2j 



to set up an energetic denitrification. In an experiment made by myself 

 in the Kothamsted laboratory in 1880, and published in the Journal of 

 the Agricultural Society of England in 1881 (2nd Series, Vol. 

 xvii.J 7lbs of dry, finely powdered loam were placed in a percolator, 

 and thoroughly saturated with water; the. column of soil was about 

 eight inches in depth. The nitrates naturally present having 

 been entirely removed by the passage of water through the soil, 

 a known quantity of nitrate of sodium was placed on the surface. 

 After a week had elapsed a sm:ill quantity of water was placed each day 

 on the surface of the soil, and the drainage water removed and analysed. 

 When nitrates no longer appeared in the drainage water it, was found 

 that only 21 per cent, of the nitrate applied to the soil has heen recover- 

 ed ; 79 per cent, had been lost. The mode in which this loss had occur- 

 red was evident to the eye ; the column of soil, though always kept in 

 a saturated condition, was cut across about the middle with large trans- 

 verse fissures, the result of the formation of gas within the soil. In 

 another experiment, made with the same soil under identical conditions, 

 in which an equivalent quantity of chloride of sodium was employed, the 

 whole of the salt applied vas recovered in the drainage water, and no 

 fissure? appeared in the column of sod. 



A recently published experiment by Breal furnishes a further ex- 

 cellent example of the active denitrification which takes place in a soil 

 kept saturated with water. Replaced some garden soil in a percolator 

 and consolidated it by pressure ; the column was about fifteen inches 

 high. Water more than sufficient for saturation, was then poured upon 

 the soil. When the water had run through, it was poured back again 

 over the soil, and this treatment was continued for some time. The soil 

 at the commencement of the experiment was in an active state of nitri- 

 fication, and the drainage water was at first rich in nitrates, but at the 

 end of three weeks the nitrate had entirely disappeared from the drain- 

 age water, though no water had been removed from the soil. 



We have now the conditions which bring about a loss of nitrogen by 

 denitrification plainly before us. There are needed : 1. The specific 

 organism. 2. The presence of a nitrate and suitable organic matter. 

 3, Such a condition as to aeration that the supply of atmospheric oxygen 

 shall not be in excess relatively to the supply of organic matter, 4. 

 The usual essential conditions of bacterial growth as plant food, and a 

 suitable temperature. 



Of these conditions, the supply of organic matter is by far the 

 most important in determining the extent to which denitrification 

 will take place. An abundance of the special organism is of 

 comparatively little importance ; for, if the conditions are made favour- 

 able for its growth and development, it will increase with such rapidity 

 that the number of organisms originally present will soon become a 

 matter of indifference. Nothing, however, will compensate for a defi- 

 ciency of organic matter ; if this is small in quantity, the action will be 

 strictly limited in extent, however large may he the supply of the speci- 

 fic organism This fundamental principle of the reaction has been 

 strangely overlooked by the German investigators when interpreting 

 the results of their experiments. They have sought to explain the deni- 

 trification in their experiments as due to the supply of additional orga- 



