capable, at this time, of killing sheep; attenuated a little
more it ceases to be fatal to sheep, but still kills rabbits
and guinea-pigs. When it no longer kills adult guinea-
pigs, it still kills young guinea-pigs or young mice. This
is also true for other microbes.

Virulence appears to us, therefore, to be an intrinsic
quality of which the microbe will be divested more and
more until it becomes harmless. But here is a fact
which proves that things are not as simple as they seem.
If virulence were only this, the different methods of
attenuation would destroy it in the same fashion, and
the order in which the different species of animals are
attacked would be always the same. But experiment
shows that this order varies according to the method of
attenuation. The anthrax bacteridium attenuated, for
example, with bichromate of potash, in the experiments
of MM. Chamberland and Roux, as we shall see at once,
may still kill the sheep or at least make them very ill,
leaving them in the latter case vaccinated, while it
produces no effect whatever on rabbits or guinea-pigs,
and does not even vaccinate them. It is exactly the
reverse of the behavior of the anthrax bacteridium
attenuated by growth at 42° to 43°C., which kills guinea-
pigs and rabbits at a stage when it is harmless for the
sheep, and does not even vaccinate them. We obtain
the same results with spores of the anthrax bacteridium
attenuated by the action of a temperature of 35°C., in a
liquid containing 2 per cent sulphuric acid.

Thus virulence is not, as we might suppose, an absolute
quality, diminishing little by little after the fashion
of reserve food; it is a relative quality, in the estimation
of which not only conditions pertaining especially to the
microbe must be taken into account, but also those
pertaining to the nature, age, and as we shall soon see, the
individuality of the animal on which it is studied.

-307-