LIBERATION OF NITROGEN, AND DE-NITRIFICATION 237 



into the exceedingly poisonous nitrites is also of interest to Pathological 

 Mycology, but we cannot further discuss the matter here. The important point, 

 so far as we are now concerned, is, that the disengagement of free nitrogen from 

 nitric salts can go on even in the presence of air. The de-nitrification occurring 

 in stored manure and in arable soil appears to be a twofold process : the 

 anaerobic nitrate destroyers acting in the lower strata away from the air, whilst 

 the symbiotic activity of the Bactarium coli commune (so plentiful in animal 

 excreta) and the Bacillus denitrificans I. comes into play at the surface. From 

 this it is evident that the theory which assumes the possibility of preventing the 

 destruction of nitrates by thoroughly loosening, and consequently aerating the 

 soil, is of little value. The bacteria in question are (for some unexplained 

 reason) present in enormous numbers in the excrement of various animals. 

 First in this respect is horse-dung, which has always been regarded by practical 

 men as a hot manure, a property which is explained by the foregoing observa- 

 tions. Consequently such manure should not be applied, especially when fresh, 

 to soil that has recently received a dressing of nitrate of soda ; otherwise a 

 serious loss of nitrogen will result. This injurious action is, however, not limited 

 merely to such fields as have been artificially manured with nitrate, since (as we 

 shall see in chapter xxxvi.) the ammonia salts in the soil are, under favourable 

 conditions, oxidised into nitrates by the activity of a special group of bacteria, 

 such nitrates then forming a welcome food for the organisms dealt with in the 

 present paragraph. That it is a question of more than insignificant quantities 

 will be evident from the discovery reported by PAUL WAGNER (I.) a discovery 

 which led to the aforesaid researches of Burri and Stutzer viz., that out of 100 

 parts by weight of nitrogen applied in the form of stall-manure to the soil, only 

 25 parts are, on an average, recovered in the crop, whilst the remaining 75 parts 

 are entirely lost. These figures do not fully represent the extent of the loss 

 occasioned in the soil and manures by the activity of the de-nitrifying bacteria, 

 and there still remains another phenomenon for consideration. We must recall 

 that the fission fungus known as Bacillus denitrificans (and probably also a number 

 of allied species not hitherto investigated) separates nitric oxide as well as nitro- 

 gen from nitrates. This oxide then escapes into the outer layers of the manure 

 heap or soil, where it is brought into contact with oxygen, and combines there- 

 with to form nitrogen trioxide 



2NO + = N 2 3 . 



This latter then reacts on the ammonia and ammonia derivatives (urea, &c.) 

 in the soil, in such a manner as to liberate both the nitrogen of the trioxide and 

 that of the ammonia as well 



N 2 3 + 2NH 3 = 2N 2 + 3 H 2 0. 



Consequently the nitrogen compounds insusceptible to the direct action of the 

 microbes in question are also included in the wasteful reaction set up. It was 

 on this account that the production of ammonia during the decomposition of 

 manure was casually referred to at the commencement of this paragraph. The 

 present is a fitting opportunity for referring to the statements of several workers 

 e.g. H. B. GIBSON (I.) who, like Reiset, thought they had observed a libera- 

 tion of nitrogen in their researches on putrefaction. Their results were all 

 obtained by the use of complex bacterial mixtures, and therefore cannot be 

 considered as reliable. In this case, also, those experiments alone are decisive in 

 which pure cultures have been employed. 



By the activity of these bacteria an enormous quantity of combined nitrogen 

 is daily set at liberty in the soil. To replace this loss, and to restore the con- 



