Ives: Quality Engineering (STEP 4b), round 2: the Ives-papers update

QE agent, 2026-10-07 (16:20 local). This round checks the review after the source folder gained Ives-papers/: 31 of Ives's own papers, 1924-1951, obtained legally under follow-up Q10. They were run through text prep (manifest now 38 documents, 42 chunks) and digested (digest files dated 16:16-16:19).

The previous QE round (verdict MINOR, then rechecked to PASS) is kept unchanged at the end of this file under "Round 1". A recheck of this round should be appended at the very end of the file under "Recheck (round 2)".

Inputs read in full: - _text/manifest.json and all 42 digests (the 31 new ones closely); - facts.json, claims.md, physicist.md, mathematician.md, contrarian.md, arbiter.md and primer.md, read for the sections each new paper bears on, plus a term-by-term search of all seven files; - sites.md, followups.json and the round-1 qe.md; - the Ruderfer appendix in Moray's Sea of Energy, 5th ed., pp.271-274, in review/Morray-TH/_text.

example.py was run. Spot checks were made against the extracted text in _text/. Tags: [read] = checked by me in the named text; [calc] = my arithmetic (Python, project venv).

Verdict: MAJOR

The review files (facts.json and claims.md 15:13, arbiter 15:15, primer 15:20, physicist 15:32, the other reviews 14:48-14:51) all predate the new papers. They still list most of those papers as "not obtained" or "not read". Three things now in hand can change a central verdict or add central claims:

1. Claim 4 (gravitation) rests on one abstract plus an inference, and the primary papers contradict that inference. - The arbiter's verdict is Never properly tested ("abstract only; low confidence"). It reasons that "the extra velocity terms were presumably chosen to reach 43" arcsec per century. - The primer repeats this: "one matched number does not test a force law with that much freedom". Its table says light bending, Shapiro delay and redshift are "not addressed in anything read". - The full papers say otherwise. JOSA 29:183 (1939) derives, "from purely optical considerations", the formula ds^2 = c^2 y dT^2 - R^2 dtheta^2 - dR^2/y with y = 1 - 2GM/(Rc^2). Ives calls it "the Schwarzschild formula of general relativity" [read: _text 1939, line 221]. He derives the gravitational red shift and the bending of light from it [read: digest p.2; text p.4]. - JOSA 38:413 (1948) gets the force equations from one postulate: a body at rest in the field has "the dimensions, clock rate, and mass of a body ... moving with the velocity acquired in falling from infinity". It shows that the starting relation "is a relation also obtained in the general relativity treatment". It states that "the force equations are actually implicit in the general relativity solution" [read: _text 1948 p.416]. - So nothing is fitted. The theory is the Schwarzschild solution of GR rewritten as forces acting in flat space with an ether. Ruderfer, the Moray link, says exactly this: "a classical derivation of the Schwarzschild metric, hence ... an identical explanation of the three famous tests of relativity"; and "any Lorentz-ether theory of gravitation which provides the static Newtonian gravitational force is necessarily equivalent observationally to general relativity" [read: Moray 5th ed. pp.271-272, review/Morray-TH/_text]. - The verdict is therefore likely to move. For predictions, to Established (they are GR's, for a single central mass). For the ether reading, to Not testable as stated. The reviewers must decide that from the papers.

2. Claim 3's remaining primary derivations are now in the manifest and carry sub-claims no review has weighed. - Phil. Mag. 36:392 (1945) derives mass, length and clock-rate changes from the conservation of energy and momentum for radiation striking a particle. It keeps "the space and time concepts of Newton and Maxwell", and it classes "the 'Special Principle of Relativity' as a superfluous hypothesis" [read: _text 1945, lines 52, 575]. - JOSA 39:757 (1949) argues that "the Special Theory of Relativity treatment leads to an internal inconsistency if clocks are admitted for distant time setting". It calls the Lorentz transformations "an approximation, owing their simplicity of form to the exclusion of clocks in the determination of epoch" [read: _text 1949, lines 346-360]. - The arbiter's sub-verdict, "SR internally contradictory: Refuted by calculation", was reached from the 1953 "Genesis" summary. It must be re-checked against the 1949 argument itself, and against the 1948 one-way-light paper (JOSA 38:879), the 1950 paper (JOSA 40:185) and "Revisions" (Proc. Am. Philos. Soc. 95:125, 1951).

3. New claims by Ives are absent from claims.md and every review. Each is listed in section 3. The ones that make physical, checkable statements beyond "same predictions, different explanation" are: - the 1951 prediction that "the photon will go the way of the 'caloric'"; - the 1942 statement that earlier published values of radiation pressure on a moving absorber (Lorentz; Poynting and Abraham) are wrong; - the 1939 rotating disk that "dishes" under the Fitzgerald contraction; - the 1941-42 ripple-tank analogue experiments, in which contraction, clock slowing and Fresnel drag appear in a classical medium.

Remedy for MAJOR (STEP 4b): run the deep reads below, update facts.json and claims.md, then redo Steps 2-4. The reviewers should weigh the new papers directly.

The owner's question in this run should be answered in the primer and recorded as a follow-up (section 9). It asks why Ives matters to Moray, and whether a Lorentz ether "contributes no new forces beyond those specified by Lorentz/Einstein".

1. Coverage

• Nothing was skipped. _textprep.log shows all 38 documents extracted. OCR was used for 11 of the papers, and the 1951 lecture has 2 OCR pages.
• README count. The README in the source folder says "34 of Ives's own papers". The folder and the manifest hold 31 [read: source folder listing; manifest]. The difference is probably the 1952-53 copyrighted papers counted by mistake. The README is a project file in the read-only source folder; flagged, not a review defect.

2. Shortcuts

Under STEP 4b, a central claim or derivation that rests on a digest alone is a defect. Every new paper is digest-only so far.

• S1 (central): claim 4 rests on an abstract. The three gravitation papers (1939 JOSA 29:183; 1947 Phys. Rev. 72:229; 1948 JOSA 38:413) and the 1939 chronotopic-interval paper (JOSA 29:294) are in the manifest. The arbiter's and primer's arithmetic argument ("7.16 of 43 arcsec ... extra terms must supply the rest") was an inference from the abstract. The 1948 text answers it directly: the same paper notes that plain SR mass increase gives "one sixth the advance" [read: _text 1948 p.416]. Deep reads 1-3.
• S2 (central): claim 3 rests on the 1937 papers plus the 1953 summary. The 1945, 1949 and 1951 derivations are unread except in digest. Deep reads 4-5.
• S3: claim 1 does not use Ives-Stilwell Part II (JOSA 31:369, 1941). The 1938 caveat, kept in every file, was that all shifts are below the 0.09 A H-beta doublet separation and that voltages "at least twice" higher would settle it. Part II reached about 43,000 V with shifts "of over 0.11 A as against the previous maximum of 0.047 A" [read: _text 1941, lines 32, 56-58]. That is above the doublet separation, so it directly answers the caveat the reviews leave open. Part II also takes up R. C. Jones's objection (JOSA 29:337, 1939) that stationary particles might be dragged along, and finds an imparted velocity of "about 750 meters per second" [read: line 264]. Deep read 6.
• S4 (peripheral, digest acceptable for now):
• the 1924-26 photoelectric papers, the 1933 Ives-Fry paper and the 1951 Rumford lecture (context for Moray, not claims to judge);
• the 1937 JOSA 27:177 and 27:389 papers, the 1940 moving-mirror paper and the 1947 historical note;
• the 1942-44 radiation-pressure and wave-packet papers.

They need facts.json entries and a primer mention. A deep read is not required unless the reviewers take up the photon or radiation-pressure claims as verdict items.

3. Missed material (from the new digests, checked in _text where noted)

Not in facts.json, claims.md, any review, the arbiter or the primer (term search of all seven files: "Schwarzschild", "transverse force", "dish", "capillary", "ripple", "rod-clock", "Gauss", "radiation pressure" each give 0 hits):

1. Ives's gravitation is the Schwarzschild solution in ether language (1939, 1948). See Verdict item 1. - The 1939 paper says the formula has "no non-Euclidean geometry, nor new law of gravitation". Instead it "calls for anisotropy of the ether" around a mass [read: _text 1939, lines 271-272]. - Ives limits it to optics: "the derivation provides no information outside optics". The 1948 paper then extends it to dynamics (planetary orbits).
2. The Birkhoff paper (Phys. Rev. 72:229, 1947): a transverse velocity-dependent force. Rewriting G. D. Birkhoff's flat-space gravitational equations with Newtonian time and ordinary force, Ives finds this: "to the Newtonian attractive force is added a force normal to the direction of motion, v^2/c^2 times the component of the gravitational force normal to the motion". He likens it to Gauss's posthumous "fundamental law" for moving charges, the v^2/c^2 reflecting "attraction ... transmitted at light speed" [read: _text 1947, lines 13-14, 207-272]. - This is the one place where Ives names an extra force. It bears directly on the owner's question (section 9). - A footnote in the 1948 paper says Birkhoff's simpler equations and Ives's own differ, and that the equivalence postulate alone gives no ground to prefer either [read: 1948 last footnote]. - Ruderfer's note (8) describes this paper as "the proper correction of Newtonian gravitation, which has an infinite speed of propagation, for propagation at the speed of light" [read: Moray p.271-272].
3. The chronotopic interval (JOSA 29:294, 1939). - Ives identifies Minkowski's interval s with "the undistorted diameter of an interferometer" that always gives a null result. - He builds a table of combinations of contraction and clock rate in a gravitational field that all give null interferometer results. He says choosing among them needs experiments "where the gravitational parameter varies", such as near the Sun [read: digest pp.5-8]. - This is an explicit statement by Ives of where his framework is underdetermined.
4. Derivation from conservation laws (1943 JOSA 33:163, 1944 JOSA 34:222, 1945 Phil. Mag.). - Mass-velocity variation M0/(1 - v^2/c^2)^(1/2) is derived from Newtonian conservation of mass, momentum and energy applied to radiation, with mass equivalent m = E/c^2 for radiation. Contraction and clock slowing are then derived from a light-clock argument (D_t = D_r = (1 - V^2/c^2)^(1/2), D_s = 1). - Ives's stated aim: the Lorentz transformations as consequences of "fifty-year-old" principles. - A hidden-assumption question for the Mathematician: is radiation's E/c^2 mass already relativistic input? Poincare 1900 is classical, so possibly not, but it should be checked.
5. One-way light speed and the "rod-clock-quotient" (JOSA 38:879, 1948; 39:757, 1949; 40:185, 1950; Proc. APS 95:125, 1951). - Ives's later vocabulary: "velocity" is reserved for ether quantities, and "rod-clock-quotient" (Q) is what moving rods and clocks measure. - His claims: Einstein's statement that c was "firmly established by experiment" was invalid, because only out-and-back light speed had been measured. The constant ratio in the revised transformations "is not equal to c and should not be called the velocity of light". And abandoning the constancy principle "removes the basis for geometric theories claiming fusion of space and time" [read: digests]. - The current files have "rod-to-clock quotient" only as a 1953 term ("not in the 1937 papers"). It is in fact defined in 1948.
6. The rotating disk "dishes" (JOSA 29:472, 1939). Ives argues that Wood and Hackett's "twist" in the double Fizeau toothed wheel follows from the Fitzgerald contraction alone. For the Ehrenfest paradox (a rotating disk's rim contracts but its radius does not), he proposes that the disk "dishes" like a saw blade, with a paraboloidal profile for small angular velocity [read: _text lines 197, 234-242]. That is a stated physical deformation. Whether it differs from what SR with real elastic stresses predicts is a question for the Physicist.
7. Ripple-tank analogues with Stilwell (JOSA 31:14, 1941; 32:25, 1942). - Capillary waves on mercury, with air-jet sources and the medium moving at v/c = 0.585 (11.28 cm/s against a wave speed of 19.20 cm/s). The papers show that interference patterns become identical to the stationary case when lengths are scaled by the Fitzgerald factor and periods by the Larmor-Lorentz factor. The Fresnel drag coefficient and aberration also appear. - One measured ratio disagrees with the ideal prediction: 0.852 measured against (1 - v^2/c^2) = 0.658 predicted. The authors attribute this to capillary-wave dispersion [read: _text 1942, lines 38-41, 216-218; calc: 1 - 0.585^2 = 0.658]. - These are Ives's only experiments on a material medium that shows "Lorentz invariance" for waves. They are the clearest illustration for the reader of what a Lorentz ether means physically, and they are not mentioned anywhere. (Round 1 listed them as "not obtained"; they are now in hand.)
8. Radiation pressure in moving systems (JOSA 32:32, 1942). Ives derives P = E_a(c+v)/(c-v) on an absorber moving toward a source. He states that Lorentz's P = E_i and Poynting-Abraham's P = E_i(1 + v/c) imply that a co-moving source-absorber system would show first-order effects, and that his result removes them [read: digest; _text line 126].
9. The photon (1951 Rumford lecture). "I venture to predict that this will be done, and that the photon will go the way of the 'caloric' that Rumford demolished" [read: _text 1951 Adventures, lines 2185-2191]. He bases this on standing waves, on Lorentz's 1910 coherence objection, on Poincare's quantum "vestibule" in the atom, and on Beck-Sommerfeld bipartition-angle data that he says favour the wave picture. This is a position against mainstream physics (later photon-counting and antibunching experiments). It is pertinent to Moray, who wrote on radiant energy. It should be kept and weighed, not dropped.

10. The standing-wave photoelectric work, now in Ives's own words and numbers. Round 1 put this in the primer from the NAS memoir only, and the primer still says the Rumford lecture "has not been read" and that Ives-Fry 1933 is "known only from the bibliography title". The primary papers add numbers:

    • selective/normal emission ratios of 15-25 against at most about 4 from absorption differences (ApJ 60:209, 1924);
    • a ratio of 70 at 70 degrees for Na-K alloy, from Elster and Geitel (ApJ 60:231, 1924);
    • photoemission peaking 5-10 times above room temperature near 650 K (ApJ 64:128, 1926);
    • the commercial gas-filled potassium cells for picture transmission (BSTJ 5:320, 1926).

    They also add Ives's own caution. The 1933 paper expects "only positional and qualitative agreement" and calls the result "very satisfactory". The 1951 lecture says the theory is "confirmed to fine detail" [read: digests]. The memoirists' "hunch was right" can now be checked against Ives's own, more guarded 1933 wording. 11. Ives's own fairness to Einstein (JOSA 37:810, 1947). In his historical note, Ives writes that Voigt offered no physical justification, that Lorentz claimed no objective reality for "local time", and that Einstein alone "believed the equations' predictions" and named the canal-ray shift "as an experimentum crucis". He adds that the clock-rate test is the most conclusive because MM can be explained by entrained ether and mass variation by charge variation [read: _text lines 212-307]. This is pertinent to claim 2 and to the reader's picture of Ives. It is not in any file. 12. The 1937 proposal papers (JOSA 27:177 and 27:389). - Ives shows that Kennedy-Thorndike alone "cannot by itself test the Larmor-Lorentz hypothesis" without the contraction exponent. - He proposes the canal-ray experiment as the way to fix it, crediting "Einstein and Ritz" with seeing its crucial role 30 years before. - This is the origin of IS in Ives's own words; the files currently quote it only via the 1938 paper.

4. Accuracy spot checks (7)

Digest errors found (none has reached a later file, because the later files predate these digests): - Ives-Stilwell 1938 digest, p.2-3: "the H-alpha line (4861 angstroms)" is wrong. The text says the "blue-green H-beta line, of wave-length 4861 angstroms" (OCR "Hf") [read: _text line 133]. - Ives-Stilwell 1938 digest, p.7-8: "He (double) and H3" should be H2 and H3; "He" is an OCR artefact. - Ives-Stilwell 1941 digest, p.2: the same "He particles" artefact. - Ives-Stilwell 1941 digest, p.1-2: the factor is printed "(1 - v/c^2)^(1/2)"; it should be (1 - v^2/c^2)^(1/2). - Ives-Stilwell 1942 digest, p.4-5: the exponent error noted in check 7.

5. Retention

• Nothing pertinent was removed because a reviewer judged it negatively. Round 1's retention fixes (Pobedonostsev-Parshin, Faraj, the neo-Lorentzian line, Dirac, the NRL bibliography, the Butterfield letters, the standing-wave work) are still in the arbiter and primer.
• New retention risk for the redo: the photon prediction (section 3.9), the radiation-pressure claim (3.8) and the disk-dishing claim (3.6) will probably be judged against Ives. They must stay in the record next to the judgement, with their sources (SOURCE HONESTY, "Keep pertinent information").
• Moray link (pertinent, under-used). Ruderfer's full appendix is in the Morray-TH project text. It says more than the two README quotations the primer uses (section 9). Using only the two quotations is a condensation loss, not a negative-judgement drop.

6. Process

• Source tags. Honest throughout the existing files. But the "not obtained"/"not read" source-gap lists in claims.md, arbiter.md, primer.md and facts.json are now false as statements of what is in hand. They must be rewritten in the redo.
• Web checks. No new checks are needed for this QE. For the redo:
• the status of Birkhoff's 1943 flat-space theory (my recollection, [memory: medium], is that it reproduced the classic tests but drew objections, e.g. that it requires matter with sound speed equal to c);
• and whether Ives's 1939/1948 results are exactly Schwarzschild or only to order 1/c^2.

Both affect claim 4 and should be web-checked or calculated. - Verdict scale. Applied consistently in round 1. Claim 4 now needs re-rating from primary text. - Primer. It follows STEP 4 and the reader profile (round 1). Its section 4.4 table and arithmetic point are overtaken (section 7 below). The redo should explain the Schwarzschild metric from scratch: the reader profile does not assume GR, so the metric must be introduced in plain language as "how rulers and clocks behave near a mass". - example.py. Runs (exit 0). Its output matches the primer (0.0475 A; k = 0.972 +/- 0.031; chi-square 1.66 for 8 points). Suggested addition: the IS II high-voltage points (43 kV, 0.11 A; 28.3 kV, 0.0717 A), and a perihelion check. With Ives's force equations (equivalently the Schwarzschild orbit equation) the advance per orbit is 6 pi GM/(c^2 a(1 - e^2)), which gives 43 arcsec per century for Mercury. Plain SR mass increase gives one sixth. This lets the reader verify the "one sixth" in the 1948 paper. - facts.json. 26 manifest documents have no entry (section 1).

7. Corrections to earlier documents (for the orchestrator)

1. arbiter.md, claim 4. "the only content read is the 1948 abstract" and "the extra velocity terms were presumably chosen to reach 43" are overtaken. The 1939 paper derives the Schwarzschild form, red shift and light bending. The 1948 paper derives the orbit from the equivalence postulate and says the equations are those of the GR treatment [read: _text 1939 lines 221, 271; 1948 p.416]. The verdict must be redone in Step 3.
2. primer.md section 4.4. Overtaken in the same way: - the table row "Light bending, Shapiro delay, redshift, pulsars: not addressed in anything read"; - the paragraph "one matched number does not test a force law with that much freedom".

Redshift and bending are addressed in 1939. Shapiro delay and pulsars are still not addressed by Ives; but if the metric is Schwarzschild's, they follow from it, a point for the reviewers to state with care. 3. primer.md, claims.md, arbiter.md, facts.json: source-gap lists. These papers are no longer "not obtained" or "not read": Ives-Stilwell II (JOSA 31:369), Phil. Mag. 36:392, JOSA 29:183, JOSA 38:413, Phys. Rev. 72:229, Proc. APS 95:125, JOSA 31 and 32:25 (moving medium), and the 1951 Rumford lecture. All are in Ives-papers/. Still not obtained: JOSA 42:540 (1952; in copyright, cited by reference), the 1953 JOSA 43 notes, Rev. Sci. Instr. 9:283 (1938), Nature 168:246 (1951), Proc. Roy. Dublin Soc. 26:9 (1952) and Hazelett and Turner (1979). 4. primer.md section 4.8. "His 1951 Rumford Medal lecture ... has not been read here" and Ives-Fry 1933 "known only from the bibliography title" are outdated. Both are in the manifest and digested. 5. claims.md claim 3 and primer section 4.3. "the 1953 term is 'rod-to-clock quotient'; it is not in the 1937 papers" is incomplete. The term "rod-clock-quotient" is defined in JOSA 38:879 (1948) and used in 1949 [read: digests 1948, 1949]. 6. facts.json, Preikschat entry. "Ives 1948 itself not read" is outdated: JOSA 38:879 is in the manifest. 7. Digests. The five errors listed under section 4 (H-alpha for H-beta; "He" for H2, twice; the 1941 factor; the 1942 exponent). 8. README (project file, read-only source folder). "34 of Ives's own papers" is 31 in the folder. Flagged only.

8. Remediation (one round)

Deep reads (most important first; the returned list). 1. Ives 1948, JOSA 38:413, pp.1-4 (Opus). Do the force equations follow from the equivalence postulate with no adjustable term? What are the exact gravitational-mass formula and orbit equation (page images; the OCR is garbled)? Is the perihelion result exact or to order 1/c^2? How does Ives's system differ from Birkhoff's (the final footnote)? Does any prediction differ from GR's Schwarzschild results? 2. Ives 1939, JOSA 29:183, pp.1-5 (Opus). Check the derivation of ds^2 = c^2 y dT^2 - R^2 dtheta^2 - dR^2/y from the optical assumptions. Which predictions does he derive (red shift, light bending: does he get the full GR deflection)? What does "anisotropy of the ether" mean, and does it imply anything measurable beyond GR? 3. Ives 1947, Phys. Rev. 72:229, pp.1-4 (Sonnet). The transverse-force theorem: is it a reinterpretation of Birkhoff's equations or a proposed new force? Does Ives endorse Birkhoff's theory? What does the Gauss analogy claim? 4. Ives 1945, Phil. Mag. 36:392, pp.1-12 (Opus). The chain of inputs (Maxwell, Doppler, m = E/c^2 for radiation, conservation laws). Is any SR result assumed? How are D_t, D_r and D_s fixed? Does Ives state any measurable difference from SR? What exactly does "superfluous hypothesis" mean in context? 5. Ives 1949, JOSA 39:757, pp.1-5 (Opus), read with the digests of JOSA 38:879 (1948), 40:185 (1950) and Proc. APS 95:125 (1951). The "internal inconsistency" argument stated exactly. Does it survive the standard SR result that slow clock transport agrees with Einstein synchronization in the q -> 0 limit? What is the size of the k terms for realistic q? Does any experiment separate the theories? 6. Ives-Stilwell 1941, JOSA 31:369, pp.1-6 (Sonnet). Table I in full (voltages, particles, observed against computed shifts). Do the high-voltage points exceed the 0.09 A doublet separation and so answer the 1938 caveat? Jones's objection and the 750 m/s finding. The "slight tendency above theoretical curve" and its attributed cause. Otting's data.

Not in the deep-read list, but to be read by the Step 2 reviewers from _text (short papers): - 1942 JOSA 32:25 and 1941 JOSA 31:14 (ripple analogues); - 1939 JOSA 29:472 (disk dishing); - 1942 JOSA 32:32 (radiation pressure); - 1951 Rumford lecture pp.32-35 (photon); - 1947 JOSA 37:810 (historical note); - the Ruderfer appendix, Moray 5th ed. pp.271-274 (review/Morray-TH/_text, Sea-of-Energy 5th ed. file, around line 16540).

Then (MAJOR): update facts.json and claims.md, and redo Steps 2-4. - facts.json. Add the 26 missing entries (section 1). Correct the items in section 7. - claims.md. - Rewrite the source-gap note. - Rewrite claim 4 from the primary papers. - Add to claim 3 the 1945 conservation-law derivation and the 1948-51 one-way-light and "rod-clock-quotient" papers, with the 1949 inconsistency charge. - Add to claim 1 the Part II data. - Add the new items: the photon prediction (1951), radiation pressure (1942), disk dishing (1939), the ripple analogues (1941-42), and the mass-velocity derivation (1943-45). Each item should say whether it is a different prediction or a different explanation. - physicist.md, mathematician.md, contrarian.md (Step 2 redo). Weigh the deep reads and the items above. The Mathematician checks the 1939 metric derivation, the 1948 orbit equation (reproduce 43 arcsec and "one sixth") and the 1945 input chain. The Physicist states the mainstream view of Birkhoff's theory and of Ives's photon rejection. The Contrarian builds the strongest case for the gravitation papers' neglect being unwarranted (Ruderfer's view), and against. - arbiter.md (Step 3 redo). New claim-4 verdict; re-check of the SR-inconsistency sub-verdict against 1949; verdicts for the new items. - primer.md (Step 4 redo). Rewrite section 4.4 (gravitation) with a from-scratch explanation of the Schwarzschild metric. Update sections 4.3 and 4.8 and the source gap. Add a short section on the ripple-tank analogues as the physical picture of a Lorentz ether. Add the owner's section (section 9). Extend example.py with the perihelion calculation and the IS II points.

9. The owner's question in this run: what the primer must cover

The owner asked: why Ives is interesting to Moray, and whether "even if Ives is correct then Aether follows Lorentz and Einstein" means "a Lorentz Aether contributes no new forces beyond those specified by Lorentz/Einstein, so it's only a difference of explanation of an observed phenomenon". The files already read answer most of this. The primer redo should state it plainly, with these sources:

• Why Moray's book cites Ives (Ruderfer's argument, his words). Ruderfer's appendix ("Neutrino Structure of the Ether", Lett. Nuovo Cimento 13:1, 1975, reprinted in Moray 5th ed. pp.271-274) makes a three-step argument [read: review/Morray-TH/_text, around lines 16540-16650]: 1. The rejection of an ether applies "to a Galilean ether, but not to a Lorentz-invariant ether since its description of observation is necessarily identical to that of relativity". 2. Ives gave "the most direct verification of the validity of a Lorentz ether ... independent of relativity": a classical derivation of the Lorentz transformations (Phil. Mag. 1945) and of "the Schwarzschild metric, hence ... an identical explanation of the three famous tests of relativity". Ives's ether is therefore admissible. 3. An admissible ether "focuses attention on its properties". Ruderfer defines it as "the totality of all the presently unobservable forms of energy that may exist in the microcosm" and proposes a sea of low-energy neutrinos whose "total unobservable energy ... may far exceed the total energy contained in matter, i.e. the vacuum is then a plenum rather than a void".

That last step is what connects to Moray's "sea of energy". It is Ruderfer's addition, not Ives's: none of the 31 Ives papers digested assigns the ether an energy content that could be tapped. Ives's interest for Moray is that he made a physical ether respectable again within the experimental record. The energy claim rests on Ruderfer and Moray. - "No new forces": the reading in the question matches Ives's own and Ruderfer's statements for the core theory. - Robertson: "completely equivalent in substance to the special theory". - Ives 1937: the ether's "elusiveness by any measurements of the sort we have been considering". - Ives 1939: "no ... new law of gravitation". - Ives 1948: the force equations "are actually implicit in the general relativity solution". - Ruderfer: "any Lorentz-ether theory of gravitation which provides the static Newtonian gravitational force is necessarily equivalent observationally to general relativity" [read: sources above]. - Qualifications the primer must add, so the reader's summary is exact. Each item below goes beyond "same physics, different words"; the reviewers must weigh them: - (a) Ives does name forces that GR hides in geometry: the transverse v^2/c^2 force (1947). These are the same physics expressed as forces, not new effects, as far as the 1948 paper's claim of identity with GR holds (deep reads 1, 3). - (b) Ives makes some statements about measurement that differ in wording but not in prediction: one-way light speed and the rod-clock-quotient (1948-51), and the 1949 inconsistency charge. - (c) Ives makes some physical claims that are not just reinterpretation: - the photon will be abandoned (1951); - earlier radiation-pressure formulas were wrong (1942); - a rotating disk dishes (1939); - the Ives-Stilwell result shows clocks slowed relative to the ether, with the lab's ether speed undetectable. - (d) Ives's own explanatory claim is the aberration of starlight: he calls it the "strongest evidence" for the ether (1937 JOSA 27:263, p.10). It is evidence by "sheer weight of enormous probability", not a measurement that separates the ether from SR. - Recommendation to the orchestrator. Record the owner's question as a follow-up (tools/followups.py add Ives "...") so that the answer is kept in followups.json and on the site, as well as in the primer.

---

Round 1 (before the Ives-papers update), kept for the record

Round 1 report (STEP 4b, first pass, then Recheck)

QE agent, 2026-10-07. Inputs read in full: _text/manifest.json, all 12 digest files, all 4 deep reads, facts.json, claims.md, sites.md, all 15 research notes, physicist.md, mathematician.md, contrarian.md, arbiter.md, primer.md, followups.json. example.py was run. Spot checks were made against _text/ (manifest documents), research/files/ocr/ (Ives's own papers) and the text layers of the saved Lalli 2014 and Preikschat PDFs. Tags: [read] = checked by me in the named text, [calc] = my arithmetic.

Verdict: MINOR

The central verdicts are well supported and consistent across the three reviews, the arbiter and the primer. They are: claim 1 Established; ether-versus-SR Not testable as stated; the gravitation theory Never properly tested; the E = mc^2 circularity charge Refuted by calculation. Every load-bearing number I checked matches the source. The defects can all be fixed by updating facts, arbiter and primer:

• Several primary or near-primary documents that the Investigator saved in research/files/ were used only through abstracts or the Investigator's notes. These are the Pobedonostsev-Parshin data, the 1937-1938 Ives theory papers behind claim 3 and verdict 2D, the Faraj paper and the Preikschat thesis.
• facts.json still carries statements the later steps have overtaken.
• One Investigator figure is wrong ("Ives cited 27 times" in the Preikschat thesis).
• Some pertinent material was dropped between the reviews and the arbiter and primer: the related-work line, later uses of Ives's work, Ives's standing-wave photoelectric work.

Nothing found here is likely to change a central verdict. The Pobedonostsev-Parshin tables could change how the dissent on claim 1 is described, but not the verdict "Established", which rests on storage-ring tests at the 10^-9 level.

1. Coverage

All 7 manifest documents were digested; none was skipped. The 12 OCR'd Ives papers and the other 7 PDFs in research/files/ are outside the manifest. They are Investigator finds and were read by the reviewers to different depths (see 2).

2. Shortcuts

Digest-only coverage of Optica and Wikipedia is acceptable. The deep reads cover every central secondary source. The defects below are all about documents the Investigator saved that carry a central claim, an experimental result or a derivation, and that were then used only at second hand.

• D1. Pobedonostsev and Parshin (research/files/Pobedonostsev-Ives-Stilwell-Doppler-ritz-btr.pdf, 6 pp.). This is the only experimental dissent on claim 1. Every review, the arbiter and the primer say "data tables unread; text layer garbled". The arbiter's "what would settle it" item 1 is "read their tables". The paper is in hand and has only 6 pages. Rendering the page images is the method READING PDFs prescribes for garbled text. A 6-page document that a verdict caveat turns on was left at its abstract: defect.
• D2. Claim 3 primary derivations. The arbiter says "primary derivations unread", and the contrarian says "no primary derivation checked". But Ives's 1937 derivations are in research/files/:
• JOSA 27:263, transported rods and clocks, 11 pp.;
• JOSA 27:310, apparent lengths and times, 5 pp.;
• JOSA 27:305, aberration of clocks and the clock paradox, 5 pp.

The mathematician only "skimmed" 27:263. The only papers missing are 1945, 1949 and 1951. Claim 3's verdict (Not testable as stated; SR "internally contradictory" Refuted by calculation) rests on the 1953 "Genesis" summary, Robertson, and the secondary authors' footnote. Leaving a key document partly unread is a defect under STEP 4b. - D3. Verdict 2D (Sagnac and rotation, Refuted by calculation). The verdict was reached from Ives's 1953 "Genesis" paragraph. His own analysis, "Light Signals Sent Around a Closed Path" (JOSA 28:296, 1938, 4 pp.), is saved and unread. The verdict is probably right, but it was reached without reading the paper it judges. Minor defect. - D4. Direction of the 1938 centre shift. The arbiter, primer and mathematician all list "red or blue never checked on page images" as an open question. The page images are in hand (JOSA 28, 12 pp.). This is a small check that was left undone. - D5. Faraj, "The Ives-Stilwell Experiment" (research/files/gsjournal-727, 15 pp.). The physicist "leave[s] it to the Mathematician"; the mathematician does not take it up, and it drops out of the arbiter and primer. The paper claims that Ives and Stilwell measured "a slight asymmetry in the observed longitudinal Doppler effect for equal positive and negative velocities" [read: text layer, lines 25 and 48]. That is an anomaly claim about the central experiment, and it was never weighed. A related item: the NAS bibliography lists an Ives paper "Unsymmetrical Doppler Effect in Hydrogen Canal Rays", Rev. Sci. Instr. 9:283 (1938) [read: deep/1956..., section 5]. No file mentions it. - D6. Lalli 2014 chapter (7 pp., saved, with a text layer). All three reviewers cite it only through the Investigator's note. I checked the two quotations that matter, and both are accurate: - The Ives letter of 5 Oct 1951 ("the classical English school ... against the paradox of the Semitic school"). Lalli's own wording is "the implicit anti-Semitism that this sentence seems to imply". - Bridgman's "complete negation of the operational attitude".

[read: Lalli 2014 text layer, lines 106-112, 195-212]. Acceptable in outcome; the primer's "reads it as implicitly anti-Semitic" is a fair condensation of "seems to imply". It should still be read directly when the primer is next revised.

3. Missed material (from the digests, the research notes and the saved files)

1. Ives's standing-wave photoelectric work. Ives called the photoelectric effect his most important contribution to science [read: NAS digest p.19-20]. He explained the "selective" photoelectric effect of thin alkali films as standing-wave patterns formed by incident and reflected light, an idea he traced to his 1908 Lippmann standing-wave colour-photography thesis. He found photoemission "proportional to the intensity of standing waves in the film" [read: NAS digest p.22-23]. The same line of work produced "standing light waves using photoelectric probe surfaces" (1933, with T. C. Fry) and his 1951 Rumford Medal lecture "Adventures with Standing Light Waves". None of facts.json, claims.md, the reviews, the arbiter or the primer mentions it. It is mainstream and uncontested, so it is not a claim to judge. But follow-up Q10 names standing waves and photoelectric conversion as Moray's interests, and this is Ives's own work on exactly those topics. The primer and facts.json should carry a short paragraph on it.
2. The neo-Lorentzian line and later dissenters (Investigator priority 3). These are G. Builder (1957-58), M. Ruderfer's 1960 Mossbauer ether-drift test (PRL 5), S. Prokhovnik, H. Dingle and L. Janossy. Some reviewers mention them, but the arbiter and primer drop them; the primer has only one passing "neo-Lorentzians". Ruderfer is the link to Moray, so his 1960 experiment and 1975 theory belong in the primer's record section.
3. Later use of Ives's work. The Investigator found citations in Nieper's free-energy book (a direct link into the literature this archive collects), in the geocentrist "Galileo Was Wrong", and by the Chinese Academy of Sciences "Relativity Criticism Group" during the Cultural Revolution. It also found Butterfield's letters in the Library of Congress (LoC) describing anti-Einstein feeling in "certain scientific circles", with Einstein "castigated as left-wing radicals" [web: research/2026-10-07-google-search.md, 2026-10-07-loc-finding-aid.md]. All of this is absent from the arbiter and primer. Robertson's 1956 worry that Ives's work was "seized upon" is kept; the evidence that it actually happened is not.
4. Dirac 1951, "Is there an aether?" (Nature 168:906). A Nobel laureate reopening the aether question, which Ives answered in 1953. It appears only in the primer's reading list. It is pertinent to claim 2 and to the reader's sense of how closed the question was in 1951.
5. US Naval Research Laboratory (NRL) bibliography (Benton, May 1959). A government record that lists Ives and Stilwell 1938 and 1941 as the experimental verification of clock retardation, and Ives 1937 as resolving the clock paradox "through motion relative to the ether" [web: research/2026-10-07-wikisource-benton-1959.md]. Only the mathematician mentions it.
6. The clock paradox. Ives wrote on it in 1937 (JOSA 27:305, saved) and 1951 (Nature 168:246). No claim or review covers his position. A sentence under claim 2 or 3 would do.
7. The moving-medium interference experiments with Stilwell. "Interference Phenomena with a Moving Medium" I and II (JOSA 31, 1941; 32:25, 1942) [read: deep/1956..., section 5]. These are Ives's other relativity experiments, and they are not mentioned anywhere. List them in the primer's further reading as "not obtained".
8. The NAS memoir bibliography stops at 1951. It omits: - JOSA 40:185 (1950), "Extrapolation from the Michelson-Morley experiment" (cited by Wikipedia and the Optica list); - the 1952 mass-energy paper; - Proc. Roy. Dublin Soc. 26:9 (1952); - the 1953 "Genesis" and "Note" papers [read: _text NAS memoir pp.43-44].

facts.json says "the bibliography lists no E=mc2 paper". That is true, but the reason should be stated: the list is incomplete after 1951, so the gap is not evidence about the paper. A reader may wonder whether the omission is deliberate, since Robertson wrote the relativity assessment. The record does not say. List it under "Absence of evidence" as a small open question. 9. Biographical discrepancies not recorded (immaterial to the physics, but facts.json repeats them unflagged): - Death place. The memoir says the Century Club of New York, "within its walls", where he was stricken with a heart attack [read: NAS p.28-29]. Wikipedia says Upper Montclair, NJ [read: Wikipedia text], and LoC says New York. - The 1927 Hoover television demonstration. NAS and Wikipedia date it 7 April 1927; Optica dates it 27 April 1927 [read]. facts.json repeats Optica's date. - Line count. Wikipedia's "185-line" television in 1927 conflicts with the 50 scanning holes (NAS) and 50 lines (Optica, 1926 demonstration) [read]. - Longstreth Medal years. NAS gives 1906, 1914, 1918; Wikipedia gives 1907, 1915, 1919 [read: digests].

4. Accuracy spot checks (6)

Also checked: - Death place. The digest of the NAS memoir, part 02, p.29, says Ives died "at the National Academy of Sciences where he had been stricken". The memoir actually says the Century Club of New York [read: NAS pp.28-29]. This is a digest error. The NAS deep read has it right and the error did not reach the later files. - Patent count. The same digest says the patent list has "approximately 170 patents". The memoir's prose says "no fewer than one hundred", and the list holds roughly 85-100 legible entries [read: NAS pp.44-47, count of patent numbers]. This is another digest error. facts.json's "about 100" is right.

5. Retention

• Kept, correctly. Arbiter and primer both keep the material that a reviewer might have been tempted to drop: the Pobedonostsev-Parshin dissent, Ives's single-source Princeton and Pauli account, Robertson's stake, the 1951 Butterfield letter and Lalli's reading of it, the Jammer-Arzelies-Miller support for Ives's E = mc^2 critique, Bridgman, Miller's "only positive proof", Ives's own 1938 fine-structure caveat, and the 3-4% mean deficit. No item judged negatively by a reviewer was removed for that reason.
• Dropped without being weighed (defect, minor): the Faraj asymmetry claim (D5). It is pertinent material on the central experiment, deferred by one reviewer, taken up by none, and absent from the arbiter and primer.
• Dropped in condensation (minor; not because of a negative judgement): items 2-5 in section 3, namely the neo-Lorentzian line, later uses of Ives's work, the Butterfield letters, Dirac, and the NRL bibliography. Preikschat is also missing from the primer, though the arbiter keeps it.

6. Process

• Source tags. Present throughout and honest. The reviews separate first-hand and second-hand use carefully, marking for example "[read: Investigator's note]" and "second hand through Giuliani". The [calc] tag is used consistently and example.py reproduces it. One honest-tag issue: the physicist tags Lalli and Bridgman material as "[read]" with "via research notes". The contrarian's "[web: research note]" style is clearer.
• Web checks. The physicist web-checked Rainville, Saathoff, Reinhardt, Bailey and Sagnac. The arbiter used 3 of 8 checks, upgrading Alvager, Sommerfeld and the existence of Mandelberg-Witten. The contrarian used 2 of 6, and the mathematician one Wikipedia page. Medium- or low-confidence [memory] facts that a verdict depends on were checked. Hafele-Keating and Mansouri-Sexl remain [memory: high]; no verdict rests on them alone. Acceptable.
• Verdict scale.
• The physicist uses "Refuted" (not a scale term) several times, and "Plausible, untested" for an interpretation with no measurable difference. The arbiter normalised both correctly, to "Refuted by calculation" and "Not testable as stated".
• Remaining inconsistency, in the arbiter's claim 2B and the primer's evidence-table row 2B: "Rivals rejected on other evidence" is not a scale verdict. The physicist's term for the entrained-ether and ballistic alternatives, "Tested, not supported", should be used.
• In the arbiter, claim 2C is "Not testable as stated" and, "as a position", "Historically superseded but instructive". This is a legitimate split, stated clearly.
• Primer. It follows the STEP 4 order 1-9 with answers at the end. It defines every symbol and acronym at first use and gives a worked numerical example with real data. It suits the reader profile well; no tensor or GR machinery is used without introduction. The quality is high.
• example.py. Runs (exit 0). Its output matches the primer: 0.0475 A; k = 0.972 +/- 0.031; chi-square 1.66 for 8 points; Voigt 32.7 standard errors away.
• Source gap. Repeated and updated in every downstream file, as required.
• claims.md and facts.json. These were not updated after the Investigator, reviewer and arbiter corrections. That is the orchestrator's job, and it is still pending (see 7).

7. Corrections to earlier documents (for the orchestrator)

1. Preikschat citation count. - sites.md (the Preikschat row and Leads 3), research/2026-10-07-lightspeedtest.md, physicist.md (claim 2 paragraph d and the "Investigator's findings" list): "Ives is cited 27 times" is wrong. - The thesis has 8 occurrences of "Ives": text on pp.49, 53-54 and 158, and references on p.166 (JOSA 28:215 and JOSA 38:879). - The count of 27 came from a case-insensitive match that includes "gives". Evidence: text layer of research/files/Preikschat-1964-...pdf [read].
2. digest/1956-NAS...02.md, p.29. "died ... at the National Academy of Sciences where he had been stricken" is wrong. The memoir says he died within the walls of the Century Club of New York, where he was stricken with a heart attack, on 13 November 1953 (printed pp.170-171) [read].
3. digest/1956-NAS...02.md, p.44-47. "approximately 170 patents" is wrong. The memoir says "no fewer than one hundred" (printed p.168), and the list's legible entries number about 85-100 [read].
4. facts.json, Ohanian entry. "Ives's own paper is not in the folder; his argument is known only through this summary" and "discrepancy flagged, not verified on the web" are outdated. The 1952 paper and the 1953 Note are in research/files/ and were read; the journal is confirmed as JOSA 42(8):540-543, August 1952.
5. facts.json, Bonizzoni-Giuliani entry. "No numbers" and "a possible sign slip" are outdated. The numbers come from the 1938 paper's Tables I-III. The exponent slip is in the printed 1938 paper itself (p.216), not in the secondary authors' quotation.
6. facts.json, NAS entry. Add: the bibliography ends in 1951 and omits JOSA 40 (1950) and all 1952-53 papers. Add the death place (Century Club, New York) and the variant places and dates (13, 15 or 3 November 1953; Upper Montclair per Wikipedia).
7. facts.json, Optica and Wikipedia entries. - Flag the 27 April (Optica) versus 7 April (NAS, Wikipedia) 1927 date. - Flag Wikipedia's "185-line" against the 50-line (Optica) and 50-hole (NAS) figures. - Flag Wikipedia's Longstreth years (1907/1915/1919) against NAS (1906/1914/1918).
8. claims.md. Apply the arbiter's corrections 1-5, which are still not applied. Add a note under claim 2 on the clock paradox (Ives 1937 and 1951).

8. Remediation (one round)

Deep reads (most important first; see the returned list): 1. Pobedonostsev-Parshin, all 6 pp. from page images (Opus). 2. Ives 1937 JOSA 27:263 + 27:310, for claim 3: do the finite-q transformations differ measurably from Lorentz's, and does Ives state a measurable difference? (Opus.) 3. Ives-Stilwell 1938, pp.217-226 page images: direction of the centre shift, Figs. 13-14, systematic-error discussion, any asymmetry between the direct and reflected lines (Sonnet). 4. Ives 1938 JOSA 28:296 + 1937 JOSA 27:305: Ives's own Sagnac and clock-paradox analyses, to re-check verdict 2D and to add the clock paradox (Sonnet). 5. Faraj, gsjournal 727: what asymmetry is claimed, from which data, and whether it is Ives's own reported result (Sonnet). 6. NAS memoir pp.19-23 + bibliography 1933, 1951: Ives's standing-wave photoelectric work (Sonnet). 7. Lalli 2014, pp.359-365: Ives's principles in his own terms (natural philosophy vs prediction, operationalism), and verification of every quotation the review uses (Sonnet). 8. Preikschat thesis, pp.49, 53-54, 158: what it says about Ives 1948 (one-way light speed) and the Ives-Stilwell parameter in the Robertson framework (Sonnet).

Then update the later files: - facts.json. Apply section 7 items 4-7. Add the photoelectric and standing-wave summary to the NAS entry. - claims.md. Apply the arbiter's corrections 1-5 and section 7 item 8. Add a one-line "Also in the record" mention of the standing-wave photoelectric work and the Faraj asymmetry claim. - arbiter.md: add a "QE revision" section. - Weigh the Pobedonostsev-Parshin tables (claim 1 dissent) and the Faraj asymmetry claim. - Re-check verdict 2D against Ives's 1938 closed-path paper. - Re-check claim 3 against the 1937 derivations and update the "primary derivations unread" note: only the 1945, 1949 and 1951 papers remain unread. - Replace "Rivals rejected on other evidence" (2B) with "Tested, not supported". - Restore to the record: the neo-Lorentzian line (Builder, Ruderfer 1960, Prokhovnik, Dingle, Janossy), Dirac 1951, the NRL bibliography (1959), the Butterfield letters, and later uses of Ives's work (Nieper, the Chinese Academy group). - Add the 1951 cut-off of the memoir bibliography to "Absence of evidence". - primer.md. - Evidence-table row 2B: use the scale term. - Section 4.7: add the related-work line, later uses, Dirac, the NRL bibliography, the Butterfield letters, Preikschat, and the Faraj claim with its weighing. - New short subsection (or 2.x refresher plus a 4.7 bullet): Ives's standing-wave photoelectric work and the 1951 Rumford lecture, framed as context for the Moray link. - Update the claim 3 source-gap sentence. Record the direction of the centre shift once it is checked. Correct the dates in 4.7. - Further reading: add the 1941-42 moving-medium papers and the 1951 clock-paradox letter (not obtained). - sites.md and research note. Correct the Preikschat count, through the orchestrator.

Recheck

QE recheck, 2026-10-07. Checked the current arbiter.md, primer.md, claims.md, facts.json, sites.md, physicist.md, research notes and digests against the defects above (grep of the files, not a re-read of every source).

Fixed: - D1 Pobedonostsev-Parshin: deep read exists; arbiter claim 1 and primer 4.7 now weigh the dissent "from the tables". - D2 Claim 3 / 1937 derivations: read (arbiter QE revision C; primer section on the 1937 derivations; "approach but never coincide" withdrawn). - D3 Verdict 2D re-checked against JOSA 28:296 and 27:305 (deep read present; primer clock-paradox paragraph added). - D4 Direction of the 1938 centre shift: resolved from page images (arbiter line 88; primer: red shift of the centre). - D5 Faraj: deep read, facts.json entry, primer bullet weighing the asymmetry claim. - D6 Lalli: deep read present. - Section 3 missed material: standing-wave photoelectric work (claims.md, facts.json, primer 2.9), neo-Lorentzian line, Dirac 1951, NRL bibliography, Butterfield letters and later uses, clock paradox, moving-medium papers, memoir bibliography cut-off at 1951, and the biographical discrepancies (death place, 1927 date, 185 lines, Longstreth years) are now in arbiter, primer or facts.json. - Section 6 scale term: arbiter 2B and physicist now use "Tested, not supported"; the primer no longer has "Rivals rejected on other evidence". - Section 7 items 4, 5, 6, 7, 8: facts.json (Ohanian, Bonizzoni-Giuliani, NAS, Optica, Wikipedia) and claims.md updated; no stale "not in the folder" or "no numbers" wording found. Preikschat date and the "8 occurrences, not 27" correction are recorded in arbiter.md and primer.md.

Remaining (all small, all in files the orchestrator was to correct; none affects a verdict): 1. sites.md line 21 still says Preikschat "cites Ives 27 times"; the correct count is 8. 2. research/2026-10-07-lightspeedtest.md line 5 still says "Ives is cited 27 times". 3. physicist.md line 37 still says "with Ives cited 27 times" (the arbiter states the correction, but the reviewer file carries the error). 4. digest/1956-NAS-...02.md still says Ives died "at the National Academy of Sciences" (p.29; should be the Century Club of New York) and "approximately 170 patents" (p.44-47; should be about 100). The corrected values are in facts.json, claims.md and primer, so no downstream error remains.

(The one "27 times" hit in primer.md is the correction itself, "8 times, ... counting error", and is correct.)

Final: OPEN (4 issues)

Orchestrator 2026-10-07: applied remaining 1-4 (sites.md line 21, research/2026-10-07-lightspeedtest.md, physicist.md: Preikschat cites Ives 8 times; NAS digest .02: Century Club of New York, about 100 patents).

Final: PASS

---

Recheck (round 2)

QE recheck, 2026-10-07 (about 17:30). Checked the current files by grep and file listing, not by re-reading every source: facts.json (43 entries) against the manifest, claims.md, physicist.md, mathematician.md, contrarian.md, arbiter.md, primer.md, example.py (run), the digests, deep/ and followups.json.

Verdict of round 2 was MAJOR. The redo has been done. The reviewers, arbiter, primer and example.py postdate the deep reads (deep reads 16:27-16:54; contrarian 17:04, physicist 17:07, mathematician 17:10, arbiter 17:15, example.py 17:18, primer 17:25).

Defects

• Coverage (26 manifest documents without a facts.json entry): FIXED. All 38 manifest documents now have an entry (facts.json has 43). The Preikschat entry no longer says "Ives 1948 itself not read".
• S1 claim 4 (gravitation) on an abstract: FIXED. Deep reads exist for JOSA 29:183, Phys. Rev. 72:229 and JOSA 38:413. The arbiter replaced the verdict with a split: Established for the single-mass numbers (Mercury, light bending, red shift), Not testable as stated for the ether and force reading, Never properly tested for extensions and the large-interferometer conjecture. The earlier "presumably chosen to reach 43" and "one matched number" are withdrawn in arbiter, physicist, mathematician, contrarian and primer, each with a correction note.
• S2 claim 3 (1945, 1949, 1951 derivations): FIXED. Deep reads exist for Phil. Mag. 36:392 and JOSA 39:757. The 1948-51 papers are covered through the 1949 deep read and the digests. The arbiter records that the Physicist-versus-others split closed once the 1945 and 1949 papers were read. The primer has routes (a), (b), (c).
• S3 Ives-Stilwell Part II (43 kV answers the 1938 caveat): FIXED. Deep read exists. Arbiter and primer carry it, and example.py Part 6 reproduces the 43 kV and 28.3 kV points.
• S4 peripheral papers need facts entries and a primer mention: FIXED. Entries exist. The primer lists the digest-only papers and the ones read in _text by a reviewer.
• Missed material 1-12: FIXED. The Schwarzschild reading, Birkhoff transverse force, chronotopic interval, conservation-law derivations, rod-clock-quotient, disk dishing, ripple tank, radiation pressure, photon prediction, standing-wave numbers, Ives's note on Einstein and the 1937 proposal papers are all in claims.md, the reviews, arbiter and primer. Term counts in the seven files are now non-zero for every term the QE found at zero (for example "ripple" 8 in arbiter and primer, "dish" 3-6, "caloric" 1-3).
• Retention risk (photon, radiation pressure, disk): FIXED. All three are kept in arbiter and primer next to the judgement. The Moray/Ruderfer appendix is used in full (primer section 4.10), not just two quotations.
• Corrections 1-6 (arbiter claim 4; primer 4.4 and 4.3 and 4.8; source-gap lists; rod-clock-quotient dated 1948; Preikschat): FIXED. The stale phrases appear only inside correction notes. The source-gap lists in claims.md, arbiter.md and primer.md were rewritten and now say what is still not obtained (JOSA 42:540 free copy, the 1953 JOSA 43 notes, Rev. Sci. Instr. 9:283, and others).
• Process: FIXED. The verdict scale is applied; the claim 4 re-rating is from primary text; the metric is introduced from scratch in the primer; web checks are recorded in the arbiter (5 of 8 used).
• example.py: FIXED. Runs (exit 0). Extended with the IS II points (Part 6), the perihelion advance and the "one sixth" check (Part 7), and the ripple-tank numbers (Part 9: 0.658, 0.811, 0.852).
• Owner's question in the primer: FIXED. Primer section 4.10 covers why Ives matters to Moray (Ruderfer's three steps, with the energy claim assigned to Ruderfer and Moray, not Ives) and the "no new forces" reading with the qualifications. Section 4.10 also adds that Ruderfer's "necessarily equivalent" is not a theorem (Nordstrom's scalar theory). The arbiter has a section on the owner's questions.
• Digest errors (QE section 7, item 7): REMAINS. The digest files are unchanged since 16:19. Still present: "Ha line (4861 angstroms)" and "Ha doublet" in the 1938 Ives-Stilwell digest (should be H-beta); "He (double)" in the 1938 digest and "He particles" in the 1941 digest (should be H2); the 1941 factor printed "(1-v/c^2)^(1/2)" (should be v^2/c^2); the 1942 ripple-tank exponent; and the JOSA 29:294 digest sentence "c_R = c y^(1/2)". The arbiter lists them as "do not use", and no later file repeats them, so no downstream error results. The digests are the record of what the digester wrote; the correction (or an explicit "known errors" note) has not been applied.
• Record the owner's question as a follow-up: REMAINS. QE section 9 and the primer both ask the orchestrator to run tools/followups.py add Ives "...". followups.json (last changed 15:52) holds only Q10. The answer is in the primer and arbiter but not in followups.json, so the site's "User follow-ups" box will not show it.

Not a defect, for the owner

• The README in the read-only source folder says "34 of Ives's own papers"; the folder holds 31. Unchanged, as it should be.
• The primer is still being edited (modified 17:25). The checks above were made on the file as it stood at 17:30.

Final: OPEN (2 issues)

Orchestrator 2026-10-07 (after round-2 recheck): issue 1, a "Known digest errors" note prepended to the five affected digests (1938 IS, 1941 IS II, 1941 and 1942 interference, JOSA 29:294); issue 2, the owner question recorded as Q13 and marked answered (research/2026-10-07-followup-Q13.md, pointing to primer 4.10).

Final: PASS