An exact six-period correction to the printed elliptic billiard assertion k108
Read the paper (PDF) · Download PDF · Supporting ZIP · Zenodo record
Plain-English explanation · Read on this page · Collaboration
Abstract
The experimental entry k108 in Table 2 of Reznik–Garcia–Koiller asserts constancy of the outer-tangent area ratio divided by the product of original half-angle sines for N ≡ 2 (mod 4). Two strictly convex primitive six-period trajectories on one confocal billiard family give 11664/3125 and 3645/1024, disproving the printed assertion. The horizontal six-period coordinates and caustic are prior work of Garcia–Reznik, printed in their November 2020 preprint. This note is an attributed corrective application of established geometry, with exact reproducibility certificates; it claims neither new six-period geometry nor historical firstness.
The idea in plain English
A billiard trajectory inside an ellipse traces a polygon as it reflects from the boundary. Some expressions stay constant when the starting point moves through a family of closed trajectories. A printed list proposed one such expression, called k108, for periods congruent to 2 modulo 4.
This note checks two six-bounce trajectories in one connected family. Their k108 values are different exact fractions. That disproves the particular printed experimental assertion, whose proof column was marked with a question mark.
The precise result
Use the ellipse x²/4 + y² = 1 and the fixed confocal caustic x²/(32/9) + y²/(5/9) = 1. The note gives two strictly convex, primitive, physical six-period trajectories, H and V. Both have perimeter 28/3 and incoming Joachimsthal constant 1/3.
Write A for the area of the original trajectory polygon and A′ for the area of the polygon formed by consecutive tangents to the ellipse at its vertices. The printed expression is k108 = (A′/A)/Π sin(θᵢ/2). Here θᵢ is the ordinary internal angle of the original polygon; its cosine sum agrees with the source normalization JL − 6.
The exact evaluations below differ by 553311/3200000 > 0. Since both trajectories lie in one connected oriented six-period family and 6 ≡ 2 mod 4, they refute the stated constancy.
| Exact quantity | Horizontal H | Vertical V |
|---|---|---|
| Outer-tangent area ratio A′/A | 36/25 | 45/32 |
| Original internal half-angle product | 125/324 | 32/81 |
| Printed quotient k108 | 11664/3125 | 3645/1024 |
Why the result holds
The coordinates verify the reflection law and put every caustic contact strictly inside its trajectory side. Stachel’s established parametrization places the two trajectories in the same connected oriented family, so the comparison tests variation within one family.
Intersecting consecutive boundary tangents gives the outer polygons. Their areas follow from the shoelace formula, and the half-angle products follow from exact internal-angle cosines. Every quantity in the comparison is evaluated exactly.
Scope and limitations
The target is the printed experimental, unproved assertion in Reznik–Garcia–Koiller’s Table 2. This note supplies an exact counterexample; it is not an official erratum.
The title’s “correction” refers to that failure of constancy. The paper does not prove a repaired invariant for the full six-period family or an all-period replacement.
The horizontal orbit geometry and the connected-family mechanism are established prior work. No new six-period construction, historical firstness, or independent-discovery claim is made.
Background and prior work
- Section 3.1 and Table 2, printed p. 345, give the quotient, the N ≡ 2 mod 4 condition, and a question mark in the proof column. The earlier arXiv v11 has the same target expression. Reznik–Garcia–Koiller, Fifty New Invariants of N-Periodics in the Elliptic Billiard
- The November 2020 preprint, Appendix B.5.1, already gives the horizontal six-period coordinates, confocal caustic and perimeter used here at a = 2, b = 1. Garcia–Reznik, Invariants of Self-Intersected and Inversive N-Periodics in the Elliptic Billiard (v1)
- The published Appendix B.5.1 retains that earlier horizontal construction. Garcia–Reznik, Exploring self-intersected N-periodics in the elliptic billiard
- Theorem 4.3 supplies the canonical parametrization and constant shift used to identify one connected oriented family. Stachel, On the motion of billiards in ellipses
- Corollary 4.2(i) gives the established even-period, odd-turn central symmetry. Stachel, The geometry of billiards in ellipses and their Poncelet grids
Status and AI use
Unrefereed preprint. AI tools assisted in preparing and checking this paper. Independent human review is welcome.
Read the paper
Open the published PDF · Download PDF · Original PDF on Zenodo
Supporting materials
Download k108_counterexample_support.zip · Original supporting ZIP on Zenodo
Collaborate on this result
I would welcome an independent check of the internal-angle convention and the identification of the two orbits within one oriented family. Authorship would reflect substantive contributions to the research or manuscript.
Cite this paper
Kriebel, Alec (2026). An exact six-period correction to the printed elliptic billiard assertion k108. Version 1.0. Zenodo. 10.5281/zenodo.23271172.
Licensing
The deposited work is available under Creative Commons Attribution 4.0.