6 
THOMPSON YATES LABORATORIES REPORT 
Table II that, when incubated at a low temperature, the carbolic acid in the 
medium has a very marked inhibitive effect (see Experiments 5 and 6), even when no 
such effect can be observed at a higher temperature. 
If B. coli commune is present in fair quantity, and a high dilution is 
used, as in sewage, the plates yield an almost pure culture of the bacillus and the 
numbers are easily counted. The plates also give the very characteristic smell which 
a culture of B. coli upon agar always has. The appearance of the colonies, both large 
and small, on the surface and in the depth of the medium is always very characteristic, 
and can usually be at once identified, especially under the low power of the microscope; 
the typical colony from these plates has, with two exceptions, always proved to be a 
typical B. coli commune. In material, such as river water, soil, &c, B. coli commune 
is present, if at all, in very small quantity, and a very low dilution, or none at all, is 
used. Here the plates by no means show pure cultures of B. coli ; they have to be 
very carefully examined and separate colonies cultivated, in order to be sure whether 
they are B. coli or not. The commonest organisms occurring along with B. coli, or in 
its place, are certain micrococci, one of which agrees with M. candicans, and some 
spore-bearing organisms of the 'subtilis' type. These, if small and in the depth, may 
look a little like colonies of B. coli, but can be at once detected as different when 
grown on agar. Others, however, even in such cases, have an appearance at once 
distinguishable by hairy mycelium-like outgrowths. Constant use has been made of 
subcultures on the various media to test the B. coli, when found, to see if it is typical. 
With these precautions the medium has been found useful to demonstrate the presence 
of B. coli and its quantity with fair accuracy, even in these latter cases. 
The B. coli isolated was in almost every case typical. For example, in 
experiments from January 12th to February 24th, 1900, made for the most part on 
river waters, &c, which were not highly polluted, 87 bacilli, resembling B. coli, were 
isolated and their characteristics observed ; 84 were typical B. coli commune ; one did not 
coagulate milk, and two gave no gas in glucose gelatine, although they showed typical 
coagulation of milk. These three bacilli were, in many other respects, similar to the 
typical colon bacillus. As to the importance of the various tests in deciding whether a 
bacillus is B. coli commune or not, a few words are necessary. The coagulation of 
milk has been considered a necessary property of typical B. coli, and many authors 
have considered that a form, which only differed in not possessing it, was a separate 
variety. Pioskowski,* however, considers it to be by no means a specific property, 
but one which can be modified by previous environment. For example, B. coli 
obtained from his urine media failed to coagulate milk ; but if grown many times on 
ordinary agar, and then in milk, would give the usual coagulation. I have found that 
all the examples of B. coli isolated gave coagulation of milk, with the exception of the 
* Cmtralbl. f. Bah. xix, p. 686. 
