of chemical action. Corrosion of iron, in the sense in which that term has been used in this section, has nothing whatever to do with electrolysis by stray electrical currents from outside sources. The currents involved in rusting under the theory of electrolytic action are almost infinitely short and minute, and originate in or on the metal itself.

The theory is valuable to the extent that it suggests reasonable and practical remedy of the defects either of the metal or its proposed covering, or both. As in the treatment of diseased animal and plant tissues, so in this case, intelligent diagnosis must precede the application of preventives of rust. Experimental work following the lines of the electrolytic theory in seeking, first, to prevent, or "inhibit" corrosion by a priming coat, and,  secondly, to diminish the