92 An Analysis of Sea-water ; 



the proportion of it being the quantity equivalent to that of the 

 sulpliate of lime which the analysis affords. 



The small portions of carbonates of magnesia and lime which 

 are obtained in the evaporation, I have already stated, are evi- 

 dently accidental products, from the decomposition of muriate of 

 magnesia and muriate or sulphate of lime. 



The large (piantity of muriate of soda leaves no doubt, if binary 

 compounds at all exist in a state of solution, that it is the chief 

 ingredient in sea-water, though the quantity of it may be af- 

 fected bv the changes which occur from the actions of some of 

 the other salts. 



A similar conclusion is to be drawn with regard to the muriate 

 of magnesia. Though the proportion of it may be affected by the 

 changes which occur in the analytic operations, still, from the 

 quantity in which it is obtained, a considerable part of it must 

 originally exist. 



The principal difficulty is with regard to the sulphate of mag- 

 nesia, and the sulphate of soda. It has always been supposed, 

 that sulphate of magnesia is an ingredient in sea-water, from its 

 being procured by evaporation ; and it is possible that it may 

 be so. But it is just as possible, a priori, that sulphate of 

 soda may be the original ingredient, and that, during the eva- 

 poration, the mutual action between it and muriate of magnesia 

 is favoured by the concentration, whence portions of both are 

 <iecomposed, and corresponding quantities of sulphate of mag- 

 jiesia and muriate of soda are formed. Nor is there any thing 

 connected with the mere results themselves, which proves which 

 of these views is just. 



If the appeal be made to experiment, it is sufficiently esta- 

 blished, that sulphate of magnesia may be formed by the action 

 of sulphate of soda on muriate of magnesia. \\'hen these two 

 salts are boiled together in solution, a double decomposition takes 

 place at least partially, and portions of sulphate of magnesia and 

 muriate of soda are formed. On the other hand, the reverse 

 combinations may also, acccording to circumstances, be esta- 

 blished. We have seen reason to conclude, that they are so 

 from the agencv of alcohol ; and even in an aqueous solution, 

 vhen certain proportions of the salts are used, they appear, 

 imder some circumstances, particularlv that of a low temperature, 

 to take place to a certain extent. But still these facts only show 

 ivhat decompoiitions mav occur from evaporation or otiier pro- 

 cesses ; they do not prove what the actual state of combination 

 5s in the original solution. 



It is obvious, that this is merely a case belonging to the more 

 general question, V/hat is the state of combination in a com- 

 pound 



