---
slug: a-minor-leonardo
passed: 2026-10-14
by: the assay line (research/assay-line/cited-1014b/), Codex fixer directed by claude-assaying-boyle-1014; source record by Codex, checked by Codex
checked: 147
confirmed: 107
wrong: 38
unverifiable: 0
observed: 0
stale: 2
---

This was the directed correction pass for the 40 triaged items in cited-1014b/a-minor-leonardo, read at f5e5f84f34. The fixer re-read the evidence before each edit, including the historical sources, saved data, generator, verifiers and dated work log, and replayed the raw readouts and cached check. The 107 CONFIRMED claims below retain the checker's source rereads with explicit provenance; they are not represented as 107 new source visits by this fixer. The nine listed MINOR items remain unchanged and are counted with the 29 WRONG items in the gate's wrong total of 38. The two STALE items use the triage's no-line exceptions. This record covers the triage and those confirmed claims, not every finding in the earlier checker report.

## Claims

- MINOR | X1 (return-date): From the theft to the day it hung again in the Louvre [...] 4 January 1914 [...] is 29 months, almost two and a half years. | https://en.wikipedia.org/wiki/Vincenzo_Peruggia; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs; recreated: audit-code.txt | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs:36-57. It sets the two endpoints and computes "const months = (1914 - 1911) * 12 + (0 - 7)", ignoring their days. The span is 867 days, or 28 full calendar months and 14 days. The narrative uses an approximate duration, so the issue is precision rather than false endpoint dates. Retain MINOR and list at tier C. This duration sentence is distinct from the empty-wall assertion fixed under X8.
- WRONG | X2 (sassoon-fame): The theft "was what made the Mona Lisa well known," in Sassoon's flat phrasing | https://www.aol.co.uk/news/day-mona-lisa-theft-made-003709636.html | Re-read the indexed full 21 August 2021 interview after the direct URL returned HTTP 429. The discovery paragraph follows the Florence recovery and arrest; the quoted fragment has no initial was. The repaired sentence refers to the 1913 recovery and links this interview.
- WRONG | X3 (sassoon-reporting): its disappearance "would soon become one of the largest features in reporting history." | https://guides.loc.gov/chronicling-america-theft-mona-lisa | Re-read the Library of Congress introduction: the reporting-history wording belongs to that guide. The repaired sentence names LoC and uses its footnote.
- MINOR | X4 (ledger-3): 29 months missing Documented (dates) + recomputed (the span). Recovered late 1913 (arrest 12 Dec 1913); rehung 4 Jan 1914 → 29 months. LoC; trial-record dates; arithmetic in the verifier. | https://en.wikipedia.org/wiki/Vincenzo_Peruggia; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs; recreated: audit-code.txt | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs:36-57 and the ledger at src/content/strata/a-minor-leonardo.md:76. The code uses "const months = (1914 - 1911) * 12 + (0 - 7)". That is a month-index difference, not the exact 867-day interval (28 months and 14 days). Calling the rounded span recomputed without naming the convention overstates precision. Retain MINOR and list at tier C; this ledger row is not the empty-wall sentence being fixed under X8. Same precision issue as X1.
- WRONG | X5 (citation-smith): R. A. Scotti / Smithsonian Magazine, "Stolen: How the Mona Lisa Became the World's Most Famous Painting" [...] the ~26-hour gap before the theft was noticed, the crowds at the empty wall (Kafka and Brod among them), the pre-1911 guidebook standing, and the postcard/film/cabaret reproduction wave. smithsonianmag.com | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Re-read the linked article: the byline is James Zug and the date is 15 June 2011. Scotti is discussed as the author of a separate retelling.
- WRONG | X6 (citation-vasari): Giorgio Vasari, Le Vite de' più eccellenti pittori, scultori e architettori (1st ed., 1550), life of Leonardo da Vinci [...] the smile "so pleasing that it seemed more divine than human," and the first record of the "Mona Lisa" identification. | https://www.uni-heidelberg.de/studium/journal/2008/02a/monalisa.html; https://collections.louvre.fr/en/ark:/53355/cl010062370 | Re-read Heidelberg and the Louvre catalogue. Vespucci's October 1503 marginal note names Leonardo's portrait of Lisa del Giocondo, before the 1550 published account. The smile quotation is retained.
- WRONG | X7 (footnote-vasari-first): the first record of the "Mona Lisa" identification. | https://www.uni-heidelberg.de/studium/journal/2008/02a/monalisa.html; https://collections.louvre.fr/en/ark:/53355/cl010062370 | Same first-identification error as X6. Heidelberg and the Louvre transcribe or describe the earlier October 1503 note; the citation now distinguishes it from Vasari's published identification.
- WRONG | X8 (summary-duration): the wall stood empty for 29 months. | https://fr.wikisource.org/wiki/Les_Masques_et_les_visages_au_Louvre/04 | Re-read the indexed 1912 Sizeranne article after direct retrieval timed out. Part IV, section I describes Raphael's Castiglione portrait replacing the Mona Lisa in the Salon Carré. This rules out an empty place throughout the absence without giving an installation date. Initial crowds at the empty place remain supported.
- WRONG | X9 (chart-title): How often U.S. newspapers named each painting, 1905–1920 | https://collections.louvre.fr/en/ark:/53355/cl010277627; https://www.louvre.fr/en/explore/visitor-trails/the-louvre-s-masterpieces/an-uplifting-sight | Re-read both Louvre descriptions. Venus de Milo is catalogued as a statue; Winged Victory is described as a statue. The shared chart title and accessible toggle label now say artwork.
- MINOR | X10 (chart-theft-window): from the middle of the pack to the top, in the exact window the painting was off its pegs. | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; git show f5e5f84f34:research/a-minor-leonardo/coverage.json | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:64-67: "peak.year >= 1911 && peak.year <= 1914". coverage.json has annual observations, not within-year dates. That supports a theft-era annual peak, not an exact August 1911 to January 1914 window. Retain MINOR and list at tier C for the standalone HTML sentence at public/strata/a-minor-leonardo/index.html:186. The similar time clause within the different, already-WRONG Markdown sentence is handled with X21.
- WRONG | X11 (chart-live): The lines are a live count from the Library of Congress Chronicling America newspaper archive [...] for each year, the number of digitized U.S. newspaper pages on which the phrase appears, divided by the total pages the corpus holds for that year (so a rising line means rising coverage, not more digitization). | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; git show f5e5f84f34:memory/log.d/2026-07-12T0245Z-a-minor-leonardo-measured.md | Read the embedded coverage-data block, generated: static-artifact, JSON.parse in the page, and the July 12 work log. The chart uses the stored series without an index fetch on page load. No new retrieval date has been invented.
- WRONG | X12 (chart-refetch): --check re-fetches and confirms the theft-window peak still stands in the live index. | git show f5e5f84f34:research/a-minor-leonardo/coverage-data.mjs; recreated: cache-test.txt and cache-positive-control.txt | Read fetchAll and all sixteen cache entries, then ran --check with fetch replaced by a throwing counter. It exited 0 with zero fetch calls. The descriptions and console output now identify cache reuse and missing-year fetching.
- WRONG | X13 (limit-ocr): It is OCR of century-old newsprint: counts are a noisy lower bound | git show f5e5f84f34:research/a-minor-leonardo/coverage-data.mjs; git show f5e5f84f34:memory/log.d/2026-07-12T0245Z-a-minor-leonardo-measured.md | Read the independent phrase queries, retained totals, and the July log's possible song/racehorse contamination. No step removes false positives. The text and generated caveats now allow both undercounts and overcounts; no contamination rate is claimed.
- WRONG | X14 (limit-control): OCR noise and wartime paper shortages hit every term alike, so a line that pulls away from its rivals is signal, not artifact. | git show f5e5f84f34:research/a-minor-leonardo/coverage-data.mjs; git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; git show f5e5f84f34:memory/log.d/2026-07-12T0245Z-a-minor-leonardo-measured.md | Read coverage-data.mjs and verify-immersive.mjs. Dividing separate phrase totals by the same denominator and comparing selected series does not measure term-specific OCR error or equal paper-shortage effects. The text now states that limitation.
- MINOR | X15 (verified-text): Every word above is the verified stratum. | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs; recreated: probe-verify-a-minor-leonardo.mjs.txt, probe-verify-immersive.mjs.txt, probe-positive-control.txt | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:100-131: CANON contains 21 selected fragments and uses "if (!htmlText.includes(needle)) missing.push(c);". The prose verifier also checks selected arithmetic and strings. I independently replayed both in memory: changing the HTML duration from 29 to 99 months still gives 19/19 and 12/12; changing the protected theft year makes the immersive check fail. Real selected checks exist, but the universal coverage claim overstates them. Relevel WRONG to MINOR under the explicit own-checking rule and list at tier C. This standalone sentence is not being fixed for a separate WRONG.
- MINOR | X16 (text-equality): Its text is checked character-for-character against that stratum | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs; recreated: probe-verify-a-minor-leonardo.mjs.txt, probe-verify-immersive.mjs.txt, probe-positive-control.txt | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:85-137. It normalizes text and tests selected CANON fragments with "if (!htmlText.includes(needle)) missing.push(c);", not full Markdown-to-HTML character equality. My in-memory 29-to-99 duration mutation still passes both verifiers; the protected theft-year mutation fails. Relevel WRONG to MINOR: this is an overstatement of the scope and precision of a real text check, covered by the user's own-checking rule. List at tier C. X17 shares this HTML sentence but is also MINOR; no separate WRONG requires that sentence to be fixed. Same overclaim as X19.
- MINOR | X17 (every-date): every date, duration, and conversion is recomputed from the documented record | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs; recreated: probe-verify-a-minor-leonardo.mjs.txt, probe-verify-immersive.mjs.txt, probe-positive-control.txt | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs:36-114 and 128-141. It calculates selected hard-coded quantities, including "const months = (1914 - 1911) * 12 + (0 - 7)", then checks selected prose strings; it does not extract and recompute every displayed date or number. My in-memory 29-to-99 HTML duration change leaves both verifiers passing. Relevel WRONG to MINOR under the explicit rule for "Every number is recomputed live" and similar checking overclaims. List at tier C; X16 is the other MINOR clause in this sentence. Same error as X18.
- MINOR | X18 (all-numbers): Every date and number below is recomputed from the documented record by the offline verifier in | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs; recreated: probe-verify-a-minor-leonardo.mjs.txt, probe-verify-immersive.mjs.txt, probe-positive-control.txt | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs:36-114 and 138-141: selected constants and arithmetic are checked, followed by presence tests such as "ok('the prose shows "361" (years since Vasari)', /361/.test(prose), 'figure present');". There is no extraction and validation of every date and number. The immersive check likewise tests selected fragments. Relevel WRONG to MINOR under the explicit all-numbers-recomputed rule and list at tier C. The Markdown sentence at line 34 is not being fixed for a separate WRONG. Same error as X17.
- MINOR | X19 (md-chart-equality): every word on that page is checked character-for-character against the prose below | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs; recreated: probe-verify-a-minor-leonardo.mjs.txt, probe-verify-immersive.mjs.txt, probe-positive-control.txt | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:85-137: "if (!htmlText.includes(needle)) missing.push(c);" checks 21 normalized CANON fragments, not equality of the complete Markdown and HTML prose. My in-memory duration mutation leaves the check passing, while a protected theft-year change fails. Relevel WRONG to MINOR because this is the overreach of a real text check, as specified by the own-checking rule. List at tier C. In src/content/strata/a-minor-leonardo.md:32 this is a separate sentence from the WRONG rival-ranking and sculpture-type claims; its other flagged clause, X20, is also MINOR. Same error as X16.
- MINOR | X20 (all-chart-claims): every claim it makes about the chart is recomputed from the data | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | NOT FIXED, as directed by triage. Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:54-80: "preYears.every((y) => ve(y) > ml(y))" checks Venus throughout the pre-period; Winged Victory is compared only for 1905 there. coverage.json:121-131 has 1910 Mona Lisa 143 and Winged Victory 65. My replay also confirms the raw-readout unit bug is not caught by the passing verifier. The verifier computes real selected chart propositions, but "every claim" overstates its coverage. Relevel WRONG to MINOR under the own-checking rule and list at tier C; X21 and X25-X40 separately retain the actual data errors as WRONG. This sentence shares X19, not those WRONG claims.
- WRONG | X21 (md-rivals): The rivals are the honest control: the Mona Lisa sits below them before 1911, then leaves them behind in the exact window it was off its pegs. | git show f5e5f84f34:research/a-minor-leonardo/coverage.json; git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs | Read and replayed the 1910 saved row: Mona Lisa 143 matching pages, Winged Victory 65; the rates are 253.7 and 115.3 per million. Venus is above Mona Lisa in every measured pre-1911 year, and Mona Lisa is above both in 1911–1914. The replacement figures come from coverage.json, not a fresh pull.
- WRONG | X22 (rival-paintings): two rival paintings (Venus de Milo, Winged Victory) | https://collections.louvre.fr/en/ark:/53355/cl010277627; https://www.louvre.fr/en/explore/visitor-trails/the-louvre-s-masterpieces/an-uplifting-sight | Same sculpture-type error as X9. Re-read the Louvre object descriptions and changed equivalent rival descriptions in prose, generator comments and generated caveats to artworks.
- STALE | X23 (verification-date): Last verified 27 June 2026, when it was written. | git show f5e5f84f34:memory/log.d/2026-06-27-human-realms-wave.md; git show f5e5f84f34:memory/log.d/2026-07-12T0245Z-a-minor-leonardo-measured.md | The June publication record and July 12 log support the original verification and a later successful run of both verifiers. This new dated claims record supplies the source for the regenerated placard. Triage explicitly exempts the generated placard date from an Updated line.
- STALE | X24 (date-modified): "dateModified":"2026-06-27" | git show f5e5f84f34:memory/log.d/2026-07-12T0245Z-a-minor-leonardo-measured.md; git show f5e5f84f34:public/strata/a-minor-leonardo/index.html | The July 12 work log and visible page stamp document a later modification than June 27. dateModified now dates this supported correction to 2026-10-14; the Corrected lines date the revision, so there is no separate Updated line.
- WRONG | X25 (readout-1905-raw): 1905 [...] per million pages: · Mona Lisa 24 · Venus de Milo 116 · Winged Victory 25 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1905. In Mona Lisa/Venus/Winged Victory order, 24/116/25 are raw matching-page totals; 51.5/249.0/53.7 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X26 (readout-1906-raw): 1906 [...] per million pages: · Mona Lisa 27 · Venus de Milo 112 · Winged Victory 41 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1906. In Mona Lisa/Venus/Winged Victory order, 27/112/41 are raw matching-page totals; 55.4/229.8/84.1 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X27 (readout-1907-raw): 1907 [...] per million pages: · Mona Lisa 19 · Venus de Milo 80 · Winged Victory 48 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1907. In Mona Lisa/Venus/Winged Victory order, 19/80/48 are raw matching-page totals; 38.0/160.0/96.0 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X28 (readout-1908-raw): 1908 [...] per million pages: · Mona Lisa 25 · Venus de Milo 119 · Winged Victory 42 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1908. In Mona Lisa/Venus/Winged Victory order, 25/119/42 are raw matching-page totals; 48.9/232.6/82.1 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X29 (readout-1909-raw): 1909 [...] per million pages: · Mona Lisa 26 · Venus de Milo 129 · Winged Victory 43 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1909. In Mona Lisa/Venus/Winged Victory order, 26/129/43 are raw matching-page totals; 48.3/239.7/79.9 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X30 (readout-1910-raw): 1910 [...] per million pages: · Mona Lisa 143 · Venus de Milo 169 · Winged Victory 65 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1910. In Mona Lisa/Venus/Winged Victory order, 143/169/65 are raw matching-page totals; 253.7/299.8/115.3 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X31 (readout-1911-raw): 1911 [...] per million pages: · Mona Lisa 1183 · Venus de Milo 154 · Winged Victory 56 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1911. In Mona Lisa/Venus/Winged Victory order, 1183/154/56 are raw matching-page totals; 2230.2/290.3/105.6 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X32 (readout-1912-raw): 1912 [...] per million pages: · Mona Lisa 650 · Venus de Milo 160 · Winged Victory 41 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1912. In Mona Lisa/Venus/Winged Victory order, 650/160/41 are raw matching-page totals; 1144.3/281.7/72.2 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X33 (readout-1913-raw): 1913 [...] per million pages: · Mona Lisa 925 · Venus de Milo 193 · Winged Victory 65 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1913. In Mona Lisa/Venus/Winged Victory order, 925/193/65 are raw matching-page totals; 1612.6/336.5/113.3 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X34 (readout-1914-raw): 1914 [...] per million pages: · Mona Lisa 798 · Venus de Milo 191 · Winged Victory 35 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1914. In Mona Lisa/Venus/Winged Victory order, 798/191/35 are raw matching-page totals; 1367.9/327.4/60.0 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X35 (readout-1915-raw): 1915 [...] per million pages: · Mona Lisa 158 · Venus de Milo 121 · Winged Victory 30 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1915. In Mona Lisa/Venus/Winged Victory order, 158/121/30 are raw matching-page totals; 280.9/215.1/53.3 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X36 (readout-1916-raw): 1916 [...] per million pages: · Mona Lisa 134 · Venus de Milo 154 · Winged Victory 71 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1916. In Mona Lisa/Venus/Winged Victory order, 134/154/71 are raw matching-page totals; 236.6/271.9/125.3 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X37 (readout-1917-raw): 1917 [...] per million pages: · Mona Lisa 67 · Venus de Milo 128 · Winged Victory 49 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1917. In Mona Lisa/Venus/Winged Victory order, 67/128/49 are raw matching-page totals; 120.9/230.9/88.4 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X38 (readout-1918-raw): 1918 [...] per million pages: · Mona Lisa 74 · Venus de Milo 58 · Winged Victory 71 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1918. In Mona Lisa/Venus/Winged Victory order, 74/58/71 are raw matching-page totals; 141.6/111.0/135.9 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X39 (readout-1919-raw): 1919 [...] per million pages: · Mona Lisa 99 · Venus de Milo 143 · Winged Victory 110 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1919. In Mona Lisa/Venus/Winged Victory order, 99/143/110 are raw matching-page totals; 172.0/248.5/191.1 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- WRONG | X40 (readout-1920-raw): 1920 [...] per million pages: · Mona Lisa 133 · Venus de Milo 95 · Winged Victory 57 | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html; git show f5e5f84f34:research/a-minor-leonardo/coverage.json; recreated: readout-results.json | Replayed the actual showYear function for 1920. In Mona Lisa/Venus/Winged Victory order, 133/95/57 are raw matching-page totals; 235.5/168.2/100.9 are the normalized rates. The raw label was per million pages. The corrected shared function selects matching pages in raw mode; the verifier now exercises all 32 year/mode readouts.
- CONFIRMED | The thief went for it partly because it was small | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-3 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 001 (small-motive). Read the source and compared the claim with its excerpt. Matching text: "Pensò dunque che avrebbe voluto riportare almeno un quadro in Italia: inizialmente aveva pensato alla “Bella giardiniera” (la “Madonna con il bambino e San Giovannino” dipinta da Raffaello e acquistata da Francesco I), ma era troppo grande. Confessò infatti di aver scelto la Gioconda la mattina stessa, guardandosi attorno nel Salon Carrè , per semplici motivi di praticità: il dipinto era di minori dimensioni rispetto agli altri, perciò più semplice da sottrarre e nascondere.".
- CONFIRMED | before 1911 it was one prized Leonardo among the Louvre's many treasures, not the singular icon it has since become. | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 002 (prior-standing). Read the source and compared the claim with its excerpt. Matching text: "Le vol de l’œuvre par Vincenzo Peruggia, ouvrier vitrier au musée du Louvre, en 1911, lui a donné une célébrité mondiale et surtout plus populaire. La Joconde est devenue, après sa redécouverte en 1913, le principal chef-d’œuvre du Louvre, l’œuvre la plus célèbre au monde, source inépuisable de détournements aussi bien artistiques que publicitaires.".
- CONFIRMED | On the morning of Monday, 21 August 1911, the Louvre was closed | https://guides.loc.gov/chronicling-america-theft-mona-lisa; https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 003 (theft-date). LoC gives "August 21, 1911"; the Florence interrogation reconstruction says "un lunedì mattina" and identifies Monday closure. Independent UTC calendar arithmetic gives Monday.
- CONFIRMED | Mondays were maintenance days. | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1; https://era.ed.ac.uk/items/5b76982c-83d1-47f2-9cd7-3ec540fd4455 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 004 (maintenance). The archive specifies "unico giorno della settimana di chiusura al pubblico". The contemporary curator statement quoted by Seaton, p. 101, says "Yesterday was cleaning day and the museum was closed to the public."
- CONFIRMED | A man in the white smock that museum workmen wore lifted the painting off its four iron pegs in the Salon Carré, carried it to a service stairway, took it out of its frame and its glazed case, wrapped the bare poplar panel in the smock, and walked out a side door with it under his arm. | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1; https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/; https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 005 (theft-method). The interrogation reconstruction supports removal, service stairs, discarded frame, smock and side exit; its "camice da lavoro" is supplemented by the Smithsonian account for the white work smock, glazed case and "four lonely iron hooks", and the Louvre catalogue for "huile sur bois (peuplier)".
- CONFIRMED | lifted the painting off its four iron pegs in the Salon Carré | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 006 (four-pegs). Read the source and compared the claim with its excerpt. Matching text: "Thousands of spectators, including Franz Kafka, flooded into the Salon Carré when the Louvre reopened after a week to stare at the empty wall with its four lonely iron hooks.".
- CONFIRMED | He was Vincenzo Peruggia | https://guides.loc.gov/chronicling-america-theft-mona-lisa; https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 007 (thief-name). The guide spells the name "Vincenzo Perugia". The primary Florence archive spells it "Vincenzo Peruggia"; the extra g in the page is correct.
- CONFIRMED | the early press often spelled him Perugia | https://tile.loc.gov/storage-services/service/ndnp/dlc/batch_dlc_romulan_ver01/data/sn83045462/00280658923/1914060601/0539.pdf; https://guides.loc.gov/chronicling-america-theft-mona-lisa | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 008 (early-spelling). The supplied excerpt is a modern guide, not an early newspaper. Supplement: the Evening Star, 6 June 1914, uses "Perugia" in its trial report; the LoC selection also links the December 1913 coverage. This confirms the historical spelling, without treating the modern guide as a period article.
- CONFIRMED | an Italian glazier who had himself worked at the Louvre fitting the protective glass [...] including, by his own account, on the Mona Lisa. | https://en.wikipedia.org/wiki/Vincenzo_Peruggia; https://collections.louvre.fr/en/ark:/53355/cl010062370; https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 009 (glazier). The full Wikipedia article says he made protective cases and discusses his account of fitting the Mona Lisa case. The Louvre calls him "ouvrier vitrier au musée du Louvre". The archive reconstruction is expressly drawn from his interrogations.
- CONFIRMED | The absence went unnoticed for about a day | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 010 (gap). Read the source and compared the claim with its excerpt. Matching text: "It would be 26 hours before someone noticed that the painting was missing.".
- CONFIRMED | roughly 26 hours | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 011 (26-hours). Read the source and compared the claim with its excerpt. Matching text: "It would be 26 hours before someone noticed that the painting was missing.".
- CONFIRMED | A painter who came to copy in the Salon Carré found a bare patch of wall and four pegs and assumed the picture was upstairs being photographed. | https://en.wikipedia.org/wiki/Vincenzo_Peruggia; https://era.ed.ac.uk/items/5b76982c-83d1-47f2-9cd7-3ec540fd4455 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 012 (copyist). The source excerpt establishes a copyist and next-day discovery, but omits the photography assumption. Supplement: Seaton, pp. 100-101, identifies Louis Beroud and the initial photographer explanation; the quoted curator statement says the guard "attributed it to the negligence of the official photographer".
- CONFIRMED | It was a guard who finally raised the alarm on Tuesday, 22 August. | https://en.wikipedia.org/wiki/Vincenzo_Peruggia; https://era.ed.ac.uk/items/5b76982c-83d1-47f2-9cd7-3ec540fd4455 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 013 (guard-alarm). The Wikipedia excerpt alone does not identify who raised the alarm. The fuller discovery account, supplemented by Seaton, pp. 100-101, supports the guard checking after Beroud asked and the alarm on the following morning. 22 August 1911 was Tuesday.
- CONFIRMED | conventionally given as 77 × 53 cm (about 30 × 21 inches), painted in oil on a poplar panel. | https://mini-site.louvre.fr/trimestriel/2021/Dossier_images_Louvre/115/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 014 (conventional-size). Read the source and compared the claim with its excerpt. Matching text: "Huile sur bois (peuplier) H. : 77 cm ; L. : 53 cm".
- CONFIRMED | The Louvre's own catalogue measures the panel itself at 79.4 × 53.4 cm | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 015 (catalogue-size). Read the source and compared the claim with its excerpt. Matching text: "Dimensions Hauteur : 0,794 m ; Hauteur avec accessoire : 1,01 m ; Largeur : 0,534 m ; Largeur avec accessoire : 0,77 m Materials and techniques huile sur bois (peuplier)".
- CONFIRMED | Either way it stands about two and a half feet tall | https://collections.louvre.fr/en/ark:/53355/cl010062370; recreated: arithmetic.json | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 017 (height). 79.4 / 30.48 = 2.605 feet; 77 / 30.48 = 2.526 feet. Both support the explicitly approximate phrase, using the Louvre dimensions.
- CONFIRMED | a portrait you could carry under one arm, which is exactly what Peruggia did. | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1; https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 018 (carried). The interrogation reconstruction describes carrying the painting wrapped in the work smock through the exit; the Smithsonian account supplies the under-arm detail.
- CONFIRMED | Donald Sassoon, whose Becoming Mona Lisa: The Making of a Global Icon (2001) | https://donaldsassoon.wordpress.com/curriculum-vitae-of-donald-sassoon/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 019 (sassoon-book). Read the source and compared the claim with its excerpt. Matching text: "Published in the USA by HarcourtBrace as Becoming Mona Lisa. The Making of a Global Icon , November 2001.".
- CONFIRMED | Peruggia "decided to steal the Mona Lisa not because it was then the most famous painting, but because it was small." | https://www.aol.co.uk/news/day-mona-lisa-theft-made-003709636.html | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 021 (sassoon-small). The indexed 2021 interview contains "He decided to steal the Mona Lisa not because it was then the most famous painting, but because it was small". The page stops before a continuation. The words are supported by the interview; the footnote instead pointing to the 2001 book is separately assessed at entries 73 and 79.
- CONFIRMED | Peruggia himself, at his 1914 trial, claimed a patriot's motive | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-3; https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/io-sono-e-mi-chiamo-peruggia-vincenzo/il-processo | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 022 (patriotic-trial). The archive describes the patriotic explanation given to investigators and doctors. Its process section dates the criminal judgment to "5 giugno 1914" and reports the motive argument at trial.
- CONFIRMED | that he was returning a stolen Italian treasure to Italy | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-3 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 023 (return-italian). Read the source and compared the claim with its excerpt. Matching text: "Come riferirà sia agli inquirenti sia al medico, fu nei giorni in cui lavorava al Louvre che maturò l’idea del furto. Aveva sentito dire che molti dei quadri esposti erano di pittori italiani e aveva visto in un libro una rappresentazione dell’arrivo al museo delle opere sottratte da Napoleone all’Italia.".
- CONFIRMED | his own letters show he hoped to sell it | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/io-sono-e-mi-chiamo-peruggia-vincenzo/il-processo | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 024 (profit-letters). Read the source and compared the claim with its excerpt. Matching text: "La sentenza confermò sostanzialmente le posizioni già espresse dall’accusa, rigettando le conclusioni della perizia psichiatrica e ribadendo che lo scopo di lucro era dimostrato dalle lettere di Peruggia al padre in cui annunciava l’imminente fortuna che stava per conquistare e il suo prossimo rientro in Italia, ed era ulteriormente provato dal tentativo di vendita a Londra.".
- CONFIRMED | serious treatments treat the patriotism as, at best, a half-truth he reached for after the fact. | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/io-sono-e-mi-chiamo-peruggia-vincenzo/il-processo; https://www.aol.co.uk/news/day-mona-lisa-theft-made-003709636.html | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 025 (motive-treatment). The judicial account rejects a purely patriotic explanation, citing letters and attempted sale. The Sassoon interview likewise treats it skeptically. These support the page's qualified treatment; they do not establish unanimity among historians.
- CONFIRMED | Peruggia kept it in a trunk in his Paris lodging, then carried it home to Italy | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1; https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 027 (storage-transport). The excerpt supports storage in a wooden case after an initial unprotected interval. The full arrest account places the case and picture in his Florence hotel, supporting the transport to Italy. This is a compressed sequence, not an assertion that the trunk was used from the first day.
- CONFIRMED | was caught only in December 1913 when he tried to sell it to a Florentine dealer, Alfredo Geri | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 028 (sale-attempt). The opening excerpt establishes a November approach to Alfredo Geri. The complete archive arrest narrative then describes the December Florence meeting, hotel inspection and arrest on 12 December. The excerpt alone omitted the latter events.
- CONFIRMED | the director of the Uffizi, Giovanni Poggi, authenticated it | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 029 (authentication). The full archive narrative follows Poggi's preliminary examination with inspection at the Uffizi and notification of the authorities. Its title and narrative identify Giovanni Poggi as director; the excerpt alone stops at the proposed fuller inspection.
- CONFIRMED | arrested in Florence on 12 December 1913 | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 030 (arrest-date). Read the source and compared the claim with its excerpt. Matching text: "Nel pomeriggio di venerdì 12 dicembre, mentre attendeva di essere ricontattato da Geri, Peruggia venne arrestato dalla polizia.".
- CONFIRMED | "the theft occurred in 1911, Mona Lisa was recovered in late 1913, and Vincenzo Perugia was on trial in 1914." | https://guides.loc.gov/chronicling-america-theft-mona-lisa/selected-articles | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 031 (loc-quotation). Read the source and compared the claim with its excerpt. Matching text: "Specifically, the theft occurred in 1911, Mona Lisa was recovered in late 1913, and Vincenzo Perugia was on trial in 1914.".
- CONFIRMED | Crowds filed into the Salon Carré to stare at the empty space and its four lonely pegs. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 032 (empty-wall-crowds). Read the source and compared the claim with its excerpt. Matching text: "Thousands of spectators, including Franz Kafka, flooded into the Salon Carré when the Louvre reopened after a week to stare at the empty wall with its four lonely iron hooks.".
- CONFIRMED | Among them, on a 1911 trip, were Franz Kafka and Max Brod | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 033 (brod-kafka). Read the source and compared the claim with its excerpt. Matching text: "Kafka and his traveling companion Max Brod marveled at the “mark of shame” at the Louvre and attended a vaudeville show lampooning the theft.".
- CONFIRMED | who went to look at the mark of shame, as Brod's diary has it. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/; https://era.ed.ac.uk/items/5b76982c-83d1-47f2-9cd7-3ec540fd4455 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 034 (brod-quote). Smithsonian confirms "mark of shame" but does not specify the diary. Seaton, p. 104, n. 165, attributes the wording to Brod and cites Hanns Zischler, Kafka Goes to the Movies (2003), p. 47. This supplies secondary attribution; the original diary was not independently inspected.
- CONFIRMED | the newspapers, which had given the picture little space before, now could not stop. | https://guides.loc.gov/chronicling-america-theft-mona-lisa; git show f5e5f84f34:research/a-minor-leonardo/coverage.json | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 035 (newspaper-shift). The Smithsonian account describes pre-theft guidebooks and the press/reproduction response; the dated LoC guide documents the coverage surge. The committed U.S. series also rises from 143 matching pages in 1910 to 1,183 in 1911.
- CONFIRMED | Reproductions followed in a flood [...] satirical postcards, a short film, cabaret songs [...] and high art tipped into mass art. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 038 (mass-reproductions). Read the source and compared the claim with its excerpt. Matching text: "Satirical postcards, a short film and cabaret songs followed".
- CONFIRMED | "It was an unknown painting until a thief made it famous" is as false as "it was always the crown jewel." | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 039 (not-unknown). Read the source and compared the claim with its excerpt. Matching text: "Le vol de l’œuvre par Vincenzo Peruggia, ouvrier vitrier au musée du Louvre, en 1911, lui a donné une célébrité mondiale et surtout plus populaire. La Joconde est devenue, après sa redécouverte en 1913, le principal chef-d’œuvre du Louvre, l’œuvre la plus célèbre au monde, source inépuisable de détournements aussi bien artistiques que publicitaires.".
- CONFIRMED | The Mona Lisa was admired for centuries before Peruggia. | https://sourcebooks.web.fordham.edu/source/vasari1.asp | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 040 (admired-centuries). Read the source and compared the claim with its excerpt. Matching text: "And in this work of Leonardo's there was a smile so pleasing, that it was a thing more divine than human to behold; and it was held to be something marvellous, since the reality was not more alive".
- CONFIRMED | Giorgio Vasari, in the first edition of his Lives of the Artists in 1550 | https://sourcebooks.web.fordham.edu/source/vasari1.asp | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 041 (vasari-1550). Read the source and compared the claim with its excerpt. Matching text: "Giorgio Vasari (1511-1574) wrote about hundreds of artists in his Lives of the Most Eminent Italian Architects, Painters, and Sculptors , which he published first in 1550, and in a revised edition in 1568.".
- CONFIRMED | 361 years before the theft | https://sourcebooks.web.fordham.edu/source/vasari1.asp | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 042 (vasari-gap). Read the source and compared the claim with its excerpt. Matching text: "Giorgio Vasari (1511-1574) wrote about hundreds of artists in his Lives of the Most Eminent Italian Architects, Painters, and Sculptors , which he published first in 1550, and in a revised edition in 1568.".
- CONFIRMED | already singled out its smile as "so pleasing that it seemed more divine than human." | https://it.wikisource.org/wiki/Le_vite_de%27_pi%C3%B9_eccellenti_pittori%2C_scultori_e_architettori_%281550%29/Lionardo_da_Vinci; https://sourcebooks.web.fordham.edu/source/vasari1.asp | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 043 (vasari-quote). The English supplied in the record is not word-for-word identical. The 1550 Italian has "un ghigno tanto piacevole che era cosa piú divina che umana a vederlo"; the page gives a faithful English translation of that phrase. Confirmed as translation, not as a quotation of the particular Fordham English wording.
- CONFIRMED | Walter Pater's essay on Leonardo, first printed in the Fortnightly Review in November 1869 (and collected in Studies in the History of the Renaissance in 1873) | https://www.gutenberg.org/files/4059/4059-h/4059-h.htm | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 044 (pater-publication). Read the source and compared the claim with its excerpt. Matching text: "1869. NOTES ON LEONARDO DA VINCI. Appeared in Fortnightly Review in November, 1869. Reprinted 1873 in Studies in the Renaissance.".
- CONFIRMED | 42 years before the theft | https://www.gutenberg.org/files/4059/4059-h/4059-h.htm | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 045 (pater-gap). Read the source and compared the claim with its excerpt. Matching text: "1869. NOTES ON LEONARDO DA VINCI. Appeared in Fortnightly Review in November, 1869. Reprinted 1873 in Studies in the Renaissance.".
- CONFIRMED | "she is older than the rocks among which she sits," | https://www.gutenberg.org/files/2398/2398-h/2398-h.htm | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 046 (pater-quote). Read the source and compared the claim with its excerpt. Matching text: "She is older than the rocks among which she sits;".
- CONFIRMED | a passage a generation of students could recite. | https://upload.wikimedia.org/wikipedia/commons/c/c4/The_study_and_criticism_of_Italian_art_%28IA_gri_33125001581517%29.pdf; https://campuspress.yale.edu/modernismlab/the-renaissance/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 047 (pater-recitation). The Yale excerpt supports influence rather than memorization. Supplement: Bernard Berenson wrote that he, "like so many of my contemporaries", had "learned by heart" Pater's famous passage, in The Study and Criticism of Italian Art, third series. This supports the generalization.
- CONFIRMED | By 1911 this was a celebrated Leonardo, written about by the most influential critics in Europe. | https://www.gadflyonline.com/03-25-02/artex-mona_lisa.html | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 048 (critics). Read the source and compared the claim with its excerpt. Matching text: "In the nineteenth century Lisa travelled on the back of the better-known Leonardo, but in the twentieth she became bigger than her creator".
- CONFIRMED | Pre-1911 travel guides to the Louvre gave it modest space [...] less than the Venus de Milo or the Winged Victory of Samothrace | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 049 (guidebooks). Read the source and compared the claim with its excerpt. Matching text: "In his 1878 guidebook to Paris, travel writer Karl Baedeker offered a paragraph of description about the portrait; in 1907 he had a mere two sentences, much less than the other gems in the museum, such as Nike of Samothrace and Venus de Milo.".
- CONFIRMED | it was far from the museum's most-visited object. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 050 (most-visited). Read the source and compared the claim with its excerpt. Matching text: "At the time of the “Mona Lisa” heist, Leonardo da Vinci's masterpiece was far from the most visited item in the museum.".
- CONFIRMED | the theft was a major, documented catalyst, but it was not the sole cause. | https://academic.oup.com/hwj/article-abstract/51/1/1/706426 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 051 (multiple-causes). The full publisher abstract lists royal collection, public museum, nineteenth-century critics, Leonardo's reputation, illustrated press, theft, parodies and later tours. Its theft-only excerpt is incomplete support, but the source as a whole supports multiple causes.
- CONFIRMED | the rise of cheap photographic reproduction that put the image on postcards | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 052 (photographs). Read the source and compared the claim with its excerpt. Matching text: "Satirical postcards, a short film and cabaret songs followed".
- CONFIRMED | Marcel Duchamp's mustachioed L.H.O.O.Q. of 1919 | https://www.moma.org/audio/playlist/352/4918 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 053 (duchamp-date). Read the source and compared the claim with its excerpt. Matching text: "In 1919, Duchamp takes a reproduction of the Mona Lisa and draws a mustache and goatee on it".
- CONFIRMED | by drawing on a cheap postcard reproduction | https://www.moma.org/audio/playlist/352/4918; https://www.moma.org/explore/inside_out/2011/04/01/five-for-friday-only-joking/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 054 (duchamp-postcard). The original MoMA audio says reproduction and names 1919. Supplementary MoMA collection commentary specifies "a cheap postcard of Da Vinci's Mona Lisa", supporting postcard, mustache and goatee. The audio excerpt alone did not establish postcard.
- CONFIRMED | proof the image was already mass-produced eight years after the theft, five after its return | https://www.moma.org/audio/playlist/352/4918; https://www.moma.org/explore/inside_out/2011/04/01/five-for-friday-only-joking/; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 055 (duchamp-gaps). Using the stated calendar years: 1919 - 1911 = 8 and 1919 - 1914 = 5. MoMA identifies the 1919 work and its postcard support. The durations are conventional year differences, not exact anniversary claims.
- CONFIRMED | disentangling their exact shares is a genuine historical argument, not a settled sum. | https://academic.oup.com/hwj/article-abstract/51/1/1/706426 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 056 (causal-uncertainty). Sassoon's publisher abstract enumerates several historical causes without assigning numerical shares. The page explicitly presents the allocation as unsettled; it does not claim to have measured a causal decomposition.
- CONFIRMED | a worldwide spike of dated coverage around an empty wall | https://guides.loc.gov/chronicling-america-theft-mona-lisa; https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/; https://academic.oup.com/hwj/article-abstract/51/1/1/706426 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 057 (worldwide-spike). The LoC guide supplies dated U.S. examples. The full Smithsonian account describes European publicity and popular reproductions; Sassoon's abstract describes the illustrated-press context and international fame. The page's worldwide characterization has this broader evidence, beyond the U.S.-only chart.
- CONFIRMED | the painting that came back was treated differently from the one that left. | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 058 (changed-reception). Read the source and compared the claim with its excerpt. Matching text: "Le vol de l’œuvre par Vincenzo Peruggia, ouvrier vitrier au musée du Louvre, en 1911, lui a donné une célébrité mondiale et surtout plus populaire. La Joconde est devenue, après sa redécouverte en 1913, le principal chef-d’œuvre du Louvre, l’œuvre la plus célèbre au monde, source inépuisable de détournements aussi bien artistiques que publicitaires.".
- CONFIRMED | It took a respected Leonardo and, over 29 months of headlines, made it the most famous painting on Earth. | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 059 (most-famous). The Louvre catalogue says the theft brought worldwide and more popular celebrity, and after rediscovery in 1913 it became "l’œuvre la plus célèbre au monde". This supports the broad reception thesis; 29 months here is rounded narrative language, unlike the explicitly recomputed span in entry 62.
- CONFIRMED | Theft & ~26-hour gap Documented. Stolen 21 Aug 1911; the absence went unnoticed for ~26 h (noticed 22 Aug). Period press; Smithsonian / LoC Chronicling America. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/; https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 060 (ledger-1). The theft date is in the LoC guide; Smithsonian gives "It would be 26 hours before someone noticed that the painting was missing." Calendar arithmetic confirms the next date, Tuesday 22 August.
- CONFIRMED | Thief: Vincenzo Peruggia Documented. A former Louvre glazier who had fitted the protective glass. Trial record, Florence Archivio di Stato ; standard accounts. | https://collections.louvre.fr/en/ark:/53355/cl010062370; https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1; https://en.wikipedia.org/wiki/Vincenzo_Peruggia | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 061 (ledger-2). The Louvre catalogue calls him "ouvrier vitrier" and the full interrogation/Wikipedia material describes the protective-case work. This confirms the composite ledger statement.
- CONFIRMED | Chosen for its small size Scholarly, well-supported. Taken partly because it was small , not because it was then the most famous. Sassoon, Becoming Mona Lisa ; the small size is a fact (Louvre). | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-3; https://www.aol.co.uk/news/day-mona-lisa-theft-made-003709636.html; https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 063 (ledger-4). The Florence archive directly states the size motive and the Louvre documents the size; the indexed Sassoon interview makes the same point. The book citation itself is not sufficient evidence of the exact book wording; see entries 73 and 79.
- CONFIRMED | Crowds at the empty wall Documented (diaries/press). Visitors came to stare at the bare wall and four pegs; Kafka and Brod among them. Brod's diary; Smithsonian. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/; https://era.ed.ac.uk/items/5b76982c-83d1-47f2-9cd7-3ec540fd4455 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 064 (ledger-5). The full Smithsonian article names both "Kafka" and "Max Brod" and the "mark of shame". Seaton supplies the Brod attribution trail. The record's single excerpt mentioned only Kafka.
- CONFIRMED | Not the supreme work pre-1911 Scholarly. Modest guidebook space, not the museum's most-visited object. Sassoon; Smithsonian on pre-1911 guidebooks. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 065 (ledger-6). The full Smithsonian article supports both guidebook space and "far from the most visited item in the museum". The record's chosen excerpt supports only the former.
- CONFIRMED | Long admired before 1911 Documented. Guards the "shift in degree, not from zero" calibration: Vasari (1550), Pater (1869). | https://sourcebooks.web.fordham.edu/source/vasari1.asp; https://www.gutenberg.org/files/4059/4059-h/4059-h.htm; https://www.gutenberg.org/files/2398/2398-h/2398-h.htm | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 066 (ledger-7). Read the 1550 Vasari text and Pater's 1869 essay, including publication history. Both directly praise the painting before 1911.
- CONFIRMED | Theft as the engine of fame Scholarly thesis, strongly held [...] not a measured proportion. Sassoon argues it forcefully; this page presents it as a major catalyst, not the sole cause . The chart shows a coverage spike, which is real but is not itself a measure of causation. | https://academic.oup.com/hwj/article-abstract/51/1/1/706426 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 067 (ledger-8). The full Sassoon abstract argues for multiple contributing circumstances. The ledger appropriately labels a thesis, not a measured causal share; the data show a coverage spike only. The record's short excerpt is narrower than this composite claim.
- CONFIRMED | Peruggia's patriotic motive Contested. His own claim; his letters suggest he hoped to profit. Trial record; standard accounts are skeptical. | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/io-sono-e-mi-chiamo-peruggia-vincenzo/il-processo | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 068 (ledger-9). Read the source and compared the claim with its excerpt. Matching text: "La sentenza confermò sostanzialmente le posizioni già espresse dall’accusa, rigettando le conclusioni della perizia psichiatrica e ribadendo che lo scopo di lucro era dimostrato dalle lettere di Peruggia al padre in cui annunciava l’imminente fortuna che stava per conquistare e il suo prossimo rientro in Italia, ed era ulteriormente provato dal tentativo di vendita a Londra.".
- CONFIRMED | "Theft of Mona Lisa: Topics in Chronicling America," Research Guides, Library of Congress [...] a guide to the digitized U.S. newspaper coverage of the theft (1911), recovery (late 1913), and trial (1914). guides.loc.gov/chronicling-america-theft-mona-lisa | https://guides.loc.gov/chronicling-america-theft-mona-lisa | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 070 (citation-loc). Read the source and compared the claim with its excerpt. Matching text: "August 21, 1911 Reports from Paris, France that the “Mona Lisa,” painted by the famous Leonardo da Vinci was stolen from the Louvre. 1912-1913 As the news industry reports the “Mona Lisa’s” theft, she gains world fame. October 1913 Authorities at the Louvre are taking heavier precautious to safeguard other famous works. December 12, 1913 The “Mona Lisa” was recovered in Florence, Italy at the arrest of her thief, Vincenzo Perugia, who claims he was avenging Italy. June 5, 1914 The court in Florence, Italy gave Vincenzo Perugia a “very light sentence,” which was one year and fifteen days.".
- CONFIRMED | "Vincenzo Peruggia," Wikipedia, citing the Florence Archivio di Stato trial and interrogation records: theft 21 Aug 1911, arrest in Florence 12 Dec 1913, painting returned to the Louvre 4 Jan 1914, 1914 trial and sentence; his patriotic claim and the contradicting letters. en.wikipedia.org/wiki/Vincenzo_Peruggia | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/io-sono-e-mi-chiamo-peruggia-vincenzo/il-processo; https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia-1; https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia; https://en.wikipedia.org/wiki/Vincenzo_Peruggia | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 071 (citation-peruggia). The judgment caption alone confirms only the 1914 judgment. The full archive theft/arrest/process accounts and Wikipedia support the remaining biography, dates, sentence and claimed motive; the page's Wikipedia link resolves.
- CONFIRMED | Musée du Louvre, collections catalogue entry for Portrait de Lisa Gherardini, épouse de Francesco del Giocondo (La Joconde) [...] oil on poplar; panel ≈ 79.4 × 53.4 cm. collections.louvre.fr | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 074 (citation-louvre). Read the source and compared the claim with its excerpt. Matching text: "Dimensions Hauteur : 0,794 m ; Hauteur avec accessoire : 1,01 m ; Largeur : 0,534 m ; Largeur avec accessoire : 0,77 m Materials and techniques huile sur bois (peuplier)".
- CONFIRMED | Walter Pater, "Leonardo da Vinci," first published in the Fortnightly Review (November 1869), collected in Studies in the History of the Renaissance (1873) [...] the celebrated passage beginning "She is older than the rocks among which she sits." | https://www.gutenberg.org/files/4059/4059-h/4059-h.htm | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 077 (citation-pater). Read the source and compared the claim with its excerpt. Matching text: "1869. NOTES ON LEONARDO DA VINCI. Appeared in Fortnightly Review in November, 1869. Reprinted 1873 in Studies in the Renaissance.".
- CONFIRMED | Marcel Duchamp, L.H.O.O.Q. , 1919 [...] a "rectified readymade" made by pencilling a mustache and goatee onto a cheap postcard reproduction of the Mona Lisa. Museum of Modern Art. moma.org | https://www.moma.org/audio/playlist/352/4918; https://www.moma.org/explore/inside_out/2011/04/01/five-for-friday-only-joking/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 078 (citation-duchamp). The original MoMA audio says reproduction and names 1919. Supplementary MoMA collection commentary specifies "a cheap postcard of Da Vinci's Mona Lisa", supporting postcard, mustache and goatee. The audio excerpt alone did not establish postcard.
- CONFIRMED | "She is older than the rocks among which she sits." | https://www.gutenberg.org/files/2398/2398-h/2398-h.htm | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 081 (footnote-pater-quote). Read the source and compared the claim with its excerpt. Matching text: "She is older than the rocks among which she sits;".
- CONFIRMED | a "rectified readymade" | https://www.moma.org/audio/playlist/352/4918 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 082 (footnote-readymade-quote). Read the source and compared the claim with its excerpt. Matching text: "Rectified readymade: pencil on reproduction of Leonardo da Vinci's Mona Lisa".
- CONFIRMED | admired since Vasari in 1550 | https://sourcebooks.web.fordham.edu/source/vasari1.asp | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 083 (summary-1550). Read the source and compared the claim with its excerpt. Matching text: "Giorgio Vasari (1511-1574) wrote about hundreds of artists in his Lives of the Most Eminent Italian Architects, Painters, and Sculptors , which he published first in 1550, and in a revised edition in 1568.".
- CONFIRMED | a literary touchstone since Pater in 1869 | https://www.gutenberg.org/files/4059/4059-h/4059-h.htm | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 084 (summary-1869). Read the source and compared the claim with its excerpt. Matching text: "1869. NOTES ON LEONARDO DA VINCI. Appeared in Fortnightly Review in November, 1869. Reprinted 1873 in Studies in the Renaissance.".
- CONFIRMED | On 21 August 1911 a Louvre workman named Vincenzo Peruggia walked out with it | https://guides.loc.gov/chronicling-america-theft-mona-lisa | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 085 (summary-theft). Read the source and compared the claim with its excerpt. Matching text: "In the early morning of August 21, 1911, former Louvre employee, Vincenzo Perugia, made off with the “Mona Lisa,” with hopes to return the painting to her rightful home, Italy.".
- CONFIRMED | Crowds came to stare at the four iron pegs. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 087 (summary-crowd). Read the source and compared the claim with its excerpt. Matching text: "Thousands of spectators, including Franz Kafka, flooded into the Salon Carré when the Louvre reopened after a week to stare at the empty wall with its four lonely iron hooks.".
- CONFIRMED | through two and a half years of headlines, turned it into the most famous picture on Earth. | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 088 (summary-fame). The Louvre catalogue explicitly links theft to worldwide popular celebrity and rediscovery in 1913 to its status as the most famous work. "Two and a half years" is a rough description of the period; exact elapsed time is handled in entries 26 and 62.
- CONFIRMED | over the next two and a half years it became the most famous painting on Earth. | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 089 (meta-fame). The Louvre catalogue explicitly links theft to worldwide popular celebrity and rediscovery in 1913 to its status as the most famous work. "Two and a half years" is a rough description of the period; exact elapsed time is handled in entries 26 and 62.
- CONFIRMED | The Mona Lisa was not always the most famous painting on Earth. | https://collections.louvre.fr/en/ark:/53355/cl010062370 | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 090 (og-fame). Read the source and compared the claim with its excerpt. Matching text: "Le vol de l’œuvre par Vincenzo Peruggia, ouvrier vitrier au musée du Louvre, en 1911, lui a donné une célébrité mondiale et surtout plus populaire. La Joconde est devenue, après sa redécouverte en 1913, le principal chef-d’œuvre du Louvre, l’œuvre la plus célèbre au monde, source inépuisable de détournements aussi bien artistiques que publicitaires.".
- CONFIRMED | The theft, early photography, Pater and Vasari's prestige, and Duchamp's 1919 mockery all fed the painting's fame | https://academic.oup.com/hwj/article-abstract/51/1/1/706426; https://www.moma.org/explore/inside_out/2011/04/01/five-for-friday-only-joking/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 091 (uncertainty-factors). Sassoon's full abstract covers critics, Leonardo's reputation, illustrated publicity, theft and avant-garde parody. MoMA identifies the 1919 Duchamp work and postcard support. These substantiate the stated plurality, without apportioning effects.
- CONFIRMED | the patriotic version is his own, contradicted by his letters. | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/io-sono-e-mi-chiamo-peruggia-vincenzo/il-processo | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 092 (uncertainty-letters). Read the source and compared the claim with its excerpt. Matching text: "La sentenza confermò sostanzialmente le posizioni già espresse dall’accusa, rigettando le conclusioni della perizia psichiatrica e ribadendo che lo scopo di lucro era dimostrato dalle lettere di Peruggia al padre in cui annunciava l’imminente fortuna che stava per conquistare e il suo prossimo rientro in Italia, ed era ulteriormente provato dal tentativo di vendita a Londra.".
- CONFIRMED | The "~26 hours unnoticed" is the figure given by careful secondary accounts citing the period record | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 093 (uncertainty-26). Read the source and compared the claim with its excerpt. Matching text: "It would be 26 hours before someone noticed that the painting was missing.".
- CONFIRMED | normalized, so the curve is coverage, not how much got digitized. | git show f5e5f84f34:research/a-minor-leonardo/coverage-data.mjs; recreated: arithmetic.json | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 097 (chart-normalization). The formula is count / corpusPages * 1e6. Recomputed all 48 rates and rounded display cells. This removes total corpus-volume scaling, though it cannot control a changing mix of newspapers.
- CONFIRMED | The chart measures coverage, not cause. | git show f5e5f84f34:research/a-minor-leonardo/coverage-data.mjs | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 102 (chart-measures). The instrument returns phrase-matching page totals and divides by yearly corpus pages. It has no causal model or causal allocation; the page correctly separates those concepts.
- CONFIRMED | It is the U.S. digitized press | git show f5e5f84f34:research/a-minor-leonardo/coverage-data.mjs | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 107 (limit-us). The query is restricted to "partof:chronicling+america" and the denominator uses the Chronicling America newspaper collection. The page correctly limits the plotted corpus to digitized U.S. newspapers.
- CONFIRMED | A coverage spike is evidence of attention, not of cause | git show f5e5f84f34:research/a-minor-leonardo/coverage.json; git show f5e5f84f34:research/a-minor-leonardo/coverage-data.mjs | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 110 (limit-cause). The stored counts show a phrase-coverage spike. The program supplies no design separating causes; the page appropriately avoids treating the spike as a causal proportion.
- CONFIRMED | Where the record is genuinely uncertain, the page states the uncertainty rather than a false precision. | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 121 (uncertainty-stated). Read the four expressly named uncertainties: causal shares, prior fame, motive, and approximate discovery interval. They are visible in the prose as claimed. This local description does not validate the separate unsupported margin and precision assertions.
- CONFIRMED | The two rivals are the ones that outranked the Mona Lisa in pre-1911 Louvre guidebooks. | https://www.smithsonianmag.com/arts-culture/stolen-how-the-mona-lisa-became-the-worlds-most-famous-painting-16406234/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 123 (rival-objects). The full Smithsonian article identifies the two guidebook rivals by name: "Nike of Samothrace and Venus de Milo". This guidebook comparison is distinct from the page's incorrect claim that both newspaper-count series always exceeded Mona Lisa before 1911.
- CONFIRMED | This layer's check reads the bytes this page serves and asserts against them. | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 125 (checks-bytes). verify-immersive reads "public/strata/a-minor-leonardo/index.html" and asserts selected content. This literal dependence is true, though it does not establish exhaustive validation.
- CONFIRMED | It is reading what you are reading. | git show f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs; recreated: live.html | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 126 (checks-reading). The immersive check reads the committed served HTML. The live hero-to-main text matched that HTML, so the limited assertion about reading page bytes is supported.
- CONFIRMED | It is published to read. Running it needs the repository around it. | git show f5e5f84f34:public/checks/manifest.json; recreated: reader-prose.txt and reader-immersive.txt | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 129 (checks-published). Both linked verifier files and the advertised coverage script downloaded with HTTP 200 and matched pinned bytes. The immersive command requires repository files and failed with ENOENT in the empty reader directory, as the page warns. The main verifier instead skipped prose checks and printed 13/13; this is not a full page verification.
- CONFIRMED | Theft [...] 21 Aug 1911 | https://guides.loc.gov/chronicling-america-theft-mona-lisa | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 170 (event-marker-0). The event date agrees with the named historical source. Scope correction to N1: these full date strings are chart metadata; drawChart displays only the part of mk.label before its separator. The full label is not all visible chart text.
- CONFIRMED | Recovered [...] arrest 12 Dec 1913 | https://archiviodistatofirenze.cultura.gov.it/asfi/mostre/vincenzo-peruggia | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 171 (event-marker-1). The event date agrees with the named historical source. Scope correction to N1: these full date strings are chart metadata; drawChart displays only the part of mk.label before its separator. The full label is not all visible chart text.
- CONFIRMED | Rehung at the Louvre [...] 4 Jan 1914 | https://en.wikipedia.org/wiki/Vincenzo_Peruggia | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 172 (event-marker-2). The event date agrees with the named historical source. Scope correction to N1: these full date strings are chart metadata; drawChart displays only the part of mk.label before its separator. The full label is not all visible chart text.
- CONFIRMED | Both correct a celebrated art-historical 'fact' from the primary record and refuse the cheap version. There, Michelangelo's horned Moses is traced to one unvowelled Hebrew root [...] and the popular 'the Septuagint did it' is shown false. Here, the Mona Lisa's supremacy is traced to a dated crime [...] and the popular 'it was a nobody before' is shown false too. The discipline is the same: name what is documented, name what is contested, overclaim neither. | git show f5e5f84f34:src/content/strata/the-horns-of-moses.md; git show f5e5f84f34:src/content/strata/a-minor-leonardo.md | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 173 (relation-the-horns-of-moses). Read the pinned linked stratum and the relationship text in context. The description accurately states that page's argument. The embedded external facts are checked separately in entries 178-183; the reference to a lower-bound warning in 177 reports what the linked page says, not an endorsement of that warning.
- CONFIRMED | Two myth-corrections of identical shape: a confident historical story that the record refutes. Columbus did not prove the Earth round (it was settled Greek science, measured by Eratosthenes); the Mona Lisa was not always the most famous painting (it became so after 1911). Both replace an eternal 'fact' with a datable event [...] and both insist the correction be calibrated, not flipped into the opposite overclaim. | git show f5e5f84f34:src/content/strata/did-columbus-prove-earth-round.md; git show f5e5f84f34:src/content/strata/a-minor-leonardo.md | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 174 (relation-did-columbus-prove-earth-round). Read the pinned linked stratum and the relationship text in context. The description accurately states that page's argument. The embedded external facts are checked separately in entries 178-183; the reference to a lower-bound warning in 177 reports what the linked page says, not an endorsement of that warning.
- CONFIRMED | Both watch a vivid, wrong object get manufactured [...] and refuse to blame a single cause. Schiaparelli's canali became a Martian civilization because a word supplied the maker and the eye supplied the geometry; the Mona Lisa became an icon because a theft supplied the headlines and a half-century of poets, photographers and mockers had already prepared the ground. Neither correction lets one factor carry the whole weight. | git show f5e5f84f34:src/content/strata/canals-of-mars.md; git show f5e5f84f34:src/content/strata/a-minor-leonardo.md | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 175 (relation-canals-of-mars). Read the pinned linked stratum and the relationship text in context. The description accurately states that page's argument. The embedded external facts are checked separately in entries 178-183; the reference to a lower-bound warning in 177 reports what the linked page says, not an endorsement of that warning.
- CONFIRMED | A record-correction sibling: the famous story (the guards turned spontaneously cruel; the painting was always a masterpiece) turns out to be assembled after the fact, not… | git show f5e5f84f34:src/content/strata/stanford-prison-experiment.md; git show f5e5f84f34:src/content/strata/a-minor-leonardo.md | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 176 (nearby-stanford-prison-experiment). Read the pinned linked stratum and the relationship text in context. The description accurately states that page's argument. The embedded external facts are checked separately in entries 178-183; the reference to a lower-bound warning in 177 reports what the linked page says, not an endorsement of that warning.
- CONFIRMED | The member that measures its attestation rather than citing one, and warns in the same breath that its counts are a noisy lower bound. | git show f5e5f84f34:src/content/strata/the-date-that-can-only-say-no.md; git show f5e5f84f34:src/content/strata/a-minor-leonardo.md | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 177 (nearby-the-date-that-can-only-say-no). Read the pinned linked stratum and the relationship text in context. The description accurately states that page's argument. The embedded external facts are checked separately in entries 178-183; the reference to a lower-bound warning in 177 reports what the linked page says, not an endorsement of that warning.
- CONFIRMED | Michelangelo's horned Moses is traced to one unvowelled Hebrew root | https://www.thetorah.com/article/moses-shining-or-horned-face | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 178 (moses-horns). Read the full linguistic discussion as well as the supplied art-history excerpt: Hebrew qrn, the Vulgate rendering and the Michelangelo reception support the claim. The source explicitly distinguishes the Greek translation.
- CONFIRMED | the popular 'the Septuagint did it' is shown false. | https://www.thetorah.com/article/moses-shining-or-horned-face | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 179 (moses-greek). Read the source and compared the claim with its excerpt. Matching text: "LXX (Greek Septuagint)[...]“the appearance of the skin of his face was charged with glory”;".
- CONFIRMED | Columbus did not prove the Earth round | https://www.history.com/articles/christopher-columbus-never-set-out-to-prove-the-earth-was-round | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 180 (columbus-round). Read the source and compared the claim with its excerpt. Matching text: "neither Columbus nor his Spanish patrons thought Earth was a finite plane instead of a round planet.".
- CONFIRMED | it was settled Greek science | https://classics.mit.edu/Aristotle/heavens.2.ii.html | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 181 (greek-round-earth). The Aristotle passage is direct pre-Columbian evidence for a spherical Earth. Read the surrounding Book II argument, not just the excerpt; it discusses observational evidence and existing circumference estimates.
- CONFIRMED | measured by Eratosthenes | https://www.aps.org/apsnews/2006/06/eratosthenes-measures-earth | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 182 (eratosthenes). Read the source and compared the claim with its excerpt. Matching text: "Eratosthenes’ most famous accomplishment is his measurement of the circumference of Earth.".
- CONFIRMED | Schiaparelli's canali became a Martian civilization because a word supplied the maker and the eye supplied the geometry | https://science.nasa.gov/mars/triumph-of-mariner-4/ | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 183 (mars-canali). The NASA history says the lines were "optical illusions" and "mis-translations branded them canals"; it discusses the resulting artificial-waterway interpretation. curl obtained the full readable article even though the online checker had classified it UNREADABLE.
- CONFIRMED | A Veronese or a David in the same museum is a wall; the Mona Lisa is hand luggage. | https://www.louvre.fr/en/explore/the-palace/from-the-mona-lisa-to-the-wedding-feast-at-cana; https://www.louvre.fr/en/explore/the-palace/think-big | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 184 (veronese-david). The source supports Veronese's monumental Wedding Feast. Supplementary Louvre room commentary describes David's Coronation as "At six metres high and almost ten metres wide". Both are vastly larger than the roughly 0.79 by 0.53 metre Mona Lisa; the page's comparison is supported.
- CONFIRMED | The cm→inch conversions are recomputed in the verifier. | git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs; recreated: verify-prose.txt | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 185 (inch-conversion-check). The verifier calculates 77 / 2.54 and 53 / 2.54, printing 30.3 and 20.9 inches. Recreated the run successfully.
- CONFIRMED | the figures it checks are marked in the prose | git show f5e5f84f34:src/content/strata/a-minor-leonardo.md; git show f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 186 (figures-marked). The checked dates, durations and dimensions are visibly bold in the MD and HTML. This statement concerns marked figures, not the separate, false promise to check every number.
- CONFIRMED | the "what's proven vs. contested" ledger at the end is the apparatus, not an afterthought. | git show f5e5f84f34:src/content/strata/a-minor-leonardo.md; git show f5e5f84f34:public/strata/a-minor-leonardo/index.html | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 187 (honesty-ledger). Read the ten-row "The record" ledger in the MD and HTML, including documented, scholarly and contested categories. This is a checkable description of the apparatus's presence.
- CONFIRMED | None of these is hidden in the prose; each is priced into the claim it supports. | git show f5e5f84f34:public/strata/a-minor-leonardo/index.html | Carried forward from the checker's source reread in research/assay-line/cited-1014b/a-minor-leonardo/N2-codex.md. Entry 188 (uncertainties-visible). Each of the four named uncertainties appears explicitly in the main prose or uncertainty paragraph. This checks their visibility, without treating the phrase as a blanket warranty for every other assertion.

## Findings

- [declined] X1: Re-read f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs:36-57. It sets the two endpoints and computes "const months = (1914 - 1911) * 12 + (0 - 7)", ignoring their days. The span is 867 days, or 28 full calendar months and 14 days. The narrative uses an approximate duration, so the issue is precision rather than false endpoint dates. Retain MINOR and list at tier C. This duration sentence is distinct from the empty-wall assertion fixed under X8.
- [fixed] X2: Restored the recovery referent and exact quoted fragment in HTML and Markdown; added the located interview citation and its Corrected line.
- [fixed] X3: Attributed the reporting-history quotation to the Library of Congress in both page copies, with its own citation and Corrected line.
- [declined] X4: Re-read f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs:36-57 and the ledger at src/content/strata/a-minor-leonardo.md:76. The code uses "const months = (1914 - 1911) * 12 + (0 - 7)". That is a month-index difference, not the exact 867-day interval (28 months and 14 days). Calling the rounded span recomputed without naming the convention overstates precision. Retain MINOR and list at tier C; this ledger row is not the empty-wall sentence being fixed under X8. Same precision issue as X1.
- [fixed] X5: Replaced the Smithsonian author credit with James Zug in both page copies and the research README; added a Corrected line.
- [fixed] X6: Replaced the first-record assertion with the distinction between Vasari's published identification and Vespucci's earlier note, with Heidelberg citation and a Corrected line shared with X7.
- [fixed] X7: Fixed the same first-record wording in both footnote copies and the README; the Corrected line is shared with X6.
- [fixed] X8: Changed the dek, plain metadata, heading, extended-absence wording, research description and verifier assertion to describe the missing painting. Added the 1912 replacement evidence and a Corrected line, retaining the initial empty-place crowds.
- [open note:f1b0bd] X9: Changed this layer's chart title and accessible legend label to artworks; shares the sculpture correction with X22. The relation note in src/content/strata/the-newsquake.md:14 still calls the sculpture controls rival paintings. That copy is in another layer and remains open for the directing instance.
- [declined] X10: Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:64-67: "peak.year >= 1911 && peak.year <= 1914". coverage.json has annual observations, not within-year dates. That supports a theft-era annual peak, not an exact August 1911 to January 1914 window. Retain MINOR and list at tier C for the standalone HTML sentence at public/strata/a-minor-leonardo/index.html:186. The similar time clause within the different, already-WRONG Markdown sentence is handled with X21.
- [fixed] X11: Replaced live-count descriptions with the saved pull documented on 2026-07-12 in HTML, Markdown and README; added a Corrected line.
- [fixed] X12: Corrected the page, README, generator and verifier descriptions of --check; its console output now says cache-or-fetch. The script retains its cache behavior. Added a Corrected line.
- [open note:d6f06f] X13: Replaced the lower-bound guarantee in this layer's HTML, README and generator caveats; regenerated coverage.json and its embedded copy. Added a Corrected line. On 2026-10-18 the portal the-date-that-can-only-say-no was corrected: its relation note (src/content/strata/the-date-that-can-only-say-no.md:21), three apparatus bullets, Nearby box, displayed quotation (now this layer's exact limit (ii) wording) and verifier (research/the-date-that-can-only-say-no/verify.mjs), with a dated Corrected 2026-10-18 line there. Apart from this layer's generated Nearby box (public/strata/a-minor-leonardo/index.html:339), which the landing rebuild regenerates from the corrected note, the only thing left is the preserved quotation in There, Here (public/strata/there-here/index.html:250 and research/lineage-edge-grammar/page.json:1, notes[1357]), which still says the counts are "a noisy lower bound" and needs an annotation.
- [fixed] X14: Removed the equal-effects and guaranteed-signal claims from the page, README and generator caveats; regenerated the data copies. Added a Corrected line.
- [declined] X15: Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:100-131: CANON contains 21 selected fragments and uses "if (!htmlText.includes(needle)) missing.push(c);". The prose verifier also checks selected arithmetic and strings. I independently replayed both in memory: changing the HTML duration from 29 to 99 months still gives 19/19 and 12/12; changing the protected theft year makes the immersive check fail. Real selected checks exist, but the universal coverage claim overstates them. Relevel WRONG to MINOR under the explicit own-checking rule and list at tier C. This standalone sentence is not being fixed for a separate WRONG.
- [declined] X16: Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:85-137. It normalizes text and tests selected CANON fragments with "if (!htmlText.includes(needle)) missing.push(c);", not full Markdown-to-HTML character equality. My in-memory 29-to-99 duration mutation still passes both verifiers; the protected theft-year mutation fails. Relevel WRONG to MINOR: this is an overstatement of the scope and precision of a real text check, covered by the user's own-checking rule. List at tier C. X17 shares this HTML sentence but is also MINOR; no separate WRONG requires that sentence to be fixed. Same overclaim as X19.
- [declined] X17: Re-read f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs:36-114 and 128-141. It calculates selected hard-coded quantities, including "const months = (1914 - 1911) * 12 + (0 - 7)", then checks selected prose strings; it does not extract and recompute every displayed date or number. My in-memory 29-to-99 HTML duration change leaves both verifiers passing. Relevel WRONG to MINOR under the explicit rule for "Every number is recomputed live" and similar checking overclaims. List at tier C; X16 is the other MINOR clause in this sentence. Same error as X18.
- [declined] X18: Re-read f5e5f84f34:research/a-minor-leonardo/verify-a-minor-leonardo.mjs:36-114 and 138-141: selected constants and arithmetic are checked, followed by presence tests such as "ok('the prose shows "361" (years since Vasari)', /361/.test(prose), 'figure present');". There is no extraction and validation of every date and number. The immersive check likewise tests selected fragments. Relevel WRONG to MINOR under the explicit all-numbers-recomputed rule and list at tier C. The Markdown sentence at line 34 is not being fixed for a separate WRONG. Same error as X17.
- [declined] X19: Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:85-137: "if (!htmlText.includes(needle)) missing.push(c);" checks 21 normalized CANON fragments, not equality of the complete Markdown and HTML prose. My in-memory duration mutation leaves the check passing, while a protected theft-year change fails. Relevel WRONG to MINOR because this is the overreach of a real text check, as specified by the own-checking rule. List at tier C. In src/content/strata/a-minor-leonardo.md:32 this is a separate sentence from the WRONG rival-ranking and sculpture-type claims; its other flagged clause, X20, is also MINOR. Same error as X16.
- [declined] X20: Re-read f5e5f84f34:research/a-minor-leonardo/verify-immersive.mjs:54-80: "preYears.every((y) => ve(y) > ml(y))" checks Venus throughout the pre-period; Winged Victory is compared only for 1905 there. coverage.json:121-131 has 1910 Mona Lisa 143 and Winged Victory 65. My replay also confirms the raw-readout unit bug is not caught by the passing verifier. The verifier computes real selected chart propositions, but "every claim" overstates its coverage. Relevel WRONG to MINOR under the own-checking rule and list at tier C; X21 and X25-X40 separately retain the actual data errors as WRONG. This sentence shares X19, not those WRONG claims.
- [fixed] X21: Corrected the ranking sentence and its time clause, related metadata and README using the saved series; added a Corrected line. The separately listed HTML precision claim X10 remains unchanged.
- [open note:f1b0bd] X22: Corrected equivalent rival-painting descriptions throughout this layer's prose, accessible labels, generator and generated caveats; shares the Corrected line with X9. The relation note in src/content/strata/the-newsquake.md:14 still calls the sculpture controls rival paintings. This is the same open copy as X9 and needs the same board note.
- [dated] X23: Dated this claims record 2026-10-14 and regenerated only this layer's placard from the existing read-only --check --sample output. It now says Last verified 14 October 2026 and links the generated record block. No Updated line, per the triage exception.
- [dated] X24: Set JSON-LD dateModified to 2026-10-14. The dated Corrected lines already document the revision; no separate Updated line is required.
- [fixed] X25: corrected the shared hover-unit selection for 1905; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X26: corrected the shared hover-unit selection for 1906; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X27: corrected the shared hover-unit selection for 1907; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X28: corrected the shared hover-unit selection for 1908; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X29: corrected the shared hover-unit selection for 1909; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X30: corrected the shared hover-unit selection for 1910; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X31: corrected the shared hover-unit selection for 1911; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X32: corrected the shared hover-unit selection for 1912; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X33: corrected the shared hover-unit selection for 1913; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X34: corrected the shared hover-unit selection for 1914; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X35: corrected the shared hover-unit selection for 1915; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X36: corrected the shared hover-unit selection for 1916; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X37: corrected the shared hover-unit selection for 1917; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X38: corrected the shared hover-unit selection for 1918; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X39: corrected the shared hover-unit selection for 1919; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.
- [fixed] X40: corrected the shared hover-unit selection for 1920; the counts are unchanged. Added one Corrected line for X25–X40 and a regression check of both units and values.