Professor J. Joly — Salt and Geological Time. 347 



otherwise allocated than to the sodium, or is, in short, in excess 

 of the CI equivalent of the sodium present. 



Now it is certainly a fair assumption that if rain receives its salts 

 from sea spray a similar excess of CI ions over Na ions will obtain 

 in rain-water. We must at least conclude (in deference to the 

 current belief that the proportions of salts in rain-water and sea- 

 water are not generally quite in accord) that the sodium equivalent 

 of the chlorine found in rain-water constitutes an excessive estimate 

 of the former. Pierre's results, to which I have already referred, 

 fully confirm this statement. He finds, in kilogrammes received 

 per hectare per annum, quantities of sodium chloride and other 

 chlorides, as well as of sodium sulphate, which afford 29*7 kilos 

 of chlorine and 17-72 kilos of sodium. Ascribing to this amount 

 of sodium its equivalent of chlorine, we have still a balance of 

 over 8 per cent, of the chlorine.' 



We now turn to what we know of the chemical constitution 

 of river salts. I take Sir J. Murray's mean analyses of nineteen chief 

 rivers of the world. Here we find the striking fact that there 

 is a large excess of sodium over chlorine : or the conditions of the 

 sea are reversed. We find, in fact, 157 X 10'' tons of sodium and 

 84 X 10^ tons of chlorine carried to the ocean per annum, and as 

 the combining weights of these elements are to one another as 

 23 : 35 it will appear that considerably the larger part of the sodium 

 of rivers must exist otherwise combined than with chlorine, were the 

 ionizing conditions removed. Let us now go so far as to assume 

 that all the chlorine in rivers is derived from rain, and that this 

 brings into the rivers its full equivalent of sodium, and we can 

 obviously calculate the maximum possible effect upon the age of 

 the Earth arising from the circulation of salt. (This can fall short 

 of the maximum only on the assumption of a selective retention of 

 chlorine in soils, for which I know of no evidence.) In this method 

 of treating the quantities at our disposal we leave, in short, the 

 whole of the sodium equivalent of the chlorine of rivers out of 

 account in deducing the age of the Earth. The result of the new 

 calculation is that the previous estimate of 96 millions of years 

 rises to under 148 millions of years. This result is, however, over 

 the maximum deducible even from the foregoing assumptions, for 

 as we diminish our estimate of the chlorine derived by solvent 

 denudation during geological time we increase the estimate we 

 necessarily make for original sodium in the ocean derived by 

 a primeval acid denudation effected by free H CI (for we must 



