Ai*g. 1, 1925 
Nitrogen Constituents of Celery Plants 
293 
over a low flame for 20 minutes. At the end of that period the 
material was again brought to vigorous boiling and the flame removed, 
the sulphuric acid containing the reduced nitrates being drawn back 
into the Kjeldahl flask. Twenty-five to 30 c. c. of distilled water 
were then poured upon the glass beads in the tower and the solution 
in the flask again brought to boiling; then the flame was removed 
and the wash water drawn into the flask. This washing was repeated 
four times, thus assuring complete transference of the sulphuric acid 
to the digestion flask. The material was then digested according to 
the Kjeldahl-Gunning method, the potassium sulphate not being 
added until all the water was boiled off. The digested material was 
diluted to 175 c. c., 25 c. c. of K 2 S solution (40 gm. K 2 S per liter) 
added, the whole made strongly alkaline with saturated NaOH and 
distilled as in the ordinary Kjeldahl procedure. 
For the determination of nitric nitrogen Strowd’s method {20) 
was employed. Eight grams of the material were extracted by shak¬ 
ing vigorously for 5 minutes with 250 c. c. of cold distilled water. 
The mixture was allowed to stand for 20 minutes and again shaken 
for 5 minutes. The extract was filtered and determinations made on 
aliquot portions. The portions were diluted to 250 c. c. in 500 c. c. 
Kjeldahl flasks and 2 drops of kerosene and 5 c. c. of saturated 
NaOH added to each. To half of the samples 1 gm. of Devarda’s 
alloy was also added. All samples were connected immediately to 
a Kjeldahl distillation apparatus, heated over a very low flame for 1 
hour and then 150 c. c. distilled over and caught in 50 c. c. of 4 per 
cent boric acid solution. Care was taken that the controls contain¬ 
ing no alloy were distilled at the same rate as those solutions con¬ 
taining the reducing metal. The distillate was titrated against 
standard acid, using. brom-phenol blue as an indicator, as in the 
Scales method. The difference between the readings of the control 
samples and those reduced by the alloy is due to the nitrate and 
nitrite nitrogen. . 
The determination of ammonia nitrogen was made by the aeration 
method of Folin {10, p. 499-500 ). One-gram samples of the healthy 
and diseased materials were placed in the aeration tubes and to each 
were added 10 c. c. of water, 1 c. c. of saturated potassium oxalate 
and 0.5 gm. of anhydrous sodium carbonate. The aeration was 
effected by means of a water pump, the rate being determined by the 
tendency of the suspensions to foam over. Aeration was continued 
for 12 hours. The ammonia was caught in 4 per cent boric acid and 
titrated as described. 
Only a qualitative test for nitrite nitrogen was made. Portions of 
the extract used in nitric nitrogen determinations as in Jodidi’s 
procedure for spinach were made neutral and to them was added 1 
c. c. quantities of the sulphanilic acid mixture. No nitrite was found 
by this method. 
A different method of extraction than that used by Jodidi’s pro¬ 
cedure was then employed for the test. Ten grams of celery were 
extracted for 5 minutes with 150 c. c. of distilled water and the in¬ 
fusion thus obtained was filtered through fine Swedish filter paper. 
The filtrate which had a brown color was partially clarified by shak¬ 
ing for 2 minutes with 10 gms. of Lloyd’s reagent with subsequent 
filtering; 25 c. c. of this filtrate was treated with 3 gms. of Lloyd’s 
