PRESIDENTIAL ADDRESS. 435 
To give emphasis to this view, I have advocated! the restriction of 
the name water to the liquid mixture and have proposed that the simple 
molecule represented by the symbol OH, be termed Hydrone. The generalised 
expression 
(H,0), + 20H, 
= 
may be considered to be representative of the state of equilibrium: in water— 
that is to say, of the character of the change which water undergoes when 
the conditions are varied either physically or by dissolving substances in it— 
in the sense that it pictures the resolution of the more complex into simpler 
forms and vice versd, without, however, taking into account the variety of 
molecular forms (x, x, x . . . .) which are present. 
It is probable that the agreement between ‘theory’ and practice on 
which reliance has been placed, particularly in interpreting osmotic pheno- 
mena, is more often than not only apparent and fictitious and but the outcome 
of counterbalancing effects which have been left out of account. We are too 
prone to believe in constants; we need to“remember that, except perhaps 
in the case of the perfectly gaseous state, constants are dependent variables. 
To take an example, it is assumed that glucose and cane sugar produce 
like osmotic effects when used in equivalent proportions ; indeed, it has been 
the fashion of late years to treat non-electrolytes as harmless neutrals: in 
point of fact they differ as much in behaviour as do electrolytes and such a 
conclusion must be viewed with the gravest suspicion. Recently Dr. Eyre 
and I have been able to show that three substances so similar as methylic, 
ethylic, and propylic alcohols produce effects in precipitating salts from 
solution which are markedly different, propylic alcohol being the most effec- 
tive although the least soluble. It is clear that the precipitant does not act, 
in direct competition with the salt, mainly by itself combining with and 
withdrawing water, but that it promotes the dissociation of water by the 
mechanical interposition of its molecules: in fact, that the ‘ dehydrating ’ 
power of the water is enhanced owing to the increase in the proportion of 
simple molecules in the liquid which is conditioned by the presence of the 
solute. 
The same effect is obvious when the reduction of the electric conductivity 
of a salt such as potassium chloride by equivalent quantities of the three 
alcohols is considered. This amounts to about 6 per cent. in the case of 
methylic, 12 in that of ethylic, and 17 in that of propylic alcohol; the 
reduction effected by glucose, however, amounts to about 27 and that 
effected by cane sugar to no less than 42 per cent. In these two latter cases 
the amount of water actually withdrawn from the solution by the sugar is 
probably considerable and the ‘ mechanical effect’ of the solute is there- 
fore exercised in a more concentrated solution—more concentrated, that is 
to say, than those in which the alcohols act. If, therefore, solutions of 
glucose and cane sugar of equivalent strenyth produce like osmotic effects, 
it is because unperceived compensating factors are at work in the solutions 
which in algebraic sum have the same aggregate influence. 
To explain the effect produced by substances which give rise to con- 
ducting solutions when dissolved in water (acids, alkalies, and salts), it is 
necessary to consider the special nature of the changes which may be sup- 
posed to attend dissolution in such cases. Why, it may be asked, is an 
aqueous solution of hydrogen chloride a conductor whilst that of alcohol 
is a non-conductor? I believe the answer to be that it is because, in the 
former case alone, the two components of the solution are reciprocally dis- 
tributed ; that it is because two correlative systems— 
H H 
HOY = and_—s HO 
\ou Not 
1 Roy. Soc. Proc., 1908, 81, 80; Science Progress, January 1909. 2 
FF 
