REVIEW 2 major objections 5 minor 3 cited by
Proofs Of Three Geode Conjectures
T0 review · 2 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read This paper proves all three Geode conjectures from the 2025 Monthly article, giving closed formulas for the two-index Geode numbers and a simple geometric-series evaluation of the alternating Geode series.
desk verdict First two theorems are solid and checkable; the third has a real but fixable divisibility gap. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the Geode series $GGG[t_1,t_2,\ldots]$, defined as the second factor in $\SSS-1=(t_1+t_2+\cdots)GGG$, where $\SSS$ is the formal power series solution of $0=1-\alpha+\sum_{k\ge1} t_k \alpha^{k+1}$. The proofs are carried by four standard operations: multinomial expansion, constant-term extraction, the Lagrange inversion coefficient formula, and rational-function telescoping identities. In the telescoping step, the paper exhibits an explicit difference equation $F(n,k)=H(n,k+1)-H(n,k)$ for each summand, which collapses the alternating sums to boundary terms and yields the closed coefficients.
What would settle it
For a small case such as $a=2$, $n=3$, compute the polynomial in $t_{2a}$ obtained by substituting $t_1=t_2=t_3=f$ into the coefficient expression for $GGG$, and evaluate it at $t_{2a}=f$; a nonzero value would invalidate the division step in the proof of Theorem 3.1. Alternatively, compute the first coefficients of $GGG[-f,f,\ldots,-f,f]$ directly from the defining equation $0=1-\alpha+\sum_{k\ge1} t_k \alpha^{k+1}$ and compare them with $a^n f^n$.
Extended reading notes
Core claim
The central discovery is that the Geode array, which is extracted from the factorization $\SSS-1=(t_1+t_2+\cdots)GGG$ of the hyper-Catalan series $\SSS$, is governed by explicit closed formulas rather than only by a generating series. Theorem 1.1 states that $$G[m_1,m_2]=\frac{(2m_1+3m_2+3)!}{(2m_1+2m_2+3)(m_1+m_2+1)(m_1+2m_2+2)!\,m_1!\,m_2!},$$ and Theorem 2.1 gives the analogous formula for $\tilde G[m_a,m_{a+1}]$ when the two nonzero indices are adjacent. Theorem 3.1 shows that the $2a$-variate Geode series evaluated at $(-f,f,\ldots,-f,f)$ collapses to the ordinary geometric series $\sum_n a^n f^n$. These are exact identities, not asymptotics. The proof route is coefficient extraction: Lagrange inversion turns the defining equation into multinomial sums, and telescoping certificates reduce those sums to single binomial or geometric terms.
Load-bearing premise
In the proof of Theorem 3.1, after $t_1,\ldots,t_{2a-1}$ are set to $f$, the resulting polynomial in $t_{2a}$ is divided by $t_{2a}-f$; the paper does not prove that this division is exact, yet the identity is then extracted from the quotient.
Editorial extensions
If this is right
- Every two-index Geode number $G[m_1,m_2]$ can now be computed directly from the closed formula, without first building the hyper-Catalan series.
- The formula for $\tilde G[m_a,m_{a+1}]$ shows that the one-index slices of the Geode array are two-parameter Fuss-Catalan-type numbers, as the proposers expected.
- The alternating specialization result gives a whole one-parameter family of simple geometric evaluations of the Geode series, indexed by the number of paired variables.
- The closed formulas imply the relation $G[m_1-1,m_2]+G[m_1,m_2-1]=C[m_1,m_2]$, directly connecting Geode numbers to the hyper-Catalan coefficients.
Reading between the lines
- The same coefficient-extraction scheme should yield closed formulas for Geode numbers with two nonzero indices in arbitrary positions, not only adjacent ones; the sparse formula in Remark 3.4 is an explicit prediction that could be checked on small indices.
- Remark 3.5 suggests a weighted alternating specialization; if the pattern holds, the Geode series should evaluate to $\sum_n (2a c_a-c_1-\cdots-c_a)^n f^n$ for arbitrary weights $c_i$.
- Because the Geode series factorizes the hyper-Catalan series, the evaluation $\sum_n a^n f^n$ may correspond to a recognizable branch of the original polynomial equation, which could yield a more direct combinatorial proof.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proves three conjectures posed by Wildberger and Rubine in their 2025 Monthly article on the Geode array. Theorems 1.1 and 2.1 give closed forms for the coefficients of the two-variable Geode series and of the Geode series with two adjacent nonzero variables, respectively. Theorem 3.1 evaluates the 2a-variate Geode series along the alternating line GGG[-f, f, ..., -f, f], claiming the result sum_n a^n f^n. The proofs use Lagrange inversion, WZ certificates, and constant-term extraction. The first two proofs are essentially complete and checkable; the third proof contains an unproved divisibility step and an apparent sign inconsistency, both of which appear repairable.
Significance. The paper would resolve all three conjectures from [2, p. 399] and demonstrates a clean combination of Lagrange inversion, WZ certificates, and constant-term extraction. Its strengths include fully explicit WZ certificates, a parameter-free derivation from the defining equation, and results that can be verified numerically term by term. Theorems 1.1 and 2.1 are proven rigorously as written. Theorem 3.1 is likely true and the overall strategy is convincing, but the proof as written has a load-bearing gap that must be fixed before the claim is established.
major comments (2)
- [Section 3, divide-out step] After substituting t1=...=t_{2a-1}=f, the proof divides the current polynomial in t_{2a} by the linear factor t_{2a}-f and then evaluates at f. This operation is valid only if the polynomial vanishes at t_{2a}=f. The manuscript states only 'divide out' and gives no justification. This is load-bearing because the quotient-and-evaluate step is exactly what converts the alpha-series coefficients into the Geode coefficients leading to (3.1). The divisibility is true: the factorization SSS-1=(t1+...+t_{2a})GGG implies that the numerator is divisible by the alternating sum, and after the specialization the denominator becomes t_{2a}-f. The authors should state this explicitly, either from the cited factorization or by providing a direct WZ certificate for the vanishing at t_{2a}=f.
- [Section 3, sign in Lagrange expansion] The first displayed expansion after writing (-1)^i t_i instead of the plain t_i has the sign (-1)^{m_1+...+m_{2a-1}}. If the relation to the original variables is (-1)^i t_i, then the monomial t_i^{m_i} carries the sign (-1)^{i m_i}, so the total exponent should be m_1+2m_2+...+(2a-1)m_{2a-1}. The later substitution line correctly reduces to (-1)^{m_1+m_3+...+m_{2a-1}}, which is the parity reduction of the weighted sign. As written, the first and second displays are inconsistent, and the derivation of (3.2)-(3.3) depends on the corrected sign. This needs correction, not just clarification.
minor comments (5)
- [Section 3, linear factor display] The displayed linear factor '-t1+t2-...-t_{2a-3}+t_{2a-1}-t_{2a-1}+t_{2a}' contains a duplicated t_{2a-1} and appears to be a typo; it should be the alternating sum -t1+t2-...-t_{2a-1}+t_{2a}, which after the substitution t1=...=t_{2a-1}=f equals t_{2a}-f.
- [Section 2, display] In the displayed formula for the two-adjacent-variable case, the bottom multinomial entry '1+(a-1)n+m3' should read with m_{a+1}, not m3.
- [Sections 1 and 2, WZ telescoping] When the telescoping identity F(n,k)=H(n,k+1)-H(n,k) is used, the boundary facts H(n,0)=0 and H(n,n+1)=0 are used implicitly; they should be stated explicitly, since they carry the summation.
- [Section 3, transition to (3.1)] The passage from the quotient-and-evaluate expression, involving a second sum over i and m_{2a}<=i, to the compact identity (3.1) is not fully derived; in particular the appearance of the factor (n-m_{2a}) should be explained.
- [Abstract] The abstract contains a typo: 'intoduced' should be 'introduced'.
Circularity Check
No circularity: the conjectures are derived from Lagrange inversion, WZ certificates, and multinomial/constant-term manipulations, not from the conjectured formulas.
full rationale
The paper's derivation chain is self-contained. The Geode coefficients G[m1,m2] are defined through the factorization SSS-1 = SSS_1 GGG cited from the external paper [2], and the proofs compute these coefficients by applying Lagrange inversion to the defining equation of alpha and then dividing by t1+t2 (or by t_{2a}-f in Section 3). In Sections 1 and 2 the required divisibility is explicitly proved via WZ certificates F(n,k)=H(n,k+1)-H(n,k), and the quotient coefficients are evaluated by telescoping, yielding Theorems 1.1 and 2.1 algebraically. No fitted parameter is renamed as a prediction, and no conjecture is used as an input. Theorem 3.1 follows from the same Lagrange expansion, with Claims 1 and 2 proved by multinomial and constant-term identities. The only noticeable gap is the asserted 'divide out' by t_{2a}-f in Section 3, which is not separately certified; this is a fillable omitted justification (vanishing follows from the defining equation beta=1), but a missing proof is not circularity. The self-citations [3] and [4] point to standard WZ and Lagrange-inversion methods and are not load-bearing self-support. Therefore there is no circular step; score 0.
Assumptions & free parameters
assumptions (4)
- standard math Multinomial theorem (1.1)
- standard math Lagrange inversion (1.2) for formal power series
- standard math Wilf-Zeilberger method certifies the binomial-sum identities
- domain assumption The Wildberger-Rubine factorization SSS - 1 = SSS_1 GGG and the definition of Geode coefficients [2, Theorem 12]
Cite this review
Pith. "Pith review of Proofs Of Three Geode Conjectures." pith.science (2026). https://pith.science/paper/PM3UAVUB
@misc{pith2026250617862,
author = {Pith},
title = {Pith review of: Proofs Of Three Geode Conjectures},
year = {2026},
howpublished = {\url{https://pith.science/paper/PM3UAVUB}},
note = {Machine review of arXiv:2506.17862}
}
read the original abstract
In the May 2025 issue of the Amer. Math. Monthly, Norman J. Wildberger and Dean Rubine intoduced a new kind of multi-indexed numbers, that they call `Geode numbers', obtained from the Hyper-Catalan numbers. They posed three intriguing conjectures about them, that are proved in this note.
Forward citations
Cited by 3 Pith papers
-
Ordered trees and the Geode
The coefficients of the Geode power series count leaves that appear before any internal node in the post-order traversal of ordered trees.
-
Lattice paths and the Geode
The Geode G equals (1 - sum_{n>=1} t_n (1+S+...+S^{n-1}))^{-1} and counts nonnegative lattice paths with steps -1,0,1,2,... under a natural weight.
-
Hyper-Catalan and Geode Recurrences and Three Conjectures of Wildberger
The Geode coefficients satisfy a recurrence, proved from the hyper-Catalan generating function, that yields closed forms for two consecutive polygon shapes and thereby resolves three conjectures.
Reference graph
Works this paper leans on
-
[1]
S. R. Mane, Multiparameter Fuss–Catalan numbers with application to algebraic equations, https://arxiv.org/pdf/1607.04144
-
[2]
N. J. Wildberger, D.Rubine, A Hyper-Catalan Series Solution to Polynomial Equa- tions, and the Geode , Amer. Math. Monthly, 132, (2025), no. 5, 383-402. https: //www.tandfonline.com/doi/full/10.1080/00029890.2025.2460966
arXiv 2025
-
[3]
H. S. Wilf, D. Zeilberger, Rational functions certify combinatorial identities, J. Amer. Math. Soc. 3 (1990), no. 1, 147-158. https://www.ams.org/journals/jams/ 1990-03-01/S0894-0347-1990-1007910-7/S0894-0347-1990-1007910-7. pdf
work page 1990
-
[4]
D. Zeilberger, Lagrange Inversion Without Tears (Analysis) (based on Henrici) , https://sites.math.rutgers.edu/~zeilberg/mamarim/mamarimPDF/lag. pdf Department of Mathematics, Tulane University, New Orleans, LA 70118, USA Email address: tamdeber@tulane.edu Rutgers University, Department of Mathematics, 110 F relinghuysen Rd, Piscataw ay, NJ 08854, USA Ema...
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.