necessary. The top of the back part can now be shaved gently down by
a small metal plane, a little filing will give the evenness and
rotundity required. The same treatment will be necessary for the under
part, which in good work is a continuation of the line of the edging
of the upper table. A section of the nut in its finished state will
be as in diagram 17.

[Illustration: DIAGRAM 15.]

[Illustration: DIAGRAM 16.]

[Illustration: DIAGRAM 17.]

The whole of the surfaces may now be finished with the finest emery
cloth and oil. This latter may be linseed, nut, poppy or castor oil
with turpentine, but do not use sweet or olive oil, it never dries,
but lurks about in the pores of the wood and turns rancid.

Before leaving this part of the instrument, it may be as well to take
a glance at the peg itself and its insertion at the centre underneath
the nut. This is in no respect an unimportant detail to be seen to in
the fitting up and regulation of a violin. In olden times the peg was
small, not half the size of those inserted in new violins of the present
day. The increase in the size seems to have been gradual and to
compensate for the hard wood of the peg pressing against the inner,
softer substance of the end block with the thin slice of maple used
for the ribs, both being insufficient to withstand the strain of the
tail-string. Consequently the peg is pulled upwards, sometimes
considerably out of position. This is especially likely to occur if
the hole has been bored too large or the peg is too thin or short. The
accurate fitting of these should have strict attention. Some modern
repairers, when they find that there are indications of a softer piece
of pine than usual having been used, enlarge the hole with a tool
specially made for the purpose, having two cutting edges, or with a
number of grooves having sharp ridges, others a rat-tailed file. The
latter is perhaps the best, as being less likely to split the fibres
of the soft pine. The tool is inserted, not directly in a line pointing
exactly midway between the upper and lower tables, but slightly upwards
or contrary to the tendency of the peg to accommodate itself to the
strain. When the parts under strain have settled down, the peg is seen
to be as near as possible horizontal with the length of the violin.
The best way, undoubtedly, is to make the peg an accurate but not too
tight fit, it should be sufficiently long to go right through the block.
In cases where