88 Proceedings of the Royal Irish Academy. 



curve there is hardly any doubt that these also were present. On the other 

 hand, the increase of moisture produced a considerable effect on the negative 

 ionization. The negative curve when plotted boldly, as in figure 3, now cut 

 the voltage axis at about 235 volts instead of 200. When the lower part of 

 the curve was closely examined, it was found that there were two very 

 shai-p bends at 223 and 236 volts. The constant Vu was 1828, hence the 

 mobilities are 7'7 and 8-2. Ions of mobility higher than this were present, 

 but only in very small quantities. As regards the upper part of the curve, 

 the current readiugs for voltages such as 400 were the same as before. But 

 the curve between 236 and 400 volts was not smooth. A number of well- 

 marked depressions were found, which came out consistently in each 

 experiment. MobUities calculated from these were in good agreement with 

 the values obtained for the positive ions over this range, except in the case 

 of the slowest. There was nothing to correspond to the well-established 

 positive ion of mobility 5"23, but instead an ion appeared of mobility 478. 

 The positive ions were once more examined, and while it was found that an 

 ion of mobility 5'18 (in good agreement with the usual value) was the most 

 prominent, a definite concavity was found in tlie curve in such a position as 

 to correspond to the presence of ions of the lower mobility value in small 

 quantities. 



The extra moisture, then, may be said to have left the positive ion 

 unaffected, except in the very small degree just mentioned. A great part of 

 the negative ionization, however, is thrown into two close-lying groups of 

 mobilities 7'7 and 82 (cf. positive, 7"9). faster ions are present in veiy small 

 and slower in moderate quantities. "While these are the results obtained from 

 experiments on moist hydrogen up to the present, the writer would prefer 

 not to regard them as final. 



It is interesting to note that in the experiments on air previously reported 

 the fastest ion recognizable in the moist gas had a mobOity of about 2. If 

 we assume that the carrier is the same in the two gases, and that it is big in 

 comparison wilh ihe gas molecules, the ordinary theory of mobility will give 

 for the mobility of this ion in hydrogen the value 



J 



Density of air ^2x3-8 = 7-6. 



Density of hydrogen 



in good agreement with the value found above. A similar relationship seems 

 to exist between the slowest negative ion found in diy air and in dry 

 hydrogen. The value for air was 2'5. This multiplied by the ratio of the 

 square root of the densities gives 9"5, which seems to correspond with 9"63 for 

 hydrogen (cun'e B, fig. 6). It is perhaps unsafe to push this very far, but 



