METHODS AND APPARATUS 9 



through a plankton net for a catch. In addition to yielding a definite 

 quantity of water this method also enables one to study the vertical dis- 

 tribution of the organisms. The chief objection to it is that the cur- 

 rents produced in the water at the intake end of the hose tend to drive 

 away the more active planktonts that are negatively rheotropic. Sev- 

 eral sets of experiments made with a plankton trap and the power 

 pumps showed an advantage in favor of the trap of 16 per cent in 

 Diaptomus and of 15 per cent in the Daphnias ; on the other hand, there 

 was a numerical advantage in favor of the power pumps amounting 

 to 12 per cent in Cyclops and 27 per cent in the nauplii. While the 

 latter forms are smaller than the former, they are usually more numer- 

 ous so that the gains and losses probably just about balance each other 

 in so far as the weight of the net plankton is concerned. In addition 

 to this objection the pump method is not practicable beyond a depth of 

 about 75 meters. 



At a comparatively early stage in the development of quantitative 

 plankton studies it was found that even the finest meshed nets did not 

 retain all of the organisms, but the extent of this loss was not fully 

 appreciated until Lohmann^ completed extensive investigations along 

 this line and published his results in 1908. In studies on marine plank- 

 ton he was able to show, by the use of the centrifuge and various filtra- 

 tion methods, that many organisms which were present in enormous 

 numbers were not represented at all in the net catches. It has since 

 been found that the same is true of the fresh-water plankton as well ; 

 in fact, results presented in this report show that the organic matter 

 of the material which is retained by the net is frequently only a small 

 percentage of the organic matter in the material which readily passes 

 through the meshes of the net. A very important problem in this in- 

 vestigation, then, has been to ascertain the relation between the quan- 

 tity of plankton material that is retained by the net and that which is 

 lost through its meshes. 



In addition to these gravimetric data, some chemical analyses of the 

 material were desired. For the latter it was necessary to obtain rather 

 large samples of dry material, at least five grams whenever such 

 amounts could be secured. It was soon found that this large a sample 

 of net plankton made it necessary to strain fairly large amounts of 

 water, such as several thousand liters in some instances, and the ap- 

 paratus employed had to meet these requirements. 



Pumps and Net 



For the net plankton, in fact, the quantity of water that was strained 

 for each sample varied from a minimum of 2,000 liters to a maximum 



2 Wissenschaftliche Meeresuntcrsuchungen. Bd. 10, 1908, pp. 131-370. Kiel. 



