80 HOW CROPS FEED. 
ler’s test,* which is of extreme delicacy, and which he con- 
stantly employed in his investigations. 
Zabelin operated in closed vessels. The apparatus he 
used consisted of two glass flasks, a larger and a smaller 
one, which were closed by corks and fitted with glass 
tubes, so that a stream of air entering the larger vessel 
should bubble through water covering its bottom, and 
thence passing into the smaller flask should stream through 
Nessler’s test. Next, he found that no ammonia and 
(by Price’s test) but doubtful traces of nitrous acid could 
be detected in the purest water when distilled alone in 
this apparatus. 
Zabelin likewise showed that cellulose (clippings of filter- 
p2per or shreds of linen) yielded no ammonia to Nessler’s 
test when heated in a current of air at temperatures of 
120° to 160° F. 
Lastly, he found that when cellulose and pure water to- 
gether were exposed to a current of air at the tempera- 
tures just named, ammonia was at once indicated by 
Nessler’s test. Nitrous acid, however, could be detected, 
if at all, in the minutest traces only. 
Views of Schinbein.—The reader should observe that 
Boettger and Schinbein, finding in the first instance by 
the exceedingly sensitive test with iodide of potassium 
and starch-paste, that nitrous acid was formed, when hy- 
drogen burned in the air, while the water thus generated 
was neutral in its reaction with the vastly less sensitive 
litmus test-paper, concluded that the nitrous acid was 
united with some base in the form of a neutral salt. Af- 
terward, the detection of ammonia appeared to demon- 
strate the formation of nitrite of ammonia. 
We have already seen that nitrite of ammonia, by ex- 
posure to a moderate heat, is resolved into nitrogen and 
water. Schénbein assumed that under the conditions of 
* See p. 54 
