be taken that too large a quantity is not put in, or an unpleasant
tone, even blackness, will be the result.

With regard to the strength of the red, the same precautions must be
taken: on comparing the varnish of a very red old master of Italy, say
a Landolphi, with some made with a fairly strong solution of resin and
dragon's blood, the violence of colour in the latter will be very
apparent. At a little distance off the old master will look very modest
while the other will seem coarse and vulgar.

For softening purposes a very small quantity of gum thrus may be used,
too much will result in tackiness.

For hardening, sandarac has its place and usefulness, although, as with
the naturally soft gums and resins which return to their original
condition after the solvent has evaporated, great care must be taken
to use a very small proportion.

A mixture of sandarac and shellac will result in an extremely hard and
almost insoluble varnish, a very undesirable covering for a musical
instrument of any kind as it wears badly, that is, suddenly and harshly
with a rough fractured edge, instead of the gentle thinning-away under
usage, seen with a delicate yielding material.

Acroydes is an Australian "grass gum," with very little recommendation,
as in any considerable amount, it impairs the transparency of the other
gums with which it may be mixed.

For a brown colour, a little burnt sugar will give a good tint, although
too much will spoil the consistency of the other ingredients, and the
whole will be easily affected by damp. Aloes, of which there are several
kinds, have been used as a colouring ingredient, but the results are
not on the whole to be considered as good.

Of the different lacs, or as it ought to be termed, condition of the
resin, as they are all from the same source, seed lac and garnet lac,
in proportion with other resins, will be found to have considerable
colouring matter and requiring very little in addition.

For our present purposes, those of varnishing fresh parts of injured
violins, the above mentioned component materials will be found, when
in good proportion, according to the experience of the operator, to
be nearly all that would be desirable in imitating the surrounding
work.

There is nothing that can be suggested to enable a careless or
incompetent repairer to achieve good results without care or
calculation, and these two are an absolute necessity when the repair
and restoration of a viol