English

A note on non-crossing path partitions in the plane

Computational Geometry 2025-09-23 v1

Abstract

In the paper ``Lower bounds on the number of crossing-free subgraphs of KNK_N'' (Computational Geometry 16 (2000), 211-221), it is shown that a double chain of nn points in the plane admits at least Ω(4.642126305n)\Omega(4.642126305^n) polygonizations, and it is claimed that it admits at most O(5.61n)O(5.61^n) polygonizations. In this note, we provide a proof of this last result. The proof is based on counting non-crossing path partitions for points in the plane in convex position, where a non-crossing path partition consists of a set of paths connecting the points such that no two edges cross and isolated points are allowed. We prove that a set of nn points in the plane in convex position admits O(5.610718614n)\mathcal{O}^*(5.610718614^{n}) non-crossing path partitions and a double chain of nn points in the plane admits at least Ω(7.164102920n)\Omega(7.164102920^n) non-crossing path partitions. If isolated points are not allowed, we also show that there are O(4.610718614n)\mathcal{O}^*(4.610718614^n) non-crossing path partitions for nn points in the plane in convex position and at least Ω(6.164492582n)\Omega(6.164492582^n) non-crossing path partitions in a double chain of nn points in the plane. In addition, using a particular family of non-crossing path partitions for points in convex position, we provide an alternative proof for the result that a double chain of nn points admits at least Ω(4.642126305n)\Omega(4.642126305^n) polygonizations.

Keywords

Cite

@article{arxiv.2509.17485,
  title  = {A note on non-crossing path partitions in the plane},
  author = {Javier Tejel},
  journal= {arXiv preprint arXiv:2509.17485},
  year   = {2025}
}
R2 v1 2026-07-01T05:49:03.793Z