• i, OBSERVATIONS ON GALVANISM. 



VII. 



Letter from a Correfpondtnt I. R. I. explaining fame Fa&s in 

 Galvanifm, and on other Ohjeds. 



To Mr. NICHOLSON. 



SIR, Edinburgh, May 19, 1801-. 



Obferyations on 1 SENT }'ou, a few days ago, fome hints towards a farther 

 improvement of the galvanic apparatus*, and I hope that the 

 following obfervations will arrive in time to be added as a fup- 

 plement to my laft communication. 



There feem to be two very marked laws in galvanifm ; tiie 

 firfl: is, that the phoenomena are produced by moift oxidation, 

 which feems a general one ; but, by the fecond law, that they 

 remain latent till one part of the condu6ting metal be made 

 perfectly dry ; this is at leafl; ftri6lly correct where two metals 

 are employed. 



From the tendency of galvanic effeds to become latent, by 

 the ufe of moift conduftors, a pile cannot be formed unlefs 

 one fide of the oxidated metal be covered from moiflure by 

 cement, befides having a condudtor of a different metallic 

 fubftance. The trough is, therefore, beft adapted to oecono- 

 mize galvanifm, as I have fiiewn, that nearly as large a furface 

 can be expofed to the aftlon of the fluid in it, as in the coU' 

 romie de tajfes, over which it has many advantages of fieadi- 

 nefs, portability, &c. 



From the neceffity of dry condu6lors to alternate with moifi: 

 to produce the more flriking and perfeft galvanic efFeds, it 

 would appear that the former give the neceffary celerity to 

 their aiflion ; it might therefore be ufeful to enquire, whether 

 making them of a confiderable length, and inclofing them in 

 fome fubftancp that would infure them from the moifture in- 

 feparable from the galvanic apparatus, might not confiderably 

 increafe that celerity. 



It is much to be widied that the end condu6lors were im- 

 proved ; jointed wires feem to lofe much of the power, by 

 not being in perfed contact ; fpiral wire feeins better adapted 

 to that ufe, and more capable of varying in its diredion. 



* See page 79. 



It ' 



