I 2 
THOMPSON YATES LABORATORIES REPORT 
were made of I c.c. of the 100 c.c. in the flask, and all found to be sterile, so that B. coli 
may be said to be absent in 7.5 litres. Also, 40 c.c. of the water was taken, 10 c.c. of 
(5 in 1,000) carbolized broth was added, and the mixture incubated for 24 hours 
at 42 0 , the proportion of carbolic acid in the mixture being equal to 1 in 1,000. 
After 24 hours there was slight turbidity in the broth, which when stroked on agar gave 
a growth which was unlike B. coli, and proved to be a microccocus. By the phenol- 
broth method one can say that B. coli was absent in 100 litres ( T 4 o%- of 250 litres). 
Experiment II. — A second experiment, in which 200 litres were drawn through 
at a rate of 10 litres per hour, gave a similar result. By plating on carbolic agar B. coli 
was found to be absent in 8 litres, and by incubation with carbolized broth in 80 litres. 
Experiment III. — 249 litres of the air outside were drawn through at a rate of 
about 1 1 litres per hour, and treated similarly. By carbolized agar plates, B. coli was 
found to be absent in 1 5 litres, and, by the phenol-broth method, absent in 199 litres. 
In the next experiment the air was drawn through at a much greater rate, so 
that a great deal more was analyzed. It was found, in the previous experiments, 
that, of the cotton wool plug between the wash flask and the meter, the end nearest 
the wash flask became very dirty, showing that the water had but little effect in 
scrubbing the air, and hence in retaining the bacteria. The wash flask was therefore 
discarded and the air was passed through a plug of glass wool, which had been tightly 
packed in a glass tube and then sterilized. Glass wool was used instead of cotton 
wool, because the former can be so much more thoroughly washed. At the conclusion 
of the experiment the plug was pushed out, and the tube was washed down with 
sterile water into a sterile flask ; 10 c.c. of water was used, and the whole thoroughly 
well shaken and then analyzed. On this occasion the air outside was examined, and in 
all 2,893 litres were drawn through in about 48 hours, from February 17th to 
February 19th, 1900. The water was incubated with carbolic broth, and also plated 
on carbolized agar ; in neither case was B. coli found. By the first method it was 
found to be absent in 1,157 litres, 4 c.c. being used for analysis; and by the second 
absent in 867 litres. Adding the two results together, one may say that B. coli was 
absent in 2,025 litres. The weather at the time was very cold, and snow was on the 
ground. I am hoping to repeat the experiment in dry, dusty weather. 
(b) Snow. — On February 10th, 1900, snow was collected off the ground in 
sterile petri dishes, brought into the laboratory and allowed to melt, *5 c.c. and 1 c.c. 
being taken for analysis. Four plates were poured, and they were all found to be 
sterile. 
(Y) Rain. — On three occasions rain water was analyzed in order to see if the 
B. coli was present, but each time no B. coli was found. On January 17th, 1900, the first 
analyses of rain water were made ; rain was collected in sterile petri dishes and 1 c.c. 
plated on carbol-agar. The plates, after incubation, were found to contain no B. coli. 
On the same day plates were poured and exposed to the rain with the same result. 
