IS 14.] On the Sea Water on the Coasts of France. 205 



white powder, which, separated, washed, and dried, weighed 0-3 

 gramme. The insoluble residue from the water of the Channel 

 weighed 0*4 gramme. 



Muriatic acid dissolved a part of this residue with effervescence. 

 The insoluble portion was sulphate of lime, and weighed 0-15 

 gramme. 



Ammonia poured into the filtered liquid occasioned a white pre- 

 cipitate, which was not acted on by potash, but was dissolved by 

 sulphuric acid. We ascertained that this muriatic solution con- 

 tained lime, magnesia, and a trace of iron. 



Thus the insoluble matter is composed of 



Sulphate of lime 0-15 gramme 



Carbonate of magnesia and of lime . .0-15 

 With a trace of iron. 



After having determined the nature of the insoluble jnattcr, wfi 

 examined the liquid which contained the soluble salts, and particu- 

 larly the muriate of soda in solution. 



The quantity of sulphate of magnesia was determined in the 

 following manner. Potash was j)oured into the solution as long as 

 it formed any precipitate. The liquid was filtered while still hot, 

 and the precipitate separated, washed, dried, and calcined. We 

 obtained 2-2 grammes of magnesia, corresponding to 57^ grammes 

 of sulphate of magnesia. 



If instead of potash we employ nitrate of barytes, we obtain a 

 precipitate of sulphate of barytes, which, after being washed and 

 calcined, contains exactly the quantity of sulphuric acid necessary 

 to saturate the magnesia. Hence it follows that the soluble sul- 

 phate is entirely sulphate of magnesia, and that sea water contains 

 no sulphate of soda. 



To determine with still greater accuracy whether sea water con- 

 tains any sulphate of soda, we calcined a portion of the salt in a 

 crucible with purified charcoal, obtained from lamp-black. The 

 residual mas« was mixed with water, and filtered. Paper stained 

 with turmeric was not altered by this solution,' which shows that it 

 contained no alkali. Nothing was found but a little hydrogureted 

 ■ulphuret of magnesia. The greatest part of the magnesia re- 

 mained on the filter with the charcoal. 



It i- <v, I| Known to be scarcely possible to separate sulphate of 

 nesia completely from common salt by crystallization ; because 

 the difii etween the solubility of the two salts is not swili- 



cici.lv g Bt. Neither docs efflorescence furnish a certain method. 

 We exposed th< I Its io tin' gentle heat of a stove. The sul- 



phate of magnesia effloresces very -lowly, and the efflorescence is 

 not :■. sulpha) .da is crystallized confusedly 



with common 



By subtracting all tl itcd salts from the total mass, we 



in the weight of tin: common salt. Put that method did not 



