f4 •'' CHFMICAL Atfiitirr, 



this wliolc decomposition must be referred to the iiiie<qiiM at* 

 tractions of the two wires; tor each wire will aitruct hoth 

 <»xii;en and hidroi^en, bitt with unequal degrees of force; 

 .•c '. and these attractions will be modified aqd counteracted by 



tjie attractions of tlie 0|>posite wire. If, for example, thi^ 

 positive wire attract oxi^en with a force equal to 20, and 

 bidrggeu with a force equal to 10; and vice terstt, if the ne» 

 gative wire attract hidroj^en with a force equal to 20, apd 

 oxii^en with & force equal to 10; the efficient attraction 

 bttween the positive y:'}i'.e ^d oxigen would be equal to 10, 

 and that b^etween the negative \vire and hidro^en would be 

 equal to JX), and .con^equje^Uly the power, tending to sepa- 

 rate the, oxij^en and hidrogen, would be equal to 20. If, 

 tjiereibre, we ikeep in mind, that th,e effect of the two wires 

 increases with the di.fferepce iu their electrical state; we 

 must, 96 might be shpwn hy numerical calculation, 8ui)pose, 

 that hidrogejn is .more positive than the positive wire, and 

 oxigen more negative than the negative wire. On this »uj>- 

 |)ositiori, ^udpn no o.ther, Jt \yill appear, that, as the excite- 

 ment ^f the tjjrp vyires is augmented, their action on >yatj;r 

 should be more powerful : for the nearer the electrical state 

 of the positive wire cpmes to ths^t of hi.drogen, and the elep* 

 t^ical state of the negative wire to ifiut of oxigen, the stronger 

 ahould be the efficieu)t attraction of the positive wire fpr oxi^ 

 gen, and of the oegatiye wire for hidrpgen. The. saiiie rear 

 «pnii'ig must apply to the decomposition pf all bodies, and 

 the coustituents of every bo;dy, decomposed by ^alxwhm, 

 ipast he x:on8 dered as huving electrical «;tates more v;idely 

 different, than ^re those of the pQsitive and n^gatiye wireg of 

 the galvanic battery. But this is shown tp be irppossible 

 by Sir Isaac Newtpn's experiment with the pendulum, .aud 

 \^y ev«fy kind of experimeot with the electrometef. 

 Whyifpotde- ^^d.m.ittipg, for a .{.oaeiit, th it the attractive and r.epulr 

 iff!^^tr/bTa ^^^ for^estQf the ro^ '^te .^^^rticles of rni^tter, aiVd the action 

 fcinglewire? ofgalvaitic ^jvires pn compound bodies,' are r.e;jl|y ppch as 

 Mr. Davy supposes, '}t would, I think, be difficult to exr* 

 plain, wiiy decoii). osi'io. e *r ^voJj c<"^< by j single 



wire, \wvie;yf^r pon^rful may bet)ic t^atterj', with which it is 

 counectec); why ,de<pm position \% never effected, either by 

 «y^:nou or j^alvauic citcuiciry, except when Itco conduc- 

 . . tors. 



