1887.] Capillary Action in some Chemical Decompositions. 69 
the liquid, it will decompose into water and a solution of oxygen, 
as distinct from gaseous oxygen. From analogy we might expect 
this decomposition to be limited by the amount of oxygen in 
solution. Water dissolves a certain amount of oxygen, and so long 
as it does so without the help of external energy there will be a 
dissipation of energy in the action. When the water is “saturated” 
it requires an external source of energy, such as is supplied for 
instance by an increase of pressure, to make it dissolve more. We 
might expect then that the spontaneous decomposition of hydrogen 
peroxide into water and a solution of oxygen would cease as soon 
as the work needed to make the water dissolve more oxygen 
exceeded the availability of the store of energy in the hydrogen 
peroxide. If the amount of oxygen brought into solution in this 
way exceeded the quantity required to “saturate” the water, at the 
particular temperature and pressure to which it is subject, a slow 
evaporation of oxygen from the surface would go on, and a gradual 
spontaneous decomposition of the peroxide. This would be much 
increased if the surface at which the evaporation could occur were | 
increased by the introduction of bubbles of air into the interior 
and the passage of such bubbles through the liquid. Such action 
may be sufficient to account for the slow decomposition of a 
neutral solution of hydrogen peroxide which occurs in glass vessels. 
Each bubble of gas seems to form at the surface of the glass and 
as it rises through the liquid visibly expands taking up oxygen as 
it goes. But it is insufficient to account for the very ditferent 
degrees of activity shewn by different substances in promoting 
decomposition; and for the very remarkable effect of a minute 
quantity of acid in increasing the stability of the compound. 
The action by which solids condense on their surfaces films of 
the gases in which they are immersed occurs with liquids too. 
The changes in the surface tension of mercury, used as an electrode, 
according to the nature of the gas disengaged upon it are remark- 
able enough. Also the miscibility of two liquids depends on the 
surface tension between them. Quincke has laid down the 
proposition that if the surface tension between two fluids be zero 
they are miscible in all proportions, and no drops or bubbles of the 
one form in the interior of the other. Moreover it appears that 
the surface tension between them is the smaller according as the 
fluids are more miscible (Poge. CXxx1x. 87). The solution of 
one fluid in another becomes then closely connected with, even 
if it be not a result of, capillary action; and the capillary constant 
for a mixture is well known to be different, in general, from the 
constant for either component. 
We may compare the solution of one fluid A in another B to a 
wetting of the internal surface of B by A. Indeed if distilled 
water containing air in solution be mixed quickly with strong 
