aw 
BACTERIA IN WATER. 627 
Later, on account of the exhaustion of organic pabulum, the num- 
ber is again reduced as the bacteria present gradually lose their 
vitality. Under these circumstances-it is evident that an estimate of 
the number of bacteria present in water from a given source can 
have no value, unless a sample is tested by bacteriological methods 
within a short time after it has been collected. Not more than an 
hour or two should be allowed to elapse, especially in warm weather. 
By placing the water upon ice the time may be extended somewhat, 
but Wolffhiigel has shown that the number of germs is gradually 
diminished when water is preserved in this way, and it will be safest 
to make an immediate examination when this is practicable. 
The collection may be made in a sterilized Erlenmeyer flask pro- 
vided with a cotton air filter, or in a bottle having a ground-glass 
stopper which has been wrapped in tissue paper and sterilized for an 
hour or more at 150° C. in the hot-air oven. Or the small flasks with 
a long neck may be used, as first recommended by Pasteur. These 
are prepared as follows: The bulb is first gently heated, and the ex- 
tremity of the tube dipped into distilled water, which mounts into 
Fie, 194, 
a7 
the bulb as it cools; the water is then made to boil, and when al] 
but a drop or two has escaped and the bulb is filled with steam the 
extremity of the tube is hermeticallyesealed. When the steam has 
condensed by the cooling of the bulb a partial vacuum is formed, 
and the tube is ready for use at any time. It is filled with water by 
breaking off the sealed extremity under the surface of the water of 
which a sample is desired. This is done with sterilized forceps, and 
care must be taken that the exterior of the tube is properly sterilized 
before the collection is made. The end is immediately sealed in the 
flame of a lamp. A difficulty with these vacuum ‘tubes is that they 
are so completely filled with water that this cannot be readily drawn 
from them again in small quantities. The writer therefore prefers 
to make the collection in a tube shaped as shown in Fig. 194, in which 
a partial vacuum is formed just before the collection by heating the 
air in the bulb. The water mounts into the tube as the air in the 
bulb cools, and is readily forced out again for making cultures by 
applying gentle heat to the bulb, Asa lamp is needed to seal the end 
of the tube in either case, there is no special advantage in having a 
vacuum formed in advance, and, as stated, the vacuum tubes are so 
