SEA-FISHERIES LABORATORY. 225 
Now, reviewing the results of these experiments we 
find that in the original sample (that is, in the mussels 
before relaying in relatively pure water) there were large 
numbers of colon-like bacteria—five of these mussels 
contaimed so many that counting was impossible, and the 
average (round) number of colonies on the rest of the 
plates was 200. In the first sample, taken after four days’ 
quarantine, five mussels had undergone no cleansing, but 
three were sterile (to the medium employed), and, 
omitting the five that had undergone no change, we find 
that the average number of colon-like bacteria isolated 
from the mussels examined was about 40. The second 
sample, taken after eight days of quarantine, contained 
six mussels that had undergone no change, but the 
average numbers of colon-like bacteria isolated from the 
remainder was about 30. he third sample, taken after 
16 days’ quarantine contained four mussels which had 
not cleansed themselves, and the remainder gave an 
average tesl count of 55 colon-like bacteria. 
We find, then, that a decided reduction in the 
numbers of bacteria, belonging to the Lacilus coli group, 
took place in these mussels after relaying in relatively 
unpolluted water. In some cases the reduction amounted 
to total disappearance; that is, some of the relaid mussels 
in each sample must be regarded as practically free from 
sewage bacteria. ‘This is certainly the report which I 
would have made with regard to mussels ‘6, 8, 10, 16, and 
18 in the sample taken after four days; in mussels 4, 5, 
10, and 15 in the sample taken after eight days; and in 
mussels 1, 2, 3, 6, 11 and 17, in the sample taken after 16 
days. The original sample, taken before the beginning 
of the quarantine period, contained no mussels that 
could be regarded as practically free from pollution. 
Mussels 8 and 25 were not very seriously polluted, but 
