e those floating masses of the polar seas, the
approach of which is perceived through the sudden cooling of the
atmosphere.

His task is a complicated one. He has not only to conduct, in the spirit
of the author's intentions, a work with which the performers have
already become acquainted, but he must also introduce new compositions
and help the performers to master them. He has to criticise the errors
and defects of each during the rehearsals, and to organize the resources
at his disposal in such a way as to make the best use he can of them
with the utmost promptitude; for, in the majority of European cities
nowadays, musical artisanship is so ill distributed, performers so ill
paid and the necessity of study so little understood, that _economy of
time_ should be reckoned among the most imperative requisites of the
orchestral conductor's art.

Let us now see what constitutes the mechanical part of this art.

The power of _beating the time_, without demanding very high musical
attainments, is nevertheless sufficiently difficult to secure; and very
few persons really possess it. The signs that the conductor should
make--although generally very simple--nevertheless become complicated
under certain circumstances, by the division and even the subdivision of
the time of the bar.

The conductor is, above all, bound to possess a clear idea of the
principal points and character of the work of which he is about to
superintend the performance or study; in order that he may, without
hesitation or mistake, at once determine the time of each movement
desired by the composer. If he has not had the opportunity of receiving
his instructions directly from the composer, or if the _times_ have not
been transmitted to him by tradition, he must have recourse to the
indications of the metronome, and study them well; the majority of
composers, nowadays, taking the precaution to write them at the
beginning, and in the course, of their pieces. I do not mean to say by
this that it is necessary to imitate the mathematical regularity of the
metronome, all music so performed would become of freezing stiffness,
and I even doubt whether it would be possible to observe so flat a
uniformity during a certain number of bars. But the metronome is none
the less excellent to consult in order to know the original time, and
its chief alterations.

If the conductor possess neither the author's instructions, tradition,
nor metronome indications,--which fr