Thank you, Bob Taylor, for the most comprehensive answer yet to this
question.  My experience with vacuum pneumatics is second to none,
my expertise being with "charge" pneumatics, as applied to keyed and
keyless book-playing organs.  I have also worked on church organs
employing the "exhaust" pneumatic system, which is akin to the keyless
book system.

In the mid-nineteenth century, when pneumatics were first applied to
church organ action, it was found that consoles were no longer tied
to the organs themselves.  There seemed to be no limit to intervening
tubing run.

The organ in London's St. Paul's Cathedral was originally placed on
a rood screen dividing the nave from the chancel.  Once constraints
were released by pneumatic action, the organ was simply sawn in half,
each half being placed on opposite sides of the chancel arch at about
the same height from the floor as before, and the screen removed.
Action tubing from the console on one side ran down and across to the
other side under the floor, and up the other side; quite a considerable
distance.

I would guess that the action employed was "exhaust", which utilizes
the "water-hammer" effect  to shut off the notes once the keys were
released.  The action bleeds would have been set fairly small for
a fast "on" time, and while on, air would be travelling along the tube
at a constant rate from the bleed in the organ to the open pallet at
the console.  As soon as the key pallet closed, the air's escape route
was blocked which caused a fast back-lash called "water-hammer" (just
like the bang often heard when water is suddenly turned off at the
kitchen sink).

Using this system it became possible for a console to be placed at
the opposite end of a church from the organ.  In these circumstances
I would doubt the length of tubing involved would have any appreciable
effect on response time.  At that kind of distance there would be more
problems associated with the speed of sound.  I would guess that with
a vacuum system the main difference is that the air in the tube would
be flowing in the opposite direction, so the effect would be similar.
Would anyone disagree?

Regards,
John Page, UK

 [ The speed of sound (about 1.13 feet per millisecond) delays
 [ both the sound of the organ pipes _and_ the signal in the hoses,
 [ so this might reduce the magnitude of the problem.  -- Robbie