AGRICULTURAL EXPERIMENTS 251 



Warington also accumulated several important facts upon 

 another bacterial process going on in the soil which works 

 in an opposite direction and transforms nitrates into nitrogen 

 gas. Somewhat similar organisms again set free nitrogen gas 

 from ammonia and the organic compounds of nitrogen ; the 

 whole group of these changes, to which the name of denitri- 

 fication is sometimes given, involves the impoverishment of the 

 soil by loss of combined nitrogen. Warington investigated 

 the conditions under which these losses are likely to be of 

 practical importance and in this connection the long-continued 

 application of particular manures to the field plots afforded 

 some interesting data. For example, it is found that when 

 nitrogen is supplied as manure to the mangold crop, as much 

 as 78 per cent, is recovered when the nitrogen is put on in 

 the form of nitrate of soda, whereas the recovery from the 

 nitrogen of ammonium salts is only 57 per cent, and from 

 farmyard manure as little as 32 per cent. Nor is the deficit 

 left behind in the soil. On the wheat field, for example, of 

 the nitrogen supplied as farmyard manure year after year until 

 the soil has become exceedingly rich, only about one quarter 

 has been recovered in the crop and another quarter accumulated 

 in the soil, whilst at least half has been lost by bacterial con- 

 version into nitrogen gas. How to tune up this conversion 

 factor and obviate these losses is one of the most important 

 questions before the soil chemist. 



Allusion has already been made to the part played by 

 bacteria in the soil in bringing nitrogen into combination and 

 as a consequence of such discoveries the whole point of view 

 of the function of the soil has been changed during the 

 history of the Rothamsted experiments. It is no longer 

 regarded as simply a vehicle for the transference of nutrients 

 to the crop but as an active laboratory, sometimes enriching, 

 sometimes merely circulating but at other times impoverishing 

 the capital embarked in the land ; the relative magnitude of 

 these processes will become more and more under control as 

 the conditions under which they operate are better known. 



Another important outcome of this vital view of the soil is 

 the question of the conservation of its fertility under different 

 systems of farming. We see from the long-continued trials at 

 Rothamsted that under any scheme of treatment the soil tends to 

 arrive at a state of equilibrium and to yield a crop which only 



