Series

Isoperimetric — the series

9 essays on one idea, from the one that introduces it to the one that assumes the rest.
  1. One perimeter of 300, spent five ways. Regular polygons all of the same perimeter, drawn to scale beside the circle of that perimeter, with the area each encloses and the ratio 4πA/L².

    The most area a fence can hold

    One length of boundary, and the question of what shape to bend it into. The answer is a circle, everybody knows it, and the argument that convinced the nineteenth century turned out to prove something slightly different.

    part 1 · geometry
  2. Fences of 300 against a straight wall. Fences made of two, three and four straight pieces with both ends on a wall, beside a half-circle with the same length of fence, each labelled with the area it holds.

    Half a circle against a wall

    Lay a fence of fixed length with both ends against a straight wall and the best shape is a half-circle, holding exactly twice what a full circle of the same fence holds. The proof is a mirror: doubled in the wall, any fence becomes a closed curve with twice the length and twice the area, and the closed-curve answer carries over. In a corner the same mirrors give a slice of a circle — until the corner's angle stops dividing a half-turn.

    part 2 · geometry
  3. Four shapes of perimeter 300 between their inner and outer circles. A square, an ellipse, a Reuleaux triangle and a stadium, each drawn with the largest circle inside it and the smallest circle around it, the ring between the two shaded, with the ring's width and the widest ring allowed.

    Nearly the most means nearly round

    A shape that holds almost as much as a circle of the same perimeter must almost be a circle. Bonnesen made that exact: the ring between a convex shape's largest inscribed circle and smallest enclosing circle is never wider than √(L² − 4πA)/π. Three quite different shapes holding 99% of the circle's area all have rings under 9.55 wide, and not one of 200 random convex shapes breaks the bound.

    part 3 · geometry
  4. Three tilings by cells of one area, and the wall each needs. Panels of triangles, squares and hexagons, all with cells of the same area, labelled with the wall length each cell needs once shared walls are split between neighbours: the hexagons need the least.

    The least wall for equal rooms

    Divide the plane into rooms of equal area using as little wall as possible, and every wall does double duty. Bees settled on hexagons long ago, and the proof that nothing does better — not even rooms with curved walls — came in 1999. The straight-walled half of it is two facts: the rooms of any division average six sides, and more sides never cost more wall.

    part 4 · geometry
  5. One symmetrisation: every chord slid to the middle. A lopsided shape with a dent beside its Steiner symmetrisation about a horizontal line, with a few vertical chords marked in both: the chords keep their lengths and are centred on the line.

    Every chord slid to the middle

    Take a shape, pick a line, and slide every chord that crosses the line at right angles until the line cuts it in half. The area cannot change, the boundary can only get shorter, and the result is symmetric. Do it again about another line, and another, and the shape is squeezed towards a disc — unless the lines are badly chosen, in which case it stops short.

    part 5 · geometry
  6. The lowest mode of the regular pentagon. Level lines of the first Dirichlet eigenfunction of a regular 5-gon; λ times area 18.9191.

    Three sides and four are proved

    Among all shapes of a given area, the disc has the smallest lowest eigenvalue. Among triangles it is the equilateral one, among quadrilaterals the square — both proved by sliding chords to an axis. For five sides the regular pentagon wins every computation, every nudge raises its value by the square of the nudge, and there is still no proof.

    part 6 · geometry
  7. The largest hexagon of diameter one. Regular hexagon of diameter 1: area 0.649519053; best hexagon: area 0.674981443, 6 pairs at distance 1; vertices (0.3438, -0.3253) (0.5000, 0.2115) (-0.0000, 0.6138) (-0.5000, 0.2115) (-0.3438, -0.3253) (-0.0000, -0.3862).

    The largest polygon of diameter one

    Among pentagons whose points are never more than one unit apart, the regular pentagon has the most area, and the same is true for every odd number of sides. For six sides it is false: a lopsided hexagon found in 1975 beats the regular one by four per cent. Symmetry wins half the cases and loses the other half, and the reason is parity.

    part 7 · geometry
  8. The longest perimeter a polygon of diameter one can have. Reinhardt bound 2n sin(pi/2n) for n=3..16 (reached unless n is a power of two), regular n-gon perimeters at diameter 1, and the best quadrilateral found 3.035273 against 2+4 sin 15 = 3.035276.

    The longest fence of diameter one

    Among polygons with n sides and diameter one, the one with the most area is regular only when n is odd. Ask for the most perimeter instead and the answer is kinder: Karl Reinhardt proved in 1922 that the longest perimeter is 2n sin(π/2n) whenever n has any odd factor at all, reached by polygons cut from curves of constant width, many of them not regular. Only when n is a power of two does every polygon fall short — and the best quadrilateral, found by searching every one, is an equilateral triangle with one corner bent out.

    part 8 · geometry
  9. Two sofas for a hallway one unit wide. Hammersley's sofa with exact area 2.20742 and raster area 2.2095; a shape from an optimised Fourier path, raster area 2.2195 with the same raster excess of 0.0021; Gerver's published value 2.2195.

    The largest sofa round a corner

    A hallway one unit wide turns a right angle. What is the largest shape that can be carried round? A half-disc gets to 1.571. John Hammersley cut a notch and, by one line of calculus, reached π/2 + 2/π = 2.2074. Joseph Gerver reached 2.2195 in 1992 with a boundary of eighteen analytic pieces and a value with no closed form. It took another thirty years for a proof that this was the end, and the problem began as a question about moving furniture.

    part 9 · geometry

All series