Density of Flagellates and Myxophyceae 223 



erally less than 50 per cent of the preceding organisms, we can 

 only give in few words an outline of their distribution: 



Syracosphaeraceae, which have 15 to 40 species in the Medi- 

 terranean euphotic layer, have only two to nine species below 

 it (for instance: Thoracosphaem heimii Lohm., Sijracosphaera 

 profunda Bernard, Corisphaera fagei Bernard . . . ) . They consist 

 generally of one-third of the number of coccolithaceae (Cocco- 

 lithits wallichi is, by far, less common than C. fragilis). 



Dinoflagellates are represented chiefly by small Gymnodin- 

 ians, not very abundant (one-tenth of the naked flagellates). 

 Here and there, Exuviella (cf. marina Cienk.) is more common 

 and prevails in the oriental MediteiTanean (sometimes 300 to 

 4,000 cells/ml at 500 and 800 m depth). It is not certain that 

 the deep form is the same Exuviella as that at surface. The 

 electron microscope is required to show the details of the shells 

 for separation. 



Nostoc is rare when water is polluted by suspended mud. 

 In very clean water, far from the bottom and from the coasts, 

 we can have thousands of Nostoc per ml; as in the Indian Ocean. 



Off western Algeria (Fig. 1) the saline oriental current, 

 running between 250 and 500 m depth, is sometimes very rich. 

 In the oriental basin, and at station six (between Sardinia and 

 Sicilia) deeper fertilities seem to be caused by local currents, 

 estimated in the following pages. 



In the Indian Ocean, 200 m is often the richest layer, and 

 1,000 m has more cells than 400 to 600 m, and much more Nostoc. 



Table 3 gives the mean deep numbers of organisms, com- 

 pared with the mean numbers in the euphotic zone, for tlie 

 eight stations described. 



The aphotic zone is very rich, except for Nostoc. The densi- 

 ties of flagellates are often greater than at the surface, and repre- 

 sent at least two-thirds of the euphotic densities. In poorer 

 stations (not included) the values are at least one-third of the 

 euphotic value. 



HYPOTHESIS ON THE CAUSES OF DEEP SEA FERTILITY 



It is actually impossible to be certain of the real factors in 

 the deep sea. Heterotrophy is the only possible process of metab- 



