Complete proposed proof, 21 November 2026. The complete written argument passed independent internal review. The priority audit records the sources checked and access limits. This work remains unrefereed. Coordination: issue 168.
1. Statement
For a nonempty integer partition , let its perimeter be its largest part plus its number of parts minus one. Let count partitions of perimeter having exactly distinct part sizes of multiplicity at least . Let count partitions of perimeter having exactly distinct occurring part sizes divisible by .
Marcus McCrea, Fixed-Perimeter Franklin Statistics and an Eventual Inequality, arXiv:2609.12005v2, dated 5 November 2026, Section 9, Conjecture 9.2, proposes the following statement. The full primary PDF was inspected.
Theorem. Fix an integer , and put
There exists a unique such that, for every fixed , the following inequalities hold for all sufficiently large integers :
This is the full conjecture. It does not prescribe the comparison at or outside the interval between the two means. The previously proved fixed- inequality is a different statement and is not counted here.
2. Positive generating functions
Fix throughout, and write
The source's Proposition 3.1, equation (7), gives, for ,
where
For the other statistic, put
The source's Proposition 3.2, equation (9), gives
where
All four series have nonnegative coefficients. Both series have positive coefficients at every power with , and no smaller powers. For this follows from the factor and the expansion of . For , the numerator already has this support, and has nonnegative coefficients and constant term one.
There is a unique with and a unique with . Positivity of and makes these zeros simple. Each is analytic on and has a simple pole at the positive real point . Each is analytic on the same open disk. Indeed, for one has , and the analogous inequality holds for on . Their remaining possible poles occur at one, beyond the stated radii.
3. A coefficient lemma, including the floor
We use an elementary large-power coefficient estimate and give its proof to make the local, rather than merely cumulative, comparison explicit.
Lemma. Let
have positive radius of convergence , and suppose has a simple pole at . Let be a nonzero power series with nonnegative coefficients, analytic on . For , there exists a unique such that
Set . Then, as ,
In particular, these coefficients are positive for every sufficiently large , and
Proof. For , define a random integer by
Analyticity on a disk larger than gives finite exponential moments. Direct differentiation shows
The variance is strictly positive because the support contains consecutive integers. Thus is strictly increasing. Its limit at zero is . Its limit as is infinity, because a simple positive pole gives
This proves existence and uniqueness in (5), and positivity of .
Fix that . If , let be an independent nonnegative integer with probabilities . Multiplying the probability generating functions gives the exact identity
where the are independent copies of .
Here is the local estimate needed for (8). Write , , and let , . Fourier inversion for integer-valued random variables yields
The consecutive support implies for : equality in the triangle inequality would force to have the same value for all consecutive , hence . Continuity then gives a uniform bound strictly below one on every compact region away from zero. At zero, finite exponential moments give
In particular, for some and sufficiently small . Also and as .
For ,
In the integral (9), put near zero. The integrand becomes
which tends to for each fixed . The bound permits dominated convergence on the expanding neighborhood; the remaining part of (9) is exponentially small by the uniform bound away from zero. Thus
This proves (6). Taking logarithms, using , proves (7). ∎
4. The two rate functions
Apply the lemma to (1) and (3), with . Both pairs satisfy all its assumptions. We obtain differentiable functions on such that
For either function, differentiation of (7) and use of (5) cancel the terms containing :
Since , differentiating gives
Consequently each rate is strictly concave. Direct substitution gives
For clarity, the first logarithmic derivative is
and , give the first mean in (14). On the other side, , , , and
give .
By (12)–(14), has its unique maximum at , while has its unique maximum at . The two maximum values coincide:
Both means belong to the lemma's interval. Also for : holds at and persists by induction. Finally, because .
5. Existence and uniqueness of the crossover
On , set . Strict maximality and (15) imply
For , equation (13) makes the saddle smaller than and the saddle larger than . Each is strictly increasing, so (12) yields
Continuity, (16), and (17) give exactly one zero . This provides an implicit exact definition of the transition density, without a numerical approximation.
For fixed in that interval, (11) gives
Both counts are eventually positive by the coefficient lemma. If , the limit is positive, so the ratio tends to infinity. If , the limit is negative, so the ratio tends to zero. This proves the two eventual strict inequalities for every allowed . Any other proposed transition density would give the opposite inequality at a point strictly between the two densities, so the density in the conjecture is unique. ∎
Scope and verification
The proof covers every integer and every fixed real density in the exact interval specified by the conjecture, apart from its expressly excluded crossover. The convergence threshold may depend on , as permitted by the source. It does not claim a single finite- sign change over all or a uniform threshold in . No numerical root or finite screen is a premise of the proof.
The positive generating-function rewrites, support, pole, mean, and floor conditions are all explicit. The coefficient lemma is proved directly rather than inferred from a large-deviation bound on cumulative probabilities. Source status and the priority audit are recorded separately.