and not very tight. Theoretically it is not necessary
to screw the eye-bolt into the barrel for a distance greater than the
diameter of the bolt, but, in practice, it is better to screw it in
for a considerably greater distance than that if a reasonable degree of
safety is to be secured.

Now about turning the barrel to secure the right adjustment. The barrel
looks solid, but, as a matter of fact, it is hollow and much more frail
than it appears. For that reason it should not be turned by seizing it
with pliers, as that may distort it and spoil the bore within it. The
best method is to pass a piece of wire through the hole in its centre,
and to use that as a lever. When the correct adjustment has been
secured, the turnbuckle must be locked to prevent it from unscrewing. It
is quite possible to lock it in such a way as to allow it to unscrew a
quarter or a half turn, and that would throw the wires out of the very
fine adjustment necessary. The proper way is to use the locking wire so
that its direction is such as to oppose the tendency of the barrel to
unscrew, thus:

WIRES.--The following points should be carefully observed where wire is
concerned:

1. Quality.--It must not be too hard or too soft. An easy practical way
of learning to know the approximate quality of wire is as follows:

Take three pieces, all of the same gauge, and each about a foot in
length. One piece should be too soft, another too hard, and the third
piece of the right quality. Fix them in a vice, about an inch apart and
in a vertical position, and with the light from a window shining upon
them. Burnish them if necessary, and you will see a band of light
reflected from each wire.

Now bend the wires over as far as possible and away from the light.
Where the soft wire is concerned, it will squash out at the bend, and
this will be indicated by the band of light, which will broaden at that
point. In the case of the wire which is too hard, the band of light will
broaden very little at the turn, but, if you look carefully, you will
see some little roughnesses of surface. In the case of the wire of the
right quality, the band of light may broaden a very little at the turn,
but there will be no roughnesses of surface.

By making this experiment two or three times one can soon learn to know
really bad wire from good, and also learn to know the strength of hand
necessary to bend the right quality.

2. It must not be damaged. That is to say, it must