586 Wisconsin Academy of Sciences , Arts, and Letters. 
the center of the bowl between conical discs which divide 
it into thin layers. The friction of these discs causes the 
water to rotate rapidly and again subjects it to very great 
centrifugal action which removes the last portion of the 
material that is obtained in the centrifugal process. This 
material is deposited on the under side of the discs and most 
of it passes down and out to the pocket at C. 
The centrifuged water passes to the center and is then 
forced upward and outward to chamber D from which it 
passes on into the filter compartment as shown by the ar¬ 
rows. The filter chamber is filled with a series of horizontal 
corrugated plates, nineteen in number, which possess per¬ 
forated retaining rims at their outer and inner margins. The 
filter papers are placed between the corrugated plates and 
the perforations in the plates are arranged so that the water 
passes through the filter papers in its course through this 
chamber. This removes the final portion of organisms ob¬ 
tained in this process. 
The material that is deposited on the conical discs and on 
the sides of the bowl is removed, together with the water 
remaining in the bowl at the end of the run, about 5.5 liters 
in all, and the whole is promptly evaporated to dryness at a 
temperature not exceeding 50° to 60° C. A little chloro¬ 
form is added to prevent fermentation during evaporation. 
The material is then weighed and analyses are made to 
determine the percentages of nitrogen, ether extract, pen¬ 
tosan, crude fiber, and ash. 
The nannoplankton shows marked variations in quantity 
during the year. The minimum amount is somewhat less 
than the quantity of net plankton and the maximum amount 
is from ten to fifteen times as great as that of the net plank¬ 
ton. 
The amount of organic material that is deposited on the 
filter paper is ascertained by determining the quantity of 
nitrogen in them in excess of that of the blank papers. 
COUNTING CELLS. 
Various methods have been used for the purpose of indi¬ 
cating the quantity of plankton in a given volume of water, 
but the counting method seems to be the most satisfactory 
one, since it gives an idea not only of the total number of 
