ON ELECTROLYSIS IN ITS PHYSICAL AND CHEMICAL BEARINGS. 361 



For stronger solutions x diminishes ; and this means simply that kjm falls off, 

 just as Kohlrausch's curves more instructively show. 



§§12 and 13. Discussion of the Tables. 



§ 14. Influence of Temperature on Conductimty. — Dilution and heating exert a 

 similar influence on molecular conductivity. 



Chapter 111, — Theory. 



§ 15. On Conductivity considered as a Function of Concentration. — Kohlrausch 

 and most authors suppose that kjm is constant for extreme dilution ; ' nevertheless 

 he seems to make the statement with a certain reserve, for in another passage he 

 Bays, " This numher [gramme-molecules per cc] can, according to aU the experi- 

 ments, be put proportional to the conductivity of attenuated solutions, provided 

 extreme attenuation be excepted." Nevertheless it is not difficult to prove that such 

 a proportionality follows from the principles postulated in his -work : and that pre- 

 cisely for solutions of extreme attenuation.' 



Then follows a proof, based on Kohkausch's surmises and Hittorf s hypotheses, 



(1) That kjm is necessarily constant ; 



(2) That luhen tiuo or moi-e salts are dissolved, the conductivity of the whole is 



the sum of the separate conductivities ; and 



(3) That the conductivity of a solution equals the sum of conductivities of 



salt and solvent. 



To all this it is necessary and sufficient to remark that the surmises of a philo- 

 sopher, which are to some extent upset by his own experimental results and 

 accordingly by him stated ' with a certain reserve,' afford a very insecure basis for 

 an elaborate proof, and for deductions therefrom.^ 



If it be once granted — 1st, that the conductivity of the water in solution is, 

 under all circumstances, nil, or extremely small ; 2nd, that every atom of a .salt 

 added conducts equallj^, and independently of every other ; — the laws stated are 

 pretty obvious without further proof. But it is of little use attempting to prove 

 these laws by begging the question. 



§ 16 contains statements numbered (4) and (5j, viz., If the three laivs are 

 not true, it must be because of chemical action between the substances. 



§ 17. Hydrates (i.e., hydrogen compomids, like acids and bases) are peculiar ; 

 either because they dissolve glass, or because impurities contained in distilled 

 water act upon them chemically, and alter them. 



§ 18. Statement No. 6. — The ' exponent of dilution ' is less than 2 for salts, 

 greater than 2 for hydrates. 



§ 19. Exce]jtions. 



§ 20. Nature of the Resistance of Electrolytes. — A hypothetical discussion of 

 the friction between atoms, the mode in which ions rub against other molecules, and 

 of the amount of rotation they produce in them ; with three conclusions, numbered 

 respectively — 



(7) The resistance of a solution is greater as the internal friction is greater. 



(8) The resistance is greater as the ions are more complex. 



(9) The resistance is greater as the molecular weight of the .lolvent is greater. 



§ 21. Properties of Solutions of Normal Salts. — Discussion of some results of 

 Hittorf, and four statements — 



(10) Salts lohich are able to form double salts are most likely to form molecular 



complexes. 



(11) Aqueous solutions contain the electrolyte dissolved, at least partially, in 



the form of molecular complexes. 



(12) Dilution diminishes complexity towards an asymptotic limit. 



(13) The limit toivard which the complexity of a dissolved nortnal salt 



approaches at extreme attenuation is the same for all nonnal salts. 



I feel that this brief analysis is not quite fair to the contents of these last 

 sections, which are ingenious and interesting; but I scarcely think a detailed 



» See § 5 of letter on p. 386. 



