Mr Kaye, The Selective Absorption of Rontgen Rays. 239 
disc which formed one of the metals of the anticathode. Villard* * * § 
showed that cathode rays reduce an oxidised copper plate to 
metallic copper. In just the same way it was noticed that the 
lead disc, which had tarnished a little before the apparatus could 
be sealed and exhausted, gradually developed a bright elliptical 
patch of clean lead at the place where the pencil of cathode rays 
impinged. 
As mentioned above, a portion of the Rontgen rays passed 
along tube B and out through the aluminium window W into 
the ionization chamber C. If a screen S was used to cut down the 
rays, it was inserted in front of the ionization chamber. The 
aluminium window was '067 mm. thick and 2 cms. diameter, and 
perceptibly sagged under the outside pressure. It was not possible 
to cement such a sheet of metal directly to the thin edge of the 
glass tube, and be sure of a joint which would stand. To get over 
this, a brass tube which would just slip over the glass tube, was 
provided with a heavy flange. A stout ring was screwed down to 
this by six screws equally spaced. The aluminium sheet was 
gripped between the ring and the flange, and the joint was com- 
pleted with sealing wax, the screws being tightened up while the 
sealing wax was still molten. The brass tube was then sealed 
to the glass tube, and this part of the apparatus gave no further 
trouble through leakage. 
The whole of the discharge tube was surrounded with sheet 
lead through a hole in which the Rontgen ray tube B penetrated. 
The ionization chamber consisted of a flat cylinder of brass 
about 4 cms. long and 9 cms. diameter. The faces were covered 
with aluminium leaf , 01 mm. thick. The front face was situated 
about 3 cms. from the aluminium window of the tube and was 
symmetrically placed with respect to it. 
A ring of aluminium wire of diameter rather less than that of 
the cylinder was mounted with its plane parallel to and midway 
between the faces of the cylinder. Over the ring was fastened 
aluminium leaf and the whole was supported by a wire passing 
through and clamped in a tube insulated by a sulphur plug which 
was inserted in an earthed guard-tube. The guard-tube was insu- 
lated from the chamber by an ebonite stopper let into the side of 
the cylinder. 
Aluminium leaf was chosen for the ends of the ionization 
chamber, partly because it is very transparent to Rontgen rays, 
and partly because it emits, as J. J. Thomson f, Townsend |, and 
Barkla§ have shown, but little secondary Rontgen radiation, and 
* Villard, Journal de Physique, 3 mo S6rie, viii. p. 140, 1899. 
t Thomson, Proc. Cainb. Phil. Soc. xxv. p. 109, Nov. 1906. 
X Townsend, Ibid. x. 1899, p. 217. 
§ Barkla, Phil. Mag. [6] 7, p. 543, 1904. 
