, the vessel _c_ was made to
contain a solution of caustic potash in place of acid, still the same
results occurred. Decomposition of the iodide was effected freely, though
there was no metallic contact of dissimilar metals, and the current of
electricity was in the _same direction_ as when acid was used at the place
of excitement.

885. Even a solution of common salt in the glass _c_ could produce all
these effects.

886. Having made a galvanometer with platina wires, and introduced it into
the course of the current between the platina plate and the place of
decomposition _x_, it was affected, giving indications of currents in the
same direction as those shown to exist by the chemical action.

887. If we consider these results generally, they lead to very important
conclusions. In the first place, they prove, in the most decisive manner,
that _metallic contact is not necessary for the production of the voltaic
current._ In the next place, they show a most extraordinary mutual relation
of the chemical affinities of the fluid which _excites_ the current, and
the fluid which is _decomposed_ by it.

888. For the purpose of simplifying the consideration, let us take the
experiment with amalgamated zinc. The metal so prepared exhibits no effect
until the current can pass: it at the same time introduces no new action,
but merely removes an influence which is extraneous to those belonging
either to the production or the effect of the electric current under
investigation (1000.); an influence also which, when present, tends only to
confuse the results.

889. Let two plates, one of amalgamated zinc and the other of platina, be
placed parallel to each other (fig. 74.), and introduce a drop of dilute
sulphuric acid, _y_, between them at one end: there will be no sensible
chemical action at that spot unless the two plates are connected somewhere
else, as at PZ, by a body capable of conducting electricity. If that body
be a metal or certain forms of carbon, then the current passes, and, as it
circulates through the fluid at _y_, decomposition ensues.

890. Then remove the acid from _y_, and introduce a drop of the solution of
iodide of potassium at _x_ (fig. 75.). Exactly the same set of effects
occur, except that when the metallic communication is made at PZ, the
electric current is in the opposite direction to what it was before, as is
indicated by the arrows, which show the courses of the currents (667.).

891. Now _both_