332 
Transactions of the Royal Society of South Africa. 
all possessed certaio common cljaracters — power of fermenting glucose, 
lactose, clotting milk and producing indol ; he found that the strain used for 
immunisation was the only one agglutinated. These common characters are, 
however, as shown above, referable to a considerable vai*iety of types or 
species. In the experiments to be recorded it is shown that among the 
commoner types of B. coli (A sub-group) even after these organisms have 
been classified and separated into different species by cultural tests, it is 
still impossible to demonstrate any species specificity of an agglutinating 
anti-serum, and that agglutinating sera are specific only for the individual 
homologous strain. 
Antisera were obtained to certain strains which represented common 
cohform types, (1) Type Al (B. MacCo7ihey No. 71), (2) Type A4 
{B. Griinthal), (3) Type A3 (B. vesiculostis), (4) TypeA2 (B . coli communis) , 
and a number of other strains corresponding in all their characters with these 
types were tested with the immune sera. 
Immime Sera. — Rabbits were immunised against the particular organisms 
by repeated intravenous injection of increasing amounts of bacillaiy emul- 
sions sterilised at a temperature of 65° C. for half an hour. For this purpose 
twenty-four hours agar slope cultures were emulsified in convenient quantities 
of 0 85 per cent, sodium chloride solution. The series of doses were as 
follows : i' 2 ' 1 2 emulsified agar slope cultures, given at 
intervals of 7 to 10 days. Ten days after the last injection the sera were 
tested with the strains used for immunisation, and if found of suitable value, 
i. e. agglutinating in a dilution of 1 : 2000 or in higher dilutions, the animal 
was bled and the serum after separation stored in sealed tubes. In the 
original experiments sterility of the serum was ensured by heating at 57° C. 
for one hour on three successive days. In some cases it was noted that there 
was a marked depreciation of the agglutinating value of the serum by 
heating, due apparently to the varying thermostability of the agglutinin. 
To obviate this the measures adopted for bleeding and collecting the serum 
were carried out with the utmost precautions to exclude contamination, and 
the serum was heated at 57° C. for only half an hour on two successive 
days. 
Method of Carrying out the Agglutination Tests. — A twenty-four hours agar 
slope culture was emulsified in 5 c.c. of 0*85 per cent, salt solution, and the 
emulsion allowed to stand in the incubator for about one hour to allow the 
larger clumps and fragments of agar to deposit. The supernatant fluid was 
then decanted and made up to 10 c.c. Varying dilutions of the antiserum 
were mixed with equal volumes (0'5 c.c.) of bacillary emulsion, and the 
mixtures placed in narrow tubes in which the agglutination could be observed 
by the naked eye. As a control 0'5 c.c. of the bacillary emulsion was mixed 
with an equal volume of salt solution and included in the test series ; this 
eliminated any fallacy due to auto-agglutination. It is to be noted, however, 
