THE CORROSION OF IRON AND OTHER METALS 317 



therefore that, whenever galvanic activity is induced, the total 

 amount of corrosion is increased, although the actual extent of 

 that increase must depend upon local circumstances " (p. 240). 



It was demonstrated by Faraday that whenever chemical 

 change takes place the energy liberated is in the form of elec- 

 trical energy and we must regard it as established that heat 

 is developed as a consequence of the frittering down of electrical 

 energy by the resistance of the circuit. Whenever a proper 

 path is provided, as in a voltaic cell, the electrical energy can 

 escape as such almost entirely; if, however, the action take 

 place in a closed circuit from which there is no outlet, as when 

 a metal is simply dissolved in an acid, the electrical energy is 

 apparent only as heat. 



It may be asserted, I believe, without hesitation that when- 

 ever corrosion takes place it takes place in a " galvanic " circuit, 

 and that galvanic activity necessarily accompanies corrosion. 



In his book Friend discusses at length the question " Is an 

 acid necessary to corrosion?" The conclusion at which he 

 arrives agrees with that stated in the earlier part of this article — 

 that it is. He is of opinion that previous observers who 

 have contended that the presence of an acid is not necessary 

 have failed to exclude carbonic acid from their apparatus. 

 But although he regards Moody's experiments with favour, he 

 ultimately discards them as evidence on the ground that Moody 

 used chromic acid and thereby rendered the iron " passive " — 

 overlooking the fact that in Moody's experiments the iron 

 rusted at once when brought into contact with the glass of 

 the tube in which it was enclosed and also when air containing 

 carbon dioxide was admitted to the apparatus. The objection 

 that the iron would be rendered "passive" by the treatment 

 to which he subjected it was foreseen by Moody and disposed 

 of by him. Friend's own experiments, to which reference has 

 been made already, confirm Moody's result in the most striking 

 manner possible. 



It is commonly supposed that iron will not rust in presence 

 of alkali. Friend in his book describes a series of experiments, 

 which are a most important contribution to the discussion of 

 this question, in which iron was kept in contact with purified 

 oxygen and solutions of caustic soda : when he used a very 



