19 



tips of the ell)ows before it returns to the axis for discharge at 

 the lower end. This machine is geared to give, with power, 

 8,000 revohitions per minute. When fitted with cranks for two 

 men, four to five thousand revohitions can he obtained. This 

 apparatus was tested with water from the river at a time when 

 it was full of water bloom, — formed principally of (drteria, — 

 and also with water from the lakes in varying kinds and amounts 

 of plankton. It proved to be more effective in the removal of 

 the plankton than any method previously tried, but the opera- 

 tion of the machine by hand was extremely laborious, and the 

 precipitation of the plankton was very slow. Furthermore, a 

 variable and oftentimes considerable amount of the plankton — 

 especially that found in the water-bloom — is at times lighter 

 than the water, and thus cannot be removed by centrifugal force 

 with the heavier constituents. 



In November, 1897, a Berkefeld army filter (system 

 Bruckner) was added to the plankton equipment. It is very 

 efficient in removing all the solid matter from the w^ater, and its 

 operation with ordinary samples is quite rax)id. It consists of 

 a force-pump and a cylinder of diatomaceous earth, upon which 

 the plankton and silt contained in the water are collected. This 

 is removed bj' washing with a l>rush, but in the process a 

 part of the suljstance of the cylinder is brushed off. This 

 debris is added to the silt of the water and renders subsequent 

 microscopic examination more difficult. The brushing is also 

 disastrous to some of the more delicate organisms, Init leaves 

 l)y far the greater part of the minute forms which escape the 

 silt intact and in suitable condition for enumeration. 



During the past two years some progress has been made 

 with an examination, measurement, and enumeration of the 

 plankton of the regular series, though much of the time has 

 l)een given to the preparation of plankton apparatus and the 

 improvement of the method. In this work the examination of 

 the test collections l)y the enumeration method has been par- 

 ticularly time-consuming. The work of enumeration has been 

 facilitated by the use of a set of six counting machines, which 

 eiuible the observer to kee]) a record of six different species at 

 once without the mental effort of carrying the count in the mind. 

 An extended amount of this work remains to \)v done before we 



