68 Prof. Liveing, On the Influence of [Feb. 14, 
and the effect is to drive the drop up the solid until its in- 
creased size and weight counter- 
act this tendency. If however 
the bubble be one of air on the 
point of a solid wetted by the 
liquid in which it is immersed, 
the bubble will be convex to the 
liquid at all points, and the curva- 
ture at a wil] have a greater radius 
than at b in the principal section. 
y, 
Now the normal pressure is pro- yy 
portional to the sum of the re- YY 
ciprocals of the radii of curvature Y; Uj Ws YY 
in the principal sections, and we 
shall have the pressure at bb greater than at a when the radius 
of curvature at b in the principal plane represented in the figure 
is less than half that at a. In this case the bubble will be driven 
towards the point and its escape facilitated. 
The thinning out of the adhering film at a point, or sharp 
edge of a solid, must facilitate the rolling up of the film into 
bubbles when the solid is immersed in a fluid which has in 
contact with the solid a smaller surface tension, since it affords 
good opportunity for the second fluid to wet the solid. 
If the liquid contain in solution some gas G, other than air, 
this gas will evaporate into the bubbles of air until the pressure of 
G in the bubble attains a certain maximum depending on the 
temperature and on the amount of G dissolved in the liquid. The 
escape of the bubbles will be hastened by their expansion and 
they will remove more or less of G from the liquid under cir- 
cumstances of temperature and pressure which would not allow of 
the independent formation of bubbles of G within the liquid. For 
if we consider that the normal pressure on a bubble of G in a 
liquid is proportional to the tension of the common surface of the 
liquid and G, and inversely proportional to the radius of curvature 
of the bubble, this pressure will be enormous when the bubble is 
very minute and must in general cause any very small bubbles, if 
such should form, to be immediately re-dissolved. If G, instead of 
being a gas merely dissolved in the liquid, be a product of 
chemical decomposition, either of the liquid itself or of something 
dissolved in it, this action which prevents the separation of gas 
within the liquid must have the same effect as an increase of 
pressure in preventing decomposition within the body of the 
liquid. The effect of pressure in increasing the stability of 
compounds which give gaseous products of decomposition is well 
known. If then we have a solution, say, of hydrogen peroxide, and 
some of it decompose, as it always tends to do, within the body of 
