Frost—Bacteriological Control of Public Milk Supplies. 1321 
number of changes in color or character of curd which are signi¬ 
ficant, and third, a more constant amount of gas production. 
It appears to be worthy of further trial, and it is to be hoped 
that other workers will test its merits. 
b) Comparison of Smith and Durham Tubes. The Smith fer¬ 
mentation tubes are inconvenient in several ways as compared 
with the Durham tubes. They are more difficult to fill; they re¬ 
quire considerably more medium; they are difficult to store and 
hence are harder to handle in the sterilizer and incubator, and 
they are more time-consuming to clean. 
If the inner tubes of the Durham apparatus are of the same 
length it is not difficult to measure the amount of gas formed; in 
fact, a gasometer can be constructed and used exactly as with the 
Smith tube. The tubes used for this wock were heavy walled 
test-tubes, y 2 inch by 6 inches, and a homeopathic vial of 1 dram 
capacity and 65 mm. long. The gasometer used is shown in Fig. 
7. 
The Smith tube was used exclusively in the early part of the 
investigation, then, for a time, the two kinds were run in dupli¬ 
cate, and finally the Durham tube was used instead of the Smith 
tube, with equal satisfaction, so far as the results obtained were 
concerned, and with much greater comfort. When the gas for¬ 
mula is desired, or a titration of the medium from the closed 
and open arm separately, then the Smith tubes must be used. 
When, however, it is sufficient to know that gas is formed, and 
about how much, the Durham tube is quite sufficient, and is very 
much more convenient, and saves medium. 
c) Use of Endo’s Medium. The use of Endo’s medium for the 
detection of B. coli in milk has been tried by a number of 
workers. Kinyoun and Deiter, for example, # are very enthu¬ 
siastic over its use. Ford,f however, says: “The results were, 
in general, not so satisfactory as those obtained from a dextrose 
tube.” It has been tried out. in the course of my work, and a 
column is provided in the Tables I. and VIII. for the data ob¬ 
tained. Towards the last its use was discontinued, not so much 
because it was thought useless, but because of the difficult tech¬ 
nique it requires when compared with the use of fermentation 
tubes. 
* Am. Jour. Pub. Health, 1912, 2, p. 979, and also in personal conver¬ 
sation. 
f Johns Hopkins Hosp. Bull., 1913, 24, p. 25. 
