hereafter note with more detail, the air next the surface of the earth
is moving in toward a kind of chimney by which it escapes to the upper
regions of the atmosphere. A study of cyclones and tornadoes, or even
of the little air-whirls which in hot weather lift the dust of our
streets, shows that the particles of the atmosphere in rushing in
toward the centre of upward movement take on the same whirling motion
as do the molecules of water in the basin--in fact, the two actions
are perfectly comparable in all essential regards, except that the
fluid is moving downward, while the air flows upward. Briefly stated,
the reason for the movement of fluid and gas in the whirling way is as
follows: If every particle on its way to the centre moved on a
perfectly straight line toward the point of escape, the flow would be
directly converging, and the paths followed would resemble the spokes
of a wheel. But when by chance one of the particles sways ever so
little to one side of the direct way, a slight lateral motion would
necessarily be established. This movement would be due to the fact
that the particle which pursued the curved line would press against
the particles on the out-curved side of its path--or, in other words,
shove them a little in that direction--to the extent that they
departed from the direct line they would in turn communicate the
shoving to the next beyond. When two particles are thus shoving on one
side of their paths, the action which makes for revolution is doubled,
and, as we readily see, the whole mass may in this way become quickly
affected, the particles driven out of their path, moving in a curve
toward the centre. We also see that the action is accumulative: the
more curved the path of each particle, the more effectively it shoves;
and so, in the case of the basin, we see the whirling rapidly
developed before our eyes.

In falling in toward the centre the particles of star dust or vapour