1822.] Weight of an Atom of Alumina. IDo 



tity which the sulphate of potash in 100 grains of alum ought to 

 contain. This 0*8 grain had been dissipated by the heat of the 

 wind furnace from the sulphate of potash, and had rendered it 

 alkaline. 



These facts suggested a method of determining the quantity 

 of water in alum, and upon putting it in practice, I found that 

 the process, when repeated carefully, gave me always the very 

 same result. Instead of 100 grains of alum, I employed in pre- 

 ference 60*875 grains. 



My method was this : I exposed 60*875 grains of alum to 

 an intense red heat in a wind furnace in a platinum crucible 

 which I had previously weighed. The loss of weight sustained 

 was 43*62 grains. The residual matter in the crucible was 

 digested in distilled water. The clear solution was separated 

 from the alumina by the filter, and the filter was washed with 

 distilled w^ater till the liquid ceased to be affected by muriate of 

 barytes. The solution thus obtained was concentrated on the 

 sand-bath, and then precipitated by muriate of barytes. The 

 sulphate of barytes obtained, after being washed, dried, and 

 heated to redness, weighed 13*28 grains, which is equivalent to 

 4*504 grains of sulphuric acid. If v^e add this weight of sul- 

 phuric acid to the 43*62 grains driven off by heat, we obtain 

 48*124 grains as the weight of the whole water and sulphuric 

 acid contained in the 60*875 grains of alum. If from this quan- 

 tity we deduct 20 grains, formerly shown to be the weight of the 

 sulphuric acid, there remain 28124 grains of water. A repeti- 

 tion of this experiment gave the very same result. 



Should any person think of repeating this experiment, he 

 must be on his guard not to use filtering paper till it has been 

 digested for some time in distilled water ; for I was once or 

 twice deceived by using a filtering paper, which exhibited traces 

 of sulphuric acid. I was puzzled at getting more sulphate of 

 barytes than I ought to have had. The excess indeed was very 

 small ; but it prevented that exact coincidence between different 

 experiments which I was anxious to obtain. 



The reader will please to observe, that 28*124 almost exactly 

 coincides with the weight of 25 atoms of water ; for 1*125 x 25 

 = 28*125. My number is only -oy^^-^ less than this quantity. 

 Surely then I am warranted in concluding that 60*875 parts of 

 alum contain exactly 25 atoms, or 28*125 parts of water : 100 

 parts of alum then contain 46*2012 parts of water, which is 

 1*2012 more than the quantity detected by Berzelius. 



3. Potash. — To determine the potash contained in alum, I 

 found that an easier process than that of Berzelius gave results 

 fully as accurate. 100 grains of alum were exposed in a plati- 

 num crucible to a moderate heat on the sand-bath till the water 

 of crystallization was dissipated. The crucible was then kept 

 for half an hour in a red heat. Distilled water was now poured 

 upon the mass remaining in the crucible in successive portions, 



