372 _ BACTERIOLOGY. 
a suitable toxin and by employing and distributing an 
antitoxin as a standard to test toxins by. In this way 
smaller testing stations can make their results corre- 
spond with those of the central station. 
In spite of the great variations in the neutralizing 
value of a fatal dose in different toxins we do not be- 
lieve ‘there has been any such great difference in the 
toxins used by the different stations for testing pur- 
poses. Most laboratories have taken the culture fluid 
at about the time of its greatest toxicity, and the 
neutralizing value of a fatal dose of this toxin would 
seldom vary more than 10 per cent. above or below the 
standard now adopted in Germany by the government 
testing station, this latter being presumably as close as 
possible to that used to establish the original Behring- 
Ehrlich unit. 
Where error has been made it has usually been by 
taking too old culture fluids, which would cause the 
antitoxin strength of samples tested to be estimated 
below and not above its real value. Culture 8, which is 
used not only by the New York Board of Health Labor- 
atory but by many other laboratories in the United 
States and Europe, fortunately produces on the sixth 
to eighth day—the time at which the culture is usually 
removed—a toxin which grades Ehrlich’s antitoxin 
within 5 per cent. of the strength given by him. 
We believe that by using such a bacillus we can, 
after gaining a fuller knowledge of its characteristics, 
obtain a toxin of a known and suitable neutralizing 
value, and thus always correctly standardize an anti- 
toxic serum. This is certainly true for the bacillus 
which we have used for the past four years. Mean- 
while, a fairly permanent preparation of a carefully 
