268 MICROBES, FERMENTS, AND MOULDS. 
Koch employs a more certain method. He makes 
his sowings on glass plates, covered with sterilized 
gelatine and kept, at a temperature of 30°, by means of 
a slender platinum wire which has been made red hot, 
then allowed to cool, and charged with a very minute 
particle of matter, which is full of bacteria. The 
colonies of different microbes isolate themselves, and 
may be plainly seen on the glass plate with the aid of 
a magnifier. Their variable size and characters often 
enable experienced observers to distinguish them by 
their aspect alone (Fig. 87, 1, 2). The test-tubes, 
containing sterilized gelatine, are then inoculated 
with the microbe which it ‘is desired to study (Figs. 
82, 105), after taking the usual precautions. 
The filters used to sterilize liquids are of Sévres 
biscuit-ware heated to 120°, or unglazed pottery. 
Such is the Chamberland filter already described. 
Cultures for Experiments on Animals.—The pro- 
cesses we have just indicated are also necessary in 
these experiments. Here likewise all the causes of 
error which would arise from the want of cleanliness, or 
from the impurity of the culture liquids, must be care- 
fully avoided; and it must also be ascertained that 
the effect produced on the animal is not due to any 
other microbe than that of the experiment, nor to 
any irritating and septic substance. The experiment 
should be repeated several times by taking some of 
the blood of the inoculated animal, and making a pure 
culture, which may be used to reproduce the disease 
in other animals. 
