12 SMITHSONIA^^ MISCELLANEOUS COLLECTIONS VOL. 56 



seems to me much more probable and better fitted to give a iirst approxi- 

 mation to tlie truth/ 



Oceanic sodium is at least chiefly derived from lime-soda feldspars 

 which as essential constituents are practically confined to Archean and 

 later igneous rocks. The original surface of the earth must have con- 

 sisted of such rocks to the exclusion of all others, while at the present day 

 the greater part of the land area is covered with sedimentaries. Now the 

 rate of decomposition of rocks is chiefly dependent on exposure. Even in 

 areas of ancient feldspathic massives decomposition does not seem to pene- 

 trate to great depths. Thus in the southern Appalachians great areas of 

 gneiss and allied rocks are now covered with a blanket of saprolite (rotten 

 rock in place) which is in many localities 50 feet in thickness, but at all 

 the points where I have observed it, less than 100 feet thick. Immediately 

 below the saprolite blanket there is incipient decomposition and the feld- 

 spars are milky, but not many yards lower down the feldspars are 

 characteristically translucent and the rock bluish in tint. A layer of decom- 

 position products 100 feet thick seems to arrest decay. Corresponding state- 

 ments are true of tertiary volcanics, excepting where the decomposition is 

 solfataric. On the other hand mesozoic and paleozoic massive rocks deeply 

 buried under sediments are in many cases found to be very free from 

 decomposition. In short, buried massives decompose at a rate which is 

 scarcely sensible. 



It is quite imaginable that in the far distant future all the massive 

 rocks might be thoroughly decomposed down to sea level or a trifle below. 

 The continents would then be exclusively detrital. Under such conditions 

 there could be no further important additions to the sodium content of the 

 ocean, for there would then be no leaching; while mere diffusion to any 

 considerable distance is too inordinately slow to produce any noteworthy 

 result even in millions of years. 



Thus in the distant past there must have been a time when a far 

 greater mass of massive rock was decomposed each year than now decays 

 in the same period; and a limit to this process can also be foreseen. The 

 total area of exposed massives has surely diminished and will continue 

 to diminish. Climate and temperature may perhaps have been in the 

 past much what they are to-day; the rate of chemical denudation per unit 

 area may not have changed considerably, but the most rigid uniformitarian 

 would not maintain that the total area of exposed massive rocks has been 

 constant. The inference seems unavoidable that sodium accumulation is 

 an asymptotic process which progressed more rapidly (though possibly not 

 with greater intensity) in the distant past and will come substantially to 

 an end when a certain very finite layer of surface material has been ex- 



» Vide Science, voL 31, 1910, p. 509. 



i 



