350 
Transactions of the Royal Society of South Africa. 
were found to correspond (Table XXIV). It is noteworthy that certain repre- 
sentatives of sub-group A whose fermentative reactions did not correspond to 
those of the accepted B. coli types deviated in the presence of 0*025 c.c. of 
antiserum over 15 doses of complement, and even by varying the amount of 
antiserum one could not differentiate them from an A4 strain other than 
strain 1, i.e. the strain used for immunisation (Table XXV). Included 
among these were two strains which according to the usual criteria would not 
have been classed as typical B. coli, viz. two non-lactose-fermenters, and a 
Table XXVI. — Lysis of 0'5 c.c. 5 jjer ceiit. Ox Blood Susjjension + 5 Doses 
immune body. 
Antiserum 
to Strain 3, 
Bl {B. Mac- 
Co nkey No. 
74). j 
Bacillary emulsion 
0 4 c.c. 
Doses 
of complement. 
Complement 
deviated by 
emulsion 
alone. 
! 
O-O05 c.c. ' 
0-001 „ 
0-0005 „ 
0-0001 „ 
] Strain 3, B f 
V (B. No. 74 \ 
MacConkey) 1 
3 D. 
0 
0 
0 
Trace 
5D. ! 
0 1 
0 j 
Dist. 
V. mk. 
10 D. 1 
0 1 
Dist. 
Mk. 
Al. com. 
15 D. 
0 
Al. com. 
Com. 
Com. 
20 D. 
0 
Com. 
Com. 
Com. 
3D. 
Mk. 
0 D. 
Com. 
0025 „ 
0-01 ,, 
0-005 „ 
1 
) Strain 2, Bl ( 
I (B. No. 74 J 
j MacConkey) { 
0 
0 
0 
0 
0 
0 
0 
0 
Trace 
0 
Trace 
Al. com. 
0 
Al. com. 
Com. 
Mk. 
Com. 
0-005 „ 
0-001 „ 
0-0005 „ 
0-0001 „ 
- Strain 1, B2 | 
0 
0 
0 
Trace 
0 
Trace 
Trace 
Mk. 
0 
Dist. 
Mk. 
V. mk. 
Trace 
Mk. 
Al com. 
Just c. 
Dist. 
V. mk. 
Com. 
Com. 
V. mk. 
Just c. 
0-005 „ 
U-001 „ 
0-0005 „ 
0-0001 „ 
1 Strain 2, B2| 
0 
0 
0 
Trace 
0 
Trace 
j Trace 
Mk. 
0 
Dist. 
Mk. 
Al. com. 
0 
Mk. 
Just c. 
Just c. 
Trace 
Al. com. 
Com. 
Com. 
Mk. 
Com. 
0-025 c.c. antiserum alone deviated 1 D. of complement. 
strain which only fermented lactose after mutation ; these strains all produced 
indol. The fact that such Gram-negative bacilli behave in complement-deviation 
experiments with an antiserum to a strain of typical B. coli similarly to other 
typical organisms strongly suggests that lactose fermentation, which has always 
been considered one of the most important characters of B. coli, can hardly be 
taken as of more importance biologically than other sugar reactions. 
Among the 31 strains of sub-groups A, B, C and D examined in this way, 
three A types (Nos. 1, 9 and 22) did not react characteristically with the A4 
and A3 antisera, i. e. the group reaction was absent in the case of the No. 1 
and 22 strains and not well marked in the case of the No. 9 strain (Table 
XXIV) . The No. 22 strain was isolated from a case of conjunctivitis in which 
it was present in pure culture ; the Nos. 1 and 9 strains were of direct faecal 
