10 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY. 



particles organic and otherwise. Kofoid finds that from 

 shrinkage alone the area of the openings in a square centi- 

 metre may decrease 50 per cent. 



A difficulty of quite another kind that has been 

 pointed out by various critics is that small as the meshes 

 of the silk are, many of the minuter organisms of the 

 plankton pass through and are lost. Kofoid has 

 determined the percentage of loss from this source for 

 certain organisms, e.g., he finds that " of Codonella as 

 many as 21 individuals may escape to one retained." 



But these difficulties, serious as they may seem, and 

 as they are regarded by some, may still be to a 

 considerable extent overcome by taking precautions and 

 by applying corrections to the result. They are trouble- 

 some difficulties, but they are not, to my mind, fatal. 

 They only make the work more difficult, more expensive, 

 and more slow. They may still, after all corrections have 

 been applied, allow of a fair approximation to a correct 

 quantitative estimate of the organisms present in the 

 particular sample of water through which the net has been 

 pulled. But can we safely apply the results so obtained 

 to any further purpose ? This brings me to my main 

 difficulty — a difficulty I have felt for many years, but 

 which was forcibly brought before me in my observations 

 at Port Erin during the present summer. It is a funda- 

 mental assumption (principle I think it is sometimes 

 called) in the Hensen method, that the organisms are 

 distributed with such uniformity over wide areas of the 

 sea that, after taking some samples per square metre, it 

 is considered justifiable to multiply up by the number of 

 square metres in a fishing district, or even in the wide 

 extent of the Baltic or the North Sea. Here we have 

 evidently a most fundamental point upon which the 

 stability of the entire superstructure depends. If your 



