308 THE ACCURACY OF THE OBSERVATIONS [APP. IV 



Now it is diflScult to separate the error due to the variation 

 in uniformity from the other errors. Absolute uniformity of the 

 plankton does not of course exist in the sea. Hensen's postulate is 

 usually misunderstood. What the Kiel school of planktologists main- 

 tain is that wherever the physical conditions {Lehenshedingungen) are 

 uniform there is also an approximate uniformity in the nature and 

 distribution of the plankton. So we find, as Cleve's charts shew, that a 

 large sea area, like the North Atlantic, or the North Sea, is to be divided 

 into a number of sub-areas, each of which is characterised by a peculiar 

 plankton-facies. The error due to experiment can only be separated 

 from those due to irregularity if we make test hauls in an area where 

 the physical conditions are as similar as possible over the entire area. 

 Now such pliysical uniformity does not exist over even comparatively 

 small areas of inshore seas. In the North Sea, as the charts published 

 in the Bulletin des Resultats of the International Fishery Investigations 

 Bureau shew, there are very great diflferences in the salinity and 

 temperature of the water, from place to place, and from time to time ; 

 and with these differences there must be differences in the plankton. 

 In the Irish Sea we have a certain homogeneity in the physical 

 conditions, but this homogeneity is due to the mechanical mixing of 

 the water coming from outside the area, with that coming from the 

 land. It is not a uniformity due to the common origin of the water. 

 One would therefore expect to find a certain degree of irregularity of 

 the plankton due to the fact that some species which were adapted to 

 live well in either of the water sources are not adapted for life in the 

 mixture of both, and therefore that some of them will die out. Still 

 less would we expect to find an inshore area adapted for such test 

 hauls. In the Sargasso Sea, however, we have a vast mass of water 

 occupying the centre of the Gulf Stream cyclonic circulation. It is 

 what Haeckel called a Halistatic area, and in it we should expect a 

 uniform plankton. Over 2000 miles of ocean the temperature during 

 the cruise varied only from 26° -2 to 25° '4 C, and the salinity from 

 36 '2 to 37. Schiitt estimated that the total mean divergence of the 

 average catch of 32 "/q was due partly to the lack of uniformity, and 

 partly to errors of experiment. The latter he estimated at 20 Yo- 

 Therefore the mean variation of the plankton in the Sargasso Sea in 

 August 1888 was about 16 7o more or less than the mean. 



