Robbie quoted the 1963 description of the recording system:

[ When the key was struck, the prong dipped into the mercury,
[ completing a circuit and signaling the _exact_ force and
[ duration of the finger stroke; in other words, every subtlety
[ of the pianist's expression."

How can this be?  A carbon fibre dipping into a pool of mercury would
only complete a circuit.  Is one of the properties of mercury having a
sensitivity to pressure, a bruise factor?  I don't think so.

Although it could register the duration of the note, it would be the
coding on the sides of the roll which determine dynamics.

It sounds like this story borrows a little from science fiction and is
a bucket of steam in disguise.

Regards,

Jon Page, piano technician
Harwich Port, Cape Cod, Mass.

 [ A carbon rod of high resistance per linear centimeter, dipping into
 [ a pool of mercury, would exhibit a resistance change in linear
 [ proportion to the distance moved.  The differentiated signal would be
 [ the key velocity, centimeters per second.  Many modern MIDI keyboards
 [ use a conductive plastic pad which is squeezed against metallic pads
 [ to produce a varying resistance, analogous to the carbon rod in a
 [ mercury pool.
 [
 [ Fascinating and tantalizing!  But did it exist?  Elsewhere in this
 [ MMDigest Douglas Henderson says that Jeffrey Morgan wrote a series
 [ of articles for the AMICA magazine on this subject.  Could someone
 [ please tell me what issues?  -- Robbie