PELAGIC PLANT LIFE 361 



arise because the conditions of existence vary in closely 

 adjoining areas. Lohmann has found that at certain seasons 

 10 to 15 c.c. of sea- water amply suffice to give a representative 

 sample of the total plankton, but it is evident that only the 

 commonest organisms floating in the sea in any locality do 

 occur so densely and regularly that we can be sure of securing 

 them, or even of catching enough for ascertaining their com- 

 parative frequency, in a water-sample consisting of only a few 

 litres of water or less. The more scattered or mobile the 

 individuals are, the larger masses of water must we examine to 

 get a knowledge of the quantity present in any locality. 



It follows, therefore, that we must abandon all thought of a No universal 

 universal method. Fine silk nets give us complete collections "sUmatin^ 

 of the larger Ceratia and diatoms, but are of no use for the quantity of 

 smallest species, for which we are obliged to have recourse to p^^"^'°"- 

 more delicate methods of filtration, and to the centrifuge. The 

 larger forms, too, will be found in our silk nets in sufficient 

 quantities, if they are at all abundant, but where they are 

 scarcer than, say, fifty specimens to the litre, the centrifuge 

 cannot be depended on. Besides amongst these larger organisms 

 some species are so scanty that even a vertical haul with the 

 big net yields insufficient material, so we have been compelled 

 to adopt the special methods described in this volume. 



Various methods have been employed for estimating the 

 quantity of plankton on the basis of catches made. We can Determina- 

 allow the whole sample to sink to the bottom of a measuring tionsof 



, 1 . ^ . • 1 • 1 •! 1 volume and 



glass, and appraise its volume, or we can weigh it while the weight. 

 organisms are saturated with water or spirit, or we can weigh 

 the dry substance. Such determinations of volume and 

 weight give us our first rough idea of the variations in the 

 quantity of plankton, but there are many sources of error 

 which it is unnecessary to discuss here. The worst fault is that 

 measurements of this kind group into a whole the most diverse 

 values, such as plants and animals, producers and consumers, 

 one-celled organisms that are constantly reproducing themselves, 

 and multicellular animals with a longer duration of life, or, again, 

 organisms with slow and others with rapid metabolism. If we 

 want to know a litde about the conditions of development of 

 organisms, we must have a method of investigation that allows 

 us to trace the growth and retrogradation of each of the different 

 species by itself, and counting then becomes the only method counting 

 possible, as Hensen has continually asserted. Counting is a necessary. 

 method that requires much time, and also absolute accuracy in 



