Professor J. JoJij — Circulation of Salt. 505 



Now this was fully pointecl out in my previous paper ; it should 

 be perfectly well known to Mr. Ackroyd ; but we find still that 

 Mr. Ackroyd maintains that " the chemist's convention of taking 

 chlorine as a measure of sodium in rain- and rivei'-water is service- 

 able, and cannot involve more final error in connection with this 

 problem than that indicated by the ratio of these elements in sea- 

 water." 



As regards the first part of this statement, we have seen that " the 

 chemist's convention " would give hopelessly erroneous results if 

 applied to river analysis. As to the ratio of the chlorine and 

 sodium in sea-water, this has nothing to do with the matter beyond 

 indicating that as there is a large excess of chlorine over sodium in 

 the sea we may expect a similar excess to obtain in rain-water. 

 We may also observe that if " the chemist's convention " were 

 applied to sea-water an entirely erroneous result would be obtained 

 on the other side ; the sodium would be greatly overestimated. 



On the strength of this convention, however, Mr. Ackroyd again 

 quotes his analyses of the Aire (a small coastal stream), and, preferring 

 it to the mean analysis of the large rivers, again arrives at some 

 thousands of millions of years. That, following similar reasoning, 

 a stream could be found giving an infinite age to the earth, goes 

 without saying. Why will Mr. Ackroyd not have the 19 rivers? 

 The only objection I have heard urged against them (and it is one 

 of considerable weight) is that they are not numerous enough. 

 This is Professor Sollas' criticism. 



Coming now to the question of the origin of the salts of inland 

 seas, a question which Mr. Ackroyd has raised in connection with 

 the allowance proper for rain-borne sodium, I see in his last paper 

 in this Magazine that Mr. Ackroyd would explain the wide differences 

 in chemical composition of these waters by the effects arising from 

 concentration. The enormous amounts of precipitated salts required 

 by this hypothesis must, however, here be considered. Let it be 

 assumed, as he desires, that sea-water reaching closed lakes in rain- 

 water has concentrated until the balance between 20 per cent, of Mg Clj 

 and 5 per cent, of Na CI is attained. In sea-water there is but 0"38 

 per cent, of MgCla. There is, on the other hand, 2-73 per cent, of 

 NaCl. To reach the required percentage of MgCL a concentration 

 of 53 times the original is necessary. This involves a concentration 

 of the Na CI amounting to 145 per cent. Deducting the 5 per cent, 

 that remains in solution, but remembering that Na CI will by no 

 means be the only salt precipitated, also allowing the small imported 

 amount of Mg S O4 as a set-off against doloinitizing actions, we 

 finall}' arrive at the conclusion that the precipitated salts amount 

 to well over 1^ times the entire mass of the existing inland sea. 

 This is the least quantity we must look for in such a case, for it 

 is the amount thrown down if the concentration had only just 

 attained the existing state. 



I do not contend that the existence of such masses is out of com- 

 parison with known salt deposits ; but their absence in the particular 

 localities would constitute a fatal objection to supposing such extreme 



