Nolan — Ionic Mobilities in Air and Hydrogen. 87 



which we are at present iTiiable to define exactly. The normal condition 

 when the hydrogen has been in the vessel for some days is represented by 

 curve A. It is possible that the positive ion of mobility 9'4 would be found 

 on close examination to resolve into two groups, corresponding to the two 

 negative ions found normally. 



If we consider now the other end of the negative curve, we find that it 

 crosses the voltage axis at about 97 volts. This indicates an ion of mobility 

 18'8, which would correspond to Haines' ion Bg of mobility 19-6. The part 

 of the curve lying between 100 volts and 190 volts is rather complex, and 

 seems to suggest a considerable number of groups of ions. Some of them we 

 have already found in the positive ionization (15"0 and 11"0), others are new; 

 for example, ions of mobility about 14, 12, and 10. In working over this part 

 of the curve it is easy to see at any one time that the ionization consists of a 

 mixture of different groups, but the identification and segregation from one 

 another of these groups from day to day is not easy. Small quantities of air 

 finding admission to the apparatus will pull down all the mobility values. 

 This, if not recognized, will lead to luuch confusion. It is advisable then, as 

 a test of purity, to locate an ion of well-established mobility. In this work 

 the author has adopted the practice of taking observations each day in the 

 region of the slowest positive ion as a test. This ion comes out very 

 distinctly. If under ordinary circumstances the bend in the curve occurs 

 about 350 volts, it is assumed that the gas is pure. As a test of purity this 

 may be illusoiy, but it is at least a method of getting corresponding conditions 

 from day to day. 



The results so far described have been obtained with hydi'ogen that was 

 not specially purified in any way, and which could be considered as only 

 moderately dry. The only drying action to which it was exposed was 

 bubbling through a layer of strong sulphuric acid about 2 cms. deep. The 

 effect on the ions of making the gas moister was next tested. 



Mohilities in Moist Hydrogen. 



Water was substituted for the strong sulphuric acid just mentioned. 

 When a slow stream of hydrogen bubbles had been passing through the 

 water and into the vessel for some time, the examination of the ionization 

 was resumed. With tlui positive ionization no change was detected. The 

 current values were the same as before for corresponding voltages. The five 

 slower ions (7-9-5-2) came out quite distinctly. The region of the three 

 faster ions (15-0-9'4) was not examined, but from the general run of the 



