CHEMISTRY IX THE OCEANS 585 



earth's surface. We have found that a disintegration rate of one 

 per minute, readily measurable, necessitates the extraction of 

 roughly 100 g of opal or silicon dioxide. Such quantities are 

 available from siliceous sponges living in environments of the 

 mixed layer waters. For every 100 g of sponge-produced opal, the 

 equivalent of 100 tons of sea water must have had its silicon 

 depleted to zero by the activity of these animals. Deep-sea sponges 

 require the equivalent of 10 tons of sea water to obtain a corre- 

 sponding amount of silicon inasmuch as these waters contain on 

 the average 10 times more silicon. 



Let us now turn to the more general problem of chemical 

 processes in the oceans. Chemical reactions take place at phase 

 discontinuities, i.e., the atmosphere-hydrosphere, biosphere-hydro- 

 sphere, and sediment-hydrosphere interfaces. We shall focus our 

 attention mainly on the reactions of sea water with the solid 

 components of the sea floor. 



We can reasonably expect that most inorganic species correspond 

 closely to those predicted from complete equilibration between all 

 reacting substances. Such thermodynamically unstable species as 

 iodide and manganous ions probably exist because the waters that 

 contain them have been inaccessible to surfaces at which reactions 

 to stable substances might occur. We shall return to this concept 

 in considerations of the formation of ferromanganese minerals on 

 the sea floor. 



Insight into the relative reactivities of elements in the marine 

 hydrosphere has been obtained from considerations of the average 

 time an element spends in the oceans. An extremely simple model 

 of the ocean is used in which it is assumed that the presently 

 observed chemical composition represents a steady state system 

 in which the amount of material introduced per unit time is com- 

 pensated by an equal amount deposited as sediments. A further 

 assumption is that there is a complete mixing of materials intro- 

 duced into the oceans in times that are short with respect to the 

 residence times. We can then define the residence time of an ele- 

 ment as the total amount of the element in sea water divided by 

 the amount of the given element introduced by the rivers or 

 precipitated to the sediments per unit time. 



About seven years ago Barth (1952), on the basis of river 



