METHODS OF PLANKTON RESEARCH, 523 



By lowering the tube it fills gradually with water out 

 of the various depths through which the lower end sinks, 

 so that finally it contains a water column, consisting of 

 water from all depths between the surface and the lowest 

 point reached. When the tube is again slowly raised 

 through this column to the surface, it is once more filled 

 by water from each layer. Thus by repeatedly lowering 

 and raising, whilst the pump is worked, any quantity of 

 water may be obtained, representing a vertical column 

 whose height is that from the lowest point reached by 

 the tube up to the surface and whose other dimensions 

 can be reckoned directly from the volume of water 

 collected. 



The water in the sea will naturally rise in the pump 

 tube of its own accord until it attains the same level as 

 the surface. It is only necessary, therefore, to use the 

 pump to lift the water from the surface of the sea into 

 the boat, and a small pump is accordingly quite sufficient. 

 It is even possible in a boat with a deep bottom, where 

 the upper end of the tube can be placed lower than the 

 surface of the sea, to siphon up the water, but usually, 

 owing to the motion of the boat, this method is not 

 successful. The whole length of the tube should be 

 fastened to a rope which will bear its weight. If currents 

 are present, the rope and pipe must be kept vertical by 

 means of a sinker. 



A very simple and cheap arrangement was con- 

 structed and used by Lohmann in the Mediterranean (13), 

 so that the simple turning of a windlass both worked 

 the pump and pulled up the tube. Thus the rate of 

 pumping and the pulling up of the tube were always in 

 the same relation, however quickly the windlass was 

 turned. Moreover, the direction in which the windlass 

 was turned had no effect on the working of the pump, 



