530 
Journal of Agricultural Research 
Vol. XXVIII, No. & 
A titration of the acid through which the hydrogen gas from the digestion of 
the ammonium salt with sodium hydroxid and Devarda alloy had passed for 24 
hours, before heat was applied, where the flask was connected with the condenser, 
showed a very small quantity of ammonia by a Nessler test. The amount 
obtained required less than 0.05 cc. of 0.1 N acid in the titration. 
Table I.— Percentage recovery of nitrogen from ammonium salts 
Method 
No. of 
determi¬ 
nations 
Maxi¬ 
mum 
Mini¬ 
mum 
Average 
A—Strowd method___ 
4 
98. 4 
97. 8 
98. 1 
B—Modified Strowd method, Bunsen valve_ 
4 
97. 1 
96. 2 
96. 6 
C—Same as B, but without £>evarda alloy_ 
3 
97. 9 
97. 1 
97. 4 
D —Same as B, but connected with condenser a __ 
E —Same as D, but without Devarda alloy 
12 
8 
97. 6 
99. 5 
92. 9 
97. 5 
95. 7 
98. 7 
a Block tin condensers were used for all distillations. A battery of same was kindly furnished for this 
work by the Feed Control Laboratory of this station. 
It would appear from Table I that for some unaccountable reason, the Devarda 
alloy had a detrimental effect on the ammonia recovery, as lower results were 
obtained in its presence. 
THE EFFICIENCY OF THE TWO PROCEDURES FOR THE REDUCTION OF RELATIVELY 
LARGE AMOUNTS OF PURE NITRATE 
Pure potassium nitrate and sodium nitrate were used and separate weighings 
of 0.25 gm. or 0.5 gm. were made, generally the latter amount. No material 
differences were found in the recovery of the nitrogen from either quantity and 
all are included in Table II. A comparison was again made by the modified 
method, connecting the flask with the condenser instead of the Bunsen valve 
during the 24-hour digestion at room temperature. The results are given in B 
and C of Table II. 
Table II.— Percentage recovery of nitrogen of pure nitrate 
Method 
No. of 
determi¬ 
nations 
Maxi¬ 
mum 
Mini¬ 
mum 
Average 
A-Strowd method__ __ 
5 
99. 2 
98. 7 
98. 9 
B-Modified Strowd method, Bunsen valve_ 
8 
99. 4 
97. 4 
98. 4 
C—Same as B, but connected with condenser_ 
6 
99.0 
92. 8 
& 95. 4 
« One determination equivalent to 89.2 per cent recovery is not included. 
& One determination equivalent to 90.9 per cent recovery is not included. 
It will be observed that good results were obtained by both procedures on 
relatively large amounts of pure nitrate. The determinations by the modified 
method where the flasks were connected at once with the condenser are not as 
good as those with the Bunsen valve. This is contrary to what would be ex¬ 
pected if there had been a loss of ammonia in the latter case. 
