Matings between compositions of rational maps and free products of finite cyclic groups

Luna Lomonaco; Miguel Ratis Laude; Shaun Bullett

Matings between compositions of rational maps and free products of finite cyclic groups

T0 review · 3 major / 3 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read The paper proves that pairs of parabolic maps can be mated with free products of cyclic groups via algebraic correspondences that, for the first time, are not conjugate to their own inverses.

desk verdict A genuine advance in parabolic pair-matings, but Theorem D has an unproved holomorphicity step and the non-time-reversibility claim is asserted without proof. read the letter →

arxiv 2607.14364 v2 pith:XDMACOHZ submitted 2026-07-15 math.DS

Shaun Bullett , Luna Lomonaco , Miguel Ratis Laude This is my paper · ORCID

classification math.DS MSC 37F1030F4037F34

keywords matingsalgebraiccorrespondencesrationalmapsfreeproductsofcyclicgroupsKleinianquasiconformalsurgerypinchedpolynomial-likedeletedcovering

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 constructs, for any degrees p,q ≥ 2, a holomorphic algebraic correspondence of bidegree (pq,pq) on the Riemann sphere that simultaneously realizes the dynamics of the two compositions g∘f and f∘g and the action of a faithful discrete representation of the free product C_{p+1}*C_{q+1}. The construction works in two settings: parabolic rational maps f,g whose compositions have connected filled Julia set, and arbitrary polynomials with connected filled Julia set mated with Kleinian representations with totally disconnected limit set. In both cases the mating correspondence factors as the composition of two 'deleted covering correspondences' attached to rational maps P,Q that are conjugate to polynomials of degrees p+1 and q+1. The authors' headline claim is that these are the first matings that are not time-reversible: unlike all previously known correspondences of this kind, they are not conjugate to their own inverse.

What carries the argument

The pinched polynomial-like map: a branched cover from a pinched polygon onto a polygon, used to localize g∘f and f∘g around their filled Julia sets; and the deleted covering correspondence Cov^0_R of a rational map R, defined by R(w)=R(z), w≠z, which sends each point to the other points in its fiber. The group side is the Fuchsian representation Γ_{p,q} of C_{p+1}*C_{q+1}, whose generators have parabolic composition; its fundamental domain is a hyperbolic triangle. The mating is assembled by quasiconformally interpolating boundary homeomorphisms that conjugate the boundary actions of the group to those of the polynomial-like maps, then applying the Measurable Riemann Mapping Theorem.

What would settle it

Compute the correspondence F for a specific example (e.g., f(z)=z^2/(z+1/2), g(z)=(2z^2-z-1/2)/(2z+1) from Figure 1) and compare the iterated images of a generic point under F and under F^{-1}; if the closures are the same, time-reversibility would hold, contradicting the paper's claim. Alternatively, verify directly on the tiles φ_-(ρ^{-m}(A_ε∩Δ_ρ)) whether φ_- is conformal; a single tile where the quasiconformal dilatation is nonzero disproves the holomorphicity step in the proof of Theorem D.

Watch

Extended reading notes

Core claim

The central discovery is that a mating between a pair of maps (f,g) and the group Γ_{p,q} ≅ C_{p+1}*C_{q+1} can be realized not just by an abstract algebraic correspondence, but concretely as F = Cov^0_Q ∘ Cov^0_P, the composition of the deleted covering correspondences of two rational maps P and Q that are each conjugate to polynomials of degrees p+1 and q+1. This factorization gives the mating an explicit algebraic form and reveals its lack of time-reversibility: reversing F introduces a commutator between the two covering correspondences, rather than landing in the same conjugacy class. The proof glues the pinched polynomial-like restrictions of g∘f and f∘g to the fundamental domains of Γ

Load-bearing premise

The proof of Theorem D assumes that the quasiconformal boundary-conjugating maps φ_- and φ_+ become conformal on the tiles where P and Q are defined, so that P and Q are holomorphic; quasiconformal conjugates of Möbius transformations are not generally holomorphic, and no conformality is established.

Editorial extensions

If this is right

Every pair f,g satisfying the hypotheses yields an explicit algebraic correspondence F of bidegree (pq,pq), given in closed form as Cov^0_Q ∘ Cov^0_P.
The filled Julia sets K(g∘f) and K(f∘g) are glued together in the mating, with only the parabolic fixed point identified; the correspondence F|Λ- is hybrid conjugate to g∘f and F^{-1}|Λ+ to f∘g.
Because F is not conjugate to F^{-1}, the standard symmetry argument used to classify previous matings does not apply; new non-reversible dynamical systems are obtained.
For polynomials, the construction extends to all faithful discrete representations of C_{p+1}*C_{q+1} with Cantor limit set, via a Kleinian perturbation of Γ_{p,q} in which all non-elliptic elements are hyperbolic.
The factorization theorem shows that the mating correspondence is determined by the two polynomial-conjugate maps P,Q, so the parameter space of matings is a quotient of the product of the degree-(p+1) and degree-(q+1) polynomial spaces.

Reading between the lines

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

If the factorization is correct, the non-reversibility should be visible numerically: the correspondence F and its inverse have different grand orbit structures, so iterating F versus F^{-1} should produce different limit sets for generic starting points; a computational check on the examples in Figure 1 would test this directly.
The construction hints at a broader principle: any pair of maps with a common 'pinching' structure and a group with matching orbifold might admit a mating correspondence of the form Cov^0_Q ∘ Cov^0_P, suggesting a systematic recipe for building matings between iterated compositions and free products.
A direct test of the proof's load-bearing step would be to check whether φ_- is conformal on the tiles φ_-(ρ^{-m}(A_ε∩Δ_ρ)); if not, the non-parabolic orbit matings of Theorem D may still exist but require a different factorization argument.
The examples with f=g (Figure 8, left) show that even self-mated pairs can yield non-time-reversible correspondences, implying the phenomenon is not merely an artifact of asymmetry between f and g.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

Let me give you the short version: the parabolic half of this paper is a genuine advance, and the construction of pair-matings with free products is new. But Theorem D has a specific unjustified holomorphicity step, and the "first non-time-reversible" claim is not proved. The paper is worth refereeing, not desk-rejecting.

The parabolic construction is the real meat. The idea of mating a pair (f,g) of parabolic maps with C_{p+1}*C_{q+1}, recovering Hecke cases at q=1, and realizing the correspondence as Cov_Q^0 ∘ Cov_P^0, is a genuine broadening of the theory. The surgery in Section 3 is detailed: pinched polynomial-like restrictions, boundary gluing with controlled angles via Pommerenke-Warschawski estimates, an invariant Beltrami form, and straightening. I found the argument plausible, though it relies on the same authors' earlier machinery (Lemma 3.2 defers to [9]); that's normal for a program like this.

The trouble is in Section 4.3, the proof of Theorem D. The maps P and Q are defined on tiles using φ_- and φ_+, which come from Lemma 4.2 as quasiconformal maps conjugating boundary actions. The text says P is holomorphic on those tiles because "φ_- conjugates the action of ρ_ε to a holomorphic map on Ĉ." That does not follow. A quasiconformal conjugate of a Möbius map is quasiconformal, not generally holomorphic. In Theorem B the analogous construction worked because there was a true conformal conjugacy on Ω; in the non-parabolic orbit mating there is only grand-orbit equivalence, and the Beltrami straightening Φ does not make φ_- conformal on the tiles. So the factorization F = Cov_Q^0 ∘ Cov_P^0 for the non-parabolic case is not established as written. This is a real, localizable gap, not a matter of taste.

There's a second, smaller issue: the headline claim of "first examples that are not time-reversible" is asserted without proof. The construction is not manifestly reversible, but the paper doesn't show F is not conjugate to F^{-1}. That's a missing argument, though it may be easy to supply.

If the holomorphicity gap is fixable—say, by giving a different argument for conformality on the tiles or by restricting the claim—the paper is a solid contribution. As it stands, Theorem D should be treated as conditional. The parabolic theorems A and B deserve serious attention on their own.

Who is this for? People working on correspondences, matings, and the complex-dynamics/Kleinian-group interface. I'd send it to a referee, but ask them to focus on §4.3 and the reversibility claim. The paper is not close to being ready as is, but it's also not a reject.

Referee Report

3 major / 3 minor

Summary. The paper constructs algebraic correspondences on the Riemann sphere that mate a pair of rational maps (f,g), each with a parabolic fixed point, with the Fuchsian group Γ_{p,q} isomorphic to C_{p+1}*C_{q+1}. The main parabolic result (Theorem A) produces a holomorphic correspondence of bidegree (pq,pq) under the hypothesis that g∘f belongs to Per^{pq}_1(1) and has connected filled Julia set. Theorem B represents such a correspondence as the composition of two deleted covering correspondences of rational maps conjugate to polynomials. Theorems C and D extend these constructions to polynomial pairs and to non-parabolic faithful discrete Kleinian representations with totally disconnected limit set, using the weaker notion of "orbit mating." The paper also claims that these are the first matings that are not time-reversible.

Significance. If the main theorems are correct, this is a substantial extension of mating theory: instead of mating a single map with a group, the paper mates a pair of maps with a free product of finite cyclic groups. The construction is explicit and uses a combination of pinched polynomial-like restrictions, group-combinatorial gluing, and quasiconformal/Beltrami surgery. The parabolic part (Theorems A and B) is detailed and draws on established techniques, and the algebraic correspondence framework is appropriate. The claimed factorizations into deleted covering correspondences are of independent interest, and the potential non-reversibility of the new correspondences would be a notable novelty. However, the proof of Theorem D contains a load-bearing gap, and a headline claim in the abstract is not proved.

major comments (3)

[§4.3, Proof of Theorem D] The factorization F = Cov_Q^0 ∘ Cov_P^0 is not established. The map P is declared to be holomorphic on the tiles φ_-(ρ^{-m}(A_ε∩Δ_ρ)) because "φ_- conjugates the action of ρ_ε to a holomorphic map on Ĉ." This does not follow. Lemma 4.2 supplies only quasiconformal extensions φ_-, φ_+ of boundary diffeomorphisms, and a quasiconformal conjugate of a Möbius transformation need not be holomorphic. The Beltrami straightening in Theorem C makes the global correspondence F holomorphic, but it does not make φ_- or φ_+ conformal on the tiles; their Beltrami coefficients are not shown to vanish. In Theorem B the analogous construction worked because the definition of mating gives a conformal conjugacy on Ω, whereas in Theorem C/D only a grand-orbit equivalence is available. Thus P and Q are not proved to be rational, and Theorem D's central claim is unsupported. The authors need either to prove th
[Theorems B and D, statements] Both theorem statements say that P and Q have degrees p and q, respectively. The proofs and the abstract require degrees p+1 and q+1. As stated, Cov_P^0 and Cov_Q^0 would have bidegrees (p-1,p-1) and (q-1,q-1), so Cov_Q^0∘Cov_P^0 would have bidegree ((p-1)(q-1),(p-1)(q-1)), not the asserted (pq,pq). This appears to be a typo, since the proof constructs P of degree p+1 and Q of degree q+1, but the statements must be corrected.
[Abstract and Introduction] The paper advertises that these matings are "the first examples that are not time-reversible." No proof of non-reversibility is supplied. Showing that the construction is not of the form J∘Cov_R^0 does not imply that the correspondence is not conjugate to its own inverse. Either prove non-reversibility for at least the constructed family, or qualify the claim as "not of the previously known reversible form."

minor comments (3)

[Theorem D statement] The group is written as C_p*C_q; it should be C_{p+1}*C_{q+1} to match Theorem C, the abstract, and the rest of the paper.
[Lemma 3.2] The proof of the quasiconformal extension lemma is condensed and refers to [9] and [14] for the main method. Since Theorem A depends on it, please expand the cusp-extension argument or give precise references to the exact statements being used.
[Proof of Theorem B] In the definition of P on Λ_-, the expression ϕ_+^{-1}∘ f∘ϕ_- is not explained; a sentence clarifying that f sends K(g∘f) to K(f∘g), so the composition lands in Λ_+, would improve readability.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the main construction is self-contained, and the self-citations are background rather than load-bearing.

full rationale

The results are constructive existence theorems. Theorem A assembles a topological correspondence G from pinched polynomial-like restrictions of g∘f and f∘g and from the explicit action of Γ_{p,q}, then straightens G with an invariant Beltrami form. Theorem C does the same with polynomial-like restrictions and an annulus quotient. The factorization results (Theorems B and D) are not predictions from fitted data: the maps P and Q are explicitly defined from the conjugacy maps φ, φ± and from f,g, and the equality Cov_Q^0 ∘ Cov_P^0 = F is verified on an open set with an accumulation point and then extended analytically. This is construction, not a fit renamed as a prediction. Self-citations appear — [9] for the pinched polynomial-like/Farey-like framework, [14] for the method in Lemma 3.2, [5,30] for the orbit-mating notion — but they provide definitions or parallel techniques; the load-bearing quasiconformal extension in Lemma 3.2 is argued directly using Pommerenke and Warschawski estimates. The genuine local weakness is in §4.3, where the paper asserts that P is holomorphic on tiles since φ_- conjugates the action of ρ_ε to a holomorphic map on Ĉ, although φ_- is only quasiconformal. This is a correctness gap in the proof of Theorem D, but it is not a circularity: it does not reduce Theorem D to its own assumptions. Hence the circularity score is minimal.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

Pure mathematics paper: no empirical data. The construction uses standard tools (Measurable Riemann Mapping, Chow, Klein combination) and a few results from earlier papers by the same group. No free parameters are fitted: the angle θ, neighborhoods U,V, and perturbation ε are auxiliary construction choices, not fitted values. The main unproved-but-needed input is the quasiconformal rigidity of the Kleinian representations and the technical gluing lemma deferred to [9].

assumptions (5)

standard math Measurable Riemann Mapping Theorem
Used in proofs of Theorems A and C to straighten the topological correspondence G into a holomorphic correspondence F.
standard math Chow's theorem: closed analytic subsets of complex projective space are algebraic
Used in §2.3 and Theorem A to conclude that the straightened holomorphic correspondence is an algebraic correspondence.
standard math Klein combination theorem for free products of finite cyclic groups
Used in §2.2 to show that Γ_{p,q} is a faithful discrete representation of C_{p+1}*C_{q+1}.
standard math Quasiconformal extension and Warschawski strip estimates
Used in Lemma 3.2 to extend boundary diffeomorphisms quasiconformally across cusps; the proof is condensed and partly deferred to [9, Lemma 5.2].
domain assumption Quasiconformal rigidity of faithful discrete representations of C_{p+1}*C_{q+1} with Cantor limit set
Invoked in §4.3 to reduce to the Fuchsian representative Γ_ε; a known theorem in Kleinian groups, not proved in this paper.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Matings between compositions of rational maps and free products of finite cyclic groups." pith.science (2026). https://pith.science/paper/XDMACOHZ

@misc{pith2026260714364,
  author       = {Pith},
  title        = {Pith review of: Matings between compositions of rational maps and free products of finite cyclic groups},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XDMACOHZ}},
  note         = {Machine review of arXiv:2607.14364}
}

abstract

Given a pair of rational maps $(f, g)$, of degrees $p$ and $q$, each with a parabolic fixed point having a fully invariant simply-connected basin of attraction, we construct an algebraic correspondence $F$ on the Riemann sphere, of bidegree $(pq, pq)$, realizing a mating between the two compositions $g\circ f$ and $f\circ g$ of the maps, and the parabolic faithful discrete representation of the free product of cyclic groups of orders $p + 1$ and $q + 1$. We also show that $F$ is the composition of a pair of deleted covering correspondences of rational maps which are conjugated to polynomials of degrees $p + 1$ and $q + 1$. We generalize our method to construct matings between compositions of pairs of polynomials and (non-parabolic) faithful Kleinian representations of the same group, now with connected regular set. As far as we are aware, these matings between pairs of maps and groups are the first examples that are not time-reversible (that is, they are not conjugate to their own inverses).