Author note (2026-11-02). The paper cited here, arXiv:2603.11311 (“Cut and project schemes in the Poincaré disc: From cocompact Fuchsian groups to chaotic Delone sets”), is by Richard A. Howat, Tony Samuel, Ayşe Yıltekin-Karataş. The name(s) this essay attributes it to (Mayero) do not appear among the listed authors, so the attribution below is wrong or refers to someone else's earlier work; I have not worked out which. Found by a mechanical check of each essay's named authors against the arXiv author list.
Quasicrystals built from hyperbolic geometry produce tiles with infinitely many sizes.
The classical cut-and-project method for constructing quasicrystals starts with a higher-dimensional lattice in Euclidean space. You slice through the lattice with an irrationally angled hyperplane, project nearby lattice points onto it, and the result is a quasiperiodic point set — aperiodic but ordered, like a Penrose tiling. The tile lengths (gaps between consecutive points in 1D) form a finite set. A Fibonacci sequence, for example, has exactly two tile lengths.
Mayero (arXiv:2603.11311) replaces Euclidean space with the Poincaré disc — the standard model of hyperbolic geometry — and asks what cut-and-project sets look like when the underlying lattice is a cocompact Fuchsian group. The answer: the resulting point sets are chaotic Delone sets with countably infinite tile lengths.
This is a qualitative shift. In Euclidean cut-and-project, the finite number of tile lengths reflects the finite-dimensional structure of the ambient lattice. Hyperbolic geometry, with its exponential volume growth and non-commutative isometry groups, injects enough complexity that the tiling can never be described by finitely many building blocks. Every scale introduces new gap sizes. The quasicrystal isn't just aperiodic — it's inexhaustibly varied.
The motivation comes from metamaterial design. Davies et al. (2023) asked whether cut-and-project schemes with non-square lattices or non-linear curves could generate graded metamaterials with superior wave-manipulation properties. Hyperbolic tilings answer definitively: yes, the structural richness is unlimited. Whether that richness is physically useful — whether countably infinite tile types produce better wave scattering than two — remains open.
The constraint that made Euclidean quasicrystals tractable was also the constraint that made them finite. Remove it, and the tiling becomes a different mathematical object entirely.
Mayero, "Cut and project schemes in the Poincaré disc: From cocompact Fuchsian groups to chaotic Delone sets," arXiv:2603.11311 (2026).