958 Journal of Agricultural Research vol. vi.No. 34 



tent made up to the optimum (two-thirds water-holding capacity). Two 

 samples were incubated for one week and two for four weeks. The 

 ammonia and nitrate nitrogen were then determined and reported as 

 milligrams of nitrogen and nitrates (N0 3 ), respectively, per ioo gm. of 

 soil. In 1 91 3 the ammonia was determined as follows: The water con- 

 tent was made up to a definite volume and the whole shaken for 45 

 minutes. Two gm. of calcium oxid were then added and the contents were 

 again shaken for a short time and allowed to stand until the supernatant 

 liquid became clear. A definite volume of this liquid was then distilled 

 in the presence of magnesium oxid, the distillate being collected in 

 standard acid and titrated. The calcium oxid was added as a clarifying 

 agent in order to obtain a solution upon which nitrate and nitrite 

 nitrogen could be determined colorimetrically. The presence of this 

 reagent caused a perceptible increase in the ammonia set free, prob- 

 ably liberating some of the loosely attached nitrogen; but since the 

 results are comparative and because of reasons already mentioned, the 

 method seemed justifiable. However, such insignificant quantities of 

 nitrate and nitrite nitrogen were found after seven days' incubation 

 that these determinations were discontinued during 1914, and the am- 

 monia was determined by direct distillation of the soil in copper flasks. 



Where the incubation lasted for four weeks, the water loss by evapora- 

 tion from the soil was replaced from time to time. Besides these experi- 

 ments, samples were also run during 191 4 with the addition of calcium 

 carbonate in excess of that required to neutralize all nitric acid that could 

 be formed from the cottonseed meal. Nitrate nitrogen was determined 

 in all cases upon an aliquot part of a solution, obtained as directed 

 above, using the phenoldisulphonic-acid colorimetric method. 



The nitrifying inoculation experiments were conducted as follows: 

 A soil possessing both a high nitrifying capacity and nitrifying efficiency, 

 in the sense that Stevens and Withers (17) use these terms, was selected 

 as a standard medium. To 100-gm. samples of this soil, cottonseed 

 meal containing 60 mgm. of nitrogen was added, sufficient water added 

 to bring it up to optimum, less 20 c. c, and the whole subjected to 20 

 pounds' pressure for one hour in the autoclave. These samples were 

 then inoculated from the various plots with 20 c. c. of a soil suspension 

 made by shaking 1 part of soil in 2 parts of water. The incubation 

 covered a period of 28 days at room temperature. Ammonia and 

 nitrates were determined as stated above. The results are reported in 

 milligrams of nitrogen and nitrate (N0 3 ), respectively, per 100 gm. of 

 soil. It will be noted that the nitrate data for 191 3 are low, in many 

 places zero. This was caused by the failure to add calcium carbonate 

 which was added in 191 4. Apparently some substance toxic to nitri- 

 fication is produced by heating. This substance gradually disappears on 

 standing, and the disappearance is materially hastened by the addition 

 of calcium carbonate. 



