628 
Proceedings of the Boyal Society 
therefore, if we take a mixture of 3 parts of water and 1 part of 
alcohol and add water to it, we get an expansion. In cases where 
I have drawn the curves for two different temperatures, the effect 
of increase of temperature is to diminish the amount of contraction. 
On the right hand side of the diagram the volume of 1 gramme 
of the salt in cubic centims is placed at the extremity of each of the 
curves, and if it be desired to obtain the increase of volume of a 
given weight of water on adding to it 1 gramme of a salt, it is only 
necessary to subtract the contraction from this number, and you get 
the increase of volume. From this it will be observed that Dalton’s 
theorem — that the volume of water was not increased by dissolving 
in it an anhydrous salt capable of taking up water of crystallisation 
— does not hold good in any of the cases taken, except that of 
sodium carbonate ; and there it only holds good provided the salt be 
dissolved in not less than twenty times its own weight of water. 
P.S . — Since writing the above I find that in 1846 a paper 
was published by John Joseph Griffin, on the “ Constitution of 
Aqueous Solutions of Acids and Alkalies,” in which he points out 
that the volume occupied by a substance in solution varies with the 
quantity of water in which it is dissolved, thus controverting 
Dalton’s theorem. In this paper he gives contractions of the com- 
mon acids and alkalies, and also of K 2 C0 3 , FTa 2 C0 3 , NH 4 C], 
and MgS0 4 + 7H 2 0 (Memoirs of the Chemical Society , vol. iii. 
p. 155). 
5. On the Increase of Electrolytic Polarisation with Time. 
By Mr W. Peddie. 
The law of variation of the electromotive force of polarisation 
with current- density has been very thoroughly investigated. It is 
found that it may be represented by an equation of the form 
F = a- be~ cx , 
where x is the current-density, and a, b, c are constants. So for 
very feeble direct currents the reverse polarisation current is pro- 
portional to the current-density ; but for strong currents it cannot 
be expressed by an algebraical formula as a function of the current- 
