g the table in
position and cramping them down.

For the other parts the thin knife will be sufficient, wetted and worked
between, the cramps being applied as before.

This way of closing up dispenses with all cause for hurry. The exact
amount of glue can be calculated without danger of over-loading, and
the next necessary opening for repairs can be effected without the
least risk of damage to the margin of the upper table. By this method
there is no occasion for wiping superfluous glue from underneath the
over-lapping edge, as there will not be any perceptible, or, indeed,
present, when tidily done.

Sufficient time being allowed for drying--a little longer for the end
blocks, these being more hidden and slowly affected by the
atmosphere--the cramps may be removed.

As a final process, the varnishing over the parts that have been
repaired may be touched upon.

The success of this so much depends upon the natural talent of the
operator for matching colours and mixing of different gums for
obtaining as close as possible resemblance to the surrounding work,
that any hard and fast rules concerning it cannot be laid down.

An alcoholic mixture is almost invariably used for the covering of raw
repairs, time scarcely ever being available for the use of an oil
varnish.

The most commonly used basis is a lac varnish. The ease with which this
is dissolved and manipulated is a temptation to use it at times when
it would be the least desirable for the imitation of old varnish. One
great fault in connection with it is its retaining a glare on the
surface when hardened, and the undesirable aspect is given of polished
spots where repairs have been going on. There is only one way of
counteracting this--by mixing other gums or resins that have less or
but little glare when hard. Those of a very astringent quality should
be avoided, as when dry their pulling power or contraction is very great,
and a cracked surface not at all like the rich fused appearance of many
of the old masters, but dry and uninteresting, will make itself too
evident.

By a carefully-calculated mixture of soft and hard gums, with a little
transparent colouring matter when necessary, a very good if not highly
successful imitation of the surrounding work can be accomplished, such
as will prevent the repair "catching the eye" too soon, for it must
do so eventually when hunted for. The density of colouring and
thickness of the principal component parts must