# Reality Revision 1.47 — water harmonics without hidden-dimension inflation

## Concrete addition

Matter Harmony Lab 0.4.0 adds an executable many-molecule water animation.
Sixty-four rigid H2O geometries are placed on an ordered periodic tetrahedral
scaffold whose box volume matches the declared 298.15 K water density.  A
320-coordinate skew system couples 192 visible displacement coordinates to
128 hidden coordinates and projects only the visible block into the rendered
three-dimensional positions.

For state `z=(q,h)` and frozen coupling `C`,

    G = [0 C; -C^T 0],       dz/dtau = Gz.

The written identity `d||z||^2/dtau = z^T(G+G^T)z = 0` supplies exact norm
preservation.  It does not turn the chosen coupling into a physical law.

## Result and counterexample

The 64-frame computation preserves state norm to 2.220e-16, rigid molecular
geometry to below 1.2e-13 degrees and 1.4e-15 Å, and the calibrated density to
reported precision.  Forty-one unit tests pass.  A separately implemented
audit checks 36,864 atomic-coordinate scalars and the principal invariants.

The important counterexample is preserved and now visible in the water model:
an arbitrary hidden-plane rotation with no visible coupling gives exactly zero
projected motion.  Integer-tick frequency aliases also remain exact to
5.519e-14 in the executed finite calculation.  Therefore neither a moving
picture nor fixed samples uniquely identify continuous hidden rotations.

## Limits

This is a synthetic harmonic network, not liquid-water molecular dynamics,
quantum electrodynamics, hydrodynamics, or measured evidence of a fourth or
fifth spatial dimension.  The ordered scaffold is not a claim about the
instantaneous structure of liquid water.  Density and geometry are matched by
construction.  The colors encode chosen latent phase, not electricity or
magnetism, and the fixed tick is a simulation discretization rather than a
discovered universal physical increment.

The next admissible test is an ordinary electrostatic water baseline with
explicit charges and a frozen time-dependent dielectric or structure-factor
target.  Only then may a prespecified latent coupling be compared on fresh
data.