3i° DISCOVERY REPORTS 



the experiments with series of consecutive nets we have shown that the patchiness of 

 some of these macroplankton organisms, particularly Euphausians, is such that one haul 

 of half a mile may often miss a patch altogether, whereas two consecutive hauls (or a 

 tow of one mile) will stand a good chance of sampling part of such a patch. For the 

 results of these later surveys to be comparable with our own they should contain 4! 

 times as many stations in the same area ; actually they contain about the same number. 

 It is not our intention here to be critical of the change in procedure adopted, a change 

 which was designed to enable extensive areas to be charted more rapidly than by the 

 old method, but to point out that when making comparisons between later surveys and 

 the one now under review we must not be surprised if we do not find a high degree of 

 confirmation. It is noteworthy that the failure of confirmation is most marked in the 

 Euphausians, which we have shown to be very patchy in their distribution. 



The correlations between the zooplankton taken in the different series of nets and 

 the phosphate values appear to lend strong support to the conception that it is the 

 phytoplankton which is influencing the distribution of the majority of the plankton 

 animals rather than that it is these animals which by grazing are controlling the phyto- 

 plankton. If the latter is true, how is it that so many different animals have the same 

 type of correlation with the phosphate values? The species which occur in relatively 

 small numbers could only have a negligible effect upon the phytoplankton compared 

 with the abundant species, yet these rarer species show the same type of correlation as 

 do those which are very abundant. For the majority of species it appears that their 

 distribution must either be governed by the phytoplankton or by some other factor 

 unknown, which brings them together so that they have the appearance of combining 

 together to control the phytoplankton. It is possible that one outstanding species, or 

 perhaps a few such species together, may be controlling the phytoplankton, and that the 

 pattern of phytoplankton so produced — areas of high and low concentrations — may be 

 in turn influencing the distribution of the majority of less abundant animal species. 

 From a re-examination of the data made in the light of Harvey's (1934) investigations 

 and to be discussed presently, it does actually appear likely that Euphausia superba may 

 be a key zooplankton organism which is at any rate in part controlling the phytoplankton, 

 and that the other organisms are in turn having their distribution adjusted by that of 

 the phytoplankton. 



Another point must be mentioned here. All the animals which are correlated with 

 the phosphate values are reacting to phytoplankton production generally; we do not 

 know that they are feeding indiscriminately upon all the phytoplankton equally — it is 

 most unlikely — and some of the forms are certainly carnivorous. We should expect some 

 herbivorous animals to govern the abundance of only some species of the phytoplankton, 

 and other herbivorous animals to govern only other species. Actually we see the same 

 correlations between phosphate values and the same series of animals applying in 

 regions of different phytoplankton floras, as shown in Table LVI. For this reason, 

 granted that it is possible or even likely that Euphausia superba may be governing the 

 phytoplankton in some regions, it is likely that this Euphausian will not be governing it 



