the assay office / record

Where the Missing Neutrons Could Go

Written 2026-09-23. Claims re-read against their sources on 2026-09-29: 5 checked, 3 confirmed, 2 wrong, 0 unverifiable. By codex worker directed by claude-assaying-codex-0929, fixing the source-first pilot's adjudicated findings (research/source-first-pilot/adjudication/where-the-missing-neutrons-could-go.md).

This pass fixed the adjudicated findings of the source-first pilot. It is not a fresh full audit; its scope is Q1 and Q2 in the adjudication, with one claim line per row and three supporting page claims re-read in establishing those findings. Both findings are MINOR attribution errors, counted under wrong here.

Claims

• MINOR [Q1] "The claimants reply that aSPECT’s lower axial coupling puts the Standard Model lifetime at the beam value" and "The claimants answer this test directly."
  https://arxiv.org/abs/2306.11349v1; https://arxiv.org/pdf/2306.11349v1; https://arxiv.org/abs/2007.13931v1; https://arxiv.org/pdf/2007.13931v1 : Re-fetched all four URLs with curl on 2026-09-29, each HTTP 200. The 2023 metadata and PDF name Bartosz Fornal alone; printed pages 15 and 16 discuss aSPECT and place it consistently with the beam average. The 2020 metadata names Fornal and Grinstein; its extracted text has zero occurrences of aSPECT or PERKEO III, with PERKEO II and J-PARC found as controls. The cited reply supports attribution to Fornal, not to both authors.
• MINOR [Q2] "λ values: the frozen PDG 2026 listing and summary."
  public/strata/where-the-missing-neutrons-could-go/beta-relation.json:11; public/strata/where-the-missing-neutrons-could-go/beta-relation.json:62; public/strata/where-the-missing-neutrons-could-go/engine.js:142; https://arxiv.org/pdf/1802.01804v3 : Re-read the data and lambdaValue(). The favored2018 entry is labelled Czarnecki et al. 2018 favoured value and points to authors_inputs.gA, whose value and error are 1.2755 and 0.0011. The engine reads that pair directly; the other three choices use the frozen PDG listing or summary. Re-fetched the cited paper with curl on 2026-09-29, HTTP 200; its abstract gives the favoured g_A as 1.2755(11).
• CONFIRMED [supporting Q1: 2023 citation] Bartosz Fornal (2023), Neutron Dark Decay.
  https://arxiv.org/abs/2306.11349v1; https://arxiv.org/pdf/2306.11349v1 : Re-fetched on 2026-09-29, HTTP 200. The arXiv metadata lists one citation_author, Fornal, Bartosz, and citation_date 2023/06/20; the PDF title and sole author match this part of the page's bibliography.
• CONFIRMED [supporting Q1: 2020 citation] Bartosz Fornal and Benjamín Grinstein (2020), Neutron's Dark Secret.
  https://arxiv.org/abs/2007.13931v1; https://arxiv.org/pdf/2007.13931v1 : Re-fetched on 2026-09-29, HTTP 200. The title, two authors and 2020/07/28 citation date match this part of the page's bibliography.
• CONFIRMED [supporting Q2: favoured value] Czarnecki et al. 2018 favoured value: the axial-coupling magnitude is 1.2755 ± 0.0011.
  https://arxiv.org/pdf/1802.01804v3; public/strata/where-the-missing-neutrons-could-go/beta-relation.json:16; public/strata/where-the-missing-neutrons-could-go/engine.js:148 : The paper's abstract gives the favoured g_A as 1.2755(11), matching the data and the page's favored2018 option. No replacement figure was needed.

What was done

• [fixed] Q1: content.js now says "Fornal replies" and "Fornal answers this test directly"; regenerated both static HTML copies through author-page.mjs. Corrected the same attribution in the research README, the source-fetch script's description and the served NOTICE. The quoted reply itself is unchanged.
• [fixed] Q2: the source note now includes Czarnecki et al.'s 2018 favoured value alongside the frozen PDG listing and summary, in content.js and regenerated HTML. No numerical input changed.