plunges on its way quite invisible to us unless we place our eyes in
its course. In other words, to be visible, light must enter the eye.
(A concentrated beam was passed through an empty tube, and then
ordinary air let in.)

The other test, that of Mr. Aitken, depends on the condensation of
steam. When a jet of steam finds itself in dusty air, it condenses
around each dust particle as a nucleus, and forms the white visible
cloud popularly called steam. In the absence of nuclei Mr. Aitken has
shown that the steam cannot condense until it is highly
supersaturated, and that when it does it condenses straight into
rain--that is, into large drops which fall. The condensation of steam
is a more delicate test for dust than is a beam of light. A curious
illustration of the action of nuclei in condensing moisture has just
occurred to me, in the experiment--well known to children--of writing
on a reasonably clean window-pane with, say, a blunt wooden point, and
then breathing on the glass; the condensation of the breath renders
the writing legible. No doubt the nuclei are partially wiped away by
the writing, and the moisture will condense into larger drops with
less light-scattering power along the written lines than over the
general surface of the pane where the nuclei are plentiful, and the
drops therefore numerous and minute. Mr. Aitken points out that if the
air were ever quite dustless, vapor could not condense, but the air
would gradually get into a horribly supersaturated condition, soaking
all our walls and clothes, dripping from every leaf, and penetrating
everywhere, instead of falling in an honest shower, against which
umbrellas and slate roofs are some protection. But let us understand
what sort of dust it is which is necessary for this condensing
process. It is not the dust and smoke of towns, it is not the dust of
a country road; all such particles as these are gross and large