DISTRIBUTION OF B. COLI COMMUNE 13 
On January 24th five samples of rain water were collected from gutters at 
different places on the roof of the laboratory. B. coli was found to be absent in .5 
c.c. in each case. 
The last analysis was made on April 3rd, 1900, three samples of rain were 
collected, as it was falling, in sterile petri dishes. In each case no B. coli was found 
in 1 c.c. 
Hail was also analyzed with the same result. Unfortunately, the storm 
(January 27th, 1900) lasted a very short time, so that only a small quantity could be 
collected. The B. coli was, however, found to be absent in 1 c.c. 
(d) Dust. — In all five samples of dust were examined ; they were obtained on 
two occasions (February 27th and April 3rd) from different places in the laboratory, 
from the tops of cupboards, &c, where it was lying thick and undisturbed. The 
samples were taken by means of a sterile swab, 10 c.c. of sterile water was then poured 
into the test tube containing the swab and the whole thoroughly well shaken, and 1 c.c. 
of the mixture taken for analysis. This was the method adopted in the case of four 
samples, and no B. coli was found. In the case of the fifth sample, the swab was itself 
shaken successively in three tubes of melted carbolized agar, and these tubes then 
poured ; these plates were also found to be free from B. coli, though by no means 
sterile. The plates had to be very carefully examined, and many subcultures made of 
colonies of which it could not definitely be said that they were not B. coli, although 
they had not the typical appearance. In every case these cultures proved to be an 
organism other than B. coli. However, for analyzing such substances as dust, earth, 
sand of filter beds, &c, it must be confessed that the method of plating with carbolized 
agar is by no means ideal, and requires a great deal of tedious work in examination of 
separate colonies before definite conclusions can be drawn as to the presence or 
absence of the B. coli. 
On one occasion also the dust underneath the laboratory floor was collected 
and analyzed, one gram being weighed and shaken up with 20 c.c. sterile water. From 
this, typical B. coli were isolated, and they were present approximately to the extent of 
10 per gram. The fact that the bacillus was found here and not in the dust lying upon 
the cupboards, etc., may probably be explained by the possibility of a culture of B. coli 
having at some time been dropped and broken on the floor, and may be traced to 
a direct infection. 
(e) Virgin Soil [moorland). — Samples of moorland soil were obtained from 
different places on the high ground of the Lake Vyrnwy and Rivington watersheds. 
In all 1 5 samples were examined, and the details are given in Table IV. The analysis 
was made as follows : 1 gram of the soil was weighed into a sterile flask with the 
help of a sterile platinum spatula ; 10 c.c. of sterile water were added in some cases, 
and 50 c.c. in others. The whole was then thoroughly well shaken and 1 c.c. taken 
for analysis, and, as usual, three plates of each were poured. The B. coli was found 
