McCrea's conjecture: source and priority audit

What Lean checks

Complete proposed proof of the unique linear-density crossover for every modulus at least three. Positive coefficient asymptotics handle the floor, and the exponent difference has exactly one zero in the stated interval. Independent internal review and exact profile-word diagnostics passed.

Audit date: 21 November 2026. Coordination: issue 168.

Exact target

Marcus McCrea, Fixed-Perimeter Franklin Statistics and an Eventual Inequality, arXiv:2609.12005v2, dated 5 November 2026 in the manuscript margin, Section 9, Conjecture 9.2, printed page 12. The full primary PDF was read, including the definitions, both bivariate generating functions, coefficient formulas, mean calculations, conjecture, and following discussion.

The conjecture asks for a unique asymptotic sign transition at a fixed linear density between the two limiting means, for every integer modulus at least three. It expressly excludes the crossover point and densities outside that open interval. The result is a single full statement across all moduli; no numerical modulus or density is counted separately.

The source's text calls uniqueness open after describing continuity and large-deviation heuristics for existence. Our proof supplies both the local coefficient asymptotics and uniqueness. The latter follows because, between their distinct maximizers, one strictly concave exponent decreases and the other increases. We do not claim a new general large-deviation method: the Fourier argument and exponential tilting are classical and are written out to discharge the exact coefficient and floor requirements.

Prior results separated from the target

  1. Gray, Payne, Swisher and Watson, Fixed perimeter analogues of some partition results, arXiv:2502.12394v1, published in Discrete Mathematics 349 (2026), 114968, DOI. Its Conjecture 1.2 keeps the number of marked part sizes fixed. That statement is already proved and is not claimed here.

  2. McCrea's original September 2026 version and November revision prove that fixed-number statement. The November revision explicitly adds the different linear-density conjecture. The author/paper record and SciRate author record were checked. Versioned dates in the actual PDF take precedence over index pages that attach the original September date to a later revision.

  3. Philip Cuthbertson, A perimeter analogue of Franklin's identity and an inequality related to the parity of parts, arXiv:2609.17389, September 2026. The primary abstract describes the fixed-number Franklin inequality and a parity-bias inequality. McCrea's later full November text expressly acknowledges that independent fixed-number proof in its introduction before posing Conjecture 9.2 in Section 9. The Cuthbertson full PDF could not be retrieved in this environment. Its existence does not supply the linear-density statement; this distinction is supported by McCrea's later primary text, not merely by an abstract omission.

  4. Pankaj Jyoti Mahanta, A proof of a conjecture on fixed perimeter partitions, arXiv:2608.11684. The inspected primary abstract addresses parity bias for distinct-part partitions, a different comparison. It does not state the Franklin-statistic crossover.

  5. The same Gray–Payne–Swisher–Watson paper's separate Conjecture 1.5, a modulus tipping point for two residue classes, was also screened. Its MathDB page already contains a full positive-pole proof posted by Shivam Patel before this run. That conjecture is excluded, even though the page's older status summary still says open. It is not McCrea's new linear-density statement.

Search coverage

Both available web search engines were used, with exact source and recent-date searches in addition to broad discovery. Queries included:

"McCrea" "linear-density" conjecture
"2609.12005" "conjecture" "crossover"
"McCrea" "crossover" proof conjecture
"linear-density single-crossover" proof
"2609.12005" "9.2" proof
"Franklin" "transition density" partitions proof
"McCrea" "Conjecture 9.2"
"Franklin" "single-crossover" proof
"linear density" "Franklin" partitions crossover
"2609.12005" proof counterexample crossover
"Gray" "Payne" "Swisher" "Watson" conjecture proof
"fixed-perimeter" "Franklin" conjecture

The independent analysis reviewer separately searched both engines for the author, exact conjecture phrase, full paper title, and coefficient/crossover terminology. The pertinent results returned the original November conjecture and the earlier fixed-number proofs. No prior proof, refutation, withdrawal, or competing crossover result was located.

Review and limits

The analysis reviewer checked the exact source statement and formulas, positive support and simple poles, the fixed-saddle Fourier argument, the bounded displacement caused by the floor, both mean values, rate differentiation, and the endpoint comparison. The complete written proof passed without a mathematical correction; one differential-formatting typo was corrected.

The result remains unrefereed. The priority conclusion is bounded by the sources and indexing available on the audit date. The novelty claim concerns the complete conjecture, not the classical coefficient-estimation tools. An earlier full resolution, if found, would supersede the priority claim.