368 SCIENCE PROGRESS 



In both cases the gases were prepared and purified with the 

 greatest care so as to eliminate moisture, dust, etc., which 

 might have effected the result. The general experimental 

 arrangement consisted in using three bulbs, which could be 

 evacuated, each bulb being fitted with an electrode of platinum- 

 iridium connected to an electrometer. The electrode could be 

 charged positively or negatively as desired. The two outer 

 bulbs were connected to the separate gas reservoirs, the reaction 

 being carried out in the central bulb. The natural rate of 

 discharge is observed and plotted, and at a certain moment 

 the gases are allowed to mix. If ionisation occurs it will 

 be shown by an abrupt change in the slope of the curve 

 obtained from the electrometer readings. The sensitivity of 

 the method was tested and shown to be satisfactory, by 

 measuring the ionisation produced on bubbling air through 

 water. 



The union of oxygen and nitric oxide had been examined 

 previously by several investigators from the point of view of 

 the production of ions, the conclusion finally arrived at being 

 that no ionisation occurs in this reaction. This conclusion has 

 been completely substantiated by Pinkus ; the time curve is 

 perfectly continuous. 



In the reaction between chlorine and nitric oxide, however, 

 quite different results are obtained. Pinkus has found that 

 when the chlorine is in excess the reaction is accompanied by 

 ionisation. Positive as well as negative ions are produced. 

 If, however, the chlorine is not in excess, or only in slight excess, 

 there is no break in the curve, i.e. apparently no ionisation 

 occurs. It is difficult to suggest an explanation of this phe- 

 nomenon. It is possible that, in presence of excess chlorine, 

 the unstable compound NOCl 2 (assumed by Trautz and his 

 collaborators) may be formed temporarily, and that the ionisa- 

 tion may be due to this. On the other hand, as Pinkus points 

 out, there is no definite evidence that such a compound is 

 formed, and even if it were, it would not necessarily give an 

 explanation of the phenomenon, for in the previous reaction 

 there is undoubtedly formation of intermediate compounds 

 such as N 2 3 , and yet this reaction is not accompanied by 

 ionisation. Pinkus concludes that the observed ionisation is 

 due simply to the gaseous reaction as a whole ; that is, we have 

 a demonstration for the first time that a homogeneous gaseous 



