CRYSTALLIZATION PRODUCED BY VOLTAIC ACTION. 417 
This remarkable fact can only be observed with an electrical apparatus of 
very feeble tension, inasmuch as, when this tension possesses a certain 
energy, the dissolved metal goes always to the negative pole, be the wire 
or plate of metal which is plunged into the solution what it may. 
To what cause are we, in this instance, to ascribe the predilection 
of a metal, combined with an acid, for a plate of the same metal ? The 
force of cohesion, whatever it may be, is the only influence to which 
it can be attributed; for that principle must be supposed to act with 
greater force on similar than on dissimilar molecules. In this case, the 
force of cohesion added to that of the electric current should determine 
the precipitation. It must not be forgotten however that the chemical 
action of the solution on the positive wire is also powerfully conducive 
to the production of the general effect. 
If it be desired to obtain continuous effects with the thermo-electric 
apparatus, the copper ring into which the platina ring is passed must 
be renewed from time to time, because, at a certain stage of the experi- 
ments, the copper being entirely oxidized, the continuity is broken and 
the electro-chemical effects consequently cease. 
An apparatus with a platina and an iron wire has not sufficient action 
to produce decompositions. This negative effect is undoubtedly to be 
attributed to the singular electric properties of iron. 
The apparatus we are about to describe is intended to produce slow 
and continuous electro-chemical actions. 
Take two small glass jars (fig. 1); having poured into the one some 
Fig. t. 
nitric acid, and into the other some potash dissolved in water, we esta- 
blish a communication between them by means of a bent glass tube 
filled with potters’ clay moistened with a solution of nitrate of potash or 
chloride of sodium, and then plunge into each liquid a plate of plaftna 
fixed to the extremity of a wire of the same metal. At the free end of 
