n representing the transport number of the anion. Two equations are thus obtained for the determination of the unknown u and v, viz.

u = (1 − n) x, v = n. λ.

Calculating by the aid of this expression the velocity of one and the same ion from the conductivity of its different compounds, satisfactory results are obtained, as Kohlrausch has shown, so long as salts of monobasic acids only are compared with one another. The following table contains the velocities of ions calculated by Kohlrausch and expressed in terms of an arbitrary standard, in which is measured in terms of the ten-millionth part of the conductivity of mercury, or the latter is set down as 107.

If these velocities are expressed in absolute terms, it is then seen that they are very small even under the influence of a strong electromotive force, and the particles move in a second through a few hundredths or a tenth of a millimetre, and consequently at a snail-like speed. It is evident that they meet with considerable opposition to their movement.

§ 104. Relation between Electrolytic Conduction and Diffusion. The intimate relationship between electrolytic conduction and the motion of the particles is shown also in the fact that those movements which take place independently of electricity vary in a similar manner. J. H. Long has proved experimentally that the velocities with which different salts of analogous composition diffuse into water stand to one another in approximately the same relation as their electrolytic con

ductivities, so that the compounds which diffuse the most easily are the best conductors. This statement is not absolutely true, but holds only for certain groups of compounds of similar composition, because in different groups the extent of the dissociation is different, and undecomposed molecules diffuse with velocities other than those of the ions.

That many salts are in reality resolved into their ions and do not, or only in part, diffuse undecomposed, is also shown by Long's observations; for the comparison of compounds containing the same anion-for example, the chlorides-has shown that the rate of diffusion is inversely proportional to the transport number of the anion; but the comparison of salts with the same cation-for example, the potassium salts --has demonstrated the rate of diffusion to be directly proportional to the transport number of the anion. This practically amounts to saying that, if two salts have the same ion in common, then the salt with the more mobile second ion is the more easily diffusible. The rate of diffusion would therefore appear to be the sum of the velocities of the ions.

§ 105. The Function of the Ions in the Production of Electric Currents. The close relationship between the electrolytic ions and the movement of electricity is seen also in the fact that electric currents are produced by the contact of unequally concentrated solutions of electrolytes simultaneously with the diffusion tending to compensate for the inequality in concentration. The intensity of the currents can be shown, both experimentally and theoretically, to be related to the velocities of the ions.

The electric currents produced ordinarily by contact of two or more metals with one another or with one or more electrolytes appear to owe their origin to the free and mobile ions. set at liberty by dissociation. The chemical affinity of the metals for the anions exercises an attraction on these. By the deposition of the anions on the metal and the giving up of their negative electricity the metal becomes so charged with electricity that the further approximation of the anions is prevented. The strength of this charge of negative electricity is greater the greater the affinity of the metal for the anion. For instance, if two metals, like copper and zinc, are

immersed in a liquid, then the metal, in this case zinc, possessing the stronger affinity will be charged more strongly with negative electricity than the other, viz. the copper. If the two metals are united by a metallic conductor, then the more strongly charged zinc will give up its negative electricity to the copper, and in return receive a charge of positive from the copper. Thus the equilibrium at the points of contact of the two metals and the electrolyte is disturbed; in consequence of the reduced negative charge of the zinc, more anion is attracted, and the increased negative charge of the copper induces a repulsion of the anion, and cations are attracted by reason of their positive charge. In addition, at the point of contact of both metals there is a separation of electricity opposite to the charge produced by the ions. Equilibrium cannot be established so long as both the metals are in contact with each other and the electrolyte. But as the ions collect more and more on the metals and cover them, the negative anion on one, and the positive cation on the other, the ions take the place of the metal and thus reverse the action completely; for the positive cation attracts the anion, and the reverse. This separation of the ions which produces a current opposed to the original is styled 'electric polarisation.'

In order, therefore, to produce a constant current the separation of the ions at the electrodes must be prevented, or in other ways made innocuous, which end can be attained by suitable choice of the electrolyte. In this way constant electric batteries can be produced. be produced. Daniell's battery is one of the oldest of this kind, and consists of a plate of copper surrounded by a solution of copper sulphate, CuSO4, and a plate of zinc immersed in dilute sulphuric acid, and separated from the copper by a porous cell. The zinc attracts to itself the anion, SO1, and repels the cation, H2, and is charged with negative electricity, which passes over to the copper on which the positive cation, Cu, collects; whilst if the copper and zinc were not in contact the copper would also be surrounded with the anion, SO,. The precipitate of copper on the copper plate leaves the latter unchanged; the zinc remains unaltered, because by combining with the anion, SO1, zinc sulphate, ZnSO4, is formed which dissolves in the water.

The combination remains, therefore, almost entirely unchanged so long as zinc, acid, and copper sulphate are present.

According to the conception, put forth recently by L. Sohncke, and developed uniformly by the use of the older representations, the source of the electric current, respecting which there has been so much discussion, is to be sought neither in the contact of the metals nor in the chemical action of the metals, but in the dissociated state of one or other of the electrolytes in contact with the metals. This completely confirms the observation made by F. Kohlrausch, that simple unmixed liquids are not as a rule electrolytes, and are therefore incapable of developing a current unless in the fused state and at high temperatures. Dissociation produced either by mixing with other liquids or by the application of heat is therefore essential to the action of electrolytes.

That in aqueous solutions hydrogen chloride exists to a large extent in a state of dissociation (cf. § 109) can be demonstrated by the depression of the freezing point (cf. § 78). Still, the mode of dissociation cannot in this case be determined; for by the electrolysis of concentrated solutions hydrogen and chlorine are the ions, whilst from dilute solutions hydrogen and oxygen are formed. Hence it would appear probable that in the first case the hydrogen chloride is decomposed into hydrogen and chlorine, and in the second case the solution contains the compound HCl + H2O, or H,ClO, the existence of which Thomsen assumes, and this is resolved into H, and HCIO, the latter yielding oxygen, O, and hydrochloric acid, HCl, at the anode.

§ 106. Dissociation a Condition Preparatory to Chemical Change. In the majority of cases it would appear that dissociation must precede chemical change; for the electrolytes which are so easily dissociated belong to the class of substances distinguished by their ability to take part in chemical actions. This ability ceases so soon as the possibility of this dissociation taking place is removed. Anhydrous hydrogen chloride, liquefied by pressure and cold, does not attack the metals, which are easily dissolved by its aqueous solutions.

This extremely interesting and remarkable phenomenon becomes perfectly clear in the light of the hypothesis that pure hydrogen chloride cannot be dissociated and remain so, because each of the separated ions must come in contact with others and be fixed by these, whilst in the aqueous solutions they would both be separated by the water, and remain apart for a short time.

The behaviour of many elements is very remarkable according as they exist in compounds which are electrolytes or non-conductors, i.e. in compounds which do not undergo dissociation. Thus, for instance, chlorine, bromine, and iodine are separated from their compounds by solutions of silver salts only when the compounds are such as easily dissociate, and these elements form the ions. The majority of organic compounds containing these halogens are either incapable of being dissociated or dissociate at high temperatures only, and then only in some cases is the dissociation such that the halogens chlorine, bromine, and iodine appear as ions. In complete agreement with these facts, the chlorine, bromine, and iodine of such compounds either do not react at all with silver nitrate or only slightly; many other compounds of these and other elements behave in a similar manner. The chlorine of chlorates and perchlorates in which the metals are the cations and the radicals CIO, and CIO, the anions, does not in solution give any silver chloride, but forms first silver chlorate and perchlorate, from which the chloride can be produced by their decomposition. The sulphates with the anion SO, in the ordinary course of things give rise to sulphates with the same anion,' and many other salts and similar compounds behave in the same way. The compounds may decompose in other ways if the manner of the dissociation, and consequently the nature of the ions, be changed by heat or by the action of other bodies.

4

6

4

If by the study of a series of compounds capable of undergoing dissociation the ions contained in them are known with any degree of certainty, the majority of the reactions of these compounds may be predicted, for the combinations and changes always result from the union of the ions with those of the other active bodies. These facts afford an explanation