NEW CLASSES OP GALVANIC CONDUCTORS. 319 



chemical effects, which depend on the presence of water, 



continue decreasing, and soon cease entirely, when all the 



water at the points of contact is decomposed. From that 



moment the soap resumes its characteristic property, and 



insulates the positive electricity. 



It will be proper to introduce here an observation of * f tliere De an ? 



. moisture in the 

 some importance to the success of experiments of this kind. soap therefore, 



They who would repeat them without being able to procure the operator 



prisms of soap exposed to the air for some years, or com- at first, the soap 



pletely dried by the action of an oven or a stove cautiously -acting as a per- 

 , .,, , /•<. feet conductor, 



conducted, might be tempted to accuse me at first of not 



having seen clearly; for a communication being established 

 between the polar wires by soap yet damp, both the elec- 

 trometers and the apparatus for decomposing water will 

 begin by indicating a more or less perfect completion of 

 the galvanic circle. y But the part that water acts in these 

 phenomena perfectly explains this want of success. It is 

 the portion of free water, interposed in the damp soap 

 between it and the positive wire, that in this case conceals 

 the characteristic property, by which this substance belongs 

 to the fifth class. To evince this nothing more is necessary, But this effect 

 than to suffer a few moments to elapse : the water foreign i ng a nJi e ^ 

 to the conditions of the experiment will be consumed with tne water in 

 more or less rapidity, according to its abundance and the the wire is de- 

 energy of the pile; and then the whole of the soap will composed, 

 insulate the positive electricity, while it will serve as a con- 

 ductor to the negative. On talcing out the positive wire, 

 that has thus been inserted into damp soap, the point will 

 be found oxided, if the metal be of a nature to admit it, 

 which never takes place in soap perfectly dry. It is scarcely Though the 



necessary to add, that, if this wire be cleaned, and insert- soap Wl " act 



' , , ' , as a c °nduct r 



ed into any other part of the damp soap, the same excep- again, if th e 



tion to the general rule will again recur, since in this new Wire be ins erted 



- m a fresh part, 



point of contact the conductor will find a fresh portion of 



free water. The oxidation of the positive wire in damp The cessation, 



soap might lead to the supposition, that the insulation of not owir?** ** 



the positive pole is owing to the production of this non- oxidation of the 



conducting coat. But the contrary may be proved by em- Wir6i 



ploying platina wires, which exhibit the phenomena in 



question, as soon as the water interposed by chance or 



design 



