ρ_crit: the Reference Density (Saturation Knee)

ρ_crit is the density scale at which C(ρ) = tanh(γ ln(1 + ρ/ρ_crit)) bends from its low-density ramp (C ≈ γρ/ρ_crit) into saturation (C → 1). It is a saturation knee, a reference scale, not a critical point: the function is smooth and nothing transitions there. At γ = 2, C(ρ_crit) = tanh(2 ln 2) = 0.88, and the midpoint C = ½ sits at ρ ≈ 0.32ρ_crit. The “crit” subscript is kept only because every script and ledger row since Session 53 uses it (see the glossary entry).

C runs from uncorrelated (C → 0) to collective (C → 1). It is not a quantum-degeneracy parameter. Earlier versions of this page said C runs “from quantum (C → 0) to classical (C → 1)” as density rises. That is backwards physics: quantum degeneracy sets in at high phase-space density (nλ_dB³ ≳ 1), and a dilute gas is the classical limit. The same labelling is why the most collective quantum states (superconductors, condensates) come out at the “wrong” end of the γ formula; see γ regimes.

In the galaxy sector ρ_crit is per-system, not a universal constant: V_flat is the observed flat rotation velocity of each specific system, so each galaxy (or other gravitationally bound object) has its own ρ_crit. A is the proportionality constant; V_flatenters as an input. The rotation-curve fit is not zero-parameter — it requires an observed V_flat per galaxy. Other sectors use different knees altogether; see one symbol, several knees below.

The A Parameter

A is calibrated against the Jeans criterion (Session 53), not derived from it: β_J = λ_Jeans / R_half is the dimensionless Jeans-length-to-galaxy-size ratio, empirically β_J ≈ 1.1 ± 0.2 across SPARC galaxies. A carries density units: since ρ_crit = A·V_flat² with ρ in M_☉ pc⁻³, A is in M_☉ pc⁻³ (km/s)⁻² — not (km/s)⁻². Earlier versions of this page dropped the density dimensions, which is exactly where a large numerical discrepancy can hide unnoticed.

Consequence — the knee is unreachable, and no length choice rescues it. If ℓ were a coarse-graining length it would have to smooth ρ and ρ_critalike — that is what a coarse-graining length is. Doing both, ℓ cancels:

x is a virial ratio. For any bound system V_c ≲ V_flat, so x ≲ 0.019β_J² ≈ 0.02 in every sector at every ℓ — the knee is out of reach by roughly 40× for free, with no fitted parameter. Verified numerically on all five plotter disks (max over ℓ: 1.7×10⁻³ to 1.1×10⁻², computed 2026-08-05 on the plotter's earlier rows; on 2026-10-26 the plotter switched to real SPARC rows, including a corrected UGC 128, and this range has not been recomputed; the 0.019β_J² bound does not depend on it) and to four digits at Cassini and wide-binary scales. It is kernel-robust: a Gaussian kernel gives coefficient 0.00505, moving the ceiling down a further 3.8×. This is the only galaxy-sector statement on this site that requires no estimator choice, no velocity definition, and no contested external measurement.The single escape found is β_J = 4.5, which lifts the self-consistent x to 0.385 (C = 0.57) — at 17σ from this page's own calibration. No refutation is added to the ledger from this (it remains at 6): it closes a question rather than opening one.

At the site's working A = 0.029, NGC 3198 gives ρ/ρ_crit = 9.45×10⁻⁴ — recomputed from the plotter's own midplaneDensity() (ρ(0) = 0.6164 M_☉ pc⁻³), so it can be re-run. Values of 0.934, 1.43×10⁻³ and 0.91 briefly shown here on 2026-08-05 were high by a factor of 1.515 and are corrected.

Symbol note: this page previously wrote the Jeans ratio as α, which forced a paragraph disclaiming confusion with the fine-structure constant. It is now β_J throughout, matching parameter derivations; the disclaimer is no longer needed.

Physical Meaning (as proposed)

Every gravitationally bound system has a characteristic rotation velocity V_flat — the velocity at which its rotation curve flattens. This velocity encodes the total mass and size of the system. The original motivation (Session 53) was a Jeans-type argument: put the knee where a system's self-gravity becomes strong enough to bind it as one unit. That argument calibrates A; it does not derive it (see the correction above).

Below ρ_crit, C is small and the proposed boost is large (1/C in the division reading). Above it, C saturates toward 1 and gravity is Newtonian. Nothing about quantum versus classical behaviour enters: this is a statement about how strongly matter is proposed to act as one collective unit, not about degeneracy.

One symbol, several knees

The symbol ρ_crit names at least six different objects across the site and archive. They span more than ten orders of magnitude. All values are in M_☉/pc³, at V = 200 km/s where V enters (explorer finding 2026-10-24, arithmetic on knees the site already prints; not a pre-registered test):

No single knee serves both galaxies and cosmology. The galactic knees sit 6.6 to 10.4 orders of magnitude above the cosmological one. Put the cosmological knee into a galaxy and C = 1 at every disc density, so gravity is Newtonian and there is no boost. Put a galactic knee into the Friedmann equation and the late-time dark energy comes out 3.7×10⁶ to 2.7×10¹⁰ times ρ_Λ. Both fail by more than six orders of magnitude.

The one cross-sector link on file is MOND's a₀ ≈ cH₀/2π. Written as a density it becomes ρ_knee(V) = (c/2πV)²·(2/3)ρ_c0, which scales as V⁻². That is an acceleration identification (Milgrom's), not a density knee. The framework asserts V⁺² and the measured knee is velocity-blind, so neither carries it. So “one equation” means one functional form with a knee calibrated separately in each sector.

Open question (untested, not refuted): is there a rule that predicts each sector's knee without fitting? For example, put the knee wherever the sector's characteristic acceleration equals a₀. Such a rule would have to land both 4×10⁻⁸ (cosmos) and the SPARC knee. Nobody has proposed one. If one exists, the unifying object is a₀, not ρ_crit.

Connection to MOND

From ρ_crit = A × V_flat² and the coherence function, two cosmological results emerge: