92 
The Opsonic Index — A Medico- Statistical Enquiry 
frequency distribution for normal serum might be very constant, in which ease it 
might not be necessary to put up a normal serum control on each occasion. It is 
doubtful, however, whether one would get in this way significant differences for the 
tubercle bacillus, although it might work with other organisms. In the following 
tables I give the calculated means and standard deviations together with the 
frequency distributions and the corresponding indices for two organisms. 
Counts were made of 100 leucocytes at a time, and these component distribu- 
tions are here recorded in order to give an idea of the variability of means, 
standard deviations, etc. with such small samples. The two organisms used were 
the Streptococcus foecalis and the Bacillus friedlander. 
I. Normal Serum. Streptococcus foecalis. 
B* 
L(l)* 
L(2) 
L(3) 
L(4) 
i(5) 
Totals 
U 
4 
6 
8 
5 
9 
32 
1 
10 
.3 
11 
5 
7 
36 
2 
9 
12 
10 
8 
10 
49 
3 
6 
6 
13 
7 
13 
45 
14 
14 
10 
13 
20 
71 
5 
6 
9 
8 
6 
2 
31 
G 
9 
8 
5 
7 
9 
38 
7 
6 
9 
8 
4 
34 
8 
11 
7 
8 
14 
2 
42 
9 
6 
8 
2 
4 
5 
25 
10 
3 
6 
2 
9 
2 
22 
11 
2 
4 
2 
1 
5 
14 
12 
6 
2 
4 
4 
4 
20 
13 
0 
2 
2 
1 
2 
U 
4 
3 
2 
1 
3 
13 
15 
1 
1 
0 
2 
1 
5 
16 
1 
1 
1 
2 
1 
6 
17 
0 
0 
0 
1 
1 
2 
18 
2 
1 
1 
1 
5 
19 
1 
0 
1 
20 
1 
0 
1 
21 
0 
0 
1 
1 
Totals 
100 
100 
100 
100 
100 
500 
Mean 
6-17 
6-16 
5-42 
6-65 
5-38 
5-96 
S. D. 
4-24 
3-99 
4-44 
4-38 
4-21 
4-28 
* 7j = Number of Bacteria ingested. L (1), L (2) etc. = successive frequency distributions of ingesting 
leucocytes. 
