Projective plane
From Free net encyclopedia
In mathematics, a projective plane has two possible definitions, one of them coming from linear algebra, and another (which is more general) coming from the combinatorics of block designs. The first definition quickly produces planes that are homogeneous spaces for some of the classical groups. The second is suitable for an exhaustive study of the simple incidence properties of plane geometry.
Contents |
Combinatorial definition
According to the more general, combinatorial definition, a projective plane consists of a set of "lines" and a set of "points" with the following properties:
- Given any two distinct points, there is exactly one line incident with both of them.
- Given any two distinct lines, there is exactly one point incident with both of them.
- There are four points such that no line is incident with more than two of them.
The last condition simply excludes some degenerate cases (see below).
Examples
A projective plane is an abstract mathematical concept, so the "lines" need not be anything resembling ordinary lines, nor need the "points" resemble ordinary points. The most common projective plane is the real projective plane, which is a topological surface with surprising geometric properties; after that is the complex projective plane of algebraic geometry, a topological four-dimensional manifold. For any field K, there is a projective plane with three homogeneous coordinates in K, which can also be thought of in terms of a three-dimensional vector space V over K, 'points' being one-dimensional subspaces and 'lines' two-dimensional subspaces.
Image:Fano.png |
The smallest possible projective plane has only seven points and seven lines. It is often called the Fano plane, and is shown in the picture on the right (see also finite geometry). In this representation of the Fano plane, the seven points are shown as small black balls, and the seven lines are shown as six line segments and a circle. However, we could equally consider the balls to be the "lines" and the line segments and circle to be the "points" — this is an example of the duality of projective planes: if the lines and points are interchanged, the result is still a projective plane.
A permutation of the seven points that carries collinear points (points on the same line) to collinear points is called a "symmetry" of the plane.
Properties
It can be shown that a projective plane has the same number of lines as it has points. This number can be infinite (as for the real projective plane) or finite (as for the Fano plane). A finite projective plane has
- n2 + n + 1 points,
where n is an integer called the order of the projective plane. (The Fano plane therefore has order 2.) For all known finite projective planes, the order n is a prime power. The existence of finite projective planes of other orders is an open question. The only general restriction known on the order is the Bruck-Ryser-Chowla theorem that if the order n is congruent to 1 or 2 mod 4, it must be the sum of two squares. This rules out n = 6. The next case n = 10 has been ruled out by massive computer calculations, and there is nothing more known, in particular n = 12 is still open. There is a projective plane of order n if and only if there is an affine plane of order n. When there is only one affine plane of order n there is only one projective plane of order n, but the converse is not true. A projective plane of order n has n + 1 points on every line, and n + 1 lines passing through every point, and is therefore a Steiner S(2, n +1, n2 + n + 1) system (see Steiner system). Conversely, one can prove that all Steiner systems of this form (<math>n\geq2</math>) are projective planes.
Linear algebra definition
One can construct projective planes (or higher dimensional projective spaces) by linear algebra over any division ring — not necessarily commutative. See for example quaternionic projective space. If we use a finite field with pn elements we get a finite projective plane with order pn. The Fano plane is then the plane over the field with two elements, Z2.
Generalizations
One can also do the reverse, and construct a coordinate "ring" - a so-called planar ternary ring (not necessarily a genuine ring) corresponding to any projective plane as defined above. Algebraic properties of this "ring" turn out to correspond to geometric incidence properties of the plane. For example, Desargues' theorem corresponds to the coordinate ring being a division ring, while Pappus's theorem corresponds to this ring being commutative. However, the "ring" need not be of this type, and there are many non-Desarguesian projective planes. Alternative, not necessarily associative division rings correspond to Moufang planes. In the case of finite projective planes, the only proof known of the purely geometric statement that Desargues theorem then implies Pappus' theorem (the converse being always true and provable geometrically) is through this algebraic route, using Wedderburn's theorem that finite division rings must be commutative.
It is possible to make analogous incidence definitions for higher dimensional projective n-spaces, for n larger than 2. These turn out to not be as interesting as the planar case, as they correspond to classical projective geometry over division rings for a very simple reason: with the extra room to work in, one can prove Desargues theorem geometrically as in its article by using incidence properties in this higher dimensional space and thus the coordinate "ring" must be a division ring.
The plane over the octonions turns out to be an interesting real manifold, which can be used for geometric constructions and understanding of the Exceptional Lie groups.
Degenerate planes
Degenerate planes do not fulfill the third condition above. There are two families of degenerate planes.
1) For any number of points P1, ..., Pn, and lines L1, ..., Lm,
- L1 = { P1, P2, ..., Pn}
- L2 = { P1 }
- L3 = { P1 }
- ...
- Lm = { P1 }
2) For any number of points P1, ..., Pn, and lines L1, ..., Ln, (same number of points as lines)
- L1 = { P2, P3, ..., Pn }
- L2 = { P1, P2 }
- L3 = { P1, P3 }
- ...
- Ln = { P1, Pn }
Connection with Latin squares
A projective plane of order n (<math>n \geq 2</math>) exists if and only if there is an affine plane of this order. The number of mutually orthogonal latin squares is at most n − 1. It turns out n − 1 is possible if and only if there is an affine plane of this order.
Construction of projective planes of prime order
Method 1
This is the standard construction using homogeneous coordinates over a finite field.
Method 2
To construct a projective plane of order N (N prime), proceed as follows:
- Create one point P
- Create N points, which we will label P(c) : c = 0, ..., (N − 1)
- Create N2 points, which we will label P(r, c) : r, c = 0, ..., (N − 1)
On these points, construct the following lines:
- One line L = { P, P(0), ..., P(N − 1)}
- N lines L(c) = {P, P(0,c), ..., P(N − 1,c)} : c = 0, ..., (N − 1)
- N2 lines L(r, c): P(c) and the points P((r + ci) mod N, i), where i = 0, .., N − 1 : r, c = 0, ..., (N − 1)
Note that the expression
- (r + ci) mod N
will pass once through each value as i varies from 0 to N − 1, but only if is N is prime.
By this construction, we have two degenerate planes: one point incident with one line (for N = 0) and a triangle consisting of three points and three lines (for N = 1). Every plane constructed with prime N (N > 1) fulfills all three conditions above.
For example, for N=2:
- One line L = { P, P(0), P(1)}
- 2 lines L(c) = {P, P(0,c), P(1,c)} : c = 0, 1
- 4 lines L(r, c): P(c) and the points P((r + ci) mod 2, i), where i = 0, 1 : r, c = 0, 1
Small orders
While the classification of all projective planes is far from done, here are some results for some orders :
- 2 : all isomorphic with PG(2,2)
- 3 : all isomorphic with PG(2,3)
- 4 : all isomorphic with PG(2,4)
- 5 : all isomorphic with PG(2,5)
- 6 : impossible as order of a projective plane, proved by Euler in his thirty-six officers problem
- 7 : all isomorphic with PG(2,7)
- 8 : all isomorphic with PG(2,8)
- 9 : PG(2,9), and three more different (up to isomorphism) non- Desarguesian planes.
- 10 : impossible as order of a projective plane, proved by heavy computer calculation.
- 11 : at least PG(2,11), others are not known but possible
- 12 : it is assumed it is impossible as order of a projective plane, but not proven
See also
External links
- Projective plane on mathworld
- Lam, Clement W.H. (1996). "The Search for a Finite Projective Plane of Order 10". Retrieved Apr. 30, 2005.
- [1] gives more information on small projective planes.