204 REPORT—1890. 
one, or at least the same number of molecules of the salt are attached to 
the anion and cation respectively. 
There is a certain amount of evidence for the existence of molecular 
aggregates in electrolytic solutions. Dr. E. Wiedemann (‘B.A. Report,’ 
1887, p. 346) has examined the conductivity of copper-chloride solution 
at different temperatures from this point of view. The solution is 
specially interesting, because it changes colour with temperature, and the 
colour change is probably due to a change in the state of hydration. The 
conductivity increases nearly at a constant rate up to 60°, and beyond this 
point the rate rapidly diminishes, and therefore indicates that the con- 
ductivity of salts varies with their degree of hydration. 
Helmholtz (Faraday Lecture, p. 289) says that it is possible that the 
majority of molecules in SO,H, may be divided into SO, and H,, some of 
them on the other hand into SO,H and H. This would account for an 
alteration in the apparent velocity of hydrogen at different concentra- 
tions, for in the latter case some of the hydrogen would be carried 
backwards. 
Bouty,! also discussing the conduction of H,SO,, says, ‘One does not 
see how to explain a variation of this kind except by a change in the 
nature of the electrolyte (i.e. of the dissolved hydrate).’ By making the 
hypothesis which Bourgoin made, that the hydrate really decomposed by 
the current was S,0,6H,O, Bouty considers that the anomaly of electro- 
lysis, as expressed by Hittorf’s values of », and also that of conductivity, is 
explained. Hydrochloric acid is in much the same condition as sul- 
phuric acid ; it conducts as if its molecules contained three equivalents 
of basic hydrogen. Other remarks of a similar bearing might also be 
quoted.” 
The electrolysis of strong solutions of CdI, in alcohol has been ac- 
cepted on all sides as involving the decomposition of complex molecules. 
Moreover, Arrhenius, in a letter to Lodge (May 17, 1886), ‘ British 
Association Report,’ 1886, p. 311, suggests the formation of double mole- 
cules and treble molecules in concentrated solutions. 
Crompton 3 has sought to prove the relation between hydrates existing 
in sulphuric acid and the conductivity of solutions by plotting the second 
differentials of the conductivity-concentration curves, and obtaining the 
result as a series of straight lines. The line of argument is that taken by 
Mendeleef in discussing the hydrates of alcohol and sulphuric acid by 
plotting the first differential of the density-concentration curves, but it 
is pushed a stage further. The method, however, is a somewhat uncertain 
one, and has been called in question. See Pickering ‘ Zeitschr. fiir phys. 
Chem.’ vol. vi. p. 10, also ‘Chem. Soc. Journ.’ 1890, p. 64. It is liable 
to represent in a foreshortened way small irregularities of the original 
curves, which may, indeed, have a corresponding experimental basis ; 
they may also depend merely upon errors of plotting of the original 
curve. 
Whatever evidence there may be for the existence of aggregates in 
comparatively strong solutions affecting the conductivity, it must be re- 
membered that it is not clear that the electricity is carried by the aggre- 
gates. Itis possible that the solution may contain a number of dissociated 
1 Ann. de Chim. [6] 3, p. 481, 1884. 
? See for instance Bouty, C.R. 104, p. 1789, 1887; and especially for the electro- 
lysis of cadmium salts, see Werschoven, Zeitschr. f. ph. Chem. vol. 5, p. 481. 
3 Jour. Chem. Soc. 53, p. 116, 1888. 
