﻿of the Ionization produced by Rontgen Rays. 119 



rays as it has for ultra-violet light, and the tube was filled 

 with air which had been babbled through water both cold 

 and warm ; but in neither case could any absorption be 

 found. 



These preliminary experiments showed that the evacuated 

 tube was unnecessary; bat the general arrangement was 

 retained without exhausting the tube for the following- 

 reason : — 



It was found that the number of ions collected per minute 

 by the small detector was liable to vary in an erratic manner 

 when the aluminiums were away; and this was discovered to be 

 due to accidental distortions of the electric field, but the field 

 could be rendered steady by keeping the left-hand aluminium 

 sil ways in place [all the tubes were kept earthed]. Even 

 then the presence of the large detector, if it were charged, in 

 the nearer position seriously affected the field of the small one; 

 and therefore to ensure the small detector always collecting- 

 ions from the same region, the large one had all its parts 

 earthed whenever the small one was in use. The presence 

 of the right-hand earthed aluminium also exercised a steadying 

 effect upon the large detector; for this, although it had a 

 guard-ring, was from its size very susceptible to induced 

 charges, and it had finally to be surrounded on all sides except 

 the front by an earthed metal case. It then worked in a very 

 consistent and satisfactory manner. 



Final Results of Experiments. 



The mean of the experiments for the shorter distance when 

 the wires of large detector were 24 cm. from anode gave that 

 the ratio of the charges collected by the large and small 

 detectors in any time was 1*08, which was correct, taking the 

 extreme variations to 5 per cent. 



For the longer distances when the wires were 98 cm. 

 from the anode, this ratio came out to 1*04, with a slightly 

 less variation of about 4 per cent. The variations were 

 generally due to the fact that the ionization produced 

 when the current was first turned on and the bulb cold, was 

 generally larger than when it was turned on for 1 minute for 

 later observations ; in some cases these first readings were 

 omitted in taking the mean ; the primary current through 

 the induction-coil too often varied in the course of a minute 

 by "1 ampere, which must also produce slight errors. This 

 current was in general 4 amperes. 



