Chloride of Aluminium and its Analysis. 



'>5I 



Alumina. 



k<^ 



0.2975* 



0.2975 



0.2975 



0.2975 



0.2975 



0.2975 



0.2975 



0.2975 



1, 

 2, 

 3, 



4, 

 5, 

 6, 



7, 

 18, 



These values k 



0.139188* 



0.139188 



0.139188 



0.139188 



0.139188 



0.139188 



0.139188 



0.139188 



Atomic weiglits 

 Alumina, ol" Aliimiiiiiim. 



:: x: i.25000t x-. 

 :: y : i.7ii6()t y-- 

 :: x' ', 1.23585.t x'-. 

 1. 236201 x"-. 

 1.21567t y'-- 

 1.2151Gty': 

 0.87617^ Z-- 

 0.8792011 z'-. 



y 



:2.67W4 

 3.65849 

 :2.64149 

 : 2.64224 

 : 2.59836 

 =2.60446 

 = 1.87272 

 = 1.8792014 



ve respectively the following composition for alu- 



mina. 



Oxygen, 

 Aluminium, 



1.42174 1 

 1.25000 1 



.40564 

 .23585 



1.40604 

 1.23620 



0.99655 

 0.87617 



\< 



Alumina, 2.67174 2.64149 2.64224 1.87272 



y y' v" ^' 



Oxygen. 1.94683 1.38269 -1.. 38930 1.00000022 



Aluniinium, 1.71166 1.21567 1.21516 .87920118 



^Alumina, 3.65849 2.-59836 2.60446 1.87920140 



The oxygen in x x' x" y' and y" makes a distant approximation 

 to a sesquioxide ; y approaches a deutoxide ; but that of z dedu- 

 duced from Berzelius' reversed numbers, (vide i' and l:,^ and of z' 

 deduced the results of my experiments, approaches very near to 

 the ratio of that in a protoxide. 



The atomic weight of alumininm, upon the supposition that the 

 chloride is a sesquichloride, would be 

 5 in No. 7 of ^- = 1.31425 



'"^and " " 8 " A-=- 1.31880 



We can now calculate the atomid weight of alumina, considering 

 the chloride a sesquichloride. 



