It was discovered in 1833 by Michael Faraday, Professor of Chemistry in the Royal Institution in London, that if an electric current be passed simultaneously through differentsolutions, the weights of metals deposited or of elements or groups of elements liberated are proportional to their equivalents ( see p. 1 5). If the same current be passed, for example, through a solution of dilute sulphuric acid, copper sulphate, and iodide of potassium, each contained in its own vessel, provided with plates of platinum or some other unattackable metal dipping into the solution, for every gram of hydrogen evolved from the kathode in the vessel containing sulphuric acid, 8 grams of oxygen are evolved from the anode ; 32.7 grams of copper are deposited on the kathode dipping into the copper solution, while 8 grams of oxygen rise in bubbles from the anode ; and lastly, 127 grams of iodine are liberated from the anode in the vessel containing potassium iodide, 1 gram of hydrogen rising from the kathode. The evolution of hydrogen instead of the deposition of potassium is due to the fact that the metal potassium is unable to exist in presence of water, but immediately displaces its equivalent of hydrogen. All these numbers are in the proportions of the equivalents of the elements. And without the liberation of these elements no current passes. The elements may, therefore, in a certain sense, be said to convey the electricity ; and as the same quantity of electricity passes through each solution, liberating equivalents of the elements in each case, it would appear that the same quantity of electricity is conveyed by quantities of elements proportional to their equivalents. The equivalent of an element, it will be remembered, is the weight of the element which can combine with or replace one part by weight of hydrogen ; it may be identical with, or it may be a fraction of the atomic weight. In the instances given above, the equivalents of iodine and of potassium are identical in numerical value with their atomic weights ; but those of oxygen and of copper, 8 and 32.7, are half their atomic weights, which are respectively 16 and 63.4. It would follow, therefore, that an atom of copper or of oxygen is capable of conveying a quantity of electricity twice as great as that conveyed by an atom of hydrogen or of iodine.

But how is it known that the atoms " convey " quantities of electricity ? Must they be imagined as like boats, taking in their load of electricity at one pole, and ferrying it over to the other, and there discharging ? It was at one time held that the process rather resembled the method of loading a barge with bricks, where a row of men, who may stand for the atoms, pass bricks, representing the electricity, from one to the other. But it was proved by Hittorf that the charged atoms actually travel or "migrate" from one pole to the other, carrying with them their electric charges. And the charged atoms, for which the name "ions," or "things which go," was devised by Faraday, do not always move at the same rates. The rate of motion depends on the friction which the ions undergo on moving through the water or other solvent in which the salt is dissolved. This friction is different for different ions ; it also depends on the particular solvent employed ; and it is diminished if the temperature is raised. The force which impels the ions is the same as that commonly known as electric attraction and repulsion ; the negatively charged atoms or "kations" being repelled from the negative and attracted by the positive electrode dipping into the solution, while the positively charged atoms or "anions" are repelled by the anode and attracted by the kathode.

When the anions touch the kathode, they are discharged ; and similarly, when the kations touch the anode, they lose their charge. And for every anion discharged, a kation must simultaneously lose its charge. The result of this is that the number of anions remaining in solution must always be equivalent to the number of kations. It need not always be the same, for it is possible for a kation like copper to carry twice the charge of an anion like chlorine; but the number of " electrons," or electric charges, must always be the same, although some ions are capable of carrying more than one electron. There can never, therefore, be an excess of, say, copper ions in solution ; for they are always balanced by the requisite number of anions. Thus, if the solution be evaporated, the remaining salt has its usual composition ; though, of course, there is less of it than if none had been decomposed.