3. Analog Gravity as Template for Ether Dynamics

This section constitutes the central theoretical innovation of the monograph. We demonstrate that gravitational phenomena can be understood as consequences of ether dynamics by establishing a rigorous mathematical identity between the effective spacetime metric experienced by waves in a flowing medium and the Schwarzschild metric of general relativity in Painlevé–Gullstrand coordinates.

The key result (Theorem 3.2) is that the Schwarzschild solution of Einstein's field equations, expressed in Painlevé–Gullstrand coordinates, is exactly the acoustic metric for a constant-density ether flowing radially inward at the Newtonian free-fall velocity. This is not an approximation, not a weak-field limit, and not a metaphor — it is a mathematical identity. All predictions of Schwarzschild geometry (gravitational redshift, light bending, Shapiro delay, perihelion precession, black hole horizons) follow directly.

3.1 The Unruh–Visser Framework: Sound as Curved Spacetime

We begin with a result that is entirely mainstream: Unruh's 1981 demonstration [10] that sound waves in a moving fluid propagate along geodesics of an effective curved spacetime metric, subsequently formalised by Visser [11] and extensively reviewed by Barceló, Liberati, and Visser [35].

Consider a fluid characterised by:

• Density $\rho(\mathbf{x}, t)$
• Velocity field $\mathbf{v}(\mathbf{x}, t)$
• Pressure $p(\mathbf{x}, t)$
• Barotropic equation of state: $p = p(\rho)$
• Specific enthalpy: $h = \int dp/\rho$
• Local sound speed: $c_s = \sqrt{dp/d\rho}$

The fluid is assumed inviscid (Euler fluid) and irrotational ($\mathbf{v} = \nabla\psi$ for some velocity potential $\psi$).

Governing equations. The fluid obeys the continuity equation and the Euler equation:

$$\frac{\partial \rho}{\partial t} + \nabla \cdot (\rho \mathbf{v}) = 0 \tag{3.1}$$ $$\frac{\partial \mathbf{v}}{\partial t} + (\mathbf{v} \cdot \nabla)\mathbf{v} = -\frac{\nabla p}{\rho} = -\nabla h \tag{3.2}$$

For irrotational flow, (3.2) integrates to the Bernoulli equation:

$$\frac{\partial \psi}{\partial t} + \frac{1}{2}|\nabla\psi|^2 + h = 0 \tag{3.3}$$

where the arbitrary function of time has been absorbed into $\psi$.

Linearisation. Now decompose all quantities into a background (subscript 0) and perturbation:

$$\rho = \rho_0 + \varepsilon\,\delta\rho, \qquad \psi = \psi_0 + \varepsilon\,\delta\psi, \qquad \mathbf{v}_0 = \nabla\psi_0 \tag{3.4}$$

where $\varepsilon \ll 1$ and the background satisfies (3.1)–(3.3) exactly.

Linearising the Bernoulli (3.3) at first order in $\varepsilon$:

$$\frac{\partial(\delta\psi)}{\partial t} + \mathbf{v}_0 \cdot \nabla(\delta\psi) + \delta h = 0 \tag{3.5}$$

Since $\delta h = (dh/d\rho)\,\delta\rho = c_s^2\,\delta\rho/\rho_0$ (using $dh = dp/\rho$ and $dp/d\rho = c_s^2$), we can express the density perturbation as:

$$\delta\rho = -\frac{\rho_0}{c_s^2}\!\left(\frac{\partial(\delta\psi)}{\partial t} + \mathbf{v}_0 \cdot \nabla(\delta\psi)\right) \tag{3.6}$$

Linearising the continuity (3.1):

$$\frac{\partial(\delta\rho)}{\partial t} + \nabla \cdot (\rho_0 \nabla(\delta\psi) + \delta\rho\,\mathbf{v}_0) = 0 \tag{3.7}$$

Substituting (3.6) into (3.7):

$$-\frac{\partial}{\partial t}\!\left[\frac{\rho_0}{c_s^2}\!\left(\frac{\partial(\delta\psi)}{\partial t} + \mathbf{v}_0 \cdot \nabla(\delta\psi)\right)\right] + \nabla \cdot \!\left[\rho_0 \nabla(\delta\psi) - \frac{\rho_0 \mathbf{v}_0}{c_s^2}\!\left(\frac{\partial(\delta\psi)}{\partial t} + \mathbf{v}_0 \cdot \nabla(\delta\psi)\right)\right] = 0 \tag{3.8}$$

This can be written compactly as:

$$\partial_\mu\!\left(f^{\mu\nu}\,\partial_\nu\,\delta\psi\right) = 0 \tag{3.9}$$

where $x^\mu = (t, x^i)$ and the tensor density $f^{\mu\nu}$ has components:

$$f^{00} = -\frac{\rho_0}{c_s^2}, \qquad f^{0i} = f^{i0} = -\frac{\rho_0\,v_0^i}{c_s^2}, \qquad f^{ij} = \rho_0\!\left(\delta^{ij} - \frac{v_0^i\,v_0^j}{c_s^2}\right) \tag{3.10}$$

Identification of the effective metric. (3.9) is a curved-spacetime wave equation. In a spacetime with metric $g_{\mu\nu}$, the covariant scalar wave equation is:

$$\frac{1}{\sqrt{-g}}\,\partial_\mu\!\left(\sqrt{-g}\,g^{\mu\nu}\,\partial_\nu\,\phi\right) = 0 \tag{3.11}$$

Comparing Eqs. (3.9) and (3.11), we identify:

$$f^{\mu\nu} = \sqrt{-g}\;g^{\mu\nu} \tag{3.12}$$

To extract the metric, we compute $\det(f^{\mu\nu})$. From the explicit components of $f^{\mu\nu}$ ((3.9)), evaluating the $4 \times 4$ determinant by cofactor expansion along the first row:

$$\det(f^{\mu\nu}) = -\frac{\rho_0^4}{c_s^2} \tag{3.13}$$

From $f^{\mu\nu} = \sqrt{-g}\;g^{\mu\nu}$ ((3.12)), taking the determinant in 3+1 dimensions and using $\det(g^{\mu\nu}) = 1/g$:

$$\det(f^{\mu\nu}) = (\sqrt{-g})^4 \cdot \frac{1}{g} = \frac{(-g)^2}{g} = g \tag{3.13a}$$

Equating with the explicit computation (3.13): $g = -\rho_0^4/c_s^2$, and therefore:

$$g = -\frac{\rho_0^4}{c_s^2} \tag{3.14}$$

and therefore:

$$\sqrt{-g} = \frac{\rho_0^2}{c_s} \tag{3.15}$$

The inverse metric is $g^{\mu\nu} = f^{\mu\nu}/\sqrt{-g}$, and inverting yields the acoustic metric:

$$\boxed{g_{\mu\nu} = \frac{\rho_0}{c_s}\begin{pmatrix} -(c_s^2 - v_0^2) & -v_{0j} \\ -v_{0i} & \delta_{ij}\end{pmatrix}} \tag{3.16}$$

where $v_0^2 = |\mathbf{v}_0|^2$.

This theorem is proved by construction: (3.9) with the identification (3.12) shows that the perturbation $\delta\psi$ satisfies the covariant wave equation on the spacetime defined by (3.16). Null geodesics of this metric define the sound cones; trapped regions (where $v_0 > c_s$) define acoustic horizons. These results are thoroughly established in the literature [10, 11, 35].

Remark on the conformal factor. The overall factor $\rho_0/c_s$ in (3.16) is a conformal factor. Under a conformal rescaling $g_{\mu\nu} \to \Omega^2 g_{\mu\nu}$, null geodesics are preserved (their paths are unchanged, though affine parameterisation changes). In 3+1 dimensions, the conformal factor also drops out of frequency ratios measured between two points, provided the factor is time-independent. Consequently, for a steady background flow, the observable predictions — ray paths, frequency ratios, horizon locations — depend only on $c_s(\mathbf{x})$ and $\mathbf{v}_0(\mathbf{x})$, not on $\rho_0$ separately.

3.2 From Acoustic Metric to Ether Metric

The acoustic metric (3.16) was derived for sound in a fluid. We now make the central identification:

Acoustic system	Ether system
Background fluid	Ether medium
Fluid density $\rho_0$	Ether density $\rho_e$
Background flow velocity $\mathbf{v}_0$	Ether flow velocity $\mathbf{u}$
Local sound speed $c_s$	Local light speed $c_\ell$
Acoustic perturbation $\delta\psi$	Electromagnetic field perturbation
Acoustic metric $g_{\mu\nu}$	Effective spacetime metric

The ether metric is:

$$g_{\mu\nu}^{\text{ether}} = \frac{\rho_e}{c_\ell}\begin{pmatrix} -(c_\ell^2 - u^2) & -u_j \\ -u_i & \delta_{ij}\end{pmatrix} \tag{3.17}$$

Light propagates along null geodesics of this metric. Material bodies follow timelike geodesics. The causal structure of spacetime — including horizons, redshift, and gravitational lensing — is determined by the ether's flow velocity $\mathbf{u}(\mathbf{x})$ and the local light speed $c_\ell(\mathbf{x})$.

This identification raises an immediate question: does it reproduce the known gravitational metric? The next subsection shows that it does — exactly.

3.3 Painlevé–Gullstrand Coordinates and the Gravity–Ether Identity

The Schwarzschild metric describing the spacetime geometry outside a spherically symmetric mass $M$ is most commonly written in Schwarzschild coordinates:

$$ds^2 = -\left(1 - \frac{r_s}{r}\right)c^2\,dt_s^2 + \frac{dr^2}{1 - r_s/r} + r^2\,d\Omega^2, \qquad r_s = \frac{2GM}{c^2} \tag{3.18}$$

These coordinates are singular at the Schwarzschild radius $r = r_s$ (the coordinate singularity, not a physical singularity). In 1921, Paul Painlevé [36] and independently in 1922 Allvar Gullstrand [37] discovered an alternative coordinate system in which the Schwarzschild metric takes a remarkably different form.

The Painlevé–Gullstrand time coordinate. Define a new time coordinate $T$ related to Schwarzschild time $t_s$ by:

$$dT = dt_s + \frac{\beta(r)}{c(1 - \beta(r)^2)}\,dr, \qquad \beta(r) = \sqrt{\frac{r_s}{r}} = \frac{1}{c}\sqrt{\frac{2GM}{r}} \tag{3.19}$$

The quantity $c\beta(r)$ has a direct physical interpretation: it is the velocity of a radially free-falling observer who starts from rest at spatial infinity and falls inward under gravity. By energy conservation in Newtonian gravity, this velocity is:

$$v_{\text{ff}}(r) = \sqrt{\frac{2GM}{r}} = c\beta(r) \tag{3.20}$$

This Newtonian expression is, remarkably, exact in GR when using PG coordinates.

The Painlevé–Gullstrand metric. Substituting (3.19) into (3.18) and expanding the differentials ($dt = dT - [v_{\text{ff}}/(c^2 - v_{\text{ff}}^2)]dr$, substituted into all $dt$ and $dt^2$ terms, then collecting by $dT^2$, $dT\,dr$, $dr^2$), the Schwarzschild metric becomes:

$$\boxed{ds^2 = -\!\left(c^2 - v_{\text{ff}}^2\right)dT^2 - 2\,v_{\text{ff}}\,dT\,dr + dr^2 + r^2\,d\Omega^2} \tag{3.21}$$

where $v_{\text{ff}}(r) = \sqrt{2GM/r}$ and we adopt the convention that the inward flow corresponds to $v_{\text{ff}} > 0$ in the cross term.

Several features of this metric are immediately noteworthy:

(i) The spatial sections ($T = \text{const}$) are flat Euclidean space: $dl^2 = dr^2 + r^2 d\Omega^2$. There is no spatial curvature. All gravitational effects are encoded in the temporal components $g_{TT}$ and $g_{Tr}$.

(ii) The metric is regular at the Schwarzschild radius $r = r_s$, where $v_{\text{ff}} = c$. The Schwarzschild coordinate singularity has been removed. The PG coordinates extend smoothly through the horizon and cover the entire Schwarzschild spacetime (exterior and interior).

(iii) The PG time $T$ is the proper time of radially free-falling observers starting from rest at infinity. This gives $T$ a direct physical interpretation absent in the Schwarzschild time coordinate.

Remark on the constant-density assumption. Theorem 3.2 requires constant ether density $\rho_e = \rho_0$, giving a constant conformal factor. In Section 4.2, we introduce the superfluid condensate component whose density varies — this variation is the source of the MOND phenomenology (Theorem 4.1). These are different regimes: Theorem 3.2 describes the "normal ether" component (approximately constant density, responsible for the Schwarzschild metric), while Theorem 4.1 describes the superfluid condensate component (variable density, responsible for the dark-matter-like enhancement). When both components are present, the conformal factor becomes position-dependent, and the exact identity of Theorem 3.2 becomes an approximation whose accuracy depends on the ratio of condensate density variation to background density. Quantifying this correction is part of the strong-field programme (Section 11.3).

3.4 Physical Interpretation: Gravity as Ether Inflow

Theorem 3.2 yields a concrete physical picture of gravity:

Gravity is the steady-state inflow of ether toward mass. Objects in free fall are carried inward by the ether flow. The "force" of gravity is the drag of the medium.

This picture can be stated precisely:

(i) Free fall. A test particle at rest in the ether frame (co-moving with the ether flow at its location) follows a geodesic of the PG metric. Since PG time $T$ is the proper time of free-falling observers, the flow velocity $\mathbf{u}(r) = -\sqrt{2GM/r}\;\hat{\mathbf{r}}$ is the velocity of free fall. A particle "dropped" from rest at infinity is simply at rest in the ether; it falls because the ether flows inward.

(ii) Hovering. A particle that remains at fixed $r$ (a "hovering" observer) moves against the ether flow. It requires a non-gravitational force (rocket thrust, normal force from a surface) to maintain position. The required acceleration is:

$$a = \frac{d v_{\text{ff}}}{dT}\bigg|_{\text{following ether}} = \frac{GM}{r^2}\!\left(1 - \frac{r_s}{r}\right)^{-1/2} \tag{3.26}$$

which reproduces the exact GR expression for the proper acceleration of a static observer in Schwarzschild spacetime, including the relativistic correction factor that diverges at the horizon.

(iii) Escape velocity. The inflow velocity at the Schwarzschild radius is $v_{\text{ff}}(r_s) = c$. At this point, even a light signal directed radially outward is carried inward by the ether flow. This is the horizon: the surface at which the ether inflow velocity equals the speed of light.

(iv) Interior. For $r < r_s$, the ether inflow velocity exceeds $c$. All future-directed trajectories — including light — are carried inward. The interior of a black hole is a region of superluminal ether flow.

This picture was noted (though not developed into a full ether theory) by Hamilton and Lisle [38], who called it the "river model" of black holes. We take it further: the river model is not merely a pedagogical visualisation — it is the mathematical content of Schwarzschild gravity expressed in its most physically transparent form.

3.5 Gravitational Predictions: Exact Results

Since the ether metric (3.24) is exactly the Schwarzschild metric in PG coordinates, all predictions of Schwarzschild geometry follow identically. We catalogue these for completeness and to make the comparison with observation explicit.

3.5.1 Gravitational Redshift

Consider two static observers at radii $r_1$ and $r_2 > r_1$, both outside the horizon ($r_1, r_2 > r_s$). The static observers must resist the ether flow, and their proper time ticks at rate:

$$d\tau = \sqrt{|g_{TT}|}\;dT = \sqrt{c^2 - v_{\text{ff}}^2(r)}\;dT = c\sqrt{1 - \frac{r_s}{r}}\;dT \tag{3.27}$$

A signal of frequency $\nu_1$ emitted at $r_1$ is received at $r_2$ with frequency:

$$\frac{\nu_2}{\nu_1} = \frac{\sqrt{1 - r_s/r_1}}{\sqrt{1 - r_s/r_2}} \tag{3.28}$$

For emission near a mass and reception at infinity:

$$\frac{\nu_\infty}{\nu_r} = \sqrt{1 - \frac{r_s}{r}} \approx 1 - \frac{GM}{rc^2} \tag{3.29}$$

Physical mechanism in the ether picture: The ether flows inward. A photon emitted upward must fight against the inflowing ether. It loses energy (is redshifted) because it is propagating against the current — precisely as a sound wave is frequency-shifted when propagating against a flowing medium.

Experimental confirmation: Pound and Rebka (1960) measured the gravitational redshift in a 22.6 m tower at Harvard, confirming (3.29) to 1% accuracy [39]. Gravity Probe A (1976) confirmed it to $7 \times 10^{-5}$ [40]. Modern atomic clocks confirm it at the $10^{-5}$ level over height differences of 33 cm [41].

3.5.2 Light Bending

The deflection of light passing a mass $M$ at impact parameter $R$ is determined by the null geodesics of the metric (3.21). The standard GR calculation (which applies identically in PG coordinates) yields:

$$\Delta\theta = \frac{4GM}{Rc^2} = \frac{2r_s}{R} \tag{3.30}$$

For light grazing the Sun ($R = R_\odot$, $M = M_\odot$):

$$\Delta\theta = \frac{4 \times 6.674\times10^{-11} \times 1.989\times10^{30}}{6.96\times10^8 \times (3\times10^8)^2} = 8.49 \times 10^{-6} \text{ rad} = 1.75'' \tag{3.31}$$

Physical mechanism in the ether picture: The ether flows radially inward toward the Sun. Light passing the Sun is partially carried by the ether flow, deflecting it toward the mass — precisely as a swimmer crossing a river is carried downstream.

The factor of 2 beyond the Newtonian prediction (which gives $2GM/Rc^2$) arises because in the PG metric, the spatial sections are flat but the cross term $g_{Tr}$ introduces an additional deflection beyond what a pure "refractive index" model would give. The ether flow deflects light both by altering the local propagation speed and by physically carrying the wavefronts.

Experimental confirmation: Dyson, Eddington, and Davidson (1919) first measured solar light bending [42], and modern VLBI measurements confirm (3.30) to $10^{-4}$ accuracy [43].

3.5.3 Shapiro Time Delay

A radar signal passing near a mass $M$ at closest approach distance $R$ and travelling between radii $r_1$ and $r_2$ experiences an excess time delay:

$$\Delta T = \frac{2GM}{c^3}\!\left[\ln\!\left(\frac{4r_1 r_2}{R^2}\right) + 1 + \mathcal{O}(r_s/R)\right] \tag{3.32}$$

Physical mechanism in the ether picture: The ether inflow slows the outward propagation of the signal (the signal must fight the current) and accelerates the inward propagation, but these effects do not cancel because the signal spends more coordinate time in the region of strong inflow. The net effect is a delay.

Experimental confirmation: Shapiro (1964) predicted this effect [44]; subsequent radar ranging to Mercury and Mars and Cassini spacecraft tracking [45] confirm it to $10^{-5}$ accuracy.

3.5.4 Perihelion Precession

A test body in a bound orbit around mass $M$ with semi-major axis $a$ and eccentricity $e$ experiences a perihelion advance per orbit of:

$$\Delta\phi = \frac{6\pi GM}{a c^2(1-e^2)} \tag{3.33}$$

For Mercury ($a = 5.79 \times 10^{10}$ m, $e = 0.2056$):

$$\Delta\phi = \frac{6\pi \times 6.674\times10^{-11} \times 1.989\times10^{30}}{5.79\times10^{10} \times (3\times10^8)^2 \times (1-0.04227)} = 5.029 \times 10^{-7} \text{ rad/orbit} \tag{3.34}$$

With Mercury's orbital period of 87.97 days, this yields $43.0''$/century, in agreement with the observed anomalous precession of $(42.98 \pm 0.04)''\text{/century}$ [46].

Physical mechanism in the ether picture: The precession arises because the ether inflow modifies the effective potential experienced by the orbiting body. In Newtonian gravity with a static ether, orbits are closed ellipses. The ether's inflow introduces a velocity-dependent correction to the effective potential (analogous to the "magnetic" part of the gravitoelectromagnetic analogy), causing the ellipse to precess.

3.5.5 Gravitational Wave Speed

Gravitational waves in GR propagate at speed $c$, confirmed to extraordinary precision by the simultaneous detection of gravitational waves and gamma rays from the neutron star merger GW170817/GRB 170817A [47]:

$$\frac{|c_{\text{gw}} - c|}{c} \leq 5 \times 10^{-16} \tag{3.35}$$

In the ether framework, gravitational effects arise from perturbations of the ether flow. We demonstrate in Section 3.7 that linearised ether perturbations propagate at speed $c$, consistent with this observation.

3.6 Ether Horizons and the Singularity Question

The PG metric (3.21) provides a particularly clean description of horizons in the ether picture.

The horizon. At $r = r_s = 2GM/c^2$, the ether inflow velocity equals $c$. The metric remains perfectly regular:

$$g_{TT}(r_s) = -(c^2 - c^2) = 0, \qquad g_{Tr}(r_s) = -c, \qquad g_{rr}(r_s) = 1 \tag{3.36}$$

The line element at $r = r_s$ is $ds^2 = -2c\,dT\,dr + dr^2 + r_s^2 d\Omega^2$, which is well-defined and non-degenerate ($\det g_{\mu\nu} \neq 0$). This regularity is a known advantage of PG coordinates [38]; in the ether interpretation, it means the ether flow passes smoothly through the sonic point with no discontinuity.

The interior. For $r < r_s$, we have $v_{\text{ff}} > c$: the ether flows superluminally. All future-directed causal curves are carried inward. The ether picture makes this physically vivid: inside the horizon, the ether "river" flows faster than light can swim against it.

The Schwarzschild singularity. The PG metric has $v_{\text{ff}} \to \infty$ as $r \to 0$, which implies infinite ether inflow velocity and hence a genuine physical singularity. This singularity is present in both the GR and ether descriptions.

However, the ether framework opens a path to singularity resolution that GR alone does not. If the ether has finite compressibility — a maximum flow velocity or a modified equation of state at extreme conditions — then the singularity may be replaced by a region of maximally compressed, maximally fast-flowing ether. Specifically, if we modify the constitutive relation at high velocities:

$$c_{\ell,\text{eff}}^2 = c^2 + \alpha_{\text{UV}}\,v_{\text{ff}}^2 \tag{3.37}$$

where $\alpha_{\text{UV}} > 0$ is a small parameter encoding ether microstructure effects, then the acoustic horizon condition $v_{\text{ff}} = c_{\ell,\text{eff}}$ has no solution — the effective light speed increases with the ether flow, preventing horizon formation or modifying its structure.

We flag this as speculative. The specific modification (3.37) is illustrative, not derived. A rigorous treatment requires a complete theory of ether microstructure, which is beyond the scope of this monograph. We note, however, that singularity resolution is generic in analog gravity systems (fluids cannot have infinite velocity) and that this provides physical motivation for expecting similar resolution in an ether theory.

3.7 Gravitational Waves as Ether Perturbations

We now show that linearised perturbations of the ether propagate as waves at speed $c$.

Background. Consider flat ether: $\rho_e = \rho_0$, $\mathbf{u} = 0$, $c_\ell = c$. The ether metric (3.17) reduces to the Minkowski metric (times a constant conformal factor).

Perturbations. Introduce small perturbations:

$$\mathbf{u} = 0 + \delta\mathbf{u}, \qquad \rho_e = \rho_0 + \delta\rho, \qquad c_\ell = c + \delta c_\ell \tag{3.38}$$

The ether satisfies the continuity and Euler equations. Linearising:

Continuity:

$$\frac{\partial(\delta\rho)}{\partial t} + \rho_0\,\nabla \cdot (\delta\mathbf{u}) = 0 \tag{3.39}$$

Euler (in the absence of external forces):

$$\rho_0\,\frac{\partial(\delta\mathbf{u})}{\partial t} = -\nabla(\delta p) = -c^2\,\nabla(\delta\rho) \tag{3.40}$$

where we used the equation of state $dp_e/d\rho_e = c^2$ evaluated at the background.

Taking the time derivative of (3.39) and the divergence of (3.40):

$$\frac{\partial^2(\delta\rho)}{\partial t^2} = -\rho_0\,\frac{\partial}{\partial t}\nabla\cdot(\delta\mathbf{u}) = -\rho_0 \!\left(-\frac{c^2}{\rho_0}\right)\nabla^2(\delta\rho) \tag{3.41}$$

yielding the wave equation:

$$\boxed{\frac{\partial^2(\delta\rho)}{\partial t^2} = c^2\,\nabla^2(\delta\rho)} \tag{3.42}$$

Ether density perturbations propagate at speed $c$, the speed of light.

An identical wave equation holds for the velocity perturbation. Taking the curl of (3.40) shows that $\nabla \times (\delta\mathbf{u})$ is constant — vorticity perturbations do not propagate. This mirrors GR, where gravitational waves are transverse-traceless (purely spatial, divergence-free) perturbations.

Comparison with observation. The LIGO/Virgo constraint (3.35) requires gravitational perturbations to travel at $c$ to within $5 \times 10^{-16}$. The ether wave (3.42) gives propagation speed exactly $c$, satisfying this constraint.

Tensorial structure. In GR, gravitational waves are described by a rank-2 tensor perturbation $h_{\mu\nu}$ with two independent polarisations (plus and cross). In the ether framework, the full perturbation involves both $\delta\rho$ (scalar) and $\delta\mathbf{u}$ (vector). The scalar mode corresponds to a longitudinal (breathing) perturbation, which is absent in GR. The vector modes, when decomposed into transverse and longitudinal parts, yield two transverse degrees of freedom matching the GR polarisations.

Prediction. The ether framework generically allows a scalar (breathing) mode that is absent in pure GR. Current gravitational wave observations constrain but do not exclude non-tensorial polarisations [48]. Future observations with multiple detectors (LIGO-Virgo-KAGRA-LISA network) will test this prediction. The scalar mode amplitude depends on the ether coupling to matter; if the coupling is conformal, the scalar mode is suppressed.

3.7.2 Gravitational Wave Generation from the Ether

The wave (3.42) establishes that the ether can carry gravitational perturbations at speed $c$. We now show that it can generate them — extending the gravitational sector from propagation to radiation.

The static ether field (3.56) is the Poisson equation $\nabla^2\Phi = 4\pi G\rho_m$, with instantaneous propagation. For time-dependent sources, the Painlevé–Gullstrand metric structure requires retardation: the metric perturbation $h_{00} = -2\Phi/c^2$ must propagate at $c$, not instantaneously. The time-dependent extension is:

$$\boxed{\frac{1}{c^2}\frac{\partial^2\Phi}{\partial T^2} - \nabla^2\Phi = -4\pi G\rho_m} \tag{3.42a}$$

Derivation. The argument proceeds in two steps. First, the PG metric identification (Theorem 3.2) establishes that the ether metric perturbation propagates at speed $c$ (not $c_s$). The free perturbation satisfies $\Box\delta\Phi = 0$ with $\Box = c^{-2}\partial_T^2 - \nabla^2$ ((3.42)). Second, the static limit must recover Newtonian gravity: $-\nabla^2\Phi = -4\pi G\rho_m$. The unique Lorentz-covariant equation that satisfies both conditions is (3.43).

More formally: in PG coordinates, the weak-field expansion of the Einstein tensor $G_{00}$ gives $G_{00} \approx -(2/c^2)\nabla^2\Phi + (2/c^4)\partial_T^2\Phi$ to leading order in $\Phi/c^2$. The linearised Einstein equation $G_{00} = 8\pi GT_{00}/c^4$ with $T_{00} = \rho_m c^2$ yields (3.43) directly.

Gravitational radiation. The retarded solution (3.44a), expanded in multipoles for a source of size $R \ll \lambda_{\text{GW}}$, gives the standard quadrupole radiation formula:

$$P = \frac{G}{5c^5}\left\langle\dddot{Q}_{ij}\dddot{Q}_{ij}\right\rangle \tag{3.42c}$$

where $Q_{ij} = \int\rho_m\,x_ix_j\,d^3x$ is the mass quadrupole moment and dots denote time derivatives. For a circular binary of masses $M_1$, $M_2$ at separation $R$:

$$P = \frac{32\,G^4}{5\,c^5}\frac{(M_1 M_2)^2(M_1 + M_2)}{R^5} \tag{3.42d}$$

This is the Peters formula [151], confirmed to $< 0.2\%$ accuracy by four decades of Hulse–Taylor binary pulsar observations [152].

Significance. The sourced wave (3.43) extends the ether's gravitational content from kinematics (Theorem 3.2: geodesic motion, horizons, redshift) to linearised dynamics (gravitational radiation, orbital energy loss, inspiral). The ether now generates, propagates, and absorbs gravitational waves, reproducing the complete linearised gravitational-wave phenomenology of GR. The nonlinear regime (binary mergers, strong-field backreaction) remains an open problem (C1), but the gap between the ether programme and full GR has been significantly narrowed.

3.8 Emergent Lorentz Invariance

A persistent objection to ether theories is that the ether defines a preferred frame, while all observations confirm Lorentz invariance to extraordinary precision. We now show that this objection, while historically influential, is physically unfounded.

3.8.1 Lorentz Invariance from Fluid Dynamics

In the acoustic analogy, low-frequency sound waves obey Lorentz invariance of the acoustic metric exactly — even though the underlying fluid manifestly has a preferred frame (its rest frame). The acoustic Lorentz invariance is exact at all wavelengths much larger than the mean free path of the fluid molecules.

This result extends directly to the ether. If the ether has a microstructure at some fundamental length scale $\ell_e$ (which may be as small as the Planck length $\ell_P = \sqrt{\hbar G/c^3} \approx 1.616 \times 10^{-35}$ m), then:

This is a well-established result in the analog gravity and quantum gravity phenomenology literature [49, 50].

3.8.2 Modified Dispersion Relations

If the ether has discrete microstructure at scale $\ell_e$, the dispersion relation for light is modified at high energies. Consider the simplest model: ether as a regular lattice with spacing $\ell_e$.

The wave equation on a one-dimensional lattice with spacing $\ell_e$ and wave speed $c$ is:

$$\ddot{\phi}_n = \frac{c^2}{\ell_e^2}(\phi_{n+1} - 2\phi_n + \phi_{n-1}) \tag{3.43}$$

where $\phi_n$ is the field at lattice site $n$. The plane wave ansatz $\phi_n = A\,e^{i(kn\ell_e - \omega t)}$ yields:

$$\omega^2 = \frac{4c^2}{\ell_e^2}\sin^2\!\left(\frac{k\ell_e}{2}\right) \tag{3.44}$$

Expanding for $k\ell_e \ll 1$ (wavelengths much larger than lattice spacing):

$$\omega^2 = c^2 k^2\!\left[1 - \frac{(k\ell_e)^2}{12} + \frac{(k\ell_e)^4}{360} - \cdots\right] \tag{3.45}$$

The leading correction is quadratic in $k\ell_e$, giving a modified dispersion relation:

$$\boxed{\omega^2 = c^2 k^2\!\left(1 + \xi_2\,(k\ell_e)^2 + \xi_4\,(k\ell_e)^4 + \cdots\right)} \tag{3.46}$$

with $\xi_2 = -1/12$ for the simple lattice model.

Remark on the linear term. A term $\xi_1\,(k\ell_e)$ would represent CPT-violating dispersion and is absent for parity-symmetric microstructures. The Fermi-LAT observation of GRB 090510 constrains $|\xi_1| < 0.01$ at the Planck scale [51], effectively ruling out linear dispersion. The quadratic term $\xi_2$ is far less constrained (see Section 9.2.1).

3.8.3 Observational Constraints on the Ether Scale

The modified dispersion relation (3.46) produces an energy-dependent group velocity:

$$v_g = \frac{\partial\omega}{\partial k} \approx c\!\left(1 + \frac{3\xi_2}{2}(k\ell_e)^2\right) = c\!\left(1 + \frac{3\xi_2}{2}\frac{E^2\ell_e^2}{\hbar^2 c^2}\right) \tag{3.47}$$

Two photons with energies $E_1$ and $E_2$ emitted simultaneously from a source at cosmological distance $d$ arrive with time separation:

$$\Delta t = \frac{3\xi_2\,d\,\ell_e^2}{2\hbar^2 c^3}(E_1^2 - E_2^2) \tag{3.48}$$

Current observational status. The Fermi-LAT Collaboration [51] and MAGIC Collaboration [52] have searched for energy-dependent time delays from gamma-ray bursts and active galactic nuclei. For the quadratic term:

$$|\xi_2|\,\ell_e^2 < 3.2 \times 10^{-26}\;\text{m}^2 \tag{3.49}$$

If $\xi_2 \sim \mathcal{O}(1)$ (as the lattice model predicts):

$$\ell_e < 5.7 \times 10^{-13}\;\text{m} \approx 10^{22}\,\ell_P \tag{3.50}$$

This constrains the ether microstructure scale to be below about $10^{-13}$ m — comparable to the nuclear scale. If $\ell_e \sim \ell_P$, the predicted time delay is:

$$\Delta t \sim \frac{\ell_P^2 E^2 d}{2\hbar^2 c^3} \sim \frac{(10^{-35})^2 \times (10^{11} \times 1.6\times10^{-19})^2 \times 10^{26}}{2 \times (10^{-34})^2 \times (3\times10^8)^3} \sim 10^{-9}\;\text{s} \tag{3.51}$$

for a 100 GeV photon from a source at $z \sim 1$ — at the threshold of detectability with next-generation instruments such as the Cherenkov Telescope Array (CTA) [53].

3.9 The Ether Field Equation

The Painlevé–Gullstrand identification (Theorem 3.2) establishes that Schwarzschild gravity corresponds to a specific ether flow profile (3.22). We now address the question: what dynamical equation determines this flow profile?

3.9.1 The Ether Inflow Equation

In the PG picture, the ether flows radially inward with velocity $v_{\text{ff}}(r) = \sqrt{2GM/r}$. This is precisely the Newtonian free-fall velocity, which satisfies:

$$\frac{1}{2}v_{\text{ff}}^2 = \frac{GM}{r} = -\Phi(r) \tag{3.52}$$

where $\Phi = -GM/r$ is the Newtonian gravitational potential satisfying Poisson's equation:

$$\nabla^2\Phi = 4\pi G\rho_m \tag{3.53}$$

with $\rho_m$ the mass density of matter.

We can therefore express the ether inflow in terms of $\Phi$:

$$\mathbf{u} = -\sqrt{-2\Phi}\;\hat{\mathbf{r}} \tag{3.54}$$

or equivalently:

$$\frac{1}{2}u^2 + \Phi = 0 \tag{3.55}$$

(3.55) is the Bernoulli equation for the steady-state ether flow, with the "total energy per unit mass" of the ether flow equal to zero (corresponding to ether starting from rest at infinity). Combined with Poisson's (3.53), this gives a complete system:

$$\boxed{\nabla^2\Phi = 4\pi G\rho_m, \qquad \mathbf{u} = -\nabla\Psi, \qquad \frac{1}{2}|\nabla\Psi|^2 = -\Phi} \tag{3.56}$$

where $\Psi$ is the ether velocity potential ($\mathbf{u} = -\nabla\Psi$, with the sign convention chosen so that $\Psi$ increases inward). The third equation is the Bernoulli condition.

Remark. The system (3.56) is the weak-field ether field equation. It determines the ether flow velocity from the matter distribution. The Schwarzschild solution (3.22) is the unique spherically symmetric solution for a point mass. For general matter distributions, (3.56) yields the ether flow pattern from which the effective metric and all gravitational predictions follow.

3.9.2 Ether Conservation and the Sink Interpretation

The steady-state ether flow (3.22) has non-zero divergence:

$$\nabla \cdot \mathbf{u} = -\frac{1}{r^2}\frac{d}{dr}\!\left(r^2\sqrt{\frac{2GM}{r}}\right) = -\frac{3}{2}\sqrt{\frac{2GM}{r^3}} \tag{3.57}$$

For constant ether density $\rho_e = \rho_0$, the continuity equation $\partial_t\rho_e + \nabla\cdot(\rho_e\mathbf{u}) = 0$ is not satisfied: $\rho_0\,\nabla\cdot\mathbf{u} \neq 0$.

There are three interpretations of this result, which we state with full transparency:

(a) Mass as ether sink. The mass $M$ continuously absorbs ether at rate $\dot{M}_e = -\rho_0\!\oint \mathbf{u}\cdot d\mathbf{S}$. The ether inflow is replenished from the cosmological background. In this picture, mass is not merely immersed in the ether — mass is a persistent disturbance (vortex, soliton, or topological defect) that continuously absorbs the ether medium. This is the most physically intuitive interpretation but requires a mechanism for ether absorption.

(b) Compressible ether. If $\rho_e$ is allowed to vary, the continuity equation becomes $\nabla\cdot(\rho_e\mathbf{u}) = 0$ in steady state, giving $\rho_e(r) \propto 1/(r^2 |u(r)|) = 1/(r^{3/2}\sqrt{2GM})$. This preserves ether conservation but introduces density variation that modifies the metric. The correction to gravitational predictions is at post-Newtonian order and may provide a testable prediction distinct from GR (see Section 9.2.2).

(c) Effective description. The PG flow pattern is an effective description valid in the region outside the mass. Inside the mass (a star, planet, or compact object), the ether dynamics differ, and global ether conservation may be maintained. This is analogous to how the vacuum Schwarzschild solution is valid only outside the matter distribution; inside, one must solve the Tolman–Oppenheimer–Volkoff equation.

We regard the question of ether conservation as an open problem for the programme, not a fatal objection. We note that analogous questions arise in GR itself: the Schwarzschild solution describes a vacuum spacetime, and the source (matter) is treated separately. The ether programme faces the same challenge of connecting interior and exterior solutions.

3.9.3 Extension Beyond Weak Field

The weak-field ether (3.56) is exact for the Schwarzschild case (because the PG metric is exact). For more general situations (binary systems, cosmology, gravitational wave generation), the ether dynamics must be extended to a fully relativistic formulation.

The natural extension is to promote the ether to a relativistic fluid with four-velocity $U^\mu$ and apply relativistic fluid dynamics (the Israel–Stewart formalism or simpler perfect fluid models). The acoustic metric then becomes a function of the relativistic flow:

$$g_{\mu\nu}^{\text{ether}} = A(n, s)\!\left[\eta_{\mu\nu} + B(n, s)\,U_\mu U_\nu\right] \tag{3.58}$$

where $A$ and $B$ are functions of the ether number density $n$ and entropy density $s$, and $\eta_{\mu\nu}$ is the Minkowski metric.

Matching with the full Einstein field equations beyond weak field requires:

$$G_{\mu\nu}[g^{\text{ether}}] = \frac{8\pi G}{c^4}\,T_{\mu\nu}^{\text{matter}} \tag{3.59}$$

which constrains the functions $A$ and $B$ and the ether equation of state. This is a well-posed mathematical problem but has not been solved in full generality. We identify it as a key theoretical challenge for the ether programme.

Open problem (flagged). The complete relativistic ether field equations that reproduce the full Einstein equations for arbitrary matter distributions have not been derived. The weak-field and spherically symmetric cases are established (this section); the general case requires extending the analog gravity framework to the strong-field, dynamical regime. This is a central task for the research programme outlined in Section 11.

Clarification on scope. The acoustic metric framework reproduces the kinematic content of Schwarzschild gravity: geodesic motion, causal structure, horizons, redshift, light bending, and orbital precession. These are properties of a given spacetime geometry. What the framework does not yet provide is the dynamical content: the field equations that determine how the spacetime geometry (equivalently, the ether flow) responds to arbitrary matter distributions. In GR, this dynamical content is the Einstein field equations $G_{\mu\nu} = 8\pi G T_{\mu\nu}/c^4$. In the ether framework, the analog is the complete set of ether field equations — of which (3.56) is the weak-field, static limit. The kinematic results are theorems; the dynamical extension is an open problem.

3.10 Summary of Part II Results

We collect the key results of this section:

1. Acoustic metric derived (Theorem 3.1): Sound in a moving fluid propagates on an effective curved spacetime determined by the fluid properties. This is a mathematical theorem, not a conjecture.

2. Painlevé–Gullstrand identity (Theorem 3.2): Schwarzschild gravity is exactly the acoustic metric for constant-density ether flowing inward at the Newtonian free-fall velocity. This identification is exact, not a weak-field approximation.

3. All Schwarzschild predictions reproduced: Gravitational redshift ((3.28)), light bending ((3.30)), Shapiro delay ((3.32)), perihelion precession ((3.33)), gravitational wave speed ((3.42)) — all follow from the PG identification.

4. Emergent Lorentz invariance (Theorem 3.3): Lorentz symmetry is exact at wavelengths $\lambda \gg \ell_e$ and violated at order $(\ell_e/\lambda)^2$, connecting to the active observational programme in quantum gravity phenomenology.

5. Modified dispersion relation ((3.46)): Specific, quantitative prediction testable with CTA observations of gamma-ray bursts.

6. Ether field equation ((3.56)): Determines the ether flow from the matter distribution in the weak-field regime.

7. Open problems identified: (a) ether conservation / sink interpretation, (b) strong-field / dynamical extension, (c) scalar gravitational wave mode.

The ether framework reproduces the empirical content of Schwarzschild GR not by reverse-engineering but by identifying an exact mathematical equivalence between a known coordinate representation of Schwarzschild spacetime and the acoustic metric for a flowing medium. This equivalence exists independently of whether one regards the ether as physically real — but if one does, it provides a concrete physical mechanism for every gravitational phenomenon.