Dear All,  The perennial question of how much to restore seems to
come up..., well, perennially!  Here's what I have always done; it's
not gotten me any complaints in 34 years of restoration.

In general, I divide the restoration into two parts:

1. The mechanism:

The primary purpose here is to restore the function of the mechanism
so that it plays as close as possible to how it played when it was
originally made.  If a music box tooth is missing, it gets replaced;
if a wheel (gear) tooth is missing or broken it gets fixed.  If a
tooth, or a pipe, or a reed or a whistle is out of tune, because of
rust or damage, it gets repaired, but none of these, if intact, ever
get re-tuned just because I think the original tuning was wrong or
could be improved.  What's there (in original condition) is left alone.

Some restorers retune, others put cement into music box cylinders
that originally had none (yes, it does sound better, but that's not
the purpose of restoration, which is to restore, not to make better!),
but none of these actions fit the underlying concept of restoring as
close as possible to original condition.  Using that principle when
recovering an organ bellows which was originally covered in leather
I would recover it in leather, not in a synthetic substitute.
Otherwise its not 'restoration.'

One of the big questions often discussed is the 'polishing' of
mechanical music instrument mechanisms.  It has long been the
convention among clock restorers to restore the clock mechanism to its
original polish levels.  Unfortunately, that has often been interpreted
insofar as mechanical music is concerned as meaning the highest level
of polishing achievable.

Often, one sees music boxes with the combs mirror-polished (not original
and it spoils the tuning), cylinders glowing from rouge polishing, and
wheels polished so that they sparkle.  The original 'level' of polish
is often still there underneath the tarnish of brass.

There are means of bringing back the original level of polish to brass
parts that have tarnished.  That's as far as polishing should go --
restore to original levels, not to what the restorer sees as pretty or
salable or possible.

Of course there are problems like rusted steel: when the rust is
removed it often leaves pits.  These are not pretty, yet to polish them
out means taking off metal from otherwise original and sound parts --
not a good idea.  One should keep what's original.  Remove dirt, rust
and tarnish as well as possible, and polish only to restore original
condition as far as this is possible.

2. Furniture:

Here, too, discretion is the better part of valor.  Finished wood gets
a patina of age which should (in my opinion) be respected as much as
possible.  Dirt should be removed, of course, and the case or furniture
thoughtfully repolished.  But certainly, taking a sound piece of
furniture and stripping it and then refinishing it with synthetic
lacquers and the like seems like a "no no" to me.  If, because of wear
or damage, it or a part of it must be refinished, one should (again,
in my opinion) try to use original fishing materials whenever possible.

Now it is true that there are exceptions; I practice some myself.
If a tortoise shell case of a miniature musical box or bird box has
been dropped and broken into a dozen pieces then the only way to
restore it is to put it together again with glues that did not exist
150 years ago.  It's either that or make a whole new case, something
which would not be appropriate as long as the old case can be put back
together.  So here I compromise my own principles.  I use a modern
method because it helps to save an antique piece.  It doesn't come out
better than original nor even as good.  But it does save it for
posterity.

There are many more aspects to the questions of restoration vs.
rebuilding vs. making it like new, etc.  Do we need to mark the new
parts indelibly to indicate that they are in fact new?  Should the
restorer's name be prominently displayed on the mechanism?  Should
restoration be even attempted if the possibility of success is not
assured?  I.e., is it better to leave the original broken condition so
that a future restorer with more skills, better tools and materials can
see the piece in its 'original' damaged state?

What about documentation of everything that was done?  If a screw was
replaced by a new screw made to look exactly as the old one must this
be noted?  Must the screw be so marked?

These and many other questions are some of the most interesting aspects
of restoration and probably have as many answers as there are
restorers.

The motto to follow, I think, is:  Whenever possible, do as little as
necessary to restore to original functioning and original appearance
plus normal aging.  Most of the time, "Less is more!"

Frank Metzger