Skip to content
— CH. 1 · INTRODUCTION —

Tessellation

~7 min read · Ch. 1 of 8
8 sections
  • Tessellation is the art and mathematics of covering a surface with shapes that fit together perfectly, leaving no gaps and allowing no overlaps. The Sumerians were doing it around 4000 BC, pressing clay tiles into wet walls to create decorative patterns. Thousands of years later, a Dutch graphic artist named M. C. Escher visited Spain and walked through the Alhambra palace. He came away so inspired by the tiled walls that tessellations became a defining obsession of his career. At roughly the same moment, mathematicians were working out a proof that all the possible repeating patterns in a flat plane fall into exactly seventeen categories. How can something so simple, a surface covered by shapes, open onto so much depth? That question runs from ancient Sumerian walls to a 2023 discovery by a hobbyist mathematician that stunned the geometry world.

  • Clay tiles pressed into Sumerian walls around 4000 BC mark the earliest known use of tessellation as a deliberate design. The impulse spread across civilisations. In classical antiquity, craftsmen cut stone, clay, and glass into small squared blocks called tesserae to create decorative mosaic floors and walls. The word "tessellation" itself traces directly to that Latin root: tessella means a small cubical piece, derived from tessera (square), which in turn comes from the Greek word for four. Ancient Rome used these techniques extensively. Islamic architecture took the craft further still, developing the elaborate Girih and Zellige tile traditions seen in buildings like the Alhambra in Granada and La Mezquita. These weren't only decorations. Tessellated surfaces served practical ends too, providing durable and water-resistant pavement, floor, and wall coverings across the ancient and medieval world.

  • In 1619, Johannes Kepler made what stands as an early documented study of tessellations. He wrote about regular and semiregular patterns in his Harmonices Mundi. Kepler may have been the first person to systematically explore and explain the hexagonal structures found in honeycombs and snowflakes. Two centuries passed before the next major breakthrough. In 1891, the Russian crystallographer Yevgraf Fyodorov proved that every periodic tiling of the plane features one of seventeen different groups of isometries. That proof is generally regarded as the unofficial beginning of tessellation as a formal mathematical discipline. Other researchers followed: Alexei Vasilievich Shubnikov and Nikolai Belov contributed through their book Colored Symmetry, published in 1964, and Heinrich Heesch and Otto Kienzle published their own work in 1963. The Swiss geometer Ludwig Schläfli expanded the field further by defining what he called polyschemes, now known as polytopes, and created the Schläfli symbol notation to describe them compactly. Under his system, a square is {4}, an equilateral triangle is {3}, and a tiling of regular hexagons with three six-sided polygons at each vertex carries the notation {6,3}.

  • Only three shapes can tile a flat plane using identical regular polygons with no gaps: the equilateral triangle, the square, and the regular hexagon. Each of these works because their interior angles divide evenly into 360 degrees at every meeting point. The pentagon, by contrast, fails this test. Its internal angle is not a divisor of 2 pi, which is why regular pentagons cannot tile the Euclidean plane. Beyond the three regular tilings, there are eight semi-regular tilings (nine if mirror-image pairs are counted separately), which mix more than one type of regular polygon while keeping the same arrangement at every corner. These are described using vertex configuration notation; a tiling of squares and regular octagons, for instance, has the notation 4.8.2 at each vertex. Irregular tessellations are far less constrained. Any triangle or quadrilateral, even a non-convex one, can serve as a prototile. The Hirschhorn tiling, published by Michael D. Hirschhorn and D. C. Hunt in 1985, demonstrated a pentagon tiling using irregular pentagons, solving a shape that the regular version cannot manage. For tiles that are neither convex nor concave in a simple sense, polyominoes offer a broad family of options for tiling puzzles and mathematical exploration.

  • Penrose tilings use two different quadrilateral prototiles and are the best-known example of a set of shapes that can only tile a plane without ever repeating. They belong to a broader class called aperiodic tilings. Though they lack translational symmetry, aperiodic tilings do possess other symmetries: infinite repetition of any bounded patch, and certain finite groups of rotations or reflections. Wang tiles, squares coloured on each edge so that touching edges must match in colour, form another category. A suitable set of Wang tiles can tile the plane, but only aperiodically. This connects to computability: any Turing machine can be represented as a set of Wang tiles that tile the plane if and only if the machine does not halt. Because the halting problem is undecidable, determining whether a given Wang tile set can tile the plane is also undecidable. Truchet tiles, introduced by Sebastien Truchet in 1704, are square tiles split into two contrasting triangles; they can tile periodically or randomly. The most recent chapter in aperiodic tiling came in 2023, when David Smith, described as a hobbyist mathematician, discovered a single shape that forces aperiodic tiling. Dubbed "the hat," it was the first confirmed einstein tile, solving a longstanding open problem in geometry.

  • M. C. Escher visited Spain in 1936, and his encounter with the Moorish tiling of the Alhambra palace changed his practice permanently. He had already been working with tessellations, but the sophistication of the Islamic geometric tradition revealed new possibilities. Escher went on to make four drawings he called the Circle Limit series, all based on tilings using hyperbolic geometry. For Circle Limit IV, completed in 1960, he prepared a detailed pencil and ink study mapping out the required geometry before committing to the woodcut. He described the tiles in that series in his own words: "No single component of all the series, which from infinitely far away rise like rockets perpendicularly from the limit and are at last lost in it, ever reaches the boundary line." The Alhambra's tilings have also attracted mathematical attention because it has been claimed, though disputed, that all seventeen wallpaper groups are represented somewhere within the palace. Of the three regular tilings, two fall into the p6m wallpaper group and one into p4m. Whether or not the Alhambra contains all seventeen, the variety and sophistication of its geometric program continue to interest researchers today.

  • Honeycombs present the most familiar natural tessellation, their hexagonal cells fitting together without waste. In botany, the word "tessellate" describes checkered patterns on flower petals, tree bark, or fruit; the fritillary flower and some species of Colchicum are characteristically tessellate. Many other natural patterns arise not from biology but from fracture. Gilbert tessellations, named after Edgar Gilbert, model the formation of mudcracks and needle-like crystals by imagining cracks that start at randomly scattered points and propagate in two opposite directions along random lines, producing irregular convex polygons. Basaltic lava flows display a related phenomenon: as lava cools, contraction forces generate crack networks that frequently resolve into hexagonal columns. The Giant's Causeway in Northern Ireland is a well-known example of this process. Tasmania's Tasman Peninsula has a rare sedimentary counterpart: Eaglehawk Neck's tessellated pavement, where rock has fractured into rectangular blocks. Even foams obey geometric constraints. Lord Kelvin proposed in 1887 that the most efficient way to pack cells of equal volume was a structure called the bitruncated cubic honeycomb, with slightly curved faces. In 1993, Denis Weaire and Robert Phelan proposed the Weaire-Phelan structure, which uses less surface area to divide cells of equal volume than Kelvin's solution.

  • Henry Dudeney invented the hinged dissection as a tiling puzzle, and Martin Gardner wrote about rep-tiles in Scientific American. Gardner's articles directly inspired Marjorie Rice, an amateur mathematician, who went on to find four new tessellations using pentagons. James and Frederick Henle proved that it is possible to tile the entire plane using squares whose side lengths are all different natural numbers, with no size repeated. In manufacturing, tessellation reduces waste when cutting shapes from sheet metal; car doors and drink cans are among the applications where nesting shapes efficiently matters economically. Despite centuries of study, no general rule exists for determining whether an arbitrary shape can tile a plane, leaving many open problems in the field. The Conway criterion offers a sufficient but not necessary test for periodic tiling without reflections; shapes can fail the criterion and still tile. The 2023 discovery of the first einstein tile shows that major surprises remain possible. David Smith's "hat" was under professional review at the time of reporting, awaiting formal confirmation before the longstanding problem it addresses would be considered officially closed.

Common questions

What is tessellation and what are the basic rules?

Tessellation is the covering of a surface using one or more geometric shapes, called tiles, with no overlaps and no gaps. Common rules require that there are no gaps between tiles and that no corner of one tile lies along the edge of another.

Which shapes can form a regular tessellation?

Only three shapes can form a regular tessellation: the equilateral triangle, the square, and the regular hexagon. Each of these can be duplicated infinitely to fill a flat plane with no gaps.

What is an aperiodic tessellation and what are Penrose tilings?

An aperiodic tessellation uses a set of tile shapes that can tile the plane but cannot form a repeating pattern. Penrose tilings, which use two different quadrilateral prototiles, are the best-known example of aperiodic tiling.

When did M. C. Escher first become inspired by tessellations?

M. C. Escher visited Spain in 1936 and was inspired by the Moorish tiling of the Alhambra palace. He later made four Circle Limit drawings based on hyperbolic geometry, including Circle Limit IV, completed in 1960.

What is the einstein tile and who discovered it?

The einstein tile is a single shape that forces aperiodic tiling of the plane. It was discovered in 2023 by David Smith, a hobbyist mathematician, and was dubbed "the hat." The discovery was under professional review at the time of reporting.

How many wallpaper groups exist and what is their connection to tessellations?

There are exactly 17 wallpaper groups, representing all possible symmetry groups for periodic tilings of the plane. This was proved by the Russian crystallographer Yevgraf Fyodorov in 1891, and it is claimed that all seventeen groups are represented in the Alhambra palace in Granada, though this is disputed.

All sources

86 references cited across the entry

  1. 1bookThe Math Book: From Pythagoras to the 57th Dimension, 250 Milestones in the History of MathematicsClifford A. Pickover — Sterling — 2009
  2. 2bookMosaics of the Greek and Roman worldKatherine M. D. Dunbabin — Cambridge University Press — 2006
  3. 3webThe Brantingham Geometric MosaicsHull City Council — 2008
  4. 4bookHarmonices MundiKepler, Johannes — 1619
  5. 5webDynamic Coverage Problems in Sensor NetworksDjidjev, Hristo et al. — Los Alamos National Laboratory — 2012
  6. 6journalSimmetrija na ploskosti Symmetry in the planeFyodorov, Y. — 1891
  7. 7bookColored SymmetryAlekseĭ Vasilʹevich Shubnikov et al. — Macmillan — 1964
  8. 8bookFlächenschluss: System der Formen lückenlos aneinanderschliessender FlächteileH. Heesch et al. — Springer — 1963
  9. 9encyclopediaTessellate
  10. 10bookThe Symmetries of ThingsConway, R. et al. — Peters — 2008
  11. 11bookMathematical ModelsCundy and Rollett — Oxford — 1961
  12. 13magazineWill It Tile? Try the Conway Criterion!Schattschneider, Doris — September 1980
  13. 14bookRegular PolytopesCoxeter, H. S. M. — Methuen — 1948
  14. 15bookM.C. Escher's Legacy: A Centennial CelebrationMichele Emmer et al. — Springer — 8 May 2007
  15. 16bookGeometric Symmetry in Patterns and TilingsClare E. Horne — Woodhead Publishing — 2000
  16. 17webSome Special Radial and Spiral TilingsDutch, Steven — University of Wisconsin — 29 July 1999
  17. 18journalEquilateral convex pentagons which tile the planeM. D. Hirschhorn — 1985
  18. 19webSchläfli TessellationsNRICH (Millennium Maths Project) — University of Cambridge — 1997–2012
  19. 20bookThe Penguin Dictionary of Curious and Interesting GeometryDavid Wells — Penguin Books — 1991
  20. 21journalEdge Tessellations and Stamp Folding PuzzlesKirby, Matthew — 2011
  21. 22bookGroups and SymmetryM.A. Armstrong — Springer-Verlag — 1988
  22. 23journalWhat symmetry groups are present in the Alhambra?Branko Grünbaum — June–July 2006
  23. 24journalDecagonal and quasi-crystalline tilings in medieval Islamic architecturePeter J. Lu — 23 February 2007
  24. 25webTwo-Dimensional Symmetry MutationDaniel H. Huson — Princeton University — 1991
  25. 26journalThe Pinwheel Tilings of the PlaneRadin, C. — May 1994
  26. 27webPenrose Tiles Talk Across MilesDavid Austin — American Mathematical Society
  27. 28webAperiodic TilingE. O. Harriss — University of London and EPSRC
  28. 29journalRaman scattering in Fibonacci superlatticesM. W. C. Dharma-wardana et al. — 1987
  29. 30journalProving theorems by pattern recognition—IIHao Wang — 1961
  30. 31magazineGames, logic and computersHao Wang — November 1965
  31. 32journalThe undecidability of the domino problemRobert Berger — 1966
  32. 33journalUndecidability and nonperiodicity for tilings of the planeRaphael M. Robinson — 1971
  33. 34journalAn aperiodic set of 13 Wang tilesKarel II Culik — 1996
  34. 35journalTruchet curves and surfacesCameron Browne — 2008
  35. 36journalThe tiling patterns of Sebastian Truchet and the topology of structural hierarchyCyril Stanley Smith — 1987
  36. 39bookThe Grammar of OrnamentOwen Jones — Bernard Quaritch — 1910
  37. 40journalVoronoi Diagrams – A Survey of a Fundamental Geometric Data StructureAurenhammer, Franz — 1991
  38. 41bookSpatial Tessellations – Concepts and Applications of Voronoi DiagramsOkabe, Atsuyuki et al. — John Wiley — 2000
  39. 42bookDelaunay Triangulation and Meshing: Application to Finite ElementsGeorge, Paul Louis et al. — Hermes — 1998
  40. 43bookLectures on Random Voronoi TessellationsJesper Moller — Springer — 1994
  41. 44journalUniform tilings of 3-spaceGrünbaum, Branko — 1994
  42. 45journalÜber Wirkungsbereichsteilungen von kubischer SymmetriePeter Engel — 1981
  43. 46bookFirefly Guide to GemsCally Oldershaw — Firefly Books — 2003
  44. 47bookMinerals and Rocks in ColourJ. F. Kirkaldy — Blandford — 1968
  45. 48bookKaleidoscopes: Selected Writings of H.S.M. CoxeterHarold Scott Macdonald Coxeter et al. — John Wiley & Sons — 1995
  46. 49bookQuasicrystals and GeometryMarjorie Senechal — CUP Archive — 26 September 1996
  47. 51arxivCoordinates for a new triangular tiling of the hyperbolic planeMaurice Margenstern — 4 January 2011
  48. 53bookThe Beauty of Geometry: Twelve EssaysCoxeter, H.S.M. — Dover Publications — 1999
  49. 54bookGeometric Patterns from Roman MosaicsField, Robert — Tarquin — 1988
  50. 55webMathematics in Art and ArchitectureNational University of Singapore
  51. 56bookSpeak the Culture: SpainAndrew Whittaker — Thorogood Publishing — 2008
  52. 57webIntroduction to Hyperbolic and Automatic GroupsS. M. Gersten — University of Utah
  53. 58webHyperbolic EscherJos Leys — 2015
  54. 59bookTessellation Quilts: Sensational Designs From Interlocking PatternsChristine Porter — F+W Media — 2006
  55. 60bookDesigning tessellations: the secrets of interlocking patternsJinny Beyer — Contemporary Book — 1999
  56. 61bookOrigami TessellationsEric Gjerde — Taylor and Francis — 2008
  57. 63journalCrack Spacing in Brittle Films on Elastic SubstratesM. D. Thouless — 1990
  58. 64journalCrack patterns in thin filmsZ. C. Xia et al. — 2000
  59. 65journalControlled mud-crack patterning and self-organized cracking of polydimethylsiloxane elastomer surfacesR. Seghir et al. — 2015
  60. 66journalHow honeycombs can build themselvesPhilip Ball — 2013
  61. 67bookShorter Oxford English DictionaryOxford University Press — 2007
  62. 68journalColchicums: autumn's best-kept secretKathy Purdy — 2007
  63. 69arxivLimit theory for planar Gilbert tessellationsTomasz Schreiber et al. — 2010
  64. 70journalTopological properties of random crack networksN. H. Gray et al. — 1976
  65. 71bookApplications of Undergraduate Mathematics in EngineeringE. N. Gilbert — Macmillan — 1967
  66. 72journalSoap, cells and statistics: Random patterns in two dimensionsDenis Weaire et al. — 1984
  67. 73bookTesselated pavementsD. F. Branagan — University of Wollongong — 1983
  68. 74bookShapesPhilip Ball — Oxford University Press — 2009
  69. 75webHistory of Jigsaw PuzzlesDaniel McAdam — American Jigsaw Puzzle Society
  70. 76bookThe Tao of TangramJerry Slocum — Barnes & Noble — 2001
  71. 77bookPolyominoesSolomon W. Golomb — Princeton University Press — 1994
  72. 78bookPolyominoes: A guide to puzzles and problems in tilingGeorge E. Martin — Mathematical Association of America — 1991
  73. 79bookHinged Dissections: Swinging and TwistingGreg N. Frederickson — Cambridge University Press — 2002
  74. 80magazineOn 'Rep-tiles,' Polygons that can make larger and smaller copies of themselvesGardner, Martin — May 1963
  75. 81bookAha! A Two Volume Collection: Aha! Gotcha Aha! InsightMartin Gardner — MAA — 14 December 2006
  76. 82newsThe Importance of Recreational MathMani Suri — 12 October 2015
  77. 83journalTiling the Plane with Congruent PentagonsDoris Schattschneider — MAA — 1978
  78. 84webSquaring the SquareW. T. Tutte
  79. 85magazineMathematical GamesMartin Gardner et al. — November 1958
  80. 86journalSquaring the planeFrederick V. Henle et al. — 2008