290 M. De la Rive's Researches 



gave scarcely any signs of tension to an electroscope. It is the 

 same with dry piles, in which the humidity with which the 

 paper is always more or less impregnated performs the office 

 of conductor, and they present in a much more obvious man- 

 ner all the phenomena produced by piles of zinc and copper, 

 charged with pure water. I must not omit to say that, in 

 operating with the last-mentioned piles, care must be taken 

 completely to insulate all the parts of which they are composed, 

 by uniting their pairs by glass rods covered with lac-varnish 

 instead of wooden rods, and by employing as troughs glasses 

 insulated by means of supports of resin. 



Dynamic effects of the pile. — For investigating the dyna- 

 mic effects of the pile, I have employed either a galvanic 

 multiplier, Breguet's metallic helix, which I placed in the 

 circuit to appreciate the heat developed by the current, or a 

 very sensible apparatus intended to measure the gases result- 

 ing from the decomposition of the water by the current in 

 passing through an acid solution. In my memoir I have given 

 a detailed description of this apparatus, and of the manner in 

 w T hich I employed it. 



In the first place, it is evident that whatever be the dyna- 

 mic effect to be produced, when the number of pairs is con- 

 stant, the quantity of electricity disengaged in a given time, 

 and consequently the intensity of the effect produced, increase 

 in proportion to the increased extent of the surface of each 

 pair. This increase, however, is far from being the same for 

 the different classes of effects, as shall be shown elsewhere. 

 But the true question which is here to engage our attention is 

 this : A surface attacked by a certain solution, and a corre- 

 sponding surface of a metal less attacked, or not at all, being 

 given, what is the number of pairs to be formed of them to 

 produce the most considerable dynamic effect? At the first 

 glance the reply does not appear doubtful. A single pair must 

 be made of them, for, according to our theory, the quantity 

 of electricity which circulates through the conductor is always 

 equal to that developed in a single pair; the electricities of the 

 interior pairs reciprocally neutralizing each other. But this 

 theoretical reply is only true so far as the conductor which 

 unites the two poles can be considered as perfect ; thus it is 

 true with relation to all the dynamic effects that can be pro- 

 duced by employing, as a conductor, a copper wire of suffi- 

 cient thickness not to be heated by the current. It is true 

 also in a less degree, that is, two or three pairs must be formed 

 of the given surfaces to produce the maximum of effect, when 

 the connecting wire is, either from its nature or its dimen- 

 sions, a worse conductor, in which case the Wire becomes 



