# n234 — The Newton-anomaly catalog, and the first cone-universality measurement: both cones bend alike to 0.02%


**CORRECTION (n240, 2026-07-12):** the 'cone universality' measurement used a UNIFORM k-shift — a units rescaling to which any channel responds with coefficient 1 trivially. A physical mass's dressing has zero spatial mean (conservation), so this probe never touched the bending question. The universality statement survives only as: both cones respond identically to uniform rescaling. The local-index (deflection) question is open and in crisis: see n239–n240.

**Date:** 2026-07-12 · **Question (the human's):** find the other anomalies relative to Newton, and where the theory stands on each.

## The new computation: the lensing response (Eddington's γ, at channel level)

The bridge's s·a² vertex made static: how do the two cones respond to a smooth background dressing? Measured on the glass (central differences, h = ±0.05):

**d ln c²_EM/dk = +0.9982 · d ln c²_g/dk = +0.9981 — ratio 1.0002.**

Two consequences: (1) **cone universality** — light's cone and gravity's cone respond *identically* (0.02%) to a gravitational background: the P3″ analog of "light and gravitational waves are deflected alike," a nontrivial output since the two channels have entirely different characters (kernel-projected vs pattern-sector) and different λ's. (2) The common response ≈ 1 (the naive c² ∝ k value survives full dressing to 0.2%) feeds the γ dictionary: with R = 1 (clusters, n189), the spatial index shift is fixed — what remains for a true γ_PPN is the operational clock-side normalization (the same wall as n232's dictionary, shared).

## The catalog: every classical departure-from-Newton, vs the theory

| Anomaly (vs Newton) | Data | Theory status |
|---|---|---|
| Mercury perihelion (43″/cy) | classic | **program open with derived source** (n232–n233: b₃/a = +0.123, zero dials, Mercury = pure test once dictionary + orbit exist) |
| Light bending (1.75″ = 2× Newton) | Eddington → VLBI | **channel-level result tonight**: cone universality 1.0002; γ dictionary = the shared wall |
| Shapiro delay | Cassini 10⁻⁵ | same γ machinery; rides the same dictionary |
| Gravitational redshift | Pound–Rebka → clocks | **the theory's cleanest sector**: g₀₀ = 1+2Φ exact (mechanism 4), survived every audit |
| Frame dragging (Lense–Thirring) | GP-B, LARES-2 | **absent** — needs g₀ᵢ, which no version of the theory has ever had (the audit's point; unchanged by P3″) |
| Geodetic precession | GP-B 0.3% | absent — same g₀ᵢ/orbit wall |
| Binary-pulsar decay (radiation) | Hulse–Taylor 10⁻⁴ | the buried crystal claim; **reopened by the glass** (n222–n224: both polarizations propagate) but quadrupole power = the flow/normalization program, unbuilt |
| Rotation curves (the galactic anomaly) | flat curves | windings (entry 10, rescoped): topology survives the glass; amplitude numbers pending |
| Strong EP / Nordtvedt | LLR 10⁻⁴ | η_N = 0 as a graph identity (survived n226); *relativistic* η_N needs the body problem — absent |
| Composition EP | MICROSCOPE 10⁻¹⁵ | exposure (universality chosen, not derived) — unchanged |

## The honest pattern

Newton's anomalies split cleanly along the theory's own fault line: everything **clock-side or static** (redshift, perihelion source, EP identities, cone responses) is derived, exact, or programmed with no dials; everything **motion-side** (frame dragging, geodetic, orbits, radiation power) hits the same single wall — no g₀ᵢ, no test-body dynamics, no matter cone: **Gate F again**. The catalog does not scatter; it points. The matter/transport sector is not one item on a list — it is the last wall in front of every remaining anomaly at once.

**Class: the lensing response [A-model, central differences, one seed — ensemble sweep owed]; cone universality [the night's result]; the catalog [status ledger, sources: the standing registry]. Gate F is now, formally, the whole classical-tests program.**