1822.] Weight of an Atom of Alumina, 163 



crucible. The sulphate of potash weighed 18*3 parts ;' but was 

 found, when dissolved in water, to contain 0*15 of sulphate of 

 strontian ; so that the real quantity of sulphate of potash 

 obtained was 18-15 parts. Now sulphate of potash is a com- 

 pound of five parts acid + six parts potash ; so that the true 

 quantity of potash in 18*15 parts is 9*9 instead of 9*81, as stated 

 by Berzehus. 



(4.) Berzelius's mode of determining the water of crystalliza- 

 tion in alum was to heat the salt in a platinum crucible over a 

 spirit of wine lamp. The loss of weight sustained was always 

 45 per cent. I find that by this method we cannot drive off the 

 whole of the water from alum. A small portion still remains 

 which cannot be dissipated, except by the apphcation of a red 

 heat. The analysis of alum by Berzelius then, when corrected, 

 gives us the following results : 



Acid , 33-82 



Alumina 10*86 



Potash . . . , 9-90 



Water 45-00 



99-58 



leaving a deficiency of almost half a per cent, which, as we shall 

 see afterwards, was owing to water not driven off by the heat of 

 a spirit of wine lamp. 



II. I shall now relate as concisely as possible the experiments 

 which I made in order to ascertain the constituents of alum ; 

 omitting, as is my usual practice, all the trials which were either 

 unsuccessful or not more successful than those which I state. 

 On the present occasion, the experiments which I omit were at 

 least 10 times more numerous than those which I give ; for I 

 employed a great variety of ways to analyze alum, partly to 

 check my results by one another, and partly to determine which 

 mode of analysis was easiest and most to be depended on. 



1. Sulphuric Acid. — When 60*875 grains of pure alum crys- 

 tals are dissolved in water, and the solution mixed with a 

 solution of 53 grains of chloride of barium, a white precipitate 

 falls, consisting of sulphate of barytes. After this precipitate 

 has subsided, if we test the clear liquid which swims over it by 

 means of solutions of glauber salt and muriate of barytes, we 

 shall find that it contains no traces either of barytes or of 

 sulphuric acid. Consequently the barytes from 53 grains of 

 chloride of barium exactly saturates the sulphuric acidui 60-875 

 grains of alum. But the barytes from 53 grains of chloride of 

 barium amounts to exactly 39 grains, or 9*75 x 4 ; but 9-75 

 barytes just saturate 5 sulphuric acid. It is obvious from this, 

 that 39 grains of barytes will just saturate 20 grains of sulphuric 

 acid; consequently 60*875 grains of alum contain exactly 20 



M 2 



