under any circumstances be allowed.
It must be understood then in what follows, that singing alto in this,
the chest voice, either before or during the break, is unqualifiedly
condemned.

But we will suppose now that boys have been permitted to sing only in
the head register, that they have been assigned to the upper part in
part singing, for notwithstanding that usage is to the contrary, this is
what should be done. As has already been suggested the voices of girls
change less, and at a younger age than do boys, and they begin to show
weight of tone and increased volume, at an age when boys are at their
best as sopranos. Girls at this period should sing the middle and lower
parts, but it must be said in passing that much of the music contained
in our text-books ranges too low in pitch for them, or any voice except
a low contralto or a tenor. They must not be permitted to use their
voices at full strength, and special care should be taken of those who
at this age show hoarseness. With girls as with boys, the change is
accompanied with periods of great relaxation of the vocal bands, and
during these periods the singing tone is either very light, or very
loud.

Returning to the subject of treatment of boys' voices during mutation,
and premising that they have sung only in the head voice during
childhood, the question arises whether they are not in many cases set to
singing bass prematurely. It is obvious that during this period the
voice is actually _broken_, divided in two. The lower notes are produced
in the chest or man's register, while more or less of the boy's voice
remains as upper tones. These tones, by the way, never are lost, they
remain as the falsetto or head voice of the man.

Now the vibratory action of the vocal ligaments is much larger for the
chest voice than for the head, or as we ordinarily call it, the
falsetto. There is then no question that during mutation a boy can
confine himself to the use of his old voice, or so much of it as is
available at any time with very little strain. The tone will be light,
in fact, during the active periods of laryngeal growth which
characterize mutation, there will perhaps be no voice at all, owing to
the congestion of the parts, but in the periods of rest separating the
periods of growth, the vocal bands will respond. The compass of the head
voice at this time varies largely, but it corresponds pretty closely to
that of the second soprano, in three part exercises,