LeBon-G: Arbiter (Step 3)

Project: Gustave Le Bon (1841-1931), The Evolution of Matter (English, 1907) and The Evolution of Forces (English, 1908), plus the 1990 Vassilatos review (Journal of Borderland Research) linking them to Thomas Moray. Date: 2026-10-08.

Inputs read: claims.md, physicist.md, mathematician.md, contrarian.md (each once, in full); sites.md and research/ notes through the reviewers' quotations; the Moray 1926 Radiant Energy extract checked myself. Tags: [read] document text; [calc] my own arithmetic (.venv Python); [web] URL; [memory] with confidence.

Source gap: none. All three documents have text (the Matter PDF is OCR throughout; tables were checked against page images by the deep reader and by the mathematician for pp.73, 74, 76). The Vassilatos review is a secondary source from an interested party (BSRF sells the books it recommends). The two Le Bon books are primary. Not read by any reviewer, so conclusions about them rest on the Investigator's notes or on Le Bon's and Legge's own accounts: the French originals, Kalahne 1905, Gates 1904, Ramsay and Spencer 1906, Campbell 1905, De Heen 1901/1912, Nye 1974, the Nobel nomination, the IMEC finding aid, Le Bon's CR notes of 5 April and 5 August 1897.

1. Where the three reviewers differ, and my recomputation

2. Corrected claim list (final verdicts)

Claim 1. Matter is perishable (all atoms slowly dissociate toward "ether")

Agreement: all three split it the same way. Weak form Established; strong form Tested, not supported; ether end-state superseded. Disagreement: none of substance (the mathematician adds "Not testable as stated" for the ether end-state because no measurable prediction exists; the others call it superseded; both are true, a concept that made no distinct prediction and was dropped). Strongest argument: the physicist's and contrarian's. Le Bon's only quantitative handle, "1 in 100,000,000 molecules" (Matter p.427), is Rutherford's ion count in a gas, not a count of destroyed atoms; and the leaks he measured were traced by J. J. Thomson to radioactive impurity (Legge's note, p.428), by Elster and Geitel to penetrating radiation, and later to radon, wall radioactivity and cosmic rays (Wilson, Hess [memory, high]). Rutherford and Cooke and McLennan and Burton found closed-vessel leak fell by up to 30% inside thick shielding, as summarised in a 1926 abstract [web] https://authors.library.caltech.edu/records/1bch2-v3x84. His own 8-day figure (1-2 degrees per minute, p.430) has no blank run in the same geometry [corrected by QE revision R1: an instrument baseline of at most 1 degree per hour with the cap on is stated on p.354]. Final verdict (split): - Some atoms disintegrate, and electrons can be freed from atoms by light, heat, chemistry: Established (credit shared with Becquerel, the Curies, Rutherford, Soddy, Thomson, Lenard; Le Bon argued the generalisation early). - Every atom of ordinary matter dissociates spontaneously: Tested, not supported (indirectly: the background ionisation he relies on was traced elsewhere, and stable nuclides show no decay products; proton partial lifetime limits above 8e33 years [web]). No test was aimed at his own protocol. - "Ether" as the end state: Historically superseded but instructive. Settles it: a blank-corrected electroscope or ionisation-chamber run on a clean ordinary metal (copper, zinc), inside and outside thick shielding, with radon monitoring, over days. Mainstream expects the same rate as an empty chamber.

Claim 2. Atoms are reservoirs of intra-atomic energy; conservation of energy is a "dogma"

Agreement: the arithmetic is right; a very large internal energy store exists; the dynamo argument fails. Disagreement: (i) how to label the 6,800-million-hp estimate (physicist: Historically superseded; mathematician and contrarian: Refuted by calculation); (ii) how to label the "dogma" argument (physicist: Not testable as stated / Tested, not supported; mathematician: Refuted by calculation; contrarian: Not testable as stated). Resolution of (i): both are right about different statements. [calc] The number is the kinetic energy of 1 g if the whole gram flew at c/3: 5.0e12 J. As a statement that matter holds an energy of order mc^2 (9.0e13 J/g) it is the right order of magnitude (Historically superseded but instructive, because he had no E = mc^2 and no mechanism; Abraham's 6e13 J and Thomson's 1.0e12 J, which he quotes, were also electromagnetic-mass estimates). As a budget for what a gram of ordinary matter can yield it is refuted by his own data: his p.427 concession that only 1 in 1e8 molecules is dissociated, his own radium figure (0.116 W/g, about 1.4e10 J per gram over the whole chain, [calc], about 350 times lower), and his p.80-81 admission that waiting millions of years makes the "millions of kilogrammetres" amount to little. The fair reading is that his number is a premise (total conversion at 1e8 m/s), not a result. Resolution of (ii): split the statement. The philosophical point (that "conservation of energy" is partly a convention, as Poincaré said in Science and Hypothesis ch.8 [web] https://en.wikisource.org/wiki/Science_and_Hypothesis/Chapter_8) is not testable as stated, and Le Bon says so himself: "these are barren discussions, since experiment is incapable of throwing light on this question" (Forces p.83, [read: contrarian]). The practical claim, that potential energy can always be given whatever value is needed, is false for conservative forces: U(x) is fixed by independent force measurement and exists only when curl F = 0 (mathematician). Poincaré rejected the reading that the law is a mere tautology for a limited system, so Le Bon's use of him goes further than the source supports. The physicist's "Tested, not supported" applies to the physics; the mathematician's "Refuted by calculation" applies to the potential-energy loophole specifically. Dynamo: Refuted by calculation. [calc] A 0.5 m conductor at 10 m/s in 1 T gives 5 V; with 5 ohm, 1 A, 5 W electrical; back force B I l = 0.5 N; shaft power F v = 5 W. Output equals input; with 2.5 ohm the drag doubles. If an atomic reservoir supplied the output, shaft torque would not rise with load; it does on every generator. The magnetic force does no work (perpendicular to velocity), which is why "a finite magnet gives unlimited electricity" is true and unremarkable. Strongest argument: the mathematician's worked example, which is complete and checkable. Final verdict (split): - Matter holds an enormous internal energy (of order mc^2 per gram, released at about 1e-4 to 1e-2 of it in real processes): Established (by post-1905 physics, not by Le Bon's argument); his 1900-1905 estimate as a magnitude: Historically superseded but instructive. - 6,800 million hp as what a gram of ordinary matter would yield: Refuted by calculation (premise contradicted by his own figures). - Conservation of energy is a dogma sustained by "potential energy": Not testable as stated in its philosophical form; the claim that potential energy can be freely assigned: Refuted by calculation. - Dynamo and magnet output come from dissociating matter: Refuted by calculation. - Light, heat or a small trigger release more energy than they cost, from ordinary bodies: Never properly tested by Le Bon (no energy-in/energy-out table in either book; stored-reserve release by a trigger, as with a spark and gunpowder, is not over-unity). [QE revision R3: the books do propose such a machine as "possible" (Matter p.231; Forces p.38) but give no device, trial or accounting.] Tesla-coil trigger release (Vassilatos): Never properly tested; the books propose the idea but show no engine. Settles it: for the dynamo, nothing new is needed (measure shaft torque against electrical load). For a "trigger release", calorimetry of total energy in versus out for a stated protocol.

Claim 3. Light, heat and chemistry release this energy in ordinary bodies

Agreement: the observations (charge leak under light, UV, flames, chemical reaction) are real and mostly known effects; the dissociation reading is not supported; the hot-metal "emanation" has not been neutrally tested. Disagreement: only on wording. The contrarian lists "Never properly tested, not dismissed without testing" for the emanation; the physicist says the same with Legge's request for independent checking (p.411). Evidence kept: the photo-leak data are good work for 1905: the cut-offs match textbook work functions (lambda_c = 1240 eV nm / phi: Al about 290-300 nm, Cu about 267, Ni about 248, Pt about 221; [memory, high]; Le Bon's table agrees: Al and Zn respond in sunlight, Cu hardly, Ni, steel, Pt only in extreme UV); he separated negative photo-emission from positive-leak air ionisation by vacuum UV (absorbed in about 2 cm of air); bell-and-coherer control against Hertzian waves; cleaning every ten minutes. Kalahne (Heidelberg, 1905) wrote "my observations absolutely confirm that the chemical phenomenon pointed out by Gustave Le Bon is the cause of the radiation"; Gates and Ramsay and Spencer (1906) repeated parts and differed on explanation; Rutherford adopted the quinine and chemical finding into Radio-Activity but classed it as not true radioactivity [read: via Legge and the deep read; the papers themselves unread]. Against: no blank leak rate anywhere; tables carry slips (lamp-black 61 vs 7 degrees, "single to triple" for a sixfold range, thorium "70"); Campbell (J. J. Thomson's lab, 1905) ascribed the chemical leak to heat, answered in an Athenaeum item of 24 March 1906 whose content nobody here has read, so who was right on that specific point is open. Energy check [calc]: discharging a 10 pF electroscope at 1000 V needs 5e-6 J and about 6e10 electrons, so absorbed light of order 1e-7 J suffices; no anomaly in scale, though Le Bon never measured lamp output. Chapter XII (mercury, magnesium, aluminium): amalgam chemistry (oxide film removed, alumina tufts, 102 degrees C, emf 0.75 to 1.65 V); not transmutation. Vassilatos's "managed to prove that transmutations can be made" is not supported by this chapter. Final verdict (split): - Charge leaks under light, spark UV, flames, hydration, oxidation: Established as electroscope phenomena (explained by photoelectric emission, vacuum-UV air ionisation, chemi-ionisation). - These show ordinary matter dissociating and releasing intra-atomic energy: Tested, not supported (Thomson, Rutherford, Campbell, Ramsay, background ionisation), except as noted for the chemical-leak/heat dispute, where the record I can read does not give the outcome. - Warmed ordinary metal expels a limited "emanation" that rest renews: Never properly tested (unchanged by QE revision R1; explanations were offered by Thomson, Wilson, McLennan and Burton, but no controlled replication of his hot-mass protocol is on record; this was not dismissed without comment, but it was not repeated). - Chapter XII as "variability of chemical species" implying transmutation: Established as chemistry; the transmutation reading: changed in QE revision R9 to Not testable as stated (no transmutation is claimed in the text; no experiment tested it). Settles it: repeat the 400 degrees C hot-mass test (cage at 35 degrees C, mass 3 cm away) with a no-source blank, a radon monitor, shielded and unshielded cages, an earthed screen against convection, and a mock-heated cage.

Claim 4. "Black light"

Agreement: the long afterglow, IR stimulation of CaS, IR quenching of ZnS, near-IR transparency of ebonite and black paper, and ZnS-screen imaging are ordinary solid-state luminescence and near-IR optics; his glass-strip, ice-trough and two-halves-of-one-screen controls are real and well designed; the 1896 single "new radiation" through iron and copper is not supported. Disagreement: the label for the 1896 form: physicist "Tested, not supported"; contrarian "Never properly tested and superseded"; the label for the whole bundle (mathematician: "Not testable as stated" for the "distinct invisible light" reading). Resolution of the 1896 form: the evidence on file: one positive reproduction of a different (1897 metal-and-light) experiment by Berget in Lippmann's laboratory; Poincare's protocol questions and presentation of four notes; Le Bon's own admission in 1897 that the experiments "do not invariably succeed" [web] https://www.ghtc.usp.br/livros%20RM%202022/Studia-2-3-web.pdf; his own 1908 re-sorting into radioactive particles, long-wave infra-red and invisible phosphorescence (Forces p.317); H. Becquerel's 1897 suggestion that it was infra-red; contemporary plate-holder-leak objections (1896 press, via an interested reprint at BSRF); Kragh's statement "black light turned out to be imagination rather than reality - most physicists failed to detect the rays" [web] https://arxiv.org/pdf/1208.3114, a historian's characterisation with no specific failed test named in what the Investigator read. I find no published neutral controlled test and no published failure with data. The physicist's "Tested, not supported" over-reads Kragh; the contrarian's "Never properly tested" fits the record. I adopt Never properly tested, with the note that Le Bon himself abandoned the single-agent form, which is the better reason not to rely on it. Points both for and against kept in the record: - For: the 1899 primary notes (CR 128, 16 and 30 January, presented by Poincare) match the book's decay series exactly (so the book is not a later alteration); Gaston Braun fils ran the statuette experiment for him in a photographer's cellar (named operator, but not independent); Izarn (professor of physics) wrote "I was stupefied with the clearness of the results" (Forces p.350, one source: Le Bon's book); Gariel's textbook summary ("almost extraordinary ... no occasion to put them on one side", Forces p.319); the red-darkroom-lamp explanation of others' failures is physically correct (red light quenches ZnS) and testable. [read] - Against or limiting: no never-insolated blank in the 18-month series; no fog level for plates kept 40-60 days; no spectrum of the afterglow; Fig. 35 candle-versus-sunlit-card ranking contradicts luminance estimates [calc: white card in full sun about 2.5e4 cd/m^2, candle flame of order 1e4]; transparency method example reversed (5 s vs 20 s means one quarter, not "twice as transparent"); "bodies under 100 degrees C hardly exceed 5-6 micrometres" wrong (Wien: 373 K peaks at 7.8 micrometres; 310 K at 9.3 micrometres); "atmosphere absorbs all above 5 micrometres" false (8-14 micrometre window); ammoniacal copper sulphate "transparent" doubtful; the 1899 note says the late emission passes 0.05 mm mica but not glass, which is hard to square with its conclusion of "identity" with solar light (contrarian; glass thickness not given, so a tension, not a refutation); statuette caption "eighteen months" contradicts text and primary note (3-4 days after going dark; the caption is the error). Decay data [calc]: intensity as 1/exposure falls roughly as t^-1.2 to t^-1.5, the power law expected from a distribution of trap depths; plate reciprocity failure biases late values downward, so the true decay is slower still. - Metal-on-plate "radiations": consistent with Russell-type chemical fogging (freshly abraded metals, wood, ink fog plates in the dark; hydrogen peroxide vapour implicated) [web] https://www.nature.com/articles/060210a0.pdf, https://en.wikipedia.org/wiki/William_James_Russell (Russell's 1897 paper itself not found). Le Bon cited Russell and accepted a chemical cause (Forces p.344). Final verdict (split): - Long invisible afterglow, photographic detection, IR stimulation and quenching, IR transparency of ebonite and black paper, ZnS-screen imaging (a darkening where IR falls, pressed onto a plate): Established in kind. His specific numbers (2 h at 3 days to traces at 18 months; CaS 300 times less sensitive than ZnS; plates 5000 times less): Plausible, untested (no independent blank-corrected repeat). - "Darkness added to darkness gives light": Not testable as stated (rhetoric for IR stimulation of stored charge). - The 1896 single new radiation passing metal and cardboard: Never properly tested (and Historically superseded but instructive, by his own re-sorting). Dismissed without testing in the US press of May-June 1896 (QE revision R9); in France it was engaged with (Poincare's questions, Academy notes). - Metal-on-plate radiations: Historically superseded but instructive (chemical fogging). - Naval night-vision and thermal-imaging ideas (Forces pp.365-368): Plausible, untested in his account (no trial reported); realised later by IR image converters [memory, high]. Settles it: replicate with a calibrated 1 micrometre IR LED, unexposed blank screens, glass/ice/ebonite filters, and a no-IR-red-lamp darkroom; for 1896, plates in sealed foil with and without freshly abraded metal and a peroxide scavenger.

Claim 5. Le Bon was first and was resisted

Correction to claims.md first (contrarian is right, checked against the quoted text of Matter ch.XIV): Le Bon is more careful than the claim as stated. He writes that in 1896-97 "I was still confusing two very different things" (p.450); that "the real author of the discovery of radio-activity was Niepce de Saint-Victor" (p.452); that Curie and Rutherford later brought it up to date. What he claims: non-polarisation of uranium rays against Becquerel; cathode-ray-like effluves from illuminated bodies (1897); universality of the phenomenon; the doctrine "intra-atomic energy". Claim 5 should read: priority for the universality of dissociation and for intra-atomic energy, and for being right against Becquerel on polarisation; he credits Niepce, the Curies and Rutherford with radioactivity proper. Agreement: the reviewers agree on the facts: Hertz, Hallwachs, Elster and Geitel, Branly precede his photo-leak; Becquerel 1896 precedes; Martins documents Le Bon first denied polarisation of uranium rays (May 1897), Becquerel's 1900 text credited "Rutherford and Le Bon", his 1903 book omitted the credit [read: research note citing Martins, [web] https://www.ghtc.usp.br/livros%20RM%202022/Studia-2-3-web.pdf]. This point is documented in one historian's paper plus the 1900 and 1903 Becquerel texts as he cites them; it stays in the record as a real point for Le Bon. Disagreement: scale labels (the scale fits a historical claim poorly). Physicist: Tested, not supported for dissociation priority; contrarian: Not testable as stated in priority terms; mathematician: Historically superseded. I use the scale only where a physical content exists. Reception, both directions kept: engagement (Poincare presented four notes and questioned protocols; d'Arsonval; about 15 notes in the Comptes rendus; Berget reproduced one experiment; Kalahne, Gates, Ramsay and Spencer acted on or repeated results; De Heen (Liege) adopted the doctrine and wrote in his favour in 1901 [Le Bon's account; unread]; Rutherford cited him; translation and 15 French editions by 1908; a 1903 Nobel physics nomination reported by one blog, unconfirmed [web failed]) and resistance (H. Becquerel: Le Bon "had no idea of the phenomena of radio-activity", p.448; popular histories by Becquerel and Berget omit him, per Le Bon and Martins; Thomson's impurity view; Campbell; Naquet, Painleve, Despaux; Poincare called his 1902 priority note against Nodon "plutot nuisible"; amateur status noted by the Poincare Papers editors). Context that bears on his standing outside physics, stated because it belongs to the reception record and not as a verdict on his data: IMEC describes controversy over craniology and racial and sexual hierarchies, and Moatti documents antisemitic writings 1888-1923 (research notes). Le Bon is also an interested party in the dispute; his statement that De Heen "alone grasped its import" sits beside his own citations of Kalahne, Gates and Ramsay, and "time has proved the correctness" is true of non-polarisation, not of identity of uranium rays with illuminated-body effluves. "Records stricken" (Vassilatos): contradicted by IMEC fonds 851GLB (open on appointment, scientific documentation and correspondence), free scans at the Internet Archive and Project Gutenberg (which BSRF's own catalogue links), Nye 1974, the 2007 Poincare correspondence volume, and Marpeau's revision of the "university rejected him" picture. I found no evidence of blocked funding or publication, and none to say there was none on the Academy side: the Academy's refereeing archive is unread. Final verdict (split): - Non-polarisation of uranium rays against Becquerel: Established (documented by Martins). - Observation of leak from illuminated bodies (1897): priority unresolved (outside the scale; QE revision R9: not Plausible, untested) (CR notes of 5 April and 5 August 1897 not obtained; Lenard's memoir came later). - Universal radioactivity and dissociation of ordinary bodies: Tested, not supported (see claims 1 and 3). - Priority for the idea of large atomic energy: Not testable as stated (statements of large atomic energy before 1905 exist from Rutherford, Thomson, Le Bon; nothing links them to the E = mc^2 derivation; Einstein replied twice in 1922 to Le Bon's priority letters, per Moatti, texts unread). - "Resisted": Historically superseded but instructive (documented disagreement and a slide out of physics); "suppressed by stricken records": not supported by the records found. Settles it: CR notes of April and August 1897, the Nobel nomination (nominator and text), De Heen 1901, Nye 1974, the IMEC finding aid, the Academy's referee records.

Claim 6. The Moray connection (secondary, from Vassilatos)

Correction to claims.md and facts.json (all three reviewers and the Investigator agree, and I verified): Moray's own writings do cite Le Bon. [read] 1926 Radiant Energy, lines 2107-2138 of the extract: "A few of these scientists were Gustave Le Bon, Davy, Ampere, Faraday ...", "Dr. Le Bon noted the emission by all bodies of particles endowed with immense speed", "Dr. Le Bon proved by experiment". The Investigator also reports the Sea of Energy bibliography listing both books and Moray's Air Force proposal crediting Le Bon with "direct conversion as early as 1897" [read: research note]. The statement "No Moray statement is quoted" holds only for the 1990 review. Also: Moray's son John E. Moray's attachment of "1,340,000 barrels of gunpowder" to fission is a misreading; the figure is Le Bon's bullet-speed illustration of 5e12 J (mathematician [calc] 6.7e7 kg powder = 1.34e6 barrels of 50 kg). Evidence by sub-claim: - Moray read and used Le Bon: Established as a historical fact from Moray's text. That the books were his sourcebook for a working method: the books contain no method. - Library removal / FBI visit 1940-41: documented in one first-person recollection by John E. Moray (1978 introduction, 1980 talk): "I can remember the FBI coming to my father's door and demanding why he was using the Salt Lake City Library copy" [read: research note]. Stake: Moray's son and promoter of the Moray process. No library or FBI record found; no one has asked for one. Against a general removal: the books are on the Internet Archive and Gutenberg. A single library withdrawing a copy is possible and unexamined. - "Very deep secrets", Tesla-coil engines, bombardment pre-dating Moray's transmutation: the books give no engine and no energy-out data. What Matter does contain is Le Bon's effluves through insulating discs with an Oudin resonator (half a millimetre per contrarian's page read; Legge reports half a centimetre with a Tesla transformer in vaseline, near PDF pp.309, 445-447). A mundane reading (displacement current in the dielectric igniting a discharge on the far face, or creepage) is possible but untested (the contrarian cannot exclude it; the physicist's page was not examined for figures). The chemical "transformations" are amalgam chemistry. Moray's own transmutation programme (Research Institute Inc., 1964-65, W. J. Hooper's endorsement) was run by interested parties; no neutral replication found. - Third book "Girard Le Beau", The Evolution of Matter and Energy: hearsay relayed by Vassilatos; no trace in HathiTrust, Open Library, Google (queries in sites.md) or BSRF's catalogue. Moray's 1926 booklet carries the subtitle "The Evolution of Energy and Matter", a plausible origin of the garbling (inference). Also Vassilatos's "Belgian laboratory": the experiments were in Paris (Springer abstract; Poincare letterhead, 29 rue Vignon); the Belgian links are the Ostend lecture (1907) and the Belgian academy title. Final verdict (split): - Moray cited and drew on Le Bon: Established (Moray's own text). - Library suppression and FBI story: Never properly tested (single interested-party recollection; no record sought). - Books hold a working method for releasing intra-atomic energy: Not testable as stated (no method given); and the quantitative content (Le Bon's p.80-81 concession) refutes practical release from ordinary matter. - Bombardment/transmutation lineage: changed in QE revision R9 to Not testable as stated (the Le Bon text contains amalgam chemistry and no transmutation claim; no experiment tested a lineage); Moray's own transmutation claims belong to the Moray project. - "Girard Le Beau": Never properly tested (no copy produced). Settles it: Salt Lake City Public Library holdings and withdrawal records, an FBI FOIA request for Moray 1940-41, a physical copy of any "Girard Le Beau" book; for the Oudin experiment, a repeat with a grounded shield, varied polarity, thickness and frequency.

3. What would settle it (single deciding item per question)

1. Is ordinary matter radioactive? Blank-corrected ionisation measurement of clean copper and zinc, shielded and unshielded, with radon monitoring. Expected by mainstream: background only.
2. Do dynamos draw on atomic energy? Torque against electrical load. Already done on every generator; the answer is no.
3. Is the black-light afterglow what Le Bon says? Independent repeat with unexposed blanks, calibrated IR source, a red-free darkroom.
4. Did the French establishment suppress his work? The Academy of Sciences' refereeing records, the Nobel nomination file, the IMEC finding aid.
5. Did Moray's lineage run through Le Bon's books? Moray's own writings already show citation; library and FBI records would test the suppression story.

4. Open questions

1. What exactly did Campbell (1905) and Le Bon's reply (Athenaeum, 24 March 1906) say about heat versus chemistry as the cause of the chemical leak? Who was right?
2. Is the "Faraday-cage emanation" real at his stated size, or background ionisation plus desorption? No blank exists.
3. Did Le Bon's 1897 CR notes anticipate Lenard and Thomson on the nature of the particles from illuminated bodies?
4. Which of 6000 V and 1125 kgm per kcal is a French misprint and which a translation error?
5. Is there any record at all of the "Girard Le Beau" book or of the Salt Lake City library episode?
6. What did Kragh mean by "most physicists failed to detect the rays"; who tried and with what result?
7. Who nominated Le Bon for the 1903 Nobel prize, if the nomination exists?
8. The 15 Nov versus 22 Nov 1902 Revue scientifique citation (mathematician correction 5): two issues of one series, or a citation error?

5. Absence-of-evidence analysis

The record does not go dark for Le Bon himself: it moves from physics journals (to about 1908) to history of science (Nye 1974, Martins, Kragh, the 2007 Poincare correspondence, Marpeau, Moatti), while Le Bon stayed prominent as a popular author. The anomalies are narrower.

A. Black light after about 1908: no replications, refutations or follow-up experiments in the sources found, though the claims were presented to the Academy and a textbook (Gariel) summarised them. Competing explanations: - Absorption into standard luminescence physics: the effects are early descriptions of phosphor trap storage and IR stimulation, known by E. and H. Becquerel before him; no separate topic remained. For: the physics matches modern IR-stimulable CaS/SrS phosphors [web] https://patents.google.com/patent/US4507562; Le Bon himself cites the Becquerels. Against: it does not explain why his photographic-through-opaque-boxes work was not taken up as a technique for decades. - Guilt by association with N-rays (1903-04) and loss of interest: for: Kragh and others group black light with N-rays; Poincare discussed N-ray measurements with Le Bon in 1903. Against: Le Bon's data were plates and exposure times, not eye judgments of faint sparks (mathematician), so the grouping is a label, not a demonstrated fault. - Failed replications by others using a red darkroom lamp (Le Bon's own account): for: physically sound; against: unpublished and a post hoc excuse until repeated. - Le Bon's turn to psychology and sociology after 1895-1910: for: his bibliography; against: nothing blocks a physicist from returning. - Suppression or blocked publication: no evidence found; he published through the Academy, Flammarion and English publishers. What would decide: the Academy's referee reports, De Heen's and Gariel's later texts, the IMEC finding aid (his laboratory notebooks may be there), a modern replication.

B. The Moray library/FBI episode (1940-41): single first-person, interested-party account; silent elsewhere. Competing readings: a real FBI inquiry about a man with Tesla-style apparatus in wartime; a misremembered childhood event; promotional embellishment. Decisive records: FBI FOIA file on Thomas Moray, Salt Lake City Public Library circulation or withdrawal records. Not sought yet.

C. "Girard Le Beau": silent in every catalogue searched. Readings: a garbling of "Gustave Le Bon" plus Moray's booklet subtitle; a private or unpublished item; a mis-attributed title. Decisive: a copy.

D. Le Bon's priority notes of 1897 and the Nobel file: documents exist (CR notes, nomination) but nobody here has read them. This is a gap in our reading, not a gap in the record.

6. Corrections to earlier documents (for the orchestrator)

1. claims.md claim 2: the dynamo and solar-heat argument is at Forces PDF pp.155-159 (Part II, magnetism); the "dogma of indestructibility" argument is at pp.80-91 (Part I, Book III). Add: the 6,800 million hp figure is a premise (total conversion at 1e8 m/s), not a measurement; his own p.427 concession is 1 in 1e8 molecules.
2. claims.md claim 4: the ZnS image is a darkening where IR falls (quenching), transferred to a plate, not a glow (Forces p.347).
3. claims.md claim 5: reword as in Section 2, claim 5 above (he credits Niepce, the Curies and Rutherford; he claims universality, intra-atomic energy and non-polarisation).
4. claims.md claim 6 and facts.json (Vassilatos entry): "No Moray statement is quoted" is true only of the 1990 review; Moray's 1926 Radiant Energy names Le Bon (verified, lines 2107-2138). "Belgian laboratory" is wrong: experiments were in Paris.
5. deep read, Forces statuette caption: the caption's "eighteen months" is the error; the 1899 primary note (CR 128 p.166) says three or four days after the statue went dark, 8-15 day exposures.
6. deep read, Forces section 2.1: Le Bon names no repeaters in the book, but the 1899 note footnote names Gaston Braun fils (photographer, Paris cellar).
7. sites.md "Poincare 1903 on intra-atomic energy": the endothermic-atom objection also appears in Le Bon's own Matter ch.IV (PDF pp.99-100), alongside Painleve and Naquet.
8. sites.md "CR 128 (30 Jan 1899) downloaded, not read": the mathematician and contrarian have read it; its figures (CaS 300 times, plates 5000 times less sensitive than ZnS) match Forces pp.337, 357, 361.
9. Matter p.73: 1125 kgm per kcal is wrong (about 427; Forces p.65 prints 425). Matter p.74: "6000 volts" for the earth is off by a factor of about 1000 (4,800 C gives 6.8e6 V). Both uncarried errors; French original needed to assign them.
10. Matter pp.78-79: "100,000 kilogrammes per second" should be kilometres; Matter p.72 footnote "per ton per second" should be per ton.
11. Physicist, mathematician: proton-decay bound was tagged [memory, medium]; web check gives partial lifetime > 8.2e33 years (2009) and a reported 2.4e34 years (2020 SK, URL not confirmed); the contrarian's "1e34" was nearer.
12. The mathematician's request that Fig. 35 and the transparency example be flagged as unsupported or reversed is confirmed by the physicist, contrarian and my reading of their numbers; keep in the primer.

7. Web checks (budget 8; 5 calls used)

1. Proton decay limit: confirmed, see Section 1. https://arxiv.org/pdf/0903.0676
2. Ra-226 chain energy: no published total found; Q-value sum of about 35 MeV from a search summary; consistent with my own estimate. Retagged as [calc] with web support for the first-step energy (4.87 MeV).
3. 1903 Nobel nomination of Le Bon: blog only; archive entry not found; fetch of the blog failed (ECONNREFUSED). Stays unconfirmed.
4. Reviewers' web checks accepted as given: Russell effect (partly confirmed), IR stimulation and quenching of sulphide phosphors, Kragh, Martins.
5. Not checked (verdict does not depend on them): ammoniacal copper sulphate near-IR absorption [memory, medium]; Ramsay-Spencer, Kalahne contents (second-hand through Legge).

Summary verdicts (as revised; see Section 8, QE revision)

1. Matter perishable: weak form Established; strong form Tested, not supported; ether end-state Historically superseded but instructive.
2. Intra-atomic energy: large internal energy Established (by later physics); his magnitude Historically superseded but instructive; as yield from ordinary matter Refuted by calculation; "dogma" Not testable as stated (philosophy), loophole Refuted by calculation; dynamo Refuted by calculation; release exceeding input Never properly tested.
3. Release in ordinary bodies: leak phenomena Established; dissociation reading Tested, not supported; hot-metal emanation Never properly tested; ch.XII chemistry Established, transmutation reading Tested, not supported.
4. Black light: afterglow, IR stimulation and quenching, IR transparency, ZnS imaging Established in kind (numbers Plausible, untested); 1896 single radiation Never properly tested and superseded; naval idea Plausible, untested.
5. First and resisted: polarisation priority Established; universal-radioactivity priority Tested, not supported; resistance Historically superseded but instructive; "records stricken" not supported.
6. Moray: Moray cited Le Bon Established; suppression/FBI Never properly tested; hidden method Not testable as stated; "Girard Le Beau" Never properly tested.

8. QE revision (Step 4b remediation, 2026-10-08)

Inputs: qe.md (verdict MINOR) and the new deep reads in deep/ (Matter: pp.159-176, 201-216, 227-243 with 444-447, 350-388; Forces: pp.137-153, 178-196, 369-391, 392-402 with 415-420). I read the new deep-read sections in full and did not re-open the page images; where a deep reader checked an image I say so. Tags: [read] document text via a deep read; [calc] my arithmetic (.venv-style Python); [web] URL; [memory] with confidence. Web calls used this revision: 1 (total 6 of 8).

Overall: the central verdicts do not change. Six verdicts are added (R4-R8, R10), one factual statement is corrected (R1), two scale-term misuses are fixed (R9), and several pieces of pertinent material that were missing are put back (R11). Section 2 and the Summary verdicts above carry the edits marked "QE revision".

R1. The "no blank leak rate" statement was too strong (qe.md 3.1)

[read: Matter p.354] "An electroscope can only be used for this kind of research when it does not give a loss greater than one angular degree in an hour after being covered with its cap." That is an acceptance criterion for the instrument, not a measured blank for a run (deep read, section I). Scale [calc]: 1 degree per hour is 1/60 degree per minute. His 8-day dark cylinder (1-2 degrees per minute, p.430) is 60-120 times that; the warmed cage (5-6 degrees per minute) is 300-360 times. His weakest reported light rates (1-2 degrees in 2 minutes, p.367-368) are 30-60 times above it. Other controls the deep read found [read]: after every spark-UV experiment he covered the light with black paper and found "several times" rapid discharge from electrical influences (Hertzian waves), fixed by earthing a Leyden-jar coating, sharpening electrodes, a coherer-and-bell monitor, and wiping the quartz every 10-15 minutes (p.381-383); a black-paper band round the sunlit face reduced the discharge to zero (p.376); metals cleaned every ten minutes (200-fold effect if skipped, p.370). These are real controls, stated without numbers. Still missing [read]: a blank run in the same open geometry, with the same cylinder and the same warming; any dark-rate table beside the sun or spark rates; humidity, temperature or atmospheric-electricity records; a blank for the salt-on-glass, lamp-black and organic samples; repeat counts or spreads. Radon progeny on surfaces and desorption therefore remain open explanations. Failed replications by others, kept in the record [read: Matter p.381]: "The importance of a large receiving surface is paramount, and it is because many observers have neglected this essential point that they have been unable to repeat my former experiments." No observer, laboratory or year is named; Le Bon is the only source for both the effect and the reason it fails elsewhere. Against him: Kalahne, Gates and Ramsay and Spencer reported partial repeats (Section 2, claim 3); for him: the Hertz leak itself and earlier partial sunlight results by Elster, Geitel and Branly are acknowledged by Le Bon (p.389). Further points from the same deep read: - Solar-UV "disappearance" (p.371-374): four dated observations (23 and 30 August, 26 October, 12 November 1901) say the UV vanished from the L or M lines on days when the leak was slow; "all intervening factors" were eliminated but are not listed; no dark rate is given for those days. Near 0.3 micrometre these lines lie in the ozone cut-off region [memory, medium]. Verdict: Plausible, untested. - "Tin 40 times uranium" (p.386-388, page image): amalgamated tin in sun 40 degrees in 10 s against uranium screen 6 degrees in 60 s; ratio 40 [calc, agrees]. The printed text says metals have an activity "four times as great" yet gives "nearly 3,000,000" particles per second; 4 x 70,000 = 280,000 and 40 x 70,000 = 2.8 million [calc], so "four" is a slip for "forty". The comparison sets UV photo-emission from a clean amalgam against alpha ionisation from a uranium nitrate film, measured in different light, with no dark or unamalgamated blank (plain tin is listed among the least sensitive bodies, p.368). Verdict: the rate comparison is Established as a measurement of that electroscope; the reading that this is "radioactivity" is Not testable as stated (a relabelling of photoelectric emission, which Le Bon's own spectral cut-offs match, Section 2 claim 3). - Electrode dependence of spark UV (aluminium points give three times the leak of gold; photographic paper shows the opposite ranking): Le Bon concludes the spectrum contains "a something with which we are not acquainted" (p.370-371). The alternative he himself tested (Hertzian influence) is handled on p.381-383. Open, no verdict. Effect on verdicts: claims 1 and 3 unchanged. Corrections 1 in Section 6 stands; add: "no blank leak rate anywhere" becomes "an instrument baseline is stated; no blank in the same geometry".

R2. Chemical theory of radioactivity and the radium prediction (new sub-claim 1b; Matter pp.201-216)

Claims, each [read: Matter p.207-214] (deep read, sections A-F): - (a) Radioactivity arises from chemical reactions between a body and an infinitesimal trace of another body; "an absolutely pure body cannot be radio-active" (p.208, 213). Examples: mercury with a trace of tin becomes "extraordinarily radio-active" under light; dried quinine salt becomes active on slight hydration (no quantities, light source or detector given in this chapter). - (b) Radium metal will never be isolated, and if it is it "will have lost all the properties which render its combinations so interesting" (p.212-213). He hedges: "very problematical", and he would "much more willingly admit the existence of an unknown compound of barium" (p.214). He argues from silence: the "complete silence" of radium holders about isolation attempts is "a strong presumption in favour of my hypothesis" (p.213; an argument from silence by an interested party). - (c) Lifetime: Heydweiler weighed 0.02 mg lost per day from 5 g (137 years to lose 1 g at a constant rate; the arithmetic is right [calc]); Le Bon corrects it upward on the grounds that spreading a body over a large surface raises its measured activity: 5 g lose a fifth in 20 years, so a gram "would last one hundred years" (p.211-212); he says only repeated experiments will settle it. - (d) If radium were spontaneously active it would have vanished in geological time (p.212). Weighing: - (b) Metallic radium was isolated by Marie Curie and Andre Debierne in 1910 by electrolysis of the chloride on a mercury cathode and distillation of the amalgam in hydrogen [web] https://webelements.co.uk/radium/history.html (one library guide gives 1911; Curie's Nobel in chemistry was 1911). That the metal is radioactive is the definition of the element [memory, high]. Both halves of the prediction failed. Verdict: Tested, not supported. His wording was hedged, which the record should show. - (a) Rutherford and Soddy, whom Le Bon quotes (p.210) as allies for the word "chemical", proposed a transformation of the atom itself with a fixed decay probability; radioactivity is independent of chemical state [memory, high]. Le Bon's quotation is selective (deep read, section G). His mercury-and-tin observation fits ordinary surface chemistry changing photoemission (as in Chapter XII, Section 2 claim 3), not radioactivity. Verdict on the theory: Tested, not supported. The observations themselves: Plausible, untested (no numbers in the chapter). - (c) [calc] Ra-226 half-life 1600 years [memory, high]: fraction lost in 20 years = 1 - 2^(-20/1600) = 0.86%, against Le Bon's 20%, about 23 times too fast; Heydweiler's 0.02 mg per day from 5 g is 0.146% per year [calc], about 3.4 times the accepted 0.043% per year. Both fail by accepted decay data. Verdict: Tested, not supported (by later decay-rate measurements; the specific papers are not read here). - (d) The argument ignores supply: radium is re-made continuously from uranium-238 (half-life about 4.5 billion years) in ores [memory, high]. Verdict: Tested, not supported. This sub-claim ties claims 1 and 3 together: it is why he ran the hydration experiments on uranium, thorium and radium (Matter p.419 per qe.md). The explanation of the "dissociation" evidence as chemistry of traces is in the record next to the evidence he cited. Credit kept: he was right that radium's chloride is a barium-like salt (p.214) and that the quantities of matter transformed are tiny.

R3. Le Bon's own over-unity statements (qe.md 3.3; Matter pp.227-243)

[read: Matter p.231] "...the creation, in the future, of such a machine - that is to say, of an apparatus giving forth a power extremely superior to that expended in setting it in motion - can be considered possible." Premises, both marked "probable": that friction dissociates atoms, and that this is why static electricity is high tension. The setting is a hand-worked machine at 50,000 V and 2 A. [calc] 50,000 x 2 = 100,000 W = 136.0 metric hp (134.1 British); "per second" is a unit slip, whose author (Le Bon or translator) cannot be told from the English. The 2 A is hypothetical: a hand-turned influence machine gives of the order of 10-100 microamperes [memory, high], so the supposed current is 1e4 to 1e5 times larger; and its output is bounded by the work put in (a charged capacitor stores only (1/2) C V^2). Also [read: Forces p.38 via qe.md, not re-read]: the energy matter emits under an excitant "may far exceed that which it has received". The proposed route is material, not mechanical: "placing matter in a state in which it can be easily dissociated", with sunlight as "a model agent" (p.232). Verdict: Never properly tested (no machine, trial, current measurement or energy table is reported). Not a separate experiment, a corollary of the intra-atomic-energy thesis. Corrects the earlier statements that the trigger-release idea is "not shown in the books" (Section 2, claims 2 and 6): the idea is Le Bon's own, written down, without a method. Related, same chapter, on the Section 1 table: the earth "6000 volts" and the 1125 kgm per kcal slips are unchanged.

R4. Electrification by influence, the aluminium arrow and friction experiments (Forces pp.137-153; was a verdict gap)

[read, deep read section 2; images of Figs 2-5 checked by the deep reader] - Apparatus: an aluminium or gold-leaf strip 0.5 cm by 2 cm, thickness not given, floats "wholly and indefinitely motionless" between a positively charged body A above and a rod B below, throws out sparks visible in the dark; with A negative the leaf sticks to it (p.140-141). No charge, voltage, current, duration, distances, mass or weighing; the only variation is the sign. Witnesses are "those physicists to whom I have shown it", unnamed. - Inference [read: p.142-143]: the strip is "in every way comparable to a fragment of radium"; "if the above experiment were continued for several centuries, we should see the aluminium become gradually less ... and finally disappear. We should then have witnessed the complete transformation of matter into energy." - Calculation [calc]: assume a 0.01 cm leaf, so m = 0.027 g (my assumption; Le Bon gives no thickness); mc^2 = 2.4e12 J. To disappear in 300 years the average power would be 2.4e12 / (300 x 3.156e7 s) = about 260 W, continuously, from a strip throwing "little sparks". Small sparks and corona carry milliwatts; at 10 mW the strip would last about 8 million years. So his own timescale is contradicted by his own observation by a factor of the order of 1e4. - Mainstream reading [memory; the deep reader flags it untested]: corona or point discharge at the strip's tip, with ion wind and recoil supplying a dynamic force; no static arrangement of fixed charges gives stable equilibrium (Earnshaw), which is why Le Bon's "iron coffin" analogy fails to apply; energy comes from the charged body A and its source. Le Bon names no test of this (tip shape, current, vacuum). The classic influence experiment: induced charge is rearranged from the cylinder's own charges (Griffiths-level), and the work of moving the cylinder supplies the energy he dismisses as "feeble" (p.140) with no measurement. - Electroscope sign reversal (Figs 3-5): after about 12 minutes under a negative rod a ball electroscope ends negative; a pointed rod does it in seconds; a plate does not (p.145-147). Dependence on shape matches charge loss by point discharge [memory, high]; no rod potential, distance or humidity given. - Insulating rod 1 m long, rubbed over 2 cm, slowly charging an electroscope at the far end (p.149): no control for surface leakage, which is the textbook cause of this. - Battery versus static tension (p.150-151): the figures are right (50,000 / 2 V = 25,000 cells [calc]); the explanation (high tension because the energy is intra-atomic) ignores V = Q/C: a small machine at 50,000 V holds very little energy; battery voltage is fixed by the cell reaction (about 1-2 V per cell). - Rubbed metal strips (Fig. 6): 1000-1500 V on a 10 cm square held in an ebonite handle and rubbed with catskin, success only in dry air; heated metal recovers after minutes (p.152-153). The ebonite handle and the catskin are the usual tribo-electric sources and the charge on the metal was not distinguished from transfer from them. Verdicts (split): - Floating strip is stable and sparks: Plausible, untested (no repeat on record; a corona or ion-wind reading fits and has not been excluded or tested by Le Bon). - Strip is converting matter into energy / "complete transformation of matter into energy": Refuted by calculation (his own several-centuries figure needs about 260 W against milliwatt sparks) and Never properly tested by him (no weighing, no accounting). - Unlimited charge generated by influence: Refuted by calculation (conservation of charge; work of displacement). - Electroscope reversal and rubbed-metal charging as observations: Established in kind (point discharge, triboelectricity); the dissociation reading: Never properly tested (no measurement separates it from charge transfer). - Battery-versus-static tension as proof of intra-atomic origin: Refuted by calculation (V = Q/C; energy supplied by the driving work). This closes the gap for claim 2's "electricity is a product of dissociation": it is now judged on its own pages, not only through the dynamo.

R5. Branly and Le Bon, 1899, Hertzian-wave shielding (Forces pp.178-196)

[read, deep read; thicknesses checked on page images by the deep reader] Joint note, Comptes rendus, 1 April 1899, pp.879 ff. (not in the PDF; unread). Findings: cubic boxes about 50 cm, walls 0.02-2 mm: the bell rang until the door was closed with screws, then stayed silent "many hundreds of times" (p.187); wall thickness made no difference (tin-lined wood 0.01 mm as opaque as 2 mm metal, page image); a 1 mm by 20 cm slit passed waves at every distance available while 100 round 1 cm holes did not beyond 50 cm; a razor-cut slit worked at one-sixth of the 1 mm slit's distance; a slit perpendicular to the radiator axis worked at 6-8 times the distance of a parallel one; 1 mm gauze acted as a sheet except within a few centimetres (p.188-189). Dielectrics: dry sand 30 cm and dry stone 40 cm passed waves at 40 m (limit of space); wet, 25 m for stone; cement 10 cm walls silent beyond 7 m, damp 30 cm cement opaque; with a 2 cm-spark coil sand 16 m, stone 13 m (p.192-195). He states the caveats himself: "nothing absolute about" the figures; dry sand and stone seemed transparent "only in appearance" because the range ran out. Interpretation. [memory, high; the deep reader tags it the same way] These are leakage through seams and slots, polarisation dependence of a slot, mesh much smaller than the wavelength acting as a sheet, and moisture raising loss in cement and stone. Those are standard shielding results. Absent: an open-air baseline at the same distances, source power, wavelength, independent repeat. The detector is a binary coherer. Verdict: Established in kind (the main finding, that apparent transparency of closed metal boxes came from poorly fitting doors, corrected earlier reports by Bose and Veillon on the same ground, p.181-182); the specific ranges: Plausible, untested. The weapon proposal (p.178-180): a parallel beam of Hertzian waves from banks of cylindro-parabolic mirrors to detonate ships and magazines at "several kilometres", which would "render war impossible". He did not carry it out "solely on account of the expense". His only evidence is an anecdote at room scale: an Oudin resonator on a 60 cm-spark coil charged all metal objects in a 10 m room so that they threw "thousands of sparks" and melted switchboard wires at several metres (single witness, himself; no date). His own text names the obstacle: diffraction at these wavelengths means "the greater part of the energy is lost". Verdict: Never properly tested (not attempted); Plausible, untested as engineering is not a fair label, because the diffraction limit he cites and the inverse-square fall off [memory, high] weigh heavily against ranges of kilometres without enormous apertures. His point that shielding a ship is "illusory" because of seams is the same Branly-Le Bon result, used the other way.

R6. Oudin-resonator effluves through glass and ebonite; Legge's repeat; X-rays from an electrodeless vessel (Matter pp.227-243, 444-447)

This is the primary-text anchor for Vassilatos's Tesla-coil and "particles through mica and ebonite" statements, and the earlier arbiter text located it wrongly (PDF p.309 contains no such passage; see Section 6 corrections below). [read: deep read, page images of Figs 25, 26, 62 and of Legge's footnotes checked by the deep reader] - Le Bon (p.235, 444-446): raising tension "by several hundred thousand volts with a high frequency resonator" (Oudin), fed by an induction coil of at least 30 cm spark through two Leyden jars and a spark gap of about 1 cm, makes the effluves pass "without deviation" through a glass or ebonite strip that must not exceed 0.5 mm. Fleming's 1900 Cantor Lectures called it "striking" (footnote p.235). Voltage is asserted, frequency never stated. The plate is a "strip" of unstated area; Fig. 62 shows nothing beyond it (no counter-electrode); Fig. 25's rays are hand-drawn; Fig. 26 is a long exposure showing a diffuse far-side glow broader than the near-side tuft. - Legge's repeat (p.446, footnotes signed F. L.): a Tesla transformer of Elster-Geitel pattern by Ernecke, Berlin, in solid vaseline, two 40 cm Leyden jars in parallel, a 1 cm gap, a 35 cm coil; effluves passed discs of glass, ebonite, sulphur and shellac 5 mm thick and paraffin wax 10 mm. "the substance to be traversed must be an insulator of electricity." Both footnotes and the digest line that attributes them to Le Bon or gives "4 cm" are wrong: the page images read 1/2 mm and 1/2 cm. Legge's polarity note (funnel electroscope "above the apparatus", p.447) shows a net positive charge near the terminal, not charge crossing the plate; Le Bon gives no method for his "positive ions". - X-rays (p.237, 446-447): effluves led into a glass vessel with no electrode at high vacuum show the bones of a hand on a platinocyanide screen; vessel size, pressure, distance and witnesses are not given. Context [memory, medium-high]: electrodeless and single-terminal bulbs excited by high-frequency fields predate this (Tesla, J. J. Thomson, 1891-97). Competing readings, each against the stated details: - (P) Le Bon: positive particles of dissociated matter physically cross the plate. - (D) Displacement current (dielectric-barrier discharge): the plate is a capacitor in series, the alternating field acts through it, and a new brush forms on the far face. It predicts what the text reports: high frequency essential (a coil spark merely pierces glass), insulators only, the plate not damaged, allowed thickness rising with tension (0.5 mm, then 5 mm with Legge's higher tension, 10 mm paraffin), a diffuse far-side glow beginning at the plate face. [calc, plate area assumed 10 cm^2, ebonite relative permittivity 2.8 [memory, medium]] C = about 50 pF at 0.5 mm and 5 pF at 5 mm; reactance 32 kilohms at 100 kHz and 3.2 kilohms at 1 MHz at 0.5 mm. A few centimetres of air to the far brush is below 1 pF, so a thin plate is nearly transparent at these frequencies and thicker plates need proportionally more voltage. The 1/thickness scaling is what the thickness-versus-tension observation shows qualitatively. - (C) Creepage round the plate's edges: "without deviation" and Fig. 26 argue against it, but plate size is never given and no edge-masking test is reported, so the text does not exclude it. No control separates (P) from (D): a metal plate (earthed or floating) in the same place, a collector beyond the plate with and without it, a direct-current source, a measurement of voltage or frequency, plate dimensions and edge masking, tilt of the plate: none is reported. His field control (an earthed gauze cage extinguishes Geissler tubes but not platinocyanide letters, p.241) was run for the UV experiment, and shows he knew the Oudin field acts through glass at metres. Verdicts: - The observation (a glow beyond a thin insulating plate in a high-frequency field, with thicker plates passing at higher tension): Plausible, untested as to Le Bon's reading, with one independent repeat on record by a sympathetic translator with different apparatus (documented in one published source, his own footnotes). I do not call it "Established" because neither a neutral repeat nor a frequency or voltage measurement exists. - Particles of dissociated matter pass through the plate: Never properly tested; the displacement-current reading fits every reported detail and is not excluded. - Electrodeless-vessel X-rays: Plausible, untested (consistent with known high-frequency excitation; no apparatus detail). - Link to Vassilatos: Legge's transformer is a Tesla transformer, so the "Tesla-coil" element has a text anchor (as a repeat, not as an engine); the books still give no device that yields more than it takes. The decisive test named in Section 2 (grounded shield, polarity, thickness, frequency) is not reported done; Legge's thickness-versus-tension data and his insulator condition are partial evidence and are kept in the record.

R7. Heated radium, the emanation, induced activity: Vassilatos checked against the book (Matter pp.159-176)

[read, deep read section 5] The book's account: heated or boiled radium "loses the greater part of its activity ... but it regains it entirely in twenty days or so"; after red heat or boiling "it still preserves a quarter of its primary activity", "solely due to the a particles" (p.168); the emanation "reduced by one-half in four days" and gone in three weeks (p.168); three quarters of radium's steady heat comes from the emanation, so a heated sample gives a quarter of its heat (p.169). The facts are credited to Rutherford, Curie, Ramsay, McClelland and J. J. Thomson, or to no one; Le Bon's own contributions here are zinc-sulphide scintillation screens (p.163) and a diamond phosphorescence test for radium samples (p.165). He doubts Ramsay's helium transmutation reading (p.169-170) and objects to Rutherford's reading of the alpha and beta data. He adds that most of his claims here are "a very new subject ... many contradictions and uncertainties" (p.160). [calc] Radon half-life 3.82 days [memory, high]: ingrowth after the emanation is driven off is 1 - exp(-0.693 t / 3.82) = 71.9% after 7 days and 97.3% after 20 days, matching "twenty days or so". Radium, radon and the two alpha-emitting daughters (Ra-226, Rn-222, Po-218, Po-214) are four alpha emitters in equilibrium, so a quarter of the alpha activity remains with only radium left [memory, high]. The book therefore reports correct mainstream radiochemistry. Against the review's three phrases: - "loses activity for about a week": not in the pages; the book says twenty days. The nearest week-like figure is the emanation's four-day half-life. [read] - "Le Bon found": the heating and recovery figures carry no name and are not presented as his experiment; the attribution to him is Vassilatos's. [read] (Memory, medium: the heating-and-recovery result is usually credited to Curie and Danne, around 1903; not checked on the web.) - "altering the half-life of radium": absent. What is removed and regrown is the emanation, a gas; the decay constant of radium is not touched. The recovery is explained by ingrowth [calc]. Verdicts: the book's account of heating and recovery: Established. Vassilatos's three statements as descriptions of the book: not supported by the text (a reading error by the reviewer). The physical claim that heating alters radium's half-life: not shown by Le Bon or by this chapter; the data he reports are explained by ingrowth. I do not assign it a scale term because no experiment on half-life is in the record here. Vassilatos's UV-makes-radium-like-materials: the book's version is the photo-leak (amalgamated tin "forty times" uranium, R1), tied to claim 3. Also kept from this chapter: induced activity (a negative plate in thorium emanation becoming 10,000 times more active than thorium, p.173), bismuth in radium bromide active for three years (p.174), and Le Bon's inconsistent "momentarily" against "fairly long time" (p.172). Condensation temperatures "120" and "150" lack minus signs in the text (p.168, image not checked).

R8. Hidden forces, Legge's vacuum-jacket test, Dufour's breath (Forces pp.392-402, 415-420)

[read, deep read] Delbruck (an engineer) suggested that soaring birds may generate a force that cancels gravitation from intra-atomic energy; Le Bon: "very difficult to verify, but is not fundamentally absurd" (p.401); he accepts neither Delbruck nor levitation ("their reality must first be rendered beyond dispute", p.401-402). The needle that a living body attracts or repels: Le Bon himself suggests heat. Legge's bracketed note (p.401) reports a double-walled evacuated bell glass over the apparatus: "the needle remains motionless, no matter what bodies are brought near", concluding "convection currents formed by heat" (the identity of "Ed." as Legge is an inference). The jacket cannot separate electrical from thermal causes; no numbers. Seances with Dastre: "nothing demonstrative" (p.402), medium unnamed, N-rays as a warning. Dufour's breath: "air that has been breathed contains radio-active particles" (p.419), given as support for dissociation in living bodies; no citation or figures; the sentence is garbled in the text layer. A mundane reading, exhaled radon or thoron from ambient air [memory, medium], is not tested in the text. Kola and caffeine as atomic excitants with Dastre's reported agreement: hypothesis only; Le Bon limits intra-atomic energy in the body to rare, "exceptional" efforts (p.420). Verdicts: Delbruck anti-gravity: Never properly tested (hypothesis). Needle moved by living bodies: Tested, not supported as a new force (the one reported control, Legge's vacuum jacket, removed the effect; a single footnote report). Seances: Tested, not supported (Le Bon and Dastre, outcome negative, conditions not described; the one account is Le Bon's own). Dufour's breath: Never properly tested (second-hand, no data). Kola and caffeine: Not testable as stated. This bears on Vassilatos's "living energies" framing and on the BSRF "electrogravitic experiment by Dr Le Bon about 1907" lead: no such experiment appears in the pages read; the nearest text is Delbruck's idea and Legge's null test. The lead stays in the record as unconfirmed.

R9. Scale terms and the 1896 black-light form (qe.md 6, 7.6, 7.7)

• Transmutation reading of Chapter XII, and the "bombardment/transmutation lineage" of claim 6: "Tested, not supported" was a misuse. No experiment tested a transmutation claim, and the text contains none. Replaced above with Not testable as stated, plus the plain statement that the text contains amalgam chemistry only.
• 1897 priority of leak from illuminated bodies: "Plausible, untested" does not fit a historical question; shown as outside the scale.
• 1896 form of black light, dismissed without testing? The record on file is mixed [read]: the 1896 press (Cosmos, 23 May 1896, via Literary Digest, reprinted at BSRF; the reprinter is an interested party): "M. Le Bon's experiments have been described before the French Academy of Sciences, a sufficient proof that they have been performed in good faith; but his conclusions have been usually looked upon with suspicion", and elsewhere in the same notice "no one apparently has thought it worth while to take the trouble" to test them (per qe.md 5.3; I did not re-read the notice itself). So in the US press of May-June 1896 it was dismissed without testing; in France Poincare questioned the protocol, Berget reproduced a different (1897) experiment, and Le Bon's own 1908 re-sorting dropped the single-agent form. Kragh's "most physicists failed to detect the rays" names no test. The verdict stays Never properly tested, with "dismissed without testing" in the 1896 press and engagement but no neutral test in France. Note: the 1896 date is the announcement; Le Bon dates the start of the work to about 1894, "studying for two years" before 1896 (Forces p.317, per qe.md).

R10. Infra-red plants, the antagonism experiments, the copper-sulphate conflict (Forces pp.369-391)

[read, deep read; Figs 38-39 and the p.385-386 table checked on page images by the deep reader] - Plant work (p.375-378): black-paper bell-glasses and wrapped fruit; results are qualitative (seeds germinate faster then die in a fortnight; artichoke tops blanch but develop better; tomatoes turn white). Le Bon says the garden was lent for one season, calls them "hints", and concedes aeration and heat were not controlled (p.376-378). No dark-only opaque control separates an infra-red effect from darkness, heat and closed air. Juvisy red-glass greenhouses are cited as others' work (p.374, no reference). Verdict: Never properly tested (by his own admission); no verdict beyond that. - Antagonism (p.379-391): a zinc sulphide screen behind an ammoniacal copper sulphate trough (2 cm) is brighter in the dark than a bare one (table: 1 bare in sun, 2 in shade, 7 under cobalt glass in sun, 14 behind the copper trough in sun); behind a quinine trough it does not glow at all; a clouded gelatin plate is cleared in one hour of sun through 1/2 mm ebonite, the unclouded half serving as control (p.389-390). The quenching of sulphide phosphors by red and infra-red light and the clearing of a fogged plate by long-wave light (the Herschel effect [memory, medium]) are known effects. Verdict: Established in kind (consistent with the Section 2 claim 4 verdict); the table's numbers: Plausible, untested. - Internal inconsistency [read]: the p.364 table calls ammoniacal copper sulphate "transparent" to invisible radiation not exceeding 3 micrometres at 1 cm, while the antagonism experiments (p.384, 387) rely on the same solution at 2 cm to stop "the great majority" of the red and infra-red extinguishers. A reading that reconciles them: the trough works by absorbing visible red and green, which Le Bon also assigns to the extinguishing group (p.381, 387), but on p.384 he says "red and infra-red". Also the "transparency" of p.364 was measured with the same phosphor screen whose response the later chapter shows depends on the mixture of rays. No acknowledgment of the conflict appears in pp.369-391. This confirms the earlier doubt (Section 2, claim 4) and moves it from "doubtful" to an internal inconsistency in the book's own tables. - Brightness unit: "0.002 B (1/2000 candle power)" but 14 x 0.002 = 0.028, matching the printed 3/100 [calc]; 0.002 is 1/500, so "1/2000" is a misprint (book or translator). - Speculative extensions kept: a retina as sensitive as ZnS would make street lighting "perfectly unnecessary" (p.388). No test.

R11. Material put back (retention; qe.md section 5)

Kept in the record, with source, whatever the verdict above: 1. Le Bon's letter to Poincare of 29 November 1902 (Poincare Papers, 2-35-18) [read: research note]: with quinine sulphate in a closed box "la decharge sera enormement reduite. Les choses se passent comme s'il fallait que les ions puissent circuler pour agir." This is Le Bon's own evidence that the quinine effect is ion transport through air, not a penetrating radiation. It bears on claim 3 and on the dispute with Rutherford's classification, in Le Bon's favour as to method (he tested and reported a limit) and against any "penetrating radiation" reading. 2. Reception in physics (Forces preface, per digest Forces 01 p.14-15 [read, digest]): P. D. Innes (Cavendish, Proc. Roy. Soc. 1907, p.442): "the only theory which can satisfactorily account for the phenomena observed is that of atomic disintegration" and "a great store of energy in the atom seems now beyond question"; quoted by the editor as approval of Le Bon, which overstates it, since atomic disintegration is not his dissociation of all matter. Kelvin's 1904 statement that the heat of radium "cannot proceed from the stored energy of radium", withdrawn at the British Association meeting in Cambridge in 1904 (Matter p.86, translator's note signed F. L. [read, digest Matter 03]). Two Cambridge critics who admitted the theory was "in the main correct" while challenging the experiments and originality (Matter p.27 translator's preface, Athenaeum, February-March 1906 [read, digest Matter 01]). These belong beside the critics already listed in claim 5. They are Legge's reports of reception (an interested party), not independent statements. 3. Investigator's related-work leads [web, per sites.md "Leads for review"; the leads are characterisations by their sources, interested parties in the alternative-science tradition; I have not opened them]: Resines (BSRF booklet) on an "electrogravitic experiment by Dr Le Bon" about 1907; Vassilatos, "Induced Radioactivity by Solar Light" (Journal of Borderland Research, 1997; magnesium plate under focused sunlight) and "Black Waves"; William R. Lyne, Pentagon Aliens (techniques said to be "rediscovered" from Le Bon); R. A. Ford, Homemade Lightning (Le Bon's calcium sulphide experiment); Oliver Lodge's views called "congruent" with Le Bon's ether cosmology (Kragh). They bear on claims 3, 4 and 6 as lineage; none is a replication. The 1997 solar-radioactivity article is the nearest to a modern repeat claim of the photo-leak and has not been weighed. 4. Failed repeats named by Le Bon himself (R1), Legge's vacuum-jacket null test (R8), and Legge's repeat of the passage experiment (R6) are now in the record. 5. Peripheral experiments, kept without verdicts (qe.md 3.9): Juvisy red-glass greenhouses; diamond phosphorescence to tell Brazilian from Cape stones, about 200 examined (Forces pp.268-272); electric sparks restoring thermoluminescence after calcination (Forces pp.290-293); Nova Persei read as dissociation with nebular masses moving at about the speed of light (Forces p.118; the 1901 light echo [memory, high]); solar-UV disappearance photographs of 1901 (R1). Digest only, not read by me.

R12. Corrections to earlier documents (additions to Section 6)

13. Matter, PDF year: the title page p.10 is dated 1909, so this copy is a 1909 printing; the first English edition was 1907 (Forces p.13). facts.json and the Matter deep read say "not on the title page".
14. Location of the Oudin passage experiment: PDF pp.235-237 (Book V, Chapter II, printed pp.206-208, Figs 25-26) and pp.444-447 (Chapter XIII, Fig. 62), not "near PDF pp.309, 445-447" (Section 2, claim 6; physicist.md). Plate limits: Le Bon 0.5 mm; Legge 5 mm (glass, ebonite, sulphur, shellac) and 10 mm (paraffin); the text-layer "4 mm"/"4 cm" is OCR for 1/2 mm and 1/2 cm; the digest line for p.446 is wrong and the footnotes there are Legge's.
15. Vassilatos radium statements: see R7; the claims.md claim 3 and the facts.json Vassilatos entry should add "twenty days", "reported from Rutherford and Curie, not Le Bon's measurement", and the ingrowth calculation.
16. claims.md: add sub-claim 1b (chemical theory of radioactivity and the radium prediction, R2) and add to claim 2 that Le Bon himself calls an over-unity machine "possible" (R3).
17. Section 2 claim 4 and the ammoniacal copper sulphate remark: now an internal inconsistency (R10); and the unit misprint "1/2000" for 1/500.
18. Matter p.388: "four times as great" is a slip for "forty" (R1); the digest p.411 attribution of the independent-check request to Le Bon is Legge's (qe.md 7.4).
19. Section 7 web-check count: 6 of 8 used.

R13. Revised verdict table (new and changed rows only)

Updated Summary verdicts (replace the six lines above where they differ): 1. Matter perishable: unchanged, plus 1b chemical theory of radioactivity and radium prediction Tested, not supported. 2. Intra-atomic energy: unchanged, plus over-unity machine Never properly tested (Le Bon proposed it as "possible"); electrification by influence as matter-to-energy Refuted by calculation; battery-versus-static argument Refuted by calculation. 3. Release in ordinary bodies: unchanged, except transmutation reading Not testable as stated; "tin 40 times uranium" a photo-emission comparison. 4. Black light: unchanged; adds Hertzian shielding Established in kind, copper-sulphate inconsistency, weapon proposal Never properly tested, 1896 form dismissed without testing in the US press. 5. First and resisted: unchanged; Innes, Kelvin 1904 and the Cambridge critics added to the reception record. 6. Moray: unchanged except the Tesla-coil and lineage wording above; Legge's Tesla-transformer repeat is the textual anchor; Vassilatos's heated-radium statements not supported by the book.