Concept

Incidence

The relation saying which points lie on which lines, taken as the whole of a geometry's structure. Taking it as the whole structure is what makes finite geometries possible, since no notion of distance or angle is required.

Named by 20 essays across 3 fields — each of them below, with the objects they name alongside it.

Two points, and everything one round of compass and straightedge adds. Two starting points with the line and circles they permit, and the four points where those objects cross.

What two points can build

A compass and a straightedge are not a craft. They are two operations on a set of points, applied over and over, and writing them that way turns "can this be drawn?" into a question with an answer.

computation · Constructible numbers
The Fano plane, and the incidence table behind it. Seven points joined by six straight lines and one circle, beside the seven-by-seven table of which point lies on which line.

Seven points, seven lines

A geometry with seven points, in which every two points lie on exactly one line and every two lines meet in exactly one point. There are no parallels, the whole thing is built out of the two-element field, and one of its lines has to be drawn as a circle.

computation · Finite geometry
A schedule on 9 points where every pair meets exactly once. Points around a circle with the triples of a Steiner system drawn between them, beside the list of triples.

A schedule where every pair meets once

Sort n people into groups of three so that every two of them share a group exactly once. Two divisions have to come out whole, that rules out most sizes — and at every size the divisions permit, a schedule exists.

computation · Finite geometry
Nine points of a triangle, on one circle. A triangle with the midpoints of its sides, the feet of its three altitudes and the midpoints from each corner to the orthocentre marked; all nine lie on a single circle of half the circumradius.

Nine points on one circle

Three midpoints, three feet of altitudes and three more midpoints. Nine points defined in three unrelated ways, on an arbitrary triangle, and all nine sit on one circle — checked here on two hundred and forty triangles as well as on the drawn one.

geometry · Triangle centres
The midpoint of a segment, drawn with a compass and no straightedge. A segment with the arcs that step its length three times round one end to reach the point twice as far away, and the further arcs that send that point back to the midpoint, every one of them a circle.

The straightedge buys nothing

Every point a compass and a straightedge can construct together can be constructed by the compass alone. The straightedge draws lines nobody needs; the compass does the work, and the proof that it does is an inversion performed with arcs.

computation · Compass-only
The nine-point grid, and the lines they force. 9 points with all 20 of their connecting lines drawn. The 12 carrying exactly two points are drawn solid and the rest faintly; the count is computed from the coordinates rather than read off the drawing.

The line with only two points on it

Scatter finitely many points on a page, not all in one line, and draw every line through two or more of them. However cunningly the points are placed, some line ends up carrying exactly two — and the proof is a minimisation with no algebra in it at all.

geometry · Ordinary lines
The affine plane of order 3, one parallel class at a time. The n² cells of a complete set of orthogonal Latin squares of order 3, with the rows, the columns and each square's symbol classes drawn as lines of a plane.

The plane hiding in the squares

A complete family of orthogonal squares is not a collection of squares that happen to agree nowhere. It is a geometry — a plane with n² points in which every two points lie on exactly one line — and reading it that way is how the impossible orders were found.

computation · Latin squares
The nine-point circle, touching four others. A triangle with its nine-point circle, its inscribed circle and its three escribed circles, each of the four tangent to the first — with the distances between centres compared against the radii.

One circle touching four

The nine-point circle touches the inscribed circle and each of the three escribed ones. Nothing in its construction mentions them, the two families of centres are built from different kinds of number, and the tangency is four exact equalities between distances and radii.

geometry · Triangle centres
The classical centres as three weights each. A table of triangle centres with the weights on the three corners that produce each, and the determinants that decide which triples of them are collinear.

A centre is three weights

Write each classical centre as a weighted average of the corners and a coincidence becomes a determinant. The Euler line is then one number rather than a construction, the whole catalogue becomes mechanical, and the reason one centre is missing from it is visible in the weights.

geometry · Triangle centres
A design on 7 points cannot have fewer than 7 blocks. The incidence matrix of a design on 7 points and 7 blocks beside the product of it with its own transpose, which has a constant off the diagonal and a determinant computed exactly.

More blocks than points

A schedule in which every pair meets once cannot use fewer groups than it has people. Nothing about the counting conditions says so, and the proof is not combinatorial at all — it is a determinant, computed over a field the schedules have nothing to do with.

computation · Finite geometry
A plane of 13 points from a list of 4 numbers. A ring of 13 points with one block of 4 of them drawn as a closed path, beside the table of the 13 blocks its shifts produce.

A plane in a list of numbers

A projective plane of order three has thirteen points and thirteen lines and fifty-two incidences. All of it is in the four numbers 0, 1, 3, 9 — because their pairwise differences hit every non-zero residue modulo thirteen exactly once, and the plane is that list's thirteen shifts.

computation · Finite geometry
The one line from which the nearfield plane looks Desarguesian. A grid of the 91 lines of the nearfield plane of order nine shaded by how many of 40 Desargues configurations with that line as axis failed; only the line at infinity has none, and every other line at least 20.

A plane no field built

Every finite field builds a projective plane, and for a long time every known plane was built that way. The plane over Dickson's nearfield of order nine has ninety-one points, ninety-one lines and every incidence right — and Desargues' theorem fails in it on most configurations tried, except for one line, from which it never fails at all.

computation · Finite geometry
The conic y = x² in the plane of order 7. A 7 by 7 grid of the affine plane over GF(7) with the points of the conic y = x² filled and its point at infinity marked: 8 points, no three collinear.

The curve that no three points in line define

In a finite plane, take as many points as possible with no three on a line. In odd order the largest such sets have one more point than the order — and every one of them, searched exhaustively in the small planes and proved by Segre for all odd orders, is a conic. In even order every tangent meets at one point, which can be added, and the curves stop being forced.

computation · Finite geometry
The most edges with no four-cycle. Points for n = 2 to 9: the largest number of edges with no four-cycle, 1, 3, 4, 6, 7, 9, 11, 13, between the counting bound above and ½n^(3/2) below, far under the complete graph's count.

The densest graph without a square

Forbid four points joined in a cycle and a graph can keep only about ½n^(3/2) of its edges — far fewer than the quarter of all pairs a triangle-free graph keeps. Counting pairs of neighbours proves the ceiling in two lines. What reaches it is not a random graph but a finite geometry: the points of a projective plane, joined when they are orthogonal.

discrete · Extremal graphs
A triangle, its midpoints and its centroid, turned into lines. The dual arrangement of 7 points: one line per point, crossing where points were collinear. 3 crossings are of exactly two lines, the dual of the ordinary lines; the others are where three or more meet.

Three ordinary lines from a count

Kelly's proof finds one line through exactly two of the points by minimising a distance. Melchior, seven years earlier, had found three — by turning every point into a line and counting the corners, edges and regions of the picture that results. Euler's formula for the projective plane does the rest, and it says exactly which configurations have no more than three.

geometry · Ordinary lines
Böröczky's 12 points and their 6 ordinary lines. A disc standing for the projective plane: the 6 corners of a regular polygon inside, and 6 points at infinity marked in pairs on the rim. All 22 connecting lines are drawn, the 6 ordinary ones solid.

The fewest ordinary lines a polygon allows

Take the corners of a regular polygon and add the points at infinity where its parallel chords meet. Every chord then carries three points, the line at infinity carries all the new ones, and the only lines left with exactly two points are the tangents at the corners — half as many as there are points. Dirac guessed in 1951 that nothing does better, and Green and Tao proved it in 2013.

geometry · Ordinary lines
Removing one end of an ordinary line from a triangle, its midpoints and its centroid. Two panels. Left: 7 points with all 9 connecting lines, one ordinary line solid and one of its ends ringed. Right: the same points with that end removed, 7 connecting lines left.

At least as many lines as points

Sylvester's theorem says some line through two of the points misses all the rest. Remove one end of that line and the line itself disappears, taking at least one line away with one point. Run that backwards and it proves that n points not all in a line determine at least n lines — and the only sets that manage exactly n are a line of n − 1 points with one point off it.

geometry · Ordinary lines
9 trees on a cubic, 10 rows of three. The curve y = 1/(1 + x²) with 8 marked points on it and a 9th at infinity. Every line through three of them is drawn: 10 in all.

Rows of three, planted on a cubic

Nine trees can be planted in ten rows of three, and the arrangement that does it is not a grid or a star but nine points on a cubic curve. On the curve three points are in line exactly when their angles add up to a right angle, so choosing the points as a cyclic group turns collinearity into addition — and the count of rows it produces is the number Green and Tao proved is the most any planting can reach.

geometry · Ordinary lines
The plane of order 3 as a table, and the table times its transpose. The 13 × 13 incidence table of the projective plane of order 3 and its product with its transpose, which has 4 on the diagonal and 1 in every other cell.

The orders a plane cannot have

Every counting condition allows a projective plane of order six, and there is none. The proof that rules it out looks at one matrix identity — each point on seven lines, each two points on one — and turns it, by way of Lagrange's four squares, into the statement that six would have to be a sum of two squares. Run on the planes that do exist, the same argument hands back their orders as sums of two squares; run on six, it asks for something no arithmetic can supply.

computation · Finite geometry
Every power of x that draws a hyperoval, in the planes of order 4 to 4096. q = 4: 1 exponents in 1 classes (conic); q = 8: 3 exponents in 1 classes (conic); q = 16: 3 exponents in 1 classes (conic); q = 32: 11 exponents in 3 classes (conic, translation/Glynn I/Glynn II, Segre); q = 64: 3 exponents in 1 classes (conic); q = 128: 23 exponents in 5 classes (conic, translation, Segre/Glynn II, translation, Glynn I); q = 256: 9 exponents in 2 classes (conic, translation); q = 512: 27 exponents in 5 classes (conic, translation, Segre, translation, Glynn I/Glynn II); q = 1024: 9 exponents in 2 classes (conic, translation); q = 2048: 45 exponents in 8 classes (conic, translation, Segre, translation, translation, Glynn II, translation, Glynn I); q = 4096: 9 exponents in 2 classes (conic, translation).

Every power of x that draws a hyperoval

In a plane of order 2^h, the graph of x^k plus two points at infinity is sometimes a hyperoval — as many points as a plane allows with no three in line. Searching every exponent in every plane from order 4 to 4096 finds hundreds that work, and once six symmetries of the problem are applied they fall into exactly the families already known: the conic, the translation curves, Segre's x⁶ and Glynn's two. Whether that list is complete in every order is open.

computation · Finite geometry

Named alongside it

The objects these essays reach for when they reach for this one.

Projective planeCounting argumentExhaustive searchFinite fieldExtremal configurationOrdinary linePoint setSymmetryCircleConjectureCounterexampleExistence proof

All concepts