200 
BULLETIN OF THE BUREAU OF FISHERIES 
showed increased growth as compared with the controls, while in 19 cases there was 
a retardation of the propagation or great mortality, accompanied by tremendous 
development of bacteria. In 14 experiments the population during some period of 
growth was greater than in the controls. These periods of increased rate of propa- 
gation occurred usually during the first few days of growth. Thereafter the controls 
continued to increase more rapidly and by the end of the experiment attained greater 
populations. These relationships may be noted by examining figure 16 representing 
the results of 3 experiments. 
In all the experiments with oil extracts, the retarding effect of the latter is appar- 
ent in concentrations of 25 percent and higher and when the extract is permitted to 
act over a considerable period of time. 
EFFECT OF OIL HELD IN COLLODION BAGS 
The method used in these experiments consisted in suspending in a flask contain- 
ing SO cc of culture medium, a collodion bag with 2 cc of crude Pelto oil. Bags were 
made by dipping a test tube in a thin solution of collodion (Merck’s reagent) and 
drying it for 45 seconds. Freshly made bags were sterilized by boiling in distilled 
water. Each bag remained suspended in a culture medium throughout the experi- 
ment. A more or less continuous dialysis through the collodion membrane was 
noticed, for water gradually collected in the bottom of the bag. The colorless and 
apparently somewhat volatile dialyzed substances have not been identified. Their 
presence was readily detected, however, by odor and irritating effect on the mucous 
membrane of the mouth while pipetting the sample for measurement. All the 
controls contained collodion bags filled with the Miquel solution or sea water. The 
end results of the experiments are given in table 22 the examination of which discloses 
that in all the cases there was a noticeable retardation in Nitzschia growth caused 
by oil held in collodion bags. Figure 17 presents a more complete record of one of 
the experiments. 
Table 22. — -Retardation of growth of Nitzschia culture caused by oil held in collodion bags 
[Each figure is an average of measurements of 6 different flasks] 
Beginning of experiment 
Duration, 
days 
Retardation 
of growth, 
percent 
Medium 
July 26 --- - 
22 
20 
Miquel. 
Do. 
12 
24 
Do 
12 
12 
Aug. 22 
14 
32 
Do. 
EFFECT OF BRINE ON NITZSCHIA 
Untreated brine (bleed water) separated from the oil as it comes out of the well 
was filtered from the heavy sediments accompanying it. The filtrate was carefully 
brought to the boiling point for sterilization. Sometimes this caused precipitation 
and the sample had to be discarded. 
In a series of experiments lasting from 14 to 31 days tests were made with various 
concentrations of brine added to sea water or Miquel solution. As can be seen by 
examining figures 18 and 19 representing two typical experiments, the retardation 
