Pith. sign in

REVIEW 2 major objections 4 minor 1 cited by

This paper proves that for a typical surface in a family of Markoff-like equations over a prime field, every non-trivial orbit under the three mutation moves has size divisible by p, extending a theorem of W.Y. Chen from the classical Marko

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-05 11:49 UTC pith:SGGBW7D2

load-bearing objection The result is likely true, but the central extension argument in Section 2.2 is flawed as written; a shifted-index recurrence repairs it, so the paper deserves a careful referee, not unconditional trust. the 2 major comments →

arxiv 2509.02187 v3 pith:SGGBW7D2 submitted 2025-09-02 math.NT math.DSmath.RA

Divisibility by p for Markoff-like Surfaces

Matthew de Courcy-Ireland , Matthew Litman , Yuma Mizuno This is my paper
classification math.NT math.DSmath.RA MSC 11D2511T2414J26
keywords Markoff-like surfacesmodular orbitsp-divisibilityfinite fieldsgeneralized cluster algebrasCayley cubic surfacequadratic characters
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper studies equations of the form x1^2 + x2^2 + x3^2 + a1 x2 x3 + a2 x1 x3 + a3 x1 x2 = (3 + a1 + a2 + a3) x1 x2 x3 over the field of p elements, together with three moves that flip one variable to the other root of the quadratic. Its main theorem says that when p ≥ 5, the sum s = 3 + a1 + a2 + a3 is non-zero, and no a_i^2 equals 4 (or a special condition holds when one does), every orbit except the singleton (0,0,0) has size divisible by p. This extends a theorem of W.Y. Chen for the classical Markoff surface (all a_i = 0) and thereby rules out small non-trivial orbits in the modular mutation graph for typical members of the family. A secondary result counts the total number of non-zero solutions, with a correction term governed by Cayley's cubic surface, and shows that in special families there must be at least two or four orbits.

Core claim

The central claim is Theorem 1.1: for a prime p ≥ 5 and parameters a_i in F_p with s = 3 + a1 + a2 + a3 ≠ 0 and, for every i, a_i^2 ≠ 4, every orbit under the three moves other than {(0,0,0)} has size divisible by p. If some a_i^2 = 4, the same conclusion holds provided 2 a_{i-1} = a_{i+1} a_i, which means the parameters are, up to permutation, (2σ, α, ασ) with σ = ±1. The proof follows Martin's elementary proof of Chen's theorem for the Markoff surface, extending the auxiliary functions Δ_i to triples with a zero coordinate by propagating their values around the dihedral cycles that the moves generate on the conic x_i = 0; the consistency of this propagation is exactly Proposition 2.3. The

What carries the argument

The load-bearing object is the family of functions Δ_i(x) = x_i/(x_{i-1} x_{i+1}) + (1/2)(a_{i-1}/x_{i-1} + a_{i+1}/x_{i+1}) defined on triples with non-zero coordinates. On a solution of the surface equation they satisfy sum_i Δ_i = s, and the moves satisfy Δ_i(x) + Δ_i(m_i x) = s. Averaging these identities over an orbit O of size V gives sV = (3s/2)V in F_p, forcing V = 0 mod p once s ≠ 0. The delicate part is extending Δ_i to triples where one coordinate vanishes: the moves m_{i-1} and m_{i+1} act on the conic x_i = 0 as a dihedral group whose rotation ρ has order N (the order of r^2 for a root r of r^2 + a_i r + 1 = 0), and the extension is defined by propagating an arbitrary starting v

Load-bearing premise

The whole proof rests on being able to define the auxiliary functions Δ_i at triples with a zero coordinate, and to keep the two identities they satisfy on non-zero triples valid there too; without that extension the averaging argument that forces p to divide the orbit size cannot even be written down.

What would settle it

Pick a prime p ≥ 5 and parameters with s ≠ 0 and all a_i^2 ≠ 4 in F_p (for example p = 11, (a1,a2,a3) = (1,2,3)), enumerate all solutions to (1) over F_p, apply the three moves (2) to closure, and check whether every orbit other than {(0,0,0)} has size divisible by p; a single orbit whose size is not a multiple of p would refute Theorem 1.1, while the paper's counterexamples show the hypotheses are needed.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • No non-trivial orbit can have size 1, 2, ..., p−1 on any allowed surface; the modular mutation graph is p-periodic in the strong sense that every non-trivial component has size a positive multiple of p.
  • For the special parameters (2σ, α, ασ) with σ = ±1, Theorem 1.2 guarantees the non-zero solutions split into at least two orbits, and into at least four when α = ±2, so the p-divisibility does not collapse the orbit structure to a single component everywhere.
  • The counterexample (2,2,−2) shows both hypotheses of Theorem 1.1 are sharp: with s ≠ 0 but condition (4) violated, there are orbits of size 1, 2, and 4, and similarly with s = 0 for parameters like (0,0,−3).
  • Proposition 4.1 gives the exact number of non-zero solutions as p^2 + p(∑_i χ(a_i^2 − 4) + C); since Theorem 1.1 makes every non-trivial orbit size a multiple of p, the non-zero solutions are partitioned into orbits each of size at least p, with total p^2 + O(p).

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the angle-function extension works for other exchange polynomials (for example f_i(z) = 1 + b_i z + c_i z^2 with c_i ≠ 1), the same averaging argument would give p-divisibility for a wider family of generalised cluster algebra mutations; the dihedral-cycle consistency condition would become a character-sum identity that can be checked explicitly.
  • The Cayley-cubic correction in the solution count points to the 27-line geometry of that cubic surface; a natural test is whether the exceptional orbits that split under the quadratic character obstruction (χ(x_i)χ(x_i') ≠ −1) correspond to the lines or double-six configurations of Cayley's cubic.
  • Because the divisibility holds for every prime in the allowed range, the method might lift to statements about reduction mod p^k or to adelic orbit closures, turning the modular result into a constraint on how integer orbits of the moves reduce modulo powers of p.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 4 minor

Summary. The paper studies the Markoff-like surfaces (1) over F_p and proves, in Theorem 1.1, that for p≥5, s=3+a1+a2+a3≠0, and either all a_i^2≠4 or condition (4) for any a_i^2=4, every nontrivial orbit under the three moves (2) has size divisible by p. The proof follows Martin's averaging argument: one introduces functions Δ_i satisfying sum_i Δ_i=s and Δ_i(x)+Δ_i(m_i x)=s, extends them to triples with a zero coordinate using the dihedral cycles generated by m_{i±1}, and then derives sV=3sV/2 for an orbit O of size V. The paper also proves quadratic obstructions giving at least two or four orbits for special parameters (Theorem 1.2), computes the total number of nonzero solutions mod p (Proposition 4.1), and gives counterexamples when the hypotheses fail.

Significance. If correct, the main theorem is a substantial extension of Chen's Markoff-surface theorem and Martin's elementary proof, with consequences for the modular mutation graph: for typical parameters all nontrivial orbits have size at least p. The explicit solution count in Proposition 4.1, the Cayley-cubic parametrization of exceptional parameters, and the numerical orbit data add useful supporting content. The proof is self-contained and does not assume the desired orbit conclusions. However, the central extension step in §2.2 is misindexed as written, and Proposition 1.3 is false as stated; both are repairable but currently invalidate the printed proof.

major comments (2)
  1. [§2.2, Eqs. (22)–(23)] The propagation of Δ_{i−1} around the dihedral cycle does not satisfy the required identity (15) for m_{i+1}. Let v_n=ρ^n x, w_n=m_{i−1}v_n, f_n=Δ_i(v_n), h_n=Δ_i(w_n), e_n=Δ_{i−1}(v_n). Equations (22)–(23) give e_{n−1}−e_n=h_{n−1}+f_{n−1}. But the required m_{i+1}-pairing at v_n is e_{n−1}−e_n=f_n+h_{n−1}; the difference is f_n−f_{n−1}, generically nonzero. Concretely, with p=7, a=(1,1,1), i=1, x=(0,1,4), δ=0, the construction gives Δ_2(x)+Δ_2(m_2x)=4 mod 7 instead of s=6. The correct recurrence is e_n=e_0−Σ_{ℓ=1}^{n}f_ℓ−Σ_{ℓ=0}^{n−1}h_ℓ; its consistency still requires Eq. (24). Thus Proposition 2.3 is the right consistency condition, but the printed (22)–(23) do not define a valid extension.
  2. [Proposition 1.3] The statement is false as written. For p=7, a=(1,1,1), s=6, take x=(3,1,1) and y=(1,3,1). Then m_1x=(1,1,1)=m_2y, but x≠y. The proof actually establishes a different statement, namely that if m_i x=m_j x for i≠j then x is fixed by both moves (or that if m_i x=y=m_j x then x=y). This corrected version is what is needed in Lemma 2.1 and Proposition 2.2, and it is elementary, but the proposition must be restated and reproved.
minor comments (4)
  1. [Proposition 2.3 proof] In the factorization after the displayed sum, the factor should be Δ_i(m_{i−1}x), not Δ_{i−1}(m_{i−1}x); otherwise the notation is inconsistent with the summands.
  2. [§2.2, N=1 discussion] The sentence 'as long as 2a_{i−1}=a_{i+1}a_i, the resulting functions Δ_i solve (14) and (15)' depends on the corrected recurrence; with the current equations it is not true.
  3. [Throughout] The proof of Lemma 2.1 uses Proposition 1.3; once Proposition 1.3 is corrected, the argument is clear but should be checked against the new statement.
  4. [§2.4] The formula for the cycle-average of (Δ_{i−1}+Δ_{i+1})/2 is stated without derivation; a short proof would improve readability.

Circularity Check

0 steps flagged

No significant circularity: the orbit-sum proof is derived, not assumed.

full rationale

The derivation of Theorem 1.1 is self-contained given standard finite-field facts. The load-bearing step is the extension of Delta_{i±1} to triples with xi=0. The paper does not assume this extension exists; it constructs it in (22)-(23) from an arbitrary starting value delta and then proves the consistency condition (24) in Proposition 2.3 by direct computation from (13) and the dihedral action on the conic xi=0. The arbitrary delta cancels in the orbit sum, and the equation sV=3sV/2 follows algebraically from (14)-(16); neither equation is an input. Theorem 1.2 rests on an explicit square identity (31), and Proposition 4.1 is a direct conic count. The only self-citations (e.g., [13], [15]) are contextual remarks about the classical Markoff surface and K3 examples; they are not used to establish any theorem here. Section 5 explicitly identifies parameter regimes where the conclusion fails, which confirms the theorem is not vacuous and the hypotheses are doing real work. No circular step was found.

Axiom & Free-Parameter Ledger

0 free parameters · 5 axioms · 0 invented entities

No fitted parameters appear; the a_i are fixed field elements. The arbitrary delta in the extension is a gauge choice that cancels in the orbit sum. All other inputs are standard finite-field facts and the orbit-move structure. The paper introduces the Delta_i angle functions and the C(a) correction, but these are technical functions, not independent postulated entities.

axioms (5)
  • standard math Quadratic character sum sum_{t in F_p} chi(t^2-c) = -1 and standard finite-field conic point counts.
    Used throughout Section 4 to count solutions to conics in two variables; cited from Carlitz and standard finite field theory.
  • standard math Burnside-Cauchy-Frobenius orbit-counting lemma.
    Used in Proposition 5.1 to count orbits of the group generated by m1 and m2 on conics.
  • domain assumption The moves m_i are involutions and mutation orbits are connected components of the move graph; the intended no-bigon version of Proposition 1.3 holds.
    This defines the object being counted and is needed so each orbit is a union of pairs and fixed points under each move. The printed statement of Proposition 1.3 is misstated, but the proof establishes the no-bigon fact actually used.
  • ad hoc to paper The functions Delta_i can be extended consistently to triples with a zero coordinate so that equations (14) and (15) hold.
    This is the central technical construction of the proof, not a standard background result. It is proved via the arbitrary starting value delta and Proposition 2.3, but it is the load-bearing mechanism of Theorem 1.1.
  • domain assumption s = 3 + a1 + a2 + a3 is nonzero in F_p and p >= 5.
    These are explicit hypotheses of Theorem 1.1. The proof divides by s and uses 1/2, and the counterexamples show divisibility can fail when s=0.

pith-pipeline@v1.4.0-alltime-deepseek-medium · 19496 in / 26791 out tokens · 287335 ms · 2026-08-05T11:49:12.082181+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Divisibility by $p$ for Markoff-like Surfaces." pith.science (2026). https://pith.science/paper/SGGBW7D2

@misc{pith2026250902187,
  author       = {Pith},
  title        = {Pith review of: Divisibility by $p$ for Markoff-like Surfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SGGBW7D2}},
  note         = {Machine review of arXiv:2509.02187}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We study orbits in a family of Markoff-like surfaces with extra off-diagonal terms over prime fields $\mathbb{F}_p$. It is shown that, for a typical surface of this form, every non-trivial orbit has size divisible by $p$. This extends a theorem of W.Y. Chen from the Markoff surface itself to others in this family. The proof closely follows and elaborates on a recent argument of D.E. Martin. We expect that there is just one orbit generically. For some special parameters, we prove that there are at least two or four orbits. Cayley's cubic surface plays a role in parametrising the exceptional cases and dictating the number of solutions mod $p$.

Figures

Figures reproduced from arXiv: 2509.02187 by Matthew de Courcy-Ireland, Matthew Litman, Yuma Mizuno.

Figure 1
Figure 1. Figure 1: Left: the quiver for the cluster algebra associated with the classical Markoff equation. Right: the quiver for the generalised cluster algebra associated with the generalised Markoff equation with parameters a1, a2, a3. The f and fi are called exchange polynomials at the corresponding vertices. Then (1) becomes (9) X i [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The action of mi−1 and mi+1 on triples with xi = 0. For each value of xi−1, they form a cycle whose length depends on the order of a solution to r 2 + air + 1 = 0. If ai = 0, then the cycle has length 4. 2.1. Vanishing coordinates. In order to define ∆i±1(x) for triples with xi = 0, it is useful to describe these triples in more detail. Throughout this section, we let x be a point with xi = 0 for some i. F… view at source ↗
Figure 3
Figure 3. Figure 3: Values of (∆i−1, ∆i+1) in the Markoff case (a1 = a2 = a3 = 0, all ∆i = 0 on {xi = 0}, and r = √ −1), where δ is an arbitrary value of ∆i−1(x) to start the cycle. By examining the factors of Equation (26), we see that the Cayley cubic also comes into play when evaluating double fixed points. If the factor u 2 + ai−1ai+1u + a 2 i−1 + a 2 i+1 − 4 vanishes, then x is fixed by both mi−1 and mi+1, and (−sxi + ai… view at source ↗
Figure 4
Figure 4. Figure 4: The number of solutions to x 2 +Bxy +y 2 +Dx+ Ey+F = 0 in Fp. Top: the number is p−χ(B2−4) for smooth conics, in terms of the quadratic character χ mod p. Bottom: if D2 + E2 + F B2 − 4F − BDE = 0, there is a correction of p times χ(B2 − 4) or χ(D2 − 4F). Blue: analogous conic sections over the reals. Assuming the constant on the right-hand side of (36) is non-zero, Proposi￾tion 4.2 shows that the number of… view at source ↗
Figure 5
Figure 5. Figure 5: Orbits of size 2 for a = (2, 2, −2) and s ̸= 0. As for orbits of size 4, there are four “tripods” of the same shape; one central triple connected to three double fixed points. In fact, these four tripods can be broken into two classes, one class of a single orbit where all three of the double fixed points contain one coordinate equal to 0 and another class containing three orbits where only one double fixe… view at source ↗
Figure 6
Figure 6. Figure 6: Orbits of size 4 for a = (2, 2, −2) and s ̸= 0 [PITH_FULL_IMAGE:figures/full_fig_p022_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: The solutions to x 2 + y 2 + z 2 = 0 mod 3. The move on i = 3 is a sign change m3 : x3 7→ −x3 which commutes with the other two moves. They are given by (41) m1 : x1 7→ −x1 + 3x2 (42) m2 : x2 7→ −x2 + 3x1 Scaling x 7→ tx also commutes with all three moves. It is therefore enough to understand the action on the two conics (∗, ∗, 1) and (∗, ∗, −1), which are linked by m3, and separately the action on (∗, ∗, … view at source ↗
Figure 8
Figure 8. Figure 8: One of the orbits for m1, m2, m3 acting on the two conics (∗, ∗, ±1) for p = 11. We write X = −1 for brevity. There are fixed points 2 7→ 2 on triples of the form (2, 5, ±1). One orbit of size 22 contains both (1, 1) and (−1, −1). This orbit is preserved by the involutions (x, y, 1) 7→ (y, x, 1) (x, y, 1) 7→ (−x, −y, 1) However, (x, y) = (6, 29) and (x, y) = (−6, −29) lie in different orbits of size 22 whi… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Generalized discrete Markov spectra

    math.NT 2025-12 unverdicted novelty 7.0

    The paper constructs generalized discrete Markov spectra for the family of equations x² + y² + z² + k1 yz + k2 zx + k3 xy = (3 + k1 + k2 + k3) xyz, with each spectrum element realized as both a Lagrange constant of a ...

Reference graph

Works this paper leans on

28 extracted references · 13 canonical work pages · cited by 1 Pith paper

  1. [1]

    Esther Banaian and Yashaki Gyoda,Cluster algebraic interpretation of generalized Markov numbers and their matrixizations, arXiv 2507.06900 https://arxiv.org/abs/ 2507.06900

    Pith/arXiv arXiv

  2. [2]

    Esther Banaian and Yadira Valdivieso,Snake Graphs and Caldero-Chapoton Functions from Triangulated Orbifolds, Journal of Algebra, Volume 674, 15 July 2025, Pages 77-116https://doi.org/10.1016/j.jalgebra.2025.03.006

    work page doi:10.1016/j.jalgebra.2025.03.006 2025

  3. [3]

    Benedetto and William M

    Robert L. Benedetto and William M. Goldman,The topology of the relative character varieties of a quadruply-punctured sphereExperiment. Math. 8(1): 85-103 (1999)

    work page 1999

  4. [4]

    Jean Bourgain, Alexander Gamburd, and Peter Sarnak,Strong approximation and Diophantine properties of Markoff triples, J. Amer. Math. Soc. DOI: https://doi. org/10.1090/jams/1061 Published electronically: August 29, 2025 arXiv:1607.01530

    Pith/arXiv arXiv 2025

  5. [5]

    Theory of Groups of Finite Order

    William Burnside (1897). Theory of Groups of Finite Order. Cambridge University Press

    work page

  6. [6]

    Leonard Carlitz,The number of points on certain cubic surfaces over a finite field, Boll. Un. Mat. Ital.,(3), 12, (1957), pages 19–21

    work page 1957

  7. [7]

    Augustin-Louis Cauchy,M´ emoire sur diverses propri´ et´ es remarquables des substitutions r´ eguli` eres ou irr´ eguli` eres, et des syst´ emes de substitutiones conjug´ ees.Comptes Rendus Acad. Sci. Paris 21, 835, 1845

    work page

  8. [8]

    Duke Math

    Serge Cantat,Bers and H´ enon, Painlev´ e and Schr¨ odinger. Duke Math. J. 149 (3) 411 - 460, 15 September 2009.https://doi.org/10.1215/00127094-2009-042

    doi:10.1215/00127094-2009-042 2009

  9. [9]

    Serge Cantat and Frank Lorary,Dynamics on Character Varieties and Malgrange irreducibility of Painlev´ e VI equationAnnales de l’Institut Fourier, Volume 59 (2009) no. 7, pp. 2927-2978. DOI10.5802/aif.2512

    2009

  10. [10]

    Serge Cantat, Christophe Dupont, and Flrestan Martin-Baillon,Dynamics on Markov surfaces: classification of stationary measures, arXiv pre-prints https://arxiv.org/ abs/2404.01721 26

    Pith/arXiv arXiv

  11. [11]

    Chekhov, Marta Mazzocco, and Vladimir N

    Leonid O. Chekhov, Marta Mazzocco, and Vladimir N. Rubtsov,Painlev´ e Monodromy Manifolds, Decorated Character Varieties, and Cluster Algebras, Int. Math. Res. Not. 2017, No. 24, 7639-7691 (2017).https://doi.org/10.1093/imrn/rnw219

    doi:10.1093/imrn/rnw219 2017

  12. [12]

    Chen,Nonabelian level structures, Nielsen equivalence, and Markoff triples, Ann

    William Y. Chen,Nonabelian level structures, Nielsen equivalence, and Markoff triples, Ann. of Math.199, pages 301–443 (2024) https://doi.org/10.4007/annals.2024. 199.1.5

    work page doi:10.4007/annals.2024 2024

  13. [13]

    London Math

    Jillian Eddy, Elena Fuchs, Matthew Litman, Daniel Martin, and Nico Tripeny,Con- nectivity of Markoff mod-p graphs and maximal divisors, Proc. London Math. Soc. Volume 130, Issue 2, February 2025, e70027https://doi.org/10.1112/plms.70027

    doi:10.1112/plms.70027 2025

  14. [14]

    Ferdinand Georg Frobenius (1887), ”Ueber die Congruenz nach einem aus zwei endlichen Gruppen gebildeten Doppelmodul”, Crelle’s Journal, 101 (4): 273–299, doi:10.3931/e-rara-18804

    work page doi:10.3931/e-rara-18804

  15. [15]

    Silverman, and Austin TranOrbits on K3 Surfaces of Markoff Type

    Elena Fuchs, Matthew Litman, Joseph H. Silverman, and Austin TranOrbits on K3 Surfaces of Markoff Type. Experimental Mathematics (2023). DOI: 10.1080/ 10586458.2023.2239265

    arXiv 2023

  16. [16]

    Yasuaki Gyoda,Positive integer solutions to(x+y) 2 + (y+z) 2 + (z+x) 2 = 12xyz, arXiv 2109.09639https://arxiv.org/abs/2109.09639

    Pith/arXiv arXiv

  17. [17]

    Yashaki Gyoda and Kodai Matsushita,Generalization of Markov Diophantine equation via generalized cluster algebra, Electronic Journal of Combinatorics 30 (2023), P4.10

    work page 2023

  18. [18]

    Yasuaki Gyoda, Shuhei Maruyama, and Yusuke Sato.SL(2,Z)-matrixizations of generalized Markov numbersarXiv:2407.08203https://arxiv.org/abs/2407.08203

    Pith/arXiv arXiv

  19. [19]

    Kazuhiro Hikami,Note on Character Varieties and Cluster Algebras, SIGMA15 (2019), 003, 32 pageshttps://doi.org/10.3842/SIGMA.2019.003,

    work page doi:10.3842/sigma.2019.003 2019

  20. [20]

    Kyungyong Lee, Li Li, Michelle Rabideau, and Ralf Schiffler.On the ordering of the Markov numbers, Advances in Applied Mathematics, Volume 143, (2023), 102453

    work page 2023

  21. [21]

    Wilhelm Magnus,Rings of Fricke Characters and Automorphism Groups of Free GroupsMath. Z. 170, 91-103 (1980)https://doi.org/10.1007/BF01214715

    doi:10.1007/bf01214715 1980

  22. [22]

    Sara Maloni, Fr´ ed´ eric Palesi, and Ser Peow Tan.On the character variety of the four-holed sphere, Groups Geom. Dyn. 9 (2015), 737–782 DOI 10.4171/GGD/326

    doi:10.4171/ggd/326 2015

  23. [23]

    Andrey Markoff,Sur les formes quadratiques binaires ind´ efinies, Volume 17, pages 379–399, (1880)https://link.springer.com/article/10.1007/BF01446234

    doi:10.1007/bf01446234

  24. [24]

    Martin,A new proof of Chen ’s theorem for Markoff graphs, Inven- tiones Mathematicae, Volume 241, pages 623–626 (2025)

    Daniel E. Martin,A new proof of Chen ’s theorem for Markoff graphs, Inven- tiones Mathematicae, Volume 241, pages 623–626 (2025). https://doi.org/10.1007/ s00222-025-01346-9

    work page 2025

  25. [25]

    Evan M O‘Dorney,Large Orbits on Markoff-Type K3 Surfaces over Finite Fields, International Mathematics Research Notices, Volume 2023, Issue 24, (2023), pages 21874-–21879,https://doi.org/10.1093/imrn/rnac341

    doi:10.1093/imrn/rnac341 2023

  26. [26]

    Michelle Rabideau and Ralf Schiffler,Continued fractions and orderings on the Markov numbers, Adv. Math. 370 (2020), 107231. https://doi.org/10.1016/j.aim.2020. 107231

    doi:10.1016/j.aim.2020 2020

  27. [27]

    Peter Swinnerton-Dyer,Cubic surfaces over finite fieldsMath. Proc. Camb. Phil. Soc. (2010), 149, 385 DOIhttps://doi.org/10.1017/S0305004110000320

    work page doi:10.1017/s0305004110000320 2010

  28. [28]

    Henry Gustave Vogt,Sur les invariants fondamentaux des ´ equations diff´ erentielles lin´ eaires du second ordre. Ann. Sci. Ecole Norm. Sup. (3) 6, Suppl. 3-72 (1889) (Th` ese, Paris). Email address:matthew.decourcy-ireland@math.su.se Department of Mathematics, Stockholm University and the Nordic Institute for Theoretical Physics Email address:matthew.lit...

    work page