218 



THE CONDITIONS OF LIFE IN THE SEA 



[PART iir 



apparently of doubtful value. Silica in solution in water can 

 only be estimated by rendering the substance completely insoluble,, 

 and when there is such a large quantity of associated soluble salts- 

 as exists in sea water the estimation is very difficult. Raben^ 

 made a series of analyses of silica in water samples collected in 

 1902 — 4 by the Poseidon and appears to have obtained results 

 which represent the actual contents of the sea in this substance. 

 In each estimation 3 litres of sea water were evaporated to dryness 

 in a platinum basin. The mass of salts was then treated with 

 hydrochloric acid and again evaporated to dryness, and this process- 

 was repeated thrice so as to render the silica completely insoluble. 

 The salts obtained were then drenched with hvdrochloric acid, 

 dissolved in water and filtered, and the residue was ignited and 

 weighed. It was regarded as SiOo, but when it had a trace of 

 colour, indicating the presence of impurity, it was treated with 

 hydrofluoric acid so that the silica could be expelled as silicon 

 tetrafluoride. The residue, if any, w^as again weighed and the 

 weight deducted from the former value. In this way Raben 

 obtained values which are quoted in the following table : 



ATYioxint of SiOo, in "inilligrammes, in solution in 

 one litre of sea tvater. 



Date of collection 

 of samples 



No. of 

 samples 



SiO. 



Thus the proportion of soluble silica in the sea is very minute, 

 at no time exceeding 1^ parts per million of water. We find also 

 that particles of clay are also present in suspension in the shallower 



^ Loc. cit. See also Brandt, above. 



