352 REPORT— 1887. 



one, but it is the application wliicli he makes of it in the consideration of all 

 manner of chemical reactions ; he attempts, in fact, an Electrolytic Theory of 

 Chemistry. 



Professor Armstrong, on the other hand, holds, provisionally at any rate, quite 

 heterodox views as to the nature of electrolysis, which, so far as I understand 

 them, appear to be these : — 



Initially a salt-solution exhibits no trace of dissociation, there are no free or 

 semi-free atoms to be acted on electrically, but so soon as an E.M.F. is applied to 

 it a locomotion of the molecules past each other begins, as evidenced by the known 

 occurrence of electric endosmose. By this process every salt molecule is brought 

 within range of a water molecule as they slide past each other, and the residual 

 affinity of some constituent of each of two molecules straining at each otber under 

 these conditions, superadded to the strain already set up by the applied E.M.F., is 

 sufficient to effect disruption of the molecule, whose separated atoms then travel 

 opposite ways to the electrodes carrying their charges with them, and conduction 

 occurs in the manner ordinarily assumed. For instance, in a solution of PICl in 

 H„0, the is straining at the CI, and this force, as the molecules flow past each 

 other, is sufficient to assist the applied E.M.F. to produce disruption and inter- 

 change, i.e., to bring about the same result as the Williamson-Clausius dissociated 

 condition ordinarily supposed to exist before the action of E.M.F. 



If, however, the molecules in the liquid are all complex molecules, or aggregates 

 of a large number of atoms, hanging together by the residual affinity of each, these 

 residual affinities are in the complex so nearly satisfied that they have little or no 

 further power to act on a water or other molecule ; and they thus resist being 

 broken up, and refuse to conduct a current. A liquid composed wholly of such 

 complex aggregates is thus not an electrolyte ; and Armstrong calls it a pseudo- 

 dielectric. 



A liquid which contains among a large proportion of such aggregates a few 

 simple molecules here and there is an electrolyte but a very badly conducting one. 

 Its conductivity increases as the aggregates get brpken down, whether by heat or 

 by dilution. 



It will be observed that this hypothesis does not dispense with dissociation ■, it 

 only denies dissociation previous to the application of E.M.F. So soon as E.M.F. 

 is applied, mutual action between the molecules, assisting the strain caused by the 

 E.M.F. itself, produces the very state of dissociation postulated by all physicists as 

 necessary for actual electrolytic conduction. 



Neither does the hypothesis dispense with a dissociating power of the solvent ; 

 it only limits its power to the breaking down of complex molecules, instead of 

 allowing it to break up the simple molecules themselves. It is not supposed able 

 to do this latter until aided by applied E.M.F. 



To summarise. — The orthodox view supposes fully combined molecules, 

 whether aggregates or not, to be undecomposable by any moderate E.M.F. ; but it 

 supposes a certain proportion of them split up or dissociated, either actually or 

 potentially, by addition of a foreign body. Not necessarily a sohent : it pictures 

 the dissociation of water by salt quite as easily as that of salt by water. Each 

 atom while in the nascent or uncombiued state has associated with it a definite 

 electric charge, and these loose electrified atoms are thus immediately amenable to 

 the smallest directive E.M.F. 



Armstrong's view supposes complex molecules to be undecomposable by any 

 moderate E.M.F., but it imagines a certain proportion of them split up or decom- 

 posed into simpler molecules by the action of a solvent. It supposes, further, a 

 tendency or endeavour on the part of the water to split up these simple molecules 

 still more into their constituent atoms ; but it asserts that the water is unable to 

 effect this until aided by an extra strain, in the shape of an externally applied 

 E.M.F., and by the locomotive disturbance or endosmose thereby set up in the 

 liquid. 



A third view there is which must probably have been held more or less dis- 

 tinctly by several physicists, and which for several reasons usually commends itself 

 to me, viz., that all simple molecules are strongly combined, and therefore intract- 



