56 C. J. WHITE. 



important in this direction, and indeed almost rendered 

 superfluous any further complete analyses of sea- water — 

 the percentage of total solids only being required. 



However, since in the seventy-seven complete analyses 

 of * ' Challenger " waters there is not included any off the 

 coast of New South Wales, 2 the present analysis was 

 undertaken — at the suggestion of Professor Liversidge and 

 in connection with his investigation on the presence of gold 

 in sea-water — as a check on the theoretical quantities 

 demanded by Dittmar's and Buchanan's calculations. 



The accurate determination of total salts is not quite 

 such a simple matter as at first sight it appears. The 

 apparently most obvious method of evaporating a known 

 quantity of water to dryness is inapplicable, since under 

 these conditions the magnesium chloride is partly decom- 

 posed according to the equation 



Mg01 2 + H 2 = Mg + 2 HC1. 

 The large excess of alkali chlorides which might be expected 

 to protect the magnesium chloride does not prevent 

 decomposition. 



Other methods of determining the total solids which 

 readily suggest themselves are 



(1) Determination of specific gravity. 



(2) Determination of some important constituent which 

 can be readily estimated with a high degree of accuracy — 

 like chlorine— and multiplying this by a certain constant 

 factor. 



The latter method was employed by Dittmar, who, work- 

 ing on total halogen estimated as chlorine, concluded that 

 the constant factor should be 1*8058. The former was 

 adopted by Buchanan who applied it to many hundreds of 



2 Four surface samples (Nos. 441 - 447) were collected on the trip from 

 Melbourne to Sydney, but seemingly were only utilised for specific gravity 

 determinations. 



