Galois lines for a canonical curve of genus 4, III: non-cyclic Galois lines

arXiv:2604.26584 · math.AG · Submitted 2026-04-29 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.

Ada: We're looking at the paper: "Galois lines for a canonical curve of genus 4, III".

Paper summary: Emmy: So wrapping up this discussion on "Galois lines for a canonical curve of genus four III: non-cyclic Galois lines," we see that the authors successfully proved the upper bounds for S3-lines at ten and K4-lines at fifteen <ref:2604.26584#pg2>.

Kolya: The work is interesting because it moves away from just listing existing examples and instead provides a proof based on analyzing how subgroups are represented in the projective group GL(four C) <ref:2604.26584#pg3>.

Ada: Exactly. The authors show that the structure of the automorphism group, specifically how it relates to the kernel of a map phi, dictates whether a line is an S3-line or K4-line <ref:2604.26584#pg1>.

Emmy: The key finding for us is that this structural analysis successfully leads to the explicit bounds, and they even found an example curve with an automorphism group of S5 that hits those exact maximum counts <ref:2604.26584#pg2>.

Kolya: It’s a solid piece of work because it gives us concrete numbers, and the method used is sound—it proves the limits without needing prior classification tables or massive computer enumeration <ref:2604.26584#pg3>.

Ada: In simple terms, the paper tells us that for these genus four curves, the complexity of these specific Galois lines is limited by the structure of their symmetry group <ref:2604.26584#pg1>.

Emmy: It shows that understanding how those automorphisms act on the curve is a more powerful tool for counting geometric objects than just looking at the geometry in isolation <ref:2604.26584#pg3>.

Conclusion: Ada: So we're wrapping up on this paper, "Galois lines for a canonical curve of genus four III: non-cyclic Galois lines."

Emmy: Basically, they finished proving the upper limits for those S3 and K4 lines on these genus four curves.

Ada: It’s important to remember the authors are looking at non-cyclic Galois lines specifically.

Kolya: They're dealing with these specific groups, the symmetric group three and the Klein four-group.

Emmy: And they show how those constraints on the curve's automorphism group directly limit how many of these lines can exist.

Ada: The implication is that you can get very tight numbers for geometric objects just by looking at the underlying algebraic structure, like that kernel of a map phi.

Kolya: That’s what I mean, it suggests this isn't just counting examples; it’s about understanding the shape of the symmetry group itself.

Ada: It really boils down to how those subgroups are represented in that projective space GL four C.

Emmy: Which is exactly what they did: they looked at the structure of S3 and K4 subgroups and found they must fit into certain types, like G two by three or G three by two.

Kolya: So the bound isn't just some arbitrary number; it’s tied to how these specific subgroups can actually show up geometrically on the curve.

Ada: That’s right, and the example they gave with an S five automorphism group actually hits those upper bounds perfectly.

Emmy: It shows that you can have a complex symmetry group and still be limited in how many of these specific Galois lines you can find.

Kolya: So what does this mean for us? It means we have a framework to predict how many of these non-cyclic lines should exist based on the curve's basic algebraic properties.

SHOTARO KATO, JIRYO KOMEDA, TAKESHI TAKAHASHI

Niigata University · Kanagawa Institute of Technology

math.AG

Submitted: 2026-04-29

Updated: 2026-11-18

Comments: 25 pages. Submitted for publication. v2: Expanded introduction and examples. Added a GAP appendix reproducing the computation of all S3-lines and K4-lines. Corrected minor errors and typos

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The gist: Let C be a canonical curve of genus 4 over an algebraically closed field k of characteristic zero, and let l be a line in P3; A line l is called an S3-line (resp.

Key concepts

S3-line
A line in projective space where the extension of function fields from a genus 4 curve to the line is Galois with a symmetric group S3 as its Galois group. This means the symmetries of the curve over that line correspond to permutations of three elements.
K4-line
A line where the extension of function fields is Galois with the Klein four-group K4 as its Galois group. This involves symmetries corresponding to a specific set of four non-identity, commuting involutions (elements of order 2).
Galois Line
A line $l$ such that the field extension $k(C)/ ext{P}^*l(k( ext{P}^1))$ is Galois. This means the algebraic structure relating the curve's function field to the line's function field is symmetric, allowing for predictable symmetries in how points on the curve relate to points on the line.
Gonality
The gonality of a curve is related to its minimal degree linear system. The paper finds that for the specific curve studied, its gonality is 3, which is relevant when analyzing potential Galois lines.

Terminology

Summary

Let C be a canonical curve of genus 4 over an algebraically closed field k of characteristic zero, and let l be a line in P3; A line l is called an S3-line (resp. p. 0 a K4-line) if the extension k(C)/π∗l(k(P1)) is Galois and its Galois group is isomorphic to the symmetric group S3 on three letters (resp. p. 0 the Klein four-group K4); The number of S3-lines is at most 10, and the number of K4-lines is at most 15

Introduction to Galois Lines

The paper investigates non-cyclic Galois lines for canonical curves of genus 4, specifically focusing on S3-lines and K4-lines, and establishes upper bounds for their numbers In [6], the authors studied non-skew cyclic Galois lines and proved that the number of C4-lines and C5-lines is at most one They also investigated skew cyclic Galois lines, proving that the number of such lines is equal to 0, 1, 3, or 9 The present work turns to non-cyclic Galois lines for canonical curves of genus 4 and establishes upper bounds for S3-lines and K4-lines

Characterization of Galois Lines

A line l is called a Galois line if the extension k(C)/π∗l(k(P1)) is Galois, and it is an S3-line (resp. a K4-line) if its Galois group is isomorphic to the symmetric group S3 on three letters (resp. p. 0 the Klein four-group K4) For a line l, if it is a Galois line, Gl is defined as the group Gl:= The group Gl:= The group Gl:= The group Gl of automorphisms of C such that πl ◦ σ = πl Proposition 2. p. 01 states that the curve C is a (2, 3)-complete intersection and its gonality gon(C) is equal to 3

Bounds for S3-lines

The proof of Theorem 1.1 divides the argument into several cases, starting with the number of S3-lines when rank Q = 3 If Ker φ = 1, then G ∼= Im φ is isomorphic to one of Cm, Dm, A4, S4, or A5 The number of C3 subgroups contained in S3 subgroups of Cm (resp. p. 6 Dm, A4, S4) is equal to 0 (resp. p. 4 at most 1). Using Proposition 2. p. 65, the number of S3-lines is at most 10 If Ker φ ≠ 1, then the number of S3-lines is at most 4

Bounds for K4-lines

For the number of K4-lines when rank Q = 3, Proposition 3.9 states that the number of K4-lines is at most 5 If Ker φ = 1, the number of K4 subgroups of Cm (resp. p. 6 Dm, A4, S4) equals 0 (resp. p. 9 0 or m/2, 1, 4). Hence, if Im φ ≠ Dm or Im φ ∼= Dm where m is odd, then the number of K4-lines is at most 5 If Ker φ ≠ 1, the number of K4-lines is at most 9

Example Attaining Maximal Numbers

The paper presents an explicit example of a canonical curve C defined by Q = XW + Y Z = 0 and F = X2Z − Y2X − Z 2W + W 2Y = 0, which has Aut(C) ∼= S5 By explicit computer-assisted computation, this curve C admits exactly ten distinct S3-lines and fifteen distinct K4-lines This example simultaneously attains the upper bounds for the numbers of S3-lines and K4-lines given in Theorem 1.1

Summary of Method

The method used to determine all S3-lines and K4-lines on the curve C involves four steps: Step 1 is the identification of automorphisms, where Aut(C) is isomorphic to G(120) ∼= S5 Step 2 involves enumerating candidate subgroups isomorphic to S3 or K4, finding that every S3-subgroup is conjugate in GL(4, C) to one of G(2 × 3), G(2 × 3, 2 × 6), or G(3 × 2) Step 3 involves selecting Galois subgroups based on Proposition 2.4 and Proposition 2. p. 16, where an S3-subgroup corresponds to an S3-line if it is conjugate in P GL(4, C) to the group defined by the representation (2) Step 4 involves recovering the Galois lines from fixed-point data, where for an S3-subgroup Gl, one chooses an element σ of order 3 and computes its fixed locus Fix(σ), which is the union of a line and two points For a K4-subgroup Gl, one chooses two elements τ1, τ2 of order 2 with nonzero trace, and the line joining their fixed points P1 and P2 is the corresponding K4-line

Conclusion on Cyclic Lines

For the curve C considered here, there is no cyclic Galois line because rank Q = 4, so that C admits exactly two trigonal pencils, and by Propositions 3.8 and 3.13, the two trigonal pencils cannot be Galois simultaneously Furthermore, since C admits a K4-line, Lemma 3. p. 811 implies that the two trigonal pencils are either both Galois or both non-Galois, which leads to neither of the two trigonal pencils being Galois Moreover, by [6, Corollary 3. p. 85], the existence of a C4-line implies rank Q = 3, and since rank Q = 4 for this curve there is no C4-line Similarly, [6, Corollary 3. p. 18] shows that there is no C5-line, and [7, Corollary 2. p. 26] shows that there is no C6-line Hence the curve C admits no cyclic Galois line

References

[1] E. Arbarello, M. Cornalba, P. A. Griffiths, and J. Harris, Geometry of algebraic curves [3] P. p. 0 p. 0 A. Griffiths and J. Harris, Principles of algebraic geometry [8] I. Kuribayashi and A. p. 1 p. 1 p. 1 p. 1 p. 1 p. 1 p. 1 p. 1 Kuribayashi, On automorphism groups of compact Riemann surfaces of genus 4 [9] I. p. 1 p. 1 p. 1 p. 1 Kuribayashi and A. Kuribayashi, Automorphism groups of compact Riemann surfaces of genera three and four [10] H. p. 1 p. 1 p. 1 p. 1 Yoshihara, Galois lines for space curves [2] The GAP Group, GAP – Groups, Algorithms, and Programming

Email Addresses

Email address: shotarokato0308@gmail.com Email address: komeda@gen. p. 0 p. 0kanagawa-it.ac.jp Email address: takeshi@eng. p. 0 p. 0niigata-u.ac.jp

Acknowledgements

The authors would like to thank Professor Akira Ohbuchi of Tokushima University for valuable discussions and helpful advice on the computational approach used in the work of the first author

References List

[1] E. Arbarello, M. Cornalba, P. A. Griffiths, and J. Harris, Geometry of algebraic curves [3] P. A. Griffiths and J. Harris, Principles of algebraic geometry [8] I. Kuribayashi and A. Kuribayashi, On automorphism groups of compact Riemann surfaces of genus 4 [9] I. Kuribayashi and A. Kuribayashi, Automorphism groups of compact Riemann surfaces of genera three and four [10] H. Yoshihara, Galois lines for space curves [2] The GAP Group, GAP – Groups, Algorithms, and Programming [4] S Kato, S3 Galois lines and K4 Galois lines for canonical curves of genus 4. p. 0 p. 0 Senior Thesis, Faculty of Engineering, Niigata University, 2024 (in Japanese) [5] S Kato, Galois lines associated with non-cyclic groups for canonical curves of genus 4. p. 0 p. 0 Master’s thesis, Graduate School of Science and Technology, Niigata University, 2026 (in Japanese) [6] J. p. 0 p. 0 Komeda and T. p. 0 p. 0 p. 0 p. 0 Takahashi, Galois lines for a canonical curve of genus 4, I: Non-skew cyclic lines. p. 0 p. 0 Rend. Semin. Mat. Univ. Padova 152 (2024), 1–20 [7] J. Komeda and T. Takahashi, Galois lines for a canonical curve of genus 4, II: Skew cyclic lines [8] I. p. 0 p. 0 Kuribayashi and A. Kuribayashi, On automorphism groups of compact Riemann surfaces of genus 4 [9] I. Kuribayashi and A. Kuribayashi, Automorphism groups of compact Riemann surfaces of genera three and four [10] H. Yoshihara, Galois lines for space curves [2] The GAP Group, GAP – Groups, Algorithms, and Programming [4] S Kato, S3 Galois lines and K4 Galois lines for canonical curves of genus 4. Senior Thesis, Faculty of Engineering, Niigata University, 2024 (in Japanese) [5] S Kato, Galois lines associated with non-cyclic groups for canonical curves of genus 4. Master’s thesis, Graduate School of Science and Technology, Niigata University, 2026 (in Japanese) [6] J. Komeda and T. Takahashi, Galois lines for a canonical curve of genus 4, I: Non-skew cyclic lines. Rend. Semin. Mat. Univ. Padova 152 (2024), 1–20 [7] J. Komeda and T. Takahashi, Galois lines for a canonical curve of genus 4, II: Skew cyclic lines [8] I. Kuribayashi and A. Kuribayashi, On automorphism groups of compact Riemann surfaces of genus 4 [9] I. Kuribayashi and A. Kuribayashi, Automorphism groups of compact Riemann surfaces of genera three and four [10] H. Yoshihara, Galois lines for space curves [2] The GAP Group, GAP – Groups, Algorithms, and Programming [3] P. p. 1 A. Griffiths and J. Harris, Principles of algebraic geometry

Email address: shotarokato0308@gmail.com Email address: komeda@gen.kanagawa-it.ac.jp Email address: takeshi@eng.niigata-u.ac.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. System can determine if a given line l in P3 is an S3-line or K4-line for a canonical curve C of genus 4 by checking if its induced extension k(C)/pi∗l(k(P1)) is Galois and identifying the Galois group as S3 or K4.

  2. System can calculate the maximum number of S3-lines (at most 10) and K4-lines (at most 15) for any given canonical curve C of genus 4 without relying on prior classification tables or computer calculations.

  3. System can classify the structure of Galois lines by analyzing a finite subgroup G of Aut(C) preserving the projection, using Lemma 3.1 to establish a short exact sequence and restricting the image Im phi to known groups like Cm, Dm, A4, S4, or A5.

  4. System can determine whether a K4-line l for C has an element in Gl with trace zero by checking if its fixed locus Fix(sigma) is the union of two lines as described in Lemma 2.7.

  5. System can distinguish between different S3-lines for a given canonical curve C by comparing the traces of elements of their associated Galois groups, using Proposition 2.5 to show that if l != l', then Gl != Gl'.

  6. System can determine the number of S3-lines for rank Q=3 curves by checking if Ker phi is trivial or non-trivial, leading to bounds of at most 10 or 4 respectively.

  7. System can determine the number of K4-lines for rank Q=3 curves by analyzing the structure of H generated by elements with trace zero, concluding that the number is at most 5.

  8. System can determine the number of K4-lines for rank Q=4 curves based on whether Ker phig and Ker ph are trivial or non-trivial, yielding bounds of at most 15 or 9 respectively.

  9. System can identify the automorphism group Aut(C) for a given canonical curve C by utilizing the classification of Kuribayashi–Kuribayashi to determine it is isomorphic to G(120) ∼= S5 for the example curve provided.

  10. System can verify if a specific line l is an S3-line by checking if its associated Galois group Gl is generated by elements matching the structure in Proposition 2.4, which involves matrices with specific forms and primitive cubic roots of unity.

Related papers