The Reproduction Test's unit is pipeline replacement versus attrition, but the current-signatures section mostly scores function and atmosphere — caseload, news deserts, approval, infringement letters — which the page already says are immature proxies for reproduction.

Brennan resignation-over-threats and the publicly-funded criminal Bar headcount series are intake/exit numbers, so the proxy objection does not cover the whole of §IV. Caseload and news deserts can remain as stress around those series.

Gallup approval and the partisan trust gap after Dobbs can be disagreement with judicial output. That is compatible with the same pipeline still producing role-bound justices, so a poll is not a replacement-rate signature of Trust-coherence decay.

Hungary is engineered demolition of contest institutions (ownership, regulator pressure, Sovereignty Protection Office), not staffing falling below replacement. The replacement-rate metaphor fits the UK Bar and misfires on an owner who buys the tabloid.

Contest already has a failing cell for coordinated press capture, so Hungary can sit on the matrix without being an attrition case. The title test then sells two audits — replacement-rate and capture — as one headline about replacement rates failing.

A yearly sample of the steelmanned contestability test already surfaces pipeline decay as a rising fail rate. The Reproduction Test is then a reporting-frequency upgrade of the five-part audit, not a different audit, unless a repeated sample still misses something only a flow test catches.

The general repair (fund, restructure, or let decay) depends on Function-Dominant versus Rent-Dominant grades the table labels as an unevidenced format sketch, so the UK criminal-Bar crisis that opens the essay cannot yet be run through the Anti-Capture Test.

Criminal defence grades from the counterfactual question alone: disappearance yields catastrophic rights failures and no migrated substitute. The five-question audit is then not what licenses the motivating repair; it is extra machinery for other units.

ART capitalisation, Ireland's 44 tenured appointments, and Colorado security grants restore stock (money, seats, cameras). Filing them as the diagnostic scoring carrier growth treats input restoration as replacement-rate recovery.

Tenured appointments buy existing people; Colorado hardens the perimeter to stop remaining clerks from quitting. Retention and purchase are not intake, so even the strongest rebuild row answers a different question than replacement-rate.

Trust-coherence is a population property (shared grammar, felt verdict-versus-preference distinction), not a professional pipeline with intake, exams, and apprenticeships. Treating it as a carrier class with a replacement rate imports a staffing metaphor onto social cohesion, so only three of four classes can be counted as pipelines.

Putnam/Skocpol already name civic habits, horizontal trust, and mass-membership hollowing. The claimed remainder is a closure-function subset, but Trust-coherence still has no measurement that can move while Office and Contest stay put, so the partition does not yet audit this class differently from social capital.