RELATION OF BACTERIA TO DISEASE. 97 
certain extent. The means usually employed are the 
frequent replanting of cultures and their growth in cap- 
sules placed in the bodies of susceptible animals. But 
with all our efforts we are usually only able to restore 
approximately the degree of toxin-formation which the 
cultures originally possessed. The adaptation of bac- 
teria to any nutritive substance, living or dead, so that 
they will grow more readily, is more easily brought 
about, provided they will grow atall. The streptococcus 
from erysipelas and the pneumococcus from pneumonia 
are typical of this class of bacteria. Inoculate a rabbit 
with a few streptococci obtained from a case of human 
sepsis, and, as a rule, no result follows ; inject a few 
million, and usually a local induration or abscess ap- 
pears; but if one hundred million are administered 
septicemia develops. From this rabbit now inoculate 
another, and we find that a dose slightly smaller suffices 
to produce the same effect; in the next animal inoculated 
from this still less is required, and so on, until in time, 
with suitable cultures, a very minute number will surely 
develop and produce death. The same increase in 
virulence can be noted when septic infection is carried 
in surgery or obstetrics from one human case to another. 
By allowing bacteria to continue to develop under cer- 
tain fixed conditions they become accustomed to them, 
and less adapted for all that differ. 
Somewhat distinct, again, from that class of bacteria 
which multiply rapidly are those which, like the tubercle 
and leprosy bacilli, develop slowly. Here increase of 
virulence is shown, as before, by the production of dis- 
ease through the introduction of very small numbers 
into the body, but increase in rapidity of development 
cannot progress except to within certain limits. A sin- 
7 
