16 Historical Sketch of Improvements in [July, 



almost always of a combination or mixture of about 94 parts of 

 protoxide of iron and six parts of oxide of titanium. — (Journ. de 

 Pharm. V. 25S.) 



6. Nickel. — It seems barely necessary to put the reader in 

 ' mind of Dr. Clarke's method of obtaining pure nickel by means 



of the blow-pipe, and of the process which I myself am accus- 

 tomed to follow to procure this metal in a state of purity. They 

 will both be found together in the Annals of Philosophy, xiv. 



142. 



7. Man^fanese. — Von Mons has stated a curious experiment, 

 which, he says, may be made upon the black oxide of manga- 

 nese. If allot solution of binoxalate of potash be poured upon, 

 the black oxide of manganese, and the mixture be thrown upon 

 the filter, we obtain a liquid of a fine red colour. During the 



■• -action of the acid on the oxide some carbonic acid is disengaged. 

 ! (Jour, de Pharm. v. 307.) It is obvious that the solution is 

 . analogous to the red sulphate of manganese. The oxide loses a 



Sortion of its oxygen, and is reduced to the state in which it 

 issolves in sulphuric acid, and forms the red sulphate. Would 

 not pure oxahc acid answer better than the binoxalate ? 



8. Lead. — I have related a fact respecting the protoxide of 

 lead which I think deserves the attention of those whose busi- 

 ness it is to reduce that metal from the ore by smelting. I 

 found that I could volatilize this oxide by an ordinary red heat ; 



: the consequence of this great volatihty is, that much of the 

 metal must make its escape during the process of smelting. 

 {Annals of Philosophy, xiv. 314.) 



A very elaborate set of experiments by Berzelius, to determine 

 the composition of oxide of lead, will be found in the Annals of 



• Philosophi/, XV. 94. They appear to have been made with the 

 most scrupulous attention to accuracy, and approach as near the 

 truth as we can come experimentally in the present state of our 

 apparatus. The result is, that protoxide of lead is a compound of 



Lead 100-000 



Oxygen 7*725 



If we suppose this oxide a compound of one atom lead and 

 one atom oxygen, and calculate from the analysis of Berzelius, 

 we obtain for the weight of an atom of lead, 12*945. The number 

 13, which I have pitched upon, is within less than half a per 

 cent, of the number obtained from the result of Berzelius's expe- 

 riments. Now I am afraid that Berzelius deceives himself when 

 he thinks that he can depend upon the result of his analyses 

 within less than half a per cent. I have no doubt whatever that 

 the difference between 13 and 12*945 is within the limits of una- 

 voidable error. The simplicity and great convenience of 13 

 gives it in my eyes a preference, which will induce me to retain 

 it till I see stronger reasons than have yet been advanced for 



