REVIEW 3 major objections 5 minor 17 references
Tilings of the sphere by congruent pentagons V: Edge combination $a^{4}b$ with rational angles
T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read The paper proves a complete classification of edge-to-edge tilings of the sphere by congruent $a^4b$ pentagons with all angles rational in degrees, with three explicit families.
desk verdict This is a verification of the a^4b classification from [6] by a different cyclotomic method, with added geometric data and pictures; the argument is credible but the 'all' statement depends on unverified CAS claims of no cyclotomic roots. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The argument is carried by a small collection of tiling lemmas plus an algebraic reduction. The parity lemma forces the number of ab-angles at every vertex to be even; the balance lemma restricts how the angles $\delta$ and $\epsilon$ can cluster; and the special-tile lemma guarantees a tile whose five vertices are nearly all of degree 3. From these, the adjacent-angle deduction (AAD) technique bookkeeps forced angle and edge arrangements around vertices and rules out countless configurations. The pivotal analytical tool is equation (2.6), a trigonometric identity that every $a^4b$-tiling's angles must satisfy; substituting $x=e^{i\theta}$ turns it into a polynomial whose rational-angle solutions are exactly cyclotomic points. The authors solve these by factoring over cyclotomic fields, taking norms to $\mathbb{Q}[x]$, applying the standard gcd tests on $f(x),f(-x),f(x^2),f(-x^2)$ to decide whether any factor is cyclotomic, and computing resultants to reduce two- and three-variable polynomials to one variable. The case analysis organizes the 258 vertex-type combinations inherited from the companion paper into three-$a^2b$-vertex, two-$a^2b$-vertex, and one-$a^2b$-vertex cases, where an $a^2b$-vertex is a vertex at which two $a$-edges and one $b$-edge meet; after filtering by simplicity and by the parity, balance, and AAD constraints, only the listed pentagons survive.
What would settle it
Re-run the appendix's cyclotomic-root computations with independent computer algebra and certified gcd tests; a single factor declared 'no cyclotomic root' that actually has one, followed through the corresponding linear angle constraints to a simple pentagon with $f\ge12$ even and $\beta\neq\gamma$, $\delta\neq\epsilon$, would be a counterexample. Conversely, an exhaustive search over rational angles up to a chosen denominator that found an $a^4b$ tiling outside the three classes would refute the theorem.
Extended reading notes
Core claim
On the paper's own terms, the central result is an exhaustive list. Every edge-to-edge tiling of the sphere by congruent $a^4b$ pentagons with rational angles falls into exactly one of: (1) the one-parameter family of symmetric $a^4b$-pentagonal subdivisions of the tetrahedron with 12 tiles; (2) for each $m\ge4$, the unique symmetric pentagon with angles $(\frac8f,1-\frac4f,1-\frac4f,\frac12+\frac2f,\frac12+\frac2f)\pi$, $f=4m$, with side cosines $\cos a=1-2\left(\frac{\sqrt5-1}{4}\cos\frac{4\pi}{f}\right)^2$ and $\cos b=2\left(\frac{(3-\sqrt5)\cos^2\frac{4\pi}{f}+\sqrt5-2}{\cos\frac{4\pi}{f}}\right)^2-1$, each admitting a symmetric 3-layer earth map tiling, plus two standard flip modifications when $m$ is odd; and (3) the unique non-symmetric degenerate pentagon with angles $(10,12,6,5,15)\pi/15$, $\cos a=\frac{\sqrt6(5+\sqrt5)^{3/2}}{60}$, $\cos b=\frac{\sqrt5}{3}$, admitting a 20-tile non-symmetric earth map tiling and a unique flip modification. The authors further report that among the 43 rational pentagons satisfying the basic combinatorial constraints inherited from the 258 vertex-type cases, only 12 are simple and only this last non-symmetric pentagon admits tilings beyond the symmetric families.
Load-bearing premise
The theorem's 'all' rests on the completeness of the 258 vertex-type enumeration taken from the companion paper and on the correctness of dozens of computer algebra claims that certain resultants and factors have no cyclotomic roots; if either assumption fails, an unlisted tiling could exist.
Editorial extensions
If this is right
- The classification is exhaustive: any edge-to-edge tiling of the sphere by congruent rational-angle $a^4b$ pentagons must appear in one of the three listed classes.
- Combined with the companion general-angle paper, the rational and irrational $a^4b$ cases together close the $a^4b$ problem and provide a shorter independent verification of the earlier long preprint.
- Every prototile in the list carries explicit angle and side-length data, so the total number of distinct tilings for each rational-angle prototile can be counted from the displayed vertex-type data.
- Because the only non-symmetric rational pentagon has one angle $\pi$, degenerating it yields new non-edge-to-edge quadrilateral tilings of the sphere.
Reading between the lines
- If the completeness proof is right, the rational-angle constraint is extremely selective: from 258 vertex-type combinations and 43 candidate pentagons only one non-symmetric simple pentagon survives to admit tilings, suggesting that rational angles almost never coexist with non-symmetric $a^4b$ geometry.
- The same norm-to-$\mathbb{Q}$ and resultant pipeline could be run fully automatically with independently generated certificates, turning the appendix's repeated 'no cyclotomic root' checks into a formally checkable proof object.
- The method should transfer to the remaining edge combinations: one could ask whether rational-angle tilings for $a^2b^2c$, $a^3bc$, $a^3b^2$, and $a^5$ satisfy comparably sparse classification statements, and whether degeneration of those families produces further quadrilateral tilings.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This manuscript completes the classification of edge-to-edge spherical tilings by congruent pentagons with edge combination a^4b and all angles rational in degrees. The main theorem asserts three families: a one-parameter symmetric subdivision of the tetrahedron with 12 tiles; a sequence of unique symmetric pentagons admitting symmetric 3-layer earth map tilings with 4m tiles for m≥4, with standard flip modifications for odd m; and a unique non-symmetric degenerate pentagon admitting two 20-tile tilings. The proof route is to derive the trigonometric identity Eq. (2.6) from known identities, convert the search for rational angle solutions into a search for cyclotomic points on polynomials, enumerate the possible a^2b-vertex types from the companion paper [14], and solve the resulting systems by resultants and the Bradford–Davenport root-of-unity test. The final sections summarize the broader classification of monohedral pentagonal spherical tilings and list induced non-edge-to-edge quadrilateral tilings.
Significance. If the classification is correct, it closes the rational-angle a^4b case and, together with the companion papers, completes the classification of monohedral pentagonal tilings of the sphere. The paper provides explicit prototile data, counts of tilings, and 3D pictures, and it derives several new non-edge-to-edge quadrilateral tilings from degenerate pentagons. The explicit angle and edge-length formulas, such as the symmetric family with angle sum 3+4/f and the closed forms for cos a and cos b, are concrete and checkable. The main weakness is not in the derivation of Eq. (2.6) but in the auditability of the massive computational case analysis that underpins the word 'all' in the theorem.
major comments (3)
- [Theorem, §4, Appendix] The completeness of the Theorem rests on dozens of unverified assertions in the Appendix that certain resultant polynomials have no cyclotomic roots. For example, in the Appendix's Case {αδϵ, β2γ}, Subcase 10, a resultant polynomial spanning several pages is printed and followed only by 'None', with no factorization, no trace of the Bradford–Davenport test, and no certificate or code. Since any root of unity in such a resultant would yield a rational angle quintuple satisfying the linear vertex constraints and Eq. (2.6), a single missed cyclotomic factor would add a prototile not listed in the Theorem. The same issue occurs in Example 3.1, items (1), (5), and (7). Please supply an executable script or certified factorizations for every 'None' answer, or at least factor the printed polynomials into cyclotomic and non-cyclotomic parts.
- [§4 (opening), [14, Tables 4-6, 10-12, 14, 21, 23, 28, 30]] The classification is also conditional on the completeness of the vertex-type enumeration in [14], which is not reproduced in this manuscript. The sentence 'Per [14] ... the 258 vertex combinations yield 43 rational pentagons' transfers a large and essential part of the 'all' statement to an overlapping-author preprint. Because the current theorem is an absolute classification, the precise logical dependence should be stated, and ideally the relevant enumeration should be summarized or independently verified.
- [Example 3.2, Example 3.3, and Appendix cases with coefficients in Q(ζ_n)] For polynomials with coefficients in cyclotomic fields, the reduction to Q[x] is not shown. In Example 3.2 the norm N(L) is displayed, but the specialized univariate polynomial N(L)(x, e^{iπ/5}) is not given; the text only asserts that it has no cyclotomic root. In Example 3.3, the text says that solving all 15 resultants yields four rational angle sets, but only one resultant is displayed and the remaining computations are relegated to raw, unfactored form in the Appendix. Without these intermediate data, the claimed absence of cyclotomic points cannot be checked from the manuscript alone, even in principle.
minor comments (5)
- [Table 1] The Contradiction column lacks line breaks; entries such as 'No rational solutionαδϵ, βδϵ, γϵ2' run together and should be reformatted for readability.
- [Lemma 8] The sentence 'The proof’s last line is verifiable symbolically (e.g., Maple)' should be replaced by the actual identity or an explicit computer-algebra check, since the displayed derivation omits the final simplification.
- [Introduction and §3] The terminology 'rational angles in degree' and 'with any irrational angle' should be clarified: the main theorem concerns angles that are rational multiples of π, while [14] treats the complementary case; the current wording can be misread.
- [Figure 6 and Section 4 tables] The twelve simple pentagons in Figure 6 are not always cross-referenced to the specific rows of Tables 2–6 that produce them; adding such cross-references would aid verification.
- [Appendix] Many Appendix blocks do not state the variable substitutions used to obtain the displayed polynomial; adding a one-line definition of x and y for each case would make the computations reproducible from the text.
Circularity Check
No significant circularity: the rational-angle a4b classification is derived from external geometric identities and a root-of-unity search, with no fitted parameter renamed as a prediction.
full rationale
The derivation chain is: Lemma 7 (cited from [6]) gives the trigonometric conditions (2.3)-(2.5) for an almost equilateral pentagon; Lemma 8 derives the angle-only equation (2.6) from it inside any a4b tiling. Section 3 converts (2.6) into polynomial equations in roots of unity and solves them by resultants and cyclotomic-root algorithms. Section 4 imports from [14] the combinatorial enumeration of a2b-vertex types, solves the corresponding linear angle sums, and rejects cases either by 'None' (no cyclotomic roots) or by parity, balance, and adjacent-angle-deduction arguments. The surviving simple rational pentagons are then matched to explicit tilings, giving the theorem's list. No step fits a parameter from the claimed output, and no quantity is defined in terms of the conclusion. The first two theorem classes are the symmetric families already classified in [14], and the third is the unique rational angle point from [14]'s non-symmetric family; citing [14] is importing a different, broader classification as a lemma, not assuming the rational-angle theorem. The 'no cyclotomic root' assertions in the Appendix are computational rather than circular: a missed root would falsify completeness, but would not make the proof self-referential. Hence no circularity is present.
Assumptions & free parameters
assumptions (6)
- domain assumption The only possible edge combinations for edge-to-edge congruent pentagonal spherical tilings are a2b2c, a3bc, a3b2, a4b, a5.
- domain assumption Angle sum, special tile, parity, and balance lemmas about a^4b tilings (Lemmas 1-6).
- domain assumption Lemma 7: necessary and sufficient trigonometric equations for an almost equilateral pentagon with given angles and edge length a.
- domain assumption Theorems 1 and 2 of [14]: classification of a^4b tilings with general angles and the enumeration of possible a2b-vertex types.
- standard math Correctness of the Bradford-Davenport, Beukers-Smyth, and Aliev-Smyth algorithms for finding cyclotomic points, and of their implementation in SageMath/Maple.
- standard math The resultant and norm computations over cyclotomic fields in the Appendix are error-free as performed.
Cite this review
Pith. "Pith review of Tilings of the sphere by congruent pentagons V: Edge combination $a^{4}b$ with rational angles." pith.science (2026). https://pith.science/paper/3B5PJEFU
@misc{pith2026250707038,
author = {Pith},
title = {Pith review of: Tilings of the sphere by congruent pentagons V: Edge combination $a^4b$ with rational angles},
year = {2026},
howpublished = {\url{https://pith.science/paper/3B5PJEFU}},
note = {Machine review of arXiv:2507.07038}
}
abstract
We classify edge-to-edge tilings of the sphere by congruent pentagons with the edge combination $a^4b$ and with rational angles in degree: they are a one-parameter family of symmetric $a^4b$-pentagonal subdivisions of the tetrahedron with $12$ tiles; a sequence of unique symmetric $a^4b$-pentagons admitting a symmetric $3$-layer earth map tiling by $4m$ tiles for any $m\ge4$, among which each odd $m$ case admits two standard flip modifications; and a unique non-symmetric and degenerate $a^4b$-pentagon admitting a non-symmetric $3$-layer earth map tiling and its standard flip modification with $20$ tiles. The full classification from this series and all induced non-edge-to-edge quadrilateral tilings from degenerate pentagons are summarized with their 3D pictures.
Figures
Figures from the paper (9 more)
Reference graph
Works this paper leans on
-
[6]
H. M. Cheung, H. P. Luk, M. Yan. Tilings of the sphere by congruent pentagons IV: Edge combination a4b. preprint, arXiv:2307.11453, 2023
arXiv 2023
-
[14]
J. Shu, Y. Liao, E. Wang. Tilings of the sphere by congruent pentagons IV: edge combination a4b with general angles. preprint, arXiv:2412.08492, 2024
work page Pith review arXiv 2024
-
[1]
C. Adams. The Tiling Book: An Introduction to the Mathematical Theory of Tilings. American Mathematical Society, Providence, RI, 2022
work page 2022
- [2]
- [3]
-
[4]
F. Beukers, C. J. Smyth. Cyclotomic points on curves. In: Number theory for the millennium , vol. I (Urbana, IL, 2000), A K Peters, Natick, MA, 2002, pp. 67–85
work page 2000
-
[5]
R. J. Bradford, J. H. Davenport. Effective tests for cyclotomic polynomials. In: Symbolic and algebraic computation (Rome, 1988), Lect. Notes Comput. Sci. , vol. 358, Springer, Berlin, 1989, pp. 244– 251
work page 1988
-
[7]
B. Gr¨ unbaum, G. C. Shephard.Tilings and Patterns . W. H. Free- man, New York, 1987 (Reprinted by Dover, 2016)
work page 1987
Show all 17 references
-
[8]
K. S. Kedlaya, A. Kolpakov, B. Poonen, M. Rubinstein. Space Vectors Forming Rational Angles. preprint, arXiv:2011.14232, 2020
2011 arXiv
-
[9]
S. Lang. Algebra, 3rd ed. Graduate Texts in Mathematics, vol. 211, Springer, New York, 2002
2002
-
[10]
Liang, E
J. Liang, E. Wang, M. Yan. Moduli of pentagonal subdivision tilings. preprint, arXiv:1907.08776, 2024
1907 arXiv
-
[11]
Y. Liao, P. Qian, E. Wang, Y. Xu. Tilings of the sphere by con- gruent quadrilaterals I: edge combination a2bc. Chin. Ann. Math. Ser. B , 45(5):733–766, 2024
2024
-
[12]
Y. Liao, E. Wang. Tilings of the sphere by congruent quadrilaterals II: edge combination a3b with rational angles. Nagoya Math. J. , 253:128–163, 2024
2024
-
[13]
Y. Liao, P. Qian, E. Wang, Y. Xu. Tilings of the sphere by con- gruent quadrilaterals III: edge combination a3b with general angles. Forum Math., 36(5):1159–1186, 2024. TILINGS OF THE SPHERE BY CONGRUENT PENTAGONS V 23
2024
-
[15]
E. Wang, M. Yan. Tilings of the sphere by congruent pentagons I: edge combinations a2b2c and a3bc. Adv. Math., 394:107866, 2022
2022
-
[16]
E. Wang, M. Yan. Tilings of the sphere by congruent pentagons II: edge combination a3b2. Adv. Math., 394:107867, 2022. 24 JINJIN LIANG, YIXI LIAO, WENCHUAN HU, AND ERXIAO W ANG Appendix: This Appendix finds all rational non-symmetric pentagons ( β ̸= γ, δ ̸= ϵ) with f ≥ 12 eve...
2022
-
[17]
= 0. None. Case {βδϵ, α2γ, αγ2δ2}, (x = e iϵ+ 2iπ f , y = e iϵ− 2iπ f ). (x + y)(ζ4x7y − ζ4x6y2 − ζ4x5y3 − ζ4x4y4 + ζ4x3y5 + 4x5y2 − 3x4y3 + x3y4 + xy6 + ζ4x6 + ζ4x4y2 − 3ζ4x3y3 + 4ζ4x2y4 + x4y − x3y2 − x2y3 − xy4 + y5) = 0. None. Case {βδϵ, α2γ, αγ2δϵ}, (x = e 2iδ, y= e 4iπ f...
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.