebonite plugs K and 
K’. These ebonite plugs fit into the ends of the brass tubes 
D and D’', and these tubes were made of considerable length 
so that the insulation might be removed as far as was con- 
venient from the cylinder which was to be heated. To make 
the junctions between the ebonite plugs and the brass 
tubes air-tight large rubber corks M and M’ were bored to 
fit tightly over them, and then they were bound round tightly 
with heavy wire. A similar joint was made at N and N’ by 
means of heavy rubber compression-tubing. These junctions 
also proved very satisfactory. 
The cylinder and the inclosed gas were heated by means of 
an electric current passing through a coil of german-silver 
wire wound tightly round the outside of the cylinder, and 
insulated from the cylinder by means of sheet-asbestos. By 
this means the gas in the cylinder could be heated up fairly 
quickly, and by regulating the current the temperature of 
the gas could be kept quite steady. The temperature of the 
gas was measured by means of a mercury pressure-gauge R 
connected to the tube T, as shown in the diagram in fig. 4, 
where the general arrangement of the apparatus is shown. 
The increase in temperature, as indicated by the expansion of 
the gas, was measured by the height to which the mercury 
in the arm L rose above a fixed point Q. The mercury in 
