OXIDATION OF IRON. 105 



in which the statements referring to it differed, and from 

 recent observations he had made he doubted the correct- 

 ness of the acknowledged composition of rust. He further 

 stated that if he took a bar of rusted wrought iron and put 

 it into violent vibrations by applying at one end the fall of 

 a hammer, scales would be separated which did not appear 

 to him to be the substance I had described. 



This conversation induced me to commence a series of 

 experiments which I shall now detail. 



I first carefully analyzed some specimens of iron-rust 

 which were procured as far as possible from any source of 

 contamination. Thus, one of these samples was supplied 

 to me by Sir Charles Fox as taken from the outside of the 

 Conway Bridge, the other secured by myself at Llangollen, 

 North Wales. 



These specimens gave the following results when sub- 

 mitted to analysis : — 



Conway Bridge. Llangollen. 



Sesquioxide of iron 92900 93'°94 



Protoxide of iron 6-177 5-810 



Carbonate of iron 0-617 0*605 



Carbonate of lime 0*295 °'~95 



Silica 0*121 0*196 



Ammonia trace trace 



IOO'OOO 100*000 



These results clearly show the correctness of Sir Charles 

 Fox's statements that the composition of the rust of iron 

 is far more complicated than it is stated in our text-books. 

 Therefore the question may be asked, is the oxidation of 

 iron due to the direct action of the atmosphere, or to the 

 decomposition of its aqueous vapour ? or does the very small 

 quantity of carbonic acid which it contains determine or 

 intensify the oxidation of metallic iron ? To reply to these 

 queries I have made a long series of experiments extend- 

 ing over two years, which I hope will tend to throw some 

 light on this very important question ; and although it 



