14 BULLETIN 953, tf. S. DEPARTMENT OF AGEICrLTUEE. 
upper half of the silo. This is only what would be expected when 
it is considered that the ammonia nitrogen is a decomposition prod- 
uct of the amino bodies and the longer stay in the silo gives more 
time for such decomposition to take place. 
The amotmts of amino and of ammonia nitrogen, expressed in 
parts per million, do not follow parallel curves, although there is 
a general rise and fall throughout the whole period of juice collec- 
tion. The proportionate increase in the amount of amino nitrogen 
is greater, as is also the later decrease. The ammonia nitrogen in 
the first sample is nearly one-fourth as much as the amino nitrogen, 
but as the amount of amino nitrogen increases much more rapidly 
than the ammonia nitrogen it drops in the fifth sample to less than 
one-fifth. The proportion remains at one-fifth, or below, up to 
the sixteenth sample and then slightly rises to the thirty-sixth sam- 
ple. Here the amount of ammonia nitrogen decreases slowly and 
the amount of amino nitrogen decreases rapidly: consequently, the 
proportion of ammonia nitrogen to amino nitrogen in the last two 
samples is raised to over one-fourth. 
The specific gravity ranges from 1.027 to 1.046 and the acidity 
from an amount requiring 21.1 cubic centimeters to an amount re- 
quiring 39.5 cubic centimeters normal alkali for 100 grams of juice. 
The total nitrogen varies from 0.139 per cent to 0.323 per cent, and 
the albuminoid nitrogen from 0.0112 per cent to 0.0496 per cent. 
The ammonia nitrogen ranges from 171.7 to 44S.7 parts per million 
and the amino nitrogen from 710 to 2.139 parts per million. 
It will be seen from the tables that the greater part of the nitrogen 
present in the juice escaping from the silo is in the form of soluble 
nonalbuminoid nitrogen compounds. Although the actual food value 
of such compounds is still somewhat a matter of controversy, yet 
it may be a matter of interest, from a practical standpoint, to ob- 
serve the possible loss of food material caused by the escape of juice 
these two years. If the total nitrogen of the 2,579 pounds of juice 
collected in 191^15 is expressed as pure protein, we have a loss 
of 28.89 pounds, which represents the protein in about 1.500 pounds 
of average silage. Expressing in the same way the results for the 
season of 1915-16. we have a loss of 150.75 pounds in the 9.194.5 pounds 
of juice collected, representing the protein hi about 7.500 pounds of 
average silage. 
On a technical basis the results of the two seasons* study of silage 
juice may furnish some explanation for the large variations in the 
losses of soluble silage constituents which are occasionally reported 
by investigators. It shows how a large amount of juice, carrying 
with it much soluble food material, may sink to the bottom of the 
silo or easilv be lost through cracks or through an earthen floor. 
