Hopper-WJ (William J. Hooper): Arbiter (Step 3)

Spelling: the folder says "Hopper"; the documents say William J. Hooper (1895-1971). Tags: [read] from documents or the reviewers' reads of them, [web] with URL, [memory] with confidence, [calc] my own computation (numpy, .venv). Where I rely on a reviewer's [read], I say "per reviewer".

Source gap note (repeated from claims.md and the reviews)

The project folder holds only US 3,610,971 and a 10-page typescript essay. Neither contains a measurement. The numbers come from documents the Investigator downloaded: Hooper's "Similarities of the Motional Electric & Gravitational Forces" (read by all three reviewers), Gibson's 1983 monograph (read by the physicist, mathematician and contrarian), NASA TM-106963 (all three), and the New Horizons manuscript (chapter 3 read only by the mathematician; the physicist and contrarian did not read it). Not read by anyone: Experimental Notes (1969-70, not online), the 1958 APS abstract, the Fisher 1957 paper, USAF AFRPL-TR-72-31, Montana State report by Lamb (1985), US 3,656,013 beyond the Investigator's summary and the mathematician's partial look. Consequence: the verdicts on the "immune to shielding" half of Claim 2 and on the unipolar-induction sub-question rest on the mathematician's reading of New Horizons ch. 3 plus mainstream theory, not on the original experimental notebooks. The Claim 4 and 5 verdicts (gravity) rest on calculation plus one NASA test, which is firmer.

1. Disagreements between reviewers that I recomputed

Other checks: the reviewers' web citations for MICROSCOPE (https://arxiv.org/abs/2209.15487), ALPHA-g, Eöt-Wash (https://ui.adsabs.harvard.edu/abs/2008PhRvL.100d1101S/abstract) and NASA (https://ntrs.nasa.gov/citations/19950022472) were tagged [web] by them; I did not re-open them. Web calls used by me: 4 of 8. Not rechecked because no verdict depends on them: Hall-effect value of n in copper [memory, high], Cullwick (1957), ELF through-earth communication.

2. Claim-by-claim arbitration

Verdict scale: Established / Plausible, untested / Never properly tested / Tested, not supported / Refuted by calculation / Internally inconsistent / Not testable as stated / Historically superseded but instructive.

Claim 1. Flux around a current moves with the electrons' drift velocity V, so a current gives a motional electric field E = B x V

• Agree: a steady current in a neutral wire has no external E apart from surface-charge fields (all three); "does flux move/rotate" has no operational meaning in a symmetric geometry (physicist, mathematician, contrarian); only Hooper tested the observable; Hooper's own formula is 103 times too small for his own reading.
• Disagree: physicist says Refuted by calculation (strong form); mathematician and contrarian say Never properly tested.
• Stronger argument: the physicist is right that accepted laws (Faraday, Gauss, charge neutrality) give zero external E, but "refuted by calculation" in the scale means no experiment needed. That is too strong here: Hooper's claim is that accepted laws miss a term, and he reports a reading. A reading that exists but whose cause is unknown cannot be dismissed by calculation alone. The mathematician and contrarian are right on the scale. The mathematician adds something the physicist lacked: Klicker's thesis shows the moving-flux idea reproduces the second-order relativistic correction (relative size (v/c)^2; with v = 1.8e-4 m/s about 4e-25), which is what mainstream physics already contains. That makes the weak idea harmless and the strong, measurable version (relative size ~1) conflict with it by 25 orders. [read per mathematician: Klicker pp.10, 52, 54; calc]
• Arbiter verdict (split):
• Weak form, motional emf E = v x B of a conductor moving through B: Established.
• "Flux moves/rotates" as a description: Not testable as stated (symmetric geometry).
• Strong form, steady flux of a current drifts with the electrons and produces a measurable external E at rest charges: Never properly tested (no neutral replication; NASA's attempt inconclusive; not dismissed without testing, since Hooper ran controls). Mainstream theory predicts zero, and Hooper's own model misses his own datum by 103x, so the weight of evidence is against; this is weight, not a test.
• What would settle it: a bench test with a copper and a nichrome (same n, ~70x the resistivity) bifilar winding and a dummy winding wound inductively, same geometry, same current. B x V (with V = J/(n e)) predicts the same P.D. for any metal of the same n at the same current and the same sign for either current direction; Joule-heat and charging artefacts change by a large factor. Alternatively Klicker's unfinished electron-beam capacitor test, if a beam and electrometer of sufficient sensitivity are available.

Claim 2. A non-inductive 4020-wire coil gives an isolated, shield-immune field that charges an external capacitor

• Agree: the sign algebra is correct; the model predicts P.D. proportional to I^2 for both DC and AC, but Hooper reports a straight line for AC and Gibson's dry-ice data follow I^3.8; the factor 100 gap; artefacts (thermal, lead charging, supply imbalance, contact potential) of the same size as the signal were recorded by Hooper and Gibson themselves; no neutral replication; NASA found the capacitor readings "too erratic" [read per reviewers: nasa-ntrs.md].
• Disagree:
• Shielding. The physicist treats "immunity" as one claim and refutes the mechanism (a static curl-free E is shielded). The mathematician showed it is two claims: (i) closed-circuit motional emf in Hooper's New Horizons ch. 3 experiments (Ironing Board, Trapeze, Aluminum Box; 1954) cannot be removed by electrostatic shielding because the circulation of an electrostatic field around a loop is zero; Hooper's observed 10.64 uV against theory ~10.8 uV is ordinary flux cutting; (ii) the generator's field charging a capacitor outside a shield, where the capacitor and electrometer sit inside the grounded cabinet. Only the mathematician read ch. 3. I adopt his split because it is checkable arithmetic and accepted physics, and the physicist's and contrarian's reasoning do not contradict it. The mathematician's recomputation found small slips in Hooper's Table 3.1 (trial 8 v = 0.690, not 0.590; mean theory 10.79, not 10.76; gap ~1.4%, not 1.11%); these do not change anything [read per mathematician; calc not repeated by me].
• Thermal versus electrostatic. The physicist puts weight on heating (375 W, 0.028 K/s bulk; I recomputed 0.028). The contrarian argues that a bulk lag is too slow for readings that track current within seconds and favours fast electrostatic pickup (37 V across the winding at 10 A), backed by Gibson's series-resistance data. Both are plausible; neither excludes the other, and Hooper reports both that stepping gave "substantially identical" values and that readings are enhanced after warm-up. The contrarian's unique point is the strongest internal evidence in the record: adding external resistance at fixed current raised P.D. (a carbon-pile rheostat 3.7 to 12.3 ohm and a 1500 W photoflood bank; at 13 A with 12.3 ohm higher than 15 A with none) [read per contrarian: Gibson]. Under B x V, P.D. depends on B and V set by the current in the winding; an external resistor cannot change them. The physicist and mathematician mention Gibson's resistance observation but do not weigh it. I weigh it as the main reason the observation looks like an artefact class (lead and housing potentials), while noting that Gibson and Hooper read it as a change in conduction-electron number, and that the pre-balance tables were "not redone".
• Montana State. Klicker reports a student duplication of the shielding demos (Lamb, EE 470, 1985, work of J. C. Stover and K. A. Klicker), report not seen. Under the mathematician's split this duplicates the established closed-circuit effect, not the generator.
• Rate. Physicist: "Never properly tested" plus "mechanism Refuted by calculation". Mathematician and contrarian: Never properly tested. The mechanism-refuted part is a physical-prior statement; it holds only if the signal is a true E field, which is the question.
• Arbiter verdict (split):
• Closed-circuit motional emf is not removed by electrostatic shielding: Established (ordinary flux rule).
• A steady unshieldable external E from the generator that charges a capacitor on the far side of a shield: Never properly tested. Not dismissed without testing (Hooper's controls, Gibson's notes, a NASA attempt), but the controls are incomplete (no dummy winding, no copper-versus-nichrome, no step-response measured at microsecond-to-second scale, shield and detector on the same side), and the internal evidence (I^2 versus AC linear, I^3.8, resistance dependence, plate-thickness polarity, day-to-day drift, factor 100) favours an artefact. The mechanism as a field outside Maxwell's equations: Refuted by calculation only if the reading is a field at all; I keep that as the physicist's prior, not as a test result.
• The odd logic that a failed WWII drift meter demonstrated immunity (cancelling fields) is an analogy, not an experiment [read: essay pp.1-2, per physicist/contrarian].
• What would settle it: (a) the dummy tests under Claim 1; (b) hold current fixed and vary series resistance, supply symmetry, lead potentials and room temperature (B x V predicts no change); (c) measure the response to a current step on a microsecond to second timescale (thermal EMF needs seconds to minutes; a true field of this type would follow the current at once); (d) put the electrometer and capacitor outside the grounded cabinet; (e) read Experimental Notes and the 16 Dec 1969 shielding entry.

Claim 3. P.D. = B V l measures drift velocity and conduction-electron density; drift speed is higher at low temperature, up to ~1e10 in superconductors

• Agree: drift speed v = J/(n e) has no temperature dependence at fixed J; the patent's 1e8 cm/s is the Fermi speed, not a drift speed; the printed 3.6e12 cm/s (above c) was a typo corrected after Hooper's death (3.6e-2 cm/s; I recompute 3.53e-2); the Fermi-Dirac "100-fold fewer electrons" rescue is wrong because Fermi-Dirac statistics and Hall data fix n at about one electron per atom.
• Disagree on the capacitor-as-meter idea: physicist Never properly tested; mathematician Internally inconsistent; contrarian Tested, not supported.
• Stronger argument: the mathematician. One reading cannot give both V and n, so "measure both" is circular, and on Hooper's own numbers the instrument disagrees with his own model by 103; the data do not support "measuring drift speed" because the reading has not been shown to be B x V. The physicist's "never properly tested" is true of the idea in general (a properly controlled capacitor drift meter has not been built), and I keep that for the idea in the abstract.
• Arbiter verdict (split):
• 1e10 enhancement and temperature dependence of drift speed at fixed J; "100-fold lower n": Refuted by calculation (recomputed above; even at the depairing ceiling the factor is 4.7e5).
• The capacitor as a simultaneous measurement of V and n, as presented: Internally inconsistent (circular; 103x off its own model).
• The capacitor as a drift-velocity meter if the effect were real: Never properly tested.
• What would settle it: the Claim 1 dummy tests plus a measurement in aluminium (n about twice copper's, so V = J/(n e) falls by about half at the same J); if the reading does not scale as 1/n, the "drift meter" reading is not B x V.

Claim 4. The field polarizes neutral atoms and attracts matter like gravity; gravity is an electric polarization effect

• Agree (all three): the dielectric force F = alpha E grad E is standard and attractive; gravity is composition-independent to ~1e-15 (MICROSCOPE, [web] https://arxiv.org/abs/2209.15487) while the polarization force scales with polarizability, not mass; the attraction experiment was only planned [read: essay p.7].
• Disagree: numerical size only (see table). Physicist 1e-23, mathematician 2e-21; I find 1e-24 (classical V) to 1e-20 (Hooper's measured V, alpha 6 A^3) to ~3e-20 (alpha 15 A^3). Either way 20 or more orders of magnitude short, so no verdict changes. The mathematician's extra points (wrong distance law: 1/r^3 for a line field and 1/r^5 for a point source against 1/r^2; no field E_g ever defined) are unchallenged and I keep them.
• Arbiter verdict (split): gravity as the polarization force: Refuted by calculation (magnitude, distance law, mass-independence). A nonuniform E field attracts neutral polarizable matter (dielectrophoresis): Established, irrelevant to gravity.
• What would settle it: done by calculation. An experiment would be a torsion-balance comparison of materials of equal mass and different polarizability next to the generator; mainstream predicts no difference to 1e-15 (a test that would also settle the strong form).

Claim 5. A strong B x V field cancels or reverses the gravity-induced polarization (weight loss, anti-gravity, lift)

• Agree: no equation links polarization to weight; the patent's five claims cover hardware only; NASA Lewis (Millis and Williamson, TM-106963, 1995) built the self-cancelling coils, weighed a 555 g acrylic disk with threshold 0.02 g (4e-5 of the weight; I recompute 0.02/555 = 3.6e-5), accelerometer threshold ~40 ug, coils also in ice, and saw nothing [read per all three; [web] https://ntrs.nasa.gov/citations/19950022472]. Caveats all three state: an agency propulsion screen interested in a large effect; on/off toggling without warm-up (Hooper took data warm); a "minuscule" effect below threshold not excluded; no superconducting coil (cost ~$10,000); the pancake coil, not the 4020-wire cylinder with which Hooper's capacitor data were taken. Hooper reports no weight data himself. Hartman (1996): Hooper "and others" witnessed gravitational forces; one secondary review, no witness named, no second source found; I keep it in the record beside Hooper's own statement that the attraction test was "planned".
• Mathematician's addition: with B = 10 T and V at the Landau speed (170 m/s), E = 1.7e3 V/m, giving 3e-9 of weight on the same force law. I recomputed 10 x 168 = 1.7e3 V/m; the force is alpha E^2/L with L = 0.1 m: alpha 1 A^3: 1.1e-40 x (1.7e3)^2 / 0.1 = 3.2e-33 N; copper weight 1.0e-24: 3e-9. Confirmed. Even at an unattainable field the force is nine orders short. FIG. 3 is an ordinary inductive coil with a conducting shell, so Faraday and eddy-current (Lenz) forces would mimic a weight change; the patent does not discuss them [read per mathematician].
• Arbiter verdict (split): weight change from room-temperature and ice-cooled coils at the tested currents: Tested, not supported. Superconducting coil and spacecraft lift: Never properly tested, and Refuted by calculation on the polarization-force estimate, because it depends on Claim 4.
• What would settle it: a superconducting self-cancelling coil on a balance with 1e-6 sensitivity, run warm and cold, with the dummy and thermal-convection controls NASA suggested. Predicted by mainstream: no effect.

Claim 6. The 1957-59 Fisher/Hooper rotor gravity-meter deflections linked the field to gravity

• Agree (unanimous): Hooper's own controls (sound absorber removed the deflection, a buzzer or high-pitched sound reproduced it; "Thus our joint project collapsed" [read: essay p.3]) show an acoustic artefact. This is the best control in the record and it is Hooper's. Kragh independently confirms Fisher (1891-1966), the APS presentations and "two US patents to Hooper" [web, per Investigator and reviewers: Kragh, Varying Gravity, Springer 2016, pp.68-69]. All note the logical point that an unshieldable field would not be stopped by an acoustic barrier, and that Hooper kept the belief on the strength of the shielding claim.
• Arbiter verdict: Tested, not supported, by the experimenters' own controls. Historically instructive: it shows how a belief survived its own refutation. Original deflection data, rotor speed and meter type are not in any document found. Fisher's 1957 Gravity Research Foundation paper and the APS 1958 abstract were not located.
• What would settle it: nothing more is needed to retire the rotor evidence; recovery of the original records would only fix details.

Claim 7. The field penetrates barriers (communication, power, detonation) and energy "available everywhere on the earth" can be tapped

• Agree: no signalling data; the patent states only "I have demonstrated that AC current can be used" [read per physicist: patent p.5]; through-barrier low-frequency magnetic induction is real but is ordinary magnetic coupling [memory, high]; Hooper's belief that Moray tapped gravity (Gibson 1983) is a belief, not data.
• Disagree: physicist adds energy-tapping is refuted (a static conservative field does no net work in a closed cycle); mathematician and contrarian say Not testable as stated.
• Stronger argument: the physicist's energy argument is right if the field is a static conservative one, but the claim never specifies a source or how energy would be extracted, so it is not a quantitative claim to refute. The contrarian adds that the barrier-communication part is a "never properly tested prediction that depends entirely on Claim 2". I accept both.
• Arbiter verdict (split): barrier-penetration signalling by Hooper's field: Never properly tested (it depends on Claim 2; nothing reported). Through-barrier signalling by low-frequency magnetic induction: Established (a different mechanism). Free energy from the Earth's field: Not testable as stated (no mechanism or magnitude); conservation of energy gives a strong prior against a static field as a source.
• What would settle it: a pulsed-coil transmitter with the capacitor receiver inside a grounded steel enclosure and a matched magnetic-pickup receiver, with Claim 2's dummy controls; also a statement of what source supplies the energy.

3. Final verdict table

4. Weighing the Investigator's findings (kept in the record)

All of these stay in the record whatever the verdict.

• US 3,656,013 (filed 19 April 1968) and French family member FR 2039172 A1: FIG. 11 yoke experiment (galvanometer follows the iron yoke; Hooper reads this as flux rotating with the system) and a counter-rotating rotor in a grounded high-permeability box. Standard unipolar induction in symmetric geometry; interpretive. The French application was withdrawn; the reason is unknown. The box experiment is the rotating analogue of Claim 2.
• Similarities paper: the only quantitative account by Hooper. Controls: quick current stepping (30-10-30-10 A, "substantially identical"), larger outer capacitor with grounded inner pair, gaussmeter (max ~0.5 G) and five-wire pickup loop (no emf), a built balanced supply, an unnamed "disinterested expert" confirming readings with an oscilloscope (his characterization). Errors: eq. (3) misplaced ln; "j = 0.04172 amp/cm2" is a cross-section; "d = 9.96" for 8.96 g/cm3; page citation 207 versus 209 for Bull. APS 15(2); "good agreement" versus a factor ~100.
• Gibson 1983: participant, financier's wife, later company president and publisher, so an interested party; but candid, including data against the claim. Her statements that the original equipment was kept (1983) and that no one continued the work are single-source.
• NASA TM-106963: the one independent test; a null on weight; capacitor readings "too erratic". The authors suggest Hooper "may have misinterpreted thermal effects", an inference from his papers and Gibson, not a measurement.
• Klicker 1986: sympathetic thesis; shows moving flux reproduces the (v/c)^2 term of special relativity when flux drift is half the carrier drift; reports duplication of shielding demos at Montana State (report not seen); states an electron-beam test was begun but not completed for lack of beam size and electrometer sensitivity. He "has not been able to disprove Hooper's claims".
• Cravens 1990 (USAF): "There has been no critical experiment which can disprove whether a magnetic flux rotates with its source" [web, excerpt: https://www.rexresearch.com/hooper/hooper1.htm]. Consistent with the "not testable in symmetric geometry" ruling.
• Hartman 1996 and later patents citing Hooper (Banduric, PST Associates, WWK Trust, Lockheed Martin homopolar): continued interest, not corroboration.

5. Absence of evidence: the record goes dark after 1971

What goes dark: Hooper (president and director of research of Electrodynamic Gravity, Inc., 1964-71) died in February 1971 [web: http://wiki.naturalphilosophy.org/index.php?title=William_J_Hooper] before either patent issued (US 3,610,971 on 5 Oct 1971); the Certificate of Correction followed in May 1972. The company (Florida) was dissolved 9 December 1991 [read per Investigator: Sunbiz]. No later experiment, replication attempt by the company, or publication is on record until Klicker (1986) and NASA (1995), and the Experimental Notes are not online.

Reading: the first two explanations fit the available record and need no extra hypothesis; the absence of any later attempt to rebuild a 4020-wire generator and test it at a neutral lab is itself a gap, since the apparatus is cheap and was said to be kept. That the gap persists means the effect is untested, not refuted. The USPTO file wrappers and Ohio and Florida filings are the cheapest decisive records; AFRPL-TR-72-31 is the lead that could show whether the US Air Force propulsion community assessed the work.

6. Open questions

1. Does the generator reading survive a dummy winding (inductive wound, same geometry), a copper-versus-nichrome swap, and a fixed-current test varying external series resistance and lead potentials?
2. What is the time response to a current step (microseconds to minutes)? That separates thermal from field from leakage.
3. What do the 1954 shielding experiments and the 16 Dec 1969 Experimental Notes entry show beyond the closed-circuit flux rule? Is there any open-loop experiment with the detector outside the shield?
4. Did the 1985 Montana State work (Lamb; Stover and Klicker) exist, and what did it measure?
5. What were the original gravity-meter deflections in 1957-59, and what was the rotor?
6. Hartman: who were the "others" who witnessed gravitational forces?
7. Why was the French application withdrawn?
8. Copper-to-stainless thermopower for the actual plates; the polarizability of copper (bulk, not free atom) is moot for the verdict.
9. Does AFRPL-TR-72-31 assess Hooper's work?

7. Summary of the arbiter's changes to Step 1 and Step 2

• claims.md should note that the "immune to shielding" experiments are described in New Horizons ch. 3 (1954, closed-circuit) as well as the WWII story (mathematician).
• The physicist's "1.4e-4 V/m" and 1e-23 figure, and "a hundred billion times", are corrected above to 1.4e-3 V/m (as measured), 1e-24 to 3e-20, and ~1e10.
• The physicist's "Refuted by calculation" for the strong form of Claim 1 is downgraded to "Never properly tested" with a strong theoretical prior against.

8. QE revision

Trigger: qe.md (verdict MINOR). The QE's remediation deep reads have now been written and I read all six: deep/research-hooper-NewHorizons-main.md, -appendix.md, research-rexresearch-gibson.md, research-rexresearch-3656013.md, research-nasa-TM-106963.md and research-archive-Klicker.md. Tags as in the header: [read] = from a deep read of the document (I did not reopen the originals; I say "per deep read"), [calc] = my own numpy computation in .venv, [memory] with confidence. Web checks in this revision: none. Two [memory] inputs (Na+ mobility, copper resistivity at 195 K) are medium confidence and no verdict turns on them. Sections 1 to 7 above stand except where this section says otherwise; the final verdict table (section 3) is already updated.

8.1 Does anything change? Summary

No headline verdict reverses. Four things move:

1. Claim 2 gains a third part: the question that started Hooper's work (an open-circuit detector at rest inside a closed, grounded shield) was never tested in anything read. That part is Never properly tested.
2. The Gibson resistance data are stronger evidence for an artefact than I stated, and Hooper's reading of them (the number of conduction electrons changes) is Refuted by calculation. The overall generator verdict stays Never properly tested because no neutral test exists.
3. Claim 7 loses a wrong rationale ("no source specified") and gets the right one.
4. Three claims from the patents that the reviews and my first pass never listed (desalination, hydrogen to helium, mechanical rotor) now have verdicts (table rows 8 to 10).

8.2 Corrections to my own first pass (numbers and facts)

8.3 Claim 1 (moving flux, E = B x V): additions

• FIG. 11 yoke experiment (US 3,656,013) read against the textbook. In the patent, a U-shaped yoke of high-permeability iron carries two axial magnets and a brass rod; a stiff conductor on the rod sweeps the gap, with a galvanometer. Reported, with no numbers [read per deep read: 3,656,013, FIG. 11]: (1) magnets fixed, conductor swept: deflection; (2) conductor fixed, magnets and yoke turned together: equal and opposite deflection; (3) magnets fixed to the conductor, yoke turned: deflection; (4) yoke and conductor fixed, magnets turned: no deflection. Hooper concludes "the yoke structure plays an important part in controlling movement of the magnetic flux".
• My reading [memory, high: standard unipolar-induction reasoning]. The U-shaped yoke makes the gap field non-symmetric about the rod, so the field pattern is set by the yoke, not by the magnets. Turning the yoke turns the pattern, which is ordinary relative motion between conductor and field; turning axially symmetric magnets inside a fixed yoke leaves the pattern where it was, so no emf. All four results are what the textbook predicts. Result (4) is the one that would be positive if the flux rotated with a spinning magnet, as Hooper's Claim 1 says; he reports zero and explains it by the yoke. So this experiment is not a test of rotating flux; if anything it is a datum for the textbook side, which Hooper reads the other way. Cravens' remark (no critical experiment) stands: the geometry cannot discriminate. The patent mentions no controls for contact or thermal emf.
• Klicker's matching (per deep read, pp.44-49). Klicker takes special relativity (and the Moon-Spencer form of Ampere's law) as the baseline and asks what flux velocity reproduces it. Hooper's assumption (flux velocity equal to carrier velocity) misses the (v/c)^2 term by a factor 2; "exactly one half" fixes it. By Klicker's own words this is a theory match, "it is assumed that the baseline is correct, and only experimental evidence can determine if this is true". So the thesis does not support Hooper's equal-velocity assumption. His own worked figure (14 uV at 30 A, p.17) uses the full drift velocity 0.0176 m/s; with half the velocity it would be 7 uV (Klicker does not comment) [per deep read].
• Two further problems in Klicker's 14 uV figure [calc]. (a) 0.0176 m/s is Hooper's value extracted from his 10 A reading, not an independent number, so "compares favourably with some of his measured results" is circular; Klicker calls 0.02 m/s "an accepted value" for copper, but the classical value at 10 A in this wire is 1.8e-4 m/s (section 1), a factor 100 lower. (b) Hooper's model has V proportional to I, so at 30 A the model gives 9 times the 10 A value (4.4e-7 V using the classical V, or 4.5e-5 V using Hooper's extracted V), not the same V as at 10 A. My recomputation of Klicker's arithmetic: 1.42e-5 V, as he says. Klicker states there are no error bars and tabulates no Hooper data [read per deep read]. This weakens it as support; it does not make it evidence against.
• NBS letter and Smythe (contemporaneous mainstream response; per deep read of NH p.45, preface p.9-10). Hooper quotes an unnamed National Bureau of Standards (NBS) colleague: two assumptions are both "not in conflict with anything observed experimentally": (i) two equal and opposite electric fields exist inside a shielded moving enclosure, or (ii) at no point is there a field or charge; "any prediction not in accord with both ... should be considered as lacking an experimental basis". Hooper's gloss is that this takes a position "without a single iota of direct experimental evidence" (p.34) and that Fermi, asked in 1944 at Principia College, replied (as Hooper reports it, paraphrase) that he could recall no such evidence. Smythe (1939, p.500) says induced emfs in the shields "will set up electric fields tending to counteract the fields induced inside". These are the mainstream statements of the textbook view and Hooper's account of how it was reached; the NBS review of his WWII drift-meter plan (restriction: inoperable inside a metal aircraft) is documented only by Hooper, with no date, report number or reviewer name. Stakes: the NBS letter is reported by the inventor; the Fermi exchange is Hooper's paraphrase only.
• Weight. This material supports the split verdict. The two textbook assumptions make identical predictions for any observable, so "does flux move?" has no test in symmetric geometry; Hooper correctly points out that it was therefore assumed, not measured. Neither side measured it.

8.4 Claim 2 (shield-immune generator): additions and a three-part verdict

• The experiments, geometry by geometry (per deep read of NH ch. 3 and 7). All five shielding experiments Hooper describes have the detector in a loop (or an electrometer) sharing the shield with the source:

Hooper says the field "readily passes through the stainless steel walls of a grounded housing chamber" (ch. 7) but no reading outside the box is described. The only "through-metal" evidence is that the field is said to pass two 1/8 inch (1/4 inch in total) stainless capacitor plates (Similarities p.88), but the reading is itself taken between those plates, so it is not a test of passing through a shield to a detector beyond it. In the book the Ironing Board objection (that Faraday's law gives an emf in a loop) is answered by Hooper with flux cutting, which is ordinary physics and agrees with my first-pass split. - New third part, the original question. The NBS restriction is about an open-circuit voltmeter at rest inside a closed grounded shield that moves as a whole. No experiment in these pages does that. The textbook answer (Smythe) is zero net inside. Verdict on that part: Never properly tested by Hooper or anyone found; mainstream prediction clear; not dismissed without testing because the NBS letter says experiments agree with both descriptions. - Inventor-side rebuttals (retention; the QE found these missing). From Gibson 1983, as her statements: 1. The supply imbalance "was always a negative factor, causing our readings to be less, never more", so the tables were not redone; Hooper "did not feel that it changed the conclusions, even though the exactness of the measurements might be questioned". 2. Hot-generator increase "was not totally due to thermoelectric effects, because there was a substantial jump back at 25 amps"; hot 30 A gave 220 uV, cold about 115 (estimated from her parabola column). 3. Up-and-down current scans on 19 Jul gave repeat values (15 A: 30, 27, 26 uV; 10 A: 10, 10), "If our readings had been due to heat, they would not go up and down the scale with the current". 4. Readings were steady while current was held, "never just a swing of the needle and then back". 5. An unnamed "disinterested expert" confirmed the readings with an oscilloscope (Similarities p.87); no readings or name given. 6. The unbalance example, 100 V and 26 A: first negative to 11 uV and then positive to 17 uV in one direction, only positive to 28 uV when reversed; balancing gave equal readings in both directions. - How I weigh them. (1): the unbalance example gives 17 and 28 uV in the two directions and does not report the balanced value. If the error is one-sided, 28 is true and "always lower" holds. If it is antisymmetric (plus or minus about 5.5 uV around 22.5 uV) the larger reading is too large. Her data cannot tell these apart [calc]. (2): fair; it shows heat is not the whole story in that run, which fits a charging or lead-potential component as well. (3) and (4) answer a thermal drift and a one-time transient but not a current-proportional pickup. (5): single report by the inventor's side, unverified. - Gibson's resistance data against the model. Under B x V the P.D. depends on B (set by the winding current) and V (set by current density and n). A resistor in series outside the winding cannot change either at fixed current. Her tables show it does. - Fixed 13 A: 15 uV (3.7 ohm, 48 V) against 100 uV (12.3 ohm, 160 V): a factor 6.7 [calc]. Rheostat and photofloods gave the same 100 uV at the same voltage and current. - Fit of 13 DC rows (Feb 4 and Fig. 5; supply voltage V, current I, P.D. in uV; her values as printed) [calc]: ln P.D. against ln I alone gives P.D. proportional to I^2.24 with a scatter of 0.70 in ln (a factor 2); against ln V alone P.D. proportional to V^2.06 with scatter 0.19; both together P.D. proportional to I^0.29 V^1.91, scatter 0.17. The best-fit current exponent collapses once voltage is allowed. At 30 A: 60, 200, 360 uV at 110, 160, 225 V: P.D. proportional to V^2.5. The rows are hand-read from an interested party's tables across several days, and V is collinear with I where R is fixed, so I call this a strong indication, not a proof. - AC rows (added resistance): 6.7 A at 120 V and 18 ohm gives 160 uV, more than 25 uV at 15 A with nothing added and than 115 uV at 25 A [read: Gibson Fig. 7]. - Heating cannot be the explanation for this particular effect: the winding carries the same current in both 13 A rows, so its own Joule heating is the same; the external resistor is a rheostat in one run and 1500 W lamps in the other and the readings agree. What does change is the voltage on the leads, which is the quantity the supply-imbalance episode already showed couples into the reading (about 11 uV at 100 V). I therefore revise my section 2 weighing: the contrarian's electrostatic-pickup argument is stronger than I credited, and the physicist's heating argument explains the warm-up drift (hot 30 A about twice cold), not the resistance dependence. - Hooper's own reading (n, the conduction-electron density, falls with added resistance, rises about tenfold in dry ice, and is 3 to 4 times larger for AC): see Claim 3, 8.5. - Dry ice and temperature. Gibson's 28 Feb 1969 run: 10 A gave 1.5 uV against 7 to 15 uV at room temperature (a fall by a factor 5 to 10); 30 A gave 95 uV after 5 h (60 to 125 uV at room temperature, ratio 0.8 to 1.6) and 37 to 45 uV after 17 h (ratio 0.3 to 0.6). The fit is I^3.78 [calc, as before]. The winding resistance fell from 3.7 to 2.3 ohm (68 V at 30 A is 2.27 ohm); the ratio 0.61 matches copper near 195 K [memory, medium], so this is the one temperature estimate for the run, and it is from resistance, not a thermometer. Hooper's book (ch. 7, p.73) says cooling to "-50 C ... approximately doubles V", then in the next sentence that V "does not increase with decreasing temperature" (the two sentences contradict each other as printed; a typo or a transcription slip; and dry ice is -78 C, not -50 C [memory, high]). Neither statement matches Gibson's tables as a doubling. The hot generator (not the cold) came nearest to doubling. - Field size in the book. Ch. 7 says the field near the generator is "approximately of the order of ~10^-2 volts/meter, with a current of 30 amperes". Scaled from my recomputed 1.43e-3 V/m at 10 A by (30/10)^2 this is 1.3e-2 V/m [calc], so Hooper's own number is the measured scale and my section 1 ruling (1.4e-3 V/m at 10 A) agrees with it. The physicist's 1.4e-4 is still the slip noted. - Detector geometry numbers. Hooper's Similarities radii are 0.1 and 0.1035 m; Hooper's lab notes of 4 Feb 1969, as quoted by Klicker, give 0.10265 and 0.10615 m (285 pF). The two sets differ by 2.5 per cent in ln(r2/r1) (0.0344 and 0.0335); Gibson says "different size capacitor plates were on the generator" at different times. Small, but the tables, the Similarities figure and Klicker's 14 uV were not all taken with one capacitor. - Arbiter verdict (split, revised): - Closed-circuit motional emf not removed by electrostatic shielding: Established (unchanged). - Open-circuit detector at rest inside a closed grounded shield: Never properly tested (new part). - Steady unshieldable external E from the generator that charges a capacitor beyond a shield: Never properly tested (unchanged). The new weight on the artefact side is the voltage dependence, which is the inventor's own data. It is still not a controlled test, because no dummy winding, no copper-versus-second-metal run and no detector outside the box exist. - Mechanism outside Maxwell's equations: Refuted by calculation only if the reading is a field (unchanged). - What would settle it (added): (f) repeat the Gibson Fig. 5 test at fixed 13 A with 0, 5 and 12 ohm in series and a floating, a grounded and a balanced supply, with leads of fixed geometry; B x V predicts no change; (g) put the electrometer and capacitor outside the box and compare with a dummy winding of the same geometry carrying the same current.

8.5 Claim 3 (drift velocity, n, temperature) and Claim 4 (gravity from polarization): additions

• Claim 3, Hooper and Gibson's interpretation of the tables. Gibson 1983: because B does not depend on temperature or on an external resistor, a change in P.D. means a change in V and hence in n: "almost ten times as many conduction electrons" in dry ice at low current, a fall in n with added resistance, "3 to 4 times as many" with AC. I rule on it. Carrier density in copper is fixed at about one conduction electron per atom (8.5e28 per m^3) by Hall-effect and band-structure data [memory, high]; drift speed at a given current density is J/(n e) with no temperature dependence. An external resistor cannot change the winding's n at all. The rise of P.D. with an external resistor (8.4) is exactly what this reading would need n to do, so the reading is Refuted by calculation; the data are better read as an artefact class. That strengthens my first-pass Claim 3 verdict and adds the sentence in table row 3.
• Claim 3 in Hooper's own cold run. The prediction "V higher at low temperature" was tested by his own dry-ice run and did not show the rise at 10 A (the fall was 5 to 10 times) or at 30 A (ratio 0.3 to 1.6). I add Tested, not supported (by his own data); it is not a neutral test and the temperature was not measured, so I do not promote it further.
• Claim 4, ch. 5 of the book (per deep read, pp.57-67). Hooper's argument: gravity is "the difference between the pull and the push of a convergent B x V electric field acting on the electrical constituents of atoms".
• The field is derived for a Bohr hydrogen atom (a model in which the electron circles the nucleus, which the quantum ground state does not do [memory, high]). The far field is printed as E = -q r0 w^2 / r (his eq. 5.2, "mks as printed"). Units: that expression has units of C per s^2, not V/m; the Biot-Savart factor mu0/(4 pi) = 1e-7 is missing. With the factor restored, E = 1e-7 q r0 w^2 / r (the flux moves at V = (r0 + r) w) [calc].
• Speed limit. With w = 2 pi x 6.8e15 per s (Hooper's figure; the Bohr value is 6.6e15 [memory, high]), c / w = 7.0e-9 m, about 135 Bohr radii. For r larger than that, V = r w exceeds the speed of light, so the rigidly rotating flux assumed in the formula is impossible beyond about 7 nm, a few tens of atomic spacings. Ch. 5 applies the field at macroscopic distances. [calc]
• Newton's law. Eqs. 5.4 to 5.12 reach F = G M1 N2 / r^2 through a chain of proportionalities: elastic restraint, amplitude proportional to the square root of the energy, and the step Q^2 = k0 M1 N2 ("the product of the two atomic masses"), which is set by assumption, not derived. So the mass product that makes the force law "gravity" is put in at the end. Hooper himself calls the derivation "very elementary and non-rigorous", and says the spin field's force goes as 1/r^4 which "obviously disqualifies it for the role of gravity".
• Status as he states it: "Whether or not this field will ultimately be identified with gravity remains to be shown experimentally" (p.67). He also states the weak form of the correspondence, which I keep: a convergent field would attract polarizable atoms (dielectrophoresis, Established).
• Verdict stays Refuted by calculation (magnitude, distance law, mass-independence, plus these three internal points).

8.6 Claim 2 and the Klicker record: Montana State and the other statements

• Reference [24], answered (open question 4). [read per deep read: Klicker p.61] "Lamb, T., 'Unique Properties of Motional Electric Fields,' EE 470 Special Topics, Montana State University, 1985 ... Work performed by J. C. Stover and K. A. Klicker on experiments with shielding motional electric fields and induction in field free regions summarized in paper by T. Lamb." So it is a 1985 course paper, written by a student who summarises work by two others, one of whom is the thesis author. It is not in the folder; the thesis gives no apparatus, magnitude, instrument or count. The experiments are described as showing that the motional field "cannot be electrostatically shielded by a Faraday cage held at a fixed potential enclosing the detection device", with the detection device described generically as "a conductor and an ammeter that would measure the induced current in the conductor when it was passed through a magnetic field" (p.13-14). That is a closed-circuit flux-cutting experiment (my 8.4 class 1-4), not the 4020-wire generator. Klicker states: "Originally an experimental approach was attempted. But limitations in available hardware and instrumentation ... has forced an analytical investigation" (p.5), and an electron-beam capacitor test "prevented the completion" for lack of beam size and electrometer sensitivity (p.54). Hooper's generator reading therefore has no independent replication in this record. Klicker also writes that Hooper "(with possibly one exception) has been the only one to actually measure this effect" (p.4).
• Klicker's magnetic-shield statement (per deep read, p.9-10). "The only way to shield a material from a motional electric field is to use a magnetic shield (high u material) around the source of the magnetic flux" and a shield around the detector merely redirects flux, "the shielded area still cuts flux". This is Klicker's reasoning with no experiment attached, and it contradicts Hooper's "immune to shielding, magnetic or electrostatic" as an unqualified statement. It is consistent with textbook physics for flux cutting. I keep both statements in the record. In the generator the winding is non-inductive, so there is no external flux for a source shield to contain; that is part of what the design asserts.
• Graneau remark (hearsay-level, one source; retention). Klicker's reference [13] relays a private conversation at the IEEE Tesla Symposium (Aug 1986) with P. Graneau, who "mentioned ... that an experiment had been done where a charged object reacted to steady state current flow in a metallic conductor". No data, no second source found. An ordinary candidate is the surface-charge electric field that exists outside any resistive current-carrying wire [memory, high]; the record does not say which was meant.

8.7 Claim 5 (weight) and NASA: additions

• "Rise of the nonferrous disk" resolved (per deep read, NASA p.5). The sentence is conditional: "It is suspected that any rise of the 'nonferrous disk' observed may have been due to upward convection currents". It refers to the disk of the patent's FIG. 4 weight device. NASA gives no Hooper weight reading, date or value anywhere. It also says: "Hooper's own experiments concentrated instead on measuring the charge build-ups", that his documented experiments "did not use the flat coil associated with the weight change claim", and "nor is it clear what experiments he conducted to explore the gravity connection". My conclusion: no weight observation by Hooper exists in the documents read, and the convection suggestion is a suspicion, not a test. The patent FIG. 3 and 4 descriptions are in the future tense ("should yield", "should weigh less"). Hartman's single secondary statement that Hooper "and others" witnessed gravitational forces stays in the record beside this absence.
• NASA caveats strengthened (per deep read): (a) the surface-charge readings were "too erratic", "highly suspect"; (b) the coils were not perfectly self-cancelling: residual field 15 to 21 mG for the thin-wire coils but 513 mG (10 AWG, 50 A) and 895 mG (4 AWG, 100 A), which moved the accelerometer by -0.8 ug per mG, so those gravity readings were corrected; (c) plus-or-minus levels "larger than most of the recorded data"; (d) weight threshold 0.02 g on a 555 g acrylic disk (3.6e-5), accelerometer 40 ug; (e) Table I's "predicted Hooper effect" column is relative (units arbitrary) and is i^2 p/A divided by about 10 (my arithmetic on their Eq. 4; the factor is not stated in the paper); (f) the on-off method was chosen over warm-up, which Hooper used.
• Size of the field in NASA's coils [calc]. Treating each coil as a pair of opposed current sheets, with the classical drift velocity: 4 AWG at 50 A gives V = 1.8e-4 m/s and E about 1.3e-6 V/m; 4 AWG at 100 A about 5e-6 V/m; 10 AWG at 50 A about 1e-5 V/m. These are crude (a flat-sheet estimate) and fall in the range of the 1.4e-5 V/m I get for the Hooper generator at 10 A with classical V; if Hooper's extracted V (100 times larger) applied, 100 times larger. Either way the polarization force estimate in Claim 4 puts the weight change 20 orders short, so NASA's null cannot test the polarization mechanism; it does test the patent's literal prediction of a weight change from this geometry, and found none at 4e-5 of the weight.
• Verdict (split, wording refined): flat coils at room temperature and iced, at the currents used: Tested, not supported. Pulsed-coil FIG. 3 version and the vertical-bundle generator in a weight test: Never properly tested. Superconducting and lift: Never properly tested; polarization estimate refutes at attainable fields.

8.8 Claim 7 and the patent applications the first pass omitted

• Claim 7, corrected rationale (per deep read, NH ch. 5-6, p.64-69). Hooper does name a source and a method.
• Source: the orbital motion of electrons in the atoms of the earth, giving induction fields (not radiation) at about 6.8e15 per second ("approximately 10^15 simultaneous pulls and pushes ... each second"); the field has never been detected as an electric field, "probably due to its very high composite frequency nature" (p.85).
• Method: "a properly designed ultra-high frequency receiving circuit incorporating a transistor valve" converting "the kinetic energy of planetary electrons in the atoms comprising the earth" to usable power (p.67); ch. 6 asks for "ultra-high frequency rectifiers" and adds "there are no roadblocks on the horizon". Support quoted: Tesla (1933), Dyson (1958), Carl Eyring (BYU) on Moray (8.9).
• Magnitude: none. The only power figure is Moray's 50 kW. Ch. 6 gives no frequency number for the receiver.
• Why I keep Not testable as stated as a device claim, with a strong prior against: the source is the Bohr-orbit picture, in which an atom in its lowest quantum state has no orbital motion to tap [memory, high]; the formula for the field breaks above about 7 nm (8.5); and no field strength, impedance or power is given.
• The reviewers' version of the claim (a static conservative field can do no net work) is no longer needed; the claim as Hooper states it is an oscillating field, so my earlier "the physicist is right if the field is a static conservative one" applies to the static case only.
• Desalination (patent FIG. 8, in the folder patent, and in 3,656,013; per deep read). A coil on a hollow tube is said to drive Na+ toward the centre and Cl- outward in flowing brine, with a concentric tube drawing off the Na-rich centre stream. No salinity, flow, separation or data. [calc, with ion mobility of Na+ in water 5.2e-8 m^2 per V per s, [memory, medium]]: at Hooper's measured field scale 1e-2 V/m, the drift speed is 5e-10 m/s, so in 10 s the ions move 5 nm while diffusion over 10 s spreads them about 160 um (D = 1.3e-9 m^2/s [memory, medium]); at 31 V/m (the rotating-flux estimate 1 T x 31 m/s, if the flux rotated at 0.1 m radius and 50 rev/s) the drift is 16 um in 10 s, still below diffusion; at the mathematician's 1.7e3 V/m ceiling it is about 0.9 mm. Concentrated brine (about 0.5 mol/L) has a screening length near 0.4 nm [memory, medium] and carries a real current in any E, so the field would drive current, not a stable separation. I judge: Never properly tested (no data; depends on Claim 1 or 2), with the magnitude estimate showing no separation at the field sizes Hooper has measured.
• Hydrogen to helium (3,656,013). The text adds "one neutron ... add one more neutron ... add one more proton" to a hydrogen nucleus and confines an arc with the B x V field. No neutron source, no yield, no measurement. Proton-proton fusion needs energies of order keV to overcome a barrier of about 0.4 MeV (for protons at nuclear distance) [memory, medium], against fields of 1e-2 to 1e3 V/m acting over centimetres, which give eV to keV. Not testable as stated. It is Hooper's "utility", not a data claim.
• Mechanical rotor (3,656,013; Similarities p.86; Gibson p.125). The Similarities paper: a squirrel-cage of electromagnets at 10,000 rpm gave 15 V AC in a pickup coil (this is ordinary induction), and a semi-cylindrical capacitor over the rotor with an electrometer gave DC deflections that reversed with rotation direction and with magnet current; no magnitudes. The patent's grounded-box rotor "might be" examined with an electrometer outside the box; no readings in the patent. Gibson (1983): "Dr. Hooper spent many years building devices to rotate magnets at high speeds, but always found that the magnetic field of the magnets, and the vibration and noise of the motors interfered with his attempts to measure the field with either a capacitor connected to an electrometer or a gravity meter". These are two accounts from the inventor's side that do not agree on whether the rotor capacitor measurements worked. Verdict Never properly tested; a rotor with vibration was the same class of apparatus as the 1957-59 acoustic artefact (Claim 6), which I now cite as consistent but not as proof.

8.9 Retention items the QE found missing (kept in the record)

• Hooper's 1963 APS paper "Rotation of Flux about a Magnetic Axis" (published as "Unipolar Electromagnetic Induction"; St. Louis, 25-28 Mar 1963) and the 1931 Physics 1(1) paper [web, per Investigator: https://www.wiki.naturalphilosophy.org/index.php?title=William_J_Hooper]. Not read. It bears on Claim 1. Its uptake: DePalma N-machine papers and US5587618A (homopolar machine) cite Hooper; Lockheed Martin Energy Research's US5844345A cites US 3,656,013 [web, per Investigator: Google Patents, sites.md]. This is continued interest, not corroboration.
• Fisher and Pascual Jordan. Kragh (Varying Gravity, Springer 2016) documents Joel E. Fisher (1891-1966) as a promoter of an expanding-Earth hypothesis influential on Pascual Jordan's varying-G proposal, and confirms the Fisher-Hooper APS work [web, per Investigator; Kragh pp.68-69]. Fisher is a named, documented person on the rotor side of the story; the rotor measurements themselves were lost to an acoustic artefact (Claim 6).
• Moray and Eyring (per deep read of NH p.68). Hooper writes of T. Henry Moray's device, "claims, which appear to have been well documented", of capturing "cosmic radiant energy continuously, to the extent of 50 kw", research from about 1926, and that "Dr. Carl Eyring (in 1925, then head of the Department of Physics of Brigham Young University)" examined it and "could come to no other conclusion but that the electrical energy was obtained (at least in part) from field energy generated '...in the earth itself'". Sources Hooper gives are Moray's own self-published books (Eyring quote at their pp.144-145); no independent document is cited; the dates (1925 examination; research "around 1926") sit oddly together. Gibson adds that Hooper was "convinced that Moray had tapped the earth's gravitational field". Hooper writes that Moray was "virtually isolated" by fear of losing patent rights. I keep it as a cross-link to the Moray reviews (Morray-TH, TernionAG); I have not weighed Moray's claims here.
• Gilstrap 2017, "A New Way to View Electrodynamics" (ResearchGate), said to build on Hooper's 1968 operational approach; not opened [web, per Investigator].
• Later patents citing Hooper: Banduric (US9337752B2, US10084395B2, US10320312B2), PST Associates (US20050030139A1, US20050099761A1), WWK Trust (GB2355599A), Lockheed Martin (US5844345A) [web, per Investigator]. Not read.
• Hooper's six "operationally unique" fields, Table 1.1 (14 properties), the switching-circuit argument against the usual statement of Faraday's law (after G. I. Cohn's IIT thesis), the linear-versus-parabolic (v/c) argument, the Slepian (Westinghouse 1949), Moon and Spencer (1954-55), Page and Adams and Bridgman citations (per deep read, NH pp.14-36). These are the supporting literature he draws on; the Cohn thesis is the same source Klicker cites for the opposing conclusion (flux does not move with its source). Kept as leads, not weighed.
• Single-source items stated as such: the NBS review and letter and Fermi exchange (Hooper only); the "disinterested expert" (Hooper only); Gibson as sole source of all raw tables and of the statements that equipment was kept (1983) and that nobody continued the work; Hartman's "others" witnessing gravitational forces; Graneau (private remark).

8.10 Corrected claim list (what changes in claims.md, the primer and the table)

The claim statements 1 to 7 stand. The list of corrections to carry forward:

1. Claim 1. Add Klicker's factor one half and the FIG. 11 reading (8.3). Keep the 100-300 fold gap.
2. Claim 2. Verdict is now three-part (8.4). Give the 13 A numbers correctly (15 to 100 uV at the same current; 60, 200, 360 uV at 30 A); P.D. tracks supply voltage; add inventor-side rebuttals; add the NBS and Smythe exchange; replace "same n as copper" for nichrome.
3. Claim 3. Add "Refuted by calculation" for the change-in-n reading; add that Hooper's own cold run did not show the predicted rise; note the book's self-contradictory sentence on temperature.
4. Claim 4. Add the ch. 5 derivation findings (8.5).
5. Claim 5. Add that Hooper reported no weight observation, NASA's "rise" is conditional, the coils' residual magnetic fields and the factor 10 in Table I; fix the polarizability conversion (6.1 A^3).
6. Claim 6. Unchanged; add Gibson's remark on rotor measurements swamped by vibration and noise as consistent.
7. Claim 7. Replace the rationale (8.8); verdict wording changed in the table.
8. New claims 8 to 10 (desalination, hydrogen to helium, rotor): added to the verdict table with verdicts.
9. Open questions (section 6) updated: Q3 (shielding): five geometries listed, none is an open-circuit detector at rest inside a closed grounded shield, the Experimental Notes still unread; Q4 (Montana State): answered (8.6); Q5 (1957-59 rotor data) and Q9 (AFRPL-TR-72-31) unchanged; new Q10: what is the balanced-supply value in Gibson's 100 V, 26 A example; new Q11: does the 1963 APS paper contain an experiment that distinguishes rotating flux?