W. Ackroyd — The Circulation of Salt. 447 



To return now to the Dead Sea. It will be at once recognized 

 that the large percentage of magnesium chloride present, arising 

 from ages of concentration, is favourable to precipitation of common 

 salt, which in the southern parts of the lake forms quite a paste 

 (Tristram). The ratio of chlorine to bromine in the surface 

 ■waters of the Dead Sea and of the Mediterranean are nearly alike 

 (100 : 2-1), and the only divergence one looks for is in the increase 

 of the bromine figure for the Dead Sea as its sodium chloride is 

 precipitated, and this one finds to be the case. It is also to be 

 noted here that all attempts to find traces of bromine in the springs 

 and rivers of Palestine have hitherto failed (Watts' Dictionary of 

 Chemistry, vol. v, p. 183). All these points are compatible with 

 the Dead Sea having derived the greater part of its salts from the 

 Mediterranean. As one recognizes, however, that solvent denudation 

 must play an important part in adding to the soluble contents of 

 inland lakes, it is to its extensive limestone gathei'ing-ground that 

 the Dead Sea must owe its calcareous character, and the contrast 

 in this respect with the Mediterranean is increased by the latter 

 possessing and the former being without a lime-secreting fauna. 



On the Proportion of Chlorides supplied by Solvent Denudation. — 

 It was of interest and of importance to know approximately what in 

 a river water was due to solvent denudation and what to atmospheric 

 transportation. To find the proportion of chlorides due to solvent 

 denudation the following line of reasoning suggested itself to me. 

 A sample of Malhara Cove water in Craven had a hardness of 10°, 

 i.e. it contained 10 grains of calcium carbonate per gallon or its 

 equivalent of dolomitic compound. Analysis showed the limestone 

 to have in it -01 per cent, of combined chlorine. The 10 grains 

 therefore contained '001 grain of chlorine. Now the whole gallon 

 of water gave '7 grain of chlorine, whence it follows that only 

 a fifth of 1 per cent, of all the combined chlorine in the water was 

 due to solvent denudation, or of the load of salt carried to the sea 

 approximately 99*8 per cent, was sea-salt. Professor Joly, in his 

 criticism in the Chemical News, does not appreciate this result ; in 

 the Geological Magazine he ignores it altogether. It is capable 

 of wide application. Thus, in some 40 full analyses of limestones 

 and dolomites published by the United States Geological Survey 

 (Bull., 148, pp. 254-274), 31 samples show no trace of chlorine, 

 2 only traces, and 9 samples from -01 up to '14 per cent. The 

 average quantity of chlorine in the last 9 samples is '06 per cent., 

 and in the whole 40 samples -01 per cent. Limestone is one of the 

 most soluble of all rocks, and will probably furnish the largest 

 share of chlorine to the rivers, and these figures demonstrate that 

 so far as the North American continent is concerned its limestones 

 are not likely to supply any greater proportion of this element than 

 the limestones of Yoikshire. Reference to other facts confirm this 

 view of things. Before proceeding farther let me give the atomic 

 proportions of sodium and chlorine in some of the bodies we have to 

 deal with, premising for the benefit of the non-chemical reader that 



