Before Sputnik · source record · preprint

Microscopy assessment of two multiple-transient candidates

Yang, J.; Villarroel, B. · 2026

Verified — held in project librarySRC-YANG-VILLARROEL-2026

identifier / locator: arXiv:2608.26371 (astro-ph.IM), posted 2026-08-26; MNRAS submission format
url: https://arxiv.org/abs/2608.26371
verification note: Full text read by the editor's agent on 2026-08-24 from a copy supplied by the authors' group ahead of publication (inbox drop, runId trn-microscopy-2026-08-24), when no public version existed. The public preprint appeared on arXiv on 2026-08-26 (arXiv:2608.26371), was surfaced by the 2026-08-31 literature watch, and was correctly matched to this record by the duplicate guard rather than imported twice. On 2026-09-01 the identifier was updated as this note pre-registered, and every quoted passage in TRN-E022, TRN-E023 and TRN-E024 was re-checked verbatim against the arXiv text: all present. Citation no longer depends on supplied provenance; the permission history is retained for the record: citation permission was granted by Beatriz Villarroel (second author; VASCO project) by email to the founder ahead of publication, recorded here 2026-08-25; the email is held by the founder and was reported by him directly to the maintainer — the same supplied-provenance standard as IMG-TRN-P03 (who granted, what channel, held where).
reliability notes: Preprint; not yet peer reviewed — figures may change before journal publication.
Second author is the hypothesis's principal proponent; first author (Harvard) is new to this dispute. The manuscript acknowledges Caltech and Harvard plate-archive staff and Avi Loeb.
Notably candid against its own side's interest: it quotes its microscopist's opinion that emulsion defects are a reasonable attribution, states plainly that microscopy cannot classify individual sources as flash versus defect, and reports the multi-observatory parallax test infeasible for these two events.
The Caltech inspection was not a proponents-only affair: the manuscript's own footnote records that the microscopist was directed to the nine objects by Mr. Hambly — the copy-defect argument's principal author.
The examination was not blinded, and no per-generation first-appearance census across the copy chain was performed; those parts of TRN-R001 remain open.

Evidence drawn from this source

Supports· moderateTRN-E022

Yang & Villarroel manuscript: all nine 1950 sources are on the original glass, and the 1952 triple is not confined to copy material

what the source states: Microscopy of the original April 1950 glass plate at the Caltech archive, registered against the published digitization using field stars as fiducials, finds all nine catalogued objects present on the physical emulsion at their published positions, "ruling out a digitization-artifact origin." For the July 1952 triple, a Harvard copy plate was examined and the candidate is present both on the copy plate and in the original survey digitization at the catalogued positions. The features in both cases are point-like, marginally sharper than the stellar images, with sharp edges and no elongation.
what we infer: The first examination of original plate material in this dispute, and it settles the lowest rung more firmly than any prior record: the marks are physical features of the emulsion that was in (or printed directly from what was in) the telescope, not artifacts of scanning or software. Moderate rather than strong solely because of the document's status — unpublished, unreviewed, and unrepeatable by outsiders until the images or the plates are generally accessible.
limitations: Preprint (arXiv:2608.26371, posted 2026-08-26); not yet peer reviewed — figures and claims may change before journal publication.
The 1952 examination was of a copy plate; presence on that event's original negative is inferred from co-presence in the original survey digitization, not observed directly.
Registration and inspection were performed by the authors' group; not blinded.

Yang, J. (2026)Verified — held in project libraryTRN-C001 TRN-C003 TRN-C024

Qualifies· moderateTRN-E023

Yang & Villarroel manuscript: the copy-stage defect mechanism cannot explain the two canonical events

what the source states: Because the nine 1950 sources are present on the original negative, and the 1952 triple appears in the original survey digitization and not only on copy material, the manuscript concludes this "disfavours the copy-plate-artefact hypothesis of Hambly & Blair (2024) for these specific sources, since the features are not confined to the copy material." It frames registration against the original digitization as "the actual discriminant" for that hypothesis: a feature intrinsic to the original exposure appears there; a feature born in copying would appear only in copy material.
what we infer: Qualifies rather than undermines, deliberately: TRN-C007's genealogy — the two digitizations really do descend from a shared photographic ancestor — remains true and remains a live concern for the ~100,000 uninspected catalog candidates. What this record removes is that argument's reach over the two flagship events. For those, "it's in two cousins" is answered with "it's on the grandparent." The claim's diagnostic weight against the canonical events, which was its sharpest use, does not survive if this manuscript's registration holds up in review.
limitations: Preprint (arXiv:2608.26371); not yet peer reviewed — the registration analysis is the load-bearing step and has not been independently checked.
Applies to the two inspected events only; the copy-defect base-rate question for the full catalog (TRN-R007) is untouched.

Yang, J. (2026)Verified — held in project libraryTRN-C007

Qualifies· moderateTRN-E024

Yang & Villarroel manuscript: microscopy cannot tell a flash from an emulsion defect, and its own microscopist leaned defect

what the source states: The manuscript states that "microscopy of individual point-like features cannot by itself separate a short-duration flash from an intrinsic emulsion defect, because the two scenarios produce qualitatively similar signatures at the single-source level." Its footnote quotes the microscopist who inspected the Caltech plate — directed to the nine objects by Mr. Hambly: the features "were not stellar in appearance," with sharp edges and no elongation, and "in my experience, it is reasonable to attribute these anomalies to emulsion defects," noting the field lies in a corner of the plate where such defects are more common. The manuscript argues this reading reflects a natural bias — comparison against steady stars favours the defect interpretation over sub-second emission — and reports that the multi-observatory parallax test is infeasible for these two events, since no simultaneous second-site plate of comparable depth exists. It concludes that resolving the events' nature requires population-level statistics and, for future events, purpose-built multi-station imaging.
what we infer: The honest ceiling of the decisive test, reported by the test's own authors. The case had framed the original-plate examination as the observation that could collapse the dispute in either direction; this record establishes that it cannot — the crux (TRN-C020) narrows from "original or copy?" to "flash or flaw in the original emulsion?", and single-source morphology cannot answer that. The microscopist's defect-leaning opinion is the closest thing this case has to an independent eye on the physical evidence, and it is recorded at face value alongside the manuscript's argument that it embeds a steady-star default. Discriminating power now rests with the coma signature (TRN-C011, TRN-R003), rebuilt population statistics, and prospective multi-station imaging (TRN-R009).
limitations: Preprint (arXiv:2608.26371); not yet peer reviewed.
The microscopist's assessment is quoted from private correspondence reproduced in the manuscript's footnote; the microscopist is not named in the text read.
The plate-corner location cuts both ways and the manuscript says so: defects congregate in corners, and the reading was made without considering transient phenomena.

Yang, J. (2026)Verified — held in project libraryTRN-C020