146 RUTH L. PHILLIPS 
showed a rather remarkable freedom from contamination after 
an interval of forty-six days. One of the bacterial mixtures 
showed complete absence of contamination when tested by 
plating at this time. These tests demonstrate that it is perfectly 
possible to avoid undue contamination of pure cultures of bac- 
teria used for feeding Paramecium, and that it is even possible 
to do this for some time without particular care. In practical 
work, however, one should always flame each pipette carefully 
before it is used, and take every possible precaution by sterilizing 
all media and glassware. 
The Paramecia used in all feeding experiments were isolated 
from a laboratory culture rich in both Paramecium aurelia and 
caudatum. Paramecium aurelia was chosen because of the 
greater ease with which it can be handled in small amounts of 
culture fluid. The usual method of growing pure-line cultures 
was employed, and stock lines were kept in vials of hay infusion. 
Perfectly fresh cultures of bacteria for feeding were secured by 
inoculating small vials of sterile 0.1 per cent infusions with pure 
cultures of bacteria. This was done each day, so that a series of 
bacterial cultures, each twenty-four hours old at the time of 
using, was obtained. Such cultures insured vigorous, rapidly 
growing bacteria, and were as free as possible from an accumula- 
tion of bacterial decomposition products. Mixtures of these 
cultures were made as desired in the watch-crystals in which the 
Paramecia were to be grown. By making the mixtures in this 
way instead of allowing the bacteria to grow in mixed culture in 
the vials of infusion, the possibility of unfavorable reactions of 
the bacteria with one another was reduced to the minimum. 
Workers with infusions of unknown bacterial content have 
supposed that the bacteria carried over in transfer of a Para- 
mecium to fresh medium would provide an ample supply of food. 
This assumption appears to have been justified. It was desired 
to use infusions of this sort as controls, but it seemed that this 
method might allow the gradual introduction of harmful bacteria 
and that the decline so frequently noted in the vitality of cultures 
of Paramecium might be due to such a change of food. To avoid 
this possibility in the chance mixtures, hay infusion was inocu- 
