382 On the Frequency Distributions of Phagocytic Coimts 
TABLE I. 
Figures from London Hospital tested on Basis of Normal Distribution. 
5 xi. 07. " Control " Mean 2 053 + -1093. 
Expt. 
No. 
(1) 
Mean 
Bacilli 
(2) 
Difference A from "Control" 
Mean 
(3) 
Opsonic Infiex 
(4) 
AjE * 
2 
1-33 - 
■72+ -16 
0-65 
4-50 
3 
2-21 
•16+ -16 
1-08 
1-02 
e 
O 
2-57 
•52+ -18 
1-25 
7 
2-27 
•21 + ^16 
1-10 
1^31 
11 
1-95 
•10+ -16 
0-95 
0^63 
2 xl. 07. " Control " Mean 3'453 ± -0324. 
k 
4-05 
+ -60+ ^25 
M7 
2^43 
5 
3-43 
- ^02+^22 
0^99 
0^09 
6 
3-95 
+ -50+ ^25 
1-14 
2^00 
7 
3-07 
- -38+ -24 
0-89 
r58 
10 
4-71 
+ r26+ •25 
136 
5-04 
11 
3-48 
+ -03 +^24 
1-01 
0-13 
12 
3-63 
+ -18 +^24 
1^05 
0^75 
13 
3-43 
- ^02+^24 
0^99 
0^08 
No statistician glancing at tlie figures could help being struck by the apparent 
skewness of the distributions. 
We give in Table II. four examples taken at random. 
TABLE 11. 
Four Examples of Phagocytic Counts from London Hospital Records. 
Number of 
Frequency of Cells 
Bacilli 
per cell 
1 
2 
3 
0 
3 
5 
9 
2 
1 
15 
4 
18 
3 
2 
22 
7 
18 
9 
3 
26 
15 
23 
11 
4 
23 
16 
31 
11 
5 
25 
13 
21 
14 
6 
17 
7 
10 
9 
7 
5 
2 
10 
9 
8 
9 
4 
3 
3 
0 
5 
1 
5 
3 
10 
0 
1 
2 
1 
11 
0 
0 
0 
0 
12 
0 
0 
0 
0 
Totals 
150 
75 
150 
75 
* iJz; probable error of A. 
