the inhabitants of Cumana, in 1799, of the
grand shower of shooting stars beheld in
1766, doubtless contributed not a little to
Olbers's belief in the periodical return of
a like exhibition every thirty–three or
thirty–four years.

But even in its reduced proportions in
the years following 1833, the November
phenomenon was not the less interesting
to study. And soon afterwards, M. Quételet
announced to the Académie of Brussels,
that the night of the 10th of August
rivalled, in respect to the number of its
shooting stars, that of the 13th of November.
The facts fully confirmed his assertion;
and the more closely they were observed,
the more importance they gave to
these periodical meteoric displays.

The first singular circumstance remarked,
was, the variation of the intensity of the
phenomenon at different epochs of the
same year. An annual variation was soon
indubitable. Afterwards, by watching
what takes place, not during the course of
an entire year, but every night, it was
found that, even in this short interval of
time, there is a manifest variation in the
frequency of shooting stars. This gives us
a diurnal variation, taking a day to mean
twenty–four hours. Moreover, although
these so–called stars are seen to shoot from
every quarter of the heavens, close examination
shows that the different quarters
do not furnish equal quantities of shooting
stars. There is also, in this respect, a
variation, which is called the azimuthal
variation. For instance, a great many
more shooting stars start from the east than
from the west: while, on the other hand,
about as many come from the north as
from the south.

The existence of these variations,
annual, diurnal, and azimuthal, was for a
long time the stumbling–block of the
astronomical, or cosmical, theory of shooting
stars; namely, the theory which attributes
the phenomenon to the earth's successively
encountering, while travelling through
space, a multitude of small bodies
dispersed in it. These variations were the
ground on which some philosophers refused
to acknowledge shooting stars to be
anything else than atmospheric meteors,
entirely originated and developed in the
atmosphere which surrounds the earth.
Thus, Humboldt, in his Cosmos, says:
"It is difficult to guess what influence a
more advanced hour of the night can
exercise on these phenomena. If it were
established that, under different meridians,
shooting stars began to be visible at a fixed
hour, we should be obliged to admit (if we
wish to maintain the astronomical theory)
the supposition improbable in itself that
certain hours of the night, or rather of the
morning, are more favourable to the
inflammation of shooting stars, and that,
during the preceding hours, some of them
remain invisible."

In fact, if the phenomenon of shooting
stars be occasioned by the earth's meeting
a multitude of small bodies dispersed in
space, what can be more natural than to
admit that these encounters take place as
much at one date as at another as much
at one hour of the night as at any other
hour of the night; in short, that the phenomenon
will occur without any periodical
variation?

M. Delaunay, however, clearly shows
that in consequence of the earth's motions
of translation and rotation, uniformity in
the appearance of shooting stars cannot
exist. On the contrary, he demonstrates
that, with the astronomical theory of
shooting stars, the annual, diurnal, and
azimuthal, variations must necessarily occur
at every locality, under the very circumstances
which are observed to show themselves;
so that these remarkable peculiarities,
which were long considered very
serious objections to the astronomical
theory, are really, on the other hand,
proofs of its truth. We learn from this
how mistrustful we ought to be of the first
impressions which strike our minds, however
probable they may appear, until they have
been submitted to scrupulous and searching
examination.

The reasoning by which M. Delaunay
works out his proposition is too lengthy
and too full of illustrative details to find
room here. The inquiring reader, who does
not care to take anything for granted,
is referred to the original "Notice,"
which is so lucid and logical as to be
easily understood by any clear–headed
person familiar with French, who will
peruse it slowly and with steady attention.
We, therefore, simply repeat the statement
that the three variations annual,
diurnal, and azimuthal observed in the
appearance of shooting stars, instead of
contradicting the astronomical or cosmical
theory of that phenomenon, and furnishing,
as was believed, capital objections to
its adoption, are, on the contrary,
completely in harmony with it. According to
that, then, there is reason to think that
shooting stars are due to the earth's