56 SCIENCE PROGRESS 



Not only is the discrepancy not inexplicable, there is no dis- 

 crepancy to explain. So much did I take this for granted that, 

 in my last article on the subject, 1 I did not think it necessary to 

 consider the sea-salt method. I therefore take this opportunity 

 to repeat the arguments, and to remedy what is apparently a 

 deficiency. 



Prof. Joly's original paper 2 was based on the supposed facts 

 (i) that, as roughly estimated by Sir John Murray, of the solid 

 matter dissolved in river water which reaches the sea 3*47 

 per cent, is sodium ; (2) that nearly all this, hypothetical sodium is 

 obtained by erosion of the rocks ; (3) that when this hypothetical 

 sodium reaches the sea, none of it returns to the rocks. On this 

 supposition, dividing the amount of sodium in the sea by the 

 amount which reaches it each year, an estimate of geologic time 

 could be made. The objection is, briefly, that the three 

 supposed facts are merely supposed facts. No single one of 

 them is reliable. 



For convenience we will take the second point first. Of the 

 sodium which actually reaches the sea, a considerable pro- 

 portion is associated with chlorine. None of the sodium chloride 

 in the rivers can be attributed to erosion. This is so for two 

 reasons. In the first place, it is well known that the proportion 

 of chlorine in the rocks, igneous or sedimentary, is infinitesimal. 

 In the second place, the sources of the chlorine have been 

 thoroughly well determined. In the main, they are two, cyclic 

 salt, carried by the wind from the sea, and salt due to human 

 contamination. It has been found possible, particularly in New 

 York State, to eliminate the cyclic salt, the amount of which is a 

 function of the distance from the coast, and to show that the 

 residual chlorine in river water is a direct function of density of 

 population. Unless you take the sewage from town and country 

 districts directly out to sea, the salt in it inevitably reaches the 

 rivers. If you obtain an abnormally high chlorine ratio when 

 the sewage is supposed to be carried out to sea, the inference is 

 leakage. The source which would naturally occur to any one, 

 brine-springs, has been shown to be negligible. Even in New 



1 This journal, Oct. 1913. 



2 Trans. Royal Society, Dublin, vol. 7, pp. 26f. Sir John Murray's paper, Scottish 

 Geographical Magazine, 1887. The results are best tabulated for the purposes of 

 this discussion in the " Data of Geochemistry," U.S.A. Geological Survey Bulletin 

 No. 330, p. 88. 



