THE CORROSION OF IRON AND OTHER METALS 319 



electrolyte and condition rusting in presence of oxygen. It is 

 illogical to assume on the evidence advanced by Kohlrausch and 

 Heydweiller that it can, as every step they took to purify water 

 further involved a diminution of its electrical conductivity and 

 it is therefore permissible to assume that ideal water — water 

 pure and simple — would be a non-conductor ; on general 

 grounds it is also to be supposed that a binary compound such 

 as water would not be an electrolyte. The conclusion that 

 water is " electrolytically dissociated " is purely hypothetical ; 

 and all the arguments used in support of the hypothesis involve 

 reasoning within a vicious circle. If chemists are wise they 

 will cease to talk of the "acid" and the "electrolytic" theory 

 of rusting as distinct "theories"; practical men will then no 

 longer be justified in calling " A plague on both your houses " 

 and in time may show some respect for theory that is brought 

 under their notice in simple and straightforward language. 



The Influence of Oxygen in promoting Corrosion 



The rusting of iron is undoubtedly a complex process and 

 rust may be formed in more than one way, according as the 

 action takes place primarily either in a circuit from which 

 oxygen is absent or in one in which it is present. 



When iron — coupled with an electro-negative conductor — is 

 brought into contact with an acid solution, it is dissolved as 

 ferrous salt probably in the manner illustrated in the equation 



Fe 4- C °3 H * - Fe / C0 3 H + h 

 t e + C0 3 H 2 r e \ COsH + rt * 



Ferrous bicarbonate 



Or it may be — especially in the case of weak solutions — that 

 a " basic salt " is the first product : 



Fe 4- C °s H 2 _ Fpl COaH 4- H 

 * e + OH 2 -* e lOH +il * 



Ferrous hydroxycarbonate 



In any case, whatever the conditions, in so far as iron is 

 dissolved at all by an acid, it may be supposed that the metal 

 is always converted into ferrous salt ; the production of rust 

 may be said to be a secondary phenomenon. 



Presumably when a ferrous salt in solution is brought into 

 contact with oxygen it is converted into a ferric salt and rust 



