Barrier-Surface Mucosal Immunity

The gut tolerance latch generalises. Across an antigen sweep the offset from a surface's own baseline is invariant at ±spinodal(1.0)=±0.38490, so the saddle-node is barrier-surface-agnostic. The gut (antigen 0.50) is the one verified point; respiratory (airway) and skin (epidermal) barriers are named candidates, each awaiting its live volume. Grade [F].

The seam layer now generalises the gut identity to other barrier surfaces under the same 2×sha256 (9357f23b…), still importing ZERO sibling code. The gut saddle-node (induction = antigen + spinodal, maintenance = antigen − spinodal) is shown barrier-surface-agnostic: across an illustrative 5-point antigen sweep at g=1.0 the surface-independent offset is invariant at +0.38490 / −0.38490, so the gut (antigen 0.50 → induction 0.8849, maintenance 0.1151) is the one vendored point on a barrier-agnostic line. Respiratory (airway) and skin (epidermal) barriers are NAMED candidates — each needs its live volume and owned absolute antigen scale [O], not claimed here. Nothing feeds the hashed engine core.

Generalising the gut identity to any barrier surface

Chapter 16 established a closed-form identity at one barrier surface: the intestinal mucosal tolerance latch that digestive's IBD module reads IS this volume's T23 saddle-node and T24 suppressor complement, on the byte-identical R19 substrate. The induction threshold equals the luminal antigen plus the substrate spinodal, and the maintenance threshold equals the antigen minus the spinodal. This chapter asks whether that identity is specific to the gut or generic to barrier surfaces, and shows — in closed form — that it is generic.

The surface-independent offset is invariant

Write the tolerance thresholds for an arbitrary barrier-surface antigen a at the shared mucosal R19 scale (γ = 1.0): induction = a + spinodal(1.0), and maintenance = a − spinodal(1.0). The quantity that does not depend on the surface is the OFFSET from that surface's own antigen baseline: induction − a =  +spinodal(1.0) = +0.38490018, and maintenance − a = − spinodal(1.0) = −0.38490018. Across an illustrative 5-point antigen sweep (0.30, 0.40, 0.50, 0.60, 0.70) the offset is invariant to the eighth decimal — it never depends on a. That is what barrier-surface agnosticism means: the tolerance switch behaves identically at every barrier surface; only the surface's own antigen baseline differs. The gut is not a special case — it is the one point on this line that happens to be vendored and live-verified (antigen 0.50 → induction 0.8849, maintenance 0.1151).

Respiratory: a named airway-mucosa candidate

A respiratory or lung VP volume would localise this volume's tolerance switch to the airway mucosa — inhaled-antigen tolerance, with airway hypersensitivity as the airway reading of a broken latch. By barrier-surface agnosticism the structural prediction is exact: the airway induction threshold would equal the airway antigen plus spinodal(1.0), and the maintenance threshold the airway antigen minus spinodal(1.0) — a shared-substrate identity exactly analogous to the gut seam, differing only in the surface-owned antigen baseline. This is declared as a NAMED CANDIDATE, not a declared seam. The honest reason is the same discipline that keeps the musculoskeletal marrow-niche identity a candidate: the respiratory volume is not on disk in this session, so neither a vendored airway snapshot nor a live airway engine exists to verify the identity or to supply the absolute airway-antigen scale. The immune-side tolerance primitive (T23/T24, barrier-surface- agnostic) is real in this archive; the airway localisation and its absolute antigen scale are [O], owned by the future respiratory volume.

Skin: a named epidermal-barrier candidate

A skin VP volume would localise the same switch to the epidermal barrier — contact tolerance, with contact hypersensitivity as the skin reading of the same broken latch. The structural prediction is identical in form: epidermal induction = epidermal antigen + spinodal(1.0), maintenance = epidermal antigen − spinodal(1.0). It too is a NAMED CANDIDATE for the same reason — no live skin volume, no owned absolute epidermal-antigen scale on disk — and it rides the same barrier-agnostic line as the gut and the airway. The immune volume owns the tolerance primitive; the skin volume owns the epidermal localisation and its absolute antigen scale, [O] here.

Why candidates and not seams, and what is owned elsewhere

The program's discipline is that a shared-substrate identity is only claimed when it can be verified against the live sibling engine, the way the gut identity is checked against digestive's live relapsing course. The barrier-surface-agnosticism proof is owned and verified here — it rests on the measured T23 saddle-node and T24 suppressor complement of this volume and is computed from this archive alone, with zero sibling imports. What is NOT owned here is each real surface's absolute antigen scale: the airway aeroallergen load and airway mucosal scale belong to the respiratory volume, the epidermal contact-antigen load and barrier scale to the skin volume. Both are openly [O] with their owner named, exactly as the circulatory transport scale and the bone-niche scale were in chapter 18. Each candidate carries a stated promotion path (add the volume to the live harness, confirm substrate drift zero and that its barrier threshold reduces to barrier-antigen ± the spinodal) and a stated falsifier (if the offset were not ±the spinodal across the antigen sweep, barrier-surface agnosticism is dead and both candidates fall — the gut seam and its vendored antigen-0.50 point survive, because they assert nothing about a surface that has not been built).

What this is — and is not

This chapter is architecture and bookkeeping, not new physics. It adds no measurement to the thirty-five stress targets and changes no number in the engine; it proves one closed-form invariance and names two structural candidates, stating honestly which is owned and verified (the barrier-agnostic offset), which is named-not-claimed (the respiratory and skin identities), and what absolute scale is owned elsewhere and is therefore [O]. The cross-package statements are direction-and-structure claims — direction/class only, not medical advice, and not a validation of VP theory. The seam layer carries its own 2×sha256 (9357f23b…), separate from the engine emit() hash, which stays byte-identical.