166 PA THOGENIC BA CTERIA.

be examined pass through a horizontal sterile tube about
70 cm. long and 3.5 cm. wide (Fig. 43), the interior of
which is coated with gelatin in the same manner as an
Esmarch tube. The tube, having been prepared, is
closed at both ends with sterile corks carrying smaller
glass tubes closed with cotton. When ready for use the
tube at one end is attached to a hand-pump, the cotton
is removed from the other end, and the air passed through
. very slowly, the bacteria having time to precipitate upon
the gelatin as they pass. When the required amount has
passed the tubes are again plugged, the apparatus stood
away for a time, and subsequently, when they have
grown, the colonies are counted. The number of colo-
nies in the tube will represent pretty accurately the
number of bacteria in the amount of air which
passed through the tube.

In such a cylindrical culture it will be noted
that if the air is passed through with the
proper slowness, the colonies will be much
more numerous near the end of entrance than
that of exit. The first to fall will probably
be those of heaviest specific gravity—i. e. the
moulds and yeasts.

A still more exact method is that of Petri,
who uses small filters of sand held in place in a
wide glass tube by small wire nets (Fig. 44).
The sand used is made to pass through a
sieve whose openings are of known size, is
heated to incandescence, then arranged in
the tube so that two of the little filters, held
in place by their wire-gauze coverings, are
FIG. 44.— superimposed. One or both ends of the tube
etiis san ^& closed with - corks having a narrow glass

filter for air- s .&.

examination, tube. The apparatus is heated and sterilized
in a hot-air sterilizer, and is then ready for
use. The method of employment is very simple. By
means of a hand-pump 100 liters of air are made to pass
through in from ten to twenty minutes. The sand from