EXPERIMENT STATION BULLETINS. 



395 



is absorbed bj potassium permanganate. While the method can ap- 

 parently be made quite accurate, one factor of uncertainity was met with 

 in the formation, by some secondary reaction, of free nitrogen ( ?) from 

 the reagents used. The earlier experimenters had this same trouble and 

 Kreusler^ decided that it was due to the dissociation of NO into ele- 

 mental nitrogen and oxygen. In running blank determinations, we 

 found that no matter what precautions were used some gas was always 

 left which could not be absorbed in potassium permanganate. This 

 gas was shown by qualitative tests to contain a small amount of oxy- 

 gen but most of it could not be absorbed in pyrogallic acid. By working 

 under similar conditions this factor may be made practically constant. 



Fig. 1. 



The following apparatus (figure I) and method of procedure were 

 used in the determination of amino nitrogen. It is practically the 

 same as was used by Van Slyke. The apparatus consists of a small 

 wide-mouthed bottle (A) of 80 c.c. capacity fitted with a three-hole 

 rubber stopper which supports (1) a glass cylinder (B) closed at the 



»Vers. Stat. 31, 284 and 302. (1885). See also Sachsse and Kormann, Vers. Stat. 17,[324.T (1874) 



