Flip-graphs of non-orientable filling surfaces

Hugo Parlier; Lionel Pournin; Pallavi Panda

REVIEW 3 major objections 4 minor 18 references

Flip-graphs of non-orientable filling surfaces

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The modular flip-graph diameter of every non-orientable filling surface grows linearly at a rate between 5/2 and 4, and the unpunctured Möbius strip attains the exact rate 5/2.

desk verdict Solid extension of the Parlier–Pournin flip-graph results to non-orientable surfaces, capped by an exact 5n/2 diameter for the Möbius strip; the few unproved structural facts are load-bearing but easy to fill. read the letter →

arxiv 2505.04074 v1 pith:3IGKMW36 submitted 2025-05-07 math.GT math.CO

Pallavi Panda , Hugo Parlier , Lionel Pournin This is my paper · ORCID

classification math.GTmath.CO MSC 57K2005C12

keywords flipgraphsfillingsurfacesMöbiusstriptriangulationsofmappingclassgroupsgraphdiameternon-orientablesimplicial

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

This paper studies how many elementary moves, called flips, are needed in the worst case to turn one triangulation of a non-orientable surface into another, when the surface is built by placing $n$ marked points on a distinguished boundary component and keeping all other topology fixed. Its central claim is that this worst-case number, the diameter of the flip-graph (or of its quotient by boundary-fixing homeomorphisms), grows linearly in $n$ with a growth rate $c_\Sigma$ lying between $5/2$ and $4$. The paper further proves that the lower endpoint is attained: for the unpunctured Möbius strip, the flip-graph diameter is exactly asymptotic to $5n/2$, satisfying $\lfloor 5n/2 \rfloor - 2 \le \operatorname{diam}(\mathcal{F}(M_n)) \le \lfloor 5n/2 \rfloor$ for every positive $n$. These results extend previously known orientable bounds to non-orientable surfaces and identify the simplest non-orientable surface as the extremal example. A third result bounds the diameter of the subgraph formed by simplicial triangulations of the Möbius strip between roughly $5n/2$ and $4n$ with additive constants.

What carries the argument

The lower bound is carried by two explicit families of triangulations, $A^-_n$ and $A^+_n$, built from a zigzag of boundary arcs together with fixed triangulations of a one-holed subsurface; every geodesic between them is forced to spend many flips on three distinguished boundary arcs, and lemmas on boundary-arc contraction convert those forced flips into recursive lower bounds that accumulate to $\lfloor 5n/2 \rfloor$. The upper bound on the Möbius strip is carried by the central triangle: a triangle bounded by at least one non-separating arc whose three bounding arc lengths sum to $n$. The paper proves that every triangulation of $M_n$ contains such a triangle, using a structural classification of triangles (each is bounded by at least two arcs, with a fixed parity pattern of non-separating arcs), and then routes any two triangulations through intermediate triangulations $C_u(v,w)$ that are all within distance about $n$ of the two endpoints.

What would settle it

For $n=5$ or $n=6$, enumerate all triangulations of the unpunctured Möbius strip and check each triangle against Proposition 4.1: any triangle with fewer than two bounding arcs, any two-arc triangle with both or neither of its loops non-separating, or any three-arc triangle with exactly one non-separating arc would refute the structural claim behind the upper bound. Alternatively, compute the exact flip distance between the paper's $A^-_n$ and $A^+_n$ by exhaustive search; a distance below $\lfloor 5n/2 \rfloor - 2$ would contradict Theorem 1.2.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the asymptotic-diameter theory of flip-graph moduli spaces for orientable surfaces carries over to non-orientable surfaces, and that the Möbius strip is the example that makes the general lower bound sharp. For any non-orientable filling surface $\Sigma$, the limit $c_\Sigma = \lim_{n\to\infty} \operatorname{diam}(\mathcal{MF}(\Sigma_n))/n$ exists and satisfies $5/2 \le c_\Sigma \le 4$; for non-orientable one-holed surfaces of demigenus $g$ (the number of cross-caps inserted into a disk), the upper bound improves to $4 - 1/(2g)$ when $g\ge 3$ and to $23/8$ when $g=2$. For the unpunctured Möbius strip $M$, whose mapping class group is trivial so that the modular flip-graph coincides with the ordinary flip-graph, the paper proves $\lfloor 5n/2 \rfloor - 2 \le \operatorname{diam}(\mathcal{F}(M_n)) \le \lfloor 5n/2 \rfloor$ for every $n$.

Load-bearing premise

The load-bearing premises are two structural rules asserted without proof: that every triangle in a Möbius-strip triangulation belongs to one of the three listed shape classes, and that a vertex-reduction lemma proven for orientable surfaces (Theorem 2.4, transposed from [8, Lemma 3.2]) still works without orientability. If either rule has an exception, the matching upper bound collapses.

Editorial extensions

If this is right

Because the limit $c_\Sigma$ exists for every non-orientable filling surface, the asymptotic flip distance between triangulations is a well-defined topological invariant of the surface type.
The Möbius strip attains $c_\Sigma = 5/2$, so the universal lower bound in Theorem 1.1 is sharp; no constant larger than $5/2$ can be valid uniformly across non-orientable filling surfaces.
For a non-orientable one-holed surface of demigenus $g\ge 3$, the general upper bound $4$ improves to $4 - 1/(2g)$, and to $23/8$ for $g=2$, so the bound tightens as the non-orientable topology grows.
On the Möbius strip, the simplicial-triangulation flip graph has diameter at least $\lfloor 5n/2\rfloor - 16$ and at most $4n+K$, so restricting to simplicial triangulations does not change the linear order of worst-case flip distance.
Because the flip graph is a quasi-isometric model of the mapping class group, the linear bounds provide explicit constants for the quasi-isometry of non-orientable mapping class groups.

Reading between the lines

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

If the answer to the paper's Question 1.4 is that $c_\Sigma$ is monotone nondecreasing in demigenus, then the Möbius strip's value $5/2$ would force every non-orientable one-holed surface to have $c_\Sigma = 5/2$ only at demigenus 1; equality at higher demigenus would contradict monotonicity.
The two-family construction $A^-_n, A^+_n$ is more flexible than the earlier orientable-only versions, so tracking flips incident to a fourth or fifth boundary arc in the same recursion could in principle yield lower bounds strictly above $5/2$ for surfaces of higher demigenus.
Proposition 4.1 is asserted without proof; a direct enumeration of all triangulations of $M_5$ and $M_6$ would serve as a mechanical check of that classification before a full proof, and would also test whether the seven-flip bound to simplicial triangulations is tight.
One could ask whether the exact $5n/2$ behaviour transfers to a punctured Möbius strip; the paper's central-triangle machinery depends on the unpunctured structure, so the punctured case is a natural testbed for whether the constant $5/2$ is stable under adding interior punctures.

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

Quick verdict: this is a genuinely useful paper. It does what the title says—extends the orientable bounds on modular flip-graph diameters to non-orientable filling surfaces, and then pins down the Möbius strip case exactly: diameter asymptotic to 5n/2, which matches the general lower bound. Theorem 1.1 and Theorem 1.2 are new results, not just restatements. The proof of Theorem 3.1 genuinely generalizes the boundary-contraction machinery from [9], and the Möbius upper bound via central triangles is real work. The paper even catches a constant error in [9], which is the kind of care that earns trust.

Soft spots are real but minor. Proposition 4.1, the structural classification of arcs bounding a triangle in the Möbius strip, is stated without proof and is load-bearing for the upper bound: Lemma 4.5 and Theorem 4.8 both rely on it. The reader flagged this, and I agree. It is not an error—the mod-2 homology parity argument works. The boundary of a triangle is null-homologous, and the unique Z/2 homology class of the Möbius strip forces the number of non-separating arc-occurrences in the boundary walk to be even. That gives exactly the two-arc and three-arc cases claimed. The paper should supply that half-page proof. Theorem 2.4 is likewise quoted from [8] with the proof omitted; here the transposition is as straightforward as the authors say, because the star of a boundary vertex is a disk, so the orientable proof carries over. Still, an editor should ask to see it.

Citations are honest: heavy use of [8] and [9] is justified because these are the prior results being extended, not window dressing. The exposition is careful, aside from a few typos ("triangilations", "sufficees"). Theorem 1.3 and Question 1.4 are reasonable side contributions.

Who is this for? People who work on flip graphs, triangulations, and mapping-class-group geometry. It will not change the world outside that niche, but inside it, the exact asymptotic for the Möbius strip is a clean benchmark. I would send it to a competent referee. With the structural lemma added, it should be accepted.

Referee Report

3 major / 4 minor

Summary. The paper extends the theory of modular flip-graphs of filling surfaces to non-orientable surfaces. A filling surface Σ_n is obtained from a non-orientable surface Σ by fixing all marked points except n marked points on a privileged boundary component; the modular flip-graph MF(Σ_n) is the quotient of the flip-graph of triangulations by the mapping class group fixing that boundary component pointwise. The main theorems assert: (1) for every non-orientable filling surface there is a constant c_Σ with diam(MF(Σ_n))/n converging to c_Σ, and 5/2 ≤ c_Σ ≤ 4, with improved upper bounds for one-holed surfaces of demigenus g; (2) for the unpunctured Möbius strip M, diam(F(M_n)) lies between floor(5n/2)-2 and floor(5n/2), so the lower bound of the general interval is sharp; and (3) the subgraph of simplicial triangulations satisfies analogous linear bounds. The proofs adapt the orientable constructions of Parlier–Pournin, introduce a new general lower bound via boundary-arc contractions, and prove an upper bound for the Möbius strip through a structural theorem on central triangles.

Significance. If the results hold, this is a substantial extension of the known asymptotics of flip-graph diameters from orientable to non-orientable filling surfaces. The exact asymptotic diam(F(M_n)) ~ 5n/2 for the Möbius strip is particularly valuable, as it identifies the first non-orientable surface for which the constant is determined and shows that the general lower bound is sharp. The paper is also methodologically strong: the lower-bound proof is a genuine induction using boundary-arc contractions and explicit pairs of triangulations, the central-triangle argument is a new structural tool, and the statements are precise and falsifiable. The proofs are mostly self-contained except for two explicitly flagged transpositions from earlier work. There are no fitted parameters or numerical computations; the arguments are combinatorial and checkable by hand.

major comments (3)

[Section 4, Proposition 4.1] Proposition 4.1 is stated without proof, yet it is load-bearing for the Möbius-strip upper bound. Lemma 4.5 uses part (iii) to identify the third side of a triangle with two non-separating sides as separating, and Theorem 4.8 uses part (ii) in the two-arc case. A proof should be supplied. The statement is correct: each triangle is a disk, hence its boundary walk is null-homologous in H_1(M_n; Z/2), and since every non-separating arc represents the unique nonzero class, the number of non-separating arc-occurrences in the boundary walk must be even. This short parity argument would close the gap, but as written the manuscript asks the reader to accept an unproved structural classification on which the main upper bound rests.
[Section 2, Theorem 2.4 and Theorem 2.5] Theorem 2.4 is quoted from [8, Lemma 3.2] with the remark that the proof can be immediately transposed to non-orientable surfaces, but no proof or verification is included. This lemma is used to prove Theorem 2.5, which in turn gives the existence of the constant c_Σ and the general 4n upper bound in Theorem 1.1. Since the lemma is load-bearing, the manuscript should either prove it or explain in detail why the orientable proof carries over. The natural justification is that the star of a boundary vertex is a disk even in a non-orientable surface, so the local flip-counting argument from [8] should apply unchanged; nevertheless, this needs to be stated rather than left to the reader.
[Section 4, proof of Theorem 4.8] In the case where T+ contains a separating loop twice incident to u, the proof asserts that "the non-separating loop contained in T+" exists, but no justification is given. The presence of a separating loop based at u does not by itself rule out the coexistence of non-separating non-loop arcs incident to u elsewhere in the same triangulation. This case split is used to define the arc γ along which T+ is cut, so the proof of the upper bound is incomplete at this point. The authors should justify the asserted existence of a non-separating loop, or restructure the argument to avoid this assumption.

minor comments (4)

[Section 2, paragraph after Theorem 2.5] The sentence "Combining Proposition 2.3 with Theorem 2.7 shows that for every non-orientable filling surface..." appears to refer to Theorem 2.5, which is the result just proved; Theorem 2.7 is defined later. The same paragraph also says the first upper bound on c_Σ follows from Theorem 2.7, where Theorem 2.5 seems intended.
[Section 3, proof of Theorem 3.1] There is a typo in the phrase "it sufficees to lower bound" in the final case; it should read "it suffices".
[Section 4, paragraph before (21)] The phrase "two (non-necessarily distinct) points" should be "two (not necessarily distinct) points", and in Lemma 3.6 the word "canot" should be "cannot".
[Section 4, Figure 13] The figure and its caption are dense, and the four cases would be easier to follow if the text explicitly named which of the four panels is being referenced in each sentence of the proof of Theorem 4.8.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the main bounds are proved from prior published lemmas and new combinatorial constructions, with no parameter fitted to the claimed result.

full rationale

The derivation chain is self-contained in the sense required by the circularity pass. The general existence of c_Sigma and the 4n upper bound lift Theorem 2.4 ([8, Lemma 3.2]) to non-orientable surfaces by transposition; this is a citation of a published, independently checkable previous result, not an assumption of the target statement. The lower bound 5n/2 is established constructively in Section 3 via the families A_n^- and A_n^+ and the boundary-contraction Lemma 3.2, whose proof is said to work as is in the non-orientable case. The sharp Möbius upper bound in Section 4 rests on the central-triangle existence (Theorem 4.3) and the explicit path bounds (22), (23), and (21); the central-triangle definition is not a re-encoding of the diameter bound, and its existence is proved from Lemmas 4.4-4.6. Proposition 4.1 is stated without proof and is load-bearing, which is a potential correctness gap, but it is a topological classification of triangles, not a quantity defined in terms of the diameter; hence it is not a circular step. The paper also corrects an error in [9] (the constants K_g and K_2), which indicates independent scrutiny of the cited self-work rather than blind reliance. No fitted parameter is renamed as a prediction, and no uniqueness theorem from the authors' prior work is invoked to force a choice. The manuscript explicitly asks an open question (Question 1.4), further showing that the claims are not definitionally closed. Overall circularity score: 0.

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

The paper relies on standard facts about triangulations and on several results from the authors' prior orientable papers, transposed to non-orientable surfaces. The ledger lists the structural assumptions carried in without a full proof here; none involve fitted constants or invented objects.

assumptions (5)

domain assumption Strong convexity of flip-graphs for orientable surfaces (Theorem 2.1 from [3]) and its adaptation to arcs parallel to the privileged boundary for possibly non-orientable surfaces (Theorem 2.2)
Used to show subgraphs of ear triangulations are strongly convex, implying diam(MF(Σ_n)) is non-decreasing in n (Proposition 2.3), which is needed for existence of c_Σ.
standard math Euler characteristic: the number of interior arcs in a triangulation of Σ_n satisfies κ(Σ_n)/n → 1
Used in Section 2 to guarantee a vertex v on the privileged boundary incident to at most 4 interior arcs in total across T^- and T^+, enabling the recursive upper bound.
domain assumption Existence of cut systems γ_1..γ_g in any triangulation of a one-holed surface, whose removal yields a disk
Invoked in Section 2 to build the upper-bound path for one-holed surfaces of demigenus g.
domain assumption Lemma 3.2 (arc-contraction distance inequality) from [8] holds for non-orientable filling surfaces, with proof 'working as is'
It is the key recursion for the lower bound in Theorem 3.1; a failure would invalidate the 5n/2 lower bound, including the Möbius strip lower bound.
domain assumption Proposition 4.1 structural classification of triangles in Möbius strip triangulations
Stated without proof in Section 4 and used to prove the existence of central triangles (Theorem 4.3), the backbone of the upper bound diam(F(M_n)) ≤ 5n/2.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Flip-graphs of non-orientable filling surfaces." pith.science (2026). https://pith.science/paper/3IGKMW36

@misc{pith2026250504074,
  author       = {Pith},
  title        = {Pith review of: Flip-graphs of non-orientable filling surfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3IGKMW36}},
  note         = {Machine review of arXiv:2505.04074}
}

abstract

Consider a surface $\Sigma$ with punctures that serve as marked points and at least one marked point on each boundary component. We build a filling surface $\Sigma_n$ by singling out one of the boundary components and denoting by $n$ the number of marked points it contains. We consider the triangulations of $\Sigma_n$ whose vertices are the marked points and the associated flip-graph $\mathcal{F}(\Sigma_n)$. Quotienting $\mathcal{F}(\Sigma_n)$ by the homeomorphisms of $\Sigma$ that fix the privileged boundary component results in a finite graph $\mathcal{MF}(\Sigma_n)$. Bounds on the diameter of $\mathcal{MF}(\Sigma_n)$ are available when $\Sigma$ is orientable and we provide corresponding bounds when $\Sigma$ is non-orientable. We show that the diameter of this graph grows at least like $5n/2$ and at most like $4n$ as $n$ goes to infinity. If $\Sigma$ is an unpunctured M\"obius strip, $\mathcal{MF}(\Sigma_n)$ coincides with $\mathcal{F}(\Sigma_n)$ and we prove that the diameter of this graph grows exactly like $5n/2$ as $n$ goes to infinity.

Figures

Figures reproduced from arXiv: 2505.04074 by the authors.

**Figure 2.** The contraction of α. T ′ of Σn that is obtained from T by flipping some arc. We say that this flip is incident to α when it modifies the triangle incident to α within T. In other words, such a flip exchanges the diagonals of a quadrilateral incident to α. The following is established in [8] (see Theorem 2.4 therein) in the case when Σ is orientable but its proof works as is in the non-orientable case. Lemma 3.2: Co… view at source ↗

**Figure 3.** If Σ is not one holed we cut it into Σ⊗ and Σ⊙ [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗

Figures from the paper (14 more)

**Figure 4.** Figure 4: The triangulations A− n (first two rows) and A+ n (third row) when n is even (left) and odd (right). The first row shows A− n when Σ is one-holed, and the second when it is not [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]

**Figure 5.** Figure 5: The triangles t −, t +, and t. possible to transform t − into t + by a single flip. As a consequence, at least two flips are incident to α along the considered path. Now assume that the first flip incident to α along that path removes ε. Since Σ ⊗ is one-holed, this fl…

**Figure 6.** Figure 6: Two flips incident to α. not cross any other arc of A− n than δ. In that case, the flip that replaces δ by ϕ is necessarily the one shown on the left of [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]

**Figure 7.** Figure 7: The two flips incident to arc α along the path considered in the proof of Proposition 3.7 [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]

**Figure 8.** Figure 8: The geodesic path in Lemma 3.9. Remark 3.8: Consider some geodesic path in MF(Σn) and some arc flipped along it. If the two triangles incident to this arc are not modified by earlier flips, then this arc can be flipped first along the considered geodesic. In order to p…

**Figure 9.** Figure 9: Three flips considered in the proof of Lemma 3.9. [PITH_FULL_IMAGE:figures/full_fig_p019_9.png]

**Figure 10.** Figure 10: The triangles t and t ′ from the proof of Lemma 4.6 represented within the surface Mq [PITH_FULL_IMAGE:figures/full_fig_p026_10.png]

**Figure 11.** Figure 11: The triangulation Cu(v, w) and four flips in this triangulation, where the introduced arcs are dotted [PITH_FULL_IMAGE:figures/full_fig_p027_11.png]

**Figure 12.** Figure 12: The arc β and the triangle t [PITH_FULL_IMAGE:figures/full_fig_p028_12.png]

**Figure 13.** Figure 13: ∆n+2 (top) and F + (bottom). Cutting T + along γ results in a triangulation U + of a disk ∆n+2 shown in the top row of [PITH_FULL_IMAGE:figures/full_fig_p030_13.png]

**Figure 14.** Figure 14: The unique simplicial triangulation of M [PITH_FULL_IMAGE:figures/full_fig_p031_14.png]

**Figure 15.** Figure 15: The two triangles incident to a non-separating loop [PITH_FULL_IMAGE:figures/full_fig_p032_15.png]

**Figure 16.** Figure 16: The triangulation T ′ (left), a flip in this triangulation shown as a function of the number of non-separating arcs incident to a and b (center), and a triangulation with three pairs of arcs with the same two vertices (right) [PITH_FULL_IMAGE:figures/full_fig_p033_16.png]

**Figure 17.** Figure 17: The disk ∆n+2 [PITH_FULL_IMAGE:figures/full_fig_p034_17.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 17 canonical work pages

[9]

Hugo Parlier and Lionel Pournin, Modular flip-graphs of one holed surfaces , European Journal of Combinatorics 67 (2018), 158–173

work page 2018
[8]

9, 2697–2737

Hugo Parlier and Lionel Pournin, Flip-graph moduli spaces of filling surfaces , Journal of the European Mathematical Society 19 (2017), no. 9, 2697–2737

work page 2017
[1]

A logarithmic bound for the chromatic number of the associahedron

Louigi Addario Berry, Bruce Reed, Alex Scott, and David R. Wood, A logarithmic bound for the chromatic number of the associahedron , arXiv:1811.08972 (2018)

work page Pith review arXiv 2018
[2]

Bell, Simplifying triangulations, Discrete & Computational Geometry 66 (2021), no

Mark C. Bell, Simplifying triangulations, Discrete & Computational Geometry 66 (2021), no. 1, 1–11

work page 2021
[3]

6, 3809– 3844

Valentina Disarlo and Hugo Parlier, The geometry of flip graphs and mapping class groups, Transactions of the American Mathematical Society 372 (2019), no. 6, 3809– 3844

work page 2019
[4]

Edelman and Victor Reiner, Catalan triangulations of the M¨ obius band, Graphs and Combinatorics 13 (1997), no

Paul H. Edelman and Victor Reiner, Catalan triangulations of the M¨ obius band, Graphs and Combinatorics 13 (1997), no. 3, 231–243

work page 1997
[5]

2, 189–194

Allen Hatcher, On triangulations of surfaces , Topology and its Applications 40 (1991), no. 2, 189–194

work page 1991
[6]

Lee, The associahedron and triangulations of the n-gon , European Journal of Combinatorics 10 (1989), no

Carl W. Lee, The associahedron and triangulations of the n-gon , European Journal of Combinatorics 10 (1989), no. 6, 551–560

work page 1989

Show all 18 references

[7]

2, 303–384

Lee Mosher, Mapping class groups are automatic , Annals of Mathematics 142 (1995), no. 2, 303–384

1995
[10]

Hugo Parlier and Lionel Pournin, Once punctured disks, non-convex polygons, and pointi- hedra, Annals of Combinatorics 22 (2018), 619–640

2018
[11]

1, 45–63

Hugo Parlier and Lionel Pournin, Counting geodesics between surface triangulations , Mathematical Proceedings of the Cambridge Philosophical Society 178 (2025), no. 1, 45–63

2025
[12]

Lionel Pournin, The diameter of associahedra , Advances in Mathematics 259 (2014), 13–42

2014
[13]

2, 111–129

Lionel Pournin, Eccentricities in the flip-graphs of convex polygons , Journal of Graph Theory 92 (2019), no. 2, 111–129

2019
[14]

Lionel Pournin and Zili Wang, Strong convexity in flip-graphs, arXiv: 2106.08012 (2021)

2021 arXiv
[15]

3, 647–681

Daniel Sleator, Robert Tarjan and William Thurston, Rotation distance, triangulations, and hyperbolic geometry, Journal of the American Mathematical Society 1 (1988), no. 3, 647–681

1988
[16]

I, Transactions of the Amer- ican Mathematical Society 108 (1963), no

James Dillon Stasheff, Homotopy associativity of H-spaces. I, Transactions of the Amer- ican Mathematical Society 108 (1963), no. 2, 275–292

1963
[17]

II, Transactions of the Amer- ican Mathematical Society 108 (1963), no

James Dillon Stasheff, Homotopy associativity of H-spaces. II, Transactions of the Amer- ican Mathematical Society 108 (1963), no. 2, 293–312

1963
[18]

Dov Tamari, Mono¨ ıdes pr´ eordonn´ es et chaˆ ınes de Malcev, Th` ese de Math´ ematiques, Paris, 1951

1951

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [9]

Hugo Parlier and Lionel Pournin, Modular flip-graphs of one holed surfaces , European Journal of Combinatorics 67 (2018), 158–173

work page 2018

[2] [8]

9, 2697–2737

Hugo Parlier and Lionel Pournin, Flip-graph moduli spaces of filling surfaces , Journal of the European Mathematical Society 19 (2017), no. 9, 2697–2737

work page 2017

[3] [1]

A logarithmic bound for the chromatic number of the associahedron

Louigi Addario Berry, Bruce Reed, Alex Scott, and David R. Wood, A logarithmic bound for the chromatic number of the associahedron , arXiv:1811.08972 (2018)

work page Pith review arXiv 2018

[4] [2]

Bell, Simplifying triangulations, Discrete & Computational Geometry 66 (2021), no

Mark C. Bell, Simplifying triangulations, Discrete & Computational Geometry 66 (2021), no. 1, 1–11

work page 2021

[5] [3]

6, 3809– 3844

Valentina Disarlo and Hugo Parlier, The geometry of flip graphs and mapping class groups, Transactions of the American Mathematical Society 372 (2019), no. 6, 3809– 3844

work page 2019

[6] [4]

Edelman and Victor Reiner, Catalan triangulations of the M¨ obius band, Graphs and Combinatorics 13 (1997), no

Paul H. Edelman and Victor Reiner, Catalan triangulations of the M¨ obius band, Graphs and Combinatorics 13 (1997), no. 3, 231–243

work page 1997

[7] [5]

2, 189–194

Allen Hatcher, On triangulations of surfaces , Topology and its Applications 40 (1991), no. 2, 189–194

work page 1991

[8] [6]

Lee, The associahedron and triangulations of the n-gon , European Journal of Combinatorics 10 (1989), no

Carl W. Lee, The associahedron and triangulations of the n-gon , European Journal of Combinatorics 10 (1989), no. 6, 551–560

work page 1989