320 Tht EffiBs of Allalit In Alum. 



or folidity. The folution here defcribed, which hsd conftantly rcfufcd to afford cryflallizcd 

 nlum alone, afforded it immediately by the addition of a few drops of the folution of pot- 

 afh -, and as I had employed thefe two fubftances in the rcquifitc proportion, the reft of the 

 folution a/Torded to the very end pure alum without any mixture of fulphatc of pot-afli. 



Into another portion of the fame folution of pure aluminc I dropped the fame quantity of 

 carbonate of foda as I had added of that of pot-aOi to the former. No cryflallization wjs 

 formed, even by the help of evaporation. 



Lime and barytes produced no better effcft. 



Thcfc experiments began to confirm the opinion I had, that the cryflallization of alum ia 

 not prevented by an exccfs of acid, as .'.as hitherto been thought, and that pot-alh was not 

 of ufe fimply to faturatc this acid, but that it performed an oflice of more importance. For 

 I re-ifoned, that if the common opinion were true, foda, lime, barytes, and all the fubftances 

 which by a more powerful force would take this acid from alum, ought to give the fainc rc- 

 fult. Another argument likewife prcfented itfelf, which feemed decifive. li the alkalis, pot- 

 afli and ammoniac do nothing more than unite to the fuperabundant acid of the alum, the 

 fulphates of pot-afh and of ammoniac ought not to occafion any change in pure alum in its 

 acidulated ftate ; but if thefc alkalis enter as a conftituent part into the alum, and are necef- 

 fary to its exiftence, they ought to produce the fame efiecls as pure pot-afh or ammuniac. 



I therefore added to a third portion of the folution of fulphate of alumine before men- 

 tioned, fome drops of the folution of fulphate of pot a Qi ; immediately upon which octahedral 

 alum was formed. The fulphatc of ammoniac prefented the fame effect. 



This rcfult gave ftill greater confirmation to my firft notions, though it did not yet afford 

 a demonftration perfc£lly witliout obje£lion. For it might have happened that the two falts 

 I made ufe of might determine the cryftallizationof the alum, fimply by abforbiagthe fupcr- 

 fluous acid, of which they are very greedy *. 



To determine this poffible fad, I mixed in the uncryftalhzable folution of alumine fame of 

 the acid fulphate of pot-afh, and obtained a cryftallization no lefs abundant than with the 

 ■neutral fulphate of pot-afli. 



Thislaft experiment docs not therefore leave any doubt with regard to the influence and 

 mode of action of pot-afli atul ammoniac in the fabrication of alum. This action is ftill 

 more ftrongly confirmed, by the. examination of the alums wiiich have been formed by the 

 proceffes above related. For in this manner it is proved that they contain notable quantities 

 of the fulphates of pot-a(li and ammoniac. 



Thcfc experiments naturally led me to an examination of the different alums of commerce. 

 Bergman had already announced, though in an indiftinfl. manner, that not only the common 

 alum, but likewife that of Rome, when decompofed by ammoniac, afforded traces of the 

 fulphate of pot-afh -, and Scheele had remarked on his fide, that alum which does not con- 

 tain pot-afh is not fit for making pyrophorus. Bergman, quoting this fa£l from Scheele, 

 ihcws likewife that he confidcred the fulphate of pot-afh in alum as a foreign fubftance f . 



* Ccterum alumcn non tantum vulgarc, fed ciiiim ronianum alcali vobcili pricipiiatum liijuorcm cxhi'uct, 

 ^i<i haud raro alcali vegetabilc viuiolalum contincc. ioid. 



f Alumen hoc inquinamcnto fpoliatum pyrophoro gencrando incptuin eft : quod f.iclle experiri licet, nam 

 onagmaatuminofuin diftirflam rcfpucns cryil.illiz.ilionem, nullqm pra-'bet pyroj>horum, modo coafucio tr,i£la- 

 tuin, quAmvis idem addito akali vcgetabilis pauxillc, cximium poirig.it, &c, 



3 Li 



