222, AN ANALYSIS 
phate of soda in a much larger quantity, with only an inferior 
proportion of sulphate of magnesia, are obtained. How is this 
diversity of result to be accounted for ? 
As the relative quantities of these salts are thus varied, and. 
are indeed nearly altogether dependent on the kind of analy- 
sis, it is obvious that one or other of them must be an origi- 
nal ingredient, and the other must be a product of decomposi- 
tion. If sulphate of magnesia is the original ingredient, 
then, when it is not obtained, or is obtained only in very in- 
ferior quantity, while sulphate of soda is procured in its place, 
it must be held that it is decomposed ; and the only decompo- 
sition that can account for these results is that from the mutual 
action of muriate of soda and sulphate of magnesia, by which, 
while portions of them are removed, corresponding portions 
of sulphate of soda and of muriate of magnesia are formed. 
On the other hand, if sulphate of soda is the original ingredi- 
ent, then, when it is not obtained, or is obtained only in small 
quantity, it must be held that it is decomposed ; and the only 
decomposition of it that can here take place, must be from the 
action of muriate of magnesia, by which, while quantities of 
both these salts are removed, corresponding quantities of sul- 
phate of magnesia and muriate of soda will be produced. 
Of the two analyses, the one in which sulphate of soda prin- 
cipally is obtained, is that in which the solvent action of alco- 
hol is employed ; the other, in which there is the mere sepa- 
ration of the salts by evaporation and crystallization, is that 
which affords scarcely any of it, but in place of it sulphate of 
magnesia. Now, it is to be observed, that in both of these the 
preliminary operation of evaporation to dryness is the same. 
Since sulphate of soda, therefore, is not obtained by this ope- 
ration, it is obvious, that, even on the assumption of its being 
the original ingredient of sea-water, it must, in the progress of 
the 
