SELECTED PAPERS 



quest in this direction has shown that Pseudomonas aeruginosa, Micro- 

 coccus denitrificans and Denitro bacillus licheniformis all produce nitrous 

 oxide under normal conditions in fairly large quantities. 



For the last-mentioned species Verhoeven [1952] has definitely 

 shown that in growing cultures in the first phase of the process up to 

 65% of the nitrate is converted into nitrous oxide, whilst in the second 

 phase the greater part of this compound disappears and is converted 

 into nitrogen. It should, however, be emphasized that at the stage of 

 the maximum yield of nitrous oxide only a small part of this product 

 was present in the gas phase, about four fifth of it has remained dis- 

 solved in the medium, from which it was almost spontaneously set 

 free in the vacuum applied in the Van Slyke gas chamber. In various 

 respects this is a remarkable observation. In the first place it means a 

 solution for a riddle already signalized by Korsakova [1927], and 

 amply confirmed in several of Verhoeven's experiments, that quite 

 considerable nitrogen losses occur in certain stages of denitrification 

 processes when the media are analysed with the aid of the usual 

 methods, taking into account the gas evolved. In making such nitrogen 

 balances until now no attention had ever been given to the nitrous 

 oxide dissolved in the medium. 



But the observation is particularly remarkable when considered 

 from a quantitative poinl of view. For it was found that the amount of 

 nitrous oxide which could be extracted from the medium surpassed 

 at least several times the solubility of this gas in pure water as deter- 

 mined by Schwab and Berninger [1928]. We must, therefore, conclude 

 that nitrous oxide originating in an aqueous medium has a strong 

 tendency to form supersaturated solutions, unless we are willing to 

 accept that in the medium nitrous oxide is present in some unstable 

 compound (hyponitrous acid?) which at once dissociates with nitrous 

 oxide evolution, as soon as the partial pressure of the nitrous oxide 

 in the gas phase is markedly reduced. Whatever the explanation may 

 be, this does not do away with the irrefutable fact that in the second 

 phase of a normal denitrification experiment streams of nitrogen escape 

 from a medium which contains neither nitrate, nor nitrite. There can, 

 therefore, be no doubt that under certain conditions either free nitrous 

 oxide itself or some unstable compound which easily dissociates under 

 nitrous oxide formation is the source from which gaseous nitrogen arises. 



A further demonstration of this statement was obtained in an exper- 



490 



