RECENT FRESHWATER INVESTIGATIONS 305 



The vertical net, planned by Hensen and first used by 

 Apstein in fresh water, is described by the latter (96). Some 

 improvements in detail were made by Reighard (94a) and by 

 Kofoid (97) while the latter adapts the vertical net by an in- 

 genious arrangement to oblique hauls in shallow water. In a 

 later paper (98a) he gives a careful account of the best method 

 for the construction of the vertical net. Vertical closable nets 

 worked by sliding weights are described by Birge (97a) and 

 Marsh (97), and a horizontal net which can be opened and 

 closed by a cord by Lakowitz (96). The former are prompt 

 and accurate in action and may be used at any depth, having 

 been tested up to 130 m., while the latter is apparently cum- 

 bersome, if not uncertain in action, and on the authority of the 

 author can be made use of only up to 20 m. in depth. As to 

 the material, fine silk gauze, of which such nets are constructed, 

 Frenzel (97a) makes some criticisms regarding its inconstancy; 

 since while a single haul does not usually close the pores to a 

 noticeable extent there are exceptions, and furthermore con- 

 tinued use is sure to modify its filtering capacity, thereby falsi- 

 fying all calculations. The pores are closed by accumulated 

 detritus, not by diatoms or other small planktonts. The net 

 should be vigorously washed and wrung out each time to clean 

 out the pores. Hensen (97) questions these statements regarding 

 clogging and objects to such drastic treatment in cleaning the 

 net. Recently Kofoid (97a) has attacked the accuracy of 

 results obtained by the vertical net on the opposite basis: that 

 it allows the escape of too many organisms since "the silk 

 retains from 5 per cent, to less than 0.1 per cent of the total 

 number of organisms present excluding bacteria, as contrasted 

 with the catch of the Berkefeld filter;" volumetrically the catch 

 equals from one-half to only one forty-fifth of the amount 

 actually present in the water. Reighard (98) calls attention to 

 the fact that the larger size of the nets used by some observers 

 makes clogging a less important factor than in studying silt- 

 laden waters with a small net. Shrinkage being largely if not 

 entirely eliminated by previous treatment of the net, leakage is 

 20 



