successively encountering a vast number of
small bodies which circulate in celestial
space which reach us from all quarters with
velocities absolutely equal among themselves,
or at least very nearly equal. Moreover
(it has been concluded from the characters
presented by the diurnal variation),
the velocity in space of shooting stars
must bo greater than that of the earth in
her orbit, and but slightly different from
the velocity which would cause a comet
travelling from the depths of space to make
a near approach to the earth.

Another observed fact: At the times
when the phenomenon of shooting stars
occurs in its greatest intensity, namely,
about the 12th and 13th of November, and
the 9th and 10th of August, the shooting
stars, instead of coming indifferently from
all the regions of space, come almost all
from determinate directions. One set,
those of November, started from the
constellation of the Lion; the others, those of
August, from the constellation Perseus.
This circumstance led to the separation
of the shooting stars into two distinct
classes. One class consisted of the
regular streams which the earth periodically
encounters every year, at epochs of
the same date; those are periodical
shooting stars. The others, on the contrary,
wandering singly in space, in all possible
directions, fall in with the earth indifferently
on all sides; they are called, after
Olbers, "sporadic" shooting stars. The
shooting stars of the periodical November
flood have received the special name of
Leonides, from the constellation Lion,
whence they seem to issue; those of the
August flood, in like manner, have received
the name of Perseides.

A further step in the inquiry, was this:
M. Schiaparelli, having found the orbit
described by the swarm of the Perseides,
afterwards discovered a remarkable and
wholly unexpected agreement between it
and the orbit of a large comet observed
in 1862, which orbit is a very elongated
ellipse. This identity of the two orbits
might have been the result of pure chance,
in which case it would have been of little
importance. But a second fact of the same
kind soon showed that the idea of an
accidental coincidence must be given up. The
orbit of the Leonides was found to coincide
with that of a comet discovered in the beginning
of 1866. The hint being thus unmistakably
given, by two remarkable instances,
of the coincidence of the orbits of a swarm
of shooting stars and of a known comet,
other analogous facts were searched for. It
was speedily seen that the shooting stars of
December 10th, describe in space the same
ellipse as the famous comet of Beila, and,
moreover, that the shooting stars of April
10th, move in the orbit of the first comet
of 1861.

These results have thrown great light
upon the question of shooting stars. A
comet which follows in space the same
route as a swarm of shooting stars, must
be regarded as forming an integral part of
that swarm. It is no other than a local
concentration of the matter of the swarm
a concentration sufficiently intense to render
its mass visible, even at great distances from
the earth. It follows that shooting stars
are of the same nature as comets. They
consist of small masses of cometary matter,
which circulate in space, unperceived by
us in consequence of their diminutive size,
and only become visible when they penetrate
the earth's atmosphere. Like comets,
or at least like the less dense portion of
those heavenly bodies, they are in the state
of gas. All observers are aware that the
fixed stars are visible, without any sensible
diminution of their brightness, through the
tails of comets. Shooting stars present
the same degree of transparency, as was
plainly stated by M. Coulvier-Gravier, long
before Schiaparelli's discovery of the identity
of comets and shooting stars. "Eight
times," he wrote in 1859, "but eight times
only, have we seen the nucleus of a fixed
star of the first magnitude through a
shooting star, also of the first magnitude. If
this fact is confirmed, as I believe, it will
result that the matter which gives birth to
a shooting meteor is transparent."

We are now, therefore, enabled to form
a clear idea of the nature and cause of the
phenomenon of shooting stars, which may
be stated in the following terms:

Masses of nebulous matter, scattered
throughout the stellar spaces, and presenting
a high degree of diffusion, are brought
within the limits of our planetary system
by the paramount influence of the sun. At
the same time, whether by the same action
of the sun or of the large planets near
which they pass, they undergo a progressive
change of form, in consequence of
which they are drawn out and lengthened
into parabolic or elliptic streams or
bands. By reason of their extreme diffuseness,
the matter of which they are composed
is far from occupying the totality of
the space throughout which their diverse
portions are scattered. Instead of that,