12 Prof. Thomson, On Chemical Composition of a Gas [Oct. 31, 



To measure the saturation current the plate a was connected 

 to one pair of quadrants of an electrometer, the other pair of 

 quadrants and the guard pieces /3 and 7 were to earth, the two 

 pairs of quadrants were at the beginning of the experiment 

 connected together. The plate 8 was connected to one terminal 

 of a battery of small storage cells the other extremity of which 

 was put to earth ; for gases in which the ionization did not greatly 

 exceed that of air, 200 cells giving a potential difference of 400 

 volts between the two plates which were 1 cm. apart were found 

 sufficient to produce the saturation current, but for gases like 

 Cl 2 , H 2 S, S0 2 , HC1 in which the ionization is very much larger 

 than in air it was found necessary to use 600 cells to make sure 

 that the maximum current was reached. 



The connection between the two pairs of quadrants was then 

 broken and when the rays were not passing through the gas and 

 care had been taken with the insulation of the plates, the needle 

 of the electrometer remained at rest ; this was used as a test for 

 the insulation. When however the rays passed through the gas 

 the needle of the electrometer was deflected, as a charge of 

 electricity passed across the gas from the plate 8 to the plate a ; 

 the quantity of electricity which passes in a given time, in these 

 experiments 30 seconds, is proportional to the saturation current. 

 Experiments were made with the positive terminal of the battery 

 connected with 8 and then the connections were reversed and the 

 negative terminal was connected with 8, when the exhausted tube 

 was working steadily, the deflections of the electrometer on 

 opposite sides of the zero position were found to be the same 

 in the two cases. 



The tube used to generate the Rdntgen rays gave out 'soft 

 rays,', bulbs giving out very penetrating rays were found to be 

 too variable for these experiments. 



The results of the experiments are given in the following 

 table, the numbers give the saturation current through the 

 various gases, the saturation current through air being unity. 



H 2 = 33 



N 2 = -89 

 2 =11 

 C0 2 = 1-4 

 CO = -86 

 NO =1-08 

 N 2 = 1-47 



The experiments with NH 3 could not be made with the same 

 accuracy as with other gases as it was found impossible to dry the 

 ammonia. The gas was passed through tubes several feet long 



