144 
Journal of Agricultural Research 
Vol. IX, No. 5 
Table II .—Comparison of direct-distillation and acid-extraction methods for determining 
ammonia in soils 
Nitrogen recovered. 
Ammonium salts added. 
Nitro¬ 
gen 
added 
per 100 
gm. of 
soil. 
Riverside soil. 
Lordsburg soil. 
Extracted with 
10 per cent of 
hydrochloric 
add. 
Extracted with 
10 gm. of mag¬ 
nesium oxid. 
Extracted with 
10 per cent of 
hydrochloric 
acid. 
Extracted with 
10 gm. of mag¬ 
nesium oxid. 
Ammonium hydroxid. 
Ammonium nitrate. 
Ammonium chlorid. 
Ammonium sulphate. 
Mgm. 
50-00 
50.00 
50.00 
50.00 
Mgm. 
26.16 
26. 28 
26. r6 
26.30 
Per ci. 
52.32 
52-56 
52.32 
52.60 
Mgm. 
26.20 
26.30 
26.40 
26.50 
Per ct . 
52.40 
52.60 
52.80 
53-00 
Mgm . 
10.10 
10.00 
10.10 
8- 50 
Per ct. 
20.20 
20.00 
20.20 
17.00 
Mgm. 
11.00 
10.80 
10. 7S 
9.10 
Per ct. 
22.00 
21.60 
21.50 
18.20 
After it became evident that the ammonia could not be removed from 
the soil by extracting with acid or distilling with magnesium oxid, the 
soil was distilled with increasing amounts of potassium and sodium hy- 
droxid. The results secured are presented in Table III. It is seen that 
the ammonia removed by boiling with potassium and sodium hydroxid 
is greater than the amount recovered by extracting with acid or dis¬ 
tilling with magnesium oxid; but that much of the ammonia was retained 
by the soil, even when boiled with large quantities of potassium and 
sodium hydroxid. One c. c. of 10N potassium-hydroxid solution 
seemed to be less effective than 2.5 c. c., but the maximum removal of 
ammonia seems to have been secured when 2.5 c. c. of solution were 
added. In some cases the sodium hydroxid seems to have been more 
effective than the potassium solution; but this is possibly due to the 
fact that duplicate samples were quite variable, and, even though the 
results given are averages of several determinations, some allowance must 
be made for the lack of uniformity in duplicates. It is obvious that 
much of the ammonia fixed by this soil can not be removed by boiling 
with caustic substances, even when the compounds are added in great 
excess. When the soils were boiled with 10 per cent hydrochloric acid, 
more of the ammonia was removed than by extracting with cold hydro¬ 
chloric acid of the same strength (Table IV). However, boiling for one 
hour left about one-third of the ammonia in the Riverside soil. When 
the boiling was continued for four hours, the ammonia removed varied 
from 86.4 to 88 per cent. The ammonia seems to have been more firmly 
fixed in the Lordsburg soil, as boiling for four hours with 10 per cent 
acid removed less than 75 per cent of the nitrogen added. 
