Russell Wattam writes in 000925 MMD:

> .... A completely chromatic scale is really a luxury,... but a
> non-chromatic scale should certainly not be seen as a detriment....

Russell is quite right in this regard.  This also extends to musical
instruments which are "hand-played".  Many legitimate instruments are
diatonic and cannot play many sharps or flats without some sort of
device.  Concert harps immediately spring to mind.

Bugles (without keys or valves) are severely limited in notes,
sounding only fundamental, harmonics, and a few "bends"; but this
limitation becomes a challenge for genius to devise arresting musical
numbers like "Taps," "Reveille," "Charge," etc.

One need only consider the "short octave" found in many historic
keyboard instruments.  In these, the lowest F#, G#, A# might be
tuned to C, D, and E; allowing performance of music reaching down
to C without having to add C, C#, D, D#, and E to the keyboard.
Historic music (which lacked/avoided much of the modern chromaticism)
rarely calls for certain pitches, especially in the lowest bass line.
(As an aside, I seem to recall reading that there are virtually no
hymns written in the key of G-flat or its enharmonic, F-sharp.)

The lowest pitches in pipe organs employ the largest and most expensive
pipes.  Generally speaking, the lowest octave in a chromatic set costs
as much as (often more than) the next four octaves of that rank.  And
when you consider all the space, the fitting, the mitering, and the
weight involved in squeezing just one set of 16' pitch pipes (even
stopped pipes) into a reasonably sized organ, one realizes the
impracticality of designing a whole instrument around pipes that may
hardly ever be used!

A fully chromatic instrument is foolishly extravagant if it will be
played only mechanically.  In a mechanical organ, one can simply
transpose the music into a key for which the organ has all the right
notes.  Wasting money for say, a low C# and low G# means one must forgo
something more useful.  Better to put one's money into the things which
count.

Oh! and let me add that this is all my own opinion, with some fact
tossed in for verisimilitude.  ;-)

Regards,
Robert B. Linnstaedt
http://members.aol.com/Linnstaedt/organ.html