304 EIGHTH REPORT — 1838. 



"preservers ;" a more destructive practice can scarcely be con- 

 ceived, or one more fatally applied. From these circumstances, 

 and lest these mischievous results should be extended elsewhere, 

 it lias been deemed right thus at length to refute it, which I 

 conceive is fully done by Schcenbein's, Professor Davy's, and 

 my own experiments. 



127. To recur again for a moment to the subject of the boxes • 

 of specimens of cast iron sunk for experiment, it was stated 

 that they were divided into separate cells for each kind of iron 

 by veneers of varnished oak. The reason of this arrangement, 

 and a deduction which has grown out of it, and is likely to prove 

 important as affording a mode of protecting cast and wrought 

 iron, remain to be stated. It having been early remarked that 

 the harder irons, whether cast or wrought, were acted on much 

 more slowly than the softer and more carbonaceous ones, it 

 appeared not impossible that if several different sorts were in- 

 closed in electrical continuity in the same box, grave errors 

 might be introduced into our results by the iron least acted on 

 standing in a negative relation to those more rapidly corroded, 

 and increasing the action of the sea or other M'ater upon them, 

 and at the same time being themselves preserved to a certain 

 extent. 



128. By a few preliminary experiments with the galvano- 

 meter, this was found to be a correct view, — it was found that 

 of any two different irons, the harder was always in a negative 

 relation to the softer, which was positive to it, and hence the 

 separation of every specimen became necessary in order to eli- 

 minate this source of error. 



129. This at once suggested to the writer the possibility of 

 preserving the hard gray cast iron and the wrought iron, &c., 

 in common use, by the application of protectors formed of the 

 softest and most highly carburetted cast iron attainable ; and as 

 the conversion of this latter into plumbago, to a great extent, 

 did not seem materially to alter its electrical relation to gray or 

 A^hite cast iron, or to wrought iron, it seemed probable that it 

 might afford an electro-chemical protector superior in many re- 

 spects even to zinc. With this view experiments are now in 

 progress, and so far are decisively in favour of the method. 



130. The intensity of current produced by soft and hard cast 

 iron is much greater than would have been anticipated. When 

 two small bars, each 4 in. long, by 0*5 in. wide, by 0"25 thick, 

 one of soft black and the other of hard gray cast iron, were both 

 broken in two, and an iron wire soldered to each half, on im- 

 mersing the two halves of either one piece in common water 

 the needle of a Melloni's galvanometer was scarcely disturbed ; 



