Ring—The Suspension of Solids in Fluids. 281 
that pressure is very great, and this has been interpreted as 
meaning that the viscosity has decreased with increased pres¬ 
sure. 
It is a significant fact in this connection that all determina¬ 
tions of the viscosity of water and of air by the capillary or 
transpiration method have given results which are lower than 
those derived from observations of oscillating disks or spheres 
or swinging pendulums. But this relation of determined 
values is what should be expected did the flow of fluids through 
capillary tubes and porous media increase faster than the pres¬ 
sure. The extremely careful work, therefore, which has been 
done to determine the viscosity of fluids it appears, may be 
legitimately placed in evidence in support of the view that, the 
flow of air and water through capillary tubes and through 
sands, soils and rock may increase more rapidly than the pres¬ 
sure. 
Were it admissible to suppose that there is a stationary or 
comparatively stationary film of fluid adhering to the wails of 
capillary tubes through which water or air is flowing and that 
this layer could become thinner as the pressure is increased,, 
then this change would have the effect of increasing the effec¬ 
tive diameter of the tube or pore, and thus of allowing the flow 
to increase faster than the pressure, while at the same time the 
viscosity might remain unchanged and yet appear to decrease 
with the pressure. 
We are inclined, therefore, to accept the fact of the flow 
through capillary tubes and pores increasing faster than the 
pressure as further proof that solids immersed in fluids retain 
about and upon them a layer of appreciable thickness which is 
restained by, or compelled to move with them, according as 
one or the other medium is in motion. Since a tube having a 
diameter not exceeding .05 millimeter permits a flow of water 
which increases more rapidly than the pressure, it appears that 
narrower tubes possess the power to prevent the formation of 
vortex or cross currents which impede flow, and it appears 
reasonable to ascribe this prevention to the specific mutual at¬ 
traction of the solid and fluid whose restraining influence 
