NOTE ON THE IRON-TURNINGS CELL. 
3 
the middle of the liquid, but only at the sides of the containing 
vessel ; also that their formation may be almost indefinitely 
retarded by carefully getting rid of all adherent air. This is 
what we should expect from the consideration that the initiation 
of a bubble of vapour where no bubble of air already exists in- 
volves the overcoming of cohesion forces or adhesion forces across 
the surfaces which are to form the opposite sides of the bubble. 
When a solid is dropped into a liquid, it generally takes down 
with it air adhering to the surface, as is shown by the immediate 
appearance of air bubbles when the atmospheric pressure is 
removed by an air pump. The manner in which the air is carried 
down ma}’’ be readily understood from the following considera- 
tions : — Every solid surface, however smooth it may appear to 
the naked eye, reveals under the microscope many scratches and 
cavities. Time is required for a liquid to spread over the surface 
of a solid, even when it is wetted by the liquid, and the spread of 
a liquid such as water may be indefinitely hindered by the 
presence of greasy particles ; so that if the air is to be completely 
expelled from the crevices, the solid must be so slowly immersed 
as to give the liquid time to spread over the sides of each cavity, 
and it will generally happen, owing to the rapidity of immersion, 
that the liquid will flow over the mouths of many pores and 
scratches without spreading into them so as to expel the air, 
which will thus be entrapped in minute bubbles.^ 
The same entrapping of air will take place on the sides of a 
vessel into which a liquid is poured, and there can be little doubt 
that the surfaces of the fragmentary matter at which bubbles 
appear contain minute invisible bubbles of air entrapped in this 
way. 
^ The influence of roughness of surface in thus retarding the 
w’etting of an immersed solid is fully discussed in a paper communi- 
cated by the author to the Eoyal Society and printed in abstract 
Proc. Boy. Soc., vol. 34, 1882. 
