the spectroscope, been watched in its ascent. On
September 7, 1871, Young examined at noon a vast hydrogen cloud 100,000
miles long, as it showed to the eye, and 54,000 high. It floated
tranquilly above the chromosphere at an elevation of some 15,000 miles,
and was connected with it by three or four upright columns, presenting
the not uncommon aspect compared by Lockyer to that of a grove of
banyans. Called away for a few minutes at 12.30, on returning at 12.55
the observer found--

"That in the meantime the whole thing had been literally blown to shreds
by some inconceivable uprush from beneath. In place of the quiet cloud I
had left, the air, if I may use the expression, was filled with flying
debris--a mass of detached, vertical, fusiform filaments, each from 10"
to 30" long by 2" or 3" wide,[643] brighter and closer together where
the pillars had formerly stood, and rapidly ascending. They rose, with a
velocity estimated at 166 miles a second, to fully 200,000 miles above
the sun's surface, then gradually faded away like a dissolving cloud,
and at 1.15 only a few filmy wisps, with some brighter streamers low
down near the photosphere, remained to mark the place."[644]

A velocity of projection of _at least_ 500 miles per second was, by
Proctor's[645] calculation, required to account for this extraordinary
display, to which the earth immediately responded by a magnetic
disturbance, and a fine aurora. It has proved by no means an isolated
occurrence. Young saw its main features repeated, October 7, 1881,[646]
on a still vaster scale; for the exploded prominence attained, this
time, an altitude of 350,000 miles--the highest yet chronicled. Lockyer,
moreover, has seen a prominence 40,000 miles high shattered in ten
minutes; while uprushes have been witnessed by Respighi, of which the
initial velocities were judged by him to be 400 or 500 miles a second.
When it is remembered that a body starting from the sun's surface at the
rate of 383 miles a second would, if it encountered no resistance,
escape for ever from his control, it is obvious that we have, in the
enormous forces of eruption or repulsion manifested in the outbursts
just described, the means of accounting for the vast diffusion of matter
in the solar neighbourhood. Nor is it possible to explain them away, as
Cornu,[647] Faye,[648] and others have sought to do, by substituting for
the rush of matter in motion, progressive illumination through electric
dischar