VII.] NITRIFICATION 123 



them in a steamer for an hour or so to the temperature 

 of boiling water. One flask is retained as a check, to 

 another about a gramme of fresh soil is added, to a 

 third and fourth a gramme of soil each and half a 

 gramme of carbonate of lime. One of these latter 

 flasks is then heated up to boiling again for half an 

 hour. The four flasks are put aside in a warm and dark 

 place for a period of three weeks or a month, and the 

 liquid in each is then tested for the presence of nitrates 

 by taking out one drop of it on a loop of platinum wire 

 and introducing it into a solution of diphenylamine in 

 sulphuric acid. The control — the nutrient solution to 

 which no soil has been added — ^will show no nitrate ; 

 sometimes the solution to which soil only has been 

 added will also show none, but in the solution to which 

 both soil and carbonate of lime have been added, 

 nitrates will be found and the ammonia may have 

 entirely disappeared. Finally, the fourth flask which 

 had been boiled after the addition of the soil will also 

 show no change, only the unaltered ammonia remains 

 without nitrates. In this simple way it is possible to 

 demonstrate that some living agency is forming nitrates 

 in almost all cultivated soils. These agents are, how- 

 ever, absent from peaty and acid soils, and do not occur 

 in subsoils taken below a depth which varies with the 

 nature of the soil. Another instructive experiment may 

 be made by filling a glass tube about 5 feet long and i 

 inch in diameter with fragments of chalk, the cork 

 which closes the bottom of the tube being supplied with 

 one exit-tube to drain off any liquid, and another rather 

 longer tube turned over inside, through which air can be 

 admitted to the interior. The tube is first of all seeded 

 with the nitrifying bacteria by running some well-water 

 through it, or by putting a little soil on the top of the 

 chalk, then a dilute ammoniacal solution or even a liquid 



