304 
Journal of Agricultural Research 
Vol. IX, No. 9 
METHOD OF SAMPLING 
For the determination of nitric nitrogen samples of soil were taken in 
foot sections to a depth of 6 feet, by means of a King soil tube. Single 
samples were taken from as near the center of the plot as possible, care 
being taken that separate borings were at least 3 feet apart. The sam¬ 
ples thus obtained were taken to the chemical laboratory, where nitric 
nitrogen and moisture determinations were made immediately upon the 
samples. The results reported here, therefore, are all referred to 
moisture-free basis. 
For the bacteriological work all possible precautions were taken, when 
collecting the samples, against the contamination of one sample by 
another. The surface soil to a depth of half an inch was scraped off by 
means of a sterile spade. A hole 12 inches deep was dug, and a slice 
of soil to this depth was taken from the side of the hole and placed in 
a sterile mixing pan. This process was repeated from four or five places 
in the field, and then the contents of the pan was carefully mixed by 
means of a sterile spatula. From this composite sample a representa¬ 
tive portion, about 5 pounds of soil, was placed in a sterile ore sack and 
conveyed to the laboratory for analysis. 
Before each sampling, the spade, mixing pan, and spatula were all 
carefully sterilized by heat from a plumber's torch, thus preventing 
the transfer of organisms from one soil to another. The samples were 
immediately transferred to the laboratory, partly air-dried in the dark, 
and then ground in a sterile mortar, all coarse rock being removed. 
The analysis was begun in all cases within 24 hours of the time of taking 
samples. 
methods of analysis 
The soil extract for the determination of nitric nitrogen was obtained 
by means of the Pasteur-Chamberland filter. For rapid work a series of 
24 Pasteur-Chamberland filters was arranged together and connected 
to a tank of compressed air filled by means of an air pump run by a % 
horsepower electric motor. Fifty gm. portions of the soil were triturated 
in a mortar with 250 c. c. of distilled water, and 2 gm. of quicklime for 2 
minutes, allowed to settle 20 minutes, and then filtered through the 
Pasteur-Chamberland filter. By this method a clear, colorless filtrate 
was readily obtained. 
An aliquot portion (50 c. c.) was immediately measured into 100-c. c. 
beakers and evaporated to dryness on the electric hot plate. The residue 
was treated with 1 c. c. of phenol-disulphonic acid equally distributed over 
it, and then allowed to stand for 10 minutes. This solution was diluted 
with water and the excess of add neutralized with dilute ammonia. 
The color produced was compared with that produced by a standard 
solution of potassium nitrate treated in the same manner. The quantity 
