Ill 



Detorminations were made of tlie anioniit of nifro<jeTi in tlie forms of 

 iiriiinoiiia, nitric acid, and total nitrogen in cacii soil at the beginning', 

 from time to time during, and at tlie end of the experiment. The 

 anionnts of gain or loss in each specimen were thus learned and the cor- 

 responding amounts i)er hectare were calculated. The author urges, 

 however, that the excess of ammonia iu the soil at the end over that at 

 the beginning of the experiment does not represent the amount of 

 ammonia absorbed, since part of it is used for the fornuition of complex 

 nitrogenous compounds, and is not transformed into ammonia and dis- 

 solved by thedilute acid used for extracting and determining the amounts 

 of ammonia in the soil. More or less of the ammonia would of course 

 be (dianged to nitric acid. 



The twenty-tive experiments were divided into four groups : 



(1) The first group included six non-calcareous soils, which bad served 

 as material in previous experiments by the author on the fixation of 

 the free nitrogen of the air. Since no gain of nitrogen had been found 

 in either of these soils in the previous experiments, it was assumed that 

 any gain which should be found in the present experiments must be 

 attributed to the absorption of nitrogen comi)ounds from the air. The 

 soils were kei)t moist during experimental periods of from 4J)5 to 622 

 days. One was nearly covered, and showed no considerable gain of 

 nitrogen. In the other five the gain ranged from 15.3 to 50.1 kg. per 

 hectare, or from 13.7 pounds to 45.1 pounds i)er acre per year. 



(2) Thesecond gronp included threeexperiments, one with a calcareous 

 soil, the other two with a non-calcareous soil and subsoil. The exjjeri- 

 ment continued 115 days from May 4. The soils were moistened daily. 

 The calcareous soil gained nitrogen at the rate of 47.1 kg. per hectare, 

 or 42.4 pounds per acre, and the non-calcareous soil and subsoil at the 

 rate of 39.1 and 31.4 kg. respectively per hectare, or 35.2 and 31.0 pounds 

 per acre per year. 



(3) The third group included eight experiments with four specimens 

 of surface soils and the corresponding subsoils, all calcareous. The 

 experimental periods varied from 1(;3 to 170 days, commencing June 

 17. In every case there was considerable gain of nitrogen, the gains 

 by the surface soils varying from 40.7 to 4G.5 kg. per hectare, or from 

 36.6 to 41.9 j)0unds per acre, and by the subsoils from 30.8 to 40.1 kg. 

 per hectare, or from 27.7 to 36.1 pounds per acre per year. 



(4) The experiments of the fourth group were made with the same 

 kinds of soil as the third, at the same time, and in the same manner, 

 except that the soil was not moistened, but was thoroughly' stirred once 

 a week. It therefore retained only hygroscoi)ic moisture and was quite 

 dry. The gains of nitrogen averaged somewhat larger than iu the third 

 group. The most noticeable ditterence between the results with the 

 two groups was in the quantities of ammonia and nitric acid at the end 

 of the experiments. In the third group the quantities of ammonia 

 changed very little, but that of nitric acid increased and at the end was 

 large. In the fourth group, on the other hand, the quantities of nitric 



