TUBERCULOSIS. 233

The mechanical fragmentation of the bacilli had been
used by Klebs in the studies of antiphthisin and tubercu-
locidin, Koch now used it with advantage in his studies,
and pulverized living, fresh, virulent, but perfectly dry
bacteria in an agate mortar, in order to liberate the ba-
cillary substance from its protecting envelope of fatty
acid. In the trituration only a very small quantity of the
bacteria could be handled at a time, and Koch seemed
thoroughly aware of the risk incurred from inhalation of
the finely pulverized bacillary mass.

Having reduced the bacilli to fragments, they were
removed from the mortar in distilled water, and collected
by centrifugation, in a small glass tube, as a muddy re-
siduum at the bottom of an opalescent, clear fluid. For
convenience he named the clear fluid TO, the sediment
TR. TO was found to contain tuberculin. In order
to separate the essential poison of the bacteria as perfectly
as possible from the irritating tuberculin, the TR or
fragments were dried perfectly, triturated once more,
re-collected in fresh distilled water and re-centrifugated.
After the second centrifugation microscopic examination
showed that the bacillary fragments had not been resolved
into a uniform mass, for when TO was subjected to stain-
ing with carbol-fuchsin and methyl-blue it was found to
exhibit a blue reaction, while in TR a cloudy violet reac-
tion was obtained.

The addition of 50 per cent, of glycerin had no effect
upon TO, but caused a cloudy white deposit to be thrown
down from TR. This last reaction showed that TR con-
tained fragments of the bacilli which are insoluble in
glycerin.

Experiment showed that TR had decided immunizing
powers. Injected into tuberculous animals in too large
dose it produced a reaction, but its effects were entirely
independent of the reaction. Koch's aim in using this
substance in therapeutics was to produce immunity in
the patient without reactions, by gradual but rapid in-
crease of the dose. In so large a number of cases did