1158 
Journal of Agricultural Research 
Vol. XXXI, No. 12 
cent tannic acid solution as long as precipitation occurred, excess of 
the precipitant being avoided. After 24 hours, or sooner, the pre¬ 
cipitate was filtered off, washed, and the filtrate and washings made 
up to 500 c. c. Two 25 c. c. portions of this were oxidized according 
to the Kjeldahl method, which gave the nitrogen of the protein-free 
extract (6). 
Each of two 200 c. c. portions of this solution received enough 
sulphuric acid to make a 5 per cent sulphuric acid solution and 
boiled under a reflux condenser in a glycerine bath for two hours. 
On cooling, the hydrolysates were nearly neutralized with sodium 
hydroxide solution, made alkaline with magnesium oxide previously 
reduced to cream, and distilled in 1,000 c. c. Kjeldahl flasks, the 
distillate being received in an Erlenmeyer flask containing N/10 
H 2 S0 4 . The titration gave the ammoniacal nitrogen corresponding 
to the nitrogen of acid amides present in the extracts (c ). 
The residues which remained in the 1-liter Kjeldahl flasks were 
thoroughly extracted with boiling hot ammonia-free water, the ex¬ 
tracts being filtered and washed with hot water. The black sub¬ 
stance which, together with the magnesium oxide, remained on the 
filter was quantitatively transferred to a 500 c. c. Kjeldahl flask and 
oxidized according to Kjeldahl’s method. The nitrogen thus ob¬ 
tained represented the humin nitrogen (d). 
All of the filtrates and washings from the black substance were 
together concentrated on the water bath and made up to 100 c. c. In 
two 10 c. c. portions of this the total nitrogen was estimated accord¬ 
ing to Kjeldahl’s method, and in another two 10 c. c.'portions the 
amino nitrogen was determined according to the method of Van 
Slyke {22, 23, 24 , 25), which yielded the nitrogen of amino acids 
present in the extracts (e ). 
To 50 c. c. of the remaining solution concentrated hydrochloric 
acid was added to a concentration of 20 per cent and boiled in a 
glycerine bath under reflux for 12 hours. The hydrolysate was next 
evaporated on the water bath to dryness in order to expel the hydro¬ 
chloric acid, taken up with hot water and made up to 50 c. c., of which 
two portions of 10 c. c. each were oxidized according to the Kjeldahl 
method; in two other 10 c. c. portions the amino nitrogen was esti¬ 
mated according to Van. Slyke’s method. Substracting from the 
result thus secured the amino nitrogen present in the solution prior 
to the hydrolysis was obtained the nitrogen of polypeptides present 
in the extracts (/). 
The residual water-soluble nitrogen, which is made up of nitrog¬ 
enous compounds other than those estimated above, constitutes the 
difference between the protein-free water-soluble nitrogen as obtained 
in (6) and the sum of the nitrogen secured as acid-amide nitrogen in 
(c), humin nitrogen in (d), amino nitrogen in (e), and peptide nitrogen 
in (f ). The results obtained by the above methods are summarized 
in Table V. 
