354 
BULLETIN OF THE BUREAU OF FISHERIES. 
beneath the second stick. The observation required two men, one on the bow 
to start the bottle some feet above the first mark and the other on the stern to 
hold the stop watch. The difference in data obtained by this method and by 
Price’s current meter did not exceed 8 per cent. The float method can be used 
only if the position of the boat is exactly parallel to the direction of the current. 
If the current is slow and the wind blows across the river, the observation can not 
be made. During calm days on Lake Pepin the author often observed the drift 
of the plankton algse alongside the boat when the current meter indicated no move- 
ment; evidently the drift was not strong enough to move the cups of the wheel. 
The Price current meter is more suitable for river observation than for lake obser- 
vation. It seems that the instrument works only when the velocity of current 
is more than 0.2 foot per second. 
The transparency of the water was measured with a round white disk, 25 cm. 
in diameter, attached to a long graduated metal rod. The results of the observa- 
tions are expressed in centimeters, representing the depth at which the plate 
disappears from view. 
Pump ■ — The greater part of the plankton collections was made with a pump. 
The water was pumped from different depths, usually at intervals of 5 feet, and 
then filtered through the plankton net made of No. 20 bolting silk. An iron 
double-acting oscillating force pump, No. 0 (manufactured by the Goulds Co., 
Seneca Falls, N. Y.), was used with a rubber hose 1 inch in diameter. The pump 
and the graduated tank of 50 liters capacity were fastened on the bow. 
The galvanized-iron tank was cylindrical in form, with a long neck on the top 
and a pipe close to the bottom. In the neck was a graduated glass window. The 
total height of the tank was 28J inches, the diameter of the bottom being 13f 
inches, the diameter of the neck 6 inches and its height 7-| inches, and the outlet 
pipe 10 inches long and 1 inch in diameter. The tank had a capacity of about 
13 gallons, the mark on the neck corresponding to 50 liters having been made 
after many careful measurements. The same kind of tank was used in previous 
work on LakeKossino near Moscow and has been already described (Galtsoff, 1914). 
The hose was suspended on a line. At the river stations where the current 
was swift a weight of about 30 or more pounds was attached to the end to keep 
the line straight. It took about three minutes to fill the tank; the same amount 
of time was required to filter the water through the plankton net. Special care 
was taken to reduce the pressure on the filtering surface, the best method being 
to keep about three-fourths of the net in the water. The flow of water when 
emptying the tank was also regulated. The same net, Apstein’s small vertical net, 
13| inches long, with an upper ring 4f inches in diameter, was used for the vertical 
plankton hauls. All collections were preserved in 3 per cent formalin. 
DETERMINATION OF PLANKTON. 
Two methods were used for volumetric determination of plankton: (a) settling 
in graduated tubes during 24 hours; (b) centrifuging for 2 minutes at the rate of 
1,000 revolutions per minute. The first method has been strongly criticized 
