o recollection of any 'method' by
means of which they themselves learnt to sing at sight, and who
therefore think their pupils can pick up the knowledge in the same
instinctive fashion. Experience proves that this is very rarely the
case.

With very little children it is well to keep entirely to hand signs and
ear tests until all the notes of the scale are known, through their
'mental effect'. One reason for this is that such children cannot read
or write, so no musical work can be done with them which implies this
knowledge. Care must be taken to vary the lessons as much as possible.

At one lesson the teacher can give the hand signs and ear tests herself.
At the next, one of the class can give the hand signs for the rest of
the class, and the teacher the ear tests. At the next, a child can give
the ear tests, and so on. An experienced teacher will find plenty of
similar ways for producing new interest in the lessons, even though the
actual amount of work done be necessarily small. Nothing is gained by
hurrying over the initial stages of ear-training. The foundation must be
securely laid, or trouble will come later. Those who have had
experience of class work in kindergartens know the special difficulties
to be met--the irregularity of attendance, the constant stream of new
pupils coming in, and so on. Unless plenty of opportunity is given for
revision the work will suffer in thoroughness.

For children who take this work between the ages of eight and twelve, no
better scheme for sight-singing can be found than that contained in
Somervell's _Fifty Steps in Sight-singing_, supplemented by the
children's books, _A Thousand Exercises_, published by Curwen. It is
essential to read carefully the appendices to this work, especially that
concerned with the minor keys. Another book of sight-singing exercises
which follows the same sequence is the _Rational Sight Reader_, by
Everett, published by Boosey.

In teaching the keys of G major and F major it is most important that
the class shall themselves discover the necessity for the F[#] and B[b]
in the respective signatures. Inexperienced teachers sometimes teach
this as a dogma, and thereby deprive the children of the delight of
discovering it for themselves.

Thus, if the scale of G major be played with F[n] instead of F[#], the
class will discover that _taw_ has been played instead of _te_, and will
soon find out how to correct the wrong sound.

Similarly, if the scale