Introduction.
In work on the migrations of planktonic organisms, one of the chief difficulties is counting the catches. Information about the depths a t which the animals are to be found a t different times during the day is obtained by catching samples a t different depths b y some sort of closing net. The catches are then preserved in dilute formalin solution, and later the numbers of the different species in each catch counted. The numbers caught a t each depth can obviously be varied by changing the size of the net, but there are several reasons why the net should not be too small. Firstly it is by no means certain that there is a uniform horizontal distribution of the plankton ( G a r d i n e r , 1926), and therefore the net should be as large as is practicable, so that as wide an area as possible is fished. Secondly, if the net is small, random variations in the numbers due to imperfections in the technique of catching can assume significant proportions. There is, in fact, every reason for having the net as large as possible, except t h a t great numbers of animals of different species are tedious t o count, especially when some of the species resemble each other quite closely. A machine which makes very considerably easier the counting of large numbers of small animals has been made and used for some time a t the Laboratory of the Freshwater Biological Association. The principle of this machine has now proved to be helpful to other workers ( W o r t h i n g t o n , 1936), so in this paper is published a short description of the details of the apparatus.