Gustave Le Bon: intra-atomic energy and "black light", a primer

Sources. This primer covers Gustave Le Bon (1841-1931) through the two books in the project folder: The Evolution of Matter (English translation by F. Legge, 1907, from L'Évolution de la matière, Paris 1905; the PDF copy is a 1909 printing, title page p.10 [read]) and The Evolution of Forces (English, 1908). It also covers one secondary source: a 1990 review by Gerry Vassilatos in the Journal of Borderland Research. That review links the books to the inventor Thomas Henry Moray. It comes from an interested party, Borderland Sciences Research Foundation (BSRF), which sells the books it recommends. No source gap: all three documents have text. The Matter PDF is OCR (machine-read) throughout, so table numbers were checked against page images (by the deep reader, and by the mathematician for PDF pp.73, 74, 76). The verdicts are the arbiter's (arbiter.md). This primer adds no new verdicts.

Not read by any reviewer. Several items are known only through Le Bon's own text, his translator Legge's footnotes, or the Investigator's notes: the French originals; the papers by Kalähne (1905), Gates (1904), Ramsay and Spencer (1906) and Campbell (1905); De Heen's 1901 and 1912 texts; Mary Jo Nye's 1974 history; the 1903 Nobel nomination file; the finding aid for Le Bon's archive at IMEC (Institut Mémoires de l'édition contemporaine); Le Bon's Comptes rendus (CR, the weekly proceedings of the French Academy of Sciences) notes of 5 April and 5 August 1897; the joint Branly-Le Bon CR note of 1 April 1899 on Hertzian-wave shielding; and the alternative-science writings that the Investigator found citing Le Bon (Resines, Vassilatos 1997, Lyne, Ford; see section 6, "Kept in the record"). Where a statement rests on one of these, this primer says so.

QE revision (2026-10-08). The quality check (qe.md) found one statement in this primer that was too strong ("no blank run anywhere"), several of Le Bon's own experiments and predictions with no verdict, and pertinent material that had been dropped. New deep reads were done and the arbiter added verdicts (arbiter.md, section 8, R1-R13). This version carries them: see the new rows in section 4 and section 6, the second worked example in section 5, and "Corrections to earlier documents" at the end.

Tags. [read] = from the books, with PDF page ("p.N" means PDF page throughout); [read: deep] = from the project's deep reads; [read: research note] = from the Investigator's notes on a primary source; [web] = with URL; [memory] = from training, with confidence; [calc] = computed in example.py (run command in its header).

---

1. The idea in one paragraph

Gustave Le Bon was a French physician who became famous for The Crowd (1895), a founding book of social psychology. From about 1894-96 to about 1908 he also ran physics experiments in his own Paris laboratory. These were the years when X-rays, radioactivity and the electron were found. From what he saw, Le Bon built a sweeping doctrine. All matter, not just radium, is slowly falling apart ("dissociating"). Atoms are huge stores of "intra-atomic energy". Light, heat and chemical action can speed the release of that energy in ordinary metals. Electricity and solar heat are this energy leaking out. He wrote that a machine "giving forth a power extremely superior to that expended in setting it in motion" could "be considered possible" (Matter p.231), though he never built or tested one. And matter eventually dissolves back into the ether, the medium that physicists then believed filled space. He backed this with hundreds of electroscope measurements of charge leaking from metals under light, flames and chemicals, and with a long study of "black light". Black light was invisible radiation from phosphorescent screens, which he photographed for months after they stopped glowing, and he showed that infrared light passes through "opaque" ebonite and black paper. Much of what he measured was real. The leaks were the photoelectric effect and the ionisation of air. The "black light" was long-lasting phosphorescence and near-infrared light. And he was right, decades before nuclear physics, that a gram of matter holds an enormous amount of energy. What fails is the step from these observations to "all matter is dissociating" and "dynamos run on atomic energy". This primer shows where that step happens, using his own numbers.

---

2. What you need to know first

Kinetic energy and old units. A mass m moving at speed v carries kinetic energy KE = ½mv² (Newton) in joules (J), with m in kg and v in m/s. Le Bon worked in French engineering units. The kilogram-metre (kgm) is the work done lifting 1 kg through 1 m: 1 kgm = 9.807 J. The metric horsepower (French "cheval-vapeur") is 75 kgm per second = 735.5 W, slightly less than the British 745.7 W. Horsepower measures power (energy per second), not energy. Read: Halliday, Resnick and Walker, Fundamentals of Physics, the chapters on kinetic energy and work, and on power.

Mass-energy, E = mc². Special relativity (Einstein, 1905) says that a body of mass m has rest energy E = mc², where c = 3.0 × 10⁸ m/s is the speed of light. For 1 g that is 9.0 × 10¹³ J. At high speed, kinetic energy is (γ − 1)mc², where γ = 1/√(1 − v²/c²) is the Lorentz factor. Real nuclear processes release only a small fraction of mc²: about 10⁻⁴ for radium decay and about 10⁻³ for fission of uranium-235 [memory, high]. Le Bon's books were written just before or alongside Einstein's paper, and they do not use E = mc². Read: Krane, Modern Physics, the special-relativity chapter; Tipler and Llewellyn, Modern Physics, same.

Radioactive decay. An unstable nucleus decays at random with a fixed probability per second, the decay constant λ. The number of nuclei left falls as N(t) = N₀e^(−λt). The half-life is T½ = ln 2/λ. The activity (decays per second) is A = λN. Each decay releases a definite energy Q (the "Q-value") set by the mass difference: Q = (M_parent − M_products)c². Radium-226 has T½ ≈ 1600 years and reaches stable lead-206 in nine decays, its own and eight daughters' (five alpha, four beta) [memory, high]. If a short-lived daughter is removed (for example radon gas, driven off by heating), the long-lived parent keeps making it, and it grows back as 1 − e^(−λ_d t), where λ_d is the daughter's decay constant, until it is back in balance with the parent (ingrowth). The decay rate of a nucleus does not depend on its chemical state: radium decays at the same rate as a metal, a chloride or a bromide, hot or cold (the only known exceptions are shifts below about 1% in a few nuclei that decay by capturing an inner electron, not radium) [memory, high]. Read: Krane, Introductory Nuclear Physics (Wiley, 1988), the chapters on radioactive decay and alpha decay.

The photoelectric effect. Light of frequency f comes in photons of energy E = hf = hc/λ, where h is Planck's constant and λ the wavelength. A handy form is E [eV] = 1240/λ [nm], where eV is the electron-volt (1.6 × 10⁻¹⁹ J) and nm the nanometre. A photon can knock an electron out of a metal only if its energy exceeds the metal's work function φ, the energy that binds the outermost electrons (about 4-6 eV for clean metals). So each metal has a cut-off wavelength λ_c = 1240 eV·nm / φ. Light longer than this does nothing, however bright. A negatively charged plate loses electrons this way and discharges. A positive plate pulls the electrons back. Read: Krane, Modern Physics, the chapter on the particle nature of light.

The gold-leaf electroscope and ionisation of air. An electroscope is a metal rod with a thin gold leaf that stands out from the rod when charged. When the charge leaks away, the leaf falls. Le Bon measured "degrees per minute" of leaf fall. Air is normally an insulator, but anything that makes ions in it (pairs of positive and negative charged particles) lets the electroscope discharge whatever the sign of its charge. Sources of ions include ultraviolet below about 185 nm, flames, some chemical reactions, radioactivity in the walls and air (radon), and cosmic rays. A sealed electroscope always leaks a little because of this background [memory, high]. Read: any history of cosmic rays. Hess's 1912 balloon flights are the classic story.

Electromagnetic induction and Lenz's law. A conductor of length l moving at speed v across a magnetic field B (in tesla, T) has an electromotive force (EMF, a voltage) ℰ = Blv along it. If a current I flows, the field pushes back on the wire with force F = BIl, opposing the motion (Lenz's law). Read: Griffiths, Introduction to Electrodynamics, Chapter 7 (Electrodynamics), the sections on motional EMF and Faraday's law.

Capacitance and point discharge. A conductor holding charge Q at voltage V has capacitance C = Q/V (in farads, F) and stores energy ½CV². A static (friction or influence) machine reaches tens of thousands of volts but charges only a few picofarads (pF, 10⁻¹² F), so it holds very little energy; a battery's 1-2 V per cell is set by its chemical reaction and it can deliver far more charge. Near a sharp point or edge the electric field is concentrated. Above about 3 million V/m air breaks down there and charge sprays off as a faint glow or tiny sparks (corona or point discharge), pushing air along as an "ion wind". The energy comes from whatever keeps the conductor charged. Read: Griffiths, Introduction to Electrodynamics, Chapter 2 (electrostatics, capacitors, energy).

Conservation of energy and potential energy. The first law of thermodynamics says energy is neither created nor destroyed, only converted. For a force F(x) you can define a potential energy U(x) = −∫F·dx only when the work around every closed loop is zero, which is the condition curl F = 0. Then U is fixed, up to a constant, by measuring the force. Read: Schroeder, An Introduction to Thermal Physics, Chapter 1; Taylor, Classical Mechanics, the chapter on energy.

Phosphorescence and traps. Some crystals, such as calcium sulphide (CaS) and zinc sulphide (ZnS) with traces of metal impurity, store energy when light falls on them. Electrons are lifted and caught in "traps", defects in the crystal. They escape slowly and emit light as they do: that is the afterglow. A trap of depth E_t at temperature T empties at a rate proportional to e^(−E_t/kT), where k is Boltzmann's constant. A crystal with many trap depths gives an afterglow that falls off as a power of time, I ∝ t^(−n), rather than as a single exponential. Infrared light or heat can empty traps early. In some materials the escaping electrons emit light (stimulation, as in CaS). In others they recombine without light (quenching, as in ZnS:Cu). Read: Yen, Shionoya and Yamamoto (eds.), Phosphor Handbook, the chapters on persistent and photostimulated luminescence [memory, high for the book; chapter numbers not certain].

---

3. The textbook view

(a) Kinetic and rest energy.

KE_Newton = ½ m v²; KE_rel = (γ − 1) m c²; E_rest = m c²; γ = 1/√(1 − β²), β = v/c

• m = mass (kg), v = speed (m/s), c = 2.998 × 10⁸ m/s.

(b) Energy actually released by decay. For N atoms each releasing Q, the total energy is N·Q, delivered at power P = A·Q = λN·Q. For 1 g of radium-226 [calc]:

• N = 6.022 × 10²³ / 226.0 = 2.66 × 10²¹ atoms
• λ = ln 2 / (1600 years) = 1.37 × 10⁻¹¹ s⁻¹, so A = 3.66 × 10¹⁰ decays per second. This is the old unit, the curie.
• Q for the whole chain from Ra-226 to Pb-206 (five alpha decays and four beta decays) comes from atomic masses [memory, high]: (226.02541 − 205.97447 − 5 × 4.00260) u × 931.5 MeV/u = 35.3 MeV per atom. Here u is the atomic mass unit and MeV = 10⁶ eV.
• Total if every atom reaches lead: 1.5 × 10¹⁰ J, or 1.7 × 10⁻⁴ of mc². Neutrinos carry a few MeV away, so the heat is a little less, about 1.4 × 10¹⁰ J.

(c) Photoelectric cut-off. λ_c = hc/φ = 1240 eV·nm / φ.

• φ = work function. Clean surfaces [memory, high]: Mg 3.7 eV, Al 4.1, Cd 4.2, Zn 4.3, Cu 4.65, Ni 5.1, Pt 5.65.
• Sunlight at ground level stops near 295-300 nm, because ozone absorbs below that [memory, high].

(d) Generator energy balance. For a rod of length l moving at v in field B, driving current I through a load resistance R:

ℰ = B l v, I = ℰ / R, P_electrical = ℰ I, F_drag = B I l, P_mechanical = F_drag · v = B I l v = ℰ I

The shaft power always equals the electrical power, minus nothing and plus losses (heating, friction). The magnetic force qv × B on a moving charge q is perpendicular to the charge's velocity, so it does no work. The magnet is not the energy source and is not "used up".

(e) Wien's displacement law. The peak wavelength of thermal radiation from a body at absolute temperature T is λ_max = 2898 µm·K / T (µm = micrometre). Boiling water (373 K) peaks at 7.8 µm and a human body (310 K) at 9.3 µm. Glass blocks beyond a few µm. Ebonite (hard rubber) and black paper pass the near infrared from about 0.8 µm upward unless they are loaded with carbon black [memory, high].

(f) Long afterglow. With a spread of trap depths, intensity falls roughly as I(t) ∝ t^(−n) with n ≈ 1-2 over long times. A photographic plate integrates light over its exposure, so the exposure needed for a fixed darkening goes roughly as 1/I [memory, high]. Plates lose sensitivity at very low light levels ("reciprocity failure"), so very long exposures overstate how weak the light is.

---

4. Where Le Bon departs

The table puts each departure next to the textbook. In every row, the observation in the left column is real.

The fair reading of the starting points. Le Bon's premises were good ones for 1900. The new radiations really did show that atoms are not eternal and indivisible. Radium really does release heat without stopping, about 100 calories per hour per gram (Curie and Laborde, which he quotes [read: p.73]). Thomson's and Abraham's electromagnetic-mass estimates, which he also quotes [read: p.75-77], really did suggest energies of order 10¹² to 10¹³ J per gram. "Matter is a prodigious reservoir of energy" [read: p.80] is correct. The physicist and contrarian both credit him with arguing the general point early (arbiter, claim 2). He was careful, too, about what he had not shown. Right after the big numbers he writes [read: p.80-81]:

"the substances which dissociate quickest, like radium, only disengage very minute quantities of energy. All those millions of kilogrammetres which a simple gramme of matter contains amount in reality to very little if, to obtain them, we have to wait millions of years."

The worked example below shows that this sentence is quantitatively right. It also shows that it undercuts the stronger claims made elsewhere in the books.

On "dogma" and Poincaré. Henri Poincaré's Science and Hypothesis (ch. 8) does say that, applied to the whole universe, the energy principle shrinks to "there is something which remains constant". But he also says that for a limited system "it has a perfectly clear meaning", and he rejects the idea that it is a mere tautology [web] https://en.wikisource.org/wiki/Science_and_Hypothesis/Chapter_8. Le Bon's use of him goes further than the source. The philosophical point is not testable as stated. The practical claim, that potential energy can be given whatever value is needed, is false. A hidden energy store must still match the forces you measure (mathematician §2; arbiter claim 2).

---

5. Worked example: the energy of a one-centime coin

Le Bon's own setup [read: Matter p.71]: "Let us take, for instance, a copper one-centime piece, weighing ... one gramme, and let us suppose that by accelerating the rapidity of its dissociation we could succeed in totally dissociating it." He uses "the lesser of these figures, that of 100,000 kilométres per second" for the particle speed. All numbers are reproduced in example.py, Part A.

Step 1. His arithmetic, in his units. m = 10⁻³ kg, v = 10⁸ m/s. KE = ½ × 10⁻³ × (10⁸)² = 5.0 × 10¹² J. In kilogram-metres: 5.0 × 10¹² / 9.807 = 5.10 × 10¹¹ kgm. Le Bon: "510 thousand millions" (the OCR prints "kitemétres", a garble of kilogrammetres) [read: p.71-72]. ✔ "Stopped in a second", this is a power of 5.10 × 10¹¹ kgm/s ÷ 75 = 6.80 × 10⁹ metric hp. Le Bon: "about six thousand eight hundred million horse-power". ✔ With British horsepower you get 6.7 × 10⁹. That is why one reviewer first saw a slip, but there is none (arbiter §1). His train check: 5.10 × 10¹¹ kgm against a 3000 kgf haulage force gives 1.70 × 10⁸ m = 170,000 km = 4.25 times the Earth's circumference. Le Bon: "a little over four times and a quarter". ✔

So the arithmetic is right throughout. The question is what it measures.

Step 2. Relativity and the ceiling. β = v/c = 0.334, so γ = 1/√(1 − 0.111) = 1.061. KE_rel = (γ − 1)mc² = 0.0608 × 10⁻³ × (3.0 × 10⁸)² = 5.46 × 10¹² J, 9% above Newton's value. The Newtonian formula is a fair approximation at c/3. Rest energy of 1 g: mc² = 8.99 × 10¹³ J. Le Bon's figure is 5.0 × 10¹² / 8.99 × 10¹³ = 5.6% of the ceiling, about one eighteenth. Lesson: the order of magnitude is right. That is why admirers say he "anticipated" E = mc². But he reached it by an assumption (whole mass at c/3), not from any relation between mass and energy. Abraham's 6 × 10¹³ J, which he quotes, is the same kind of estimate [read: p.76].

Step 3. What the best natural source actually gives. Radium is the substance that "dissociates quickest", in Le Bon's words. - Measured heat, as Le Bon quotes Curie: 100 small calories per hour per gram = 100 × 4.184 J / 3600 s = 0.116 W per gram. - Over the "1000 years" he assumes for radium's life [read: p.73]: 0.116 W × 3.16 × 10¹⁰ s = 3.7 × 10⁹ J. - With modern data, if every atom in 1 g went all the way to lead: 2.66 × 10²¹ atoms × 35.3 MeV × 1.602 × 10⁻¹³ J/MeV = 1.5 × 10¹⁰ J, or 1.7 × 10⁻⁴ of mc². - Check on the 0.116 W: fresh radium alone gives A × 4.87 MeV = 0.029 W. Radium in balance with its three short-lived alpha-emitting daughters gives 0.143 W [calc]. Curie and Laborde's sample lay between the two.

So a gram of radium, over its whole life, yields about 1/330 of Le Bon's coin figure. And radium is itself about 2.7 million times more active per gram than uranium-238 [calc: ratio of half-lives 4.47 × 10⁹ y / 1600 y, times 226/238].

Step 4. How long would the coin take? Suppose, generously, that the copper coin released energy as fast as radium: 5.0 × 10¹² J / 0.116 W = 4.3 × 10¹³ s = 1.4 million years. That is Le Bon's own "wait millions of years", in numbers. Copper is not radium, though. It shows no measurable decay at all, and its nuclei are stable. Searches for decay of the proton, the building block of every nucleus, set partial lifetimes above 8 × 10³³ years for the main channel [web] https://arxiv.org/pdf/0903.0676. So the real figure for ordinary matter is not millions of years but effectively never.

Step 5. What this shows. - What holds: matter contains an energy of order 10¹³ J per gram. Le Bon's sums are correct, and his concession at pp.80-81 is correct. - What fails: "6,800 million horse-power" as what a gram of ordinary matter can yield. That figure needs total conversion, which no process observed then or now gives for ordinary matter. His own data contradict it: the radium figure, and the "1 in 100,000,000" he quotes at p.427. Arbiter verdict: magnitude Historically superseded but instructive; as a yield from ordinary matter, Refuted by calculation.

Companion check: the dynamo (example.py, Part B). Take a rod l = 0.5 m moving at v = 10 m/s through B = 1 T, closed through R = 5 Ω (ohms). ℰ = Blv = 5 V; I = 1 A; P_electrical = 5 W; drag force F = BIl = 0.5 N; mechanical power needed = Fv = 5 W. Halve the load resistance to 2.5 Ω: I = 2 A, P_electrical = 10 W, F = 1.0 N, mechanical power 10 W. The pusher pays exactly for every watt drawn. If the electricity came from atoms in the magnet or the wire, the drag would not depend on R. Le Bon himself grants that "a determinate quantity of movement and consequently of work is, then, required" [read: Forces p.158], but he gives no measurement of it. Every generator's fuel consumption follows its electrical load. Arbiter verdict: Refuted by calculation.

Two small slips you can catch in one line each (example.py, Part C). - Matter p.73 converts heat "at the rate of 1125 kilogrammetres per great calorie". The right figure is 4186 J / 9.807 = 427 kgm per kcal, and his other book prints 425 [read: Forces p.65]. - Matter p.74: a twentieth of 96,000 coulombs (4,800 C) would charge "a globe the size of the earth to a potential of 6000 volts". The Earth's capacitance is 4πε₀R = 7.1 × 10⁻⁴ F (ε₀ = permittivity of free space, R = Earth's radius), so 4,800 C gives 6.8 million volts. The error is a factor of 1000 and understates his own point. Neither slip is carried into any conclusion. Whether they are misprints in the French or translation errors is unknown until someone checks the French original.

Second worked example: heated radium, the book against its reviewer (example.py, Part F). Vassilatos's 1990 review says: "Le Bon found that heating radium to red hot was sufficient to depress its radioactivity for a week", and calls him "a scientist who discovered the means of altering the half-life of radium and measurably proved it" [read: Vassilatos review, p.1-2]. Here is what the book says [read: Matter p.168]: heated radium "loses the greater part of its activity ... but it regains it entirely in twenty days or so"; after red heat "it still preserves a quarter of its primary activity", "solely due to the α particles"; and the emanation (radon gas) is "reduced by one-half in four days". The figures are credited to Rutherford and Curie or to no one, not to Le Bon's own experiments. - Step 1. What heating does. It drives the radon gas out. Radium stays and keeps making radon. - Step 2. Ingrowth. Radon's half-life is 3.82 days, so λ_d = 0.693/3.82 = 0.181 per day. The fraction regrown after t days is 1 − e^(−0.181 t): 52% at 4 days, 72% at 7 days, 97% at 20 days. "Twenty days or so" is right. - Step 3. The quarter. In balance, radium and its three short-lived alpha-emitting daughters (radon-222, polonium-218, polonium-214) each give the same number of alphas per second. Remove radon and its daughters and only radium's share is left: 1/4. "A quarter" is right. - Step 4. Le Bon's own lifetime estimate. He says 5 g of radium lose a fifth in 20 years, so a gram "would last one hundred years" [read: Matter p.211-212]. With T½ = 1600 years the fraction lost in 20 years is 1 − 2^(−20/1600) = 0.86%, about 23 times less than his 20%.

Lesson: the book reports correct radiochemistry from others, and the recovery is a gas growing back, not a change in radium's half-life. Vassilatos's "a week", "Le Bon found" and "altering the half-life" are not in the pages; they are the reviewer's misreading. Where Le Bon made his own estimate (the 100-year life), it was far off. Arbiter: the book's account Established; Vassilatos's three statements not supported by the text; the lifetime Tested, not supported (arbiter R2, R7).

---

6. Evidence table

"Replicated?" means repeated by someone other than Le Bon. Verdicts are the arbiter's.

Kept in the record although no verdict rests on it. - Le Bon's "effluves" through insulators now has its own row (6e). The test that would decide between particles and displacement current is still not done: a metal plate (earthed or floating) in the same place, a collector beyond the plate with and without it, a direct-current source, measured voltage and frequency, and masked plate edges. - Le Bon's letter to Poincaré, 29 November 1902 (Poincaré Papers) [read: research note]: with quinine sulphate in a closed box, "la décharge sera énormément réduite. Les choses se passent comme s'il fallait que les ions puissent circuler pour agir" (the discharge is enormously reduced; things happen as if the ions had to be able to move about to act). This is Le Bon's own evidence that the quinine effect is ions carried through air, not a penetrating radiation. It bears on claim 3 and on Rutherford's classification of the effect. - Legge's vacuum-jacket test (Forces p.401, row 6g): a documented control in this project's orbit that removed a claimed "living force". The jacket cannot by itself separate a thermal from an electrical cause, and no numbers are given. - Reception in physics, both directions (all reported by Legge, the translator and an interested party): P. D. Innes (Cavendish Laboratory, Proc. Roy. Soc. 1907, p.442): "a great store of energy in the atom seems now beyond question" (Forces preface, p.14-15), which Legge presents as approval of Le Bon although atomic disintegration is not his dissociation of all matter; Kelvin's 1904 statement that radium's heat "cannot proceed from the stored energy of radium", withdrawn later that year at the British Association (Matter p.86, Legge's note); two Cambridge critics who "admitted the theory was in the main correct" while attacking the experiments and their originality (Athenaeum, February-March 1906; Matter p.27). These stand beside the critics in row 5d. - Related work found by the Investigator (sites.md "Leads for review"; each is its source's characterisation, from writers in the alternative-science tradition; none opened by the reviewers; none is a replication): J. Resines (BSRF booklet B0086) on "an electrogravitic experiment by Dr Le Bon carried on by about 1907" (no such experiment appears in the pages read; row 6g); G. Vassilatos, "Induced Radioactivity by Solar Light" (Journal of Borderland Research 53(3), 1997), a magnesium plate under focused sunlight with an anecdote about W. G. Lehr, the nearest thing to a modern repeat claim of the photo-leak, not yet weighed; Vassilatos, "Black Waves" (JBR 48(4)), with Dan Winter's galena-and-carnotite experiment and Paul Brown's claim that Moray's device was a nuclear battery; W. R. Lyne, Pentagon Aliens (techniques said to be "rediscovered" from Le Bon); R. A. Ford, Homemade Lightning (Le Bon's calcium sulphide experiment); and Oliver Lodge's ether views, which Kragh calls congruent with Le Bon's. - Peripheral experiments, no verdicts (digest only): diamond phosphorescence used to tell Brazilian from Cape stones, about 200 examined (Forces pp.268-272); electric sparks restoring thermoluminescence after heating (Forces pp.290-293; radiation-induced thermoluminescence [memory, medium]); Nova Persei read as atomic dissociation, with "nebulous masses moving at about the speed of light" (Forces p.118; in fact the 1901 light echo, light from the outburst reflecting off surrounding dust [memory, high], an instructive misreading); Juvisy red-glass greenhouses (others' work, no reference); induced activity, a negative plate in thorium emanation becoming 10,000 times more active than thorium (Matter p.173). - Reception context. IMEC records controversy over Le Bon's craniology and his racial and sexual hierarchies. Moatti documents antisemitic writings from 1888 to 1923 [read: research notes]. These bear on his standing outside physics, not on his data. - Le Bon as an interested party. He wrote that De Heen "alone grasped its import", yet his own chapter V cites confirmations by Kalähne, Gates and Ramsay. "Time has proved the correctness" holds for non-polarisation. It does not hold for the identity of uranium rays with emissions from illuminated bodies. - Einstein. Le Bon wrote to Einstein in 1922 pressing his priority for intra-atomic energy. Einstein replied twice, per Moatti, a critic of "alterscience". The replies are unread here. - "1,340,000 barrels of gunpowder". John E. Moray attached this figure to fission. It is actually Le Bon's illustration of the energy needed to drive a bullet at 10⁸ m/s (mathematician [calc]).

---

7. What a learner should take away

Solid. - A gram of matter does hold an energy of order 10¹³ J. Le Bon said so, with correct arithmetic, before E = mc² was available. What he lacked was the relation between mass and energy, and any means of release. - His photo-leak tables are good 1905 work. The order of metals follows the work functions (example.py, Part E: Mg and Al respond in sunlight; Zn and Cd weakly; Cu, Ni and Pt only to spark ultraviolet). He separated negative photoemission from positive leaks caused by air ionisation, using vacuum UV absorbed within about 2 cm of air. - His infrared and phosphor work is real physics, with well-designed controls: the glass strip, the ice troughs, and opposite effects on two halves of one ZnS screen. It covers IR stimulation versus quenching, near-IR transparency of "opaque" materials, and an idea for infrared night signalling decades early. His explanation of why others failed, a red darkroom lamp quenching the ZnS, is physically sound and testable. - He was right against Henri Becquerel on the polarisation of uranium rays. - With Édouard Branly in 1899 he showed that closed metal boxes stop radio waves and that earlier reports of "transparent" boxes came from badly fitting doors; slits leak, and their direction matters. These are correct shielding results (row 4f).

Wrong, and why it is instructive. - A correct calculation can rest on an unmeasured premise. All his energy sums check out. The flaw is "suppose the whole gram is dissociated". Always ask what the number is a calculation of. - A phenomenon is not its interpretation. Electrons leaving a surface, ions in air and a leaking electroscope do not show atoms being destroyed. He states an instrument baseline (a capped electroscope must lose no more than 1° per hour, Matter p.354), and his effects are 60-360 times that. But a baseline taken with the cap on is not a blank run in the same open geometry with the same warming, so background sources that the geometry itself brings in (radon products on surfaces, desorbed gas) are not excluded. Always ask whether the control matches the experiment. - A theory that makes a sharp prediction can be checked. Le Bon's chemical theory of radioactivity predicted that pure metallic radium would be inactive. Curie and Debierne isolated it in 1910 and it was radioactive. He had hedged ("very problematical"), but the prediction was clear, and it failed (row 1c). - Ask what each hypothesis predicts that its rival does not. "Dynamo output comes from atoms" predicts no extra shaft drag when you draw current. Lenz's law predicts drag proportional to current. The measurement exists on every generator. - Carry units through every sentence. 1125 versus 427 kgm per kcal, 6000 V versus 6.8 MV, a transparency ratio stated backwards (a 20 s exposure against a 5 s standard means one quarter as transparent, not "twice"), a Wien-law sentence printed wrong (Forces p.362), "four times" where his own numbers give forty (Matter p.388), and "1/2000 candle power" for 0.002, which is 1/500 (Forces p.385). Each is catchable in one line. - Read the source behind the secondary account. Vassilatos's "a week" and "altering the half-life" are not in the book, which says "twenty days or so" and reports a gas growing back (section 5, second example).

Genuinely open. - Did his hot-metal "emanation" (400 °C mass, cage at 35 °C, fatigue and recovery after rest) have a size beyond background ionisation and desorption? Nobody has repeated it with a blank, a radon monitor and shielding. - Who was right in the 1905-06 dispute with Campbell over heat versus chemistry as the cause of the chemical leak? - Do his CR notes of April and August 1897 anticipate Lenard and Thomson on the particles emitted by illuminated metals? - What did "most physicists failed to detect the rays" (Kragh) rest on? Who tried, and with what result? - Who nominated him for the 1903 Nobel prize, and why? - What would the IMEC archive's laboratory notebooks show about his blanks and controls? - Is there any record of the Salt Lake City library episode, an FBI file on Moray for 1940-41, or a "Girard Le Beau" book? - Does the effluve passage survive a test that separates particles from displacement current (row 6e)? - What do the 1899 Branly-Le Bon note and Vassilatos's 1997 solar-radioactivity article actually report, and does the 1997 experiment include a dark blank? - Is there any record of the "electrogravitic experiment by Dr Le Bon" about 1907 that BSRF's Resines describes?

---

8. Check yourself

1. Le Bon gives "6,800 million horse-power" for his gram. (a) Which horsepower makes his figure exact? (b) Why does he need the words "if stopped in a second"?
2. Using his own radium figure (100 small calories per hour per gram) and his own assumed 1000-year life, how much energy does a gram of radium give? How many times smaller is that than his coin figure?
3. A dynamo delivers 2 kW to a load. Its operator lowers the load resistance so that the current doubles at the same speed. What happens to the torque needed at the shaft, and what would Le Bon's hypothesis predict instead?
4. In sunlight, Le Bon found a strong negative leak from clean aluminium and almost none from copper. Using λ_c = 1240 eV·nm / φ with φ(Al) ≈ 4.1 eV and φ(Cu) ≈ 4.65 eV, and the fact that ground-level sunlight stops near 295 nm, explain the result. Why did a positively charged plate barely leak?
5. Le Bon's afterglow series needed 2 h of exposure at 3 days and 12 h at 15 days. Assuming intensity ∝ 1/exposure and I ∝ t^(−n), find n. Why is a power law, rather than an exponential, a sign of ordinary phosphor physics?

---

9. Further reading

Mainstream textbooks - Halliday, Resnick and Walker, Fundamentals of Physics: work, energy and power; induction. - Griffiths, Introduction to Electrodynamics, Chapter 2 (electrostatics, capacitors, energy) and Chapter 7 (motional EMF, Faraday's law, Lenz's law). - Krane, Modern Physics: special relativity, the photoelectric effect. - Krane, Introductory Nuclear Physics (1988): decay laws, Q-values, alpha decay. - Schroeder, An Introduction to Thermal Physics, Chapter 1 (energy, the first law). - Yen, Shionoya and Yamamoto (eds.), Phosphor Handbook: persistent and photostimulated luminescence. - Poincaré, Science and Hypothesis, ch. 8, "Energy and Thermodynamics" [web] https://en.wikisource.org/wiki/Science_and_Hypothesis/Chapter_8.

History - Mary Jo Nye, "Gustave LeBon's Black Light: A Study in Physics and Philosophy in France at the Turn of the Century", Historical Studies in the Physical Sciences 4 (1974) 163-195 [web: listing] https://online.ucpress.edu/hsns/article/doi/10.2307/27757330/47934. This is the standard history; paywalled and not read here. - Helge Kragh, "Eight Physicists and Astronomers: Biographical Portraits" (2012), Le Bon chapter [web] https://arxiv.org/pdf/1208.3114. - Roberto de Andrade Martins, "Becquerel's experimental mistakes", Studia 2-3 [web] https://www.ghtc.usp.br/livros%20RM%202022/Studia-2-3-web.pdf. - Henri Poincaré Papers, Le Bon correspondence 1896-1911 [web] https://henripoincarepapers.univ-nantes.fr/chp/text/lebon01.html (letters lebon01-16).

Le Bon's own works - The Evolution of Matter (tr. F. Legge, 1907); the French original L'Évolution de la matière (Flammarion, 1905). Free scans are on the Internet Archive. - The Evolution of Forces (1908); French L'Évolution des forces (1907). - La Naissance et l'Évanouissement de la Matière (1908, from his Ostend lecture), Project Gutenberg #67253 [web] https://www.gutenberg.org/ebooks/67253. - Comptes rendus notes: "La lumière noire", CR 122 (1896) 188-190; CR 128, 16 and 30 January 1899 (invisible residual luminescence; transparency of opaque bodies). Copies are in research/files/. - É. Branly and G. Le Bon, note on the absorption of Hertzian waves by metal enclosures and dielectrics, CR, 1 April 1899, p.879 ff. (cited in Forces; not read here). - French texts online at gustavelebon.org [web] https://gustavelebon.org/livres/levolution-de-la-matiere. - Archive: IMEC, fonds 851GLB [web] https://www.imec-archives.com/archives/fonds/851GLB.

The Moray link (secondary) - G. Vassilatos, review of the two books, Journal of Borderland Research (1990). Interested party: BSRF sells the books. - G. Vassilatos, "Induced Radioactivity by Solar Light", Journal of Borderland Research 53(3) (1997), and "Black Waves", JBR 48(4). Interested party (BSRF); not read here. - On the isolation of metallic radium (Curie and Debierne, 1910): [web] https://webelements.co.uk/radium/history.html. - T. H. Moray, Radiant Energy (1926) and The Sea of Energy in Which the Earth Floats. See the owner's Morray-TH project.

---

Corrections to earlier documents

1. mathematician.md, §1 hidden assumption D: "Counting all alphas in its decay chain to lead (about 28 MeV per atom, [memory, medium])". The alpha Q-values sum to 29.8 MeV. The whole chain from Ra-226 to Pb-206, betas included, releases 35.3 MeV per atom, computed from atomic masses and checked by summing the individual steps [calc: example.py, Part A3]. That gives 1.5 × 10¹⁰ J per gram before neutrino losses. The arbiter's "32-35 MeV, about 1.3-1.5e10 J per gram" already brackets this. No verdict changes.
2. This primer, earlier version, §4 row "Leak in sealed vessels" and §7 said "No blank run (an empty instrument) is reported anywhere in the books". That was too strong. Matter p.354 states an instrument baseline: an electroscope is usable only when, capped, it loses no more than "one angular degree in an hour" [read: p.354]. What is missing is a blank run in the same open geometry, with the same cylinder and warming (qe.md §3.1; arbiter R1). Verdicts unchanged.
3. This primer, earlier version, §1 said the experiments ran "from 1896". Le Bon dates the start to about 1894, "studying for two years" before his 1896 note [read: Forces p.317]. Now "from about 1894-96".
4. This primer, earlier version, §6 rows 3d and 6c gave "Tested, not supported" for the transmutation reading of Chapter XII and the bombardment lineage to Moray. No experiment tested either reading; the text simply contains none. Now Not testable as stated (arbiter R9). Row 5b's "Plausible, untested" for a priority question is now marked as outside the scale.
5. This primer, earlier version, row 2e treated the "trigger release" idea only as Vassilatos's. Le Bon proposed it himself (Matter p.231; Forces p.38), without any device or test (arbiter R3).
6. This primer, earlier version, row 4b omitted whether the 1896 form was dismissed without testing, which the verdict scale requires. The 1896 US press shows it was (row 4b); in France it was questioned but not neutrally tested (arbiter R9).

---

Answers

1. (a) The metric horsepower (75 kgm/s = 735.5 W): 5.10 × 10¹¹ kgm ÷ 75 = 6.80 × 10⁹. With the British 746 W you get 6.7 × 10⁹. (b) Horsepower is a rate (power), and 5 × 10¹² J is an energy. Turning one into the other needs a time, and he chose one second. There is no error, but the "horsepower" figure is really a statement about energy.
2. 100 cal/h = 0.116 W. Over 1000 years (3.16 × 10¹⁰ s) that is 3.7 × 10⁹ J, about 1,360 times less than 5.0 × 10¹² J. With modern data (complete decay to lead) a gram gives 1.5 × 10¹⁰ J, still about 330 times less.
3. Doubling the current at the same speed doubles the electrical power to 4 kW. The drag force BIl doubles, so the shaft torque and the mechanical power double too (plus losses). The engine burns more fuel. On Le Bon's hypothesis the energy comes from atoms, so the shaft would need only the same torque as before. Measurement shows the torque doubles.
4. λ_c(Al) = 1240/4.1 ≈ 302 nm, which lies inside sunlight, so photons can free electrons. λ_c(Cu) = 1240/4.65 ≈ 267 nm, which is shorter than any sunlight reaching the ground, so no photon has enough energy, however bright the sun. A positive plate attracts the emitted electrons straight back, so there is no net loss of charge. The positive leak he did see near a spark came from air ionised by vacuum ultraviolet.
5. The intensity ratio is 2/12 = 1/6 while time grows by 15/3 = 5, so n = ln 6 / ln 5 ≈ 1.11. Fitting the first four points gives n ≈ 1.5 (example.py, Part D). A single trap depth gives an exponential decay with one time constant. Here the apparent time constant grows from about 7 days to about 45 days, which is what a spread of trap depths in an impure crystal produces. It is ordinary phosphor behaviour and needs no new radiation.