REGENERATION OF PHOSPHATE AND SILICATE 109 



showing that regeneration of silicate is taking place in situ mainly as a result of a rain 

 of dead diatoms and faecal pellets from the surface. Thus the oxygen content at 

 the surface is 7-86 c.c. /litre. Below the surface the oxygen content falls at least as 

 far as 60 m. with a probable minimum at 80 m. The silicate content is 2500 mg. at 

 10 m., below which it increases to a maximum for the Antarctic surface layer of 

 3400 mg. at 80 m. It will be noticed that at 150 and 200 m., in the bottom of the 

 Antarctic surface layer, the oxygen content has increased and the silicate decreased. 

 The definite increase in oxygen content at 150 and 200 m., together with the 

 lowered silicate content would seem to indicate that the water in the layer between 

 150 and 200 m. represents water which has sunk to these depths from the surface 

 sometime previously. It must be admitted, however, that the phosphate content 

 of 128 mg. at these depths does not agree with this conclusion unless silicate is 

 selectively absorbed by the phytoplankton before the phosphate. Silicate is obviously 

 being regenerated between 20 and 100 m. with a maximum at 80 m., this regeneration 

 being accompanied by a fall in oxygen content as might be expected. When the 

 observations were made this regeneration had not progressed so far at depths below 

 100 m., as the results at 150 and 200 m. undoubtedly show. In this respect it is signi- 

 ficant that the euphausian faecal pellets were chiefly in the upper part of the surface 

 layer. 



Thus the results at St. 1543 afford an excellent example of regeneration of silicate 

 in situ. Although the actual methods of decomposition and regeneration are not known, 

 the fact that a large number of siliceous skeletons had obviously passed through the 

 digestive system of the zooplankton tends to suggest that perhaps the siliceous skeletons 

 are excreted in such a condition that re-solution is facilitated. Obviously a knowledge 

 of the pH of the digestive system of the zooplankton would be of service in this con- 

 nection. It is possible that silicate from the excreted faecal pellets of the zooplankton is 

 directly dissolved in the reasonably high pH of the upper part of the surface water and 

 that this may explain the very rapid regeneration of silicate that appears to take place. 



It has been noted (Earland, 1936, p. 8) that in bottom samples from the Weddell Sea, 

 diatoms are almost entirely absent whereas farther north diatom skeletons form the 

 diatom ooze. Of course the concentration of diatoms in the Weddell Sea is less 

 than that in the northern region. It is possible that the diatom ooze is formed by 

 those diatoms whose mortality is due to some cause other than their being eaten by 

 zooplankton, i.e. the belt of diatom ooze is found south of or close to the Antarctic 

 convergence, where a sudden change of temperature occurs and where the sinking 

 surface water may cause a high mortality of diatoms which sink and are not redissolved 

 as fast as they are deposited. It is also possible that the operation of a limiting factor 

 in the densely populated surface water in the northern region of the Antarctic zone 

 may be responsible for a high mortality and the formation of the belt of diatom ooze. 



The suggestion has been made earlier in this report that when the zooplankton con- 

 sumes the phytoplankton part of the phosphate content of the latter is incorporated in 

 the former, whilst the siliceous skeletons of the diatoms and other undigested parts of 



