DIPHTHERIA BACILLUS. 347 
well understood, not for from two to four days. In 
neutral bouillon the culture fluid frequently becomes 
slightly acid and toxin production may be delayed for 
from one to three weeks. The greatest accumulation 
of toxin is on the fourth day, on the average, after the 
rapid production of toxin has commenced. After that 
time the number of living bacilli rapidly diminishes 
in the culture, and the conditions for those remaining 
alive are not suitable for the rapid production of toxin. 
As the toxin is not stable, the deterioration taking place 
in the toxin already produced is greater than the amount 
of new toxin still forming. 
Bacilli, when repeatedly transplanted from bouillon 
to bouillon, gradually come to grow on the surface 
only. This characteristic seems to aid in the develop- 
ment of toxin. 
The relations of toxin to antitoxin will be described 
after the subject of antitoxin has been considered. 
Non-virulent Diphtheria Bacilli. Xerosis Bacilli. In 
the very large number of tests for virulence of the 
bacilli obtained from hundreds of cases of suspected 
diphtheria which have been carried out during the 
past six years in the laboratories of the Health De- 
partment of New York City, in over 95 per cent. of 
cases the bacilli derived from exudates or pseudomem- 
branes and possessing the characteristics of the Léffler 
bacilli have been found to be virulent, that is producers 
of diphtheria toxin. But there are, however, in inflamed 
throats as well as in healthy throats, either alone or 
associated with the virulent bacilli, occasionally bacilli, 
which though morphologically and in their behavior on | 
culture media identical with the Klebs-Léffler bacillus, 
yet producers, at least in artificial culture media and 
