A new Theory of Galvanism. 211 



stated induced me to prefer for a first experiment a more ma- 

 nageable arrangement. 



Since writing the above, I find that when, in the apparatus of 

 twenty copper and twenty zinc plates, ten copj)cr plates on one 

 side are connected with ten zinc on the other, and a communica- 

 tion made between the remaining twenty bv a piece of iron wire 

 about the eighth of an inch in diameter, the wire enters into a 

 vivid state of combustion on the immersion of the plates. Pla- 

 tina wire eqxial to No. 1 S (the largest I had at hand) is rapidly 

 fused if substituted for the iron. 



This arrangement is equivalent to a battery of two large Gal- 

 vanic pairtj ; excepting that there is no insulation, all the plates 

 being plunged in one vessel. 1 have usually separated the pairs 

 by a board extending across the frame merely. 



Indeed, when the forty plates were successively associated in 

 pairs, of copper and zinc, though suspended in a fluid held in a 

 common recijjient without partitions, there was considerable in- 

 tensity of Galvanic action. This shows that, independently of 

 any power of conducting electricity, there is some movement in 

 the solvent fluid which tends to carry forward the Galvanic prin- 

 ciple from the copper to the zinc end of the scries. I infer that 

 electro-caloric is communicated in this case by circulation, and 

 that in non-elastic fluids the same diiificulty exists as to its retro- 

 cession from the positive to the negative end of the series, as is 

 found in the downward passage of caloric through them. 



It ought to be mentioned, that the connecting wire should be 

 placed between the heterogeneous surfaces before their immersion, 

 as the most intense ignition takes place immediately afterwards. 

 If the connexion be made after the plates are immersed, the 

 effect is much less powerful; and sometimes after two or three 

 immersions the apparatus loses its power, though the action of 

 the solvent should become in the interim much more violent. 

 Without any change in the latter, after the plates have been for 

 some time suspended in the air, they regain their efficacy. I had 

 observed in a Galvanic pile of three hundred pairs of two inches 

 sfjuare, a like consequence resulting from a simultaneous immer- 

 sion of the whole*. The bars holding the plates were balanced 

 by weights, as window sashes are, so that all the plates could be 

 very quickly dipped. A platina wire. No, IS, was fused into a 

 globule, while tlie evolution of potassium was demonstrated by a 

 rose-coloured flame arising from some potash which had been 

 placed iietween the poles. The heat liowever diminished in a 

 few seconds, though the greater extrication of hydrogen from the 

 plates indicated a more intense chemical action. 



Agreeably to an observation of Dr. Patterson, electrical ex- 



• Sccl'latc III. fig. o. 



O 2 citemcnt 



