230 
Dissolve casein by heating to boiling. It is 
desirable to let this solution stand for several 
hours to get a perfect solution. This is not neces- 
sary, however. Make up volume to 500 ec.c. and 
bring the reaction of the solution to between 
+ 0.1 and + 0.2 Fuller’s seale. Do not allow this 
solution to become alkaline to phenolphthalein or 
over —0.2. If the casein is weighed accurately 
and the normal solution accurate the reaction will 
be about + 0.2. 
The agar solution is prepared by dissolving 
10 gm. agar in 500 ¢.e. of water. Both casein and 
agar solution should be filtered, then mixed. Tube 
and sterilize in autoclave under pressure for 20 
minutes; then cool the tubes quickly in cold water 
or ice water. The final reaction of the medium 
will be about + 0.1 Fuller’s scale. If the medium 
is alkaline, the bacterial growth will be restricted. 
If the medium is more than + 0.1 some of the 
casein may be precipitated during sterilization. 
The casein agar should be clear and almost color- 
less when poured in a Petri dish. Sometimes the 
casein will be slightly precipitated during steriliza- 
tion on the cooling, but it is of no consequence, 
since on pouring into plates the precipitate on 
account of its finely divided condition becomes 
invisible. 
The study of the bacterial growth on casein 
agar and infusion agar shows the following points: 
1. The 24 hours’ count at 37° on casein agar 
was almost always lower than on infusion agar 
when raw milk is being examined. When pas- 
teurized milk was examined the casein plates 
showed a higher count in 37 per cent. of the 
samples. 
2. After 6 days’ incubation at 30° C., out of 50 
samples of raw milk plated, 44 per cent. of the 
samples showed higher counts on casein agar. 
With 50 samples of pasteurized milk, 78 per cent. 
of the samples showed a higher count on casein 
agar. 
3. From a study of the bacteria from about 50 
samples of both raw and pasteurized milk it seems 
that acid-forming bacteria do not develop quite as 
well on casein agar. It does, however, favor the 
growth of the alkali formers, the peptonizers and 
inert bacteria. 
4. The number of peptonizing bacteria in a 
sample of milk may be determined directly from a 
casein agar plate. After counting the plate it 
should be flowed with N/10 lactic acid; this causes 
the precipitation of the casein, giving a white 
opaque plate except where the casein has been 
dissolved about a colony of peptonizing bacteria. 
SCIENCE 
[N.S. Von. XXXV. No. 893 
There is then left a clear zone around the colonies 
of peptonizing bacteria which enables one to de- 
termine their numbers in the sample of milk under 
examination. It has been found from a study of 
a large number of samples that this method of 
determination is accurate. 
Sugars may be added to the casein agar or 
the casein solution may be used as a liquid me- 
dium without agar. It is believed that these 
media using casein will be of considerable value in 
bacteriological milk analysis. 
The Analysis of the Gases Produced by One Hun- 
dred Cultures of Bacteria: WM. MANSFIELD 
CLARK. 
The purpose of these analyses was to furnish 
data for the identification of gas-producing bac- 
teria isolated from dairy products. 
The bacteria were grown in a special form of 
culture bulb, evacuated with a mercury pump after 
inoculation, sealed up and incubated seven days 
at 30°C. The culture medium was a bouillon 
containing 1 per cent. dextrose. Exactly 5 cc. of 
this was used in each bulb. 
The collection of the gas was made with an 
Antropoff mercury pump and the analyses were 
made with special burettes and Hempel pipettes 
adapted for accurate analyses of small volumes. 
The majority of the cultures analyzed gave a 
ratio of CO., H, similar to that of B. colt com- 
munis. Certain other distinct ratios were found. 
These depend in large measure upon the volume of 
CO., the hydrogen tending to remain constant. 
Certain other relationships are suggested tenta- 
tively, pending further investigation. 
A Study of Gas-forming Bacteria in Milk: L. A. 
Rogers and B. J. Davis. 
Cultures of gas-forming organisms have been 
isolated from milk and other dairy products ob- 
tained in various parts of the country. These have 
been studied with special reference to the relation 
between certain physiological reactions, as the 
fermentation of carbohydrates and the amount of 
gas and ratio of H, to CO.. Plotted on the fre- 
queney basis the H,: CO, ratio has given four 
more or less distinct nodes, one at the ratio 
1:1.1, one at 1: 1.8, one at 1: 2.2, and one at 
WS Pot 
Arranged in a similar way, the amount of gas 
produced under given conditions shows nodes at 
4 c.c., between 7 and 8 cc. and 17 «cc. 
Proper classification of the cultures shows a 
close correlation between the H,: CO, ratio and 
the amount of gas. 
