322 SCIENCE PROGRESS 



it is easy to understand that as these accumulate on the iron 

 surface they may well exercise a considerable influence in 

 lowering the resistance in the circuit of change by increasing 

 the area of the electro-negative element. 



Influence of Salts 



On the assumption that iron rusts onty when " acid " is 

 present, it is not to be expected that strictly neutral salts will 

 promote rusting; this conclusion appears to be a correct one. 

 As Adie and others have shown, in the absence of air, solutions 

 of the sulphates and chlorides of the alkali metals are without 

 action on iron but it is more or less attacked by magnesium 

 chloride, especially if the temperature of the solution be raised ; 

 magnesium chloride, it is well known, is more or less acted 

 upon by water according to the temperature in the sense of 

 the equation 



MgCL + OH 3 = HC1 + MgCl . OH 



and the solution is therefore in some degree acid. 



The evidence as to the action of neutral salts when air has 

 free access to the solutions is somewhat conflicting. In the 

 case of concentrated solutions iron seems always to be less 

 attacked than it is when exposed merely in water, the reason 

 being apparently that such solutions contain less oxygen than 

 is present in ordinary water. That corrosion occurs at all in 

 such cases must be attributed to the presence of carbonic acid 

 in the solutions. 



According to experiments made by Heyn and Bauer, 

 accounts of which were published in 1908-1910, even dilute 

 solutions of the chloride and sulphate of potassium and 

 sodium have less corrosive action on iron than distilled water 

 alone. Friend and Brown came to an opposite conclusion 

 but Friend points out in his book than an explanation of the 

 difference is probably to be found in the fact they have 

 established, that whereas at io° C. certain solutions of sodium 

 chloride have greater corrosive action on iron than distilled 

 water at the same temperature, at 14 the solution and water 

 have the same corrosive action, whilst at 21 the water has the 

 greater action, as observed by Heyn and Bauer. It is to be 

 supposed that the extent to which corrosion takes place will 

 depend not only on the amount of carbonic acid but also on 





