Septemker 27, 1900] 



NA TURE 



525 



In consequence of this mobility of these dififerently elec- 

 trified ions, it is easy to understand their chemical activity 

 in these conducting solutions, and thus these two 

 important properties of these solutions receive a 

 common explanation. No really satisfactory explana- 

 tion of how the solvent water confers on substances 

 dissolved in it this wonderful independent mobility of 

 the ions has yet been proposed. Some writers have 

 described the phenomena as if all that was needed was 

 to assert that the ions move about independently, that 

 the material, CuSO^ for example, is simply dissociated 

 into Cu and SO4, and that these ions gad about freely 

 and independently in the liquid. Such writers conse- 

 quently speak of the substance as being dissociated in 

 solution. In what is recognised as ordinary chemical 

 dissociation there is no different electrification of the con- 

 stituents into which the substance is dissociated, and 

 there is thus an essential distinction between the 

 independent mobility of electrified ions in solution and 

 what is recognised as chemical dissociation. Whatever 

 be the true account of the matter, it is almost certainly 

 very much more complicated than ordinary chemical 

 dissociation, and the action of the water is evidently of 

 the first importance. There is great difficulty in explain- 

 ing this independent mobility, on account of two things 

 which have not been satisfactorily explained as yet. In 

 the first place, it is very hard to understand why these 

 oppositely electrified ions do not combine together in 

 pairs as they would do if they were merely under the 

 action of the electric forces due to their opposite electri- 

 fication. In the second place, it is very hard to under- 

 stand where the energy comes from that is required to 

 separate these independent ions and keep them free 

 from one another. When copper sulphate is dissolved 

 in water there is very little change of temperature ; if it 

 be the anhydrous salt, there is a rise of temperature ; 

 while we would naturally expect an enormous absorption 

 of heat to account for all the energy that would be 

 required to separate the Cu and SO4 from one another. 

 This all shows how very different this phenomenon is 

 from ordinary dissociation, and in consequence this 

 peculiar action of water, and in a less degree of some 

 other solvents, has been called " ionisation." The two 

 most important properties of an ionised fluid are con- 

 ductivity for electricity, and a remarkably chemical activity 

 which has been shown to be directly proportional to the 

 conductivity. 



Besides liquids there are gaseous conductors of electri- 

 city. Gases do not usually conduct electricity at all. 

 Even under circumstances when one would naturally 

 expect them to carry electric charges on their molecules, 

 they seem quite incapable of doing so. When the 

 surface of a liquid is electrified and it evaporates, one 

 would naturally expect the escaping molecules of gas to 

 carry away with them some of the electric charge from 

 the surface of the liquid. It is not known whether an 

 electrified metal volatilising would or would not carry 

 away with it some of the superficial electric charge, but 

 ordinary liquids volatilising certainly do so to a very 

 small extent, if at all. This may be, of course, because 

 the charge is carried by superficial ions^ and it is not the 

 ions that escape, but the molecules of the liquid itself. 

 But why these extremely movable ions cannot escape 

 as a gas from the surface of the liquid is a matter still 

 requiring explanation. 



Under a great variety of circumstances, however, 

 gases are able to conduct electricity. Leaving aside the 

 spark, glow and arc discharges in gases at high 

 pressures, and the well-known discharges in gases at low 

 pressures, in all which cases there is evidently something 

 like a breaking up of the gas itself under intense electric 

 force, there are a number of cases in which a gas 

 can conduct electricity under quite feeble electric forces. 



NO. 1 61 3, VOL. 62] 



Within any space at all close to a spark discharge of any 

 kind, in flames and in the gases escaping from them, in 

 the neighbourhood of surfaces of solids illuminated by 

 ultra-violet light, in the neighbourhood of surfaces of 

 solids being acted on chemically by the gas, in a gas 

 traversed by kathode rays, and in a gas traversed by 

 X-radiations and by those various, most curious and re- 

 markable radiations that have been classed under the 

 name of Becquerel rays — in all these cases gases conduct 

 electricity, apparently quite freely. 



Under these circumstances it has been usual to de- 

 scribe a gas in this condition as " ionised," and to seek 

 for a separate and independent mobility of its ions. 

 Great success has attended these investigations. The 

 difficulties that surround ionisation in liquids are mostly 

 absent here. The ions, if left for a time to themselves, 

 do combine together in pairs, and it requires a continuous 

 and considerable expenditure of energy to keep them 

 apart. The diffusion of the independent electricities 

 has been studied, and many quantitative results that 

 were to be expected from the theory have been proved 

 to exist. 



There is, however, an important difference between 

 the conductivity of a gas and of a liquid. In the case of 

 a liquid, the electricity always travels along with matter 

 in the form of an atom or a group of atoms ; in a gas there 

 is every reason to believe that we are often dealing with 

 electric charges which, if connected with matter at all, 

 are connected with masses which are about 500 times 

 smaller than a hydrogen atom. So far no good reason 

 has been given for believing that the electric charges 

 that move about among the molecules of a gas carry any 

 matter along with them. There does not seem any 

 difficulty in supposing that the electric charge of an atom 

 can exist independently of the atom. All theories of 

 electrolysis have supposed that electric charges are 

 transferred within the liquid along with material atoms, 

 but from the liquid to the electrodes all theories have 

 supposed that the electric charges jump from the liquid 

 atom to the electrode ; and if it can jump from one atom 

 to another, there seems no reasonable objection to 

 believing in its independent existence. On account of 

 the difference between the nature of the conductivity of 

 a gas and of a liquid, it would be well to confine the 

 term ionisation to the case of conductivity due to the 

 mobility of charged atoms or groups of atoms, and to 

 call the conductivity due to the existence of these mobile 

 electric charges which are not connected with atoms by 

 another name, such as " electronisation." 



One of the most remarkable results of the study of 

 these mobile electric charges which are unconnected 

 with atoms is that only negative electric charges have as 

 yet been discovered to be free from atoms ; the corre- 

 sponding positive charges seem to be always attached to 

 atoms or groups of atoms. This has naturally led to a 

 rehabilitation of the old single fluid theory of electricity 

 in which matter plays the part of the positive fluid in the 

 old double fluid theories, and the phenomena of electron- 

 isation, so far known, certainly lend support to this 

 hypothesis. But we really know so little about the 

 subject, that it is rather too soon to form anything 

 beyond a rough working hypothesis. So long as we 

 know that there exist outstanding, second order effects 

 like gravitation, we are premature in concluding that 

 the connection between positive electrification and 

 matter proves any first order difference between posi- 

 tive and negative electricity, as it may be a second 

 order effect. 



The conductivity of gases produced by the pressure of 

 these movable electric charges is in some respects more 

 analogous to that of metals than to that of liquids. In 

 liquids an electric current is accompanied by streams of 

 matter, while in the electronised gas, so far as it has been 



