120 PROFESSOR JOHN GIBSON ON 
concentrated ultramaximal solutions it favours abstraction of water and changes which 
are accompanied by the formation of water, such as esterification. These different pro- 
perties may all be summed up by simply stating that solutions of hydrochloric acid 
tend to gain in specific conductivity. 
Saturated solutions of certain chlorides may be used as indicators of this tendency. 
If, for example, a small quantity of a solution of hydrogen chloride be added to a 
saturated solution of common salt containing a few minute crystals of salt in suspension, 
either some of the undissolved salt will dissolve, or else some salt will be thrown out of 
solution. It was found that precipitation occurs only when the hydrochloric acid added 
is in ultramaximal solution. The same condition holds for precipitation by hydrochloric 
acid of the chlorides of potassium, ammonium, and rubidium from their saturated 
solutions.* Premaximal solutions of hydrochloric acid may be distinguished in this 
way from ultramaximal solutions even when their concentration differs from the 
maximal solution by only +1 per cent. HCl. Thus, precipitation was not observed 
with 17 per cent. HCl or less, and invariably observed with 19 per cent. and more. 
Ture Benaviour oF AQueous SoLtutions oF HypRoGEN CHLORIDE TOWARDS DISSOLVED 
OxYGEN AND DISSOLVED CHLORINE RESPECTIVELY. 
The behaviour of solutions of hydrogen chloride towards dissolved chlorine on the 
one hand, and towards dissolved oxygen on the other, presents another case in point. 
Chlorine decomposes water, especially under the influence of light. The action 
is partially illustrated by the equation 
2H,0 + 2Cl, —= 4HCl + 0,. 
This reaction is a reversible one, or, in other words, under certain conditions dissolved 
oxygen and hydrogen chloride react so as to produce free chlorine and water. Current 
theories do not enable us to predict the concentration at which equilibrium would be 
established. Assuming a tendency towards increased conductivity, the progress of the 
reaction in either direction should be associated with a gain in specific conductivity. It 
is owing to this tendency towards increased specific conductivity that concentrated ultra- 
maximal solutions of hydrogen chloride, after long keeping in partially filled clear glass 
bottles, contain free chlorine, due to the oxidation of the hydrogen chloride. Numerous 
solutions were examined, but free chlorine was never found to persist in solutions which 
had been diluted to below 18 per cent. HCl, ze. to below the concentration corresponding 
to maximum specific conductivity. So long as the solution is premaximal, free chlorine 
disappears from it, with formation of hydrogen chloride and free oxygen, the 
solution thereby gaining in specific conductivity. On the other hand, when the hydro- 
chlorie acid is in ultramaximal solution, hydrogen chloride is oxidised by the oxygen 
* Gipson and Denison, Proc. Roy. Soc. Hdin., vol. xxx. p. 562, 1909-10. 
