maxima. And he may well feel encouraged by
the fortunate combination of many experimental details into one
explanatory whole, no less than by the hopeful prospect of further
developments, both practical and theoretical, along the same lines.

What we really know about the corona can be summed up in a few words. It
is certainly _not_ a solar atmosphere. It does not gravitate upon the
sun's surface and share his rotation, as our air gravitates upon and
shares the rotation of the earth; and this for the simple reason that
there is no visible growth of pressure downwards (of which the
spectroscope would infallibly give notice) in its gaseous constituents;
whereas under the sole influence of the sun's attractive power, their
density should be multiplied many million times in the descent through a
mere fraction of their actual depth.[594]

They are apparently in a perpetual state of efflux from, and influx to
our great luminary, under the stress of opposing forces. It is not
unlikely that some part, at least, of the coronal materials are provided
by eruptions from the body of the sun;[595] it is almost certain that
they are organized and arranged round it through electro-magnetic
action. This, however, would seem to be influential only upon their
white-hot or reflective ingredients, out of which the streamers and
aigrettes are composed; since the coronal gases appear, from
observations during eclipses, to form a shapeless envelope, with
condensations above the spot-zones, or at the bases of equatorial
extensions. The corona is undoubtedly affected both in shape and
constitution by the periodic ebb and flow of solar activity, its
low-tide form being winged, its high-tide form stellate; while the rays
emitted by the gases contained in it fade, and the continuous spectrum
brightens, at times of minimum sun-spots. The appendage, as a whole,
must be of inconceivable tenuity, since comets cut their way through it
without experiencing sensible retardation. Not even Sir William
Crookes's vacua can give an idea of the rarefaction which this fact
implies. Yet the observed luminous effects may not in reality bear
witness contradictory of it. One solitary molecule in each cubic inch of
space might, in Professor Young's opinion, produce them; while in the
same volume of ordinary air at the sea-level, the molecules number
(according to Dr. Johnstone Stoney) 20,000 trillions!

The most important lesson, however, derived from eclipses is