Patterns Within the Markov Tree

Robert A. Gore

REVIEW 3 major objections 4 minor 3 references

Patterns Within the Markov Tree

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper claims that the largest entry of a Markov triplet uniquely determines the other two entries, and that this follows from a Pell equation attached to each Markov region.

desk verdict A serious pattern-driven exploration of the Markov tree with some correct elementary algebra, but the claimed proof of the Frobenius Uniqueness Conjecture rests on an unproved completeness assertion and is circular at the critical step. read the letter →

arxiv 2506.04299 v2 pith:OUKKJKAH submitted 2025-06-04 math.GM

Robert A. Gore This is my paper

classification math.GM MSC 11D2511D0911B3911B50

keywords MarkovtripletsFrobeniusUniquenessConjecturePellequationLucassequencestreeregionnumberspalindromicrepeatcyclessumoftwosquares

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims to settle the Frobenius Uniqueness Conjecture: for every positive integer solution $\{x,R,z\}$ of $x^2+R^2+z^2=3xRz$, the largest entry $R$ determines the two smaller entries $x$ and $z$. The route goes through the Markov tree: each region $R$ has edge triplets whose region numbers obey a Lucas-sequence recurrence with kernel $\{3R,-1\}$, and those sequences satisfy the Pell equation $V^2 - ((3R)^2-4)U^2 = -(2R)^2$. The paper argues that the Pell equation has solutions exactly when $R$ is a Markov number, and that the two smallest companion values are precisely $x$ and $z$, making the triplet unique. The same machinery also produces palindromic last-digit cycles along region edges and a self-referential algorithm that writes every Markov number as a sum of two special squares.

What carries the argument

The object that carries the argument is the Lucas pair $U_k(3R,1)=\sinh(k\theta)/\sinh\theta$ and $V_k(3R,1)=2\cosh(k\theta)$ with $\cosh\theta=3R/2$, together with the edge sequence functions $H_{sf}[\{x,R,z\},n]$ built from them. These functions reduce every edge of a Markov region to a constant-coefficient recurrence with kernel $\{3R,-1\}$, give generating functions for all triplets in the region, and produce the Pell equation whose solution structure is the bridge from $R$ back to $x$ and $z$.

What would settle it

Search the Pell equation $K^2-D(R)J^2=-(2R)^2$ for a Markov region number $R$ and find a positive solution $J$ that is smaller than both $x$ and $z$ but equal to neither, or find any positive solution for a non-Markov integer $R$; either would contradict the paper's characterization.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that every Markov region $\{x,R,z\}$ carries a pair of Lucas sequences $U_k(3R,1)$ and $V_k(3R,1)$ from which the region numbers of all left- and right-edge triplets can be written as $U[\{x,R,z\},n+1]=zU_{n+1}(3R,1)-xU_n(3R,1)$ and $V[\{x,R,z\},n+1]=zV_{n+1}(3R,1)-xV_n(3R,1)$. These satisfy $V[\{x,R,z\},n+1]^2 - D(R) U[\{x,R,z\},n+1]^2 = -(2R)^2$ with $D(R)=(3R)^2-4$. The paper asserts that the full solution set of this Pell equation is generated from the baseline solution $\{V[\{x,R,z\},1], U[\{x,R,z\},1]\}$ by powers of $(3R+\sqrt{D(R)})/2$, so the displayed sequences are the only solutions. Consequently, for a fixed $R$, the two smallest positive solutions $J$ of $K^2-D(R)J^2=-(2R)^2$ are the two entries $x$ and $z$ that complete a Markov triplet with $R$; hence the region number $R$ uniquely determines the triplet and the Frobenius Uniqueness Conjecture is true.

Load-bearing premise

The proof depends on the assertion that every integer solution of $X^2-D(R)Y^2=-(2R)^2$ is obtained from the single baseline solution by multiplying by powers of $(3R+\sqrt{D(R)})/2$, and therefore that no other solution orbit exists.

Editorial extensions

If this is right

If the proof is correct, the Frobenius Uniqueness Conjecture becomes a theorem: each non-singular Markov triplet is uniquely identified by its largest entry.
The Pell equation supplies a direct computational criterion for Markov numbers: an integer $R$ is a Markov region number exactly when $K^2-((3R)^2-4)J^2=-(2R)^2$ has an integer solution.
The sequence functions produce generating functions for every triplet in every region, so all Markov triplets at any depth can be listed without recursive tree traversal.
For each Markov number $R$, the sibling-recursion algorithm yields a canonical pair $\{\sigma,\Lambda\}$ with $\sigma^2+\Lambda^2=R$, and along each edge these pairs satisfy the same recurrence kernel $\{3R,-1\}$ as the region numbers.
The last-digit tables imply periodic palindromic structure for edge region numbers and square terms, with cycle lengths determined by $R$ modulo 20.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A non-Markov $R$ should have no solution of $K^2-((3R)^2-4)J^2=-(2R)^2$; checking this for a dense range of non-Markov integers would be a fast computational stress test of the paper's main equivalence.
The recursive square-term algorithm is essentially a descent through the Markov tree via the sibling number $s=3xz-R<R$; it could be turned into an algorithm that computes the canonical square decomposition of a Markov number in time proportional to its depth.
The observed palindromic cycles built from odd-indexed Fibonacci digits suggest that each region's edge labels inherit modular rotations from a single Fibonacci-type sequence; a testable extension is whether every region's repeat cycle is a cyclic rotation of the Fibonacci cycle rather than an independent pattern.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

You should know this paper before spending time on it: the pattern observations are genuine and sometimes pretty, but the headline result—a proof of the Frobenius Uniqueness Conjecture—does not hold up. The gap is not a minor technicality; it is the load-bearing step.

What the paper does well: the derivation of recurrence kernels and sequence functions for region numbers along edges of an arbitrary Markov region is correct as far as I checked. The generating functions, the Lucas-sequence representations, and the observation that negative indices interchange left and right edges are all clean and useful. The digit-cycle and palindromic data are carefully tabulated, and the connection to odd-indexed Fibonacci numbers in certain palindromes is a nice computational find. The special-square algorithm (Q) is an interesting construction, and the claim that the square terms also satisfy the {3R,-1} recurrence is plausible and likely correct.

The soft spot is exactly where the Uniqueness Conjecture is supposed to be proved. In Section 2 the paper asserts that all solutions to X^2 - D(R)Y^2 = -(2R)^2 are generated from one baseline solution by powers of (1/2)(3R + sqrt(D(R))). That requires the unit to be fundamental and the relevant ideal classes to collapse to one orbit. For a fixed R, the norm equation in the order of Q(sqrt(D(R))) can have finitely many orbits, and the paper gives no argument that only one occurs. The phrase "one can prove" appears without proof, and "these are the only solutions" is simply stated. The Mathematica table is finite and, as the paper itself admits near Table 2, could have been produced by sorting the paper's own sequence values—so it is not independent evidence. Worse, the conclusion that the two smallest solutions are x and z is essentially a restatement of the Uniqueness Conjecture, because the Pell equation is just the Markov equation rearranged. The reasoning is circular at the decisive point.

There are also smaller issues: the RegionSign exceptions and the self-referential startup values for algorithm Q are not fully justified, and the paper cites only a Quanta article, which is fine for inspiration but not for locating the conjecture in the literature. These are fixable in principle; the missing completeness proof is not.

My recommendation: this should not be published as a resolution of the conjecture. A serious referee would spot the gap quickly, so I would not desk-reject it out of hand—it is important enough that the record should be clear—but the verdict would be Reject. If the conjecture claim were removed, the remaining pattern sections might support a modest exploratory note, and the sequence-function framework could be worth citing for those who work on Markov tree parametrizations. As it stands, I would not cite it for any result, only perhaps as an example of how computational pattern-hunting can create the illusion of a proof.

Referee Report

3 major / 4 minor

Summary. The paper studies the Markov tree and its regions, introducing Lucas-type sequence functions for edge triplets. It derives a Pell equation V[{x,R,z},n+1]^2 - D(R) U[{x,R,z},n+1]^2 = -(2R)^2, with D(R)=(3R)^2-4, and uses this equation to claim a proof of the Frobenius Uniqueness Conjecture. The remainder of the paper catalogs palindromic last-digit cycles for region numbers and for special square decompositions, gives an algorithm for the two square terms of each Markov number, and connects Markov regions to Farey fractions.

Significance. If the main claim were established, the paper would resolve a longstanding open problem, so the potential significance is very high. The paper contains concrete, reproducible observations: the Pell identity itself is verifiable, the sequence parametrizations of edge triplets are explicit, and the cycle tables and palindromic patterns are presented with enough data to be checked. However, the claimed proof of uniqueness rests on a single unproved completeness assertion for the generalized Pell equation, and the finite Mathematica table is not an independent verification. As it stands, the paper is an extensive collection of well-documented patterns rather than a proof of the Uniqueness Conjecture.

major comments (3)

[Section 2] The step labeled 'one can prove that V and U are the only solutions' is the entire substance of the uniqueness proof, but no proof is given. The baseline solution and the unit (3R+sqrt(D(R)))/2 demonstrate that the sequences V and U lie on one orbit of the Pell equation K^2 - D(R)J^2 = -(2R)^2; they do not show that this orbit exhausts all solutions. A generalized Pell equation can have multiple solution classes modulo the unit group, and the paper supplies no fundamental-unit or class-number argument ruling this out. The subsequent conclusion that the two smallest positive J are always x and z, and therefore that R determines the pair {x,z}, is valid only if this exhaustiveness statement is true. This is the load-bearing gap in the claimed proof of the Frobenius Uniqueness Conjecture.
[Section 2, Table 2] The table and the surrounding statement that solutions exist 'if and only if the parameter R is equal to a Markov number' are not a proof. The table lists only Markov parameter values, and the text itself concedes that the table could have been created by sorting values of U[{x,R,z},n+1]. Consequently the table provides no independent evidence for exhaustiveness or for the converse direction, and it cannot certify a statement about all R. The uniqueness conclusion therefore depends entirely on the unproved completeness claim.
[Section 4.2] The algorithm Q[{x,R,z}] is asserted to always find one and only one special square pair for every triplet, but no proof of existence, uniqueness, or termination is supplied. The recursion relies on a precomputed MarkovList, on three startup values, and on an ad hoc RegionSign function with two exceptional cases; no argument shows that these ingredients determine a well-defined map on all triplets. This is a separate unsupported claim in the paper's second main thread, although it does not bear directly on the Uniqueness Conjecture.

minor comments (4)

[Section 1] The notation for the several sequence objects (U_n, V_n, U[{...}], V[{...}], Hsf, Ksf) is introduced informally; a consolidated notation table would substantially improve readability.
[Section 3.3] The statement that appending the left and right edge sequences 'clearly' creates a palindromic list is an observed pattern, not a proof for all regions; a short congruence argument using U_{-n} and V_{-n} would make the claim rigorous.
[Section 5.3] The claim that the ratio of the upper to lower oscillation bounds 'analytically converges' to a region-dependent limit is not demonstrated, and the stated Fibonacci and Pell limiting values are given without derivation.
[References] The reference list contains only a Quanta Magazine article; standard references on Markov's theorem, the Frobenius uniqueness conjecture, Lucas sequences, and generalized Pell equations should be included.

Circularity Check

2 steps flagged · score 8.0 of 10

The purported proof of the Frobenius Uniqueness Conjecture rests on an unproved assertion that one Lucas-sequence orbit exhausts the Pell equation's solutions; the paper's own table is conceded to be a sorting of that same sequence.

self definitional [Section 2, from 'Now, one can prove...' through '...true if and only if...']
"Now, one can prove that V [{x, R, z}, n+ 1] and U [{x, R, z}, n+ 1] are the only solutions to this Pell equation by choosing a convenient baseline solution ... Since the results, {Xm, Ym}, have the same recurrence kernels, generating functions, and values as {V [{x, R, z}, m+ 1], U[{x, R, z}, m+ 1]} respectively, these are the only solutions: K2 − D(R) J2 = −(2R)2 is true if and only if K = V [{x, R, z}, n+ 1] and J = U [{x, R, z}, n+ 1]."

The orbit {V_m, U_m} is generated from a baseline solution defined by one particular Markov triplet {x,R,z}. The claim that this single orbit contains every integer solution of X^2 − D(R)Y^2 = −(2R)^2 is a nontrivial completeness statement about the unit group and ideal classes of the order Z[sqrt(D(R))]; the paper supplies no proof of it. But that completeness is exactly the Frobenius Uniqueness Conjecture for R: any other companion pair {x',z'} for the same R would yield another solution (or another orbit) of the same Pell equation, and conversely any additional Pell solution would give another companion pair for R.
fitted input called prediction [Section 2, paragraph beginning 'Such an analysis using Mathematica shows...' and the note after Table 2]
"Such an analysis using Mathematica shows that if and only if the parameter R is equal to a Markov number does a solution exist and that the two smallest solutions, J, for a given parameter R are always the x and z values which complete the Markov triplet {x, R, z} for region number R. ... This statement follows from the analysis of the full Pell equation, V [{x, R, z}, n+ 1]2 − D(R) U [{x, R, z}, n+ 1]2 = −(2R)2, not the above, more dramatic, Mathematica analysis since the table could have been created by sorting the n = {... − 3, −2, −1, 0, 1, 2 ...} values of U [{x, R, z}, n+ 1]."

The paper explicitly concedes that Table 2 could have been produced by sorting values of U[{x,R,z}, n+1]. Since the sequence function is defined with U[{x,R,z},0] = x and U[{x,R,z},1] = z, the two smallest sorted solutions are the input triplet members by construction. Thus the Mathematica 'finding' is not an independent verification: the x and z values are presented as predicted solutions of the Pell equation, but they are already embedded in the sequence used to generate the table. The text first offers the table as evidence that the two smallest solutions complete the triplet, then disclaims it, but the main proof still depends on the same orbit-completeness assertion, so the disclaimer does not remove the circularity.

full rationale

Section 2 correctly derives the Pell identity V[{x,R,z},n+1]^2 − D(R) U[{x,R,z},n+1]^2 = −(2R)^2 from the Markov equation via hyperbolic substitutions; that algebraic identity itself is not circular. The circularity enters at the completeness step: the paper jumps from the identity to 'these are the only solutions' with no proof that the unit (3R + sqrt(D(R)))/2 is fundamental or that the norm equation has no additional solution orbits. That completeness claim is logically equivalent to the uniqueness of the companion pair for a fixed Markov number R, i.e., to the Frobenius Uniqueness Conjecture itself. The subsequent Mathematica table is explicitly disclaimed as being reproducible by sorting U-values, so the 'two smallest solutions are x and z' observation merely reads off the initial values built into the sequence function. No self-citation or external benchmark is involved; the failure is internal to the argument. The other sections on palindromic digit cycles, Farey indexing, and special square terms are independent pattern observations and do not affect the circularity score. Because the paper's central claim reduces the conjecture to an asserted restatement of itself, the score is 8.

Assumptions & free parameters 2 free parameters · 3 assumptions · 1 invented entities

The paper relies on standard Lucas sequence theory and Pell equation theory, but it also uses an unproven assertion that a single orbit generates all Pell equation solutions. The claimed resolution of the Uniqueness Conjecture depends on this assertion and on the implicit assumption that the Markov tree contains all triplets.

free parameters (2)

Startup values for algorithm Q = Q[{1,1,1}]={{0,1},{1,1}}; Q[{1,2,1}]={{1,1},{0,1}}; Q[{1,5,2}]={{1,2},{0,1}}
The recursive algorithm Q requires three hand-chosen initial outputs to terminate. These values are not derived and are needed for the special-square-term computations.
RegionSign exceptions = RegionSign[{1,2,1}]=-1, RegionSign[{1,5,2}]=-1; otherwise LeftRightSign*ParitySign
The sign rule in the Q algorithm is defined by the position of the triplet in a hand-ordered MarkovList, with two explicit exceptions. It is fitted so that the algorithm produces the desired square terms.

assumptions (3)

domain assumption The Markov child formulas generate all non-singular Markov triplets from {1,5,2}.
Used throughout Section 1 to derive edge sequence functions; the paper cites it as well known but does not prove it.
ad hoc to paper The element 1/2(3R+sqrt(D(R))) generates all solutions to the residual Pell equation X^2-D(R)Y^2=-(2R)^2.
Invoked in Section 2 as the simplest solution to the unit Pell equation. The paper asserts that this yields the set of solutions and hence these are the only solutions, but this is exactly the unproven step on which the Uniqueness Conjecture resolution rests.
domain assumption The sequence functions Hsf and Ksf list all triplets and special square terms along the edges of every region for all integers n.
The paper extends formulas derived for positive n to all n at the start of Section 1.2 and claims this works for all regions, but no proof of completeness is given.

invented entities (1)

Region sibling number
purpose: A new value s=3xz-R associated to each non-singular Markov triplet, used to structure the Q algorithm and the sequence functions.
This is a reformulation of the secondary solution of the Markov equation for fixed x and z. It is a mathematical definition, not a new physical entity, and has no independent falsifiable handle.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Patterns Within the Markov Tree." pith.science (2026). https://pith.science/paper/OUKKJKAH

@misc{pith2026250604299,
  author       = {Pith},
  title        = {Pith review of: Patterns Within the Markov Tree},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OUKKJKAH}},
  note         = {Machine review of arXiv:2506.04299}
}

read the original abstract

An analysis of the Markov tree is presented. Markov triplets, {x,R,z}, are the positive integer solutions to the Diophantine equation x2 + R2 + z2 = 3xRz. Inspired by patterns of the Fibonacci and Pell triplets in Region 1 and Region 2 of the tree, an investigation of interior regions of the Markov tree finds generating functions and sequence functions for all triplets of all regions. These sequence functions lead to the discovery of a Pell equation for the Markov region numbers along the edges of all regions. Analysis of this Pell equation leads to the resolution of the Uniqueness Conjecture. Further analysis using these sequence functions finds palindromic repeat cycles of the last digits of region numbers along the edges of all regions. Then, since all Markov numbers are the sum of the squares of two integers and again inspired by the patterns of the two unique squares which sum to form the region numbers of certain Fibonacci triplets in Region 1, an investigation of interior regions of the Markov tree finds generating functions and sequence functions for the two special square terms which sum to form the region numbers of the triplets along the edges of all regions. Further analysis using these sequence functions finds palindromic repeat cycles of the last digits of these two special square terms for all regions.

Figures

Figures reproduced from arXiv: 2506.04299 by the authors.

**Figure 2.** Patterns in the frequency of occurrence of last digits of Markov numbers. [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗

**Figure 3.** Markov Region Numbers as a Function of their Farey Number [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Markov Region Numbers as a Function of their Farey Number [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]

**Figure 5.** Figure 5: The Oscillatory Behavior of ΛR as a Function of σR along the Left and Right Edges of Sample Region 5. 6 Inspiration This work was begun after reading an article in Quanta Magazine. The abundance of patterns of the Fibonacci and Pell numbers in Region 1 and Region 2 of …

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

3 extracted references · 3 canonical work pages

[1]

Patterns Within the Markov Tree

Patterns Within the Markov Tree Robert A. Gore Cheyney, PA August 27, 2025 Abstract An analysis of the Markov tree is presented. Markov triplets, {x, R, z}, are the positive integer solutions to the Diophantine equation x2 + R2 + z2 = 3xRz. Inspired by patterns of the Fibonacci and Pell triplets in Region 1 and Region 2 of the tree, an investigation of in...

work page Pith review arXiv 2025
[5]

The abundance of patterns of the Fibonacci and Pell numbers in Region 1 and Region 2 of the Markov tree provided signposts for further analysis using Mathematica

6 Inspiration This work was begun after reading an article in Quanta Magazine. The abundance of patterns of the Fibonacci and Pell numbers in Region 1 and Region 2 of the Markov tree provided signposts for further analysis using Mathematica. The hope is that others will find these results as interesting as I have. 7 References Kramer, A., & Kramer, A. (20...

work page 2023
[89]

Also note that {1,4,1} and {9,6,9} are tens complements of each other. Interestingly, although the lengths of repeat cycles of even-indexed Fibonacci numbers are the same as the lengths of repeat cycles of odd-indexed Fibonacci numbers, none of the even-indexed Fibonacci number repeat cycles are palindromic since its first repeat cycle, {0, 1, 3, 8, 1, 5,...

work page 1985

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Patterns Within the Markov Tree

Patterns Within the Markov Tree Robert A. Gore Cheyney, PA August 27, 2025 Abstract An analysis of the Markov tree is presented. Markov triplets, {x, R, z}, are the positive integer solutions to the Diophantine equation x2 + R2 + z2 = 3xRz. Inspired by patterns of the Fibonacci and Pell triplets in Region 1 and Region 2 of the tree, an investigation of in...

work page Pith review arXiv 2025

[2] [5]

The abundance of patterns of the Fibonacci and Pell numbers in Region 1 and Region 2 of the Markov tree provided signposts for further analysis using Mathematica

6 Inspiration This work was begun after reading an article in Quanta Magazine. The abundance of patterns of the Fibonacci and Pell numbers in Region 1 and Region 2 of the Markov tree provided signposts for further analysis using Mathematica. The hope is that others will find these results as interesting as I have. 7 References Kramer, A., & Kramer, A. (20...

work page 2023

[3] [89]

Also note that {1,4,1} and {9,6,9} are tens complements of each other. Interestingly, although the lengths of repeat cycles of even-indexed Fibonacci numbers are the same as the lengths of repeat cycles of odd-indexed Fibonacci numbers, none of the even-indexed Fibonacci number repeat cycles are palindromic since its first repeat cycle, {0, 1, 3, 8, 1, 5,...

work page 1985