194 
BULLETIN OF THE BUREAU OF FISHERIES 
Experiments carried out with diatom cultures and discussed in the present paper 
were devised to test the effect of oil (1) as a heavy layer covering the surface of the 
water; (2) absorbed by some neutral substance and held on the bottom of the culture 
flasks; and (3) as a soluble extract. 
METHOD 
Cultures of the single diatom species, Nitzschia closterium, E., were grown in the 
laboratory in solutions prepared according to the Miquel formula. (Solution A, 
potassium nitrate 20.2 g in 100 cc distilled water; Solution B, calcium chloride 4 g, 
sodium phosphate (secondary crystals) 4 g, ferric chloride 2 g, and 1 cc concentrated 
hydrochloric acid in 80 cc of water. Add 2 cc solution A and 1 cc solution B to 
one liter of carefully filtered sea water; sterilize by bringing just to the boilding point, 
then cool and filter.) Eighty cc of this solution were poured in round pyrex flasks 
of 150-cc capacity and inoculated by adding 2 or 5 cc of Nitzschia stock culture. 
Flasks covered with inverted small beakers were placed in front of the laboratory 
window where they were protected from direct sunlight. In some of the experi- 
ments instead of Miquel solution plain filtered sea water was used. Sea water used 
during winter experiments was received from Woods Hole, Mass., and stored in 
paraffined oak barrels. It contained noticeable amounts of hydrogen sulphide. 
Before using, the water was aerated for at least 24 hours and the precipitated sulphur 
filtered off. During the summer months the sea water from the laboratory supply 
was used. The salinity of the water varied between 31 and 32 parts per thousand. 
Twice a day temperature readings were made by means of a thermometer kept in 
one of the flasks filled with water. Fluctuations in the light and temperature con- 
ditions which were not controlled, are undoubtedly responsible for certain variations 
in the growth rates. To avoid this difficulty a comparison was always made between 
the experimental and the corresponding control flasks which stood by its side. The 
relative position of each pair of flasks was changed every other day to compensate 
for a possible difference in illumination. 
Diatom counts were made by means of a photoelectric set-up (fig. 11) which con- 
sisted of Weston photronic cell ( Ph ) connected to a Weston D. C. microammeter (A) 
of 0-100 ranges, accurate within 0.5 percent of the full scale value at any part of it. 
For measuring the turbidity of a sample, 50 cc of it were poured in a cylindrical glass 
container C, having a diameter of 27 mm, 90 mm high, with fused bottom made of 
optical glass. The cylinder rested on a diaphragm, the opening of which was slightly 
smaller than the inside diameter of the cylinder. Both the diaphragm and the 
cylinder were placed directly over the photoelectric cell the surface of which was 
covered with thick black paper in such a manner as to cut ail the light except that 
which passed through the column of water in the cylinder. A 21-candlepower, 
6-volt Mazda bulb ( L ) fed by a 108 ampere-hour storage battery served as a source 
of light. The photoelectric cell was placed in a wooden box 4 by 4 by 12 inches. 
The bulb was mounted at the lower end of an adjustable arm which could be moved 
up and down and fixed in a desired position by a set screw. A piece of fine ground 
glass ( Gl ) separated the two compartments of the box. The front part of the box 
had a door through which the glass container could be placed in position. The box 
was painted inside with black paint. Great care was exercised in avoiding the 
fluctuations in the voltage and in insuring uniformity in the intensity of light. It was 
noticed that there was a drop in voltage during the first 10 minutes following the 
