218 NITRIFICATION 



in a soil increased when pure air was led through it, but that 

 no increase was observable when the air contained a trace 

 of an antiseptic like chloroform or carbon bisulphide. Further 

 ex23erinients cast Hght on the conditions under which the 

 nitrogen in ammonium - salts would thus pass over into 

 nitrates — a preliminary seeding from a previously nitrifying 

 solution or from soil or natural waters is necessary — bright 

 hght inhibits the process, and the drying up of a soil, even 

 at the ordinary temperature of a room, is sufficient to destroy 

 the agent of the change. All these facts showed that the 

 change to nitrate was effected by living organisms present in 

 the soil and in natural waters. It was also shown that certain 

 food substances, particularly phosphoric acid, are required in 

 the nitrifying solution. About the same time also, Munro 

 sho\ve(l that the organism can obtain its carbon from purely 

 inorganic sources like the carbonates of ammonia or calcium, 

 acquiring the necessary energy for splitting up the carbon 

 dioxide from the combustion of the ammonia to nitrous and 

 nitric acid. This remarkable fact was afterwards more 

 rigorously demonstrated by Winogradsky, who established a 

 relation of about thirty-five to one l)etween the nitrogen oxidised 

 and the carbon assimilated. 



In the course of Warington's experiments he observed that 

 when a comparatively strong ammoniacal solution (containing 

 also phosphates, etc.) was seeded from a soil, the first product 

 of the oxidation was largel}', if not wholly, nitrites, and that 

 these nitrites were converted into nitrates at a later stage 

 when most of the original ammonia had been oxidised. This 

 seemed to indicate that the reaction takes place in two stages, 

 a preliminary oxidation to nitrite being followed by a second 

 independent change of the nitrite into nitrate. Warington 

 succeeded in separating by relocated cultivations one agent that 

 >^ould only carry on the first oxidation from ammonia to nitrite, 

 and a second that would oxidise nitrite into nitrate but would 

 not attack the original ammoniacal solution. Although 

 Warington was not able at that time to demonstrate in a pure 



