DISCUSSION OF RESULTS OF DETERMINATIONS. 23 



of ammonia per million and toward the end of the season the amounts 

 fell to about 3.6 to 4.7 parts per million. No such gradual decrease 

 in amounts of nitrites toward the end of the season occurred. The 

 variations in the quantities of ammonia and of nitrites in the indi- 

 vidual soil layers were apparently not then sufficient to account for 

 the marked and general and consistent changes observed in the case 

 of the nitrates. 



It may be remarked that although the soil is heavy in texture and 

 likely to compact under bad treatment, the soil on the plats was quite 

 loose and open, it being the first year after breaking. Also, toward 

 the middle of summer the soils cracked quite freely, so that there was 

 fair opportunity for soil aeration all season, and, as the precipitation 

 curve will show, there was no trouble with water-logging of the soil. 

 All conditions considered, it is apparent that conditions were not 

 extremely favorable for active denitrification. 



SEASONAL VARIATIONS IN NITRIFICATION. 



Referring again briefly to all of the figures, one is struck by the 

 remarkable uniformity of direction of all the curves, especially in 

 figures 1, 2, and 3. The facts that the same general variations exist 

 in all cases and that denitrification is probably not responsible for 

 the nitrate changes point to a general seasonal phenomenon which is 

 affected in degree but not in kind by the presence of the growing 

 crops. 



It must be kept in mind that the nitrates found were those soluble 

 in water and that the curves thus represent only the water-soluble 

 nitrates. . It would appear that nitrification began earliest in the 

 surface layer of the soil and that the maximum activity of this proc- 

 ess gradually extended downward into the successively lower 6-inch 

 layers and reached its maximum in any one layer in general at a 

 period of about two weeks later than in the next 6-inch layer above. 

 In other words, conditions of temperature, aeration, etc., were favor- 

 able for nitrification in the surface layer earliest, as would of course 

 be expected, and these favorable conditions obtained later in the 

 lower soil layers. By June 12 these requisite conditions had in 

 general reached their optimum throughout the 2-foot soil section 

 studied, and it might well be conceived that the physiological activity 

 of the bacterial flora necessary for nitrification reached its maximum 

 in the comparatively short period of about two or three weeks in each 

 soil layer after the conditions had once become most favorable and 

 that the general "crop" of nitrifying bacteria matured by the middle 

 of June. 



Under this interpretation it would be inferred that the nitrifying 

 bacteria had an active growing season, so to speak, of two to three 



173 



