84 FRESH-WATER BIOLOGY 



The centrifuge is the most convenient as well as the most ef- 

 ficient instrument for obtaining the nannoplankton. A rather 

 high speed machine is best, one which makes 2500 or more revo- 

 lutions per minute, and the electrically driven type is most satis- 

 factory. For most fresh-water organisms the sedimentation is 

 complete in five to eight minutes at this speed, but occasionally 

 for some forms a second centrifuging is necessary. In bodies of 

 fresh water the nannoplankton is usually so abundant that only a 

 small quantity of water, not more than 15 cc, is required for a 

 sample. Thus the standard makes of centrifuges will serve for 

 such investigations. The glass tube which holds the sample of water 

 should be well tapered at the bottom. This form concentrates the 

 material on a small area from which it can be removed more con- 

 veniently as well as more completely. The material is taken up 

 together with one cubic centimeter of water in a long pipette and 

 is then transferred to a Sedgwick-Rafter counting cell. This cell 

 and its use are fully described by Whipple. Sometimes it is de- 

 sirable to centrifuge 50 or even 100 cc. in order to study the rarer 

 forms. For enumeration studies a combination of the direct count- 

 ing and the centrifuge methods gives the most satisfactory results. 



Whenever possible, living material should be used for the count- 

 ing. The samples may be preserved in formaldehyde neutrahzed 

 with sodium carbonate and then centrifuged at a later time, but 

 some of the monads are recognized with difficulty after preserva- 

 tion. Most of the flagellates do not move rapidly enough to 

 offer any serious difficulty in counting them alive but the ciHates 

 do. When the latter are present, it is best to make a special count 

 for them. They are readily killed by placing a drop of iodine 

 solution in the corner of the counting cell before the sample is in- 

 troduced. 



Material for a study of the dry weight as well as the organic 

 matter of the nannoplankton may be obtained either by filtering 

 a relatively small sample of water through a coarse-grained alundum 

 cone or by passing a large sample of water through a power centri- 

 fuge that acts continuously. In the former process the sample of 

 water, from one to five Kters, is filtered through the cone and the 

 material and cone are then thoroughly dried in an oven. The 



