4 BULLETIN 782, XJ. S. DEPARTMENT OF AGRICULTURE. 
4.5 cubic centimeters of permanganate solution is sufficient. After 
boiling from 1 to 2 minutes the precipitate becomes almost white. 
If the precipitate remains colored, add from 2 to 3 drops of hydrogen 
peroxid solution and shake well 1 or 2 minutes. When cold, filter by 
weak suction in a Gooch crucible, wash with distilled water, once 
with alcohol, dry at 100° C, and weigh. The weight of the precipitate 
multiplied by 0.271 equals anhydrous citric acid. The method is 
accurate in milk, as shown by the fact that triplicate determinations 
upon a sample of milk gave the following results: No. 1, 0.2246 gram, 
No. 2, 0.2270 gram, and No. 3, 0.2285 gram of citric acid to 100 
cubic centimeters of milk. The amount of citric acid found in the 
three analyses differs only by 4 milligrams. 
That ammonia played little or no part in causing the alkaline 
reaction, especially during the first 7 days of the incubation, is shown 
in Table 2. Experiments were carried on to determine the amount 
of ammonia produced after periods of 7, 14, and 30 days of incubation 
at 30° C. Three sets of tubes of skim milk were inoculated with 
cultures of the alkali-forming bacteria and the ammonia was deter- 
mined together with the reaction of the milk in each of the sets after 
the various periods of incubation. Each tube of milk contained 10 
cubic centimeters and was marked so as to show the total volume. 
The amount of water lost by evaporation was made up at the time 
of the determ in ation of ammonia and acidity. Five cubic centimeters 
or one-half of the 10 cubic centimeters was then removed for the 
ammonia test and the remaining 5 cubic centimeters portion was 
then titrated in order to determine the reaction. The ammonia was 
determined by the Folin method (6) with 0.2 cubic centimeter of oxyl 
alcohol added to prevent foaming. When the quantity of ammonia 
and the change in reaction of the milk was known, it was possible 
to calculate the amount of change in reaction due to the ammonia 
formed, and the difference between that and the total change was 
assumed to be due to formation of sodium carbonate or bicarbonate. 
As may be seen from the results in Table 2, the reduction of the 
acidity in 58 of the 68 cultures during the first 7 days' incubation 
was due entirely to the formation of alkaline carbonates. In 10 
cultures there was slight amount of ammonia produced, which as 
the results show accounted for only a very small part of the alkali 
produced. Generally speaking, the small amount of ammonia pro- 
duced by the alkali-forming bacteria shows that the reduction of 
acidity in milk after 7 days at 30° C. was due almost entirely to 
carbonates or bicarbonates. 
