THE ACCOMPANYING SUBSTANCES. 21 



ftated, for filica, for magnefia, and for lime. All ftones in 



Which there is but little alumina, and a great quantity of filica, 



leave, after fufion with potafli, a light and flocculent fub- 



ftance, which cannot be diflblved by the acids : this fubftance, 



however, which is filica, has been in folution in the alkali. 



But, if a greater proportion of alumina be prefent, none of and promotes its 



this flocculent precipitate appears j hence it is evident, that ° u ' * y * 



alumina mud determine its folution. Its eafy folubility, in the 



latter cafe, cannot depend upon the divifion of the particles 



of the filica in the ftone ; for, in the firft place, after being 



fufed with potafh, the tenuity of the particles of every ftone 



muft be nearly the fame ; and, in the next place, I have not 



obferved, that any earth, except alumina, can promote the 



chemical folution of the filica, though they muft all occafion 



its mechanical divifion. 



As to the affinity of alumina for magnefia, it is by much Alumina very 

 themoft powerful of all thofe which any of the earths have powerfully at- 

 tor each other. I attempted to precipitate magnefia from 

 muriatic acid, by ammonia, even in excefs; but found that 

 the whole muriate of magnefia had not been decompofed, and 

 that a triple fait, or an ammoniacal muriate of magnefia *, 

 had been formed. I then poured an excefs of ammonia into a 

 folution of muriate of magnefia, mixed with a large proportion 

 of a folution of muriate of alumina. All the earth was preci- 

 pitated ; and nothing remained in folution, except muriate of 

 ammonia. The liquor was then filtered, and the precipitate 

 wafhed and dried. I diflblved it in muriatic acid, and boiled 

 it with a great excefs of potafli. Some alumina was taken up, 

 but by no means all the quantity that had been ufed. The 

 precipitate which had refifted the action of potafli, was again 

 diffolved in muriatic acid, and precipitated by carbonate of 

 potafh. The carbonate of magnefia was held in folution by 

 the excefs of carbonic acid ; and, by ufing potafli and carbo- 

 nic acid alternately, (the firft to diffolve alumina, the fecond 

 to diflblve carbonate of magnefia,) I effected a feparation of 

 the earths. Thefe experiments (hew, that there is an affinity 

 between alumina and magnefia, and a certain point of falu- 

 ration, where the action of potafh upon alumina is wholly 

 counteracted by the affinity of that earth for magnefia. 



* This fait is well known in chemiftry. 



When 



