43

deficiency of soda in sedimentary rocks is so characteristic of secondary rocks that it may with some safety be applied to discriminate between the two classes of substances in cases where petrological distinctions of other kinds break down.

To what is this so marked deficiency of soda to be ascribed?  It is a result of the extreme solubility of the salts of sodium in water.  This has not only rendered its deposition by evaporation a relatively rare and unimportant incident of geological history, but also has protected it from abstraction from the ocean by organic agencies.  The element sodium has, in fact, accumulated in the ocean during the whole of geological time.

We can use the facts associated with the accumulation of sodium salts in the ocean as a means of obtaining additional support to the view, that the processes of solvent denudation are responsible for the saltness of the ocean.  The new evidence may be stated as follows:  Estimates of the amounts of sedimentary rock on the continents have repeatedly been made.  It is true that these estimates are no more than approximations.  But they undoubtedly are approximations, and as such may legitimately be used in our argument; more especially as final agreement tends to check and to support the several estimates which enter into them.

The most recent and probable estimates of the sediments on the land assign an average thickness of one mile of

44

secondary rocks over the land area of the world.  To this some increase must be made to allow for similar materials concealed in the ocean, principally around the continental margins.  If we add 10 per cent. and assign a specific gravity of 2.5 we get as the mass of the sediments 64 x 1016 tonnes.  But as this is about 67 per cent. of the parent igneous rock—­i.e. the average igneous rock from which the sediments are derived—­we conclude that the primary denuded rock amounted to a mass of about 95 x 1016 tonnes.

Now from the mean chemical composition of the secondary rocks we calculate that the mass of sediments as above determined contains 0.72 x1016 tonnes of the sodium oxide, Na2O.  If to this amount we add the quantity of sodium oxide which must have been given to the ocean in order to account for the sodium salts contained therein, we arrive at a total quantity of oxide of sodium which must be that possessed by the primary rock before denudation began its work upon it.  The mass of the ocean being well ascertained, we easily calculate that the sodium in the ocean converted to sodium oxide amounts to 2.1 x 1016 tonnes.  Hence between the estimated sediments and the waters of the ocean we can account for 2.82 x 1016 tonnes of soda.  When now we put this quantity back into the estimated mass of primary rock we find that it assigns to the primary rock a soda percentage of 3.0.  On the average analysis given above this should be 3.41 per cent.  The agreement,