[ 1003 ] 
XXII. Revision of the Atomic Weight of Aluminum. 
By J. W. Mallet, F.R.S., Professor of Chemistry in the University of Virginia. 
Received March. 13,—Read Ajml 22, 1880. 
Need for a recletermination of the atomic weight of aluminum. 
There is probably no one of the so-called chemical elements equally abundant in 
nature with aluminum, and occurring in as numerous compounds, with regard to the 
atomic weight of which our knowledge has long rested upon so slender a foundation 
of accurate experiment. The following brief statement includes, I believe, all the 
determinations of this constant which are on record. 
Former Determinations. 
1. Experiments of Berzelius, 1812 .—Berzelius* precipitated a solution of alum 
by addition of ammonia, dissolved the precipitate in sulphuric acid to saturation, 
filtered, concentrated the filtrate by evaporation, and threw down aluminum sulphate 
by alcohol. This salt was well washed with alcohol, to separate as far as possible any 
excess of acid, and was then heated in a platinum crucible over an alcohol lamp, 
weighing from time to time, until no further loss of weight occurred. The anhydrous 
sulphate so obtained was but slowly soluble in water on heating, but left no insoluble 
residue. 10 grins, of this salt was now raised to a higher temperature in a weighed 
platinum crucible, and strongly heated as long as any loss of weight could be detected. 
The residue of loose, light, white alumina found in the crucible weighed 2 - 9934 grms. 
Consequently the salt consisted of— 
“ Sulphuric acid” (S0 3 ) .... 70 - 066 or lOO’OOO 
Alumina.29‘934 „ 42722 
100-000 „ 142-722 
Several small arithmetical errors have been made in the discussion of this single expe¬ 
riment of Berzelius, upon which for nearly half a century the value assigned to the 
atomic weight of aluminum may be said to have rested. 
In the original paper it is calculated that if 42722 parts of alumina contain 19"90 
* Gilbert’s ‘ Annalen der Physik,’ xl. (1812), 260. 
6 N 2 
