Now starts the speculation, mixed with a little factual evidence.
The comb has been cut, the teeth profiles shaped for the lead weights
and damper platforms, the comb hardened and tempered to turn teeth into
springs (the solid root of the comb is often less hard that its teeth).

That particular comb would now be ready to accommodate a whole variety
of arrangements.  Leads would be soldered on and sized to anticipate
the gamme, teeth would be initially set after hardening and tempering
so that tips align in a horizontal plane at equal tip spacing.  Also,
tips would be of uniform thickness throughout for smooth damper action
and for interaction with cylinder pins.

Tooth tip profiles would be the maker's desired profile, sometimes with
complex hooked profiles.  David Lecoultre and his brother Henri Lecoultre
both had different shaped hooks!  Other makers such as Falconnet also
used hooked teeth but they were probably liaising with either David or
Henri Lecoultre.

The late H.A.V. Bulleid wrote many articles showing how the brass comb
base was scratched with marks to indicate how the comb was to be tuned.
Other gamme and serial number information may be found on the lowest
lead weight and cylinder cap, all quite useful information when tuning
a comb.

The comb is now ready to be tuned to its chosen gamme.  Again there
is little information, but the high quality movements were almost
certainly tuned by selected outworkers, some highly skilled and others
perhaps 'jobbing' tuners.  Nobody seems to know if there were standard
reference combs in the early years.  (The years of, say, pre-1830 to
about 1860 -- Bulleid's 'golden era' of the musical box, at the end of
which the change from key-wind to lever-wind spring motors took place.)

Tuning probably started at or near the mid-point of the comb where
most of the melody and harmonies are developed.  This meant choosing
a pitch to start the octaves, equivalent to the 'middle C' of the piano
but not necessarily any specific pitch.  All the required intervals
such as octaves, 3rds, 5ths the incidentals, etc., would then be
developed up and down the scale, finally arriving once more at the
'stretched' ends of the comb as the final elements of tuning.

A tune played in the scale of, say, C on a modern piano sounds
different if transposed to another scale because the [equally tempered]
intervals are standard and the ears are accustomed to the result.
Back in Handel's time the intervals were of a different standard and
transposition had its problems and musical limitations.  We know,
from the work of Keith Harding and others that there were a number of
standards such as 'Mean Tone Temperaments' to aid tuning, but there is
no evidence to date, as far as I am aware, that it was ever used to
tune combs!

A comb tooth has far greater harmonic limitations than an organ pipe
or stringed instrument.  Its fundamental is effectively its pitch with
few discernable higher harmonics.  The skill of the arranger and tuner
is to achieve harmony by combining the melody with chords.  The tuner
would then finish with stretching at bass and treble ends.  They do so
in a manner such that these higher or lower frequencies are not swamped
by the mid-range ones.  The ear does not (for most people) detect that
a higher note or lower note is not an exact equal octave to its
counterparts in the mid range.

The intensity of sound when a tooth is released is another factor.
The extreme treble notes are quite intense but of very short duration.
The bass end extremes add to the mid-range harmonies and underpin the
melody.  The nearest equivalent, perhaps, is the choir, where the high
descants sing out above all others and the deep bass singers' voices
linger on as the music progresses.  I once measured a stretch rising
to between 1.5 and 2 tones above the 'linear' octave of its lower
counterpart.  This may 'sound' extreme but it sounded okay until
measured!

Paul Bellamy
Kent, UK

 [ Part 3 of 4 will follow tomorrow.  -- Robbie