Dr. Thomson on the ♦^ [March, 



gmtnl^ of sulphuric acid: 100 grains of alum must of course 

 Entail! 32*854*2 grains of sulphuric acid. 



: In this determination, from the great insolubility of sulphate 

 of barytes, I believe the number to be correct at least to as many 

 decimal places as 1 have given. Now Berzelius's number (even 

 when corrected) exceeds this by 0*9658 grain. 



2. Mater. — To determine the weight of water in alum by 

 direct experiment is attended with difficulties, which, for some 

 time, I found insurmountable. Indeed after a vast number of 

 attempts, I found such discordance between my results, that I 

 was induced to suspect that the water in alum was not a constant 

 quantity, and that of consequence an analysis of this salt per- 

 fectly correct was impossible. By degrees, however, 1 began 

 to suspect the sources of the variance, and a closer inspection 

 put me on a way of obtaining the water, if not by one simple 

 process, yet by uniting two together. 1 shall give in the first 

 place one or two of my early results, that the reader may per- 

 ceive the want of agreement between them. 



(1.) 100 grains of alum exposed for three hours to a heat of 

 600^ lost 44*04 grains ; or almost a grain less than Berzehus 

 disengaged by means of a spirit lamp. We see from this that 

 Berzelius had applied a heat exceeding 600° in intensity. 



(2.) 100 grains of alum exposed to a low red heat lost 70*72 

 grains. This obviously exceeded all the water in the alum, 

 while it appeared to fall short of the water and sulphuric acid 

 added together. For if the water amounts to 45 grains (and 

 from Berzehus's experiments it cannot be less) ; the water and 

 sulphuric acid united amount to 77*8542 grains. 



(3.) 100 grains of alum exposed to a strong heat in a wind 

 furnace lost 71*66 grains. This also falls short of the water and 

 siilphuric acid. 



tlpon examining the alumina which remained when the resi- 

 dual matter of experiment (2) was digested in water ; by dissolv- 

 ing it in muriatic acid, and mixing the solution with muriate of 

 barytes, I obtained a precipitate of sulphate of barytes indicating 

 the presence of little more than 1-lOth of a grain of sulphuric 

 acid. The alumina from experiment (3) examined in the same 

 way gave no traces whatever of sulphuric acid. The solution 

 made by digesting distilled water on it (which contained the 

 sulphate of potash from the alum) being examined, was found 

 to give a strong purple tinge to cudbear paper. Of course it 

 contained an excess of alkali. It thus became evident that the 

 heat of a wind furnace is sufficient not only to drive off all the 

 water and all the sulphuric acid united to the alumina, but like- 

 wise a portion of the sulphuric acid of the sulphate of potash. 

 On precipitating the whole sulphuric acid from this solution by 

 means of muriate of barytes, I found that the quantity of sulphu- 

 ric acid which it contained was about 0*8 graiu below the quan- 



