264 CHAMBERLIN— THE AGE OF THE EARTH. 



From this it appears that the mean per cent, of sodium in average 

 igneous rock is about thirty thnes as great as their content of chlorine. 

 This is a large difference, but it does not represent the full dis- 

 crepancy, for the chlorine in the ocean exceeds the sodium in about 

 the proportion of 1.8 to i. Taking this into account, the discrepancy- 

 rises to somewhat above ^o to K This is a formidable discrepancy. 

 How is it to be met on the assumption that the sodium in solution is 

 not reconverted into sodium solids, but remains in perpetual solution ? 

 The dilemma is not much relieved by reckoning in the sediments and 

 the ocean salts, for Clarke and Washington also give^^ the ratio of 

 sodium to chlorine when the atmosphere and hydrosphere are reck- 

 oned in with the outer ten miles of the lithosphere. The discrepancy, 

 corrected for actual oceanic proportions, is even then nearly 20 to i. 

 Quite naturally volcanoes have been thought to be a source of excess 

 of chlorine, but any contribution from the volcanoes is covered by 

 this inclusion of the whole atmosphere and hydrosphere in the aver- 

 age. Besides, the later studies of volcanic gases do not sustain the 

 earlier views that they contained a specially high content of chlorine.^^ 

 The observed differences between the sodium and the chlorine appear 

 to have grown mainly out of the normal processes of cyclic change 

 when these are viewed in their largest aspects. If the sodium returns 

 to the solid state in due (though lesser) proportion to the potassium, 

 magnesium, calcium, and chlorine, as these constituents are found 

 mixed in the solutes and the sediments, there is no necessary dis- 

 crepancy in these great differences. The discrepancy is constructive 

 and is imposed by the assumption that the sodium does not take its 

 proportional part in cyclic action. Under the alternative interpreta- 

 tion, the amounts of the several elements present in the ocean are 

 primarily functions of their own cyclic histories; their proportions 



sition of Igneous Rocks," Proc. Nat. Acad, of Sci., Vol. 8, No. 3 (May, 1922), 

 pp. 108-13. In the paper as read at the Symposium I used the then latest and 

 most authoritative figures, viz. : those of H. S. Washington, " The Chemistry 

 of the Earth's Crust," Jour. Franklin Inst., in which the sodium was given as 

 2.85 and the chlorine as 0.055. 



18 Ibid., p. 114. 



19 E. T. Allen, " Chemical Aspects of Vulcanism with a Collection of the 

 Analyses of Volcanic Gases." Papers from the Geophysical Laboratory, Car- 

 negie Institution of Washington, No. 440 (1922), pp. 1-80. 



