168 H. Rose on a new mode of preparing Aluminium. 
must have volatilized by the fusion. stevia metals comport them- 
selves similarly, as copper and even silv 
In this way the loss ¥ metal is none or very minute, at the most 
only a few milligram 
If the aluminium is i sale’ under a cover of chlorid of potadae 
sium, its surface is not thoroughly even, but shows minute exca-_ 
vations, which is oe the case if it is melted under chlorid of alu- 
minium and sodiu 
he quantities wg chlorid of aluminium and sodium to be em- 
ployed for this purpose, are most easily prepared by bringing the 
mixture of alumina and charcoal ina glass tube of as large a 
oa diameter as possible, and by introducing in it another glass tube . 
_-open at both ends of a smaller diameter filled with pulverized ta- 
ble-salt. By heating the space containing the mixture of alumina — 
and charcoal very strongly, and that which contains the chlorid 
of sodium less, while passing the chlorine gas through the tube, 
the vapors of the chlorid of aluminium will be abso orbed sO ea- 
gerly by the chlorid of sodium, that no chlorid of aluminium or 
only a trace of it deposits on the other parts of the apparatus. 
If the smaller glass tube with the chlorid of sodium has been 
weighed beforehand, the quantity of the chlorid of aluminium 
it has absorbed can be easily determined. The latter however is 
not combined equally with the chlorid of sodium; that part 
which was next to the mixture of clay and charcoal, contains the 
most of it. 
I have varied the process for the reduction of aluminium in 
many ways, but have returned to that described. Often the so- 
in consequence of the action of the chlorid of potassitim ape 
_ ihe slorendie, Neither will the globules of aluminium 
‘bya longer glowing heat, which was tried, up to two an 
- this effect was produced only through the most intense heat pos- 
sible. If, however, after the highest incandescence of the cru- 
cible, that is after five or ten minutes, the heating ceases, the 
bs 
. 
; 
1 
Bit 
