This page intentionally left blank

A course in combinatorics
This is the second edition of a popular book on combinatorics, a subject
dealing with ways of arranging and distributing objects, and which involves
ideas from geometry, algebra and analysis. The breadth of the theory is
matched by that of its applications, which include topics as diverse as codes,
circuit design and algorithm complexity. It has thus become essential for
workers in many scientic elds to have some familiarity with the subject.
The authors have tried to be as comprehensive as possible, dealing in a uni-
ed manner with, for example, graph theory, extremal problems, designs,
colorings and codes. The depth and breadth of the coverage make the book
a unique guide to the whole of the subject. The book is ideal for courses
on combinatorial mathematics at the advanced undergraduate or beginning
graduate level. Working mathematicians and scientists will also nd it a
valuable introduction and reference.
J. H. VAN LI NT is Emeritus Professor of Mathematics at the Technical Uni-
versity of Einhoven.
R. M. WI LSON is Professor of Mathematics at the California Institute of
Technology.
A Course in
Combinatorics
SECOND EDITION
J. H. van Lint
Technical University of Eindhoven
and
R. M. Wilson
California Institute of Technology
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore,
So Paulo, Delhi, Dubai, Tokyo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
First published in print format
ISBN-13 978-0-521-80340-3
ISBN-13 978-0-521-00601-9
ISBN-13 978-0-511-67289-7
Cambridge University Press 1992, 2001
2001
Information on this title: www.cambridge.org/9780521803403
This publication is in copyright. Subject to statutory exception and to the
provision of relevant collective licensing agreements, no reproduction of any part
may take place without the written permission of Cambridge University Press.
Cambridge University Press has no responsibility for the persistence or accuracy
of urls for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Paperback
eBook (EBL)
Hardback
CONTENTS
Preface to the rst edition xi
Preface to the second edition xiii
1. Graphs 1
Terminology of graphs and digraphs, Eulerian cir-
cuits, Hamiltonian circuits
2. Trees 12
Cayleys theorem, spanning trees and the greedy
algorithm, search trees, strong connectivity
3. Colorings of graphs and Ramseys theorem 24
Brooks theorem, Ramseys theorem and Ramsey
numbers, the L ovasz sieve, the Erd osSzekeres
theorem
4. Turans theorem and extremal graphs 37
Tur ans theorem and extremal graph theory
5. Systems of distinct representatives 43
Bipartite graphs, P. Halls condition, SDRs, K onigs
theorem, Birkhos theorem
6. Dilworths theorem and extremal set theory 53
Partially ordered sets, Dilworths theorem, Sperners
theorem, symmetric chains, the Erd osKoRado
theorem
7. Flows in networks 61
The FordFulkerson theorem, the integrality theorem,
a generalization of Birkhos theorem, circulations
8. De Bruijn sequences 71
The number of De Bruijn sequences
vi A Course in Combinatorics
9. Two (0, 1, ) problems: 77
addressing for graphs and
a hash-coding scheme
Quadratic forms, Winklers theorem, associative
block designs
10. The principle of inclusion and exclusion; 89
inversion formulae
Inclusionexclusion, derangements, Euler indica-
tor, M obius function, M obius inversion, Burnsides
lemma, probl`eme des menages
11. Permanents 98
Bounds on permanents, Schrijvers proof of the Minc
conjecture, Feketes lemma, permanents of doubly
stochastic matrices
12. The Van der Waerden conjecture 110
The early results of Marcus and Newman, Londons
theorem, Egoritsjevs proof
13. Elementary counting; Stirling numbers 119
Stirling numbers of the rst and second kind, Bell
numbers, generating functions
14. Recursions and generating functions 129
Elementary recurrences, Catalan numbers, counting
of trees, Joyal theory, Lagrange inversion
15. Partitions 152
The function p
k
(n), the partition function, Ferrers
diagrams, Eulers identity, asymptotics, the Jacobi
triple product identity, Young tableaux and the hook
formula
16. (0, 1)-Matrices 169
Matrices with given line sums, counting (0, 1)-
matrices
17. Latin squares 182
Orthogonal arrays, conjugates and isomorphism,
partial and incomplete Latin squares, counting Latin
squares, the Evans conjecture, the Dinitz conjecture
18. Hadamard matrices, ReedMuller codes 199
Hadamard matrices and conference matrices, re-
cursive constructions, Paley matrices, Williamsons
method, excess of a Hadamard matrix, rst order
ReedMuller codes
Contents vii
19. Designs 215
The ErdosDe Bruijn theorem, Steiner systems,
balanced incomplete block designs, Hadamard designs,
counting, (higher) incidence matrices, the Wilson
Petrenjuk theorem, symmetric designs, projective
planes, derived and residual designs, the Bruck
RyserChowla theorem, constructions of Steiner triple
systems, write-once memories
20. Codes and designs 244
Terminology of coding theory, the Hamming bound,
the Singleton bound, weight enumerators and
MacWilliams theorem, the AssmusMattson theorem,
symmetry codes, the Golay codes, codes from projec-
tive planes
21. Strongly regular graphs and partial geometries 261
The BoseMesner algebra, eigenvalues, the integrality
condition, quasisymmetric designs, the Krein condi-
tion, the absolute bound, uniqueness theorems, partial
geometries, examples, directed strongly regular graphs,
neighborhood regular graphs
22. Orthogonal Latin squares 283
Pairwise orthogonal Latin squares and nets, Eulers
conjecture, the BoseParkerShrikhande theorem,
asymptotic existence, orthogonal arrays and transver-
sal designs, dierence methods, orthogonal subsquares
23. Projective and combinatorial geometries 303
Projective and ane geometries, duality, Paschs
axiom, Desargues theorem, combinatorial geometries,
geometric lattices, Greenes theorem
24. Gaussian numbers and q-analogues 325
Chains in the lattice of subspaces, q-analogue of
Sperners theorem, interpretation of the coecients of
the Gaussian polynomials, spreads
25. Lattices and Mobius inversion 333
The incidence algebra of a poset, the M obius func-
tion, chromatic polynomial of a graph, Weisners
theorem, complementing permutations of geometric
lattices, connected labeled graphs, MDS codes
26. Combinatorial designs and projective geometries 351
Arcs and subplanes in projective planes, blocking
sets, quadratic and Hermitian forms, unitals, general-
ized quadrangles, M obius planes
viii A Course in Combinatorics
27. Dierence sets and automorphisms 369
Blocks lemma, automorphisms of symmetric de-
signs, PaleyTodd and StantonSprott dierence sets,
Singers theorem
28. Dierence sets and the group ring 383
The Multiplier Theorem and extensions, homomor-
phisms and further necessary conditions
29. Codes and symmetric designs 396
The sequence of codes of a symmetric design,
Wilbrinks theorem
30. Association schemes 405
Examples, the eigenmatrices and orthogonality re-
lations, formal duality, the distribution vector of a
subset, Delsartes inequalities, polynomial schemes,
perfect codes and tight designs
31. (More) algebraic techniques in graph theory 432
Tournaments and the GrahamPollak theorem, the
spectrum of a graph, Homans theorem, Shannon
capacity, applications of interlacing and Perron
Frobenius
32. Graph connectivity 451
Vertex connectivity, Mengers theorem, Tutte connec-
tivity
33. Planarity and coloring 459
The chromatic polynomial, Kuratowskis theorem,
Eulers formula, the Five Color Theorem, list-colorings
34. Whitney Duality 472
Whitney duality, circuits and cutsets, MacLanes
theorem
35. Embeddings of graphs on surfaces 491
Embeddings on arbitrary surfaces, the RingelYoungs
theorem, the Heawood conjecture, the Edmonds embed-
ding technique
36. Electrical networks and squared squares 507
The matrix-tree theorem, De Bruijn sequences, the
network of a squared rectangle, Kirchhos theorem
37. Polya theory of counting 522
The cycle index of a permutation group, counting
orbits, weights, necklaces, the symmetric group, Stir-
ling numbers
Contents ix
38. Baranyais theorem 536
One-factorizations of complete graphs and complete
designs
Appendix 1. Hints and comments on problems 542
Hints, suggestions, and comments on the problems in
each chapter
Appendix 2. Formal power series 578
Formal power series ring, formal derivatives, inverse
functions, residues, the LagrangeB urmann formula
Name Index 584
Subject Index 590
Preface to the rst edition
One of the most popular upper level mathematics courses taught
at Caltech for very many years was H. J. Rysers course Combina-
torial Analysis, Math 121. One of Rysers main goals was to show
elegance and simplicity. Furthermore, in this course that he taught
so well, he sought to demonstrate coherence of the subject of com-
binatorics. We dedicate this book to the memory of Herb Ryser,
our friend whom we admired and from whom we learned much.
Work on the present book was started during the academic year
198889 when the two authors taught the course Math 121 together.
Our aim was not only to continue in the style of Ryser by showing
many links between areas of combinatorics that seem unrelated,
but also to try to more-or-less survey the subject. We had in mind
that after a course like this, students who subsequently attend a
conference on Combinatorics would hear no talks where they are
completely lost because of unfamiliarity with the topic. Well, at
least they should have heard many of the words before. We strongly
believe that a student studying combinatorics should see as many
of its branches as possible.
Of course, none of the chapters could possibly give a complete
treatment of the subject indicated in their titles. Instead, we cover
some highlightsbut we insist on doing something substantial or
nontrivial with each topic. It is our opinion that a good way to
learn combinatorics is to see subjects repeated at intervals. For
this reason, several areas are covered in more than one part of the
book. For example, partially ordered sets and codes appear several
times. Enumeration problems and graph theory occur throughout
xii A Course in Combinatorics
the book. A few topics are treated in more detail (because we like
them) and some material, like our proof of the Van der Waerden
permanent conjecture, appears here in a text book for the rst
time.
A course in modern algebra is sucient background for this book,
but is not absolutely necessary; a great deal can be understood
with only a certain level of maturity. Indeed, combinatorics is well
known for being accessible. But readers should nd this book
challenging and will be expected to ll in details (that we hope
are instructive and not too dicult). We mention in passing that
we believe there is no substitute for a human teacher when trying
to learn a subject. An acquaintance with calculus, groups, nite
elds, elementary number theory, and especially linear algebra will
be necessary for some topics. Both undergraduates and graduate
students take the course at Caltech. The material in every chapter
has been presented in class, but we have never managed to do all
the chapters in one year.
The notes at the end of chapters often include biographical re-
marks on mathematicians. We have chosen to refrain from any
mention of living mathematicians unless they have retired (with
the exception of P. Erd os).
Exercises vary in diculty. For some it may be necessary to
consult the hints in Appendix 1. We include a short discussion of
formal power series in Appendix 2.
This manuscript was typeset by the authors in A
M
S-T
E
X.
J. H. v. L., R. M. W.
Eindhoven and Pasadena, 1992
Preface to the 2nd edition
The favorable reception of our book and its use for a variety of
courses on combinatorial mathematics at numerous colleges and
universities has encouraged us to prepare this second edition. We
have added new material and have updated references for this ver-
sion. A number of typographical and other errors have been cor-
rected. We had to change this century to the last century in
several places.
The new material has, for the most part, been inserted into the
chapters with the same titles as in the rst edition. An exception
is that the material of the later chapters on graph theory has been
reorganized into four chapters rather than two. The added material
includes, for example, discussion of the Lov asz sieve, associative
block designs, and list colorings of graphs.
Many new problems have been added, and we hope that this last
change, in particular, will increase the value of the book as a text.
We have decided not to attempt to indicate in the book the level
of diculty of the various problems, but remark again that this
can vary greatly. The diculty will often depend on the experience
and background of the reader, and an instructor will need to decide
which exercises are appropriate for his or her students. We like the
idea of stating problems at the point in the text where they are
most relevant, but have also added some problems at the end of
the chapters. It is not true that the problems appearing later are
necessarily more dicult than those at the beginning of a chapter.
A number of the hints and comments in Appendix 1 have been
improved.
xiv A Course in Combinatorics
Preparation of the second edition was done during a six-month
visit to the California Institute of Technology by the rst author
as Moore Distinguished Scholar. He gratefully acknowledges the
support of the Moore Foundation.
1
Graphs
A graph G consists of a set V (or V (G)) of vertices, a set E (or
E(G)) of edges, and a mapping associating to each edge e E(G)
an unordered pair x, y of vertices called the endpoints (or simply
the ends) of e. We say an edge is incident with its ends, and that
it joins its ends. We allow x = y, in which case the edge is called
a loop. A vertex is isolated when it is incident with no edges.
It is common to represent a graph by a drawing where we repre-
sent each vertex by a point in the plane, and represent edges by line
segments or arcs joining some of the pairs of points. One can think
e.g. of a network of roads between cities. A graph is called planar
if it can be drawn in the plane such that no two edges (that is, the
line segments or arcs representing the edges) cross. The topic of
planarity will be dealt with in Chapter 33; we wish to deal with
graphs more purely combinatorially for the present.
edge ends
a x, z
b y, w
c x, z
d z, w
e z, w
f x, y
g z, w
Figure 1.1
Thus a graph is described by a table such as the one in Fig. 1.1
that lists the ends of each edge. Here the graph we are describing
2 A Course in Combinatorics
has vertex set V = x, y, z, w and edge set E = a, b, c, d, e, f, g;
a drawing of this graph may be found as Fig. 1.2(iv).
A graph is simple when it has no loops and no two distinct edges
have exactly the same pair of ends. Two nonloops are parallel
when they have the same ends; graphs that contain them are called
multigraphs by some authors, or are said to have multiple edges.
If an ordered pair of vertices is associated to each edge, we have
a directed graph or digraph. In a drawing of a digraph, we use an
arrowhead to point from the rst vertex (the tail) towards the sec-
ond vertex (the head) incident with an edge. For a simple digraph,
we disallow loops and require that no two distinct edges have the
same ordered pair of ends.
When dealing with simple graphs, it is often convenient to iden-
tify the edges with the unordered pairs of vertices they join; thus
an edge joining x and y can be called x, y. Similarly, the edges
of a simple digraph can be identied with ordered pairs (x, y) of
distinct vertices.
(i) graph (ii) graph with loop (iii) digraph (iv) multiple edges
Figure 1.2
There are several ways to draw the same graph. For example,
the two graphs of Fig. 1.3 are essentially the same.
We make this more precise, but to avoid unnecessarily technical
denitions at this point, let us assume that all graphs are undirected
and simple for the next two denitions.
We say two graphs are isomorphic if there is a one-to-one cor-
respondence between the vertex sets such that if two vertices are
joined by an edge in one graph, then the corresponding vertices are
joined by an edge in the other graph. To show that the two graphs
in Fig. 1.3 are the same, nd a suitable numbering of the vertices
1. Graphs 3
in both graphs (using 1, 2, 3, 4, 5, 6) and observe that the edge sets
are the same sets of unordered pairs.
Figure 1.3
A permutation of the vertex set of a graph G with the property
that a, b is an edge if and only if (a), (b) is an edge, is called
an automorphism of G.
Problem 1A. (i) Show that the drawings in Fig. 1.4 represent the
same graph (or isomorphic graphs).
(ii) Find the group of automorphisms of the graph in Fig. 1.4.
Remark: There is no quick or easy way to do this unless you are
lucky; you will have to experiment and try things.
Figure 1.4
The complete graph K
n
on n vertices is the simple graph that
has all
_
n
2
_
possible edges.
Two vertices a and b of a graph G are called adjacent if they are
distinct and joined by an edge. We will use (x) to denote the set
of all vertices adjacent to a given vertex x; these vertices are also
called the neighbors of x.
4 A Course in Combinatorics
The number of edges incident with a vertex x is called the degree
or the valency of x. Loops are considered to contribute 2 to the
valency, as the pictures we draw suggest. If all the vertices of a
graph have the same degree, then the graph is called regular.
One of the important tools in combinatorics is the method of
counting certain objects in two dierent ways. It is a well known
fact that if one makes no mistakes, then the two answers are the
same. We give a rst elementary example. A graph is nite when
both E(G) and V (G) are nite sets. We will be primarily con-
cerned with nite graphs, so much so that it is possible we have
occasionally forgotten to specify this condition as a hypothesis in
some assertions.
Theorem 1.1. A nite graph G has an even number of vertices
with odd valency.
Proof: Consider a table listing the ends of the edges, as in Fig.
1.1. The number of entries in the right column of the table is twice
the number of edges. On the other hand, the degree of a vertex x
is, by denition, the number of times it occurs in the table. So the
number of entries in the right column is
(1.1)

xV (G)
deg(x) = 2[E(G)[.
The assertion follows immediately.
The equation (1.1) is simple but important. It might be called
the rst theorem of graph theory, and our Theorem 1.1 is its rst
corollary.
A subgraph of a graph G is a graph H such that V (H) V (G),
E(H) E(G), and the ends of an edge e E(H) are the same
as its ends in G. H is a spanning subgraph when V (H) = V (G).
The subgraph of G induced by a subset S of vertices of G is the
subgraph whose vertex set is S and whose edges are all the edges
of G with both ends in S.
A walk in a graph G consists of an alternating sequence
x
0
, e
1
, x
1
, e
2
, x
2
, . . . , x
k1
, e
k
, x
k
1. Graphs 5
of vertices x
i
, not necessarily distinct, and edges e
i
so that the ends
of e
i
are exactly x
i1
and x
i
, i = 1, 2, . . . , k. Such a walk has length
k. If the graph is simple, a walk is determined by its sequence of
vertices, any two successive elements of which are adjacent.
If the edge terms e
1
, . . . , e
k
are distinct, then the walk is called a
path from x
0
to x
k
. If x
0
= x
k
, then a walk (or path) is called closed.
A simple path is one in which the vertex terms x
0
, x
1
, . . . , x
k
are
also distinct, although we say we have a simple closed path when
k 1 and all vertex terms are distinct except x
0
= x
k
.
If a path from x to y exists for every pair of vertices x, y of G,
then G is called connected. Otherwise G consists of a number of
connected components (maximal connected subgraphs). It will be
convenient to agree that the null graph with no vertices and no
edges is not connected.
Problem 1B. Suppose G is a simple graph on 10 vertices that is
not connected. Prove that G has at most 36 edges. Can equality
occur?
The length of the shortest walk from a to b, if such walks exist, is
called the distance d(a, b) between these vertices. Such a shortest
walk is necessarily a simple path.
Example 1.1. A well known graph has the mathematicians of the
world as vertices. Two vertices are adjacent if and only if they
have published a joint paper. The distance in this graph from
some mathematician to the vertex P. Erd os is known as his or her
Erd os-number.
Figure 1.5
A polygon is the graph of a simple closed path, but more pre-
cisely it can be dened as a nite connected graph that is regular
of degree 2. There is, up to isomorphism, exactly one polygon P
n
6 A Course in Combinatorics
with n vertices (often called the n-gon) for each positive integer n.
The sequence of polygons is shown in Fig. 1.5.
A connected graph that contains no simple closed paths, i.e. that
has no polygons as subgraphs, is called a tree.
Problem 1C. Show that a connected graph on n vertices is a tree
if and only if it has n 1 edges.
Problem 1D. The complete bipartite graph K
n,m
has n + m ver-
tices a
1
, . . . , a
n
and b
1
, . . . , b
m
, and as edges all mn pairs a
i
, b
j
.
Show that K
3,3
is not planar.
No introduction to graph theory can omit the problem of the
bridges of K onigsberg (formerly a city in Prussia). The river Pregel
owed through this city and split into two parts. In the river was
the island Kneiphof. There were seven bridges connecting dierent
parts of the city as shown in the diagram of Fig. 1.6.
Figure 1.6
In a paper written in 1736 by L. Euler (considered the rst paper
on graph theory) the author claims that the following question was
considered dicult: Is it possible to make a walk through the city,
returning to the starting point and crossing each bridge exactly
once? This paper has led to the following denition. A closed path
through a graph using every edge once is called an Eulerian circuit
and a graph that has such a path is called an Eulerian graph.
Theorem 1.2. A nite graph G with no isolated vertices (but pos-
sibly with multiple edges) is Eulerian if and only if it is connected
and every vertex has even degree.
1. Graphs 7
Proof: That G must be connected is obvious. Since the path
enters a vertex through some edge and leaves by another edge, it
is clear that all degrees must be even. To show that the conditions
are sucient, we start in a vertex x and begin making a path. We
keep going, never using the same edge twice, until we cannot go
further. Since every vertex has even degree, this can only happen
when we return to x and all edges from x have been used. If
there are unused edges, then we consider the subgraph formed by
these edges. We use the same procedure on a component of this
subgraph, producing a second closed path. If we start this second
path in a point occurring in the rst path, then the two paths can
be combined to a longer closed path from x to x. Therefore the
longest of these paths uses all the edges.
The problem of the bridges of K onigsberg is described by the
graph in Fig. 1.6. No vertex has even degree, so there is no Eulerian
circuit.
One can consider a similar problem for digraphs. The necessary
and sucient condition for a directed Eulerian circuit is that the
graph is connected and that each vertex has the same in-degree
as out-degree.
Example 1.2. A puzzle with the name Instant Insanity concerns
four cubes with faces colored red, blue, green, and yellow, in such a
way that each cube has at least one face of each color. The problem
is to make a stack of these cubes so that all four colors appear on
each of the four sides of the stack. In Fig. 1.7 we describe four
possible cubes in attened form.
R
R Y G B
R
cube 1
R
R Y B G
Y
cube 2
G
B B R Y
G
cube 3
B
G Y R G
Y
cube 4
Figure 1.7
It is not a very good idea to try all possibilities. A systematic
approach is as follows. The essential information about the cubes
is given by the four graphs in Fig. 1.8.
8 A Course in Combinatorics
Figure 1.8
An edge indicates that the two adjacent colors occur on opposite
faces of the cube. We obtain a graph G by superposition of the
four graphs and number the edges according to their origin. It is
not dicult to see that we need to nd in G two subgraphs that
are regular of degree 2, with edges numbered 1, 2, 3, 4 and such that
they have no edge in common. One of the subgraphs tells us which
pairs of colors to align on the left side and right side of the stack.
The other graph describes the colors on front and back. Of course
it is easy to rotate the cubes in such a way that the colors are where
we wish them to be. The point of the example is that it takes only
a minute to nd two subgraphs as described above. In this example
the solution is unique.
We mention a concept that seems similar to Eulerian circuits but
that is in reality quite dierent. A Hamiltonian circuit in a graph
G is a simple closed path that passes through each vertex exactly
once (rather than each edge). So a graph admits a Hamiltonian
circuit if and only if it has a polygon as a spanning subgraph. In
the mid-19th century, Sir William Rowan Hamilton tried to popu-
larize the exercise of nding such a closed path in the graph of the
dodecahedron (Fig. 1.9).
Figure 1.9
1. Graphs 9
The graph in Fig. 1.4 is called the Petersen graph (cf. Chapter 21)
and one of the reasons it is famous is that it is not Hamiltonian;
it contains n-gons only for n = 5, 6, 8, 9, and not when n = 7 or
n = 10.
By Theorem 1.2, it is easy to decide whether a graph admits an
Eulerian circuit. A computer can easily be programmed to check
whether the degrees of a graph are even and whether the graph is
connected, and even to produce an Eulerian circuit when one exists.
In contrast to this, the problem of deciding whether an arbitrary
graph admits a Hamiltonian circuit is likely intractable. To be
more precise, it has been proved to be NP-completesee Garey
and Johnson (1979).
Problem 1E. Let A
1
, . . . , A
n
be n distinct subsets of the n-set
N := {1, . . . , n}. Show that there is an element x N such that
the sets A
i
\{x}, 1 i n, are all distinct. To do this, form a graph
G on the vertices A
i
with an edge with color x between A
i
and A
j
if and only if the symmetric dierence of the sets A
i
and A
j
is {x}.
Consider the colors occurring on the edges of a polygon. Show that
one can delete edges from G in such a way that no polygons are
left and the number of dierent colors remains the same. Then use
1C. (This idea is due to J. A. Bondy (1972).)
Problem 1F. The girth of a graph is the length of the smallest
polygon in the graph. Let G be a graph with girth 5 for which all
vertices have degree d. Show that G has at least d
2
+ 1 vertices.
Can equality hold?
Problem 1G. Show that a nite simple graph with more than one
vertex has at least two vertices with the same degree.
Problem 1H. A graph on the vertex set {1, 2, . . . , n} is often de-
scribed by a matrix A of size n, where a
ij
and a
ji
are equal to
the number of edges with ends i and j. What is the combinatorial
interpretation of the entries of the matrix A
2
?
Problem 1I. Let Q := {1, 2, . . . , q}. Let G be a graph with the
elements of Q
n
as vertices and an edge between (a
1
, a
2
, . . . , a
n
) and
(b
1
, b
2
, . . . , b
n
) if and only if a
i
= b
i
for exactly one value of i. Show
that G is Hamiltonian.
10 A Course in Combinatorics
Problem 1J. Let G be a simple graph on n vertices (n > 3) with
no vertex of degree n 1. Suppose that for any two vertices of G,
there is a unique vertex joined to both of them.
(i) If x and y are not adjacent, prove that they have the same
degree.
(ii) Now show that G is a regular graph.
Notes.
Paul Erd os (19131996) (cf. Example 1.1) was probably the most
prolic mathematician of the 20th century with well over 1400 pa-
pers having been published. His contributions to combinatorics,
number theory, set theory, etc., include many important results.
He collaborated with many mathematicians all over the world, all
of them proud to have Erd os-number 1, among them the authors
of this book; see J. W. Grossman (1997).
Leonhard Euler (17071783) was a Swiss mathematician who
spent most of his life in St. Petersburg. He was probably the most
productive mathematician of all times. Even after becoming blind
in 1766, his work continued at the same pace. The celebration in
1986 of the 250th birthday of graph theory was based on Eulers
paper on the K onigsberg bridge problem. K onigsberg is now the
city of Kaliningrad in Russia.
For an elementary introduction to graph theory, we recommend
R. J. Wilson (1979), and J. J. Watkins and R. J. Wilson (1990).
Sir William Rowan Hamilton (18051865) was an Irish mathe-
matician. He was considered a genius. He knew 13 languages at
the age of 12 and was appointed professor of astronomy at Trinity
College Dublin at the age of 22 (before completing his degree). His
most important work was in mathematical physics.
References.
M. Garey and D. S. Johnson (1979), Computers and Intractability;
A Guide to the Theory of NP-completeness, W. H. Freeman and
Co.
J. W. Grossman (1997), Paul Erd os: The Master of Collaboration,
pp. 467475 in The Mathematics of Paul Erd os, R. L. Graham
and J. Nesetril (eds.), Springer-Verlag.
1. Graphs 11
J. J. Watkins and R. J. Wilson (1990), Graphs (An Introductory
Approach), J. Wiley & Sons.
R. J. Wilson (1979), Introduction to Graph Theory, Longman.
2
Trees
We come to the rst not so easy theorem. It is due to A. Cayley
(1889). We shall give three dierent proofs here. Two more proofs
will occur in later chapters; see Example 14.14 and Example 38.2.
The rst two proofs illustrate a method that is used very often in
combinatorics. In order to count certain objects that seem hard to
count, one nds a one-to-one mapping onto a set of other objects
whose number is easier to determine.
Theorem 2.1. There are n
n2
dierent labeled trees on n vertices.
The term labeled emphasizes that we are not identifying isomor-
phic graphs. We have xed the set of vertices, and two trees are
counted as the same if and only if exactly the same pairs of vertices
are adjacent. A spanning tree of a connected graph G is a spanning
subgraph of G that is a tree. The theorem could have been stated:
the complete graph K
n
has n
n2
spanning trees.
Example 2.1. Here are the 16 labeled trees on four vertices:
2. Trees 13
Example 2.2. There are three nonisomorphic trees on ve ver-
tices:
The number of spanning trees in K
5
isomorphic to a specic tree
T on ve vertices is 5! divided by the order of the automorphism
group of T (why?). Thus there are 5!/4! = 5 trees in K
5
isomorphic
to the rst tree above, and 5!/2 = 60 trees isomorphic to either of
the other two trees, for a total of 125 spanning trees.
Problem 2A. Find the six nonisomorphic trees on 6 vertices, and
for each compute the number of distinct spanning trees in K
6
iso-
morphic to it.
Before starting the proofs, we make the following observations.
(Probably the reader has already noticed these things in solving
Problem 1C.) Firstly, every tree with n 2 vertices has at least
two monovalent vertices (vertices of degree 1). This is immediate,
for example, from Problem 1C and equation (1.1): the sum of the
degrees d
1
, d
2
, . . . , d
n
, all of which are at least 1, is 2n2. Secondly,
if a monovalent vertex and its incident edge are deleted from a tree,
the resulting graph is still a tree. Finally, given a tree T, if we
introduce a new vertex x and a new edge joining x to any vertex
of T, the new graph is again a tree.
Figure 2.1
Proof 1: The rst proof we present, due to H. Pr ufer (1918), uses
an algorithm that associates to any tree T a name T(T) (called
the Pr ufer code) that characterizes the tree.
14 A Course in Combinatorics
For the vertices of K
n
, we take the ordered set V = 1, 2, 3, . . . , n.
Given a spanning tree T in K
n
, we let T
1
= T and generate a se-
quence of trees T
1
, T
2
, . . . , T
n1
and two sequences of vertices as
follows: Given the tree T
i
with ni +1 vertices, i = 1, 2, . . . , n1,
let x
i
be the least monovalent vertex of T
i
and delete x
i
and its
incident edge x
i
, y
i
from T
i
to obtain a tree T
i+1
on ni vertices.
The name of T is to be
T(T) = (y
1
, y
2
, . . . , y
n2
).
We claim that the mapping T, from the set of all spanning trees
in K
n
to the set V
n2
of all possible names, is one-to-one and onto
(bijective). This will prove that the number of spanning trees in
K
n
is n
n2
.
For the tree in Fig. 2.1, where n = 10, we have (x
1
, y
1
) = (3, 2),
(x
2
, y
2
) = (4, 2), (x
3
, y
3
) = (2, 1),. . . , (x
9
, y
9
) = (9, 10); these edges
are the columns of the matrix below.
_
3 4 2 5 6 7 1 8 9
2 2 1 1 7 1 10 10 10
_
So T(T) = (2, 2, 1, 1, 7, 1, 10, 10). Dont include y
9
= 10.
To understand why T is bijective, we rst note some simple facts
about the x
i
s and y
i
s. First, y
n1
= n, always. This is because
every tree (with at least two vertices) has at least two monovalent
vertices, so the vertex n will never be the least monovalent vertex.
Second, x
k
, x
k+1
, . . . , x
n1
and n are the vertices of the tree T
k
.
Third, x
i
, y
i
, k i n 1, are exactly the edges of T
k
, in some
order.
The number of times a vertex v occurs among y
1
, y
2
, . . . , y
n2
is
deg
T
(v)1. This is because v occurs deg
T
(v) times among the edges
x
i
, y
i
, 1 i n 1, and exactly once in x
1
, x
2
, . . . , x
n1
, y
n1
.
Similarly, the number of times a vertex v of T
k
occurs among
y
k
, y
k+1
, . . . , y
n2
is its degree in the tree T
k
less 1. In particular,
the monovalent vertices of T
k
are those elements of V not in
x
1
, x
2
, . . . , x
k1
y
k
, y
k+1
, . . . , y
n1
,
and this means that x
k
, the least monovalent vertex of T
k
, is the
least element of 1, 2, . . . , n not in the above set. In particular,
2. Trees 15
x
1
is the least element of V not in the name T(T), and we can
uniquely determine x
k
from T(T) and x
1
, . . . , x
k1
.
Problem 2B. How many trees T are there on the set of vertices
1, 2, 3, 4, 5, 6, 7 in which the vertices 2 and 3 have degree 3, vertex
5 has degree 2, and hence all others have degree 1? Do not just
draw pictures but consider the possible Pr ufer codes of these trees.
Proof 2: We give another proof, again by a reversible algorithm.
Consider any mapping f from 2, 3, . . . , n 1 to 1, 2, . . . , n.
There are n
n2
such mappings f. Construct a digraph D on the
vertices 1 to n by dening (i, f(i)), i = 2, . . . , n1, to be the edges.
Fig. 2.2 shows an example with n = 21.
D consists of two trees rooted at 1 and n and a number (say k) of
circuits (directed polygons) to which trees are attached. (Directed
trees with all the edges pointing in the direction of one vertex, called
the root, are called arborescences.) These circuits are placed as in
Fig. 2.2 where the rightmost vertex in the i-th component, denoted
by r
i
, is its minimal element (and l
i
is the vertex on the left). The
circuits are ordered by the condition r
1
< r
2
< < r
k
. To D we
adjoin the tree obtained by adding the edges 1, l
1
, r
1
, l
2
, . . . ,
r
k1
, l
k
, r
k
, n and deleting the edges r
i
, l
i
as in Fig. 2.3.
Figure 2.2
If the tree of Fig. 2.3 is given, consider the path from 1 to n
(=21). Let r
0
:= 1. Dene r
1
to be the minimal number on this
path (excluding r
0
= 1) and in general r
i
as the minimal number
on the path from r
i1
to n. It is easily seen that we recover the
function f in this way.
16 A Course in Combinatorics
Figure 2.3
Generalizations of this proof may be found in E gecioglu and Rem-
mel (1986).
Problem 2C. Let G be a directed graph with vertices x
1
, . . . , x
n
for which a (directed) Eulerian circuit exists. A spanning arbores-
cence with root x
i
is a spanning tree T of G, with root x
i
, such
that for all j ,= i there is a directed path from x
j
to x
i
in T. Show
that the number of spanning arborescences of G with root x
i
does
not depend on i. (This is dicult; see the hints.)
Proof 3: We now give a proof by counting since it is useful to
have seen this method. We remind the reader of the denition of
a multinomial coecient. Let r
1
, r
2
, . . . , r
k
be nonnegative integers
with sum n. Then
_
n
r
1
,...,r
k
_
is dened by
(2.1) (x
1
+x
2
+ +x
k
)
n
=

_
n
r
1
, . . . , r
k
_
x
r
1
1
x
r
2
2
. . . x
r
k
k
,
where the sum is over all k-tuples (r
1
, . . . , r
k
) with sum n.
Since (x
1
+ +x
k
)
n
= (x
1
+ +x
k
)
n1
(x
1
+ +x
k
), we have
(2.2)
_
n
r
1
, . . . , r
k
_
=
k

i=1
_
n 1
r
1
, . . . , r
i
1, . . . , r
k
_
.
We denote the number of labeled trees with n vertices for which the
degrees are d
1
, d
2
, . . . , d
n
by t(n; d
1
, d
2
, . . . , d
n
). Clearly this number
is 0 if one of the d
i
is 0. The value of t(n; d
1
, d
2
, . . . , d
n
) depends
only on the multiset of numbers d
i
and not on their order. We
may assume without loss of generality that d
1
d
2
d
n
,
so d
n
= 1. Take the vertex v
n
corresponding to d
n
. It is joined to
some vertex v
i
of degree d
i
2, and any of the remaining vertices
2. Trees 17
is a candidate. Therefore
(2.3) t(n; d
1
, . . . , d
n
) =
n1

i=1
t(n 1; d
1
, . . . , d
i
1, . . . , d
n1
).
It is trivial to check by hand that
(2.4) t(n; d
1
, . . . , d
n
) =
_
n 2
d
1
1, . . . , d
n
1
_
for n = 3. Since the numbers on the left-hand side, respectively
right-hand side, of (2.4) satisfy the same recurrence relation ( (2.3),
respectively (2.2) ) it follows by induction that (2.4) is true for all n.
In (2.1), we replace n by n 2, k by n, r
i
by d
i
1 and x
i
by 1.
We nd
n
n2
=

t(n; d
1
, d
2
, . . . , d
n
).

Compare (2.4) with Problem 2B.

A spanning tree is easily constructed by starting at any vertex,
taking the edges to vertices at distance 1, then one edge to each
vertex at distance 2, etc. Several other constructions are possible
(e.g. by starting with G and deleting suitable edges).
A graph with no polygons as subgraphs is called a forest. Each
component C
1
, C
2
, . . . , C
k
of a forest G is a tree, so if a forest with
n vertices has k components, it has
([V (C
1
)[ 1) + ([V (C
2
)[ 1) + + ([V (C
k
)[ 1) = n k
edges.
A weighted graph is a graph G together with a function associat-
ing a real number c(e) (usually nonnegative) to each edge e, called
its length or cost according to context. Let us use the term cost
here. Given a weighted connected graph G, dene the cost of a
spanning tree T of G as
c(T) :=

eE(T)
c(e).
18 A Course in Combinatorics
The graph may represent a network of cities where c(x, y) is the
cost of erecting a telephone line joining cities x and y, and so it
is clear that nding a cheapest spanning tree in G is a problem of
practical importance.
The following method is often called the greedy algorithm. In fact,
it is only one of a number of algorithms which can be called greedy
algorithms, where one does not plan ahead but takes what seems
to be the best alternative at each moment and does not look back.
It is surprising that such a simple procedure actually produces a
cheapest spanning tree, but this is proved in Theorem 2.2 below.
Let us say that a set S of edges of a graph G is independent when
the spanning subgraph with edge set S (denoted G:S) is a forest.
Greedy algorithm. Let G be a connected weighted graph with
n vertices. At each point, we will have a set e
1
, e
2
, . . . , e
i
of i
independent edges (i = 0 to start), so that G:e
1
, e
2
, . . . , e
i
has ni
components. If i < n1, let e
i+1
be an edge with ends in dierent
components of G:e
1
, e
2
, . . . , e
i
and whose cost is minimum with
respect to this property. Stop when we have chosen n 1 edges.
Theorem 2.2. With e
1
, . . . , e
n1
chosen as above, the spanning
tree T
0
:= G:e
1
, . . . , e
n1
has the property that c(T
0
) c(T) for
any spanning tree T.
Proof: Let a
1
, a
2
, . . . , a
n1
be the edge set of a tree T, numbered
so that c(a
1
) c(a
2
) c(a
n1
). We claim something much
stronger than c(T
0
) c(T); namely, we claim that c(e
i
) c(a
i
) for
each i = 1, 2 . . . , n 1. If this is false, then
c(e
k
) > c(a
k
) c(a
k1
) c(a
1
)
for some k. Since none of a
1
, a
2
, . . . , a
k
was chosen at the point
when e
k
was chosen, each of these k edges has both ends in the
same component of G:e
1
, e
2
, . . . , e
k1
. Then the number of com-
ponents of G:a
1
, a
2
, . . . , a
k
is at least the number n k + 1 of
components of G:e
1
, e
2
, . . . , e
k1
and this contradicts the fact that
a
1
, a
2
, . . . , a
k
is independent.
Problem 2D. Here is a variation on the above greedy algorithm.
Let x
1
be any vertex of a weighted connected graph G with n
2. Trees 19
vertices and let T
1
be the subgraph with the one vertex x
1
and
no edges. After a tree (subgraph) T
k
, k < n, has been dened,
let e
k
be a cheapest edge among all edges with one end in V (T
k
)
and the other end not in V (T
k
), and let T
k+1
be the tree obtained
by adding that edge and its other end to T
k
. Prove that T
n
is a
cheapest spanning tree in G.
In many practical situations, it is necessary to search through a
tree starting from a particular vertex. (A tree with a distinguished
vertexthe rootis called a rooted tree.) There are two well known
methods known as depth-rst search and breadth-rst search. We
explain the terminology by the example of Fig. 2.4.
Figure 2.4
In a depth-rst search starting at a, one essentially considers the
tree as a fence and walks around it, keeping the fence on the left,
i.e. along the walk abdidjdbebfk . . . lhca. If one decides to number
the vertices in accordance with the search, one nds the numbering
a = 1, b = 2, d = 3, i = 4, . . . , l = 12. In this description, we rely
on a planar drawing of the tree; but see below.
In a breadth-rst search, one proceeds as in the construction of
a spanning tree mentioned above. The vertices are then numbered
in the order of Fig. 2.4, i.e. alphabetically.
These ideas apply, more generally, to searching through the ver-
tices of a connected graph.
Given a nite connected graph G, we can obtain a numbering
of the vertices of G and a spanning tree T of G, called a depth-
rst search tree for G, in the following manner. Pick a vertex v
0
and start with the tree T
0
with vertex v
0
and no edges. Proceed
inductively: once vertices v
0
, v
1
, v
2
, . . . , v
k
and a tree T
k
with exactly
those vertices and some of the edges of G have been chosen, let
20 A Course in Combinatorics
be the largest index k so that v

is adjacent to some vertex not

in T
k
. Such a vertex will exist if T
k
is not yet a spanning tree;
otherwise stop and let T = T
k
. Call that new vertex v
k+1
and add
it and the edge v

, v
k+1
to T
k
to obtain a tree T
k+1
. We consider
T as a rooted tree with root v
0
.
We give two properties of depth-rst search trees and use them
to give a simple constructive proof of a theorem on orientations of
graphs.
Given a vertex x of a rooted tree with root v
0
, the ancestors of x
are the vertices traversed by the (unique) path from x to the root
v
0
. The rst vertex other than x on that path is the parent of x. If
x is an ancestor of y, we also say that y is a descendant of x. We
will count x as a descendent and ancestor of itself.
Proposition 2.3. If vertices x and y are adjacent in G, then one
of them is a descendant of the other in any depth-rst search tree
T of G.
Proof: Suppose x is numbered with a smaller index than y in the
depth-rst search, say x = v
k
.
At the stage when v
0
, v
1
, v
2
, . . . , v
k
have been chosen, the largest
index so that v

is adjacent to an as-yet-unnumbered vertex is

clearly = k. So v
k+1
is joined to v
k
(and not to some v
i
with i < k)
in T. If v
k+1
= y, we are done; y is a descendant of v
k
. Otherwise,
since v
k
is still adjacent to an as-yet-unnumbered vertex, namely
y, the choice of will be = k or = k +1, and v
k+2
is adjacent to
either v
k
or v
k+1
in T. If v
k+2
= y, we are done; y is a descendant
of v
k
.
Inductively, as long as y remains unnumbered, v
k
, v
k+1
, . . . , v
k+j
will be descendants of v
k
and the next choice of will be as one of
k, k + 1, . . . , k + j. Then the vertex v
k+j+1
numbered next will be
adjacent to one of v
k
, v
k+1
, . . . , v
k+j
and hence will be a descendant
of v
k
. Since the graph is nite, eventually this newly numbered
vertex must be y.
An isthmus of a connected graph G is an edge whose deletion
results in a disconnected graph. (Some authors call this a bridge.)
Proposition 2.4. Let x, y be an edge of T which is not an
isthmus in G; say x is the parent of y. Then there is an edge in G
2. Trees 21
but not in T joining some descendant a of y and some ancestor b
of x.
Proof: Let D be the set of descendants of y. So y D and x / D.
Since G with x, y deleted is still connected, there is some edge
a, b ,= x, y with one end a D and the other end b / D. This
edge a, b is certainly not in T and by Proposition 2.3, b is an
ancestor of a (since it cannot be a descendant of a because then it
would be a descendant of y too and hence would be in D). The
(unique) path in T from a to v
0
passes through y (an ancestor of
a) and then x (the parent of y); but b must be on this path (since
it is an ancestor of a), so b must be an ancestor of x too.
Any directed graph obtained from an undirected graph G by
assigning a direction to each edge of G is called an orientation of
G. A walk in a digraph D may be called strong when each edge is
traversed by the walk according to its direction, i.e. from its tail to
its head, and the digraph D is strongly connected when for any two
vertices x, y, there is a strong walk from x to y.
Theorem 2.5. Let G be a nite connected graph without isth-
muses. Then G admits a strong orientation, i.e. an orientation
that is a strongly connected digraph.
Proof: We will construct a digraph D from G by choosing a direc-
tion for each edge. Find a depth-rst search tree T and numbering
v
0
, v
1
, . . . of the vertices of G. Let v
i
, v
j
be an edge of G with
i < j. If v
i
, v
j
is in T, direct it from v
i
to v
j
, i.e. (v
i
, v
j
) is an
edge of D. If v
i
, v
j
is in not T, direct it from v
j
to v
i
, i.e. (v
j
, v
i
)
is an edge of D.
It remains to show that D is strongly connected. There is a
strong walk from v
0
to any vertex x of G (using only edges of the
tree T), so it will suce to show that we can nd a strong walk
from any vertex x to v
0
.
Given a vertex x
k
, k > 0, Proposition 2.4 says that some edge
a, b in G but not in T joins some descendant a of y to some
ancestor b = v
i
of v
k
. We get a strong walk in D from v
k
to v
i
by
appending the directed edge (a, v
i
) to a strong walk in T from v
k
to its descendant a. Of course, i < k since v
i
is an ancestor of v
k
.
If i = 0 we are done. Otherwise, we repeat the argument to nd
22 A Course in Combinatorics
a strong walk from v
i
to some v
j
with j < i and concatenate the
walks to get a strong walk from v
k
to v
j
. Continue in this manner
until you reach v
0
.
Problem 2E. A graceful labeling of a tree T on n vertices is a
mapping f : V (T) 1, 2, . . . , n so that the numbers [f(x)f(y)[
computed across edges x, y are all dierent. Show that the path-
graphs (trees with exactly two monovalent vertices) admit graceful
labelings. (It is conjectured that all trees admit graceful labelings.)
Problem 2F. Suppose a tree G has exactly one vertex of degree
i for 2 i m and all other vertices have degree 1. How many
vertices does G have?
Problem 2G. Let G be a graph with exactly one vertex of degree
i for 2 i m and k other vertices, all of degree 1. Prove that
k
m+3
2
|. Give a construction for such a graph.
Problem 2H. Consider labeled trivalent rooted trees T with 2n
vertices, counting the root labeled 2n; see Figure 14.3. The labels
are chosen in such a way that the procedure leading to T(T) has
1, 2, 3, . . . , 2n 1 as rst row. How many possible codes T(T) are
there?
Notes.
A. Cayley (18211895), professor at Cambridge from 1863 until
his death, was one of the great mathematicians of the 19th century.
His work includes important contributions to the theory of elliptic
functions, analytic geometry and algebra, e.g. the theory of invari-
ants. His paper on trees appeared in 1889 but it did not contain
what we would consider a proof. Of the many proofs (ve of which
are treated in this book) the one by Prufer is the best known.
H. Prufer (18961934) was one of I. Schurs many pupils. He was
professor at Munster.
References.
A. Cayley (1889), A theorem on trees, Quart. J. Pure and App.
Math. 23, 376378.
2. Trees 23

O. E gecioglu and J. B. Remmel (1986), Bijections for Cayley trees,

spanning trees, and their q-analogues, J. Combinatorial Theory
(A) 42, 1530.
H. Pr ufer (1918), Neuer Beweis eines Satzes uber Permutationen,
Archiv der Math. und Phys. (3) 27, 142144.
3
Colorings of graphs and
Ramseys theorem
We shall rst look at a few so-called coloring problems for graphs.
A proper coloring of a graph G is a function from the vertices to
a set C of colors (e.g. C = 1, 2, 3, 4) such that the ends of every
edge have distinct colors. (So a graph with a loop will admit no
proper colorings.) If [C[ = k, we say that G is k-colored.
The chromatic number (G) of a graph G is the minimal number
of colors for which a proper coloring exists.
If (G) = 2 (or (G) = 1, which is the case when and only when
G has no edges), then G is called bipartite. A graph with no odd
polygons (equivalently, no closed paths of odd length) is bipartite
as the reader should verify.
The famous Four Color Theorem (K. Appel and W. Haken,
1977) states that if G is planar, then (G) 4.
Clearly (K
n
) = n. If k is odd then (P
k
) = 3. In the following
theorem, we show that, with the exception of these examples, the
chromatic number is at most equal to the maximum degree (R. L.
Brooks, 1941).
Theorem 3.1. Let d 3 and let G be a graph in which all vertices
have degree d and such that K
d+1
is not a subgraph of G. Then
(G) d.
Proof 1: As is the case in many theorems in combinatorial analy-
sis, one can prove the theorem by assuming that it is not true, then
considering a minimal counterexample (in this case a graph with
the minimal number of vertices) and arriving at a contradiction.
We shall use the technique of recoloring: it is possible to change
3. Colorings of graphs and Ramseys theorem 25
the colors of certain vertices to go from one proper coloring to an-
other. For example, let S be a subset of the set C of colors. On any
connected component of the subgraph induced by the vertices with
colors from S, we arbitrarily permute the colors (without changing
those of the vertices with colors in CS). Clearly we again have a
proper coloring.
So let G be a counterexample with the minimum number of ver-
tices. Let x G and let (x) = x
1
, . . . , x
l
, l d. Since G is
a minimal counterexample, the graph H, obtained by deleting x
and the edges incident with x, has a d-coloring, say with colors
1, 2, . . . , d. If one of these colors is not used in the coloring of (x),
then we can assign this color to x and obtain a d-coloring of G. It
follows that l = d and every d-coloring of H must use all the colors
on the set (x). Let us assume that x
i
has color i for i = 1, 2, . . . , d.
Now consider x
i
and x
j
and the induced subgraph H
ij
of H with
colors i and j. If x
i
and x
j
were in dierent connected components
of H
ij
, then we could interchange the colors in one of these com-
ponents, after which x
i
and x
j
would have the same color, which
is impossible. So x
i
and x
j
are in the same component (say C
ij
)
of H
ij
. We shall now show that this component is (the graph of)
a simple path (with alternating colors i and j) from x
i
to x
j
. If
two neighbors of x
i
in H had color j, then the neighbors of x
i
in H
would have at most d 2 dierent colors. Then we could recolor
x
i
and that is impossible. Suppose y is the rst vertex on a path
from x
i
to x
j
in C
ij
that has degree 3. The neighbors of y in H
have at most d2 colors, so we can recolor y to some color / i, j
and then x
i
and x
j
are no longer connected in H
ij
, which we know
to be impossible. So such a y does not exist, proving that C
ij
is a
path.
Suppose that z is a vertex ,= x
i
on C
ij
and on C
ik
. Then z has two
neighbors with color j and two with color k. Again the neighbors
of z in H have at most d 2 colors and z can be recolored to some
color / i, j, k, again a contradiction. Hence C
ij
C
ik
= x
i
.
Our assumption that K
d+1
_ G shows that there are two vertices
in (x), say x
1
and x
2
, that are not connected by an edge. We have
the situation of Fig. 3.1. The vertex a is the neighbor of x
1
with
color 2 on C
12
.
26 A Course in Combinatorics
We recolor H by interchanging the colors 1 and 3 on the subgraph
C
13
. For the new coloring, we have new paths that we call C
/
ij
.
Clearly a C
/
23
(since x
1
now has color 3). However, on C
12
no point
except x
1
has changed color, so a C
/
12
. Hence C
/
12
C
/
23
,= x
2
,
contradicting what we proved above. The contradiction shows that
our assumption that a minimal counterexample exists is false.
Figure 3.1
In preparation for a second proof of Brooks Theorem, the reader
should do the following problem without applying the theorem.
Problem 3A. Fix an integer d 3. Let H be a simple graph with
all degrees d which cannot be d-colored and which is minimal
(with the fewest vertices) subject to these properties. (We claim H
is complete on d+1 vertices, but we dont know that yet.) (i) Show
that H is nonseparable (this means that every graph obtained from
H by deleting a vertex is connected). (ii) Then show that if the
vertex set V (H) is partitioned into sets X and Y with [Y [ 3,
then there are at least three vertices a, b, c Y each of which is
adjacent to at least one vertex in X.
Proof 2: Let d and H be as in Problem 3A. If H is not complete,
there are vertices x
1
, x
n1
, x
n
so that x
1
is adjacent to x
n1
and x
n
but so that these last two vertices are not adjacent. We want to
number the other n 3 vertices so that the sequence
x
1
, x
2
, . . . , x
n1
, x
n
has the property that each x
k
, k 2, is adjacent to at least one of
the vertices x
i
preceding it, i.e. with i < k. This is simple: when
x
1
, x
2
, . . . , x
k
have been chosen, k < n 2, choose x
k+1
to be any
3. Colorings of graphs and Ramseys theorem 27
vertex other than x
n1
or x
n
adjacent to one of x
1
, x
2
, . . . , x
k
; since
there are at least three such vertices, this is always possible.
Once this is done, we d-color the vertices starting at the end of
the sequence. Assign x
n1
and x
n
the same color. When the last
k vertices x
k+1
, . . . , x
n1
, x
n
have been colored, k 2, there is a
color available for x
k
since x
k
is adjacent to at most d 1 of the
already-colored vertices. Finally, there is a color available for x
1
since two of the vertices adjacent to x
1
have been given the same
color.
We now consider a coloring problem of a completely dierent
nature. It serves as an introduction to a very important theorem of
combinatorics, namely Ramseys theorem. Before reading on, the
reader should try the following problem.
Problem 3B. Let the edges of K
7
be colored with the colors red
and blue. Show that there are at least four subgraphs K
3
with all
three edges the same color (monochromatic triangles). Also show
that equality can occur.
The example that is always used to introduce this subject is K
6
with its edges colored red or blue. We shall show that there is at
least one monochromatic triangle. A proof is as follows. Let a be
any vertex. Because a has degree 5, it is incident with at least three
edges of the same color, say red edges to the vertices b, c, d. If one
of the edges between these three vertices is red, then we have a red
triangle; if not, then they form a blue triangle.
The idea of the proof is the same as for the more dicult situation
of Ramseys theorem. However, it does not show as much as the
following counting argument.
Theorem 3.2. If the edges of K
n
are colored red or blue, and r
i
,
i = 1, 2, . . . , n, denotes the number of red edges with vertex i as an
endpoint, and if denotes the number of monochromatic triangles,
then
(3.1) =
_
n
3
_

1
2
n

i=1
r
i
(n 1 r
i
).
Proof: Every triangle in K
n
that is not monochromatic has ex-
actly two vertices where a red and a blue edge meet. On the i-th
28 A Course in Combinatorics
vertex, two such edges can be chosen in r
i
(n1 r
i
) ways. So the
sum in (3.1) counts the bichromatic triangles twice.
Corollary.
(3.2)
_
n
3
_

n
2
(
n 1
2
)
2
||.
Proof: From (3.1) we see that is minimized if r
i
= (n 1 r
i
)
for all i when n is odd, or if r
i
=
n
2
or r
i
=
n
2
1 for all i in the
case that n is even. Since is an integer, the rst situation cannot
always arise. It is easy to show that (3.2) cannot be improved.
Note that this argument shows that a red-blue coloring of K
6
must always have at least two monochromatic triangles.
We now treat Ramseys theorem (Ramsey, 1930).
Theorem 3.3. Let r 1 and q
i
r, i = 1, 2, . . . , s be given.
There exists a minimal positive integer N(q
1
, q
2
, . . . , q
s
; r) with the
following property. Let S be a set with n elements. Suppose that
all
_
n
r
_
r-subsets of S are divided into s mutually exclusive families
T
1
, . . . , T
s
(colors). Then if n N(q
1
, q
2
, . . . , q
s
; r) there is an i,
1 i s, and some q
i
-subset of S for which every r-subset is in
T
i
.
(The reader should compare this with our introductory example
and show that N(3, 3; 2) = 6.)
Proof: We give the proof only for s = 2. The general case only
involves a little more bookkeeping.
(a) Trivially, the theorem is true for r = 1 and N(p, q; 1) =
p +q 1.
(b) For any r and p r it is also obvious that N(p, r; r) = p and
similarly N(r, q; r) = q for q r.
(c) We proceed by induction on r. So assume the theorem is true
for r 1. We now use induction on p + q, using (b). So we can
dene p
1
= N(p 1, q; r), q
1
= N(p, q 1; r). Let S be a set with
n elements, where n 1 + N(p
1
, q
1
; r 1). Let the r-subsets of S
be colored with two colors, say red and blue. As in the proof for
K
6
, we pick an arbitrary element a of S. We now dene a coloring
of the (r 1)-subsets of S
/
:= Sa by giving X S
/
the same
3. Colorings of graphs and Ramseys theorem 29
color as X a. By induction S
/
either contains a subset A of
size p
1
such that all its (r 1)-subsets are red or a subset B of size
q
1
such that all its (r 1)-subsets are colored blue. Without loss
of generality the rst situation occurs. Since A has N(p 1, q; r)
elements, there are two possibilities. The rst is that A has a subset
of q elements with all its r-subsets blue, in which case we are done.
The other possibility is that A has a subset A
/
of p 1 elements
with all its r-subsets red. The set A
/
a also has this property
because A
/
A. This proves the theorem and furthermore we have
shown
(3.3) N(p, q; r) N(N(p 1, q; r), N(p, q 1; r); r 1) + 1.

A special case of (3.3) occurs when we go back to the coloring of

edges (r = 2) of a graph with two colors. Using (a) from the proof,
we nd
(3.4) N(p, q; 2) N(p 1, q; 2) +N(p, q 1; 2).
Problem 3C. Show that equality cannot hold in (3.4) if both
terms on the right-hand side are even.
Theorem 3.4.
N(p, q; 2)
_
p +q 2
p 1
_
.
Proof: Since N(p, 2; 2) = p, the result follows from (3.4) because
binomial coecients satisfy the same relation with equality.
Let us look at what we now know about N(p, q; 2). By Problem
3C, we have N(3, 4; 2) 9. To show that equality holds, we have
to color K
8
such that there is no red triangle and no blue K
4
. We
do this as follows: number the vertices with the elements of Z
8
. Let
the edge i, j be red if and only if i j 3 or i j 4 (mod 8).
One easily checks that this coloring does the job.
Problem 3D. Use the same method to show that N(4, 4; 2) = 18
and that N(3, 5; 2) = 14.
With a lot more work it has been shown that
N(3, 6; 2) = 18, N(3, 7; 2) = 23, N(3, 8; 2) = 28, N(3, 9; 2) = 36,
30 A Course in Combinatorics
N(4, 5; 2) = 25.
No other values of N(p, q; 2) are known.
One of the interesting problems in this area that has seen virtu-
ally no progress for 30 years is the asymptotic behavior of N(p, p; 2).
We know by Theorem 3.4 that
(3.5) N(p, p; 2)
_
2p 2
p 1
_
2
2p2
.
We now show that N(p, p; 2) grows exponentially, using a method
that is used quite often in combinatorics. It is often referred to
as probabilistic since it estimates the probability that a random
coloring has a monochromatic K
p
. Consider a K
n
. There are 2
(
n
2
)
dierent ways of coloring the edges red or blue. Now x a subgraph
K
p
. There are 2
(
n
2
)

(
p
2
)
+1
colorings for which that K
p
is monochro-
matic. The number of colorings for which some K
p
is monochro-
matic is at most
_
n
p
_
times as large (because we may count some
colorings more than once). If this number is less than the total
number of colorings, then there exist colorings with no monochro-
matic K
p
. Using the fact that
_
n
p
_
< n
p
/p!, we nd that such a
coloring certainly exists if n < 2
p/2
(unless p = 2). This proves the
following theorem.
Theorem 3.5. N(p, p; 2) 2
p/2
.
From (3.5) and Theorem 3.5, we know that

2
p
_
N(p, p; 2) 4 (p 2).
It would be very nice if one could show that this p-th root has a
limit for p .
To give a considerable improvement of Theorem 3.5, we discuss a
probabilistic method that is useful in many parts of combinatorics.
We consider events A
1
, A
2
, . . . , A
n
in a probability space. Denote
by Pr[A
i
] the probability of the event A
i
and, as usual, let A
i
denote the complement of A
i
, i.e. non-occurrence of A
i
. We are
interested in applications where the A
i
denote situations we do not
wish to occur and where we would like to assert that there is a
3. Colorings of graphs and Ramseys theorem 31
positive probability that none of the events A
i
occurs. In some
easy counting situations one can use
n

i=1
Pr[A
i
] < 1 A
i
,= .
Cf. Problem 5E. However, in general, dependence among the unde-
sirable events leads to multiple counting which results in a sum that
is much larger than 1. Of course, if the events A
i
are independent,
then it suces that each has probability less than one to guarantee
that non-occurrence of all of them has positive probability. The
Lovasz Sieve handles situations where there is indeed some depen-
dence but simultaneously there are many obviously independent
combinations of events.
We dene what we shall call a dependency graph for the events
A
1
, . . . , A
n
. This is a graph G on the set of indices 1, 2, . . . , n
with the property that for every i the event A
i
is independent of
every subset of A
j
: i, j / E(G). Note that we require a lot
more than that A
i
is independent of each of the A
j
in this subset.
Theorem 3.6. Let G be some dependency graph for the events
A
1
, . . . , A
n
. Suppose that Pr[A
i
] p, i = 1, . . . , n and that every
vertex in G has degree d. If 4dp < 1, then A
i
,= .
Proof: We rst show that for every subset i
1
, i
2
, . . . , i
m
of the
index set,
(3.6) Pr[A
i
1
[A
i
2
. . . A
i
m
]
1
2d
.
The case m = 1 is trivial and for m = 2 we have
Pr[A
1
[A
2
]
p
1
1 p
2

1
4d 1
<
1
2d
,
where for convenience of notation we have taken i
j
= j and p
i
:=
Pr[A
i
]. We proceed by induction.
Suppose that in G, 1 is adjacent to 2, 3, . . . , q and not adjacent
to q + 1, . . . , m. We have
Pr[A
1
[A
2
. . . A
m
] =
Pr[A
1
A
2
. . . A
q
[A
q+1
. . . A
m
]
Pr[A
2
. . . A
q
[A
q+1
. . . A
m
]
.
32 A Course in Combinatorics
The numerator is (by denition of G) at most
Pr[A
1
[A
q+1
. . . A
m
] = Pr[A
1
]
1
4d
.
Using the induction hypothesis, we nd that the denominator is at
least
1
q

i=2
Pr[A
1
[A
q+1
. . . A
m
] 1
q 1
2d

1
2
.
This proves (3.6). We now have
Pr[A
1
. . . A
n
] =
n

i=1
Pr[A
i
[A
1
. . . A
i1
] (1
1
2d
)
n
> 0,
where we have used (3.6) for each term in the product.
We apply this method to obtain a lower bound for N(p, p; 2).
Theorem 3.7. N(p, p; 2) c p 2
p/2
, where c is a constant.
Proof: Consider K
n
and color the edges randomly with two col-
ors. For each set S of k vertices let A
S
be the event that the
subgraph on S is colored monochromatically. We wish to assert
that among the random colorings, there is at least one in which no
monochromatic subgraph on k vertices occurs. We dene a depen-
dency graph by making S and T adjacent if and only if [S T[ 2,
i.e. the subgraphs on S and T have an edge in common. The degree
d of G is clearly at most
_
k
2
__
n
k2
_
. The events A
S
all have prob-
ability 2
1
(
k
2
)
. From Theorem 3.6, Stirlings formula and a little
manipulation, we nd the result (and if we wish an estimate for c).

We have given some examples of an area of combinatorics known

as Ramsey theory. We just mention one more example, namely
a theorem due to B. L. van der Waerden (1927). It states that
there exists a number N(r) such that if N N(r) and the integers
from 1 to N are colored red or blue, then there is a monochromatic
arithmetic progression of length r in the set. For a short (but not
easy) proof see Graham and Rothschild (1974). A general reference
3. Colorings of graphs and Ramseys theorem 33
for this area is the book Ramsey Theory by R. L. Graham, B. L.
Rothschild and J. L. Spencer (1980).
An interesting application of Ramseys theorem is the following
theorem due to Erd os and Szekeres (1935).
Theorem 3.8. For a given n, there is an integer N(n) such that
any collection of N N(n) points in the plane, no three on a line,
has a subset of n points forming a convex n-gon.
Proof: (i) First we observe that if we have n points, no three on a
line, then they form a convex n-gon if and only if every quadrilateral
formed by taking four of the points is convex.
(ii) We now claim that N(n) = N(n, n; 3) will do the job. Let
S be a set of N(n) points. Number the points and then color tri-
angles red, respectively blue, if the path from the smallest number
via the middle one to the largest number is clockwise, respectively
counterclockwise. There is an n-subset with all its triangles the
same color, say red. We shall show that this set cannot contain the
conguration of Fig. 3.2.
Figure 3.2
Without loss of generality a < b < c. From triangle adc, we see
that a < d < c. Then from triangle abd it follows that a < b < d.
But then triangle bcd is blue, a contradiction. So all quadrilaterals
formed from the n-subset are convex and by (i), we are done.
Problem 3E. A tournament on n vertices is an orientation of K
n
.
A transitive tournament is a tournament for which the vertices can
be numbered in such a way that (i, j) is an edge if and only if i < j.
(a) Show that if k log
2
n, every tournament on n vertices has
a transitive subtournament on k vertices.
(b) Show that if k > 1 +2 log
2
n, there exists a tournament on n
vertices with no transitive subtournament on k vertices.
34 A Course in Combinatorics
Problem 3F. Prove that for all r N there is a minimal number
N(r) with the following property. If n N(r) and the integers in
1, 2, . . . , n are colored with r colors, then there are three elements
x, y, z (not necessarily distinct) with the same color and x +y = z.
(A result due to I. Schur.) Determine N(2). Show by an elementary
argument that N(3) > 13.
Problem 3G. Let m be given. Show that if n is large enough,
every n n (0, 1)-matrix has a principal submatrix of size m, in
which all the elements below the diagonal are the same, and all the
elements above the diagonal are the same.
Problem 3H. Show that if the edges of K
17
are colored with three
colors, there must be a monochromatic triangle.
Problem 3I. Let 1, ,
2
, . . . ,
14
be the multiplicative group
of F
16
. Number the vertices of K
16
with the elements of F
16
. We
will color the edges of K
16
with three colors. Edge i, j is to be
colored with a color that depends only on , where i j =

. Do
this in such a way that there is no monochromatic triangle. (In the
notation of Theorem 3.3, this problem and the previous one show
that N(3, 3, 3; 2) = 17.)
Problem 3J. The edges of K
n
are colored red and blue in such a
way that a red edge is in at most one red triangle. Show that there
is a subgraph K
k
with k

2n| that contains no red triangle.

Problem 3K. Let G satisfy the conditions of Theorem 3.1. Show
that by removing at most n/d edges we can nd a subgraph G
/
with chromatic number d 1.
Notes.
The four color conjecture was considered one of the most famous
open problems in combinatorics until 1976. Its solution by Appel
and Haken has caused much debate because the proof depends on
an extensive computer analysis of many cases. The validity of the
argument depends on ones trust in computers and programs. (See
Chapter 34 for two proofs of the Five Color Theorem.)
The theorem by Brooks, Theorem 3.1, which he discovered while
an undergraduate at Cambridge, is a typical example of the inge-
3. Colorings of graphs and Ramseys theorem 35
nious arguments that are quite often necessary in that part of graph
theory where algebraic methods do not apply.
F. P. Ramsey (19021928) died too young to produce the results
that he probably would have. He was interested in decision pro-
cedures for logical systems and, strangely enough, this led to the
theorem that in turn led to so-called Ramsey theory.
Theorem 3.5 is due to P. Erd os (1947). For more on probabilistic
methods, see Erdos and Spencer (1974).
Values of and estimates for the numbers N(p, q; 2) can be found
in Radziszowski (1999). The value of N(3, 9; 2) is from Grinstead
and Roberts (1982).
The proof of Theorem 3.6 is not the original proof by Erd os and
Szekeres that one usually nds in books. This proof was produced
by a student in Haifa (M. Tarsy) during an examination! He had
missed the class in which the proof had been presented. See Lewin
(1976). A proof in a similar vein was given by Johnson (1986).
References.
K. Appel and W. Haken (1977), Every planar map is four-color-
able, Illinois J. Math. 21, 429567.
R. L. Brooks (1941), On colouring the nodes of a network, Cam-
bridge Philos. Soc. 37, 194197.
P. Erd os (1947), Some remarks on the theory of graphs, Bull. Amer.
Math. Soc. 53, 292294.
P. Erd os and J. L. Spencer (1974), Probabilistic Methods in Com-
binatorics, Academic Press.
P. Erd os and G. Szekeres (1935), A combinatorial problem in ge-
ometry, Compositio Math. 2, 463470.
R. L. Graham and B. L. Rothschild (1974), A short proof of van
der Waerdens theorem on arithmetic progressions, Proc. Amer.
Math. Soc. 42, 385386.
R. L. Graham, B. L. Rothschild, and J. L. Spencer (1980), Ramsey
Theory, Wiley.
C. M. Grinstead and S. M. Roberts (1982), On the Ramsey numbers
R(3, 8) and R(3, 9), J. Combinatorial Theory (B) 33, 2751.
S. Johnson (1986), A new proof of the Erd osSzekeres convex k-gon
result, J. Combinatorial Theory (A) 42, 318319.
36 A Course in Combinatorics
M. Lewin (1976), A new proof of a theorem of Erd os and Szekeres,
The Math. Gazette 60, 136138, 298.
S. P. Radziszowski (1999), Small Ramsey Numbers, The Electronic
Journal of Combinatorics 1 DS 1.
F. P. Ramsey (1930), On a problem of formal logic, Proc. London
Math. Soc. (2) 30, 264286.
B. L. van der Waerden (1927), Beweis einer Baudetschen Vermu-
tung, Nieuw Archief voor Wiskunde 15, 212216.
4
Turans theorem and
extremal graphs
As an introduction, we rst ask the question how many edges a
simple graph must have to guarantee that the graph contains a
triangle. Since K
m,m
and K
m,m+1
do not contain triangles, we see
that if the graph has n vertices, then n
2
/4| edges are not enough.
We claim that if there are more edges, then the graph contains a
triangle (W. Mantel, 1907). The following proof is surprising. Let
G have n vertices, numbered from 1 to n, and no triangles. We
give vertex i a weight z
i
0 such that

z
i
= 1 and we wish
to maximize S :=

z
i
z
j
, where the sum is taken over all edges
i, j. Suppose that vertex k and vertex l are not joined. Let the
neighbors of k have total weight x, and those of l total weight y,
where x y. Since (z
k
+ )x + (z
l
)y z
k
x + z
l
y, we do not
decrease the value of S if we shift some of the weight of vertex l
to the vertex k. It follows that S is maximal if all of the weight
is concentrated on some complete subgraph of G, i.e. on one edge!
Therefore S
1
4
. On the other hand, taking all z
i
equal to n
1
would yield a value of n
2
[E[ for S. Therefore [E[
1
4
n
2
.
Note that Ramseys theorem states that if a graph on n vertices
has n N(p, q; 2), then the graph either has a complete subgraph
on p vertices or a set of q vertices with no edges between them
(called an independent set). We now ask the question whether some
condition on the number of edges guarantees a K
p
as a subgraph.
We saw above what the answer is for p = 3. We also already have
an idea of how to avoid a K
p
. Divide the vertices into p1 subsets
S
1
, . . . , S
p1
of almost equal size, i.e. r subsets of size t + 1 and
p 1 r subsets of size t, where n = t(p 1) + r, 1 r p 1.
Within each S
i
there are no edges but every vertex in S
i
is joined to
38 A Course in Combinatorics
every vertex in S
j
if i ,= j. (This is a complete multipartite graph.)
The number of edges is
M(n, p) :=
p 2
2(p 1)
n
2

r(p 1 r)
2(p 1)
.
Theorem 4.1. (Turan, 1941) If a simple graph on n vertices has
more than M(n, p) edges, then it contains a K
p
as a subgraph.
Proof: The proof is by induction on t. If t = 0, the theorem
is obvious. Consider a graph G with n vertices, no K
p
, and the
maximum number of edges subject to those properties. Clearly G
contains a K
p1
(otherwise adding an edge would not produce a
K
p
), say H. Each of the remaining vertices is joined to at most
p2 vertices of H. The remaining np+1 vertices do not contain
a K
p
as subgraph. Since np+1 = (t 1)(p1) +r, we can apply
the induction hypothesis to this set of points. So the number of
edges of G is at most
M(n p + 1, p) + (n p + 1)(p 2) +
_
p 1
2
_
and this number is equal to M(n, p).
Remark. The argument that was used for Mantels theorem would
show that if there is no K
p
, then [E[
p2
2(p1)
n
2
.
Problem 4A. Let G be a simple graph with 10 vertices and 26
edges. Show that G has at least 5 triangles. Can equality occur?
Tur ans paper on graph theory that contains Theorem 4.1 is con-
sidered the starting point of what is now known as extremal graph
theorysee Bollobas (1978). A simple instance of an extremal
problem will ask for the maximum number of edges a graph with a
certain property may have. The graphs whose number of edges is
maximum are called the extremal graphs with respect to the prop-
erty.
The extremal graphs for Tur ans problem are only the complete
multipartite graphs described above. This follows from an analysis
of the proof of Theorem 4.1; we ask the reader to do this at least
in the case p = 3 in the problem below.
4. Tur ans theorem 39
Problem 4B. Show that a simple graph on n vertices with n
2
/4|
edges and no triangles is a complete bipartite graph K
k,k
if n = 2k,
or K
k,k+1
if n = 2k + 1.
Problem 4C. If a simple graph on n vertices has e edges, then it
has at least
e
3n
(4e n
2
) triangles.
The girth of a graph G is the size of a smallest polygon P
n
in G.
(A forest has innite girth.) By denition, a graph is simple if and
only if it has girth 3. By Mantels theorem, a graph with more
than n
2
/4 edges has girth 3.
Theorem 4.2. If a graph G on n vertices has more than
1
2
n

n 1
edges, then G has girth 4. That is, G is not simple or contains a
P
3
or a P
4
(a triangle or a quadrilateral).
Proof: Suppose G has girth 5. Let y
1
, y
2
, . . . , y
d
be the vertices
adjacent to a vertex x, where d := deg(x). No two of these are
adjacent since G has no triangles. Moreover, no vertex (other than
x) can be adjacent to more than one of y
1
, . . . , y
d
since there are no
quadrilaterals in G. Thus (deg(y
1
)1)+ +(deg(y
d
)1)+(d+1)
cannot exceed the total number n of vertices. That is,

y adjacent to x
deg(y) n 1.
Then
n(n 1)

y adjacent to x
deg(y) =

y
deg(y)
2

1
n
_

y
deg(y)
_
2
=
1
n
(2[E(G)[)
2
.

The number
1
2
n

n 1 in Theorem 4.2 is only a boundit is not

the exact answer for all n. Determination of the extremal graphs
for this problem (maximum number of edges subject to girth 5)
for all values of n is impossibly dicult; a determination of the
graphs for which equality holds has, however, been almost possible.
40 A Course in Combinatorics
Perhaps surprisingly, there are at most four graphs with n > 2
vertices, girth 5, and
1
2
n

n 1 edges: The pentagon (n = 5),

the Petersen graph (n = 10), one with n = 50, and possibly one
with n = 3250. See the notes and Chapter 21.
Problem 4D. Suppose G is regular of degree r and has girth g
or greater. Find a lower bound for [V (G)[. (Consider the cases g
even and odd separately.)
It is not so interesting to ask how many edges are required to
force a Hamiltonian circuit. But we can ask what bound on the
minimum degree will do the job.
Theorem 4.3. If a simple graph G on n vertices has all vertices
of degree at least n/2, then it contains a P
n
as a subgraph, i.e. it
has a Hamiltonian circuit.
Proof: Suppose the theorem is not true and let G be a graph
satisfying the hypothesis for some n but having no Hamiltonian
circuits. We may take G to be such a counterexample with the
maximum number of edges; then the addition of any edge to G (i.e.
joining two nonadjacent vertices by an edge) creates a Hamiltonian
circuit.
Let y and z be nonadjacent vertices. Since adding y, z creates
a Hamiltonian circuit, there exists a simple path from y to z with
vertex terms, y = x
1
, x
2
, . . . , x
n
= z, say. The sets
i : y is adjacent to x
i+1

and
i : z is adjacent to x
i

each have cardinality n/2 and are contained in 1, 2, 3, . . . , n1,

so they must meet; let i
0
belong to both. Then
y = x
1
, x
2
, . . . , x
i
0
, z = x
n
, x
n1
, . . . , x
i
0
+1
, x
1
= y
is the vertex sequence of a simple closed path of length n in G,
contradicting our choice of G as a counterexample.
Theorem 4.3 is due to G. A. Dirac and is best possible at least
in the sense that it does not remain true if we replace n/2 by
4. Tur ans theorem 41
(n 1)/2. For example, the complete bipartite graphs K
k,k+1
have
no Hamiltonian circuits. But it does admit improvements and
generalizationssee e.g. Lovasz (1979), Problem 10.21.
Problem 4E. A 333 cube of cheese is divided into 27 111
small cubes. A mouse eats one small cube each day and an adjacent
small cube (sharing a face) the next day. Can the mouse eat the
center small cube on the last day?
Problem 4F. Let G be a simple graph with n vertices. Prove: If
each vertex of G has degree (n+1)/2, then for any edge e, there
exists a Hamiltonian circuit of G that passes through e.
Problem 4G. Prove the remark following Theorem 4.1.
Problem 4H. Show that a graph on n vertices that does not
contain a circuit on four vertices has at most
n
4
(1+

4n 3) edges.
Notes.
P. Tur an (19101976), one of the famous Hungarian mathemati-
cians of the 20th century, is best known for his work in analytic
number theory and real and complex analysis.
For every r 2 and g 2, there exists a graph that is regular of
degree r and has girth g. See Lovasz (1979), Problem 10.12.
Analysis of the proof of Theorem 4.2 shows that a graph with
n > 2 vertices, girth 5, and
1
2
n

n 1 edges is regular of degree

k :=

n 1 and also that any two vertices are joined by a (unique)

path of length 2 if they are not adjacent. With the notation of
Chapter 21, such a graph is an srg(n, k, 0, 1) and the methods of
that chapter show that k = 2, 3, 7, or 57. This was rst shown
in Homan and Singleton (1960) where in addition an example
with k = 7 and n = 50 was described (that is now known as the
Homan-Singleton graph). It is not known at this time whether
there exists an srg(3250, 57, 0, 1).
References.
B. Bollobas (1978), Extremal Graph Theory, Academic Press.
A. J. Homan and R. R. Singleton (1960), On Moore graphs with
diameters two and three, IBM J. Res. Develop. 4, 497504.
42 A Course in Combinatorics
L. Lovasz (1979), Combinatorial Problems and Exercises, North
Holland.
W. Mantel (1907), Problem 28, Wiskundige Opgaven 10, 6061.
P. Tur an (1941), An extremal problem in graph theory (in Hungar-
ian), Mat. Fiz. Lapok 48, 435452.
5
Systems of distinct
representatives
We rst give two dierent formulations of a theorem known as
P. Halls marriage theorem. We give a constructive proof and an
enumerative one. If A is a subset of the vertices of a graph, then
denote by (A) the set

aA
(a). Consider a bipartite graph G
with vertex set X Y (every edge has one endpoint in X and one
in Y ). A matching in G is a subset E
1
of the edge set such that
no vertex is incident with more than one edge in E
1
. A complete
matching from X to Y is a matching such that every vertex in X is
incident with an edge in E
1
. If the vertices of X and Y are thought
of as boys and girls, respectively, or vice versa, and an edge is
present when the persons corresponding to its ends have amicable
feelings towards one another, then a complete matching represents
a possible assignment of marriage partners to the persons in X.
Theorem 5.1. A necessary and sucient condition for there to
be a complete matching from X to Y in G is that [(A)[ [A[ for
every A X.
Proof: (i) It is obvious that the condition is necessary.
(ii) Assume that [(A)[ [A[ for every A X. Let [X[ = n,
m < n, and suppose we have a matching M with m edges. We shall
show that a larger matching exists. (We mean larger in cardinality;
we may not be able to nd a complete matching containing these
particular m edges.)
Call the edges of M red and all other edges blue. Let x
0
X
be a vertex not incident with an edge of the matching. We claim
that there exists a simple path (of odd length) starting with x
0
and
a blue edge, using red and blue edges alternately, and terminating
44 A Course in Combinatorics
with a blue edge and a vertex y not incident with an edge of the
matching. If we nd such a path p, we are done because we obtain
a matching with m + 1 edges by deleting the red edges of p from
M and replacing them with the blue edges of p. In other words,
we switch the colors on the edges of p.
Since [(x
0
)[ 1, there is a vertex y
1
adjacent to x
0
(obviously
by a blue edge since x
0
is not incident with any red edges). If y
1
is also not incident with any red edges, we have the required path
(of length one); if y
1
is incident with a red edge, let x
1
be the other
end of that red edge. Recursively, dene x
0
, x
1
, . . . and y
1
, y
2
, . . . as
follows. If x
0
, x
1
, . . . , x
k
and y
1
, . . . , y
k
have been dened, then since
[(x
0
, x
1
, . . . , x
k
)[ k+1, there exists a vertex y
k+1
, distinct from
y
1
, . . . , y
k
, that is adjacent to at least one vertex in x
0
, x
1
, . . . , x
k
.
If y
k+1
is not incident with a red edge, stop; otherwise, let x
k+1
be
the other end of that red edge.
When the procedure terminates, we construct the path p by start-
ing with y
k+1
and the blue edge joining it to, say, x
i
1
, i
1
< k + 1.
Then add the red edge x
i
1
, y
i
1
. By construction, y
i
1
is joined by
an edge (necessarily blue) to some x
i
2
, i
2
< i
1
. Then add the red
edge x
i
2
, y
i
2
. Continue in this way until x
0
is reached.
Problem 5A. A perfect matching in a graph G (not necessarily
bipartite) is a matching so that each vertex of G is incident with
one edge of the matching. (i) Show that a nite regular bipartite
graph (regular of degree d > 0) has a perfect matching. (ii) Find a
trivalent (regular of degree 3) simple graph which does not have a
perfect matching. (iii) Suppose G is bipartite with vertices X Y
(every edge having one end in X and one in Y ). Further assume
that every vertex in X has the same degree s > 0 and every vertex
in Y has the same degree t. (This condition is called semiregular-
ity.) Prove: If [X[ [Y [ (equivalently, if s t), then there is a
complete matching M of X into Y .
Example 5.1. A parlor trick involving a standard deck of 52 cards
is as follows. You are dealt ve cards at random. You keep one and
put the other four (in a specic order) into an envelope which is
taken to your partner in another room. Your partner looks at these
and announces the name of the fth card, that you had retained.
Using the values of suits and ranks, it is possible to think of clever
5. Systems of distinct representatives 45
or relatively simple ways to determine which card to keep and which
to pass, and for the partner to determine the retained card; see
the notes. Ignoring complexity of solution, however, the relevant
general mathematical problem, stated for a deck of N cards, is
does there exist an injective mapping f from the set X of the
_
N
5
_
5-element subsets of the N cards into the set Y of N(N 1)(N
2)(N3) ordered 4-tuples of distinct cards, subject to the condition
that if f(S) = (c
1
, c
2
, c
3
, c
4
), then c
1
, c
2
, c
3
, c
4
S.
In terms of matchings, we consider the bipartite graph G whose
vertices are X Y as dened above, and where there is an edge
joining S X to c
1
, c
2
, c
3
, c
4
Y exactly when c
1
, c
2
, c
3
, c
4

S. We require a complete matching M of X into Y . The reader
may check that G is semiregular and that [X[ [Y [ if and only if
N 124. So by Problem 5A(iii), for N 124, there exists such a
matching.
We now reformulate Theorem 5.1 in terms of sets and not only
prove the theorem but also give a lower bound for the number of
matchings. We consider subsets A
0
, A
1
, . . . , A
n1
of a nite set S.
We shall say that this collection has property H (Halls condition)
if (for all k) the union of any k-tuple of subsets A
i
has at least k
elements. If the union of some k-tuple of subsets contains exactly
k elements (0 < k < n), then we call this k-tuple a critical block.
We dene a system of distinct representatives (SDR) of the sets
A
0
, . . . , A
n1
to be a sequence of n distinct elements a
0
, . . . , a
n1
with a
i
A
i
, 0 i n 1.
Let m
0
m
1
m
n1
. We dene
F
n
(m
0
, m
1
, . . . , m
n1
) :=
n1

i=0
(m
i
i)

,
where (a)

:= max1, a.
From now on, we assume that the sequence m
i
:= [A
i
[ is nonde-
creasing.
For the proof of the main theorem, we need a lemma.
Lemma 5.2. For n 1, let f
n
: Z
n
N be dened by
f
n
(a
0
, a
1
, . . . , a
n1
) := F
n
(m
0
, m
1
, . . . , m
n1
)
46 A Course in Combinatorics
if (m
0
, . . . , m
n1
) is a nondecreasing rearrangement of the n-tuple
(a
0
, . . . , a
n1
). Then f
n
is nondecreasing with respect to each of the
variables a
i
.
Proof: Let
m
0
m
k1
a
i
= m
k
m
k+1

m
l
m
l+1
m
n1
be a nondecreasing rearrangement of (a
0
, . . . , a
n1
). If a
/
i
a
i
and
m
0
m
k1
m
k+1
m
l
a
/
i
m
l+1
m
n1
is a nondecreasing rearrangement of (a
0
,. . ., a
i1
, a
/
i
, a
i+1
,. . ., a
n1
)
then
f
n
(a
0
, . . . , a
i1
, a
/
i
, a
i+1
, . . . , a
n1
)
f
n
(a
0
, . . . , a
n1
)
=
=
(m
k+1
k)

(a
i
k)

(a
/
i
l)

(m
l
l)

l1

j=k+1
(m
j+1
j)

(m
j
j)

and this is 1 since a

i
m
k+1
, a
/
i
m
l
, and m
j+1
m
j
for
j = k + 1, . . . , l 1.
We now come to the second form of Halls theorem. We denote
by N(A
0
, . . . , A
n1
) the number of SDRs of (A
0
, . . . , A
n1
).
Theorem 5.3. Let (A
0
, . . . , A
n1
) be a sequence of subsets of a set
S. Let m
i
:= [A
i
[ (i = 0, . . . , n1) and let m
0
m
1
m
n1
.
If the sequence has property H, then
N(A
0
, . . . , A
n1
) F
n
(m
0
, . . . , m
n1
).
Proof: The proof is by induction. Clearly the theorem is true for
n = 1. We distinguish two cases.
Case 1. There is no critical block. In this case, we choose any
element a of A
0
as its representative and then remove a from all the
other sets. This yields sets, that we call A
1
(a), . . . , A
n1
(a), and
5. Systems of distinct representatives 47
for these sets property H still holds. By the induction hypothesis
and by the lemma, we nd
N(A
0
, . . . , A
n1
)

aA
0
f
n1
([A
1
(a)[, . . . , [A
n1
(a)[)

aA
0
f
n1
(m
1
1, . . . , m
n1
1)
= m
0
f
n1
(m
1
1, . . . , m
n1
1)
= F
n
(m
0
, m
1
, . . . , m
n1
).
Case 2. There is a critical block (A

0
, . . . , A

k1
) with
0
< <

k1
and 0 < k < n. In this case, we delete all elements of A

o

A

k1
from all the other sets A
i
which produces A
/

0
, . . . , A
/

l1
,
where
0
, . . . ,
k1
,
0
, . . . ,
l1
= 0, 1, . . . , n 1, k +l = n.
Now both (A

0
, . . . , A

k1
) and (A
/

0
, . . . , A
/

l1
) satisfy property
H and SDRs of the two sequences are always disjoint. Hence by
the induction hypothesis and the lemma, we have
N(A
0
, . . . , A
n1
) = N(A

0
, . . . , A

k1
)N(A
/

0
, . . . , A
/

l1
)
(5.1)
f
k
(m

0
, . . . , m

k1
)f
l
([A
/

0
[, . . . , [A
/

l1
[)
f
k
(m

0
, . . . , m

k1
)f
l
(m

0
k, . . . , m

l1
k)
f
k
(m
0
, . . . , m
k1
)f
l
(m

0
k, . . . , m

l1
k).
Now we remark that
m

k1
[A

0
A

k1
[ = k,
and therefore we have
(m
r
r)

= 1 if k r
k1
,
and
(m

i
k i)

= 1 if
i

k1
.
This implies that
f
k
(m
0
, . . . , m
k1
) =

0i
k1
(m
i
i)

,
f
l
(m

0
k, . . . , m

l1
k) =

k1
<j<n
(m
j
j)

,
48 A Course in Combinatorics
i.e. the product (5.1) is equal to F
n
(m
0
, . . . , m
n1
), which proves
the theorem.
Problem 5B. Show that Theorem 5.3 gives the best lower bound
for the number of SDRs of the sets A
i
that only involves the num-
bers [A
i
[.
We now come to a theorem known as Konigs theorem. It is
equivalent (whatever that means) to Halls theorem. In the theo-
rem, A is a (0,1)-matrix with entries a
ij
. By a line, we mean a row
or a column of A.
Theorem 5.4. The minimum number of lines of A that contain
all the 1s of A is equal to the maximum number of 1s in A, no
two on a line.
Proof: Let m be the minimum number of lines of A containing all
the 1s of A and let M be the maximum number of 1s, no two on
a line. Clearly m M. Let the minimum covering by lines consist
of r rows and s columns (r + s = m). Without loss of generality,
these are the rst r rows and the rst s columns. We now dene
sets A
i
, 1 i r, by A
i
:= j > s : a
ij
= 1. If some k-tuple of
the A
i
s contained less than k elements, then we could replace the
corresponding k rows by k 1 columns, still covering all the 1s.
Since this is impossible, we see that the A
i
s satisfy property H. So
the A
i
s have an SDR. This means that there are r 1s, no two on a
line, in the rst r rows and not in the rst s columns. By the same
argument there are s 1s, no two on a line, in the rst s columns
and not in the rst r rows. This shows that M r + s = m and
we are done.
The following theorem of G. Birkho is an application of Halls
theorem.
Theorem 5.5. Let A = (a
ij
) be an n n matrix with nonnegative
integers as entries, such that every row and column of A has sum
l. Then A is the sum of l permutation matrices.
Proof: Dene A
i
, 1 i n, by A
i
:= j : a
ij
> 0. For any
k-tuple of the A
i
s, the sum of the corresponding rows of A is kl.
Since every column of A has sum l, the nonzero entries in the
5. Systems of distinct representatives 49
chosen k rows must be in at least k columns. Hence the A
i
s satisfy
property H. An SDR of the A
i
s corresponds to a permutation
matrix P = (p
ij
) such that a
ij
> 0 if p
ij
= 1. The theorem now
follows by induction on l.
Problem 5C. In the hypothesis of Theorem 5.5, we replace inte-
gers by reals. Show that in this case, A is a nonnegative linear
combination of permutation matrices. (Equivalently, every dou-
bly stochastic matrixsee Chapter 11is a convex combination of
permutation matrices.)
Problem 5D. Let S be the set 1, 2, . . . , mn. We partition S into
m sets A
1
, . . . , A
m
of size n. Let a second partitioning into m sets
of size n be B
1
, . . . , B
m
. Show that the sets A
i
can be renumbered
in such a way that A
i
B
i
,= .
Problem 5E. Let A
i
= i1, i, i+11, 2, . . . , n, i = 1, 2, . . . , n.
Let S
n
denote the number of SDRs of the collection A
1
, . . . , A
n
.
Determine S
n
and lim
n
S
1/n
n
.
Let G be a bipartite graph, nite or innite; say the vertex set
is partitioned into sets X, Y of vertices so that every edge of G has
one end in X and one end in Y . We say that a matching M in G
covers a subset S of the vertices when every vertex in S is incident
with one of the edges in M.
Theorem 5.6. If there exists a matching M
1
that covers a subset
X
0
of X and there exists a matching M
2
that covers a subset Y
0
of
Y , then there exists a matching M
3
that covers X
0
Y
0
.
Proof: Think of the edges of M
1
as red edges and the edges of
M
2
as blue edges. If an edge belongs to both M
1
and M
2
, it is
purple.
A connected graph all of whose vertices have degree at most
two is easily seen to be one of: a nite path-graph (allowing the
trivial case of length 0, when the component has one vertex and
no edges), a nite polygon, an innte one-sided path-graph (with
one monovalent vertex), or an innite two-sided path-graph. The
graph H whose vertices are those of G and whose edges M
1
M
2
has the property that every vertex has degree at most two, so its
connected components are of the types enumerated above. The
50 A Course in Combinatorics
edges of any of these components, other than the graphs consisting
of a purple edge and its two ends, are alternately colored red and
blue; in particular, all polygons have even length. Every vertex of
X
0
Y
0
is in one of these nontrivial components.
For the matching M
3
, we will take all purple edges, and either
all red edges or all blue edges from every other component of H.
From the cycles and innite two-sided paths, it doesnt matter;
take all red or all blue edges, and they will cover all vertices of the
component. From the paths of odd length and innite one-sided
paths, take all red or blue edges depending on whether the rst
edge is red or blue (for a path of odd length, the rst and last edge
have the same color, so it doesnt matter what side you start on).
Again, the selected edges will cover all vertices of the component.
We have to think just a tiny bit harder for a component that is a
nite path P of even length, with vertices v
0
, v
1
, . . . , v
k
, say. There
is an odd number of vertices and they alternate between X and Y ,
so v
0
and v
k
are both in X or both in Y . If they are both in X,
take all red edges of P (those of M
1
) and put them in M
3
; if they
are both in Y , take all blue edges of P and put them in M
3
. Only
one end of the path is not covered by the chosen edges.
Consider the case that both v
0
, v
k
X (the case when they are
in Y is completely analogous). If the rst edge of P is red, then the
last is blue and it follows that v
k
/ X
0
since no edge of M
1
covers
v
k
. Thus the red edges of P still cover all vertices of X
0
and Y
0
that
were in P. Similarly, if the rst edge of P is blue, then v
0
/ X
0
and
the red edges of P still cover all vertices of X
0
and Y
0
that were in
P.
For the case X
0
= X, Y
0
= Y , and when the graph G is complete
bipartite, matchings that cover X
0
or Y
0
correspond to, or can be
interpreted as, injective mappings X Y or Y X, respectively.
Theorem 5.6 says:
Corollary. If X and Y are sets and there exist injective map-
pings f : X Y and g : Y X, then there exists a bijective
mapping from X to Y , or from Y to X, i.e. there is a one-to-one
correspondence between the two sets.
In terms of cardinality of sets, this says that if [X[ [Y [ and
5. Systems of distinct representatives 51
[Y [ [X[, then [X[ = [Y [. This is the Schr oder-Bernstein Theo-
rem; see Section 22 of P. R. Halmos (1974). It is trivial for nite
sets, of course.
Problem 5F. Let A
1
, A
2
, . . . , A
n
be nite sets. Show that if

1i<jn
[A
i
A
j
[
[A
i
[ [A
j
[
< 1,
then the sets A
1
, A
2
, . . . A
n
have a system of distinct representatives.
Problem 5G. (i) From Problem 5A we know that a bipartite
graph on 2n vertices that is regular of degree 3 has a perfect match-
ing. How many dierent perfect matchings are there if n = 4? (ii)
The same question for a bipartite graph on 10 vertices that is reg-
ular of degree 4.
Notes.
Philip Hall published his result in 1935 (with a rather dicult
proof). The proof that we gave is a generalization of ideas of Hal-
mos and Vaughan, Rado, and M. Hall. The proof is due to Ostrand
(1970) and Hautus and Van Lint (1972). See Van Lint (1974). The
problem of complete matchings is often referred to as the marriage
problem.
D. K onig (18841944) was professor at Budapest. He wrote the
rst comprehensive treatise on graph theory (Theorie der endlichen
und unendlichen Graphen, 1936). K onig (1916) contains the rst
proof of (one of the theorems called) K onigs theorem.
Just before Theorem 5.4, we referred to the equivalenceof these
theorems. This expression is often used when each of two theorems
is more or less an immediate consequence of the other.
The theorem by Birkho (1946), i.e. Theorem 5.5, is extremely
useful and will be applied a number of times in later chapters.
For the card problem mentioned in Example 5.1, here is one so-
lution. In a set of ve cards, some suit must be represented at
least twice. The rst card you pass to your partner should be
the larger of two cards of the same suit, and you will retain the
smaller, where we think of the ranks 2, 3, . . . , 10, J, Q, K, A as ar-
ranged clockwise on a circle (modulo 13), and by smaller we mean
52 A Course in Combinatorics
the card from which we must travel the least distance clockwise
to get to the other. For example, if S = 3, Q, 6, 3, 7,
you pass either the 7 or 3 to your partner. This already tells
your partner the suit of the retained card and limits that card to
six possibilities in that suit. To determine how far to count back
(counterclockwise) from the rank of the rst card, you and your
partner use the order of the remaining three cards (the 52 cards
are ordered lexicographically), and agree on some correspondence
between the six permutations of three objects and the integers from
1 to 6.
References.
G. Birkho (1946), Tres observaciones sobre el algebra lineal, Univ.
Nac. Tucum an, Rev. Ser. A, 5, 147151.
P. Hall (1935), On representatives of subsets, J. London Math. Soc.
10, 2630.
P. R. Halmos (1974), Naive Set Theory, Springer-Verlag.
D. K onig (1916),

Uber Graphen und ihre Anwendung auf Determi-
nantentheorie und Mengenlehre, Math. Annalen 77, 453465.
J. H. van Lint (1974), Combinatorial Theory Seminar Eindhoven
University of Technology, Lecture Notes in Mathematics 382,
Springer-Verlag.
P. Ostrand (1970), Systems of distinct representatives, J. of Math.
Analysis and Applic. 32, 14.
6
Dilworths theorem and
extremal set theory
A partially ordered set (also poset) is a set S with a binary relation
(sometimes is used) such that:
(i) a a for all a S (reexivity),
(ii) if a b and b c then a c (transitivity),
(iii) if a b and b a then a = b (antisymmetry).
If for any a and b in S, either a b or b a, then the partial
order is called a total order, or a linear order. If a b and a ,= b,
then we also write a < b. Examples of posets include the integers
with the usual order or the subsets of a set, ordered by inclusion.
If a subset of S is totally ordered, it is called a chain. An antichain
is a set of elements that are pairwise incomparable.
The following theorem is due to R. Dilworth (1950). This proof
is due to H. Tverberg (1967).
Theorem 6.1. Let P be a partially ordered nite set. The min-
imum number m of disjoint chains which together contain all ele-
ments of P is equal to the maximum number M of elements in an
antichain of P.
Proof: (i) It is trivial that m M.
(ii) We use induction on [P[. If [P[ = 0, there is nothing to
prove. Let C be a maximal chain in P. If every antichain in PC
contains at most M 1 elements, we are done. So assume that
a
1
, . . . , a
M
is an antichain in PC. Now dene S

:= x P :

i
[x a
i
], and dene S
+
analogously. Since C is a maximal chain,
the largest element in C is not in S

and hence by the induction

hypothesis, the theorem holds for S

. Hence S

is the union of M
disjoint chains S

1
, . . . , S

M
, where a
i
S

i
. Suppose x S

i
and
54 A Course in Combinatorics
x > a
i
. Since there is a j with x a
j
, we would have a
i
< a
j
,
a contradiction. This shows that a
i
is the maximal element of the
chain S

i
, i = 1, . . . , m. We do the same for S
+
. By combining the
chains the theorem follows.
A dual to Dilworths theorem was given by Mirsky (1971).
Theorem 6.2. Let P be a partially ordered set. If P possesses no
chain of m+ 1 elements, then P is the union of m antichains.
Proof: For m = 1 the theorem is trivial. Let m 2 and assume
that the theorem is true for m1. Let P be a partially ordered set
that has no chain of m+1 elements. Let M be the set of maximal
elements of P. M is an antichain. Suppose x
1
< x
2
< < x
m
were a chain in PM. Then this would also be a maximal chain in
P and hence we would have x
m
M, a contradiction. Hence PM
has no chain of m elements. By the induction hypothesis, PM is
the union of m1 antichains. This proves the theorem.
The following famous theorem due to Sperner (1928) is of a sim-
ilar nature. This proof is due to Lubell (1966).
Theorem 6.3. If A
1
, A
2
, . . . , A
m
are subsets of N := 1, 2, . . . , n
such that A
i
is not a subset of A
j
if i ,= j, then m
_
n
n/2|
_
.
Proof: Consider the poset of subsets of N. / := A
1
, . . . , A
m
is
an antichain in this poset.
A maximal chain ( in this poset will consist of one subset of each
cardinality 0, 1, . . . , n, and is obtained by starting with the empty
set, then any singleton set (n choices), then any 2-subset containing
the singleton (n 1 choices), then any 3-subset containing the 2-
subset (n 2 choices), etc. Thus there are n! maximal chains.
Similarly, there are exactly k!(nk)! maximal chains which contain
a given k-subset A of N.
Now count the number of ordered pairs (A, () such that A /,
( is a maximal chain, and A (. Since each maximal chain (
contains at most one member of an antichain, this number is at
most n!. If we let
k
denote the number of sets A / with [A[ = k,
then this number is

n
k=0

k
k!(n k)!. Thus
n

k=0

k
k!(n k)! n!, or equivalently,
n

k=0

k
_
n
k
_
1.
6. Dilworths theorem and extremal set theory 55
Since
_
n
k
_
is maximal for k = n/2| and

k
= m, the result
follows.
Equality holds in Theorem 6.3 if we take all n/2|-subsets of N
as the antichain.
We now consider the poset B
n
(with 2
n
elements) of the subsets
of the n-set N, ordered by inclusion. The set of i-subsets of N is
denoted by /
i
. We dene a symmetric chain in B
n
to be a sequence
P
k
, P
k+1
, . . . , P
nk
of vertices such that P
i
/
i
and P
i
P
i+1
for
i = k, k + 1, . . . , n k 1. We describe an algorithm due to De
Bruijn, Van Ebbenhorst Tengbergen and Kruyswijk (1949), that
splits B
n
into (disjoint) symmetric chains.
Algorithm: Start with B
1
. Proceed by induction. If B
n
has
been split into symmetric chains, then for each such symmetric
chain P
k
, . . . , P
nk
dene two symmetric chains in B
n+1
, namely
P
k+1
, . . . , P
nk
and P
k
, P
k
n+1, P
k+1
n+1, . . . , P
nk
n+1.
It is easy to see that this algorithm does what we claim. Fur-
thermore it provides a natural matching between k-subsets and
(n k)-subsets in B
n
(cf. Theorem 5.1). Also, see Problem 6D
below.
Problem 6A. Let a
1
, a
2
, . . . , a
n
2
+1
be a permutation of the inte-
gers 1, 2, . . . , n
2
+1. Show that Dilworths theorem implies that the
sequence has a subsequence of length n + 1 that is monotone.
A nice direct proof of the assertion of Problem 6A is as follows.
Suppose there is no increasing subsequence of n + 1 terms. De-
ne b
i
to be the length of the longest increasing subsequence that
starts with the term a
i
. Then by the pigeonhole principle, there are
at least n + 1 terms in the b
i
-sequence that have the same value.
Since i < j and b
i
= b
j
imply that a
i
> a
j
, we have a decreasing
subsequence of n + 1 terms.
To show a connection between Chapters 5 and 6, we now prove
that Theorem 5.1 immediately follows from Theorem 6.1. We
consider the bipartite graph G of Theorem 5.1. Let [X[ = n,
[Y [ = n
/
n. We introduce a partial order by dening x
i
< y
j
if and only if there is an edge from vertex x
i
to vertex y
j
. Suppose
that the largest antichain contains s elements. Let this antichain be
x
1
, . . . , x
h
, y
1
, . . . , y
k
, where h + k = s. Since (x
1
, . . . , x
h
)
56 A Course in Combinatorics
Y y
1
, . . . , y
k
, we have h n
/
k. Hence s n
/
. The partially
ordered set is the union of s disjoint chains. This will consist of
a matching of size a, the remaining n a elements of X, and the
remaining n
/
a elements of Y . Therefore n +n
/
a = s n
/
, i.e.
a n, which means that we have a complete matching.
Theorem 6.3 is a (fairly easy) example of an area known as ex-
tremal set theory in which the problems are often quite dicult.
We rst give one more example as an easy exercise.
Problem 6B. Let the sets A
i
, 1 i k, be distinct subsets of
1, 2, . . . , n. Suppose A
i
A
j
,= for all i and j. Show that
k 2
n1
and give an example where equality holds.
We now give one more example of the method that we used to
prove Sperners theorem. We prove the so-called Erd osKoRado
theorem (1961).
Theorem 6.4. Let / = A
1
, . . . , A
m
be a collection of m distinct
k-subsets of 1, 2, . . . , n, where k n/2, with the property that
any two of the subsets have a nonempty intersection. Then m
_
n1
k1
_
.
Proof: Place the integers 1 to n on a circle and consider the family
T := F
1
, . . . , F
n
of all consecutive k-tuples on the circle, i.e. F
i
denotes i, i + 1, . . . , i +k 1 where the integers should be taken
mod n. We observe that [/T[ k because if some F
i
equals A
j
,
then at most one of the sets l, l +1, . . . , l +k1, l k, . . . , l 1
(i < l < i +k) is in /. The same assertion holds for the collection
T

obtained from T by applying a permutation to 1, . . . , n.

Therefore
:=

S
n
[/ T

[ k n!.
We now count this sum by xing A
j
/, F
i
T and observing
that there are k!(nk)! permutations such that F

i
= A
j
. Hence
= m n k!(n k)!. This proves the theorem.
By a slight modication of the proof, one can show that the
theorem also holds if the sets in / are assumed to have size at
most k and they form an antichain. However we shall give a proof
using Theorem 5.1.
6. Dilworths theorem and extremal set theory 57
Theorem 6.5. Let / = A
1
, . . . , A
m
be a collection of m subsets
of N := 1, 2, . . . , n such that A
i
_ A
j
and A
i
A
j
,= if i ,= j
and [A
i
[ k n/2 for all i. Then m
_
n1
k1
_
.
Proof: (i) If all the subsets have size k, then we are done by
Theorem 6.4.
(ii) Let A
1
, . . . , A
s
be the subsets with the smallest cardinality,
say l
n
2
1. Consider all the (l +1)-subsets B
j
of N that contain
one or more of the sets A
i
, 1 i s. Clearly none of these is in /.
Each of the sets A
i
, 1 i s, is in exactly n l of the B
j
s and
each B
j
contains at most l +1 n l of the A
i
s. So by Theorem
5.1, we can pick s distinct sets, say B
1
, . . . , B
s
, such that A
i
B
i
.
If we replace A
1
, . . . , A
s
by B
1
, . . . , B
s
, then the new collection /
/
satises the conditions of the theorem and the subsets of smallest
cardinality now all have size > l. By induction, we can reduce to
case (i).
By replacing the counting argument of the proof of Theorem 6.4
by an argument in which the subsets are counted with weights, we
can prove the following generalization due to B. Bollob as (1973).
Theorem 6.6. Let / = A
1
, . . . , A
m
be a collection of m distinct
subsets of 1, 2, . . . , n, where [A
i
[ n/2 for i = 1, . . . , m, with the
property that any two of the subsets have a nonempty intersection.
Then
m

i=1
1
_
n1
[A
i
[1
_ 1.
Proof: Let be a permutation of 1, 2, . . . , n placed on a circle
and let us say that A
i
if the elements of A
i
occur consecutively
somewhere on that circle. By the same argument as in the proof
of Theorem 6.4 we see that if A
i
, then A
j
for at most [A
i
[
values of j.
Now dene
f(, i) :=
_
1
[A
i
[
, if A
i

0, otherwise.
By the argument above

S
n

m
i=1
f(, i) n!. Changing the
order of summation we have to count, for a xed A
i
, the number of
permutations placed on a circle such that A
i
. This number
58 A Course in Combinatorics
(by the same argument as in Theorem 6.4) is n [A
i
[!(n[A
i
[)!. So
we have
m

i=1
1
[A
i
[
n [A
i
[!(n [A
i
[)! n!,
which yields the result.
Problem 6C. Let / = A
1
, . . . , A
m
be a collection of m distinct
subsets of N := 1, 2, . . . , n such that if i ,= j then A
i
_ A
j
,
A
i
A
j
,= , A
i
A
j
,= N. Prove that
m
_
n 1

n
2
| 1
_
.
Problem 6D. Consider the decomposition of B
n
into symmetric
chains as described above. Show that Theorem 6.3 is an immediate
consequence of this decomposition. Show that Theorem 6.5 reduces
to Theorem 6.4 via this decomposition. How many of the chains
have their smallest element in /
i
?
Problem 6E. Here is an algorithm to construct a symmetric chain
in the poset B
n
which contains a given element S (a subset of
1, 2, . . . , n). Consider the characteristic vector x of S; for ex-
ample, if n = 7 and S = 3, 4, 7, then x = 0011001. Mark
all consecutive pairs 10, temporarily delete these pairs and again
mark all consecutive pairs 10, and repeat until only a string of the
form 00 01 11 remains. In our example, we obtain 00

01,
where the i-th coordinates are marked for i = 3, 4, 5, 6; when these
are deleted, the string 001 remains. The characteristic vectors of
the subsets in the chain are obtained by xing all marked coordi-
nates and letting the remaining coordinates range over the strings
0 000, 0 001, 0 011, . . . , 1 111. In our example, these
characteristic vectors are
00

00,
00

01,
01

01,
11

01,
6. Dilworths theorem and extremal set theory 59
which correspond to the subsets
3, 4, 3, 4, 7, 2, 3, 4, 7, 1, 2, 3, 4, 7.
Show that this algorithm produces exactly the same symmetric
chain containing S as is produced by the inductive algorithm of De
Bruijn et al. described above.
Notes.
We shall return to partially ordered sets in Chapters 23 and 25.
E. Sperner (19051980) is best known for a lemma in combina-
torial topology known as Sperners lemma, which occurred in his
thesis (1928). It was used to give a proof of Brouwers xed point
theorem. (Another connection to combinatorics: his rst professor-
ship was in K onigsberg!) He was one of the pioneers of the famous
Oberwolfach research institute.
For a survey of extremal set theory, we refer to Frankl (1988).
The short proof of the Erd osKoRado theorem is due to Katona
(1974). Theorem 6.5 is due to Kleitman and Spencer (1973) and
Schonheim (1971). The proof of Theorem 6.6 is due to Greene,
Katona and Kleitman (1976).
References.
B. Bollobas (1973), Sperner systems consisting of pairs of comple-
mentary subsets, J. Combinatorial Theory (A) 15, 363366.
N. G. de Bruijn, C. van Ebbenhorst Tengbergen and D. Kruyswijk
(1949), On the set of divisors of a number, Nieuw Archief v.
Wisk. (2) 23, 191193.
R. P. Dilworth (1950), A decomposition theorem for partially or-
dered sets, Annals of Math. (2) 51, 161166.
P. Erd os, Chao Ko, and R. Rado (1961), Extremal problems among
subsets of a set, Quart. J. Math. Oxford Ser. (2) 12, 313318.
P. Frankl (1988), Old and new problems on nite sets, Proc. Nine-
teenth S. E. Conf. on Combinatorics, Graph Th. and Computing,
Baton Rouge, 1988.
C. Greene, G. Katona, and D. J. Kleitman (1976), Extensions of
the Erd osKoRado theorem, Stud. Appl. Math. 55, 18.
60 A Course in Combinatorics
G. O. H. Katona (1974), Extremal problems for hypergraphs, in
Combinatorics (edited by M. Hall, Jr. and J. H. van Lint), Reidel.
D. J. Kleitman and J. Spencer (1973), Families of k-independent
sets, Discrete Math. 6, 255262.
D. Lubell (1966), A short proof of Sperners lemma, J. Combina-
torial Theory 1, 299.
L. Mirsky (1971), A dual of Dilworths decomposition theorem,
Amer. Math. Monthly 78, 876877.
J. Sch onheim (1971), A generalization of results of P. Erd os, G.
Katona, and D. J. Kleitman concerning Sperners theorem, J.
Combinatorial Theory (A) 11, 111117.
E. Sperner (1928), Ein Satz uber Untermengen einer endlichen
Menge, Math. Zeitschrift 27, 544548.
H. Tverberg (1967), On Dilworths decomposition theorem for par-
tially ordered sets, J. Combinatorial Theory 3, 305306.
7
Flows in networks
By a transportation network, we will mean a nite directed graph
D together with two distinguished vertices s and t called the source
and the sink, respectively, and which is provided with a function
c associating to each edge e a nonnegative real number c(e) called
its capacity. We may further assume that there are no loops, no
multiple edges, and that no edges enter the source s or leave the sink
t (although there would be no harm in admitting any of these types
of edges other than our having to be more careful in a denition or
two).
Figure 7.1
In Fig. 7.1 we give an example. We could think of a network
of pipes through which some liquid could ow in the direction of
the arrows. The capacity would indicate the maximal possible ow
(per time unit) in that section of pipe.
A ow in a transportation network is a function f assigning a
real number f(e) to each edge e such that:
(a) 0 f(e) c(e) for all edges e (the ow is feasible);
(b) for each vertex x (not the source or the sink) the sum of the
62 A Course in Combinatorics
values of f on incoming edges equals the sum of the values of f on
outgoing edges (conservation of ow).
The sum of the values of a ow f on the edges leaving the source
is called the strength of the ow (denoted by [f[). It seems obvious
that the strength of the ow is also equal to the sum of the values
of f on edges entering the sink; the reader might try to verify this
formally before reading further.
One of our objectives will be to nd a method for constructing
a maximum ow, that is, a ow with maximum strength. Before
we begin, it will be good to have a goal or an upper bound for the
strength of a ow; for example, the sum of the capacities of all edges
leaving the source is clearly such an upper bound. More generally,
by a cut separating s and t (or simply a cut), we mean here a pair
(X, Y ) of subsets of the vertex set V := V (D) which partition V
and such that s X and t Y . We dene the capacity c(X, Y ) of
the cut to be the sum of the capacities of the edges directed from
X to Y (that is, edges e = (x, y) with x X and y Y ). We claim
that the capacity of any cut is an upper bound for the strength of
any ow. More strongly, we claim that the conservation law implies
(see below) that the strength of a ow f can be computed as
(7.1) [f[ = f(X, Y ) f(Y, X),
where f(A, B) denotes the sum of the values of f on all edges
directed from A to B; then the feasibility of f immediately implies
that [f[ c(X, Y ). Thus the minimum capacity of all cuts in a
network (e.g. in Fig. 7.1 the minimum cut capacity is 20) is an
upper bound for the strength of a ow in that network.
To establish (7.1), we introduce the function by dening for
each pair (x, e), where x is a vertex incident with the edge e,
(x, e) := 1 if the edge is incoming, and (x, e) := +1 if the
edge is outgoing; (x, e) is to be 0 if x is not incident with e. (We
remark that is essentially the incidence matrix of the directed
graphsee Chapter 36.) The conservation law is equivalent to

eE
(x, e)f(e) = 0 for x ,= s, t. Notice that

xX
(x, e) is +1
if e is directed from X to Y , 1 if e is directed from Y to X, and
7. Flows in networks 63
0 if e has both endpoints in X or both in Y . Then
[f[ =

eE
(s, e)f(e) =

xX

eE
(x, e)f(e)
=

eE
f(e)

xX
(x, e) = f(X, Y ) f(Y, X).
(In the rst double sum above, the inner sum is 0 for all terms x
other than s.)
A special instance of (7.1) is [f[ = f(V t, t), the assertion
which we invited the reader to reect on earlier.
We now construct ows. Fix a ow f, possibly the 0-ow. We
shall say that the sequence x
0
, x
1
, . . . , x
k1
, x
k
of distinct vertices is
a special path from x
0
to x
k
if for each i, 1 i k, either
(i) e = (x
i1
, x
i
) is an edge with c(e) f(e) > 0, or
(ii) e = (x
i
, x
i1
) is an edge with f(e) > 0.
Edges e with f(e) = c(e) are said to be saturated and conditions (i)
and (ii) can be stated in words as requiring that forward edges of
the path are unsaturated while backward edges are positiveall
with respect to a given ow f. Suppose there exists such a special
path from s to t. Dene
i
as c(e) f(e) in the rst case and as
f(e) in the second case (picking one of the edges to use if both cases
hold) and let be the minimum of these positive numbers
i
. On
each edge of type (i) increase the ow value by , and on each edge
of type (ii) decrease the ow by . It is easy to check that the two
conditions for a ow (feasibility and conservation of ow) are still
satised. Clearly the new ow has strength [f[ +.
This idea for obtaining a stronger ow becomes an algorithm
when we iterate it (starting with the 0-ow) and incorporate a
systematic procedure for searching for special paths from s to t
with respect to the current ow. We make brief remarks concerning
termination in the notes to this chapter. But what happens when
we can go no further?
Suppose that no special path from source to sink exists with re-
spect to some ow f
0
. Let X
0
be the set of vertices x which can be
reached from s by a special path, Y
0
the set of remaining vertices.
64 A Course in Combinatorics
In this way we produce a cut. If x X
0
, y Y
0
and e = (x, y) is
an edge, then e must be saturated or we could adjoin y to a special
path from s to x to get a special path from s to y, contradicting
the denitions of X
0
and Y
0
. If, on the other hand, e = (y, x) is an
edge, then, for a similar reason, f(e) must be 0. In view of (7.1),
we have then
[f
0
[ = f
0
(X
0
, Y
0
) f
0
(Y
0
, X
0
) = c(X
0
, Y
0
).
Now it is clear that not only can no stronger ow be obtained by
our method of special paths, but that no stronger ows exist at all
because [f[ c(X
0
, Y
0
) for any ow f.
If f
0
is chosen to be a maximum ow (which exists by continuity
reasons in case one is unsure of the termination of the algorithm),
then surely no special paths from s to t exist. Note that the con-
structed cut (X
0
, Y
0
) is a minimum cut (i.e. a cut of minimum ca-
pacity), since c(X, Y ) [f
0
[ for any cut (X, Y ). Our observations
have combined to prove the following famous theorem of Ford and
Fulkerson (1956).
Theorem 7.1. In a transportation network, the maximum value
of [f[ over all ows f is equal to the minimum value of c(X, Y ) over
all cuts (X, Y ).
This theorem is usually referred to as the maxow-mincut the-
orem. The procedure for increasing the strength of a ow that we
used above shows somewhat more.
Theorem 7.2. If all the capacities in a transportation network
are integers, then there is a maximum strength ow f for which all
values f(e) are integers.
Proof: Start with the 0-ow. The argument above provides a way
to increase the strength until a maximum ow is reached. At each
step is an integer, so the next ow is integer valued too.
Problem 7A. Construct a maximum ow for the transportation
network of Fig. 7.1.
Problem 7B. An elementary ow in a transportation network is
a ow f which is obtained by assigning a constant positive value
7. Flows in networks 65
to the set of edges traversed by a simple (directed) path from s
to t, and 0 to all other edges. Show that every ow is the sum of
elementary ows and perhaps a ow of strength zero. (This means
we can arrive at a maxow by starting from the 0-ow and using
only special paths with forward edges.) Give an example of a
network and a ow which is not maximum, but with respect to
which there are no special paths using only forward edges.
Problem 7C. Let (X
1
, Y
1
) and (X
2
, Y
2
) be minimum cuts (i.e. cuts
of minimum capacity) in a transportation network. Show that (X
1

X
2
, Y
1
Y
2
) is also a minimum cut. (This can be done either from
rst principles, or with an argument involving maximum ows.)
Problem 7D. Prove P. Halls marriage theorem, Theorem 5.1,
from Theorems 7.1 and 7.2.
It should be clear that the topic of this chapter is of great practi-
cal importance. Routing schemes for all kinds of products depend
on algorithms that produce optimal ows through transportation
networks. We do not go into the algorithmic aspect of this area.
Instead, we shall show a beautiful application of Theorem 7.2 to a
problem related to Birkhos theorem, Theorem 5.5. Before giving
the theorem and its proof, we observe that several attempts were
made to prove it by reducing it to Theorem 5.5 but with no success.
The proof below is due to A. Schrijver. (If b = v in Theorem 7.3,
then we have the situation of Theorem 5.5.)
Theorem 7.3. Let A be a b v (0, 1)-matrix with k ones per row
and r ones per column (so bk = vr). Let be a rational number,
0 < < 1, such that k
/
= k and r
/
= r are integers. Then there
is a (0, 1)-matrix A
/
of size b v with k
/
ones per row and r
/
ones
per column such that entries a
/
ij
of A
/
are 1 only if the corresponding
entries of A are 1, i.e. A
/
can be obtained from A by changing some
ones into zeros.
Proof: We construct a transportation network with vertices s (the
source), x
1
, . . . , x
b
(corresponding to the rows of A), y
1
, . . . , y
v
(cor-
responding to the columns of A), and t (the sink). Edges are (s, x
i
)
with capacity k, 1 i b, (x
i
, y
j
) with capacity 1 if and only if
a
ij
= 1, and (y
j
, t) with capacity r, 1 j v. The denition
66 A Course in Combinatorics
ensures that there is a maximum ow with all edges saturated.
We now change the capacities of the edges from the source to k
/
and those of the edges to the sink to r
/
. Again, all the capacities
are integers and clearly a maximum ow exists for which the ows
f((x
i
, y
j
)) are equal to . By Theorem 7.2 there is also a maximum
ow f

for which all the ows are integers, i.e. f

((x
i
, y
j
)) = 0 or
1. From this ow, we immediately nd the required matrix A
/
.
The theorem above can be generalized in several ways with es-
sentially the same proof idea, but see below for a slightly dierent
approach.
For some combinatorial applications, it is convenient to use the
following theorem, which does not require the introduction of ca-
pacities or the concept of strength. It can be derived from Theorem
7.2see Ford and Fulkerson (1956)but we choose to give a direct
proof.
A circulation on a digraph D is a mapping f from E(D) to the
reals satisfying conservation of ow at every vertex. We do not
require nonnegativity. Circulations may be identied with vectors
in the null space of the incidence matrix of the digraph.
Theorem 7.4. Let f be a circulation on a nite digraph D. Then
there exists an integral circulation g such that for every edge e, g(e)
is equal to one of f(e)| or f(e)|.
We may say that the values of g are those of f rounded up or
down. Of course, if f(e) is already an integer, then g(e) = f(e).
Proof: Given a circulation f, consider a circulation g that satises
(7.2) f(e)| g(e) f(e)|
and for which the number of edges e with g(e) an integer is as large
as possible subject to (7.2).
Let H be the spanning subgraph of D with edge set consisting of
those edges of D for which g(e) is not an integer, i.e. for which strict
inequality holds both times in (7.2). Conservation of ow implies
that no vertex can have degree 1 in H, so if g is not integral, then
H contains a polygon.
Let P be a polygon in H and traverse P with a simple closed
path; let A be the set of edges of P that are forward edges of the
7. Flows in networks 67
path in D, and B the set of edges of P that are backward edges in
this path. For any constant c, we obtain a new circulation g
/
by
g
/
(e) :=
_
_
_
g(e) +c if e A,
g(e) c if e B,
g(e) if e / E(P).
If c is small, (7.2) will still hold with g replaced by g
/
. Now choose
c := min
_
min
eA
_
f(e)| g(e)
_
, min
eB
_
g(e) f(e)|
__
.
Then g
/
still satises (7.2), yet g
/
(e) is an integer for at least one
more edge (any edge for the which term in the expression above
achieves the minimum). This would contradict the choice of g,
were g not integral.
Corollary. Let f be an integral circulation on a nite digraph
D and d any positive integer. Then f can be written as the sum
g
1
+g
2
+ +g
d
of integral circulations such that for each index j
and each edge e,
(7.3) f(e)/d| g
j
(e) f(e)/d|.
Proof: By induction on d. For d = 1, there is nothing to prove.
Given d 2, apply Theorem 7.4 to f/d to nd an integral circu-
lation g
1
satisfying (7.3) for j = 1. Apply the induction hypothesis
to nd
f g
1
= g
2
+g
3
+ +g
d
where for each j = 2, 3, . . . , d, g
j
is an integral circulation satisfying
(f(e) g
1
(e))/(d 1)| g
j
(e) (f(e) g
1
(e))/(d 1)|.
An easy exercise is that if a is an integer and b is either a/d| or
a/d|, then

a
d
|
a b
d 1
| and
a b
d 1
|
a
d
|,
so that the above inequalities imply (7.3) for j = 2, 3, . . . , d.
68 A Course in Combinatorics
From an m n matrix A of real numbers a
ij
, not necessarily
nonnegative or integers, we obtain a circulation on a digraph with
m + n + 2 vertices and mn + m + n + 1 edges. The digraph is
similar to the one used in the proof of Theorem 7.3. There are ver-
tices x
1
, . . . , x
m
corresponding to the rows, vertices y
1
, . . . , y
n
cor-
responding to the columns, and two others called s and t. There
is an edge from x
i
to y
j
with circulation value a
ij
, an edge from s
to x
i
with circulation value equal to the i-th row-sum r
i
, an edge
from y
j
to t with circulation value equal to the j-th column-sum k
j
(i = 1, . . . , m, j = 1, . . . , n), and an edge from t to s with circula-
tion value equal to the sum of all entries of M. If we multiply this
circulation f by any scalar , apply Theorem 7.4 to f, and rein-
terpret the resulting integral circulation as a matrix, we obtain part
(i) of the following theorem. Part (ii) follows from the corollary.
Theorem 7.5. (i) Given a matrix A and a real number , there
is an integral matrix B so that the entries of B, the row-sums of
B, the column-sums of B, and the sum of all entries of B, are the
corresponding values for A rounded up or down. (ii) If A is an
integral matrix and d any positive integer, then
A = B
1
+B
2
+ +B
d
where each B
i
is an integral matrix whose entries, row-sums, column-
sums, and sum of all entries, are those of (1/d)A, rounded up or
down.
Problem 7E. Show that the following results are quick conse-
quences of Theorem 7.5: (i) Problem 5A(iii); (ii) Theorem 5.5; (iii)
Theorem 7.3; (iv) A nite graph all of whose vertices have even de-
gree has a balanced orientation, where the in-degree and out-degree
of each vertex are equal; (v) If a bipartite graph has minimum de-
gree d and maximum degree d, then its edges may be colored with
d colors so that the colors that appear at every vertex are distinct,
and with d colors so that all colors appear at each vertex.
Problem 7F. Show that the dimension of the vector space of all
circulations on a connected digraph D is [E(D)[ [V (D)[ + 1.
7. Flows in networks 69
Notes.
The term augmenting path is often used instead of special path.
If the capacities of a transportation network are integers, the spe-
cial path method for constructing maximum ows will terminate af-
ter nitely many iterations, since the strength increases by at least
one each time. But Ford and Fulkerson (1962) give an example with
irrational capacities where certain contrived choices of special paths
lead to an innite sequence of ows whose strengths convergebut
only to one-fourth of the actual maximum ow strength! If one
is careful to pick shortest special paths, however, then it can be
shown that a maximum ow is reached after at most O(n
3
) itera-
tions, where n is the number of vertices. See Edmonds and Karp
(1972).
The problem of nding a maximum ow is an example of a linear
programming problem and can be solved e.g. by the simplex algo-
rithm. The network ow problem is special in that its matrix is
totally unimodular, and this is one way of explaining why Theorem
7.2 holds. See the references below for more discussion of linear and
integer programming. Graphical methods are usually faster than
the simplex algorithm, and add insight.
Circulations on a digraph are called 1-cycles in algebraic topol-
ogy. An analogue of Theorem 7.4 holds for vectors f in the null
space of any totally unimodular matrix.
Theorems 7.1, 7.2, 7.4, and the algorithm have many further
combinatorial applications, since certain combinatorial problems
can be phrased in terms of transportation networks. For example,
nding a maximum matching in a bipartite graph is equivalent to
nding a maximum (integer valued) ow in a certain associated
networksee the referencesand thus a good algorithm exists to
nd a maximum matching. We give further applications of these
theorems in Chapter 16 to an existence problem on (0,1)-matrices,
and in Chapter 38 to a problem on partitions of sets.
References.
J. Edmonds and R. M. Karp (1972), Theoretical improvements
in algorithm eciency for network ow problems, J. Assn. for
Computing Machinery 19, 248264.
70 A Course in Combinatorics
L. R. Ford, Jr. and D. R. Fulkerson (1962), Flows in Networks,
Princeton University Press.
T. C. Hu (1969), Integer Programming and Network Flows, Addi-
son-Wesley.
V. Chv atal (1983), Linear Programming, W. H. Freeman.
8
De Bruijn sequences
The following problem has a practical origin: the so-called rotating
drum problem. Consider a rotating drum as in Fig. 8.1.
Figure 8.1
Each of the segments is of one of two types, denoted by 0 and 1.
We require that any four consecutive segments uniquely determine
the position of the drum. This means that the 16 possible quadru-
ples of consecutive 0s and 1s on the drum should be the binary
representations of the integers 0 to 15. Can this be done and, if yes,
in how many dierent ways? The rst question is easy to answer.
Both questions were treated by N. G. de Bruijn (1946) and for this
reason the graphs described below and the corresponding circular
sequences of 0s and 1s are often called De Bruijn graphs and De
Bruijn sequences, respectively.
We consider a digraph (later to be called G
4
) by taking all 3-
tuples of 0s and 1s (i.e. 3-bit binary words) as vertices and joining
the vertex x
1
x
2
x
3
by a directed edge (arc) to x
2
x
3
0 and x
2
x
3
1.
The arc (x
1
x
2
x
3
, x
2
x
3
x
4
) is numbered e
j
, where x
1
x
2
x
3
x
4
is the
binary representation of the integer j. The graph has a loop at 000
and at 111. As we saw before, the graph has an Eulerian circuit
because every vertex has in-degree 2 and out-degree 2. Such a
72 A Course in Combinatorics
closed path produces the required 16-bit sequence for the drum.
Such a (circular) sequence is called a De Bruijn sequence. For
example the path 000 000 001 011 111 111 110
100 001 010 101 011 110 101 010 100 000
corresponds to 0000111100101101 (to be read circularly). We call
such a path a complete cycle.
We dene the graph G
n
to be the directed graph on (n1)-tuples
of 0s and 1s in a similar way as above. (So G
n
has 2
n
edges.)
The graph G
4
is given in Fig. 8.2. In this chapter, we shall call a
digraph with in-degree 2 and out-degree 2 for every vertex, a 2-in
2-out graph. For such a graph G we dene the doubled graph G

as follows:
(i) to each edge of G there corresponds a vertex of G

;
(ii) if a and b are vertices of G

, then there is an edge from a to

b if and only if the edge of G corresponding to a has as terminal
end (head) the initial end (tail) of the edge of G corresponding to
b.
Clearly G

n
= G
n+1
.
Figure 8.2
Theorem 8.1. Let G be a 2-in 2-out graph on m vertices with M
complete cycles. Then G

has 2
m1
M complete cycles.
Proof: The proof is by induction on m.
(a) If m = 1 then G has one vertex p and two loops from p to p.
Then G

= G
2
which has one complete cycle.
8. De Bruijn sequences 73
(b) We may assume that G is connected. If G has m vertices
and there is a loop at every vertex, then, besides these loops, G is
a circuit p
1
p
2
p
m
p
1
. Let A
i
be the loop p
i
p
i
and
B
i
the arc p
i
p
i+1
. We shall always denote the corresponding
vertices in G

by lower case letters. The situation in G

is as in
Fig. 8.3.
Figure 8.3
Clearly a cycle in G

has two ways of going from b

i
to b
i+1
. So
G

has 2
m1
complete cycles, whereas G has only one.
(c) We now assume that G has a vertex x that does not have
a loop on it. The situation is as in Fig. 8.4, where P, Q, R, S are
dierent edges of G (although some of the vertices a, b, c, d may
coincide).
From G we form a new 2-in 2-out graph with one vertex less by
deleting the vertex x. This can be done in two ways: G
1
is obtained
by the identication P = R, Q = S, and G
2
is obtained by P = S,
Q = R. By the induction hypothesis, the theorem applies to G
1
and to G
2
.
Figure 8.4
There are three dierent types of complete cycle in G

, depending
on whether the two paths leaving r and returning to p, respectively
q, both go to p, both to q, or one to p and one to q. We treat one
74 A Course in Combinatorics
case; the other two are similar and left to the reader. In Fig. 8.5
we show the situation where path 1 goes from r to p, path 2 from
s to q, path 3 from s to p, and path 4 from r to q.
Figure 8.5
These yield the following four complete cycles in G

:
1, pr, 4, qs, 3, ps, 2, qr
1, ps, 2, qr, 4, qs, 3, pr
1, ps, 3, pr, 4, qs, 2, qr
1, ps, 2, qs, 3, pr, 4, qr
In G

1
and G

2
the situation reduces to Fig. 8.6.
Figure 8.6
In each of G

1
and G

2
one complete cycle using the paths 1, 2, 3, 4
is possible. In the remaining two cases, we also nd two complete
cycles in G

1
and G

2
corresponding to four complete cycles in G

.
Therefore the number of complete cycles in G

is twice the sum of

the numbers for G

1
and G

2
. On the other hand, the number of
complete cycles in G is clearly equal to the sum of the correspond-
ing numbers for G
1
and G
2
. The theorem then follows from the
induction hypothesis.
We are now able to answer the question how many complete
cycles there are in a De Bruijn graph.
8. De Bruijn sequences 75
Theorem 8.2. G
n
has exactly 2
2
n1
n
complete cycles.
Proof: The theorem is true for n = 1. Since G

n
= G
n+1
, the
result follows by induction from Theorem 8.1.
For a second proof, see Chapter 36.
Problem 8A. Let be a primitive element in F
2
n . For 1 i
m := 2
n
1, let

i
=
n1

j=0
c
ij

j
.
Show that the sequence
0, c
10
, c
20
, . . . , c
m0
is a De Bruijn sequence.
Problem 8B. Find a circular ternary sequence (with symbols
0, 1, 2) of length 27 so that each possible ternary ordered triple
occurs as three (circularly) consecutive positions of the sequence.
First sketch a certain directed graph on 9 vertices so that Eulerian
circuits in the graph correspond to such sequences.
Problem 8C. We wish to construct a circular sequence a
0
, . . . , a
7
(indices mod 8) in such a way that a sliding window a
i
, a
i+1
, a
i+3
(i = 0, 1, . . . 7) will contain every possible three-tuple once. Show
(not just by trial and error) that this is impossible.
Problem 8D. Let m := 2
n
1. An algorithm to construct a De
Bruijn sequence a
0
, a
1
, . . . , a
m
works as follows. Start with a
0
=
a
1
= = a
n1
= 0. For k > n, we dene a
k
to be the maximal
value in 0, 1 such that the sequence (a
kn+1
, . . . , a
k1
, a
k
) has
not occurred in (a
0
, . . . , a
k1
) as a (consecutive) subsequence. The
resulting sequence is known as a Ford sequence. Prove that this
algorithm indeed produces a De Bruijn sequence.
Notes.
Although the graphs of this chapter are commonly called De
Bruijn graphs, Theorem 8.1 was proved in 1894 by C. Flye Sainte-
Marie. This went unnoticed for a long time. We refer to De Bruijn
(1975).
76 A Course in Combinatorics
N. G. de Bruijn (1918), one of the best-known Dutch mathe-
maticians, worked in many dierent areas such as analysis, number
theory, combinatorics, and also computing science and crystalogra-
phy.
We mention a peculiarity concerning the spelling of some Dutch
names. When omitting the initials of N. G. de Bruijn, one should
capitalize the word de and furthermore the name should be listed
under B. Similarly Van der Waerden is correct when the initials are
omitted and he should be listed under W.
For a proof of Theorem 8.1 using algebraic methods, we refer to
Chapter 36.
References.
N. G. de Bruijn (1946), A combinatorial problem, Proc. Kon. Ned.
Akad. v. Wetensch. 49, 758764.
N. G. de Bruijn (1975), Acknowledgement of priority to C. Flye
Sainte-Marie on the counting of circular arrangements of 2
n
zeros
and ones that show each n-letter word exactly once, T. H. report
75-WSK-06, Eindhoven University of Technology.
C. Flye Sainte-Marie (1894), Solution to question nr. 48, Intermedi-
aire des Mathematiciens 1, 107110.
9
Two (0,1,*) problems:
addressing for graphs and
a hash-coding scheme
The following problem originated in communication theory. For
a telephone network, a connection between terminals A and B is
established before messages ow in either direction. For a network
of computers it is desirable to be able to send a message from A
to B without B knowing that a message is on its way. The idea is
to let the message be preceded by some address of B such that
at each node of the network a decision can be made concerning the
direction in which the message should proceed.
A natural thing to try is to give each vertex of a graph G a
binary address, say in 0, 1
k
, in such a way that the distance of two
vertices in the graph is equal to the so-called Hamming distance of
the addresses, i.e. the number of places where the addresses dier.
This is equivalent to regarding G as an induced subgraph of the
hypercube H
k
, which has V (H
k
) := 0, 1
k
and where k-tuples are
adjacent when they dier in exactly one coordinate. The example
G = K
3
already shows that this is impossible. We now introduce a
new alphabet 0, 1, and form addresses by taking n-tuples from
this alphabet. The distance between two addresses is dened to be
the number of places where one has a 0 and the other a 1 (so stars
do not contribute to the distance). For an addressing of a graph
G, we require that the distance of any two vertices in G is equal to
the distance of their addresses. It is trivial to show that this can be
done if n is large enough. We denote by N(G) the minimum value
of n for which there exists an addressing of G with length n.
78 A Course in Combinatorics
For a tree we can do without the stars as follows. We use in-
duction. For a tree with two vertices, we have a trivial addressing
with length 1. Suppose that we can address trees with k vertices.
If x
0
, x
1
, . . . , x
k
are the vertices of the tree T and x
0
is a monova-
lent vertex, then consider an addressing for the tree obtained by
removing x
0
. Let x
i
be the address of x
i
and suppose x
0
is joined
to x
1
. We change all addresses to (0, x
i
), 1 i k, and give x
0
the address (1, x
1
). Clearly this is now an addressing for T. So for
a tree, we have N(T) [V (T)[ 1.
As a second example, consider K
m
. In the identity matrix of
size m 1, we replace the zeros above the diagonal by stars and
add a row of zeros. Any two rows now have distance 1 and hence
N(K
m
) m1.
As a third example, we consider the graph of Fig. 9.1.
Figure 9.1
A possible (though not optimal) addressing is
1 1 1 1
2 1 0 1
3 0 0 0 1
4 0 0 1
5 0 0 0 0 0
We now show a correspondence between addressings of a graph
and quadratic forms (an idea of Graham and Pollak, 1971). Con-
sider the graph G of Fig. 9.1 and the addressing given above.
To the rst column of the addressing, we associate the product
(x
1
+x
2
)(x
4
+x
5
). Here x
i
is in the rst, respectively second, factor
if the address of i has a 1, respectively a 0, in the rst column.
If we do the same thing for each column and then add the terms,
we obtain a quadratic form

d
ij
x
i
x
j
, where d
ij
is the distance of
9. Two (0,1,*) problems 79
the vertices i and j in G. Thus an addressing of G corresponds to
writing the quadratic form

d
ij
x
i
x
j
as a sum of n products
(x
i
1
+ +x
i
k
)(x
j
1
+ +x
j
l
)
such that no x
i
occurs in both of the factors. The number of vari-
ables is [V (G)[.
Theorem 9.1. Let n
+
, respectively n

, be the number of positive,

respectively negative, eigenvalues of the distance matrix (d
ij
) of the
graph G. Then N(G) maxn
+
, n

.
Proof: Each of the quadratic forms mentioned above can be rep-
resented as
1
2
x

Ax, where x := (x
1
, x
2
, . . . , x
n
) and A has entry
a
ij
= 1 if the term x
i
x
j
occurs in the quadratic form and 0 other-
wise. Such a matrix has rank 2 and trace 0. Therefore it has one
positive and one negative eigenvalue. Since (d
ij
) is the sum of the
matrices corresponding to the quadratic forms, it can have at most
n positive (respectively negative) eigenvalues.
Theorem 9.2. N(K
m
) = m1.
Proof: We have already seen that N(K
m
) m1. Since J I,
of size m, is the distance matrix of K
m
and the eigenvalues of J I
are m1, with multiplicity 1, and 1, with multiplicity m1, the
result follows from Theorem 9.1.
With slightly more work, we shall now show that the shortest
addressing for a tree T has length [V (T)[ 1.
Theorem 9.3. If T is a tree on n vertices, then N(T) = n 1.
Proof: We rst calculate the determinant of the distance matrix
(d
ij
) of T. We number the vertices p
1
, . . . , p
n
in such a way that p
n
is
an endpoint adjacent to p
n1
. In the distance matrix, we subtract
row n 1 from row n, and similarly for the columns. Then all
the entries in the new last row and column are 1 except for the
diagonal element which is equal to 2. Now renumber the vertices
p
1
, . . . , p
n1
in such a way that the new vertex p
n1
is an endpoint
80 A Course in Combinatorics
of Tp
n
adjacent to p
n2
. Repeat the procedure for the rows and
columns with numbers n 1 and n 2. After n 1 steps, we have
the determinant

0 1 1 . . . 1
1 2 0 . . . 0
1 0 2 . . . 0
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1 0 0 . . . 2

.
From this we nd the remarkable result that the determinant D
n
of the distance matrix of a tree on n vertices satises
D
n
= (1)
n1
(n 1)2
n2
,
i.e. it depends only on [V (T)[. If we number the vertices accord-
ing to the procedure described above, then the k k principal
minor in the upper left-hand corner of the distance matrix is the
distance matrix of a subtree on k vertices. Therefore the sequence
1, D
1
, D
2
, . . . , D
n
, where D
k
is the determinant of the k k minor,
is equal to
1, 0, 1, 4, 12, . . . , (1)
n1
(n 1)2
n2
.
If we consider the sign of 0 to be positive, then this sequence has
only one occurrence of two consecutive terms of the same sign. By
an elementary theorem on quadratic forms this implies that the
corresponding quadratic form has index 1, and hence (d
ij
) has one
positive eigenvalue; see B. W. Jones (1950), Theorem 4. Now the
result follows from Theorem 9.1.
The conjecture that in fact N(G) [V (G)[1 for all (connected)
graphs G was proved by P. Winkler in 1983. The proof is construc-
tive. In order to describe the addressing, we need some preparation.
Consider the graph of Fig. 9.2.
9. Two (0,1,*) problems 81
Figure 9.2
We pick a vertex x
0
, then construct a spanning tree T by a
breadth-rst search, and then number the vertices by a depth-rst
search. The result is shown on the right-hand side of Fig. 9.2, where
edges of E(G)E(T) are dashed.
Let n := [V (G)[ 1. We need several denitions.
For i n, we dene
P(i) := j : x
j
is on a path from x
0
to x
i
in T.
For example, P(6) = 0, 3, 4, 6. Let
ij := max(P(i) P(j)).
We describe the general situation in Fig. 9.3.
Figure 9.3
Note that in Fig. 9.3, we have i < j if and only if k < l.
For i n, we dene
i
/
:= max(P(i)i).
For example, 7
/
= 3 in Fig. 9.2. Dene
i j P(i) P(j) or P(j) P(i).
82 A Course in Combinatorics
We denote distances in G, respectively T, by d
G
, respectively d
T
.
The discrepancy function c(i, j) is now dened by
c(i, j) := d
T
(x
i
, x
j
) d
G
(x
i
, x
j
).
For example, in Fig. 9.2, c(6, 9) = 4.
Lemma 9.4.
(i) c(i, j) = c(j, i) 0;
(ii) if i j, then c(i, j) = 0;
(iii) if i , j, then c(i, j
/
) c(i, j) c(i, j
/
) + 2.
Proof: (i) is trivial; (ii) follows from the denition of T since
d
G
(x
i
, x
j
) [d
G
(x
j
, x
0
) d
G
(x
i
, x
0
)[ = d
T
(x
i
, x
j
);
(iii) follows from the fact that [d
G
(x
i
, x
j
)d
G
(x
i
, x
j
)[ 1 and that
d
T
(x
i
, x
j
) = 1 +d
T
(x
i
, x
j
).
Now we can dene the addressing. For 0 i n the vertex x
i
is given the address a
i
0, 1,
n
, where
a
i
= (a
i
(1), a
i
(2), . . . , a
i
(n))
and
a
i
(j) :=
_

_
1 if j P(i),
if
_
_
_
c(i, j) c(i, j
/
) = 2, or
c(i, j) c(i, j
/
) = 1, i < j, c(i, j) even, or
c(i, j) c(i, j
/
) = 1, i > j, c(i, j) odd,
0 otherwise.
Theorem 9.5. d(a
i
, a
k
) = d
G
(x
i
, x
k
).
Proof: We may assume i < k.
(i) Suppose i k. Then d
G
(x
i
, x
k
) = [P(k)P(i)[. The values of
j such that j P(k)P(i) are exactly the positions where a
k
(j) =
1, a
i
(j) ,= 1. For these values of j we see that c(i, j) = 0, hence
a
i
(j) = 0 and we are done.
(ii) The hard case is when i , k. The key observation is the
following. Let n
1
n
2
n
l
be a nondecreasing sequence of
9. Two (0,1,*) problems 83
integers such that [n
i+1
n
i
[ 2 for all i. If m is an even integer
between n
1
and n
l
that does not occur in the sequence, then there
is an i such that n
i
= m 1, n
i+1
= m + 1. Now consider the
sequence
c(i, k) c(i, k
/
) c(i, k
//
) c(i, ik) = 0.
By the denition of a
i
(j) and the observation above, a
i
(j) = and
a
k
(j) = 1 exactly as many times as there are even integers between
c(i, ik) and c(i, k). Similarly a
k
(j) = and a
i
(j) = 1 as many
times as there are odd integers between c(i, ik) and c(i, k). So
d(a
i
, a
k
) = [P(k)P(i)[ +[P(i)P(k)[ c(i, k)
= d
T
(x
i
, x
k
) c(i, k) = d
G
(x
i
, x
k
).

Therefore we have proved the following theorem.

Theorem 9.6. N(G) [V (G)[ 1.
Problem 9A. If we use the scheme dened above, what are the
addresses of x
2
and x
6
in Fig. 9.2?
Problem 9B. Let G be a cycle (polygon) on 2n vertices. Deter-
mine N(G).
Problem 9C. Let G be a cycle (polygon) on 2n+1 vertices. Prove
that N(G) = 2n. Hint: if C
k
is the permutation matrix with
entries c
ij
= 1 if and only if j i 1 (mod k) and
k
= 1, then
(1, ,
2
, . . . ,
k1
) is an eigenvector of C
k
.
* * *
We now look at a second problem involving k-tuples from the
alphabet 0, 1, . The objects we shall study were introduced by
Rivest (1974) and given the (unfortunate) name associative block
design; cf. Chapter 19 for block designs. An ABD(k, w) is a set of
b := 2
w
elements of 0, 1,
k
with the following properties: if the
elements are the rows of a b k matrix C, then
84 A Course in Combinatorics
(i) each row of C has k w stars;
(ii) each column of C has b(k w)/k stars;
(iii) any two distinct rows have distance at least 1.
Note that the denition implies that each vector in F
k
2
has dis-
tance 0 to exactly one row of C.
The origin of the problem is the following. Consider a le of k-bit
binary words. Each sequence in 0, 1,
k
is called a partial match
query. The partial match retrieval problem is to retrieve from the
le all words agreeing with the query in those positions where the
query species a bit. So-called hash-coding schemes divide a le
into b disjoint lists L
1
, L
2
, . . . , L
b
. A record x will be stored in the
list with index h(x), where h is the hash-function mapping 0, 1
k
onto 1, 2, . . . , b. For a given partial match query, some of the lists
must be searched. An analysis of the worst-case number of lists to
be searched led to the concept of ABD. In this case h(x) is the
index of the unique row of C which has distance 0 to x.
Example 9.1. The following matrix is an ABD(4, 3):
_
_
_
_
_
_
_
_
_
_
_
0 0 0
0 1 0
0 0 1
0 1 0
1 1 1
1 0 1
1 1 0
1 0 1
_
_
_
_
_
_
_
_
_
_
_
.
We rst prove some elementary properties of an ABD.
Theorem 9.7. If an ABD(k, w) exists, then:
(1) it has exactly bw/(2k) zeros and bw/(2k) ones in each col-
umn;
(2) for each x in F
k
2
it has exactly
_
w
u
_
rows which agree with x
in u positions;
(3) the parameters satisfy
w
2
2k
_
1
1
b
_
;
9. Two (0,1,*) problems 85
(4) for any row, the number of rows with stars in the same
positions is even.
Proof: Let C be the ABD(k, w).
(1) A row of C with a star, respectively a zero, in column j rep-
resents (i. e. has distance 0 to) 2
kw1
, respectively 2
kw
, elements
of F
k
2
. From (i) and (ii) of the denition, it follows that column j
must contain bw/(2k) zeros.
(2) Let x F
k
2
. Denote by n
i
the number of rows of C which
agree with x in i positions. There are
_
k
l
_
vectors in F
k
2
which agree
with x in exactly l positions. Therefore
_
k
l
_
=

n
i
_
kw
li
_
, i. e.
(1 +z)
k
= (1 +z)
kw

n
i
z
i
.
This proves that n
i
=
_
w
i
_
.
(3) The sum of the distances between pairs of rows of C is k(
bw
2k
)
2
by (1). Since any two rows have distance at least 1, this sum is at
least
_
b
2
_
.
(4) Consider a row of C. Count vectors in F
k
2
which have zeros
in the positions where the row has stars. Each row with a dierent
star pattern represents an even number of such vectors whereas a
row with the same star pattern represents exactly one such vector.

Note that property (1) in Theorem 9.7(3) implies that a neces-

sary condition for the existence of an ABD(k, w) is that k divides
w 2
w1
.
The following strengthening of Theorem 9.7(3) is due to A. E.
Brouwer (1999).
Theorem 9.8. Let C be an ABD(k, w) with w > 3.
(1) If two rows of C agree in all but one position, then
_
w
2
_
k;
(2) otherwise w
2
> 2k.
86 A Course in Combinatorics
Proof: Suppose c
1
and c
2
are two rows of C which dier only
in position one. Then all the other rows of C must dier from c
1
in some other position. So, by (i) of the denition and Theorem
9.7(3), we nd
b 2 (w 1)
bw
2k
.
To prove the assertion, we must show that the right-hand side of
the inequality cannot be equal to b 2 or b 1. In both cases,
equality would imply 2
w1
[k which contradicts Theorem 9.7 unless
w = 4, which is excluded by substitution.
(ii) Consider two rows of C which have the same star pattern.
By hypothesis, they dier in more than one position. Again, count
the sum of the distances of all the rows from one of this pair. This
sum is at least 2 + (b 2) = b and, by Theorem 9.7.1, it is equal
to w (bw)/(2k). So w
2
2k. We must show that equality cannot
hold. By the argument above, equality would imply that rows with
the same star pattern occur in pairs which have distance 2, and
furthermore all the other rows have distance 1 to each row of such
a pair. Without loss of generality, such a pair would be
( 00 . . . 000) and ( 00 . . . 011).
The bw/(2k) 1 other rows ending in a 1 would have to end in
01, for otherwise they would have distance 0 to the second row or
distance > 1 to the rst row. Similarly, there would be bw/(2k) 1
rows ending in 10. Since we now have rows with distance 2, we
nd that necessarily bw/(2k) 1 = 1. Therefore 2
w
= 2w, which is
impossible if w 3.
Corollary. An ABD(8, 4) does not exist.
Using these results, it is easy to nd all ABD(k, w) with w 4.
Of course, w = 0 is trivial. For w = 1, 2, or 4, we must have k = w
(no stars). If w = 3, then either k = 3 (no stars) or k = 4. In that
case there are two types of ABD, one given in Example 9.1.
Problem 9D. Construct an ABD(4, 3) that has the same rst
four rows as Example 9.1 but diers in the others.
In 1987 La Poutre and Van Lint proved that an ABD(10, 5)
does not exist but the smaller possibility ABD(8, 5) turned out to
9. Two (0,1,*) problems 87
be quite dicult to handle. In 1999 D. E. Knuth asked Brouwer
whether any progress had been made in this area since the early
results and this made Brouwer decide it was time the question was
settled. An example can be found in Brouwer (1999). It does not
seem to have any structure. Now the smallest open case is the
question whether an ABD(12, 6) exists.
We shall now describe some construction methods. Some of the
ideas will be used in other chapters.
Theorem 9.9. If an ABD(k
i
, w
i
) exists for i = 1, 2, then an
ABD(k
1
k
2
, w
1
w
2
) exists.
Proof: We can assume w
2
> 0. Partition the rows of ABD(k
2
, w
2
)
into two classes R
0
and R
1
of equal size. In ABD(k
1
, w
1
) we replace
each star by a row of k
2
stars, each 0 by a row from R
0
and each 1
by a row from R
1
in all possible ways. A trivial calculation shows
that the resulting matrix is an ABD(k
1
k
2
, w
1
w
2
).
Corollary. An ABD(4
t
, 3
t
) exists.
For the proof of the next theorem, we introduce a new symbol,
namely . A k-tuple consisting of the symbols 0, 1, , and rep-
resents all possible words with only 0, 1, that can be obtained by
replacing each by a 0 or a 1 in all possible ways.
Theorem 9.10. Let w > 0. Suppose an ABD(k, w) exists, where
k = k
0
2
l
, k
0
odd. Then an ABD(k, w + ik
0
) exists for 0 i
(k w)/k
0
.
Proof: It is sucient to consider i = 1. Let C be the ABD(k, w).
Dene a matrix A of the same size by requiring a
ij
= 1 if C
ij
=
and a
ij
= 0 otherwise. By Theorem 7.3, A is the sum of two
matrices A
1
and A
2
, where A
1
has k
0
ones in each row and 2
wl
ones in each column. In a row of C, replace stars by if the star
occurs in a position where A
1
has a one. This produces the required
ABD(k, w +k
0
).
Theorem 9.11. If ABD(k, w) exists and 1 is a number such
that k and w are integers, then an ABD(k, w) exists.
Proof: It is sucient to show that ABD(k +l, w +m) exists for
(k + l)/(w + m) = k/w and (l, m) = 1. Let k = k
0
2
e
, k
0
odd.
88 A Course in Combinatorics
From (ii) of the denition we know that k
0
[w. Therefore wl = mk
and (l, m) = 1 imply that l is a power of 2. Consider the l l
circulant matrix with a row of l m stars and m minus signs as
rst row. Since l divides b we can adjoin a column of b/l copies
of this circulant to the matrix C of the ABD(k, w). It is easy to
check that this larger matrix is an ABD(k +l, w +m).
Example 9.2. From the corollary to Theorem 9.9 we have an
ABD(64, 27). Theorem 9.10 then shows that an ABD(64, w) exists
for 27 w 64. In particular, there is an ABD(64, 32). Then
Theorem 9.11 implies that an ABD(2w, w) exists for all w 32.
As mentioned before, nonexistence has been shown for w = 4 and
w = 5 and the case w = 6 is still open.
Notes.
The rst problem considered in this chapter was introduced by
J. R. Pierce at Bell Laboratories as the loop switching problem.
Several people (including one of the present authors) tried in vain
to solve it. Shortly after R. L. Graham raised the reward for the
solution to $200, it was solved by P. Winkler. It is worth noting that
Winkler stated that the idea of numbering the vertices as was done
in the proof was a regular habit due to his background in computer
science. Going over the proof, one sees that this numbering indeed
played a crucial role.
References.
A. E. Brouwer (1999), An Associative Block Design ABD(8, 5),
SIAM J. Comput. 28, 19701971.
R. L. Graham and H. O. Pollak (1971), On the addressing problem
for loop switching, Bell System Tech. J. 50, 24952519.
B. W. Jones (1950), The Theory of Quadratic Forms, Carus Math.
Monogr. 10, Math. Assoc. of America.
J. A. La Poutre and J. H. van Lint (1985), An associative block
design ABD(10, 5) does not exist, Utilitas Math. 31, 219225.
P. Winkler (1983), Proof of the squashed cube conjecture, Combi-
natorica 3, 135139.
10
The principle of inclusion
and exclusion;
inversion formulae
As we have seen in several of the previous chapters, many problems
of combinatorial analysis involve the counting of certain objects.
We now treat one of the most useful methods for counting. It
is known as the principle of inclusion and exclusion. The idea is
as follows. If A and B are subsets of S and we wish to count
the elements of SA B, then the answer is not [S[ [A[ [B[
because the elements of AB have been subtracted twice. However
[S[[A[[B[+[AB[ is correct. The following theorem generalizes
this idea.
Theorem 10.1. Let S be an N-set; E
1
, . . . , E
r
not necessarily
distinct subsets of S. For any subset M of 1, . . . , r, we dene
N(M) to be the number of elements of S in

iM
E
i
and for 0
j r, we dene N
j
:=

[M[=j
N(M). Then the number of elements
of S not in any of the subsets E
i
, 1 i r, is
(10.1) N N
1
+N
2
N
3
+ + (1)
r
N
r
.
Proof: (i) If x S and x is in none of the E
i
, then x contributes
1 to the expression (10.1).
(ii) If x S and x is in exactly k of the sets E
i
, then the contri-
bution to (10.1) equals
1
_
k
1
_
+
_
k
2
_
+ (1)
k
_
k
k
_
= (1 1)
k
= 0.

90 A Course in Combinatorics
Remark. If we truncate the sum in (10.1) after a positive (respec-
tively, negative) term, then we have an upper (respectively, lower)
bound for the number of elements of S not in any of the E
i
.
Because this method is of great importance, we shall give several
examples as illustration.
Example 10.1. Let d
n
denote the number of permutations of
1, 2, . . . , n such that (i) ,= i for all i (these are called derange-
ments). Let S := S
n
, and let E
i
be the subset of those permutations
with (i) = i. By (10.1) we nd
(10.2) d
n
=
n

i=0
(1)
i
_
n
i
_
(n i)! = n!
n

i=0
(1)
i
i!
.
From this formula, we see that for large values of n the probability
that a permutation is a derangement is nearly e
1
. From (10.2) for
n and n 1, we nd a recursion formula for d
n
:
(10.3) d
n
= nd
n1
+ (1)
n
.
The formula (10.2) can also be obtained by inversion as follows.
Consider the power series D(x) :=

n=0
d
n
x
n
n!
(d
0
= 1). Now if
F(x) := e
x
D(x), then
F(x) =

m=0
_
m

r=0
_
m
r
_
d
mr
_
x
m
m!
and since

m
r=0
_
m
r
_
d
mr
= m!, we nd F(x) = (1 x)
1
. It follows
that D(x) = e
x
(1 x)
1
and by multiplying the power series for
the two factors, we nd (10.2) again.
Example 10.2. Let X be an n-set, Y = y
1
, . . . , y
k
a k-set. We
count the surjections of X to Y . Let S be the set of all mappings
from X to Y , E
i
the subset of mappings for which y
i
is not in the
image of X. By (10.1) we nd the number of surjections to be

k
i=0
(1)
i
_
k
i
_
(k i)
n
. Now this number is trivially 0 if k > n and
clearly n! if k = n. So we have proved:
(10.4)
k

i=0
(1)
i
_
k
i
_
(k i)
n
=
_
n! if k = n,
0 if k > n.
10. Inclusion-exclusion; inversion formulae 91
There are many formulae like (10.4) that are often quite hard to
prove directly. The occurrence of (1)
i
is usually a sign that count-
ing the right kind of objects using the principle of inclusion and ex-
clusion can produce the formula, as in this example. Nevertheless
it is useful in this case to see another proof.
Let P(x) be a polynomial of degree n, with highest coecient a
n
.
We denote the sequence of values P(0), P(1), . . . by P. We now
consider the sequence of dierences P(1) P(0), P(2) P(1), . . . .
This is Q
1
, where Q
1
(x) := P(x + 1) P(x), a polynomial of
degree n 1 with highest coecient na
n
. By repeating this pro-
cedure a number of times, we nd a sequence Q
k
whose terms are

k
i=0
(1)
i
_
k
i
_
P(x +k i), corresponding to the polynomial Q
k
(x)
of degree n k with highest coecient n(n 1) . . . (n k + 1)a
n
.
If k = n, then all the terms of Q
k
are n!a
n
and if k > n, then they
are all 0. Take P(x) = x
n
. We again nd (10.4).
Example 10.3. The following identity is a well known relation
between binomial coecients:
(10.5)
n

i=0
(1)
i
_
n
i
__
m+n i
k i
_
=
_
_
m
k
_
if m k,
0 if m < k.
We see that if we wish to prove this using inclusion-exclusion, then
the sets E
i
that we wish to exclude involve choosing from an n-
set, and after choosing i of them, we must choose k i elements
from some set of size m + n i. This shows us that the following
combinatorial problem will lead us to the result (10.5). Consider a
set Z = X Y , where X = x
1
, . . . , x
n
is an n-set of blue points
and Y is an m-set of red points. How many k-subsets consist of red
points only? The answer is trivially the right-hand side of (10.5).
If we take S to be all the k-subsets of Z and E
i
those k-subsets
that contain x
i
, then (10.1) gives us the left-hand side of (10.5).
Again we can ask whether this result can be proved directly. The
answer is yes. To do this, we use the following expansion:
(10.6)

j=0
_
a +j
j
_
x
j
= (1 x)
a1
.
92 A Course in Combinatorics
Note that (1)
i
_
n
i
_
is the coecient of x
i
in the expansion of (1
x)
n
. From (10.6) we nd that
_
m+ni
ki
_
is the coecient of x
ki
in
the expansion of (1 x)
kmn1
. So the left-hand side of (10.5) is
the coecient of x
k
in the expansion of (1 x)
km1
. If m k 1,
this is obviously 0 and if m k, it is
_
m
k
_
, again by (10.6).
Example 10.4. (The Euler function) Let n = p
a
1
1
p
a
2
2
. . . p
a
r
r
be a
positive integer. We denote by (n) the number of integers k with
1 k n such that the g.c.d. (n, k) = 1. We apply Theorem 10.1
with S := 1, 2, . . . , n and E
i
the set of integers divisible by p
i
,
1 i r. Then (10.1) yields
(10.7) (n) = n
r

i=1
n
p
i
+

1i<jr
n
p
i
p
j
= n
r

i=1
(1
1
p
i
).
The next theorem is used quite often.
Theorem 10.2.

d[n
(d) = n.
Proof: Consider 1, 2, . . . , n = N. For each m N, we have
(m, n)[n. The number of integers m with (m, n) = d, i.e. m = m
1
d,
n = n
1
d and (m
1
, n
1
) = 1 clearly equals (n
1
) = (n/d). So
n =

d[n
(n/d) which is equivalent to the assertion.
At this point, it is useful to introduce the so-called Mobius func-
tion:
(10.8) (d) :=
_
_
_
1 if d = product of an even number of distinct primes,
1 if d = product of an odd number of distinct primes,
0 otherwise, i.e. d not squarefree.
Theorem 10.3.

d[n
(d) =
_
1 if n = 1,
0 otherwise.
Proof: If n = 1, there is nothing to prove. If n = p
a
1
1
. . . p
a
r
r
, then
10. Inclusion-exclusion; inversion formulae 93
by (10.8) we have

d[n
(d) =
r

i=0
_
r
i
_
(1)
i
= (1 1)
r
= 0.

Note how similar the proofs of Theorems 10.1 and 10.3 are.
Using the M obius function, we can reformulate (10.7) as follows:
(10.9)
(n)
n
=

d[n
(d)
d
.
Problem 10A. How many positive integers less than 1000 have
no factor between 1 and 10?
Problem 10B. How many monic polynomials of degree n are
there in F
p
[x] that do not take on the value 0 for x F
p
?
Problem 10C. Determine

nx
(n)
x
n
|.
Problem 10D. One of the most famous functions in complex
analysis is the so-called Riemann -function (s) :=

n=1
n
s
, de-
ned in the complex plane for Re(s)> 1. Prove that 1/(s) =

n=1
(n)n
s
.
Problem 10E. Let f
n
(z) be the function that has as its zeros all
numbers for which
n
= 1 but
k
,= 1 for 1 k < n. Prove that
f
n
(z) =

k[n
(z
k
1)
(n/k)
.
Theorem 10.3 makes it possible to derive a very useful inversion
formula known as the Mobius inversion formula.
Theorem 10.4. Let f(n) and g(n) be functions dened for every
positive integer n satisfying
(10.10) f(n) =

d[n
g(d).
94 A Course in Combinatorics
Then g satises
(10.11) g(n) =

d[n
(d)f(
n
d
).
Proof: By (10.10) we have

d[n
(d)f(
n
d
) =

d[n
(
n
d
)f(d)
=

d[n
(
n
d
)

[d
g(d
/
) =

[n
g(d
/
)

m[(n/d

)
(m).
By Theorem 10.3 the inner sum on the right-hand side is 0 unless
d
/
= n.
Remark. The equation (10.11) also implies (10.10).
Example 10.5. We shall count the number N
n
of circular se-
quences of 0s and 1s, where two sequences obtained by a rotation
are considered the same. Let M(d) be the number of circular se-
quences of length d that are not periodic. Then N
n
=

d[n
M(d).
We observe that

d[n
dM(d) = 2
n
since this counts all possible cir-
cular sequences. By Theorem 10.4 we nd from this equation that
nM(n) =

d[n
(d)2
n/d
and therefore
(10.12) N
n
=

d[n
M(d) =

d[n
1
d

l[d
(
d
l
)2
l
=

l[n
2
l
l

k[
n
l
(k)
k
=
1
n

l[n
(
n
l
)2
l
.
The nal expression has the advantage that all the terms are posi-
tive. This raises the question whether we could have obtained that
expression by some other counting technique. We shall see that
the following theorem, known as Burnsides lemma (although the
theorem is actually due to Cauchy and Frobenius; see the notes),
provides the answer.
10. Inclusion-exclusion; inversion formulae 95
Theorem 10.5. Let G be a permutation group acting on a set X.
For g G let (g) denote the number of points of X xed by g.
Then the number of orbits of G is equal to
1
[G[

gG
(g).
Proof: Count pairs (g, x), where g G, x X, x
g
= x. Starting
with g, we nd

gG
(g). For each x X there are [G[/[O
x
[
such pairs, where O
x
is the orbit of x. So the total number equals
[G[

xX
1/[O
x
[. The orbits of G partition X, and if we sum the
terms 1/[O
x
[ over all x in a particular orbit, we obtain 1. Thus

xX
1/[O
x
[ is the number of orbits.
Example 10.5 (continued). Let G be the cyclic group of order
n, i.e. the group of rotations of a circular sequence of 0s and 1s.
If d[n there are (n/d) integers g such that (n, g) = d and for each
such g there are 2
d
circular sequences that are xed by the rotation
over g positions. So Theorem 10.5 immediately yields the result
(10.12).
Example 10.6. The following problem, introduced by Lucas in
1891, is known as the probl`eme des menages. We wish to seat n
couples at a circular table so that men and women are in alternate
places and no husband will sit on either side of his wife. In how
many ways can this be done? We assume that the women have
been seated at alternate places. Call the ladies 1 to n and the
corresponding men also 1 to n. The problem amounts to placing
the integers 1 to n on a circle with positions numbered 1 to n such
that for all i the integer i is not in position i or position i + 1
(mod n). Let E
i
be the set of seatings in which husband i is sitting
next to his wife. We now wish to use inclusion-exclusion and we
must therefore calculate in how many ways it is possible to seat r
husbands incorrectly. Call this number A
r
. We do this as follows.
Consider a circular sequence of 2n positions. Put a 1 in position
2i 1 if husband i is sitting to the right of his wife; put a 1 in
position 2i if he is sitting to the left of his wife. Put zeros in
the remaining positions. The congurations that we wish to count
therefore are circular sequences of 2n zeros and ones, with exactly r
ones, no two adjacent. Let A
/
r
be the number of sequences starting
with a 1 (followed by a 0). By considering 10 as one symbol, we
see that we must choose r 1 out of 2nr 1 positions. To count
96 A Course in Combinatorics
the number A
//
r
of sequences starting with a 0, we place the 0 at the
end, and then it amounts to choosing r out of 2nr places. Hence
A
r
= A
/
r
+A
//
r
=
_
2n r 1
r 1
_
+
_
2n r
r
_
=
2n
2n r
_
2n r
r
_
.
By (10.1) we nd that the number of ways to seat the men is
(10.13)
n

r=0
(1)
r
(n r)!
_
2n r
r
_
2n
2n r
.
Problem 10F. We color the integers 1 to 2n red or blue in such
a way that if i is red then i 1 is not blue. Prove that
n

k=0
(1)
k
_
2n k
k
_
2
2n2k
= 2n + 1.
Can you prove this identity directly?
Problem 10G. Count the number of permutations x
1
, x
2
, . . . , x
2n
of the integers 1 to 2n such that x
i
+ x
i+1
,= 2n + 1 for i =
1, 2, . . . , 2n 1.
Problem 10H. Prove that for 0 k n
k

i=0
_
k
i
_
D
ni
=
nk

j=0
(1)
j
_
n k
j
_
(n j)!.
Notes.
The principle of inclusion and exclusion occurred as early as 1854
in a paper by Da Silva and later in a paper by Sylvester in 1883.
For this reason (10.1) and similar formulae are sometimes called the
formula of Da Silva, respectively Sylvester. A better name that is
also often used is sieve formula. The formula is indeed an example
of a principle that is used extensively in number theory, referred to
as sieve methods. An example that is probably familiar to most
readers is the sieve of Eratosthenes: to nd the primes n
2
, take
the integers n
2
and sieve out all the multiples of primes n.
10. Inclusion-exclusion; inversion formulae 97
The derangements treated in Example 10.1 occur again in Exam-
ple 14.1 and Example 14.10. The rst occurrence of this question
is in one of the early books on games of chance: Essai danalyse
sur les jeux de hazard by P. R. de Montmort (16781719). It is
still often referred to by the name that he gave it: probl`eme des
rencontres. Formula (10.2) is sometimes stated as follows. If n
persons check their umbrellas (a typical Dutch example; its always
raining in Holland) and subsequently pick one at random in the
dark after a power failure, then the probability that nobody gets
his own umbrella is roughly e
1
(if n is large).
The second proof in Example 10.2 is an example of the use of
calculus of nite dierences, used extensively in numerical analy-
sis.
A. F. M obius (17901868) was an astronomer (and before that an
assistant to Gauss) who made important contributions to geometry
and topology (e.g. the M obius band).
G. F. B. Riemann (18261866) was professor in G ottingen, where
he also obtained his doctorate under Gauss. He is famous for many
of his ideas, which include the Riemann integral, Riemann surfaces
and manifolds, and of course the so-called Riemann hypothesis on
the location of the zeros of the -function. One wonders what he
would have left us if he had not died so young.
In most books in which it occurs, Theorem 10.5 is called Burn-
sides lemma. This is just one of many examples of theorems, etc.
attributed to the wrong person. For a history of this misnomer, we
refer to Neumann (1979).
F. E. A. Lucas (18421891) was a French number theorist. He
is known for his books on number theory and mathematical recre-
ations. The former book contained the problem of Example 10.6.
The Fibonacci numbers were given this name by Lucas. See the
notes to Chapter 14.
References.
F. E. A. Lucas (1891), Theorie des nombres, Gauthier-Villars,
Paris.
P. M. Neumann (1979), A lemma that is not Burnsides, Math.
Scientist, 4, 133141.
11
Permanents
Before introducing the main topic of this chapter, we present a
generalization of Theorem 10.1. As in Theorem 10.1, let S be an
n-set, E
1
, . . . , E
r
(not necessarily distinct) subsets of S. Let F be
any eld. To each element a S, we assign a weight w(a) in F.
For any subset M of 1, 2, . . . , r, we dene W(M) to be the sum
of the weights of the elements of S in

iM
E
i
. For 0 j r, we
dene W
j
:=

[M[=j
W(M) (so W
0
=

aS
w(a)).
Theorem 11.1. If E(m) denotes the sum of the weights of the
elements of S that are contained in exactly m of the subsets E
i
,
1 i r, then
(11.1) E(m) =
rm

i=0
(1)
i
_
m+i
i
_
W
m+i
.
Proof: The proof is nearly the same as for Theorem 10.1. If
x S and x is contained in exactly m of the subsets E
i
, then the
contribution of x to the sum in (11.1) is w(x). If x S and x is
contained in exactly m+k of the subsets E
i
, then the contribution
to the sum equals
w(x)
k

i=0
(1)
i
_
m+i
i
__
m+k
m+i
_
= w(x)
_
m+k
k
_
k

i=0
(1)
i
_
k
i
_
= 0.

11. Permanents 99
We now give the denition of a permanent. Let A = (a
1
, . . . , a
n
)
be an n n matrix with columns a
j
= (a
1j
, . . . , a
nj
)

. Then per
A, the permanent of A, is dened by
(11.2) per A :=

S
n
a
1(1)
a
n(n)
.
So the permanent is dened in the same way as the determinant
but without the signs depending on whether the permutation is
even or odd.
From the denition, the following properties of the permanent
are obvious.
(11.3) per A = per A

;
(11.4)
if P and Q are permutation matrices, then per A = per PAQ;
(11.5) per A is a linear function of a
j
, 1 j n.
Of course per A is also a linear function of each of the rows of
A. The permanent of A is much more dicult to calculate than
its determinant. However, it is clear from (11.2) that expansion by
rows or columns is possible. So dene A
ij
to be the matrix obtained
from A by deleting row i and column j. Then
(11.6) per A =
_

n
i=1
a
ij
per A
ij
, 1 j n,

n
j=1
a
ij
per A
ij
, 1 i n.
The following method of calculating a permanent (due to Ryser) is
an application of Theorem 11.1.
Theorem 11.2. Let A be an n n matrix. If A
r
is obtained from
A by deleting r columns, then S(A
r
) denotes the product of the
row-sums of A
r
. We dene
r
to be the sum of the values of S(A
r
)
for all possible choices of A
r
. Then
(11.7) per A =
n1

r=0
(1)
r

r
.
100 A Course in Combinatorics
Proof: Let S be the set of all products p = a
1i
1
. . . a
ni
n
and dene
w(p) := p. Dene E
j
to be the set of products p for which j /
i
1
, . . . , i
n
. Then the permanent of A is the sum of the weights of
the elements of S that are not in any of the subsets E
j
. So (11.7)
is an immediate consequence of (11.1).
Problem 11A. Prove (10.4) using Theorem 11.2.
Remark. If A
1
, . . . , A
n
are subsets of 1, . . . , n and a
ij
= 1 if
j A
i
, 0 otherwise, then per A counts the number of SDRs of the
sets A
1
, . . . , A
n
.
Example 11.1. We nd another formula for the number of de-
rangements of 1, 2, . . . , n. The permanent of the matrix J I of
size n is clearly d
n
. From (11.7) we nd
(11.8) d
n
=
n1

r=0
(1)
r
_
n
r
_
(n r)
r
(n r 1)
nr
.
By expanding the term (n 1 r)
nr
and applying (10.4) after
changing the order of summation, we nd a complicated proof of
(10.2).
During the 1970s, several well known conjectures on permanents
of (0,1)-matrices were proved, often by ingenious arguments. In
fact, much of the research on permanents was motivated by these
conjectures. Therefore we will devote attention to a number of
these results in this and the next chapter. As an introduction, we
consider (0,1)-matrices with two ones in each row and column.
Theorem 11.3. If A is a (0, 1)-matrix in which all row-sums and
column-sums are 2, then
per A 2

1
2
n|
.
Proof: Consider the graph G whose vertices correspond to the
rows of A, whose edges correspond to the columns of A, and where
vertex i and edge j are incident exactly when A(i, j) = 1. This
graph is regular of degree 2, and hence is the disjoint union of
polygons. The submatrix corresponding to the vertices and edges
11. Permanents 101
of a polygon is, after reordering rows and columns if necessary, a
circulant
_
_
_
_
_
_
1 1 0 0 0 0
0 1 1 0 0 0
0 0 1 1 0 0
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
0 0 0 0 1 1
1 0 0 0 0 1
_
_
_
_
_
_
.
(This may degenerate into a 2 by 2 matrix of 1s.) The matrix A
is the direct sum of such matrices, each of which has permanent
2. The number of factors is at most
1
2
n| and we see that equality
holds in the theorem if A is the direct sum of
1
2
n| matrices J of
size 2.
This elementary theorem is concerned with the relationship be-
tween the row-sums of a matrix and the permanent of that matrix,
and the same is true for many of the following theorems. This
brings us to the rst dicult question. It was conjectured by H.
Minc in 1967 that if A is a (0,1)-matrix with row-sums r
1
, . . . , r
n
,
then
(11.9) per A
n

j=1
(r
j
!)
1/r
j
.
Observe that the proof of Theorem 11.3 shows that equality can
hold in (11.9) and in fact, we have equality if A is the direct sum of
matrices J
m.
Several results that were weaker than(11.9) were proved,
often by intricate and long arguments. The conjecture was nally
proved in 1973 by L. M. Bregman. All the more surprising is the fact
that A. Schrijver came up with an extremely elegant and very short
proof of Mincs conjecture in 1977. The proof depends on the
following lemma.
Lemma 11.4. If t
1
, t
2
, . . . , t
r
are nonnegative real numbers, then
_
t
1
+ +t
r
r
_
t
1
++t
r
t
t
1
1
t
t
r
r
.
Proof: Since xlog x is a convex function, we have
t
1
+ +t
r
r
log
_
t
1
+ +t
r
r
_

t
1
log t
1
+. . . t
r
log t
r
r
,
102 A Course in Combinatorics
which proves the assertion.
In the following, we use (11.6) in the following form:
per A =

k,a
ik
=1
per A
ik
.
Theorem 11.5. Let A be an n n (0, 1)-matrix with r
i
ones in
row i, 1 i n. Then
per A
n

i=1
(r
i
)!
1/r
i
.
Proof: The proof is by induction on n. For n = 1, the theorem is
trivial. We assume that the theorem is true for matrices of size n1.
The idea is to estimate (per A)
nper A
and to split this expression
into several products. Now note that r
i
is the number of values of
k for which a
ik
= 1 and apply the lemma. We nd:
(11.10) (per A)
nper A
=
n

i=1
(per A)
per A

i=1
_
_
r
per A
i

k,a
ik
=1
per A
per A
ik
ik
_
_
.
Now, let S denote the set of all permutations of 1, . . . , n for
which a
i
i
= 1 for i = 1, . . . , n. So [S[ = per A. Furthermore,
the number of S such that
i
= k is per A
ik
if a
ik
= 1 and 0
otherwise. So the right-hand side of (11.10) is equal to
(11.11)

S
__
n

i=1
r
i
_

_
n

i=1
per A
i
i
__
.
We now apply the induction hypothesis to each A
i
i
. This yields
(11.12) (per A)
nperA

S
_

_
_
n

i=1
r
i
_

i=1
_

j=i,
a
j
i
=0
(r
j
!)
1/r
j

j=i,
a
j
i
=1
((r
j
1)!)
1/(r
j
1)
_

_
_

_
.
11. Permanents 103
Since the number of i such that i ,= j and a
j
i
= 0 is nr
j
, and the
number of i such that i ,= j and a
j
i
= 1 is r
j
1, we can replace
the right-hand side of (11.12) by

S
__
n

i=1
r
i
_

_
n

j=1
(r
j
!)
(nr
j
)/r
j
(r
j
1)!
__
=

S
n

i=1
(r
i
)!
n/r
i
=
_
n

i=1
(r
i
!)
1/r
i
_
nper A
and the assertion is proved.
We now shall consider a special class of (0,1)-matrices, namely
the (0,1)-matrices that have exactly k ones in each row and column.
We denote this class by /(n, k). We dene:
(11.13) M(n, k) := maxper A : A /(n, k),
(11.14) m(n, k) := minper A : A /(n, k).
By taking direct sums, we nd the following inequalities:
(11.15) M(n
1
+n
2
, k) M(n
1
, k)M(n
2
, k),
(11.16) m(n
1
+n
2
, k) m(n
1
, k)m(n
2
, k).
These two inequalities allow us to introduce two more functions
using the following result, known as Feketes lemma.
Lemma 11.6. Let f : N N be a function for which f(m+n)
f(m)f(n) for all m, n N. Then lim
n
f(n)
1/n
exists (possibly
).
Proof: Fix m and x l, l m. By induction, we nd from the
inequality for f that f(l +km) f(l)[f(m)]
k
. Therefore
liminf f(l +km)
1/(l+km)
f(m)
1/m
104 A Course in Combinatorics
and since there are m possible values for l, we in fact have
liminf f(n)
1/n
f(m)
1/m
.
Now let m . We nd that
liminf f(n)
1/n
limsupf(m)
1/m
,
but then these are equal.
The assertion of the lemma is also true if in the inequality for f
we replace by . By applying the lemma to (11.15) and (11.16),
we can dene:
(11.17) M(k) := lim
n
M(n, k)
1/n
,
(11.18) m(k) := lim
n
m(n, k)
1/n
.
Problem 11B. Prove that M(n, k) k!. Prove that M(k)
(k!)
1/k
. Show by example that M(k) (k!)
1/k
. This shows that
M(k) = (k!)
1/k
.
The function m(n, k) is much more dicult to handle. What we
should expect is based on a famous problem still referred to as the
Van der Waerden conjecture, although in 1981 two dierent proofs
of the conjecture appeared (after nearly 50 years of research on this
question!). We formulate the conjecture below, and the proof will
be given in the next chapter.
Conjecture. If A is an n n matrix with nonnegative entries in
which all row-sums and column-sums are 1, then
(11.19) per A n! n
n
.
The matrices considered in the conjecture are usually called dou-
bly stochastic matrices. If A /(n, k), then dividing all the ele-
ments of A by k yields a doubly stochastic matrix. Therefore the
conjecture (now a theorem) shows that m(k) k/e. This is re-
markable because the value of M(k) given in Problem 11B tends
11. Permanents 105
to k/e for k (see the notes). This means that for large n,
and A an arbitrary element of /(n, k), the value of (per A)
1/n
is
nearly k/e. For a long time, the best lower bound for m(n, 3) was
n + 3 and even that was not easy to prove. Once again, the next
improvement was both considerable and elementary. We now give
the proof of that result, due to Voorhoeve (1979).
Theorem 11.7. m(n, 3) 6 (
4
3
)
n3
.
Proof: Let U
n
denote the set of n n matrices with nonnegative
integers as entries and all row-sums and column-sums 3; u(n) :=
minper A : A U
n
. Denote by V
n
the set of all matrices obtained
from elements of U
n
by decreasing one positive entry by 1; v(n) :=
minper A : A V
n
. We rst show that
(11.20) u(n)
_
3
2
v(n)
_
.
Let A be an element of U
n
with rst row a = (
1
,
2
,
3
, 0, . . . , 0),
where
i
0 for i = 1, 2, 3. Since
2a =
1
(
1
1,
2
,
3
, 0, . . . , 0) +
2
(
1
,
2
1,
3
, 0, . . . , 0)
+
3
(
1
,
2
,
3
1, 0, . . . , 0),
we nd from (11.5) that 2u(n) (
1
+
2
+
3
)v(n) = 3v(n),
proving the assertion.
Next, we show that
(11.21) v(n)
_
4
3
v(n 1)
_
.
We must distinguish between two cases. In the rst one, A is an
element of V
n
with rst row (1, 1, 0, . . . , 0) and the matrix obtained
from A by deleting the rst row has the form (c
1
, c
2
, B). The
column-sum of c
3
:= c
1
+c
2
is either 3 or 4. By (11.6), we have
per A = per (c
1
, B) + per (c
2
, B) = per (c
3
, B).
If the column-sum of c
3
is 3, then the matrix (c
3
, B) is in U
n1
and
we are done by (11.20). If the sum is 4, then we use the same trick as
106 A Course in Combinatorics
above: write 3c
3
as a linear combination of four vectors d
i
such that
each matrix (d
i
, B) is in V
n1
and we nd that 3 per A 4v(n1).
The second case that we have to consider is that A has (2, 0, . . . , 0)
as rst row. If we delete the rst row and column of A, then there
are again two possibilities. We obtain a matrix B that is either in
U
n1
or in V
n1
. So we have per A 2 minu(n1), v(n1) and
we are done by (11.20). By combining (11.20) and (11.21) with the
trivial value v(1) = 2, the assertion of the theorem follows.
We now consider a larger class of n n matrices, namely those
with nonnegative integers as entries and all row-sums and column-
sums equal to k. We denote this class by (n, k) and the minimal
permanent within the class by (n, k). Again we have (m+n, k)
(m, k)(n, k) and by Feketes lemma, we can dene
(11.22) (k) := lim
n
((n, k))
1/n
.
From Theorem 11.3 and Theorem 11.7, we know that (n, 2) = 2
and (n, 3) 6 (
4
3
)
n3
. From (11.19), we have seen above that
(n, k) n!(
k
n
)
n
. We have also seen that there is a connection
between permanents and SDRs. We now show a proof in which
this connection is exploited.
Theorem 11.8. (n, k) k
2n
/
_
nk
n
_
.
Proof: We denote by P
n,k
the collection of all ordered partitions
of the set 1, 2, . . . , nk into classes of size k. We have
(11.23) p
n,k
:= [P
n,k
[ =
(nk)!
(k!)
n
.
Now let / := (A
1
, . . . , A
n
) be such a partition. The number of
SDRs of the subsets A
1
, . . . , A
n
is k
n
. Consider a second partition
B := (B
1
, . . . , B
n
). We denote by s(/, B) the number of common
SDRs of / and B. We dene an n n matrix A with entries
ij
by
ij
:= [A
i
B
j
[. The point of the proof is the fact that per A
counts the number of common SDRs of / and B. Furthermore, by
denition of the partitions, the matrix A is in (n, k). Therefore
s(/, B) = per A (n, k).
11. Permanents 107
If / P
n,k
is given and some SDR of / is given, then there are
n!p
n,k1
ordered partitions B that have this same SDR. Hence we
have

BP
n,k
s(/, B) = k
n
n! p
n,k1
.
Combining this with (11.23) and the inequality for (n, k), we nd
(n, k)
k
n
n!p
n,k1
p
n,k
=
k
2n
_
nk
n
_.

This proof is due to Schrijver and Valiant (1980) who also gave
the following corollary.
Corollary. (k)
(k1)
k1
k
k2
.
Proof: This follows in the usual way from the previous theorem
by using Stirlings formula: n! n
n
e
n
(2n)
1/2
.
The corollary combined with Theorem 11.7 gives us one more
value of (k), namely (3) =
4
3
.
Problem 11C. Consider the set of integers 1, 2, . . . , 64. We rst
remove the integers 1 (mod 9), i.e. x
1
= 1, . . . , x
8
= 64. Then
we remove the integers x
i
+ 8, where 72 is to be interpreted as
8. This leaves us with a set S of 48 elements. We partition S
into subsets A
1
, . . . , A
8
and also into subsets B
1
, . . . , B
8
, where A
i
contains integers in the interval (8(i 1), 8i] and B
i
contains the
integers i (mod 8). How many common SDRs are there for the
systems A
1
, . . . , A
8
and B
1
, . . . , B
8
?
Problem 11D. We return to Problem 5G. Again, consider a bi-
partite graph on 2n vertices that is regular of degree 3. Give a
lower bound for the number of perfect matchings.
Problem 11E. On a circular arrangement of the integers 1, 2, . . . , n
consider the subsets i, i + 1, i + 2 where i = 1, 2, . . . , n and inte-
gers are interpreted mod n. How many SDRs does this collection
of sets have?
108 A Course in Combinatorics
Notes.
In his book Permanents, H. Minc (1978) mentions that the name
permanent is essentially due to Cauchy (1812) although the word as
such was rst used by Muir in 1882. Nevertheless, a referee of one of
Mincs earlier papers admonished him for inventing this ludicrous
name! For an extensive treatment of permanents, we refer to Mincs
book. There, one can nd much of the theory that was developed
mainly to solve the Van der Waerden conjecture (without success
at the time of writing of the book).
Theorem 11.2 is from Ryser (1963).
For a number of results related to the Minc conjecture, we refer
to Van Lint (1974).
The lemma known as Feketes lemma occurs in Fekete (1923).
For another application, we refer to J. W. Moon (1968).
The term doubly stochastic matrix can be motivated by consid-
ering the entries to be conditional probabilities. However, perma-
nents do not seem to play a role of importance in probability theory.
The remarks concerning m(k) and M(k) preceding Theorem 11.7
are based on Stirlings formula and the related inequality n!
n
n
e
n
. This inequality is easily proved by induction, using the fact
that (1+n
1
)
n
is increasing with limit e. Actually Stirlings formula
was rst given by de Moivre. Stirling derived an asymptotic series
for the gamma function which leads to the estimate
(x) = x
x
1
2
e
x
(2)
1
2
e
/(12x)
,
where 0 < < 1. (n! = (n + 1).)
A. Schrijver (1998) has established equality in the Corollary to
Theorem 11.8 by proving that
(n, k)
_
(k 1)
k1
k
k2
_
n
.
References.
L. M. Bregman (1973), Certain properties of nonnegative matri-
ces and their permanents, Dokl. Akad. Nauk SSSR 211, 2730
(Soviet Math. Dokl. 14, 945949).
11. Permanents 109
M. Fekete (1923),

Uber die Verteilung der Wurzeln bei gewissen
algebraischen Gleichungen mit ganzzahligen Koezienten, Math.
Zeitschr. 17, 228249.
J. H. van Lint (1974), Combinatorial Theory Seminar Eindhoven
University of Technology, Lecture Notes in Mathematics 382,
Springer-Verlag.
H. Minc (1967), An inequality for permanents of (0,1) matrices, J.
Combinatorial Theory 2, 321326.
H. Minc (1978), Permanents, Encyclopedia of Mathematics and
its Applications, vol. 6, Addison-Wesley, reissued by Cambridge
University Press.
J. W. Moon (1968), Topics on Tournaments, Holt, Rinehart and
Winston.
H. J. Ryser (1963), Combinatorial Mathematics, Carus Math.
Monograph 14.
A. Schrijver (1978), A short proof of Mincs conjecture, J. Combi-
natorial Theory (A) 25, 8083.
A. Schrijver and W. G. Valiant (1980), On lower bounds for per-
manents, Proc. Kon. Ned. Akad. v. Wetensch. A 83, 425427.
A. Schrijver (1998), Counting 1-factors in regular bipartite graphs,
J. Combinatorial Theory (B) 72, 122135.
M. Voorhoeve (1979), A lower bound for the permanents of certain
(0,1)-matrices, Proc. Kon. Ned. Akad. v. Wetensch. A 82, 8386.
12
The Van der Waerden
conjecture
In this chapter, we denote the set of all doubly stochastic matrices
of size n
n
. The subset consisting of matrices for which all
entries are positive is denoted by

n
. We dene J
n
:= n
1
J, where
J denotes the n n matrix for which all entries are 1. The vector
(1, 1, . . . , 1)

is denoted by j.
In 1926, B. L. van der Waerden proposed as a problem to deter-
mine the minimal permanent among all doubly stochastic matrices.
It was natural to assume that this minimum is per J
n
= n! n
n
(as
stated in (11.19)). The assertion
(12.1) (A
n
and A ,= J
n
) (per A > per J
n
)
became known as the Van der Waerden conjecture (although in
1969 he told one of the present authors that he had not heard this
name before and that he had made no such conjecture). In 1981 two
dierent proofs of the conjecture appeared, one by D. I. Falikman,
submitted in 1979, and one by G. P. Egoritsjev, submitted in 1980.
We shall give our version of Egoritsjevs proof which had a slightly
stronger result than Falikmans, cf. Van Lint (1981).
In the following, we shall use the term minimizing matrix for a
matrix A
n
such that per A = minper S : S
n
. As usual,
the matrix obtained from A by deleting row i and column j is
denoted by A
ij
. We often consider A as a sequence of n columns and
write A = (a
1
, . . . , a
n
). Later on, we shall consider permanents of
matrices of size n1 but we wish to use the notation for matrices of
size n. The trick is to write per (a
1
, . . . , a
n1
, e
j
), where e
j
denotes
the j-th standard basis vector. This permanent does not change
12. The Van der Waerden conjecture 111
value if the j-th row and n-th column are deleted. We remind the
reader that by Problem 5C (Birkho), the set
n
is a convex set
with the permutation matrices as vertices.
We need a few elementary results on matrices in
n
. The rst
statement is the same as Theorem 5.4.
Theorem 12.1. If A is an n n matrix with nonnegative entries,
then per A = 0 if and only if A contains an s t zero submatrix
such that s +t = n + 1.
We shall call an n n matrix partly decomposable if it contains
a k n k zero submatrix. So A is partly decomposable if there
exist permutation matrices P and Q such that
PAQ =
_
B C
O D
_
,
where B and D are square matrices. If a matrix is not partly
decomposable, then we shall say that it is fully indecomposable. If
A
n
and A is partly decomposable, then, in the representation
given above, we must have C = O, because the sum of the entries
of B equals the number of columns of B and the sum of the entries
of B and C is equal to the number of rows of B. So in that case,
A is the direct sum B D of an element of
k
and an element of

nk
.
Problem 12A. Let A be an n n matrix with nonnegative entries
(n 2). Prove that A is fully indecomposable if and only if per
A
ij
> 0 for all i and j.
Problem 12B. Let A be an n n matrix with nonnegative en-
tries. Prove that if A is fully indecomposable, then AA

and A

A
are also fully indecomposable.
Theorem 12.2. A minimizing matrix is fully indecomposable.
Proof: Let A
n
be a minimizing matrix and suppose that A
is partly decomposable. Then, as we saw above, A = BC, where
B
k
and C
nk
. By Theorem 12.1, we have per A
k,k+1
= 0
and per A
k+1,k
= 0. By Birkhos theorem, we may assume that B
and C have positive elements on their diagonals. In A we replace
112 A Course in Combinatorics
b
kk
by b
kk
and c
11
by c
11
and we put an in the positions
k, k + 1 and k + 1, k. The new matrix is again in
n
, if is small
enough. The permanent of the new matrix is equal to
per A per A
kk
per A
k+1,k+1
+O(
2
).
Since per A
kk
and per A
k+1,k+1
are both positive, this new perma-
nent is smaller than per A if is suciently small. This contradic-
tion proves the assertion.
Corollary. (i) A row of a minimizing matrix has at least two pos-
itive entries.
(ii) For any a
ij
in a minimizing matrix, there is a permutation
such that (i) = j and a
s,(s)
> 0 for 1 s n, s ,= i.
Proof: Clearly (i) is trivial, and (ii) follows from Problem 12A.

Let us now look at how far we can get with calculus. A very
important step in the direction of a proof of (12.1) is the following
surprising result due to Marcus and Newman (1959).
Theorem 12.3. If A
n
is a minimizing matrix and a
hk
> 0,
then per A
hk
= per A.
Proof: Let S be the subset of
n
consisting of the doubly stochas-
tic matrices X for which x
ij
= 0 if a
ij
= 0. Then A is an interior
point of the set S, which is a subset of 1
m
for some m. If we denote
the set of pairs (i, j) for which a
ij
= 0 by Z, we can describe S by
the relations:
n

i=1
x
ij
= 1, j = 1, . . . , n;
n

j=1
x
ij
= 1, i = 1, . . . , n;
x
ij
0, i, j = 1, . . . , n;
x
ij
= 0, (i, j) Z.
Since A is minimizing, the permanent function has a relative mini-
mum in the interior point A of the set S and we can use Lagrange
12. The Van der Waerden conjecture 113
multipliers to describe the situation. So we dene:
F(X) := per X
n

i=1

i
_
n

k=1
x
ik
1
_

j=1

j
_
n

k=1
x
kj
1
_
.
For (i, j) / Z, we have:
F(X)/x
ij
= per X
ij

i

j
.
It follows that per A
ij
=
i
+
j
and from this we nd that for
1 i n
(12.2) per A =
n

j=1
a
ij
per A
ij
=
n

j=1
a
ij
(
i
+
j
) =
i
+
n

j=1
a
ij

j
,
and similarly for 1 j n,
(12.3) per A =
j
+
n

i=1
a
ij

i
.
We introduce the vectors = (
1
, . . . ,
n
)

and = (
1
, . . . ,
n
)

.
From (12.2) and (12.3), we nd
(12.4) (per A)j = +A = +A

.
Multiplying by A

gives us
(per A)j = A

+A

A,
and hence = A

A, and similarly = AA

. The matrices AA

and A

A are both in
n
and by Problem 12B and Theorem 12.2,
they have eigenvalue 1 with multiplicity one corresponding to the
eigenvector j. So we see that both and are multiples of j. By
(12.4), we have
i
+
j
= per A and since per A
ij
=
i
+
j
, we are
nished.
114 A Course in Combinatorics
Remark. It was shown by Marcus and Newman that Theorem
12.3 implies that a minimizing matrix in

n
must be J
n
. The proof
depends on the following idea. Let A be an element of
n
with
the property that per A
hk
= per A for all h, k. If we replace any
column of A by a vector x for which

n
i=1
x
ij
= 1, then the value of
the permanent does not change (by (11.6)). We shall refer to this
idea as the substitution principle. If A is a minimizing matrix in

n
,
then the substitution principle allows us to replace any two columns
of A by their average and thus obtain a new minimizing matrix. In
this way, one constructs a sequence of minimizing matrices which
tends to J
n
. The uniqueness of the minimum takes a little extra
work.
A nal result that uses ideas from calculus is the following gen-
eralization of Theorem 12.3 due to London (1971).
Theorem 12.4. If A
n
is a minimizing matrix, then per A
ij

per A for all i and j.
Proof: Let i and j be given. By Corollary (ii) of Theorem 12.2,
there is a permutation such that (i) = j and a
s,(s)
> 0 for
1 s n, s ,= i. Let P be the corresponding permutation matrix.
For 0 1 we dene f() := per ((1 )A +P). Since A is a
minimizing matrix, f
/
(0) 0, i.e.
0
n

i=1
n

j=1
(a
ij
+p
ij
)per A
ij
= nper A+
n

s=1
per A
s,(s)
.
By Theorem 12.3 we have per A
s,(s)
= per A for s ,= i and therefore
per A
ij
per A.
Problem 12C. Show that Theorem 12.3 implies that if A

5
is a minimizing matrix, then there is a minimizing matrix B

5
that has aJ of size 4 as a principal submatrix. Then show that a
must be
1
5
.
We now come to the main tool in the proof of the Van der Waer-
den conjecture. This time we need linear algebra. We shall give a
direct proof of a theorem on symmetric bilinear forms which leads
12. The Van der Waerden conjecture 115
to the inequality that was derived by Egoritsjev from the so-called
AlexandroFenchel inequalities (which we do not treat).
Consider the space 1
n
with a symmetric inner product x, y) =
x

Qy. If Q has one positive eigenvalue and n 1 negative eigen-

values, we shall speak of a Lorentz space. We use the following
standard terminology: a nonzero vector x is isotropic if x, x) = 0,
positive, respectively negative, if x, x) is positive, respectively neg-
ative.
Every nonzero vector in the (n1)-dimensional span of the neg-
ative eigenvectors is negative, so if a is positive and b is not a
scalar multiple of a, then the plane spanned by a and b must con-
tain a negative vector. Therefore the quadratic form in given by
a +b, a +b) must have a positive discriminant. Thus we have
the following inequality, which is like the Cauchy inequality but the
other way around.
Theorem 12.5. If a is a positive vector in a Lorentz space and b
is arbitrary, then
a, b)
2
a, a)b, b)
and equality holds if and only if b = a for some constant .
The connection with permanents is provided by the following
denition. Consider vectors a
1
, . . . , a
n2
in 1
n
with positive coor-
dinates. As usual, let e
1
, . . . , e
n
be the standard basis of 1
n
. We
dene an inner product on 1
n
by
(12.5) x, y) := per (a
1
, a
2
, . . . , a
n2
, x, y),
i.e.
x, y) = x

Qy,
where Q is given by
(12.6) q
ij
:= per (a
1
, a
2
, . . . , a
n2
, e
i
, e
j
).
Note that if A is a matrix with columns a
1
, . . . , a
n
and we delete
the last two columns and the rows with index i and j, then the
reduced matrix has permanent equal to q
ij
.
116 A Course in Combinatorics
Theorem 12.6. The space 1
n
with the inner product dened by
(12.5) is a Lorentz space.
Proof: The proof is by induction. For n = 2, we have Q =
_
0 1
1 0
_
and the assertion is true. Now assume the theorem is true for 1
n1
.
In the rst step of the proof, we show that Q does not have the
eigenvalue 0. Suppose Qc = 0, i.e.
(12.7) per (a
1
, . . . , a
n2
, c, e
j
) = 0 for 1 j n.
By deleting the last column and the j-th row, we can consider (12.7)
as a relation for vectors in 1
n1
. We consider the inner product
given by
(12.8) per (a
1
, . . . , a
n3
, x, y, e
j
)
jn
and apply the induction hypothesis, (12.7) and Theorem 12.5. Sub-
stitution of x = a
n2
, y = a
n2
in (12.8) gives a positive value, and
x = a
n2
, y = c gives the value 0. Therefore
(12.9) per (a
1
, . . . , a
n3
, c, c, e
j
) 0 for 1 j n
and for each j equality holds if and only if all coordinates of c
except c
j
are 0. If we multiply the left-hand side of (12.9) by the
j-th coordinate of a
n2
and sum over j, we nd c

Qc. Therefore
the assumption Qc = 0 implies that c = 0.
For 0 1, we dene a matrix Q

by taking (12.5) and

replacing every a
i
by a
i
+ (1 )j. From what we have shown
above, it follows that for every in [0,1] the matrix Q

does not
have the eigenvalue 0. Therefore the number of positive eigenvalues
is constant. Since this number is one for = 0, it is also one for
= 1, which proves our assertion.
We formulate the combination of Theorem 12.5 and Theorem
12.6 as a corollary. (The nal assertion follows by continuity.)
Corollary. If a
1
, . . . , a
n1
are vectors in 1
n
with positive coordi-
nates and b 1
n
, then
(per (a
1
, . . . , a
n1
, b))
2
per (a
1
, . . . , a
n1
, a
n1
) per (a
1
, . . . , a
n2
, b, b)
12. The Van der Waerden conjecture 117
and equality holds if and only if b = a
n1
for some constant .
Furthermore, the inequality also holds if some of the coordinates
of the a
i
are 0, but the assertion about the consequence of equality
then cannot be made.
We are now able to generalize Theorem 12.3.
Theorem 12.7. If A
n
is a minimizing matrix, then per A
ij
=
per A for all i and j.
Proof: Suppose that the statement is false. Then by Theorem
12.4, there is a pair r, s such that per A
rs
> per A.
Choose t such that a
rt
> 0. Consider the product of two factors
per A. In the rst of these we replace a
s
by a
t
, and in the second,
we replace a
t
by a
s
. Subsequently, we develop the rst permanent
by column s and the second permanent by column t. By Theorem
12.5 and the Corollary to Theorem 12.6, we have
(per A)
2

_
n

k=1
a
kt
per A
ks
__
n

k=1
a
ks
per A
kt
_
.
By Theorem 12.4, every subpermanent on the right-hand side is at
least per A and per A
rs
> per A. Since per A
rs
is multiplied by
a
rt
which is positive, we see that the right-hand side is larger than
(per A)
2
, a contradiction.
We now use the substitution principle as follows. Take a mini-
mizing matrix A and let u and v be two columns of A. Replace u
and v by
1
2
(u +v). The new matrix is again a minimizing matrix
by Theorem 12.7.
Let A be any minimizing matrix and let b be any column of A,
say the last column. From Corollary (i) to Theorem 12.2, we know
that in every row of A there are at least two positive elements. We
now apply the substitution principle (as sketched above) a num-
ber of times but we never change the last column. In this way,
we can nd a minimizing matrix A
/
= (a
/
1
, . . . , a
/
n1
, b) for which
a
/
1
, . . . , a
/
n1
all have positive coordinates. Now apply the Corollary
to Theorem 12.6. By the substitution principle, equality must hold.
Hence b is a multiple of a
/
i
for any i with 1 i n1. This implies
that b = n
1
j and therefore A = n
1
J
n
, which completes the proof
of the Van der Waerden conjecture.
118 A Course in Combinatorics
Theorem 12.8. The implication (12.1) is true.
Notes.
For a survey of the two proofs of the Van der Waerden conjec-
ture, we refer to Van Lint (1982). There one also nds historical
comments and a nice anecdote concerning the conjecture.
B. L. van der Waerden (19031996), a Dutch mathematician,
was known mostly for his work in algebra, although he published
in several elds. His work Moderne Algebra (1931) set the trend
for many decades.
Mincs book on permanents is the best reference for all the work
that was done in relation to the Van der Waerden conjecture up to
1978.
The name Lorentz space is related to relativity theory and the
group of transformations that leave the quadratic form x
2
+ y
2
+
z
2
t
2
invariant. H. A. Lorentz was a Dutch physicist who won the
Nobel prize for his work.
References.
J. H. van Lint (1981), Notes on Egoritsjevs proof of the Van der
Waerden conjecture, Linear Algebra and its Applications 39, 18.
J. H. van Lint (1982), The van der Waerden Conjecture: Two proofs
in one year, The Math. Intelligencer 39, 7277.
D. London (1971), Some notes on the van der Waerden conjecture,
Linear Algebra and its Applications 4, 155160.
M. Marcus and M. Newman (1959), On the minimum of the per-
manent of a doubly stochastic matrix, Duke Math. J. 26, 6172.
H. Minc (1978), Permanents, Encyclopedia of Mathematics and
its Applications, vol. 6, Addison-Wesley, reissued by Cambridge
University Press (1984).
B. L. van der Waerden (1926), Jber. D. M. V. 35.
13
Elementary counting;
Stirling numbers
The next few chapters will be devoted to counting techniques and
some special combinatorial counting problems. We start with sev-
eral elementary methods that are used quite often. Consider map-
pings from 1, 2, . . . , n to 1, 2, . . . , k. Their total number is k
n
.
In Example 10.2, we studied the case where the mappings were
required to be surjective. We return to this question in Theorem
13.5. If the mappings are injections, then their number is the falling
factorial
(13.1) (k)
n
:= k(k 1) . . . (k n + 1) = k!/(k n)!.
We now consider a similar problem. The n objects to be mapped
are no longer distinguishable but the images are. We formulate this
as follows. We have n indistinguishable balls that are to be placed
in k boxes, marked 1, 2, . . . , k. In how many dierent ways can this
be done? The solution is found by using the following trick. Think
of the balls as being colored blue and line them up in front of the
boxes that they will go into. Then insert a red ball between two
consecutive boxes. We end up with a line of n+k 1 balls, k 1 of
them red, describing the situation. So the answer to the problem
is
_
n+k1
k1
_
. We formulate this as a theorem.
Theorem 13.1. The number of solutions of the equation
(13.2) x
1
+x
2
+ +x
k
= n
in nonnegative integers is
_
n+k1
k1
_
.
Proof: Interpret x
i
as the number of balls in box i.
120 A Course in Combinatorics
Corollary. The number of solutions of the equation (13.2) in pos-
itive integers is
_
n1
k1
_
.
Proof: Replace x
i
by y
i
:= x
i
1. Then

y
i
= n k. Apply
Theorem 13.1.
Example 13.1. By analogy with the question we encountered in
Example 10.6, we consider the problem of selecting r of the integers
1, 2, . . . , n such that no two selected integers are consecutive. Let
x
1
< x
2
< < x
r
be such a sequence. Then x
1
1, x
2
x
1

2, . . . , x
r
x
r1
2. Dene
y
1
:= x
1
, y
i
:= x
i
x
i1
1, 2 i r, y
r+1
:= n x
r
+ 1.
Then the y
i
are positive integers and

r+1
i=1
y
i
= n r + 2. By the
Corollary to Theorem 13.1, we see that there are
_
nr+1
r
_
solutions.
Problem 13A. On a circular array with n positions, we wish to
place the integers 1, 2, . . . , r in order, clockwise, such that consecu-
tive integers, including the pair (r, 1), are not in adjacent positions
on the array. Arrangements obtained by rotation are considered
the same. In how many ways can this be done?
Example 13.2. In how many ways can we arrange r
1
balls of color
1, r
2
balls of color 2, . . . , r
k
balls of color k in a sequence of length
n := r
1
+r
2
+ +r
k
? If we number the balls 1 to n, then there are
n! arrangements. Since we ignore the numbering, any permutation
of the set of r
i
balls of color i, 1 i k, produces the same
arrangement. So the answer to the question is the multinomial
coecient
_
n
r
1
,...,r
k
_
; see (2.1).
Example 13.3. We wish to split 1, 2, . . . , n into b
1
subsets of size
1, b
2
subsets of size 2, ..., b
k
subsets of size k. Here

k
i=1
ib
i
= n.
The same argument as used in Example 13.2 applies. Furthermore,
the subsets of the same cardinality can be permuted among them-
selves without changing the conguration. So the solution is
(13.3)
n!
b
1
! . . . b
k
!(1!)
b
1
(2!)
b
2
. . . (k!)
b
k
.
Several counting problems (often involving binomial coecients)
can be done in a more or less obvious way that leads to a (sometimes
13. Elementary counting; Stirling numbers 121
dicult) calculation. Often there is a less obvious combinatorial
way to do the counting that produces an immediate answer. We
give a few examples.
Example 13.4. Let A run through all subsets of 1, 2, . . . , n.
Calculate S =

[A[. Since there are
_
n
i
_
subsets of size i, we
apparently must calculate

n
i=0
i
_
n
i
_
. By dierentiating (1 + x)
n
,
we nd
n

i=1
i
_
n
i
_
x
i1
= n(1 +x)
n1
and substitution of x = 1 yields the answer S = n 2
n1
. If we had
spent a little more time thinking, then this answer would have been
obvious! A set A and its complement together contain n elements
and there are exactly 2
n1
such pairs.
Example 13.5. In Chapter 10 we saw some examples of formulae
involving binomial coecients for which a combinatorial proof was
easier than a direct proof. The familiar relation
(13.4)
n

k=0
_
n
k
_
2
=
_
2n
n
_
is another example. Of course one can calculate this sum by de-
termining the coecient of x
n
in (1 + x)
n
(1 + x)
n
and using the
binomial formula. However, each side of (13.4) just counts (in two
ways) the number of ways of selecting n balls from a set consisting
of n red balls and n blue balls.
Problem 13B. Show that the following formula for binomial co-
ecients is a direct consequence of (10.6):
_
n + 1
a +b + 1
_
=
n

k=0
_
k
a
__
n k
b
_
.
Give a combinatorial proof by considering (a + b + 1)-subsets of
the set 0, 1, . . . , n, ordering them in increasing order, and then
looking at the value of the integer in position a + 1.
Example 13.6. We consider a slightly more complicated example
where our previous knowledge could lead us to an involved solu-
tion. How many sequences A
1
, . . . , A
k
are there for which A
i

122 A Course in Combinatorics
1, 2, . . . , n, 1 i k, and

k
i=1
A
i
= 1, 2, . . . , n? Since we wish
to avoid that j, 1 j n, is not an element of the union of the
A
i
s, we are tempted to use inclusion-exclusion. If we choose i ele-
ments from 1, 2, . . . , n and consider all sequences A
1
, . . . , A
k
not
containing any of these i elements, then we nd (2
ni
)
k
sequences.
So by Theorem 10.1, the solution to the problem is
n

i=0
(1)
i
_
n
i
_
2
(ni)k
= (2
k
1)
n
.
This answer shows that another approach would have been better.
If we describe a sequence A
1
, . . . , A
k
by a (0, 1)-matrix A of size
k n, with the characteristic functions of the subsets as its rows,
then the condition on the sequences states that A has no column
of zeros. So there are (2
k
1)
n
such matrices!
Problem 13C. Give a solution involving binomial coecients and
a combinatorial solution to the following question. How many pairs
(A
1
, A
2
) of subsets of 1, 2, . . . , n are there such that A
1
A
2
= ?
Problem 13D. Consider the set S of all ordered k-tuples / =
(A
1
, . . . , A
k
) of subsets of 1, 2, . . . , n. Determine

/S
[A
1
A
2
A
k
[.
Problem 13E. The familiar relation
l

m=k
_
m
k
_
=
_
l + 1
k + 1
_
is easily proved by induction. The reader who wishes to, can nd a
more complicated proof by using (10.6). Find a combinatorial proof
by counting paths from (0,0) to (l+1, k+1) in the X-Y plane where
each step is of type (x, y) (x + 1, y) or (x, y) (x + 1, y + 1).
Then use the formula to show that the number of solutions of
x
1
+x
2
+ +x
k
n
13. Elementary counting; Stirling numbers 123
in nonnegative integers is
_
n+k
k
_
. Can you prove this result combi-
natorially?
It is often possible to prove relations between binomial coe-
cients by considering
_
a
k
_
formally as a polynomial in a dened by
_
a
k
_
:=
a(a 1) (a k + 1)
k!
.
If two polynomials of degree k agree for k + 1 values of the
variable, they are identical. We give an example.
Let
F(a) :=
n

k=0
_
a
k
_
x
k
y
nk
, G(a) :=
n

k=0
_
n a
k
_
(x)
k
(x +y)
nk
.
From the binomial theorem we know that if a is an integer in the
interval [0, n], then F(a) = (x + y)
a
y
na
but so is G(a), again by
the binomial theorem. So, the polynomials are identical and we
can substitute any numbers for a,x, and y to obtain relations for
binomial coecients. For instance, y = 2x and a = 2n + 1 yields:
n

k=0
_
2n + 1
k
_
2
nk
=
n

k=0
_
n +k
k
_
3
nk
,
a relation that would be extremely dicult to prove by counting.
Two kinds of numbers that come up in many combinatorial prob-
lems are the so-called Stirling numbers of the rst and second kind.
The numbers are often dened by the formulae (13.8) and (13.12)
given below. We prefer a combinatorial denition.
Let c(n, k) denote the number of permutations S
n
with ex-
actly k cycles. (This number is called a signless Stirling number of
the rst kind.) Furthermore dene c(0, 0) = 1 and c(n, k) = 0 if
n 0 or k 0, (n, k) ,= (0, 0). The Stirling numbers of the rst
kind s(n, k) are dened by
(13.5) s(n, k) := (1)
nk
c(n, k).
124 A Course in Combinatorics
Theorem 13.2. The numbers c(n, k) satisfy the recurrence rela-
tion
(13.6) c(n, k) = (n 1)c(n 1, k) +c(n 1, k 1).
Proof: If is a permutation in S
n1
with k cycles, then there
are n 1 positions where we can insert the integer n to produce
a permutation
/
S
n
with k cycles. We can also adjoin (n) as a
cycle to any permutation in S
n1
with k 1 cycles. This accounts
for the two terms on the right-hand side of (13.6).
Theorem 13.3. For n 0 we have
(13.7)
n

k=0
c(n, k)x
k
= x(x + 1) . . . (x +n 1)
and
(13.8)
n

k=0
s(n, k)x
k
= (x)
n
,
where (x)
n
is dened as in (13.1).
Proof: Write the right-hand side of (13.7) as
F
n
(x) =
n

k=0
b(n, k)x
k
.
Clearly b(0, 0) = 1. Dene b(n, k) := 0 if n 0 or k 0, (n, k) ,=
(0, 0). Since
F
n
(x) = (x +n 1)F
n1
(x)
=
n

k=1
b(n 1, k 1)x
k
+ (n 1)
n1

k=0
b(n 1, k)x
k
,
we see that the numbers b(n, k) satisfy the same recurrence relation
as the c(n, k), namely (13.6). Since the numbers are equal if n 0
or k 0, they are equal for all n and k.
13. Elementary counting; Stirling numbers 125
To prove (13.8) replace x by x and use (13.5).
We remark that it is possible to give a combinatorial proof of
(13.7) by showing that both sides of the equation count the same
objects.
We now dene the Stirling numbers of the second kind: denote
by P(n, k) the set of all partitions of an n-set into k nonempty
subsets (blocks). Then
(13.9) S(n, k) := [P(n, k)[.
Again we have S(0, 0) = 1 and take the numbers to be 0 for all
values of the parameters not covered by the previous denition.
Again we have an easy recurrence relation.
Theorem 13.4. The Stirling numbers of the second kind satisfy
the relation
(13.10) S(n, k) = kS(n 1, k) +S(n 1, k 1).
Proof: The proof is nearly the same as for Theorem 13.3. A
partition of the set 1, 2, . . . , n 1 can be made into a partition
of 1, 2, . . . , n by adjoining n to one of the blocks or by increasing
the number of blocks by one by making n a block.
We dene the Bell number B(n) to be the total number of parti-
tions of an n-set, i.e.
(13.11) B(n) :=
n

k=1
S(n, k), (n 1).
For the Stirling numbers of the second kind there is a formula
similar to (13.8).
Theorem 13.5. For n 0 we have
(13.12) x
n
=
n

k=0
S(n, k)(x)
k
.
Proof: We rst remark that by (13.9) the number of surjective
mappings from an n-set to a k-set is k!S(n, k) (a block of the parti-
tion is the inverse image of an element of the k-set). So by Example
10.2, we have
126 A Course in Combinatorics
(13.13) S(n, k) =
1
k!
k

i=0
(1)
i
_
k
i
_
(k i)
n
=
1
k!
k

i=0
(1)
ki
_
k
i
_
i
n
.
Now let x be an integer. There are x
n
mappings from the n-set
N := 1, 2, . . . , n to the x-set 1, 2, . . . , x. For any k-subset Y of
1, 2, . . . , x, there are k!S(n, k) surjections from N to Y . So we
nd
x
n
=
n

k=0
_
x
k
_
k!S(n, k) =
n

k=0
S(n, k)(x)
k
.

In Example 10.1, we saw that it can be useful to associate a so-

called generating function with a sequence a
1
, a
2
, . . . of numbers.
In Chapter 14, we will see many uses of the generating function.
At this point we treat the generating functions for the Stirling
numbers.
Theorem 13.6.

nk
S(n, k)
x
n
n!
=
1
k!
(e
x
1)
k
(k 0).
Proof: Let F
k
(x) denote the sum on the left-hand side. By (13.10)
we have
F
/
k
(x) = kF
k
(x) +F
k1
(x).
The result now follows by induction. Since S(n, 1) = 1, the as-
sertion is true for k = 1 and the induction hypothesis yields a
dierential equation for F
k
, which with the condition S(k, k) = 1
has the right-hand side of the assertion as unique solution.
For the Stirling numbers of the rst kind, it is slightly more
dicult to nd the generating function.
Theorem 13.7.

n=k
s(n, k)
z
n
n!
=
1
k!
(log(1 +z))
k
.
Proof: Since
(1 +z)
x
= e
xlog(1+z)
=

k=0
1
k!
(log(1 +z))
k
x
k
,
13. Elementary counting; Stirling numbers 127
the right-hand side in the assertion is the coecient of x
k
in the
expansion of (1 +z)
x
. On the other hand, we have for [z[ < 1,
(1 +z)
x
=

n=0
_
x
n
_
z
n
=

n=0
1
n!
(x)
n
z
n
=

n=0
z
n
n!
n

r=0
s(n, r)x
r
=

r=0
x
r

n=r
s(n, r)
z
n
n!
.
This completes the proof.
Problem 13F. Show directly that the number of permutations of
the integers 1 to n with an even number of cycles is equal to the
number of permutations with an odd number of cycles (n > 1).
Also show that this is a consequence of Theorem 13.7.
Finally we mention that the Stirling numbers of the rst and
second kind are related by
(13.14)
n

k=m
S(n, k)s(k, m) =
mn
.
This follows immediately if we substitute (13.8) in (13.12). Since
the functions x
n
, respectively (x)
n
, with n 0 both form a basis
of the vector space C[x], the formula (13.14) is just the standard
relation between the matrices for basis transformation.
Problem 13G. Show that (13.12) leads to B(n) =
1
e

k=0
k
n
k!
.
Problem 13H. Let A be the n n matrix with a
ij
:=
_
i
j
_
for
i, j = 0, . . . , n 1. Determine A
1
.
Problem 13I. From (10.6) one nds, by taking x =
1
2
, that

j=0
_
a+j
j
_
2
j
= 2
a+1
. Using a method from this chapter prove
that
a

j=0
_
a +j
j
_
2
j
= 2
a
.
Prove the same result directly by dening
a
n
=
a

j=0
_
a +j
j
_
2
aj
,
128 A Course in Combinatorics
substituting the basic recurrence for binomial coecients, thus nd-
ing that a
n
=
1
2
a
n
+
1
2
a
n1
.
Problem 13J. Suppose we call a set nice if its cardinality is di-
visible by 3. How many nice subsets does an n-set have?
Problem 13K. Prove that

n=1
S(n, n 2)x
n
=
x(1+2x)
(1x)
5
.
Notes.
Stirling numbers of the rst and second kind appear in many
areas of mathematics; for example, they play a role in a number
of interpolation formulae and in the calculus of nite dierences.
There are tables of the numbers in several books on tables of math-
ematical functions.
Later, in Chapter 37, these numbers will reappear.
James Stirling (16921770), a Scottish mathematician, studied
at Oxford. He taught mathematics at Venice and London but
switched to a career in business at the age of 43.
14
Recursions and
generating functions
Many combinatorial counting problems with a solution a
n
depend-
ing on a parameter n, can be solved by nding a recursion relation
for a
n
and then solving that recursion. Sometimes this is done by
introducing an ordinary generating function
f(x) :=

n0
a
n
x
n
,
or an exponential generating function
f(x) :=

n0
a
n
x
n
n!
,
and using the recursion to nd an equation or a dierential equation
for f(x), and solving that equation. We shall demonstrate several
of the techniques involved.
Example 14.1. As an introduction, consider once again Example
10.1. Let be a derangement of 1, 2, . . . , n + 1. There are n
choices for (n + 1). If (n + 1) = i and (i) = n + 1, then is
also a derangement on the set 1, 2, . . . , ni. If (n+1) = i and
(i) ,= n +1 = (j), then replacing (j) by i yields a derangement
on the set 1, 2, . . . , n. Therefore
(14.1) d
n+1
= n(d
n
+d
n1
),
which is also an immediate consequence of (10.3). Let D(x) be
the exponential generating function for the sequence d
0
= 1, d
1
=
0, d
2
, . . . . From (14.1) we immediately nd
(1 x)D
/
(x) = xD(x),
130 A Course in Combinatorics
and from this we nd D(x) = e
x
/(1 x) and (10.2).
In many cases, we use the generating functions only as a book-
keeping device, and our operations of addition, multiplication (and
even substitution and derivation, as we shall see below) are to be
interpreted formally. It is possible to give a completely rigorous
theory of formal power series (as algebraic objects) and we give
an introduction to this theory in Appendix 2. In most cases, it is
intuitively clear and easy to check that the operations are legiti-
mate. If the series that we use actually converge, then we can use
all appropriate knowledge from analysis concerning these series, as
we did in Example 14.1. We give another elementary example.
Example 14.2. Suppose that we have k boxes numbered 1 to k
and suppose that box i contains r
i
balls, 1 i k. A formal
bookkeeping device to list all possible congurations is to let the
named one correspond to the term x
r
1
1
x
r
2
2
x
r
k
k
in the product
(1 +x
1
+x
2
1
+ )(1 +x
2
+x
2
2
+ ) (1 +x
k
+x
2
k
+ ).
We can collect all the terms involving exactly n balls by taking
x
i
= x for all i, and considering the terms equal to x
n
. Therefore we
nd that the number of ways to divide n balls over k distinguishable
boxes is the coecient of x
n
in the expansion of (1 x)
k
, and by
(10.6) this is
_
k1+n
n
_
, giving a second proof of Theorem 13.1.
In many cases, the combinatorial problem that we are interested
in leads to a linear recursion relation with constant coecients,
which is easily solved by standard methods.
Example 14.3. We consider paths of length n in the X-Y plane
starting from (0,0) with steps R : (x, y) (x +1, y), L : (x, y)
(x1, y), and U : (x, y) (x, y+1) (i.e. to the right, to the left, or
up). We require that a step R is not followed by a step L and vice
versa. Let a
n
denote the number of such paths. First observe that
if we denote by b
n
the number of paths of length n starting with a
step U, then b
n
= a
n1
and furthermore trivially b
n+m
b
n
b
m
and
b
n
3
n1
. So by Feketes lemma, Lemma 11.6, lim
n
b
1/n
n
exists
and is at most 3. Next, note that a
0
= 1 and a
1
= 3. We split
the set of paths of length n into subsets depending on the last one
14. Recursions and generating functions 131
or two steps. Clearly there are a
n1
paths ending with the step U.
Take a path of length n 1 and repeat the last step if it is L or R,
and adjoin a step L if the last step was U. In this way, we obtain
all the paths of length n that end in LL, RR, or UL. So there are
a
n1
of these. It remains to count the paths ending with UR and
again it is trivial that there are a
n2
of these. We have shown that
a
n
= 2a
n1
+a
n2
(n 2).
Let f(x) =

n=0
a
n
x
n
. Then the recursion implies that
f(x) = 1 + 3x + 2x(f(x) 1) +x
2
f(x),
i.e.
f(x) =
1 +x
1 2x x
2
=
1
2

1 x
+
1
2

1 x
,
where = 1 +

2, = 1

2. Therefore
a
n
=
1
2
(
n+1
+
n+1
)
and we nd lim
n
a
1/n
n
= 1 +

2.
Problem 14A. (i) Let a
n
denote the number of sequences of 0s
and 1s that do not contain two consecutive 0s. Determine a
n
.
(ii) Let b
n
denote the number of sequences of 0s and 1s with no
two consecutive 1s and for which a run of 0s always has length 2
or 3, including possibly at the beginning or end of the sequence.
Show that b
1/n
n
c for some c and approximate c.
Example 14.4. Let a(r, n), where 0 r n, denote the number
of solutions of the problem of Example 13.1 (a(0, 0) = 1). We
divide the set of possible sequences into two subsets: those with
x
1
= 1 and those with x
1
> 1. The rst subset clearly contains
a(r 1, n 2) elements, the second one a(r, n 1) elements. So
(14.2) a(r, n) = a(r, n 1) +a(r 1, n 2) (n > 1).
132 A Course in Combinatorics
From this recursion, we can prove the result a(r, n) =
_
nr+1
r
_
by
induction. Using the generating function is more dicult. Try
f(x, y) :=

n=0

r=0
a(r, n)x
n
y
r
.
From (14.2) we nd
f(x, y) = 1 +x +xy +x(1 +f(x, y)) +x
2
yf(x, y),
i.e.
f(x, y) =
1 +xy
1 x x
2
y
=
1
1 x
+

a=1
x
2a1
y
a
(1 x)
a+1
.
Substitution of (10.6) for (1x)
a1
produces the required binomial
coecient for a(r, n).
As we saw in Example 14.3 (and Problem 14A) a linear recursion
with constant coecients leads to a rational function as generating
function (and vice versa). Indeed, if a
n
=

l
k=1

k
a
nk
(n > l) and
f(x) =

n=0
a
n
x
n
, then (1

l
k=1

k
x
k
)f(x) is a power series for
which the coecient of x
n
is 0 if n > l.
The following example due to Klarner (1967) shows an inter-
esting case of a linear recursion that is found with the help of the
generating function and that is not at all obvious in a combinatorial
way.
Example 14.5. Consider congurations in the plane, called poly-
ominoes, as in Fig. 14.1. The conguration consists of layers, each
consisting of consecutive squares. Two consecutive layers meet
along the edges of a number ( 1) of squares. (More precisely,
we are considering horizontally convex polyominoes.)
12
9 10 11
6 7 8
1 2 3 4 5
Figure 14.1
14. Recursions and generating functions 133
Let a
n
denote the number of polyominoes with n squares and
dene f(x) :=

n=1
a
n
x
n
. To nd f, we introduce a(m, n) for the
number of polyominoes for which the bottom layer has m squares
(and a total of n). We dene a(m, n) := 0 if m > n. Clearly
(14.3) a(m, n) =

l=1
(m+l 1)a(l, n m).
We dene
(14.4) F(x, y) :=

n=1

m=1
a(m, n)x
n
y
m
.
Then f(x) = F(x, 1). Because the series will turn up below, we
also dene
g(x) :=

n=1

m=1
ma(m, n)x
n
.
We would like to write
(14.5) g(x) =
_
F
y
_
y=1
.
Even though we have a theory of formal power series, it may be
instructive to show that the righthand side of (14.4) converges in
a suciently large region. This gives us the opportunity to show a
quick way of getting a rough estimate for a
n
. Number the squares
of the polyomino in the obvious way as in Fig. 14.1. To each square
associate a quadruple (x
0
, x
1
, x
2
, x
3
) of 0s and 1s, where x
0
= 1
means that there is a square of the polyomino below this square,
x
1
= 1 means that there is a square of the polyomino to the left,
x
2
= 1 means that there is a square above, and x
3
= 1 means
that there is a square to the right. For example, in Fig. 14.1 the
rst quadruple is (0,0,1,1). The sequence of quadruples uniquely
determines the polyomino (e.g. the fth quadruple is the rst one
in the sequence that ends in a 0, showing that m = 5, etc.). This
shows that a
n
15
n
. From this and (14.3) we nd a(m, n)
134 A Course in Combinatorics
n 15
nm
which is enough to justify (14.5). From (14.4) we nd by
substituting (14.3) and a straightforward calculation
(14.6) F(x, y) =
xy
1 xy
+
(xy)
2
(1 xy)
2
f(x) +
xy
1 xy
g(x).
Dierentiation of both sides of (14.6) with respect to y and taking
y = 1 yields (using (14.5)):
(14.7) g(x) =
x
(1 x)
2
+
2x
2
(1 x)
3
f(x) +
x
(1 x)
2
g(x).
From (14.7) we can nd g(x) and substitute this in (14.6); then
take y = 1, which yields
(14.8) f(x) =
x(1 x)
3
1 5x + 7x
2
4x
3
.
From (14.8) we see that a
n
satises the recurrence relation
(14.9) a
n
= 5a
n1
7a
n2
+ 4a
n3
(n 5).
As we remarked above, it is not at all clear how one could prove
this directly; it has been done however.
Remark. From (14.9) we nd that lim
n
a
1/n
n
= , where is the
zero with largest absolute value of the polynomial x
3
5x
2
+7x4
( 3.2).
The following example produces a result that is important for the
theory of nite elds. In this example, we combine a generalization
of the idea of Example 14.2 with the method of formal operations
with power series. The reader should convince herself that the
operations with logarithms are correct without using convergence.
Example 14.6. We shall count the number of irreducible polyno-
mials of degree n over a eld of q elements.
Number all the monic irreducible polynomials of degree at least
one over a eld of q elements:
f
1
(x), f
2
(x), f
3
(x), . . .
14. Recursions and generating functions 135
with, say, respective degrees d
1
, d
2
, d
3
, . . . . Let N
d
denote the num-
ber of degree d, d = 1, 2, 3, . . . .
Now for any sequence i
1
, i
2
, i
3
, . . . of nonnegative integers (all but
nitely many of which are zero), we get a monic polynomial
f(x) = (f
1
(x))
i
1
(f
2
(x))
i
2
(f
3
(x))
i
3
. . .
whose degree is n = i
1
d
1
+ i
2
d
2
+ i
3
d
3
+ . By unique factor-
ization, every monic polynomial of degree n arises exactly once in
this way. To repeat, there is a one-to-one correspondence between
monic polynomials of degree n and sequences i
1
, i
2
, . . . of nonneg-
ative integers satisfying n = i
1
d
1
+i
2
d
2
+i
3
d
3
+ .
Of course, the number of monic polynomials of degree n is q
n
,
i.e. the coecient of x
n
in
1
1 qx
= 1 +qx + (qx)
2
+ (qx)
3
+ .
Clearly (cf. Example 14.2), the number of sequences i
1
, i
2
, . . . with
n = i
1
d
1
+ i
2
d
2
+ . . . is the coecient of x
n
in the formal power
series
(1 +x
d
1
+x
2d
1
+x
3d
1
+ )(1 +x
d
2
+x
2d
2
+x
3d
2
+ ) . . . .
Thus we conclude
1
1 qx
=

i=1
1
1 x
d
i
=

d=1
_
1
1 x
d
_
N
d
.
Recalling that log
1
1z
= z +
1
2
z
2
+
1
3
z
3
+ , we take (formal)
logarithms of both extremes of the above displayed equation to
nd

n=1
(qx)
n
n
=

d=1
N
d

j=1
x
jd
j
.
Then, comparing coecients of x
n
on both sides, we get
q
n
n
=

d[n
N
d
1
n/d
,
136 A Course in Combinatorics
i.e.
q
n
=

d[n
dN
d
.
This last equation was our goal. We have derived it combinatori-
ally from unique factorization, but we remark that it has in fact an
elegant interpretation in terms of the factorization of x
q
n
x over
the eld of q elements. Applying M obius inversion (cf. Theorem
10.4), we obtain the following theorem.
Theorem 14.1. For a prime power q, the number of monic irre-
ducible polynomials of degree n over the eld of q elements is given
by
N
n
=
1
n

d[n
(
n
d
)q
d
.
We note that a direct consequence of Theorem 14.1 is that N
d
>
0, that is, there exist irreducible polynomials of every degree d.
This immediately leads to a proof of the existence of elds of p
d
elements for every prime p, without the usual expedient of reference
to the algebraic closures of the prime elds.
* * *
We now come to a large class of counting problems, all with
the same solution. The class is known as the Catalan family be-
cause Catalan (1838) treated Example 14.7. Actually, the equiva-
lent problem called Question 3 in Example 14.9 below was treated
by von Segner and Euler in the 18th century. We denote the so-
lutions to the problems by u
n
. We shall show that u
n
=
1
n
_
2n2
n1
_
.
These numbers are called Catalan numbers.
Example 14.7. Suppose that we have a set S with a nonasso-
ciative product operation. An expression x
1
x
2
. . . x
n
with x
i
S
does not make sense, and brackets are needed to indicate the order
in which the operations are to be carried out. Let u
n
denote the
number of ways to do this if there are n factors x
i
. For example,
u
4
=
1
4
_
6
3
_
= 5 corresponding to the products (a(b(cd))), (a((bc)d)),
((ab)(cd)), ((a(bc))d), and (((ab)c)d). Each product contains within
the outer brackets two expressions, the rst a product of m factors
14. Recursions and generating functions 137
and the second a product of n m factors, where 1 m n 1.
It follows that the numbers u
n
satisfy the recurrence relation
(14.10) u
n
=
n1

m=1
u
m
u
nm
(n 2).
From (14.10) and u
1
= 1, we nd that the generating function
f(x) :=

n=1
u
n
x
n
satises the equation
(14.11) f(x) = x +

n=2
_
n1

m=1
u
m
u
nm
_
x
n
= x + (f(x))
2
.
Solving the quadratic equation and taking into account that f(0) =
0, we nd
f(x) =
1

1 4x
2
.
From the binomial series we then obtain
(14.12) u
n
=
1
n
_
2n 2
n 1
_
.
The operations carried out above can all be justied as formal op-
erations on power series, and the binomial series can also be treated
formally without being concerned about convergence. If one feels
uneasy about this approach, there are two remedies. The rst is
to nd the solution as we did above and afterwards prove that it is
correct by using (14.10) and induction. If one really wishes to be
sure that the generating function is dened by a convergent power
series, then use the result of the calculation to nd some rough es-
timate for u
n
and prove that the estimate is correct using (14.11),
and then show that the series converges. For example, we could
try u
n
c
n1
/n
2
. This is true for n = 1 and any positive c and it
is true for n = 2 if c 4. The induction step with (14.11) yields
u
n
c
n2
n1

m=1
1
m
2
(n m)
2
c
n2
2
(n/2)
2

m=1
1
m
2
<
c
n1
n
2
138 A Course in Combinatorics
if c >
4
3

2
. So the radius of convergence of

n=1
u
n
x
n
is positive.
Remark. Formula (14.10) shows that u
n
u
m
u
nm
, and then
Feketes lemma implies that lim
n
u
1/n
n
exists. In fact, (14.12)
now shows that this limit is 4.
As we remarked above, the Catalan numbers reappear regularly
in combinatorial counting problems. This has led to the question
of proving the other results combinatorially, i.e. by showing that
there is a one-to-one correspondence between the objects to be
counted and those of Example 14.7. Some of these will be treated
in Example 14.9. First we look at another famous counting problem
that has the Catalan numbers as solution.
Example 14.8. Consider walks in the X-Y plane where each step
is U : (x, y) (x+1, y+1) or D : (x, y) (x+1, y1). We start
at (0,0) and ask in how many ways we can reach (2n, 0) without
crossing the X-axis. The solution of this problem uses an elegant
trick known as Andres reection principle (1887). In Fig. 14.2 we
consider two points A and B in the upper halfplane and a possible
path between them which meets and/or crosses the X-axis.
By reecting the part of the path between A and the rst meeting
with the X-axis (i.e. C in Fig. 14.2) with respect to the X-axis, we
nd a path from the reected point A
/
to B. This establishes a
one-to-one correspondence between paths from A
/
to B and paths
from A to B that meet or cross the X-axis.
Figure 14.2
It follows that if A = (0, k) and B = (n, m), then there are
_
n
l
1
_
paths from A to B that cross or meet the X-axis, where 2l
1
:= n
km. Since there are
_
n
l
2
_
paths fromA to B, where 2l
2
:= nm+k,
14. Recursions and generating functions 139
we nd
_
n
l
2
_

_
n
l
1
_
paths from A to B that do not meet the X-axis.
Any path from (0,0) to (2n, 0) in the upper halfplane that does not
meet the X-axis between these points goes from (0,0) to (1,1):=A,
from A to B := (2n 1, 1) without meeting the X-axis, and then
from (2n 1, 1) to (2n, 0). By the argument above, we nd that
there are u
n
such paths. If we allow the paths to meet the X-axis
without crossing, then there are u
n+1
such paths.
We remark that the number of paths from (0,0) to (2n, 0) in the
upper halfplane that do not meet the X-axis between these points is
equal to the number of sequences of zeros and ones (x
1
, x
2
, . . . , x
2n
)
with
(14.13) x
1
+x
2
+ +x
j
_
<
1
2
j for 1 j 2n 1,
= n for j = 2n.
The correspondence is given by letting a 1 correspond to a step D
of the path.
We now show by combinatorial arguments that several counting
problems lead to the Catalan numbers.
Example 14.9. Consider the following three questions.
Question 1. A tree that is drawn in the plane is called a plane
tree. How many rooted plane trees are there with n vertices and a
root with degree 1 (so-called planted plane trees)?
Question 2. A planted plane tree is called trivalent or a binary
tree if every vertex has degree 1 or 3. How many trivalent planted
plane trees are there with n vertices of degree 1?
Question 3. In how many ways can one decompose a convex
n-gon into triangles by n 3 nonintersecting diagonals?
The rst correspondence we shall show is between Question 2
and Example 14.7.
140 A Course in Combinatorics
For this, Fig. 14.3 suces.
Figure 14.3
It follows that the solution to Question 2 is u
n1
.
The correspondence between Question 1 and Example 14.8 is also
seen from a gure, namely Fig. 14.4.
Figure 14.4
In the gure, the tree can be described by a walk around it as
shown by the dotted line. This walk can be described by calling
the steps U for a step going up, and D for a step going down. This
yields a sequence of twelve steps. Replacing U by 0 and D by 1
yields a sequence that obviously satises (14.13). This shows that
the solution to Question 1 is also u
n1
.
Finally, we show a correspondence between Question 3 and Ques-
tion 2. Consider an n-gon decomposed into triangles and distin-
guish some edge. We now construct a tree as in Fig. 14.5.
14. Recursions and generating functions 141
Figure 14.5
The tree is a binary tree with vertices of degree 1 corresponding
to the sides of the n-gon, and vertices of degree 3 corresponding to
the triangles. The tree is planted at the vertex corresponding to
the special edge. So again we nd that the number of solutions is
u
n1
.
Problem 14B. Describe (with a gure) a correspondence between
the graphs of Question 1 and those of Question 2.
Problem 14C. Find a direct one-to-one mapping from the n-gons
of Question 3 to the products of Example 14.7.
We now turn to several problems in which the exponential gen-
erating function turns out to be useful. We shall give a somewhat
more systematic treatment for this case. For a complete treatment
of the methods, we refer to A. Joyal (1981).
Let M denote a type of combinatorial structure. For example,
trees, polygons, sets (the uniform structures), permutations, etc.
Let m
k
be the number of ways of giving a labeled k-set such a
structure. In each separate case we shall specify whether we take
m
0
= 0 or m
0
= 1. We use capital letters for the structure and
lower case letters for the counting sequence. We dene
(14.14) M(x) :=

k=0
m
k
x
k
k!
.
So if T denotes the structure (labeled) tree, then we know from
142 A Course in Combinatorics
Theorem 2.1 that
T(x) =

k=0
k
k2
x
k
k!
.
If S denotes the uniform structure (a set) then s
k
=1 for all k and
therefore S(x) = e
x
. If C denotes oriented circuit, then we can
start at vertex 1 and there are clearly (k 1)! ways to proceed.
Therefore C(x) = log(1 x).
Suppose we wish to consider the number of ways a labeled n-set
can be partitioned into two parts, one with a structure of type A
and the other with a structure of type B. It is easy to see that the
number of ways that we can do this is

n
k=0
_
n
k
_
a
k
b
nk
. It follows
that if we call this a structure of type A B, then
(14.15) (A B)(x) =

n=0
_
n

k=0
_
n
k
_
a
k
b
nk
_
x
n
n!
= A(x) B(x).
The reader should be careful with the result (14.15). The asser-
tion is only true if, from the compound structure there is a unique
way of nding the partition into the two parts that led to this struc-
ture. For example, if A and B are the same, the result is clearly
false.
Example 14.10. Once again we consider derangements. Call this
a structure of type D and let denote the structure permutation.
Clearly (x) = (1x)
1
. Any permutation consists of a set of xed
points (that we interpret as just a set) and a derangement on the
remaining points. So by (14.15) we have
(1 x)
1
= D(x) S(x), i.e. D(x) = e
x
(1 x)
1
as was shown in Example 10.1.
Example 14.11. In how many ways can a labeled n-set be split
into a number of pairs (= P) and a number of singletons (= S)?
First, observe that if we wish to split 2k points into pairs, we must
choose a point x
1
to go with point 1 and then split the remaining
2k2 points into pairs. Therefore p
2k
= (2k1)!! and we nd that
P(x) =

k=0
(2k 1)!!
x
2k
(2k)!
= exp(
1
2
x
2
).
14. Recursions and generating functions 143
It follows that
(14.16) (P S)(x) = exp(x +
1
2
x
2
).
Let us now try to nd the same result using a recursion relation.
We denote the structure P S by B. In the set 1, 2, . . . , n we can
either let n be a singleton or make a pair x, n with 1 x n1.
So
b
n
= b
n1
+ (n 1)b
n2
(n 1).
It follows that
B
/
(x) = (1 +x)B(x),
and since B(0) = 1, we again nd (14.16) as solution.
We return to the recurrence of Example 14.11 in Example 14.15.
Problem 14D. Consider the structures M
0
:= mapping of a set
to itself with no xed point, M
1
:= mapping of a set to itself with
exactly one xed point, and A:= arborescence. Find a relation
between M
0
(x), M
1
(x), and A(x) and check the rst few terms to
see if the result is correct.
We make things slightly more dicult. We wish to partition an
n-set into parts and then impose structure N on each of the parts,
where n
0
= 0. We claim that the exponential generating func-
tion for the compound structure is exp(N(x)). A simple example
is the case where N is the uniform structure (with the convention
n
0
= 0) and hence N(x) = e
x
1. The compound structure is of
course partition for which we know from Theorem 13.6 that the
exponential generating function is equal to

n=0
_
n

k=0
S(n, k)
_
x
n
n!
=

k=0
(e
x
1)
k
k!
= exp(e
x
1) = exp(N(x)).
We formulate this method as a theorem.
Theorem 14.2. If the compound structure S(N) is obtained by
splitting a set into parts, each of which gets a structure of type N,
then
S(N)(x) = exp(N(x)).
144 A Course in Combinatorics
Proof: By a slight generalization of (13.3) we see that if the par-
tition of the n-set consists of b
1
parts of size 1, b
2
parts of size 2,
. . . , b
k
parts of size k, where b
1
+2b
2
+ +kb
k
= n, then there are
_
n
1
1!
_
b
1
. . .
_
n
k
k!
_
b
k

n!
b
1
! . . . b
k
!
ways to make the compound structure, and this has to be divided
by n! to obtain the contribution to the coecient of x
n
in the
exponential generating function. If b
1
+ +b
k
= m, then this same
contribution is found in exp(N(x)) from the term (N(x))
m
/m!,
namely as
1
m!
_
m
b
1
, . . . , b
k
_
_
n
1
x
1!
_
b
1
. . .
_
n
k
x
k
k!
_
b
k
.
This proves the assertion.
In fact, it is not dicult to see that this theorem is a special
case of a more general method. Interpret the previous situation
as follows. We have a uniform structure on a k-set of points and
we replace each point by some structure of type N. The resulting
conguration is an element of the compound structure that we just
discussed. If the rst of the structures is not uniform but, say, of
type R, then the exponential generating function for the compound
structure will be R(N(x)) by the same argument as we used above.
We sometimes call this procedure the substitution of N into R,
and it is nice that we then must do the same for the generating
functions!
Example 14.12. If we substitute the structure oriented cycle
into the uniform structure, then we are considering the compound
structure consisting of a partition of an n-set into oriented cycles,
i.e. the structure with as usual
0
= 1. So we must have (x) =
exp(C(x)) and, indeed, (x) = (1 x)
1
and C(x) = log(1 x).
Example 14.13. Let us look at Fig. 2.2 again, adding loops at 1
and 21. We then have a description of a mapping of an n-set (with
n = 21 in this case) to itself. But we can also interpret the gure as
an element of the structure , namely (1)(4, 5, 3)(7)(20, 12)(21) in
14. Recursions and generating functions 145
which each point has been replaced by an arborescence with that
point as root. Letting A denote arborescence, we nd from Cayleys
theorem, Theorem 2.1, that
(14.17) A(x) =

n=1
n
n1
x
n
n!
.
(The extra factor n comes from the choice of the root.) Since there
are n
n
mappings from an n-set to itself, the method of Theorem
14.2 shows that the following relation must hold:
(14.18) (A(x)) =
1
1 A(x)
=

n=0
n
n
x
n
n!
.
We shall verify this after the next example.
The following well known result from complex analysis often
plays a role in combinatorial problems involving generating func-
tions. It is known as the Lagrange inversion formula. It may be
found in textbooks on analysis with analytic proofs; see the notes.
It is possible to prove the theorem within the theory of formal power
series (using formal derivation, etc.) and we give such a formal proof
in Appendix 2.
Theorem 14.3. Let f be analytic in a neighborhood of z = 0 and
f(0) ,= 0. Then, if w = z/f(z), z can be expressed as a power series
z =

k=1
c
k
w
k
with a positive radius of convergence, in which
(14.19) c
k
=
1
k!
_
_
d
dz
_
k1
(f(z))
k
_
z=0
.
Example 14.14. We shall nd a fourth proof of Theorem 2.1. Let
T be the structure labeled tree. We wish to show that t
n
= n
n2
for n 1. If A denotes arborescences as before, then obviously
a
n
= nt
n
, i.e. A(x) = xT
/
(x). Furthermore, from Theorem 14.2
we see that exp(A(x)) is the exponential generating function for
the structure rooted forest (= F). Consider a labeled tree on
n +1 vertices as an arborescence with vertex n +1 as its root and
146 A Course in Combinatorics
then delete the root and all incident edges. The result is a rooted
forest on n vertices. Since we can reverse this process, we have a
one-to-one correspondence. Therefore
(14.20) e
A(x)
= 1 +

n=1
f
n
x
n
n!
=

n=0
t
n+1
x
n
n!
= T
/
(x) = x
1
A(x),
i.e.
(14.21)
A(x)
e
A(x)
= x.
We apply Theorem 14.3 to (14.21) with z = A(x), f(z) = e
z
=
e
A(x)
, w = x. We nd A(x) =

k=1
c
k
x
k
with
c
k
=
1
k!
_
_
d
dz
_
k1
e
kz
_
z=0
=
k
k1
k!
and it follows that t
n
= n
n2
. Furthermore we see that the number
of labeled rooted forests on n vertices is (n + 1)
n1
.
Remark. We now have verication of (14.18) because (14.20) im-
plies that A
/
(x) = e
A(x)
+xe
A(x)
A
/
(x), i.e.

n=1
n
n
x
n
n!
= xA
/
(x) =
xe
A(x)
1 xe
A(x)
=
A(x)
1 A(x)
.
Remark. The procedure of removing a point from a combinatorial
structure with n+1 points that we used in Example 14.14 is called
derivation and it indeed corresponds to the derivation of generating
functions. Another example is removing a vertex from an oriented
cycle on n +1 vertices. This yields an oriented path on n vertices,
of which there are n!. So their exponential generating function is
(1 x)
1
and that is C
/
(x).
Problem 14E. Let a
n
denote the number of ways of decomposing
a convex n + 1-gon into quadrilaterals by inserting a number of
nonintersecting chords. By convention a
0
= 0, a
1
= 1. Show that

k+l+m=n
a
k
a
l
a
m
= a
n
for n 3. If f(x) is the ordinary generating
14. Recursions and generating functions 147
function for the sequence a
n
, then nd a functional equation for
f(x) and solve this equation using Theorem 14.3. (We remark that
the result that is obtained can also be proved combinatorially.)
Although we have stressed that in most cases the power series
can be considered as formal power series, it has been clear in a
number of examples that analysis can be an important tool in many
combinatorial problems. We give one more example that shows a
method that can be used for many recurrences.
Example 14.15. In Example 14.11 we encountered the recursion
(14.22) a
n
= a
n1
+ (n 1)a
n2
and the corresponding exponential generating function. In previous
examples we showed that the counting function grows more or less
as c
n
for some constant c. Clearly a
n
grows more rapidly, but what
is a good approximation for its asymptotic behavior? The rst step
in obtaining the answer to this question involves a method that is
applicable to many recurrences. In (14.22) we substitute
(14.23) a
n
=
_
C
(z)z
n
dz,
where C is a path in C that we can still choose and is a function
also still to be determined. We substitute (14.23) into (14.22).
The term (n 1)a
n2
yields
_
C
(z)(n 1)z
n2
dz, and we require
that integration by parts yields only the integral involving
/
(z),
the other term being 0 by a suitable choice of C. Then (14.22)
becomes
(14.24)
_
C
_
(z)[z
n
z
n1
] +
/
(z)z
n1
_
dz = 0,
which is true for all n N if (z)(1z) =
/
(z), i.e. (z) = e
z
1
2
z
2
.
Once we know , the requirement on C shows that the real axis
from to is a good choice for the path of integration. From
a
0
= 1 we nd that = (2e)

1
2
.
To nd the asymptotic behavior of a
n
, we must now analyze the
behavior of
(14.25) I :=
_

e
x
1
2
x
2
x
n
dx.
148 A Course in Combinatorics
Since the integrand is maximal near x =

n, we substitute x =
y +

n and nd
(14.26)
I = e

1
2
n+

n
n
1
2
n
_

e
y
1
2
y
2
exp
_
y

n +nlog(1 +
y

n
)
_
dy.
Now note that if u and v are negative, then [e
u
e
v
[ < [u v[ and
use the representation log(1 + t) =
_
t
0
ds
1+s
= s
s
2
2
+
_
t
0
s
2
1+s
ds to
obtain

n +nlog(1 +
y

n
) +
y
2
2

[y[
3

n
.
Substitution in (14.26) shows that the integral tends to
_

e
yy
2
dy =

e
1
4
.
We have proved that
(14.27) a
n

e

1
4

2
n
1
2
n
e

1
2
n+

n
.
It is clear that this result cannot be found in an easy way!
Problem 14F. Let F
n
(x) denote the expansion of (1 x
n
)
(n)/n
in a power series. Also consider the expansion of e
x
as a formal
power series. Prove that
e
x
=

n=1
F
n
(x)
is true as a relation between formal power series.
Problem 14G. Find the exponential generating function for the
number of symmetric n n permutation matrices.
Problem 14H. On a circle we place n symbols 0 and n symbols
1 in an arbitrary order. Show that it is possible to number the
positions on the circle consecutively from 1 to 2n such that the
14. Recursions and generating functions 149
sequence satises (14.13) if we replace the strict inequality < with
.
Problem 14I. Let the points 1, . . . , 2n be on a circle (consecu-
tively). We wish to join them in pairs by n nonintersecting chords.
In how many ways can this be done?
Problem 14J. Find the exponential generating function for the
number of labeled regular graphs of valency 2, with n vertices.
Problem 14K. Consider walks in the X-Y plane where each step
is R : (x, y) (x+1, y) or U : (x, y) (x, y +1). We start at (0,0)
and ask in how many ways we can reach (2n, 2n) without passing
through one of the points (2i 1, 2i 1) , i = 1, . . . , n. Prove that
this number is the Catalan number u
2n+1
.
Problem 14L. Consider walks in the X-Y plane where each step
is R : (x, y) (x + 1, y) or U
a
: (x, y) (x, y + a), with a a
positive integer. There are ve walks that contain a point on the
line x+y = 2, namely: RR, RU
1
, U
1
R, U
1
U
1
, and U
2
. Let a
n
denote
the number of walks that contain a point on the line x +y = n (so
a
2
= 5). Show that a
n
= F
2n
, where F
n
are the Fibonacci numbers
starting with F
0
= F
1
= 1.
Problem 14M. Count pairs (f, g) of functions f, g, where f maps
1, 2, . . . , r to 1, 2, . . . , n and g is a permutation of 1, 2, . . . , n
that xes the image of f pointwise. Doing this in two ways prove
that for n r
n

k=1
_
n
k
_
k
r
d
nk
= B(r) n!,
where d
m
denotes the number of derangements of 1, 2, . . . , m and
B(r) is a Bell number.
Problem 14N. Consider walks in the X-Y plane where each step
is R :(x, y) (x + 1, y), U :(x, y) (x, y + 1), or D: (x, y)
(x +1, y +1). We wish to count walks from (0,0) to (n, n) that are
below or on the line x = y. Find a recurrence for the number of
walks like (14.10).
150 A Course in Combinatorics
Notes.
For an extensive treatment of the material in this chapter, we
refer to Goulden and Jackson (1983) and Stanley (1986). A very
readable survey is Stanley (1978).
The rst of the references mentioned above has much on the the-
ory of formal power series. For this topic also see Niven (1969) and
Appendix 2. Generally speaking, operations with power series are
formal and require no discussion of convergence when the coe-
cient of any monomial x
n
, say, in a sum or product is determined
by a nite number of operations (i.e. no limits). For example,

n=0
(
1
2
+ x)
n
is not allowed in formal operations since even the
constant term is not a nite sum; but

n=0
(n!x + x
2
)
n
is allowed
(even though it does not converge for x ,= 0) since the coecient
of any power of x requires only a nite computation (cf. Problem
14F).
The rst linear recurrence that one usually learns to solve is
a
n+1
= a
n
+ a
n1
. It leads to the famous Fibonacci sequence; cf.
Problem 14A. As we observed in Chapter 10, this name is due to
Lucas. The sequence is related to a problem occurring in the book
Liber abaci (1203) by Leonardo of Pisa (known as Fibonacci).
The number a
n
of polyominoes of size n, treated in Example
14.5, was shown to be less than c
n
(where we used c = 15). Note
that this assertion also follows from Feketes lemma. We derived
the inequality by using a suitable coding for the cells. The reader
may wish to try to prove that a(m, n) < m c
n
for a suitable
c by using (14.3) and induction. The remark following (14.9) is
based on a statement in Stanley (1986), attributing an unpublished
combinatorial proof to D. Hickerson.
The result of Example 14.6 is related to the fact that F
p
n has
as subelds the elds F
p
d, where d divides n. From this, one nds
that x
p
n
x is the product of all the irreducible polynomials over
F
p
with a degree d dividing n. This yields the formula proved in
the example.
Despite the fact that there is extensive literature on all the prob-
lems with the Catalan numbers as solution, the problems reappear
regularly (e.g. in problem sections of journals). As stated earlier,
the Belgian mathematician E. Catalan (18141894) studied well-
14. Recursions and generating functions 151
formed bracketings (as in Example 14.7).
The reection principle of Fig. 14.2 was used by the French com-
binatorialist D. Andre (18401917) in his solution of Bertrands
famous ballot problem: if, at the end of an election, candidate P
has p votes and Q has q votes, p < q, then the probability that Q
was ahead all the time during the election is (q p)/(q +p).
The Lagrange inversion formula is one of the many contributions
to analysis by J. L. Lagrange (see notes to Chapter 19). For a
proof see 7.32 in Whittaker and Watson (1927). The theorem was
published in 1770. We also refer the reader to G. N. Raney (1960).
References.
E. Catalan (1838), Note sur une equation aux dierences nies, J.
M. Pures Appl. 3, 508516.
I. P. Goulden and D. M. Jackson (1983), Combinatorial Enumera-
tion, Wiley-Interscience.
A. Joyal (1981), Une theorie combinatoire des series formelles, Ad-
vances in Mathematics 42, 182.
D. A. Klarner (1967), Cell growth problem, Canad. J. Math. 19,
851863.
I. Niven (1969), Formal power series, Amer. Math. Monthly 76,
871889.
G. N. Raney (1960), Functional composition patterns and power
series reversion, Trans. Amer. Math. Soc. 94, 441451.
R. P. Stanley (1978), Generating functions, pp. 100141 in Studies
in Combinatorics (G.-C. Rota, ed.), Studies in Math. 17, Math.
Assoc. of America.
R. P. Stanley (1986), Enumerative Combinatorics, Vol. I, Wads-
worth and Brooks/Cole.
E. T. Whittaker and G. N. Watson (1927), A Course in Modern
Analysis, Cambridge University Press.
15
Partitions
We have considered several partition problems in the previous chap-
ters. We now come to the most dicult one, namely the problem
of unordered partitions of n into k parts, respectively any number
of parts. We dene p
k
(n) as the number of solutions of
(15.1) n = x
1
+x
2
+ +x
k
, x
1
x
2
x
k
1.
For example, 7 = 5 + 1 + 1 = 4 + 2 + 1 = 3 + 3 + 1 = 3 + 2 + 2, so
p
3
(7) = 4.
Using the same idea as in the Corollary to Theorem 13.1, we see
that p
k
(n) equals the number of solutions of n k = y
1
+ + y
k
with y
1
y
k
0. If exactly s of the integers y
i
are positive,
then by (15.1) there are p
s
(n k) solutions (y
1
, . . . , y
k
). Therefore
(15.2) p
k
(n) =
k

s=1
p
s
(n k).
Problem 15A. Show that p
k
(n) = p
k1
(n 1) + p
k
(n k) and
use this to prove (15.2).
Since we have the trivial initial conditions p
k
(n) = 0 for n < k
and p
k
(k) = 1, we can recursively calculate the numbers p
k
(n).
Clearly p
1
(n) = 1 and p
2
(n) = n/2|.
Example 15.1. We shall show that p
3
(n) =
n
2
12
, i.e. the number
nearest to
n
2
12
, using a method that can be used for other values of
k. Let a
3
(n) denote the number of solutions of n = x
1
+ x
2
+ x
3
,
x
1
x
2
x
3
0. Then a
3
(n) = p
3
(n + 3) and writing y
3
= x
3
,
y
2
= x
2
x
3
, y
1
= x
1
x
2
, we see that a
3
(n) is the number of
15. Partitions 153
solutions of n = y
1
+ 2y
2
+ 3y
3
, y
i
0, i = 1, 2, 3. Therefore (cf.
Example 14.2)
(15.3)

n=0
a
3
(n)x
n
= (1 x)
1
(1 x
2
)
1
(1 x
3
)
1
.
Let = e
2i/3
. The partial fraction decomposition of (15.3) yields
(15.4)

n=0
a
3
(n)x
n
=
1
6
(1 x)
3
+
1
4
(1 x)
2
+
17
72
(1 x)
1
+
1
8
(1 +x)
1
+
1
9
(1 x)
1
+
1
9
(1
2
x)
1
.
Using (10.6), we nd from (15.4)
a
3
(n) =
1
12
(n + 3)
2

7
72
+
(1)
n
8
+
1
9
(
n
+
2n
),
and this implies that
[a
3
(n)
1
12
(n + 3)
2
[
7
72
+
1
8
+
2
9
<
1
2
.
Therefore
(15.5) p
3
(n) =
1
12
n
2
.
Problem 15B. We choose three vertices of a regular n-gon and
consider the triangle they form. Prove directly that there are

1
12
n
2
mutually incongruent triangles that can be formed in this
way, thereby providing a second proof of (15.5).
We shall now show that (15.5) is an example of a more general
result.
Theorem 15.1. If k is xed then
p
k
(n)
n
k1
k!(k 1)!
(n ).
154 A Course in Combinatorics
Proof: (i) If n = x
1
+ + x
k
, x
1
x
k
1, then the
k! permutations of (x
1
, . . . , x
k
) yield solutions of (13.2) in positive
integers, not necessarily all dierent. So
(15.6) k! p
k
(n)
_
n 1
k 1
_
.
If n = x
1
+ + x
k
, x
1
x
k
1, then if y
i
:= x
i
+ (k i),
1 i k , the integers y
i
are distinct and y
1
+ +y
k
= n+
k(k1)
2
.
Therefore
(15.7) k! p
k
(n)
_
n +
k(k1)
2
1
k 1
_
.
The result follows from (15.6) and (15.7).
Problem 15C. Let a
1
, a
2
, . . . , a
t
be positive integers (not neces-
sarily distinct) with greatest common divisor 1. Let f(n) denote
the number of solutions of
n = a
1
x
1
+a
2
x
2
+ +a
t
x
t
in nonnegative integers x
1
, . . . , x
t
. What is the generating function
F(x) :=

f(n)x
n
? Show that f(n) cn
t1
for some constant c
and explicitly give c as a function of a
1
, . . . , a
t
.
The problem that we considered in the Corollary to Theorem
13.1 concerns what is called the number of compositions of n into
k parts. By this corollary the total number of compositions of n
equals

n
k=1
_
n1
k1
_
= 2
n1
. The method that we used in Theorem
13.1 shows this directly as follows: consider n blue balls with a
space between any two of them in which we can place a red ball
if we wish. The total number of congurations that can be made
in this way is clearly 2
n1
and they represent all compositions of
n. We could have proved the same result without our knowledge of
Chapter 13 as follows. Let c
nk
denote the number of compositions
of n into k parts. Dene
c
k
(x) :=

n=k
c
nk
x
n
.
15. Partitions 155
Then, using a method that we have used several times before, we
nd
c
k
(x) = (x +x
2
+x
3
+ )
k
= x
k
(1 x)
k
=

n=k
_
n 1
k 1
_
x
n
and

k=1
c
k
(x) =

k=1
x
k
(1 x)
k
=
x
1 2x
,
i.e. n has 2
n1
compositions.
Problem 15D. Show that in a list of all 2
n1
compositions of n,
n 4, the integer 3 occurs exactly n 2
n5
times.
The most dicult and mathematically the most interesting and
important function of all in this area of combinatorics is dened by
(15.8) p(n) := number of unordered partitions of n.
For example, p(5) = 7, the partitions of 5 being
1+1+1+1+1 = 2+1+1+1 = 3+1+1 = 2+2+1 = 4+1 = 3+2 = 5.
Clearly, p(n) is the number of solutions of
(15.9) n = x
1
+x
2
+ +x
n
(x
1
x
2
. . . x
n
0).
As we saw above, this can also be formulated as the number of
solutions of n = y
1
+2y
2
+ +ny
n
, where y
i
0, 1 i n. (Here
y
i
denotes the number of terms x
k
equal to i in the rst denition.)
The study of the so-called partition function p(n) has led to some
of the most fascinating areas of analytic number theory, analysis,
algebraic geometry, etc. We shall obviously only be able to look at
properties that are proved by combinatorial arguments.
Theorem 15.2. The generating function for the partition function
is
P(x) :=

n=0
p(n)x
n
=

k=1
(1 x
k
)
1
.
156 A Course in Combinatorics
Proof: We again use the idea of Example 14.2. The partition of
n =

m
i=1
ir
i
with r
i
terms equal to i (for 1 i m) corresponds
to the term x
r
1
1
x
r
m
m
in the (formal) innite product

(1 +x
k
+
x
2
k
+ ). Replacing x
k
by x
k
, we nd the result.
Many theorems about partitions can be proved easily by repre-
senting each partition by a diagram of dots, known as a Ferrers
diagram. Here we represent each term of the partition by a row
of dots, the terms in descending order, with the largest at the top.
Sometimes it is more convenient to use squares instead of dots (in
this case the diagram is called a Young diagram by some authors
but we shall use Ferrers diagram as the name for the gure). For
example, the partition (5,4,2,1) of 12 is represented by each of the
diagrams of Fig. 15.1.

Figure 15.1
The partition we get by reading the Ferrers diagram by columns
instead of rows is called the conjugate of the original partition. So
the conjugate of 12 = 5 + 4 + 2 + 1 is 12 = 4 + 3 + 2 + 2 + 1. The
relationship is symmetric. We rst show a few easy examples of
the use of a Ferrers diagram.
Theorem 15.3. The number of partitions of n into parts, the
largest of which is k, is p
k
(n).
Proof: For each partition for which the largest part is k, the
conjugate partition has k parts (and vice versa).
Problem 15E. Show with a Ferrers diagram that the number of
partitions of n + k into k parts equals the number of partitions of
n into at most k parts. (This is (15.2).)
In some cases a proof using generating functions requires less
ingenuity than using a Ferrers diagram would. The next theorem
can be proved with a diagram similar to a Ferrers diagram though
we shall use generating functions instead.
15. Partitions 157
Theorem 15.4. The number of partitions of n into odd parts
equals the number of partitions of n into unequal parts.
Proof: The generating function for the number of partitions of n
into odd parts is (by an obvious generalization of Theorem 15.2)
equal to

m=1
(1 x
2m1
)
1
, and the generating function for the
number of partitions of n into unequal parts is (by the same argu-
ment as used in the proof of Theorem 15.2)

k=1
(1 +x
k
). Since

k=1
(1 +x
k
) =

k=1
(1 x
2k
)
(1 x
k
)
=

k=1
(1 x
2k
)

l=1
(1 x
l
)
1
=

m=1
(1 x
2m1
)
1
,
the proof is complete.
Now consider the function (P(x))
1
=

k=1
(1 x
k
). In the ex-
pansion of this product, a partition of n into unequal parts con-
tributes +1 to the coecient of x
n
if the number of parts is even,
and 1 if the number of parts is odd. So the coecient of x
n
is
p
e
(n) p
o
(n), where p
e
(n), respectively p
o
(n), denotes the number
of partitions of n into an even, respectively odd, number of unequal
parts. It was shown by Euler that p
e
(n) = p
o
(n) unless n has the
form n = (m) := (3m
2
m)/2 or n = (m) = (3m
2
+ m)/2,
in which case p
e
(n) p
o
(n) = (1)
m
. The numbers (m) and
(m) are called pentagonal numbers. This name is explained by
the relation (m) =

m1
k=0
(3k + 1) and Fig. 15.2.
Figure 15.2
The following extremely elegant pictorial proof of Eulers identity
is due to Franklin (1881).
158 A Course in Combinatorics
Theorem 15.5. We have

k=1
(1 x
k
) = 1 +

m=1
(1)
m
_
x
(m)
+x
(m)
_
.
Proof: Consider a Ferrers diagram of a partition of n into unequal
parts as for 23 = 7 + 6 + 5 + 3 + 2 in Fig. 15.3.
Figure 15.3
The last row is called the base of the diagram. The number of
dots in the base is denoted by b. In Fig. 15.3 the base is indicated by
a line segment with b = 2 dots. The longest 45

line segment joining

the last point in the top row with other points of the diagram is
called the slope of the diagram. Its number of points is denoted
by s. In Fig. 15.3 the slope is indicated by a segment with s = 3
points. We now dene two operations on the diagram, which we
call A and B.
A: If b s, then remove the base and adjoin it to the diagram
at the right to form a new slope parallel to the original one, unless
b = s and the base and the slope have a point in common. This
exception is considered below.
B: If b > s, remove the slope and adjoin it at the bottom of the
diagram as a new base, except if b = s + 1 and the base and the
slope have a point in common.
In Fig. 15.3 operation A can be carried out and it yields the
partition 23 = 8+7+5+3. An example of the exceptional situation
for operation B is shown in Fig. 15.4.
Figure 15.4
15. Partitions 159
The exception for operation A can occur only if
n = b + (b + 1) + + (2b 1),
i.e. n = (b). The exception for operation B occurs if n = (s +
1) + + 2s, i.e. n = (s). In all other cases exactly one of
the operations can be carried out and we thus have a one-to-one
correspondence between partitions of n into an even number of
unequal parts and partitions of n into an odd number of unequal
parts; and hence for these values of n, we have p
e
(n) p
o
(n) = 0.
In the exceptional cases the dierence is +1 or 1.
Euler used Theorem 15.5 to nd the following recursion formula
for p(n).
Theorem 15.6. Let p(n) = 0 for n < 0. Then for n 1,
(15.10) p(n) =

m=1
(1)
m+1
p(n (m)) +p(n (m)) .
Proof: This is an immediate consequence of Theorem 15.2 and
Theorem 15.5.
Note that the sum in (15.10) is nite. With this recursion one
can rapidly generate a table for p(n) for small values of n.
The rst two terms of the recursion formula are the same as in
the famous Fibonacci recursion (cf. Problem 5E). Although these
are followed by two negative terms, one might think that p(n) will
increase in a way similar to the Fibonacci numbers, i.e. as c
n
for
some constant c. This is not the case. We shall show that p(n)
grows much more slowly. The actual asymptotics of p(n) involve
complicated methods from analytic number theory. We mention
only the main term of the asymptotic formula:
lim
n
n

1
2
log p(n) =
_
2
3
.
One can show that in fact p(n) is much smaller than exp
_

_
2
3
n
_
.
160 A Course in Combinatorics
Theorem 15.7. For n > 2 we have
p(n) <

_
6(n 1)
e

2
3
n
.
Proof: Let f(t) := log P(t). From Theorem 15.2 we nd
f(t) =

k=1
log(1 t
k
) =

k=1

j=1
t
kj
j
=

j=1
j
1
t
j
1 t
j
.
From now on let 0 < t < 1. Then from
(1 t)
1
(1 t
j
) = 1 +t + +t
j1
> jt
j1
,
we nd
f(t) <
t
1 t

j=1
j
2
=
1
6

2
t
1 t
.
Since p(n) is increasing, we have P(t) > p(n)t
n
(1 t)
1
. By com-
bining the two inequalities and then substituting t = (1 +u)
1
, we
nd
log p(n) < f(t) nlog t + log(1 t)
<

2
6

t
1 t
nlog t + log(1 t)
=

2
6
u
1
+nlog(1 +u) + log
u
1 +u
.
Therefore
log p(n) <
1
6

2
u
1
+ (n 1)u + log u.
We get the required inequality by substituting u = 6(n 1)

1
2
.

Eulers identity is a special case of a famous identity of Jacobi

that is important in the theory of theta functions. We would not
mention this so-called Jacobi triple product identity if we had not
known of the beautiful combinatorial proof using a Ferrers diagram,
due to Wright (1965).
15. Partitions 161
Theorem 15.8. We have

n=1
(1 q
2n
)(1 +q
2n1
t)(1 +q
2n1
t
1
) =

r=
q
r
2
t
r
.
Proof: We rewrite the assertion as

n=1
(1 +q
2n1
t)(1 +q
2n1
t
1
) =

r=
q
r
2
t
r

n=1
(1 q
2n
)
1
,
and substitute x = qt, y = qt
1
. This yields the relation
(15.11)

n=1
_
(1 +x
n
y
n1
)(1 +x
n1
y
n
)
_
=

r=
x
1
2
r(r+1)
y
1
2
r(r1)

n=1
(1 x
n
y
n
)
1
.
We shall prove this relation by interpreting both sides as generating
functions corresponding to the counting of appropriate combinato-
rial objects, and then showing a one-to-one mapping between these
objects. For the left-hand side, we nd as a combinatorial interpre-
tation the generating function for the number of partitions of the
Gaussian integer n + mi into parts a + (a 1)i and (b 1) + bi
(a 1, b 1), with no two parts equal. Call this number (n, m).
So the left-hand side of (15.11) is

n=1

m=1
(n, m)x
n
y
m
. On
the right-hand side, we use Theorem 15.2 and replace the product
by

k=1
p(k)x
k
y
k
. We must therefore prove that (n, m) = p(k),
where n = k +
1
2
r(r + 1) and m = k +
1
2
r(r 1). Without loss of
generality, we may assume that n m, i.e. r 0. A partition of
n +mi must have v 0 terms of type (b 1) +bi and v +r terms
of type a +(a 1)i, and therefore n
1
2
r(r +1), so k 0. In Fig.
15.5 below, we consider the example (n, m) = (47, 44), so r = 3
and k = 41. We have taken the Ferrers diagram of the partition
41 = 12 + 10 + 8 + 5 + 2 + 2 + 2 and above the top row we have
added rows of length r, r 1, . . . , 1.
162 A Course in Combinatorics
Figure 15.5
The resulting diagram is split into two parts. The shaded part on
the left is determined by the staircase starting with the part that
was added at the top. We read the shaded part by columns. By the
construction, the sequence is decreasing. These numbers are the as
for terms of type a + (a 1)i. The unshaded part is read by rows
and again the numbers are decreasing. These are the b 1s for
terms of type (b1)+bi (the last b could be 1, corresponding to one
empty unshaded row). The number of as exceeds the number of
b1s by r and the sum of the as and b1s is clearly k+
1
2
r(r+1).
So we have indeed produced a partition of n + mi of the required
type. It is easy to see that the procedure can be reversed, i.e. we
have dened the one-to-one mapping we were looking for.
Problem 15F. Prove that the number of self-conjugate partitions
of n equals the number of partitions of n into unequal odd parts.
As a nal problem related to partitions we shall consider objects
known as Young tableaux or standard tableaux. A Young tableau of
shape (n
1
, n
2
, . . . , n
m
) is a Ferrers diagram (or Young diagram) of
squares in which the integers 1 to n have been inserted (one in each
square) in such a way that all rows and columns are increasing. For
example, in Fig. 15.6 we display a Young tableau of shape (5,4,2,1).
15. Partitions 163
1 3 4 7 11
2 5 10 12
6 9
8
Figure 15.6
We are interested in determining the number of Young tableaux
of a given shape. At rst, this may seem like a somewhat unnatural
problem. However, Young tableaux play an important role in the
theory of group representations (and other areas). One of the inter-
esting facts concerning these tableaux is that there is a one-to-one
correspondence between the Young tableaux with n squares (that
we shall call cells) and involutions of 1 to n, where we include
the identity as an involution. (Note that we can therefore count
the total number of tableaux using Problem 14G.) For a treatment
of this one-to-one correspondence and several related problems we
refer to D. Knuth (1973).
In order to count the Young tableaux of a given shape, we need
to introduce the function (x
1
, . . . , x
m
) dened by
(15.12) (x
1
, . . . , x
m
) :=

1i<jm
(x
i
x
j
).
(Note that is the value of the Vandermonde determinant.)
Lemma 15.9. Let
g(x
1
, . . . , x
m
; y) := x
1
(x
1
+y, x
2
, . . . , x
m
)
+x
2
(x
1
, x
2
+y, . . . , x
m
) + +x
m
(x
1
, x
2
, . . . , x
m
+y).
Then
g(x
1
, . . . , x
m
; y) =
_
x
1
+ +x
m
+
_
m
2
_
y
_
(x
1
, . . . , x
m
).
Proof: Clearly the function g is a homogeneous polynomial of
degree 1 + deg (x
1
, . . . , x
m
) in the variables x
1
, . . . , x
m
, y. If we
interchange x
i
and x
j
, then g changes sign. So if x
i
= x
j
, then
164 A Course in Combinatorics
g must be 0, i.e. g is divisible by x
i
x
j
and hence by the poly-
nomial (x
1
, . . . , x
m
). If y = 0, the assertion is obvious. There-
fore we only have to prove the coecient
_
m
2
_
for y. If we ex-
pand g, then the terms of degree 1 in y are
x
i
y
x
i
x
j
(x
1
, . . . , x
m
) and
x
j
y
x
i
x
j
(x
1
, . . . , x
m
) for all pairs (i, j) with 1 i < j m. The sum
of these terms is clearly
_
m
2
_
(x
1
, . . . , x
m
).
We introduce a function f dened on all m-tuples (n
1
, . . . , n
m
),
m 1, with the following properties:
f(n
1
, . . . , n
m
) = 0 unless n
1
n
2
0;
(15.13)
f(n
1
, . . . , n
m
, 0) = f(n
1
, . . . , n
m
);
(15.14)
f(n
1
, . . . , n
m
) = f(n
1
1, n
2
, . . . , n
m
)+
(15.15)
f(n
1
, n
2
1, . . . , n
m
) + +f(n
1
, n
2
, . . . , n
m
1),
if n
1
n
2
n
m
0;
f(n) = 1 if n 0. (15.16)
Clearly f is well dened. We claim that f(n
1
, . . . , n
m
) counts the
number of Young tableaux of shape (n
1
, . . . , n
m
). Condition (15.13)
is obvious and so are (15.14) and (15.16). To see that the number of
Young tableaux of shape (n
1
, . . . , n
m
) satises (15.15), we consider
the entry n. It must be the last entry in one of the rows, and if
we remove the square with n, then we obtain a Young tableau for
n 1. In fact, if two or more rows of the tableau have the same
length, then n can only be the last entry in the lowest of these; but
including all terms in (15.15) does not matter since if, say, n
1
= n
2
,
then f(n
1
1, n
2
, . . . , n
m
) = 0.
Theorem 15.10. The number of Young tableaux that have shape
(n
1
, . . . , n
m
) satises
(15.17) f(n
1
, . . . , n
m
) =
(n
1
+m1, n
2
+m2, . . . , n
m
) n!
(n
1
+m1)!(n
2
+m2)! . . . n
m
!
,
and in fact this formula for f is correct if n
1
+m1 n
2
+m2
n
m
.
15. Partitions 165
Proof: We rst observe that, if for some i we have n
i
+ mi =
n
i+1
+ m i 1, then the expression on the right-hand side
of (15.17) is 0, in accordance with the fact that the shape is not
allowed. We must show that the right-hand side of (15.17) satises
(15.13) to (15.16). All but (15.15) are trivial. We nd the relation
(15.15) by substituting x
i
= n
i
+mi and y = 1 in Lemma 15.9.

There is a formulation of this theorem that makes it more in-

teresting. We introduce the concept of hook in a Young tableau.
The hook corresponding to a cell in a tableau is the union of the
cell and all cells to the right in the same row and all cells below it
in the same column. The hook-lengthis the number of cells in the
hook. In Fig. 15.7 we show a Young diagram with in each cell a
number equal to the length of the corresponding hook. The shaded
area is the hook corresponding to the cell in row 2 and column 3.
Figure 15.7
We now have the following remarkable theorem due to J. S.
Frame, G. de Beauregard Robinson and R. M. Thrall (1954).
Theorem 15.11. The number of Young tableaux of a given shape
and a total of n cells is n! divided by the product of all hook-lengths.
Proof: If we have a Young tableau of shape (n
1
, . . . , n
m
), then the
hook-length for the cell in row 1 and column 1 is n
1
+m1. Consider
the hook-lengths for the cells in row 1. In Fig. 15.7 the second
column has the same length as column 1 and hence the hook-length
for the next cell is n
1
+ m 2. For the next cell the length drops
by 2 because of the missing cell indicated by a dot in the last row.
Similarly, there is a jump of 3 for the column with two dots. So in
row 1 we nd as hook-lengths the numbers from 1 to n
1
+m1 with
166 A Course in Combinatorics
the exception of the numbers (n
1
+m1)(n
j
+mj), 2 j m.
Similarly row i contains the integers from 1 to n
i
+mi except for
(n
i
+mi)(n
j
+mj), i+1 j m. It follows that the product
of the hook-lengths is

m
i=1
(n
i
+mi)!/(n
1
+m1, . . . , n
m
).
By Lemma 15.9 we are done.
Problem 15G. Consider a Young tableau of shape (n, n). Dene
a sequence a
k
, 1 k 2n, by a
k
:= i if k is in row i of the tableau,
i = 1, 2. Use this to show that the number of Young tableaux of
this shape is the Catalan number u
n+1
(in accordance with Theorem
15.11).
Problem 15H. How many Ferrers diagrams are there that t in
a box of size k n k?
Problem 15I. Prove that the number of partitions of n into parts
not divisible by d equals the number of partitions in which no part
occurs more than d 1 times.
Problem 15J. Dene P
n
:=

n
k=1
k!. Prove that P
2
n1
(n
2
)! is
divisible by P
2n1
.
Notes.
The use of Ferrers diagrams to prove theorems such as Theorem
15.3 was introduced by Sylvester in 1853. He wrote that the proof
had been communicated to him by N. M. Ferrers.
For other proofs involving Ferrers diagrams and analogues (e.g.
a proof of Theorem 15.4) we refer to Hardy and Wright (1954),
MacMahon (1916), and Van Lint (1974).
Eulers proof of Theorem 15.5 was by induction. Polygonal num-
bers occurred in ancient mathematics, e.g. triangular numbers were
considered by Pythagoras before 500 BC.
Theorem 15.7 is due to Van Lint (1974).
The asymptotic behavior of the partition function p(n) was given
by G. H. Hardy and S. Ramanujan (1918) in the paper in which
they developed the famous circle method that has had so many ap-
plications in number theory. A further asymptotic result was given
by H. Rademacher (1937). The proofs depend on the remarkable
15. Partitions 167
functional equation for the famous Dedekind -function
(z) := e
iz
12

n=1
_
1 e
2inz
_
.
For an exposition of these proofs see Chandrasekharan (1970).
C. G. J. Jacobi (18041851) is best known for developing the
theory of elliptic functions. He became professor at K onigsberg at
the age of 23. Two years later he published his book Fundamenta
Nova Theoriae Functionum Ellipticarum. In 64 we nd what is
now known as the triple product identity.
The rst to use tableaux such as in Fig. 15.6 was Frobenius! A
year later (1901) A. Young independently introduced them in his
work on matrix representations of permutation groups. The name
Young tableaux has become standard terminology.
References.
K. Chandrasekharan (1970), Arithmetical Functions, Springer-Ver-
lag.
J. S. Frame, G. de Beauregard Robinson, and R. M. Thrall (1954),
The hook graphs of S
n
, Canad. J. Math. 6, 316324.
F. Franklin (1881), Sur le developpement du produit inni (1
x)(1x
2
)(1x
3
)(1x
4
) , Comptes Rendus Acad. Sci. (Paris)
92, 448450.
G. H. Hardy and S. Ramanujan (1918), Asymptotic formulae in
combinatory analysis, Proc. London Math. Soc. (2) 17, 75115.
G. H. Hardy and E. M. Wright (1954), An Introduction to the
Theory of Numbers, 3d edn., Clarendon Press.
D. Knuth (1973), The Art of Computer Programming, Vol. 3, Addi-
son-Wesley .
J. H. van Lint (1974), Combinatorial Theory Seminar Eindhoven
University of Technology, Lecture Notes in Math. 382, Springer-
Verlag.
P. A. MacMahon (1916), Combinatory Analysis Vol. II, Cambridge
University Press.
H. Rademacher (1937), On the partition function p(n), Proc. Lon-
don Math. Soc. 43, 241254.
168 A Course in Combinatorics
E. M. Wright (1965), An enumerative proof of an identity of Jacobi,
J. London Math. Soc. 40, 5557.
A. Young (1901), On quantitative substitutional analysis, Proc.
London Math. Soc. 33, 97146.
16
(0,1)-Matrices
In the previous chapters (0,1)-matrices have turned up a number
of times; and especially those with constant line-sums led to in-
teresting problems. In this chapter we shall consider the existence
problem for (0,1)-matrices with given line-sums, and try to count
or estimate how many there are in the case of constant line-sums.
In the rst problem, the matrices are not necessarily square. If
the row-sums of a matrix A are r
1
, r
2
, . . . , r
k
, then we shall call the
vector r := (r
1
, r
2
, . . . , r
k
) the row-sum of A, and similarly for the
column-sums.
We consider the problem of the existence of a (0, 1)-matrix with
given row-sum r and column-sum s. For convenience, we shall
assume that the coordinates of r and s are nonincreasing; that is,
r and s are partitions (we may allow trailing zero coordinates in
partitions for the purposes of this chapter).
Given two partitions r = (r
1
, r
2
, . . . , r
n
) and s = (s
1
, s
2
, . . . , s
m
)
of the same integer N, we say that r majorizes s when
r
1
+r
2
+ +r
k
s
1
+s
2
+ +s
k
for all k, interpreting r
k
or s
k
as 0 when k exceeds n or m, respec-
tively. Recall that the conjugate of a partition r is the partition r

where r

i
is the number of j such that r
j
i.
Problem 16A. Prove the following assertion: r majorizes s if and
only if the latter can be obtained from the former by a sequence of
operations pick two parts (coordinates) a and b so that a b + 2
and replace them by a 1 and b +1. (That is, s is more average
than r.) For example, (5, 4, 1) majorizes (3, 3, 3, 1) and the latter
170 A Course in Combinatorics
can be obtained from the former in several ways with sequences of
such operations, including
(5, 4, 1, 0) (5, 3, 2, 0) (5, 3, 1, 1) (4, 3, 2, 1) (3, 3, 3, 1)
and
(5, 4, 1, 0) (4, 4, 1, 1) (4, 3, 2, 1) (3, 3, 3, 1).
Theorem 16.1. Let r
1
, r
2
, . . . , r
n
and s
1
, s
2
, . . . , s
m
be two nonin-
creasing sequences of nonnegative integers each summing to a com-
mon value N. There exists an n m (0, 1)-matrix with row-sum
r and column-sum s if and only if r

majorizes s.
Proof: To prove necessity, let k be given and consider the rst k
columns of a hypothetical (0, 1)-matrix with row-sumr and column-
sum s. The number of 1s in these columns is
s
1
+s
2
+ +s
k

n

i=1
min(k, r
i
) =
k

j=1
r

j
.
The latter equality is most evident from the Ferrers diagram of the
partition r (see Fig. 16.1) where we see both expressions count the
number of cells in the rst k columns.
Figure 16.1
We now introduce a transportation network with a source S,
a sink T, one vertex x
i
for each row, and one vertex y
j
for each
column. There is to be an edge of capacity r
i
directed from S to
x
i
, 1 i n, an edge of capacity s
j
from y
j
to T, 1 j m, and
an edge of capacity 1 from x
i
to y
j
, 1 i n, 1 j m. We claim
16. (0, 1)-Matrices 171
that a (0, 1)-matrix M = (a
ij
) with row-sum r and column-sum s
exists if and only if this network admits a ow of strength N (there
are at least two cuts of capacity N, so this would be a maxow).
Given such a matrix, we get a ow of strength N by saturating the
edges incident with S and T and assigning ow a
ij
to the edge from
x
i
to y
j
. Conversely, if the network admits a ow of strength N,
then there exists an integral ow of that strength (cf. Theorem 7.2);
clearly, the edges incident with either S or T must be saturated,
and on the other edges the ow must be either 0 or 1.
Consider a cut A, B in this network which separates S and T.
Say A consists of S, n
0
vertices of X := x
i
: i = 1, . . . , n, and m
0
vertices of Y := y
i
: i = 1, . . . , m. The edges crossing from A to
B include nn
0
edges leaving S, m
0
edges into T, and n
0
(mm
0
)
edges from X to Y . The capacity of this cut is at least
r
n
0
+1
+r
n
0
+2
+ +r
n
+s
mm
0
+1
+s
mm
0
+2
+ +s
m
+n
0
(mm
0
).
Figure 16.2
It is again convenient to refer to the Ferrers diagram of the parti-
tion r (see Fig. 16.2). The number of cells in the Ferrers diagram is
N. So clearly N is at most n
0
(mm
0
) plus the number of cells in
the last nn
0
rows plus the number of cells in the last m
0
columns.
The number of cells in the last m
0
columns is the sum of the last
m
0
parts of the conjugate r

. But under the assumption that r

majorizes s,
r

mm
0
+1
+r

mm
0
+2
+ +r

m
s
mm
0
+1
+s
mm
0
+2
+ +s
m
,
172 A Course in Combinatorics
and we conclude that the capacity of any cut is at least N. By
the maxow-mincut theorem, Theorem 7.1, the required ow of
strength N exists.
We have given the above proof in order to establish the connec-
tion between ows and (0, 1)-matrices, but it is possible to give a
more direct proof of suciency of the condition in Theorem 16.1.
Our exposition is similar to that given by Krause (1996).
First suppose we have a (0, 1)-matrix A
1
with row-sum r and
column-sum s
1
, and that s
2
is obtained from s
1
by the operation
of Problem 16A. Then we claim it is easy to construct a (0, 1)-
matrix A
2
with row-sum r and column-sum s
2
: If the parts a, b
with a b +2 in s
1
(that are replaced by a 1 and b +1 to obtain
s
2
) are the sums of columns j and k, respectively, there must be a
row i so that the (i, j)-entry in A
1
is 1 and the (i, k)-entry in A
1
is 0; if we interchange these two entries, we obtain a matrix A
2
as
desired.
Now assume r

majorizes s. Start with the matrix A

0
correspond-
ing to the Ferrers diagram of r, adding columns of 0s if necessary
to obtain size n m. The column-sum of A
0
is r

. If r

majorizes
s, we can nd a sequence of partitions
r

= s
0
, s
1
, s
2
, . . . , s

= s
so that each partition in the sequence is obtained from the pre-
ceding one by the operation of Problem 16A. Then a (0, 1)-matrix
A with row-sum r and column-sum s can, by the observation of
the preceding paragraph, be obtained from A
0
by a sequence of
interchanges.
Example 16.1. Take r = (3, 2, 2, 2, 1) and s = (3, 3, 3, 1). Then
r

= (5, 4, 1), and this majorizes s. The algorithm described above

might give rise to the following sequence of matrices:
_
_
_
_
1 1 1 0
1 1 0 0
1 1 0 0
1 1 0 0
1 0 0 0
_
_
_
_

_
_
_
_
1 1 1 0
0 1 1 0
1 1 0 0
1 1 0 0
1 0 0 0
_
_
_
_

_
_
_
_
1 1 1 0
0 1 1 0
0 1 1 0
1 1 0 0
1 0 0 0
_
_
_
_

_
_
_
_
1 0 1 1
0 1 1 0
0 1 1 0
1 1 0 0
1 0 0 0
_
_
_
_
16. (0, 1)-Matrices 173
Here the column-sum vectors are
(5, 4, 1, 0) (4, 4, 2, 0) (3, 4, 3, 0) (3, 3, 3, 1).
(We do not care if the vectors are not always nonincreasing.)
We extend the idea of the algorithm in the theorem below, and
nd a (rough) estimate for the number of (0, 1)-matrices with cer-
tain line-sums as a consequence.
Theorem 16.2. Given partitions r and s of an integer N, let
M(r, s) denote the number of (0, 1)-matrices A with row-sum r
and column-sum s. If r majorizes r
0
and s majorizes s
0
, then
M(r
0
, s
0
) M(r, s).
Proof: We will establish the following simple observation. Let us
x a row-sum vector r = (r
1
, r
2
, . . . , r
n
) of length n and a column-
sum vector s = (s
1
, s
2
, . . . , s
m
) of length m that have equal sums.
(We do not require them to be nonincreasing.) If s
1
> s
2
, then we
claim that
M(r, (s
1
1, s
2
+ 1, s
3
, . . . , s
m
)) M(r, (s
1
, s
2
, s
3
, . . . , s
m
)).
Of course, the same will hold for any two columns (we have used the
rst two only for notational convenience). The same idea applies
to the transpose, where keeping s constant and replacing r
i
and
r
j
(r
i
> r
j
) by r
i
1 and r
j
+ 1 will not decrease the number
of associated (0, 1)-matrices. The inequality of the theorem then
follows from this observation and the result of Problem 16A.
To prove the observation, consider (0, 1)-matrices A of size n
(m 2) with column-sum (s
3
, s
4
, . . . , s
m
). For a given matrix A,
it may or may not be possible to prepend two columns to A to
get row-sum r; if it is possible, we need to add two 1s to a rows,
one 1 to b rows, and no 1s to c rows, say, where a + b + c = n
and 2a +b = s
1
+s
2
. But the number of ways to prepend the two
columns to get new column-sums s
1
and s
2
is
_
b
s
1
a
_
=
_
b
s
2
a
_
,
174 A Course in Combinatorics
and this is at most the number
_
b
s
1
1 a
_
=
_
b
s
2
+ 1 a
_
of ways to prepend the two columns to get new column-sums s
1
1
and s
2
+ 1 since the latter sums are closer to b/2.
The suciency of the condition of Theorem 16.1 is a quick corol-
lary of Theorem 16.2. If r

majorizes s, then
M(r, s) M(r, r

) 1.
Problem 16B. Prove that M(r, r

) = 1.
We will use A(n, k) to denote the number of n n (0, 1)-matrices
with all line-sums equal to k.
Corollary.
A(n, k)
__
n
k
_
/2
k+1
_
n
.
Proof: The number of (0, 1)-matrices with row-sum (k, k, . . . , k)
is
_
n
k
_
n
. Each has a column-sum s = (s
1
, s
2
, s
3
, . . . , s
n
) satisfying
0 s
i
n, s
1
+ + s
n
= nk. Ignoring the restriction s
i
n,
we see that by Theorem 13.1 there are at most
_
nk+n1
n1
_
2
n(k+1)
such column-sums. Since the greatest number of associated (0, 1)-
matrices occurs for column-sum (k, k, . . . , k), the number of such
matrices is at least the average number.
Problem 16C. Prove the following theorem.
Theorem 16.3. Let d and d
/
be two partitions of an (even) integer
N. If d majorizes d
/
, then there are at least as many labeled simple
graphs with degree sequence d
/
as with degree sequence d.
Problem 16D. If n is even, then
A(n,
1
2
n)
2
n
2
n
2n
.
Show this by modifying the proof of the Corollary to Theorem 16.2.
(Substituting k =
1
2
n in the corollary would give a poor result.)
Problem 16E. Let d = (d
1
, d
2
, . . . , d
n
) be a partition (possibly
including 0s) of
_
n
2
_
. Prove that there exists a tournament on the
16. (0, 1)-Matrices 175
vertex set 1, 2, . . . , n (an orientation of K
n
) so that vertex i has
outdegree d
i
for all i if and only if d is majorized by
(n 1, n 2, . . . , 2, 1, 0).
Problem 16F. (i) Given an integer m
_
n
2
_
, consider the graph
with vertex set 1, 2, . . . , n and whose edges are the rst m of the
2-subsets in lexicographical order:
1, 2, 1, 3, . . . , 1, n, 2, 3, 2, 4, . . . 2, n, 3, 4, . . . .
This graph has a certain degree sequence r, which is a partition of
2m. For example, when n = 8 and m = 20, r = (7, 7, 7, 5, 4, 4, 3, 3).
Given n, m
_
n
2
_
, and a partition d (possibly including 0s) of
2m, prove that there exists a simple graph on n vertices with de-
gree sequence d if and only if d is majorized by the partition r as
described above.
(ii) Generalize for simple k-uniform hypergraphs (a set of points
and a set of k-element subsets).
* * *
As in Chapter 11, let /(n, 2) denote the set of (0, 1)-matrices of
size n with all line-sums equal to 2. A method that we learned in
Chapter 14 allows us to count the number A(n, 2) of elements of
/(n, 2).
Let /(n, 2)

denote the subset of /(n, 2) consisting of indecom-

posable matrices (cf. Chapter 11). Dene a
n
:= [/

(n, 2)[. It is
easy to see that a
n
=
1
2
n! (n 1)! as follows. We have
_
n
2
_
choices
for the rst row. If we choose (1 1 0 . . . 0) as rst row, then there
is one more row (out of n 1) with a 1 in the rst column and a
second 1 not in the second column (so n2 choices). This gives us
(n 1)(n 2) possibilities, etc.
Now dene
m
n
:= A(n, 2)/(n!)
2
,
b
k
:= a
k
/(k!)
2
.
176 A Course in Combinatorics
Then by the same argument as was used in Theorem 14.2, we have
(16.1) 1 +

n=2
m
n
x
n
= exp
_

k=2
b
k
x
k
_
.
Therefore we nd
(16.2) 1 +

n=2
m
n
x
n
= exp
_
x log(1 x)
2
_
= e

1
2
x
(1 x)

1
2
.
It is an easy exercise to show that from the expansion
(1 x)

1
2
=

n=0
_
2n
n
_
_
x
4
_
n
,
we can conclude that
m
n
e

1
2
_
2n
n
_
4
n
for n .
We have thus proved the following theorem.
Theorem 16.4.
A(n, 2) e

1
2
(2n)!
(2!)
2n
.
This theorem is a special case of the following theorem (that we
shall only prove for k = 3, using a method that can be applied for
other values of k).
Theorem 16.5.
A(n, k) =
(nk)!
(k!)
2n
exp
_

(k 1)
2
2
_ _
1 +O
_
1
n
3
4
__
(n ),
uniformly in k for 1 k < log n.
For more information on this question and generalizations to the
case where the line-sums are not constant we refer to B. D. McKay
(1984).
16. (0, 1)-Matrices 177
For the case k = 3, we now prove a result slightly stronger than
obtained by setting k to 3 above. We shall use the truncated form of
the principle of inclusion and exclusion (cf. Theorem 10.1, Remark).
Consider N := 3n elements, numbered
1
a
, 1
b
, 1
c
, 2
a
, 2
b
, 2
c
, . . . , n
a
, n
b
, n
c
.
Form a permutation of these N elements and subsequently form the
corresponding ordered partition into triples: (x, y, z)(u, v, w) . . . .
We shall say that a repetition occurs if, in at least one of these
triples, some number occurs more than once, as in (5
a
, 3
b
, 5
c
). As-
sume that the chosen permutation results in a sequence of n
triples with no repetition. In that case, we associate an n n
(0, 1)-matrix with the permutation as follows. If the i-th triple is
(x

, y

, z

), where , , a, b, c, then the matrix has ones in

row i in the columns numbered x, y, z, and zeros elsewhere. Since
the indices a, b, c do not inuence the matrix, and the order of the
three elements within one triple is also irrelevant, each such matrix
corresponds to (3!)
2n
dierent permutations. Clearly the matrix is
in /(n, 3). We must nd an estimate for the number P of permu-
tations with no repetition among all N
0
:= N! permutations of the
3n elements.
Let 1 r n. Specify r triples and count the permutations that
have a repetition in these triples. Then sum this over all choices of
the r triples. As in Theorem 10.1, we call this number N
r
. Then if
R is even, we have
(16.3)
R+1

r=0
(1)
r
N
r
P
R

r=0
(1)
r
N
r
.
The dicult part of the proof is nding a suitable upper estimate
and lower estimate for N
r
.
We start with the upper estimate. The r triples can be chosen
in
_
n
r
_
ways. Within each triple we specify two positions, where we
require a repetition. This can be done in 3
r
ways. Now we choose
r numbers that will occur in the repetitions, and the indices a, b, c.
This can also be done in
_
n
r
_
3
r
ways. Subsequently we distribute
the chosen numbers over the chosen positions, and that can be
178 A Course in Combinatorics
done in 2
r
r! ways. Finally, we distribute the remaining N 2r
numbers arbitrarily. Clearly, a permutation with a repetition of
type (5
a
, 5
b
, 5
c
) has been counted more than once. So we have
N
r

_
n
r
_
2
3
2r
2
r
r! (N 2r)!
(16.4)

2
r
r!
(3
2
n
2
)
r
(N 2r)! N!
2
r
r!
_
N 2r
N
_
2r
N!
2
r
r!
_
1 +
8r
2
N
_
,
if r <
1
2

n. Here the last step is based on the power series expan-

sion of (1
2r
N
)
2r
, and choosing r in such a way that each term is
at most half of the previous term.
For the lower estimate, we start in the same way. After the r pairs
have been chosen and distributed over the chosen positions, we rst
complete the r triples that have a repetition, using elements with a
number that diers from the two repeated numbers for each triple.
This can clearly be done in more than (N 3r)
r
ways. Then we
distribute the remaining numbers. Some permutations that should
be counted are not, i.e.
N
r

_
n
r
_
2
3
2r
2
r
r! (N 3r)
r
(N 3r)!
(16.5)
N!
2
r
r!
(3
2
n
2
)
r
_
n(n 1) . . . (n r + 1)
n
r
_
2
(N 3r)
r
N
3r
N!
2
r
r!
_
1
r
n
_
2r
_
1
3r
N
_
r
N!
2
r
r!
_
1
3r
2
n
_
if r <

n. Here in the last step, we used the well known inequality

_
1
r
n
_
r
1
r
2
n
if 1 r

n.
16. (0, 1)-Matrices 179
We now combine (16.3), (16.4) and (16.5). We take R =
1
2

n|.
We nd
P
N!
= e
2
+ , where [[ <
2
R+1
(R + 1)!
+
3
n

r=0
2
r
r
2
r!
,
and hence
[[ <
48
n
+
18e
2
n
<
200
n
.
We have therefore found the following estimate (in accordance with
Theorem 16.5).
Theorem 16.6. We have
A(n, 3) =
(3n)!
(3!)
2n
e
2
_
1 +O(
1
n
)
_
(n ).
Remark. Note that for k = 3, we nd from the Corollary to The-
orem 16.2 that A(n, 3) grows at least as fast as (cn
3
)
n
, for a suitable
constant c, and this diers from Theorem 16.6 only in the constant.
Problem 16G. Determine A(5, 3).
We now describe another connection between (16.2) and Chapter
14. Note that squaring the right-hand side of (16.2) yields the
exponential generating function for the number of derangements
as given in Example 14.10. This means that we should be able to
show combinatorially that
(16.6) n!d
n
=
n

k=0
_
n
k
_
2
A
k
A
nk
,
where we have used A
n
for A(n, 2).
As a preparation, consider an indecomposable component of an
element of /(n, 2), for example the following 5 5 submatrix for
the case n = 9:
1 3 4 5 9
1 1 1 0 0 0
2 1 0 1 0 0
3 0 1 0 0 1
7 0 0 1 1 0
9 0 0 0 1 1
180 A Course in Combinatorics
Starting with the two 1s in the top row, there are two ways to
describe the cyclic structure:
(1, 3), (3, 9), (9, 5), (5, 4), (4, 1)
corresponding to the permutation (1 3 9 7 2) of the rows and
(3, 1), (1, 4), (4, 5), (5, 9), (9, 3)
corresponding to the reversed permutation (1 2 7 9 3).
The rst case (1 < 3) we call the red variety, the second the blue
variety.
Consider the set S consisting of elements of /(n, 2) with the
1s colored red or blue in such a way that in an indecomposable
component all 1s have the same color. Clearly
[S[ =
n

k=0
_
n
k
_
2
A
k
A
nk
.
We dene a one-to-one mapping from S to the set of pairs of per-
mutations (a
1
, a
2
, . . . , a
n
), (b
1
, b
2
, . . . , b
n
) of 1, 2, . . . , n in which the
rst is a derangement. There are n!d
n
such pairs. We do this by
example.
Let
a := (a
1
, a
2
, . . . , a
9
) = (2, 7, 1, 8, 4, 5, 9, 6, 3),
b := (b
1
, b
2
, . . . , b
9
) = (3, 1, 4, 5, 9, 2, 6, 8, 7).
The canonical cycle decomposition of a is (1 2 7 9 3)(4 8 6 5).
We split b accordingly into (3 1 4 5 9) and (2 6 8 7). The rst
pair describes the 5 5 blue submatrix treated above; the second
describes a red 4 4 submatrix.
This establishes (16.6).
The following problem gives one more relation with Chapter 14.
Problem 16H. Again let A
n
:= A(n, 2). Consider a matrix in
/(n, 2). There are
_
n
2
_
ways to choose the two 1s in row 1. Suppose
these are in the rst two columns. Possibly some other row also has
16. (0, 1)-Matrices 181
two 1s in the rst two columns. Now, consider the other possibility
and show that
(16.7) A
n
=
n(n 1)
2
(2A
n1
+ (n 1)A
n2
).
Problem 16I. Let f(x) be the function dened by the left-hand
side of (16.1). From (16.7) derive the dierential equation
2(1 x)f
/
xf = 0
and give a second proof of (16.2).
Notes.
Theorem 16.1 is due to D. Gale (1957) and to H. J. Ryser (1957).
For a more extensive treatment of the classes of (0,1)-matrices that
we have considered, we refer to Ryser (1963).
The relation between (16.2) and the generating function for de-
rangements was pointed out to us by D. G. E. D. Rogers. The
proof of (16.6) is due to P. Diaconis and D. E. Knuth (private
communication).
The relation (16.7) was announced without proof by R. Bricard
(1901).
References.
R. Bricard (1901), Probl`eme de combinaisons, LIntermediaire des
Mathematiciens 8, 312313.
D. Gale (1957), A theorem on ows in networks, Pacic J. Math.
7, 10731082.
M. Krause (1996), A simple proof of the Gale-Ryser theorem, Amer.
Math. Monthly 103, 335337.
B. D. McKay (1984), Asymptotics for (0, 1)-matrices with pre-
scribed line sums, in: Enumeration and Design (D. M. Jackson
and S. A. Vanstone, eds.), Academic Press.
H. J. Ryser (1957), Combinatorial properties of matrices of zeros
and ones, Canad. J. Math. 9, 371377.
H. J. Ryser (1963), Combinatorial Mathematics, Carus Math.
Monograph 14.
17
Latin squares
A Latin square of order n is a quadruple (R, C, S; L) where R, C,
and S are sets of cardinality n and L is a mapping L : RC S
such that for any i R and x S, the equation
L(i, j) = x
has a unique solution j C, and for any j C, x S, the same
equation has a unique solution i R. That is, any two of i R,
j C, x S uniquely determine the third so that L(i, j) = x.
Elements of R are called rows, elements of C are called columns,
and elements of S are called the symbols or entries of the Latin
square. A Latin square is usually written as an n n array for
which the cell in row i and column j contains the symbol L(i, j).
In Fig. 17.1 we have an example of a Latin square of order 5.
a b c d e
b a e c d
c d b e a
d e a b c
e c d a b
Figure 17.1
The terminology Latin square originated with Euler who used
a set of Latin letters for S.
A quasigroup is a Latin square (X, X, X; ) with a common row,
column, and symbol set X. Here we abbreviate the quadruple and
simply denote the quasigroup by (X, ). The mapping is now a
binary operation on X and the image of (x, y) under is denoted
17. Latin squares 183
by x y. So as a special case, we obtain Latin squares that are the
multiplication tables of groups.
Problem 17A. If we x the rst two rows in Fig. 17.1, then there
are many ways to ll in the remaining three rows to obtain a Latin
square (in fact 24 ways). Show that none of these Latin squares is
the multiplication table of a group.
Note that if (R, C, S; L) is a Latin square and the mappings :
R R
/
, : C C
/
, : S S
/
are bijections, and if we dene
L
/
((i), (j)) := (L(i, j)), then (R
/
, C
/
, S
/
; L
/
) is a Latin square.
The two Latin squares are called equivalent. So with this notion of
equivalence, we may assume that a Latin square of order n on the
set S := 1, 2, . . . , n has the integers from 1 to n in that order as its
rst row and as its rst column. Note that two Latin squares that
are normalized in this way and dierent, can still be equivalent.
Sometimes we denote a Latin square as just L : R C S.
An orthogonal array OA(n, 3) of order n and depth 3 is a 3 by
n
2
array with the integers 1 to n as entries, such that for any two
rows of the array, the n
2
vertical pairs occurring in these rows are
dierent. Suppose that we have such an array. Call the rows r, c,
and s, in any order. For any pair (i, j), there is a k such that r
k
= i,
c
k
= j. We make a square with entry s
k
in the i-th row and j-th
column (for all i, j). The denition of orthogonal array ensures that
this is a Latin square and we can reverse the procedure. Therefore
the concepts of Latin square and orthogonal array are equivalent.
Fig. 17.2 shows an OA(4, 3) and two corresponding Latin squares.
Two Latin squares for which the corresponding orthogonal arrays
have the same three rows (possibly in dierent order) are called con-
jugates. For example, one of the conjugates of a Latin square (with
R = C) is its transpose. As an exercise the reader should write out
the six conjugates of the Latin square in Fig. 17.2. Two orthogonal
arrays are called isomorphic if one can be obtained from the other
by permutations of the elements in each of the rows and permuta-
tions of the rows and columns of the array. Two Latin squares for
which the corresponding orthogonal arrays are isomorphic are also
called isomorphic. This means that one is equivalent to a conjugate
of the other.
184 A Course in Combinatorics
rows
columns
symbols
1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
3 2 4 1 1 4 2 3 4 3 1 2 2 1 3 4
3 2 4 1
1 4 2 3
4 3 1 2
2 1 3 4
symbols
columns
rows
1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
3 2 4 1 1 4 2 3 4 3 1 2 2 1 3 4
2 4 3 1
4 1 2 3
1 3 4 2
3 2 1 4
Figure 17.2
Problem 17B. Consider the two Latin squares obtained from Fig.
17.3 by setting a = 1, b = 2, respectively a = 2, b = 1.
1 2 3 4 5
2 1 4 5 3
3 5 a b 4
4 3 5 a b
5 4 b 3 a
Figure 17.3
Show that these two squares are equivalent. Show that a square
that is not equivalent to these corresponds to the cyclic group of
order 5.
One can also interpret a Latin square as a three-dimensional
array of 0s and 1s with exactly one 1 in each line of the array that
is parallel to one of the sides. For example, in Fig. 17.3 the entry
4 in row 2 and column 3 should then be interpreted as an entry 1
in position (2, 3, 4) of the array.
In a later chapter we shall consider orthogonal arrays with depth
greater than 3. These are often much more dicult to construct
than the arrays OA(n, 3).
Problem 17C. Construct an OA(4, 4), i.e. a 4 by 16 matrix A
with entries 1,2,3,4, such that for any two rows of A, say row i and
row j, the 16 pairs (a
ik
, a
jk
), 1 k 16, are all dierent.
17. Latin squares 185
A sub-Latin square (or subsquare) of a Latin square (R, C, S; L)
is a Latin square (R
1
, C
1
, S
1
; L
1
) with R
1
R, C
1
C, S
1
S,
and L
1
(i, j) = L(i, j) for (i, j) R
1
C
1
.
Problem 17D. Let m and n be positive integers, m < n. Show
that m
1
2
n is a necessary and sucient condition for the existence
of a Latin square of order n containing a subsquare of order m.
The following problems are all of a similar nature. They concern
a problem similar to the situation of Problems 17B and 17D, namely
the completion of a square that has been partly lled. An n
n array A with cells which are either empty or contain exactly
one symbol is called a partial Latin square if no symbol occurs
more than once in any row or column. (It might be better to say
that the array has the Latin property, but we will use the former
terminology.) We are interested in conditions that ensure that a
partial Latin square can be completed to a Latin square of order n,
i.e. when the empty cells can be lled with symbols so that a Latin
square is obtained. For example, if A is a partial Latin square with
lled (i.e. nonempty) cells everywhere except in the last row, there
is clearly a unique way to complete A to a Latin square. On the
other hand, the two partial Latin squares of Fig. 17.4 are clearly
not completable.
1 2 3 4
5
1
1
1
1
2
Figure 17.4
To a partial Latin square we can associate an array in the same
way that we did for Latin squares by only considering lled cells.
This allows us to dene the conjugate of a partial Latin square.
The two partial Latin squares in Fig. 17.4 are not really essen-
tialy dierent but are conjugates, as is evident from their partial
186 A Course in Combinatorics
OA(5, 3)s shown below:
rows
columns
symbols
_
1 1 1 1 2
1 2 3 4 5
1 2 3 4 5
_ _
1 2 3 4 5
1 2 3 4 5
1 1 1 1 2
_
If the rst k rows of a partial Latin square (k n) are lled
and the remaining cells are empty, then A is called a k n Latin
rectangle.
Theorem 17.1. A k n Latin rectangle, k < n, can be extended
to a k + 1 n Latin rectangle (and hence it can be completed).
Proof: Let B
j
denote the set of positive integers that do not occur
in column j of A. Each of the numbers 1 to n occurs k times in
A and therefore n k times in the sets B
j
. Any l of the sets B
j
together contain l(nk) elements, and therefore at least l dierent
ones. So the sets B
j
, 1 j n, have property H of Chapter 5,
and by Theorem 5.1 they therefore have an SDR. This SDR can be
adjoined as the (k + 1)-st row of the Latin rectangle.
We denote by L(n) the total number of dierent Latin squares
of order n. The exact value of L(n) is known only for n 9.
The numbers grow very rapidly, e.g. although there are only two
inequivalent Latin squares of order 5, there are 5! 4! 56 dierent
ones.
Theorem 17.2. L(n) (n!)
2n
/n
n
2
.
Proof: To construct a Latin square of order n, we can take any
permutation of 1 to n as the rst row. If we have constructed a
Latin rectangle with k rows, then by the proof of Theorem 17.1,
the number of choices for the next row is per B where b
ij
= 1 if
i B
j
. By Theorem 12.8 (the Van der Waerden conjecture) this
permanent is at least (n k)
n
n!/n
n
. So we nd
(17.1) L(n) n!
n1

k=1
(n k)
n
n!/n
n
= (n!)
2n
/n
n
2
.

17. Latin squares 187

Remark. If we had applied Theorem 5.3, we would have found
the estimate L(n) n!(n 1)! 1! (H. J. Ryser, 1969), which is
better for small n but asymptotically much smaller than (17.1).
For n = 8, the estimate (17.1) is less than 10
4
times the true
value but asymptotically (17.1) is best possible as the next theorem
shows.
Theorem 17.3. If we dene L(n) := L(n)
1/n
2
, then
L(n) e
2
n for n .
Proof: From (17.1) and Stirlings formula we nd n
1
L(n) e
2
.
By the same argument as in the proof of Theorem 17.2, and by
estimating per B using Theorem 11.5, we nd
L(n)
n

k=1
M(n, k)
n

k=1
(k!)
n/k
.
Hence, again using Stirlings formula with C some constant >

2,
we have
log L(n)
1
n
n

k=1
1
k
log k!

1
n
n

k=1
_
log k 1 +
1
2k
log k +
1
k
log C
_
=
1
n
n

k=1
log k 1 +o(1)
= 2 + log n +o(1) for n .
Combining this with the lower bound, we nd the result.
The following theorem due to H. J. Ryser (1951) is a generaliza-
tion of Problem 17D.
Theorem 17.4. Let A be a partial Latin square of order n in
which cell (i, j) is lled if and only if i r and j s. Then A can
188 A Course in Combinatorics
be completed if and only if N(i) r +s n for i = 1, . . . , n, where
N(i) denotes the number of elements of A that are equal to i.
Proof: First, observe that in a Latin square of order n, the rst
r rows contain exactly r elements equal to i of which at most ns
are in the last n s columns. So the condition on N(i) is trivially
necessary. We show that the condition is also sucient. Let B
be the (0, 1)-matrix of size r n with b
ij
= 1 if and only if the
element j does not occur in row i of A. Clearly every row of B has
sum n s. The j-th column of B has sum r N(j) n s. By
Theorem 7.5 (with d := n s) we have
B = L
(s+1)
+ +L
(n)
where each L
(t)
is an r n (0, 1)-matrix with one 1 in each row
and at most one 1 in each column.
As an example, suppose r = s = 4, n = 7, and the rst four rows
of A are
1 2 3 4
5 3 1 6
3 1 5 2
7 4 2 5
Then e.g.
B :=
_
_
0 0 0 0 1 1 1
0 1 0 1 0 0 1
0 0 0 1 0 1 1
1 0 1 0 0 1 0
_
_
= L
(5)
+L
(6)
+L
(7)
=
_
_
0 0 0 0 0 0 1
0 0 0 1 0 0 0
0 0 0 0 0 1 0
1 0 0 0 0 0 0
_
_
+
_
_
0 0 0 0 0 1 0
0 0 0 0 0 0 1
0 0 0 1 0 0 0
0 0 1 0 0 0 0
_
_
+
_
_
0 0 0 0 1 0 0
0 1 0 0 0 0 0
0 0 0 0 0 0 1
0 0 0 0 0 1 0
_
_
.
Say L
(t)
= [l
(t)
ij
]. Then we ll the cell in position (i, j) of A,
i = 1, . . . , r, j = s + 1, . . . , n, by k if l
(j)
ik
= 1. In our example, we
would ll in the last three columns with
7
4
6
1
,
6
7
4
3
,
5
2
7
6
.
17. Latin squares 189
Thus A is changed into a partial Latin square of order n with r
complete rows, i.e. a Latin rectangle. By Theorem 17.1 this can be
completed to a Latin square of order n.
The examples of partial Latin squares that cannot be completed
that were given in Fig. 17.4 both have n lled cells. The conjecture
that a partial Latin square with less than n lled cells is completable
to a Latin square was known as the Evans conjecture until it was
nally proved by B. Smetaniuk (1981). Perhaps the most important
part of Smetaniuks proof is the construction of Theorem 17.5.
Theorem 17.5. Let A be a Latin square of order n. Let B be an
n + 1 n + 1 array whose (i, j)-entry is the (i, j)-entry of A for
i, j 1, i + j n + 1, which has a new symbol on the back-
diagonal, and whose cells below the back-diagonal are empty. Then
B can be completed to a Latin square of order n + 1.
An example of a Latin square A of order 5 and the corresponding
array B are shown below.
1 2 3 4 5
4 3 5 1 2
2 5 1 3 4
5 1 4 2 3
3 4 2 5 1
1 2 3 4 5
4 3 5 1
2 5 1
5 1
3

Proof: Let C be the n n array obtained from B by dropping

the last row and column. In our example, C is the array below.
1 2 3 4 5
4 3 5 1
2 5 1
5 1
3
We describe an algorithm that lls in the empty cells of C row
by row subject to the constraint that after the rst r rows are
lled in, we will have r n a Latin rectangle on n + 1 symbols
190 A Course in Combinatorics
1, 2, . . . , n, so that the j-the column contains the same symbols,
in some order, as there were in the rst r rows of A, except for
one so-called missing symbol in each of the columns containing x,
and such that the r 1 missing symbols are all distinct. This is
trivial for r = 1 and r = 2. Assume this has been done for some r,
2 r < n and that the missing symbols at this point are
x
nr+2
, x
nr+3
, . . . , x
n
.
To ll in the (r + 1)-st row, we proceed as follows.
Let the last r symbols in row r +1 of A (that have been removed
in constructing C) be
y
nr+1
, y
nr+2
, y
nr+3
, . . . , y
n
.
Here y
nr+1
is the symbol displaced by and must be the new
missing element in that column. We may consider lling in row
r + 1 with
y
nr+2
, y
nr+3
, . . . , y
n
,
and this is satisfactory if
y
nr+1
, x
nr+2
, x
nr+3
, . . . , x
n
(which would be the new missing symbols) are distinct. If not,
consider the sequence
y
nr+1
= x
k
1
y
k
1
= x
k
2
.
.
.
y
k
m1
= x
k
m
extended as far as it can go, i.e. so that y
k
m
is not equal to any
of the x
j
s. We then ll in row r + 1 with y
nr+2
, y
nr+3
, . . . , y
n
except that y
k
1
, y
k
2
, . . . , y
k
m
are omitted and replaced, respectively,
by x
k
1
, x
k
2
, . . . , x
k
m
. It must be checked that the entire row r + 1
contains distinct sysmbols. The new missing symbols will be y
nr+1
and x
nr+2
, x
nr+3
, . . . , x
n
except that x
k
1
, x
k
2
, . . . , x
k
m
are omitted
17. Latin squares 191
and replaced, respectively, by y
k
1
, y
k
2
, . . . , y
k
m
. It must be checked
that these are distinct.
Once the empty cells of C are lled, the rst n cells of the last
row of B are to be lled with x and the missing sysmbols (when
r = n). We now have an n+1 n Latin rectangle, which completes
uniquely to a Latin square of order n + 1.
Problem 17E. Let A denote the Latin square of order 10 on the
left below, and let C be the partial 10 by 10 Latin rectangle on
the 11 symbols 0, 1, 2, . . . , 9, on the right (not including the last
row and column). Use the algorithm of the proof of Theorem
17.5 to complete C to a 10 by 10 Latin rectangle on the sym-
bols 0, 1, 2, . . . , 9, . Then complete this Latin rectangle to a Latin
square of order 11.
4 1 3 5 9 6 0 2 8 7
5 8 9 2 6 7 3 4 1 0
9 4 5 6 0 3 7 8 2 1
1 3 4 7 5 2 8 9 0 6
2 6 1 3 7 4 5 0 9 8
7 9 0 1 4 8 2 3 6 5
6 7 8 9 1 0 4 5 3 2
8 5 6 0 2 9 1 7 4 3
3 0 2 4 8 5 6 1 7 9
0 2 7 8 3 1 9 6 5 4
4 1 3 5 9 6 0 2 8 7
5 8 9 2 6 7 3 4 1
9 4 5 6 0 3 7 8
1 3 4 7 5 2 8
2 6 1 3 7 4
7 9 0 1 4
6 7 8 9
8 5 6
3 0
0

(For example, the algorithm lls in the rst ve rows of C as

indicated on the right below. The elements missing from the sev-
enth through tenth columns at this point are, respectively, 5, 9, 2, 0.
Now consider the sixth row. The in column six forces 8 to be
the missing element from column six. But 8 is distinct from the
elements currently missing from the other columns. So there is
nothing more to do; we just ll in row six with the entries 2, 3, 6, 5
from the sixth row of A.)
192 A Course in Combinatorics
4 1 3 5 9 6 0 2 8 7
5 8 9 2 6 7 3 4 1 0
9 4 5 6 0 3 7 8 2 1
1 3 4 7 5 2 8 9 0 6
2 6 1 3 7 4 5 0 9 8
7 9 0 1 4 8 2 3 6 5
4 1 3 5 9 6 0 2 8 7
5 8 9 2 6 7 3 4 1
9 4 5 6 0 3 7 8 1
1 3 4 7 5 2 8 0 6
2 6 1 3 7 4 0 9 8
7 9 0 1 4
Problem 17F.
(i) Let A and B be Latin squares of order n on symbols 1, 2, . . . , n.
Let A
/
and B
/
be the Latin squares on symbols 1, 2, . . . , n+1 (with
n + 1s on their back diagonals) that agree with A and B, respec-
tively, above their back diagonals as constructed in the proof of
Theorem 17.5. Show that if A and B are distinct, then A
/
and B
/
are distinct. This proves that the number of Latin squares of order
n + 1 with n + 1 on the back diagonal is greater than or equal to
the number N(n) of Latin squares of order n.
(ii) Explain how part (i) leads to the result
L(n) n!(n 1)! 2!1!
mentioned in the Remark following Theorem 17.2.
Theorem 17.6. A partial Latin square of order n with at most
n 1 lled cells can be completed to a Latin square of order n.
Proof: The theorem will be proved by induction. Assume it holds
for n and let L be a partial Latin square of order n+1 with at most
n lled cells.
First, suppose there is a symbol x of L that occurs only once in
L. We propose to permute rows and columns to move the lled
cells above the back-diagonal, except that the single symbol x will
be be on the back diagonal. Say that the rows with lled cells have
f
1
, f
2
, . . . , f
k
lled cells, where f
1
+ + f
k
n, and where x is
in the row with f
1
lled cells. Put the row with f
1
lled cells in
position n +1 f
1
; permute columns to push the lled cells to the
left. Put the row with f
2
lled cells in position n + 1 f
1
f
2
;
permute columns, other than the rst f
1
, to move the lled cells
to the left; the rst f
1
+ f
2
cells in that row will contain the lled
cells. Put the row with f
3
lled cells in position n+1f
1
f
2
f
3
,
17. Latin squares 193
etc., and continue until all lled cells are above the back-diagonal.
Then, if the cell containing x in row n + 1 f
1
is in column j,
interchange column j and column f
1
+ 1.
By the induction hypothesis, the part above the back diagonal,
which has at most n 1 lled cells, can be completed to a Latin
square A of order n based on symbols other than x. The array B as
described in the statement of Theorem 17.5 contains the original
lled cells of a partial Latin square that diers from L only by
row and column permutations. The theorem says that B can be
completed, and hence so can L.
At this point we can also assert that L can be completed if there
is a row that contains exactly one lled cell, or a column that
contains exactly one lled cell, because a conjugate of L will have
the property that some symbol appears exactly once (for example,
every row of the partial square on the right of Fig 17.4 contains
exactly one lled cell, and its conjugate on the left has the property
that every symbol occurs exactly once) and a conjugate of L can
be completed if and only if L can be completed.
So we may now assume that no row or column contains exactly
one lled cell. Thus the lled cells are contained in at most m rows
and m columns, where m := n/2|. We may permute rows and
columns so that all lled cells lie in the upper left subarray of order
m. Theorem 17.4 implies that every m m Latin rectangle can be
completed to a Latin square of order n+1. So it will suce to ll in
the unlled cells of the the upper left subarray of order m to get a
Latin rectangle. But this is easy, as we have n+1 symbols available
and we may proceed in any order, lling in a cell of the subsquare
with a symbol not yet used in the row or column containing the
cell.
We now come to a generalization of the idea of a Latin square.
Suppose we are given a collection ( of n-sets C(i, j), i, j = 1, 2, . . . , n.
One can ask the question whether there exists a matrix A of size
n n in which in each row, respectively column, all the entries a
ij
are dierent and for which a
ij
C(i, j) for all i and j. If all the
n-sets are 1, 2, . . . , n, the answer is yes since any Latin square is
a solution. If the elements of ( are pairwise disjoint, the question
is trivial. It is perhaps surprising that the increased freedom of not
194 A Course in Combinatorics
requiring the n-sets to be the same makes the question harder. In
fact, an analog of Theorem 17.1 does not hold in this case. Until it
was proved by F. Galvin (1995), the conjecture that such a matrix
A indeed exists was known as the Dinitz conjecture.
We shall use the terminology of colorings of graphs (a problem
treated extensively in Chapter 33). Let G be a simple graph. A
list assignment of G is a function C that assigns a set C(v) to each
vertex v of G. A C-coloring of G is a function A dened on the
vertex set such that A(v) C(v) and such that A(v) and A(w) are
dierent if v and w are adjacent in G.
Let f be a function from the vertex set of G to the integers. We
say that G is f-colorable if a C-coloring of G exists for every C
with the property [C(v)[ f(v) for all v G. If k is an integer,
a graph is called k-list-colorable if G is f-colorable for the constant
function f(v) = k. Clearly, if G is k-list-colorable, then k is at least
the chromatic number (G), dened in Chapter 3.
The line graph L = L(G) of a simple graph G is the simple graph
with vertex set V (L) := E(G) and where a, b E(G) are adjacent
as vertices of L if and only if, as edges in G, a and b have a common
incident vertex. Now, consider the line graph L
2
(n) of K
n,n
(see
Chapter 21). It has n
2
vertices i, j, i, j = 1, 2, . . . , n, where
pairs are joined if and only if they have an element in common.
The Dinitz conjecture can be formulated as follows: L
2
(n) is n-
list-colorable. The theorem proved by Galvin states that for a
bipartite graph G, the line graph L = L(G) is (L)-list-colorable.
Since Latin squares exist, L
2
(n) has chromatic number n and the
Dinitz conjecture is therefore true.
We shall prove two propositions from which the theorem will
follow.
A subset K of the vertex set of a digraph D is called a kernel of
D if no two vertices of K are joined by an edge and for every v / K
there is a w K such that there is an edge in D from v to w.
Proposition 17.7. Let D be a digraph for which every induced
subgraph has a kernel. Let f(v) := 1+outdegree(v) for every vertex
v of D. Then D is f-colorable.
17. Latin squares 195
Proof: We use induction on the number of vertices. Small cases
are easily checked. (Note that the only digraph on three vertices
for which the condition and the assertion of the problem do not
hold is the directed circuit.)
Let C be a list assignment of D with [C(v)[ f(v) for all v. Pick
an element x from one of the sets in C and consider all vertices
v such that x C(v). We call the induced subgraph on these
vertices D
1
. By our assumption, D
1
has a kernel K. Let D
2
be
the induced subgraph on the vertices in DK. For a vertex v / K
dene C
/
(v) := C(v)x. By the induction hypothesis D
2
has a
C
/
-coloring A
/
. If we dene A on D by
A(v) :=
_
A
/
(v) if v / K,
x if v K,
then A is clearly a C-coloring of D.
For the next proposition we dene an orientation of a graph G
to be normal if every clique in G is a transitive tournament.
Proposition 17.8. Every normal orientation of the line graph L =
L(G) of a bipartite graph G has a kernel.
Proof: We use induction on the number of edges of G. For small
numbers, the statement is easily checked. Suppose V (G) = X Y
(and all edges between X and Y ). We may assume that every
vertex of G has a positive degree. Let D be a normal orientation
of L(G). The cliques in L(G) correspond to vertices in X or in Y .
For every vertex x X, we consider the corresponding clique
in L(G) and dene e(x) := x, y(x) to be the sink of this clique
(with respect to D). First, assume that for every x X, this vertex
e(x) of L(G) is the source of the clique dened by y(x) Y . It is
then obvious that all the y(x) for x X are distinct. Therefore
K := x, y(x) : x X is a kernel of L(G).
Next, assume that for some x X, e(x) := x, y(x) is the sink
of the clique corresponding to x but that e
/
:= x
/
, y(x), where
x
/
,= x, is the source of the clique corresponding to y(x). Remove
the edge x
/
, y(x) from G. The resulting line graph has a kernel K
by the induction hypothesis. We claim that K is also a kernel for
196 A Course in Combinatorics
D. We have to show that there is an edge from e

to K. If e(x) K
we are done (because e

is a source). However, if e(x) / K, then

there is an edge from e(x) to some element in K, which must be
of type {x

, y(x)} by denition of e(x). Because the tournament

dened by y(x) is transitive, we are again done.
We can now prove the main result.
Theorem 17.9. If G is a bipartite graph, then L = L(G) is (L)-
colorable.
Proof: Let G be a bipartite graph on the vertex set XY . Let g
be a coloring of the vertices of L = L(G) with colors 1, 2 . . . , (L).
We orient L to form a digraph D as follows: if e
1
= {x
1
, y
1
} and
e
2
= {x
2
, y
2
} are adjacent vertices of L and g(e
1
) < g(e
2
), then

the edge is from e

1
to e
2
if x
1
= x
2
,
the edge is from e
2
to e
1
if y
1
= y
2
.
Since every clique in L corresponds to a vertex x X or y Y , this
orientation is normal. By Proposition 17.8 every induced subgraph
of D has a kernel. Clearly, 1+outdegree(e) < (L) for every vertex
e D. So, by Proposition 17.7 the proof is complete.

Problem 17G.
(a) Let n be even. Find a permutation x
1
, x
2
, . . . , x
n
of the ele-
ments of Z
n such that the dierences x
i+1
x
i
, 1 i < n, are all
dierent.
(b) Show that this is not possible if n is odd.
(c) Consider the permutation of (a). Dene a
ij
:= x
i
+x
j
. Show
that the square with entries a
ij
, 1 i, j n, is a Latin square that
has the following property: The n(n1) adjacent pairs (a
ij
, a
i,j+1
)
are dierent. Such a square is called row-complete. This square is
also column-complete.
Problem 17H. Describe a one-to-one correspondence between
symmetric Latin squares of order n in which all symbols appear
on the main diagonal and symmetric Latin squares of order n + 1
with all (n + 1)s on the diagonal.

17. Latin squares 197

Problem 17I. Let 1, 2, . . . , n be the rst row of a Latin square
and 2, 3, . . . , n, 1 the second row. How many candidates are there
for the third row?
Problem 17J. Let 1, 2, . . . , 2n 1, 2n be the rst row of a Latin
square of order 2n. Let 2, 1, 4, 3, . . . , 2n, 2n 1 be the second row.
(i) Find a formula for the number of candidates for the third row.
(ii) Give a permanent that has this number as value.
Notes.
Although Latin squares have been studied for more than 200
years, much of the material in this chapter is recent. Theorem 17.1
is the oldest; it is due to M. Hall, Jr. (1945).
Theorem 17.2 and the remark following it can be found in H. J.
Ryser (1969). At the time that paper was written, the Van der
Waerden conjecture was still open.
Theorem 17.3 is new.
Theorem 17.4 is again due to H. J. Ryser (1951).
The so-called Evans conjecture occurs as a question (not as a
conjecture) in a paper by T. Evans (1960). Before the proof by B.
Smetaniuk appeared, several partial results were published.
Our treatment of the Galvin proof of the Dinitz conjecture is
based on some unpublished notes of D. Homan.
The most extensive treatment of Latin squares can be found in
the book Latin Squares by J. Denes and A. D. Keedwell (1991).
References.
J. Denes and A. D. Keedwell (1991), Latin squares. New develop-
ments in the theory and applications., Annals of Discrete Math-
ematics 46, North-Holland.
T. Evans (1960), Embedding incomplete Latin squares, Amer.
Math. Monthly 67, 958961.
F. Galvin (1995), The list chromatic index of a bipartite multi-
graph, J. Combin. Theory Ser. B 63, 153158.
M. Hall, Jr. (1945), An existence theorem for Latin squares, Bull.
Amer. Math. Soc. 51, 387388.
H. J. Ryser (1951), A combinatorial theorem with an application
to Latin rectangles, Proc. Amer. Math. Soc. 2, 550552.
198 A Course in Combinatorics
H. J. Ryser (1969), Permanents and systems of distinct representa-
tives, in: Combinatorial Mathematics and its Applications, Uni-
versity of North Carolina Press.
B. Smetaniuk (1981), A new construction on Latin squares I: A
proof of the Evans conjecture, Ars Combinatoria 11, 155172.
18
Hadamard matrices,
ReedMuller codes
Hadamard considered the following question. Let A be an n n
matrix with real entries of absolute value at most 1. How large can
the determinant of A be (in absolute value)? Since each row of A
is a vector with length

n, the determinant cannot be larger
than n
n/2
. (The absolute value of the determinant is the n-dimen-
sional volume of the parallelopiped spanned by the row vectors in
Euclidean n-space.) Can equality hold? In that case all entries
must be +1 or 1 and furthermore any two rows must be orthog-
onal, i.e. they have inner product 0. This leads to the following
denition.
A Hadamard matrix of order n is an n n matrix H with entries
+1 and 1, such that
(18.1) HH

= nI.
Of course, any two columns of H are also orthogonal. This property
does not change if we permute rows or columns or if we multiply
some rows or columns by 1. Two such Hadamard matrices are
called equivalent. For a given Hadamard matrix, we can nd an
equivalent one for which the rst row and the rst column consist
entirely of +1s. Such a Hadamard matrix is called normalized.
Clearly the remaining rows (if any) have as many +1s as 1s, i.e.
if n ,= 1 then n must be even. Some small examples are
(1),
_
1 1
1 1
_
,
_
_
_
+ + + +
+ +
+ +
+ +
_
_
_
,
200 A Course in Combinatorics
where, in the last example, we have only indicated the sign of the
entry.
Problem 18A. Show that any two Hadamard matrices of order
12 are equivalent.
Theorem 18.1. If H is a Hadamard matrix of order n, then n = 1,
n = 2, or n 0 (mod 4).
Proof: Let n > 2. Normalize H. We can permute columns in
such a way that the rst three rows of H become
+ + + + + + + + + + + + + + + +
+ + + + + + + +
+ + + +
. .

. .
+ + + +
. .

. .
a columns b columns c columns d columns
We have a + b + c + d = n and the three inner products formed
from these rows yield a + b c d = 0, a b + c d = 0, and
abc+d = 0. If we add these equations, we nd n = 4a, proving
the theorem. (In a similar way we see that 4b = 4c = 4d = n.)
One of the famous conjectures in the area of combinatorial de-
signs states that a Hadamard matrix of order n exists for every
n 0 (mod 4). We are still very far from a proof of this conjec-
ture. The smallest n for which a Hadamard matrix could exist but
no example is known is presently 428. There are very many con-
struction methods for Hadamard matrices, of which we shall treat
a few. First we dene a second class of matrices, very similar to
Hadamard matrices.
A conference matrix C of order n is an n n matrix with 0s on
the diagonal, +1 or 1 in all other positions and with the property
(18.2) CC

= (n 1)I.
The name conference matrix originates from an application to con-
ference telephone circuits. V. Belevitch (1950) studied so-called
ideal nondissipative networks consisting of ideal transformers, to
be used to set up a conference telephone network. The theory led
to a necessary condition for the existence of such a network, namely
the existence of a conference matrix of order n, where n is the num-
ber of terminals of the network. This explains the name of these
matrices.
18. Hadamard matrices, ReedMuller codes 201
Problem 18B. Let C be a conference matrix of order n ,= 1.
Show that n is even. Show that by permuting rows and columns,
and multiplying certain rows and columns by 1, we can nd an
equivalent conference matrix that is symmetric if n 2 (mod 4),
and antisymmetric if n 0 (mod 4).
Theorem 18.2. If C is an antisymmetric conference matrix, then
I +C is a Hadamard matrix.
Proof: (I +C)(I +C)

= I +C+C

+CC

= I +(n1)I = nI.

Theorem 18.3. If C is a symmetric conference matrix of order n,

then
H =
_
I +C I +C
I +C I C
_
is a Hadamard matrix of order 2n.
Proof: Again calculate HH

and the result follows.

One of the most common construction methods for several com-
binatorial structures is the so-called recursive method in which a
large object of the required type is made by applying some pro-
cedure to two or more smaller objects of the same type. In the
next theorem we demonstrate such a method for the construction
of Hadamard matrices.
Let A be an m n matrix with entries a
ij
and B another matrix.
The matrix
_
_
_
_
a
11
B a
12
B . . . a
1n
B
a
21
B a
22
B . . . a
2n
B
.
.
.
.
.
.
.
.
.
.
.
.
a
m1
B a
m2
B . . . a
mn
B
_
_
_
_
consisting of mn blocks with the size of B is called the Kronecker
product AB of the matrices A and B.
Theorem 18.4. If H
m
and H
n
are Hadamard matrices of order m
and n respectively, then H
m
H
n
is a Hadamard matrix of order
mn.
Proof: By straightforward calculation one sees that
(AB)(C D) = (AC) (BD)
202 A Course in Combinatorics
and that
(AB)

= A

.
We take A = C = H
m
and B = D = H
n
. The result follows from
the denition of a Hadamard matrix and the fact that I
m
I
n
= I
mn
.

By repeatedly applying this theorem to H

2
:=
_
+ +
+
_
, we nd a
sequence of Hadamard matrices that we shall denote by H
n
, where
n = 2
m
, m = 1, 2, . . . .
We now come to a direct construction method for conference
matrices, and these can then be used with Theorems 18.2 and 18.3
to construct Hadamard matrices. In the following, q is a power of
an odd prime. On the eld F
q
, we dene a function (a so-called
character) by
(x) :=
_
_
_
0 if x = 0,
1 if x is a nonzero square,
1 if x is a nonsquare.
For any x and y in F
q
, we have (x)(y) = (xy), and since there
are as many nonzero squares as nonsquares we also have
(18.3)

xF
q
(x) = 0.
Now let 0 ,= c F
q
. Then (18.3) implies that
(18.4)

bF
q
(b)(b +c) = 1.
This is seen by ignoring the term with b = 0, which is 0, and then
writing (b + c) = (b)(1 + cb
1
); (note that (b)
2
= 1 if b ,= 0).
If b runs through all nonzero elements of the eld, then 1 + cb
1
takes on every value except 1.
Number the elements of F
q
: 0 = a
0
, a
1
, . . . , a
q1
. We dene a q
q matrix Q by
q
ij
:= (a
i
a
j
), 0 i, j < q.
18. Hadamard matrices, ReedMuller codes 203
Note that Q is symmetric if q 1 (mod 4), and antisymmetric
if q 3 (mod 4). As a direct consequence of the elementary
properties of and of (18.4), we nd that QQ

= qI J and
QJ = JQ = O. A matrix C of size q + 1 q + 1 is dened by
(18.5) C :=
_
_
_
_
0 1 1 . . . 1
1
.
.
. Q
1
_
_
_
_
,
where the signs of the terms 1 are chosen in such a way that
C is symmetric or antisymmetric. From the properties of Q, it
now follows that C is a conference matrix of order q + 1. This
construction is due to Paley (1933) and the conference matrices of
this type are usually called Paley matrices. For the special case
that q is a prime, the matrix Q is a circulant. We summarize the
constructions as a theorem.
Theorem 18.5. If q is a power of an odd prime, then a Hadamard
matrix of order q + 1 exists if q 3 (mod 4), and a Hadamard
matrix of order 2(q + 1) exists if q 1 (mod 4).
+ + + + + + + + + + + +
+ ++ + ++
+ ++ + ++
++ ++ + +
++ ++ + +
++ ++ + +
++ ++ + +
+++ ++ +
+ +++ ++
+ + +++ +
+ + +++ +
++ + +++
Figure 18.1
204 A Course in Combinatorics
+ + + + + + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + +
+ + + + +
+ + + + +
+ + + + +
+ + + + +
+ + + + +
Figure 18.2
Figures 18.1 and 18.2 illustrate Hadamard matrices of order 12
constructed from Paley matrices of orders 11 +1 and 5 +1, respec-
tively.
Problem 18C. Show that a Hadamard matrix of order n exists if
n 0 (mod 4), n 100, except possibly for n = 92.
There was a period of thirty years between Paleys result and the
discovery of a Hadamard matrix of order 92 by L. D. Baumert, S.
W. Golomb, and M. Hall. The method that they used was devel-
oped by Williamson in 1944 but a computer search was necessary
to nd the actual matrix. Williamsons method is based on the
following observation. Let the matrices A
i
, 1 i 4, of order n, n
odd, be symmetric and assume that they commute with each other.
Consider the matrix H dened by
(18.6)
_
_
_
A
1
A
2
A
3
A
4
A
2
A
1
A
4
A
3
A
3
A
4
A
1
A
2
A
4
A
3
A
2
A
1
_
_
_
.
We then have
(18.7) HH

= I
4
(A
2
1
+A
2
2
+A
2
3
+A
2
4
).
18. Hadamard matrices, ReedMuller codes 205
To construct a Hadamard matrix in this way, we have to nd ma-
trices A
i
that satisfy the conditions stated above, have entries 1,
and furthermore satisfy
(18.8) A
2
1
+A
2
2
+A
2
3
+A
2
4
= 4nI
n
.
Let U be the permutation matrix of order n corresponding to
the permutation (12 . . . n), i.e. u
ij
= 1 if and only if j i 1
(mod n). Then U
n
= I and any circulant is a linear combination
of powers of U. If we assume that the matrices A
i
have the form
A
i
=

n1
j=0
a
ij
U
j
with a
i0
= 1 and a
ij
= a
i,nj
, then the matrices
do indeed commute and they are symmetric. From now on, we also
assume that all the a
ij
are 1 and that (18.8) is satised. A simple
example of a Hadamard matrix constructed in this way is found
by taking n = 3, A
1
= J, and A
i
= J 2I, i = 2, 3, 4. We nd a
Hadamard matrix of order 12; see Fig. 18.3.
+ + + + + + + + +
+ + + + + + + + +
+ + + + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
+ + + + + + +
Figure 18.3
Example 18.1. Note that our construction implies that A
i
has
constant row-sum a
i
(odd) and by (18.8) we have a
2
1
+ +a
2
4
= 4n.
If we wish to use this method to nd a Hadamard matrix of order
20, we rst write 20 as a sum of four odd squares: 20 = 9+9+1+1.
From this we see that two of the matrices A
i
must be 2I J and
then it is not dicult to see that the other two have as rst row
++ +, respectively + ++.
206 A Course in Combinatorics
To analyze the situation a little further, we introduce matrices
W
i
and P
i
as follows:
(18.9) 2W
i
:= (A
1
+ +A
4
) 2A
i
,
(18.10) A
i
= 2P
i
J.
Our conventions on the coecients a
ij
imply that if p
i
is dened
by p
i
J := P
i
J, then p
i
is an odd integer. Furthermore (18.8) and
(18.9) imply that
(18.11) W
2
1
+W
2
2
+W
2
3
+W
2
4
= 4nI.
By substituting (18.10) in (18.8), we nd
(18.12)
4

i=1
P
2
i
= (
4

i=1
p
i
n)J +nI.
Suppose the term U
k
, k ,= 0, occurs in of the matrices P
i
. Con-
sidering the equation (18.12) mod 2, we nd U
2k
on the left-hand
side with coecient and on the right-hand side with coecient 1.
So is odd and this implies that U
k
occurs in exactly one of the
matrices W
i
(with coecient 2). From Example 18.1 and (18.9),
we see that the constant row-sums w
i
of the matrices W
i
satisfy
w
2
1
+ +w
2
4
= 4n. These facts reduce the number of possibilities
for matrices W
i
also satisfying (18.11) suciently to make a com-
puter search for such matrices feasible. The A
i
are then found from
(18.9). We list the rst rows of the four matrices of order 23 that
produced the rst Hadamard matrix of order 92:
A
1
: + ++++ ++++
A
2
: ++ ++ ++ + + + + ++ ++ +
A
3
: + + ++ +++ +++ ++ +
A
4
: + + ++ + ++++ + ++ +
Problem 18D. Construct a Hadamard matrix of order 28 using
Williamsons method.
We look at a dierent problem concerning Hadamard matrices.
If a Hadamard matrix of order n is normalized, then it obviously
18. Hadamard matrices, ReedMuller codes 207
has exactly n more entries +1 than entries 1. We dene the excess
of a Hadamard matrix to be the sum of all the entries, and then
dene (n) as the maximal value of the excess of all Hadamard
matrices of order n. The following bound due to Best (1977) shows
that (n) grows as n
3/2
.
Theorem 18.6. n
2
2
n
_
n
1
2
n
_
(n) n

n.
Proof: (a) Let H be a Hadamard matrix of order n. Let s
k
be
the sum of the k-th column of H. Let c
i
be the i-th row of H,
1 i n. We calculate

1i,jn
c
i
, c
j
) in two ways. By the
denition of Hadamard matrix this sum is n
2
. On the other hand

1i,jn
n

k=1
c
ik
c
jk
=
n

k=1
s
2
k
.
From this and the Cauchy-Schwarz inequality we nd
n

k=1
s
k
(n
n

k=1
s
2
k
)
1/2
= n

n,
so (n) n

n.
(b) Let x be any vector in +1, 1
n
. We multiply column j of
H by x
j
, 1 j n. Subsequently, multiply those rows that have
more terms 1 than +1 by 1. Call the resulting matrix H
x
and
dene (H
x
) :=

n
i=1
[x, c
i
)[. Clearly (n) is at least equal to the
average value of (H
x
). So we nd
(n) 2
n

x+1,1
n
(H
x
) = 2
n

x
n

i=1
[x, c
i
)[
= 2
n
n

i=1
n

d=0

x,d(x,c
i
)=d
[n 2d[
= 2
n
n
n

d=0
[n 2d[
_
n
d
_
= n
2
2
n
_
n
1
2
n
_
.

208 A Course in Combinatorics

Corollary. 2
1/2
n
3/2
(n) n
3/2
.
Proof: That the left-hand side of the inequality in Theorem 18.6
is asymptotically equal to 2
1/2

1/2
n
3/2
follows from Stirlings for-
mula. To obtain an inequality that holds for all n, we must replace
the constant by 2
1/2
.
Example 18.2. Consider a square of size 4 by 4, divided into
16 cells that we number from 1 to 16. A matrix A with rows a
i
,
1 i 16, is dened by taking a
ij
= 1 if j occurs in the square
in the same row or column as i, but j ,= i; otherwise a
ij
= 1. For
any two rows of A, there are two positions where they both have
entry 1. It follows that A is a Hadamard matrix of order 16 with
excess 64, i.e. for this matrix we have equality for the upper bound
in Theorem 18.6.
This Hadamard matrix can also be constructed as follows. Dene
H := J 2I, a Hadamard matrix of order 4 with maximal excess,
namely 8. Every row of H has the same number of terms +1,
namely 3. The matrix H H is a Hadamard matrix of order 16,
with 10 terms +1 in each row, and the maximal excess 64. If we
take the Kronecker product of this matrix with H, we again nd a
Hadamard matrix with constant row-sums and maximal excess, etc.
In general, a Hadamard matrix of order 4u
2
, all row-sums equal to
2u, and hence maximal excess, is called a regular Hadamard matrix.
Other constructions of such matrices are known.
One of the very interesting and successful applications of Had-
amard matrices was their use as so-called error-correcting codes.
Many readers will have seen the excellent pictures that were taken
of Mars, Saturn and other planets by satellites such as the Mariners,
Voyagers, etc. To transmit such a picture to Earth, it is rst di-
vided into very small (pixels) and for each such square the degree
of blackness is measured and expressed, say in a scale of 0 to 63.
These numbers are expressed in the binary system, i.e. each pixel
produces a string of six 0s and 1s (bits). The bits are transmitted
to the receiver station on Earth (the Jet Propulsion Laboratory
at Caltech). Due to a number of sources of noise, one of which
is thermal noise from the amplier, it happens occasionally that a
signal that was transmitted as a 0, respectively a 1, is interpreted
18. Hadamard matrices, ReedMuller codes 209
by the receiver as a 1, respectively a 0. If each sextuple correspond-
ing to a pixel is transmitted as such, then the errors made by the
receiver would make the pictures very poor. Since there is only a
limited amount of time to transmit the pictures and it is known
how much energy is available from the solar panels, we know the
average energy per user bit available to generate signals to be sent
to the receiver. From this we can calculate the error probability p
(per bit).
Suppose that for a useful picture, each sextuple may be wrong
with a probability P
E
at most 10
4
. To achieve this by simply
transmitting the bits, we need p 10
4
/6 since P
E
= 1(1p)
6

10
4
. Let us rst assume that the required energy is available
and try to improve the picture by doing the following. Instead
of sending a bit, say 0, we could send ve 0s and the receiver
translates a received vetuple into the bit that occurs most. We say
that we are using a code, called the repetition code, with two words
of length 5, namely 00000 and 11111. The number 1/5 is called
the information rate of this code. The power constraint allows us
to calculate how much energy we have available per channel bit.
Subsequently, it is possible to calculate the new error rate p
/
(per
channel bit). Of course p
/
> p and in our example we in fact have
p
/
= 0.035 (nearly 2000 times as large as without coding). The
coding only makes sense if error correction more than makes up for
the loss of energy per bit.
The probability that a transmitted pixel is received correctly now
becomes
[(1 p
/
)
5
+ 5p
/
(1 p
/
)
4
+ 10(p
/
)
2
(1 p
/
)
3
]
6
and this is 0.997. In other words, we have completely spoiled
our picture!
Let us now look at what was actually done in the Mariner 1969
expedition. The 64 possible information strings (corresponding to
the possible degrees of blackness of a pixel) were mapped onto the
rows of the matrices H
32
and H
32
. Since we now have codewords
of length 32 representing information words of length 6, the infor-
mation rate is 6/32, nearly the same as for the repetition code.
Note that any two of these codewords either dier in all 32 posi-
tions or dier in exactly 16 of the positions. It follows that if a
210 A Course in Combinatorics
received word contains at most 7 errors, then it resembles the cor-
rect word more than any of the 63 other words. (We have changed
our symbols to 1 instead of 0 and 1.) We say that this code is a 7-
error-correcting code. The new error probability now is p
/
0.036
(again larger). The probability that a received word is decoded
incorrectly now is
32

i=8
_
32
i
_
(p
/
)
i
(1 p
/
)
32i
and this is roughly 1.4 10
5
which is an order of magnitude better
than P
E
.
In practice, there was not enough energy to transmit good pic-
tures but by using this code, that problem was solved.
What we have seen is one example of a sequence of (low rate)
codes, known as rst order ReedMuller codes. Consider once again
the construction of the Hadamard matrices H
n
, where n = 2
m
. In
Fig. 18.4 we show H
8
.
In the matrix H
n
we replace each +1 by 0 and each 1 by 1.
Number the rows from 0 to n 1 = 2
m
1. What happens in the
Kronecker product construction when we go from m to m+ 1, i.e.
double the order of the matrix? The new rows (now interpreted as
vectors in F
2n
2
) have the form (c
i
, c
i
) for 0 i < n and they have
the form (c
i
, c
i
+ 1) for n i < 2n, where 1 denotes the all-one
vector.
+ + + + + + + +
+ + + +
+ + + +
+ + + +
+ + + +
+ + + +
+ + + +
+ + + +
Figure 18.4
Theorem 18.7. Let R
/
(1, m) denote the set of row vectors in F
n
2
18. Hadamard matrices, ReedMuller codes 211
obtained from H
n
as described above. Then R
/
(1, m) is an m-
dimensional subspace of F
n
2
.
Proof: For m = 1, the assertion is trivial. We proceed by induc-
tion. Let v
1
, v
2
, . . . , v
m
be a basis of R
/
(1, m). Our observation
made above shows that R
/
(1, m+1) consists of all linear combina-
tions of the vectors (v
i
, v
i
) and the vector (0,1). This completes
the proof.
We now dene the rst order ReedMuller code R(1, m) of length
n = 2
m
and dimension m + 1 to be the subspace of F
n
2
spanned
by the space R
/
(1, m) and the all-one vector of length n. In this
terminology, the code used by Mariner 1969 was R(1, 5). From
the properties of Hadamard matrices, we immediately see that any
two codewords in R(1, m) dier in at least
1
2
n places. We say that
the code has minimum distance d = 2
m1
. Therefore the code can
correct up to 2
m2
1 errors.
We can give the codewords a nice geometric interpretation. Let
us number the points of the space F
m
2
by considering a vector as
the binary representation of its number. For example, (0, 1, 1, 0, 1)
is considered as the point P
22
in F
5
2
. Let these representations be
the columns of an m n matrix. Let v
i
be the i-th row of this
matrix, 1 i m. Then from Fig. 18.4 and the observations
above, we see that the vectors v
i
are the natural basis for R
/
(1, m).
The basis vector v
i
is the characteristic function of the hyperplane
(x
1
, x
2
, . . . , x
m
) F
m
2
: x
i
= 1. By taking linear combinations, we
see that the codewords of R(1, m) are exactly all the characteristic
functions of the ane hyperplanes in the vector space and the char-
acteristic function of the space itself (for 1) and of the empty set
(for 0). Any two ane hyperplanes are either parallel or they meet
in an ane subspace of dimension m 2, in accordance with the
fact that two rows of the Hadamard matrix have the same entries
in exactly half of the positions.
As our denition of R(1, m) suggests, ReedMuller codes of higher
order are also dened in the theory of error-correcting codes. They
also have a geometric interpretation that we shall not go into here
(but see Example 26.4).
Problem 18E. For n = 2
m
and 1 i m, we dene the matrix
212 A Course in Combinatorics
M
(i)
n
by
M
(i)
n
:= I
2
mi H
2
I
2
i1.
Prove that
H
n
= M
(1)
n
M
(2)
n
. . . M
(m)
n
.
A received word x is decoded by calculating xH

n
. If there are not
too many errors, then all entries of this product will be nearly 0
except one with an absolute value close to n, telling us what the
true message was. A multiplication by 1 is called an operation.
Compare the number of operations that are necessary for decoding
when H
n
is used and when the representation of H
n
as a product
of matrices M
(i)
n
is used. (This is an example of what is known as
a Fast Fourier Transform.)
Problem 18F. Let v
i
, 0 i m, where v
0
= 1, be the basis
of R(1, m) given above. We consider the subspace R(2, m) of F
n
2
,
where n = 2
m
, spanned by all the vectors
v
i
v
j
:= (v
i0
v
j0
, . . . , v
i,n1
v
j,n1
).
What is the dimension of this space? Show that any two vectors of
R(2, m) dier in at least
1
4
n places.
Problem 18G. Suppose M is an m n (0, 1)-matrix so that the
Hamming distance between any two distinct rows is at least d. (If
M is the result of stacking a Hadamard matrix H on top of H
and changing the symbols to 0s and 1s, we have an example with
m = 2n and d = n/2.)
(1) Count the number of ordered triples (i, j, k) such that i and
j are (indices of) distinct rows and k is a column where M(i, k) ,=
M(j, k) in two waysone yielding an inequality involving d and
the other involving the column-sums of M. Prove that if 2d > n,
then
m
2d
2d n
.
(This is known as Plotkins bound in coding theory.) What condi-
tions ensure equality?
18. Hadamard matrices, ReedMuller codes 213
(2) Suppose d = n/2. Prove that m 2n and show that equality
implies the existence of a Hadamard matrix of order n.
Problem 18H. Let H be a Hadamard matrix of order m and
let C be a symmetric conference matrix of order n. Let P be the
matrix of size m with the form
_
O I
I O
_
where the submatrices have
size m/2. Prove that (H C) + (PH I) is a Hadamard matrix
of size mn.
Problem 18I. Let q be a prime power 1 (mod 4). Let C be the
matrix of (18.5). Choose H and P as in Problem 18H. Prove that
(H C Q) + (PH C I
q
) + (H I
q+1
J
q
)
is a Hadamard matrix of size mq(q + 1).
Notes.
J. Hadamard (18651963) was a leading mathematician around
the turn of the century. His most important work was in the theory
of analytic functions and in mathematical physics. He is best known
for the proof of the so-called prime number theorem, jointly with
C. J. De La Vallee-Poussin.
R. E. A. C. Paley (19071933) died in an avalanche while skiing
at the age of 26. In his short life he produced 26 papers of excellent
quality (mostly on Fourier Theory).
A long outstanding conjecture of H. J. Ryser asserts that no
Hadamard matrix of order n > 4 can be a circulant.
J. S. Wallis (aka J. Seberry) (1976) proved that for any integer s,
Hadamard matrices of orders 2
t
s exist whenever t > 2 log
2
(s 3).
However, it is still not known whether the set of orders of Hadamard
matrices has positive density.
For more on error-correcting codes, see Chapter 20.
The codes that are now called ReedMuller codes were (surpris-
ingly?) indeed rst treated by D. E. Muller (1954) and I. S. Reed
(1954).
For a detailed account of the coding and decoding for the Mariner
missions, see E. C. Posner (1968).
214 A Course in Combinatorics
References.
L. D. Baumert, S. W. Golomb, and M. Hall, Jr. (1962), Discovery
of a Hadamard matrix of order 92, Bull. Amer. Math. Soc. 68,
237238.
V. Belevitch (1950), Theory of 2n-terminal networks with appli-
cations to conference telephony, Electrical Communication 27,
231244.
M. R. Best (1977), The excess of a Hadamard matrix, Proc. Kon.
Ned. Akad. v. Wetensch. 80, 357361.
D. E. Muller (1954), Application of Boolean algebra to switching
circuit design and to error detection, IEEE Trans. Computers 3,
612.
R. E. A. C. Paley (1933), On orthogonal matrices, J. Math. Phys.
12, 311320.
E. C. Posner (1968), Combinatorial structures in planetary recon-
naissance, in: Error Correcting Codes (H. B. Mann, ed.), J. Wiley
and Sons.
I. S. Reed (1954), A class of multiple-error-correcting codes and the
decoding scheme, IEEE Trans. Information Theory 4, 3849.
J. S. Wallis (1976), On the existence of Hadamard matrices, J.
Combinatorial Theory (A) 21, 188195.
J. Williamson (1944), Hadamards determinant theorem and the
sum of four squares, Duke Math. J. 11, 6581.
19
Designs
In this chapter we give an introduction to a large and important
area of combinatorial theory which is known as design theory. The
most general object that is studied in this theory is a so-called
incidence structure. This is a triple S = (T, B, I), where:
(1) T is a set, the elements of which are called points;
(2) B is a set, the elements of which are called blocks;
(3) I is an incidence relation between T and B (i.e. I T B).
The elements of I are called ags.
If (p, B) I, then we say that point p and block B are incident.
We allow two dierent blocks B
1
and B
2
to be incident with the
same subset of points of T. In this case one speaks of repeated
blocks. If this does not happen, then the design is called a simple
design and we can then consider blocks as subsets of T. In fact,
from now on we shall always do that, taking care to realize that
dierent blocks are possibly the same subset of T. This allows us
to replace the notation (p, B) I by p B, and we shall often say
that point p is in block B instead of incident with B.
It has become customary to denote the cardinality of T by v and
the cardinality of B by b. So the incidence structure then is a set
of v points and a collection of b not necessarily distinct subsets of
the point set. The structure obtained by replacing each block by
its complement is, of course called the complement of the structure.
(This means that we replace I by its complement in T B.)
To obtain an interesting theory, we must impose some regularity
conditions on the structure S. As a rst example, we mention
incidence structures that have the confusing name linear spaces.
Here the blocks are usually called lines and the regularity conditions
216 A Course in Combinatorics
are that every line contains (i.e. is incident with) at least two points
and any two points are on exactly one line. Example 19.6 below
shows a simple but important linear space. The following theorem
is due to De Bruijn and Erd os (1948). The elegant proof is due to
Conway.
Theorem 19.1. For a linear space we have b = 1 or b v, and
equality implies that for any two lines there is exactly one point
incident with both.
Proof: For x T, denote by r
x
the number of lines incident with
x, and similarly for B B, let k
B
be the number of points on B.
Let there be more than one line. If x / L then r
x
k
L
because
there are k
L
lines joining x to the points on L. Suppose b v.
Then b(v k
L
) v(b r
x
) and hence
1 =

xT

L,x
1
v(b r
x
)

LB

x/ L
1
b(v k
L
)
= 1
and this implies that in all the inequalities, equality must hold.
Therefore v = b, and r
x
= k
L
if x / L.
A trivial example of equality in Theorem 19.1 is a so-called near
pencil, a structure with one line that contains all the points but
one, and all pairs containing that point as lines of size two. Much
more interesting examples are the projective planes that we shall
dene later in this chapter. These are the only possibilities. See
Problem 23C. The reader can prove this fact for linear spaces as an
exercise.
In the rest of this chapter, we shall be interested in highly regular
incidence structures called t-designs. Let v, k, t and be integers
with v k t 0 and 1. A t-design on v points with
block-size k and index is an incidence structure T = (T, B, I)
with:
(i) [T[ = v,
(ii) [B[ = k for all B B,
(iii) for any set T of t points, there are exactly blocks incident
with all points in T.
So all blocks have the same size and every t-subset of the point set
is contained in the same number of blocks. Two dierent notations
19. Designs 217
for such a design are widely used, namely t-(v, k, ) design and
S

(t, k, v). We shall use both of them. A Steiner system S(t, k, v)

is a t-design with = 1, and we suppress the index in the notation.
Most of the early theory of designs originated in statistics, where 2-
designs are used in the design of experiments for statistical analysis.
These designs are often called balanced incomplete block designs
(BIBDs). Usually trivial designs are excluded from the theory: a
design with one block that contains all the points or a design that
has all the k-subsets of the point set as blocks is of course a t-design
for t k, but is not very interesting.
We give a few examples; more will follow further on.
Example 19.1. Let the nonzero vectors of F
4
2
be the points. As
blocks we take all triples x, y, z with x + y + z = 0. Any pair
x, y with x ,= y uniquely determines a third element z, dierent
from both, satisfying this equation. So we have constructed an
S(2, 3, 15). The blocks are the 2-dimensional subspaces of F
4
2
with
0 deleted.
We construct a second design by taking all the vectors as point
set and dening blocks to be 4-tuples w, x, y, z for which w +
x +y +z = 0. This denes an S(3, 4, 16). Note that if we take the
blocks that contain 0 and delete this vector, we nd the blocks of
the previous design.
Example 19.2. We take the ten edges of a K
5
as point set. Each
of the three kinds of 4-tuples shown in Fig. 19.1 will be a block.
Figure 19.1
There are 5+10+15 = 30 blocks. No triple (of edges) is contained
in more than one block. Therefore the blocks contain 120 dierent
triples, i.e. all the triples. We have constructed an S(3, 4, 10).
218 A Course in Combinatorics
Example 19.3. Let H be a normalized Hadamard matrix of order
4k. Delete the rst row and the rst column. We now identify
points with rows of this matrix. Each column denes a subset of
the rows, namely those rows for which there is a + in that column.
These subsets are the blocks. From the argument in Theorem 18.1,
we see that any pair of points is contained in exactly k1 blocks and
clearly all blocks have size 2k1. We have a 2-(4k1, 2k1, k1)
design and such a design is called a Hadamard 2-design.
Consider the same matrix H but now delete only the rst row.
Each of the other rows determines two 2k-subsets of the set of
columns. This partition is unaected if we change the sign of the
row. The argument of Theorem 18.1 now shows that for any three
columns, there are exactly k 1 of the subsets that have three
elements in these columns. So these 2k-sets are the blocks of a
3-(4k, 2k, k 1) design called a Hadamard 3-design.
Example 19.4. Consider a regular Hadamard matrix of order 4u
2
(see Example 18.2). If we replace + by 1 and by 0, we nd
a (0,1)-matrix with 2u
2
+ u ones in every row and column, and
furthermore, any two rows or columns have inner product u
2
+ u.
Let the columns be the characteristic functions of the blocks of a
design on 4u
2
points. The properties of the matrix show that this
is a 2-(4u
2
, 2u
2
+u, u
2
+u) design. One usually prefers considering
the complement of this design, i.e. a 2-(4u
2
, 2u
2
u, u
2
u) design.
Problem 19A. Here are two more examples in the spirit of Ex-
ample 19.2.
(i) Take the edges of K
6
as points of an incidence structure. The
blocks are to be all sets of three edges that either are the edges of
a perfect matching, or the edges of a triangle. Show that this is an
S(2, 3, 15) and show that it is isomorphic to the design in Example
19.1.
(ii) Take the edges of K
7
as points of an incidence structure.
The blocks are to be all sets of ve edges of these three types: (a)
claws with ve edges incident with a common vertex, (b) edge sets
of pentagon subgraphs, and (c) ve edges that form a triangle and
two disjoint edges. Show that this is an S
3
(3, 5, 21).
We now give two elementary theorems on t-designs.
19. Designs 219
Theorem 19.2. The number of blocks of an S

(t, k, v) is
(19.1) b =
_
v
t
_
/
_
k
t
_
.
Proof: Count in two ways the number of pairs (T, B), where T is
a t-subset of T and B is a block incident with all points of T. We
nd
_
v
t
_
= b
_
k
t
_
.
Theorem 19.3. Given i, 0 i t, the number of blocks incident
with all the points of an i-subset I of T is
(19.2) b
i
=
_
v i
t i
_
/
_
k i
t i
_
.
That is, every S

(t, k, v) is also an i-design for i t.

Proof: Count in two ways the number of pairs (T, B), where T
is a t-subset of T that contains I and B is a block that is incident
with all the points of T.
Corollary. If T is a t-design with point set T and block set B and
if I is a subset of T with [I[ t, then the point set TI and the
blocks BI : I B form an S

(t i, k i, v i). This design is

called the derived design T
I
.
In Example 19.1 we already saw an example of a derived design.
If we take I = 0, the derived design for S(3, 4, 16) is S(2, 3, 15).
Problem 19B. Show that if an S(3, 6, v) exists, then v 2 or 6
(mod 20).
The number of blocks incident with any point, i.e. b
1
, is usually
denoted by r (replication number). Two special cases of Theorem
19.3 are the following relations for the parameters of a 2-design:
(19.3) bk = vr,
(19.4) (v 1) = r(k 1).
220 A Course in Combinatorics
Theorem 19.4. Let 0 j t. The number of blocks of an
S

(t, k, v) that are incident with none of the points of a j-subset J

of T is
(19.5) b
j
=
_
v j
k
_
/
_
v t
k t
_
.
Proof: For x T, let B
x
be the set of blocks incident with x. We
use inclusion-exclusion, Theorem 10.1. We nd that
b
j
=
j

i=0
(1)
i
_
j
i
_
b
i
.
The result follows by substitution of (19.2) and then using (10.5).
It is quicker to observe that b
j
apparently does not depend on
the particular set J and then count in two ways the pairs (J, B),
where J is a j-subset of T and J B = . So
_
v
j
_
b
j
= b
_
vk
j
_
. Then
the result follows from Theorem 19.2.
Corollary. If i +j t, then the number of blocks of an S

(t, k, v)
that are incident with all of a set of i points and none of a disjoint
set of j points is a constant
(19.6) b
j
i
=
_
vij
ki
_
_
vt
kt
_ .
Proof: The result follows upon application of Theorem 19.4 to
the (t i)-design T
I
, where I is the set of i points.
Corollary. If J is a j-subset of T, j t, then the point set TJ
and the blocks B with B J = form an S

(t j, k, v j) called
the residual design T
J
.
Problem 19C. (i) Prove that an S

(t, k, v) with v k+t is trivial.

(ii) Prove that the complement of an S

(t, k, v) with v k +t is a
t-design and determine its parameters.
Example 19.5. Consider a Hadamard 3-design 3-(4k, 2k, k 1)
and form the residual with respect to a set with one point. We nd
19. Designs 221
a Hadamard 2-design 2-(4k 1, 2k, k), i.e. the complement of the
design of Example 19.3.
An obvious necessary condition for the existence of an S

(t, k, v)
is that the numbers b
i
of (19.2) are integers. However, this condition
is not sucient. An S(10, 16, 72) does not exist, as is demonstrated
by the following theorem due to Tits (1964).
Theorem 19.5. In any nontrivial Steiner system S(t, k, v),
v (t + 1)(k t + 1).
Proof: In a Steiner system, any two distinct blocks have at most
t1 points in common. Choose a set S of t+1 points not contained
in any block. For each set T S with [T[ = t, there is a unique
block B
T
containing T. Each such B
T
is incident with k t points
not in S, and any point not in S is incident with at most one
such block B
T
since two such blocks already have t 1 points of
S in common. This shows that the union of all blocks B
T
contains
(t + 1) + (t + 1)(k t) points and the result follows.
Given an incidence structure with [T[ = v and [B[ = b, the
incidence matrix N is the v b matrix with rows indexed by the
elements p of T, columns indexed by the elements B of B, and
with the entry N(p, B) = 1 if p is incident with B, N(p, B) = 0
otherwise. Note that the entry in row p and column q of NN

is the sum of N(p, B)N(q, B) over all blocks B, and this is the
number of blocks that contain both p and q. Dually, the entry in
row A and column B of N

N is the cardinality of A B.
Two designs T and T
/
with incidence matrices N and N
/
are
called isomorphic or equivalent if there are permutation matrices P
and Q such that N
/
= PNQ.
We shall often identify N with the design, i.e. we refer to the
columns as blocks instead of as characteristic functions of blocks.
Now if N is the incidence matrix of a 2-design, then NN

has
the entry r everywhere on the diagonal and entries in all other
positions, i.e.
(19.7) NN

= (r )I +J,
where I and J are v v matrices.
222 A Course in Combinatorics
Problem 19D. Let N be an 11 by 11 (0,1)-matrix with the fol-
lowing properties: (i) every row of N has six ones; (ii) the inner
product of any two distinct rows of N is at most 3. Show that N
is the incidence matrix of a 2-(11,6,3) design. Furthermore show
that this design is unique (up to isomorphism).
The following theorem is known as Fishers inequality.
Theorem 19.6. For a 2-(v, k, ) design with b blocks and v > k
we have
b v.
Proof: Since v > k, we have r > by (19.4). Since J has one
eigenvalue v and its other eigenvalues are 0, the matrix on the right-
hand side of (19.7) has v 1 eigenvalues (r ) and one eigenvalue
(r ) + v = rk. So it has determinant rk(r )
v1
,= 0 and
hence N has rank v. This implies that b v.
From the argument in the preceding proof, we can make a very
important conclusion, given in the next theorem.
Theorem 19.7. If a 2-(v, k, ) design has b = v blocks and v is
even, then k must be a square.
Proof: Since b = v, we have r = k. Now N is a v v matrix and
by (19.7)
(det N)
2
= k
2
(k )
v1
.
Since det N is an integer, we are done.
Theorem 19.6 was generalized by A. Ya. Petrenjuk (1968) to
b
_
v
2
_
for any S

(4, k, v) with v k +2 and nally generalized to

arbitrary t-designs by Ray-Chaudhuri and Wilson (1975).
Theorem 19.8. For an S

(t, k, v) with t 2s and v k + s, we

have b
_
v
s
_
.
Proof: We introduce the higher incidence matrices of the t-design
T = S

(t, k, v). For i = 0, 1, 2, . . . , let N

i
denote the
_
v
i
_
b ma-
trix with rows indexed by the i-element subsets of points, columns
indexed by the blocks, and with entry 1 in row Y and column B
if Y B, 0 otherwise. For 0 i j v, we use W
ij
to denote
the i-th incidence matrix of the incidence structure whose blocks
19. Designs 223
are all the j-element subsets of a v-set. Thus W
ij
is a
_
v
i
_

_
v
j
_
matrix.
We claim that
N
s
N

s
=
s

i=0
b
i
2si
W

is
W
is
.
To see this, note that N
s
N

s
has rows indexed by s-element subsets
E and columns indexed by s-element subsets F of the points, and
for given E and F, the entry in row E and column F of N
s
N

s
is
the number of blocks that contain both E and F. This number is
b
2s
, where := [E F[. The entry in row E and column F of
W

is
W
is
is the number of i-subsets of the points contained in both
E and F, i.e.
_

i
_
. So the (E, F)-entry on the right-hand side of
the equation is

s
i=1
b
i
2si
_

i
_
, and from (19.6) it follows that this is
b
2s
.
The
_
v
s
_

_
v
s
_
matrices b
i
2si
W

is
W
is
are all positive semidenite,
and b
s
s
W

ss
W
ss
= b
s
s
I is positive denite since b
s
s
> 0 (v k + s).
Therefore N
s
N

s
is positive denite and hence nonsingular. The
rank of N
s
N

s
is equal to the rank of N
s
, i.e. N
s
has rank
_
v
s
_
, and
this cannot exceed the number of columns of N
s
, which is b.
If equality holds in the WilsonPetrenjuk inequality, Theorem
19.8, then the 2s-design is called tight. The only known examples
with s > 1 and v > k +s are the unique Steiner system S(4, 7, 23)
that we treat in the next chapter, and its complement.
It is useful to give some idea of the history of t-designs. Only
nitely many Steiner systems S(t, k, v) with t 4 are known. The
most famous are the designs S(5, 8, 24) and S(5, 6, 12) found by E.
Witt (1938) and the derived 4-designs. These will appear in the
next chapter. R. H. F. Denniston (1976) constructed S(5, 6, 24),
S(5, 7, 28), S(5, 6, 48), and S(5, 6, 84). W. H. Mills (1978) con-
structed an S(5, 6, 72). Again, the derived designs are Steiner
systems. M. J. Granell and T. S. Griggs (1994) constructed an
S(5, 6, 108). Since then, no others have been found. In 1972, W. O.
Alltop constructed the rst innite sequence of 5-designs without
repeated blocks. We remark that it is easy to show that t-designs
with repeated blocks exist for any t, but for a long time many de-
sign theorists believed that nontrivial t-designs without repeated
224 A Course in Combinatorics
blocks did not exist for t > 6. The rst simple 6-design was found
by D. W. Leavitt and S. S. Magliveras in 1982, and in 1986 D. L.
Kreher and S. P. Radziszowski found the smallest possible simple
6-design, an S
4
(6, 7, 14). The big sensation in this area was the pa-
per by L. Teirlinck (1987) proving that nontrivial simple t-designs
exist for all t. His construction produces designs with tremendously
large parameters and hence the construction of small examples is
still an open problem. For a number of special parameter sets, it
has been shown that the corresponding designs do not exist.
For the remainder of this chapter we shall mainly be interested
in 2-designs. When t = 2, we often omit this parameter in the
t-(v, k, ) notation and speak of (v, k, )-designs.
A class of designs of special interest are the 2-designs with b = v.
In this case the incidence matrix N of the design is a square matrix
and these designs should be called square designs. However, the
confusing name symmetric designs is standard terminology. (Note
that N is not necessarily symmetric.) For a symmetric 2-(v, k, )
design (19.4) becomes
(v 1) = k(k 1).
Some authors use the name projective design, a name derived from
the fact that a 2-(v, k, 1) design with b = v is called a projective
plane (see Example 19.7). Despite the fact that we are not happy
with the name, we shall use the terminology symmetric designs for
these designs.
Problem 19E. Let T be a 2-(v, k, ) design with b blocks and r
blocks through every point. Let B be any block. Show that the
number of blocks that meet B is at least
k(r 1)
2
/[(k 1)( 1) + (r 1)].
Show that equality holds if and only if any block not disjoint from
B meets it in a constant number of points.
Example 19.6. Take as points the elements of Z
7
and as blocks
all triples B
x
:= x, x + 1, x + 3 with x Z
7
. It is easy to check
that this yields an S(2, 3, 7). The following Fig. 19.2 is often drawn.
19. Designs 225
The lines repesent blocks, but one block must be represented by the
circle.
Figure 19.2
This design is known as the Fano plane. The idea of the construc-
tion will be extended in Chapter 27. It is based on the fact that the
six dierences among the elements of 0, 1, 3 are exactly all the
nonzero elements of Z
7
. If we wish to nd the block containing say
1, 6, we observe that 6 1 = 1 3 and we therefore take x = 5
and nd x + 1 = 6, x + 3 = 1, i.e. the pair is indeed in B
5
. The
reader should have no diculty nding an S(2, 4, 13) in the same
way, using Z
13
.
A symmetric design with = 1 is called a projective plane. If k is
the size of the blocks, then n = k1 is called the order of the plane
(why this is done will become clear in Example 19.7). Expressed
in n, the parameters of a projective plane of order n are:
v = n
2
+n + 1, k = n + 1, = 1.
The blocks are usually called lines. The Fano plane is the (unique)
projective plane of order 2.
Example 19.7. Consider the vector space F
3
q
. This vector space
contains (q
3
1)/(q 1) = q
2
+q +1 1-dimensional subspaces and
the same number of 2-dimensional subspaces. We now construct an
incidence structure (T, B, I), where T and B are these two classes
of subspaces of F
3
q
. If a 1-dimensional subspace is contained in a
2-dimensional subspace, we say they are incident. It is immediately
clear that we have thus dened a projective plane of order q, i.e.
a 2-(q
2
+ q + 1, q + 1, 1) design. This design is usually denoted
226 A Course in Combinatorics
by PG(2, q) or PG
2
(q), which stands for projective geometry of
dimension 2 and order q.
The construction dened above can also be applied if we re-
place F
q
by 1. We then obtain the classical real projective plane,
where points are the 1-dimensional subspaces and lines are the
2-dimensional subspaces. This geometry contrasts with classical
ane geometry in the fact that no two lines are parallel. When
speaking about the designs dened above, we use terminology from
geometry.
Problem 19F. Find a subset S = s
1
, . . . , s
5
of Z
21
such that
the elements of Z
21
as points and the 21 blocks S + x (x Z
21
)
form a projective plane of order 4. (Hint: there is a solution S for
which 2S = S.)
Problem 19G. Let (R, C, S; L) be a Latin square of order 6. De-
ne T := R C. Let B be the set of blocks
B
ij
:=
(x, y) R C : x = i or y = j or L(x, y) = L(i, j) (i, j)
for (i, j) R C.
(1) Show that this denes a 2-(36,15,6) design.
(2) Show that a regular Hadamard matrix of order 36 exists.
Problem 19H. Let T be a 3-(v, k, ) design. Suppose that the
derived design of T with respect to a point p (i.e. the case i = 1 in
the Corollary to Theorem 19.3) is a symmetric design.
(1) Show that (v 2) = (k 1)(k 2).
(2) Show that any two blocks of T meet in 0 or + 1 points.
(3) Show that the set of points not on a block B together with
the blocks disjoint from B form a 2-design T
B
.
(4) Apply Fishers inequality to the design T
B
and deduce that
v = 2k or otherwise k = (+1)(+2) or k = 2(+1)(+2).
What are the possibilities for the design T? Do we know any
designs with these properties?
Problem 19I. Let O be a subset of the points of a projective
plane of order n such that no three points of O are on one line.
19. Designs 227
Show that [O[ n+1 if n is odd and that [O[ n+2 if n is even.
A set of n + 1 points, no three on a line, is called an oval; a set of
n + 2 points, no three on a line, is a hyperoval. Two constructions
of PG
2
(4) were given in Example 19.7 and Problem 19F. In each
case, construct a hyperoval.
Problem 19J. Let O be a hyperoval (with q+2 points) in PG
2
(q),
q = 2
m
. Any of the q
2
1 points p / O has the property that there
are exactly
1
2
(q +2) secants of O through p. Take ve points on O
and split them into
p
1
, p
2
, p
3
, p
4
, p
5
.
This can be done in 15 ways. The two pairs determine two secants
that meet in a point p / O. The line through p and p
5
meets O in
a point, that we call p
6
. This denes 15 (not necessarily distinct)
6-tuples of points on O, containing the given ve points. This
denes an S

(5, 6, q + 2), a construction due to D. Jungnickel and

S. A. Vanstone (1987). Show that the points with coordinates (as
in Example 19.7) (1, t, t
2
), t F
q
, with (0,1,0) and (0,0,1) form a
hyperoval O in PG
2
(q). Show that a suitable choice of p
1
to p
5
on O with the construction explained above yields repeated blocks,
i.e. the 5-design is not simple.
Any 2-(n
2
, n, 1) design is called an ane plane. The points and
lines of the plane (= 2-dimensional vector space) F
2
q
form an ane
plane of order q. For such a design we use the notation AG
2
(n)
(2-dimensional ane geometry of order n).
Example 19.8. Let T be a projective plane of order n. If we
delete one line and all the points on that line, we nd an ane
plane of order n.
Problem 19K. Let T be any ane plane of order n. If B
1
and
B
2
are two blocks, then we write B
1
B
2
if the two blocks are
the same or if they have no points in common. Show that is
an equivalence relation. A class of this relation is called a parallel
class. Show that there exists a projective plane of order n such that
T can be obtained from that plane by the construction of Example
19.8.
228 A Course in Combinatorics
We shall now show that if N is the incidence matrix of a symmet-
ric design T, then N

is also the incidence matrix of a symmetric

design T

, called the dual of T.

Theorem 19.9. Let N be the incidence matrix of a symmetric 2-
(v, k, ) design. Then N

is also the incidence matrix of a design.

Proof: Consider any block B of the design. For 0 i k let a
i
be the number of blocks (,= B) that have i points in common with
B. Then counting blocks, pairs (p, B
/
) with p B B
/
and triples
(p, q, B
/
) with p ,= q and p, q B B
/
we nd:
k

i=0
a
i
= v 1,
k

i=0
ia
i
= k(k 1),
k

i=0
_
i
2
_
a
i
=
_
k
2
_
( 1),
from which we nd

k
i=0
(i )
2
a
i
= 0. Hence, any block B
/
,= B
has points in common with B, i.e. N

N = (k )I +J.
Note that in Example 19.7, we did not need to specify whether
the set T was the 1-dimensional subspaces or the 2-dimensional
subspaces. In the latter situation, we have the dual of the former.
In many cases the designs T and T

are not isomorphic.

Let T be a symmetric 2-(v, k, ) design. There are two other
ways to obtain a design from T. These two designs are called the
derived design and residual design of T. This could be somewhat
confusing since we have already introduced that terminology. We
shall always indicate which of the two we mean. Take any block B
of T. The residual of T with respect to B has TB as point set and
as blocks all B
/
B with B
/
,= B. It is a 2-(v k, k , ) design.
The derived design has B as point set and as blocks all B
/
B with
B
/
,= B. It is a 2-(k, , 1) design. If a design with parameters
v, k, b, r, is the residual of a symmetric design, then r = k + .
Any 2-design for which this equation holds is called a quasiresidual
design. If such a design is not the residual of a symmetric design,
then we say that it is nonembeddable. The assertion of Problem
19K is that every ane plane is embeddable in a projective plane.
A theorem due to W. S. Connor (1952), that we shall leave until
Chapter 21, states that every quasiresidual design with = 2 is
embeddable.
19. Designs 229
Example 19.9. Let C :=
_
0 1 0
0 0 1
1 0 0
_
and let E
i
denote a 3 by 3
matrix with ones in column i and zeros elsewhere. Then
N :=
_
_
E
1
I I I
E
2
I C C
2
E
3
I C
2
C
_
_
is the 9 by 12 incidence matrix of AG
2
(3). Dene
A :=
_
_
_
_
_
_
1 1 1 1 1 1 0 0 0 0 0 0
1 1 1 0 0 0 1 1 1 0 0 0
1 1 1 0 0 0 0 0 0 1 1 1
0 0 0 1 1 1 1 1 1 0 0 0
0 0 0 1 1 1 0 0 0 1 1 1
0 0 0 0 0 0 1 1 1 1 1 1
_
_
_
_
_
_
, B :=
_
_
_
_
_
_
_
_
_
_
_
1 1 0 0
1 1 0 0
1 1 0 0
1 0 1 0
1 0 1 0
1 0 1 0
1 0 0 1
1 0 0 1
1 0 0 1
_
_
_
_
_
_
_
_
_
_
_
.
Form the 24 by 16 matrix
D :=
_
_
A O
N B
N J B
_
_
.
One easily checks that D

is the 16 by 24 incidence matrix of a 2-

(16,6,3) design. This is a quasiresidual design. However, it cannot
be the residual of a 2-(25,9,3) symmetric design because the inner
product of row i + 6 and row i + 15 of D, 1 i 9, equals 4 and,
by Theorem 19.9, the inner product of the columns of the incidence
matrix of a 2-(25,9,3) design is 3. This shows that nonembeddable
designs with = 3 exist.
The combination of a counting argument and a suitable quadratic
form that we used to prove Theorem 19.9 is widely used in combina-
torics. However, sometimes it is easier to use algebraic methods as
we shall demonstrate in the following theorem, due to Ryser. (The
reader can try to prove the theorem by using counting arguments.)
230 A Course in Combinatorics
Theorem 19.10. Let T = (T, B, I) be an incidence structure with
[T[ = [B[ = v, block-size k, such that any two blocks meet in
points. Then T is a symmetric 2-design.
Proof: Let N be the incidence matrix of T. Then
(19.8) N

N = (k )I +J,
and
(19.9) JN = kJ.
By Theorem 19.9, we are done if we can show that NJ = kJ. From
(19.8) we see that N is nonsingular and hence (19.9) can be read
as J = kJN
1
. From (19.8) we nd JN

N = (k + v)J and
therefore
JN

= (k +v)JN
1
= (k +v)k
1
J,
i.e. N has constant row-sums. Then these row-sums must be k.
This proves the theorem and yields (k +v)k
1
= k as was to
be expected from (19.3) and (19.4).
As a preparation for the best known nonexistence theorem for
designs, we need two results, both due to Lagrange. For the rst,
consider the matrix H of (18.6) with n = 1, i.e. A
i
= (a
i
). Dene
y = (y
1
, y
2
, y
3
, y
4
) by y := xH, where x = (x
1
, x
2
, x
3
, x
4
). Then
from (18.7) we nd
(19.10) (a
2
1
+a
2
2
+a
2
3
+a
2
4
)(x
2
1
+x
2
2
+x
2
3
+x
2
4
) = (y
2
1
+y
2
2
+y
2
3
+y
2
4
).
Using this identity, Lagrange proved that every integer is the sum
of four squares. Clearly the identity shows that it is sucient to
prove this for primes. For an elegant proof that a prime is the sum
of four squares we refer to Chandrasekharan (1968).
The following nonexistence theorem is known as the Bruck
RyserChowla theorem.
Theorem 19.11. If v, k, are integers such that (v 1) = k(k
1), then for the existence of a symmetric 2-(v, k, ) design it is
necessary that:
(i) if v is even then k is a square;
(ii) if v is odd, then the equation z
2
= (k)x
2
+(1)
(v1)/2
y
2
has a solution in integers x, y, z, not all zero.
19. Designs 231
Proof: Assertion (i) was proved in Theorem 19.7. So assume that
v is odd. Let T be a symmetric 2-(v, k, ) design with incidence
matrix N = (n
ij
) and write n := k . We now introduce v linear
forms L
i
in the variables x
1
, . . . , x
v
by
L
i
:=
v

j=1
n
ij
x
j
, 1 i v.
Then the equation N

N = (k )I +J implies that
(19.11) L
2
1
+ +L
2
v
= n(x
2
1
+ +x
2
v
) +(x
1
+ +x
v
)
2
.
By Lagranges theorem, n can be written as n = a
2
1
+ +a
2
4
. This
and (19.10) allow us to take four of the variables x
j
and write
(19.12) n(x
2
i
+x
2
i+1
+x
2
i+2
+x
2
i+3
) = (y
2
i
+y
2
i+1
+y
2
i+2
+y
2
i+3
),
where each y
j
is a linear form in the four variables x
i
, . . . , x
i+3
.
We now rst assume that v 1 (mod 4). By applying this to
(19.11), four variables at a time, and introducing w for x
1
+ +x
v
,
we reduce (19.11) to
(19.13) L
2
1
+ +L
2
v
= y
2
1
+ +y
2
v1
+nx
2
v
+w
2
.
Since H in (18.6) is invertible, we can express the variables x
j
for
1 j v 1 as linear forms in the corresponding y
j
and hence
w is a linear form in these variables and x
v
. Next, we reduce the
number of variables in the following way. If the linear form L
1
,
expressed in y
1
, . . . , y
v1
, x
v
, does not have coecient +1 for y
1
,
then we set L
1
= y
1
, and if the coecient is +1, we set L
1
= y
1
,
and in both cases, we subsequently solve this equation for y
1
as
a linear expression in the remaining variables y
j
and x
v
. This is
substituted in the expression w. So (19.11) has been reduced to
L
2
2
+ +L
2
v
= y
2
2
+ +y
2
v1
+nx
2
v
+w
2
.
We proceed in this way for y
2
, . . . , y
v1
. In each step, w is replaced
by another linear form in the remaining variables, and hence we
end up with
L
2
v
= nx
2
v
+w
2
,
232 A Course in Combinatorics
in which both L
v
and w are rational multiples of the variable x
v
.
If we multiply this by the common denominator of the factors, we
nd an equation
z
2
= (k )x
2
+y
2
in integers. This proves the assertion if v 1 (mod 4). If v
3 (mod 4), the same procedure is applied to (19.13) after adding
nx
2
v+1
to both sides, where x
v+1
is a new variable. The equation is
then nally reduced to nx
2
v+1
= y
2
v+1
+ w
2
and again we multiply
by a common denominator to nd an equation of type
(k )x
2
= z
2
+y
2
in accordance with assertion (ii).
Example 19.10. From Example 19.7, we know that a projective
plane of order n exists for 2 n 9, except possibly for n = 6.
By Theorem 19.11, a necessary condition for the existence of a
projective plane of order 6 is that the equation z
2
= 6x
2
y
2
has a
nontrivial solution. If such a solution exists, then also one for which
x, y, and z have no prime factor in common, i.e. z and y are both
odd. Then z
2
and y
2
are both 1 (mod 8). Since 6x
2
(mod 8) is
either 0 or 6, we see that the equation has only the trivial solution
(0,0,0). Therefore a projective plane of order 6 does not exist.
If we try the same thing for a plane of order 10, we nd the
equation z
2
= 10x
2
y
2
, which has the solution x = 1, y = 1,
z = 3. In this case Theorem 19.11 tells us nothing. In 1989, Lam
et al. announced that a computer search involving several hundred
hours on a Cray 1 had excluded the existence of a projective plane
of order 10. This is the only case where the nonexistence of a
symmetric 2-design has been shown using something other than
Theorem 19.11.
Corollary. If there exists a projective plane of order n 1 or 2
(mod 4), then n is the sum of two integral squares.
Proof: The condition n 1 or 2 (mod 4) implies that v = n
2
+
n+1 3 (mod 4). Theorem 19.11 asserts that n is the sum of two
rational squares. It is well known that n is the sum of two rational
squares if and only if n is the sum of two integral squares. (This
19. Designs 233
follows from the condition that n is the sum of two integral squares
if and only if no prime divisor of the square-free part of n is 3
(mod 4).)
Problem 19L. Show that a symmetric 2-(29,8,2) design does not
exist.
Problem 19M. Suppose M is a rational square matrix of order v
and that MM

= mI. Show that if v is odd, then m is a square.

Show that if v 2 (mod 4), then m is the sum of two rational
squares.
(Note that one consequence of this latter result is that the exis-
tence of a conference matrix of order n 2 (mod 4) implies that
n 1 is the sum of two squares.)
A great deal of work has been done on the construction of 2-
designs. We shall only treat a number of examples that will give
some idea of the kind of methods that have been used. The smallest
nontrivial pair (k, ) to consider is (3, 1). A 2-(v, 3, 1) design is
called a Steiner triple system. One uses the notation STS(v) for
such a design. By (19.3) and (19.4), a necessary condition for the
existence of such a design is that v 1 (mod 6) or v 3 (mod 6).
We shall show that this condition is also sucient. This will be done
by direct construction in Examples 19.11 and 19.15. However, it is
useful to see a number of examples of a more complicated approach.
The methods that we demonstrate can be used for the construction
of other designs than Steiner triple systems. Furthermore, they can
be used to produce designs with certain subdesigns (see Problem
19N) or prescribed automorphism group. The idea of this approach
is to nd direct constructions for small examples and some recursive
constructions, and subsequently show that, for any v that satises
the necessary conditions, an STS(v) can be constructed by the
recursive methods, using the list of known small examples. We
shall see below that this in fact reduces to a (not very dicult)
problem in number theory. As stated above, we restrict ourselves
to a number of examples. The reader may wish to try to show that
our examples suce to nd an STS(v) for all possible values of v,
without using Examples 19.11 and 19.15.
We consider the trivial design with only one block of size 3
234 A Course in Combinatorics
as STS(3). We have already seen constructions of STS(7) =
PG
2
(2) and STS(9) = AG
2
(3). In Example 19.1, we constructed
an STS(15).
Example 19.11. Let n = 2t + 1. We dene T := Z
n
Z
3
. As
blocks we take all triples (x, 0), (x, 1), (x, 2) with x Z
n
and all
triples (x, i), (y, i), (
1
2
(x +y), i + 1) with x ,= y in Z
n
and i Z
3
.
This simple construction provides an STS(6t + 3) for every t.
Example 19.12. Let q = 6t + 1 be a prime power and let be
a primitive element in F
q
, i.e. F

q
is a cyclic group generated by .
We dene
(19.14) B
i,
:=
i
+,
2t+i
+,
4t+i
+, 0 i < t, F
q
.
We claim that the elements of F
q
as points and the blocks B
i,
form
an STS(q). The idea of the proof is the same as in Example 19.6.
Note that
6t
= 1,
3t
= 1 and dene s by
s
= (
2t
1). We
consider the six dierences of pairs from B
0,0
. These are:

2t
1 =
s
, (
2t
1) =
s+3t
,

4t

2t
=
s+2t
, (
4t

2t
) =
s+5t
,

6t

4t
=
s+4t
, (1
4t
) =
s+t
.
It follows that for any ,= 0 in F
q
, there is a unique i, 0 i < t,
such that occurs as the dierence of two elements of B
i,0
. Hence
for any x and y in F
q
, there is a unique i and a unique F
q
such
that the pair x, y occurs in the block B
i,
.
The method of Examples 19.6 and 19.12 is known as the method
of dierences. Example 19.15 will show a more complicated use of
the same idea.
We now know that an STS(v) exists for v = 13, 19, 25, 31, 37, 43
and 49 as well as the values mentioned above. This includes all
v 1 (mod 6) less than 50. In fact, we now know at least one
STS(v) for each feasible value of v less than 100, except v = 55,
v = 85, v = 91.
19. Designs 235
Example 19.13. Let there be an STS(v
i
) on the point set V
i
(i = 1, 2). We take V
1
V
2
as a new point set and dene as blocks
all triples: (x
1
, y
1
), (x
2
, y
2
), (x
3
, y
3
) for which
(1) x
1
= x
2
= x
3
and y
1
, y
2
, y
3
is a block of STS(v
2
);
(2) x
1
, x
2
, x
3
is a block of STS(v
1
) and y
1
= y
2
= y
3
;
(3) x
1
, x
2
, x
3
is a block of STS(v
1
) and y
1
, y
2
, y
3
is a block
of STS(v
2
).
It is practically obvious that this denes an STS(v
1
v
2
). The reader
should check that we have dened the correct number of blocks.
This construction provides us with an STS(91).
Example 19.14. We show a slightly more complicated construc-
tion. Suppose that we have an STS(v
1
) on the point set V
1
=
1, 2, . . . , v
1
with block set S
1
, and furthermore suppose that the
blocks that are completely contained in V = s +1, . . . , v
1
, where
s = v
1
v, form an STS(v). Let S
2
be the set of triples of an
STS(v
2
) on the point set V
2
= 1, 2, . . . , v
2
.
We consider a new point set
T := V (x, y) : 1 x s, 1 y v
2
.
This set has v + v
2
(v
1
v) points. We introduce a set B of four
kinds of blocks:
(1) those of the subsystem STS(v);
(2) (a, y), (b, y), c with c V , a, b, c S
1
and y V
2
;
(3) (a, y), (b, y), (c, y) with a, b, c a block in S
1
with no point
in V , and y V
2
;
(4) (x
1
, y
1
), (x
2
, y
2
), (x
3
, y
3
), where y
1
, y
2
, y
3
is a block in S
2
and the integers x
1
, x
2
, x
3
satisfy
x
1
+x
2
+x
3
0 (mod s).
Again, one easily checks that any two points of T uniquely deter-
mine a block in B. Hence T and B are the points and blocks of a
Steiner triple system on v + v
2
(v
1
v) points. A simple example
is obtained by letting the subsystem be just one block, i.e. v = 3.
Taking v
1
= 7, v
2
= 13, we nd an STS(55).
We have thus constructed an STS(v) for every feasible value of
v less than 100, except v = 85.
236 A Course in Combinatorics
Problem 19N. (a) Show that if an STS(v
1
) and an STS(v
2
) both
exist, then there is an STS(v
1
v
2
v
2
+1). Use this construction to
nd an STS(85).
(b) Construct an STS(15) on the set 0, 1, . . . , 14 such that it
contains a Fano plane on 0, 1, . . . , 6 as a subsystem.
Example 19.15. Consider as point set Z
2t
Z
3
. Addition
of elements is coordinatewise with the extra convention +(x, i) =
. For notational convenience we sometimes write the second co-
ordinate as an index, i.e. x
i
instead of (x, i). We now dene four
types of base blocks:
(1) 0
0
, 0
1
, 0
2
;
(2) , 0
0
, t
1
, , 0
1
, t
2
, , 0
2
, t
0
;
(3) 0
0
, i
1
, (i)
1
, 0
1
, i
2
, (i)
2
, 0
2
, i
0
, (i)
0
, i = 1, . . . , t1;
(4) t
0
, i
1
, (1i)
1
, t
1
, i
2
, (1i)
2
, t
2
, i
0
, (1i)
0
, i = 1, . . . , t.
We have 6t + 1 base blocks. For a = 0, 1, . . . , t 1 we add the
element (a, 0) (i.e. a
0
) to each of the elements of every base block,
thus producing t(6t +1) blocks. We claim that these are the triples
of an STS(6t+1). It is trivial that the base blocks of type 2 yield a
set of blocks in which every pair of points, one of which is , occurs
exactly once. The cyclic nature of the denition of the base blocks
shows that it is sucient for us to check that all pairs a
0
, b
0
with
a ,= b and all pairs a
0
, b
1
occur in the triples we have dened. If
a < b and b a = 2s, then the pair a
0
, b
0
occurs in the triple
obtained from 0
2
, s
0
, (s)
0
translated by the element (b s, 0).
Similarly, if b a is odd, we nd the required pair by translating a
base block of type 4. Now consider a pair a
0
, b
1
. If a = b t 1,
we nd the pair by translating the base block of type 1 by (a, 0).
If a ,= b and a < t, we have to look for the pair in a translate of a
base block of type 2 or of type 3. We must search for a base block
in which the dierence ba occurs as y x for two elements y
1
, x
0
.
For type 2, this dierence is t and in the blocks 0
0
, i
1
, (i)
1
we
nd the dierences i, 1 i t 1, and i = 2t i, 1 i t 1,
indeed every dierence once! Now, the rest of the details can be
left as an exercise.
This example shows that if v = 6t + 1, then an STS(v) exists.
Combined with Example 19.11 we have a construction for every
feasible value of v.
19. Designs 237
We end this chapter with an amusing application of the Fano
plane. At present the idea is not used in practice but the problem
itself has a practical origin, and maybe some day generalizations of
the following method will be used. Suppose one wishes to store one
of the integers 1 to 7 in a so-called write-once memory. This is a
binary memory, originally lled with zeros, for which it is possible
to change certain bits to ones but not back again, i.e. the state 1 is
permanent. This happens in practice with paper tape, into which
holes are punched, or compact discs, where a laser creates pits in
certain positions. In both cases, we cannot erase what was written
in the memory. To store the integers 1 to 7, we need a memory of
three bits. What if one wishes to use the memory four consecutive
times? The simplest solution is to have a 12-bit memory that is
partitioned into four 3-bit sections, one for each consecutive usage.
We assume that the memory is very expensive and we would like
to be able to use a shorter memory for the same purpose. We shall
now show that seven bits suce, a saving of more than 40%.
Figure 19.3
Let T = 1, 2, . . . , 7 be the set of points of PG
2
(2) and let
L denote the set of lines. To store one of the integers 1 to 7 in a
memory with positions numbered 1 to 7, we use the following rules.
As a general rule: if we wish to store i and the memory is in a state
corresponding to i (from a previous usage), then we do nothing.
Otherwise the rules are:
(1) if the memory is empty, store i by putting a 1 in position i;
(2) to store j when the memory is in state i, put a 1 in position
k, where i, j, k L;
(3) to store i when the memory contains two 1s, not correspond-
ing to i, put in two more 1s, such that i is one of the four
238 A Course in Combinatorics
1s and the other three form a line in L. No matter what
the two original 1s were, this is possible (sometimes in two
ways);
(4) if the memory contains four 1s, we may assume that we are
in the situation of Fig. 19.3. To store 3, we do nothing (by
the general rule); to store one of the missing numbers, we
put 1s in the other two positions; to store 1, 2, or 4, store a 1
in the empty position on the line through 3 and the number
we wish to store.
We leave it as an exercise for the reader to formulate the rules
for reading the memory. Note that the memory uniquely reads
the integer presently stored in the memory but it cannot see how
often an integer has been stored or what was stored on the previous
usage.
Problem 19O. (i) Suppose / is a family of subsets of an n-set
X, so that any member of / has odd cardinality, but such that
any two distinct members of / meet in an even number of points.
Show that [/[ n.
(ii) Suppose members of / have even cardinality but any two
intersect in an odd number of points. Show that [/[ n + 1. Can
you nd examples where equality holds?
Problem 19P. Consider 2-(v, k, = 2) designs where v =
_
k+1
2
_
.
(i) Find an example with k = 3. (You might try ve points on a
circle and one in the middle.)
(ii) Let A
1
, A
2
, . . . , A
b
be the blocks and for i = 2, 3, . . . , b, let

i
= [A
i
A
1
[. Compute
b

i=2

i
,
b

i=2

i
(
i
1),
b

i=2
(
i
1)(
i
2)
in terms of k. What can you say about the
i
s?
Problem 19Q. A Generalized Steiner System is an incidence struc-
ture with a point set T of size v and a set B of subsets of T, again
called blocks, where [B[ = b, such that every t-subset of T is in a
unique block. We do not require the blocks to have constant size
19. Designs 239
but we rule out the trivial case where b = 1. Denote by b
t,v
the min-
imal value of b for a nontrivial Generalized Steiner System. Prove
that for t 2
b
t,v
(b
t,v
1) t
_
v
t
_
.
Note that for t = 2, this result is the same as Theorem 19.1. Much
stronger bounds are known but not as easy to derive.
Problem 19R. We shall say that a point in the Fano plane is a
representative of a line if it is incident with that line. How many
SDRs does the Fano plane have?
Problem 19S. Construct 3-(2
k
, 4, 1) designs, for all k 2.
Problem 19T. We say that a design can be extended if it is a
derived design of a design, called the extended design. Prove that
if a symmetric design can be extended twice, it must be a 2-(21, 5, 1)
design.
Problem 19U. Consider the graph G of Problem 1J. Dene an
incidence structure I with as points the vertices of G and as lines
the sets (x), for vertices x G. Show that the properties of G
imply that I is a projective plane. (Cf. Problem 21Q.)
Problem 19V. We use the hypothesis of, and the notation intro-
duced in the proof of, Theorem 19.8.
(i) Check that W
is
N
s
=
_
ki
si
_
N
i
.
(ii) Show that the product of any two of W

is
W
is
, i = 0, 1, . . . , s,
is a linear combination of these s + 1 matrices. So the linear span
/ of these matrices is closed under multiplication.
(iii) Suppose equality holds in Theorem 19.8 so that the matrix
N
s
is square. Then N

s
M
1
N
s
= I where M :=

s
i=0
b
i
2si
W

is
W
is
.
Since M /, part (ii) implies M
1
/. Use this to prove that
there is a polynomial f(x) of degree s so that f([A B[) = 0 for
all distinct blocks A, B. (Hence there are at most s intersection
numbers.) This generalizes Theorem 19.9.
Notes.
The rst occurrence of a 2-design may be AG
2
(3) in a paper by
Pl ucker (1839). One usually attributes the introduction of Steiner
240 A Course in Combinatorics
systems to Woolhouse (1844); of course not to Steiner! Quite often
they are said to originate with a problem of T. P. Kirkman (1847).
T. P. Kirkman (18061895), a self-educated man, was a minister of
the Church of England. He was an amateur mathematician with
many contributions to the subject. Probably the best known is his
15 schoolgirls problem. The problem is to arrange 15 schoolgirls
in parties of three for seven days walks such that every two of
them walk together exactly once. This amounts to constructing an
STS(15) for which the set of triples can be partitioned into seven
parallel classes.
Jakob Steiner (17961863) was an important geometer of his
time. He became interested in what we now call Steiner systems in
1853 when he studied the conguration of 28 double tangents of a
plane quartic curve.
Sir Ronald A. Fisher (18901962) is considered to be one of the
most prominent statisticians. Besides important contributions to
statistics (multivariate analysis) and genetics, he is known for his
work on the application of statistical theory to agriculture and the
design of experiments. The applications to the design of agricul-
tural experiments account for our usage of v for the number of
points of a design (varieties) and r for the number of blocks through
a point (replication number).
Theorem 19.7 is due to M. P. Schutzenberger (1949).
R. A. Fisher was not the only statistician to contribute to the
mathematical theory of designs. In fact, we should probably con-
sider the Indian mathematician R. C. Bose (19011987) to be the
most important one. Many of the construction methods described
in this chapter (such as the method of dierences) are due to him.
G. Fano (18711952), whose name has become attached to the
plane PG
2
(2), was important in the Italian school of projective
geometry.
Projective planes over nite elds were rst studied by K. G. C.
von Staudt (17981867) in his book Geometrie der Lage (1856).
The rst example of a nonembeddable quasiresidual design was
given by Bhattacharya (1944). It was also a 2-(16,6,3) design. How-
ever, Example 19.9 is a much simpler example.
J. L. Lagrange (17361813) was born and educated in Italy but
19. Designs 241
he is considered a French mathematician (who studied in Berlin).
Besides many important contributions to analysis, he is known for
several theorems in number theory.
The BruckRyserChowla theorem, Theorem 19.11, is so well
known that it is often referred to as just BRC.
The idea of Example 19.15 is due to Skolem (1958). His method
was actually slightly dierent. It has led to the term Skolem se-
quences. These have other applications, e.g. in radioastronomy.
Here is the idea. Partition the set 1, 2, . . . , 2n into pairs a
i
, b
i

such that b
i
a
i
= i, 1 i n. This is a Skolem sequence. For
example, 9, 10, 2, 4, 5, 8, 3, 7, 1, 6 is such a partition for
n = 5. Now form the triples 0, a
i
+n, b
i
+n and consider these as
base blocks (mod 6n+1). Since all the dierences 1, 2, . . . , 3n and
their negatives occur exactly once, the blocks form an STS(6n+1).
For an interesting application of the Golay code (see Chapter 20)
to write-once memories, see Cohen et al. (1986). They show the
possibility of three successive writings of 11 bits on 23 positions.
References.
W. O. Alltop (1972), An innite class of 5-designs, J. Combinatorial
Theory (A) 12, 390395.
K. N. Bhattacharya (1944), A new balanced incomplete block de-
sign, Science and Culture 9, 108.
R. H. Bruck and H. J. Ryser (1949), The non-existence of certain
nite projective planes, Canad. J. Math. 1, 8893.
N. G. de Bruijn and P. Erd os (1948), On a combinatorial problem,
Proc. Kon. Ned. Akad. v. Wetensch. 51, 12771279.
K. Chandrasekharan (1968), Introduction to Analytic Number The-
ory, Springer-Verlag.
S. Chowla and H. J. Ryser (1950), Combinatorial problems, Canad.
J. Math. 2, 9399.
G. D. Cohen, P. Godlewski, and F. Merkx (1986), Linear binary
codes for write-once memories, IEEE Trans. Information Theory
32, 697700.
W. S. Connor, Jr. (1952), On the structure of balanced incomplete
block designs, Ann. Math. Stat. bf 23, 5771; correction ibid. 24,
135.
242 A Course in Combinatorics
R. H. F. Denniston (1976), Some new 5-designs, Bull. London Math.
Soc. 8, 263267.
G. Fano (1892), Giornale di Matimatiche 30, 114124.
M. J. Granell and T. S. Griggs (1994), A Steiner systemS(5, 6, 108),
Discrete Mathematics 125, 183186.
D. Jungnickel and S. A. Vanstone (1987), Hyperfactorizations of
graphs and 5-designs, J. Univ. Kuwait (Sci) 14, 213223.
T. P. Kirkman (1847), On a problem in combinations, Cambridge
and Dublin Math. J. 2, 191204.
D. L. Kreher and S. P. Radziszowski (1986), The existence of simple
6-(14,7,4) designs, J. Combinatorial Theory (A) 41, 237243.
C. W. Lam, S. Swiercz, and L. Thiel (1989), The nonexistence of
nite projective planes of order 10, Canad. J. Math. 41, 1117
1123.
D. W. Leavitt and S. S. Magliveras (1982), Simple 6-(33,8,36)-
designs from PL
2
(32), pp. 337352, in: Computational Group
Theory, Proc. Durham 1982.
W. H. Mills (1978), A new 5-design, Ars Combinatoria 6, 193195.
A. Ya. Petrenjuk (1968), On Fishers inequality for tactical cong-
urations (in Russian), Mat. Zametki 4, 417425.
D. K. Ray-Chaudhuri and R. M. Wilson (1975), On t-designs, Os-
aka J. Math. 12, 737744.
H. J. Ryser (1963), Combinatorial Mathematics, Carus Math.
Monograph 14.
M. P. Schutzenberger (1949), A non-existence theorem for an in-
nite family of symmetrical block designs, Ann. Eugenics 14,
286287.
Th. Skolem (1958), Some remarks on the triple systems of Steiner,
Math. Scand. 6, 273280.
J. Steiner (1853), Combinatorische Aufgabe, J. f. d. reine u. angew.
Mathematik 45, 181182.
L. Teirlinck (1987), Nontrivial t-designs without repeated blocks
exist for all t, Discrete Math. 65, 301311.
J. Tits (1964), Sur les syst`emes de Steiner associes aux trois grands
groupes de Mathieu, Rend. Math. e Appl. (5) 23, 166184.
E. Witt (1938), Die 5-fach transitiven Gruppen von Mathieu, Abh.
Math. Sem. Univ. Hamburg 12, 256264.
19. Designs 243
W. S. B. Woolhouse (1844), Prize question 1733, Ladys and Gen-
tlemans Diary.
20
Codes and designs
We introduce some more terminology from the theory of error-
correcting codes. In the most general sense, a code of length n
is simply a subset C S
n
, where S is a nite set (the alphabet).
Elements of C are called codewords. A binary code is one with
alphabet S = 0, 1; a ternary code is one with S = 0, 1, 2.
The distance d(x, y) between two words (vectors) x and y in S
n
is
dened to be the number of positions in which they dier, i.e.
(20.1) d(x, y) := [i : 1 i n, x
i
,= y
i
[.
This is indeed a distance function in the usual sense; check that it
satises the triangle inequality.
The concept of distance has led to the usage of geometric termi-
nology, e.g. the set B
r
(x) := y F
n
q
: d(x, y) r is called the
sphere with radius r and center x, though actually the name ball
would be better.
The minimum distance d of the code C is
(20.2) d := mind(x, y) : x C, y C, x ,= y.
Much of coding theory is concerned with linear codes. By a q-
ary [n, k] code, we mean a linear subspace C of dimension k of the
vector space F
n
q
.
The weight w(x) of x is dened by
(20.3) w(x) := d(x, 0).
This can be dened whenever 0 is one of the symbols (elements of
the alphabet), but is especially meaningful for linear codes. When
20. Codes and designs 245
C is linear, the distance between codewords x and y is equal to the
weight of x y, which is another codeword, and so the minimum
distance of C is equal to the minimum weight, i.e. the minimum of
the weights of nonzero codewords. We use the notation [n, k, d] code
for an [n, k] code with minimum distance at least d. If d = 2e + 1,
then C is called an e-error-correcting code.
The covering radius (C) of the code C is dened to be the
minimal R such that the spheres with radius R and codewords as
centers cover S
n
, i.e.
(20.4) (C) := maxmind(x, c) : c C : x S
n
.
In Chapter 18, we mentioned the repetition code in 0, 1
n
, i.e. the
1-dimensional subspace of F
n
2
containing only 0 and 1. If n = 2e+1
then every word has distance e to exactly one codeword. So this
code has covering radius e. The two spheres of radius e around the
two codewords are disjoint and they cover the space.
In general, we call a not necessarily linear code C S
n
an (e-
error-correcting) perfect code when [C[ > 1 and when every x
S
n
has distance e to exactly one codeword. This is equivalent
to C having minimum distance d = 2e + 1 and covering radius
e. Clearly, perfect codes are combinatorially interesting objects.
However, they are extremely rare.
Theorem 20.1. If C is a code in S
n
with distance d 2e + 1,
then
(20.5) [C[
e

i=0
_
n
i
_
(q 1)
i
q
n
.
Proof: The sum on the left-hand side of (20.5) counts the number
of words in a sphere of radius e.
The bound given in this theorem is known as the sphere packing
bound or as the Hamming bound. If equality holds, then the code
is perfect.
Problem 20A. Show that if a [23, 12, 7] binary code exists, then
this code is perfect.
246 A Course in Combinatorics
Problem 20B. By (20.5), a binary code of length 6 and minimum
distance 3 has at most 9 codewords. Show that equality cannot
hold. (However, 8 is possible.)
Two codes are called equivalent if one is obtained from the other
by some permutation of the coordinate positions in S
n
. Sometimes
this denition is extended by also allowing a permutation of the
elements of S such as interchanging +1 and 1 when S = F
3
.
A k n matrix G is called a generator matrix of the [n, k] code C
if C is spanned by the rows of G. Elementary linear algebra shows
that C is equivalent to a code with a generator matrix G = (I
k
P),
where P is some k nk matrix. This is called the reduced echelon
form for a generator.
The dual C

of C is dened by
(20.6) C

:= x F
n
q
:
cC
x, c) = 0.
If H is a generator matrix for C

, then clearly
(20.7) C = x F
n
q
: xH

= 0.
H is called a parity check matrix for the code C. If G = (I
k
P) is a
generator matrix, then H = (P

I
nk
) is a parity check matrix.
If C = C

, then C is called a self-dual code. If C C

, then C is
called self-orthogonal.
If C is a linear code in F
n
q
, then the extended code C is dened
by
(20.8)
C := (c
1
, . . . , c
n
, c
n+1
) : (c
1
, . . . , c
n
) C, c
1
+ +c
n+1
= 0.
The symbol c
n+1
is called the parity check symbol.
Example 20.1. Let n = (q
k
1)/(q 1). Consider a matrix
H of size k n, with entries in F
q
, for which the columns are
pairwise linearly independent. Note that this is the maximal value
of n for which this is possible. Then H is clearly the parity check
matrix of an [n, n k] code with minimum distance 3. Such a
code is called a q-ary Hamming code. If c is a codeword, then
20. Codes and designs 247
[B
1
(c)[ = 1+n(q 1) = q
k
. Since [C[ = q
nk
, we see that this code
is perfect (by (20.5)).
Problem 20C. Let H be the ternary [4,2] Hamming code. Dene
a (nonlinear) ternary code C of length 9 with codewords
(x
0
, x
1
, . . . , x
4
; y
1
, . . . , y
4
)
by requiring that

4
i=0
x
i
,= 0 and that (y
1
x
1
, . . . , y
4
x
4
) is a
codeword in H. Show that C has covering radius 1. (No ternary
code of length 9 with covering radius 1 and fewer than 2 3
6
code-
words is known.)
Example 20.2. Consider the binary Hamming code C of length
n = 2
k
1. By denition, the dual code C

has as generator matrix

the k n matrix that has all possible nonzero vectors of length k
as columns. Therefore, the extension of C

is the code R
/
(1, k) of
Chapter 18. C

is usually called the simplex code of length n. We

remark that the [8,4] extended binary Hamming code is self-dual.
Problem 20D. Show that the covering radius of the code R(1, 2k)
is 2
2k1
2
k1
. (Hint: use the 1 representation instead of (0,1)
and work over to show that the covering radius is at most this
large. Consider the word z that is the characteristic function of
x F
2k
2
: x
1
x
2
+ +x
2k1
x
2k
= 1
to show equality.)
We mention one easily proved bound for arbitrary codes, known
as the Singleton bound.
Theorem 20.2. Let C be any code of length n and minimum
distance d over F
q
. Then
[C[ q
nd+1
.
Proof: From each codeword, we delete the last d 1 symbols.
The set of shortened words consists of words that are pairwise
dierent! There can be no more than q
nd+1
of them.
248 A Course in Combinatorics
Example 20.3. We give an example that shows a nice connection
with Chapter 19. Let q = 2
a
. Let S be a hyperoval in PG
2
(q).
Using the terminology of Example 19.7, the elements of S are vec-
tors in F
3
q
with the property that no three of them are linearly
dependent. We take the elements of S as the columns of a 3 by
q + 2 matrix H. If we interpret H as the parity check matrix of
a [q + 2, q 1] code over F
q
, then this code has minimum distance
at least 4. So by Theorem 20.2, the distance is equal to 4. This
is one of the rare examples of equality in Theorem 20.2. Codes for
which equality holds in the Singleton bound are called maximum
distance separable codes (MDS codes), another unfortunate name!
There are many open problems concerning these codes.
If C is a q-ary [n, k] code, and if A
i
denotes the number of code-
words in C with weight i, then
(20.9) A(z) :=
n

i=0
A
i
z
i
is called the weight enumerator of C. Of course, A
0
= 1 and A(1) =
[C[ = q
k
.
Problem 20E. Suppose C is a binary code, not necessarily linear,
with length 23, minimum distance 7, and [C[ = 2
12
. Assume 0 C.
First show that C is a perfect code. Let the weight enumerator of
C be given by (20.9). Count pairs (x, c) with c C, w(x) = 4, and
d(x, c) = 3, and show that A
7
= 253. Then show that the weight
enumerator of C is in fact completely determined by the fact that
C is a perfect code and 0 C.
The following theorem is one of the most useful in the theory of
error-correcting codes. It is due to F. J. MacWilliams (1963).
Theorem 20.3. Let C be an [n, k] code over F
q
with weight enu-
merator A(z) and let B(z) be the weight enumerator of C

. Then
(20.10) B(z) = q
k
(1 + (q 1)z)
n
A
_
1 z
1 + (q 1)z
_
.
Proof: We only give the proof for the case q = 2. For other
values of q, the proof is essentially the same (instead of (1)
u,v)
used below, one must use (u, v)), where is a character on F
q
).
20. Codes and designs 249
Dene
g(u) :=

vF
n
2
(1)
u,v)
z
w(v)
.
Then

uC
g(u) =

uC

vF
n
2
(1)
u,v)
z
w(v)
=

vF
n
2
z
w(v)

uC
(1)
u,v)
.
Here, if v C

, the inner sum is [C[. If v / C

, then half of the

terms in the inner sum have value +1, the other half 1. Therefore
(20.11)

uC
g(u) = [C[ B(z).
Now
g(u) =

(v
1
,v
2
,...,v
n
)F
n
2
n

i=1
((1)
u
i
v
i
z
v
i
)
=
n

i=1
(1 + (1)
u
i
z)
= (1 z)
w(u)
(1 +z)
nw(u)
.
The result follows by substituting this in (20.11).
Corollary. If we write B(z) =

n
j=0
B
j
z
j
, then for q = 2, we nd
from (20.10):
(20.12) B
j
= 2
k
n

i=0
A
i
j

l=0
(1)
l
_
i
l
__
n i
j l
_
.
These relations, known as the MacWilliams relations, are linearly
independent equations for the coecients A
i
, given the coecients
B
j
.
Many of the known nontrivial 5-designs were found by the follow-
ing elegant application of MacWilliams theorem, usually referred
to as the AssmusMattson theorem (1969). Again we restrict the
250 A Course in Combinatorics
proof to the binary case. For other q, the theorem has an obvious
generalization with nearly the same proof. We identify the posi-
tions of a code of length n with the set T := 1, 2, . . . , n. This
allows us to interpret a codeword in a binary code as a subset of
T (i.e. as the characteristic function of a subset). The support of a
codeword is the set of coordinate positions where the codeword is
not zero.
Problem 20F. Let C be a perfect binary e-error-correcting code
of length n. Assume 0 is a symbol and that 0 is a codeword. Show
that T together with the supports of codewords of weight d = 2e+1
is an S(e + 1, 2e + 1, n).
Theorem 20.4. Let A be a binary [n, k, d] code and let B := A

be the dual code, an [n, n k] code. Let t < d. Suppose the

number of nonzero weights in B, that are less than or equal to
nt, is dt. Then for each weight w, the supports of the words
of weight w in A form a t-design and the supports of the words of
weight w in B form a t-design.
Proof: If C is any code and a t-subset T of T has been xed, then
we denote by C
/
the code of length n t obtained by deleting the
coordinates in T from the codewords in C. We denote by C
0
the
subcode of C
/
obtained by deleting the coordinates in T from the
codewords in C that have zeros at all positions of T.
The proof is in four steps.
(i) Let T be a subset of T of size t. Since t is less than the
minimum distance of A, the code A
/
has as many codewords as A,
i.e. has dimension k also. In fact, A
/
still has minimum distance
d t. So the dual (A
/
)

has dimension n k t. Clearly, B

0
is
a subcode of (A
/
)

. Since the dimension of B

0
is at least n k t,
we must have B
0
= (A
/
)

.
(ii) Let

i
z
i
and

i
z
i
be the weight enumerators for A
/
and
B
0
, respectively. We claim that these weight enumerators do not
depend on the particular t-subset T, but only on the numbers t, n,
k, and the weights of words in B.
Let 0 <
1
<
2
< <
r
nt, where r dt, be the nonzero
weights nt for the code B. These are the only possible weights
20. Codes and designs 251
for B
0
. Then (20.12) gives
[B
0
[
j
=
_
n t
j
_
+
r

i=1

i
j

m=0
(1)
m
_

i
m
__
n t
i
j m
_
.
By hypothesis, the minimum distance of A
/
is r, so we know the
values of
j
for j < r, namely
0
= 1,
1
= =
r1
= 0. Thus
we have r linear equations in r unknowns

i
. These unknowns
are uniquely determined if the r r coecient matrix M that has
(i, j)-entry p
j
(
i
) where
p
j
(x) :=
j

m=0
(1)
m
_
x
m
__
n t x
j m
_
,
1 i r, 0 j r 1, is nonsingular. But p
j
(x) is a poly-
nomial in x of exact degree j (the coecient of x
j
is (1)
j
2
j
/j!),
so elementary column operations reduce M to the Vandermonde
matrix
_
_
_
_
1
1

2
1

r1
1
1
2

2
2

r1
2
.
.
.
.
.
.
.
.
.
.
.
.
1
r

2
r

r1
r
_
_
_
_
,
which is nonsingular since the
i
s are distinct.
Thus the weight enumerator

i
z
i
does not depend on the
choice of the subset T. Since A
/
is the dual of B
0
,

i
z
i
is also
independent of T.
(iii) Let c be the collection of words of weight w in B, interpreted
as subsets of T. The number of members of c that miss all coor-
dinates in T is the number of words of weight w in B
0
, and this is
independent of T. That is, the complement of (T, c) is a t-design.
By Problem 19C, the sets in c also form the blocks of a t-design.
This proves the second assertion of the theorem. (Remark: This
might be criticized if w > n t. But in this case our argument
when applied to t
/
:= nw shows that either every w-subset is the
support of a codeword, or no w-subset is the support of a codeword;
so the words of weight w, if any, form a trivial t-design.)
252 A Course in Combinatorics
(iv) To prove the rst assertion of the theorem, we proceed by
induction. We start with w = d. Let T be the collection of words
of weight d in A. The number of sets in T that contain a given
t-subset T of T is equal to the number of words of weight dt in A
/
,
and as we saw above, this number does not depend on the choice
of T. So T is a t-design. Let w > d and suppose the assertion is
true for all w
/
with w w
/
> d. Now let T denote the collection
of words of weight w in A. In this case the number of subsets in
T that contain a given t-subset T, is equal to the number of words
of weight w t in A
/
corresponding to codewords of weight w in
A. By (iii), the total number of words of weight w t in A
/
does
not depend on the choice of T. By the induction hypothesis and
(19.6), the number of words of weight wt in A
/
corresponding to
codewords of weight less than w in A does not depend either on T.
This proves the assertion.
Problem 20G. What is the weight enumerator of the dual of the
binary Hamming code of length 2
r
1? Derive an expression for
the weight enumerator of the Hamming code itself.
Example 20.4. Let A be the extended [8, 4] binary Hamming
code. We know that A = A

. In the notation of Theorem 20.4,

we have d = 4. Take t = 3. The conditions of Theorem 20.4 are
satised. Hence the words of weight 4 form a 3-design, which is of
course the Hadamard 3-(8,4,1) design corresponding to R(1, 3).
Example 20.5. After the next example, we shall treat the famous
binary Golay code and show that the corresponding extended code
G
24
is a self-dual [24, 12, 8] code with weights 0, 8, 12, 16 and 24. So
Theorem 20.4 shows that the words of weight 8 in this code form
a 5-design. From Problem 20E, we know that there are 759 blocks
and then (19.1) shows that = 1. This also follows from the fact
that the code has distance 8, so two blocks have at most four points
in common. This design is the Witt design S(5, 8, 24) mentioned
in Chapter 19. The words of other weights also yield 5-designs.
Example 20.6. The following 5-designs, and several others, were
found by V. Pless (1972). Consider the 18 by 18 Paley matrix C
given by (18.5). We consider a ternary [36, 18] code Sym
36
with
generator G = (I
18
C). Such a code is called a symmetry code.
20. Codes and designs 253
Since C is a Paley matrix, we have GG

= O, i.e. Sym
36
is a self-
dual code. This implies that all weights in the code are divisible
by 3. We claim that all the words in Sym
36
have weight at least
12. Observe that since C is symmetric, the matrix (C I
18
) is a
parity check matrix for the code, and, because the code is self-dual,
this means that that matrix is also a generator matrix for the code.
If (a, b), where a and b are vectors in F
18
3
, is a codeword, then
(b, a) is also a codeword. This shows that if there is a codeword
with weight less than 12, there is such a codeword that is a linear
combination of at most four rows of G. These are easily checked by
hand as follows; the reader should do it as an exercise. The fact that
C is a Paley matrix and the argument used in the proof of Theorem
18.1 (and in Problem 18B) show that a linear combination of 1, 2,
or 3 rows of G has weight 18, respectively 12, respectively 12 or 15.
It remains to check combinations of four rows. Now use the fact
that Q in (18.5) is cyclic, which implies that only a few essentially
dierent combinations have to be examined. One can also extend
the array used in the proof of Theorem 18.1 by one row. Both
methods involve very little work and produce the result that the
minimum weight is 12. (Originally this was done by computer.)
We now use the generalization of Theorem 20.4 to ternary codes.
If one considers the words of some xed weight in Sym
36
(the gen-
eralization holds for weights 12, 15, 18, and 21), and if we replace
each word by the set of positions where the nonzero coordinates
occur, we nd 5-designs. Since the codewords c and 2c yield the
same set, we only consider this set as one block.
We now come to the most famous of all binary codes: the binary
Golay code G
23
. There are very many constructions of this code,
some of them quite elegant and with short proofs of its properties.
We show only one of these constructions related to design theory.
We consider the incidence matrix N of the (unique) 2-(11,6,3)
design; see Problem 19D. We have NN

= 3I + 3J. Consider N
as a matrix with entries in F
2
. Then NN

= I +J. So N has rank

10 and the only vector x with xN = 0 is 1. The design properties
imply trivially that any row has weight 6 and that the sum of two
rows of N has weight 6. We also know that the sum of three or
four rows of N is not 0.
254 A Course in Combinatorics
Next, let G be the 12 by 24 matrix over F
2
given by G = (I
12
P),
where
(20.13) P =
_
_
_
_
0 1 1
1
.
.
. N
1
_
_
_
_
.
Every row of G has a weight 0 (mod 4). Any two rows of G have
inner product 0. This implies that the weight of any linear com-
bination of the rows of G is 0 (mod 4); prove this by induction.
The observations made about N then show that a linear combina-
tion of any number of rows of G has weight at least 8. Consider
the code generated by G and call it G
24
. Delete any coordinate to
nd a binary [23, 12] code with minimum distance at least 7. By
Problem 20A, this code must have minimum distance equal to 7
and, furthermore, it is a perfect code! We denote it by G
23
. These
two notations will be justied below, where we show that G
24
, the
extended binary Golay code, is unique, from which the uniqueness
of G
23
can be shown.
Theorem 20.5. If C is a binary code of length 24, with [C[ = 2
12
,
minimum distance 8, and if 0 C, then C is equivalent to G
24
.
Proof: (i) The dicult part of the proof is to show that C must
be a linear code. To see this, observe that deleting any coordinate
produces a code C
/
of length 23 and distance 7 with [C
/
[ = 2
12
.
By Problem 20E, the weight enumerator of this perfect code is de-
termined: A
0
= A
23
= 1, A
7
= A
16
= 253, A
8
= A
15
= 506,
A
11
= A
12
= 1288. From the fact that this is the case, no mat-
ter which of the 24 positions is deleted from C, it follows that
all codewords in C have weight 0, 8, 12, 16, or 24. Furthermore a
change of origin (i.e. adding a xed codeword to all the codewords)
shows that we can also infer that the distance of any two code-
words is 0, 8, 12, 16, or 24. Since all weights and all distances are
0 (mod 4), any two codewords have inner product 0. Therefore
the words of C span a code that is self-orthogonal. However, such
a code can have at most 2
12
words. Therefore, C itself must be a
linear and self-dual code.
20. Codes and designs 255
(ii) We form a generator matrix of C by taking as rst row any
word of weight 12. So after a permutation of positions, we have:
G =
_
1 . . . 1 0 . . . 0
A B
_
.
We know that any linear combination of the rows of B must have
even weight ,= 0, so B has rank 11. Therefore the code generated
by B is the [12, 11, 2] even-weight code. We conclude that we may
assume that B is the matrix I
11
, bordered by a column of 1s.
Another permutation of columns yields a generator matrix G
/
of
the form (I
12
P), where P has the same form as in (20.13). What
do we know about the matrix N? Clearly any row of N must have
weight 6. Furthermore, the sum of any two rows of N must have
weight at least 6. By Problem 19D, N is the incidence matrix of
the unique 2-(11,6,3) design, i.e. C is equivalent to G
24
.
As we saw in Example 20.5, the words of weight 8 in G
24
form the
blocks of the Witt design T = S(5, 8, 24). Denote by 0, 1, . . . , 23
the point set of T and consider I = 21, 22, 23. By the Corollary
to Theorem 19.3, T
I
is an S(2, 5, 21), i.e. a projective plane of order
4 (which is also known to be unique). The following problem shows
the beautiful combinatorial structure of the design T.
Problem 20H. Let B be a block of T with [BI[ = and dene
B

:= BI. We saw that if = 3, then B

is a line in PG
2
(4).
Show that
(i) = 2 implies that B

is a hyperoval in PG
2
(4);
(ii) = 0 implies that B

is the symmetric dierence of two lines.

(The ambitious reader may wish to show that if = 1, then
the seven points of B

and the lines of PG

2
(4) containing at least
two such points, form a Fano plane. Such a plane is called a Baer
subplane of PG
2
(4).)
By counting the hyperovals, pairs of lines, etc., one can show that
each of the geometric congurations mentioned above is one of the
sets B

. In fact, one of the well known constructions of S(5, 8, 24)

starts with these objects and produces the design by appending
suitable subsets of I. If one does not use the automorphism group
of PG
2
(4) in the argument, then this construction is a nontrivial
256 A Course in Combinatorics
combinatorial problem. (Again: the ambitious reader should try
it.)
Problem 20I. Let F
4
= 0, 1, , . Let C be the [6, 3] code over
F
4
with codewords (a, b, c, f(1), f(), f()), where f(x) := ax
2
+
bx +c.
(i) Show that C has minimum weight 4 and no words of weight
5.
(ii) Let G be the binary code with as codewords all 4 by 6 (0,1)-
matrices A with rows a
0
, a
1
, a

, a

such that:
(1) every column of A has the same parity as its rst row a
0
,
(2) a
1
+a

+a

C.
Show that G is a [24, 12, 8] code, i.e. G = G
24
.
Problem 20J. As in Example 20.6, we construct a ternary code
Sym
12
by using the Paley matrix C of order 6. Show that Sym
12
is a [12, 6, 6] self-dual code. Puncture the code (i.e. delete some
coordinate) to obtain a [11, 6, 5] ternary code G
11
. Show that this
code is perfect. It is the ternary Golay code.
We have now given several examples of designs constructed by
using a suitable code. We reverse the procedure and study codes
generated by the (characteristic functions of the) blocks of a design.
Let N be the incidence matrix of a projective plane of order n. We
consider the subspace C of F
v
2
, where v = n
2
+n +1, generated by
the rows of N. If n is odd, C is not very interesting. Namely, if we
take the sum of the rows of N that have a 1 in a xed position, the
result is a row with a 0 in that position and 1s elsewhere. These
vectors generate the [v, v 1, 2] even-weight code and this must be
C, since C obviously has no words of odd weight. If n is even, the
problem becomes more interesting. We restrict ourselves to n 2
(mod 4).
Theorem 20.6. If n 2 (mod 4), the rows of the incidence ma-
trix N of a projective plane of order n generate a binary code C
with dimension
1
2
(n
2
+n + 2).
Proof: (i) Since n is even, the code C is self-orthogonal because
every line has an odd number of points and any two lines meet in
one point. Therefore dimC
1
2
(n
2
+n + 2).
20. Codes and designs 257
(ii) Let dimC = r and let k := n
2
+ n + 1 r = dimC

. Let
H be a parity check matrix for C. Assume that the coordinate
places have been permuted in such a way that H has the form
(I
k
P). Dene A :=
_
I
k
P
O I
r
_
. Interpret the (0,1)-matrices N and A
as matrices over . Then
det NA

= det N = (n + 1)n
1
2
(n
2
+n)
.
Since all entries in the rst k columns of NA

are even integers,

det N is apparently divisible by 2
k
. So
1
2
(n
2
+ n) k, i.e. r
1
2
(n
2
+n + 2).
The result follows from (i) and (ii).
The theorem shows that the code C generated by the rows of N
has the property that C is self-dual. We shall now show an even
more interesting property of this code, namely that one can recover
the plane from the code.
Theorem 20.7. The code C of Theorem 20.6 has minimum weight
n+1 and every codeword of minimum weight corresponds to a line
in the plane of order n.
Proof: As before, we interpret a codeword as a subset of the
plane. So it will be clear what we mean by saying that a point is
on the codeword c. Let c be a codeword with w(c) = d. Since n is
even, the codewords corresponding to lines in the plane have a 1 as
parity check symbol in the extended code C. This code is self-dual
and this implies:
(1) if d is odd then c meets every line at least once;
(2) if d is even then every line through a xed point of c meets
c in a second point.
In case (2) we immediately see that d > n +1. In case (1) we nd:
(n + 1)d n
2
+ n + 1, i.e. d n + 1. If w(c) = n + 1, then there
is a line L of the plane that meets c in at least 3 points. If some
point of L is not a point of c, then every line ,= L through that
point must meet c by (1). This would imply that d n + 3. So c
must be the line L.
Recall that in a projective plane of even order n, a hyperoval is
a set of n + 2 points such that no three are on one line.
258 A Course in Combinatorics
Theorem 20.8. The codewords of weight n + 2 in the code C of
Theorem 20.6 are precisely all the hyperovals of the plane.
Proof: (i) Let v C and w(v) = n + 2. Every line meets v
in an even number of points. Let L be a line of the plane and
suppose that v and L have 2a points in common. Each of the n
lines ,= L through one of these 2a points meets v at least once
more. Therefore 2a +n n + 2, i.e. a = 0 or a = 1.
(ii) Let V be a hyperoval. Let S be the set of
_
n+2
2
_
secants of V .
Each point not in V is on
1
2
(n + 2) such lines; each point of V is
on n + 1 secants. Since n 2 (mod 4), the sum of the codewords
corresponding to secants is the characteristic function of V . So this
is a codeword.
Theorems like the previous two yielded enough information about
the code corresponding to a projective plane of order 10 to make
possible the computer search that we mentioned earlier. The rst
important step was due to F. J. MacWilliams, N. J. A. Sloane,
and J. G. Thompson (1973) who showed that this code could have
no words of weight 15. Presently, the only known projective plane
with order n 2 (mod 4) is the Fano plane.
Problem 20K. Find a 439 matrix of 0s, 1s, and 1s so that (i)
no column is all zero, and no column is 1 times another column,
and (ii) each row contains 13 of each of 0, 1, 1.
Generalize to construct r by
1
2
(3
r
1) 1 matrices with property
(i) and (ii) (with 13 replaced by an appropriate value).
(The matrix above can be used for the following puzzle. Of 39
coins, one is counterfeit and weighs either more or less than the
others. You have access to a balance with two pans; some coins
can be put in one pan and some in the other, and the balance will
tell you which subset has greater total weight. You must determine
the counterfeit coin, and whether it is lighter or heavier, using a
total of four weighings.)
Problem 20L. Korner and Simonyi called a code over an alphabet
of q > 2 symbols trierent if the codewords were not only dierent
but for each triple of codewords there is a position where they have
three dierent entries. Dene F(n, q) to be the maximal number
of codewords in a trierent code of length n. Prove that F(n, q)
20. Codes and designs 259
6
1/2
(q/3)
n/2
. (Hint: Use Theorem 3.6. Consider a random code
with N codewords. For each 3-subset S of 1, 2, . . . , n let A
S
be the
event that the three codewords with indices in S are not all dierent
or not trierent. Show that A
S
has probability (3q(q 1)+q)
n
/q
3n
,
etc.)
Notes.
We have not been concerned with the error-correcting properties
of codes in this chapter. For a systematic treatment of coding
theory we refer to Van Lint (1999). Much more material on the
relation between design theory and coding theory can be found in
Cameron and Van Lint (1991). The best reference book for coding
theory is MacWilliams and Sloane (1977).
For some of the history of the origin of coding theory, we refer to
Thompson (1983). There is some controversy about priority, but it
is clear that both R. W. Hamming and M. J. E. Golay contributed
to the discovery of this fascinating topic in 1947 and 1948. The
monumental paper by C. E. Shannon (1948) really started things
going. It seems that Hamming was irritated by the fact that his
computer kept stopping when it detected an error. He correctly
decided that if it could detect errors, it should be able to locate
them and correct them and then get on with the job! Golay pub-
lished the two Golay codes by giving their generator matrices but
without proof of the properties.
E. F. Assmus (19311998) died tragically at a meeting on Codes
and Designs in a place he loved: the Mathematisches Forschungsin-
stitut in Oberwolfach. A large part of this chapter is based on his
many contributions to the relations between coding theory and de-
sign theory (often written in collaboration with H. F. Mattson).
M. J. E. Golay (19021989) was a Swiss physicist who worked in
many dierent elds. He is known for his work on infrared spec-
troscopy and the invention of the capillary column but to mathe-
maticians mainly for his discovery of the two Golay codes.
For more about MDS codes, we refer to MacWilliams and Sloane
(1977).
F. J. MacWilliams (19171990) made many contributions to cod-
ing theory. Of these, the theorem known by her name is the most
260 A Course in Combinatorics
important. Her book with N. J. A. Sloane is the most important ref-
erence book on coding theory. As is the case with many important
coding theorists, she spent most of her career at Bell Laboratories.
As mentioned earlier, there are many constructions of the Golay
code G
24
. Each of them shows that some group of permutations
is contained in the automorphism group of the code. For example,
one of the most common constructions displays an automorphism
of order 23. Since the code is unique, its full automorphism group
must contain all these groups as subgroups. In this way, one can
prove that this automorphism group is the famous Mathieu group
M
24
. This group is of order 24 23 22 21 20 16 3 and acts
5-transitively on the 24 coordinate positions of the codewords.
It has been shown that there are no other perfect e-error-correct-
ing codes with e > 2 than the ones mentioned in this chapter. For
more on this subject, see Van Lint (1999). Also see Chapter 30.
References.
E. F. Assmus, Jr. and H. F. Mattson, Jr. (1969), New 5-designs, J.
Combinatorial Theory 6, 122151.
P. J. Cameron and J. H. van Lint (1991), Designs, Graphs, Codes
and their links, London Math. Soc. Student Texts 22, Cambridge
University Press.
J. H. van Lint (1999), Introduction to Coding Theory, Third edition,
Springer-Verlag.
F. J. MacWilliams (1963), A theorem on the distribution of weights
in a systematic code, Bell Syst. Tech. J. 42, 7994.
F. J. MacWilliams and N. J. A. Sloane (1977), The Theory of Error-
Correcting Codes, North-Holland.
F. J. MacWilliams, N. J. A. Sloane and J. G. Thompson (1973), On
the existence of a projective plane of order 10, J. Combinatorial
Theory (A) 14, 6678.
V. Pless (1972), Symmetry codes over GF(3) and new 5-designs,
J. Combinatorial Theory (A) 12, 119142.
C. E. Shannon (1948), A mathematical theory of communication,
Bell Syst. Tech. J. 27, 379423 and 623656.
T. M. Thompson (1983), From Error-Correcting Codes through
Sphere Packings to Simple Groups, Carus Math. Monograph 21.
21
Strongly regular graphs
and partial geometries
A strongly regular graph srg(v, k, , ) is a graph with v vertices
that is regular of degree k and that has the following properties:
(1) For any two adjacent vertices x, y, there are exactly ver-
tices adjacent to x and to y.
(2) For any two nonadjacent vertices x, y, there are exactly
vertices adjacent to x and to y.
A trivial example is a pentagon, an srg(5, 2, 0, 1). Perhaps the
most famous example is the graph of Fig. 1.4, the Petersen graph,
an srg(10, 3, 0, 1).
Clearly a graph that is the union of m complete graphs K
k
is an
srg(km, k1, k2, 0). Sometimes we shall exclude trivial examples
by requiring that a strongly regular graph and its complement are
connected, i.e. we assume
(21.1) 0 < < k < v 1.
(We remark that the fact that = 0 implies that the graph is a
union of complete graphs is most easily seen from (21.4) below.) It
is not dicult to see that the complement G of an srg(v, k, , ) G
is an
(21.2) srg(v, v k 1, v 2k + 2, v 2k +)
and since the parameters are nonnegative, we nd a simple condi-
tion on the parameters, namely
(21.3) v 2k + 2 0.
262 A Course in Combinatorics
Another relation between the parameters is easily found as follows.
Consider any vertex x and partition the other vertices into the set
(x) of vertices joined to x and the set (x) of vertices not joined
to x. By the denition of strongly regular graphs, (x) consists of
k vertices, each of which is joined to vertices of (x). Each vertex
in (x) is joined to vertices in (x). Counting edges with one
end in (x) and one end in (x) in two ways, we nd
(21.4) k(k 1) = (v k 1).
Problem 21A. Show that a strongly regular graph is extremal
in the following sense. Let G be a graph with v vertices, each of
degree at most k. Suppose that any two adjacent vertices, respec-
tively nonadjacent vertices, have at least , respectively , common
neighbors. Then
k(k 1 ) (v k 1)
and equality implies that G is strongly regular.
Before going into the fascinating theory of these graphs, we men-
tion several classes of examples.
Example 21.1. The triangular graph T(m), m 4, has as ver-
tices the 2-element subsets of a set of cardinality m; two distinct
vertices are adjacent if and only if they are not disjoint. T(m) is
an srg(
_
m
2
_
, 2(m 2), m 2, 4). The Petersen graph is T(5) (see
Problem 1A).
Example 21.2. The lattice graph L
2
(m), m 2, has as vertex set
S S, where S is a set of cardinality m; two distinct vertices are
adjacent if and only if they have a common coordinate. L
2
(m) is
an srg(m
2
, 2(m 1), m 2, 2). L
2
(2) is a quadrangle which is a
trivial example because its complement is not connected.
Example 21.3. Let q be a prime power with q 1 (mod 4). The
Paley graph P(q) has the elements of F
q
as vertices; two vertices
are adjacent if and only if their dierence is a nonzero square in
F
q
. That this is an srg(q,
1
2
(q 1),
1
4
(q 5),
1
4
(q 1)) is a direct
21. Strongly regular graphs and partial geometries 263
consequence of (18.4), but it is easier to show this using the matrix
Q of (18.5) as we shall see below. Note that P(5) is the pentagon.
Example 21.4. The Clebsch graph has as vertices all subsets of
even cardinality of the set 1, . . . , 5; two vertices are joined if and
only if their symmetric dierence has cardinality 4. This is an
srg(16, 5, 0, 2). One can also describe the vertices as the words of
even weight in F
5
2
, with an edge if the distance is 4. For any vertex
x, the induced subgraph on (x) is the Petersen graph. See Fig.
21.1 below.
Figure 21.1
Let us dene the adjacency matrix A of a graph G with v vertices
1, . . . , v to be the v v (0,1)-matrix with a
ij
= a
ji
= 1 if and only
if the vertices i and j are joined. Clearly A is symmetric with
zeros on the diagonal. The statement that G is an srg(v, k, , ) is
equivalent to
(21.5) AJ = kJ, A
2
+ ( )A+ ( k)I = J.
If Q is the matrix occurring in (18.5), then by (18.4) we have Q
2
=
qI J, and the adjacency matrix of the graph P(q) dened in
Example 21.3 is A =
1
2
(Q + J I). So A satises (21.5) with
k =
1
2
(q 1), =
1
4
(q 5), =
1
4
(q 1).
We have already seen one connection between strongly regular
graphs and topics treated in earlier chapters. Several more will
264 A Course in Combinatorics
follow. The theory of strongly regular graphs shows a completely
dierent aspect of graph theory from what we studied in Chapters
14 and 8. We now rely heavily on algebraic methods. Nevertheless,
we also wish to show some nice counting arguments that are used
in this theory.
We dene the BoseMesner algebra A of an srg G to be the 3-
dimensional algebra A of linear combinations of I, J, and A. That
this is indeed an algebra is a consequence of (21.5). This algebra
consists of symmetric commuting matrices and therefore there is
an orthogonal matrix that simultaneously diagonalizes them. This
can also be seen in an elementary way from (21.5). In fact, we shall
see in the next theorem that A has three distinct eigenspaces in 1
v
and each of them is an eigenspace for any element of A.
Theorem 21.1. If there is an srg(v, k, , ), then the numbers
f :=
1
2
_
v 1 +
(v 1)( ) 2k
_
( )
2
+ 4(k )
_
and
g :=
1
2
_
v 1
(v 1)( ) 2k
_
( )
2
+ 4(k )
_
are nonnegative integers.
Proof: Let A be the adjacency matrix of the graph. By (21.5)
the all-one vector j := (1, 1, . . . , 1)

is an eigenvector of A, with
eigenvalue k, and of course it is also an eigenvector of I and of
J. Application of (21.5) yields a second proof of (21.4). The mul-
tiplicity of this eigenvalue is one because the graph is connected.
Any other eigenvector, say with eigenvalue x, is orthogonal to j and
therefore we nd from (21.5),
x
2
+ ( )x + ( k) = 0.
This equation has two solutions
(21.6) r, s =
1
2
_

_
( )
2
+ 4(k )
_
21. Strongly regular graphs and partial geometries 265
Let f and g be the multiplicities of r and s as eigenvalues of A.
Then we have
1 +f +g = v and tr(A) = k +fr +gs = 0.
If we solve these two linear equations, we nd the assertion of the
theorem.
Note that the multiplicities can also be expressed as
(21.7) f =
k(s + 1)(k s)
(k +rs)(r s)
and g =
k(r + 1)(k r)
(k +rs)(r s)
.
From (21.6) we can draw a further (surprising) conclusion. If
f ,= g, then the square root in the denominator of the expressions
for f and for g must be an integer, i.e. ( )
2
+ 4(k ) is a
perfect square. It then follows from (21.6) that the eigenvalues r
and s are integers!
The other case, i.e. when f = g, is usually called the half-case.
We then have an srg(4 + 1, 2, 1, ). The Paley graphs are
examples of the half-case. In his paper on conference telephony,
mentioned in Chapter 18, Belevitch observed that a necessary con-
dition for the existence of a conference matrix of order n is that
n 1 is the sum of two squares. See Problem 19M. We note that
the parameters v = 21, k = 10, = 4, = 5 satisfy all the neces-
sary conditions for the existence of a strongly regular graph that we
stated above, but the graph does not exist because, using (18.5), it
would imply the existence of a conference matrix of order 22 and
since 21 is not the sum of two squares, this is impossible.
The condition of Theorem 21.1 is known as the integrality con-
dition. We shall call a parameter set (v, k, , ) that satises these
conditions and the earlier necessary conditions, a feasible set.
Problem 21B. Show that if an srg(k
2
+ 1, k, 0, 1) exists, then
k = 1, 2, 3, 7 or 57. (See the notes to Chapter 4.)
We have seen that the adjacency matrix of an srg(v, k, , ) has
three eigenvalues, one of which is k. There is a partial converse:
If G is a connected regular graph of degree k with an adjacency
matrix A with exactly three distinct eigenvalues, then G is strongly
regular. This is Problem 31F of Chapter 31.
266 A Course in Combinatorics
To obtain some more examples of strongly regular graphs, we
consider another connection with an earlier topic, namely designs.
This idea is due to J.-M. Goethals and J. J. Seidel (1970). A 2-
design is called quasisymmetric if the cardinality of the intersection
of two distinct blocks takes only two distinct values, say x > y. We
introduce a graph, called the block graph of the design; the vertices
are the blocks of the design and two vertices are adjacent if and
only if their intersection has cardinality y.
Theorem 21.2. The block graph of a quasisymmetric design is
strongly regular.
Proof: Let N be the v b incidence matrix of the design and
A the adjacency matrix of its block graph G. We have (using the
parameters v, k, b, r, of the 2-design):
NN

= (r )I +J,
N

N = kI +yA+x(J I A).
(The rst equation is (19.7), the second is the denition of A.)
We know that both NN

and N

N have all-one eigenvectors j (of

dierent lengths!) with eigenvalue kr. Also, we know that NN

has only the eigenvalue r on j

, with multiplicity v1. Therefore

N

N has this same eigenvalue, with the same multiplicity, and the
eigenvalue 0 with multiplicity b v. Since x ,= y, A is a linear
combination of I, J, and N

N. Therefore A has eigenvector j and

only two eigenvalues on the space j

. They are (r+k+x)/(yx)

with multiplicity v 1 and (x k)/(y x) with multiplicity b v.
By our observation above, G is strongly regular.
Now that we know that G is strongly regular, it is easy to calcu-
late the parameters. We do not give the distasteful formulae.
Example 21.5. Consider the S(5, 8, 24) of Chapter 20. We x
two points and consider the residual design with respect to these
two points. This is a 3-(22,6,1). From Problem 20F we know that
the 21 blocks of this design that contain a given point are the lines
of PG
2
(4) (i.e. after the point is removed) and the 56 blocks that
do not contain that point are hyperovals in this plane. They form a
2-(21,6,4) design. From the properties of S(5, 8, 24) it follows that
21. Strongly regular graphs and partial geometries 267
any two of these hyperovals meet in 0 or in 2 points. Therefore,
this derived design is a quasisymmetric design. From Theorem 21.2
it follows that the block graph of the design is a (56, k, , ). Here,
k = 10 because there are 45 = (4 1)
_
6
2
_
hyperovals that meet
a given hyperoval in 2 points. From the design properties or from
the conditions on strongly regular graphs, one then easily nds that
= 0, = 2. The graph constructed in this example is known as
the Gewirtz graph. Note that if A is the adjacency matrix of the
Gewirtz graph, then by (21.5) we have (I + A)
2
= 9I + 2J, and
that means that N := I +A is the incidence matrix of a 2-(56,11,2)
design (a so-called biplane). So in this case, we also nd a new
design from the strongly regular graph.
Problem 21C. Let T := (T, B, I) be the 3-(22,6,1) design men-
tioned above; it is unique. We form a graph with as vertex set
T B . The vertex is joined to every element of T. An
element of T is joined to an element of B if they are incident. Fi-
nally, two elements of B are joined if and only if the blocks are
disjoint. Show that this denes an srg(100, 22, 0, 6). This graph is
called the Higman-Sims graph.
If at this point we were to make a list of feasible parameter sets
for strongly regular graphs, then the list would contain many sets
that do not correspond to such a graph. We shall prove a few of the
theorems that exclude these sets. Again, the methods are algebraic.
As preparation, the reader should try to do the following problem
using purely combinatorial arguments.
Problem 21D. Show that an srg(28, 9, 0, 4) does not exist.
That this graph does not exist is a consequence of the next the-
orem, known as the Krein condition.
Theorem 21.3. Let G be a strongly regular graph with adjacency
matrix A, having eigenvalues k, r, and s. Then
(r + 1)(k +r + 2rs) (k +r)(s + 1)
2
,
and
(s + 1)(k +s + 2sr) (k +s)(r + 1)
2
.
268 A Course in Combinatorics
Proof: Let B := J I A. We know that the matrices of A have
three common eigenspaces of dimension 1, f, and g, respectively.
Call these spaces V
0
, V
1
, and V
2
. Here V
0
is the space spanned by
j, and V
1
and V
2
correspond to the eigenvalues r and s of A. For
i = 0, 1, 2 let E
i
be the matrix of the projection onto V
i
, i.e. E
i
has eigenvalue 1 on V
i
and 0 on the other two eigenspaces. These
matrices are what is called a basis of minimal idempotents of A.
We now consider the same set A, but as multiplication we take the
Hadamard product (see Problem 21E). It is obvious that any two
of I, A, and B have product O. Any (0,1)-matrix is idempotent for
Hadamard multiplication. So we conclude that A is closed under
Hadamard multiplication and that the matrices I, A, and B are a
basis of minimal idempotents.
Note that the denition of E
i
(i = 0, 1, 2) implies that
I = E
0
+E
1
+E
2
, A = kE
0
+rE
1
+sE
2
,
and
B = (v k 1)E
0
+ (r 1)E
1
+ (s 1)E
2
.
From this, we can express the E
i
in I, A, and B.
Consider the behavior of the matrices E
i
under the Hadamard
product. Since they form a basis of A, we have
E
i
E
j
=
2

k=0
q
k
ij
E
k
,
where q
k
ij
is the eigenvalue of E
i
E
j
on V
k
. It is a tedious cal-
culation, but the numbers q
k
ij
can be expressed in terms of the
parameters of G using the relations given above.
At this point we need the result of Problem 21E. The Hadamard
product E
i
E
j
is a principal submatrix of the Kronecker product
E
i
E
j
. This matrix is idempotent, hence its eigenvalues are 0 and
1. By the theorem alluded to in Problem 21E, the eigenvalues q
k
ij
have to be between 0 and 1. It turns out (after one has done all the
calculations) that all except two of the inequalities that one nds
in this way are satised. These two are q
1
11
0 and q
2
22
0. These
are the two equations of the assertion.
21. Strongly regular graphs and partial geometries 269
Problem 21E. Let A and B be two symmetric n n matrices
with eigenvalues
1
, . . . ,
n
, respectively
1
, . . . ,
n
. Determine the
eigenvalues of A B. We dene the Hadamard product A B of
A and B to be the matrix with entries a
ij
b
ij
. Show that A B is
a principal submatrix of AB. What can you conclude about the
eigenvalues of A B?
To appreciate the next theorem, the reader should rst convince
himself that the parameter set (50,21,4,12) is feasible and that it
satises the Krein conditions. What should one count to show
that the set nevertheless does not correspond to a strongly regular
graph?
Theorem 21.4. Let k, r, s be the eigenvalues of the adjacency
matrix A of an srg(v, k, , ) and let the multiplicities be 1, f and
g. Then
v
1
2
f(f + 3) and v
1
2
g(g + 3).
Proof: Let B be J I A and let the matrices E
i
(i = 0, 1, 2)
be as in the proof of the previous theorem. Let
E
1
= I +A+B.
Since E
1
is symmetric, there is an orthogonal matrix (H
1
K
1
) such
that
E
1
= (H
1
K
1
)
_
I O
O O
__
H

1
K

1
_
= H
1
H

1
.
Here H
1
is an v f matrix with H

1
H
1
= I. We consider the rows
of H
1
as v vectors in 1
f
. It follows that each of these vectors has
length
1
2
and any two distinct vectors from this set (call it S) have
inner product or . Such a set is called a spherical 2-distance set
because we can interpret S as a set of points on a sphere with only
two distinct (angular) distances. We must show that the cardinality
of S is at most
1
2
f(f + 3).
We normalize and obtain a set S
/
of v vectors on the unit sphere
in 1
f
with only two inner products, say b and c. For every v S
/
we dene a function f
v
: 1 by
f
v
(x) :=
(v, x) b)(v, x) c)
(1 b)(1 c)
.
270 A Course in Combinatorics
These functions are polynomials of degree 2 in the coordinates of
x. If v S, w S, v ,= w, then f
v
(v) = 1 and f
v
(w) = 0.
Therefore these functions are linearly independent. The spaces of
homogeneous linear and quadratic forms on have dimensions f
and
1
2
f(f + 1) respectively; since x
2
1
+ + x
2
f
= 1 on , we can
express constants in forms of degree 2 and 1. From the linear
independence of the functions f
v
, it follows that there can be at
most f +
1
2
f(f + 1) =
1
2
f(f + 3) of them.
This theorem is known as the absolute bound. It was shown by
A. Neumaier (1980) that the absolute bound can be improved to
v
1
2
f(f + 1) unless q
1
11
= 0
(and similarly for the other inequality).
Before turning to the relation between strongly regular graphs
and certain incidence structures, we show that nice counting argu-
ments also play a role in this area.
Theorem 21.5. Let G be a strongly regular graph with the same
parameters as T(n), i.e. G = srg(
_
n
2
_
, 2(n 2), n 2, 4). If n > 8,
then G is isomorphic to T(n).
Proof: Fix a vertex x and denote by the induced subgraph on
(x). This is a regular graph on 2(n 2) vertices, with degree
n 2. Let y and z be nonadjacent vertices of and let there be m
vertices in adjacent to both. Since = 4 and x is adjacent to y
and z, we have m 3. In the graph , there are n2 m vertices
adjacent to y but not to z, and the same number adjacent only to
z. Hence there are m2 vertices adjacent to neither of them. So
m 2. Suppose m = 3. Consider the unique vertex w adjacent
to neither y nor z. Every vertex adjacent to w in is adjacent to
y or to z, which implies that n 2 3 + 3 = 6, a contradiction.
Hence m = 2, and we also see that there are no triangles in the
complement .
We now show that is bipartite. On the contrary, assume there
is a circuit of odd length in this graph. Choose such a circuit
C = (x
0
, x
1
, . . . , x
k
= x
0
) with k minimal. From the argument
above we know that k ,= 3. In the graph , the vertices x
0
and x
1
21. Strongly regular graphs and partial geometries 271
are nonadjacent and are both adjacent to x
3
, x
4
, . . . , x
k2
. By the
argument above, k 6 and since k is odd, k = 5. The vertex x
0
has degree n 3 in , i.e. it is adjacent to n 5 vertices, besides
x
1
and x
4
. These must be nonadjacent in to both x
1
and x
4
.
A similar assertion can be made for x
2
, yielding n 5 vertices
outside C nonadjacent in to both x
1
and x
3
. There are exactly
n 4 vertices outside C nonadjacent to x
1
; hence at least n 6
vertices nonadjacent in to both x
3
and x
4
. The result of the rst
paragraph implies that n 6 1, a contradiction.
The result of the second paragraph means that contains two
disjoint cliques (sets of vertices, any two of which are adjacent) of
size n 2. Since x was arbitrary, we have shown that any vertex
in G lies in two cliques of size n 1 (cliques of this size will be
called grand cliques). The same argument shows that any edge
is in one of these grand cliques. The number of grand cliques is
2
_
n
2
_
/(n 1) = n. Since any two grand cliques have at most one
vertex in common, they must have exactly one vertex in common.
If we consider grand cliques as points and vertices as blocks, then
we have just shown that these points and blocks form a 2-(n, 2, 1)
design, i.e. the trivial design of all pairs from an n-set. G is the
block graph of this design, i.e. G is indeed isomorphic to T(n).
We remark that the theorem can also be proved, by case analysis,
for n < 8 and that for n = 8 there are three other graphs with the
parameters of T(8), known as the Chang graphs.
Problem 21F. Let G be a strongly regular graph with the same
parameters as L
2
(n), i.e. G = srg(n
2
, 2(n1), n2, 2). Prove that
if n > 4, then G is isomorphic to L
2
(n).
R. C. Bose (1963) studied large cliques in more general strongly
regular graphs. This led him to the concept of a partial geometry.
A partial geometry pg(K, R, T) is an incidence structure of points
and lines with the following properties:
(1) every line has K points and every point is on R lines;
(2) any two points are incident with at most one line;
(3) if point p is not on line L, then there are exactly T lines
through p that meet L.
272 A Course in Combinatorics
If two points x and y are on a line, we say that they are collinear
and write x y.
Problem 21G. Determine the number of points and lines of a
partial geometry pg(K, R, T).
By interchanging the roles of points and lines in a pg(K, R, T),
we nd the so-called dual partial geometry, a pg(R, K, T).
We introduce the point graph of a partial geometry as the graph
with the points of the geometry as vertices, and an edge x, y if
and only if x y.
Problem 21H. Show that the point graph of a pg(K, R, T) is a
(possibly trivial) srg(v, k, , ), with:
v = K
_
1 +
(K 1)(R 1)
T
_
, k = R(K 1),
= (K 2) + (R 1)(T 1), = RT,
r = K 1 T, s = R.
If a strongly regular graph has parameters such that it could be
the point graph of some partial geometry, then the graph is called
pseudo-geometric, and it is called geometric if it is indeed the point
graph of a partial geometry. The idea of introducing grand cliques
and then showing that the grand cliques and the points form a
design, as used above, was used by Bose to prove the following
theorem.
Theorem 21.6. If a strongly regular graph is pseudo-geometric,
corresponding to pg(K, R, T), and if 2K > R(R1)+T(R+1)(R
2

2R + 2), then the graph is geometric.

We do not give a proof here. The ideas of this proof were ex-
tended by A. Neumaier (1979), and, after an improvement by A.
E. Brouwer, his result obtained the following form (known as the
claw bound).
21. Strongly regular graphs and partial geometries 273
Theorem 21.7. If, for a strongly regular graph with the usual
parameters, ,= s
2
and ,= s(s+1), then 2(r+1) s(s+1)(+1).
The idea of the proof is to show that if r is large, then the graph
is the point graph of a partial geometry; this is done by counting
arguments. Then the absolute bound and Krein conditions are
applied to the point graph of the dual partial geometry. These
show the inequality stating that r cannot be too large.
The parameters (2058, 242, 91, 20) are feasible and satisfy all the
other necessary conditions stated in this chapter, except the claw
bound. So a graph with these parameters does not exist.
Problem 21I. Consider an srg(v, k, , 1). Show that the induced
subgraph on (x) is a union of cliques. Count the number of (+2)-
cliques in the graph. Thus show that both k/( +1) and vk/( +
1)( + 2) are integers. Apply to the set (209,16,3,1).
The partial geometries can be divided into four classes:
(1) A partial geometry with T = K is a 2-(v, K, 1) design.
(2) A partial geometry with T = R1 is called a net; dually, if
T = K 1 we speak of a transversal design.
(3) A partial geometry with T = 1 is called a generalized quad-
rangle; the notation GQ(K1, R1) is commonly used for
a pg(K, R, 1).
(4) If 1 < T < minK 1, R 1, then we call the partial
geometry proper.
Example 21.6. Consider the ane plane AG
2
(n). From Problem
19K we know that the lines can be divided into equivalence classes
of parallel lines, each containing n lines. Consider all the points of
the plane and take the lines of m parallel classes. Then it is clear
that these form a pg(n, m, m1), i.e. a net.
Example 21.7. Consider a Latin square of order n. Let the n
2
cells be the vertices of a graph; two vertices are joined by an edge
if and only if they are in the same row or in the same column,
or if they have the same entry. The graph is regular, with degree
3(n 1). If two vertices are joined, we may assume without loss
of generality that they are in the same row (think of the Latin
square as an OA(n, 3)). Then they have n 2 mutual neighbors.
274 A Course in Combinatorics
If two vertices are not joined, then they clearly have six common
neighbors. So this is an srg(n
2
, 3(n 1), n 2, 6). This is called a
Latin square graph and the notation is L
3
(n), in accordance with
the notation of Example 21.2. Generalizations will occur in a later
chapter. The graph L
3
(n) is geometric, corresponding to a partial
geometry pg(n, 3, 2), i.e. a net. Of course lines correspond to rows,
columns, and symbols of the square. We shall see in Chapter 22
that in fact nets and orthogonal arrays are equivalent concepts.
Example 21.8. A quadrangle satises the conditions for a partial
geometry. It is a pg(2, 2, 1) which accounts for the name generalized
quadrangle.
Consider a hyperoval O in PG
2
(4). We take as point set the 15
points of the plane that are not on O. Lines will be the secants of
O. Then every line has 3 points, and every point is on 3 lines. In
fact this is a pg(3, 3, 1), i.e. GQ(2, 2).
Note that the same construction applied to PG
2
(q) for q even
yields a pg(q 1,
1
2
(q + 2),
1
2
(q 2)).
Example 21.9. Consider PG
2
(q) where q is even. As in Exam-
ple 19.7, this is an incidence structure whose objects are the 1-
dimensional subspaces and 2-dimensional subspaces of F
3
q
. Again,
let O be a hyperoval in the projective plane. Let all 1-dimensional
subspaces that are in O and their cosets in F
3
q
be lines; points are
the points of the vector space. Every line has q points; every point
is on q +2 lines. The fact that O is a hyperoval implies that for any
point p not on line L, there is a unique line through p that meets
L. So we have dened a GQ(q 1, q + 1).
Example 21.10. Consider the subgroup G of Z
6
3
generated by
the element (1,1,1,1,1,1). For each coset a + G, the sum of the
coordinates of the points is a constant i. We say that the coset is
of type i. Let /
i
be the set of cosets of G of type i. We dene a
tripartite graph by joining the coset a+G to the coset a+b+G for
each b that has only one nonzero coordinate. Clearly, any element
of /
i
has six neighbors in /
i+1
and six in /
i+2
. We construct
a partial geometry by taking some /
i
as point set and one of the
other two classes /
j
as line set. Incidence corresponds to adjacency.
That K = R = 6 is clear. It is an easy exercise to show that T = 2.
21. Strongly regular graphs and partial geometries 275
This denes a pg(6, 6, 2).
We now give a nice application of Theorem 21.5. We shall prove
the theorem, mentioned in Chapter 19 without proof, which states
that a quasiresidual design with = 2 is a residual design. Since
the cases k 6 require separate treatment, we restrict ourselves to
blocksize > 6.
Theorem 21.8. Let T be a 2-(v, k, 2) design with k > 6 and
v =
1
2
k(k + 1) (so T is quasiresidual). Then T is a residual of a
symmetric 2-design.
Proof: Let B be a block and let a
i
denote the number of blocks
(,= B) that meet B in i points. Just as in Theorem 19.9 we nd

a
i
=
1
2
k(k +3),

ia
i
= k(k +1),

i(i 1)a
i
= k(k 1).
This implies that

(i 1)(i 2)a
i
= 0. Therefore any two distinct
blocks meet in 1 or in 2 points, i.e. the design T is quasisymmetric.
From Theorem 21.2 we nd that the block graph G of T is strongly
regular. Calculation of the parameters shows that G has the same
parameters as T(k + 2). So by Theorem 21.5, G is isomorphic to
T(k + 2). This means that we can label the blocks of T with 2-
subsets of S := 1, 2, . . . , k +2 in such a way that two blocks meet
in i points whenever their labels meet in 2 i points (i = 1, 2). We
adjoin the set S to the point set of T, and we adjoin to each block
its label, a 2-subset of S. Finally, we consider S as a new block.
This produces the required symmetric design with = 2 that has
T as residual with respect to the block S.
We list below a table of feasible parameter sets with v < 30 and
what we have learned about them in this chapter. The only one
left as a challenge for the reader is no. 14, i.e. GQ(2, 4) for which
no construction was given. The corresponding graph is known as
the Schlaei graph.
276 A Course in Combinatorics
No. v k Example
1 5 2 0 1 P(5)
2 9 4 1 2 L
2
(3)
3 10 3 0 1 Petersen, T(5)
4 13 6 2 3 P(13)
5 15 6 1 3 GQ(2, 2)
6 16 5 0 2 Clebsch
7 16 6 2 2 L
2
(4)
8 17 8 3 4 P(17)
9 21 10 3 6 T(7)
10 21 10 4 5 does not exist, conference
11 25 8 3 2 L
2
(5)
12 25 12 5 6 L
3
(5)
13 26 10 3 4 STS(13), Theorem 21.2
14 27 10 1 5 GQ(2, 4)
15 28 9 0 4 does not exist, Thms. 21.3, 21.4
16 28 12 6 4 T(8)
17 29 14 6 7 P(29)
We give some remarks on directed strongly regular graphs and a
few problems.
The condition on paths of length 2 in the denition of strongly
regular graphs is best seen by writing (21.5) as follows
(21.5) AJ = kJ, A
2
= kI +A+(J I A).
The denition was generalized to directed graphs by A. Duval (1988).
We consider a directed graph G without loops or multiple edges.
If there is an edge from a to b, we write a b. We allow edges
in both directions, indicated by a b. In that case we call (a, b)
an undirected edge. The generalization of (21.5) will be the same
with respect to paths of length 2, but we now require that each
vertex has indegree k, outdegree k and that among these edges
there are t undirected edges (producing paths of length 2 from a
vertex to itself). So, G is called a directed stronly regular graph if
its adjacency matrix A satises
(21.8) AJ = JA = kJ, A
2
= tI +A+(J I A).
21. Strongly regular graphs and partial geometries 277
For such a graph on v vertices, the notation dsrg(v; k, t; , ) will
be used.
Example 21.11. Let G be a graph on six vertices consisting of a
directed triangle 1 2 3 1, a directed triangle 1
/
3
/

2
/
1
/
, and undirected edges (i, i
/
), i = 1, 2, 3. If C is the 3 3
circulant matrix with c
ij
= 1 for (i, j) = (1, 2), (2, 3), (3, 1) and 0
otherwise, then the adjacency matrix A of G is
A =
_
C I
I C
2
_
.
One easily checks that A satises (21.8) with k = 2, t = 1, = 0,
and = 1. So G is a dsrg(6; 2, 1; 0, 1).
Problem 21J. State and prove an analog of Theorem 21.1 for
directed strongly regular graphs.
Problem 21K. Exclude strongly regular graphs and graphs with
adjacency matrix J A. Show that the eigenvalues calculated in
Problem 21J lead to an analog of the half-case if ()
2
+4(t
) is not the square of an integer. Prove that in this case, the
adjacency matrix A of the dsrg is of the type of Q in (18.5).
Problem 21L. Consider the case = 1, = 0, t = k 1. Prove
that only three values of v are possible. Show that in the second
case, an example can be obtained from K
3,3
by replacing vertices
by directed triangles, and edges by copies of the graph of Example
21.11 in a suitable way.
We now give some comments and problems on neighborhood reg-
ular graphs.
In a strongly regular graph G the neighborhoods (x) and (x)
of x in G, respectively in the complement G, are regular graphs. C.
D. Godsil and B. D. McKay (1979) called any graph that has this
property a neighborhood regular graph. It is not dicult to show
that a neighborhood regular graph that is also regular is in fact a
strongly regular graph. If G or G is not connected, trivial situations
arise. In the following problems, we assume that G is neighborhood
regular, not regular, and that G and G are connected.
278 A Course in Combinatorics
Problem 21M.
(i) Show that the degree of (x) does not depend on x. Call this
number a.
(ii) Show that there is a number a such that for each x, the degree
of (x) is a.
Problem 21N. If X and Y are subsets of the vertex set of G, we
denote by [XY [ the number of edges (x, y) in G with x X and
y Y . Let x
1
and x
2
be two non-adjacent edges in G with degree
k
1
, respectively k
2
, k
1
,= k
2
. Let d
i
:= deg (x
i
), i = 1, 2. Note
that d
1
,= d
2
. Now consider the following subsets of the vertex set
of G : A := (x
1
) (x
2
), B := (x
2
) (x
1
), C := (x
1
) (x
2
),
and D := (x
1
) (x
2
.
(i) Find relations between the numbers [XY [ and [X[, where X
and Y are among B, C, D.
(ii) Deduce that
[DD[ = (a +a + 1)d
1
d
2
1
a[C[ +[CC[.
(iii) Prove that d
1
+d
2
= a +a + 1.
(iv) Show that all the vertices in G have degree k
1
or k
2
.
Example 21.12. Let G be the octagon on vertices 1 to 8 with the
two diagonals (15) and (26). This is a neighborhood regular graph
with k
1
= 2, k
2
= 3, a = 0, a = 2, d
1
= 1, and d
2
= 2.
Problem 21O. Given a Steiner triple systemo on n points, dene
a graph G whose vertices are the triples / of the system and where
A, B / are adjacent when [A B[ = 1.
(i) Show by elementary means (i.e. do not use Theorem 21.2
or matrix methods) that G is strongly regular and calculate its
parameters.
(ii) Show that if C is a clique in G and [C[ > 7, then C is a
subset of the triples that contain a xed point x for some point x
of o.
(iii) Use this to show that if the graphs G
1
and G
2
arising from
two Steiner triple systems on n > 15 points are isomorphic, then
the two systems are also isomorphic. (We remark that there are
more than 163929929318400 nonisomorphic Steiner triple systems
21. Strongly regular graphs and partial geometries 279
on 25 pointssee R. M. Wilson (1973/74)so we have at least
that many nonisomorphic strongly regular graphs on 100 vertices.)
Problem 21P. Show that equality in Problem 4H cannot occur
unless n = 3.
Problem 21Q. Prove the so-called Friendship Theorem: At a
party with n people (n > 3) every two persons have exactly one
mutual friend. Then there is a unique person at the party who is
a friend of all the others. Use problem 1J.
Notes.
Strongly regular graphs and partial geometries were introduced
by R. C. Bose in 1963. A much more general concept, namely
that of an association scheme, had been introduced by Bose and
Shimamoto in 1952; see Chapter 30 here and Cameron and Van
Lint (1991), Ch. 17.
For a survey of construction methods for strongly regular graphs,
see Hubaut (1975) and Brouwer and Van Lint (1982).
The Clebsch graph may be dened by the 16 lines in the Clebsch
quartic surface, a pair of lines being adjacent if and only if they
are skew. Alfred Clebsch (18331872) was a mathematician and
physicist who worked in Karlsruhe, Giessen, and G ottingen. He was
one of the founders of the famous journal Mathematische Annalen.
The algebra A was introduced by R. C. Bose and D. M. Mesner
in 1959.
The reference Goethals and Seidel (1970) contains, besides The-
orem 21.2, many interesting connections between strongly regu-
lar graphs and other combinatorial designs. Hadamard matrices,
Steiner systems and the Golay code all contribute to the theory.
For these connections, again see Cameron and Van Lint (1991).
The Higman-Sims graph is connected to their famous nite sim-
ple group (see Higman and Sims (1968)).
Theorem 21.3 became known as the Krein condition because a
special case of this theorem was proved by L. L. Scott by applying
a result of M. G. Krein concerning topological groups to a problem
on nite groups. The simple proof given in this chapter is due to D.
G. Higman (1975) and P. Delsarte (1973). The case of equality in
280 A Course in Combinatorics
the Krein conditions was treated by Delsarte, Goethals and Seidel
(1977) and Cameron, Goethals and Seidel (1978).
The idea of the proof of Theorem 21.4 is due to T. H. Koorn-
winder (1976).
Theorem 21.5 was proved several times: see for example, L. C.
Chang (1959), W. S. Connor (1958), A. J. Homan (1960). The
three exceptional graphs for n = 8 are known as the Chang graphs.
For surveys on partial geometries, see Van Lint (1983), De Clerck
and Van Maldeghem (1995), and De Clerck (2000).
Nets were introduced by R. H. Bruck in 1951. Boses result on
pseudo-geometric graphs was inspired by Brucks work (e.g. the
idea of grand cliques).
For all that one would like to know about generalized quadrangles
(and more) we refer to the book Finite Generalized Quadrangles by
S. E. Payne and J. A. Thas (1984).
Only three proper partial geometries with T = 2 are known. The
geometry pg(6, 6, 2) of Example 21.9 was rst constructed by Van
Lint and Schrijver (1981); the construction that we described is due
to Cameron and Van Lint (1982). The other known examples are
a pg(5, 18, 2) due to Haemers and a pg(8, 20, 2) found by Mathon
(with the aid of a computer).
A big step in the direction of the proof of the nonexistence of
a projective plane of order 10, mentioned in Chapter 19, was the
proof of the nonexistence of pg(6, 9, 4). This implied (see Example
21.8) that, if the plane existed, it would not have any hyperovals.
References.
R. C. Bose (1963), Strongly regular graphs, partial geometries, and
partially balanced designs, Pacic J. Math. 13, 389419.
R. C. Bose and D. M. Mesner (1959), On linear associative alge-
bras corresponding to association schemes of partially balanced
designs, Ann. Math. Stat. 30, 2138.
A. E. Brouwer and J. H. van Lint (1982), Strongly regular graphs
and partial geometries, in: Enumeration and Design (D. M. Jack-
son and S. A. Vanstone, eds.), Academic Press.
R. H. Bruck (1951), Finite nets I, Canad. J. Math. 3, 94107.
R. H. Bruck (1963), Finite nets II, Pacic J. Math. 13, 421457.
21. Strongly regular graphs and partial geometries 281
P. J. Cameron (1978), Strongly regular graphs, in: Selected Top-
ics in Graph Theory (L. W. Beineke and R. J. Wilson, eds.),
Academic Press.
P. J. Cameron, J.-M. Goethals, and J. J. Seidel (1978), The Krein
condition, spherical designs, Norton algebras and permutation
groups, Proc. Kon. Ned. Akad. v. Wetensch. 81, 196206.
P. J. Cameron and J. H. van Lint (1991), Designs, Graphs, Codes
and their links, London Math. Soc. Student Texts 22, Cambridge
University Press.
P. J. Cameron and J. H. van Lint (1982), On the partial geometry
pg(6, 6, 2), J. Combinatorial Theory (A) 32, 252255.
L. C. Chang (1959), The uniqueness and nonuniqueness of trian-
gular association schemes, Sci. Record Peking Math. 3, 604613.
F. De Clerck and H. Van Maldeghem (1995), Some classes of rank
2 geometries. In F. Buekenhout (editor) Handbook of Incidence
geometry, Buildings and Foundations, North-Holland.
F. De Clerck (2000), Partial and semipartial geometries: an update,
Combinatorics 2000 (Gaeta, Italy).
W. S. Connor (1958), The uniqueness of the triangular association
scheme, Ann. Math. Stat. 29, 262266.
P. Delsarte (1973), An algebraic approach to the association
schemes of coding theory, Philips Res. Repts. Suppl. 10.
P. Delsarte, J.-M. Goethals and J. J. Seidel (1977), Spherical codes
and designs, Geometriae Dedicata 6, 363388.
A. M. Duval (1988), A Directed Graph Version of Strongly Regular
Graphs, J. Combinatorial Theory (A)47, 71100.
A. Gewirtz (1969), The uniqueness of g(2, 2, 10, 56), Trans. New
York Acad. Sci. 31, 656675.
C. D. Godsil and B. D. McKay (1979), Graphs with Regular Neigh-
borhoods, Combinatorial Mathematics VII, 127140, Springer.
J.-M. Goethals and J. J. Seidel (1970), Strongly regular graphs
derived from combinatorial designs, Canad. J. Math. 22, 597
614.
D. G. Higman (1975), Invariant relations, coherent congurations
and generalized polygons, in: Combinatorics (M. Hall, Jr. and J.
H. van Lint, eds.), D. Reidel.
D. G. Higman and C. C. Sims (1968), A simple group of order
282 A Course in Combinatorics
44,352,000, Math. Z. 105, 110113.
A. J. Homan (1960), On the uniqueness of the triangular associ-
ation scheme, Ann. Math. Stat. 31, 492497.
X. Hubaut (1975), Strongly regular graphs, Discrete Math. 13, 357
381.
T. H. Koornwinder (1976), A note on the absolute bound for sys-
tems of lines, Proc. Kon. Ned. Akad. v. Wetensch. 79, 152153.
J. H. van Lint (1983), Partial geometries, Proc. Int. Congress of
Math., Warsaw.
J. H. van Lint and A. Schrijver (1981), Construction of strongly
regular graphs, two-weight codes and partial geometries by nite
elds, Combinatorica 1, 6373.
J. H. van Lint and J. J. Seidel (1969), Equilateral point sets in
elliptic geometry, Proc. Kon. Ned. Akad. v. Wetensch. 69, 335
348.
A. Neumaier (1979), Strongly regular graphs with smallest eigen-
value m, Archiv der Mathematik 33, 392400.
A. Neumaier (1980), New inequalities for the parameters of an as-
sociation scheme, in: Combinatorics and Graph Theory, Lecture
Notes in Math. 885, Springer-Verlag.
S. E. Payne and J. A. Thas (1984), Finite Generalized Quadrangles,
Pitman.
R. M. Wilson (1973/74), Nonisomorphic Steiner triple systems,
Math. Z. 135, 303313.
22
Orthogonal Latin squares
Two Latin squares L
1
: R C S and L
2
: R C T (with the
same row and column sets) are said to be orthogonal when for each
ordered pair (s, t) S T, there is a unique cell (x, y) RC so
that
L
1
(x, y) = s and L
2
(x, y) = t.
The use of the word orthogonal is perhaps unfortunate since it has
other meanings in mathematics, but it has become far too common-
place to try to change now.
For example, let
A :=
_

_
A K Q J
Q J A K
J Q K A
K A J Q
_

_
, B :=
_

_




_

_
Figure 22.1
where R = C = 1, 2, 3, 4, S = A, K, Q, J, T = , , , .
The orthogonality of A and B is evident when the two squares are
superposed and we see that each element of S T appears exactly
once.
_

_
A K Q J
Q J A K
J Q K A
K A J Q
_

_
Orthogonality does not depend on the particular symbol set used.
We could replace the playing card suits in our example by any
four symbols in any order, and the resulting square would still be
284 A Course in Combinatorics
orthogonal to A. We have used Latin letters for the symbols of A;
if we had used Greek letters for B, it would have been clear why
the superposition of two orthogonal Latin squares is often called a
Grco-Latin square.
The example we have given with playing cards is similar to what
is known as Eulers 36 ocers problem. According to folklore,
Euler was asked by Catherine the Great (at whose court he was in
residence) to arrange 36 ocers from six dierent regiments and
of six dierent ranks (one ocer of each rank from each regiment)
in a 6 by 6 array so that each row and each column contained one
ocer of each rank and one ocer of each regiment. A solution
requires a pair of orthogonal Latin squares of order 6.
Surprisingly, there is no solution. Whether he was asked by
Catherine or not, Euler did consider the problem in 1779 and con-
vinced himself that it was impossible. Euler was capable of monu-
mental calculation, but it is not clear that he had really examined
all cases. This was systematically done in 1900 by G. Tarry. Today
a computer can do this easily. A short proof (still involving several
cases to be analyzed) was given by D. R. Stinson (1984).
Euler knew that a pair of orthogonal Latin squares of order n
existed for all odd values of n, and all n 0 (mod 4); see Theorem
22.3 below. For examples of orthogonal Latin squares of odd orders
n, let G be any group of order n and dene squares L
1
and L
2
with
row, column, and symbol sets G by
L
1
(x, y) := xy, L
2
(x, y) := x
1
y.
For completely trivial reasons, there is no pair of orthogonal
Latin squares of order 2. Euler asserted that he was of the opin-
ion that this impossibility for n = 2, 6 extended as well to n =
10, 14, 18, . . . , i.e. to orders n 2 (mod 4). This statement was
known as Eulers conjecture for 177 years until it was suddenly
and completely disproved by Bose, Parker, and Shrikhande. We will
prove later their theorem that pairs of orthogonal Latin squares of
orders n exist for all n except n = 2, 6.
At various times, a more general question has been raised: can
one nd sets of many Latin squares, any two of which are orthogo-
nal? We will denote by N(n) the largest integer k for which there
22. Orthogonal Latin squares 285
exist k Latin squares of order n that are pairwise orthogonal. For
example, we may add
C :=
_

_
0
0
0
0
_

_
(the addition table of the Klein four-group) to the squares in Fig.
22.1 to obtain three pairwise orthogonal Latin squares of order 4;
thus N(4) 3. We remark that the denition of orthogonality is
trivial or vacuous for n = 1 or 0, and so N(1) = N(0) = .
Here is a construction that proves N(q) q1 for a prime power
q. All row, column, and symbol sets are to be the elements of the
eld F
q
. For each nonzero element a in F
q
, dene L
a
(x, y) := ax+y;
these q 1 squares are pairwise orthogonal. As we all know, the
systems
ax +by = s
cx +dy = t
have unique solutions (x, y) when ad bc ,= 0, so it is easy to tell
when two squares dened by linear equations are orthogonal. Here
are the squares for q = 5:
_

_
0 1 2 3 4
1 2 3 4 0
2 3 4 0 1
3 4 0 1 2
4 0 1 2 3
_

_
,
_

_
0 1 2 3 4
2 3 4 0 1
4 0 1 2 3
1 2 3 4 0
3 4 0 1 2
_

_
,
_

_
0 1 2 3 4
3 4 0 1 2
1 2 3 4 0
4 0 1 2 3
2 3 4 0 1
_

_
,
_

_
0 1 2 3 4
4 0 1 2 3
3 4 0 1 2
2 3 4 0 1
1 2 3 4 0
_

_
.
Theorem 22.1. For n 2, we have 1 N(n) n 1.
Proof: We may change all row, column, and symbol sets in a set
of k pairwise orthogonal Latin squares L
i

k
i=1
to 1, 2, . . . , n (for
286 A Course in Combinatorics
notational convenience), and we might as well rename the symbols
if necessary so that the rst row of each square is 1, 2, . . . , n in that
order. Now we consider the elements
L
1
(2, 1), L
2
(2, 1), . . . , L
k
(2, 1)
in the second row, rst column. None of these is 1, since 1 occurs
already in the rst column. They are distinct because if L
i
(2, 1) =
L
j
(2, 1) = s, say, then the fact that also L
i
(1, s) = L
j
(1, s) = s
would contradict the orthogonality of L
i
and L
j
. The upper bound
follows.
In view of our construction from nite elds and the simple the-
orem above, we have
(22.1) N(q) = q 1 if q is a prime power.
At this point, we want to show that the concept of pairwise or-
thogonal Latin squares is equivalent to something the reader has
already seen in the previous chapter! The assertion of the theo-
rem will be more immediately apparent if we rst notice that the
property of being a Latin square can itself be described in terms of
orthogonality. For example, a square of order 5 is Latin if and only
if it is orthogonal to the two squares
_

_
0 0 0 0 0
1 1 1 1 1
2 2 2 2 2
3 3 3 3 3
4 4 4 4 4
_

_
and
_

_
0 1 2 3 4
0 1 2 3 4
0 1 2 3 4
0 1 2 3 4
0 1 2 3 4
_

_
.
Theorem 22.2. A set of k pairwise orthogonal Latin squares of
order n exists if and only if an (n, k + 2)-net exists.
Proof: Suppose L
i
: R C S
i
, 1 i k, are pairwise orthog-
onal Latin squares. Let T = R C, the set of n
2
cells. Roughly
speaking, we take as lines the rows, the columns, and the iso-
symbol lines in each square. More formally, let
/
1
= (x, b) : b C : x R,
22. Orthogonal Latin squares 287
/
2
= (a, y) : a R : y C,
/
i+2
= (x, y) : L
i
(x, y) = c : c S
i
, 1 i k,
and let B =
k+2
i=1
/
i
. That (T, B) is an (n, k + 2)-net follows from
the denitions of Latin and orthogonal.
Given an (n, k+2)-net (T, B), where B =
k+2
i=1
/
i
is the partition
into parallel classes, dene L
i
: /
1
/
2
/
i+2
by declaring that
L
i
(A, B) is to be the unique line in /
i+2
containing the point of
intersection of A and B. As usual we leave to the reader the details
of checking the Latin and orthogonal properties of these squares.

Corollary. N(n) = n 1 if and only if there exists a projective

(or ane) plane of order n.
The value of N(n) tells us, in some very imprecise sense, how
close to the existence of a projective plane of order n we may get.
But we know terribly little about this function. The only value of
n other than prime powers for which N(n) is known is n = 6.
Problem 22A. (i) Consider the complement of the graph of an
(n, n)-net and prove that N(n) n2 implies that N(n) = n1.
(ii) Use the result of Problem 21F to prove that if n > 4, then
N(n) n 3 implies that N(n) = n 1.
Theorem 22.3.
(i) N(nm) minN(n), N(m).
(ii) If n = p
e
1
1
p
e
2
2
p
e
r
r
is the factorization of n into powers of
primes, then
N(n) min
1ir
(p
e
i
i
1).
Proof: Part (ii) follows from part (i) and (22.1), used inductively.
To prove part (i), we rst dene a Kronecker-like product for
squares in the natural way: Given L
i
: R
i
C
i
S
i
, i = 1, 2,
dene
L
1
L
2
: (R
1
R
2
) (C
1
C
2
) (S
1
S
2
)
by
(L
1
L
2
)((x
1
, x
2
), (y
1
, y
2
)) := (L
1
(x
1
, y
1
), L
2
(x
2
, y
2
)).
288 A Course in Combinatorics
It is straightforward to verify that if A
i

k
i=1
are pairwise orthogonal
Latin squares of order n and B
i

k
i=1
are pairwise orthogonal Latin
squares of order m, then A
i
B
i

k
i=1
are pairwise orthogonal Latin
squares of order nm. We leave this task to the reader, but advise
you just to read on and not bother.
Theorem 22.3 is known as MacNeishs theorem. MacNeish con-
jectured in 1922 that equality held in Theorem 22.3(ii). This would
imply Eulers conjecture. But combinatorial problems of this type
rarely have such simple answers.
The MacNeish conjecture was disproved rst for n = 21 when
three Latin squares of order 21 were constructed with the aid of
the 21 point projective plane of order 4; see Example 22.1 below.
This was an example of composition methods, which also provided
the rst counterexample to Eulers conjecture (n = 22). The rst
pair of orthogonal Latin squares of order 10 was found by Parker by
what might be called dierence methods. We discuss composition
methods rst; they have proved more powerful for large values of
n.
Bose, Parker and Shrikhande gave many constructions for sets
of orthogonal Latin squares. We choose to describe two basic and
elegant ones in terms of quasigroups. Recall that a quasigroup is a
Latin square whose row, column, and symbol sets are the same set
X. A quasigroup L is idempotent when L(x, x) = x for all x X.
For example, if X is the nite eld F
q
of order q, then
L
a
(x, y) := ax + (1 a)y
denes an idempotent quasigroup L
a
if a ,= 0, 1. Any two squares
of this form are orthogonal.
Problem 22B. For a prime power q 4, construct two Latin
squares A, S of order q with row and column sets equal to F
q
so
that A is orthogonal to its transpose, and to S, which is to be
symmetric. In addition, ensure that S is idempotent if q is odd
and unipotent, i.e. has constant diagonal, if q is even.
Suppose we have a linear space, cf. Chapter 19, on a point set X
with line set /. Further suppose that for each line A in /, we have k
pairwise orthogonal idempotent quasigroups L
A
1
, L
A
2
, . . . , L
A
k
on the
22. Orthogonal Latin squares 289
set A. Then we can construct k pairwise orthogonal idempotent
quasigroups L
1
, L
2
, . . . , L
k
on the entire set X by declaring, for
each i = 1, 2, . . . , k, that L
i
(x, x) := x for x X, and for distinct
x, y X, L
i
(x, y) := L
A
i
(x, y), where A is the unique line in /
which contains both x and y. It is a simple matter to check that
L
1
, L
2
, . . . , L
k
are Latin squares and that they are orthogonal; for
example, given i, j, s and t, s ,= t, a cell (x, y) for which
L
i
(x, y) = s and L
j
(x, y) = t
exists, and x, y may be found in that line B which contains s and
t, because of the orthogonality of L
B
i
and L
B
j
.
Theorem 22.4. For every linear space with line set / on an n-set
X, we have
N(n) min
A/
N([A[) 1.
Proof: Let k be the indicated minimum. This means we have at
least k + 1 pairwise orthogonal quasigroups on each A /. We
will describe how to obtain k pairwise orthogonal idempotent quasi-
groups on each A. Then N(n) k follows from the construction
above.
In general, let H
1
, H
2
, . . . , H
k+1
be pairwise orthogonal quasi-
groups on an m-set B. Pick any b B. There will be m cells
(x, y) for which H
k+1
(x, y) = b, one in each row and one in each
column. Simultaneously permute the columns, say, of all squares
so that these cells are on the diagonal of the (k + 1)-th square.
Orthogonality implies that for each i k, all m symbols occur on
the diagonal of the i-th square. Finally, we permute the symbols
independently in each of the rst k squares so that the resulting
squares are idempotent.
Theorem 22.5. If / is the set of lines of a linear space on an n-set
X and B / is a set of pairwise disjoint lines, then
N(n) min (N([A[) 1 : A /B N([B[) : B B) .
Proof: Let k be the indicated minimum. As we saw in the proof
of the preceding theorem, there exist k pairwise orthogonal idem-
potent quasigroups L
A
i
on each A in /B. If necessary, we add
290 A Course in Combinatorics
singleton sets to B so that B becomes a partition of X. We have
k pairwise orthogonal quasigroups L
B
i
(not necessarily idempotent)
on each B B. We dene L
1
, L
2
, . . . , L
k
on the entire set X by
declaring, for each i = 1, 2, . . . , k, that L
i
(x, x) := L
B
i
(x, x) where
B is the unique member of B that contains x, and for distinct x, y,
L
i
(x, y) := L
A
i
(x, y), where A is the unique line in / (whether or
not in B) that contains both x and y. The quasigroups L
i
are or-
thogonal (the easy details are left to the reader) and thus N(n) k.

Example 22.1. Let n = 21 and consider the projective plane of

order 4 with 21 lines of size 5. Theorem 22.4 implies that N(21) 3
and this disproves MacNeishs conjecture. Delete three noncollinear
points. We have a linear space with 18 points, and lines of sizes 5,
4, and 3; the three lines of size 3 are pairwise disjoint. Theorem
22.5 implies that N(18) 2 and this disproves Eulers conjecture.
Example 22.2. To see how far wrong Euler was, we consider two
more values of n 2 (mod 4). Delete three noncollinear points
from a projective plane of order 8 to obtain a linear space on 70
points with lines of sizes 9, 8, and (pairwise disjoint lines of size) 7.
Theorem 22.5 shows N(70) 6. It is possible to nd seven points
in the plane constructed from F
8
such that no three are collinear
(even 10, see Problem 19I), and deleting these produces a linear
space which can be used in Theorem 22.4 to give us N(66) 5.
The next construction would be dicult to describe without the
use of transversal designs. Transversal designs provide a compact
and conceptually convenient language with which to manipulate
sets of pairwise orthogonal Latin squares. (They also provide a
large family of linear spaces useful for the construction of other
combinatorial designs or even more orthogonal Latin squares.) A
TD(n, k) can be dened as the dual incidence structure of an (n, k)-
net, i.e. as an (n, k, k 1)-partial geometry. So there are nk points
and n
2
blocks (we will use blocks rather than lines when dis-
cussing transversal designs). Each point is in n blocks; each block
contains k points. The points fall into k equivalence classes (called,
confusingly, groups) of size n so that two points in the same group
are not contained in a block while two points in dierent groups
22. Orthogonal Latin squares 291
belong to exactly one block. In particular, each block contains ex-
actly one point from each group. We need to refer to the groups
so often that we incorporate them into the notation and speak of
a transversal design (X, (, /) where the three coordinates are the
sets of points, groups, and blocks, respectively. (X, ( /) is a
linear space with block-sizes k and n.
Example 22.3. Here is a TD(2,3):
Groups: a
1
, a
2
, b
1
, b
2
, c
1
, c
2

Blocks: a
1
, b
1
, c
1
, a
1
, b
2
, c
2
, a
2
, b
1
, c
2
, a
2
, b
2
, c
1
.
In view of Theorem 22.2, the existence of a TD(n, k+2) is equiva-
lent to the existence of k pairwise orthogonal Latin squares. At the
risk of boring some readers, we review the connection by quickly
describing how to get the squares from a TD(n, k + 2): number
the groups G
1
, G
2
, . . . , G
k+2
and dene L
i
: G
1
G
2
G
i+2
by
declaring L
i
(x, y) to be the point of intersection of G
i+2
and the
block which contains x and y.
A TD(n, k) is said to be resolvable when the set / of blocks
can be partitioned into n parallel classes /
1
, /
2
, . . . , /
n
, i.e. each
/
i
is a set of n blocks of size k which partition the point set X.
From a TD(n, k) (X, (, /), we can always construct a resolvable
TD(n, k 1) by deleting the n points of one of the groups G
0
=
x
1
, x
1
, . . . , x
n
, which removes exactly one point from each block;
for each i, the blocks which have had x
i
removed form a parallel
class on the point set XG
0
of the TD(n, k 1). (Conversely, a
resolvable TD(n, k 1) can be extended to a TD(n, k).)
Theorem 22.6. If 0 u t, then
N(mt +u) minN(m), N(m+ 1), N(t) 1, N(u).
Proof: Let k be the right-hand side above, plus 2. This means
that transversal designs TD(m, k), TD(m+1, k), TD(t, k +1), and
TD(u, k) exist; and to prove the theorem, we must construct a
TD(mt +u, k).
As the construction is rather technical, we warm up by rst de-
scribing the construction of a TD(mt, k) from a TD(t, k) and vari-
ous TD(m, k)s. This is the degenerate case u = 0 in the theorem.
292 A Course in Combinatorics
The construction does not require a TD(m+1, k) or a TD(t, k +1)
and thus it reproves Theorem 22.3(i).
Let (X, (, /) be a TD(t, k). To each x X, associate a set M
x
of
m new elements so that any two sets M
x
are disjoint. For S X,
let M
S
:=
xS
M
x
.
We construct a TD(mt, k) on the point set M
X
of size kmt with
groups M
G
: G ( each of size mt; the blocks B are obtained as
follows. For each A /, choose blocks B
A
so that
(M
A
, M
x
: x A, B
A
)
is a TD(m, k), and let B =
A/
B
A
. The verication is straight-
forward.
To return to the general case, recall that the existence of a
TD(t, k + 1) implies the existence of a resolvable TD(t, k). So we
have a TD(t, k) (X, (, /) where / admits a partition into parallel
classes /
1
, /
2
, . . . , /
t
. We treat the blocks of the rst u parallel
classes in one way, and the blocks in B :=
t
i=u+1
/
i
in another way.
Let (U, H, () be a TD(u, k). We also require a partition of U
into u k-subsets K
1
, K
2
, . . . , K
u
; each is to consist of exactly one
point from each set H H, but these k-subsets are not required to
be blocks in (.
Let ( = G
1
, G
2
, . . . , G
k
and H = H
1
, H
2
, . . . , H
k
be number-
ings of the groups. We construct a TD(mt + u, k) with point set
Y := M
X
U and groups
:= M
G
1
H
1
, M
G
2
H
2
, . . . , M
G
k
H
k
.
The blocks are obtained as follows. For each block B B as before,
let
(M
B
, M
x
: x B, T
B
)
be a TD(m, k). For each block A CalA
i
, let
(M
A
K
i
, (M
A
M
G
j
) (K
i
H
j
) : j = 1, 2, . . . , k, T
A
)
be a TD(m+1, k) in which K
i
occurs as a block, and let T
/
A
denote
the remaining (m+1)
2
1 blocks. Then we claim that (Y, , c) is
the required TD(mt +u, k), where
c := (
_
_
BB
T
B
_

_
_
A/
1
/
u
T
/
A
_
.
22. Orthogonal Latin squares 293
Verication requires consideration of several cases.
Example 22.4. With m = 3 in Theorem 22.6, we see that
(22.2)
N(3t +u) 2 whenever 0 u t, N(t) 3, and N(u) 2.
We may take (t, u) = (5, 3), (7, 1), (7, 5), and (9, 3) to nd that
N(n) 2 for n = 18, 22, 26, and 30.
Theorem 22.7. N(n) 2 for all n ,= 2, 6.
Proof: We need only consider n 2 (mod 4). For n = 10, 14,
see Examples 22.6 and 22.7 below. For n = 18, 22, 26, and 30, see
Example 22.4 above. We now assume n 34. One of
n 1, n 3, n 5, n 7, n 9, n 11
is divisible by 3 but not by 9, so we can write n = 3t + u where
u = 1, 3, 5, 7, 9, or 11, and t is not divisible by 3. Since n is even,
t is also odd, so N(t) 4 by Theorem 22.3(ii). A consequence of
n 34 is t 11, so 0 u t, and then N(n) 2 by (22.2).
Theorem 22.8. N(n) as n .
Proof: Let x be a positive integer. We claim that N(n) x 1
whenever
(22.3) n 2
_

px
p
_
2x+1
,
where the product is extended over all primes p x.
Given n satisfying (22.3), let m be chosen (Chinese remainder
theorem) so that 0 m (

px
p)
x
and such that for all primes
p x,
m
_
1 (mod p
x
) if p divides n,
0 (mod p
x
) if p does not divide n.
Then choose an integer t 1 (mod

px
p) so that
0 u := n mt <
_

px
p
_
x+1
.
The remainder of the proof is provided by Problem 22C below.
294 A Course in Combinatorics
Problem 22C. With m, t, and u chosen as above, show that u t
and that Theorem 22.3(ii) implies that N(mt +u) x 1.
* * *
We have seen constructions by so-called dierence methods in
Chapter 19. We use this idea to construct Latin squares here. It
is convenient to describe our constructions in terms of orthogonal
arrays. An OA(n, k) is a k n
2
array (or we prefer here to think of
a set of n
2
column vectors of height k since the order of the columns
is immaterial) whose entries are taken from a set S of n symbols, so
that for any distinct i, j, 1 i, j k, and any two symbols s, t from
S, there is a unique column whose i-th coordinate is s and whose
j-th coordinate is t. This generalizes the OA(n, 3)s introduced in
Chapter 17.
This is yet another equivalent formulation of orthogonality; the
existence of an OA(n, k + 2) is equivalent to the existence of k
pairwise orthogonal Latin squares. For example, to get the array
from squares L
1
, L
2
, . . . , L
k
, assume without loss of generality that
all row, column, and symbol sets are the same set S and take all
columns
[i, j, L
1
(i, j), L
2
(i, j), . . . , L
k
(i, j)]

, i, j S.
Readers should immediately be able to write out two orthogonal
squares from the OA(3, 4) in Fig. 22.2 (of course, any two rows may
be used to coordinatize the squares).
_

_
x x x y y y z z z
x y z x y z x y z
x y z z x y y z x
x y z y z x z x y
_

_
Figure 22.2
Example 22.5. The following matrix of elements of Z
15
was found
with the aid of a computer by Schellenberg, Van Rees, and Vanstone
22. Orthogonal Latin squares 295
(1978):
_

_
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 2 5 7 9 12 4 1 14 11 3 6 8 10 13
0 6 3 14 10 7 13 4 11 2 8 5 1 12 9
0 10 6 1 11 2 7 12 3 8 13 4 14 9 5
_

_
It has the property that for any two rows, the dierences between
the coordinates of those two rows comprise all elements of Z
15
, each
with multiplicity one. It follows that we obtain an OA(15, 5) by
taking all translates of these columns by elements of Z
15
. Thus three
pairwise orthogonal Latin squares of order 15 can be constructed.
But the OA(15, 5) is resolvable: the 15 translates of any column
have the property that in each coordinate, all symbols occur once.
As with transversal designs, we can add a new row to obtain an
OA(15, 6) and hence four pairwise orthogonal Latin squares of order
15. It is not known whether N(15) 5.
The last three rows of the above matrix are what we call below
pairwise orthogonal orthomorphisms of Z
15
.
In 1960, Johnson, Dulmage, Mendelsohn and a computer proved
by this method that N(12) 5, using the group Z
2
Z
2
Z
3
. The
cyclic group of order 12 is of no use; see Theorem 22.9 below.
Example 22.6. We will describe a construction that proves
N(m) 2 N(3m+ 1) 2.
Let G be an abelian group of order 2m+1 and let M be a system of
representatives for the pairs i, i (i ,= 0) in G. That is [M[ = m
and G = 0 M M. For each i M, introduce a new symbol

i
and take S := G
i
: i M as the symbols set for an
OA(3m+ 1, 4).
Consider the set of all column vectors obtained from
_

_
_

_
0
0
0
0
_

_
_

_
_

i
0
i
i
_

_
,
_

_
0

i
i
i
_

_
,
_

_
i
i

i
0
_

_
,
_

_
i
i
0

i
_

_
: i M
_

_
by translating them by the elements of G with the understand-
ing that the s are xed, i.e.
i
+ g =
i
. This gives us
296 A Course in Combinatorics
(4m + 1)(2m + 1) columns. Add to these the m
2
columns of an
OA(m, 4) to obtain an OA(3m+ 1, 4).
To be more explicit in the case m = 3, let us use symbols Z
7

x, y, z and M = 1, 2, 3. We take the seven translates of the
following 13 columns, and adjoin the OA(3, 4) from Fig. 22.2:
_

_
0 x 0 1 6 y 0 2 5 z 0 3 4
0 0 x 6 1 0 y 5 2 0 z 4 3
0 1 6 x 0 2 5 y 0 3 4 z 0
0 6 1 0 x 5 2 0 y 4 3 0 z
_

_
Verication of the construction requires special consideration of
the symbols x, y, z (which we have used rather than the s), and
the observation that the dierences between any two of the above
rows (when both coordinates are in Z
7
) comprise all the elements
of Z
7
, each with multiplicity one. The resulting squares are shown
below.
_

_
0 z 1 y 2 x 3 6 5 4
4 1 z 2 y 3 x 0 6 5
x 5 2 z 3 y 4 1 0 6
5 x 6 3 z 4 y 2 1 0
y 6 x 0 4 z 5 3 2 1
6 y 0 x 1 5 z 4 3 2
z 0 y 1 x 2 6 5 4 3
1 2 3 4 5 6 0 x y z
2 3 4 5 6 0 1 z x y
3 4 5 6 0 1 2 y z x
_

_
_

_
0 4 x 5 y 6 z 1 2 3
z 1 5 x 6 y 0 2 3 4
1 z 2 6 x 0 y 3 4 5
y 2 z 3 0 x 1 4 5 6
2 y 3 z 4 1 x 5 6 0
x 3 y 4 z 5 2 6 0 1
3 x 4 y 5 z 6 0 1 2
6 0 1 2 3 4 5 x y z
5 6 0 1 2 3 4 y z x
4 5 6 0 1 2 3 z x y
_

_
Example 22.7. We will describe an OA(14, 4) on the symbols
Z
11
x, y, z. We take the 11 translates by elements of Z
11
of the
following 17 columns, and adjoin the OA(3, 4) from Fig. 22.2:
_

_
0 0 6 4 1 x 1 4 0 y 2 6 0 z 8 9 0
0 1 0 6 4 0 x 1 4 0 y 2 6 0 z 8 9
0 4 1 0 6 4 0 x 1 6 0 y 2 9 0 z 8
0 6 4 1 0 1 4 0 x 2 6 0 y 8 9 0 z
_

_
Once again, verication of the construction requires special consid-
eration of the symbols x, y, z (which we have used rather than the
s), and the observation that the dierences between any two of
22. Orthogonal Latin squares 297
the above rows (when both coordinates are in Z
11
) comprise all the
elements of Z
11
, each with multiplicity one.
Problem 22D. Show that the existence of k pairwise orthogonal
idempotent quasigroups is equivalent to the existence of a TD(n, k+
2) in which there may be found a parallel class of blocks.
An orthomorphism of an abelian group G is a permutation of
the elements of G such that
x (x) x
is also a permutation of G. The reader can check that the square
L(x, y) := (x) +y is Latin if and only if is a permutation, and is
orthogonal to the addition table A(x, y) := x + y of G if and only
if is an orthomorphism.
We remark that if there is any orthogonal mate to A, then there
is an orthomorphism: the cells in the positions where an orthogonal
mate contains a given symbol have the property that there is ex-
actly one in each row and column, and so are of the form (x, (x)),
x G, for some permutation . Since these cells must contain
dierent symbols in A, the mapping dened by (x) := x +(x)
is a permutation. In view of this remark, the following theorem
proves that cyclic squares of even order have no orthogonal mates,
i.e. there are no Latin squares orthogonal to it.
Theorem 22.9. If an abelian group G admits an orthomorphism,
then its order is odd or its Sylow 2-subgroup is not cyclic.
Proof: If the Sylow 2-subgroup is cyclic and nontrivial, then there
is exactly one element z of G with order 2. When we add all
elements of G, each element pairs with its additive inverse except
z and 0; so the sum of all elements of G is z. But if were an
orthomorphism, then
z =

xG
((x) x) =

xG
(x)

xG
x = z z = 0,
a contradiction to our choice of z.
298 A Course in Combinatorics
Remarks. If G is abelian and has odd order, then it has automor-
phisms xing only the identity (this is true for some abelian groups
of even order also), and these are orthomorphisms. Even if G is
not abelian, one can dene orthomorphisms (also called complete
mappings) and it remains true that these do not exist if the Sylow 2-
subgroup is cyclic and nontrivialsee Hall and Paige (1955), where
complete mappings are also shown to exist for solvable groups with
trivial or noncyclic Sylow 2-subgroups.
* * *
H. B. Mann (1950) observed that a Latin square of order 4t + 1
with a subsquare of order 2t has no orthogonal mates. Similarly, a
Latin square of order 4t +2 with a subsquare of order 2t +1 has no
orthogonal mates. Both these results are corollaries of part (ii) of
the following theorem, which also proves that none of the pairs of
orthogonal squares of orders 12t + 10 we constructed in Example
22.6 can be extended to a set of three or more pairwise orthogonal
Latin squares. We leave it to the reader to convince himself that
subsquares correspond to subtransversal designs.
Theorem 22.10. Let (X, (, /) be a TD(n, k) which contains a
sub-TD(m, k) (Y, H, B) with m < n. (This means Y X, H =
G Y : G (, and B /.) Then
(i) m(k 1) n;
(ii) if (X, (, /) is resolvable, then
m
2
n
_
mk n
k 1
_
.
Proof: Pick a point x
0
XY . Say x
0
belongs to G
0
(. For
each of the m(k 1) points y Y G
0
, there is a unique block
A
y
/ with x
0
, y A
y
. A block containing two points of Y
must belong to B (and could not contain x
0
), so these m(k 1)
blocks are distinct. This number cannot exceed the total number
of blocks on x
0
, which is n.
Suppose that (X, (, /) is resolvable and that /
1
, /
2
, . . . , /
n
is
a partition of / into parallel classes. Let s
i
denote the number of
22. Orthogonal Latin squares 299
blocks of /
i
which are contained in Y and let t
i
denote the number
of blocks of /
i
which meet Y in exactly one point. Then
ks
i
+t
i
= [Y [ = mk and s
i
+t
i
n,
from which it follows that (k 1)s
i
mk n, or equivalently,
s
i

_
mk n
k 1
_
.
Part (ii) now follows from m
2
= [B[ =

n
i=1
s
i
.
Problem 22E. Prove that N(24) 3 and N(33) 3.
Problem 22F. Suppose we have c 1 mutually orthogonal Latin
squares of order k Then we have an OA(k, c+1) which we represent
as follows:
_
_
_
_
1 1 . . . 1 2 2 . . . 2 . . . k k . . . k
1 2 . . . k
.
.
. A
1
A
k1
1 2 . . . k
_
_
_
_
,
where every row of each matrix A
i
is a permutation of 1, 2, . . . , k.
Let A be the c k(k 1) matrix formed by the matrices A
i
. Use
this matrix and the matrix S of Problem 19U to prove the following
theorem.
Theorem. If there exists a symmetric (v, k, 1)-design, then N(v)
N(k).
Problem 22G. Generalize Theorem 22.6 to the following.
Theorem. If 0 u t, 0 v t, then
N(mt +u +v)
minN(m), N(m+ 1), N(m+ 2), N(t) 2, N(u), N(v).
Problem 22H. Show that N(21) 4.
Problem 22I. Show that N(51) 4.
300 A Course in Combinatorics
Problem 22J. Suppose that by giving examples one has shown
that N(n) 3 for a n 7a + 3. Show that N(n) 3 for
7a n 7
2
a and conclude that N(n) 3 for n a.
Notes.
The term mutually orthogonal Latin squares rather than pair-
wise orthogonal Latin squares and its acronym MOLS are in com-
mon usage.
We have already mentioned Euler in the notes to Chapter 1.
Eulers paper of 1782 was titled A new type of magic square,
because Euler had noticed that special magic squares arise from
a pair of orthogonal Latin squares. As an example, change the
symbol sets of A and B in Fig. 22.1 to 4, 3, 2, 1 and 12, 8, 4, 0,
respectively, and add, rather than superpose, to obtain
_
_
_
16 11 6 1
2 5 12 15
9 14 3 8
7 4 13 10
_
_
_
.
Each line of this matrix sums to 34. (The Latin property does not
in general imply that the diagonals sum to the magic number, but
our example is a particularly nice square: the diagonals, the four
corners, and many other sets of four entries sum to 34.)
The authors are aware that A is used in Fig. 22.1 both as the
name of a matrix and an entry in that same matrix. We are just
checking whether readers are on their toes.
Orthogonal Latin squares, often in the form of orthogonal ar-
rays, are of great importance in the statistical theory of design of
experiments.
MacNeishs paper of 1922 actually includes a false proof of Eulers
conjecture.
The connection between orthogonal Latin squares and nite pro-
jective planes was observed by R. C. Bose (1938).
Many results on orthogonal Latin squares were anticipated by
E. H. Moore (1896), as has been pointed out to the authors by R.
D. Baker. Unfortunately, the term Latin square is nowhere to be
found in Moores paper, so no one noticed his results for a long
22. Orthogonal Latin squares 301
time. The results in his paper include special cases of Theorem
22.3 and Theorem 22.4 (albeit in entirely dierent forms) as well as
descriptions of projective planes of prime power order, so he could
have disproved the MacNeish conjecture had he known of it.
Theorem 22.6 is a special case of a construction of Wilson (1974)
which in turn has been generalized in several ways. See T. Beth,
D. Jungnickel, and H. Lenz (1986).
Using methods similar to Problem 22J, the second of the present
authors proved that N(n) 3 for n 47. Presently it is known
that N(n) 3 for n > 10. See Colbourn and Dinitz (1996) for
recent lower bounds on N(n) for various n.
Theorem 22.8 is due to Chowla, Erd os, and Straus (1960). They
proved more: N(n) n
1/91
for suciently large n. Their work
is number-theoretical and is based on the BoseParkerShrikhande
constructions. Their result has been improved several times, but
the best result to date is that of T. Beth (1983). This seems far
from the real truth. It is quite possible that N(n) n/10, say, for
all, or for all but nitely many, values of n, but at the present time
this appears to be incredibly hard to prove.
References.
T. Beth (1983), Eine Bemerkung zur Absch atzung der Anzahl
orthogonaler lateinischer Quadrate mittels Siebverfahren, Abh.
Math. Sem. Hamburg 53, 284288.
T. Beth, D. Jungnickel, and H. Lenz (1986), Design Theory, Bibli-
ographisches Institut.
R. C. Bose (1938), On the application of the properties of Ga-
lois elds to the problem of construction of hyper-Graeco-Latin
squares, Sanhkya 3, 323338.
R. C. Bose, S. S. Shrikhande, and E. T. Parker (1960), Further
results in the construction of mutually orthogonal Latin squares
and the falsity of a conjecture of Euler, Canad. J. Math. 12,
189203.
S. Chowla, P. Erd os, and E. G. Straus (1960), On the maximal
number of pairwise orthogonal Latin squares of a given order,
Canad. J. Math. 12, 204208.
C. J. Colbourn and J. H. Dinitz, editors (1996), The CRC Handbook
of Combinatorial Designs, CRC Press.
302 A Course in Combinatorics
A. L. Dulmage, D. Johnson, and N. S. Mendelsohn (1961), Ortho-
morphisms of groups and orthogonal Latin squares, Canadian J.
Math. 13, 356372.
M. Hall and L. J. Paige (1955), Complete mappings of nite groups,
Pacic J. Math. 5, 541549.
H. F. MacNeish (1922), Euler squares, Ann. Math. 23, 221227.
H. B. Mann (1950), On orthogonal Latin squares, Bull. Amer.
Math. Soc. 50, 249257.
E. H. Moore (1896), Tactical memoranda IIII, Amer. J. Math. 18,
264-303.
P. J. Schellenberg, G. M. J. Van Rees, and S. A. Vanstone (1978),
Four pairwise orthogonal Latin squares of order 15, Ars Comb.
6, 141150.
D. R. Stinson (1984), Nonexistence of a Pair of Orthogonal Latin
Squares of Order Six, J. Combinatorial Theory (A) 36, 373376.
R. M. Wilson (1974), Concerning the number of mutually orthog-
onal Latin squares, Discrete Math. 9, 181198.
23
Projective and combinatorial
geometries
A combinatorial geometry is a pair (X, T) where X is a set of points
and where T is a family of subsets of X called ats such that
(1) T is closed under (pairwise) intersection,
(2) there are no innite chains in the poset T,
(3) T contains the empty set, all singletons x, x X, and the
set X itself,
(4) for every at E T, E ,= X, the ats that cover E in T
partition the remaining points.
Here, F covers E in T means that E, F T, E _ F, but that
E _ G _ F does not hold for any G T. This latter property
should be familiar to the reader from geometry: the lines that
contain a given point partition the remaining points; the planes
that contain a given line partition the remaining points.
A trivial example of a geometry consists of a nite set X and all
subsets of X as the ats. This is the Boolean algebra on X.
We remark that (1) and (2) imply that T is closed under arbitrary
intersection.
Example 23.1. Every linear space (as introduced in Chapter 19)
gives us a combinatorial geometry on its point set X when we take
as ats , all singletons x : x X, all lines, and X itself. The
fact that the lines on a given point partition the remaining points
is another way of saying that two points determine a unique line.
Example 23.2. Every Steiner system S(t, k, v) gives us a com-
binatorial geometry on its point set X when we take as ats all
subsets of cardinality < t, all blocks, and X. To get a geometry by
304 A Course in Combinatorics
this construction, it is not necessary that all blocks have the same
size, but only that each t-subset is contained in a unique block
(cf. Problem 19R, where we have called these Generalized Steiner
Systems).
Example 23.3. Let V be an n-dimensional vector space over a
eld F. By an ane subspace of V , we mean either the empty
set or a coset (or translate) of an (ordinary, i.e. linear) subspace
of V in the additive group. For example, for a, b F, the subset
(x, y) : y = ax+b is an ane subspace of F
2
. The set V together
with all ane subspaces forms a combinatorial geometry called the
ane geometry AG
n
(F). We write AG
n
(q) for AG
n
(F
q
); the case
n = 2 was introduced in Chapter 19.
Example 23.4. The projective geometry PG
n
(F), while more fun-
damental than AG
n
(F), can be a little more awkward to dene.
Let V be an (n+1)-dimensional vector space V over a eld F. The
point set X of PG
n
(F) is to consist of all the 1-dimensional (linear)
subspaces of V . For example, if F is the eld F
q
of q elements, then
the number of projective points of PG
n
(F
q
) is
q
n+1
1
q 1
= q
n
+ +q
2
+q + 1.
To each linear subspace W of V , we associate a at F
W
consisting
of all the 1-dimensional subspaces of V that are contained in W,
and take T to be the set of all such ats F
W
. We write PG
n
(q) for
PG
n
(F
q
); the case n = 2 was introduced in Chapter 19.
Projective geometries PG
n
(F) can be dened also for F a division
ring (lacking commutativity of multiplication); see Crawley and
Dilworth (1973).
Let Y be any subset of the elements X of a combinatorial geom-
etry (X, T) and let
c := F Y : F T.
Then (Y, c) is also a combinatorial geometry, the subgeometry on
Y . For example, AG
n
(F) is a subgeometry of PG
n
(F). There is
a standard embedding of the point set of AG
n
(F) into the point
23. Projective and combinatorial geometries 305
set of PG
n
(F): to each vector x in an n-dimensional vector space
V , associate the 1-dimensional subspace that is the span of (x, 1)
in the (n + 1)-dimensional space V F. The image consists of all
projective points not contained in the hyperplane V 0.
The set of ats T of a combinatorial geometry, when ordered by
inclusion, has several important properties.
A lattice L is a partially ordered set with the property that any
nite subset S L has a meet (or greatest lower bound), that is an
element b in L so that

aS
[b a] and
aS
[c a] c b,
as well as a join (or least upper bound), that is an element b in L
so that

aS
[b a] and
aS
[c a] c b.
The meet and join of a two-element set S = x, y are denoted,
respectively, by x y and x y. It is easily seen that and are
commutative, associative, idempotent binary operations; moreover,
if all two-element subsets have meets and joins, then any nite
subset has a meet and a join.
The lattices we will consider have the property that there are no
innite chains. Such a lattice has a (unique) least element (that we
denote by 0
L
) because the condition that no innite chains exist
allows us to nd a minimal element m, and any minimal element
is minimum since if m _ a, then m a would be less than m.
Similarly, there is a unique largest element 1
L
.
For elements a and b of a poset, we say that a covers b and write
a b when a > b but there are no elements c so that a > c > b.
For example when U and W are linear subspaces of a vector space,
U W when U W and dim(U) = dim(W) + 1. Recall that a
chain in a partially ordered set P is a totally ordered subset of P.
So a nite chain can be thought of as a sequence a
0
< a
1
< < a
n
.
A point of a lattice L with miminum element 0
L
is an element that
covers 0
L
.
A geometric lattice is a lattice L that has no innite chains, and
such that
(1) L is atomic, that is, each element of L is the join of points
of L, and
306 A Course in Combinatorics
(2) L is semimodular, that is, if a and b are distinct and both
cover c in L, then a b covers both a and b.
Theorem 23.1. The set of ats of a combinatorial geometry, or-
dered by inclusion, is a geometric lattice. Conversely, given a geo-
metric lattice L with points X, then (X, F
y
: y L) is a combi-
natorial geometry, where
F
y
:= x X : x y.
Proof: Since the set T of ats of a combinatorial geometry is
closed under intersection, the poset of ats is an atomic lattice
(the meet of two ats is their intersection; the join of two ats is
the intersection of all ats containing both). Suppose that ats F
1
and F
2
cover a at E in a combinatorial geometry. Let x be a point
in F
1
F
2
. There is a at G
2
that covers F
2
in T and contains the
point x. G
2
contains F
1
since otherwise E _ F
1
G
2
_ F
1
. But by
a symmetric argument, there is a at G
1
that covers F
1
and also
contains F
1
and F
2
. The join F
1
F
2
must be contained in both G
1
and G
2
. Since G
i
covers F
i
, F
1
F
2
= G
1
= G
2
.
Let L be a geometric lattice and dene the ats F
y
as subsets
of the point set X of L as in the statement of the theorem. It is
clear that the empty set, any singleton, and X are ats (when y
is taken to be 0
L
, a point of L, or 1
L
, respectively). Since x y
and x z if and only if x y z, we have F
y
F
z
= F
yz
; thus
T := F
y
: y L is closed under intersection. No point x not in a
given at F
y
can be in two ats that cover F
y
, so it remains only to
show that some at that covers F
y
contains x. We show that F
xy
covers F
y
to complete the proof.
This is equivalent to showing that xy y in L whenever x _ y
in L and x is a point of L. Choose a maximal chain
0
L
< y
1
< y
2
< < y
k
= y.
Since x and y
1
cover 0
L
, x y
1
covers y
1
. Since x y
1
and y
2
cover
y
1
, x y
2
covers y
2
(clearly x y
1
y
2
= x y
2
). Inductively, we
nd that x y
k
covers y
k
.
In some sense, the dierence between geometric lattices and com-
binatorial geometries is the same as that between incidence struc-
tures on the one hand and families of subsets of a set on the other.
23. Projective and combinatorial geometries 307
Incidence structures and lattices are more abstract and must be
used to avoid confusion in certain cases (e.g. when discussing du-
ality or intervals), but we prefer the language and notation of sets
and subsets for many arguments.
For example, for elements a and b of a partially ordered set,
the interval [a, b] is dened as [a, b] := x : a x b. Any
interval in a lattice is again a lattice. The reader should check
that any interval of a geometric lattice is again a geometric lattice.
It would be awkward to state the corresponding fact in terms of
combinatorial geometries.
In view of Theorem 23.1 (more precisely, the simple correspon-
dences decribed in the proof), it is often convenient (though some-
times confusing) to mix the notation and language of combinatorial
geometries and geometric lattices. For example, we can use the
same symbol PG
n
(F) to denote the combinatorial geometry and
the corresponding lattice of subspaces of a vector space.
Problem 23A. (i) Any interval of PG
n
(F) is isomorphic as a poset
to PG
m
(F) for some m n. (ii) PG
n
(F) is isomorphic to its dual
poset (where order is reversed).
Example 23.5. The partition lattices
n
, whose elements are all
partitions of an n-set X, provide another family of geometric lat-
tices, and hence give us combinatorial geometries. The partitions
are ordered by renement; / is a renement of B if each block of
B is the union of blocks of /. So the least element (the nest par-
tition) is all singletons x : x X and the greatest element is
X, consisting of a single block. A moments thought shows that
one partition covers another if the former is obtained by coalescing
two blocks of the latter into one. The points are the partitions that
consist of a single block of size 2 and n 2 singletons.
For example,
4
has 15 elements:
1234
123, 4, 124, 3, 134, 2, 234, 1, 12, 34, 13, 24, 14, 23
12, 3, 4, 13, 2, 4, 14, 2, 3, 23, 1, 4, 24, 1, 3, 34, 1, 2
1, 2, 3, 4
308 A Course in Combinatorics
To simplify the notation, we have written 123, 4, for example,
for what should be 1, 2, 3, 4.
Example 23.6. The points of
n
are in one-to-one correspondence
with the edges of K
n
. Given any simple graph G on n vertices, we
obtain a geometric lattice L(G) whose points correspond to the
edge set E(G) of G as follows. The elements of L are to be all
those partitions / of V (G) such that the subgraph of G induced
by each block of / is connected. The partitions with this property
are exactly those that are the join (in
n
) of points (in
n
) corre-
sponding to some subset of the edges of G. The lattices L(G) are
sometimes called the lattices of contractions of G (see Chapter 33).
Fig. 23.1 is an attempt to illustrate the combinatorial geometries
L(K
4
) and L(K
5
). L(K
5
), for example, has ten points; the dot in
the gure labeled 12 represents the point (partition) 12, 3, 4, 5.
The partitions of the type 123, 4, 5, for example, contain three
points and are represented by line segments; lines with two points
are not shown. (We note that the six point/seven line L(K
4
) is the
geometry that arises from deleting a point from the Fano congu-
ration.)
Figure 23.1
For a subset S of the points of a combinatorial geometry, the
closure S is the intersection of all ats containing S. For example,
in a linear space, the closure of a two point set is the line containing
the points. Exercise: show that S S, A B A B, and
S = S. It will be occasionally convenient to use the symbol for join
in the lattice of ats; note that
E F := E F.
23. Projective and combinatorial geometries 309
A subset S X is independent when for each x S, x / S x.
In Example 23.1, three points in a linear space are independent
if and only if they are not contained in a line; no four points are
independent. A set of points in PG
n
(F) is independent if and only
if representative vectors are linearly independent. A set of points
in AG
n
(F) is independent if and only if the vectors are anely
independent. One viewpoint of combinatorial geometries is that
they represent the study of abstract independence.
By the rank of a combinatorial geometry with point set X, we
mean the maximum size of an independent subset of X. PG
n
(F)
and AG
n
(F) both have rank n + 1. We will avoid the word di-
mension, but if we forget and use it, it will mean the rank minus
one. The maximum size of an independent subset of a at F is the
rank of that at. The ats of rank 1 are called points, the ats of
rank 2 are called lines, the ats of rank 3 are called planes. The
ats of rank 1 less than rank(X) are called hyperplanes or copoints.
Occasionally, we need to use the term coline for a at for which the
rank equals rank(X) 2.
From any combinatorial geometry (X, T), we obtain a linear
space (X, L) where L is the set of all lines (ats of rank 2).
In Fig. 23.1, any two points determine a line that may have two
or three points; only the latter are drawn in the gure. The planes
in L(K
5
) have either four or six points; the latter contain four
three-point lines.
We collect some simple facts:
Lemma 23.2.
(1) If x / A but x A y, then y A x (the exchange
axiom).
(2) If S is an independent set of points in a geometry and x / S,
then x S is independent.
(3) If F = S, then F = A for any maximal independent subset
A of S.
Proof: F
1
:= A x is the at that covers E := A and contains
x; F
2
:= A y is the at that covers E and contains y. If x F
2
,
it must be that F
1
= F
2
. This proves (1).
If x S is not independent, there is some y S such that
310 A Course in Combinatorics
y (S x)y. Let A := Sy. Since S is independent,
y / A. But (1) then implies that x A y, i.e. x S. This
proves (2).
Suppose F = S and that A is an independent subset of S that
is maximal with respect to that property. If S _ A, we could nd
a larger independent subset of S by (2). So S A and since F is
the smallest at containing S, F A. This proves (3).
A basis for a at F is an independent subset B F so that
B = F, i.e. any maximal independent subset of F. As an exercise,
the reader should check that we could also have dened a basis of
F as a minimal spanning set, i.e. a subset B of F so that B = F
and that is minimal with respect to this property.
Problem 23B. Let G be a connected simple graph. Show that
the bases of the combinatorial geometry L(G) are exactly the edge
sets of spanning trees in G.
Theorem 23.3. All bases of a at F in a combinatorial geometry
have the same nite cardinality (called the rank of F). For ats E
and F,
(23.1) rank(E) + rank(F) rank(E F) + rank(E F).
Proof: It should be clear that the condition that there are no
innite chains of ats forces all independent sets to be nite.
If it is not true that all bases of F have the same cardinality,
choose an ordered pair (B
1
, B
2
) of bases with [B
1
[ > [B
2
[ but so
that [B
2
B
1
[ is as small as possible subject to this constraint. Pick
x B
1
B
2
. Then B
1
x is independent and has a closure which
does not contain F and hence does not contain B
2
. Pick y B
2
B
1
such that y / B
1
x. Then (B
1
x) y is independent by
Lemma 23.2 and is contained in a basis B
3
for F. Now we have an
ordered pair (B
3
, B
1
) of bases with [B
3
[ > [B
2
[, but B
2
B
3
is a
proper subset of B
2
B
1
, a contradiction.
To prove the second part of the theorem, rst note that Lemma
23.2(2) implies that a basis for a at E can be extended to a basis
for any at E
/
containing E.
Now let B be a basis for EF. Extend B to bases B
1
and B
2
for
E and F, respectively. Then any at containing B
1
B
2
contains E
23. Projective and combinatorial geometries 311
and F and hence E F; that is, B
1
B
2
= E F and so B
1
B
2
contains a basis for E F. Then
rank(E F) [B
1
B
2
[ = [B
1
[ +[B
2
[ [B
1
B
2
[
= rank(E) + rank(F) rank(E F).

The inequality (23.1) is called the semimodular law.

We remark that the proof of Theorem 23.3 shows that all maxi-
mal independent subsets of any subset S, not necessarily a at, have
the same cardinality. This is a corollary of the statement for ats,
however, because we can apply Theorem 23.4 to the subgeometry
on S.
As an exercise, make sure that you understand the following: Let
E and F be ats in a geometry with E F. If rank(E) = rank(F),
then E = F; F covers E if and only if rank(F) = rank(E) + 1.
For the ane geometries AG
n
(q), the number of points on a at
of rank r is q
r
. For the projective geometries PG
n
(q), the number
of points on a at of rank r is
(23.2)
q
r+1
1
q 1
= q
r
+ +q
2
+q + 1.
Theorem 23.4. Suppose that every at of rank i in a combina-
torial geometry on v points has exactly k
i
points, i = 0, 1, . . . , n.
Then the total number of ats of rank r is
(23.3)
r1

i=0
(v k
i
)
(k
r
k
i
)
.
Furthermore, the set of points together with the family of rank r
ats is a 2-design.
Proof: Since the ats of rank r + 1 that contain a at of rank
r partition the remaining v k
r
points into sets of size k
r+1
k
r
,
there must be exactly (v k
r
)/(k
r+1
k
r
) such ats. The formula
(23.3) now follows by induction on r when we count the ordered
pairs (E, F) of rank r and rank r +1 ats with E F . Note that
k
0
= 0 and k
1
= 1, so that (23.3) is valid for r = 1.
312 A Course in Combinatorics
Any two points are contained in a unique rank 2 at. The above
argument implies that any rank 2 at is contained in the same
number of rank r ats. Hence the rank r ats give a 2-design.
The numbers of rank r ats of PG
n
(q) are called Gaussian num-
bers; see Chapter 24. Equations (23.2) and (23.3) imply that the
numbers of points and hyperplanes of PG
n
(q) are equal (there are
other ways of seeing this) and so we have the following corollary.
Corollary. The points and hyperplanes of the projective geometry
PG
n
(q) form a (v, k, )-symmetric design with
v = (q
n+1
1)/(q 1),
k = (q
n
1)/(q 1),
= (q
n1
1)/(q 1).
Problem 23C. Show that the incidence structure whose points
are the points of AG
r
(2) and whose blocks are the planes of AG
r
(2)
is a Steiner system S(3, 4, 2
r
).
The following theorem is due to C. Greene (1970).
Theorem 23.5. The number of hyperplanes in a nite combina-
torial geometry is at least the number of points.
Proof: The proof will be similar to that of Theorem 19.1. First,
we shall show that if a point x is not on a hyperplane H, then
the number r
x
of hyperplanes on x is at least the number k
H
of
points on H by induction on the rank. The assertion is trivial for
geometries of rank 2. By the induction hypothesis, we know
that the number of hyperplanes of the subgeometry on H (i.e. the
number of colines C contained in H) is at least k
H
. But for each
such coline C, we get a hyperplane (the join of C and x) on x.
Now we repeat from the proof of Theorem 19.1: Let H denote
the set of hyperplanes, v := [X[, b := [H[. Suppose b v. Then
1 =

xX

H,x
1
v(b r
x
)

HH

x/ H
1
b(v k
H
)
= 1,
and this implies that in all the inequalities, equality must hold.
Therefore v = b.
23. Projective and combinatorial geometries 313
See the remark following Lemma 23.8 below concerning the case
of equality in Theorem 23.5.
In studying PG
n
(F), the dimension formula
dim(U W) + dim(U +W) = dim(U) + dim(W)
for linear subspaces of a vector space plays a crucial role. Equiva-
lently,
(23.4) rank(E F) + rank(E F) = rank(E) + rank(F)
for ats E and F of PG
n
(F). This is a stronger version of the
semimodular law. Indeed, a combinatorial geometry in which (23.4)
holds is said to be modular. In a modular combinatorial geometry,
for example, (23.4) implies that any two lines contained in a plane
must meet nontrivially (cf. Theorem 19.1).
Problem 23D. Let (X, B) be the linear space consisting of the
points and lines of a nite modular combinatorial geometry of rank
3. Show that (X, B) is either a nearpencil as dened following
Theorem 19.1 or else is a projective plane as dened in Chapter
19that is, show that any two lines have the same cardinality
n +1 and that there are exactly n
2
+n +1 points for some integer
n 2.
We have introduced the projective geometries PG
n
(F) and have
dened projective planes in Chapter 19. Here is the general de-
nition: A projective geometry is a modular combinatorial geometry
that is connected in the sense that the point set cannot be expressed
as the union of two proper ats.
We do not have the space or the time to prove here the following
fundamental result. See Veblen and Young (1907) or Crawley and
Dilworth (1973) for a proof.
Theorem 23.6. Every projective geometry of rank n 4 is iso-
morphic to PG
n
(F) for some division ring F.
Combinatorial geometries of rank 2 are without great interest,
but all are modular and, except for the two point line, are projective
geometries. The rank 3 projective geometries are, by Problem 23D,
314 A Course in Combinatorics
equivalent to projective planes as introduced in Chapter 19 (the
nearpencils are not connected).
Every modular combinatorial geometry can be put together from
projective geometries. This is the content of the following problem.
Problem 23E.
(i) Let (X
1
, T
1
) and (X
2
, T
2
) be modular combinatorial geome-
tries; X
1
X
2
= . Show that (X
1
X
2
, F
1
F
2
: F
1
T
1
, F
2
T
2
)
is a modular combinatorial geometry.
(ii) Let (X, T) be a modular combinatorial geometry such that
X is the union of two ats X
1
and X
2
where X
1
X
2
= . Let
T
i
:= F X
i
: F T, i = 1, 2. Show that (X
i
, T
i
), i = 1, 2, is
a modular combinatorial geometry and that T = F
1
F
2
: F
1

T
1
, F
2
T
2
.
Problem 23F. Prove that every at F of a combinatorial geome-
try (X, T) has a modular complement, i.e. there exists a at E such
that E F = , E F = X, and rank(E) + rank(F) = rank(X).
The linear spaces consisting of the points and lines of modular
combinatorial geometries satisfy the following condition known as
the Pasch axiom. A line that meets two sides of a triangle also
meets the third. More precisely, suppose A, B, C are distinct lines
and a, b, c distinct points with incidence as in Fig. 23.2. We require
that any line L that contains a point other than a or b from the
line C and a point other than a or c from the line B also contains a
point other than b or c from the line A. See Fig. 23.2. To see that
this holds in a modular geometry, notice that L and A must both
be contained in the plane P = a, b, c, and the modular equation
implies that two lines in a plane must meet nontrivially.
Figure 23.2
23. Projective and combinatorial geometries 315
Problem 23G. Consider the incidence structure whose points are
the k-dimensional subspaces of a vector space V and whose lines
are the (k + 1)-dimensional subspaces of V , with incidence being
containment. Show that this incidence structure satises the Pasch
axiom.
Theorem 23.7. A nite linear space (X, /) for which the Pasch
axiom holds, consists of the points and lines of some modular com-
binatorial geometry on X.
Proof: Suppose the Pasch axiom holds. We must construct the
ats of a combinatorial geometry, which we do as follows. We say
that a subset S X is a at when, for any line L,
[L S[ 2 implies L S.
Let T denote the set of all such ats; this includes the empty
set, all singletons, and X, as well as all lines in /. Check that
the denition implies that the intersection of any number of ats
is again a at.
Let x be a point not in a at S. Let T be the union of all
lines joining x to the points s S. We claim that this simple
construction produces a at T, and that T covers S.
To this end, let L be a line containing two points t
1
, t
2
of T. We
want to show L T. If t
1
, t
2
both belong to one of the lines on x,
then L is that line and so L T. Otherwise both t
1
and t
2
are in
S; then L T, so assume t
1
/ S. Then t
1
and t
2
belong to distinct
lines M
1
and M
2
on x. Say M
i
meets S in s
i
, i = 1, 2, and let N
be the line joining s
1
and s
2
. We have N S. The Pasch axiom
guarantees that L meets N in some point z of S.
Consider any other point t
3
L. Let M
3
be the line joining x
and t
3
. Since M
3
meets two sides of the triangle with sides M
1
,
L, and N (with vertices z, t
1
, and s
1
), it must meet the third side
N in some point s
3
of S. Then M
3
is one of the lines joining x to
points of S, so t
3
T. We have now proved that T is a at.
That T covers S is easy to see: If U is a at, S _ U T, pick
y US. By construction, the line joining x and y is one of the
lines on x that meets S; that line contains two points of U and
hence is contained in U. This means x U, whence T U.
316 A Course in Combinatorics
Since any point lies in a at that covers a at S, (X, T) is a
combinatorial geometry. (The hypothesis that X is nite is only
used here to ensure that there are no innite chains of ats.)
Let x be a point on a line L and let H be any hyperplane. Since
(if x is not already in H) the join of H and x must be X, L must
be one of the lines on x that meet H. That is, any line L and
any hyperplane H meet nontrivially. This implies modularity by
Lemma 23.8 below.
Lemma 23.8. A combinatorial geometry is modular if and only if
any line and any hyperplane meet nontrivially.
Proof: The modular law (23.4) shows that any rank 2 at and
any rank n1 at in a rank n geometry meet in a at of rank 1.
For the converse, we use induction on the rank. Assume that
all lines and hyperplanes in a rank n combinatorial geometry meet
nontrivially. Suppose H is a hyperplane and that there exists a line
L H and a rank n2 at C H that are disjoint. But then for
any point x / H, the hyperplane C x would be disjoint from
the line L. So by induction, the subgeometry on any hyperplane H
is modular.
Thus if there is a pair of ats E and F that provide a coun-
terexample to the modular law (23.4), then E F is equal to the
entire point set X. Let H be a hyperplane containing E and let
F
/
:= HF. Then E F
/
H. But rank(F
/
) rank(F) 1 since
otherwise a modular complement of F
/
in F would have rank 2
and so would contain a line, one that would be disjoint from H. We
see that E and F
/
provide a counterexample to (23.4) contained in
H, contradicting the modularity of the subgeometry H.
Remark. Suppose the number of hyperplanes of a nite combi-
natorial geometry (X, T) is equal to the number of points. Then,
reviewing the proof of Theorem 23.5, r
x
= k
H
whenever x / H. If
we apply Theorem 23.5 to the interval [x, X] in any combinato-
rial geometry, we nd that r
x
l
x
, where l
x
denotes the number of
lines containing x (since these lines are the points of the geomet-
ric lattice [x, X]). For x / H we obviously have l
x
k
H
since
the lines x y, y H, are distinct. In summary, equality in
Theorem 23.5 implies that l
x
= k
H
when x / H. This means that
23. Projective and combinatorial geometries 317
every line meets every hyperplane and hence (X, T) is modular by
Lemma 23.8.
Two triangles a
1
, b
1
, c
1
and a
2
, b
2
, c
2
are said to be perspective
from a point if there exists a point p (the point of perspectivity) such
that p, a
1
, a
2
are collinear, p, b
1
, b
2
are collinear, and p, c
1
, c
2

are collinear. Two triangles a

1
, b
1
, c
1
and a
2
, b
2
, c
2
are said to
be perspective from a line if there exists a line L (the line or axis
of perspectivity) such that L, A
1
, A
2
are concurrent, L, B
1
, B
2

are concurrent, and L, C

1
, C
2
are concurrent. Here we are using
triangle to mean three noncollinear points, and A
i
, B
i
, and C
i
to
denote the sides of the triangle opposite a
i
, b
i
, and c
i
, respectively.
That is, A
i
:= b
i
, c
i
, i = 1, 2, etc.
Figure 23.3
In Fig. 23.3, the triangles are perspective from the point p and
also perspective from the line L. The incidence stucture so illus-
trated, with ten points and ten lines, each line on three points and
each point on three lines, is called the Desargues conguration. The
triangles need not lie in a plane. The right diagram in Fig. 23.1
is isomorphic (!) to that in Fig. 23.3 as an incidence structure;
the former is easier to imagine as 3-dimensional. It can be viewed
in many ways, e.g. as showing that the triangles 25, 35, 45 and
12, 13, 14 are perspective from the point 15 and also perspective
from the line 23, 24, 34.
If we choose a basis for the (n+1)-dimensional vector space whose
subspaces comprise PG
n
(F), we can describe projective points by
so-called homogeneous coordinates. By x
0
, x
1
, . . . , x
n
), we mean
the 1-dimensional subspace spanned by the vector (x
0
, x
1
, . . . , x
n
).
318 A Course in Combinatorics
The projective point x
0
, x
1
, . . . , x
n
) is the same as y
0
, y
1
, . . . , y
n
)
if and only if the vector (y
0
, y
1
, . . . , y
n
) is a nonzero scalar multiple
of (x
0
, x
1
, . . . , x
n
). The hyperplanes may also be described by ho-
mogeneous (n+1)-tuples: [c
0
, c
1
, . . . , c
n
] will denote the hyperplane
consisting of points with homogeneous coordinates x
0
, x
1
, . . . , x
n
)
so that c
0
x
0
+x
1
c
1
+ +x
n
c
n
= 0.
Theorem 23.9 (Desargues theorem). In PG
n
(F), if two tri-
angles are perspective from a point, then they are perspective from
a line.
Proof: Let triangles a
1
, b
1
, c
1
and a
2
, b
2
, c
2
be perspective from
a point p. We are going to assume that no three of p, a
1
, b
1
and
c
1
are collinear so that we do not spend too much time on details.
We will also assume that all points lie in a plane, i.e. we prove the
theorem for PG
2
(F). The other cases are left for the reader.
We choose a basis for a 3-dimensional vector space so that the
basis vectors x, y, z span the 1-dimensional subspaces a
1
, b
1
, and
c
1
, respectively. Then the homogeneous coordinates for a
1
, b
1
, and
c
1
are 1, 0, 0), 0, 1, 0), and 0, 0, 1), respectively. Then p has ho-
mogeneous coordinates , , ) where all coordinates are nonzero.
We replace the original basis vectors by x, y and z; then a
1
,
b
1
, and c
1
are still represented by 1, 0, 0), 0, 1, 0), and 0, 0, 1), re-
spectively, while p now has homogeneous coordinates 1, 1, 1). This
will simplify our computations.
It follows that a
2
= , 1, 1), b
2
= 1, , 1), c
2
= 1, 1, ), for some
, , F. The line joining a
1
and b
1
is [0, 0, 1]; the line joining
a
2
and b
2
is [1 , 1 , 1] (check that both a
2
and b
2
are
on this line). The point on both lines is 1 , 1, 0). Similarly,
the point of intersection of the lines joining b
1
and c
1
with the line
joining b
2
and c
2
is found to be 0, 1 , 1), and the point of
intersection of the lines joining c
1
and a
1
with the line joining c
2
and
a
2
is found to be 1, 0, 1 ). The three points of intersection
are collinear since their coordinates are linearly dependent.
(If we are careful, the proof should go through also for division
rings F.)
Problem 23H. Prove that the linear space of PG
n
(F) also satis-
es Pappus theorem when F is a (commutative) eld, which is as
23. Projective and combinatorial geometries 319
follows. Let a
i
, b
i
, and c
i
be collinear points, say on a line L
i
, for
i = 1, 2. Suppose L
1
and L
2
meet in a point (not one of the six
named points). Then the lines a
1
, b
2
and a
2
, b
1
meet in a point
c, the lines b
1
, c
2
and b
2
, c
1
meet in a point a, and the lines
a
1
, c
2
and a
2
, c
1
meet in a point b; and the three points a, b, c
are collinear.
Figure 23.4
Projective planes in which the statement of Theorem 23.9 is valid
(if two triangles are perspective from a point, then they are per-
spective from a line) are called Desarguesian planes. Projective
planes in which the conclusion of Problem 23H holds are called
Pappian planes. We do not have the space or time to prove the
further following fundamental result. See Crawley and Dilworth
(1973).
Theorem 23.10.
(1) A projective geometry of rank 3 is isomorphic to PG
2
(E) for
some division ring E if and only if it is a Desarguesian plane.
(2) A projective geometry of rank 3 is isomorphic to PG
2
(F) for
some eld F if and only if it is a Pappian plane.
A corollary of Theorem 23.10 is that every Pappian plane is De-
sarguesian. For a synthetic (i.e. coordinate-free) proof that Pap-
pus theorem implies Desargues theorem, see D. Pedoe (1963). We
remark that every nite division ring is a eld (Wedderburns the-
orem), so that every nite Desarguesian plane is Pappian; see M.
Hall, Jr. (1972).
We describe one method of obtaining nite projective planes that
are not Desarguesian.
320 A Course in Combinatorics
Let G be an abelian group of order n
2
, and suppose that there
exist subgroups H
0
, H
1
, . . . , H
n
of order n that partition the nonzero
elements of G, that is such that H
i
H
j
= 0 for i ,= j and,
therefore,
n
_
i=0
H
i
0 = G 0.
Then, we claim, the incidence structure whose points are the el-
ements of G and whose lines are all cosets (translates) of all the
subgroups H
i
is an ane plane. In general, the intersection of cosets
of two subgroups H and K is either empty or a coset of H K,
so lines meet in at most one point. We could now count covered
pairs to show that every two distinct points are contained in a line,
but it is just as simple to note that given points x, y G, the coset
H

+ y contains both points where H

is the subgroup containing

the dierence x y. An ane plane obtained from a group G and
n+1 subgroups in this way is called an ane translation plane. It
is known that G must be elementary abelian for such subgroups to
exist; see J. Andre (1954). The ane plane may be completed to a
projective plane as in Problem 19K; any projective plane obtained
in this way is a translation plane.
Example 23.7. Here is a partition of the nonzero elements of F
4
3
into ten subgroups H
0
, H
1
, . . . , H
9
. (We have denoted the elements
as strings, dropping commas and parentheses.)
0000, 1000, 2000, 0100, 0200, 1100, 2200, 2100, 1200
0000, 0010, 0020, 0001, 0002, 0011, 0022, 0021, 0012
0000, 1010, 2020, 0101, 0202, 1111, 2222, 2121, 1212
0000, 2010, 1020, 0201, 0102, 2211, 1122, 1221, 2112
0000, 0110, 0220, 2001, 1002, 2111, 1222, 2221, 1112
0000, 0210, 0120, 1001, 2002, 1211, 2122, 1121, 2212
0000, 1110, 2220, 2101, 1202, 0211, 0122, 1021, 2012
0000, 2210, 1120, 1201, 2102, 0111, 0222, 2021, 1012
0000, 2110, 1220, 2201, 1102, 1011, 2022, 0121, 0212
0000, 1210, 2120, 1101, 2202, 2011, 1022, 0221, 0112
.
The ane plane corresponding to this spread of subgroups is the
Desarguesian plane.
23. Projective and combinatorial geometries 321
But note that the elements of the rst four subgroups, for exam-
ple, can be partitioned into subgroups in another way, as indicated
by the columns below.
0000, 0000, 0000, 0000,
1000, 2000, 0100, 0200, 1100, 2200, 2100, 1200,
0010, 0020, 0001, 0002, 0011, 0022, 0021, 0012,
1010, 2020, 0101, 0202, 1111, 2222, 2121, 1212,
2010, 1020 0201, 0102 2211, 1122 1221, 2112
The above example is the special case q = 3 of a construction we
now describe. Let V be a 2-dimensional vector space over F
q
2 and
let H
0
, H
1
, . . . , H
q
2 be the 1-dimensional subspaces over F
q
2 . If we
consider V as a 4-dimensional vector space over F
q
, the subspaces
H
i
are 2-dimensional over F
q
(they form a spread of lines in PG
3
(q);
see Theorem 24.3). Let U be any 2-dimensional subspace over F
q
other than one of the H
i
s. Then U meets any H
i
in 0 or in a
1-dimensional subspace over F
q
. Say U meets H
0
, H
1
, . . . , H
q
in q
points each and H
q+1
, H
q+2
, . . . , H
q
2 in the zero vector only. Con-
sider the F
q
-subspaces U as ranges over the nonzero elements of
F
q
2 ; since multiplying by the nonzero elements of F
q
xes U, there
are only (q
2
1)/(q 1) distinct such subspaces, say U
0
, U
1
, . . . , U
q
.
These q +1 subspaces U
i
meet pairwise only in the zero vector and
their union is H
0
H
1
H
q
. Thus
U
0
, U
1
, . . . , U
q
, H
q+1
, H
q+2
, . . . , H
q
2
is another partition of V into subgroups of order q
2
.
We shall not prove here that the projective planes resulting from
this construction are not Desarguesian (for q > 2). See Theorem
10.9 in D. R. Hughes and F. C. Piper (1973). The reader can nd
counterexamples to Desargues theorem in Example 23.7 by hand.
We conclude the chapter with proof of an important step in the
dicult proof of Theorem 23.6. We need a preliminary observation.
Proposition 23.11. The subgeometry on any at of a projective
geometry is again a projective geometry.
Proof: That the modular law holds in the subgeometry on a at
of a modular projective geometry is immediate. We must show that
322 A Course in Combinatorics
these subgeometries are connected. To this end, it will suce to
show that if H is a hyperplane in a projective geometry with point
set X, then the subgeometry on H is connected.
If not, H = EF where E and F are ats. By (23.4), rank(E)+
rank(F) = n1 where n := rank(X). Let E
1
, . . . , E
s
and F
1
, . . . , F
t
be the ats that cover E and F, respectively, that contain points
outside of H; each family partitions the points outside of H. Sup-
pose that both s, t 2. By (23.4), there exists x
i
E
i
F
i
, i = 1, 2.
The line x
1
, x
2
, again by (23.4), meets H in some point e E.
But then x
2
, being on the line joining e and x
1
, must be contained
in E
1
, a contradiction. Thus one of s or t must be 1, say s = 1, and
then X is the union of the disjoint ats F and E
1
, contradicting
the connectivity of the original projective geometry.
A corollary of Proposition 23.11 is that every line of a projective
geometry has at least three points (since a two-point line is not
connected). This is used at a critical point in the proof below. We
remark that with the aid of Problem 23D, it is easy to see that all
lines in a nite projective geometry have the same size n + 1.
Theorem 23.12. Desargues theorem holds in any projective ge-
ometry of rank 4.
Proof: Consider two triangles a
1
, b
1
, c
1
and a
2
, b
2
, c
2
that are
perspective from a point x.
We rst assume that the planes P
1
and P
2
spanned by a
1
, b
1
, c
1

and a
2
, b
2
, c
2
, respectively, are distinct. Let T := x, a
1
, b
1
, c
1
.
The lines x, a
1
, x, b
1
, and x, c
1
contain, respectively, the
points a
2
, b
2
, and c
2
, so P
1
and P
2
are contained in T, that evi-
dently has rank 4. So by the modular law, P
1
and P
2
must meet in
a line L. We claim that the two original triangles are perspective
from the line L.
The plane Q := p, a
1
, b
1
contains a
2
and b
2
and hence contains
both the lines a
1
, b
1
and a
2
, b
2
that therefore intersect in a point
q, say. The point q belongs to both P
1
and P
2
, and hence q L.
Similarly, the lines b
i
, c
i
, i = 1, 2, meet in a point on L; and the
lines a
i
, c
i
, i = 1, 2, meet in a point on L. The two triangles are
thus perspective from L.
Now assume that the two original triangles lie in a plane P.
23. Projective and combinatorial geometries 323
Let x
1
be any point not in P and x
2
any other point on the line
containing x
1
and p. The lines x
1
, a
1
and x
2
, a
2
are contained
in the plane p, x
1
, a
1
and so meet in a point a

. Similarly, the
lines x
1
, b
1
and x
2
, b
2
meet in a point b

and the lines x

1
, c
1

and x
2
, c
2
meet in a point c

. The plane P

:= a

, b

, c

and P
are contained in a rank 4 at and so meet in a line L := P P

.
The triangles a
1
, b
1
, c
1
and a

, b

, c

are perspective from the

point x
1
and lie in dierent planes, and so are perspective from the
line L. Similarly, the triangles a
2
, b
2
, c
2
and a

, b

, c

are per-
spective from x
2
, and so are perspective from L. The line a
1
, b
1

meets L in the same point as the line a

, b

; the line a
2
, b
2
meets
L in the same point as the line a

, b

; hence the lines a

1
, b
1
and
a
2
, b
2
meet in a point of L. Similarly, the other corresponding
sides of the two original triangles meet in points of L. In summary,
the original triangles are perspective from the line L.
Problem 23I. Determine the number of ats of rank nk in the
partition lattice
n
.
Problem 23J. For an arbitrary block design (with parameters
v, k, , b, r) we dene a line to be the intersection of all blocks con-
taining a pair of points x, y. Prove:
(i) Any two points are on a unique line;
(ii) If L is a line, then 2 [L[
b
r
;
(iii) If a line meets a block in two points, it is contained in the
block.
Notes.
Girard Desargues (15931662) was an architect and military engi-
neer who introduced the term involution in his work on perspective.
He worked in Lyons and Paris. The work of Pappus of Alexandria
(. 320), the last signicant Greek geometer, was considerably ear-
lier. The mathematical discipline of projective geometry did not
develop until the 19th century.
Moritz Pasch was originally an algebraist but then became inter-
ested in non-euclidean geometry. He came to Giessen at the age of
27 as Privatdozent and stayed there until his death (60 years later).
He was Rektor of the university 18931894.
324 A Course in Combinatorics
An interesting family of combinatorial geometries called transver-
sal geometries arises from bipartite graphs; see Crapo and Rota
(1970). The independent sets of these geometries are those subsets
of one part X of a bipartition (X, Y ) with the property that they
can be matched into Y .
We have seen that the concept of combinatorial geometries and
geometric lattices are cryptomorphic, i.e. they are essentially dif-
ferent axiom systems for the same structures. Crapo and Rota
(1970) give further cryptomorphic versions. For example, ma-
troids are more-or-less the same thing as combinatorial geometries,
though the emphasis of the theory of matroids is dierent.
Some characterizations of the symmetric design consisting of the
points and hyperplanes of a nite projective space may be found
in Section 2.1 of P. Dembowski (1968).
References.
J. Andre (1954),

Uber nicht-Desarguessche Ebenen mit transitiver
Translationgruppe, Math. Zeitschr. 60, 156186.
L. M. Batten (1986), Combinatorics of Finite Geometries, Cam-
bridge University Press.
H. Crapo and G.-C. Rota (1970), Combinatorial Geometries, MIT
Press.
P. Crawley and R. P. Dilworth (1973), Algebraic Theory of Lattices,
Prentice-Hall.
P. Dembowski (1968), Finite Geometries, Springer-Verlag.
C. Greene (1970), A rank inequality for nite geometric lattices, J.
Combinatorial Theory 9, 357364.
M. Hall, Jr. (1972), The Theory of Groups, 2nd edn., Chelsea.
D. R. Hughes and F. C. Piper (1973), Projective Planes, Springer-
Verlag.
D. Pedoe (1963), An Introduction to Projective Geometry, Macmil-
lan.
O. Veblen and J. W. Young (1907), Projective Geometries (2 vols.),
Ginn Co.
24
Gaussian numbers and
q-analogues
There exist many analogies between the partially ordered set of all
subsets of a nite set and the partially ordered set of all subspaces
of a nite vector space. This is primarily because they are both
examples of matroid designs as dened in the previous chapter.
Let V
n
(q) denote an n-dimensional vector space over the eld F
q
of q elements. We use the term k-subspace as a short form of k-
dimensional subspace.
We begin with some counting. To obtain a maximal chain (i.e.
a chain of size n + 1 containing one subspace of each possible di-
mension) in the partially ordered set of all subspaces of V
n
(q), we
start with the 0-subspace. After we have chosen an i-subspace U
i
,
0 i < n, we can choose an (i + 1)-subspace U
i+1
that contains
U
i
in (q
n
q
i
)/(q
i+1
q
i
) ways since we can take the span of U
i
and any of the (q
n
q
i
) vectors not in U
i
but any (i +1)-subspace
will arise exactly (q
i+1
q
i
) times in this manner. In summary, the
number of maximal chains of subspaces in V
n
(q) is
M(n, q) =
(q
n
1)(q
n1
1)(q
n2
1) (q
2
1)(q 1)
(q 1)
n
.
We may consider M(n, q) as a polynomial in q for each integer
n. When the indeterminate q is replaced by a prime power, we
have the number of maximal chains in the poset PG
n
(q). When
q is replaced by 1, we have M(n, 1) = n!, which is the number of
maximal chains in the poset of subsets of an n-set.
The Gaussian number
_
n
k

q
can be dened as the number of k-
subspaces of V
n
(q). These are called Gaussian coecients by some
326 A Course in Combinatorics
authors to emphasize the analogy with binomial coecients. To
nd an expression for
_
n
k

q
, we count the number N of pairs (U, ()
where U is a k-subspace and ( is a maximal chain that contains
U. Of course, every maximal chain contains exactly one subspace
of dimension k, so N = M(n, q). On the other hand, we get each
such maximal chain uniquely by appending to a maximal chain
in the partially ordered set of all subspaces of U, of which there
are M(k, q), a maximal chain in the partially ordered set of all
subspaces of V
n
(q) that contain U; there are M(n k, q) of these,
since the partially ordered set W : U W V is isomorphic to
the partially ordered set of subspaces of the factor space V/U that
has dimension n k. Thus
_
n
k
_
q
=
M(n, q)
M(k, q)M(n k, q)
=
(q
n
1)(q
n1
1) (q
nk+1
1)
(q
k
1)(q
k1
1) (q 1)
.
For some purposes, it is better to think of
_
n
k

q
as a polynomial
in an indeterminate q rather than as a function of a prime power
q. That the rational function above is in fact a polynomial can
be seen in several ways. For example, it is an easy exercise to see
that a rational function in x which is integral for innitely many
integral values of x must be a polynomial in x. Perhaps Gaussian
polynomial is a better term than Gaussian number or coecient.
As an example,
_
6
3
_
q
= q
9
+q
8
+ 2q
7
+ 3q
6
+ 3q
5
+ 3q
4
+ 3q
3
+ 2q
2
+q + 1.
When the indeterminate q is replaced by 1 in
_
n
k

q
, we obtain
_
n
k
_
. This explains a small part of a tendency for results concerning
nite vector spaces to reduce to the corresponding results for sets
when q is replaced by 1. It is also possible to have so-called q-
analoques of results on sets where we try to replace k-subset by
k-subspace. Sometimes these statements are true and have proofs
that are similar to the results on sets.
The following is the q-analogue of Sperners theorem, Theorem
6.3.
24. Gaussian numbers and q-analogues 327
Theorem 24.1. If / is an antichain in the partially ordered set
of all subspaces of V
n
(q), then
[/[
_
n
n/2|
_
q
.
Proof: Let / be an antichain and count the number N of pairs
(U, () where U / and ( is a maximal chain that contains U.
Every maximal chain contains at most one subspace in /, so N
M(n, q). On the other hand, each k-subspace in / lies in exactly
M(k, q)M(n k, q) maximal chains (. Thus
M(n, q) N =
n

k=0
c
k
M(k, q)M(n k, q),
where c
k
is the number of k-dimensional subspaces belonging to
/. The proof is completed in a manner analogous to the proof of
Theorem 6.3 if we believe that
_
n
k

q

_
n
n/2|
_
q
for all k, and this is
left to the reader to verify.
The following theorem gives a combinatorial interpretation of the
coecients of the
_
n
k

q
as a polynomial in q, and thus proves they
are all positive integers.
Theorem 24.2. Let
_
n
k
_
q
=
k(nk)

=0
a

.
Then the coecient a

is the number of partitions of whose Ferrers

diagrams t in a box of size k n k.
Proof: We can work with the vector space F
n
q
of n-tuples over
F
q
. It is well known that every k-subspace of F
n
q
arises uniquely as
the rowspace of a k n matrix over F
q
that (i) has rank k and
(ii) is a so-called row-reduced echelon form. This means that the
rst nonzero entry in each row is a 1, the entries above a leading 1
are 0s, and the leading 1 in row i is further to the right than the
leading 1 in row i 1, i = 2, 3, . . . , k.
328 A Course in Combinatorics
Suppose the leading 1 in row i occurs in column c
i
, i = 1, 2, . . . , k.
Then (nk +1c
1
, nk +2c
2
, . . . , n1c
k1
, nc
k
) is a non-
increasing sequence of nonnegative numbers and so, when terminal
zeros are dropped, corresponds to a partition of some number into
at most k parts of size at most n k. Conversely, such a partition
gives us the positions of leading ones in a class of echelon forms.
For example, there are 20 classes of echelon forms in the case
n = 6, k = 3; see Fig. 24.1.
_
1 0 0
0 1 0
0 0 1
_ _
1 0 0
0 1 0
0 0 0 1
_ _
1 0 0
0 1 0
0 0 0 0 1
_ _
1 0 0
0 1 0
0 0 0 0 0 1
_ _
1 0 0
0 0 1 0
0 0 0 1
_
_
1 0 0
0 0 1 0
0 0 0 0 1
_ _
1 0 0
0 0 1 0
0 0 0 0 0 1
_ _
1 0 0
0 0 0 1 0
0 0 0 0 1
_ _
1 0 0
0 0 0 1 0
0 0 0 0 0 1
_ _
1 0 0
0 0 0 0 1 0
0 0 0 0 0 1
_
_
0 1 0 0
0 0 1 0
0 0 0 1
_ _
0 1 0 0
0 0 1 0
0 0 0 0 1
_ _
0 1 0 0
0 0 1 0
0 0 0 0 0 1
_ _
0 1 0 0
0 0 0 1 0
0 0 0 0 1
_ _
0 1 0 0
0 0 0 1 0
0 0 0 0 0 1
_
_
0 1 0 0
0 0 0 0 1 0
0 0 0 0 0 1
_ _
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1
_ _
0 0 1 0 0
0 0 0 1 0
0 0 0 0 0 1
_ _
0 0 1 0 0
0 0 0 0 1 0
0 0 0 0 0 1
_ _
0 0 0 1 0 0
0 0 0 0 1 0
0 0 0 0 0 1
_
Figure 24.1
In each matrix of Fig. 24.1, the positions of the dots, when the
intervening columns are deleted and we reect through a vertical
axis, describe the Ferrers diagram of a partition (of some number
9) that ts in a 3 by 3 box. The class represented by the last
matrix in the rst row, for example, contains q
7
echelon forms.
In general, the class of echelon forms where the leading 1 in row
i occurs in column c
i
, i = 1, 2, . . . , k, contains q

matrices for some

since the positions not containing 1s or required to be 0s may
be lled arbitrarily with elements of F
q
. To be precise,
= (n k + 1 c
1
) + + (n 1 c
k1
) + (n c
k
)
because for each i = 1, 2, . . . , k, there are n (k i) c
i
positions
in the i-th row that may be arbitrarily lled. That is, the class
consists of q

matrices where the partition into at most k parts of

24. Gaussian numbers and q-analogues 329
size at most nk that corresponds to the class is in fact a partition
of the number .
Thus when a

is dened as the number of partitions of whose

Ferrers diagrams t in a box of size k nk, we have a polynomial

k(nk)
=0
a

that agrees with the polynomial

_
n
k

q
when evaluated
at any prime power q; hence the two polynomials are equal.
We remark that when we set q = 1 in the statement of Theorem
24.2, we obtain as a corollary the result that the total number of
partitions whose Ferrers diagram ts in a box of size k n k is
_
n
k
_
.
Problem 24A. Show this directly.
Next, we derive a recursion (24.1) for the Gaussian numbers.
This provides another way to see that they are polynomials in q
(by induction on n, say). Pick a hyperplane H, i.e. an (n 1)-
subspace, of V
n
(q). Some k-subspaces are contained in H (their
number is
_
n1
k

q
) and the rest meet H in a (k 1)-subspace. Each
of the
_
n1
k1

q
(k 1)-subspaces in H is contained in
_
n k + 1
1
_
q
=
q
nk+1
1
q 1
k-subspaces of V , of which
_
n k
1
_
q
=
q
nk
1
q 1
are contained in H; this leaves q
nk
that are not contained in H.
Thus
(24.1)
_
n
k
_
q
=
_
n 1
k
_
q
+q
nk
_
n 1
k 1
_
q
.
Problem 24B. Determine the exponents e
i
(they may be func-
tions of m, n, and k as well as i) so that the following identity is
valid:
_
n +m
k
_
q
=
k

i=0
q
e
i
_
n
i
_
q
_
m
k i
_
q
.
330 A Course in Combinatorics
(One way to solve this involves echelon forms; another uses (24.1).)
The equation of Problem 24B is a q-analoque of the equation
_
n +m
k
_
=
k

i=0
_
n
i
__
m
k i
_
for binomial coecients. The q-analogue of inclusion-exclusion ap-
pears in the next chapter; see Theorem 25.2 for an application.
What is the q-analogue of (simple) t-designs? This would be
a family B of k-subspaces of V
v
(q) such that each t-subspace is
contained in exactly members of B. No nontrivial examples of
such q-S

(t, k, v)s were known with t 2 until 1986 when S.

Thomas described a family with
q = 2, = 7, t = 2, k = 3, v 1 (mod 6).
The case t = = 1 is already nontrivial for vector spaces. For
sets, an S(1, k, v) is a partition of a v-set into k-subsets, and this
exists if and only if k divides v.
Theorem 24.3. There exists a family of k-subspaces of V
v
(q) such
that every 1-subspace is contained in exactly one member of the
family (a so-called spread of k-subspaces) if and only if k divides v.
Proof: In other words, we are interested in families of k-subspaces
such that any two of them intersect in the 0-subspace but such that
their union is the entire v-dimensional vector space.
The number of k-subspaces that we require is the total number
of nonzero vectors divided by the number of nonzero vectors in a
k-subspace, and this is (q
v
1)/(q
k
1). This is an integer if and
only if k divides v.
Suppose v = km where m is an integer. As a v-dimensional
vector space over F
q
, we take an m-dimensional vector space V
over F
q
k. Let B be the family of 1-subspaces of V as a vector space
over F
q
k. (So there are q
k(m1)
+q
k(m2)
+ +q
k
+1 of these and
each contains q
k
vectors.) Now think of V as a vector space over
F
q
, a subeld of F
q
k; the members of B are k-dimensional subspaces
over F
q
and provide the required partition.
24. Gaussian numbers and q-analogues 331
Problem 24C. Let B be a spread of k-subspaces in V
v
(q). Let /
consist of all cosets of all members of B. Show that / is the block
set of an S(2, q
k
, q
v
) on the set of points V
v
(q). (The case v = 2k
is particularly interesting because we obtain ane planes of order
q
k
. Planes obtained from spreads are called translation planes.)
Problem 24D. Prove that
n

k=0
(q
n
1) . . . (q
nk+1
1)q
(
k
2
)
_
n
k
_
q
= q
n
2
.
The Erd os-Ko-Rado Theorem (see Theorem 6.4) can be formu-
lated in two (equivalent) ways. In one formulation, the bound for
m is
_
n1
k1
_
and in the other it is
k
n

_
n
k
_
, where
_
n
k
_
counts the total
number of k-subsets of the n-set.
Now consider a possible q-analogue. Start with a collection / :=
A
1
, A
2
, . . . , A
m
of m distinct k-dimensional subspaces of the n-
dimensional vectorspace V (n, q) over F
q
such that any two of the
A
i
intersect in a subspace of dimension 1. Then one can either
conjecture that if k
1
2
n
m
_
n 1
k 1
_
q
or m
k
n

_
n
k
_
q
.
In this case, the bounds dier. The rst one is stronger than the
second one. For a proof of the rst bound see W. N. Hsieh (1975).
This result was strengthened by P. Frankl and R. M. Wilson (1986).
The second bound can be proved by generalizing the proof of The-
orem 6.4 which is the next problem.
Problem 24E. Consider the situation described above. Let :=
(x
1
, x
2
, . . . , x
n
) run through the set B
n
of all ordered bases of V (n, q).
We place on a circle. If a sequence of k consecutive vectors x
i
(on the circle) is a basis for A
j
, we say A
j
.
(i) Show that for a xed , we have [/ [ k.
(ii) For a xed A
j
, count the number of bases B
n
such that
A
j
.
(iii) Prove that
m
k
n

_
n
k
_
q
.
332 A Course in Combinatorics
Notes.
It has been suggested that one should use the notation n
q
. for
M(n, q) (which is to be pronounced n q-torial). Then we have
_
n
k
_
q
=
n
q
.
k
q
. (n k)
q
.
.
Also note
lim
q1
n
q
. = n
1
. (that is, n!).
We do not know who rst proposed this.
The q-analogue of Ramseys theorem was a long outstanding
problem until it was solved by R. L. Graham, K. Leeb, and B.
L. Rothschild in 1972.
Carl Friedrich Gauss (17771855) was possibly the greatest math-
ematician (scientist) of all time. He made extremely important
contributions to number theory. He was the rst to study proper-
ties of the expressions that are now called Gaussian polynomials or
Gaussian numbers.
References.
P. Frankl and R. M. Wilson (1986), The Erd os-Ko-Rado theorem
for vector spaces, J. Combinatorial Theory (A) 43, 228236.
R. L. Graham, K. Leeb, B. L. Rothschild (1972), Ramseys theorem
for a class of categories, Adv. Math. 8, 417433.
R. L. Graham, B. L. Rothschild, and J. Spencer (1980), Ramsey
Theory, Wiley.
W. N. Hsieh (1975), Intersection theorems for systems of nite
vector spaces, Discrete Math. 12, 116.
S. Thomas (1986), Designs over nite elds, Geometriae Dedicata
24, 237242.
25
Lattices and Mobius inversion
One of the techniques belonging to the foundations of combinatorics
is the principle of M obius inversion over partially ordered sets. This
can be thought of as a generalization of inclusion-exclusion as well
as an extension of the inversion with the classical M obius function
of number theory, which was discussed in Chapter 10.
Let P be a nite partially ordered set. We will consider matrices
whose rows and columns are indexed by the elements of P, that
is mappings from P P to the rationals or complex numbers. The
incidence algebra A(P) consists of all matrices such that (x, y) =
0 unless x y in P. By the denition of matrix multiplication,
()(x, y) =

zP
(x, z)(z, y).
If , A(P), then the above sum need be extended over only
those z in the interval [x, y] := x z y. It is easily seen that
A(P) is closed under multiplication, as well as addition and scalar
multiplication.
An element of A(P) that will play an important role in the fol-
lowing theory is (the zeta function of P) which is dened by
(x, y) =
_
1 if x y in P,
0 otherwise.
We claim that is invertible and that its inverse, which is called
the Mobius function of P and is denoted by , is integral and lies
in A(P).
This is simple to verify. The equation = I (the identity)
334 A Course in Combinatorics
requires that
(25.1)

xzy
(x, z) =
_
1 if x = y,
0 otherwise,
and this can be ensured by simply dening inductively, by declar-
ing (x, x) := 1, (x, y) := 0 if x _ y, and
(25.2) (x, y) :=

xz<y
(x, z) for x < y in P.
For example, when P = 1, 2, 3, 4, 6, 12, the lattice of all (positive)
divisors of the integer 12,
=
_
_
_
_
_
_
_
1 1 1 1 1 1
0 1 0 1 1 1
0 0 1 0 1 1
0 0 0 1 0 1
0 0 0 0 1 1
0 0 0 0 0 1
_
_
_
_
_
_
_
, =
_
_
_
_
_
_
_
1 1 1 0 1 0
0 1 0 1 1 1
0 0 1 0 1 0
0 0 0 1 0 1
0 0 0 0 1 1
0 0 0 0 0 1
_
_
_
_
_
_
_
.
The top row of was computed as follows:
(1, 1) = +1,
(1, 2) = (1, 1) = 1,
(1, 3) = (1, 1) = 1,
(1, 4) = (1, 1) (1, 2) = 0,
(1, 6) = (1, 1) (1, 2) (1, 3) = +1,
(1, 12) = (1, 1) (1, 2) (1, 3) (1, 4) (1, 6) = 0.
We remark that a somewhat dierent equation concerning re-
sults if we consider instead the relation = I:
(25.3)

xzy
(z, y) =
_
1 if x = y,
0 otherwise.
Another much more complicated way of looking at the matter
of the invertibility of is as follows. We rst remark that a nite
25. Lattices and Mobius inversion 335
partial order can always be dominated by a total order; that is,
there exists an indexing P = x
1
, x
2
, . . . , x
n
such that x
i
x
j
in
P implies that i j. (Proof: Let x be any maximal element of P,
index Px by induction, and put x at the end.) With respect to
such an indexing of P, the matrices in A(P) are upper triangular.
The zeta function has 1s on the diagonal and so has determinant 1
and hence an integral inverse by Cramers formula. Then
1
=
lies in A(P) because the inverse of any matrix (or element of a
nitedimensional commutative algebra) is a polynomial in that
matrix (or element).
The equation (25.2) implies that (x, y) = 1 if x < y. Also note
that (x, y) = k 1 if the interval [x, y] is a k-point line, i.e. the
interval consists of x, y, and k pairwise noncomparable elements
z
1
, . . . , z
k
, each with x < z
i
< y.
In the following theorem, we list the values of the Mobius function
of some common partially ordered sets. The proof of Theorem 25.1
will not be given until the end of the chapter because we wish to
talk about the implications of M obius inversion rst. Clever and/or
industrious readers will be able to prove some parts of Theorem
25.1 with the recursion (25.2) and simple induction arguments, but
we prefer to wait until we have Weisners theorem, Theorem 25.3,
available. We do not prove part (v) heresee the notes.
Theorem 25.1. (i) For the lattice of all subsets of an n-set X,
(A, B) =
_
(1)
[B[[A[
if A B,
0 otherwise.
(ii) For the lattice of all (positive) divisors of an integer n,
(a, b) =
_
(1)
r
if
b
a
is the product of r distinct primes,
0 otherwise, i.e. if a [ b or
b
a
is not squarefree.
(That is, (a, b) = (
b
a
) where the latter is the classical function
of a single variable in (10.8).)
(iii) For the lattice of all subspaces of a nite-dimensional vector
space V over the eld F
q
of q elements,
(U, W) =
_
(1)
k
q
(
k
2
)
if U W and dim(W) dim(U) = k,
0 if U _ W.
336 A Course in Combinatorics
(iv) For the lattice
n
of all partitions of an n-set X,
(/, B) = (1)
[/[[B[

BB
(n
B
1)!
for elements / and B of
n
with / _ B, where n
B
denotes the
number of blocks of / that are contained in a block B of B.
(v) For the lattice of faces of a convex polytope,
(A, B) =
_
(1)
dim(B)dim(A)
if A B,
0 otherwise.
We now state the principle of Mobius inversion. Let P be a nite
partially ordered set and its Mobius function. Let f, g, h : P 1
(or into any additive group) be functions such that the relations
g(x) =

a:ax
f(a) and h(x) =

b:bx
f(b)
hold for all x P. Then, for all x P,
(25.4) f(x) =

a:ax
(a, x)g(a),
and
(25.5) f(x) =

b:bx
(x, b)h(b).
These are readily veried by direct substitution. For example, the
right-hand side of (25.4) is

a:ax
(a, x)
_

b:ba
f(b)
_
=

b:bx
f(b)
_

a:bax
(a, x)
_
.
But the inner sum on the right is zero by (25.3) unless b = x, so only
the term f(x) survives. Or, in matrix notation (when we regard
25. Lattices and Mobius inversion 337
f, g, and h as row or column vectorswhichever is appropriate
whose coordinates are indexed by P), the relations assumed for g
and h are equivalent to g = f, so f = g, and h = f, so f = h.
When the principle of M obius inversion is applied to the lattice of
subsets of a set X, we recover inclusion-exclusion. Let (A
i
: i I)
be a nite family of subsets of a nite set X. For J I, let f(J)
equal the number of elements of X that belong to exactly the sets
A
j
with j J and to no others. (Think of the size of one of the
regions in a Venn diagram of the sets.) Let g(J) equal the number
of elements in
jJ
A
j
. Then g(J) =

K:JK
f(K), so Mobius
inversion gives
f(J) =

K:JK
(1)
[KJ[
g(K).
In the case J = , the above can be written
[X
_
iI
A
i
[ =

KI
(1)
[K[
[

jK
A
j
[,
which is a cryptic way of stating inclusion-exclusion.
When the principle of M obius inversion is applied to the lattice of
(positive) divisors of an integer n, we recover the classical number-
theoretic Mobius inversion. If f and g satisfy
g(m) =

k[m
f(k) for all m dividing n,
then M obius inversion gives
f(m) =

k[m
(k, m)g(k) for all m dividing n,
and this, in view of Theorem 25.1(ii), is equivalent to Theorem
10.4.
We give a q-analogue of Example 10.2, where we used inclusion-
exclusion to nd the expression

m
k=0
(1)
mk
_
m
k
_
k
n
for the number
of surjections from an n-set to an m-set.
338 A Course in Combinatorics
Theorem 25.2. The number of surjective linear transformations
from an n-dimensional vector space to an m-dimensional vector
space V over F
q
is
m

k=0
(1)
mk
_
m
k
_
q
q
nk+
(
mk
2
)
.
Proof: For a subspace U V , let f(U) denote the number of
linear transformations whose image is U. Let g(U) denote the
number of linear transformations whose image is contained in U.
Clearly,
g(U) =

W:WU
f(W)
and g(U) is q
nr
if dim(U) = r. By M obius inversion on the lattice
of subspaces of V ,
f(U) =

W:WU
(W, U)q
ndim(W)
.
Take U = V and use Theorem 25.1(iii) to produce the stated result.

Corollary. The number of n m matrices over the eld F

q
that
have rank r is
_
m
r
_
q
r

k=0
(1)
rk
_
r
k
_
q
q
nk+
(
rk
2
)
.
We note that the number of injective linear transformations has
a relatively simple form. If we x a basis for an n-dimensional
vector space and consider injections into an m-dimensional vector
space, the image of the i-th basis vector must be chosen as one of
the (q
m
q
i1
) vectors not in the span of the images of the previous
basis vectors. In summary, there are (q
m
1)(q
m
q) (q
m
q
n1
)
injective linear transformations. Since Theorem 25.2 with m = n
also gives an expression for this number, we have proved an identity.
25. Lattices and Mobius inversion 339
Problem 25A. (i) Use Mobius inversion to derive an expression
for the number of k-subspaces that meet trivially a given r-subspace
of an n-dimensional vector space over F
q
. This should give a q-
analogue to Eq. (10.5). (ii) For the special case when r + k = n,
show, from another point of view, that there are exactly q
rk
such
subspaces by consideration of r n matrices over F
q
of the form
(I M), where I is an identity of order r.
Problem 25B. Use Mobius inversion to derive an expression for
the number of nonsingular linear mappings of an n-dimensional
vector space over F
q
to itself that x no vectors other than 0, i.e.
that do not have 1 as an eigenvalue. (These will be orthomorphisms
of the additive group; see the Remarks following Theorem 22.9.)
* * *
The partially ordered sets mentioned in the statement of Theo-
rem 25.1 are all lattices. The following useful theorem was found
by L. Weisner (1935).
Theorem 25.3. Let be the Mobius function of a nite lattice L
and let a L with a > 0
L
. Then

x:xa=1
L
(0
L
, x) = 0.
Proof: Fix a and consider
S :=

x,yL
(0, x)(x, y)(a, y)(y, 1) =

xL

yx,
ya
(0, x)(y, 1).
Now on the one hand,
S =

x
(0, x)

yx
ya
(y, 1);
but y a and y x if and only if y x a, and the inner sum is

yxa
(y, 1) =
_
1 if x a = 1,
0 if x a < 1.
340 A Course in Combinatorics
Thus S is the sum in the statement of the theorem. On the other
hand,
S =

ya
(y, 1)

0xy
(0, x),
and the inner sum is always 0 since y > 0.
Corollary. For elements x, y of a geometric lattice L with x y,
(x, y) has sign (1)
rank(y)rank(x)
and, in particular, is never 0.
Proof: We show that (0
L
, 1
L
) has sign (1)
rank(L)
by induction
on the rank of L. Pick a point p L. By semimodularity, ap = 1
L
if and only if a = 1
L
or a is a copoint not on p; so Weisners theorem,
Theorem 25.3, gives
(25.6) (0
L
, 1
L
) =

h:h<1
L
,h_p
(0
L
, h).
Since all terms on the right-hand side have sign (1)
rank(L)1
by the
induction hypothesis, the proof is complete.
The number of proper colorings
G
(x) of a graph G with x colors
can be found by M obius inversion on the lattice L(G) introduced
in Chapter 23, although this is not necessarily a very practical
method. Recall that the elements of L(G) are the partitions / of
the vertex set of G, all of whose blocks induce connected subgraphs
of G. For / in L(G), let g(/) denote the number of mappings
from the vertex set of G into a set of x colors (i.e. colorings) so
that all vertices in each block of / receive the same color. Clearly,
g(/) = x
[/[
. Let f(/) denote the number of mappings that are
constant on each block of /, but such that the endpoints of edges
joining distinct blocks of / receive dierent colors. Given a coloring
counted by g(/), a moments thought shows that there is a unique
coarser partition B so that the coloring is counted by f(B) (one
must coalesce two blocks of / of the same color if an edge joins
them). Thus, g(/) =

B_/
f(B), so Mobius inversion gives
f(/) =

B_/
(/, B)g(B).
25. Lattices and Mobius inversion 341
The number of proper colorings is f evaluated on 0
L(G)
(the parti-
tion into singletons), and this is

G
(x) =

B
(0
L(G)
, B)x
[B[
=
n

k=1
_
_

[B[=k
(0
L(G)
, B)
_
_
x
k
.
The polynomial
G
(x) is called the chromatic polynomial of the
graph. The corollary above leads directly to the following state-
ment.
Theorem 25.4. The number of proper colorings of a graph G on n
vertices in x colors is given by a monic polynomial
G
(x) of degree
n whose coecients alternate in sign.
Problem 25C. Let G be a simple graph with n vertices and m
edges. Show that the coecient of x
n1
in
G
(x) is m, and that
the coecient of x
n2
is m(m1)/2 minus the number of triangles
in G.
T. Dowling and R. M. Wilson (1975) proved the following the-
orem and its corollary, the latter being a generalization of the in-
equality of Theorem 19.1 on linear spaces.
Theorem 25.5. If L is a nite lattice so that (x, 1
L
) ,= 0 for all
x L, then there exists a permutation of the elements of L so
that x (x) = 1
L
for all x L.
Proof: Before we begin the proof, we remark that an example
of a lattice that does not satisfy our hypothesis is a chain with
more than two elements. And, of course, there is no permutation
with the property stated above. On the other hand, the lattice
of all subsets of an n-set admits a unique permutation with the
property above, namely the permutation that takes each subset to
its complement.
All matrices here will have rows and columns indexed by L. Let
(x, y) :=
_
1 if x y = 1
L
,
0 otherwise.
342 A Course in Combinatorics
Let
1
be the diagonal matrix with
1
(x, x) := (x, 1
L
). Then

= because

(x, y) =

a,b
(x, a)
1
(a, b)(y, b)
=

a:ax and ay

1
(a, a)
=

a:axy
(a, 1
L
)
and the last sum is, by (25.1), 1 if x y = 1
L
, and 0 otherwise.
Under our hypothesis that (x, 1
L
) ,= 0, the matrix
1
is nonsin-
gular. Since is nonsingular too, we conclude that is also. Hence
some term in its determinant expansion does not vanish, and this
implies the conclusion of the theorem.
Corollary. In a nite geometric lattice of rank n, the number of
elements of rank nk is at least the number of elements of rank
k, 0 k n.
Proof: Consider a permutation as in Theorem 25.5 (which ap-
plies because of the Corollary to Theorem 25.3). The semimodular
law
rank(x) + rank((x)) rank(x (x)) + rank(x (x)) n
implies that the image of an element of rank k is one of rank
n k.
We give a similar matrix proof of a theorem of T. Dowling (1977)
on complementing permutations below.
Theorem 25.6. If L is a nite lattice such that (x, 1
L
) ,= 0 and
(0
L
, x) ,= 0 for all x L, then there exists a permutation of the
elements of L so that
x (x) = 1
L
and x (x) = 0
L
for all x L.
Proof: Let
1
be as in the proof of the previous theorem and let

0
be the diagonal matrix with
0
(x, x) := (0
L
, x). Now consider
25. Lattices and Mobius inversion 343
:=
1

0
. Our hypotheses imply that is nonsingular. We
claim that (x, y) = 0 unless x and y are complements. Then any
permutation corresponding to a nonvanishing term in the determi-
nant expansion of will be a complementing permutation.
To establish our claim, rst note that =
0
, so
(x, y) =

z:zx=1
L
,zy
(0
L
, z).
If this sum is not zero, then there exists some z with z x = 1
L
and z y, which implies that y x = 1
L
. By duality,
/
=

0
,
where

/
(x, y) :=
_
1 if x y = 0
L
,
0 otherwise.
Now note that =
1

/
and, similarly, (x, y) ,= 0 will imply that
x y = 0
L
.
Finally, we give the proof we promised at the beginning of the
chapter.
Proof of Theorem 25.1: (i) Since the interval [A, B] is isomor-
phic to the lattice of subsets of BA, it suces to show (, C) =
(1)
[C[
. We use the equation (25.6) and proceed by induction on
[C[. Let p be a point of C. There is only one copoint not on p,
namely the complement of p. So (25.6) implies that
(, C) = (, Cp) = (1)
([C[1)
= (1)
[C[
.
(ii) Again, it will suce to calculate (1, m), and we use induc-
tion on m. Let p be a prime divisor of m. Weisners theorem
asserts
(1, m) =

lcm(a,p)=m,a<m
(0, a).
If p
2
divides m, then the sum on the right is empty, and thus
(1, m) = 0. If p
2
does not divide m, then there is one term,
namely when a = m/p.
344 A Course in Combinatorics
(iii) It will suce to show that (0, V ) = (1)
k
q
(
k
2
)
for V of
dimension k. We proceed by induction on k. Let P be a 1-
dimensional subspace of V . By Weisners theorem,
(0, V ) =

U:UP=V,U,=V
(0, U).
By the induction hypothesis, each term (0, U) in the above sum
is equal to (1)
k1
q
(
k1
2
)
. The only subspaces U other than V such
that U P = V are those U of dimension k 1 that do not contain
P; their number is
_
k
1
_
q

_
k 1
1
_
q
= q
k1
.
This completes the proof of part (iii).
(iv) Consider two partitions / and B with / _ B. Say B has k
blocks that, when numbered, are the unions of n
1
, n
2
, . . . , n
k
blocks
of /, respectively. Then the interval [/, B] is isomorphic to the
direct product of partition lattices

n
1

n
2

n
k
,
since a partition ( with / _ ( _ B is specied by a partition of the
n
i
blocks of / that lie in the i-th block of B, for each i = 1, 2, . . . , k.
We now mention the fact that the M obius function
PQ
of the
direct product P Q of two partially ordered sets is the (Kronecker
or tensor) product of the M obius functions
P
and
Q
of P and Q;
that is,

PQ
((a
1
, b
1
), (a
2
, b
2
)) =
P
(a
1
, a
2
)
Q
(b
1
, b
2
).
(We leave the reader to think about this; just check that dening

PQ
as above produces an inverse of
PQ
.) Thus (/, B) is the
product of (0

n
i
, 1

n
i
) for i = 1, 2, . . . , k.
We now show that (0

n
, 1

n
) = (1)
n1
(n 1)! by induction
on n. Let T be a point of
n
, i.e. a partition of an n-set into a 2-set
25. Lattices and Mobius inversion 345
x, y and n 2 singletons. The partitions / so that T / = 1

n
are the 2
n2
partitions with two blocks that separate x and y; there
are
_
n2
i
_
such partitions where the block containing x has size i +1
and the block containing y has size n 1 i. From (25.6) and the
induction hypothesis,
(0

n
, 1

n
) =
n2

i=0
_
n 2
i
_
(1)
i
(i)!(1)
n2i
(n 2 i)!
= (1)
n1
(n 1)!.

Problem 25D. The dual of Weisners theorem (derived by apply-

ing Theorem 25.3 to the dual lattice) asserts

x:xa=0
L
(x, 1
L
) = 0 for all a L with a < 1
L
.
Take a to be a partition with one block of size n1 and one of size
1 (in the partition lattice) and use the above to give a somewhat
dierent proof of (0

n
, 1

n
) = (1)
n1
(n 1)!.
We use Mobius inversion over the lattice of partitions to get an
expression for the number of connected labeled simple graphs on
n vertices. Let X be an n-set. To each graph G with vertex set
X, there corresponds the partition (
G
of X whose blocks are the
vertex sets of the connected components of G. We let g(B) denote
the number of simple graphs G with V (G) = X and (
G
= B. We
let f(B) denote the number of simple graphs G with V (G) = X
and where (
G
is a renement of B. Clearly,
f(B) =

/_B
g(/).
We are interested in g(1

n
), where 1

n
is the partition with one
block. But what is easy to calculate is f(B); we get a graph counted
by f(B) by arbitrarily choosing a simple graph on each block of B,
so if B has k
i
blocks of size i, then
f(B) = 2
k
2(
2
2
)
2
k
3(
3
2
)
2
k
n(
n
2
)
.
346 A Course in Combinatorics
By Mobius inversion,
g(B) =

/_B
(/, B)f(/).
Specializing to B = 1

n
, we have
(25.7) g(1

n
) =

/
(/, 1

n
)f(/).
Recall that the number of partitions of X of type (k
1
, k
2
, . . . , k
n
) is
n!
(1!)
k
1
k
1
!(2!)
k
2
k
2
! (n!)
k
n
k
n
!
;
cf. (13.3). And using Theorem 25.1(iv), we nd the number of
connected labeled graphs on n vertices to be

(1)
k
1
+k
2
++k
n
1
n! (k
1
+ +k
n
1)!
(1!)
k
1
k
1
! (n!)
k
n
k
n
!
2
k
2(
2
2
)
+k
3(
3
2
)
++k
n(
n
2
)
where the summation is extended over all partitions (k
1
, k
2
, . . . , k
n
)
of the integer n, that is all nonnegative n-tuples with 1k
1
+ 2k
2
+
+nk
n
= n.
We do not mean to imply that this is a particularly nice answer
to our original question. Indeed, it is a summation over a large
collection of objects, whose number exceeds e

n
for large n. But it
is a natural illustration of M obius inversion and does greatly reduce
the problem. For n = 5, we have
partition of 5 number of / f(/) (/, 1

5
)
5 1 1024 1
41 5 64 1
32 10 16 1
311 10 8 2
221 15 4 2
2111 10 2 6
11111 1 1 24
25. Lattices and Mobius inversion 347
from which table and (25.7) we see that 728 out of the 1024 simple
labeled graphs on 5 vertices are connected.
* * *
We give an application of M obius inversion to coding theory.
Recall the denition of MDS codes given in Example 20.3. Let C
be such a code, i. e. an [n, k, d] code over F
q
with d = n k + 1.
If we consider a set of d positions and then look at the subcode
of C with zeros in all other positions, this subcode has dimension
k (n d) = 1. Since this subcode has minimum distance d, it
must have dimension exactly 1 by Theorem 20.2. It follows that for
n d
/
> d, specifying a set of d
/
positions and requiring codewords
to be zero in all other positions, will dene a subcode of C with
dimension d
/
d+1. We use this in the proof below. We shall show
that the weight enumerator of an MDS code is determined by its
parameters.
Theorem 25.7. Let C be an [n, k] code over F
q
with distance
d = n k + 1. Then the weight enumerator of C is 1 +

n
i=d
A
i
z
i
,
where
A
i
=
_
n
i
_
(q 1)
id

j=0
(1)
j
_
i 1
j
_
q
ijd
, i = d, d + 1, . . . , n.
Proof: Let R be a subset of N := 0, 1, . . . , n. Dene f(R) to
be the number of codewords (c
0
, c
1
, . . . , c
n1
) for which i : c
i
,=
0 = R. For a subset S of N, we dene g(S) :=

RS
f(R). As
remarked above, we have
g(S) =
_
1, if [S[ d 1,
q
[S[d+1
, if n [S[ d.
The denition of f implies that A
i
=

RN,[R[=i
f(R). We now
348 A Course in Combinatorics
apply M obius inversion with Theorem 25.1 (i). We nd
A
i
=

RN,[R[=i

SR
(S, R)g(S)
=
_
n
i
_
_
_
_
d1

j=o
_
i
j
_
(1)
ij
+
i

j=d
_
i
j
_
(1)
ij
q
jd+1
_
_
_
=
_
n
i
_
i

j=d
_
i
j
_
(1)
ij
(q
jd+1
1).
The result follows if we replace j by i j and then use
_
i
j
_
=
_
i1
j1
_
+
_
i1
j
_
.
Theorem 25.7 implies rather severe restrictions on the alphabet
size of an MDS code.
Corollary. If there exists an MDS code C over F
q
with length n
and dimension k, then
_
(i) q n k + 1 or k 1,
(ii) q k + 1 or d = n k + 1 2.
Proof: (i) Let d = n k + 1. From Theorem 25.7 we nd for
d < n that
0 A
d+1
=
_
n
d + 1
_
(q 1)(q d).
(ii) Let G := (I
k
P) be the generator matrix of C. Since C has
minimum weight d, every set of d1 = nk columns of the parity
check matrix H := (P

I
nk
) is linearly independent. Hence every
square submatrix of H is nonsingular. So, no codeword of C

has
n k zeros, i.e. C

is also MDS. Apply the result of (i) to C

.
Problem 25E. Let P be a partially ordered set. Remember that
if x < y, then a sequence x = x
0
< x
1
< < x
k
= y is called a
chain of length k from x to y. Let c
k
(x, y) denote the number of
such chains (so c
1
(x, y) = 1). Prove that
(x, y) =

k1
(1)
k
c
k
(x, y).
25. Lattices and Mobius inversion 349
Problem 25F. We consider the lattice of subspaces of the vector
space V := F
n
q
. Let S be another vector space over F
q
. Dene for
a subspace U:
f(U) := the number of linear maps from V to S with kernel U,
g(U) := the number of linear maps from V to U with kernel U.
(i) Determine g(U) and then apply M obius inversion to
g(U) =

W:WU
f(W).
(ii) Show that
f( 0) =
n

k=0
_
n
k
_
q
(1)
k
q
(
k
2
)
[S[
nk
.
(iii) Prove the polynomial identity
n1

k=0
(x q
k
) =
n

k=0
(1)
k
_
n
k
_
q
q
(
k
2
)
x
nk
.
Notes.
The expression of the chromatic polynomial of a graph in terms
of the M obius function of L(G) is due to G.-C. Rota (1964).
Theorem 25.1 (v) is essentially Eulers formula for polytopes:
f
0
f
1
+f
2
f
3
+ + (1)
n
f
n
= 0,
where f
i
is the number of faces of rank i, or dimension i 1 (f
0
=
f
n
= 1). See B. Gr unbaum (1967). Also see R. Stanley (1986),
where posets with the property that (x, y) = (1)
d
whenever the
ranks of x and y dier by d are called Eulerian posets.
References.
T. Dowling (1977), A note on complementing permutations, J.
Combinatorial Theory (B) 23, 223226.
350 A Course in Combinatorics
T. Dowling and R. M. Wilson (1975), Whitney number inequalities
for geometric lattices, Proc. Amer. Math. Soc. 47, 504512.
B. Gr unbaum (1967), Convex Polytopes, J. Wiley (Interscience).
G.-C. Rota (1964), On the foundations of combinatorial theory
I. Theory of M obius functions, Z. Wahrscheinlichkeitstheorie 2,
340368.
R. P. Stanley (1986), Enumerative Combinatorics, Vol. 1, Wads-
worth.
L. Weisner (1935), Abstract theory of inversion of nite series,
Trans. Amer. Math. Soc. 38, 474484.
26
Combinatorial designs and
projective geometries
Geometries over nite elds are a rich source of combinatorial de-
signs and related combinatorial congurations. We begin with two
topics (arcs and subplanes) that we choose to mention only in pro-
jective planes before going on to discuss quadrics and other cong-
urations in general projective spaces.
An (m, k)-arc in a projective plane is a set of m points, no k+1 of
which are collinear. We were concerned with (m, 2)-arcs in Problem
19I.
Let A be an (m, k)-arc in a projective plane of order n and let x
be a point in A. The n + 1 lines on x each contain at most k 1
other points of A, so
m 1 + (n + 1)(k 1).
An (m, k)-arc A is called perfect when equality holds above. Any
line that contains a point of a perfect (m, k)-arc evidently contains
exactly k points of the arc; that is,
[L A[ = 0 or k
for any line L. Clearly, the nonempty intersections of lines with a
perfect (m, k)-arc provide the blocks of a Steiner system S(2, k, m).
A single point is a perfect (1,1)-arc. The set of n
2
points not on
a xed line of a projective plane of order n is a perfect (n
2
, n)-arc
and the corresponding Steiner system is an ane plane of order n.
The hyperovals in Problem 19I are perfect (q +2, 2)-arcs. The cor-
responding designs are trivial. But these have dual arcs where the
corresponding Steiner systems are interestingsee Problem 26A.
352 A Course in Combinatorics
The construction of perfect (m, k)-arcs in Desarguesian planes
of even order given in part (2) of the following theorem is due to
R. H. F. Denniston (1969). No examples of perfect (m, k)-arcs,
1 < k < n, in projective planes of odd orders n are known at this
time.
Theorem 26.1.
(1) If there exists a perfect (m, k)-arc in a projective plane of
order n, then k divides n.
(2) If q is a power of 2 and k divides q, then there exists a
perfect (m, k)-arc in PG
2
(q).
Proof: Let x be a point not in a perfect (m, k)-arc A in a pro-
jective plane of order n. The lines on x partition the remaining
points, but each line on x contains exactly 0 or k points of A. So
k must divide m = 1 + (n + 1)(k 1). It follows that k divides n.
Now let q be a power of 2 and k a divisor of q. Let f(x, y) =
x
2
+ xy + y
2
be any irreducible quadratic over F
q
, and let H
be any subgroup of the additive group of F
q
with order k. In the
ane plane AG(2, q), let
A := (x, y) : f(x, y) H.
We claim that any (ane) line meets A in 0 or k points. When the
ane plane is embedded in a PG
2
(q), A will be a perfect (m, k)-arc.
Consider a line L = (x, y) : y = mx + b with b and m both
nonzero. (We leave consideration of lines of the forms (x, y) : y =
mx and (x, y) : x = c to the reader.)
The intersection L A is the set of points (x, mx +b), where
x
2
+x(mx +b) +(mx +b)
2
H, or
F(x) H, where F(x) := ( +m+m
2
)x
2
+bx +b
2
.
We are working over a eld of characteristic 2, so
x ( +m+m
2
)x
2
+bx
is a linear mapping; it has kernel of order 2 since the irreducibility
of f(x, y) ensures that both and the coecient of x
2
are nonzero.
26. Designs and projective geometries 353
The image K
F
of this mapping is thus a subgroup of order q/2 of
the additive group of F
q
. The image of F is a coset of K
F
, but since
F(x) is never 0 (again by the irreducibility of f(x, y)), the image
of F is the complement F
q
K
F
. In summary,
[x : F(x) = a[ =
_
2 if x / K
F
,
0 if x K
F
.
Thus [LA[ = 2[H(F
q
K
F
)[. The subgroup H is either contained
in K
F
, in which case L A = , or intersects K
F
in a subgroup of
order k/2, in which case [L A[ = k.
As a corollary of this theorem (with n = 2
m+1
, k = 2
m
), we
obtain Steiner systems
S(2, 2
m
, 2
2m+1
2
m
).
Steiner systems S(2, k, v) are in many ways most interesting when v
is close to k
2
, as is the case for ane and projective planes (Fishers
inequality, Theorem 19.6, shows v k
2
k + 1). These examples
have v < 2k
2
, which is still impressive.
Problem 26A. Let A be a perfect (m, k)-arc in a projective plane
P of order n, 1 k n. Let A

be the set of lines that do not

meet A. Show that A

is a perfect (m

, [n/k])-arc in the dual plane

P

, and calculate m

in terms of m, k, n.
Problem 26B. A parallel class in a Steiner system S(2, k, m) is
a set / of m/k blocks so that each point is incident with exactly
one block in /. An S(2, k, m) is resolvable when the blocks can be
partitioned into parallel classes (cf. Problem 19K).
Let A be a perfect (m, k)-arc in a projective plane P of order
n. Explain why the Steiner system S(2, k, m) whose blocks are the
nontrivial intersections of lines with A is resolvable.
A subplane S of a projective plane P is a substructure of P that
is a projective plane in its own right. Recall that a substructure of
an incidence structure (T, B, I) is an incidence structure (T
0
, B
0
, I
0
)
where T
0
T, B
0
B, and I
0
= I (T
0
B
0
). Note that given
an automorphism (or collineation) of a projective plane P, the
354 A Course in Combinatorics
substructure S consisting of the points of P xed by and the lines
of P xed by has the property that two points of S are incident
with a unique line of S, and two lines of S are incident with a unique
point of S. If S contains four points, no three collinear, then S is
a subplane; but it may be a nearpencil or have one or no lines.
Example 26.1. Let V := F
3
q
n. The points and lines of PG
2
(q
n
) are
the 1-dimensional and 2-dimensional F
q
n -subspaces of V . Dene a
substructure S of PG
2
(q
n
) whose points and lines are those 1- and
2-dimensional F
q
n -subspaces of V that admit bases consisting of
vectors whose entries are in the subeld F
q
. Then S is a subplane.
A line of S is incident with q
n
+1 points in PG
2
(q
n
), but only q +1
of those are points of S.
Theorem 26.2. If a projective plane P of order n contains a sub-
plane S of order m < n, then either
(i) n = m
2
, or
(ii) n m
2
+m.
Proof: Let L be a line of the subplane S and x a point on L in
P but not in S. The n other lines on x can contain at most one
point of S (since a line M in P containing at least two points of
S would necessarily belong to the subplane, and then the point x
in common to M and L would also belong to the subplane). The
line L contains m+1 points of S, which has a total of m
2
+m+1
points, each of which is on some line through x, so m
2
n.
Equality will imply that every line of P will meet S (evidently
in one or m + 1 points) since if there were a line N disjoint from
S, and x were taken as the point of intersection of L and N, each
of the m
2
points of S not on L would belong to one of n 1 lines.
Now assume m
2
< n, so that there does exist a line N incident
with no points of S. Each of the m
2
+ m + 1 lines of S contains
one point of N and no two such lines can contain the same point,
so m
2
+m+ 1 n + 1.
Problem 26C. Show that if PG
2
(F) contains the Fano congu-
ration PG
2
(2), then F has characteristic 2. Suggestion: Without
loss of generality, four points of the Fano conguration are 1, 0, 0),
0, 1, 0), 0, 0, 1), and 1, 1, 1). Calculate the homogeneous coordi-
nates of the other three points (that must lie on a line).
26. Designs and projective geometries 355
Problem 26D. A blocking set in a projective plane is a set S of
points that contains no line but such that every line meets S in at
least one point. Show that a blocking set in a projective plane of
order n contains at least n+

n+1 points, and that equality holds

if and only if S is the set of points of a Baer subplane.
We illustrate a polynomial method with the proof of the follow-
ing theorem on a type of ane blocking sets, due to Brouwer and
Schrijver (1978).
Theorem 26.3. If V is a subset of AG(k, q) that meets all the
hyperplanes, then
[V [ k(q 1) + 1.
Proof: Let A be the blocking set. Without loss of generality
0 A. Let B := A0. Then B intersects all hyperplanes that do
not contain 0. These are dened by equations w
1
x
1
+ +w
k
x
k
= 1,
where w runs through all non-zero vectors. Therefore
F(x
1
, x
2
, . . . , x
k
) :=

bB
(b
1
x
1
+b
2
x
2
+ +b
k
x
k
1)
is identically 0 on the space except in 0.
It is easily shown by induction that a polynomial that is identi-
cally 0 on the space must be in the ideal generated by the polyno-
mials x
q
i
x
i
, i = 1, . . . , k. Write F(x) as
F(x
1
, . . . , x
k
) =
k

i=1
F
i
(x
1
, . . . , x
k
)(x
q
i
x
i
) +G(x
1
, . . . , x
k
),
where the highest degree of x
i
in G is at most q 1. For each i,
the polynomial x
i
Fx
1
, . . . , x
k
) is identically 0, so this is also true
for x
i
G(x
1
, . . . , x
k
). Therefore G is divisible by (x
q1
i
1). Since
F(0) ,= 0, also G(0) ,= 0 and therefore G has degree k(q 1). So,
the degree of F, which is [B[, must be at least k(q 1).
A quadratic form in indeterminates x
1
, x
2
, . . . , x
n
over a eld F
is a homogeneous polynomial of degree 2 in those indeterminates,
i.e.
(26.1) f(x) = f(x
1
, x
2
, . . . , x
n
) =
n

i,j=1
c
ij
x
i
x
j
356 A Course in Combinatorics
where the coecients c
ij
are in F. The same quadratic form is
dened by more than one matrix C = (c
ij
) of coecients, since it
is only the diagonal entries and the sums c
ij
+ c
ji
that are impor-
tant. We could require that c
ij
= 0 for i > j and then quadratic
forms in x
1
, x
2
, . . . , x
n
are in one-to-one correspondence with upper
triangular n n matrices C over F.
A quadric in PG
n
(F) is a set of projective points
Q = Q(f) := x) : f(x) = 0
where f is a quadratic form in n + 1 indeterminates. We must
choose a basis and identify the vectors with (n + 1)-tuples over F
in order to make this denition.
Two quadratic forms f and g are projectively equivalent when
g can be obtained from f by an invertible linear substitution,
or in matrix notation, g(x) = f(xA) for some nonsingular n n
matrix A over F. So if f is given by a matrix C as in (26.1), i.e.
f(x) = xCx

, then g is given by the matrix ACA

. For example,
for any positive integer n, nx
2
1
+ nx
2
2
+ nx
2
3
+ nx
2
4
is projectively
equivalent to x
2
1
+ x
2
2
+ x
2
3
+ x
2
4
over the rationals; see equation
(19.12).
We remark that the matrix notation does not lend itself well to
changes of the names or number of the indeterminates, which we
wish to allow. For example, replacing x
1
in a quadratic form f
by y
1
+ y
2
, where y
1
and y
2
are new indeterminates, produces a
projectively equivalent form.
The rank of a quadratic form is the least number of indeter-
minates that occur (with nonzero coecients) in any projectively
equivalent quadratic form. For example, (x
1
+ + x
n
)
2
has rank
1. Projectively equivalent quadratic forms have the same rank.
A quadratic form in r indeterminates is said to be nondegenerate
when it has rank r.
Example 26.2. Consider quadratic forms in two indeterminates
f(x, y) = ax
2
+bxy +cy
2
.
The form has rank 0 if and only if a = b = c = 0. The corresponding
quadric contains all points of PG
1
(F) (the projective line). If not
26. Designs and projective geometries 357
zero, it will have rank 1 if and only if it is a scalar multiple of the
square of a linear form dx +ey, which is the case if and only if the
discriminant b
2
4ac is 0. The corresponding quadric in PG
1
(F)
will then consist of a single point. A rank 2 quadratic form in
two indeterminates is either irreducible or can be factored into two
distinct linear forms. In the rst case, the corresponding quadric
in PG
1
(F) will be empty, and in the second case, the corresponding
quadric in PG
1
(F) will consist of two points.
It is clear that an irreducible quadratic form of rank 2 is not pro-
jectively equivalent to a reducible quadratic form, since the latter
has zeros in F and the former does not. A reducible quadratic form
is projectively equivalent to x
1
x
2
.
Problem 26E. Show that over a eld F of odd characteristic, a
quadratic form f as in (26.1) is degenerate if and only the sym-
metric matrix C + C

is singular. For a eld F of characteristic

2, show that f is degenerate if and only if C + C

is singular and
x(C +C

) = 0 for some x with f(x) = 0.

It is important to understand that the intersection of a quadric
with a at U of PG
n
(F) is a quadric in that at. For example,
suppose that U is a projective line PG
1
(F) and that f is a quadratic
form in x
0
, x
1
, . . . , x
n
. The homogeneous coordinates of points of
U are ya + zb) where a = (a
0
, a
1
, . . . , a
n
) and b = (b
0
, b
1
, . . . , b
n
)
are linearly independent vectors. These points are in one-to-one
correspondence with the homogeneous coordinates (y, z)) of points
of PG
1
(F). Say f =

0ijn
c
ij
x
i
x
j
. Then
g(y, z) :=

1ijn
c
ij
(ya
i
+zb
i
)(ya
j
+zb
j
)
is a quadratic form in y and z and denes a quadric in PG
1
(F).
The quadratic form g may be degenerate even if f is not. In view
of Example 26.2, we see that a line is either contained fully in a
quadric Q, or meets Q in 0, 1, or 2 points.
Lemma 26.4. Any quadratic form f of rank n 3 is projectively
equivalent to
(26.2) x
1
x
2
+g(x
3
, . . . , x
n
)
358 A Course in Combinatorics
for some quadratic form g(x
3
, . . . , x
n
).
Proof: First assume q is odd. It is easy to see that a quadratic
form f is projectively equivalent to some

1ijn
c
ij
x
i
x
j
where
c
11
,= 0. Let y := x
1
+
1
2c
11
(c
12
x
2
+ + c
1n
x
n
). Then f = c
11
y
2
+
g(x
2
, . . . , x
n
). Inductively, if the rank of f is at least 3, we nd that
f is projectively equivalent to
h(x) = ax
2
1
+bx
2
2
+cx
2
3
+g
/
(x
4
, . . . , x
n
)
with a, b, c all nonzero. The three scalars can be changed by any
nonzero square factor and a projectively equivalent quadratic form
results; some pair must dier by a square factor, so we can assume
that b = c, say.
We claim that there exist s, t F
q
so that s
2
+ t
2
= b
1
a.
One way to see this is to consider the addition table of F
q
, which
is a Latin square. The number of squares in F
q
, including 0, is
(q + 1)/2. The element b
1
a (or any other) occurs (q 1)/2
times in the (q 1)/2 columns of the Latin square indexed by the
nonsquares, and (q 1)/2 times in the (q 1)/2 rows indexed by
the nonsquares; so b
1
a must occur at least once in the submatrix
with rows and columns indexed by the squares. With s and t so
chosen, h is projectively equivalent to
ax
2
1
+b(sx
2
+tx
3
)
2
+b(tx
2
sx
3
)
2
+g
/
(x
4
, . . . , x
n
)
= ax
2
1
ax
2
2
ax
2
3
+g
/
(x
4
, . . . , x
n
)
= (ax
1
+ax
2
)(x
1
x
2
) ax
2
3
+g
/
(x
4
, . . . , x
n
),
and the latter is clearly projectively equivalent to (26.2).
The case q even is a little more tedious; see the proof of Theorem
5.1.7 in J. W. P. Hirschfeld (1979).
Theorem 26.5.
(i) Any quadratic form f of odd rank n is projectively equivalent
to
(26.3) f
0
(x) := x
1
x
2
+ +x
n2
x
n1
+cx
2
n
for some scalar c.
26. Designs and projective geometries 359
(ii) Any quadratic form f of even rank n is projectively equivalent
to either
(26.4) f
1
(x) := x
1
x
2
+ +x
n3
x
n2
+x
n1
x
n
or
(26.5) f
2
(x) := x
1
x
2
+ +x
n3
x
n2
+p(x
n1
, x
n
)
where p(x
n1
, x
n
) is an irreducible quadratic form in two indeter-
minates.
Proof: This follows from Lemma 26.3 and induction.
Quadratic forms (and the corresponding quadrics) of odd rank
are called parabolic. Quadratic forms of even rank projectively
equivalent to (26.4) are called hyperbolic, and those equivalent to
(26.5) are called elliptic. Any two hyperbolic quadratic forms of a
given rank are projectively equivalent to (26.4) and hence to each
other. It is also true that all parabolic quadratic forms of a given
even rank are projectively equivalent, i.e. we may take c = 1 in
(26.3), and that all elliptic quadratic forms of a given rank are pro-
jectively equivalent. See J. W. P. Hirschfeld (1979) for this and
more on canonical forms. That hyperbolic and elliptic quadratic
forms are not projectively equivalent is a consequence of the fol-
lowing theorem.
Theorem 26.6. A nondegenerate quadric Q in PG
n
(q) has cardi-
nality
_

_
q
n
1
q1
if n is even, i.e. Q is parabolic,
(q
(n+1)/2
1)(q
(n1)/2
+1)
q1
if n is odd and Q is hyperbolic,
(q
(n+1)/2
+1)(q
(n1)/2
1)
q1
if n is odd and Q is elliptic.
Proof: In general, if
f(x
1
, . . . , x
r
) = x
1
x
2
+g(x
3
, . . . , x
r
)
and there are N vectors (x
3
, . . . , x
r
) such that g(x
3
, . . . , x
r
) = 0,
then there are (2q 1)N + (q 1)(q
r2
N) vectors (x
1
, . . . , x
r
)
360 A Course in Combinatorics
such that f(x
1
, . . . , x
r
) = 0. This allows us to verify by induction
the following formulae for the number of zeros in F
r
q
of a rank r
quadratic form f in r indeterminates as the following:
_
_
_
q
r1
if r is odd, i.e. f is parabolic,
q
r1
+q
r/2
q
r/21
if r is even and f is hyperbolic,
q
r1
q
r/2
+q
r/21
if r is even and f is elliptic.
Of course, the number of projective points in PG
n
(q) on the cor-
responding quadrics is found from the above numbers by replacing
r by n + 1, subtracting 1 (the zero vector) and then dividing by
q 1.
Theorem 26.7. Let Q be a nondegenerate quadric in PG
n
(q).
The maximum projective dimension of a at F with F Q is
_
_
_
n/2 1 if n is even, i.e. Q is parabolic,
(n 1)/2 if n is odd and Q is hyperbolic,
(n 3)/2 if n is odd and Q is elliptic.
Proof: Let f be a nondegenerate quadratic form in r indetermi-
nates. The statement of the theorem is equivalent to the statement
that the maximum dimension of a subspace U of F
r
q
such that f
vanishes on U is
_
_
_
(r 1)/2 if r is odd, i.e. f is parabolic,
r/2 if r is even and f is hyperbolic,
r/2 1 if r is even and f is elliptic.
First note that if f = f
0
in (26.3), then f(x) = 0 for any
x span(e
2
, e
4
, . . . , e
r1
), where e
1
, e
2
, . . . , e
r
is the standard ba-
sis for F
r
q
. If f = f
1
in (26.4), then f(x) = 0 for any x
span(e
2
, e
4
, . . . , e
r
). If f = f
2
in (26.5), then f(x) = 0 for any x
span(e
2
, e
4
, . . . , e
r2
). These subspaces have dimensions (r 1)/2,
r/2, and r/2 1, respectively.
It remains to show that f cannot vanish on any subspace of
larger dimension in any of these cases. We will use Theorem 26.5
and induction. The cases r = 1 and r = 2 are trivial. Suppose
26. Designs and projective geometries 361
f(x
1
, . . . , x
r
) = x
1
x
2
+ g(x
3
, . . . , x
r
) and that f(x) = 0 for all x
in some subspace U F
r
q
of dimension k. To complete the proof,
we will show that there exists a subspace U
/
F
r2
q
of dimension
k 1 so that g(y) = 0 for all y U
/
.
Clearly, g(y) = 0 for all y in
U
0
:= (x
3
, . . . , x
r
) : (0, 0, x
3
, . . . , x
r
) U.
We are done if dim(U
0
) k 1, so assume dim(U
0
) = k 2. Then
there exist vectors
(1, 0, a
3
, . . . , a
r
) and (0, 1, b
3
, . . . , b
r
)
in U. Then (1, 1, a
3
+b
3
, . . . , a
r
+b
r
) U so g(a
3
+b
3
, . . . , a
r
+b
r
) =
1. The form g evidently vanishes on all vectors in
span((a
3
, . . . , a
r
)) +U
0
and span((b
3
, . . . , b
r
)) +U
0
.
One of these subspaces must have dimension > k2 because other-
wise (a
3
+b
3
, . . . , a
r
+b
r
) U
0
, which contradicts g(a
3
+b
3
, . . . , a
r
+
b
r
) = 1.
Example 26.3. Let f be a quadratic form in three indeterminates
and consider the quadric Q(f) in the projective plane PG
2
(q). If f
is nondegenerate, there are q +1 projective points on Q(f). As we
remarked above, any line L of PG
2
(q) meets Q(f) in a quadric in
L, which in this case consists of 0, 1, or 2 points since Q(f) cannot
contain L by Theorem 26.7. So Q(f) is a set of q + 1 points, no
three of which are collinear, i.e. an oval. See Problem 19I.
If f has rank 0, Q(f) is all of PG
2
(q). If f has rank 1, Q(f)
consists of the points of a line in PG
2
(q). If f has rank 2, there
are two cases. If f is reducible, say f is projectively equivalent to
xy, then Q(f) consists of the points on the union of two lines of
PG
2
(q); if f is irreducible, then Q(f) is a single point in PG
2
(q).
Example 26.4. The rst order Reed-Muller codes were intro-
duced in Chapter 18. Here is one way to introduce the entire
family: Let V = F
m
2
. We consider vectors of length 2
m
whose coor-
dinates are indexed by elements of V ; for concreteness, write V =
362 A Course in Combinatorics
v
0
, v
1
, . . . , v
2
m
1
. The k-th order Reed-Muller code RM(k, m) is
dened to consist of all vectors (of length 2
m
)
(f(v
0
), f(v
1
), . . . , f(v
2
m
1
))
where f ranges over all polynomials in x
1
, . . . , x
m
of degree at most
k.
Since a linear form x
i
1
+ +x
i
k
gives the same function as the
quadratic form x
2
i
1
+ +x
2
i
k
over F
2
, the words in the second order
code RM(2, m) are given by binary quadratic forms f(x
1
, . . . , x
n
)
and their complements f(x
1
, . . . , x
n
) +1. Theorems 26.4 and 26.5
are useful for determining the weights that occur for codewords in
RM(2, m).
We must consider degenerate forms also; e.g. x
1
x
2
+ x
3
x
4
corre-
sponds to a codeword of weight 24 in RM(2, 6). It will be found
that the weights of codewords in RM(2, 6) are 0, 16, 24, 28, 32, 36,
40, 48, and 64.
Problem 26F. Let f(x) := x
1
x
2
+ +x
2m1
x
2m
. Then f deter-
mines a coset C of RM(1, 2m) in RM(2, 2m) consisting of words
corresponding to f(x) + a(x), where a(x) ranges over the 2
2m+1
ane functions (linear plus possibly a constant term) in 2m vari-
ables. Show that in C half of the words have weight 2
2m1
+ 2
m1
and half have weight 2
2m1
2
m1
.
Example 26.5. Let Q
3
be a nondegenerate elliptic quadric in pro-
jective 3-space PG
3
(q). By Theorem 26.6, [Q
3
[ = q
2
+1. By Theo-
rem 26.7, Q
3
contains no lines. Any plane P meets Q
3
in a quadric
in that plane; in view of Example 26.3, every plane must meet Q
3
in either an oval in that plane or a single point. Any three points of
Q
3
are contained in a unique plane, so it follows that the nontriv-
ial intersections of planes with Q
3
provide the blocks of a Steiner
system
S(3, q + 1, q
2
+ 1).
In general, a set of q
2
+ 1 points, no three collinear, in PG
3
(q) is
called an ovoid and an S(3, n + 1, n
2
+ 1) is called a Mobius plane
or an inversive plane.
Example 26.6. Let Q
4
be a nondegenerate quadric in projective
4-space PG
4
(q). By Theorem 26.6, [Q
4
[ = q
3
+ q
2
+ q + 1. Let
26. Designs and projective geometries 363
Q be the incidence structure whose points are the elements of Q
4
and whose blocks are the lines of PG
4
(q) fully contained in Q
4
.
Each point of Q is on exactly q +1 blocks of Q (see Problem 26F).
Given a point x and a block L of Q, the intersection of the plane
P := xL with Q
4
is a quadric Q
2
in P that contains, obviously,
a line and a point o that line. From Example 26.3, Q
2
must consist
of the points on two (intersecting) lines. This implies that Q is a
partial geometry pg(r, k, t), as dened in Chapter 21, where
r = q + 1, k = q + 1, t = 1.
Example 26.7. Let Q
5
be a nondegenerate elliptic quadric in
PG
5
(q). By Theorem 26.6, [Q
5
[ = (q + 1)(q
3
+ 1). By Theo-
rem 26.7, Q
5
contains no planes. Let Q be the incidence structure
whose points are the elements of Q
5
and whose blocks are the lines
of PG
5
(q) fully contained in Q
5
. By Problem 26F, each point of Q
is on q
2
+1 lines of Q. By an argument similar to that of Example
26.6, Q is a partial geometry pg(r, k, t) where
r = q
2
+ 1, k = q + 1, t = 1.
Partial geometries with t = 1 are called generalized quadrangles.
See L. M. Batten (1986) for further results and references.
Problem 26G. Let f be a nondegenerate quadratic form in n
indeterminates given by a matrix C = (c
ij
) over F
q
as in (26.1).
Let Q = Q(f) be the corresponding quadric in PG
n1
(F
q
). Let
p = x) be one of the points on Q. Let
T
p
:= y) : x(C +C

)y

= 0.
Then T
p
is a hyperplane in PG
n1
(F
q
) by Problem 26E. Show that
T
p
Q consists exactly of p and the union of any lines on p that are
contained in Q. Further show that if W is any hyperplane of T
p
not containing p, then Q
/
:= W Q is a nondegenerate quadric in
W (= PG
n3
(F
q
)) and that Q
/
is parabolic, hyperbolic, or elliptic
according to whether Q is parabolic, hyperbolic, or elliptic. We
see, in particular, that the number of lines on p that lie entirely in
Q is equal to [Q
/
[.
364 A Course in Combinatorics
A hermitian form in indeterminates x
1
, . . . , x
n
over F
q
2 is an ex-
pression of the form
(26.6) h(x) = h(x
1
, . . . , x
n
) =
n

i,j=1
c
ij
x
i
x
q
j
,
where the coecients c
ij
are from F
q
2 and where c
ji
= c
q
ij
. In
particular, the diagonal coecients c
ii
, being xed by the Frobenius
automorphism x x
q
, lie in F
q
.
Two hermitian forms f and g over F
q
2 are projectively equivalent
when g can be obtained from f by an invertible linear substitution,
or, in matrix notation, g(x) = f(xA) for some nonsingular n n
matrix A over F
q
2 . So if f is dened by a matrix C as in (26.6),
i.e. f(x) = xCx

, then g is dened by the matrix ACA

, where A

is the conjugate transpose of A; if A = (a

ij
), then A

:= (a
q
ji
).
The rank of a hermitian form is the least number of indeter-
minates that occur (with nonzero coecients) in any projectively
equivalent hermitian form.
A hermitian variety in PG
n
(q
2
) is a set of projective points
H = H(f) := x) : f(x) = 0
where f is a hermitian form in n + 1 indeterminates. We must
choose a basis and identify the vectors with (n+1)-tuples over F
q
2
in order to make this denition. It can be seen that the intersection
of a hermitian variety in PG
n
(q
2
) with a at is a hermitian variety
in that at.
Theorem 26.8. A hermitian form of rank n is projectively equiv-
alent to
(26.7) x
q+1
1
+x
q+1
2
+ +x
q+1
n
.
Proof: It is not hard to see that any nonzero hermitian form is
projectively equivalent to h as in (26.6) where c
11
,= 0; we leave
this to the reader. Let y := c
11
x
1
+ c
12
x
2
+ + c
1n
x
n
. Then
h = c
1
11
yy
q
+g(x
2
, . . . , x
n
) where g is a hermitian form in x
2
, . . . , x
n
.
Since c
11
F
q
, there exists a F
q
2 such that a
q+1
= c
1
11
and then
h = z
q+1
+g(x
2
, . . . , x
n
), where z = ay.
The theorem follows from this step and induction.
26. Designs and projective geometries 365
Theorem 26.9. The number of points on a nondegenerate hermi-
tian variety H in PG
n
(q
2
) is
(q
n+1
+ (1)
n
)(q
n
(1)
n
)
q
2
1
.
Proof: For each nonzero a F
q
, there are q + 1 values of x F
q
2
such that x
q+1
= a. If
f(x
1
, . . . , x
n
) = x
q+1
1
+g(x
2
, . . . , x
n
)
and there are N vectors (x
2
, . . . , x
n
) such that g(x
2
, . . . , x
n
) = 0,
then there are N+(q+1)(q
2(n1)
N) vectors (x
1
, . . . , x
n
) such that
f(x
1
, . . . , x
n
) = 0. Inductively, there are q
2n1
+ (1)
n
(q
n
q
n1
)
vectors in F
n
q
2
that are zeros of (26.7) and the theorem follows.
Example 26.8. Consider hermitian varieties in the projective line
PG
1
(q
2
). If a hermitian form f in two indeterminates has rank 0,
then the variety H(f) contains all q
2
+ 1 points; if f has rank 1,
H(f) consists of one point; if f has rank 2, H(f) contains q + 1
points. It follows that if H is a hermitian variety in PG
n
(q
2
) for
any n, then lines of PG
n
(q
2
) meet H in 1, q + 1, or q
2
+ 1 points.
Now consider a nondegenerate hermitian variety H
2
in the pro-
jective plane PG
2
(q
2
). It has q
3
+ 1 points by Theorem 26.9. Any
line L meets H
2
in a hermitian variety L H
2
in that line. As an
exercise, the reader should check that no lines are contained in H
2
,
so [LH
2
[ = 1 or q +1. It follows that the nontrivial intersections
of lines of PG
2
(q
2
) with H
2
provide the blocks of a Steiner system
S(2, q + 1, q
3
+ 1)
on the point set H
2
.
Designs with these parameters are called unitals. Further analy-
sis shows that the designs constructed above are resolvable; see R.
C. Bose (1959).
We conclude this chapter with a construction of higher dimen-
sional analogues of the Steiner systems S(3, q+1, q
2
+1) of Example
26.5; those constructed below have been called circle geometries.
366 A Course in Combinatorics
Theorem 26.10. If q is a power of a prime and n a positive integer,
then there exists a Steiner system S(3, q + 1, q
n
+ 1).
Proof: Let V be a 2-dimensional vector space over F
q
n and let A
be the set of 1-dimensional subspaces of V over F
q
n (i.e. the q
n
+1
points of PG
1
(q
n
), the projective line of order q
n
).
Now think of V as a 2n-dimensional vector space over F
q
. The
blocks of our Steiner system will come from the 2-dimensional sub-
spaces U over F
q
of V that are not contained in any member of A.
For every such subspace U, let
B
U
:= P A : P U ,= 0.
Note that each P B
U
meets U in a 1-dimensional subspace over
F
q
(that contains q 1 of the q
2
1 nonzero vectors of U), so
[B
U
[ = q + 1. If W = U for some nonzero scalar F
q
n, then
B
U
= B
W
; we take only the distinct sets B
U
as blocks.
Consider three distinct points P
i
A, i = 1, 2, 3. Say P
i
is the
F
q
n-span of a vector x
i
, i = 1, 2, 3. These three vectors are linearly
dependent over F
q
n, say
x
3
= x
1
+x
2
with , F
q
n. But then it is clear that U := span
F
q
x
1
, x
2

meets P
1
, P
2
, and P
3
all nontrivially. Suppose some 2-dimensional
subspace W over F
q
meets each of the P
i
nontrivially, say W con-
tains
i
x
i
with 0 ,=
i
F
q
n, i = 1, 2, 3. Then these vectors are
linearly dependent over F
q
, say

3
x
3
= a
1
x
1
+b
2
x
2
where a, b F
q
. Since x
1
and x
2
are linearly independent over F
q
n,
we have
3
= a
1
and
3
= b
2
. It follows that
3
U = W, and
we see that three distinct points are contained in a unique block.

Problem 26H. Let be a collineation of a projective plane P.

(i) Show that if xes all points on a line l and two points not
on l, then xes P pointwise.
26. Designs and projective geometries 367
(ii) If xes all points on a line l, then there is a point P such
that xes P and every line through P.
(Such an automorphism of P is called a central collineation.)
Problem 26I. In Chapter 23 we saw that AG(3, q) is a subgeom-
etry of PG(3, q), where the missing part is a projective plane P,
sometimes called the plane at innity for AG(3, q). Let q = 2
m
and let O be a hyperoval in P. We dene an incidence structure
I with as points the points of AG(3, q) and as lines those lines of
AG(3, q) that meet P in a point of O. Show that I is a generalized
quadrangle.
Problem 26J. Consider a hyperoval O in PG(2, 4). Dene a
graph G with as vertices the points of the plane not on O and
edges (x, y) if the line through x and y meets O. Prove:
(i) Every edge is in a unique triangle.
(ii) For every triangle (x, y, z) all other vertices are joined to
exactly one of x, y, z.
(iii) G is a generalized quadrangle.
Problem 26K. Consider an arbitrary symmetric (v, k, )-design.
Dene an arc to be a set S of points, no three on a block. A block
is called a tangent of S if it meets S in one point. Find an upper
bound on [S[ depending on whether S has a tangent or not.
Problem 26L. An oval in a symmetric design is an arc that meets
the bound in Problem 26J. Consider a symmetric design with pa-
rameters (4 1, 2, ). Show that the ovals of this design are the
lines of size 3 of the complementary design.
Notes.
The term maximal (m, k)-arc is often used in the literature for
what we have called perfect (m, k)-arcs, but this is too much of
an abuse of the word maximal for us to tolerate.
The combinatorial properties of quadrics and hermitian varieties
are discussed in more detail in D. K. Ray-Chaudhuri (1962) and R.
C. Bose and I. M. Chakravarti (1966).
The polynomial method illustrated in the proof of Theorem 26.3
has further uses. E.g. A. Blokhuis (1994) proved that a nontrivial
368 A Course in Combinatorics
blocking set in PG(2, p) has at least (3p+1)/2 points, a conjecture
of J. di Paola. The idea is to note that the blocking set S must have
a tangent, which without loss of generality is the line at innity. Let
the other points be (a
i
, b
i
) in the ane plane with i = 1, . . . , p +k.
A one page complicated study of the polynomial
F(t, u) =
p+k

i=1
(t +a
i
+ub
i
)
shows that k
p+1
2
.
References.
L. M. Batten (1986), Combinatorics of Finite Geometries, Cam-
bridge University Press.
A. Blokhuis (1994), On the size of a blocking set in PG(2, p), Com-
binatorica 14 (1994), 111114.
R. C. Bose (1959), On the application of nite projective geom-
etry for deriving a certain series of balanced Kirkman arrange-
ments, Golden Jubilee Commemoration Volume (195859), Cal-
cutta Math. Soc., pp. 341354.
R. C. Bose and I. M. Chakravarti (1966), Hermitian varieties in
a nite projective space PG(N, q
2
), Canad. J. Math 18, 1161
1182.
A. E. Brouwer and A. Schrijver, The blocking number of an ane
space, J. Combinatorial Theory (A) 24 (1978), 251253.
P. Dembowski (1968), Finite Geometries, Springer-Verlag.
R. H. F. Denniston (1969), Some maximal arcs in nite projective
planes, J. Combinatorial Theory 6, 317319.
J. W. P. Hirschfeld (1979), Projective Geometries over Finite Fields,
Clerendon Press.
D. K. Ray-Chaudhuri (1962), Some results on quadrics in nite
projective geometry based on Galois elds, Canad. J. Math. 14,
129138.
27
Dierence sets and
automorphisms
A class of symmetric designs arises from dierence sets (dened be-
low) in abelian groups. One such design appeared in Example 19.6.
The group reappears in the automorphism group of the design.
In a simple incidence structure, we may identify the blocks with
sets of points, i.e. the block set / is a family of subsets of the point
set X. An automorphism of a symmetric design (X, /), or any
simple incidence structure, is a permutation of X that takes /
to /, i.e. for A X, A / if and only if (A) /. We begin
with a theorem on automorphisms of symmetric designs in general.
Theorem 27.1. Let o = (X, /) be a symmetric (v, k, )-design
and an automorphism of o. Then the number of points xed by
is equal to the number of blocks xed by .
Proof: Let N be the incidence matrix of o. Dene a permutation
matrix P whose rows and columns are indexed by the points and
where
P(x, y) :=
_
1 if (x) = y,
0 otherwise.
Dene a permutation matrix Q whose rows and columns are in-
dexed by the blocks and where
Q(A, B) :=
_
1 if (A) = B,
0 otherwise.
Note that the trace of P is equal to the number of xed points, and
the trace of Q is equal to the number of xed blocks of .
370 A Course in Combinatorics
Now we have
PNQ

(x, A) =

yX,B/
P(x, y)N(y, B)Q(A, B)
= N((x), (A)) = N(x, A).
That is, PNQ

= N. Equivalently, P = NQN
1
. Thus P and
Q, being similar matrices, have the same trace and the theorem is
proved.
Corollary. The type of the cycle decomposition of on the point
set X is the same as the type of the cycle decomposition of on
the block set /.
Proof: By Theorem 27.1,
i
has the same number of xed points
as xed blocks, for each i = 1, 2, . . . .
Suppose a permutation has c
i
cycles of length i on some set
S, i = 1, 2, . . . , [S[. Let f
j
denote the number of xed points of
j
.
Then
f
j
=

i[j
ic
i
,
and by M obius inversion, Theorem 10.4,
jc
j
=

i[j
(
j
i
)f
i
.
The point is that the numbers of cycles of each length (i.e. the type
of ) are determined completely by the numbers of xed points of
the powers of .
Corollary. If is a group of automorphisms of a symmetric de-
sign, then the number of orbits of on the point set X is the same
as the number of orbits of on the block set /. In particular, is
transitive on the points if and only if is transitive on the blocks.
Proof: By Burnsides lemma, Theorem 10.5, the number of orbits
of a group of permutations of a set S is determined exactly by
the multiset (f() : ) where f() is the number of elements
of S xed by .
27. Dierence sets and automorphisms 371
Before introducing dierence sets, we digress to give one theo-
rem on orbits of automorphism groups of arbitrary 2-designs. This
has been called Blocks Lemma; see Block (1967). This provides
another proof of the above corollary, as the dual of a symmetric
design is also a 2-design according to Theorem 19.9.
Theorem 27.2. If is a group of automorphisms of a 2-(v, k, )
design with v > k, then the number of orbits of on the point set
X is less than or equal to the number of orbits of on the block
set /.
Proof: Let X
1
, X
2
, . . . , X
s
be the orbits of on X and let /
1
, /
2
,
. . . , /
t
be the orbits of on /. Dene two s t matrices C, D
as follows. C(i, j) is to be the number of points x X
i
that are
incident with a xed block A /
j
(this number is the same for any
block A in /
j
). D(i, j) is to be the number of blocks A /
j
that
are incident with a xed point x X
i
(this number is the same for
any point x in X
i
).
Consider the s s matrix product CD

. CD

(i, ) is the sum

over j of the number of ordered pairs (x, A) X
i
/
j
where x is
incident with A and A is incident with a xed point y in X

. So
CD

(i, ) = [X
i
[ if i ,= , and CD

(i, i) = (r ) + [X
i
[; that
is,
CD

= (r )I +diag([X
1
[, [X
2
[, . . . , [X
s
[)J.
This is similar to the matrix equation (19.7). Since v > k, we have
r > and the matrix on the right can be seen to be nonsingular
in several ways, e.g. by calculating its determinant. It follows that
the rank of C (and D) is s, and this cannot be greater than the
number t of columns.
* * *
Let G be an abelian group of order v. A (v, k, )-dierence set in
G is a k-subset D G such that each nonzero g G occurs exactly
times in the multiset (x y : x, y D) of dierences from D.
More formally, we are requiring that the number of ordered pairs
(x, y) with x, y D and xy = g is when g ,= 0 and this number
is k for g = 0. Evidently, (v 1) = k(k 1).
372 A Course in Combinatorics
Example 27.1. Examples of dierence sets include:
(7, 3, 1) 1, 2, 4 in Z
7
(13, 4, 1) 0, 1, 3, 9 in Z
13
(11, 5, 2) 1, 3, 9, 5, 4 in Z
11
(16, 6, 2) 10, 20, 30, 01, 02, 03 in Z
4
Z
4
(16, 6, 2) 0000, 0001, 0010, 0100, 1000, 1111 in Z
2
Z
2
Z
2
Z
2
A dierence set is nontrivial when 1 < k < v 1. A dierence
set with = 1 is sometimes called planar or simple.
Let G be an abelian group of order v. For S G, g G, we
denote by S +g the translate, or shift,
S +g := x +g : x S
of S by g. Let D be a k-subset of G, and x, y G. In general, we
claim that the number of shifts D+g that contain both x and y is
equal to the number of times d := xy occurs as a dierence within
D. This is because g (xg, yg) is a one-to-one correspondence
between the set g G : x, y D + g and the set of ordered
pairs (a, b) of elements of D such that a b = x y. (The reader
may check that this common number is also equal to the cardinality
of the intersection (D +x) (D +y).)
In particular, (G, D+g : g G) is a symmetric (v, k, )-design
if and only if D is a (v, k, )-dierence set.
Problem 27A. A (v, k, )-quotient set in an arbitrary group G of
order v (written multiplicatively) is a k-subset D G such that
any one of the following conditions holds:
(1) Each nonidentity element g G occurs exactly times in
the list (xy
1
: x, y D) of right quotients from D.
(2) Each nonidentity element g G occurs exactly times in
the list (x
1
y : x, y D) of left quotients from D.
(3) [D (Dg)[ = for each nonidentity g G.
(4) [D (gD)[ = for each nonidentity g G.
(5) (G, Dg : g G) is a symmetric (v, k, )-design.
(6) (G, gD : g G) is a symmetric (v, k, )-design.
Show that the above six conditions are equivalent.
Theorem 27.3. Let G be a group of order v. The existence of a
(v, k, )-quotient set in G is equivalent to the existence of a sym-
27. Dierence sets and automorphisms 373
metric (v, k, )-design that admits a group

G of automorphisms
that is isomorphic to G and regular, i.e. sharply transitive, on the
points of the design.
Proof: Let D be a (v, k, )-quotient set in G. Then (G, gD : g
G) is a symmetric (v, k, )-design. For g G, dene a permuta-
tion g of G by g(x) = gx. Then each g is in fact an automorphism
of (G, gD : g G) and the group

G = g : g G of automor-
phisms is clearly isomorphic to G and regular on the points.
Conversely, let G be given and let (X, /) be a symmetric (v, k, )-
design with regular group

G of automorphisms of (X, /) that is
isomorphic to G. It will be sucient to exhibit a (v, k, )-quotient
set in

G.
Fix a point x
0
X and a block A
0
/. Let
D :=

G : (x
0
) A
0
.
We claim that D is a (v, k, )-quotient set in

G. Since

G is regular
and [A
0
[ = k, we have [D[ = k. Let be a nonidentity element of

G. Then D = : (x
0
) A
0
= : (x
0
) (A
0
), so
D (D) = : (x
0
) A
0
(A
0
).
Now since

Gis regular, has no xed points and hence, by Theorem
27.1, xes no blocks. Thus the block (A
0
) is distinct from A
0
, so
[A
0
(A
0
)[ = , and by regularity, [D(D)[ = . This holds for
all nonidentity elements and establishes our claim.
In particular, the existence of a cyclic (v, k, )-dierence set, i.e.
a dierence set in Z
v
, is equivalent to the existence of a symmet-
ric (v, k, )-design that admits a cyclic automorphism, i.e. an au-
tomorphism with cycle decomposition on the pointsor blocks
consisting of one cycle of length v.
We now restrict our attention to dierence sets in abelian groups.
Ideally, we would like to describe and classify all dierence setsto
nd out which groups have dierence sets, how many there are, etc.
This we shall do for various small parameter triples, but in general
the existence problem alone is already extremely dicult.
374 A Course in Combinatorics
Observe that D G is a (v, k, )-dierence set if and only if
G D is a (v, v k, v 2k +)-dierence set. Thus we may conne
our attention to the case k <
1
2
v.
Also note that D is a dierence set if and only if every translate
of D is a dierence set.
In the case that (v, k) = 1, it happens that we can choose a
natural representative from the class of all translates; this can be a
help in classication and will be useful in the next chapter. Call a
subset of an abelian group G normalized in the case that the sum
of its elements is zero.
Proposition 27.4. Let D be a k-subset of an abelian group G of
order v. If (v, k) = 1, then D has a unique normalized translate.
Proof: Let h be the sum of the elements of D. Then the sum of
the elements of a translate D +g is h +kg. Since (v, k) = 1, there
is a unique group element g with h +kg = 0.
The reader is invited to verify to his or her satisfaction that
the normalized dierence sets with parameters (v, k, ) = (7, 3, 1),
(13, 4, 1), (11, 5, 2), (21, 5, 1) are, respectively:
1, 2, 4, 3, 5, 6 in Z
7
;
0, 1, 3, 9, 0, 2, 5, 6, 0, 4, 10, 12, 0, 7, 8, 11 in Z
13
;
1, 3, 4, 5, 9, 2, 6, 7, 8, 10 in Z
11
;
7, 14, 3, 6, 12, 7, 14, 9, 15, 18 in Z
21
.
(This will be easy after Theorem 28.3see Example 28.2.) Of
course, for (v, k) = 1, the total number of dierence sets is v times
the number of normalized dierence sets, e.g. there are 52 dierence
sets with parameters (13, 4, 1) in Z
13
.
A nal preliminary observation is that if is any automorphism
of the group G, then a subset D G is a dierence set if and only
if (D) is a dierence set. Thus from a given dierence set we can
obtain others by taking translates and by means of the symmetries
of G. We say that dierence sets D
1
, D
2
in G are equivalent when
there exists Aut(G) and g G such that D
2
= (D
1
) + g.
(Check that this is an equivalence relation.) The normalized dif-
ference sets shown above are, for each parameter triple, equivalent;
27. Dierence sets and automorphisms 375
indeed, each can be obtained from the others by multiplication by
some integer relatively prime to the order of the respective group.
Problem 27B. Show that all (16, 6, 2)-dierence sets in Z
2
Z
2

Z
2
Z
2
are equivalent.
Problem 27C. Recall from Example 19.4 that symmetric designs
with parameters v = 4t
2
, k = 2t
2
t, = t
2
t are related to regular
Hadamard matrices. Let A and B be, respectively, (4x
2
, 2x
2

x, x
2
x)- and (4y
2
, 2y
2
y, y
2
y)-dierence sets in groups G and
H (admit x or y = 1). Show that
D := (A(H B)) ((G A) B)
is a (4z
2
, 2z
2
z, z
2
z)-dierence set in G H where z = 2xy.
(Thus if G is the direct product of m groups of order 4, then there
is a (4
m
, 2 4
m1
2
m1
, 4
m1
2
m1
)-dierence set in G.)
We describe several known families of dierence sets. The known
constructions all seem to involve nite elds and/or vector spaces.
The rst of these examples is essentially contained in Chapter 18
on Hadamard matrices in the discussion of Paley matrices, but it
is worth stating explicitly. Dierence sets with parameters
(v, k, ) = (4n 1, 2n 1, n 1)
are often called Hadamard dierence sets.
Theorem 27.5 (Paley, Todd). Let q = 4n1 be a prime power.
Then the set D of nonzero squares in F
q
is a (4n1, 2n1, n1)-
dierence set in the additive group of F
q
.
Proof: Clearly, [D[ = 2n 1.
Since D is invariant under multiplication by elements of the set
S of nonzero squares, the multiset M of dierences from D also has
this property. Also, M is obviously invariant under multiplication
by 1. Since q 3 (mod 4), 1 / S and every nonzero element of
F
q
is either in S or of the form s for some s S. In summary, M
is invariant under multiplication by all nonzero elements of F
q
, and
so D is a (4n 1, 2n 1, )-dierence set for some . The relation
(v 1) = k(k 1) forces = n 1.
376 A Course in Combinatorics
We obtain dierence sets 1, 2, 4 in Z
7
, 1, 3, 4, 5, 9 in Z
11
, and
1, 4, 5, 6, 7, 9, 11, 16, 17 in Z
19
from Theorem 27.5. The (27, 13, 6)-
dierence set later in the series will be in the elementary abelian
group of order 27, not Z
27
.
Problem 27D. Show that the PaleyTodd dierence sets are nor-
malized if q > 3.
Stanton and Sprott (1958) found another family of Hadamard
dierence sets.
Theorem 27.6. If q and q + 2 are both odd prime powers, then
with 4n1 := q(q+2), there exists a (4n1, 2n1, n1)-dierence
set in the additive group of the ring R := F
q
F
q+2
.
Proof: Let U := (a, b) R : a ,= 0, b ,= 0 be the group of
invertible elements of R. Let V be the subgroup of U consisting of
those pairs (a, b) such that both a and b are squares in the respective
elds F
q
and F
q+2
or both a and b are nonsquares. Check that V
is an index 2 subgroup of U, and also that (1, 1) / V . Put
T := F
q
0. We claim that D := T V is a dierence set as
required. We do have [D[ = q +
1
2
(q 1)(q + 1) = 2n 1.
Since D is invariant under multiplication by elements of V , the
multiset of dierences from D also has this property as well as the
property that it is invariant under multiplication by (1, 1). So
the multiset of dierences from D is invariant under multiplication
by elements of all of U. Thus every element of U occurs as a
dierence the same number of times, say
1
. Every element (x, 0)
of R with x ,= 0 will occur, say,
2
times; and every element (0, y)
of R with y ,= 0 will occur, say,
3
times as a dierence from D.
Of course, we have
(27.1) k(k 1) = (q 1)(q + 1)
1
+ (q 1)
2
+ (q + 1)
3
.
It is easy to evaluate
2
and
3
. Dierences of the form (x, 0),
x ,= 0 (there are q 1 elements of this form) arise q(q 1) times
from T, never as a dierence between an element of T and V , and
(q +1) (
1
2
(q 1))(
1
2
(q 1) 1) times as dierences of two elements
of V ; thus
(q 1)
2
= q(q 1) + (q + 1) (
1
2
(q 1))(
1
2
(q 1) 1)
27. Dierence sets and automorphisms 377
from which we conclude
2
=
1
4
(q + 3)(q 1). In a similar manner
(Problem 27E), the reader will nd that
3
=
1
4
(q +3)(q 1). Then
(27.1) implies that
1
=
1
4
(q + 3)(q 1) also.
Problem 27E. With the notation as in the proof of Theorem 27.6
prove that
3
=
1
4
(q + 3)(q 1).
In the case that q and q + 2 are both primes (twin primes),
Theorem 27.6 yields cyclic dierence sets. For if s, t are relatively
prime integers, Z
s
Z
t
and Z
st
are isomorphic as additive groups
and also as rings. An isomorphism Z
st
Z
s
Z
t
is provided by
x (mod st) (x (mod s), x (mod t)) . We obtain the (15, 7, 3)-
dierence set 0, 5, 10, 1, 2, 4, 8 in Z
15
.
Problem 27F. List the elements of a cyclic (35, 17, 8)-dierence
set.
As a special case of Singers theorem, Theorem 27.7, we will
obtain another family of Hadamard dierence sets: cyclic dierence
sets with
v = 2
t
1, k = 2
t1
1, = 2
t2
1.
Recall from Chapter 23 that the points and the hyperplanes of
PG(n, q) form a symmetric design with
(27.2) v =
q
n+1
1
q 1
, k =
q
n
1
q 1
, =
q
n1
1
q 1
.
Theorem 27.7. For any prime power q and positive integer n,
there is a dierence set D with parameters as in (27.2) in the cyclic
group of order v so that the resulting symmetric design is isomor-
phic to the points and hyperplanes of PG(n, q).
Proof: In view of Theorem 27.3, we need only show that there
exists an automorphism of PG(n, q) that permutes the points in a
single cycle of length v, or equivalently so that the powers of the au-
tomorphism act transitively on the projective points. The points of
PG(n, q) are the 1-dimensional subspaces of an (n+1)-dimensional
vector space V over F
q
. Any nonsingular linear transformation T
378 A Course in Combinatorics
from V to itself will take subspaces to subspaces of the same di-
mension and thus gives us an automorphism of PG(n, q).
As an (n + 1)-dimensional vector space over F
q
, we choose V :=
F
q
n+1 as a vector space over its subeld F
q
. Let be a primitive
element of F
q
n+1 and consider the linear transformation T : x x
of V over F
q
. It is clear that T is nonsingular and that its powers are
transitive on the projective points (even on the nonzero vectors!).

The dierence sets constructed in the proof of Theorem 27.7

above are called Singer dierence setssee Singer (1938). We give
a more concrete discussion and an example. Let be a primitive
element of F
q
n+1 and dene v := (q
n+1
1)/(q 1). The cyclic
multiplicative group ) of F
q
n+1 has a unique subgroup of order
q 1, namely,

v
) =
0
= 1,
v
,
2v
, . . . ,
(q2)v
.
But the multiplicative group of the subeld F
q
has order q 1, so
we conclude F
q
= 0,
0
,
v
,
2v
, . . . ,
(q2)v
.
Now two vectors
i
and
j
in F
q
n+1 , considered as a vector space
over F
q
, represent the same 1-dimensional subspace of F
q
n+1 if and
only if
i
=
j
for some 0 ,= F
q
, that is, if and only if i j
(mod v). Thus we have a one-to-one correspondence between the
set X of 1-dimensional subspaces (projective points) and the group
Z
v
of residue classes modulo v:
0 x
0
= 0,
0
,
v
,
2v
, . . . ,
(q2)v

1 x
1
= 0,
1
,
v+1
,
2v+1
, . . . ,
(q2)v+1

.
.
.
i x
i
= 0,
i
,
v+i
,
2v+i
, . . . ,
(q2)v+i

.
.
.
v 1 x
v1
= 0,
v1
,
2v1
,
3v1
, . . . ,
(q1)v1
.
The map x
i
x
i+1
(subscripts modulo v) is an automorphism of
the projective space. To obtain a dierence set, let U be any n-
dimensional subspace of F
q
n+1 and let D := i Z
v
: x
i
U. A
normalized dierence set is obtained if U is taken to be the subspace
of elements with trace zero (the trace from F
q
n+1 to F
q
).
27. Dierence sets and automorphisms 379
Example 27.2. Consider n = 2, q = 5. We construct a (31, 6, 1)-
dierence set. A zero of the polynomial y
3
+y
2
+ 2 (coecients
in F
5
) is a primitive element of F
5
3 and 1, ,
2
furnish a basis for
F
5
3 as a vector space over F
5
. The 124 nonzero elements of F
5
3 fall
into 31 cosets modulo the subgroup
31
) = 3, 4, 2, 1 = F
5
0,
each coset being the nonzero elements of a 1-dimensional subspace.
Let us take U := span1, as our 2-dimensional subspace. Rep-
resentatives of the projective points on U are 1, , + 1, + 2,
+ 3, + 4 and, after some computation,
1 =
0
=
1
+ 1 =
29
+ 2 =
99
+ 3 =
80
+ 4 =
84
.
The resulting Singer dierence set is
0, 1, 29, 6, 18, 22 in Z
31
.
Problem 27G. Find a (57, 8, 1)-dierence set in Z
57
. (Some will
probably not want to do this by hand.)
Consider n = 3, q = 2. Here the points of PG(3, 2) are in one-to-
one correspondence with the nonzero elements of F
2
4 that in turn
are in one-to-one correspondence with the residues modulo 15. It
may be instructive to write out all lines and planes of this smallest
projective 3-space so that the cyclic automorphism is clear.
A zero of y
4
+y +1 (coecients in F
2
) is a primitive element of
F
2
4 and
3
,
2
, , 1 from a basis for F
2
4 over F
2
. Any element of F
2
4
can be written uniquely as a
3

3
+a
2

2
+a
1
+a
0
that we abbreviate
as a
3
a
2
a
1
a
0
below. We have
4
+ +1 = 0, or
4
= 0011. We rst
construct a table of vector representations of powers of .
380 A Course in Combinatorics
F
2
4

0
= 0001
5
= 0110
10
= 0111

1
= 0010
6
= 1100
11
= 1110

2
= 0100
7
= 1011
12
= 1111

3
= 1000
8
= 0101
13
= 1101

4
= 0011
9
= 1010
14
= 1001
PG(3, 2)
Lines Planes
0,5,10 0,1,4 0,2,8 1,2,4,8,0,5,10
1,6,11 1,2,5 1,3,9 2,3,5,9,1,6,11
2,7,12 2,3,6 2,4,10 3,4,6,10,2,7,12
3,8,13 3,4,7 3,5,11 4,5,7,11,3,8,13
4,9,14 4,5,8 4,6,12 5,6,8,12,4,9,14
5,6,9 5,7,13 6,7,9,13,5,10,0
6,7,10 6,8,14 7,8,10,14,6,11,1
7,8,11 7,9,0 8,9,11,0,7,12,2
8,9,12 8,10,1 9,10,12,1,8,13,3
9,10,13 9,11,2 10,11,13,2,9,14,4
10,11,14 10,12,3 11,12,14,3,10,0,5
11,12,0 11,13,4 12,13,0,4,11,1,6
12,13,1 12,14,5 13,14,1,5,12,2,7
13,14,2 13,0,6 14,0,2,6,13,3,8
14,0,3 14,1,7 0,1,3,7,14,4,9
Note that for q = 2, the Singer dierence sets have Hadamard
parameters. We obtain (31, 15, 7)-dierence sets both from The-
orem 27.5 and Theorem 27.7. These two dierence sets are not
equivalentand not even isomorphic. (Two dierence sets are iso-
morphic when the corresponding symmetric designs are isomorphic.
Equivalent dierence sets are surely isomorphic, but the converse
is not true.) The dierence set D of quadratic residues modulo 31
has the property that
D (D + 1) (D + 3) = 5, 8, 10, 19.
27. Dierence sets and automorphisms 381
The design (Z
31
, D + g : g Z
31
) cannot be isomorphic to the
points and hyperplanes of PG(4, 2) since the intersection of ats is
again a at and PG(4, 2) has no ats with precisely four points.
Gordon, Mills, and Welch (1962) have shown that the construc-
tion of Singers theorem can be modied in some cases to produce
many nonequivalent dierence sets with the same parameters.
Problem 27H. Let D be an (n
2
+ n + 1, n + 1, 1)-dierence set
in an abelian group G. Show that D is an oval in the associated
projective plane.
Problem 27I. Let G = 0, a
0
, a
1
, . . . , a
q
be any group of order
q+2, where q is a power of a prime. Let V be a 2-dimensional vector
space over F
q
and let U
0
, U
1
, . . . U
q
be its 1-dimensional subspaces.
Show that
D :=
q
_
i=0
a
i
U
i
is a dierence set in G V . For example, we obtain a (45,12,3)-
dierence set.
Notes.
The idea of group dierence sets, as a generalization of cyclic
dierence sets, is due to R. H. Bruck (1955).
Problem 27I is a result of McFarland (1973).
Theorem 27.2 holds also for arbirary 2-designs, simple or not. Of
course, one must dene an automporphism of an arbitrary incidence
structures S = (T, B, I). Perhaps the most precise denition is as
an ordered pair (, ), where is a permutation of T, is a permu-
tation of B, and such that (x, A) I if and only if ((x), (A)) I.
The automporphisms form a group under coordinate-wise composi-
tion. The permutations that occur as rst coordinates of a group
of automorphisms form a group
1
of permutations of the points;
the permutations that occur as second coordinates of a group
of automorphisms form a group
2
of permutations of the blocks.
By the orbits of on the points, respectively blocks, we mean the
orbits of
1
, respectively
2
.
382 A Course in Combinatorics
References.
R. E. Block (1967), On the orbits of collineation groups, Math Z.
96, 3349.
R. H. Bruck (1955), Dierence sets in a nite group, Trans. Amer.
Math. Soc. 78, 464481.
B. Gordon, W. H. Mills, and L. R. Welch (1962), Some new dier-
ence sets, Canad. J. Math. 14, 614625.
R. L. McFarland (1973), A family of dierence sets in non-cyclic
groups, J. Combinatorial Theory (A) 15, 110.
J. Singer (1938), A theorem in nite projective geometry and some
applications to number theorey, Trans. Amer. Math. Soc. 43,
377385.
R. G. Stanton and D. A. Sprott (1958), A family of dierence sets,
Canad. J. Math. 10, 7377.
28
Dierence sets and the
group ring
The group ring provides a natural and convenient setting for the
study of dierence sets. The existence of a dierence set will be
seen to be equivalent to the existence of a solution to a certain
algebraic equation in a group ring. We shall use the group ring
to derive the celebrated Multiplier Theorem of M. Hall, Jr. as well
as number-theoretic criteria on the parameters of a dierence set
which are stronger than the Bruck-Ryser-Chowla theorem.
Let R be a ring (commutative with one) and G a nite abelian
group (written additively). The elements of the group ring R[G]
are all formal sums
A =

gG
a
g
x
g
,
where a
g
R for each g G. (Here the symbol x is just a place
holder. The important thing is that we have an element a
g
of R
for each element g of G, i.e. the elements of the group ring are in
one-to-one correspondence with mappings G R.)
We dene addition and scalar multiplication in the obvious way:

gG
a
g
x
g
+

gG
b
g
x
g
:=

gG
(a
g
+b
g
)x
g
,
c

gG
a
g
x
g
:=

gG
(ca
g
)x
g
.
Multiplication in R[G] is dened by

gG
a
g
x
g

gG
b
g
x
g
:=

gG
_
_

h+h

=g
a
h
b
h

_
_
x
g
.
384 A Course in Combinatorics
With these denitions, R[G] is a commutative, associative R-
algebra. The notation we have chosen is appropriate for abelian
groups and emphasizes an analogy with polynomials. Note that
when G is the additive group of the residues modulo v, then the
group ring R[Z
v
] consists of all sums

v1
i=0
r
i
x
i
(exponents modulo
v) and is isomorphic to the factor ring R[x]/(x
v
1) of the polyno-
mial ring R[x]. In general, we shall even take to denoting elements
of arbitrary group rings R[G] as A(x), B(x), . . . when convenient.
The element x
0
R[G] is the multiplicative identity in R[G] and
we denote x
0
by 1.
We shall be concerned almost exclusively with the group rings
Z[G] over the integers. For a subset A G, we dene A(x) Z[G]
by
A(x) =

gA
x
g
.
In particular, G(x) =

gG
x
g
. For A, B G,
A(x)B(x) =

gG
c
g
x
g
,
where c
g
is the number of times g occurs in the multiset
(h +h
/
: h A, h
/
B)
of sums of elements of A and B.
For A(x) =

gG
a
g
x
g
Z[G], we write
A(x
1
) :=

gG
a
g
x
g
.
So for a k-subset D of a group G of order v, D is a (v, k, )-dierence
set in G if and only if the equation
D(x)D(x
1
) = n +G(x)
holds in the group ring Z[G], where n := k .
An important homomorphism from Z[G] to Z is given by
A(x) A(1) :=

gG
a
g
Z.
28. Dierence sets and the group ring 385
Proposition 28.1. Let v, k, be positive integers such that
(v 1) = k(k 1)
and let G be an abelian group of order v. The existence of a
(v, k, )-dierence set in G is equivalent to the existence of an ele-
ment A(x) Z[G] satisfying the equation
(28.1) A(x)A(x
1
) = n +G(x),
where n := k .
Proof: We have already pointed out that for a subset D of G,
D(x) satises (28.1) if and only if D is a (v, k, )-dierence set. It
remains to show that if there is a solution A(x) to (28.1), then we
can nd a solution B(x) =

gG
b
g
x
g
where the coecients b
g
are
0s and 1s.
Assume that A(x) satises (28.1) and apply the homomorphism
x 1 from Z[G] Z. We nd
(A(1))
2
= n +v = k
2
,
so A(1) = k or k. Now if A(x) satises (28.1), then so does
B(x) = A(x), so we may assume A(1) =

a
g
= k.
The coecient of 1 = x
0
in A(x)A(x
1
) is k =

gG
a
2
g
. Thus

gG
a
g
(a
g
1) = 0. But a(a 1) is strictly positive unless the
integer a is 0 or 1.
For any integer t, g tg is a homomorphism of the group G into
itself and induces a ring homomorphism Z[G] Z[G], namely
A(x) =

gG
a
g
x
g
A(x
t
) :=

gG
a
g
x
tg
.
For A, B Z[G] and n Z, we say A B (mod n) when AB =
nC for some C Z[G]. Here is an easy lemma we will use often.
Lemma 28.2. Let p be a prime and A Z[G]. Then
(A(x))
p
A(x
p
) (mod p).
386 A Course in Combinatorics
Proof: We proceed by induction on the number of nonzero co-
ecients of A(x). The lemma holds for A(x) = 0. Now if A(x) =
cx
g
+B(x) where B
p
(x) B(x
p
) (mod p), then
A
p
(x) = (cx
g
+B(x))
p
(cx
g
)
p
+B
p
(x)
= c
p
x
pg
+B
p
(x) cx
pg
+B(x
p
) = A(x
p
),
where the congruences are modulo p.
Let G be an abelian group and D a dierence set in G. An
automorphism of G is said to be a multiplier of D if and only if
the dierence set (D) is in fact a translate of D, i.e. if and only
if (D) = D + g for some g G. For example, the automorphism
x 3x of Z
13
is a multiplier of the (13, 4, 1)-dierence set 0,2,3,7
since 0,6,9,8 = 0,2,3,7 + 6.
If t is an integer relatively prime to the order of G, then x tx
is an automorphism of G since the maps x t
1
x and x t
2
x
coincide if and only if t
1
t
2
(mod v

) where v

is the exponent
of G, i.e. the least common multiple of the orders of the elements
of G. If x tx is a multiplier of a dierence set D in G, we say
that t is a numerical multiplier or Hall multiplier of D. It is at rst
surprising that many dierence sets (e.g. all known cyclic dierence
sets) must necessarily have a nontrivial numerical multiplier.
Problem 28A. Let q = p
t
, p prime. Show that p is a multiplier
of the Singer dierence sets described in the previous chapter.
Observe that an automorphism of G is a multiplier of a dif-
ference set D in G if and only if in the group ring Z[G], D(x

) =
x
g
D(x) for some g G.
Theorem 28.3 (Multiplier Theorem, rst version). Let D
be a (v, k, )-dierence set in an abelian group G of order v. Let p
be a prime, p[n, (p, v) = 1, p > . Then p is a numerical multiplier
of D.
It will be convenient to isolate part of the proof as a lemma.
Lemma 28.4. Let be an automorphism of G and let D be a
(v, k, )-dierence set in G. Consider
S(x) := D(x

)D(x
1
) G(x).
28. Dierence sets and the group ring 387
Then is a multiplier of D if and only if S(x) has nonnegative
coecients.
Proof: We begin by remarking that if is a multiplier, then we
have D(x

) = x
g
D(x) for some g G, and then
D(x

)D(x
1
) = x
g
D(x)D(x
1
) = x
g
(n +G(x)) = nx
g
+G(x).
So in this case, S(x), as dened above, is equal to nx
g
for some
g G. In particular, it has nonnegative coecients. Note that,
conversely, if D(x

)D(x
1
) = nx
g
+G(x), we can multiply this by
D(x) to nd
D(x

) (n +G(x)) = nx
g
D(x) +D(x)G(x),
nD(x

) +kG(x) = nx
g
D(x) +kG(x).
So D(x

) = x
g
D(x) and is a multiplier.
Now x x

is an automorphism of Z[G], so D(x

)D(x

) =
n +G(x), i.e. (D) is also a dierence set, and
S(x)S(x
1
) = D(x

)D(x
1
) G(x)D(x

)D(x) G(x)
= n +G(x)
2
2k
2
G(x) +
2
vG(x)
= n
2
+ 2(n +v k
2
)G(x) = n
2
.
Suppose that S(x) =

gG
s
g
x
g
with nonnegative coecients s
g
.
If s
g
> 0 and s
h
> 0 for g, h G, then the coecient of x
gh
in S(x)S(x
1
), = n
2
, is at least s
g
s
h
, i.e. strictly positive, and so
x
gh
= x
0
, i.e. g = h. So S(x) can have only one positive coecient,
say S(x) = s
g
x
g
. The equation S(x)S(x
1
) = n
2
forces s
g
= n and
we have shown S(x) = nx
g
. As noted above, we may conclude that
is a multiplier.
Proof of Theorem 28.3: Let S(x) := D(x
p
)D(x
1
) G(x).
By Lemma 28.4, it will suce to show that S(x) has nonnegative
coecients. By Lemma 28.2,
D(x
p
)D(x
1
) D
p
(x)D(x
1
) D
p1
(x)D(x)D(x
1
)
D
p1
(x) (n +G(x)) nD
p1
(x) +k
p1
G(x)
G(x) (mod p),
388 A Course in Combinatorics
since p divides n and k
p1

p
(mod p). Thus the co-
ecients of D(x
p
)D(x
1
), which are clearly nonnegative, are all
congruent to modulo p. Since p > , it must be that the coe-
cients of D(x
p
)D(x
1
) are greater than or equal to , i.e. S(x) has
nonnegative coecients.
Problem 28B. Find an element S(x) in Z[Z
7
] with the property
that S(x)S(x
1
) = 4, but with S(x) ,= 2x
g
.
Note that the hypothesis p > was essential for our proof. Yet in
every known dierence set, every prime divisor of n (not dividing
v) is a multiplier. It may be, then, that the hypothesis p >
is unnecessary and by now the question of whether this is so has
acquired the status of a classical unsolved problem. There have also
been several generalizations of the original Multiplier Theorem, all
of which would be trivial if we could eliminate the condition p > ,
and we state such a generalization below after rst giving some
applications of the current version.
Corollary. For = 1, every prime divisor of n, and hence every
divisor, is a multiplier of every (n
2
+n + 1, n + 1, 1)-dierence set.
Example 28.1. We claim that there are no (n
2
+n +1, n +1, 1)-
dierence sets with n 0 (mod 6). Let D be such a hypothetical
dierence set, and without loss of generality, assume D is normal-
ized so that it is xed by all multipliers. Both 2 and 3 would be
multipliers. Then for x D, 2x and 3x also are in D. Then the
dierence x occurs twice, once as 2x x, and once as 3x 2x;
these are dierent occurrences as long as 3x ,= 2x, i.e. x ,= 0. This
contradicts = 1.
Problem 28C. Show that there are no (n
2
+ n + 1, n + 1, 1)-
dierence sets with n divisible by any of 10, 14, 15, 21, 22, 26, 34,
35.
It has been possible to prove, with the Multiplier Theorem and
other techniques, that no planar dierence sets exist for n 3600
unless n is a prime power. But the conjecture that n must be a
prime power remains open.
Example 28.2. Consider a normalized (21, 5, 1)-dierence set D
in Z
21
. By the Multiplier Theorem, 2 is a multiplier and hence
28. Dierence sets and the group ring 389
2D = D. Thus D must be the union of the cycles of x 2x on
Z
21
. These are
0, 1, 2, 4, 8, 16, 11, 3, 6, 12, 5, 10, 20, 19, 17, 13,
7, 14, and 9, 18, 15.
But D has 5 elements, so if there is such a dierence set, D must be
7, 14, 3, 6, 12 or 7, 14, 9, 18, 15. It turns out that both of these
are dierence sets. One consists of the negatives of the other, so
the 42 dierence sets with these parameters are all equivalent.
Problem 28D. Find all normalized dierence sets with parame-
ters (7, 3, 1), (11, 5, 2), (13, 4, 1), (19, 9, 4), (31, 10, 3), and (37, 9, 2).
(Note: no dierence sets exist for one of the parameter triples.)
We give two more easy lemmas on the group ring at this point.
Lemma 28.5. Let G be an abelian group of order v and p a prime,
p [ v. Let A Z[G] and suppose A
m
0 (mod p) for some positive
integer m. Then A 0 (mod p).
Proof: Choose a power q = p
e
of p such that q m and q 1
(mod v). Then surely A
q
(x) 0 (mod p). But by Lemma
28.2, A
q
(x) A(x
q
) (mod p), so A(x
q
) 0 (mod p). Since q 1
(mod v), qg = g for every g G and A(x) = A(x
q
).
Note that x
g
G(x) = G(x) in Z[G]. It follows that
A(x)G(x) = A(1)G(x).
For n Z and A, B Z[G], we say A B (mod n, G) when AB
is an element of the ideal in Z[G] generated by n and G = G(x), or
equivalently, when
AB = nC +mG
for some C Z[G].
Lemma 28.6. Let G be an abelian group of order v and p a prime,
p [ v. If A Z[G] and
A
m
0 (mod p, G)
for some positive integer m, then A 0 (mod p, G).
Proof: Choose q = p
e
with q 1 (mod v) and q m. Then
A
q
(x) 0 (mod p, G) and A
q
(x) A(x
q
) = A(x) (mod p).
390 A Course in Combinatorics
Theorem 28.7 (Multiplier Theorem, second version). Let
D be a (v, k, )-dierence set in an abelian group G of exponent v

.
Let t be an integer, (t, v) = 1, and suppose we can nd a divisor
m of n := k such that m > and for every prime divisor p
of m, there is an integer f for which p
f
t (mod v

). Then t is a
numerical multiplier of D.
Proof: The proof will use Lemmas 28.2, 28.4, 28.5 and the fol-
lowing observation:
Let D be a (v, k, )-dierence set in G, an automorphism of G,
and put S(x) := D(x

)D(x
1
) G(x). Assume that has order
e, so that
e
= identity. Then, we assert, in the group ring Z[G],
S(x)S(x

)S(x

2
) S(x

e1
) = n
e
.
To see this, note that for any integer i, we have
D(x

i
)D(x

i
) = n +G(x) n (mod G)
and
S(x

i
) = D(x

i+1
)D(x

i
) G(x)
D(x

i+1
)D(x

i
) (mod G).
Then
S(x)S(x

)s(x

2
) S(x

e1
)
D(x

)D(x
1
)D(x

2
)D(x

) D(x)D(x

e1
)
D(x)D(x
1
)D(x

)D(x

) D(x

e1
)D(x

e1
)
n
e
(mod G).
Thus S(x)S(x

) S(x

e1
) = n
e
+G(x) for some integer . But
S(1) = (D(1))
2
G(1) = k
2
v = n; so applying the homomor-
phism x 1, we nd n
e
= n
e
+v, and hence = 0.
To continue with the proof, let S(x) := D(x
t
)D(x
1
) G(x).
By Lemma 28.4, to show t is a multiplier, it will be sucient to
prove that S(x) has nonnegative coecients. Each coecient of
28. Dierence sets and the group ring 391
S(x) is at least in value and since m > , the nonnegativity
of the coecients will follow if we can establish that S(x) 0
(mod m). To establish this, it will suce to show that S(x) 0
(mod p
i
) whenever p is prime and p
i
divides m. This we do below.
Let e be the order of t modulo v

, so that t
e
1 (mod v

). As
shown above,
S(x)S(x
t
)S(x
t
2
) S(x
t
e1
) = n
e
.
Let p be a prime divisor of m and let f be such that p
f
t
(mod v

). Then
S(x)S(x
p
f
)S(x
p
2f
) S(x
p
f(e1)
) = n
e
.
Let p
i
be the highest power of p dividing n and let p
j
be the
highest power of p dividing (all coecients of) S(x), so S(x) =
p
j
T(x) where T(x),0 (mod p). Then
p
j
T(x) p
j
T(x
p
f
) p
j
T(x
p
f(e1)
) = n
e
,
from which it follows that p
j
divides n (so j i) and
T(x)T(x
p
f
) T(x
p
f(e1)
) = (
n
p
j
)
e
.
Suppose that j < i, so that (
n
p
j
)
e
is divisible by p. Then
0 T(x)T(x
p
f
) T(x
p
f(e1)
)
T(x)T
p
f
(x) T(x)T
p
f(e1)
(x)
(T(x))
1+p
f
++p
f(e1)
(mod p).
But then, by Lemma 28.5, T(x) 0 (mod p), contradicting the
choice of j. Thus i = j.
Corollary. If n = p
e
, p prime, (p, v) = 1, then p is a numerical
multiplier of every (v, k, )-dierence set.
Proof: A dierence set D and its complement G D have the
same multipliers. Thus we may assume that k <
1
2
v and hence
n > . In Theorem 28.7, take m = n, t = p.
392 A Course in Combinatorics
Example 28.3. Consider a hypothetical (25,9,3)-dierence set.
Take t = 2 and m = 6 in the above theorem. Since 3
3
2
(mod 25), we may conclude that 2 is a multiplier.
Problem 28E. Find all normalized dierence sets with parame-
ters (15, 7, 3), (25, 9, 3), (39, 19, 9), (43, 15, 5), and (61, 16, 4). De-
termine, for each parameter set, whether these normalized dier-
ence sets are equivalent. (Note: dierence sets do not exist for most
parameter sets.)
Problem 28F. Let D be a nontrivial (v, k, )-dierence set. Prove
that if 1 is a multiplier of D, then v is even.
Problem 28G. Find a (36,15,6)-dierence set in Z
6
Z
6
with 1
as a multiplier.
* * *
Suppose D is a (v, k, )-dierence set in an abelian group G of
even order v. By Theorem 19.11(i), we know that n is a square.
But in the case of dierence sets, we can say what n is the square
of ! Let A be any subgroup of G of index 2. Say D contains a
elements of A and b elements of B := G A. Since every element
of B occurs times as a dierence from D, and only dierences of
one element in A and one in B lie in B, 2ab =
1
2
v. From this and
a +b = k, we nd that (a b)
2
= n.
Now suppose v is divisible by 3, let A be a subgroup of G of
index 3, and let B and C be the cosets of A in G. Say D contains a
elements of A, b elements of B, and c elements of C. The number
of dierences from D which lie in B, say, is ba +cb +ac, which on
the other hand must be
1
3
v. From this and a +b +c = k, we can
check that 4n = (b + c 2a)
2
+ 3(b c)
2
. Now it is not true that
every integer can be written as the sum of a square and three times
another square, so this condition rules out the existence of certain
dierence sets. For example, there are no (39, 19, 9)-dierence sets,
even though there are symmetric designs with these parameters,
because 4n = 40 cannot be written as above.
We generalize the above necessary conditions on the parameters
of dierence sets by considering homomorphisms of the group ring.
28. Dierence sets and the group ring 393
If is a homomorphism G H, then induces a ring homomor-
phism Z[G] Z[H]:
A(x) =

a
g
x
g
A(x

) :=

a
g
x
(g)
Z[H].
Theorem 28.8. Let D be a (v, k, )-dierence set in an abelian
group G. Let u > 1 be a divisor of v. If p is a prime divisor of n
and
p
f
1 (mod u)
for some integer f, then p does not divide the squarefree part of n.
Proof: We prove something stronger. Let : G H be a homo-
morphism onto a group H of order u and exponent u

, say. Suppose
p is prime and p
f
1 (mod u

) for some f. We will show that n

is exactly divisible by an even power p
2j
of p and that for this j,
D(x

) 0 (mod p
j
, H) in Z[H].
In other words, all the coecients of D(x

), which are the numbers

of elements of D belonging to the various cosets of the kernel of ,
are congruent to each other modulo p
j
.
Let p
i
be the highest power of p which divides n. Our hypothesis
p
f
1 (mod u

) means that D(x

) = D(x
p
f

) in Z[H] (where
we will do our calculations), so we have
(28.2) D(x

)D(x
p
f

) =
D(x

)D(x

) = n +
v
u
H(x) 0 (mod p
i
, H).
Let p
j
be the highest power of p so that D(x

) 0 (mod p
j
, H);
say D(x

) p
j
A(x) (mod H). The reader may check that (28.2)
implies that 2j i and that if 2j < i, then A(x)A(x
p
f
) 0
(mod p, H). But then Lemmas 28.2 and 28.6 imply in turn that
A(x)
1+p
f
0 (mod p, H) and then A(x) 0 (mod p, H). This
contradicts the choice of j, so 2j = i.
Consequences of Theorem 28.8 are: if v is divisible by 3, then
all prime divisors of the squarefree part of n are congruent to 0
or 1 (modulo 3); if v is divisible by 5, then all prime divisors of
the squarefree part of n are congruent to 0 or 1 (modulo 5); if v
is divisible by 7, then all prime divisors of the squarefree part of n
are congruent to 0, 1, 2, or 4 (modulo 7).
394 A Course in Combinatorics
Example 28.4. We give an application of the stronger claim given
in the proof of Theorem 28.8. Consider a hypothetical (154, 18, 2)-
dierence set D in G. Let : G H be a homomorphism onto a
group H of order u := 11. Take p := 2. Since 2
5
1 (mod 11),
we conclude that D(x

) 0 (mod 4, H). This means the 11 co-

ecients of D(x

) are all congruent to some integer h modulo 4;

since the 11 coecients sum to 18, h is 2. But then the coecients
would sum to at least 22 and this contradiction shows that no such
dierence sets exist.
Problem 28H. Let D be a (q
4
+q
2
+1, q
2
+1, 1)-dierence set in an
abelian group G. Let : G H be a homomorphism onto a group
H of order u := q
2
q +1. Show that any coset of the kernel of is
the point set of a Baer subplane of the projective plane of order q
2
arising from D. For example, the 21 points of the projective plane
of order 4 are partitioned into three Fano congurations.
Problem 28I. Let D be a (q
3
+q
2
+q+1, q
2
+q+1, q+1)-dierence
set in an abelian group G. Let : G H be a homomorphism
onto a group H of order u := q +1. Show that the translates of D
meet the cosets of the kernel of in q +1 points or a single point.
Show further that if the symmetric design arising from D consists
of the points and planes of PG
3
(q), then the cosets are ovoids (see
Example 26.5).
Notes.
The celebrated Multiplier Theorem was rst proved in the case
of cyclic planar dierence sets in Hall (1947). It was generalized to
> 1 by Hall and Ryser (1951) and since then has been extended
in many wayssee e.g. Mann (1965), Baumert (1971), and Lander
(1983). Algebraic number theory and characters of abelian groups
often play a role in the proofs of these results, but we have chosen
here to give proofs of Theorem 28.7 and 28.8 using only the group
ring.
Marshall Hall (19101990) did fundamental work in group theory
and coding theory as well as on combinatorial designs. He has
been a tremendous inuence on many mathematicians, including
the authors.
28. Dierence sets and the group ring 395
Theorem 28.8 is due to K. Yamamoto (1963).
References.
L. D. Baumert (1971), Cyclic Dierence Sets, Lecture Notes in
Math. 182, Springer-Verlag.
M. Hall (1947), Cyclic projective planes, Duke J. Math. 14, 1079
1090.
M. Hall and H. J. Ryser (1951), Cyclic incidence matrices, Canad.
J. Math. 3, 495502.
E. S. Lander (1983), Symmetric Designs: An Algebraic Approach,
London Math. Soc. Lecture Note Series 74, Cambridge Univer-
sity Press.
H. B. Mann (1965), Addition Theorems, Wiley.
K. Yamamoto (1963), Decomposition elds of dierence sets, Pa-
cic J. Math. 13, 337352.
29
Codes and symmetric designs
In this chapter, we elaborate on some of the material introduced in
Chapter 20. We saw in that chapter that the rows of the incidence
matrix of a projective plane of order n 2 (mod 4) span a binary
code which, when extended, is self-dual. It was observed by H. A.
Wilbrink (1985) and others that that result can be used to show
that planar dierence sets do not exist for values of n > 2 with
n 2 (mod 4); see Theorem 29.7.
One can consider the code spanned by the rows of the incidence
matrix over other prime elds F
p
as well. With essentially the same
proof as that of Theorem 20.6, we have the following theorem.
Theorem 29.1. If p divides n := k , then the F
p
-span C of the
rows of the incidence matrix N of a symmetric (v, k, )-design has
dimension at most (v + 1)/2 over F
p
. If (p, k) = 1 and p
2
does not
divide n, then the dimension of this p-ary code is exactly (v +1)/2.
Problem 29A. Prove Theorem 29.1.
It is not possible, in general, to extend the code C to get a code
of length v + 1 that is self-orthogonal with respect to the standard
dot product on F
v
p
. But let us consider other scalar products for
odd primes p.
To a nonsingular mmmatrix B over a eld F, we may associate
the scalar product (or bilinear form)
x, y) := xBy

for x, y F
m
. For a subspace C of F
m
, let
C
B
:= x : x, y) = 0 for all y C.
29. Codes and symmetric designs 397
Then C and C
B
have complementary dimensions and (C
B
)
B
=
C. We say that C is totally isotropic when C C
B
; this is the
appropriate terminology for this generalization of self-orthogonal.
We have use for the following theorem of Witt, and we include a
proof here as it is essentially a rephrasing of part of Theorem 26.6.
Theorem 29.2. Given a symmetric nonsingular matrix B over a
eld F of odd characteristic, there exists a totally isotropic subspace
of dimension m/2 in F
m
if and only if (1)
m/2
det(B) is a square
in F.
Proof: A totally isotropic subspace of dimension m/2 is, when
considered as a set of projective points in PG(m 1, 2), a at of
projective dimension (m/2) 1 that is contained completely in the
quadric Q dened by the nondegenerate quadratic form
f(x) = xBx

.
By Theorem 26.6, such a subspace exists if and only if Q is hyper-
bolic. It remains only to verify that Q is hyperbolic if and only if
(1)
m/2
det(B) is a square in F.
If f
1
is a form projectively equivalent to f, then f
1
(x) = xB
1
x

where B
1
= UBU

for some nonsingular matrix U. Of course,

(1)
m/2
det(B) is a square if and only if (1)
m/2
det(B
1
) is a square.
If Q is hyperbolic, then f is equivalent to f
1
= x
1
x
2
+ +x
m1
x
m
and B
1
is block diagonal with m/2 matrices
W =
1
2
_
0 1
1 0
_
on the diagonal. Then (1)
m/2
det(B
1
) = 1/2
m
, which is a square.
If Q is elliptic, then f is equivalent to f
1
= x
1
x
2
+ +x
m3
x
m2
+
p(x
m1
, x
m
) where p(x, y) = ax
2
+ 2bxy + cy
2
is an irreducible
quadratic form over F, and here B
1
is block diagonal with (m/2)1
matrices W and one matrix
_
a b
b c
_
on the diagonal. Then (1)
m/2
det(B
1
) = (b
2
ac)/2
m2
, and this
is a nonsquare in F since p is irreducible.
398 A Course in Combinatorics
It will be convenient to retain the term self-dual for a totally
isotropic subspace of dimension equal to half its length. E. S. Lan-
der (1983) has shown how to associate a family of p-ary codes of
length v + 1 and a scalar product to a symmetric design in such a
way that one of the codes is self-dual when n is exactly divisible
by an odd power of p. Theorem 29.2 then gives us a condition on
the parameters of the symmetric design. It turns out that these
conditions are already consequences of the BruckRyserChowla
theorem, Theorem 19.11. In some sense, then, these self-dual codes
provide a combinatorial interpretation of part of the BRC theo-
rem. In any case, these codes carry information about the design
and have further applications in the theory of symmetric designs
see Lander (1983).
Theorem 29.3. Suppose there exists a symmetric (v, k, )-design
where n is exactly divisible by an odd power of a prime p. Write
n = p
f
n
0
(f odd) and = p
b

0
with (n
0
, p) = (
0
, p) = 1. Then
there exists a self-dual p-ary code of length v + 1 with respect to
the scalar product corresponding to
B =
_
diag(1, 1, . . . , 1,
0
) if b is even,
diag(1, 1, . . . , 1, n
0

0
) if b is odd.
Hence from Theorem 29.2,
_
(1)
(v+1)/2

0
is a square (mod p) if b is even,
(1)
(v+1)/2
n
0

0
is a square (mod p) if b is odd.
Towards the proof of Theorem 29.3, we rst prove two proposi-
tions.
Given any integral m m matrix A, we may consider the Z-
module M(A) consisting of all integral linear combinations of its
rows; that is,
M(A) := yA: y Z
m
.
Fix a prime p and for any positive integer i dene modules
M
i
:= x Z
m
: p
i
x M(A),
N
i
:= y Z
m
: Ay

0 (mod p
i+1
).
29. Codes and symmetric designs 399
We have M
0
= M(A), M
i
M
i+1
and N
i
N
i+1
for all i. Let
(29.1) C
i
:= M
i
(mod p), D
i
:= N
i
(mod p).
That is, read all the integer vectors in M
i
or N
i
modulo p to obtain
C
i
or D
i
. Then each C
i
and D
i
is a subspace of the vector space
F
m
p
, i.e. a p-ary linear code. Clearly,
C
0
C
1
C
2
. . . and D
0
D
1
D
2
. . . .
Proposition 29.4. We have
C

i
= D
i
for all nonnegative integers i.
Proof: Let x and y be integral vectors such that x (mod p) C
i
and y (mod p) D
i
. This means
p
i
(x +pa) = zA and A(y +pb)

0 (mod p
i+1
)
for some integral vectors a, b, and z. Then
p
i
(x +pa) (y +pb)

= zA(y +pb)

0 (mod p
i+1
)
which implies that x y

= 0 over F
p
.
We complete the proof by showing that C
i
and D
i
have dimen-
sions which add to m. There exist unimodular matrices E and
F (integral matrices with integral inverses) such that S := EAF
is diagonal with integral diagonal entries d
1
, d
2
, . . . , d
m
which suc-
cessively divide one another: d
1
[d
2
[ . . . [d
m
. (S is the Smith normal
form of A and the d
i
s are the invariant factors.) The reader should
verify that the modules M
i
and N
i
are equivalent to
M
/
i
:= x : p
i
x M(S) and
N
/
i
:= y : Sy

0 (mod p
i+1
),
respectively, in the sense that either can be obtained from the other
by application of a unimodular transformation. Hence the dimen-
sions over F
p
of M
i
and M
/
i
are equal, as well as the dimensions of
N
i
and N
/
i
.
400 A Course in Combinatorics
Suppose p
i+1
does not divide d
1
, . . . , d
t
but that p
i+1
does divide
d
t+1
, . . . , d
m
. Then y N
/
i
implies that the rst t coordinates of y
are divisible by p; any vector y with 0s in the rst t coordinates
is in N
/
i
. Thus N
/
i
(mod p) is the span of the last m t standard
basis vectors and has dimension mt. Also, x M
/
i
implies that
the last d t coordinates of x are divisible by p and a little more
thought shows that M
/
i
(mod p) is the span of the rst t standard
basis vectors and has dimension t.
Proposition 29.5. Let A, B, and U be m m integral matrices
with
(29.2) ABA

= nU,
and where U and B are nonsingular modulo a prime p. Write
n = p
e
n
0
where (p, n
0
) = 1. Dene the sequence C
i
of p-ary codes
from A as in (29.1). Then C
e
= F
m
p
and
C
B
i
= C
ei1
for i = 0, 1, . . . , e 1.
In particular, if e is odd, then C1
2
(e1)
is a self-dual p-ary code with
respect to the scalar product given by B on F
m
p
.
Proof: Let x and y be integral vectors such that x (mod p) C
i
and y (mod p) C
ei1
. This means
p
i
(x +pa
1
) = z
1
A and p
ei1
(y +pa
2
) = z
2
A
for some integral vectors z
1
, z
2
, a
1
, and a
2
. Then
p
e1
x, y) = p
e1
xBy

z
1
ABA

z
2

0 (mod p
e
)
in view of (29.2). Thus x, y) = 0 in F
p
and we see C
ei1
C
B
i
.
Now let x C
B
i
. This means xB C

i
, which is D
i
by Propo-
sition 29.4, and so for some integral vector x
/
which reduces to x
when read modulo p,
x
/
BA

0 (mod p
i+1
).
29. Codes and symmetric designs 401
From (29.2), A BA

U
1
= nI, and since a matrix commutes with
its inverse,
(29.3) BA

U
1
A = nI.
Since U is nonsingular modulo p, dU
1
is integral for some d prime
to p, e.g. d := det(U). We multiply (29.3) on the left by dx
/
to get
x
/
BA

(dU
1
)A = p
e
dn
0
x
/
,
and then
zA = p
ei1
dn
0
x
/
where z :=
1
p
i+1
x
/
BA

(dU
1
) is integral. This means p
ei1
dn
0
x
/
is in M
ei1
and hence x C
ei1
.
The assertion that C
e
= F
m
p
is left as an easy problem.
Problem 29B. Prove that C
e
= F
m
p
.
Proof of Theorem 29.3: Let N be the incidence matrix of a
symmetric (v, k, )-design and let p be a prime. Assume = p
2a

0
where (
0
, p) = 1 and a 0; we will explain later what to do when
is exactly divisible by an odd power of p. Let
(29.4)
A :=
_
_
_
_
p
a
N
.
.
.
p
a
p
a

0
p
a

0
k
_
_
_
_
, B :=
_
_
_
_
1 0
.
.
.
1
0
0
_
_
_
_
.
The reader should verify, using the properties of N and the relation
(v 1) = k(k 1), that ABA

= nB.
In the case is exactly divisible by an even power of p, we apply
Proposition 29.5 with the matrices A and B as in (29.4), and where
U := B.
If is exactly divisible by an odd power of p, we apply the above
case to the complement of the given symmetric design, which is a
symmetric (v, v k,
/
)-design where
/
= v 2k+. Say
/
= p
c

/
0
where (
/
0
, p) = 1. From
/
= n(n 1), it follows that c is odd
and that

/
0
= n
0
(n 1) n
0
(mod p).
402 A Course in Combinatorics
We have replaced what would be
/
0
in the conclusion by
0
n
0
,
which is allowed since they dier by a square factor modulo p, in
order to express the result in terms of the original parameters.
The following theorem is a consequence of Problem 19M, but we
give a proof similar to that of Theorem 29.3.
Theorem 29.6. If there exists a conference matrix of order n 2
(mod 4), then no prime p 3 (mod 4) can divide the square-free
part of n 1.
Proof: A conference matrix of order n is, in particular, an integral
matrix A such that AA

= (n 1)I
n
. By Proposition 29.5 with
A = U = I, there exists a self-dual p-ary code (self-dual with
respect to the standard inner product) of length n for every prime
divisor p of the square-free part of n1. Theorem 29.2 then implies
that every such prime is 1 (mod 4).
* * *
When p is a prime not dividing the order v of an abelian group
G, the group ring F
p
[G] is a semi-simple algebra over F
p
= Z
p
. This
means that there exist no nonzero nilpotents, i.e. nonzero elements
a such a
m
= 0 for some positive integer m. The nonexistence
of nonzero nilpotents is proved in Lemma 28.5. By Wedderburns
theorem, every nite-dimensional commutative semi-simple algebra
with identity / over a eld F is isomorphic to the direct product
of elds, each an extension of F. It follows that each ideal 1 of /
is principal and is generated by an idempotent e, i.e. an element e
with e
2
= e. See any advanced text on algebra for proofs.
We do not need all of this information, but here are some facts
about principal ideals which are generated by idempotents that we
will need. These make good exercises.
First, if 1 = e
1
) and also 1 = e
2
), where both e
1
and e
2
are
idempotent, then e
1
= e
2
. Suppose 1
1
= e
1
) and 1
2
= e
2
), where
e
1
and e
2
are idempotent. Then
1
1
1
2
= e
1
e
2
) and 1
1
+1
2
= e
1
+e
2
e
1
e
2
).
Note that e
1
e
2
and e
1
+e
2
e
1
e
2
are again idempotents.
29. Codes and symmetric designs 403
Theorem 29.7. Let D be an (n
2
+n+1, n+1, 1)-dierence set in
an abelian group G of order v := n
2
+n +1. If n 0 (mod 2) but
n , 0 (mod 4), then n = 2. If n 0 (mod 3) but n , 0 (mod 9),
then n = 3.
Proof: Let D be a (n
2
+n+1, n+1, 1)-dierence set in an abelian
group G and let p be a prime divisor of n. By Theorem 28.3, p is
a multiplier of D and we will assume from now on that D is xed
by p. We work in the F
p
-algebra F
p
[G].
Let 1
1
be the ideal in F
p
[G] generated by D(x) and 1
2
the ideal
generated by D(x
1
).
We have D(x
p
) = D(x), but D
p
(x) = D(x
p
) by Lemma 28.2, so
D
p
(x) = D(x) in F
p
[G]. Then D
p1
(x) is idempotent and will also
generate 1
1
. Similarly, D
p1
(x
1
) is an idempotent generator for
1
2
. So the idempotent generator of 1
1
1
2
is
D
p1
(x)D
p1
(x
1
) = (n +G(x))
p1
= G(x),
and the idempotent generator of 1
1
+1
2
is
D
p1
(x) +D
p1
(x
1
) G(x).
We now wish to consider the dimensions of 1
1
and 1
2
over F
p
.
In general, the rank of a principal ideal generated by A(x) is the
rank of the v v matrix whose rows are the coecients of x
g
A(x),
g G. This matrix is the incidence matrix of a symmetric (n
2
+
n + 1, n + 1, 1)-design when A(x) = D(x) or D(x
1
). If we now
assume that p
2
does not divide n, then by Theorem 29.1, 1
1
and 1
2
have dimension (v + 1)/2. The intersection 1
1
1
2
has dimension
1, so their sum must have dimension v. The idempotent generator
of the whole group ring, as an ideal in itself, is 1; so we conclude
that
(29.5) D
p1
(x) +D
p1
(x
1
) G(x) = 1 in F
p
[G].
We are only able to exploit the above equation when p = 2 or 3.
When p = 2, (29.5) asserts D(x) +D(x
1
) 1 +G(x) (mod 2) in
Z[G]. The number of odd coecients of 1 +G(x) is v 1 = n
2
+n,
404 A Course in Combinatorics
but the number of odd coecients of D(x) +D(x
1
) cannot exceed
2(n + 1); it follows that n 2.
When p = 3, (29.5) asserts D
2
(x) +D
2
(x
1
) 1+G(x) (mod 3)
in Z[G]. We claim that the nonzero coecients of D
2
(x) and
D
2
(x
1
) consist of n + 1 coecients that are 1s and
_
n+1
2
_
coef-
cients that are 2s. If C is any planar dierence set, there will
be a term x
2g
in C
2
(x) for each g C and a term 2x
g+h
for each
unordered pair g, h C; note that = 1 implies, e.g., that
g
1
+ h
1
,= g
2
+ h
2
unless g
1
, h
1
= g
2
, h
2
. The sum of two such
group ring elements cannot have more than
_
n+1
2
_
+ 2(n + 1) coef-
cients that are 1 (mod 3), but 1 +G(x) has n
2
+n coecients
that are = 1; it follows that n 4.
Problem 29C. Suppose D is a dierence set with n 2 (mod 4)
and that 2 is a multiplier of D. What are the parameters of D, as
functions of n?
Notes.
Some of the material preceding the proof of Theorem 29.3 is more
elegant when p-adic numbers are introduced as in Lander (1983),
but we have chosen to present the material without them.
References.
D. Jungnickel and K. Vedder (1984), On the geometry of planar
dierence sets, European J. Combinatorics 5, 143148.
E. S. Lander (1983), Symmetric Designs: An Algebraic Approach,
London Math. Soc. Lecture Note Series 74, Cambridge Univer-
sity Press.
V. Pless (1986), Cyclic projective planes and binary extended cyclic
self-dual codes, J. Combinatorial Theory (A) 43, 331333.
H. A. Wilbrink (1985), A note on planar dierence sets, J. Combi-
natorial Theory (A) 38, 9495.
30
Association schemes
Given two k-subsets A, B of an n-set, n 2k, there are k + 1
possible relations between them: they may be equal, they may
intersect in k 1 elements, they may intersect in k 2 elements,
. . . , or they may be disjoint.
Given two words (k-tuples) a, b A
k
, where A is an alphabet of
size at least 2, there are k+1 possible relations between them: they
may be equal, they may agree in k 1 coordinates, they may agree
in k 2 coordinates, . . . , or they may disagree in all coordinates.
These instances of a set together with a list of mutually exclu-
sive and exhaustive binary relations are examples of association
schemes, which we dene shortly. Association schemes provide one
of the foundations of combinatorics and so we include this chapter
even though it will be dicult reading. They have been implicit in
many of the previous chapters; we have explicitly discussed 2-class
association schemes, as they are equivalent to the strongly regular
graphs discussed in Chapter 21. This chapter elaborates on some
of the material of Chapter 21 but has dierent goals.
Association schemes arose rst in the statistical theory of design
of experiments, but the work of Ph. Delsarte (1973) has shown how
they serve to unify many aspects of our subject. In particular, cer-
tain results of coding theory and the theory of t-designswhich
were originally discovered independentlyare now seen to be for-
mally dual aspects of the same ideas in association schemes. For
example, Fishers inequality and its generalization, Theorem 19.8,
is formally dual to the sphere packing bound, Theorem 21.1. We
use the machinery of association schemes in this chapter to give
proofs of Lloyds theorem on perfect codes and its formal dual theo-
rem for tight designs and orthogonal arrays. Delsartes inequalities,
406 A Course in Combinatorics
Theorem 30.3, on the distribution vector of a subset of an associ-
ation scheme provide a linear programming bound on the size of
codes and are also of interest in extremal set theory.
By a binary relation on a set X, we mean a subset of X X. A
k-class association scheme, sometimes we say just scheme, on a set
X of points consists of k + 1 nonempty symmetric binary relations
R
0
, R
1
, . . . , R
k
on X which partition X X, where R
0
= (x, x) :
x X is the identity relation, and such that for some nonnegative
integers p

ij
, 0 , i, j k, the following system of axioms holds:
given any (x, y) R

, there are exactly p

ij
elements z X such
that (x, z) R
i
and (z, y) R
j
. We say x, y X are i-th associates
when (x, y) R
i
.
The numbers p

ij
, 0 , i, j k, are the parameters of the
scheme. That p
0
ii
exists means that there is a constant number
of i-th associates of any element of X, which is usually denoted by
n
i
. We have
(30.1) p
0
ii
= n
i
and p
0
ij
= 0 for i ,= j
and
n
0
= 1, n
0
+n
1
+ +n
k
= N.
where N := [X[. The numbers n
0
, n
1
, . . . , n
k
are called the degrees
of the scheme.
Example 30.1. The Johnson schemes J(v, k). The points of the
scheme J(v, k) are the
_
v
k
_
k-subsets of a v-set S. Two k-subsets
A, B are declared to be i-th associates when [AB[ = k i. Thus
0-th associates are equal. The parameters p

ij
exist by symmetry
and may be expressed as sums of products of binomial coecients,
but we will not bother to write these out in general.
The scheme J(6, 3), for example, is a 3-class scheme with 20
points. The reader should check that n
1
= n
2
= 9, n
3
= 1. A few
of the other parameters are p
1
11
= 4, p
2
11
= 4, p
3
11
= 0.
Example 30.2. The Hamming schemes H(n, q). The points of
H(n, q) are the q
n
words of length n over an alphabet of size q.
Two n-tuples x, y are declared to be i-th associates when they dis-
agree in exactly i coordinates. Thus 0-th associates are equal. The
30. Association schemes 407
parameters p

ij
exist by symmetry and may be expressed as sums
of products of binomial coecients and powers of q 1, but we will
not bother to write these out in general.
The scheme H(5, 3), for example, is a 5-class scheme with 125
points. The reader should check that n
1
= 5 2, n
2
= 10 4, etc.
Each of the relations R
i
may be thought of as the adjacency rela-
tion of a graph G
i
on the vertex set X. (A scheme is a special kind
of partition of the edgesor coloring of the edgesof a complete
graph.) It should be clear that if we start with a 2-class scheme,
then G
1
is a strongly regular graph with degree n
1
, = p
1
11
, and
= p
2
11
. In fact, any strongly regular graph gives rise to a 2-
class association scheme when we declare that two distinct vertices
are 1st associates when they are adjacent in G and 2nd associates
if not adjacent; the other parameters exist, i.e. are constant, and
can be computed from the parameters of the graph. For example,
p
1
12
= k 1.
A distance regular graph is a graph G such that the number of
vertices at distance i to x and distance j to y depends only on the
distance between the vertices x and y, not on the particular ver-
tices. That is, dening two vertices to be i-th associates if and only
if their distance in G is i produces an association scheme (where the
number of classes will be the diameter of the graph). See Brouwer,
Cohen, and Neumaier (1989). Schemes which arise in this way are
called metric. The schemes mentioned in Examples 30.1 and 30.2
above are metric, as are those in Examples 30.3 and 30.4 below.
It is important to know that the parameters p

ij
exist; it is not so
important to know their exact values. This is lucky because there is
often no convenient expression for these parameters. In the scheme
J(v, k), for example, a triple sum of binomial coecients seems to
be required for the general p

ij
. In the examples we have just given,
we know that the p

ij
s exist because of symmetry. More precisely,
in each case there exists a group G of permutations of X so that two
ordered pairs of points (x
1
, y
1
) and (x
2
, y
2
) are in the same relation
R
i
if and only if there exists G such that (x
1
) = x
2
and
(y
1
) = y
2
. That is, the relations R
0
, R
1
, . . . , R
k
are the orbits of G
on XX (with R
0
the trivial orbit of all (x, x), x X). In Example
30.1, G is the symmetric group S
v
acting on all k-subsets of a v-set;
408 A Course in Combinatorics
we can nd a permutation which takes an ordered pair of k-subsets
to another if and only if the size of the intersection of each pair is
the same. In Example 30.2, G is the wreath product of S
q
with S
n
(that is, we allow any permutation of the n coordinates followed by
independent permutations of the q symbols in each coordinate); we
can nd such a transformation which takes an ordered pair of n-
tuples to another if and only if the pairs agree in the same number
of coordinates.
In general, if G is a transitive group of permutations on a set X
such that the orbits of G on X X are symmetric, they may be
taken as the relations of an association scheme on X. Our next
three examples also arise in this way.
Example 30.3. This is the q-analogue of the Johnson scheme:
Take the k-subspaces of a v-space V over F
q
as points. Two k-
subspaces A, B are declared to be i-th associates when dim(A
B) = k i.
Example 30.4. Take the k m matrices over F
q
as the points,
where k m, say. Two matrices A, B are declared to be i-th
associates when the rank of AB is k i.
To put this example into the framework above, we let X be the
set of all k m matrices and we may take G to be the set of all
permutations X UXW + C as U ranges over the nonsingular
k k natrices, W over the nonsingular m m matrices, and C
over the k m matrices. If (A, B) and (A
/
, B
/
) are pairs of k
m matrices such that the ranks of A B and A
/
B
/
are equal,
then U(A B)W = A
/
B
/
for some nonsingular U and W, and
then X UXW + (B
/
UBW) will map the rst pair onto the
second.
Example 30.5. The cyclotomic schemes are obtained as follows.
Let q be a prime power and k a divisor of q 1. Let C
1
be
the subgroup of the multiplicative group of F
q
of index k, and let
C
1
, C
2
, . . . , C
k
be the cosets of C
1
. The points of the scheme are to
be the elements of F
q
, and two points x, y are declared to be i-th
associates when x y C
i
(and 0-th associates when x y = 0).
In order to have a scheme with the above denition, we require
1 C
1
so that the relations will be symmetric, i.e. 2k must di-
30. Association schemes 409
vide q 1 if q is odd. Cf. Example 21.3 which is the case k = 2.
We introduce the association matrices A
0
, A
1
, . . . , A
k
(also called
the adjacency matrices) of an association scheme. These matrices
are square with both rows and columns indexed by the elements of
the point set X of a scheme. For i = 0, 1, . . . , k, we dene
A
i
(x, y) =
_
1 if (x, y) R
i
,
0 otherwise.
The matrices A
i
are symmetric (0, 1)-matrices and
A
0
= I, A
0
+A
1
+ +A
k
= J
where J is the all-one matrix of size N N. We denote by A
the linear span over the reals of A
0
, A
1
, . . . , A
k
. These matrices
are linearly independent since each contains at least one 1; and a
position in which A
i
has a 1 contains a 0 in every other association
matrix. The axioms of an association scheme are exactly what is
required to ensure that A is closed under matrix multiplication. To
see this, it suces to show that the product of any two of the basis
matrices is in A and, in fact, we have
(30.2) A
i
A
j
=
k

=0
p

ij
A

because A
i
A
j
(x, y) is the number of z such that A
i
(x, z) = 1 and
A
j
(z, y) = 1, and this number is p

ij
where is such that A

(x, y) =
1.
The algebra A is called the BoseMesner algebra of the scheme;
this algebra was introduced for strongly regular graphs in Chapter
21. We note at this point that not only is A closed under normal
matrix multiplication, but it is also closed under Hadamard mul-
tiplication as introduced in Problem 21E. The Hadamard product
A B of two matrices is the matrix obtained by coordinate-wise
multiplication:
(A B)(x, y) := A(x, y)B(x, y).
410 A Course in Combinatorics
As an algebra with respect to Hadamard multiplication, A is
almost trivial. We have
A
i
A
j
=
_
A
i
if i = j,
O if i ,= j
(that is, A
0
, A
1
, . . . , A
k
are orthogonal idempotents), and the sum of
the A
i
s is J, the identity with respect to Hadamard multiplication.
So Hadamard multiplication is extremely simple when matrices in
A are expressed with respect to the basis A
0
, A
1
, . . . , A
k
of A.
However, a well known result of matrix theory (an extension
of the spectral theorem which says that a symmetric real matrix
has an orthogonal basis of eigenvectors) asserts that a commuta-
tive algebra of real symmetric matrices has a basis of orthogonal
idempotents with respect to ordinary matrix multiplication which
sum to the identity. More geometrically, there exists an orthogonal
decomposition
1
X
= V
0
V
1
V
k
of the Euclidean space 1
X
, the space of all vectors whose coordi-
nates are indexed by the elements of X with the standard inner
product, such that the orthogonal projections E
0
, E
1
, . . . , E
k
from
1
X
onto the subspaces V
0
, V
1
, . . . , V
k
, respectively, are a basis for
A. We have
E
i
E
j
=
_
E
i
if i = j,
O if i ,= j,
and
E
0
+E
1
+ +E
k
= I.
Of course, when matrices are expressed with respect to the basis
E
0
, E
1
, . . . , E
k
, ordinary multiplication is also extremely simple.
The subspaces V
0
, V
1
, . . . , V
k
are called the eigenspaces of the
scheme: in a linear combination M =

k
i=0

i
E
i
, each vector in
V
i
is an eigenvector of value
i
for M. There is no natural number-
ing of the eigenspaces in general with one exception: since J A,
and J has the vector j of all 1s as an eigenvector of value N and all
vectors orthogonal to j as eigenvectors of value 0, it must be that
one of the eigenspaces consists of scalar multiples of j alonewe
shall always assume that this is V
0
. Then the orthogonal projection
30. Association schemes 411
onto V
0
(which has j as an eigenvector of value 1 and all vectors
orthogonal to j as eigenvectors of value 0) is
E
0
=
1
N
J.
We let m
i
denote the dimension of V
i
. Then
m
0
= 1, m
0
+m
1
+ +m
k
= N.
Note that m
i
is the trace of E
i
since the eigenvalues of E
i
are 1
(with multiplicity equal to the dimension of V
i
) and 0. The numbers
m
0
, m
1
, . . . , m
k
are called the multiplicities of the scheme.
Example 30.6. We can explicitly describe the eigenspaces of the
Hamming scheme H(n, 2). The points of this scheme are the binary
n-tuples (or words) a of F
n
2
.
For each a F
n
2
, dene a vector v
a
with coordinates indexed by
the point set by
v
a
(b) := (1)
a,b)
.
These vectors are orthogonal and are the rows of a Hadamard ma-
trix; see Fig. 30.1 and Chapter 18. We claim that V
i
may be taken
to be the span of all vectors v
a
as a ranges over the words of weight
i, i = 0, 1, . . . , n.
We check that each v
a
is an eigenvector of all association matrices
A
j
. Let a have weight . Then
(v
a
A
j
)(b) =

c
v
a
(c)A
j
(c, b)
=

c:d(b,c)=j
(1)
a,c)
= (1)
a,b)

c:d(b,c)=j
(1)
a,b+c)
= v
a
(b)

u:wt(u)=j
(1)
a,u)
= v
a
(b)
n

i=0
(1)
i
_

i
__
n
j i
_
.
This calculation shows, with E
i
the matrix of the orthogonal
projection onto V
i
, that
A
j
=
n

=0
_
n

i=0
(1)
i
_

i
__
n
j i
_
_
E

412 A Course in Combinatorics

since the two sides give the same result when premultiplied by any
v
a
. So A spanE
0
, . . . , E
n
. Equality holds because both spaces
have dimension n + 1.
The eigenvalues of the schemes H(n, q) for general q are given in
Theorem 30.1 below.
Figure 30.1
For example, if we order the words of length n = 3 as
000, 100, 010, 001, 011, 101, 110, 111,
then the eigenspaces V
0
, V
1
, V
2
, V
3
of the cube (= A
1
of H(3, 2))
shown at the right of Fig. 30.1 are spanned respectively by the rst
row, the next three rows, the next three rows, and the last row of
the Hadamard matrix at the left of Fig. 30.1.
Example 30.7. We now describe the eigenspaces of the Johnson
scheme J(v, k). We omit the proofs; see Delsarte (1973) or Wilson
(1984b).
The point set of this scheme is the set X of k-subsets S of a v-set
X. For each subset T X of size k, let e
T
be the vector of
length
_
v
k
_
in 1
X
where
e
T
(S) :=
_
1 if T S,
0 otherwise.
Let U
i
be the span of e
T
: T X, [T[ = i for i = 0, 1, . . . , k. We
claim that U
0
U
1
U
k
= 1
X
and that U
i
has dimension
_
v
i
_
. Let V
0
:= U
0
(the constant vectors) and for i > 0, let V
i
be the
orthogonal complement of U
i1
in U
i
, i.e.
V
i
:= U
i
U

i1
.
30. Association schemes 413
It is clear that V
0
, V
1
, . . . , V
k
are orthogonal and sum to 1
X
. It can
be shown that each vector in V
i
is an eigenvector of A
j
of value
P
j
(i) as displayed in Theorem 30.1(i) below.
Since we have two bases of the vector space A, we may consider
the transformation matrices between them, which are called the
eigenmatrices of the scheme. Dene P (the rst eigenmatrix) and
Q (the second eigenmatrix) as the k +1 k +1 matrices with rows
and columns indexed by 0, 1, . . . , k such that
(A
0
, A
1
, . . . , A
k
) = (E
0
, E
1
, . . . , E
k
) P
and
N (E
0
, E
1
, . . . , E
k
) = (A
0
, A
1
, . . . , A
k
) Q.
We write P

(i) for the (i, ) entry of P, and similarly, we let Q

(i)
be the (i, ) entry of Q, so that
(30.3) A

= P

(0)E
0
+P

(1)E
1
+ +P

(k)E
k
,
and
(30.4) N E

= Q

(0)A
0
+Q

(1)A
1
+ +Q

(k)A
k
.
Of course, we have
Q = N P
1
, P = N Q
1
.
The -th column of P consists of the eigenvalues of A

.
While it is not so important to know the parameters p

ij
of a
scheme, it is important for applications to know the eigenmatrices
P and Q. See Bannai and Ito (1984) or Delsarte (1973) for proofs of
the following theorem and the determination of the eigenmatrices
of other schemes.
Theorem 30.1. (i) For the Johnson scheme J(v, k), the degrees
are n

=
_
k

__
vk

_
and the multiplicities are m

=
_
v

_
v
1
_
, =
0, 1, . . . , k. The entries of the rst eigenmatrix are P

(i) where
P

(x) =

=0
(1)

_
k

__
k x

__
v k + x

_
.
414 A Course in Combinatorics
(ii) For the Hamming scheme H(n, q), the degrees and multiplic-
ities are n

= m

=
_
n

_
(q 1)

, = 0, 1, . . . , n. The entries of the

rst eigenmatrix are P

(i) where
P

(x) =

=0
(q)

(q 1)

_
n

__
x

_
.
Problem 30A. Calculate the eigenmatrix P of J(8, 3) in the fol-
lowing way. First calculate A
1
A
2
as a linear combination of A
0
, A
1
,
A
2
, and A
3
, and then ll in the missing line in the following table:
A
0
1
= A
0
,
A
1
1
= A
1
,
A
2
1
= 15A
0
+ 6A
1
+ 4A
2
,
A
3
1
= ,
A
4
1
= 1245A
0
+ 1036A
1
+ 888A
2
+ 720A
3
.
From this table, it is only mildly tedious to derive the minimal
polynomial of A
1
. Find the eigenvalues of A
1
. Now express A
2
and
A
3
as polynomials in A
1
, and so nd their eigenvalues. Find the
multiplicities. Check your work by verifying the orthogonality rela-
tions of Theorem 30.2 or just calculating the values from Theorem
30.1(i).
Problem 30B. Show how to calculate all the parameters p

ij
of
a scheme given the eigenmatrix P. That is, prove that they are
uniquely determined by P.
Problem 30C. The Latin square graphs are srg(v, k, , )s where
for some integers n and r,
v = n
2
, k = r(n1), = (n2)+(r1)(r2), = r(r1).
These were introduced in Example 21.7 for r = 3. Find the eigen-
matrices P and Q for the 2-class schemes corresponding to these
graphs.
The so-called negative Latin square graphs are strongly regular
graphs srg(v, k, , ) whose parameters are obtained from the above
30. Association schemes 415
by replacing n by n and r by r. So v = (n)
2
, k = (r)(n1),
etc. (It is strange that this should yield parameters satisfying
(21.4), but it does.) Find the eigenmatrices P and Q for the corre-
sponding 2-class schemes.
Theorem 30.2. The eigenmatrices of a scheme satisfy the orthog-
onality relations
P

_
_
_
1 0 0
0 m
1
0
.
.
.
.
.
.
.
.
.
0 0 m
k
_
_
_
P = N
_
_
_
1 0 0
0 n
1
0
.
.
.
.
.
.
.
.
.
0 0 n
k
_
_
_
and
Q

_
_
_
1 0 0
0 n
1
0
.
.
.
.
.
.
.
.
.
0 0 n
k
_
_
_
Q = N
_
_
_
1 0 0
0 m
1
0
.
.
.
.
.
.
.
.
.
0 0 m
k
_
_
_
.
Proof: The vector space A can be equipped with an inner product
in a more-or-less natural way: We dene A, B) to be the sum of
the entries of the Hadamard product A B. Check that this is the
same as the trace of the matrix product AB

(which is the same

as tr(AB) when B is symmetric, of course).
With respect to this inner product, the basis A
0
, A
1
, . . . , A
k
is an
orthogonal basis (but not orthonormal); we have A
i
, A
i
) = N n
i
.
But E
0
, E
1
, . . . , E
k
is also an orthogonal basis since E
i
E
j
= 0 for
i ,= j; we have E
i
, E
i
) = tr(E
i
) = m
i
. The derivation of the
theorem from this is now elementary linear algebra.
Consider the rst relation. The entry in the (, ) position on
the right is
A

, A

) =

i
P

(i)E
i
,

j
P

(j)E
j
)
=

i,j
P

(i)P

(j)E
i
, E
j
) =

i
m
i
P

(i)P

(i),
and this last expression is the entry in the (, ) position on the
left. The second relation is derived similarly.
416 A Course in Combinatorics
Another way to express the content of the above theorem is
Q =
_
_
_
1 0 0
0 n
1
0
.
.
.
.
.
.
.
.
.
0 0 n
k
_
_
_
1
P

_
_
_
1 0 0
0 m
1
0
.
.
.
.
.
.
.
.
.
0 0 m
k
_
_
_
.
Equivalently,
(30.5) m
j
P
i
(j) = n
i
Q
j
(i)
for all i, j = 0, 1, . . . , k.
With Theorems 30.1 and 30.2, we can nd the second eigenmatrix
Q for the schemes J(v, k) and H(n, q). It is somewhat surprising
that P = Q for H(n, q); the reader should check this.
Problem 30D. Explain why the zeroth row and column of P and
Q are as indicated below.
(30.6) P =
_
_
_
1 n
1
... n
k
1
.
.
.
1
_
_
_
, Q =
_
_
_
1 m
1
... m
k
1
.
.
.
1
_
_
_
.
Delsarte (1973) observed that the columns of the second eigenma-
trix Q provide a system of linear constraints, which we will call Del-
sartes inequalities, on, what he calls, the inner distribution vector
of a nonempty subset Y of the point set X of an association scheme.
We dene the distribution vector of Y to be a = (a
0
, a
1
, . . . , a
k
)
where
a
i
:=
1
[Y [
[(Y Y ) R
i
[;
that is, a
i
is the average number of i-th associates y Y of an
element x Y . We have
a
0
= 1, a
0
+a
1
+ +a
k
= [Y [.
For many interesting subsets Y , the number of i-th associates
y Y of an element x Y is constant, i.e. it does not depend on
which x in Y is chosen. For example, this is true if C is a linear
30. Association schemes 417
code in H(n, q) in which case the distribution vector coincides with
what we called the weight enumerator of C in Chapter 21, and
also for other beautiful congurations which have been discussed
in previous chapters and whose distribution vectors we list below.
Here Hamming(7) is the Hamming code of length 7, Golay is short
for Golay code, X means extended, and a * denotes dual code.
Object Scheme Distribution vector
S(2, 3, 7) J(7, 3) (1,0,6,0)
S
2
(2, 5, 11) J(11, 5) (1,0,0,10,0,0)
S
2
(3, 6, 12) J(12, 6) (1,0,0,20,0,0,1)
S(5, 6, 12) J(12, 6) (1,0,45,40,45,0,1)
S(4, 7, 23) J(23, 7) (1,0,0,0,140,0,112,0)
S(5, 8, 24) J(24, 8) (1,0,0,0,280,0,448,0,30)
Hamming(7) H(7, 2) (1,0,0,7,7,0,0,1)
X-Hamming(7) H(8, 2) (1,0,0,0,14,0,0,0,1)
Binary Golay H(23, 2) (1,0,0,0,0,0,0,253,506,0,0,1288, . . . )
X-Binary Golay H(24, 2) (1,0,0,0,0,0,0,0,759,0,0,0,2576,0, . . . )
Ternary Golay H(11, 3) (1,0,0,0,0,132,132,0,330,110,0,24)
*-Ternary Golay H(11, 3) (1,0,0,0,0,0,132,0,0,110,0,0)
Theorem 30.3. The distribution vector a of a nonempty subset
of an association scheme satises
aQ 0
where 0 is the row vector of k + 1 zeros.
Proof: Let 1
X
be the characteristic vector of Y . That is,
(x) =
_
1 if x Y,
0 if x / Y.
Then
a
i
=
1
[Y [
A
i

.
Since E

is idempotent and symmetric,

0 |E

|
2
= (E

)(E

= E

=
1
N

_
k

i=0
Q

(i)A
i
_

=
[Y [
N
k

i=0
Q

(i)a
i
.

418 A Course in Combinatorics

We note for further reference that the -th inequality holds with
equality if and only if the projection E

is the zero vector. The

zeroth inequality is trivial since the zeroth coordinate of aQ is a
0
+
a
1
+ +a
k
, which of course is nonnegative.
Example 30.8. For J(8, 3), Delsartes inequalities when fractions
are cleared are
15 + 7a
1
a
2
9a
3
0,
30 + 2a
1
5a
2
+ 9a
3
0,
10 2a
1
+a
2
a
3
0.
These can be viewed as necessary conditions for the existence of a
family T of 3-subsets of an 8-set with a given distribution vector
(1, a
1
, a
2
, a
3
).
We can ask what they imply about [T[ when we assume some of
the a
i
s are zero. For example, assume a
3
= 0 (no two members of
T are disjoint). The maximum value of 1 +a
1
+a
2
(= [T[) subject
to these inequalities is 21; this is a linear programming problem.
We already knew [T[ 21 because of the Erd os-Ko-Rado theorem,
Theorem 6.4. Theorem 30.3 implies Theorem 6.4; see the notes.
Theorem 30.3 leads to a linear programming bound on the car-
dinalities of codes. For the Hamming scheme H(n, q), given an
integer d, we may consider the linear programming problem
maximize 1 +a
d
+a
d+1
+ +a
n
subject to a
i
0 and
(1, 0, . . . , 0, a
d
, a
d+1
, . . . , a
n
) Q 0
where Q is the n+1 by n+1 second eigenmatrix of the scheme. If
lpb denotes the maximum value of 1 + a
d
+ a
d+1
+ + a
n
under
these conditions and C is any q-ary code of length n with minimum
distance at least d, then [C[ lpb since the distribution vector a
of C satises the conditions and the coordinates sum to [C[.
We mention that for the Hamming schemes, if a is the distri-
bution vector of a linear code C, then
1
[C[
aQ is the distribution
vector of the dual code C

. To see this, just compare the formulae

in MacWilliams theorem, Theorem 20.3, with those in Theorem
30.1(ii). This certainly explains why aQ 0 for the distribution
30. Association schemes 419
vector a of a linear code, but we have proved above that it holds
even if C is not linear.
n d SPB LPBa
0
a
1
a
2
a
3
a
4
a
5
a
6
a
7
a
8
a
9
a
10
a
11
a
12
a
13
a
14
a
15
11 3 170.7 170.7 1 0 0 18.3 36.7 29.3 29.3 36.7 18.3 0 0 1
11 5 30.6 24 1 0 0 0 0 11 11 0 0 0 0 1
11 7 8.8 4 1 0 0 0 0 0 0 2 1 0 0 0
12 3 315.1 292.6 1 0 0 20 45 48 56 65.1 40.7 11.4 3.4 1.7 0.1
12 5 51.9 40 1 0 0 0 0 15 17.5 0 0 5 1.5 0 0
12 7 13.7 5.3 1 0 0 0 0 0 0 2.7 1.7 0 0 0 0
13 3 585.1 512 1 0 0 22 55 72 96 116 87 40 16 6 1 0
13 5 89.0 64 1 0 0 0 0 18 24 4 3 10 4 0 0 0
13 7 21.7 8 1 0 0 0 0 0 0 4 3 0 0 0 0 0
14 31092.31024 1 0 0 28 77 112 168 232 203 112 56 28 7 0 0
14 5 154.6 128 1 0 0 0 0 28 42 8 7 28 14 0 0 0 0
14 7 34.9 16 1 0 0 0 0 0 0 8 7 0 0 0 0 0 0
15 3 2048 2048 1 0 0 35 105 168 280 435 435 280 168 105 35 0 0 1
15 5 270.8 256 1 0 0 0 0 42 70 15 15 70 42 0 0 0 0 1
15 7 56.9 32 1 0 0 0 0 0 0 15 15 0 0 0 0 0 0 1
Figure 30.2
It is dicult to extract useful information in general from the
linear programming bound (LPB), but it can be shown that it is al-
ways at least as good as the sphere packing bound (SPB), Theorem
21.1; see Delsarte (1973). For small values of the parameters, one
can use the simplex algorithm and expicitly evaluate this bound.
We have done so for some values of n and d in the case of binary
codes and display the results in the table of Fig. 30.2. The table
gives a nonnegative vector (1, 0, . . . , 0, a
d
, a
d+1
, . . . , a
n
) whose sum
is maximum subject to aQ 0. It is a good exercise for the reader
to try to decide whether codes meeting the LPB as in the table
actually exist. We mention that several can be obtained from an
interesting binary code of length 16 called the Nordstrom-Robinson
code, which has 256 codewords and minimum distance 6. It is not
linear. See problems 30I and 30J.
There appears to be a kind of duality between the two bases of,
and between ordinary and Hadamard multiplication on, the Bose
420 A Course in Combinatorics
Mesner algebra. It sometimes happens that there exists a second
association scheme so that the E
i
s of the latter under Hadamard
multiplication act like the A
i
s of the former under ordinary multi-
plication (apart from a scalar factor), and vice versa. Two schemes
are said to be formally dual when the rst eigenmatrix P of one is
equal to the second eigenmatrix Q of the other (cf. Problem 30E
below). Many examples of pairs of formally dual strongly regular
graphs are known. The Hamming scheme, as mentioned above, is
formally dual to itself. The Johnson scheme usually has no formal
dual, because in general the Krein parameters dened below are
not integers.
The Krein parameters of an association scheme are the (k + 1)
3
numbers q

ij
dened by
(30.7) N E
i
E
j
=
k

=0
q

ij
E

.
If the scheme has a formal dual, these are the parameters p

ij
for
that formally dual scheme and hence are nonnegative integers. We
always have
q

ij
0, for all 0 i, j, k,
because the q

ij
s are eigenvalues of the Hadamard product of two
positive semidenite matrices and are hence nonnegative; cf. Prob-
lem 21E. In principle, the Krein parameters are functions of the
original parameters p

ij
(see Problem 30E) and their nonnegativity
can be viewed as a necessary condition for the existence of a scheme
with given parameters; we did this for strongly regular graphs in
Theorem 21.3.
Problem 30E. Show how to calculate all the parameters q

ij
of
a scheme given the eigenmatrix Q. That is, prove that they are
uniquely determined by Q.
We need to know later that
(30.8) q
0
ii
= m
i
and q
0
ij
= 0 for i ,= j.
This follows from (30.7) when we consider the sum of all entries of
the matrices on both sides; the sum of the entries of the left-hand
30. Association schemes 421
side is N times
E
i
, E
j
) =
_
m
i
if i = j,
0 otherwise.
For some applications to codes and designs, we need to know
more about a scheme. However, here is one interesting result we
can prove for general schemes before we specialize.
Given a k-class scheme and a subset K of 1, 2, . . . , k, a subset
Y of the point set is called a K-clique when for every pair of distinct
elements x, y Y , x and y are j-th associates for some j K. The
subset Y is a K-coclique when no pair of elements x, y Y are
j-th associates for any j K. If G is the graph whose adjacency
matrix is

jK
A
j
, then cliques in G are the same as K-cliques
in the scheme, and cocliques (independent sets of vertices) are the
same as K-cocliques in the scheme.
Theorem 30.4. Let A X be a K-coclique and B X a K-clique
in a k-class association scheme on a set X where K 1, 2, . . . , k.
Then
[A[ [B[ N.
Proof: Let a = (a
0
, a
1
, . . . , a
k
) be the distribution vector of A
and let b = (b
0
, b
1
, . . . , b
k
) be the distribution vector of B. Take
the inverse of the rst equation of Theorem 30.2, premultiply by a
and postmultiply by b

to get
(30.9) aQ
_
_
_
1 0 0
0 m
1
0
.
.
.
.
.
.
.
.
.
0 0 m
k
_
_
_
1
(bQ)

= N a
_
_
_
1 0 0
0 n
1
0
.
.
.
.
.
.
.
.
.
0 0 n
k
_
_
_
1
b

.
By Theorem 30.3, both aQ and bQ are nonnegative vectors. Their
zeroth coordinates are [A[ and [B[, respectively, so the scalar on
the left of (30.9) is at least [A[ [B[. Our hypothesis implies that
a
i
b
i
= 0 for i > 0, so the scalar on the right of (30.9) is N.
An example of equality in Theorem 30.4 in J(v, k) occurs when
there exists an S(t, k, v). Take K = 1, 2, . . . , k t. Then the
block set of the S(t, k, v) is a K-coclique. The set of all k-subsets
containing a xed t-subset is a K-clique. An example of equality
422 A Course in Combinatorics
in Theorem 30.4 in H(n, q) occurs when there exists a perfect e-
error-correcting code C. Take K = 1, 2, . . . , 2e. Then C is a
K-coclique. A sphere of radius e about a xed word is a K-clique.
Problem 30F. Find an example of a regular graph G, a clique A,
and a coclique B such that [A[ [B[ > [G[.
Problem 30G. Prove that if A is a clique and B a coclique in a
graph G which admits a transitive group of automorphisms, then
[A[ [B[ [G[.
We now dene polynomial schemes, of which there are two types.
An association scheme (with a particular numbering A
1
, A
2
, . . . , A
k
of its association matrices) is said to be P-polynomial when A
i
is a polynomial of degree i in A
1
for i = 0, 1, 2, . . . , k. An asso-
ciation scheme (with a particular numbering E
1
, E
2
, . . . , E
k
of its
idempotents) is said to be Q-polynomial when E
i
is a Hadamard
polynomial of degree i in E
1
for i = 0, 1, 2, . . . , k (we mean that
there is a polynomial of degree i so that E
i
results by applying the
polynomial element-wise to, i.e. to each entry of, E
1
).
Problem 30H. Prove that an association scheme is P-polynomial
if and only if it is a metric scheme.
The Q-polynomial schemes are also called cometric, but there
seems to no simple geometric interpretation of cometric. We use
the terms metric and cometric rather than P- and Q-polynomial.
It can be seen that the Hamming and Johnson schemes are co-
metric; see Delsarte (1973).
We dene a d-code in a metric scheme to be a subset S X
whose characteristic vector satises
A
i

= 0 for i = 1, 2, . . . , d 1.
We dene a t-design in a cometric scheme to be a subset S X
whose characteristic vector satises
E
i

= 0 for i = 1, 2, . . . , t.
The combinatorial signicance of d-codes is straightforward: S
is a d-code if and only if no two distinct elements of S are i-th
30. Association schemes 423
associates for i < d. So in the Hamming scheme, S is a (2e + 1)-
code if and only if it is e-error correcting, i.e. has minimum distance
at least 2e + 1. The combinatorial signicance of t-designs is not
so clear and can only be understood with more knowledge of the
particular scheme. Note that S is a t-design if and only if the rst
t nontrivial inequalities on the distribution vector of S in Theorem
30.3 hold with equality, i.e. that E
i
= 0 for i = 1, . . . , t.
The following theorem explains that designs in the Johnson and
Hamming schemes correspond to the classical concepts of t-designs
and orthogonal arrays, respectively. An orthogonal array of index
and strength t on a set A of q symbols is a subset C A
n
of
q
t
words so that for every choice of t of the n coordinates, each
possible t-tuple of elements of A occurs exactly times in those t
coordinates among the members of C.
Theorem 30.5.
(i) A family S of k-subsets of a v-set is a t-design when considered
as a subset of J(v, k) if and only if it is the set of blocks of a t-design
in the classical sense.
(ii) A family S of n-tuples from an alphabet A of q elements is a
t-design when considered as a subset of H(n, q) if and only if it is
the set of columns of an orthogonal array of strength t.
Partial proof: We will show that the denition of t-design in
the schemes implies that the family S is a classical t-design or
orthogonal array and leave the converses of both parts (i) and (ii)
for the reader.
Let S be a t-design in J(v, k) in the sense of association schemes
and its characteristic vector. In Example 30.7, we saw that E
0
+
E
1
+ + E
t
is the orthogonal projection onto the space spanned
by e
T
as T ranges over the t-subsets of the v-set. The number of
members of S containing a t-subset T is
e

T
= (E
0
+E
1
+ +E
t
)e

T
= E
0
e

T
.
Since E
0
is a scalar multiple of the all-one matrix J, this number
is independent of the particular t-subset T.
Let S be a t-design in H(n, q) in the sense of association schemes
and its characteristic vector. In Example 30.6, we described the
424 A Course in Combinatorics
eigenspaces for q = 2, and we prove the theorem only in this case.
A complete proof may be found in Delsarte (1973).
With the notation of Example 30.6,
, v
a
) =

bS
(1)
a,b)
,
so S is a t-design if and only if

bS
(1)
a,b)
= 0 for all nonzero a of weight t.
For example, when a is taken of weight 1, the above equation
implies that in any given coordinate position, half the members of
S have entry 0 and half have entry 1. When a is taken of weight
2, e.g. as (1, 1, 0, 0, . . . , 0), the equation implies that the number of
members of S that begin with 00 or 11 is equal to the number that
begin with 10 or 01; it follows from this and the previous sentence
that exactly 1/4 of the n-tuples in S begin with each of 00, 10, 01,
11.
In general, consider a t-subset T 1, 2, . . . , n of coordinate
positions. For each subset I of T, let
I
denote the number of
a S with entry 1 in coordinates of I and entry 0 in coordinates of
T I. For each subset I of T, let a
J
denote the binary n-tuple with
1s in coordinates of J and 0s in the remaining n[J[ coordinates.
We have 2
t
linear equations, one for each J T:

[IJ[0 (mod 2)

[IJ[1 (mod 2)

I
=
_
[S[ if J =
0 otherwise.
Clearly
I
= [S[/2
t
for all I T is a solution of this system. But
the coecient matrix of this system is a Hadamard matrix of order
2
t
(see Chapter 18) and in particular is nonsingular, so the solution
is unique. Thus S is an orthogonal array of strength t.
The next two theorems are formal duals. That is, their proofs
are similar but the roles of the bases of A
i
s and E
i
s are inter-
changed, as is the role of ordinary and Hadamard multiplication
of matrices. For the Hamming schemes, Theorem 30.6(i) reduces
30. Association schemes 425
to the sphere packing bound, Theorem 21.1, and part (iii) gives a
very strong condition, due to S. P. Lloyd (1957), for equality to hold
(perfect codessee Chapter 20). For the Johnson schemes, Theorem
30.7(i) reduces to Theorem 19.8, and part (iii) gives a very strong
condition, due to Ray-Chaudhuri and Wilson (1975), for equality
to hold (tight designssee Chapter 19).
Theorem 30.6. Let C be a (2e +1)-code in a k-class metric asso-
ciation scheme on a set X. Let 1
X
be the characteristic vector
of C.
(i) We have
[C[ N/(1 +n
1
+n
2
+ +n
e
).
(ii) There are at least e indices i 1, 2, . . . , k such that
E
i

,= 0.
(iii) Equality holds in (i) if and only if equality holds in (ii), in
which case the e indices i such that E
i

,= 0 are exactly those

integers i for which
e

=0
P

(i) = 0.
Proof: Let be the characteristic vector of a (2e+1)-code C and
consider the expression
:= (c
0
A
0
+c
1
A
1
+ +c
e
A
e
)
2

where c
0
, c
1
, . . . , c
e
are scalars. We shall evaluate in two ways.
Introduce
f(i) := c
0
P
0
(i) +c
1
P
1
(i) + +c
e
P
e
(i)
as a function of the c
j
s. By (30.3), c
0
A
0
+ c
1
A
1
+ + c
e
A
e
=

k
i=0
f(i)E
i
so
=
_
k

i=0
f(i)E
i
__
k

i=0
f(i)E
i
_

=
k

i=0
f(i)
2
E
i

.
426 A Course in Combinatorics
On the other hand, A
i
is a polynomial of degree i in A
1
, so (A
0
+
A
1
+ + A
e
)
2
is a polynomial of degree 2e in A
1
and hence a
linear combination of A
0
, A
1
, . . . , A
2e
. Our hypothesis implies that
A
i

= 0 for i = 1, 2, . . . , 2e. Thus to evaluate , we need only

the coecient of A
0
when (A
0
+A
1
+ +A
e
)
2
is written as a linear
combination of A
0
, A
1
, . . . , A
2e
; by (30.2) and (30.1),
=
_
e

i,j=0
c
i
c
j
p
0
ij
_
A
0

=
_
e

i=0
c
2
i
n
i
_
[C[.
Note that by (30.6), f(0) = c
0
n
0
+ c
1
n
1
+ + c
e
n
e
. We now
combine the two values for , remembering that E
0
=
1
N
J so that
E
0

=
1
N
[C[
2
, to obtain
(30.10) (c
2
0
+c
2
1
n
1
+ +c
2
e
n
e
)[C[ =
k

i=0
f(i)
2
E
i

1
N
(c
0
+c
1
n
1
+ +c
e
n
e
)
2
[C[
2
.
Everything will follow from (30.10). Part (i) follows when we
take all c
i
:= 1.
To prove part (ii), suppose for contradiction that there are fewer
than e indices i 1 such that E
i

,= 0. By elementary linear
algebra, there exist scalars c
0
, . . . , c
e
, not all zero, such that f(i) = 0
for i = 0 and all i such that E
i

,= 0. But then (30.10) gives

[C[

e
i=0
n
i
c
2
i
= 0, a contradiction.
Assume equality holds in (i). Then (30.10), with all c
i
s equal
to 1, shows that f(i) =

e
=0
P

(i) = 0 for the e or more values of

i for which E
i

,= 0. We claim that f(i) = 0 cannot hold for

more than e values of i. This is because f(0), f(1), . . . , f(k) are the
eigenvalues of A
0
+A
1
+ +A
e
, which is a polynomial of degree
e in A
1
; that is, f(0), . . . , f(k) arise from evaluating a polynomial
of degree e on the eigenvalues P
1
(0), . . . , P
1
(k) of A
1
. (The matrix
A
1
has k + 1 distinct eigenvalues since it generates an algebra of
dimension k+1.) Our claim follows from the fact that a polynomial
of degree e has at most e zeros. So (ii) holds with equality and (iii)
is proved in this case.
30. Association schemes 427
Finally, assume equality holds in (ii). Choose scalars c
0
, . . . , c
e
so that f(i) = 0 for all i such that E
i

,= 0 and such that

f(0) = 1. Then (30.10) yields [C[ = N(c
2
0
+c
2
1
n
1
+ +c
2
e
n
e
). The
Cauchy-Schwartz inequality shows that
1 =
_
e

i=0
c
i
n
i
_
2

_
e

i=0
c
2
i
n
i
__
e

i=0
n
i
_
so [C[ N/(1 +n
1
+n
2
+ +n
e
). Thus equality holds in (i).
Corollary (Lloyds theorem). If a perfect e-error-correcting code
of length n over an alphabet of size q exits, then
L
e
(x) :=
e

i=0
(1)
i
_
n x
e i
__
x 1
i
_
(q 1)
ei
has e distinct integral zeros.
Proof: This is just Theorem 30.6(iii) stated for the Hamming
scheme. The sum

e
=0
P

(x) where P

(x) is as in Theorem 30.1(ii)

simplies to L
e
(x) as stated above.
Example 30.9. We show that there are no nontrivial perfect 2-
error-correcting binary codes. If such a code has length n > 2, then
its cardinality is 2
n
/(1+n+
_
n
2
_
), so 1+n+
_
n
2
_
= 2
r
for some integer
r. (That n has this property is unusualbut it may happen, e.g.
for n = 90.) From Theorem 30.1(ii), for H(n, 2),
P
0
(x) +P
1
(x) +P
2
(x) = 2x
2
2(n + 1)x + 1 +n +
_
n
2
_
and Theorem 30.6(iii) then asserts that
x
2
(n + 1)x + 2
r1
= 0
has two integral roots x
1
and x
2
. We must have x
1
= 2
a
and x
2
= 2
b
for positive integers a and b where a+b = r 1 and 2
a
+2
b
= n+1.
Then it may be easily veried that
(2
a+1
+ 2
b+1
1)
2
= 2
a+b+4
7.
428 A Course in Combinatorics
If a and b are both 2, then the left-hand side is 1 (mod 16)
while the right-hand side is 9 (mod 16), a contradiction. The
reader may consider other cases and will nd that the only possi-
bility is a, b = 1, 2 in which case n = 5.
Theorem 30.7. Let D be a 2s-design in a k-class cometric asso-
ciation scheme on a set X. Let 1
X
be the characteristic vector
of D.
(i) We have
[D[ 1 +m
1
+m
2
+ +m
s
.
(ii) There are at least s indices i 1, 2, . . . , k such that
A
i

,= 0.
(iii) Equality holds in (i) if and only if equality holds in (ii), in
which case the s indices i such that A
i

,= 0 are exactly those

integers i for which
s

=0
Q

(i) = 0.
Proof: Let be the characteristic vector of a 2s-design D and
consider the expression
:= (c
0
E
0
+c
1
E
1
+ +c
s
E
s
) (c
0
E
0
+c
1
E
1
+ +c
s
E
s
)

where c
0
, c
1
, . . . , c
s
are scalars. We shall evaluate in two ways.
Introduce
g(i) :=
1
N
(c
0
Q
0
(i) +c
1
Q
1
(i) + +c
s
Q
s
(i))
as a function of the c
i
s. By (30.4), c
0
E
0
+ c
1
E
1
+ + c
s
E
s
=

k
i=0
g(i)A
i
so
=
_
k

i=0
g(i)A
i
_

_
k

i=0
g(i)A
i
_

=
k

i=0
g(i)
2
A
i

.
30. Association schemes 429
On the other hand, E
i
is a Hadamard polynomial of degree i in
E
1
, so the Hadamard square of E
0
+ E
1
+ + E
s
is a Hadamard
polynomial of degree 2s in E
1
and hence a linear combination of
E
0
, E
1
, . . . , E
2s
. Our hypothesis implies that E
i

= 0 for i =
1, 2, . . . , 2s. Thus to evaluate , we need only the coecient of E
0
when the Hadamard square of E
0
+ E
1
+ + E
s
is written as a
linear combination of E
0
, E
1
, . . . , E
2s
; by (30.7) and (30.8),
=
_
1
N
s

i,j=0
c
i
c
j
q
0
ij
_
E
0

=
1
N
2
_
s

i=0
c
2
i
m
i
_
[C[
2
.
Note that by (30.6), g(0) =
1
N
(c
0
m
0
+ c
1
m
1
+ + c
s
m
s
). We
now combine the two values for to obtain
(30.11)
1
N
2
(c
2
0
+c
2
1
m
1
+ +c
2
s
m
s
)[C[
2
=
k

i=0
g(i)
2
A
i

1
N
(c
0
+c
1
m
1
+ +c
s
m
s
)
2
[C[.
Everything will follow from (30.11) in a manner similar to that
of the proof of Theorem 30.6.
Problem 30I. Consider the extended Golay code G
24
. For no-
tational convenience, arrange the coordinates so that the word
c = (1, . . . , 1, 0 . . . , 0) with 1s in the rst eight positions, and 0s
elsewhere, is a codeword.
(i) Use counting arguments to show that G
24
contains 30 words
of weight 8 that have no 1s in the rst eight positions (so: distance
16 to c). Then show that all the words in G
24
that have no 1s in
the rst eight positions form a subcode that is essentially R(1, 4).
(ii) Let A be the set of 16-tuples (a
9
, a
10
, . . . , a
24
) obtained by
dropping the rst eight coordinates from words a = (a
1
, a
2
, . . . , a
24
)
in G
24
such that the weight of (a
1
, a
2
, . . . , a
7
) is 1. Show that A
is a code of length 16 with 256 codewords and minimum distance
6.
(iii) Use A to construct some examples of codes where equality
holds in Figure 30.2.
430 A Course in Combinatorics
Problem 30J. Show that a binary linear code with parameters
[13, 6, 5] does not exist. (And hence there are no binary linear
codes with parameters [16, 8, 6] either.)
Problem 30K. The odd graph has as vertices the k-subsets of a
(2k + 1)-set, joined by an edge if they are disjoint. (For k = 2 this
is the Petersen graph.)
(i) Show that this graph is distance regular.
(ii) Show that if k = 3, the vertices corresponding to the Fano
plane form a perfect code in the graph.
Problem 30L. Find all examples of perfect 3-codes in the Johnson
scheme J(v, 3). (All the examples are trivial.)
Notes.
Association schemes were introduced by statisticians at the same
time as partially balanced designs. An incidence structure is par-
tially balanced with respect to an association scheme on its points
when each pair of i-th associates occurs in a constant number
i
of blocks. Examples include the partial geometries of Chapter 21.
Partially balanced designs were introduced to get around Fishers
inequality b v, or equivalently r k, which need not hold for
them. Replications cost money (r stands for replication number)
or take time, so for practical reasons, one wants r small.
To do the analysis of an experiment for which a certain design
(incidence structure) has been used, it is necessary to do calcula-
tions with NN

where N is the incidence matrix; these calculations

are vastly simplied for partially balanced designs because in an m-
class scheme
NN

=
0
A
0
+
1
A
1
+ +
m
A
m
lies in the small-dimensional BoseMesner algebra even though the
size of the matrices may be very much more than the dimension
m+ 1.
Delsartes inequalities imply the general Erd os-Ko-Rado theo-
rem: If n (t + 1)(k t + 1) and T is a family of k-subsets of an
n-set so that any two of members of T meet in at least t points,
then [T[
_
nt
kt
_
. See Wilson (1984a).
30. Association schemes 431
See Van Lint (1999) for more on the nonexistence of perfect
codes.
References.
E. Bannai and T. Ito (1984), Association Schemes, Benjamin/
Cummings.
A. E. Brouwer, A. M. Cohen, and A. Neumaier (1989), Distance
Regular Graphs, Springer-Verlag.
Ph. Delsarte (1973), An Algebraic Approach to the Association
Schemes of Coding Theory, Philips Res. Rep. Suppl. 10.
Ph. Delsarte (1975), The association schemes of coding theory, in:
Combinatorics (Proc. Nijenrode Conf.), M. Hall, Jr. and J. H.
van Lint, eds., D. Reidel.
J. H. van Lint (1999), Introduction to Coding Theory, Third edition,
Springer-Verlag.
S. P. Lloyd (1957), Binary block coding, Bell System Tech. J. 36,
517535.
D. K. Ray-Chaudhuri and R. M. Wilson (1975), On t-designs, Os-
aka J. Math. 12, 737744.
R. M. Wilson (1984a), The exact bound in the Erd os-Ko-Rado
theorem, Combinatorica 4, 247257.
R. M. Wilson (1984b), On the theory of t-designs, pp. 1950 in:
Enumeration and Design (Proceedings of the Waterloo Silver
Jubilee Conference), D. M. Jackson and S. A. Vanstone, eds.,
Academic Press.
31
(More) algebraic techniques
in graph theory
We have used linear algebraic techniques on the adjacency matrices
of graphs in Chapter 9, and extensively in Chapter 21. We collect
here several other elegant applications. Also see Chapter 36.
A tournament is an orientation of a complete graph; that is, a
directed graph such that for any two distinct vertices x and y, there
is either an edge from x to y, or an edge from y to x, but not both.
Tournaments were introduced briey in Problem 3D. The adjacency
matrix of a digraph has a 1 in position (x, y) when there is an edge
from x to y, and 0 otherwise.
Lemma 31.1. The rank of the adjacency matrix A of a tourna-
ment on n vertices is either n or n 1.
Proof: The denition of tournament ensures that A+A

= J I,
where all matrices are n n. Suppose the rank of A is at most
n 2. Then there exists a nonzero row vector x such that both
xA = 0 and xJ = 0. We then compute
0 = x(A+A

)x

= x(J I)x

= xx

< 0,
which contradicts the existence of x.
The following theorem is due to R. L. Graham and H. O. Pollak.
Their original proof applied Sylvesters law, which they had also
used for their proof of Theorem 9.1.
Theorem 31.2. Suppose that the complete graph K
n
can be ex-
pressed as the union of k edge-disjoint subgraphs H
1
, H
2
, . . . , H
k
where each H
i
is a complete bipartite graph. Then k n 1.
31. (More) algebraic techniques in graph theory 433
Proof: Orient each complete bipartite subgraph H
i
by directing
all edges from one of the two color classes to the other. This pro-
duces a tournament on n vertices whose adjacency matrix A is the
sum of the k adjacency matrices A
i
of the digraphs H
i
(augmented
to n n matrices by including all vertices). The adjacency ma-
trix of such a complete directed bipartite subgraph, after a suitable
renumbering of the vertices, has the block form
_
_
O J O
O O O
O O O
_
_
and in particular has rank 1. This implies that the rank of A is at
most k, and Lemma 31.1 completes the proof.
This brings to mind the De BruijnErd os theorem, Theorem 19.1,
which asserts, in a quite dierent terminology, and which was given
a quite dierent proof, that if K
n
is the union of k edge-disjoint
complete subgraphs, then k n. But no proof of Theorem 31.2 is
known which does not use linear algebra.
As a real symmetric matrix, the adjacency matrix A = A(G) of
a nite graph G has an orthogonal basis of eigenvectors. (Eigen-
vectors corresponding to dierent eigenvalues are necessarily or-
thogonal.) By the eigenvalues of G we mean the eigenvalues of
A(G). Here is a short table of the spectra, i.e. the complete lists of
eigenvalues, of several graphs:
graph spectrum
K
5
4, 1, 1, 1, 1
K
3,3
3, 0, 0, 0, 0, 3
Cube 3, 1, 1, 1, 1, 1, 1, 3
Pentagon 2,
1
2
(1 +

5),
1
2
(1 +

5),
1
2
(1

5),
1
2
(1

5)
Petersen graph 3, 1, 1, 1, 1, 1, 2, 2, 2, 2
L
2
(3) 4, 1, 1, 1, 1, 2, 2, 2, 2
Heawood graph 3,

2,

2,

2,

2,

2,

2,

2,

2,

2,

2,

2,

2, 3
434 A Course in Combinatorics
(The Heawood graph is the bipartite incidence graph of the
Fano conguration, with seven vertices representing the points and
seven others representing the lines. It appears again in Chapter
35.)
Nonisomorphic graphs may have the same spectra. For example,
we mentioned in Chapter 21 that there exist four nonisomorphic
strongly regular graphs with the parameters of T(8); cf. Theorem
21.5.
Problem 31A. Show that it is not possible to nd three edge-
disjoint copies of the Petersen graph in K
10
. Hint: if A
1
, A
2
, and
A
3
are the adjacency matrices of three edge-disjoint Petersen sub-
graphs of K
10
, then all three matrices have the same spectrum, all
have the all-one vector j as an eigenvector, and A
1
+A
2
+A
3
= JI.
Let S be a symmetric matrix. The expression xSx

/xx

, for x ,=
0, is called a Raleigh quotient. Let e
1
, e
2
, . . . , e
n
be an orthonormal
basis of eigenvectors with corresponding eigenvalues

1

2

n
.
If we write x in this basis, say x = a
1
e
1
+ +a
n
e
n
, then
(31.1)
xSx

xx

=

1
a
2
1
+
2
a
2
2
+ +
n
a
2
n
a
2
1
+a
2
2
+ +a
2
n
.
In particular, for any nonzero x,

1

xSx

xx

n
.
While most graphs we will consider are simple, we remark that
there is nothing to prevent the consideration of adjacency matrices
of multigraphs below (where the entry in row x and column y is
the number of edges joining x and y), or the adjacency matrices
of weighted graphs (where the entry in row x and column y is a
weight associated to the edge joining x and y).
It is at rst surprising that the spectrum of a graph has anything
at all to do with the more geometric or combinatorial properties of
the graph. One of the rst connections observed was the following
theorem of A. J. Homan. A coclique, or an independent set of
vertices, in a graph G is a set of vertices no two of which are
adjacent.
31. (More) algebraic techniques in graph theory 435
Theorem 31.3. Let G be a graph on n vertices which is regular
of degree d and let
min
be the least eigenvalue of G, so
min
is
negative. Then for any coclique S in G,
[S[
n
min
d
min
.
Proof: Let A be the adjacency matrix of G and :=
min
. Then
AI is positive semidenite, i.e. all eigenvalues are nonnegative;
note that the eigenvectors of A I are the same as those of A.
One of the eigenvectors is j := (1, 1, . . . , 1), and we have
j(AI) = (d )j.
Then with
M := AI
d
n
J,
we will have jM = 0. Every other eigenvector e of A may be
taken to be orthogonal to j, so eJ = 0, and is seen also to be an
eigenvector of M with nonnegative eigenvalue. That is, M is also
positive semidenite.
Now let be the characteristic vector of a coclique S consisting
of m vertices of G, i.e. (x) is 1 if x S and 0 otherwise. Then
A

= 0 and we have
0 M

d
n
J

= m
d
n
m
2
.
The stated inequality follows.
There are extensions and variations on Theorem 31.3 for nonreg-
ular graphssee Haemers (1979).
Problem 31B. Apply Theorem 31.3 to the complement of a sim-
ple regular graph G to obtain an upper bound on the size of a clique
in G in terms of the spectrum. Find several examples where your
bound is met.
Problem 31C. The line graph L = L(G) of a simple graph G
was introduced in Chapter 17. For example, L
2
(m) as dened in
Chapter 21 is the line graph of K
m,m
, T(m) is the line graph of K
m
,
436 A Course in Combinatorics
and the line graph of a pentagon is a pentagon. (i) Show that the
Petersen graph is not the line graph of any graph. (ii) Prove that
if G has more edges than vertices, then the minimum eigenvalue
of the line graph L(G) is 2. (Suggestion: let N be the incidence
matrix of G and consider N

N.)
L. Lov asz (1979) observed that the bound on the size of a coclique
in Theorem 31.3 is also a bound on what is called the Shannon
capacity of a graph. This is a concept that arose in information
theory. Suppose a set of letters is to be used for transmitting
messages. Some pairs of letters are assumed to be confusable or
confoundable. We say two potential messages (words or strings
of n of these letters) are confoundable when in each coordinate the
letters are either identical or confoundable. We desire a set of words
no two of which are confoundable.
To state this in more graph-theoretic terminology, we introduce
the graph G whose vertices are the letters and where two vertices
are adjacent if and only if the letters are confoundable. Thus a set
of letters, no two of which are confoundable, is exactly a coclique in
this graph. The strong product GH of simple graphs G and H is
the simple graph dened by V (GH) := V (G) V (H) and where
distinct vertices (x
1
, y
1
) and (x
2
, y
2
) are adjacent in GH when x
1
and x
2
are equal or adjacent, and y
1
and y
2
are equal or adjacent.
The vertices of G
n
:= G G G (n factors) correspond to
words of length n, and two are adjacent in G
n
just when the words
are confoundable. Thus we are interested in the size of a largest
coclique in G
n
.
The independence number (G) of a graph G is the largest car-
dinality of a coclique in G. The Shannon capacity of a graph G
is
(G) := lim
n
((G
n
))
1/n
= sup
n
((G
n
))
1/n
.
That the limit exists and is equal to the supremum follows from the
observation that the product of cocliques in G and H is a coclique
in G H, and thus that (G
k+m
) (G
k
)(G
m
). (See Feketes
lemma in Chapter 11.)
Example 31.1. The problem of evaluating the Shannon capacity
of such simple graphs as the pentagon P
5
was open for many years
31. (More) algebraic techniques in graph theory 437
before the relevance of algebraic techniques was noticed by Lov asz.
It is easy to nd 2
n
words of length n, no two of which are con-
foundable, e.g. all words of 1s and 3s. This shows (P
5
) 2. This
is not so good: we get only four words of length 2 this way, when
it is not hard to nd a set of ve:
(1, 1), (2, 3), (3, 5), (4, 2), (5, 4).
For even values of n, we can take the product of n/2 copies of the
above ve words to get (

5)
n
pairwise nonconfoundable words of
length n; this shows (P
5
)

5. Can we do better, perhaps for

very large values of n? No, as we shall see after Theorem 31.6,
(P
5
)

5.
Here is the approach of Lov asz (1979). First, observe that with
the denition of Kronecker product of Chapter 18, we have, for
vectors x, y 1
n
, v, w 1
n
,
(31.2) (x v)(y w)

= x, y)v, w).
Let G be a graph. For simplicity we shall always assume that its
vertices are 1, 2, . . . , n. An orthonormal representation of G is a
system (v
1
, . . . , v
n
) of unit vectors in a Euclidean space such that
if i and j are nonadjacent vertices, then v
i
and v
j
are orthogo-
nal. Clearly, every graph has an orthonormal representation, for
example by pairwise orthogonal vectors.
Lemma 31.4. Let (u
1
, . . . , u
n
) and (v
1
, . . . , v
m
) be orthonormal
representations of G and H respectively. Then the vectors u
i
v
j
form an orthonormal representation of GH.
Proof.: This is a consequence of (31.2).
Dene the value of an orthonormal representation (u
1
, . . . , u
n
) to
be
min
c
max
1in
1
c, u
i
)
2
,
where c ranges over all unit vectors. The vector c yielding the min-
imum is called the handle of the representation. Let (G) denote
the minimum value over all repesentations of G. It is easy to see
that this minimum is attained. A representation with value (G)
is called optimal.
438 A Course in Combinatorics
Lemma 31.5. (GH) (G)(H).
Proof: Let (u
1
, . . . , u
n
) and (v
1
, . . . , v
m
) be optimal orthonormal
representations of G and H, with handles c and d, respectively.
Then, by (31.2), c d is a unit vector, and again by (31.2),
(GH) max
i,j
1
c d, u
i
v
j
)
2
= max
i,j
1
c, u
i
)
2

1
d, v
j
)
2
= (G)(H).
(One can prove that equality holds in this lemma.)
Theorem 31.6. (G) (G).
Proof: We rst show that (G) (G). Let (u
1
, . . . , u
n
) be an
optimal orthonormal representation of G with handle c. Suppose
that 1, 2, . . . , k is a maximum independent set in G and hence
u
1
, . . . , u
k
are pairwise orthogonal. Therefore
1 = [[c[[
2

i=1
c, u
i
)
2
(G)/(G).
From this and Lemma 31.5 we nd that (G
n
) (G
n
) (G)
n
.

Example 31.1, continued. We can now show that (C

5
) =

5
by an elegant idea from Lov asz (1979). Consider an umbrella whose
handle and ve ribs have unit length. If we open such an umbrella,
we will pass a point where the angles between nonadjacent ribs are
all /2. If we call the handle c and the ribs u
1
to u
5
, all oriented
away from their common point, then u
1
, . . . , u
5
is an orthonormal
representation of C
5
in 1
3
. It is an easy calculation to show that
c, u
i
) = 5
1/4
. Theorem 31.6 and the denition of (C
5
) then yield
the result.
The result also follows form the corollary below.
Corollary. Let G be a graph on n vertices which is regular of
degree d and let
min
be the least eigenvalue of G. Then
(G)
n
min
d
min
.
31. (More) algebraic techniques in graph theory 439
Proof: Let A be the adjacency matrix of G. In the proof of
Theorem 31.3, we saw that with :=
min
, M := A I
d
n
J
was positive semidenite (and singular). Thus M = BB

for some
real matrix B of rank < n. Let the rows of B be x
1
, . . . , x
n
; we
have
x
i
, x
i
) =
d
n
, x
i
, x
j
) =
d
n
,
the latter when i, j are not adjacent. Let c be any unit vector
orthogonal to the rows x
i
of B and dene
v
i
:=
1

x
i
+
1
_
n/(d )
c.
Check that v
1
, . . . , v
n
is an orthogonal representation of G. Finally
note that for any i,
1
c, v
i
)
2
=
n
d
.
The result follows from Theorem 31.6.
Next, we give two applications of interlacing of eigenvalues.
Lemma 31.8. Let A be a symmetric matrix of order n with eigen-
values

1

2

n
.
Suppose N is an m n real matrix such that NN

= I
m
, so m n.
Let B := NAN

, and let

1

2

m
be the eigenvalues of B. Then the eigenvalues of B interlace those
of A, in the sense that

i

i

nm+i
for i = 1, 2, . . . , m.
Proof: Let e
1
, . . . , e
n
be an orthonormal basis of eigenvectors of
A corresponding to
1
, . . . ,
n
, and let f
1
, . . . , f
m
be an orthonormal
basis of eigenvectors of B corresponding to
1
, . . . ,
m
.
440 A Course in Combinatorics
Fix i and consider U := spanf
1
, . . . , f
i
. By (31.1),
xBx

xx

i
for every nonzero x U. Let W := xN : x U. Then W is an
i-dimensional subspace and
(31.3)
yAy

yy

i
for every y W. Choose y ,= 0 in W spane
i
, e
i+1
, . . . , e
n
.
Then in addition to (31.3), we have
yAy

yy

i
from (31.1) and this proves
i

i
.
A similar argument with U := spanf
i
, . . . , f
m
will prove that

i

nm+i
.
The term interlacing seems most natural in the case m = n 1
where

1

1

2

2

3

n1

n1

n
.
An important special case of Lemma 31.8 occurs when the rows
of N are m distinct standard basis vectors of length n, i.e. with a
single 1 in each row. In this case the matrix B is just an m m
principal submatrix of A. The following observation is due to D.
M. Cvetkovic.
Theorem 31.9. The size of a coclique in a graph G cannot exceed
the number of nonnegative eigenvalues, or the number of nonposi-
tive eigenvalues, of G.
Proof: If G has a coclique of size m, then the adjacency matrix A
has an m m principal submatrix of all zeros. If the eigenvalues
of A are
1

2

n
, then by interlacing,
m
0 and
0
nm+1
, so A has at least m eigenvalues which are 0 and at
least m which are 0.
The next theorem is a result of A. J. Homan on the chromatic
number of a graph. We remark that for regular graphs, this would
follow from Theorem 31.3.
31. (More) algebraic techniques in graph theory 441
Theorem 31.10. Let
1

n
be the eigenvalues of a graph
G. Then
(G) 1 +
1
/(
n
).
Proof: Suppose G can be properly colored with m colors. The
color classes induce a partition of the adjacency matrix
A =
_
_
A
11
A
1m
.
.
.
.
.
.
A
m1
A
mm
_
_
,
where A
ij
is the submatrix consiting of the rows indexed by the
vertices of color i and columns indexed by the vertices of color j.
Each of the diagonal blocks A
ii
is a square zero matrix.
Let e be an eigenvector of A corresponding to the maximum
eigenvalue
1
and write e = (e
1
, . . . , e
m
) where e
i
has coordinates
indexed by the vertices of color i. Let
N :=
_
_
_
_
_
1
|e
1
|
e
1
0 0
0
1
|e
2
|
e
2
0
0 0
1
|e
3
|
e
3

.
.
.
.
.
.
.
.
.
_
_
_
_
_
(an m n matrix) and let B := NAN

. By Lemma 31.8, the

eigenvalues
1
, . . . ,
m
of B interlace those of A and hence are be-
tween
1
and
n
. On the other hand, we have constructed B so
that
1
is an eigenvalue since
(|e
1
|, . . . , |e
m
|)B = eAN

=
1
eN

=
1
(|e
1
|, . . . , |e
m
|).
Finally note that the diagonal entries of B are zeros, so
0 = trace(B) =
1
+ +
m

1
+ (m1)
n
.
This proves the theorem. (But perhaps we should have said that if
some of the e
i
are zero, those rows should not be included in N.)

442 A Course in Combinatorics

Problem 31D. Show that for any nite graph G, (G) 1 +
1
where
1
is the largest eigenvalue of G. Hint: consider the degrees
of an induced subgraph of G that is minimal with respect to having
the same chromatic number as G.
In addition to being symmetric, the adjacency matrix of a graph
is also a nonnegative matrix. There is a useful theory of such matri-
ces, at the center of which is the PerronFrobenius theorem, The-
orem 31.11 below. A complete proof may be found in Gantmacher
(1959).
A square matrix A with rows and columns indexed by a set X,
say, is called irreducible when it is not possible to nd a proper
subset S of X so that A(x, y) = 0 whenever x S and y X S.
Equivalently, A is not irreducible if and only if it is possible to apply
a simultaneous row and column permutation to obtain a matrix of
the form
_
B O
C D
_
where B and D are square and of order at least 1. It should be clear
that if A is the incidence matrix of a graph, then it is irreducible
if and only if the graph is connected.
Problem 31E. Let D be a nite digraph. Let A the adjacency
matrix of D, or any matrix with rows and columns indexed by the
vertices of D and where A(x, y) = 0 if there are no edges directed
from x to y and A(x, y) > 0 if there are such edges. (i) Show that
A
k
(x, y) > 0 if and only if there is a directed walk of length k from
x to y in D. (ii) Show that A is irreducible if and only if D is
strongly connected, that is, if and only if for any two vertices of D,
there is a directed path from x to y.
Theorem 31.11 (PerronFrobenius). Let A be an irreducible
n n nonnegative matrix. There is, up to scalar multiples, a
unique eigenvector a = (a
1
, . . . , a
n
) all of whose coordinates a
i
are
nonnegative. The coordinates of this vector a are, in fact, strictly
positive. The corresponding eigenvalue (which is called the dom-
inant eigenvalue of A) has algebraic multiplicity 1 and has the
property that [[ for any eigenvalue of A.
31. (More) algebraic techniques in graph theory 443
Partial proof: We omit the proof of the existence of a nonnega-
tive eigenvector, but show how irreducibility is used and give much
of the rest of the proof.
We will show that any nonnegative eigenvector a of A, corre-
sponding to eigenvalue , say, is positive in all coordinates. Con-
sider A as the matrix of a strongly connected digraph as in Problem
31E. Since there is a directed path from any vertex to any other of
length n 1,
I +A+A
2
+ +A
n1
> O.
Here and below, an inequality for matrices or vectors is understood
to hold in all coordinates. Since a 0 but a ,= 0,
0 < a(I +A+A
2
+ +A
n1
) = (1 + + +
n1
)a,
and so 0 < a.
Next we show the eigenvalue has geometric multiplicity 1, i.e.
that the corresponding eigenspace has dimension 1. Let a
/
be any
eigenvector corresponding to eigenvalue , with coordinates that
are positive, negative, or zero. If we add or subtract small scalar
multiples of a
/
from a, we obtain other positive or nonnegative
eigenvectors a ca
/
corresponding to eigenvalue . We can choose
c so that a+ca
/
0, but so that at least one coordinate is zero. This
contradicts the result of the previous paragraph unless aca
/
= 0.
We omit the proof that has algebraic multiplicity 1, i.e. is a simple
zero of the characteristic polynomial.
Let a be the positive eigenvector for A, with eigenvalue . Let b
be the positive eigenvector of A

; say Ab

= b

. Then
ab

= (aA)b

= a(Ab

) = ab

.
This is a contradiction unless = , so Ab

= b

. Suppose u is
an eigenvector of A that is not a scalar multiple of a, say uA = u
with ,= . Then
ub

= u(Ab

) = (uA)b

= ub

.
444 A Course in Combinatorics
which implies ub

= 0. So any other eigenvector u cannot have

nonnegative coordinates. Now let u := ([u
1
[, . . . , [u
n
[) and check
that uA [[ u follows easily from A O. Then
ub

= u(Ab

) = ( uA)b

[[ ub

,
from which we have [[.
Theorem 31.12. Let be the dominant eigenvalue of a connected
graph G. Then G is bipartite if and only if is also an eigenvalue
of G.
Proof: First suppose G is bipartite. Then the adjacency matrix
A of G is, after renumbering the vertices if necessary so that the
color classes consist of the rst k and last n k, of the form
A =
_
O B
B

O
_
where B is k n k. Let e be an eigenvector corresponding to
an eigenvalue and write e = (e
1
, e
2
) where e
1
consists of the
rst k coordinates. It may be checked that (e
1
, e
2
) is then an
eigenvector of eigenvalue . So if G is bipartite (connected or
not), its spectrum is in fact symmetric about 0.
Now suppose is the dominant eigenvalue of a connected graph
G and that is also an eigenvalue. Let e = (e
1
, e
2
) be an
eigenvector of unit length corresponding to , where we have
numbered the vertices so that e
1
and e
2
both have all nonnegative
coordinates. Partition the adjacency matrix A accordingly:
A =
_
B C
D E
_
.
Then
= eAe

= e
1
Be

1
+e
2
Ee

2
e
1
Ce

2
e
2
De

1
,
and hence
(31.4) (e
1
, e
2
)A(e
1
, e
2
)

31. (More) algebraic techniques in graph theory 445
with equality if and only if
(31.5) e
1
Be

1
= e
2
Ee

2
= 0.
The equation (31.1) and the fact that is the dominant eigenvalue
imply that equality must hold in (31.4) and that (e
1
, e
2
) is an eigen-
vector corresponding to eigenvalue . Then Theorem 31.11 implies
that all coordinates of (e
1
, e
2
) are positive. Finally, (31.5) shows
that B = E = O and this means that G is bipartite.
Theorem 31.13. Let G be a nite graph and A its adjacency
matrix. There exists a polynomial f(x) such that f(A) = J if and
only if G is connected and regular.
Proof: Suppose f(A) = J for some polynomial f(x). Then, since
polynomials in a given matrix commute, AJ = JA. The entry in
row x and column y of AJ is deg
G
(x); the entry in that position of
JA is deg
G
(y). Hence G is regular.
If f(A) = J, then, in particular, for any x and y, we must have
A
k
(x, y) ,= 0 for some k. The entry in row x and column y of A
k
is the number of walks of length k from x to y in G. Hence G is
connected.
Now suppose G is connected and regular of degree d. Then j is an
eigenvector of A corresponding to eigenvalue d. By Theorem 31.11,
the eigenvalue d has multiplicity one. For any symmetric matrix M,
the matrix of the orthogonal projection onto any of its eigenspaces
is a polynomial in M; explicitly, if (x
1
)(x
2
) (x
k
) is
the minimal polynomial of M, then it is an exercise to show that
1
(
1

2
) (
1

k
)
(M
2
I) (M
k
I)
is the orthogonal projection onto the eigenspace a : aM =
1
a.
Since the orthogonal projection onto the span of j is
1
v
J, this matrix
is a polynomial in A.
Isomorphic graphs will have the same eigenvalues, but, as we
mentioned earlier, the converse is not true: nonisomorphic graphs
may have the same spectra (multisets of eigenvalues). See Problem
21O. But we may have a partial converse, in particular for circulant
graphs with a prime number of vertices.
446 A Course in Combinatorics
Let G be an abelian group, written additively. For S G, the
Cayley graph (G, S) has vertex set G, and there is an edge directed
from x to y in (G, S) if and only if y x S. Cayley graphs have
arisen numerous times in the preceding chapters. A Cayley graph
may be thought of as an undirected graph when S = S, and will
have no loops when 0 , S. A circulant graph is a Cayley graph
(G, S) where G is cyclic.
(Cayley graphs can also be dened for nonabelian groups. But
we consider only the abelian case below.)
A G-matrix is a square matrix A whose rows and columns are
indexed by the elements of G and where for some vector of objects
(a
g
: g G), A(i, j) = a
ji
. A circulant matrix (or simply a
circulant) is a G-matrix where G is cyclic. In general, the adjacency
matrix of a Cayley graph based on G is a G-matrix.
We remark that G-matrices over a ring R are closed under mul-
tiplication and the algebra of such matrices is isomorphic to the
group ring R[G] as in Chapter 28.
Proposition 31.14. Let be a primitive nth root of unity in some
eld F and U the matrix with rows and columns indexed by Z
n
with
U(i, j) =
ij
. Let A be a circulant based on elements (a
i
: i Z
n
)
of F. Then UAU
1
is the diagonal matrix with diagonal entries

j
:=
n1

i=0
a
i

ji
, j = 0, 1, . . . , n 1.
That is, the eigenvalues of A are
0
,
1
, . . . ,
n1
(and the corre-
sponding eigenvectors are the rows of U).
Problem 31F. (i) Prove Proposition 31.14. (ii) For those familiar
with the characters of an abelian group: Show that the eigenvalues
of a rational G-matrix A are

=

gG
a
g
(g), as ranges over
the characters of G. Cf. Example 30.6.
The following is from Elspas and Turner (1970).
Theorem 31.15. Let A and B be rational circulant matrices of
prime order p > 2, based on vectors (a
i
: i Z
p
) and (b
i
: i Z
p
),
respectively. If A and B have the same spectrum, then for some
31. (More) algebraic techniques in graph theory 447
t ,= 0 in Z
p
, a
i
= b
ti
for all i Z
p
, where the subscripts are read
modulo p.
Proof: Let be a complex primitive pth root of unity. Then, in
view of Proposition 31.14, :=

n1
i=0
a
i

i
is an eigenvalue of A and
so is one of the eigenvalues of B, and so
n1

i=0
a
i

i
=
n1

i=0
b
i

si
for some s.
First, assume s ,= 0, and choose t so that st 1 (mod p). Then
p1

i=0
(a
i
b
ti
)
i
= 0.
But as the minimal polynomial of over the rationals is 1+x+x
2
+
+x
p1
, we may conclude that for some constant c, a
i
b
ti
= c for
all i. The traces of A and B, which must be equal, are, respectively,
pa
0
and pb
0
. Hence a
0
= b
0
, and then c = 0.
If s = 0, then is rational and
(a
0
) +a
1
+ +a
p1

p1
= 0.
We may concude that all coecients above are equal to some con-
stant c; this means A = I +cJ.
We repeat the argument with the roles of A and B interchanged,
and we nd that either the statement of the theorem holds or that
B = I +dJ for some rational numbers and d.
The eigenvalues of aI +bJ are a +pb of multiplicity 1 and a +b
of multiplicity p 1. If A = I + cJ and B = I + dJ have the
same spectra and p > 2, it follows that A = B and the theorem
holds for any t.
We remark that for p = 2, the statement of the theorem is not
valid, since e.g. the circulants
_
1 1
1 1
_
and
_
1 1
1 1
_
have the same spectrum 2, 0. However, it is valid if A and B are
both nonnegative.
448 A Course in Combinatorics
Corollary. Circulant graphs of prime order are isomorphic if and
only if they (their adjacency matrices) have the same eigenvalues.
Proof: We have asserted before that isomorphic graphs have the
same spectra.
Suppose the circulant graphs (Z
p
, S) and (Z
p
, T), for some
S, T Z
p
, p prime, have adjacency matrices A and B with the
same spectra. Here A is the circulant based on (a
i
: i Z
p
) where
a
i
= 1 if i S and a
i
= 0 otherwise; similarly, and B is based on
(b
i
: i Z
p
) where b
i
= 1 if i T and b
i
= 0 otherwise. Theorem
31.15 says (for p = 2, we need the remark) a
i
= b
ti
for some t ,= 0.
Then the permutation : i ti of Z
p
is an isomorphism of (Z
p
, S)
onto (Z
p
, T) because
i j in (Z
p
, S) i j S
t(i j) T ti tj in (Z
p
, T),
where x y means there is an edge directed from x to y.
Corollary. Given S, T Z
p
, the circulant graphs (Z
p
, S) and
(Z
p
, T) are isomorphic if and only if T = tS for some multiplier
t ,= 0 in Z
p
.
Proof: If (Z
p
, S) and (Z
p
, T) are isomorphic, then they have
the same spectra, and as in the proof of the preceding corollary, we
have a
i
= b
ti
for some t ,= 0, which means T = tS.
Conversely, if T = tS, then as we saw above, : i ti of Z
p
is
an isomorphism.
It had been conjectured that the corollary above holds when p
is replaced by any integer n, and the condition t ,= 0 replaced
by (t, n) = 1, but counterexamples exist for n = 8, 9 and other
values. In fact, the corollary with p replaced by n holds exactly
when n is square-free or 4 times an odd square-free number. See
M. Muzychuk (1997).
Problem 31G. Prove that the vertices of any nite simple graph
may be colored red and blue so that each red vertex is adjacent to an
even number of red vertices, and each blue vertex is adjacent to an
odd number of red vertices. First prove the lemma: if S = (a
ij
) is a
31. (More) algebraic techniques in graph theory 449
symmetric (0, 1)-matrix, then the diagonal a = (a
11
, a
22
, . . . , a
nn
),
considered as a column-vector, is in the span of the columns of S
over the eld F
2
.
Problem 31H. Let G be a connected regular graph with exactly
three distinct eigenvalues. Show that G is strongly regular.
Problem 31I. Let G be a nontrivial strongly regular graph and
for x V (G), let (x) denote the subgraph induced by the vertices
nonadjacent to x. Prove that (x) is connected by (1) explaining
why k would be an eigenvalue of G if (x) were not connected,
and (2) showing by calculation that k is not an eigenvalue of G.
Problem 31J. Suppose the Petersen graph P is the union of span-
ning subgraphs H, K, that are regular of degrees 1 and 2, respec-
tively. Let A be the adjacency matrix of P and B, C the adjacency
matrices of H and K, so that A = B +C.
Show that there exists a vector u that is an eigenvector of Acorre-
sponding to eigenvalue 1 and also an eigenvector of B corresponding
to eigenvalue 1 (and hence an eigenvector of C corresponding to
eigenvalue 2). Next, explain why K cannot be a connected graph.
(This shows that P is not Hamiltonian.)
Problem 31K. Let G be a directed graph on n vertices without
multiple edges. We allow edges a b and b a to occur simul-
taneously in which case we call a, b an undirected edge. Suppose
that for any two vertices a, b there is exactly one walk of length 3
from a to b if a ,= b and no such walk if a = b. If A is the adjacency
matrix of G, then this property is expressed by the equation
(31.6) A
3
= J I.
(i) Prove that G is regular;
(ii) If c is the valency of G, show that n = c
3
+ 1;
(iii) Prove that G has exactly
1
2
(c
2
+c) undirected edges;
(iv) Construct an example of such a graph if n = 9.
Notes.
Theorem 31.3 was never published by Homan, one of the rst
algebraic graph theorists, but caught the fancy of many of those
who heard of it, and has become a classic.
450 A Course in Combinatorics
Problem 31D is a result of H. Wilf. Problem 31A is an unpub-
lished result of A. J. Schwenk.
References.
N. Biggs (1974), Algebraic Graph Theory, Cambridge University
Press.
D. M. Cvetkovic, M. Doob, and H. Sachs (1979), Spectra of Graphs,
a Monograph, V. E. B. Deutscher Verlag der Wissenschaften.
B. Elspas and J. Turner (1970), Graphs with circulant adjacency
matrices, J. Combinatorial Theory 9, 297307.
F. R. Gantmacher (1959), The Theory of Matrices, Chelsea.
W. Haemers (1979), Eigenvalue Techniques in Design and Graph
Theory, Mathematisch Centrum, Amsterdam.
L. Lov asz (1979), On the Shannon capacity of a graph, IEEE Trans.
Information Theory 25, 17.
M. Muzychuk (1997), On Adams conjecture for circulant graphs,
Discrete Math. 176, 285298.
32
Graph connectivity
For k 2, a graph G is said to be k-vertex connected, or simply k-
connected, when [V (G)[ k+1 and the removal of any k1 vertices
(and any incident edges) from G does not result in a disconnected
graph. We use 1-connected as a synonym for connected.
If a graph G with at least k + 1 vertices is not k-connected and
the deletion of a set S of k 1 vertices disconnects it, there is
a partition of V (G) S into nonempty sets X, Y with no edges
crossing (one end in X, one in Y ). Let H and K be the subgraphs
induced by X S and Y S, except that edges with both ends in
S are to be put in one and only one of H or K. Then we obtain
edge-disjoint subgraphs H and K whose union is G and such that
[V (H)V (K)[ = k1. Conversely, if such subgraphs H and K exist
and each contains at least one more vertex than their intersection,
then G is not k-connected.
A graph is said to be nonseparable when it is 2-connected and has
no loops, or when it is a bond-graph (with two vertices and any pos-
itive number of edges joining them, including the link-graph with
one such edge), a loop-graph (one edge joining a single vertex to
itself), or a vertex-graph. All polygons, for example, are nonsepa-
rable; path-graphs or other trees with at least two edges are not.
(We do not want loops, if there are other edges, because while
the deletion of the incident vertex may not disconnect the graph in
a combinatorial sense, it does in a topological sense. With another
type of 2-connectivity, that of Tutte, we would not need to prohibit
loops, or allow the small graphs, explicitly; see Problem 32D.)
Here is a small observation about the structure of nonseparable
graphs.
452 A Course in Combinatorics
Lemma 32.1. Let G be a nite nonseparable graph with at least
two edges and H a maximal proper nonseparable subgraph of G.
Then G is the union of H and a path-graph P that joins distinct
vertices of H and has none of its internal vertices in H.
The path in the statement of the lemma may be called a handle.
Figure 32.1 is suggestive.
H
P
Figure 32.1
Proof: It may be that V (H) = V (G). In this case, let e be an
edge in G but not H and let P be the link-graph consisting of e
and its ends. Since adding e to H produces a nonseparable graph
and H is maximal among the proper nonseparable subgraphs, the
union of H and P is G.
If V (H) ,= V (G), the connectivity of G implies that there exists
an edge e with one end x V (H) and the other y / V (H). By the
connectivity of Gx, there exists a simple path that does not pass
through x, joining y to some other vertex w of H. If z is the rst
vertex of H on this path (as we proceed from y to w), the edge e
and the initial part of this path provide a path-graph P joining x
to z that has no edge terms in H and no vertex terms in H other
than its ends. Then H P is a nonseparable graph larger than H,
and so is equal to G.
As a consequence of the lemma, given a nite nonseparable graph
G with at least two edges, there is a sequence of nonseparable
subgraphs
H
0
H
1
H
2
H
k
= G
where H
0
is a link-graph and where each H
i+1
is the union of H
i
and a handle.
A subdivision of a graph G is, informally, a graph H obtained
by inserting extra vertices into some edges. More precisely, we
32. Graph connectivity 453
can replace the edges of G by (nite) path-graphs whose internal
vertices are new, i.e. not vertices of the original G; and we allow
the replacing of loops by polygons. Two graphs are said to be
homeomorphic if they are isomorphic to subdivisions of the same
graph. Subdivision of a nonseparable graph (other than the link-
graph) preserves nonseparability.
Problem 32A. Let us dene the rank of a connected graph G as
[E(G)[ [V (G)[ + 1. (In Chapter 34, we will see that this is the
dimension of what is called there the cycle space of G.) When we
add a handle to a nonseparable graph, the rank goes up by one.
Connected graphs of rank 0 are trees. The nonseparable graphs
of rank 1 are the polygons. The nonseparable graphs of rank 2
are the -graphs, homeomorphic to the letter , i.e. subdivisions of
the bond-graph with three edges, since that is all that can arise by
adding a handle to a polygon. Describe all nonseparable graphs of
rank 3 as subdivisions of four graphs.
It is clear that if G is not nonseparable and H, K are edge-disjoint
subgraphs with at most one vertex in common, whose union is G,
then an edge of H and an edge of K cannot be contained together
in any polygon of G.
Theorem 32.2. Any two edges in a nonseparable graph G are
contained in the edge set of a polygon subgraph.
Proof: We assert that given distinct vertices x, y and an edge e
of a nonseparable graph G, there exists a path from x to y which
traverses e. So if x, y are the ends of another edge e
/
, we obtain a
polygon containing e and e
/
. We prove the assertion by induction
on the total number of edges (or the rank).
The link-graph with edge e is a nonseparable subgraph. Let H be
a maximal proper nonseparable subgraph containing e. If both x
and y are in H, we are done by the induction hypothesis. Otherwise
G is the union of H and a path-graph P joining vertices a and b of
H, say, and at least one of x or y is an internal vertex of P. If both
x and y are vertices of P, a path from a to b in H that traverses
e together with segments of P provides the required path. Finally,
assume x is internal to P and y V (H). Say y ,= a. Then a path
454 A Course in Combinatorics
in H from y to a that traverses e together with the segment of P
from x to a provides the required path.
Corollary. Any two vertices of a nonseparable graph G (other
than the link-graph) are contained in some polygon in G.
Proof: Choose an edge on each vertex and apply the theorem,
if the edges are distinct. If the edges are the same, there is a
polygon containing that edge, as long as the graph has a second
edge somewhere.
The corollary is actually equivalent to Theorem 32.2, since the
theorem can be derived by inserting new vertices into the middle
of the given edges, and applying the corollary to the new vertices.
The corollary is the special case k = 2 of the following theorem
of H. Whitney (1932).
Theorem 32.3. A graph G with at least k +1 vertices is k-vertex
connected if and only if, for any two distinct vertices x and y of
G, there exist k internally disjoint paths joining x and y, i.e. paths
which share no vertices other than their ends x and y.
This will be an immediate consequence of Mengers Theorem
below. We give the latter in the version for directed graphs. As
usual, an undirected graph may be considered as a digraph when
we replace each undirected edge by a pair of directed edges, one in
each direction.
We will apply Theorem 7.2 on integral ows, and require the
following observation. (This is almost identical to Problem 7B.)
Problem 32B. Let f be a nonnegative ow from s to t in a di-
graph D (capacities are not required).
(i) Show that if f has positive strength, then there exists a di-
rected path from s to t all of whose edge terms e have f(e) > 0.
(ii) Conclude that if f is integral and has strength k, then there
exist k directed paths p
1
, p
2
, . . . , p
k
from s to t in D such that for all
edges e E(D), the number of paths p
i
that traverse e is f(e).
Problem 32C. Let s and t be distinct vertices in a digraph D.
Show that the maximum number of edge-disjoint directed paths
from s to t is equal to the minimum number of edges in a set E
32. Graph connectivity 455
whose deletion disconnects s and t in that direction, i.e. such that
every directed path from s to t contains at least one edge from E.
Let s, t be two distinct vertices in a digraph D. An (s, t)-sep-
arating set is a subset S V (D) s, t so that any directed path
from s to t contains at least one vertex of S.
Theorem 32.4 (Mengers Theorem). Let s, t be two distinct
vertices in a digraph D and assume there is no edge directed from
s to t. If no (s, t)-separating set has size less than k, then there
exist k internally disjoint directed paths from s to t in D.
Proof: We construct a transportation network N whose vertices
are s, t, and a pair of new vertices x
1
, x
2
for each vertex x of D
other than s, t. For each edge of D directed from s to a vertex x of
D, there is an edge of N directed from s to x
1
of innite capacity
(or any integer capacity k). For each edge of D directed from a
vertex x of D to t, there is an edge of N directed from x
2
to t of
innite capacity. For each edge e of D directed from x to y, where
x, y are vertices of D other than s, t, there is an edge directed from
x
2
to y
1
of innite capacity. Finally, for each vertex x of D other
than s, t, there is an edge directed from x
1
to x
2
of capacity 1.
Note that, by construction, a directed path from s to t in N must
have vertex terms of the form
s, a
1
, a
2
, b
1
, b
2
, . . . , z
1
, z
2
, t
where s, a, b, . . . , z, t are the vertex terms of some directed path
from s to t in D.
Suppose we have an integral ow f from s to t in N of strength k.
The values of f can only be 0 and 1, because every edge of N either
enters a vertex x
1
for which the only edge leaving has capacity 1,
leaves a vertex x
2
for which the only edge entering has capacity 1,
or is an edge x
1
x
2
of capacity 1. By Problem 32B, there exists a
set of k directed paths from s to t in N that pass through any edge
e at most once. These paths are edge-disjoint and hence internally
disjoint, because two directed paths sharing a vertex x
1
or x
2
must
share that single edge x
1
x
2
. So the corresponding k paths in D
are internally disjoint.
456 A Course in Combinatorics
To ensure that such a ow f exists, we consider a cut (S, T) in
N separating s and t and suppose, for contradiction, that the cut
has capacity < k. The only edges of N with capacity possibly less
than k are those directed from x
1
to x
2
for some x V (D). If we
let V be the set of vertices x of D such that x
1
S and x
2
T,
then the capacity of the cut is [V [. Every directed path from s to t
in N must pass through one of these edges x
1
x
2
. Equivalently,
every directed path in D must pass through a vertex x of V , so V
is an (s, t)-separating set. If the hypothesis of the theorem holds,
we have a contradiction.
* * *
A convex polytope P in 1
n
is the convex hull of nitely many
points, i.e.
P =
1
x
1
+ +
k
x
k
:
i
0,
1
+ +
k
= 1.
The dimension of a polytope is the dimension of its ane span.
A supporting hyperplane of P is a hyperplane
H = x : (x) = c
(where is a nonzero linear functional 1
n
1 and c a scalar)
with H P ,= but such that P lies entirely in one of the closed
halfspaces of H, say
P x : (x) c.
A face of P is the intersection of a supporting hyperplane of P with
P (though the empty set is often included as a face). A vertex of P
is a 0-dimensional face; an edge of P is a 1-dimensional face. The
graph G of a polytope P (also called the 1-skeleton of P) has V (G)
and E(G) equal to the vertices and edges, respectively, of P with
incidence determined by inclusion.
A face F is a convex polytope in its own right; indeed, it is the
convex hull of the vertices of P that it contains (see Gr unbaum,
1967). The graph of F is a subgraph of the graph of P.
32. Graph connectivity 457
The following result is due to M. Balinski (1961). For the proof,
we require the following fact from the theory of linear programming;
see Gr unbaum (1967). If s is a vertex of a polytope P in 1
m
and
a linear functional, either (s) (x) for all x P or there exists
a vertex t of P adjacent to s such that (t) > (s).
Theorem 32.5. The graph Gof an n-dimensional convex polytope
is n-connected.
Proof: We will use induction on n. A 0-dimensional polytope
consists of a single point. The graph of a 1-dimensional polytope
consists of two vertices and an edge joining them.
Let x
1
, x
2
, . . . , x
n1
, a, b be distinct vertices in the graph G of
a convex polytope P of dimension n. We may assume P 1
n
.
We want to nd a walk in G from a to b that avoids the vertices
x
1
, . . . , x
n1
. Choose an (n2)-dimensional ane space S that con-
tains x
1
, . . . , x
n1
. Some hyperplane H, i.e. an (n1)-dimensional
ane subspace, on S is such that both a and b are on the same
side of H, i.e. H = z : (z) = and (a) , (b) , for some
linear functional on 1
n
and some scalar . (We allow the case
that one or both points are in H.)
Since P _ H, we can assume P contains points z with (z) >
(otherwise a and b are in H and we can replace and by their
negativesi.e. consider the other side of H).
By the above fact from linear programming, there exist paths p
and q in G from a and b to vertices a
/
and b
/
, respectively, at which
attains its maximum value
/
, and such that no vertices of p and
q other than their initial vertices lie in H. The intersection P H
/
,
where H
/
:= z : (z) =
/
, is a face of P of dimension at most
n 1, and in particular contains a connected graph, so there is a
path r in G from a
/
to b
/
with vertices in H
/
. We concatenate p, r,
and the reverse of q to nd a walk from a to b.
Problem 32D. A graph G is said to be k-Tutte-connected when
for every < k, it is not possible to nd a pair (H, K) of nonempty
edge-disjoint subgraphs of G, each containing at least edges, such
that H K = G and [V (H) V (K)[ = . Such a pair of sub-
graphs may be called an -separation of G. The concepts of vertex-
connectivity and Tutte-connectivity dier. But it is easy to see, for
458 A Course in Combinatorics
a nonnull graph, that 1-Tutte-connectivity is the same as ordinary
connectivity and that 2-Tutte-connectivity is the same as nonsepa-
rability. A 2-Tutte-connected graph with at least two edges cannot
have any loops, since otherwise we have two edge-disjoint subgraphs
(one a loop-graph and the other with all other edges), each with at
least one edge, but with only one vertex in common. Prove that for
a graph G with [V (G)[ k +1, G is k-Tutte-connected if and only
if G is k-vertex connected and has no polygons with fewer than k
edges.
Notes.
See the notes to Chapter 34 for some remarks on H. Whitney.
Karl Menger (19021985) was an Austrian mathematician who
left Austria in 1937 for the United States, where he spent the rest
of his life. He worked in many elds, including logic, didactics, and
economics, but is best known for his work on dimension theory and
curve theory. After his Ph. D., he worked with L. E. J. Brouwer
in Amsterdam to return to Vienna as professor in 1927. There
he founded the Vienna Mathematical Colloquium that had such
famous members as K. Godel and A. Wald and a great number of
visitors. Their results were published in eight volumes before the
seminar was forced to stop because of its many Jewish contributors.
References.
M. Balinski (1961), On the graph structure of convex polyhedra in
n-space, Pacic J. Math. 11, 431434.
L. R. Ford, Jr. and D. R. Fulkerson (1962), Flows in Networks,
Princeton University Press.
B. Gr unbaum (1967), Convex Polytopes, J. Wiley (Interscience).
H. Whitney (1932), Nonseparable and planar graphs, Trans. Amer.
Math. Soc. 34, 339362.
33
Planarity and coloring
We start with some material on deletion and contraction. We shall
see in the next chapter that these are dual concepts; cf. Problem
34C.
Let G be a graph and e E(G). We dene two graphs G
/
e
and
G
//
e
with the same edge set E(G)e. The rst G
/
e
is obtained from
G by deleting the edge e. (It is sometimes convenient subsequently
to remove any isolated vertices that may arise, but we shall allow
them to remain.) The second G
//
e
is obtained by identifying the
endpoints of e and removing e itself. We can think of reeling in
or shrinking an edge in a drawing of a graph as in Fig. 33.1.
Figure 33.1
The chromatic polynomial
G
of a nite graph G was introduced
in Chapter 25:
G
() is the number of proper colorings of G with
colors. For example, note that

K
n
() = ( 1)( 2) ( n + 1).
That
G
is a polynomial in was shown in Chapter 25also see
the remark following (33.1).
Example 33.1. We claim that for a tree T on n vertices,

T
() = ( 1)
n1
.
460 A Course in Combinatorics
To see this, observe that if x is a monovalent vertex of a tree T,
there are 1 colors available for x after the tree on n1 vertices,
obtained by removing x, is given any proper coloring. The claim
follows by induction on the number of vertices.
If e is an edge of G, not a loop, we can break the proper colorings
of G
/
e
into two classes: those in which the endpoints of e have
dierent colors (these are proper colorings of G) and those in which
the endpoints of e have the same color (and these are in one-to-one
correspondence with the proper colorings of G
//
e
). This establishes
(33.1)
G
() =
G

e
()
G

e
().
The equation (33.1) provides another way of seeing that
G
() is
a polynomial in . A graph H with loops has no proper colorings,
i.e.
H
() = 0. For an edgeless graph H on n vertices,
H
() =
n
.
These are both polynomials. Now use (33.1) and induction on the
number of edges.
The chromatic polynomial of a tree T could also be obtained by
induction and (33.1): T
//
e
is a tree and T
/
e
has two components that
are trees.
Problem 33A. Let (G) denote the number of spanning trees in
a graph G (this is sometimes called the complexity of G). Show
that for any nonloop e E(G), (G) = (G
/
e
) +(G
//
e
).
Problem 33B. Find the chromatic polynomial of the n-gon C
n
.
Find the chromatic polynomial of the n-wheel W
n
(this is the graph
obtained from C
n
by adding a new vertex and joining it to all
vertices of C
n
).
It makes no sense to ask for the number of proper colorings of
a graph with 1 colors, but since
G
() is a polynomial, we can
evaluate
G
(1). R. P. Stanley (1973) discovered the combinatorial
interpretation of [
G
(1)[ that follows. An orientation of a graph
is one of the directed graphs arising by choosing for each edge one
of its endpoints to be the head; the other is the tail. (So the number
of orientations of a graph is 2
m
where m is the number of edges that
are not loopsalthough for topological purposes, there are reasons
why a loop should also be considered to admit two directions.)
33. Planarity and coloring 461
Theorem 33.1. The number of acyclic orientations, i.e. orien-
tations in which there are no directed circuits, of a graph G is
(1)
[V (G)[

G
(1).
Proof: If G has loops, then
G
() = 0 and G has no acyclic
orientations, so the theorem holds in this case.
Consider an acyclic orientation of G
/
e
where e is not a loop. There
are two ways to direct e to get an orientation of G. We claim that
one or both of them are acyclic, and that the cases where both are
acyclic are in one-to-one correspondence with acyclic orientations
of G
//
e
. This is because directing e from one end x to the other y
produces a nonacyclic orientation if and only if there is a directed
path from y to x in G
/
e
; there cannot be directed paths from x to y
and also from y to x in G
/
e
since we assumed an acyclic orientation;
and there is no directed path from y to x in G
/
e
if and only if
identifying x and y produces an acyclic orientation of G
//
e
.
Thus
(G) = (G
/
e
) +(G
//
e
)
when (H) is taken to be the number of acyclic orientations of a
graph H. But by (33.1), this same recursion is also satised when
(H) is taken to be (1)
[V (H)[

H
(1). The theorem follows by
induction on the number of edges after we check that it is valid for
edgeless graphs.
By an embedding or proper drawing of a graph G on a surface
o, we mean a drawing where the edges do not cross but meet only
at vertices. More precisely, we mean a representation of the graph
where the vertices correspond to points in o, the edges correspond
to Jordan arcs (continuous one-to-one images of the unit interval)
in o that join the points of o corresponding to the ends of the edge
in G, and where no internal point of any of the Jordan arcs is a
point of any other Jordan arc or one of the points corresponding
to the vertices of G. We consider only embeddings of nite graphs.
While the plane is not compact, it can be naturally embedded in
the sphere (its one-point compactication) and a graph is planar if
and only if it admits an embedding on the sphere.
Examples of graphs that are not planar include K
3,3
and K
5
; we
will mention several ways to see this later. Subdivisions of these
462 A Course in Combinatorics
graphs, and any graph containing one of these subdivisions, will also
not be planar. An important theorem that we shall not prove here
is the following. Proofs may be found in Chartrand and Lesniak
(1986), Tutte (1984), and Diestel (1997).
Theorem 33.2 (Kuratowski). A graph G is planar if and only
if it has no subgraphs isomorphic to subdivisions of K
5
or K
3,3
.
Problem 33C. Find a subdivision of K
3,3
as a subgraph of the
Petersen graph.
A minor of a graph G is a graph H obtained from G by delet-
ing and contracting edges, and then possibily deleting some or all
isolated vertices.
It can be seen that that the order in which edges are deleted
or contracted is not important. For example, if the edges of a
subset S E(G) are contracted in any order, we obtain a graph
isomorphic to the graph G
//
S
, called a contraction of G, dened as
follows. The vertices of G
//
S
are, formally, the components of the
spanning subgraph, denoted G:S, of G with edge set S. The edge
set of G
//
S
is E(G) S. The ends of an edge e E(G) S are, in
G
//
S
, to be the components of G:S that contain the ends of e as an
edge of G. Informally, the components of G:S are being shrunk to
single vertices.
We use the notation G
/
S
for the graph obtained by deleting the
edges in S, i.e. G:(E(G) S), but it is convenient to require that
any isolated vertices are also deleted when using this notation.
Example 33.2. Let A be the ve spokes in the usual drawing of
the Petersen graph P, on the left of Fig. 1.4. The subgraph P:A
has ve components, so P
//
A
will have ve vertices. The contraction
G
//
S
in this case is K
5
. If B is the set of ve edges in the interior
pentagram, then P:B has six components and P
//
B
is the wheel W
5
.
If C is the edge set of a spanning tree of P, then P
//
C
consists of a
single vertex and six loops. (A graph with one vertex may be called
a cotree.)
Note that it is quite likely that G
//
S
will have loops and/or parallel
edges even if G is simple. (Those contractions of a simple graph
G that have no loops are in one-to-one correspondence with the
33. Planarity and coloring 463
elements of the lattices (G), sublattices of the partition lattices,
introduced in Example 23.6).
A graph G is a contraction of any of its subdivisions H, since
contracting any k 1 edges of a path-graph with k edges produces
a link-graph.
If we have a drawing of a graph G on a surface o and e E(G), it
is clear that there is a natural drawing of G
/
e
on o. If e is a nonloop,
there is also a natural induced drawing of G
//
e
on o: the ends of the
edge can be continuously brought closer on the surface (see Fig.
33.1). (According to our combinatorial denition, contracting a
loop is the same as deleting it.) In particular, if G admits an
embedding on a surface o, then so does every minor of G.
Problem 33D. Show that a graph G that has a minor isomorphic
to K
5
or K
3,3
also contains a subdivision of K
5
or K
3,3
. Somewhat
more generally, prove:
(i) if a trivalent graph H occurs as a minor of a graph G, then G
contains (as a subgraph) a subdivision of H. So if K
3,3
is a minor,
then a subdivision of K
3,3
is a subgraph;
(ii) if K
5
is a minor of G, then G contains a subdivision of K
5
or
a subdivision of K
3,3
.
A deep and extremely important theorem due to N. Robertson
and P. Seymour states that in every innite set of graphs there are
two such that one is a minor of the other. See Chapter 12 of Diestel
(1997) for a discussion of part of this work. It follows that the set of
graphs that do not embed on a given surface and are minimal with
respect to this property (every proper minor being embeddable) is
nite. For the plane, there are two.
Given an embedding of a graph G on a surface o, we dene
the faces or regions with respect to the embedding to be the topo-
logically connected components that result when the vertices and
edges (more precisely, the points corresponding to vertices and sets
of points in the Jordan arcs corresponding to edges) are removed
from o. We denote the set of faces by F(G). We realize that this
is horrible notation, since the faces depend on the embedding (in
particular, on the surface) and are not in general determined by G.
But we will use it anyway.
464 A Course in Combinatorics
In general, these faces may or may not be 2-cells (homeomorphic
to the open unit disk, i.e. without holes). But for an embedding of
a connected graph on the sphere, the regions will be 2-cells. This
fact is related to the Jordan curve theorem which asserts that if
the points of a one-to-one continuous image of the unit circle in the
sphere are removed, the resulting topological space has exactly two
simply connected components that are 2-cells.
Theorem 33.3 (Eulers formula). For a drawing of a connected
planar graph G on the sphere or plane,
f e +v = 2
where f, e, and v are, respectively, the numbers of faces, edges, and
vertices. More generally, for any graph G embedded on the sphere,
(33.2) 1 [F(G)[ +[E(G)[ [V (G)[ +[C(G)[ = 0,
where C(G) denotes the set of components of G.
Proof: We prove the latter assertion by induction on the num-
ber of edges of G. For an edgeless graph, [C(G)[ = [V (G)[ and
[F(G)[ = 1, so (33.2) holds.
Suppose G has a nonloop a. Consider an embedding of G and
note that contracting an edge will not aect the number of faces.
Apply the induction hypothesis to G
//
a
to get
1 [F(G
//
a
)[ +[E(G
//
a
)[ [V (G
//
a
)[ +[C(G
//
a
)[ = 0.
We have [E(G)[ = [E(G
//
a
)[ + 1, [V (G)[ = [V (G
//
a
)[ + 1, [C(G)[ =
[C(G
//
a
)[, and also [F(G)[ = [F(G
//
a
)[. So (33.2) holds for G.
Suppose G has loops; let a be one of these, and apply the induc-
tion hypothesis to G
/
a
to get
1 [F(G
/
a
)[ +[E(G
/
a
)[ [V (G
/
a
)[ +[C(G
/
a
)[ = 0.
Of course, [E(G)[ = [E(G
/
a
)[ + 1, [V (G)[ = [V (G
/
a
)[, and [C(G)[ =
[C(G
/
a
)[. On the sphere, the Jordan curve theorem implies that the
two faces incident with e are distinct, so [F(G)[ = [F(G
/
a
)[ +1 and
we obtain (33.2) for G.
33. Planarity and coloring 465
Example 33.3. We show that K
5
and K
3,3
are not planar. The
set of edges incident with a face in a planar embedding of a non-
separable graph G (other than the link-graph) contains a polygon
subgraph; see Chapter 34. The degree of a face is the number of
incident edges (i.e. its degree as a vertex of the dual graph of Chap-
ter 34). The sum of the degrees of the faces is twice the number of
edges, since every edge is incident with two faces.
Consider a hypothetical embedding of K
5
in the plane. By Eu-
lers formula, there are seven faces. But each has degree at least
3, so the sum of the degrees is at least 21, which, of course, is a
contradiction to [E(K
5
)[ = 10.
Similarly, there would be ve faces in a hypothetical embedding
of K
3,3
in the plane, each of degree at least 4 since K
3,3
has girth
4. Then the sum of the degrees would be at least 20, contradicting
[E(K
3,3
)[ = 9.
(A similar argument shows that the Petersen graph is not planar.
But we already knew this since both K
5
and K
3,3
occur as minors
of the Petersen graph and they are not planar.)
Problem 33E. Determine all pairs (d
1
, d
2
) of integers with d
i
2,
i = 1, 2, so that there exists a planar graph (not necessarily simple)
that is regular of degree d
1
and such that all faces have degree d
2
.
(This will include the degrees of the vertices and faces of the ve
Platonic solids.)
E. Steinitz (1922) characterized the graphs of 3-dimensional con-
vex polytopes.
Theorem 33.4. A graph G is the graph of a 3-dimensional convex
polytope if and only if G is simple, planar, and 3-connected.
That the graph of a 3-dimensional polytope is 3-connected is a
special case of Theorem 32.5. We do not prove the rest of the
theorem.
The question Can the countries of a map on the sphere be col-
ored with four or fewer colors so that adjacent countries are colored
dierently?, occurs in a letter dated October 23, 1852, from Au-
gustus de Morgan to Sir William Rowan Hamilton. It is likely that
the originator of the question was the brother, Francis Guthrie, of
a student, Frederick Guthrie, of de Morgan.
466 A Course in Combinatorics
Coloring the countries (regions) of a planar map is equivalent to
coloring the vertices of the dual graph that we will introduce in the
next chapter. So for us, the four color problem is the question of
whether every planar graph is 4-colorable.
Four colors are clearly necessary since K
4
is planar, but it is easy
to nd many other planar graphs G that cannot be 3-colored. This
problem has motivated a great deal of graph theory. More on the
history of the four color problem may be found in Biggs, Lloyd,
and Wilson (1976).
In 1890, P. J. Heawood proved the Five Color Theorem, Theo-
rem 33.6 below: a loopless planar graph G has chromatic number
(G) 5. Heawood used ideas of A. B. Kempe (in particular, the
recoloring idea reviewed below). In 1976, K. Appel and W. Haken
nally settled the four color problem with an announcement of a
proof of the Four Color Theorem: (G) 4. Surprisingly, perhaps,
their proof did not use the vast quantity of work and theories de-
veloped in this century but went back to the ideas of Kempe and
Heawood. Another surprise, perhaps, is that their proof required
over 1000 hours of computer time. It was not clear when they be-
gan that the computer would ever nish. As an oversimplication,
there were many cases to consider but it was not certain that their
number would be nite; the computer was programmed to gener-
ate cases itself and could conceivably have gone on forever breaking
cases into subcases. See Appel, Haken, and Koch (1977), and Ap-
pel and Haken (1977). We have no space to discuss this work in
any detail, but will prove the Five Color Theorem below.
Proposition 33.5. Every nonnull simple planar graph G has a
vertex of degree at most 5.
Proof: This follows quickly from Eulers formula. It suces to
prove that a component has such a vertex, so assume G is con-
nected. Suppose a drawing of G has f regions, e edges, and v
vertices. If no vertex of degree less than six exists, then the sum of
the v degrees of the vertices, which is 2e, is at least 6v; that is,
v
e
3
.
Since G is simple (and we may assume it is not a link-graph with
33. Planarity and coloring 467
two vertices joined by a single edge), each region is incident with
at least three edges, so the sum of the f degrees of vertices of G

,
which is 2e, is at least 3v; that is
f
2e
3
.
But then we have the contradiction
2 = f e +v
e
3
e +
2e
3
= 0.

Of course, coloring a loopless graph is equivalent to coloring a

simple graph (just suppress the multiple edges). An immediate
consequence of the above result is the Six Color Theorem: a loopless
planar graph G has (G) 6. When a vertex x of degree at most
ve is deleted from a simple planar graph and the result colored
with six colors inductively, there is always a color available for x.
We recall the idea of recoloring introduced in Chapter 3. Suppose
G is properly colored and let and be two of the colors. The
colors and can be switched on any component of the subgraph
of G induced by the vertices of colors and to obtain another
proper coloring. So given two vertices x and y of colors and
, respectively, either there is a path (of odd length) with vertex
terms a
0
= x, a
1
, . . . , a
n
= y that alternate colors and (a so-
called Kempe chain), or there is a proper coloring in which both x
and y have color , say, while vertices of colors other than and
retain the same color.
Theorem 33.6. If G is a loopless planar graph, then (G) 5.
Proof: We use induction on the number of vertices. The assertion
is surely true for graphs with at most ve vertices. Consider a
planar drawing of a loopless graph G and assume the theorem holds
for all graphs with one vertex less. We can assume G is simple.
By Proposition 33.5, there exists a vertex x V (G) of degree
5. Let G
x
denote the graph obtained from G by deleting the
vertex x and the incident edges. We have a natural planar drawing
of G
x
obtained from the drawing of G. By the induction hypothesis,
468 A Course in Combinatorics
G
x
has a proper coloring with ve colors. There is a color available
for x, and we are done, unless x has degree exactly ve and the
vertices y
1
, y
2
, . . . , y
5
(which we write in the cyclic order given by
the drawing) adjacent to x in G have received ve distinct colors
in the coloring of G
x
. So we assume that y
i
has received color i,
i = 1, 2, 3, 4, 5. See Fig. 33.2.
Consider colors 1 and 3. Either there exists a proper coloring of
G
x
in which y
1
and y
3
both have color 1 (while y
2
, y
4
, and y
5
retain
their original colors) or there exists a simple path with vertex terms
(33.3) y
1
, a
1
, b
1
, a
2
, b
2
, . . . , a
s
, b
s
, y
3
,
say, in G
x
with vertices alternately colored 1 and 3. In the rst
case, x may be assigned color 3 and we are done. Similarly, either
there exists a proper coloring of G
x
in which y
2
and y
4
both have
color 2 (while y
1
, y
3
, and y
5
retain their original colors) or there
exists a simple path with vertex terms
(33.4) y
2
, c
1
, d
1
, c
2
, d
2
, . . . , c
t
, d
t
, y
4
,
say, in G
x
with vertices alternately colored 2 and 4. In the rst
case, x may be assigned color 4 and we are done.
To complete the proof, we note that both paths above cannot
exist. If the path in (33.3) exists, then
(33.5) x, y
1
, a
1
, b
1
, a
2
, b
2
, . . . , a
s
, b
s
, y
3
, x
is the vertex sequence of a simple closed path, which divides the
plane into two regions. The vertex y
2
is in one region while y
4
is
in the other. Any path from y
2
to y
4
in G
x
must necessarily have
a vertex in common with the path in (33.4) and in particular must
have a vertex of color 1 or 3; thus a path as in (33.5) cannot exist.

33. Planarity and coloring 469

Figure 33.2
In Chapter 17, we treated list-colorings of graphs to explain
Galvins proof of the Dinitz conjecture. We shall now treat a result
of C. Thomassen (1994) that provides a second proof of the ve
color theorem by rst giving each vertex of the graph a list of ve
dierent colors, not necessarily the same list for each vertex. One
of the ideas of the proof should be especially noted because it often
can be used to make proofs easier. It involves making an induction
hypothesis more stringent with the eect that the induction step
goes more easily.
Theorem 33.7. Let G be a planar graph with N vertices v
1
,
v
2
, . . . , v
N
. For i = 1, . . . , N let S
i
be a set of ve elements (which
we call colors). There exists a mapping f on the vertices, with
f(v
i
) S
i
such that f(v
i
) ,= f(v
j
) for all adjacent pairs v
i
, v
j
(a ve
list-coloring of G).
Proof: We rst make the problem harder by adding extra edges
which change G into a triangulated polygon. Let the numbering be
such that v
1
, v
2
, . . . v
k
are the succesive vertices of the polygon (see
Fig. 33.3). For each of these vertices remove two colors from its list
so that vertices on the perimeter of the polygon only have lists of
three available colors. Again, the problem is more dicult. Finally,
restrict matters even further by selecting two adjacent vertices on
the perimeter and assigning to each a color from their list. We claim
that with these initial conditions, the coloring can be completed.
The proof is by induction on the number of vertices (small cases
are trivial).
470 A Course in Combinatorics
v

v
k
v
k
u

u
m
Figure 33.3
For the induction, we distinguish two cases. In the rst, we
assume that the polygon has an edge connecting two vertices on
the perimeter that are not adjacent on the perimeter, i.e. a chord.
The chord divides the polygon into two parts, one of which has
both of the precolored vertices (one possibly as an endpoint of
the chord). The coloring of that part can be completed by the
induction hypothesis. From this coloring, the other part inherits a
(pre)coloring of two adjacent vertices on the perimeter and again
induction nishes the job.
Now assume that there is no chord. Let v
1
and v
2
be the pre-
colored vertices on the perimeter and let v
1
, u
1
, u
2
, . . . , u
m
, v
k1
, in
that order (clockwise, see Fig. 33.3) be the vertices adjacent to v
k
.
The vertices u
i
are interior points and because of the triangulation,
there are edges (v
1
, u
1
), (u
1
, u
2
), . . . (u
m1
, u
m
), (u
m
, v
k1
). The list
of v
k
has at least two colors not equal to the color of v
1
. We pick
(these) two and remove them from the lists of u
1
to u
m
and, if nec-
essary, further shorten these lists to size 3. By induction, we can
complete the coloring of Gv
k
. Possibly v
k1
has obtained one
of the two colors on v
k
s list but there is one left to complete the
coloring of G.
Notes.
A. B. Kempe (18491922) took a degree in 1872 with special dis-
tinction in mathematics at Trinity College, Cambridge where he
also gained a musical reputation with his ne countertenor voice.
33. Planarity and coloring 471
He chose law as his profession while not relinquishing his mathe-
matical studies.
The Polish mathematician Kazimierz (or Casimir) Kuratowski
(18961980) taught at Warsaw University (for nearly 40 years) and
Lwow Polytech before retiring in 1966. His celebrated theorem
characterizing planar graphs was proved in 1930.
Ernst Steinitz (18711928) was professor of mathematics at Bres-
lau (now Wroclaw) and Kiel. He made important contributions to
eld theory (with Hilbert).
In 1993, M. Voigt found an example of a planar graph with 238
vertices that is not four list-colorable.
References.
K. Appel and W. Haken (1977), The solution of the four-color map
problem, Scientic American 237, 108121.
K. Appel, W. Haken, and J. Koch (1977), Every planar map is four
colorable, Illinois J. Math. 21, 429567.
N. L. Biggs, E. K. Lloyd, and R. J. Wilson (1976), Graph Theory
17361936, Oxford University Press.
G. Chartrand and L. Lesniak (1986), Graphs and Digraphs, 2nd
edn., Wadsworth.
R. Diestel (1997), Graph Theory, Springer-Verlag Graduate Texts
in Mathematics 173.
R. P. Stanley (1973), Acyclic orientations of graphs, Discrete Math.
5, 171178.
E. Steinitz (1922), Polyeder und Raumeinteilungen, Enzykl. Math.
Wiss. 3, 1139.
C. Thomassen (1994), Every planar graph is 5-choosable, J. Com-
bin. Theory Ser. B 62, 180181.
W. T. Tutte (1984), Graph Theory, Encylopedia of Math. and its
Appl. 21, Addison-Wesley. Reissued by Cambridge University
Press.
34
Whitney duality
A Jordan arc divides a small neighborhood of one of its internal
points into two halves. Thus in addition to being incident with
two vertices (which may coincide), an edge in a map is incident
with two regions (which may coincide). See Fig. 34.1.
Figure 34.1
This allows us to dene the dual graph G

of a graph G with
respect to a 2-cell embedding of G on a surface o. The vertices
V (G

) are to be the regions with respect to the embedding; the

edges E(G

) are to be the same as the edges of G; the ends of an

edge e in G

are to be the regions with which e is incident. Two

examples are shown in Fig. 34.2.
We have dened G

abstractly, but it is important to note that

it admits an embedding on the same surface o: we can represent a
region by a point in its interior; the interior points in two regions
sharing an edge e can be joined by a Jordan arc crossing only the
Jordan arc representing e in G. The drawing of the dual of the
cube in Fig. 34.2 is shown in Fig. 34.3.
Examples will convince the reader that (G

is isomorphic to G.
Two other facts that we invite the reader to think about but do
34. Whitney duality 473
not attempt to prove formally are: a graph that is not connected
cannot have a 2-cell embedding; the dual graph with respect to a
2-cell embedding is always connected.
Figure 34.2
The planar duals of the series of polygons are the bond-graphs,
i.e. the graphs with two vertices and any positive number of edges
joining them. In particular, the loop-graph and the link-graph are
dual. The planar dual of a tree is a cotree, and a planar dual of a
cotree is a tree.
474 A Course in Combinatorics
Figure 34.3
In general, a graph may have several planar dual graphs. Fig.
34.4 shows two isomorphic cotrees and their nonisomorphic planar
duals with respect to the two embeddings. Fig. 34.5 shows two
embeddings of a nonseparable graph G and the dual graphs with
respect to each embedding, which are not isomorphic.
Figure 34.4
Figure 34.5
The goal of this chapter is to understand the relation between
a planar graph and its dual graph(s) combinatorially. We will
use this understanding (Whitney duality) to prove some theorems
34. Whitney duality 475
about planar graphs and their duals (without any reference to pla-
nar drawings or topology).
Recall that the incidence matrix N = N(G) of a graph G is
the matrix whose rows are indexed by V (G), whose columns are
indexed by E(G), and where
N(x, e) =
_
_
_
1 if x is incident with a nonloop e,
2 if e is a loop incident with x
0 if x is not incident with e.
We are going to work over the eld F
2
of two elements, so loops
essentially correspond to columns of all zeros. Every other column
contains two ones and, in particular, the sum modulo 2 of all rows
of N is the zero vector.
We may consider the code ((G) of a graph G, which is the binary
code generated by the incidence matrix N(G). A graph H is a
Whitney dual of a graph G when E(G) = E(H) and the code ((G)
of G is equal to the dual code ((H)

of H. Clearly, H is a Whitney
dual of G if and only if G is a Whitney dual of H. This denition
requires that G and H have the same edge set, which is convenient
notationally. But, informally, it is only necessary that the edges of
G and H be labeled by the same set, or otherwise put in one-to-one
correspondence.
Note that the code of a polygon with n edges consists of all even-
weight words of length n; the code of a bond-graph with n edges
consists of the two constant words. The code of a tree consists of
all words; the code of a cotree consists of only the zero word.
Some terms in graph theory are used in many contexts, with
varying meanings. In this chapter, we use cycle, circuit, cutset,
and bond of a graph G to mean certain subsets, described below,
of the edge set E(G). Other uses of these terms the reader may
know must be temporarily forgotten.
As in Chapter 20, we use the term support of a codeword x to
denote the subset of the indexing set for its coordinates that are
nonzero in x. For binary codes, we can identify the codewords with
their supports. The supports of words in ((G) are called cutsets of
G; supports of words in ((G)

are called cycles of G. The terms

circuits and bonds are used, respectively, for minimal nonempty
476 A Course in Combinatorics
cycles and cutsets. We use cycle space and cutset space to mean
the set of all cycles and cutsets, respectively. Of course, these are
vector spaces over the eld F
2
, with addition of two sets equal to
their symmetric dierence.
It is easy enough to understand when a subset S E(G) is a
cycle, or a cutset. The binary sum of the rows of the incidence ma-
trix indexed by vertices in X has support equal to the set (X, Y )
of edges that cross from X to Y := V (G) X. i.e. that have one
end in X and one end in Y . That is, S is a cutset if and only if
S = (X, Y ) for some X, Y . We use this notation only when X
and Y partition the vertex set; we allow X or Y to be empty, in
which case S is empty.
A vector, with support S, say, is orthogonal over F
2
to a row of
N indexed by a vertex x when and only when x is incident with an
even number of edges in S, loops being counted twice. So the cycles
are those subsets of E(G) so that every vertex has even degree in
G:S, the spanning subgraph of G with edge set S.
Example 34.1. Let G be the graph below.
1
d
4
e
c
3
b
2
a
The incidence matrix of G is
a b c d e
1
2
3
4
_
_
_
1 0 0 1 1
1 1 0 0 0
0 1 1 0 1
0 0 1 1 0
_
_
_
.
There are four cycles in G: , a, b, e, c, d, e, and a, b, c, d.
The latter three are, in fact, circuits. There are eight cutsets. The
34. Whitney duality 477
set a, b, c, d is both a cycle and a cutset; it is the support of the
sum of rows 2 and 4. But a, b, c, d is not a bond; it contains a
smaller cutset, e.g. the support a, b of row 2. The set a, b is a
bond. Another example of a bond is b, d, e; it is the support of
the sum of rows 1 and 2, or of 3 and 4.
Problem 34A. (i) Let G be connected. Prove one of the equiva-
lent statements: the number of cutsets of G is exactly 2
[V (G)[1
; the
dimension of the cutset space of G is [V (G)[ 1.
(ii) Prove, in general, that the dimension of the cutset space of G
is [V (G)[ [C(G)[, where C(G) is the set of components of G. (As
a corollary, the dimension of the cycle space is [E(G)[ [V (G)[ +
[C(G)[.)
Example 34.2. The cycle spaces of K
6
and of the Petersen graph
are binary codes of length 15 of dimensions 10 and 6, respectively.
(Fig. 35.1 in the next chapter shows that K
6
is the dual of the
Petersen graph with respect to an embedding in the real projective
plane, but they are not Whitney duals, since the dimensions of
their codes sum to more than the number of edges.) The minimum
distances of these codes are 3 and 5, respectively. We remark that
the vector of all 1s can be added to a generating set for these
two codes and the minimum distances remain 3 and 5 while the
dimension goes up by 1 in both cases.
Theorem 34.1. Let G be a graph.
(i) A set S E(G) is a circuit if and only if it is the edge set of
a polygon subgraph of G.
(ii) Assume G is connected. A cutset (X, Y ) is a bond of G
if and only if the subgraphs of G induced by X and Y are both
connected.
Proof: Clearly the edge set of a polygon is a minimal nonempty
cycle. A nonempty cycle is certainly not the edge set of a forest
(since every tree has monovalent vertices, say) and so contains the
edge set of a polygon; so a minimal nonempty cycle is the edge set
of a polygon.
Let S = (X, Y ). If the subgraphs induced by X and Y are con-
nected, then S cannot properly contain a nonempty cutset because
deleting any proper subset of the edges in S does not produce a
478 A Course in Combinatorics
disconnected graph. If the subgraph induced on Y , say, is not con-
nected, there is a partition of Y into nonempty subsets Y
1
and Y
2
with no edges between them. Then S
/
:= (XY
1
, Y
2
) is contained
in S. Connectivity implies that there are edges between X and Y
1
and X and Y
2
, and this means S
/
is both nonempty and a proper
subset of S.
The reader may check that a set of edges of an arbitrary graph G
is a bond of G if and only if it is a bond in one of the components
of G. Also note that a subset S of edges of a graph G is a bond
of G if and only if the deletion of the edges in S results in a graph
with more components than G (e.g. a disconnected graph, if G
is connected to start with) and S is minimal with respect to this
property.
Given a set S of edges of a graph G, S is a circuit if and only if
deleting the edges not in S (and any isolated vertices that result)
produces a polygon; S is a bond if and only if contracting the edges
not in S (and deleting any isolated vertices that result) produces a
bond-graph.
Notice that the set S(x) of edges of a graph G that are incident
with a vertex x, not including loops, is always a cutset. These
span the cutset spacethey are the supports of the rows of the
incidence matrix. In view of Theorem 34.1(ii), S(x) is a bond for
every x V (G) if and only if G is 2-connected (or is a connected
graph with two vertices).
Problem 34B. Show that the circuits of a graph span the cycle
space, and the bonds span the cutset space. To do this, just verify
that the nonzero codewords in a binary linear code (, whose sup-
ports are minimal among the supports of all nonzero codewords,
generate the code. (This is true for linear codes over any eld.)
Hence G and H are Whitney duals when the circuits of G are
exactly the bonds of H, or when the bonds of G are exactly the
circuits of H.
Example 34.3. The two graphs described by each table in Fig.
34.2 are Whitney duals. This is because each pair consists of
planar duals (see Theorem 34.2 below). For example, a, e, g, c
and a, j, b are bonds in the cube G and circuits in G

; the set
34. Whitney duality 479
a, b, c, d is a circuit in G and a bond in G

.
Problem 34C. Let G and H be Whitney duals with E(G) =
E(H) = E. Explain why G
/
e
and H
//
e
are Whitney duals for any
e E.
Problem 34D. Let G and H be Whitney duals with E(G) =
E(H) = E, both of which are connected. Show that S is the edge
set of a spanning tree in G if and only if E S is the edge set of a
spanning tree in H.
Theorem 34.2. A graph G is planar if and only if it has a Whitney
dual. A dual G

of G with respect to any planar drawing of G is a

Whitney dual of G.
We prove only the second sentence (which implies the only if
part of the rst sentence) at this time, and postpone the other part
until the very end of the chapter.
Proof: Let G

be a planar dual of G with respect to some planar

drawing of G.
Let S be a circuit of G. So S is the edge set of a polygon subgraph
P of G that in the drawing, by the Jordan curve theorem, divides
the sphere or plane into two parts; let T
1
be the set of faces inside,
and T
2
the set of faces outside, P. This is a partition of the vertices
of G

. Clearly, each edge of P is incident with one face of T

1
and
one of T
2
. Any other edge of G lies (except possibly for its ends) in
one of the parts and so is incident only with members of T
1
or T
2
.
In summary, S = (T
1
, T
2
) is a cutset in G

. The circuits generate

the cycle space of G, so we have shown that the cycle space of G is
contained in the cutset space of G

.
Now we note the fact that the cutset space of G

is contained in
the cycle space of G for an embedding on any surface. There is a
cyclic ordering to the edges incident with with a given vertex (loops
appear twice). Partition the vertices of G

(the regions or faces)

say they are colored red and blueand consider the cutset C of G

consisting of edges incident with one red and one blue region. The
number of edges of C incident with a vertex of G is certainly even;
see Fig. 34.6. (This is the same as saying that if the vertices of a
polygon are colored red and blue, then the number of edges with
ends of dierent colors is even.) So C is a cycle in G.
480 A Course in Combinatorics
Hence the cycle space of G is equal to the cutset space of G

.
Figure 34.6
Eulers formula is a corollary of what we have proved so far.
Problem 34A says the dimension of the cycle space of G is [E(G)[
([V (G)[ [C(G)[), while the dimension of the cutset space of G

is
[F(G)[ 1 (we have remarked that G

is always connected). When

we equate these, we obtain (33.2).
Whitneys Theorem allows a quick proof of the following theorem
of S. MacLane (1937).
Theorem 34.3. A graph G is planar if and only if the cycle space
of G has a basis B = B
1
, B
2
, . . . , B
k
so that each edge of G occurs
in at most two members of B.
Proof: If G is planar, the boundaries of all but one face will pro-
vide such a basis. For if G

is the dual of G with respect to some

embedding on the plane, then, since G

is connected, the sets of

all nonloops incident with all but one of the vertices of G

(faces
of G) provide a basis for the cutset space of G

(cycle space of G)
and each edge is an element of at most two of these sets.
Now assume the cycle space of G has a basis as in the statement
of the theorem. Let B
0
= B
1
+B
2
+ +B
k
. Any edge of G is in 0
or 2 of the sets B
i
, 0 i k. Dene a graph H whose vertices are
0, 1, . . . , k, whose edge set is E(G) and where an edge e is incident
in H with the two integers j and so that e B
j
and e B

when
e is in two of the sets B
i
, and where e is a loop in H (at any vertex
i) if e is in none of the sets B
i
. We claim H is a Whitney dual of
G.
The set of nonloops of H that are incident in H with a vertex i is
B
i
, of course, and this is a cutset in H. Since the B
i
s generate the
cycles of G, the cycle space of G is contained in the cutset space of
34. Whitney duality 481
H. But we know that the cycle space of G has dimension k, while
the cutset space of H has dimension k, since H has k+1 vertices.
We conclude that the cycle space of G is equal to the cutset space
of H, i.e. they are Whitney duals. Hence G is planar by Theorem
34.2.
As promised, we now give some results about planar graphs and
their duals using Whitney duality (and not referring to the plane
or sphere at all).
Theorem 34.4. If G and H are Whitney duals and H is con-
nected, then G is bipartite if and only if H is Eulerian (admits an
Eulerian circuit).
Proof: G is bipartite if and only if the entire edge set E(G) is
a cutset of G. Since G

is connected, it is Eulerian if and only if

E(G) is a cycle of G

; see Theorem 1.2. The two conditions are

equivalent since G

is a Whitney dual of G.
A Tait coloring (after the English mathematician P. G. Tait) of a
trivalent graph is a coloring of the edges with three colors, say , ,
and , so that each color appears at every vertex. Fig. 34.7 shows
such a coloring. The existence of a Tait coloring of a trivalent graph
is equivalent to partitioning the edges into three perfect matchings
(cf. Chapter 5) and so a bipartite trivalent graph has a Tait coloring
by Problem 5A.
Figure 34.7
Theorem 34.5. If G and H are Whitney duals and G is trivalent,
then G admits a Tait coloring if and only if H is 4-colorable.
482 A Course in Combinatorics
Proof: We claim that a trivalent graph G admits a Tait coloring
if and only if there exist two cycles S
1
and S
2
in G so that E(G) =
S
1
S
2
. Given the coloring, take S
1
to be the union of the edges of
colors and , and S
2
to be the union of the edges of colors and
. These are both cycles since every vertex is incident with two
edges of S
1
and two edges of S
2
. Conversely, if E(G) is the union
of two cycles S
1
and S
2
, color an edge e with color
_
_
_
if e S
1
, e / S
2
,
if e / S
1
, e S
2
,
if e S
1
S
2
.
Every vertex x has even degree (between 0 and 3) in the spanning
subgraphs with edge sets S
1
and S
2
; but E(G) = S
1
S
2
implies
that both degrees are 2 and that there is one edge of each color
incident with x.
Our second claim is that a graph H can be properly 4-colored if
and only if E(H) can be expressed as the union of two cutsets of H.
If E(H) is the union of (X
1
, X
2
) and (Y
1
, Y
2
), then X
1
Y
1
, X
1

Y
2
, X
2
Y
1
, X
2
Y
2
are the color classes of a proper 4-coloring. If
A, B, C, D are the color classes of a proper 4-coloring, then E(H)
is the union of (A B, C D) and (A C, B D).
The two observations and the denition of Whitney duals com-
plete the proof.
An isthmus of a graph is an edge whose deletion increases the
number of components (i.e. disconnects the graph if it was originally
connected). This is the dual concept of a loop. An isthmus is an
edge e so that the singleton set e is a cutset; a loop is an edge e
so that the singleton set e is a cycle. (Some authors use acyclic
edge for an isthmus since it is an edge contained in no circuits. The
term bridge has also been used.)
Problem 34E. Show that a trivalent graph with an isthmus does
not have a Tait coloring.
In view of the Four Color Theorem, every isthmus-free trivalent
planar graph does admit a Tait coloring. (An isthmus in G would
be a loop in the dual which could then not be properly colored.)
34. Whitney duality 483
Planarity is necessary, since the Petersen graph is an example of
an isthmus-free trivalent graph which has no Tait coloring.
Conversely, we claim that to prove the Four Color Theorem it
would be sucient to establish that every isthmus-free trivalent
planar graph does admit a Tait coloring. For suppose that the re-
gions of every isthmus-free trivalent planar graph G can be properly
4-colored. Then, given a planar graph G, replace each vertex x of
degree ,= 3 by the conguration indicated for the case of a vertex of
degree 5 in Fig. 34.8. (This may be done even for vertices of degrees
1 and 2.) This creates more regions (more vertices in the dual G

)
but preserves adjacency of the original regions, so if the regions of
this larger but trivalent graph can be 4-colored, then we obtain a
coloring for the original graph. By Theorem 34.5, the regions of
the larger but trivalent graph can be 4-colored if and only if it has
a Tait coloring.
Figure 34.8
Tait conjectured in 1880 that every 3-connected planar trivalent
graph has a Hamiltonian circuit, which would have implied that the
graph admits a Tait coloring. W. T. Tutte (1956) found a coun-
terexample but did prove that every 4-connected planar trivalent
graph is Hamiltonian.
Theorem 34.6. If G and H are Whitney duals without isolated
vertices and G is nonseparable, then H is nonseparable.
Proof: Theorem 32.2 shows that any two edges of G are contained
in a circuit of G. Hence any two edges of H are contained in a bond
of H. But if two edges a and b are contained in a bond (X, Y ), we
can nd a simple path in the subgraph induced by X, joining one
end of a and one of b, and a simple path in the subgraph induced
484 A Course in Combinatorics
by Y joining the other ends, and we nd a polygon containing a
and b. So any two edges of H are contained in a circuit of H. This
implies that H is nonseparable.
Corollary. The set of edges incident with a face in a planar em-
bedding of a nonseparable graph G, not a link-graph, is a circuit
in G.
Proof: The set S of edges incident with a face is the set of edges
incident with a vertex in G

. But G

is nonseparable by Theorem
34.5, so S is a bond in G

. Then S is a circuit in G.
Problem 34F. Two edges a and b of a connected graph G that
are contained together in a circuit of G are also contained in some
bond of G.
Theorem 34.7. Let G and H be simple graphs without isolated
vertices that are Whitney duals. If G is 3-connected, then H is
3-connected.
Proof: Assume G is 3-connected. For x V (G), let S(x) denote
the set of edges incident with x. Each set S(x) is a bond with
the property that G
/
S(x)
is nonseparable, and these bonds span the
cutset space of G by denition.
Thus in H, there exists a family T of circuits that spans the cycle
space of H and where each member S of T is a circuit with the
property that H
//
S
is nonseparable.
We know H is nonseparable; suppose for contradiction that it is
not 3-connected. Then H is the edge-disjoint union of subgraphs
A and B of H, each with at least two edges, and where A B is a
graph consisting of two isolated vertices, say x and y. If a polygon
in H contains edges of both A and B, it must have x and y among
its vertices. If the edges of such a polygon are contracted, we would
obtain a separable graph since it would be the union of the graphs
obtained from A and B by contraction of the edges of the polygon
in each (which would identify x and y); these contracted subgraphs
will have only this vertex resulting from the identication of the
vertices of the polygon in common (and the reader can check that
there still remains at least one edge in each contracted subgraph).
34. Whitney duality 485
Thus the circuits in T are contained completely in either E(A)
or E(B). Let C be any circuit of H. Since T spans the cycle space
of H, C E(A) and C E(B) are, respectively, the modulo 2 sum
of members of T that are contained in E(A) and E(B). Hence each
of CE(A) and CE(B) would be cycles, and then the minimality
of C would imply that C E(A) or C E(B) is empty. It follows
that every circuit of H is contained in either E(A) or E(B), which
contradicts the nonseparability of H.
Problem 34G.
(i) Let G be a simple 3-connected graph and S a bond in G.
Show that G
/
S
is nonseparable if and only if S is equal to S(x), the
set of edges incident with x, for some vertex x of G.
(ii) Prove: If G and H are simple 3-connected graphs with equiv-
alent codes, then G is isomorphic to H.
Theorem 34.8. Let G be a simple 3-connected graph embedded
on the sphere. A circuit C of G is the set of edges incident with
some face in this embedding if and only if G
//
C
is nonseparable.
Proof: The circuit C is the set of edges incident with a face if
and only if it is the set of edges incident with a vertex in the dual
H := G

. By Problem 34G(i), this is the case if and only if H

/
C
is
nonseparable. This latter graph is a Whitney dual of G
//
C
.
Thus the polygons that are boundaries of faces in an embedding
of a simple 3-connected planar graph are uniquely determined. This
is not true for nonseparable graphs in generalsee Fig. 34.5. If we
know the faces, i.e. the circuits that bound faces, the embedding is
determined (this is explained somewhat in the next chapter), and
we have the following result.
Corollary. A simple 3-connected planar graph has a unique em-
bedding on the sphere.
Example 34.4. If K
5
were planar, no 4-gon would be a face, since
contracting the four edges of a 4-gon produces a graph with two
vertices, four links, and two loops at one of the vertices, and this
graph is not nonseparable. On the other hand, a triangle would
necessarily be a face.
486 A Course in Combinatorics
We can obtain yet another proof that the simple 3-connected
graphs K
5
, K
3,3
, and the Petersen graph are not planar: contracting
the edges of any 3-gon in K
5
, any 4-gon in K
3,3
, and any 5-gon in
the Petersen graph, always yields a nonseparable graph. So if these
graphs were planar, every 3-gon, 4-gon, and 5-gon, respectively,
would bound a face in the embedding in view of Theorem 34.8.
But every edge belongs to three 3-gons, to four 4-gons, and to four
5-gons in the respective graphs, a contradiction to planarity.
E. Steinitz (1922) characterized the graphs of 3-dimensional con-
vex polytopes.
Theorem 34.9. A graph G is the graph of a 3-dimensional convex
polytope if and only if G is simple, planar, and 3-connected.
That the graph of a 3-dimensional polytope is 3-connected is a
special case of Theorem 32.5. We do not prove the rest of the
theorem here.
We need a lemma for the if part of Theorem 34.2.
Lemma 34.10. Let G and H be graphs with the same edge set
E. Suppose G is the union of edge-disjoint subgraphs G
1
and G
2
so
that [V (G
1
) V (G
2
)[ 1. Let H
1
and H
2
be the subgraphs of H
that have edge sets E(G
1
) and E(G
2
), respectively, with isolated
vertices deleted.
(i) If G and H are Whitney duals, then G
i
and H
i
are Whitney
duals for both i = 1 and i = 2.
(ii) If G
i
and H
i
are Whitney duals for both i = 1 and i = 2 and
[V (H
1
) V (H
2
)[ 1, then G and H are Whitney duals.
Proof: From Theorem 34.1, it is easy to see that every bond of
G is contained fully in either E(G
1
) or E(G
2
). Moreover, a subset
S E(G
i
) is a bond of G
i
if and only if it is a bond of G.
Suppose G and H are Whitney duals. If S is a bond of G
i
, then
S is a circuit of H contained in E(H
i
) and so is a circuit of H
i
.
And if S is a circuit of H
i
, then it is a circuit of H, hence a bond
of G contained in E(G
i
) and so is a bond of G
i
.
Suppose that G
i
and H
i
are Whitney duals for i = 1, 2 and that
[V (H
1
) V (H
2
)[ 1. Let S be a bond of G. Then S is a bond of
34. Whitney duality 487
G
1
or G
2
and hence a circuit in either H
1
or H
2
. But then certainly
S is a circuit in H. If S is a circuit of H, then it is a circuit in
either H
1
or H
2
. Hence it is a bond of G
1
or G
2
and so is a bond
of G.
Proof of Theorem 34.2, continued: For the proof of the con-
verse, we rst establish the following assertion. Given a pair G, H
of Whitney duals with the same edge set E and no isolated ver-
tices, we claim that there exists a possibly dierent Whitney dual
H
0
which admits what we will call a system of walks for G. This is
a family w
x
: x V (G) of walks w
x
in H
0
, one for each vertex x
of G, such that w
x
traverses in H
0
exactly the edges incident with
x in G, loops being traversed twice and nonloops once. We will
use induction on the number of edges. We may assume G has no
isolated vertices.
First note that the assertion is easy if G is nonseparable. Of
course, it is true for a loop-graph. Otherwise there are no loops
and for every vertex x, the set S(x) of edges incident with x in G
is a bond in G and hence a circuit in H. We need only let w
x
be a
simple closed path in H traversing the edges of the circuit S(x) in
H.
If G is not nonseparable, there exist edge-disjoint subgraphs G
1
and G
2
, each with at least one edge, and that have no vertices or
one vertex in common, such that their union is G. With H
1
and
H
2
as in Lemma 34.10, H
i
is a Whitney dual of G
i
, i = 1, 2. By
the induction hypothesis, there exist graphs H
0
i
that are Whitney
duals of G
i
and that admit systems of walks w
(i)
x
: x V (G
i
)
for G
i
, i = 1, 2. We may take H
0
1
and H
0
2
to be vertex disjoint. If
G
1
and G
2
are disjoint, let H
0
be the disjoint union of H
0
1
and H
0
2
;
then the union of the two systems of walks provides a system of
walks for G. If G
1
and G
2
have one vertex z in common, there are
walks w
(1)
z
and w
(2)
z
in H
0
1
and H
0
2
that traverse the edges incident
with z in G
1
and G
2
, respectively. Then let H
0
be obtained as the
union of H
0
1
and H
0
2
but where we identify some vertex in the walk
w
(1)
z
in H
0
1
with some vertex in the walk w
(2)
z
in H
0
2
. We combine
these two walks into one which traverses the edges incident with z
in G; this together with the remaining walks in the systems in H
1
and H
2
provides a system of walks for G. In either case, H
0
is a
488 A Course in Combinatorics
Whitney dual of G by Lemma 34.10.
Finally, we prove the following. Let G be a graph with a Whitney
dual H which admits a system of walks w
x
: x V (G) for G.
We assert that G has an embedding in the sphere so that the edges
incident with x in G occur in the same cyclic order as they occur
in the walk w
x
. We proceed by induction on the number of edges.
If G has only loops, then H is a forest. Let e be a loop in G
such that e is incident with a monovalent vertex in H. Let z be the
vertex incident with e in G. The walk w
z
in H evidently traverses
e twice in succession, say
w
z
= (s
0
, e, s
1
, e, s
2
= s
0
, a
3
, s
3
, a
4
, s
4
, . . . , a
k
, s
0
),
where the s
i
s are vertices of H. The graphs G
/
e
and H
//
e
are Whitney
duals and suppressing e in w
z
will provide a system of walks in H
//
e
for G
/
e
, so that by the induction hypothesis, G
/
e
admits a planar
drawing such that the edges incident with z in G
/
e
occur in the same
cyclic order as they occur in the walk w
z
with e suppressed. In a
small neighborhood of the point representing z in the embedding,
we can insert the loop e to obtain the required embedding of G (see
Fig. 34.9).
Figure 34.9
Now let e be a nonloop in G with ends y and z. Then G
//
e
and
H
/
e
are Whitney duals and we claim that there is a natural system
of walks u
x
: x V (G
//
e
) in H
/
e
for G
//
e
. For x ,= y, z, we take
u
x
:= w
x
. For the new vertex x
0
, obtained by identifying y and z,
we combine w
y
and w
z
omitting e as follows. Say
w
y
= (s
0
, e, s
1
, a
2
, s
2
, . . . , s
k1
, a
k
, s
0
)
34. Whitney duality 489
and
w
z
= (t
0
, e, t
1
, b
2
, t
2
, . . . , t
m1
, b
m
, t
0
),
where e, a
2
, a
3
, . . . , a
k
and e, b
2
, b
3
, . . . , b
m
are the edges incident
with y and z, respectively, in G (loops appearing twice) and the
s
i
s and t
j
s are vertices of H. By reversing one of these walks if
necessary, we can assume s
0
= t
0
and s
1
= t
1
. Then take
u
x
0
:= (s
1
, a
2
, s
2
, . . . , s
k1
, a
k
, s
0
, b
m
, t
m1
, b
m1
, . . . , b
3
, t
2
, b
2
, s
1
).
By the induction hypothesis, G
//
e
admits an embedding in the plane
so that the edges incident with x in G
//
e
occur in same cyclic order
as they occur in the walk u
x
. In a small neighborhood of the point
representing x
0
in the embedding, we can uncontract e to obtain
the required embedding of G (see Fig. 34.10).
Figure 34.10
Problem 34H. In a k-Tutte-connected graph (dened in Prob-
lem 32D) with at least 2k 2 edges, every circuit and every bond
contains at least k edges.
Problem 34I. A matroid consists of a set X (which we assume
here to be nite) and a family ( of nonempty subsets called circuits
such that no circuit is a proper subset of another circuit and, if
A, B (, x A B, y A B, then there exists C ( so that
C A B, y C, x / C. This latter property is called the
elimination axiom.
(i) Prove that the edge set of a nite graph together with the
family of all circuits (edge sets of polygons) is a matroid.
(ii) Prove that the family of minimal nonempty supports of code
words in a linear code over any eld provides the circuits of a ma-
troid.
490 A Course in Combinatorics
(iii) Let (X, T) be a combinatorial geometry as dened in Chap-
ter 23, and let ( be the family of minimal dependent subsets of X.
Show that (X, () is a matroid.
(To do all parts of this problem, it can save work to do part (iii)
rst and to derive the others from it.)
Notes.
Hassler Whitney (19071989), a pioneer in topology who was
also intensely concerned with mathematical education for the last
twenty years of his life, was a professor at Harvard and later at
the Institute for Advanced Study in Princeton. He was awarded
the Wolf Prize in 1982. Whitney had a degree in music from Yale,
played the violin, viola and piano, and was concertmaster of the
Princeton Community Orchestra.
Matroids were introduced by Hassler Whitney in 1935 in order to
abstract certain properties of linear independence, of circuits and
bonds in graphs, and of duality. Their theory is cryptomorphic to
that of geometric lattices. See Welsh (1976) and Crapo and Rota
(1971).
References.
H. Crapo and G.-C. Rota (1971), On the Foundations of Combina-
torial Theory: Combinatorial Geometries, M. I. T. Press.
S. MacLane (1937), A structural characterization of planar combi-
natorial graphs, Duke math. J. 3, 460472.
P. G. Tait (1880), Remarks on the colouring of maps, Proc. Roy.
Soc. Edinburgh 10, 501503.
W. T. Tutte (1956), A theorem on planar graphs, Trans. Amer.
Math. Soc. 82, 99116.
W. T. Tutte (1984), Graph Theory, Encylodpedia of Math. and its
Appl. 21, Addison-Wesley.
D. J. A. Welsh (1976), Matroid Theory, Academic Press.
H. Whitney (1932), Nonseparable and planar graphs, Trans. Amer.
Math. Soc. 34, 339362.
H. Whitney (1935), On the abstract properties of linear depen-
dence, Amer. J. Math. 57, 509533.
35
Embeddings of graphs
on surfaces
We consider embeddings of graphs on surfaces other than the sphere.
The classication of surfaces (see Frechet and Fan, 1967) shows that
there are two innite families of surfaces, up to homeomorphism.
We use T
g
, g 0, to denote the orientable surface of genus g (the
g-torus). This can be realized as a sphere with g handles or g
holes; T
0
is the sphere. We use ^
n
, n 1, to denote the nonori-
entable surface which can be constructed by inserting n cross-caps
on the sphere (these cannot be embedded in 1
3
). We state the ex-
tension of Eulers formula. A proof may be found in Frechet and
Fan (1967), or in Chartrand and Lesniak (1986) for the orientable
case.
Theorem 35.1. For a 2-cell embedding of a graph G on T
g
,
[F(G)[ [E(G)[ +[V (G)[ = 2 2g.
For a 2-cell embedding of a connected graph G on ^
n
,
[F(G)[ [E(G)[ +[V (G)[ = 2 n.
For any embedding, not necessarily a 2-cell embedding, the inequal-
ities which result by replacing = by in the equations above will
hold.
Example 35.1. Fig. 35.1 shows an embedding of the Petersen
graph P on the real projective plane ^
1
. In this diagram, diamet-
rically opposite points on the boundary are to be identied. There
are six faces, each a pentagon. More precisely, the dual graph with
respect to this embedding is the complete graph K
6
.
492 A Course in Combinatorics
Figure 35.1
(This establishes a curious one-to-one correspondence between
the 15 edges of P and the 15 edges of K
6
. We have previously seen
that there is a one-to-one correspondence between the 10 vertices
of P and the 10 edges of K
5
.)
(The real projective plane can be thought of as the surface of the
sphere where diametrically opposite points have been identied.
The Petersen graph can be thought of as the graph of a regular
dodecahedron where diametrically opposite points have been iden-
tied.)
Example 35.2. Fig. 35.2 shows an embedding of the complete
graph K
7
on the torus T
1
. We use the usual representation of the
torus; here the horizontal and vertical boundary lines are to be
identied. There are 14 triangular faces. The dual graph with
respect to this embedding has 14 vertices of degree 3 and seven
hexagonal faces, any two of which are adjacent. This is the Heawood
graph (dened in Chapter 31), which is given as an example of a
map on the torus whose faces require seven colors.
Figure 35.2
35. Embeddings of graphs 493
Theorem 35.2. If a loopless graph G admits an embedding in T
g
,
g > 0, then
(G)
7 +

1 + 48g
2
.
Proof: Suppose G can be emdedded in T
g
. We can assume that
G is simple, connected and has at least three vertices. We claim
that there exists a vertex x of degree at most N 1 where N :=
(7 +

1 + 48g)/2. Once this is established, the theorem will follow

by induction on the number of vertices; for if this vertex x is deleted
and the remaining graph properly N|-colored, there remains a
color free for x.
By Theorem 35.1, we have
2 2g f e +v
where f := [F(G)[, e := [E(G)[, v := [V (G)[. Since every face is
incident with at least three edges, 3f 2e. Let d be the average
degree of the vertices of G. We assume for contradiction that d >
N 1. This implies that v > N. Also d > 6 since N 7 for g 1,
and we have
2 2g
2
3
e e +v =
1
3
e +v =
1
6
vd +v =
1
6
v(6 d),
12g 12 v(d 6) > N(N 7) = 12g 12.
This provides the promised contradiction.
Here is a small table of bounds from Theorem 35.2:
genus g 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
7 8 9 10 11 12 12 13 13 14 15 15 16 16 16
The assertion that Theorem 35.2 was the best possible result, i.e.
that
7+

1+48g
2
| colors were required for some graph which could be
embedded in T
g
, was known as the Heawood conjecture until it was
proved by Ringel and Youngs in 1968. See Ringel (1974). What
they did is to prove the following stronger result.
494 A Course in Combinatorics
Theorem 35.3 (RingelYoungs). Given g 0, let
n :=
7 +

1 + 48g
2
|.
Then the complete graph K
n
can be embedded in T
g
.
We will prove here only a small part of this result (see Theorem
35.4 below). For example, Theorem 35.3 shows that (G) 19 for
graphs G emdedded in T
20
, T
21
, and T
22
. Theorem 35.4 shows that
K
19
can be embedded in T
20
(and hence in T
21
and T
22
), verifying
the Heawood conjecture for g = 20, 21, 22.
We need to understand 2-cell embeddings combinatorially, in
terms of the graph itself. Here is what is important.
Given a 2-cell embedding of a graph G on a surface o, we may
traverse the boundary of any face F and we obtain a closed walk w
F
in G. The starting vertex is not important, nor is the direction. In
many cases, this may be a simple closed path in G, but, e.g., when G
is a tree embedded in the sphere, there is a single face and this walk
passes through every edge twice (see Fig. 2.4). In general, every
edge will either belong to exactly two walks in w
F
: F F(G) or
occur twice in one of these walks. A system / of closed walks in
G with this property that every edge of G occurs twice among the
walks in / will be called a mesh in G.
Example 35.3. Here are two meshes in K
5
. The walks are de-
scribed by their vertex sequences:
/
1
= (1, 2, 3, 4, 5, 1),
(1, 2, 4, 1), (2, 3, 5, 2), (3, 4, 1, 3), (4, 5, 2, 4), (5, 1, 3, 5),
/
2
= (1, 2, 4, 5, 1), (1, 3, 2, 5, 3, 4, 2, 3, 5, 4, 1), (1, 4, 3, 1, 5, 2, 1).
The mesh /
1
consists of simple paths (one pentagon and ve
triangles), but the second walk in /
2
traverses two edges twice.
A triangular mesh is a mesh consisting of simple closed paths of
length 3. So a triangular mesh for K
n
is equivalent to a 2-design
S
2
(2, 3, n) with blocks of size 3 and index 2. But these designs
rarely correspond to embeddings of K
n
on surfaces because there
is an important additional condition that must hold.
35. Embeddings of graphs 495
Given a mesh /in a connected graph G, we may try to construct
a 2-cell embedding of G in a surface so that these walks bound the
faces. There is really no choicewe must proceed as follows. For
each walk w M of length , we need a corresponding closed disk
F
w
, for example a convex -gon and its interior in the plane (even
for = 1 and 2, although we cannot use straight line segments as
the edges in these cases). Label the vertices and edges of F
w
with
the vertices of the walk w in the order they occur in w. (So some
vertices and perhaps edges of F
w
will receive the same label if w
is not a simple closed path.) We will paste or sew these disks
together at their boundaries.
Consider the disjoint union of all disks F
w
. For each vertex x of
G, identify all the points labeled with an x. For each edge e of G,
identify the interior points of the edges of the disks labeled with
e in a one-to-one continuous manner in the direction indicated by
their ends. (It is necessary to agree that loops have two directions.)
Figure 35.3
Figure 35.4
Figs. 35.3 and 35.4 show, respectively, polygonal regions in the
plane (disks) labeled by the vertices of the walks in the meshes of
496 A Course in Combinatorics
Example 35.3. In Fig. 35.5, we have started the pasting process for
the disks in Fig. 35.3; it remains to identify diametrically opposite
points on the boundary (and we obtain the real projective plane
with K
5
drawn on it).
Figure 35.5
In general, the topological space constructed by this pasting has
neighborhoods homeomorphic to the open disk (2-cell) about points
x in the interior of the disks F
x
and also for points in the interior
of an edge on the boundary of the disks (because every edge is
in two of the walks of the mesh). But there may be trouble at
vertices, where many of the disks F
w
meet. For example, suppose
that the walks that traverse a vertex x in a certain mesh have vertex
sequences
(. . . , 1, x, 2, . . . ), (. . . , 2, x, 3, . . . ), (. . . , 3, x, 1, . . . ),
(. . . , 4, x, 5, . . . ), (. . . , 5, x, 6, . . . ), and (. . . , 6, x, 4, . . . ).
The disks corresponding to the rst three walks t together to
make a neighborhood of x which is a 2-cell, but with the other
three, a neighborhood of x will look like two cones touching at
their apexes. The graph /
x
dened in the next paragraph would
consist of two disjoint triangles in this case.
For a vertex x V (G), consider the graph /
x
where V (/
x
) is
the set of edges of G incident with x and where two members a, b of
V (/
x
) are adjacent in /
x
when they occur as consecutive edges
of some walk in / (with the vertex x between them). Of course,
join a and b by two edges in /
x
if they occur consecutively twice,
which probably will not often be the case. Clearly, /
x
is regular
35. Embeddings of graphs 497
of degree 2. The vertex condition requires that /
x
is connected
(i.e. a single polygon) for every vertex x.
The topological space constructed by this pasting from a mesh
in a connected graph is a surface if and only if the mesh satises
the vertex condition. There is a natural embedding of the graph
G in this surface. The meshes in Example 35.3 satisfy the vertex
condition. See Example 35.4 and Fig. 35.6.
Figure 35.6
To discover what surface o has been constructed from a mesh
/, we rst compute the Euler characteristic h := [/[ [E(G)[ +
[V (G)[. In view of Theorem 35.1, if h is odd, we must have
o = ^
2h
. But if h is even, we must decide whether the sur-
face is orientable or not (unless h = 2, when the sphere is the only
possibility): o = T
1h/2
if o is orientable, o = ^
2h
otherwise. We
state without proof that o is orientable if and only if the walks can
be directed (one of the two possible directions being chosen for each
walk) so that every edge is traversed exactly once in each direction,
in which case we say the mesh is orientable.
It is easy to check whether a mesh satisfying the vertex condition
is orientable. We give one walk a direction and that determines the
directions of all walks sharing an edge, which in turn determines
the directions of further walks. The vertex condition implies that
the direction of every walk in the mesh is determined (why?). Then
we check each edge to see in what directions it is traversed.
Example 35.3 (continued). The mesh /
1
provides an embed-
ding of K
5
on the real projective plane ^
1
. The mesh /
2
is
orientablethe walks may be directed as they have been written
downand we have a 2-cell embedding of K
5
on the double torus
T
2
.
498 A Course in Combinatorics
In describing walks in simple graphs in the examples that follow,
we list the vertex sequences in square brackets but suppress the
terminal vertex since it is the same as the initial one. And it is
convenient to refer to paths of length 3, 4, . . . as triangles, quadri-
laterals, etc., ignoring the dierence between walks in a graph and
subgraphs that are polygons.
Example 35.4. We show that K
8
can be embedded on the double
torus T
2
. This shows that at least eight colors are required for T
2
and proves the Heawood conjecture for g = 2.
There must be 18 faces (so that f e+v = 2), and their degrees
must add to 56. We display a solution with two quadrilaterals and
16 triangles as the faces. We use dierence methods (cf. Chapter
19).
As the vertices of K
8
, we take Z
8
, the integers modulo 8. The
quadrilaterals will have vertex sequences [0, 2, 4, 6] and [1, 3, 5, 7].
The triangles are to be the translates modulo 8 of the two initial
triangles [0, 1, 4] and [0, 3, 2]. The reader should check that every
edge appears in exactly two of these walks. Let us check the vertex
condition. Since translation modulo 8 preserves the polygons, we
need only check the vertex 0. The walks traversing 0 are [0, 2, 4, 6],
[0, 1, 4], [7, 0, 3], [4, 5, 0], [0, 3, 2], [5, 0, 7], and [6, 1, 0]. They t to-
gether to form a single polygon as indicated in Fig. 35.6.
It remains to check whether the mesh is orientable. It turns out
that if the quadrilaterals are directed as [0, 2, 4, 6] and [7, 5, 3, 1],
and every other triangle is also reversed (i.e. whenever the initial
triangles are translated by an odd element of Z
8
), we obtain an
oriented mesh.
Problem 35A. Find a mesh of ve 4-gons in K
5
, a mesh of six
5-gons in K
6
, and a mesh of seven 6-gons in K
7
, all of which satisfy
the vertex condition. What surfaces have you constructed?
Problem 35B. Find a mesh of 16 pentagons in the Clebsch graph
(see Example 21.4). Check the vertex condition and also check for
orientability. What surface, if any, has been constructed? What
is the dual graph with respect to this embedding? (Cf. Problem
35G.)
Problem 35C. (i) Show that if a simple graph G is embedded in
35. Embeddings of graphs 499
the nonorientable surface ^
g
, g 1, then
(G)
7 +

1 + 24g
2
.
(ii) Show that if a simple graph is embedded in the Klein bottle
^
2
, then either it has a vertex of degree 5 or it is regular of
degree 6. Show that there is a unique mesh of 14 triangles in K
7
that satises the vertex condition and that it is orientable, and so
K
7
does not embed in the Klein bottle. Conclude that (G) 6
for graphs G embedded in ^
2
.
(Thus K
7
does not embed in the Klein bottle after all. We con-
clude that (G) 6 for graphs embedded in ^
2
.)
Problem 35D. Before we continue with surfaces other than the
sphere, we remark that our understanding of meshes allows a quick
proof of the fact that if G and H are nonseparable Whitney duals,
then there exists an embedding of G in the sphere with respect to
which the dual G

is isomorphic to H. As in the proof of Theorem

34.2, the set of edges incident with a vertex of H is a circuit in
G. The system of walks we get in G is a mesh /. Show that /
satises the vertex condition. Explain why the surface constructed
is the sphere and why the dual G

is isomorphic to H.
Problem 35E. Consider a connected graph G and a mesh /
in G satisfying the vertex condition (i.e. a map). To keep things
simple, assume that G is simple and that the walks in /are simple
closed pathswhich we identify with polygons and call faces. An
automorphism of this map is a permutation of V (G) which takes
edges to edges and faces to faces (i.e. an automorphism of G that
preserves /). Show that if an automorphism xes a vertex x,
an edge e incident with x, and a face F incident with e, then it is
the identity.
(It follows, for example, that a simple map as above with graph
G has at most 4[E(G)[ automorphisms. Examples where equality
occurs include the Platonic solids and the maps in Examples 35.1
and 35.2.)
Theorem 35.8. If n 7 (mod 12), then the complete graph K
n
500 A Course in Combinatorics
admits a triangular embedding on an orientable surface of genus
g := (n 3)(n 4)/12.
Proof: We construct an orientable triangular mesh in K
n
satisfy-
ing the vertex condition. Write n = 12s +7. As the vertices of K
n
,
we use Z
n
.
The construction is encoded in the diagrams of Fig. 35.7. The
diagrams are for the cases n = 7, n = 19, n = 31, and n = 12s + 7
in general. (In the general case, the vertical edges are directed
alternately and are labeled 1, 2, . . . , 2s, consecutively.) We could
describe this construction simply as an application of dierence
methods and not give any gures, but that would be a shame since
the diagrams were conceptually very helpful in nding this and
other embeddings.
Note that the edges are labeled with the integers from 1 to 6s+3.
The conservation law, or Kirchhos current law, holds: at every
vertex, the sum of values on the incoming edges is equal to the sum
of the values on the outgoing edges. (Except that we have no source
or sink, this is a ow of strength zero as dened in Chapter 7.)
Actually, it will be useful to think that the reverses of the indi-
cated directed edges are also present; the reverse of the edge shown
in Fig. 35.7 will carry a label equal to the negative of the shown
edge. Then all nonzero values modulo n = 12s + 7 appear on the
directed edges, each exactly once. The vertices are of two types:
solid (representing clockwise) and hollow (representing counter-
clockwise).
Every vertex in the diagrams of Fig. 35.7 will provide a family of
n directed triangles in K
n
which are translates of one another mod-
ulo n. For each solid vertex that is the tail of directed edges with
values a, b, c in clockwise order, we take the n directed triangles
[x, x +a, x +a +b], x Z
n
.
Remember that a + b + c = 0 in Z
n
, so we get the same triangles
(but with dierent initial vertices) if we take [y, y + b, y + b + c],
y Z
n
or [z, z + c, z + c + a], z Z
n
. For each hollow vertex that
35. Embeddings of graphs 501
is the tail of directed edges with values a, b, c in counter-clockwise
order, we take the n directed triangles
[x, x +a, x +a +b], x Z
n
.
For example, in the third diagram in Fig. 35.7 (when n = 31),
there is a hollow vertex with incoming edges labeled 7 and 2 and an
outgoing edge labeled 9. We regard this as outgoing edges labeled
7, 2, and 9 (counter-clockwise), and take the 31 directed trian-
gles obtained as the translates of [0, 7, 9] modulo 31 (or trans-
lates of [0, 9, 2] or [0, 2, 7]). Another vertex in that diagram yields
the 31 directed triangles obtained as the translates of [0, 14, 13]
modulo 31.
It is not hard to see that the collection of directed triangles ob-
tained from the diagrams form an oriented triangular mesh. For
example, when n = 31, to nd a triangle traversing the edge joining
4 to 23 in that direction, notice that 23 4 = 12 in Z
n
. The edge
labeled 12 leaves a solid vertex with other outgoing edges labeled
15 and 3, and the triangle [4, 4+(12), 4+(12)+15] = [4, 23, 7]
is found. The triangle traversing the edge joining 23 to 4 in that
order is found to be [23, 4, 6].
Figure 35.7
502 A Course in Combinatorics
It will suce to check the vertex condition only at 0. Let e
i
denote the edge 0, i of K
n
. Consider the triangle of the mesh
that traverses 0 and enters 0 on edge e
a
. In the diagram, the edge
labeled a will enter a vertex that is the tail of edges with labels
a, p, q, say, in clockwise or counter-clockwise order, depending
on whether the vertex is solid or hollow. That is, the triangle is
[a, 0, p] and leaves 0 on the edge e
p
. To summarize, if a triangle
enters 0 on an edge e
i
of K
n
, the value j such that the triangle
leaves 0 on e
j
is found from the diagram as follows: nd the edge
labeled i in the diagramturn clockwise or counter-clockwise, as
appropriate, at the head of that edge and leave on the next edge
the label on that edge of the diagram will be j.
For example, when n = 31, we nd the sequence of edges e
i

of
K
n
so that the triangle which enters 0 on e
i

will leave on e
i
+1
to
be
e
1
, e
13
, e
8
, e
9
, e
7
, e
10
, e
6
, e
5
,
e
13
, e
14
, e
12
, e
15
, e
11
, e
6
, e
4
, e
15
, e
3
, e
7
, e
2
, e
14
,
e
1
, e
8
, e
5
, e
11
, e
4
, e
10
, e
3
, e
12
, e
2
, e
9
, e
1
, . . . .
Consecutive pairs above will be adjacent in /
0
, which is seen to
be a single polygon. The reader should check that all diagrams
in Fig. 35.7 will similarly yield under this procedure a sequence
traversing each edge of the diagram once in each direction. This is
quite amazing.
Problem 35F. (i) If the complete graph K
n
admits an embed-
ding on the orientable surface T
g
so that all faces are bounded by
triangles, then g = (n 3)(n 4)/12.
(ii) If the complete bipartite graph K
n,n
admits an embedding on
the orientable surface T
g
so that all faces are bounded by quadri-
laterals, what is g in terms of n? What is the least value of n > 2
for which such an embedding exists?
In order to present the proof of Theorem 35.8 as quickly and
as simply as possible, we have taken the construction out of its
original context. We close this chapter with a brief discussion of
some of the ideas that allowed Ringel and Youngs to come up with
constructions for all values of n.
35. Embeddings of graphs 503
A technique for embedding graphs on orientable surfaces was de-
scribed by J. Edmonds (1960). It may be thought of as arising
from the mesh in the dual graph with respect to the embedding.
If we have a connected simple graph G embedded in an orientable
surface o, then the choice of an orientation on o will induce at
every vertex a cyclic permutation of the edges incident with that
vertex. Edmonds pointed out that any such local cyclic permuta-
tions determine an embedding. Let G be a simple graph. Suppose
a cyclic permutation
x
of the set S(x) of edges incident with x is
given for each vertex x. Then there is a natural oriented mesh /
determined as follows. In the directed walk traversing an edge e
directed from one of its ends y to the other x, the edge following e
is to be
x
(e).
The examples below concern simple graphs, so the local permuta-
tions may be abbreviated as permutations of the adjacent vertices
and walks may be described by their vertex sequences.
Example 35.5. Consider K
5
with vertex set 1, 2, 3, 4, 5 and the
following local permutations:
1 : (2435)
2 : (1435)
3 : (4125)
4 : (1325)
5 : (1234).
It might be good to think of the resulting walks as a decomposition
of the edges of the complete directed graph on 5 vertices. Let e
ij
denote the edge directed from i to j. Here are the edge terms of
the walks:
(e
12
, e
24
, e
45
, e
51
),
(e
13
, e
32
, e
25
, e
53
, e
34
, e
42
, e
23
, e
35
, e
54
, e
41
),
(e
14
, e
43
, e
31
, e
15
, e
52
, e
21
).
The vertex sequences of these walks are exactly those of /
2
in
Example 35.3.
In the family of graphs with 4s + 2 vertices in Fig. 35.7, we
specied a cyclic order at each vertex by the use of a solid vertex
to indicate that the local permutation is to be clockwise and a
504 A Course in Combinatorics
hollow vertex to indicate counter-clockwise. As part of the proof of
Theorem 35.8, we had to check that the resulting embeddings have
exactly one face.
Given a nite group and a subset S of nonidentity elements of
such that S implies that
1
S, the Cayley graph G(, S)
is the simple graph with vertex set and where vertices and
are adjacent if and only if
1
S. A complete graph K
n
is a Cayley graph with respect to any group of order n, where S
consists of all nonidentity elements. Often it is required that S
generates the groupthis ensures that the corresponding Cayley
graph is connected.
W. Gustin (1963) introduced the theory of quotient manifolds
and developed methods to embed Cayley graphs into orientable
surfaces. We have no space to describe this precisely, but give
some examples. Consider the case where the local permutations
at the vertices of a Cayley graph are all shifts of an initial cyclic
permutation (s
1
s
2
. . . s
k
) of the set S of vertices adjacent to 0;
that is, the local permutation at is
(s
1
+ s
2
+ . . . s
k
+)
(with additive notation).
Example 35.6. Consider K
7
with vertex set 0, 1, 2, 3, 4, 5, 6 and
with local permutation (1 +i 3 +i 2 +i 6 +i 4 +i 5 +i) (mod 7)
at i as below:
0 : (132645)
1 : (243056)
2 : (354160)
3 : (465201)
4 : (506312)
5 : (610423)
6 : (021534).
The walks are:
[1, 2, 4] [2, 3, 5] [3, 4, 6] [4, 5, 0] [5, 6, 1] [6, 0, 2] [0, 1, 3]
[3, 5, 6] [4, 6, 0] [5, 0, 1] [6, 1, 2] [0, 2, 3] [1, 3, 4] [2, 4, 5].
This is the triangular embedding of K
7
on the torus we have seen
in Example 35.2.
35. Embeddings of graphs 505
If we try this method on K
31
, for example, with vertex set Z
31
and initial local permutation
(1, 13, 8, 9, 7, 10, 6, 5, 13, 14, 12, 15, 11, 6, 4,
15, 3, 7, 2, 14, 1, 8, 5, 11, 4, 10, 3, 12, 2, 9),
we obtain the same triangular mesh as in the proof of Theorem
35.8. The diagrams of Fig. 35.7 actually represent the quotient
manifolds of the dual graphs of K
n
with respect to the prescribed
embeddings. For n = 12s + 7, the embedded graph represented
by the diagram has 4s + 2 vertices, 6s + 3 edges, and 1 face; the
resulting embedding of K
n
has (4s+2)n faces, (6s+3)n edges, and
(1)n vertices.
Problem 35G. Let be a nonzero element in F
q
where q is a
power of a prime. Say has order m in the multiplicative group
and assume that 1 is a power of . Consider the Cayley graph
G(F
q
, )) where ) = 1, ,
2
, . . . . We are using the additive
group of F
q
for the Cayley graph, so a and b are adjacent if and
only if ab ). (This is a complete graph when is a primitive
element in F
q
; it is the Clebsch graph when q = 16 and m = 5.)
Let us take the local permutation at 0 to be (1, ,
2
, . . . ,
m1
)
and obtain the other local permutations as shifts. What is the size
and number of the faces in the resulting mesh?
Notes.
P. J. Heawood (18611955) spent most of his career at Durham,
England as professor of mathematics, nally to become vice-
chancellor.
References.
G. Chartrand and L. Lesniak (1986), Graphs and Digraphs, 2nd
edn., Wadsworth.
J. Edmonds (1960), A combinatorial representation for polyhedral
surfaces, Notices Amer. Math. Soc. 7, 646.
M. Frechet and K. Fan (1967), Initiation to Combinatorial Topol-
ogy, Prindle, Weber and Schmidt.
506 A Course in Combinatorics
W. Gustin (1963), Orientable embedding of Cayley graphs, Bull.
Amer. Math. Soc. 69, 272275.
G. Ringel (1974), Map Color Theorem, Springer-Verlag.
G. Ringel and J. W. T. Youngs (1968), Solution of the Heawood
map coloring problem, Proc. Nat. Acad. Sci. U.S.A. 60, 438445.
A. T. White (1973), Graphs, Groups, and Surfaces, Mathematical
Studies 8, North-Holland.
36
Electrical networks and
squared squares
We start with the matrix-tree theorem, which expresses the number
of spanning trees in a graph as the determinant of an appropriate
matrix.
Theorem 36.1. The number of spanning trees in a connected
graph G on n vertices and without loops is the determinant of
any n 1 n 1 principal submatrix of the matrix DA, where
A is the adjacency matrix of G and D is the diagonal matrix whose
diagonal contains the degrees of the corresponding vertices of G.
For multigraphs G without loops we agree that the adjacency
matrix A is dened to have A(x, y) equal to the number of edges
joining x and y, for distinct vertices x and y. We postpone the proof
until we have developed some tools, but we give some examples now.
Figure 36.1
Example 36.1. Let G be the graph in Fig. 36.1. Then
D A =
_
_
_
2 1 0 1
1 3 1 1
0 1 2 1
1 1 1 3
_
_
_
508 A Course in Combinatorics
and G has 8 spanning trees.
Example 36.2. Take G to be the complete graph K
n
in Theorem
36.1. In this case, the matrix D A is nI J, where I is the
identity matrix of order n and J is the n n matrix of all 1s. The
calculation of the determinant of a n1 n1 principal submatrix
of this simple matrix can be done in several ways (row operations,
consideration of eigenvalues) and is left as an exercise. Once this is
done, we obtain yet another proof of Cayleys theorem, cf. Chapter
2, that the complete graph K
n
has n
n2
spanning trees.
Problem 36A. Let M be an n n matrix all of whose line-sums
are 0. Then one of the eigenvalues of M is
1
= 0; let
2
,
3
, . . . ,
n
denote the other eigenvalues. Show that all principal n1 n1
submatrices of M have the same determinant, and this value is
equal to the product
1
n

3

n
.
Example 36.3. We can use the observation of the above problem
to calculate the number of spanning trees in regular graphs whose
spectrum is known. For example, let A be the adjacency matrix of
the Petersen graph. Then A has eigenvalues
3, 1, 1, 1, 1, 1, 2, 2, 2, 2
(see Chapter 21). The matrix M is 3I A, which has eigenvalues
0, 2, 2, 2, 2, 2, 5, 5, 5, 5.
We conclude that the Petersen graph has 2000 spanning trees.
The following lemma is known as the CauchyBinet theorem. It
is more commonly stated and applied with the diagonal matrix
below taken to be the identity matrix.
Lemma 36.2. Let A and B be, respectively, r m and m r
matrices. Let be the m m diagonal matrix with entry e
i
in
row i, column i. For an r-subset S of 1, 2, . . . , m, let A
S
and B
S
denote, respectively, the r r submatrices of A and B consisting
of the columns of A, or the rows of B, indexed by elements of S.
Then
det(AB) =

S
det(A
S
) det(B
S
)

iS
e
i
36. Electrical networks and squared squares 509
where the sum is extended over all r-subsets S of 1, 2, . . . , m.
Proof: We will prove this assuming that e
1
, . . . , e
m
are indepen-
dent indeterminates. Of course, it will then hold for all values of
e
1
, . . . , e
m
.
The entries of the r r matrix AB are linear forms in the
indeterminates e
1
, . . . , e
m
; explicitly, if A = (a
ij
) and B = (b
ij
),
then the (i, k) entry of AB is

m
j=1
a
ij
b
jk
e
j
. Thus det(AB) is a
homogeneous polynomial of degree r in e
1
, . . . , e
m
.
Consider a monomial e
t
1
1
e
t
2
2
. . . where the number of distinct in-
determinates e
i
that occur, i.e. that have exponent t
i
> 0, is less
than r. Substitute 0 for the indeterminates e
i
which do not occur
in e
t
1
1
e
t
2
2
. . . . The monomial e
t
1
1
e
t
2
2
. . . and its coecient are unaf-
fected by this substitution. But after this substitution, the rank of
is less than r so the polynomial det(AB) evaluates to the zero
polynomial.
So we see that the coecient of a monomial in the polynomial
det(AB) is 0 unless the monomial is a product of r distinct inde-
terminates e
i
, i.e. unless it is of the form

iS
e
i
for some r-subset
S. The coecient of a monomial

iS
e
i
in det(AB) can be ob-
tained by setting e
i
, i S, equal to 1 and e
j
, j / S, equal to 0.
When this substitution is made in , AB evaluates to A
S
B
S
. So
the coecient of

iS
e
i
in det(AB) is det(A
S
) det(B
S
).
We will deal extensively with directed graphs H in this chapter.
All graph-theoretic terms (trees, components, circuits, etc.) will
be applied to a directed graph with the same meaning as for the
underlying undirected graph we obtain by ignoring the directions
of the edges. Thus a path in a digraph will traverse some edges in
a forward direction, other edges will be traversed backward. It
is really undirected graphs we are interested in, but it is convenient
for the theory to choose an orientation and produce a digraph; any
orientation will do and produces essentially the same theory.
As we have often done in previous chapters, we use the functional
notation M(i, j) to denote the entry in row i and column j of a
matrix M, and f(i) to denote the ith coordinate of a vector f.
The incidence matrix N of a directed graph H is the matrix
whose rows are indexed by V (H), whose columns are indexed by
510 A Course in Combinatorics
E(H), and where
N(x, e) =
_
_
_
0 if x is not incident with e, or e is a loop,
1 if x is the head of e,
1 if x is the tail of e.
We mention that
(36.1) rank(N) = [V (H)[ [C(H)[
where C(H) is the set of components of H. To see this, suppose
that g is a row vector with coordinates indexed by V (H) and that
gN = 0. This means for every edge e, directed from x to y, say,
that g(y) g(x) = 0. It is clear then, that gN = 0 if and only if g
is constant on the vertex set of every component of H, and hence
the dimension of the space of all g such that gN = 0 is [C(H)[.
We also mention that the determinant of any square matrix that
has at most one 1 and at most one 1 in any column, all other
entries being 0, is equal to 0 or 1. This follows by induction:
if every column has a +1 and a 1, then the sum of all rows is
the zero vector, so the matrix is singular. Otherwise, expand the
determinant by a column with one nonzero entry to nd it is 1
times the determinant of a smaller matrix with the same property.
So every square submatrix of the incidence matrix of a digraph has
determinant 0 or 1. (Matrices with this property are called totally
unimodular.)
Proof of Theorem 36.1: Let H be a connected digraph with n
vertices, with incidence matrix N. Let S be a set of n 1 edges
and, with the notation of the CauchyBinet theorem, consider the
n n1 submatrix N
S
, whose columns are indexed by elements of
S, of the incidence matrix N. By (36.1), N
S
has rank n 1 if and
only if the spanning subgraph of H with edge set S is connected,
i.e. if and only if S is the edge set of a tree in H. Let N
/
be obtained
by dropping any single row from the incidence matrix N. Since the
sum of all rows of N (or N
S
) is the zero vector, the rank of N
/
S
is the same as the rank of N
S
. The observations of this and the
preceding paragraph prove that
(36.2)
det(N
/
S
) =
_
1 if S is the edge set of a spanning tree in H,
0 otherwise.
36. Electrical networks and squared squares 511
Let a connected loopless graph G on n vertices be given, let H
be any orientation of G, and let N be the incidence matrix of H.
Then NN

= D A because
NN

(x, y) =

eE(G)
N(x, e)N(y, e)
=
_
deg(x) if x = y,
t if x and y are joined by t edges in G.
An n 1 n 1 principal submatrix of D A is of the form
N
/
N
/

where N
/
is obtained from N by dropping any one row. By
CauchyBinet,
det(N
/
N
/

) =

S
det(N
/
S
) det(N
/
S

) =

S
( det(N
/
S
))
2
,
where the sum is extended over all (n 1)-subsets S of the edge
set. By (36.2), this is the number of spanning trees of G.
Remark. If we view E(G) as a set of indeterminates and apply
Lemma 36.2 to n 1 n 1 principal submatrices of NN

where here is the diagonal matrix whose rows and columns are
indexed by the edges and where the diagonal entries are the edges
themselves (i.e. (e, e) := e), we nd that det(NN

) is the sum
of monomials corresponding to the edge sets of the trees in G. For
example, for the graph G in Fig. 36.1,
NN

=
_
_
_
a +b b 0 a
b b +c +e c e
0 c c +d d
a e d a +d +e
_
_
_
.
Of course, NN

is a singular matrix (all line-sums are 0), but

the determinant of the 3 3 matrix obtained by dropping, e.g.,
the last row and column is
(a +b)(b +c +e)(c +d) b
2
(c +d) c
2
(a +b) =
abc +abd +acd +ace +ade +bcd +bce +bde,
512 A Course in Combinatorics
representing the eight spanning trees in G.
We now give a generalization of Theorem 36.1 due to W. T.
Tutte (1948), with a proof that is shorter than the one we gave
for the rst theorem! (The generalized statement allows a type
of induction that is not possible in the original.) The statement
of Theorem 36.1 is recovered when we apply Theorem 36.3 to the
directed graph obtained from an undirected graph by replacing each
undirected edge with a pair of directed edges, one in each direction.
Theorem 36.3. Let Gbe a directed graph with vertices x
1
, . . . , x
n
.
Dene the matrix M, or M(G), as follows: M = (m
ij
) where m
ii
is
the number of edges pointing away from x
i
, not counting loops, and
for i ,= j, m
ij
is the negative of the number of edges directed from
x
i
to x
j
. Then the number A

, or A

(G), of spanning arborescences

of G with x

as root is the minor of position (, ) in M, i.e. the

determinant of the n1 n1 principal submatrix of the matrix
M obtained by dropping row and column .
Proof: For notational convenience, we take = 1. If, for some
i > 1, there are no edges directed away from x
i
, then the statement
of the theorem holds: row i of M(G) is the zero-vector; both the
minor of position (1, 1) and the number of arborescences are zero.
If, for some i > 1, there is more than one edge directed away
from x
i
, let G
1
and G
2
be directed graphs equal to G except that
some ( 1) of those edges are put into G
1
and the others ( 1)
are put into G
2
. Since every spanning arborescence of G with root
x
1
must contain exactly one edge directed away from x
i
, A
1
(G) =
A
1
(G
1
) + A
1
(G
2
). Also note that M(G), M(G
1
), and M(G
2
) are
identical except for the ith rows, and the ith row of M(G) is the
sum of the ith rows of M(G
1
) and M(G
2
). So the minor of position
(1, 1) in the rst matrix is the sum of those minors in the latter
two matrices.
Hence it will be sucient to verify the statement of the theorem
for digraphs G such that m
ii
= 1 for all i > 1. That is, with the
property that there is a unique edge directed away from any vertex
x
i
, i > 1.
If such a digraph Ghas a directed closed path not passing through
x
1
, say x
2
x
3
, x
3
x
4
, . . . , x
k
x
2
, then the sum of rows 2
36. Electrical networks and squared squares 513
through k is 0 and hence the minor of position (1, 1) is 0. There
are no directed paths from any of these vertices to x
1
, so there are
no spanning arborescences.
If such a digraph G has no directed closed paths not passing
through x
1
, the directed path starting at any vertex x
j
, and fol-
lowing the edge directed away from each vertex that is met, must
reach x
1
. We see that the digraph G, when edges leaving x
1
are
deleted (these edges may not occur in any spanning arborescence),
is itself a spanning arborescence; so A
1
(G) = 1. The minor of posi-
tion (1, 1) in M(G) is also seen to be 1, e.g. by induction: expand
the determinant by a column j corresponding to a vertex x
j
that
has indegree 0 in the arborescence (and so a 1 in position (j, j) and
0s in positions (i, j), i > 1, i ,= j).
We remark that if the edges of a digraph G are viewed as in-
dependent indeterminates and a matrix like M(G) is formed but
where the entries are formal sums of edges rather than integers,
then the minors are the formal sums of all arborescences, each rep-
resented by the product of its edges. The reader may easily modify
the above proof. For example, if G is the digraph on vertices 1, 2, 3
with edge a directed from 2 to 1, edge b directed from 3 to 1, and
edge c directed from 3 to 2, then we consider the matrix
M =
_
_
0 0 0
a a 0
b c b +c
_
_
.
The minor of position (1, 1) is ab + ac, which represents the two
arborescences rooted at 1.
Problem 36B. (Easy.) Let T
n
be the transitive tournament on
vertices 1, 2, . . . , n with edges directed from j to i whenever i < j.
What is M(T
n
)? What is the number of spanning arborescences
rooted at 1 in T
n
? Describe all such arborescences.
We can now give a second proof of Theorem 8.2 on the number of
De Bruijn sequences. Let the digraph G
n
be as dened in Chapter 8.
All the indegrees and outdegrees of G
n
are equal to 2. The solution
of Problem 2B suggested in Appendix 1 shows that the number of
(directed) Eulerian cycles in G
n
is equal to the number of spanning
514 A Course in Combinatorics
arborescences in G
n
rooted at any vertex. By Theorem 36.3 and
Problem 36A, this number can be obtained from knowledge of the
eigenvalues of the matrix M(G
n
).
Number the rows and columns of a matrix B
n
of size 2
n1
by the
integers 0 to 2
n1
1. Let b
ij
= 1 if j = 2i or 2i + 1 (mod 2
n1
).
(Note that the top half and bottom half of B
n
are the same.) Then
M(G
n
) = 2I B
n
. We now calculate the determinant D
n
(x) of
xI M(G
n
). By the obvious row and column operations on the
lower half of this determinant we nd
D
n
(x) = (x 2)
2
n2
D
n1
(x).
Since D
2
(x) = x(x 2), we see that all the nonzero eigenvalues of
M
n
are equal to 2 and hence by Problem 36A, the number of De
Bruijn cycles of length 2
n
is
1
2
n1
2
2
n1
1
= 2
2
n1
n
.
* * *
Let N be the incidence matrix of a digraph H. The rowspace
of N is called the coboundary space of H; the nullspace of N

,
i.e. the space of all row vectors f such that fN

= 0, is called the
cycle space of H. These are of importance in the theory of electrical
networks and algebraic topology. From (36.1), the dimension of the
coboundary space of a digraph H is [V (H)[ [C(H)[, and hence
the dimension of the cycle space is [E(H)[ [V (H)[ +[C(H)[.
From the denition, a vector f is a cycle if and only if for each
vertex x, the sum of the values f(e) on outgoing edges is equal to
the sum of the values f(e) on incoming edges. An example of a
cycle f (a so-called elementary cycle) is obtained from any simple
closed path p in H by dening f(e) to be +1 on the forward edges
of the path, 1 on the backward edges, and 0 on all edges not in
the path. A vector is a coboundary if and only if it is orthogonal to
all cycles, so in particular, the sum of the values of a coboundary
on the forward edges of a simple closed path will equal the sum of
the values on the backward edges of the path.
36. Electrical networks and squared squares 515
Problem 36C. A vector g with coordinates indexed by E(H) is
a coboundary on a digraph H if and only if for every closed walk
w, the signed sum of the values of g on the edges (the sum of the
values of g on the forward edges of w minus the sum of the values
of g on the backward edges of w) is zero.
Let us agree that an electrical network consists of a digraph H,
together with a function r associating to each edge e a resistance
r(e) 0, and a function s associating to each edge e an impressed
electromotive force or voltage source s(e). That is, each edge is
to be thought of as a resistor or a battery or both. If batteries and
resistors are connected, a current ow f(e) will pass through each
edge and a potential dierence or voltage g(e) will be measurable
across each edge. All that we need to know here is that the vectors
f and g are determined from r and s by the rules called Kirchhos
laws and Ohms law. Kirchhos current law and voltage law assert,
respectively, that f is a cycle and that g is a coboundary. Ohms
law says that g(e) = s(e) +r(e)f(e), or, in matrix notation, g =
s +fR where R is the diagonal matrix with the numbers r(e) on
the diagonal and where we think of s, g, and f as row vectors.
Problem 36D. Suppose e : r(e) = 0 contains no circuits, i.e.
is the edge set of a forest. Show that the electrical network has a
unique solution, by which we mean that for any s there exist unique
vectors f and g satisfying the above laws.
18
14
15
10
4
7
8
9
1
Figure 36.2
516 A Course in Combinatorics
By a squared rectangle we mean a rectangle which has been par-
titioned into a nite number of squares. Two examples appear in
Figs. 36.2 and 36.3. These rectangles have dimensions 33 by 32,
and 177 by 176, respectively. Squared rectangles are thought to be
elegant when all internal squares have distinct sizes and when they
contain no smaller squared rectangle.
77
99
34
9
16
43
41
57
25
78
21
Figure 36.3
Problem 36E. Find integral dimensions for the small squares in
Fig. 36.4 so that the diagram represents a squared rectangle.
x
y
z
Figure 36.4
36. Electrical networks and squared squares 517
For many years, it was not known whether a square could be par-
titioned into a nite number of unequal smaller squares. In the late
1930s, such squared squares were nally discovered. One example
was shown on the cover of the Journal of Combinatorial Theory
until a smaller example was found by A. J. W. Duijvestijn (1978);
the cover illustration was updated in January of 1979. This is a
partition into the least possible number (21) of squares. Comput-
ers are used for these results, but an important tool in the study of
squared rectangles is the connection with electrical networks which
was discovered by four undergraduates at Cambridge in 19361938.
A popular account of some of their work is given by W. T. Tutte
(1961) and (1965) who was one of them.
We give an informal description of this connection with electrical
networks. Let us use the term special network for an electrical
network with a distinguished edge e
0
so that s(e) = 0 for e ,= e
0
,
and where
r(e) =
_
1 if e ,= e
0
,
0 if e = e
0
.
That is, we have a system of unit resistors and a single battery.
33
28
36
16
36
25
9
2
5
28
16
33
61
9
7
2
5
25
Figure 36.5
A special electrical network arises from a squared rectangle as
follows. Dene a digraph H whose vertices are the maximal hor-
izontal line segments in the diagram. The edges of H are to be
the small squares, and we include one special edge e
0
for the entire
518 A Course in Combinatorics
rectangle. We direct each edge (except e
0
) downward, i.e. towards
the lower line segment. An example is shown in Fig. 36.5. The
digraph H is clearly planar. There is a real number f(e) naturally
associated to edge e, namely the size of the corresponding small
square; we take f(e
0
) to be the width of the large rectangle. It is
not hard to see that f is a cycle on H: for each maximal horizontal
line segment, the sum of the sizes of the squares sitting on that
segment is equal to the sum of the squares hanging from it. More-
over, were it not for the special edge, f would also be a coboundary:
if for each vertex (horizontal line segment) x, we take h(x) to be
the distance of the line segment from the top side of the large rect-
angle, then for e ,= e
0
, f(e) is equal to h(head of e) h(tail of e).
So if we let s(e
0
) be the height of the rectangle and take g to be
the coboundary of h, then f and g are the current ow and voltage
vectors which solve this special network, i.e. g = s +Rf.
Conversely, a special electrical network where the digraph is pla-
nar leads to the construction of a squared rectangle. We do not
give a proof. The ratio of its width to its height will be the ratio of
the current ow f(e
0
) to the voltage g(e
0
) in the solution to the
network. (Some of the edges may have negative current ow when
the network is solved, but we could reverse the direction on these
edges to get a positive value. The edges with zero current ow may
be deletedor contracted; they will not give rise to squares.) Thus
a squared square corresponds to a planar network of resistors which
have the net eect of a resistance of one ohm.
Problem 36F. Solve the special networks in Fig. 36.6 below by
any means you wish, and sketch the corresponding squared rectan-
gles. The dual graphs may also be thought of as special networks
what are the squared rectangles that arise from the dual networks?
Figure 36.6
36. Electrical networks and squared squares 519
Let N be the incidence matrix of a digraph D and a diagonal
matrix indexed by E(D) with (e, e) = r(e) > 0, the resistance of
the edge e; let C := N
1
N

. If an additional edge e
0
is directed
from one vertex y to another vertex x of D, and a sucient voltage
source is impressed on that single edge so that a current f(e
0
) ows
through it, then Kirchhos laws imply that the potential dierence
g(e) across an edge e joining vertex a to vertex b, say, is given by
(36.3) g(e) =
f(e
0
)
(D)
(xy.ab)
where (xy.ab) denotes the cofactor of the element c
yb
in the cofactor
of the element c
xa
of C = (c
ij
) and (D) denotes the number of
spanning trees of D (the complexity of D). See Jeans (1908).
A consequence of this is the following theorem.
Theorem 36.4. If the voltage source s(e
0
) on the distinguished
edge e
0
of a special network with graph G is taken to be the number
of spanning trees on the edge e
0
, then all values f(a) of the resulting
current ow will be integral, and the current ow f(e
0
) through the
distinguished edge will be equal to the number of spanning trees
o the edge e
0
.
Proof: In the discussion preceding the statement of the theorem,
take D to be an orientation of G
/
e
0
and the identity matrix. If
the current ow f(e
0
) is (D) = (G
/
e
0
), then it is clear from (36.3)
that all values of f(a) = g(a) are integral. It remains only to show
that g(e
0
) is equal to the number of spanning trees on the edge
e
0
.
By (36.3), g(e
0
) = (xy.yx) = (xy.xy). By denition, (xy.xy)
is the determinant of the matrix N
//
N
//

where N
//
consists of N
with rows x and y deleted. By the CauchyBinet theorem, this is
equal to

S
( det(N
//
S
))
2
where the sum is extended over all (n 2)-subsets S of the edge
set and N
//
S
is the square submatrix of N
//
whose rows are indexed
by elements of S.
520 A Course in Combinatorics
The proof is completed by the observation that det(N
//
S
) is 0
unless S e
0
is the edge set of a spanning tree in G, in which
case it is 1. This is similar to the proof of (36.2) and is left as an
exercise.
In summary, Theorem 36.4 says that the resistance between the
ends of an edge e
0
of a graph G when all other edges are thought of
as unit resistors, is the ratio (G
//
e
0
)/(G
/
e
0
). Readers might check
that this agrees with their knowledge of electrical networks when
G is a polygon or a bond-graph. A squared square arises when we
nd a planar graph G and an edge e
0
so that (G
//
e
0
) = (G
/
e
0
). This
statement alone does not necessarily help in the search for squared
squares (and electrical engineers will think this would be a waste
of resistors anyway). Such graphs appear to be quite rare.
Notes.
The rst appearance in the literature of the problem of using in-
congruent squares to make a rectangle seems to be M. Dehn (1903).
The rst extensive tables of squared squares were published by
C. J. Bouwkamp, A. J. W. Duijvestijn, and P. Medema (1960).
W. T. Tutte (1918) has made numerous signicant contributions
to graph theory. He is Emeritus Professor of Mathematics in the
Department of Combinatorics and Optimization of the University
of Waterloo, one of the few academic departments in the world
whose name includes the word combinatorics.
For the proof of Theorem 36.3, we have followed De Bruijn and
Van Aardenne-Ehrenfest (1951).
One of the Cambridge students mentioned earlier, C. A. B. Smith,
wrote a very nice account of the history of their collaboration, called
Did Erd os save Western Civilization? (see Graham and Nesetril
(1997)). Apparently, Erd os conjectured that any dissection of a
square into smaller squares must contain at least two squares of
equal size. One thing led to another and Tutte ended up at Bletch-
ley Park instead of in the army. It is rumored that there he supplied
the vital clue to the deciphering of the Enigma, which explains the
title of the paper.
References.
N. G. de Bruijn and T. van Aardenne-Ehrenfest (1951), Circuits
36. Electrical networks and squared squares 521
and trees in oriented linear graphs, Simon Stevin 28, 203217.
C. J. Bouwkamp, A. J. W. Duijvestijn, P. Medema (1960), Tables
relating to simple squared rectangles of order nine through fteen,
T. H. Eindhoven.
A. J. W. Duijvestijn (1978), Simple perfect squared square of lowest
order, J. Combinatorial Theory (B) 25, 555558.
M. Dehn (1903), Zerlegung von Rechtecke in Rechtecken, Math.
Ann. 57.
R. L. Graham and J. Nesetril (Eds.) (1997), The Mathematics of
P. Erdos, Springer.
J. H. Jeans (1908), The Mathematical Theory of Electricity and
Magnetism, Cambridge University Press.
W. T. Tutte (1948), The dissection of equilateral triangles into
equilateral triangles, Proc. Cambr. Phil. Soc. 44, 463482.
W. T. Tutte (1961), Squaring the square, in: M. Gardner, The
2nd Scientic American Book of Mathematical Puzzles and Di-
versions, Simon and Schuster.
W. T. Tutte (1965), The quest of the perfect square, Amer. Math.
Monthly 72, No. 2, 2935.
37
Polya theory of counting
We return to counting in this chapter. There are many instances
when we are not interested in the number of primitive objects, but
rather the number of equivalence classes of objects with respect to
an appropriate equivalence relation. Moreover, these equivalence
relations are often induced by certain permutation groups in a nat-
ural way.
Question 1. What is the number of essentially dierent neck-
laces which can be made with n beads of two dierent colors? For
n = 6, this number is, by inspection, 13.

Note that we choose to regard the necklaces

(37.1)

and

as essentially the same, since one can be reected through a ver-

tical axis (or ipped over) to get the other. Evidently, underlying
our concept of essentially dierent is the dihedral group D
n
of
automorphisms of the n-gon.
Question 2. What is the number of nonisomorphic simple graphs
on n vertices? For n = 4, there are 11.
Question 3. What is the number of essentially dierent ways to
paint the faces (or edges, or vertices) of the cube with n colors?
37. Polya theory of counting 523
Question 4. Given three white balls and one black ball, what is
the number of ways to distribute them into two square boxes and
one round box?
[
s
][ ]( ) [

][
s
]( ) [

][
s
]( ) [
s
][

]( )
[

][ ](
s
) [

][

](
s
) [
s
][ ](

) [

][
s
](

)
[
s
][

](

) [
s
][ ](

) [

][
s
](

) [

][ ](
s
)
[

][

](
s
) [
s
][ ](

) [

][ ](
s
) [ ][ ](
s
)
Let A and B be nite sets and G a group of permutations of A
(or more generally, a nite group acting on A). The elements of B
will be called colors. The group G acts on the set B
A
of mappings
f : A B when for G and f B
A
, we dene (f) B
A
by
((f))(x) := f(
1
(x)).
It is the orbits of an appropriate G on the set B
A
that we wish to
count in the rst three questions above.
(The use of
1
rather than on the right-hand side above is
not an error. It is required to ensure that ((f)) = ()(f),
which is necessary in order to have a legitimate action of G on
B
A
. Formally, we have a homomorphism of G into the symmetric
group on B
A
.)
In the necklace problem, we would take A to be the vertex set
of an n-gon, and G as its automorphism group represented as a
permutation group on the vertices (i.e. the dihedral group D
n
of
order 2n in its usual representation). A slightly dierent problem
arises when the beads to be used are not round, but, say, at on
one side. Then the necklaces cannot be ipped over and those in
(37.1) are regarded as distinct.
Here, we would take G to be the cyclic group of order n in its
regular representation; cf. Example 10.5.
Asking for the number of orbits of mappings is the simplest ques-
tion we can pose. We might further want to know the number of
orbits of a given weight; for example, the number of necklaces of
n beads, k of which are to be white, or the number of graphs on
n vertices with k edges. We might introduce a permutation group
524 A Course in Combinatorics
H on B and, with respect to a more general equivalence relation,
ask for the number of congurations. (This would be necessary
for the distribution problem.)
To begin, we recall Burnsides lemma, Theorem 10.5: the number
of orbits of a nite group G on a set X is the average number of
xed points:
(37.2)
1
[G[

G
(),
where () denotes the number of points of S that are xed by the
permutation . Recall that this formula followed from counting
ordered pairs (x, ) X G such that (x) = x.
Theorem 37.1. Let A and B be nite sets and let G act on A.
Denote by c
k
(G) the number of permutations in G that have exactly
k cycles in their cycle decomposition on A. Then the number of
orbits of G on the set B
A
of all mappings f : A B is
1
[G[

k=1
c
k
(G)[B[
k
.
Proof: By Burnsides lemma, the number of orbits is given by
(37.2) where () is here the number of mappings f : A B such
that (f) = f, i.e. f(a) = f(
1
(a)) for all a A. But a mapping
f is xed by if and only if f is constant on every cycle of ; the
reader should check this. Such mappings are obtained by assigning
an element of B to each cycle of , and thus if has k cycles, the
number of mappings f xed by is [B[
k
.
We encourage the reader to pause at this point and to verify that
Theorem 37.1 produces the answer 13 when applied to the necklace
counting problem raised as Question 1 above.
Since it is to be used extensively later, it is appropriate to intro-
duce the cycle index of a permutation group at this point. First,
in this chapter, it will be convenient to use the venerable notation
(1
k
1
2
k
2
. . . n
k
n
) for the partition of the integer n which has k
i
parts
of size i, i = 1, 2, . . . , n. Do not compute powers or multiply; this
is just notation.
37. Polya theory of counting 525
For a permutation of a set A, let z
i
() denote the number
of cycles of having length i; so (1
z
1
()
2
z
2
()
. . . ) is a partition of
n = [A[, called the type of . Given a group G acting on A, we
dene the cycle index Z
G
as a polynomial in n letters X
1
, X
2
, . . . , X
n
by
Z
G
(X
1
, X
2
, . . . , X
n
) :=
1
[G[

G
X
z
1
()
1
X
z
n
()
n
.
The assertion of Theorem 37.1 is that the number of orbits of G
on B
A
is
Z
G
(b, b, . . . , b) =
1
[G[

G
b
z
1
()+z
2
()++z
n
()
where b := [B[.
Example 37.1. A cyclic group C
n
of order n has (d) elements
of order d for each divisor d of n. As a permutation in the regular
representation of C
n
, an element of order d has n/d cycles of length
d. Thus, for the regular representation of C
n
, we have
Z
C
n
(X
1
, . . . , X
n
) =
1
n

d[n
(d)X
n/d
d
.
We can now count the number of one-sided necklaces of n beads
of two colors. The answer is Z
C
n
(2, 2, . . . , 2) =
1
n

d[n
(d)2
n/d
; cf.
equation (10.12). For n = 6, we get 14, one more than with dihedral
equivalence (the two necklaces in (37.1) are now distinguished). For
n = 10, we get 108.
Example 37.2. To nd the cycle index of the dihedral group D
n
in its usual representation on n points, we must account for the
permutations (reections) outside of its cyclic subgroup of order n.
There are two cases depending on the parity of n:
Z
D
n
=
_
_
_
1
2n
_

d[n
(d)X
n/d
d
+nX
1
X
n1
2
2
_
for n odd,
1
2n
_

d[n
(d)X
n/d
d
+
n
2
X
2
1
X
n
2
1
2
+
n
2
X
n
2
2
_
for n even.
526 A Course in Combinatorics
The number of two-sided necklaces with n beads of two dierent
colors is Z
D
n
(2, 2, . . . , 2). For n = 6, we get 13 (verifying the list
following Question 1). For n = 10, we get 78.
Problem 37A. Compute the cycle index of the group of rotations
of the cube represented as permutations of the six faces. (There
are 24 rotationsincluding the identitythat map the cube, as a
rigid body in 3-space, onto itself. We are ignoring the 24 reections
of the cube.) What is the number of essentially dierent ways to
paint the faces of the cube in 2 colors? In 3 colors? In n colors?
Before going further, let us write down the cycle index of the
symmetric groups. We noted the formula for the number of parti-
tions of a given type (1
k
1
2
k
2
. . . n
k
n
) in equation (13.3). Each block
of size i may be equipped with a cyclic permutation in (i1)! ways,
so we have
Z
S
n
=

(1
k
1
2
k
2
... )
1
1
k
1
2
k
2
. . . n
k
n
k
1
!k
2
! . . . k
n
!
X
k
1
1
X
k
2
2
. . . X
k
n
n
.
We list the rst several polynomials below:
1!Z
S
1
= X
1
,
2!Z
S
2
= X
2
1
+X
2
,
3!Z
S
3
= X
3
1
+ 3X
1
X
2
+ 2X
3
,
4!Z
S
4
= X
4
1
+ 6X
2
1
X
2
+ 3X
2
2
+ 8X
1
X
3
+ 6X
4
,
5!Z
S
5
= X
5
1
+ 10X
3
1
X
2
+ 15X
1
X
2
2
+ 20X
2
1
X
3
+ 20X
2
X
3
+ 30X
1
X
4
+ 24X
5
,
6!Z
S
6
= X
6
1
+ 15X
4
1
X
2
+ 45X
2
1
X
2
2
+ 40X
2
3
+ 40X
3
1
X
3
+ 15X
3
2
+ 120X
1
X
2
X
3
+ 90X
2
1
X
4
+ 90X
2
X
4
+ 144X
1
X
5
+ 120X
6
.
Example 37.3. Theorem 37.1 can be used for the enumeration of
graphs. Let V be a xed set of n vertices and let E consist of all
2-subsets of V . Then the simple graphs with vertex set V can be
viewed as mappings f : E 0, 1, the graph corresponding to f
having edge set f
1
(1) = e : f(e) = 1.
37. Polya theory of counting 527
Two graphs on V are isomorphic if and only if there exists a
permutation of V taking the edges of one onto the edges of the
other. To describe this in a manner suitable for the application
of Theorem 37.1, let S
n
be the symmetric group on V and let
S
(2)
n
=
(2)
: S
n
be the induced group, isomorphic to S
n
, of
permutations of E where

(2)
: x, y (x), (y).
Then S
(2)
n
acts on 0, 1
E
and we note that f, g : E 0, 1 corre-
spond to isomorphic graphs if and only if they lie in the same orbit
of S
(2)
on 0, 1
E
.
It is not a pleasant task to compute the types of all permutations
in S
(2)
n
. For example, if S
5
has type (2
1
3
1
), then it turns out
that
(2)
has type (1
1
3
1
6
1
) on the 10 edges of K
5
. The cycle index
of S
(2)
5
is
1
120
(X
10
1
+ 10X
4
1
X
3
2
+ 15X
2
1
X
4
2
+ 20X
1
X
3
3
+ 20X
1
X
3
X
6
+ 30X
2
X
2
4
+ 24X
2
5
).
In general, we will have a sum over all partitions of n. The number
of nonisomorphic graphs on ve vertices is found, by substituting
2 for all indeterminates, to be 34.
We now generalize Theorem 37.1 to allow for weights. Let A be
an n-set, let G act on A, and let B be a nite set of colors. Let R
be a commutative ring containing the rationals and let w : B R
assign a weight w(b) R to each color b B. For f : A B,
dene
W(f) :=

aA
w(f(a)) R;
here W(f) is the weight of the function f. Note that two mappings
representing the same orbit of G on B
A
have the same weight, i.e.
W((f)) = W(f) for every G. The sum

fR
W(f),
528 A Course in Combinatorics
extended over a system of representatives R for the orbits, is called
the conguration counting series. (The orbits are often called con-
gurations; the term series becomes appropriate when the weights
are monomials.) If all weights w(b) are 1, then the following theo-
rem reduces to Theorem 37.1.
Theorem 37.2. With the terminology as above, the conguration
counting series is given by

W(f) = Z
G
_

bB
w(b),

bB
[w(b)]
2
, . . . ,

bB
[w(b)]
n
_
.
Proof: Let N =

W(f), the sum extended over all pairs (, f)
with G, f B
A
, and (f) = f. We have
N =

fB
A
W(f)[G
f
[,
where G
f
is the stabilizer of f. Consider the terms W(f)[G
f
[ as f
ranges over an orbit O of G on B
A
: each is equal to W(f
0
)[G[/[O[
where f
0
is a representative of the orbit O, so these terms sum to
[G[W(f
0
). It is now clear that N is [G[ times the conguration
counting series.
On the other hand,
N =

G
_
_

(f)=f
W(f)
_
_
.
Recalling the denition of Z
G
, we see that the proof will be com-
plete if we show that

(f)=f
W(f) =
_

bB
w(b)
_
k
1
_

bB
[w(b)]
2
_
k
2

_

bB
[w(b)]
n
_
k
n
whenever is a permutation of type (1
k
1
2
k
2
n
k
n
).
A mapping f : A B is xed by if and only if f is constant
on every cycle of on A. Let
C
1
, C
2
, . . . , C
k
, k := k
1
+k
2
+ +k
n
37. Polya theory of counting 529
be the cycles of . The mappings f B
A
xed by are in one-to-
one correspondence with k-tuples (b
1
, b
2
, . . . , b
k
) of elements of B
(the corresponding mapping is the one associating b
i
to all elements
of C
i
). The weight of the mapping f corresponding to (b
1
, . . . , b
k
)
is
W(f) =
k

i=1
[w(b
i
)]
[C
i
[
,
and summing over all k-tuples (b
1
, b
2
, . . . , b
k
),

(f)=f
W(f) =

b
1
,b
2
,...,b
k
[w(b
1
)]
[C
1
[
[w(b
2
)]
[C
2
[
[w(b
k
)]
[C
k
[
=
_

b
1
[w(b
1
)]
[C
1
[
__

b
2
[w(b
2
)]
[C
2
[
_

=
n

c=1
_

b
[w(b)]
c
_
k
c
,
as required.
Example 37.4. Consider the necklace problem with cyclic equiv-
alence. Here G = C
n
, B = black, white. We take R to be the
polynomial ring Q[X] and dene w(black) = 1, w(white) = X.
Then the weight of a coloring f : A B is
W(f) = X
k
where k is the number of white beads.
Thus the number of essentially dierent cyclic necklaces with n
beads, k of which are white, is the coecient of X
k
in the congu-
ration counting series
Z
C
n
(1 +X, 1 +X
2
, 1 +X
3
, . . . , 1 +X
n
)
=
1
n

d[n
(d)[1 +X
d
]
n/d
=
1
n

d[n
(d)
n/d

r=0
_
n/d
r
_
x
rd
.
530 A Course in Combinatorics
This coecient is
1
n

d[(k,n)
(d)
_
n/d
k/d
_
.
So if k and n are relatively prime, the number is simply
1
n
_
n
k
_
,
while for n = 12, k = 4, there are
1
12
_
(1)
_
12
4
_
+(2)
_
6
2
_
+(4)
_
3
1
_
_
=
43 necklaces.
Problem 37B. What is the number of essentially dierent ways
to paint the faces of a cube such that one face is red, two are blue,
and the remaining three are green? Do this by hand (which may
be quicker) as well as by using Theorem 37.2, and compare your
answers.
Let A and B be nite sets, [A[ = n, and let G and H be nite
groups, G acting on A and H on B. The direct product GH acts
on B
A
when for f B
A
and (, ) GH we dene (, )(f) by
((, )(f))(a) = (f(
1
(a)).
Theorem 37.3. The number of orbits of GH on B
A
is
1
[H[

H
Z
G
(m
1
(), m
2
(), . . . , m
n
())
where
m
i
() :=

j[i
jz
j
(), i = 1, 2, . . . , n.
Proof: By Burnsides lemma, the number of orbits is
1
[G[[H[

G,H
(, )
where (, ) is the number of mappings f B
A
with (, )(f) = f.
To complete the proof, it will suce to show that for each H,
(37.3)
1
[G[

G
(, ) = Z
G
(m
1
(), m
2
(), . . . , m
n
()).
37. Polya theory of counting 531
Fix G and H, and let
C
1
, C
2
, . . . , C
k
, k := z
1
() +z
2
() + +z
n
()
be the cycles of on A. A mapping f : A B is xed by (, ) if
and only if all the restrictions f
i
:= f[C
i
are xed, 1 i k. So
(, ) is the product, over i in the range 1 i k, of the numbers
of mappings f
i
: C
i
B satisfying
f
i
((a)) = (f
i
(a)) for all a C
i
.
Let C
i
have length and x a
o
C
i
. Suppose that f
i
: C
i
B
is xed by ([C
i
, ) and that f
i
(a
o
) = b. Then f
i
is completely de-
termined: f
i
(
t
(a
o
)) =
t
(b). Moreover, b = f
i
(a
o
) = f
i
(

(a
o
)) =

(b) and so we see that the cycle of containing b must have length

/
, a divisor of .
Conversely, if b is an element of B lying in a cycle of having
a length dividing := [C
i
[, then we can dene a mapping f
i
on C
i
by f
i
(
t
(a
o
)) :=
t
(b) (check that f
i
is well dened) and this f
i
is
xed by ([C
i
, ).
In summary, when [C
i
[ = , the xed mappings f
i
: C
i
B are
equinumerous with the elements of B lying in cycles of of length
dividing . The number of such elements is, of course,
m

() =

j[
jz
j
().
Then
(, ) =
k

i=1
m
[C
i
[
() = [m
1
()]
z
1
()
[m
2
()]
z
2
()
[m
n
()]
z
n
()
,
from which the desired equation (37.3) is immediate.
Example 37.5. What is the number of ways to distribute 2 red,
2 yellow, and 4 green balls into 1 round and 3 square boxes? We
take
A = R
1
, R
2
, Y
1
, Y
2
, G
1
, G
2
, G
3
, G
4
,
532 A Course in Combinatorics
G = S
2
S
2
S
4
, B = r, s
1
, s
2
, s
3
, H = S
1
S
3
.
It is easy to see that the cycle index of G is the product of the
cycle indices of the symmetric groups of which it is the product, so
(37.4) Z
G
= Z
S
2
Z
S
2
Z
S
4
=
1
2!2!4!
(X
2
1
+X
2
)
2
(X
4
1
+ 6X
2
1
X
2
+ 3X
2
2
+ 8X
1
X
3
+ 6X
4
).
There are three types of permutations in H. If is the identity,
then
m
1
() = 4, m
2
() = 4, m
3
() = 4, m
4
() = 4;
if transposes two of s
2
, s
2
, s
3
, then
m
1
() = 2, m
2
() = 4, m
3
() = 2, m
4
() = 4;
if xes only r, then
m
1
() = 1, m
2
() = 1, m
3
() = 4, m
4
() = 1.
By Theorem 37.3, the number of distributions is
1
3!
1
2!2!4!
_
(4
2
+ 4)
2
(4
4
+ 6 4
2
4 + 2 4
2
+ 8 4 4 + 6 4)
+ 3(2
2
+ 4)
2
(2
4
+ 6 2
2
4 + 3 4
2
+ 8 2 2 + 6 4)
+2(1
2
+ 1)
2
(1
4
+ 6 1
2
1 + 3 1
2
+ 8 1 4 + 6 1)

= 656 (if we did the arithmetic correctly).

Problem 37C. Given G on A, H on B, nd an expression for the
number of orbits of injective mappings f : A B.
We give, without proof, a statement of the extension of Theorem
37.3 to include weights. See De Bruijn (1964) for a proof. If all
weights are 1, then this reduces to Theorem 37.3.
Theorem 37.4. Let G act on A, H on B, let R be a commutative
ring and w : B R. Assume that w is constant on each orbit of
H on B. With W(f) :=

aA
w[f(a)], the sum

W(f), extended
37. Polya theory of counting 533
over a system of representation for the orbits of G H on B
A
, is
equal to
1
[H[

H
Z
G
_
M
1
(), M
2
(), . . . , M
n
()
_
where
M
i
() =

i
(b)=b
[w(b)]
i
.
Example 37.6. We continue Example 13.5. Take
w(r) = r, w(s
1
) = w(s
2
) = w(s
3
) = s,
where R = [r, s]. The number of distributions with t balls in the
round box and the remaining 8t in square boxes is the coecient
of r
t
s
8t
in
1
6
_
Z
G
(r + 3s, r
2
+ 3s
2
, r
3
+ 3s
3
, r
4
+ 3s
4
)
+3Z
G
(r +s, r
2
+3s
2
, r
3
+s
3
, r
4
+3s
4
) +2Z
G
(r, r
2
, r
3
+3s
3
, r
4
)
_
where Z
G
is the polynomial in (37.4).
Problem 37D. Identify necklaces of two types of colored beads
with their duals obtained by switching the colors of the beads. (We
can now distinguish between the two colors, but we cant tell which
is which.) Now how many of these reduced congurations are there?
(For example, with n = 6 and dihedral equivalence, there are 8
distinct congurations.) How many necklaces are self-dual?
Problem 37E. The faces of the cube are to be colored with s col-
ors, and the vertices with t colors (which can be assumed dierent,
though it is not important). In how many essentially dierent ways
(with respect to the rotation group of the cube) can this be done?
* * *
The theorems of this chapter are also sources of algebraic iden-
tities.
534 A Course in Combinatorics
Consider the case of distribution of like objects into unlike cells.
(We are recalling something which was discussed in Chapter 13.)
For example, in how many ways can we distribute 23 apples to
Fred, Jane, and George? More generally, consider the distributions
of a set A of n apples to a set B of x individuals (the elements
of A are like, whereas the elements of B remain distinct, i.e. are
unlike). Such a distribution amounts to the selection of a family
(k
b
: b B) of nonnegative integers k
b
with

bB
k
b
= n, k
b
apples
going to individual b.
Thus the number of distributions, as shown in Theorem 13.3, is
_
n +x 1
n
_
.
However, the best formal denition of such a distribution is as
an orbit of mappings f : A B with respect to the symmetric
group S
n
acting on A. (Two mappings f, g B
A
determine the
same distribution if and only if they dier by a permutation of
the apples.) By Theorem 37.1, the number of distributions is
1
n!
n

k=0
c
k
(S
n
)x
k
,
where c
k
(S
n
) is the number of permutations of n letters with ex-
actly k cycles. In Chapter 13, the numbers c
k
(S
n
) were called the
signless Stirling numbers of the rst kind, and denoted by c(n, k).
By Theorem 37.2, for every nonnegative integer x,
(x +n 1)
(n)
=
n

k=0
c
k
(S
n
)x
k
,
and hence this must hold as a polynomial identity. We have there-
fore re-proved the formula (13.7).
Take G to be S
n
on an n-set A, let B be nite and let w be
the insertion of B into the polynomial ring [B]. Here the weight
of a mapping f : A B is a monomial and two mappings are
equivalent, that is, represent the same conguration, if and only if
37. Polya theory of counting 535
they have the same weight. The conguration counting series
(37.5)

W(f) = Z
S
n
_

bB
b,

bB
b
2
, . . . ,

bB
b
n
_
expresses the so-called homogeneous-product-sum symmetric func-
tions as a polynomial in the power-sum symmetric functions. For
example, with n = 3 and B = X, Y ,
6(X
3
+X
2
Y +XY
2
+Y
3
) =
(X +Y )
3
+ 3(X +Y )(X
2
+Y
2
) + 2(X
3
+Y
3
).
With n = 4 and B = X, Y, Z, (37.5) expresses

i+j+k=4
X
i
Y
j
Z
k
as a polynomial in
X +Y +Z, X
2
+Y
2
+Z
2
, X
3
+Y
3
+Z
3
, and X
4
+Y
4
+Z
4
.
Notes.
We remark that Z
S
n
(X
1
, . . . , X
n
) is the coecient of Y
n
in
exp
_
X
1
Y +
1
2
X
2
Y
2
+
1
3
X
3
Y
3
+
_
as an element of ([X
1
, X
2
, . . . ])[[Y ]].
The answers provided by P olya theory have the advantage (over
inclusion-exclusion, say) that the formula produced is the sum of
positive terms rather than an alternating sum.
G. P olya (18871985) was a Hungarian mathematician whose
book Problems and Theorems in Analysis written in 1924 with G.
Szego is still a classic. But he is perhaps best known for How to
Solve It, which has sold more than one million copies. He wrote
papers on number theory, complex analysis, combinatorics, proba-
bility theory, geometry and mathematical physics.
References.
N. G. de Bruijn (1964), P olyas theory of counting, in: E. F. Beck-
enbach (ed.), Applied Combinatorial Mathematics, Wiley.
F. Harary and E. D. Pulver (1966), The power group enumeration
theorem, J. Combinatorial Theory 1.
38
Baranyais theorem
In this chapter we shall give an elegant application of the integrality
theorem on ows to a problem in combinatorial design.
Example 38.1. Suppose we have been entrusted to draw up a
schedule for the Big Ten football teams. Each weekend they are
to divide into 5 pairs and play. At the end of 9 weeks, we want
every possible pair of teams to have played exactly once.
Here is one solution. Put 9 teams on the vertices of a regular
9-gon and one in the center. Start with the pairing indicated below
and obtain the others by rotating the gure by multiples of 2/9.
0
1
2
3
4 5
6
7
8
x
Explicitly, here is the schedule:
x 0
1 8
2 7
3 6
4 5
x 1
2 0
3 8
4 7
5 6
x 2
3 1
4 0
5 8
6 7
x 3
4 2
5 1
6 0
7 8
x 4
5 3
6 2
7 1
8 0
x 5
6 4
7 3
8 2
0 1
x 6
7 5
8 4
0 3
1 2
x 7
8 6
0 5
1 4
2 3
x 8
0 7
1 6
2 5
3 4
There are many other ways to do this. For example, start with the
initial pairing shown below.
38. Baranyais theorem 537
0
1
2
3
4 5
6
7
8
x
A perfect matching in a graph is also called a 1-factor. A parti-
tion of the edge set of a graph into 1-factors is a 1-factorization. In
the above example, we constructed 1-factorizations of K
10
. There
are 396 nonisomorphic 1-factorizations of K
10
, a computer result of
E. N. Gelling (1973). See also Mendelsohn and Rosa (1985).
Example 38.2. Let be any abelian group of odd order. Consider
the complete graph on the vertex set . For g , let
/
g
:= g, a, b : a +b = 2g, a ,= b.
Then /
g
: g is a 1-factorization of the complete graph.
We consider the following generalization of this problem: let us
use the term parallel class of k-subsets of an n-set to mean a set
of n/k k-subsets which partition the n-set. Can the set of all k-
subsets be partitioned into parallel classes of k-subsets? Of course,
this necessitates that k divides n. The number of parallel classes
required would be
k
n
_
n
k
_
=
_
n1
k1
_
.
This was not hard for k = 2. It is much harder for k = 3,
but was done by R. Peltesohn in 1936, and for k = 4 by J.-C.
Bermond (unpublished). The reader might try to nd appropriate
parallel classes for n = 9, k = 3. It is certainly not clear that our
generalization always admits a solution and so it was surprising
when Zs. Baranyai proved Theorem 38.1 in 1973. All known proofs
use a form or consequence of Theorem 7.2 or 7.4. The proof we
give is due to A. E. Brouwer and A. Schrijver (1979).
Theorem 38.1. If k divides n, the set of all
_
n
k
_
k-subsets of
an n-set may be partitioned into disjoint parallel classes /
i
, i =
1, 2, . . . ,
_
n1
k1
_
.
Proof: In this proof, we will use the term m-partition of a set X
for a multiset / of m pairwise disjoint subsets of X, some of which
538 A Course in Combinatorics
may be empty, whose union is X. (The normal use of partition
forbids the empty set, but it is important here to allow it to occur,
perhaps with a multiplicity, so that the total number of subsets is
m.)
In order to get an inductive proof to work, we prove something
seemingly stronger than the original statement. Let n and k be
given, assume k divides n, and let m := n/k, M :=
_
n1
k1
_
. We
assert that for any integer , 0 n, there exists a set
/
1
, /
2
, . . . , /
M
of m-partitions of 1, 2, . . . , with the property that each subset
S 1, 2, . . . , occurs in exactly
(38.1)
_
n
k [S[
_
of the m-partitions /
i
. (The binomial coecient above is inter-
preted as zero if [S[ > k, of course, and for S = , the m-partitions
containing are to be counted with multiplicity equal to the num-
ber of times the empty set appears.)
Our assertion will be proved by induction on . Notice that it is
trivially true for = 0, where each /
i
will consist of m copies of
the empty set. Also notice that the case = n will prove Theorem
38.1, since the binomial coecient in (38.1) is then
_
0
k [S[
_
=
_
1 if [S[ = k,
0 otherwise.
Remark. This somewhat technical statement is not really more
general, but would follow easily from Theorem 38.1. If M parallel
classes exist as in the statement of Theorem 38.1, then for any
set L of points of X, the intersections of the members of the
parallel classes with L will provide m-partitions of L with the above
property.
Assume for some value of < n that m-partitions /
1
, . . . , /
M
exist with the required property. We form a transportation network
as follows. There is to be a source vertex , another named /
i
38. Baranyais theorem 539
for each i = 1, 2, . . . , M, another vertex named S for every subset
S 1, 2, . . . , , and a sink vertex . There is to be a directed edge
from to each /
i
with capacity 1. There are to be directed edges
from/
i
to the vertices corresponding to members of /
i
(use j edges
to , if occurs j times in /
i
); these may have any integral capacity
1. There is to be a directed edge from the vertex corresponding
to a subset S to of capacity
_
n 1
k [S[ 1
_
.
We exhibit a ow in this network: assign a ow value of 1 to the
edges leaving , a ow value of (k [S[)/(n ) to the edges from
/
i
to each of its members S, and a ow value of
_
n1
k[S[1
_
to the
edge from S to . This is easily checked to be a ow: the sum of
the values on edges leaving a vertex /
i
is

S/
i
k [S[
n
=
1
n
_
mk

S/
i
[S[
_
=
1
n
(mk ) = 1.
The sum of the values on the edges into a vertex S is

i:S/
i
k [S[
n
=
k [S[
n
_
n
k [S[
_
=
_
n 1
k [S[ 1
_
.
Since all the edges leaving are saturated, this is a maximum ow
and has strength M. The edges into are also saturated in this
ow, and hence in any maximum ow.
540 A Course in Combinatorics
By Theorem 7.2, this network admits an integral-valued maxi-
mum ow f. All edges leaving will be saturated, so it is clear
that for each i, f assigns the value 1 to one of the edges leaving
/
i
and 0 to all others. Say f assigns 1 to the edge from /
i
to its
member S
i
. For each subset S, the number of values of i such that
S
i
= S is
_
n1
k[S[1
_
.
(The desired integral ow f could also have been obtained with-
out explicitly introducing capacities and using Theorem 7.2, but
simply by applying Theorem 7.4 to the exhibited rational ow com-
pleted to a circulation with the aid of an extra edge from to
with ow value M.)
Finally, we obtain a set of m-partitions /
/
1
, . . . , /
/
M
of the set
1, 2, . . . , +1 by letting /
/
i
be obtained from /
i
by replacing the
distinguished member S
i
by S
i
+ 1, i = 1, . . . , M. Readers
should check that each subset T of 1, 2, . . . , + 1 occurs exactly
_
n ( + 1)
k [T[
_
times among /
/
1
, . . . , /
/
M
. This completes the induction step.
Problem 38A. Let v and u be integers with v 2u and v even,
and consider the complete graph K
u
as a subgraph of K
v
. Suppose
the edges of K
u
are colored with v1 colors so that distinct edges of
the same color are disjoint. Show that this coloring can be extended
to a coloring of E(K
v
) with v 1 colors so that edges of the same
color are disjoint. (Such a coloring of E(K
v
) is equivalent to a
1-factorization of K
v
, with the edges of any given color forming a
1-factor.)
Notes.
Zsolt Baranyai (19481978) was a Hungarian mathematician who
was also a professional recorder player. He toured Hungary with
the Barkfark Consort giving concerts and died in a car accident on
a country road after one of them. His mathematical work included
many other wonderful results on complete uniform hypergraphs.
One-factorizations of complete graphs are related to symmetric
Latin squaressee Chapter 17. The result of Problem 38A is due
38. Baranyais theorem 541
to Cruse (1974). Generalizations may be found in Baranyai and
Brouwer (1977).
References.
Zs. Baranyai and A. E. Brouwer (1977), Extension of colourings of
the edges of a complete (uniform hyper) graph, Math. Centrum
Dep. Pure Math. ZW. 91, 10 pp.
A. E. Brouwer and A. Schrijver (1979), Uniform hypergraphs, in:
A. Schrijver (ed.), Packing and Covering in Combinatorics,
Mathematical Centre Tracts 106, Amsterdam.
A. Cruse (1974), On embedding incomplete symmetric Latin
squares, J. Combinatorial Theory (A) 16, 1822.
E. N. Gelling (1973), On 1-factorizations of the complete graph and
the relationship to round robin schedules, M.Sc. Thesis, Univer-
sity of Victoria.
E. Mendelsohn and A. Rosa (1985), One-factorizations of the com-
plete grapha survey, Journal of Graph Theory 9.
R. Peltesohn (1936), Das Turnierproblem f ur Spiele zu je dreien,
Dissertation Berlin, August Pries, Leipzig.
Appendix 1
Hints and comments
on problems
Problem 1A. Show that the
_
5
2
_
pairs from 1, . . . , 5 can be used
to label the vertices in such a way that a simple rule determines
when there is an edge. To nd the full automorphism group, con-
sider the subgroup that xes a vertex and its three neighbors. This
graph is known as the Petersen graph. It was rst studied by the
Danish mathematician J. P. C. Petersen (18391910). See also
Chapter 21.
Problem 1B. Let the vertex set V be the disjoint union of V
1
and
V
2
, with no edges from V
1
to V
2
. What is the maximum number of
edges possible?
Problem 1C. Use (1.1). (i) Consider a vertex of degree 1; use
induction. (ii) A circuit has as many edges as vertices; if the graph
is connected, every vertex has degree 1.
Problem 1D. Call the vertices a
1
, a
2
, a
3
and b
1
, b
2
, b
3
. First, omit
a
3
and show that there is only one way to draw the graph with the
remaining six edges in the plane. A better way is to use Eulers
formula (see Chapter 33).
Problem 1E. Each color occurs an even number of times in a cir-
cuit. See J. A. Bondy, Induced subsets, J. Combinatorial Theory
(B) 12 (1972), 201202. The problem can also be formulated as
follows. Given a square (0, 1)-matrix with distinct rows, it is pos-
sible to delete a column so that the resulting matrix has distinct
rows. This fact can also easily be proved by induction.
Problem 1F. Consider the vertices at distance 2 from a xed
Hints and comments on problems 543
vertex. See Fig. 1.4.
Problem 1G. Use the pigeonhole principle.
Problem 1H. Consider when a
ij
= a
jk
= 1.
Problem 1I. Find an inductive procedure for constructing the
required circuit.
Problem 1J. For (i), establish a one-to-one correspondence be-
tween the vertices adjacent to x and those adjacent to y.
Problem 2B. Any ordering of 22335 works.
Problem 2C. Let A
i
be the number of spanning arborescences
with x
i
as root and denote by r
i
the number of edges out of x
i
.
The assertion can be proved by showing that the number of dierent
Eulerian circuits in G is
A
i

n
j=1
(r
j
1)!.
To do this, take i = 1 and consider some spanning arborescence
with x
1
as root. Fix some edge out of x
1
as the rst edge of an
Eulerian circuit. Number the other edges out of x
1
arbitrarily. For
i = 2, 3, . . . , n, number the edges out of x
i
arbitrarily but such
that the edge of the spanning arborescence has number r
i
. This
numbering leads in a natural way to the construction of an Eulerian
circuit and this process can be reversed. See Chapter 9 of J. H.
van Lint, Combinatorial Theory Seminar Eindhoven University of
Technology, Lecture Notes in Mathematics 382, Springer-Verlag,
1974.
Problem 2D. One approach is as follows. Show that T
n
has the
property that for each of its edges a, a has ends in dierent com-
ponents of G : e E(G) : c(e) < c(a). Then show that any
spanning tree T with this property is a cheapest spanning tree.
Problem 2E. There is only one way to obtain the dierence n1,
etc.
Problem 2F. Use Problem 1C.
Problem 2G. Show that if G is a graph with this property, there
is a graph G
/
with the same property, the same number of edges,
544 A Course in Combinatorics
and in which the vertex of degree m is connected to all other ver-
tices with degree > 1. Then use induction (m m + 2). The
construction can be done inductively by the same idea.
Problem 3A. For part (ii), if the conclusion fails, then H can
be written as the union of two edge-disjoint subgraphs A, B which
intersect in two vertices s, t. Apply the induction hypothesis to
graphs A
/
, B
/
obtained by adding an edge joining s, t to A, B, re-
spectively.
Problem 3B. It is possible to use the same type of argument as
given for K
6
(below this problem). Show that there must be a
vertex that is in two monochromatic triangles. Then, argue on the
remaining six vertices. The argument will be long and tricky. The
reader who has cheated and has read on, knows the Corollary to
Theorem 3.2.
Problem 3C. Call the terms, both even, on the right-hand side n
1
and n
2
. Suppose equality holds and consider a graph with n
1
+n
2
1
vertices. If there is no red K
p
and no blue K
q
, what is the red-degree
of any vertex? How many red edges are there in this graph?
Problem 3D. In Z
17
consider 2
i
, with i = 0, 1, 2, 3. For the
other problem use Z
13
. Apply (3.4).
Problem 3E. (a) Fix a vertex and consider all edges to, respec-
tively from, this vertex. Pick the larger set and use induction.
(b) Use the probabilistic method. A solution can be found in
Theorem 1.1 of P. Erd os and J. Spencer (1974), where this problem
is used to illustrate the method.
Problem 3F. Take as vertices 1, 2, . . . , n. Color i, j with the
color of [i j[. To avoid a triple for two colors, if 1 is red, then 2
must be blue and hence 4 must be red, etc. N(2) = 5. Separate two
such 2-colored congurations by ve numbers with a third color to
estimate N(13). See I. Schur,

Uber die Kongruenz x
m
+ y
m
z
m
(mod p), Jber. Deutsche Math. Ver. 25 (1916), 114116.
Problem 3G. Color with four colors as follows:
i, j (a
ij
, a
ji
) (i < j).
Hints and comments on problems 545
Problem 3H. Use the argument that was used for K
6
and two
colors.
Problem 3I. Let the color depend on (mod 3). Show that the
sum of two cubes in F

16
is not a cube.
Problem 3J. Consider a maximal subgraph with the required
property and then count the vertices not in that subgraph.
Problem 3K. Consider vertices of a xed color and delete edges
in such a way that recoloring is possible.
Problem 4A. (i) Color K
10
in such a way that red corresponds
to an edge of G, blue to a nonedge. There are triangles with 3 red
edges, 2 red and 1 blue, 2 blue and 1 red, and nally with 3 blue
edges. Let a
i
(i = 1, 2, 3, 4) be the numbers of these. Set up a
system of equations and inequalities for these numbers, expressed
in the degrees of the vertices of G. These should show that there
are at least four triangles in G and equality can then be excluded,
again by looking at the equations.
(ii) A second solution is as follows. Show there is a triangle.
Then consider a triangle, the ary 7-set, and edges in between. This
gives a number of new triangles of two types; estimate it.
Problem 4B. Use induction. First review the proof of Theorem
4.1.
Problem 4C. The number of triangles on an edge a with ends x
and y is at least deg(x) + deg(y) n. Sum this over all edges a.
Problem 4D. Fix any vertex x and let D
i
denote the set of ver-
tices at distance i from x in G.
For g = 2t +1, [V (G)[ 1 +r +r(r 1) + +r(r 1)
t1
since
it is easy to see that [D
i
[ = r(r 1)
i1
for i = 1, 2, . . . , t.
For g = 2t, we still have [D
i
[ = r(r 1)
i1
for i = 1, 2, . . . , t 1.
We can also get a lower bound [D
t
[
r1
r
D
t1
since each vertex
y D
t1
is adjacent to r 1 vertices in D
t
while a vertex y D
t
can of course be adjacent to at most r vertices in D
t1
.
Problem 4E. The graph is bipartite.
Problem 4F. By Theorem 4.3, there exists a Hamiltonian circuit
H. Assume that H does not contain an edge e with ends x and y.
546 A Course in Combinatorics
Let x
/
, respectively y
/
, be the successors of x, respectively y, on H.
Consider the path P from x
/
to y along H, then along y, x, and
from x to y
/
along H in the reverse direction. If the edge x
/
, y
/
is
in G, we are done. If not, argue as in the proof of Theorem 4.3 to
show that P can be completed to a circuit by removing one edge
and adding two others.
For a further generalization, see L. Lov asz (1979).
Problem 4G. Show that the maximum value of S occurs when
z
i
> 0 only on points of a clique.
Problem 4H. Count pairs of edges with a common vertex in two
ways.
Problem 5A. (i) Given A X, one side of a bipartition, count
the number of edges that have one end in A and the other in (A).
(ii) A trivalent graph with no perfect matching is the graph on
four vertices and six edges with one vertex incident with three non-
loops and loops incident with the other three vertices. Replace the
loops by appropriate simple graphs.
(iii) Same hint as for part (i).
Problem 5B. Construct a sequence of sets A
i
of size m
i
using new
elements only when it is necessary to keep condition H satised. A
solution is given on p. 41 of Van Lint (1974).
Problem 5C. Use induction on the number of nonzero entries of
the matrix. The argument of Theorem 5.5 can be used similarly.
Problem 5D. Theorem 5.5.
Problem 5E. Express the number of SDRs with 1, respectively
2, representing A
1
in terms of S
i
with i < n. See Problem 14A.
Problem 5F. Count the number of sets x
1
, x
2
, . . . , x
n
with x
k

A
k
for all k and with x
i
= x
j
.
Problem 5G. (i) is easy but (ii) gets messier. Consult Example
10.1
Problem 6A. Dene a poset using both the index and the size of
the integers a
i
.
Hints and comments on problems 547
Problem 6B. (i) Consider A and A. (ii) Let x be in all the sets.
Problem 6C. Show that large sets can be replaced by their com-
plements, and apply Theorem 6.5.
Problem 6D. (i) No two of the subsets are in the same chain. (ii)
Use induction.
Problem 6E. See C. Greene and D. Kleitman, Strong versions of
Sperners Theorem, J. Combinatorial Theory (A) 20 (1976), 8088.
Problem 7A. This is a straightforward application of the algo-
rithm. The minimum cut capacity is 20.
Problem 7B. Given a ow f with nonzero strength, show that
the edges e with f(e) ,= 0 contain the edges of a simple directed
path from s to t. Subtract a scalar multiple of the corresponding
elementary ow to obtain a ow f
/
which vanishes on more edges
than f and proceed by induction.
Four vertices suce for the required example.
Problem 7C. Consider a maximum ow f. From Equation (7.1),
edges from X
i
to Y
i
are saturated and edges from Y
i
to X
i
have
zero ow value.
Problem 7D. Given the bipartite graph G with vertices X Y ,
construct a network by adding two vertices s and t, edges directed
from s to elements of X of capacity 1, and edges directed from
elements of Y to t of capacity 1. Direct all original edges from X
to Y ; it is convenient to assign them a large capacity, say [X[ + 1,
so that no such edges are present in a minimum cut. Explain why
there exists a complete matching from X to Y in G if and only if
this network has maximum ow strength X.
Problem 7E. For (iv), apply Theorem 7.2, with d = 2, to the
incidence matrix of the graph. (This result is also a quick conse-
quence of Theorem 1.2.) For (v), consider the bipartite incidence
matrix M of a bipartite graph G, which is dened as follows. Say
G has bipartition (X, Y ). The rows of M are indexed by the ver-
tices in X, and the columns of M are indexed by the vertices in Y .
The entry in row x X and column y Y is the number of edges
joining x and y (0 or 1 if G is a simple graph).
548 A Course in Combinatorics
Problem 7F. Let T be a spanning tree in D. Show that for any
assignment of real numbers to the [E(D)[ [V (D)[ +1 edges of D
not in T, there is a unique way to extend this to a circulation f on
D. It may be useful to remember that a tree with at least one edge
has a monovalent vertex.
Problem 8A. Assume that some subsequence of n consecutive
elements occurs twice. Show that satises an equation of degree
less than n.
Problem 8B. The vertices of the digraph are the ordered pairs
from 0, 1, 2; the edges are the 27 ordered triples.
Problem 8C. The sequence must contain four 0s and four 1s.
Three consecutive 0s force 00010111 which does not work; the pair
00 must occur twice and this then forces 00110011 which does not
work either.
Problem 8D. Since every vertex in G
n
has indegree 2 and outde-
gree 2, we do get a closed path from 00 . . . 0 to 00 . . . 0. The nal
edge must come from 10 . . . 0 (which we identify with the integer
1 in binary). The fact that this edge is used implies that the edge
from 1 = 10 . . . 0 to 0 . . . 01 has already been used. So, we came into
the vertex 1 twice, i. e. from 2 = 010 . . . 0 and from 3 = 110 . . . 0.
For the same reason as above, both 2 and 3 were entered twice.
Proceeding by induction, we see that all vertices were visited twice
on the path through G
n
, proving the assertion.
Problem 9A. Apply the algorithm.
Problem 9B. A trivial addressing of length n as in Theorem 9.2
is easily found. What is the diameter of the graph?
Problem 9C. Use Theorems 9.1 and 9.6. Somewhere in the calcu-
lation

k cos(kx) will appear. This is the derivative of

sin(kx)
and that sum is determined by multiplying by sin(
1
2
x).
Problem 9D. See Theorem 9.7.4.
Problem 10A. Use Theorem 10.1, where E
i
(i = 1, 2, 3, 4) are the
integers 1000 divisible by 2,3,5,7, respectively.
Hints and comments on problems 549
Problem 10B. Use inclusion-exclusion and the fact that if f(i) =
0, then (x i) divides f(x).
Problem 10C. Use x| =

kx
1 and Theorem 10.3.
Problem 10D. Multiply

a
n
n
s
and

b
m
m
s
. Determine the
coecient of k
s
, and use Theorem 10.3.
Problem 10E. Determine

d[n
log f
d
(z); use Theorem 10.3 or
10.4.
Problem 10F. There are clearly 2n + 1 colorings. Use inclusion-
exclusion where a set E
i
consists of colorings with i red and i 1
blue. To determine N
j
, use Example 10.6.
For a direct solution, calculate

n=0
n

k=0
(1)
k
_
2n k
k
_
2
2n2k
x
2n
.
To do this, use (10.6) and the fact that

(2n + 1)x
2n
= (
x
1x
2
)
/
.
Problem 10G. Use Theorem 10.1.
Problem 10H. Count permutations of 1, 2, . . . , n that x none of
1, 2, . . . , n k.
Problem 11A. Consider an all-one matrix of size n by k. Adjoin
a column of 0s and then complete to a square matrix by adding
rows of type (0, . . . , 0, 1).
Problem 11B. (i) Theorem 5.3; (ii) Theorem 11.5; (iii) use a
direct product of matrices J
k
.
Problem 11C. Determine A = (a
ij
) with a
ij
= [A
i
B
j
[. To
nd the permanent of this matrix, compare with the probl`eme des
menages.
Problem 11D. Use Theorem 11.7.
Problem 11E. Calculate the corresponding permanent B
n
by ex-
panding by the rst row and column. Show that B
n
= B
n1
+
550 A Course in Combinatorics
B
n2
2. So B
n
2 [ n 3 is a Fibonacci sequence. Also see
Problem 5E.
Problem 12A. Use Theorem 12.1.
Problem 12B. Suppose AA

is decomposable. Explain the block

of 0 entries by considering some row of A, its nonzero entries, and
the inner products with other rows.
Problem 12C. Do four averaging operations on pairs of rows and
similarly for columns. Note that the element a
55
does not change
during these operations. Determine the permanent and minimize.
This shows that the matrix was
1
5
J from the start.
Problem 13A. Use Theorem 13.1.
Problem 13B. (i) Multiply two powers of (1 x). (ii) Let k be
in position a + 1; what can precede this term?
Problem 13C. (i) Choose A
1
and then choose A
2
as a subset
of its complement; determine the sum of products of binomial co-
ecients. (ii) Consider a 2 by n matrix for which the rows are
the characteristic functions of A
1
, respectively A
2
. Translate the
condition on A
1
and A
2
to a condition on this matrix.
Problem 13D. First, x the union U of the sets in /. Apply
Example 13.6 and then sum over all choices of U. The answer
shows that we could have found a better solution. To nd this
better solution, again depict / by a (0,1)-matrix of size k by n.
To this matrix adjoin a special row, namely the characteristic
function of the union of the sets of /. Calculate the number of
these (0,1)-matrices that have a 1 in a specied position in their
special row.
Problem 13E. In each path, consider the last step of type (x, y)
(x + 1, y + 1). Use Theorem 13.1. For the last question, insert k
balls into n+k possible positions; let x
k
be the length of the interval
preceding the k-th ball.
Problem 13F. (i) Prove by induction. If a permutation of 1 to
n 1 is written as a product of cycles, there are n ways to adjoin
Hints and comments on problems 551
the element n, one of which has one more cycle. (ii) In Theorem
13.7, sum both sides over k; use (13.5).
Problem 13G. Insert (13.13) into (13.11) and change the order
of summation.
Problem 13H. Use (10.6).
Problem 13I. Use the relation F(a) = G(a) of page 123.
Problem 13J. Use the binomial formula with x = , where
3
=
1.
Problem 13K. Count partitions of an n-set into n 2 parts di-
rectly.
Problem 14A. (i) The sequences end in 10 or in 1. Establish the
Fibonacci recursion and solve by trying a
n
= t
n
as a solution (use
linearity). The method of Example 14.3 also can be used.
(ii) Express b
n
in terms of the numbers c
n
of such sequences end-
ing in a 1. Find a recurrence for c
n
. The constant c will be the
largest root of a polynomial equation of degree 4; it is approxi-
mately 1.22075.
Problem 14B. Consider a trivalent planted plane tree T. Make
a new tree T
/
as follows. Take as vertices the root of T and all its
vertices of degree 3. Join two vertices in T
/
if there is a path from
one to the other in T using exactly one edge with slope to the left
(see Figure 14.3). Finally, add a new root at the bottom. Show
how to reverse this mapping. Show that the number of vertices of
T
/
equals the number of vertices of degree 1 in T. See p. 25. of J. H.
van Lint Combinatorial Theory Seminar, Eindhoven University of
Technology, Lecture Notes in Mathematics 382, SpringerVerlag,
1974.
Problem 14C. Fix one edge; number others cyclicly: a, b, c, . . . .
Give the chords the obvious names; be careful about the order in
which these names are given. See also p. 25. of Van Lint (1974),
ibid.
Problem 14D. Use the method of Fig. 2.2. One xed point pro-
duces an arborescence; the remainder is a mapping with no xed
points. See (14.15). Take M
1
(0) = 1, A(0) = 0.
552 A Course in Combinatorics
Problem 14E. Fix an edge and then consider the quadrilateral
that it belongs to. The other three sides divide the (n+1)-gon into
possibly empty polygons; see (14.10). Solve f = x+f
3
by applying
Theorem 14.3.
A combinatorial proof that the number of ways of dividing an
(n + 1)-gon into quadrangles by nonintersecting diagonals is 0 if n
is even, and
1
3k+1
_
3k+1
k
_
if n = 2k +1, can be given as follows. Since
the case of even n is trivial, we assume n = 2k +1. There are k 1
diagonals and k quadrangles in the dissection. Put a vertex in each
quadrangle and on each edge; x one edge-vertex as a root. Vertices
of quadrangles are joined by an edge if the quadrangles share a
diagonal; an edge-vertex is joined to the quadrangle-vertex of its
quadrangle. Describe a walk around the tree, that was just dened,
by a sequence of symbols x and y, where we use an x every time we
reach a quadrangle vertex not visited before, and similarly a y for
edge-vertices. We nd a sequence of k symbols x and 2k+1 symbols
y. We know, see Example 10.6, that there are
1
3k+1
_
3k+1
k
_
circular
sequences of this type. It remains to show a one-one correspondence
between these circular sequences and the sequences describing the
trees. Describe the x, y sequence as follows: an x corresponds to a
step in the X-Y plane one unit to the right and one unit upwards;
a y to a step one unit to the right and k/(2k +1) steps downward.
The walk starts in (0,0) and ends in (3k + 1, 0), never meeting
the X-axis in between. If we describe a circular sequence of the
given type in the same way, it will have a unique minimum because
(k, 2k +1) = 1. This will give us the unique starting point to make
it correspond to one of the trees.
Problem 14F. See Appendix 2 on formal operations. Take loga-
rithms and use Theorem 10.3.
Problem 14G. Show equivalence with Example 14.11.
Problem 14H. Make a path as in Example 14.8 and nd a lowest
point.
Problem 14I. (i) See (14.10). (ii) Put a vertex on the circle, say
between 1 and 2n. This will be the root of a tree. Put a vertex in
each region and join vertices in the obvious way. Show equivalence
with one of the Catalan problems. See Van Lint (1974), p.26.
Hints and comments on problems 553
Problem 14J. A labeled regular graph of valency 2 is a union of
(labeled) polygons. Apply Theorem 14.2.
Problem 14K. Call the number of walks A
4n
and dene A(z) :=
1 +

n=1
A
4n
z
4n
. If B
2n
is the number of walks from (0,0) to (n, n)
that avoid points (i, i), then we know from (14.12) that B(z) :=

n=1
B
2n
z
2n
= 1

1 4z
2
. Find a relation between A(z) and
B(z) by considering the rst crossing of x = y for each walk.
It is also possible to prove the result by nding a one to one
mapping from the walks counted by A
4n
to the walks counted in
Example 14.8. This is dicult. For a solution see the solution to
Problem 3096 by W. Nichols in American Mathematical Monthly
94 (1987).
Problem 14L. Show that a
n
= 3a
n1
a
n2
and that F
2n
satises
the same recurrence.
Problem 14M. For g with k xed points, count their number and
the number of appropriate f and sum over k; for f with image of
size i, split 1, 2, . . . , r into preimages, count the f and the number
of appropriate g; sum over i. Use (13.9) and (13.11). This is the so-
lution to Problem 3057 of the American Mathematical Monthly 94
(1988), given by J. M. Freeman, S. C. Locke, and H. Niederhausen.
Problem 14N. Introduce b
k,n
:= the number of walks from (0,0)
to (n, n) that meet the line x = y for the second time in (k, k).
Problem 15A. Distinguish between x
k
= 1 and x
k
> 1, in which
case take y
i
= x
i
1. Use induction.
Problem 15B. Let E, I, and S denote the number of equilateral,
isosceles, and scalene triangles, respectively. Calculate E, I + E,
and express
_
n
3
_
in E, I, and S.
(This proof is due to J. S. Frame, Amer. Math. Monthly 47,
(1940), 664.)
Problem 15C. Use partial fractions. Find c using the coecient
of (1 x)
t
and (10.6).
Problem 15D. Count the compositions of n into k parts which
have a 3 in position j. Do this for all j and sum; (a composition
with m parts 3 is thus counted m times).
554 A Course in Combinatorics
A second solution with a more general result is as follows. First
show that in a list of all possible compositions there are exactly
(n + 1) 2
n2
integers. This can be done using the argument with
the red balls. Then prove by induction on n that for 1 m < n
a list of the 2
n1
compositions of n contains the integer m exactly
(n m + 3)2
nm2
times. For this, use the formula 1 +

n1
k=1
(k +
3)2
k2
= (n + 1) 2
n2
which is also shown by induction.
Problem 15E. Leave out the rst column.
Problem 15F. For the unequal odd parts, use hook shaped g-
ures to form a Ferrers diagram and read this in the usual way.
Problem 15G. Show a correspondence with Example 14.8.
Problem 15H. See Chapter 13.
Problem 15I. Generalize Theorem 15.4.
Problem 15J. Apply Theorem 15.10 to an n n square.
Problem 16A. Use induction; compare two initial segments of r
and s.
Problem 16B. Determine the rst column of a matrix with row-
sum r and column-sum r

. Use induction.
Problem 16C. Use Problem 16A and switch edges. Another pos-
sibility is to describe the graph by its adjacency matrix and use the
same idea as in the proof of Theorem 16.2.
Problem 16D. Estimate the number of solutions of s
1
+ +s
n
=
1
2
n
2
by just using the fact that each s
i
is at most n.
Problem 16E. For the if part, rst note that there exists a tour-
nament with degree sequence (n1, . . . , 2, 1, 0). Then use Problem
16A. Show that if in a tournament the outdegree of x is greater
than the outdegree of y, then there is a tournament on the same
vertex set and with the same outdegrees except that outdegree of x
is one less than it was originally, and the outdegree of y is 1 greater.
One can do this by changing the directions on at most two edges.
Problem 16F. For the generalization, one considers the hyper-
graph given by choosing the rst m of the k-subsets in lexicograph-
ical order. For the if parts, again use Problem 16A.
Hints and comments on problems 555
Problem 16G. A(5, 3) = A(5, 2). See (16.2). Counting directly is
possible as follows: (i) Fix the rst row as (1, 1, 0, 0, 0); this means
that the nal result must be multiplied by 10. (ii) Put a 1 in
position (2,1); this means that we should also multiply by 4. (iii)
Split into two cases: a 1 in position (2,2), etc. or a 1 in position
(2,3), in which case a factor 3 is introduced, etc.
Another solution is obtained by observing that /

(5, 2) has car-

dinality
1
2
5!4! = 1440, and that there are (
_
5
2
_
)
2
3! = 600 decom-
posable matrices in /(5, 2).
Problem 16H. In case 1, there are n1 possible choices. In case
2, remove row 1 and column 1, substitute an element of /(n1, 2)
and in two ways then produce an element of /(n, 2).
Problem 16I. Straightforward substitution.
Problem 17A. Index the rst two rows by x and y, the rst two
columns by u and v, all under the assumption that we have the
multiplication table of a group of order 5. Show that (v
1
u)
2
= 1.
Problem 17B. This is already dicult, despite the small size of
the square! Take a = 1. Note that the 1 in the upper left-hand
corner is in subsquares of type
_
1 i
i 1
_
for i = 2, 3, 4, 5. Consider
any Latin square with this property. If the (2,3) entry is not 4,
then permuting rows 4 and 5, columns 4 and 5, and symbols 4 and
5, changes the square into an equivalent one with a 4 in position
(2,3). At this point, there is only one way to complete the square.
So, if a Latin square has any cell with a property like the one we
mentioned for the upper left hand 1, then it is equivalent to this
one.
Using the same kind of argument one can show that there are
only a few possibly inequivalent squares of order 5. For some, one
then has to establish the equivalence, a tedious task.
Problem 17C. At this point, trial and error is prescribed. How-
ever, see Chapter 22.
Problem 17D. Suciency is shown by an example (you can, but
do not have to, use Theorem 17.1). For necessity, observe that just
556 A Course in Combinatorics
appending one more column to a Latin square of order m requires
m new symbols.
Problem 17E. Apply the algorithm.
Problem 17F. For part (i), describe an algorithm that recovers
the original Latin square of order n from the constructed Latin
square of order n+1 with constant back diagonal. For (ii), explain
why the number of Latin squares of order n with ns on the back
diagonal is equal to N(n)/n!.
Problem 17G. (a) For n = 6, an example is 6, 1, 5, 2, 4, 3.
(b) Add the dierences.
(c) Assume two pairs are equal and solve.
Problem 17H. Start by replacing the diagonal entries of the
smaller square by (n + 1)s.
Problem 17I. See Example 10.6.
Problem 17J. (i) Use inclusion exclusion. If k integers are in the
wrong place, write k = 2i +j where there are i pairs in the wrong
place, etc.
(ii) Use J I of size n.
Problem 18A. To show this directly, argue as in Theorem 18.1
concerning the rst three rows. Show that no further row can start
with three +1s. Hence, rows 5,6, and 7 all start with + +
(why?). These can then be completed in a unique way, etc. For
another solution, see Problem 19D.
Problem 18B. Use the method of Theorem 18.1. Note that a
permutation of rows and columns, that preserves the 0-diagonal,
does not help us to make a matrix symmetric or antisymmetric.
So, only multiplications by 1 will do the job. Normalize the rst
row and make the rst column (except for the 0th entry) all 1 or
all 1. It is now sucient to show that the (2,3) entry and the
(3,2) entry are the same, respectively opposites, whenever n 2
(mod 4), respectively n 0 (mod 4).
Problem 18C. A trivial exercise!
Problem 18D. We have 28 = 1+9+9+9 and 28 = 1+1+1+25.
Use the rst of these. Note that each W
i
has 1 on the diagonal
Hints and comments on problems 557
and if it has w
ij
= 2, then w
7i,j
is also 2; similarly for 2. The
observation about occurrences of the U
i
in the matrices W
j
then
completely forces the choice of the four matrices W
i
. They prove
to work.
Problem 18E. Show by induction that M
(i)
n
. . . M
(m)
n
= H
2
mi+1
I
2
i1. For the second part, count the number of non-zero elements
of M
(i)
n
. See the reference E. C. Posner (1968).
Problem 18F. For both questions use induction.
Problem 18G. For (2), show that the number of rows with rst
coordinate 0 (or 1) is at most n.
Problem 18H. Straightforward; see Theorem 18.4.
Problem 18I. Review the properties of Q.
Problem 19A. For the isomorphism in part (i): Label the vertices
of K
6
with 0000, 1000, 0100, 0010, 0001, 1111 in F
4
2
and label an
edge with the sum in F
4
2
of the labels on its ends. This gives a one-
to-one correspondence between the edges of K
6
and the 15 nonzero
vectors in F
4
2
.
For part (ii), consider congurations consisting of three edges
(there are four types). The construction is due to E. S. Kramer
(1990), Discrete Math. 81, 223224.
Problem 19B. Calculate b
i
, i = 3, 2, 1, 0; b
1
/ Z.
Problem 19C. This is Theorem 19.4.
Problem 19D. (i) Let N have column-sums c
i
. Then

c
i
=
66. The sum of all inner products of rows is
_
c
i
2
_
. Use the
Cauchy-Schwarz inequality. Conclude that all column-sums are 6
and all inner products are 3. (ii) To show uniqueness, consider the
complement: a 2-(11,5,2) design. Without loss of generality, the
rst row is (11111000000); the remaining ten rows start with the
_
5
2
_
pairs (110 . . . ), (1010 . . . ), etc. Then show that the rows 2 to
5 can be completed in one way. This leaves a 6 by 6 square to be
completed. At rst sight, there seem to be two ways to do this.
Find a permutation that maps one into the other.
558 A Course in Combinatorics
Problem 19E. Denote by a
i
the number of blocks that meet B in
i points. Calculate

a
i
,

ia
i
, and
_
i
2
_
a
i
. Derive an inequality
for a
0
using the form

(i c)
2
a
i
. Compare with Theorem 19.9.
Problem 19F. The mapping x 2x has six orbits, two of which
are too long to use (e.g. (1,2,4,8,16,11)). See Example 28.2.
Problem 19G. Use the corresponding orthogonal array. Use the
common 0, 1 to 1, 1 mapping.
Problem 19H. For (1) use (19.2) or (19.4). For (3) count pairs
p, q / B that are contained in blocks that meet p. Thus nd the
parameters of T
B
. For (4), rst suppose that T
B
is a single block.
Consider Example 19.3. In the other case Fishers inequality yields
k ( + 1)( + 2). Now, express b
0
as a fraction in k and and
deduce that k must divide 2( + 1)( + 2).
Problem 19I. (i) Fix a point x on O. Count pairs (y, L), y O,
L a line through x and y. (ii) From (i) we see that if [O[ = n + 2,
then every line that meets O, meets it twice. Take a point z / O
and count the lines through z that meet O. (iii) Take any four
points, no three on a line. These determine six lines, that together
contain 19 points of the plane.
Problem 19J. For p
1
to p
5
take (1,0,0), (0,1,0), (0,0,1), (1,1,1)
and (1, t, t
2
) and show that there are two ways to group these to
produce p
6
= (1, x, x
2
), where x = t/(1 +t).
Problem 19K. Let B
1
B
2
. How many blocks meet B
1
? Take
x B
1
, y B
2
; how many blocks meet both B
1
and B
2
? There are
n +1 equivalence classes, each with n lines (blocks). To each class
append a point that is to be added as an extra point to each line
of that class, etc.
Problem 19L. Use Theorem 19.11. Consider the equation mod-
ulo 3.
Problem 19M. For v even, use quadratic forms, as in the proof
of Theorem 19.11. See J. H. van Lint and J. J. Seidel (1966),
Equilateral point sets in elliptic geometry, Proc. Kon. Nederl. Akad.
Wetensch. 69, 335348.
Hints and comments on problems 559
Problem 19N. (a) The method of Example 19.14 also works if
[V [ = 1; (85 = 7(13 1) + 1). (b) 15 = 7(3 1) + 1.
Problem 19O. Consider the n b incidence matrix N, where
b = [/[. Check that in case (i), N

N is nonsingular modulo 2. For

part (ii), let N
1
be obtained by adding column of all 1s to N, and
show that N

1
N
1
has rank at least b modulo 2.
Problem 19P. The last of the displayed sums should evaluate to
zero. This implies each
i
is 1 or 2.
Problem 19Q. Use induction; start the induction with t = 2.
Count, in two ways, ordered triples (x, A, B) where A and B are
distinct blocks, both of which contain the point x.
Problem 19R. Represent lines as in Example 19.6 and use the
fact that this is a cyclic representation. Distinguish dierent cases
by the number of rst elements used in the SDR.
Problem 19S. It is possible to give a construction using induction
on k, but it is quicker to give a direct construction using F
k
2
as the
point set. Review Example 19.1.
Problem 19T. Use the result of Problem 19H. The extensions
will appear in Chapter 20.
Problem 19U. Apply Theorem 5.5.
Problem 20A. Use (20.5).
Problem 20B. Apply the pigeonhole principle to the last two bits
of the words. A binary code with three words is easily handled!
Problem 20C. For each of the three ways in which the conditions
for a codeword can be violated, change a suitable coordinate to
remedy the situation.
Problem 20D. (i) The rows of the generator matrix correspond
to a
i
, i = 1, . . . , n, where the a
i
are the rows of a Hadamard
matrix. Let u F
n
2
, where n = 2
2k
. By u

we denote the 1
representation. Calculate

u

, a
i
)
2
to nd the upper bound.
(ii) For equality, rst nd the weight of z. Then use the fact that
the codewords correspond to linear functions and adding a linear
function to x yields an equivalent quadratic form.
560 A Course in Combinatorics
Problem 20E. (i) See Problem 20A. (ii) Count words of weight 5
covered by codewords of weight 7; the other words of weight 5 are
covered by codewords of weight 8. Using this idea, set up a system
of linear equations for the A
i
.
Problem 20F. A codeword at distance e from a word of weight
e + 1 must have weight 2e + 1.
Problem 20G. The codewords in the dual of the binary Hamming
code of length 2
r
1 may be identied with linear functionals on
F
r
2
(cf. the geometric description of Reed-Muller codes in Chapter
18); each nonzero codeword has weight 2
r1
.
Problem 20H. Use (19.6) and the fact that G
24
is linear and has
minimum weight 8; note that there are
_
21
2
_
pairs of lines. If = 1,
consider the 7-point conguration B

in the plane. Any line that

meets B

must meet it in 1 point or in 3 points. How many meet

B

in 3 points?
Problem 20I. (i) Find the standard generator matrix for C. (ii)
The conditions (1) and (2) clearly dene a linear code. In (1)
we have two choices for parity; in (2) we have 4
3
choices for the
codeword; in ve of the columns of A we then have two possibilities
(not in the last one!). For the choice even, it is clear that a
nonzero codeword in C forces weight 8; 0 has to be treated
separately. For the choice odd, it is clear that we obtain a weight
of at least 6; condition (1) ensures that equality cannot hold. For
more on this wonderful description of the Golay code and many
consequences see: J. H. Conway, The Golay codes and the Mathieu
groups, Chapter 11 in J. H. Conway and N. J. A. Sloane, Sphere
Packings, Lattices and Groups, Springer-Verlag, 1988.
Problem 20J. Self-duality is done as in Example 20.6. All weights
in Sym
12
are divisible by 3. Consider linear combinations of two
rows of the generator matrix and then conclude what happens if
we combine with a third row. Use (20.5).
Problem 20K. Prove the generalization by induction on r. For
each r, nd such a matrix none of whose columns is a constant
vector. For the induction step, concatenate three copies of the
matrix with r rows, to which have been appended a row of all 0s,
Hints and comments on problems 561
all 1s, and all 2s, respectively, and then add three appropriate
columns.
Problem 20L. The hint in the chapter should be sucient.
Problem 21A. The argument is the same as for (21.4), now with
inequalities.
Problem 21B. Use the integrality condition. Also see the refer-
ence A. J. Homan and R. R. Singleton (1960) from Chapter 4.
Problem 21C. Use (19.2); see Problem 20F. If two blocks are
disjoint, count the number of blocks that meet both of them, each
in two points. How many meet only one of them? Similar counting
arguments (distinguish a few cases!) are used to show that = 6.
Problem 21D. Suppose such a graph exists. Pick a vertex x and
consider the sets (x) and (x). We consider these as the points
and blocks of a design. Show that this is a 2-(9,4,3) design. Next,
x a block of this design and let a
i
be the number of blocks that
meet this block in i points. Show that a
0
1 (see Problem 19E)
and show that the fact that (x) has degree 5 implies that a
0
should be at least 5.
Problem 21E. The rst two questions are straightforward. For
the third we refer to textbooks on matrix theory (see interlacing)
or to Chapter 31. The theorem we are alluding to states that the
eigenvalues of a principal submatrix B of a symmetric matrix A are
bounded by the largest and the smallest eigenvalue of the matrix A.
To prove this, just consider any vector x and calculate x

Ax/x

x.
Specialize by letting x be an eigenvector of B with 0s appended.
Problem 21F. Review the proof of Theorem 21.5. Show that
(x) = K
n1,n1
. Dene a grand clique to be a clique of size n and
show that any edge is in exactly one grand clique. Show that there
are exactly 2n grand cliques, which fall into two classes of n so that
grand cliques in the same class are disjoint. See S. S. Shrikhande,
The uniqueness of the L
2
association scheme, Ann. Math. Stat. 30
(1959), 781798.
Problem 21G. Count ags and other congurations, e.g. two in-
tersecting lines and a point on one of them.
562 A Course in Combinatorics
Problem 21H. See Problem 21E for v; k is obvious. Take two
points on line L. Points joined to both of them are on L or on one
of the other R 1 lines through one of them. A similar argument
yields . For r and s see (21.6).
Problem 21I. On (x) use the fact that = 1. Then consider
x (x).
Problem 21J. Eq. (21.8) leads to an equation like (21.6) with t
instead of k. Although A is not symmetric the eigenvectors for
eigenvalue ,= k are orthogonal to j.
Problem 21K. Show that if ( )
2
+ 4(t ) is not a square,
then the eigenvalues have multiplicities 1,f, and g, where f = g =
1
2
(v 1). Then show that A = A

= J I and AA

= lJ +(l +1)I,
where v = 4l + 3. Dene Q := AA

.
Problem 21L. Let d :=
_
( )
2
+ 4(t ). Apply (21.8) to j
and then show that d divides (k 1)
2
. Using 4k 7 = d
2
, prove
that d must divide 9.
For v = 18, use the matrix
_
cI (c 1)I +J
(c 1)I +J cI
_
and replace c by the matrix C of Example 21.11 and d by C
2
to
obtain a matrix A of size 18 18.
Problem 21M. (i) Consider (x) (y) where (xy) is an edge.
(ii) Prove that G is also neighborhood regular.
Problem 21N. (i) (x
1
) = C B is regular with degree a, so
[CB[ = a[C[ [CC[, etc.
(ii) Eliminate [BC[ and [BD[ from the relations in (i).
(iii) Interchange x
1
and x
2
. Use d
1
,= d
2
.
(iv) Express d
i
in k
i
, a, a, and the number of vertices n of G.
Then prove that k
1
+k
2
= 2n 3a a 5.
Problem 21O. Part (i) is straightforward. For example, =
(n + 3)/2. (ii) If four triples of a clique C contain a point x, then
all triples of the clique must contain x. (iii) If n > 15, the cliques
Hints and comments on problems 563
of size (n 1)/2 in the graph of a triple system are in one-to-one
correspondence with the points of the triple system.
Problem 21P. Equality would imply the existence of a srg(a
2
+
a + 1, a + 1, 1, 1). Use Theorem 21.1. Only a = 1 is possible and a
K
3
obviously has no circuit on four vertices.
Problem 21Q. The graph G representing the friendship relation
is either a union of triangles with one vertex common to all of them
or G is regular. Use Theorem 21.1.
Problem 22A. Show, by computation, that the complement of
the graph of an (n, n)-net has the parameters of the graph of an
(n, 1)-net. Then show that it is the graph of an (n, 1)-net, i.e. there
is a partition of the points into n sets of size n, no two points of
which are collinear.
(More generally, an (n, r)-net may be completed to an (n, n+1)-
net if and only if the complement of the graph of the (n, r)-net is
the graph of an (n, n + 1 r)-net.)
Problem 22B. Use the constructions from elds described before
the problem. Take S(x, y) := (x +y)/2 for q odd, for example.
For even q, we can use A and S to schedule mixed-doubles tennis
matches. Suppose we have q couplesMrs. i and Mr. i for i
F
q
. Let Mrs. i and Mrs. j play against each other with respective
partners Mr. A(i, j) and Mr. A(j, i) during round S(i, j). There
will be q 1 rounds, labeled by the o-diagonal symbols in S,
during which everyone is playing in exactly one match. It will be
found that no one meets her or his spouse as partner or opponent
the entire tournament, but meets everyone else exactly once as an
opponent, and has everyone of the opposite sex exactly once as a
partner. For q odd, we can have q rounds in which one couple sits
out each round.
Problem 22C. Show that m, m+1, t and u each have the property
that for each prime p < x, they are either prime to p or divisible
by p
x
.
Problem 22D. Review the connection between sets of pairwise
orthogonal Latin squares and transversal designs.
564 A Course in Combinatorics
Problem 22E. Use Theorem 22.6.
Problem 22F. Columns of A consist of integers between 1 and k.
For every column of A and of S replace an integer i in the column
of A by the i-th element in the column of S. By adding suitable
columns and one row, complete to an OA(v, c + 1).
Problem 22G. The proof is similar to the proof of Theorem 22.6.
One now also needs a (1, /, T), a TD(v, k).
Problem 22H. Use Problem 22F.
Problem 22I. Use Problem 22G.
Problem 22J. Use Problem 22G.
Problem 23A. Note that the closure of a set S of edges consists
of all edges that have both ends in the same connected component
of the spanning subgraph of G with edge set S.
Problem 23B. The lines of AG
r
(2) have size 2.
Problem 23C. Dispose rst of the cases when the union of two
lines is the whole point set. Then show how to set up a one-to-one
correspondence between the points of any two lines.
Problem 23D. The hardest part is showing that T = F
1
F
2
:
F
1
T
1
, F
2
T
2
. Prove that F
1
F
2
= F
1
F
2
using the fact that
the rank of F
1
F
2
is the sum of the ranks of F
1
and F
2
.
See H. Crapo and G.-C. Rota (1970) for a complete discussion
of connectedness of combinatorial geometries and irreducibility of
geometric lattices.
Problem 23E. Suppose A F = and rank(A) + rank(F) =
rank(A F). Show that these equations remain valid if A is re-
placed by A
/
:= A x where x is any point not in A F.
Problem 23F. Use (23.4).
Problem 23G. We may work in PG
2
(F) and assume without loss
of generality that L
1
= [1, 0, 0] and L
2
= [0, 1, 0]. Then (explain
why) a
1
= 0,
1
, 1), b
1
= 0,
1
, 1), and c
1
= 0,
1
, 1), while a
2
=

2
, 0, 1), b
2
=
2
, 0, 1), and c
2
=
2
, 0, 1).
Hints and comments on problems 565
Problem 23H. See Chapter 13.
Problem 23I. Just use the denition of .
Problem 24A. See Chapter 13.
Problem 24B. One answer is e
i
= i(m k + i). Consider k by
n + m matrices of rank k that are in echelon form. How many of
these have i leading 1s in the rst n columns?
Problem 24C. Check that x and y are contained in a coset of a
subspace U if and only if x y is contained in U.
Problem 24D. Count all n n matrices.
Problem 24E. See Theorem 6.4.
Problem 25A. For a subspace U of the r-dimensional space W,
let f(U) be the number of k-subspaces which intersect W in U.
Use (25.5).
Problem 25B. For a subspace U, let f(U) denote the number of
nonsingular mappings whose set of xed vectors is (exactly) U, and
h(U) the number of nonsingular linear mappings that x (at least)
the vectors in U. It is easy to give a formula for h(U) in terms of
n, q and the dimension of U. We require a formula for f(0).
Problem 25C. Use (25.2) to calculate (0, x) for x of rank 1 or
2.
Problem 25D. There are only n 1 partitions x ,= 0
L
(each
consisting of one block of size 2 and n 2 singletons) that satisfy
x a = 0
L
.
Problem 25E. For x < z < y a chain of length k from x to z
extends to a chain of length k + 1 from x to y.
Problem 25F. See Theorem 25.1 (iii). Count injective mappings
from V to S and then vary [S[.
Problem 26A. The number of lines on a point x not in A which
meet A is a constant, i.e. independent of x.
Problem 26B. Any line disjoint from the arc can be used to ob-
tain a partition of the blocks into parallel classes, one parallel class
for each point of the line.
566 A Course in Combinatorics
Problem 26C. The point of intersection of the line though 1, 0, 0)
and 0, 1, 0) and the line though 0, 0, 1) and 1, 1, 1), for example,
is 1, 1, 0).
Problem 26D. Let
i
be the cardinality of the intersection of the
i-th line with S. The inequality is easy if some
i


n + 1, so
assume otherwise and consider

i
(
i
1)(
i

n 1).
Problem 26E. Whether the characteristic is even or odd, f is
degenerate if and only if x(C +C

) = 0 for some x with f(x) = 0.

To nd such an x when f is degenerate, take the last row of a
nonsingular matrix A such that ACA

has a last row and column

of all zeros, which exists because x
n
, say, does not occur in some
projectively equivalent form. Also see J. W. P. Hirschfeld (1979).
Problem 26F. By Theorem 26.6 with q = 2 and n = 2m 1,
nondegenerate quadratic forms in 2m variables have either 2
2m1

2
m1
or 2
2m1
+ 2
m1
zeros, where we are counting the zero vector
here. So the codewords (vectors of length 2
m
) corresponding to such
forms have weight 2
2m1
+2
m1
or 2
2m1
2
m1
. Since, as functions
over F
2
, x
2
i
= x
i
, we may think of the polynomials f(x) + a(x) as
quadratic forms, or their complements when there is a constant
term 1. Show that all these forms are nondegenerate.
Alternatively, show that there is an invertible ane substitution
(where e.g. x
i
may be replaced by x
i
+1) that takes f(x) +a(x) as
a function on F
2m
2
, to f(x) or f(x) + 1.
Problem 26G. It is relatively straightforward to show that Q
/
is
a nondegenerate quadric in W. There is nothing more to do if n is
odd. Theorem 26.5 can be used to complete the proof for n even.
It is easy to see that if Q
/
is hyperbolic, then so is Q. Next use
equation (26.4) to show that every point p on a hyperbolic quadric
Q in PG
n1
(q) is contained in a at F Q of projective dimension
n/2 1. Check that such a at is contained in T
p
so that the at
F W Q
/
has dimension one less than that of F.
Problem 26H. If point x is not on line , consider the plane
determined by x and .
Problem 26I. The points of a triangle are on a line. For (iii) see
Chapter 21.
Hints and comments on problems 567
Problem 26J. See the argument for arcs in projective planes.
Problem 26K. Consider three points. Let x
i
be the number of
blocks of the complementary design that contain i of these points.
Use the usual counting argument to show that x
0
+x
3
= .
Problem 27A. It is not dicult to see that the even numbered
items are equivalent to each other, and similarly for the odd num-
bered items. To show (3) and (4) are equivalent, for example,
introduce an incidence matrix N and show that one is equivalent
to the matrix equation NN

= (k )I + J, and the other to

N

N = (k )I +J.
Problem 27B. View (Z
2
)
4
as a vector space over the eld F
2
. We
may assume the dierence set contains the zero vector. It will span
(Z
2
)
4
and so contains a basis.
Problem 27C. Count separately the number of times a nonzero
element of GH can be written as a dierence of two elements of
A(H B), a dierence of two elements of (G A) B, and as a
dierence of one element of each of these sets.
Problem 27D. Multiply the sum of the elements, that we want
to show is 0, by any square.
Problem 27E. We already said that this problem is similar to the
illustrated calculation of
2
.
Problem 27F. Use Theorem 27.5. (A table of cyclic dierence
sets may be found in L. D. Baumert, Cyclic Dierence Sets, Lecture
Notes in Math. 182, Springer-Verlag, 1971.)
Problem 27G. A zero of the polynomial y
3
+3y+2 (coecients
in F
7
) is a primitive element of F
7
3 .
Problem 27H. The cardinality of the intersection of a translate
D + g of D with the set D is the number of times g arises as a
sum from D.
Problem 27I. For distinct i and j, the dierences x y, x
U
i
, y U
j
, comprise all vectors in V , each exactly once.
568 A Course in Combinatorics
Problem 28A. The Frobenius automorphism of F
p
t(n+1) is an
automorphism of the symmetric design of 1-dimensional and n-
dimensional subspaces of F
p
s(n+1) over F
p
t .
Problem 28B. We may take S(x) as in Lemma 28.4 with D =
1, 2, 4 but where is not a multiplier.
Problem 28C. If n 0 (mod 10), say, then for any x D,
x, 2x, 4x, 5x D and the dierence x is seen to occur twice
unless 3x = 0. If 3x = 0 for all x D, then 3x = 0 for all x in
the group, and then n
2
+ n + 1 must be a power of 3. Show that
n
2
+n + 1 is never divisible by 9.
Problem 28D. The numbers of normalized dierence sets with
these parameters are, respectively: 2, 2, 4, 2, 0, 4.
Problem 28E. There are two normalized (15, 7, 3) dierence sets;
they are equivalent since one consists of the negatives of the ele-
ments of the other. There are no normalized dierence sets for any
of the other parameters. To prove this in each case, rst nd a
multiplier by Theorem 28.7.
For e.g. (v, k, ) = (25, 9, 3), Example 28.3 shows 2 is a multiplier
of a hypothetical normalized dierence set D. If the group is cyclic,
D is the union of cycles of x 2x on Z
25
which are
0, 5, 10, 20, 15 and Z
25
0, 5, 10, 15, 20;
but no such union has nine elements. If the group is elementary
abelian (exponent 5), then D must consist of 0 and two cycles of
the form x, 2x, 4x, 8x = 3x; but the dierence x occurs ve times
already among 0 and the elements of such a cycle.
Problem 28F. Assume D(x
1
) = D(x) and use Lemma 28.2 with
p = 2.
Problem 28G. Recall the construction of a symmetric design
with those parameters from a Latin square of order 6 in Problem
19G.
Problem 28H. Assume D(x
q
) = D(x). Note that q
3
1
(mod [H[). Show D(x

) 0 (mod q, H) and conclude that one

of the coecients of D(x

) is q + 1 and all others are 1.

Hints and comments on problems 569
Problem 28I. The rst part is similar to Problem 28H. Then
show that if, in PG
3
(q), a line meets a set S of size q
2
+1 in k 3
points, then one of the q +1 planes on that line contains more than
q + 1 points of S.
Problem 29A. Review the proof of Theorem 20.6.
Problem 29B. Review the latter part of the proof of Proposition
29.5.
Problem 29C. The answer is (v, k, ) = (4n 1, 2n 1, n 1)
or (4n 1, 2n, n), i.e. Hadamard or the complement of Hadamard.
Review the proof of Theorem 29.7.
Problem 30A. A
1
A
2
= 8A
1
+8A
2
+9A
3
and A
3
1
= 90A
0
+83A
1
+
56A
2
+ 36A
3
.
Problem 30B. By (30.3), A
i
A
j
=

k
=0
P
i
()P
j
()E

. Then by
(30.4),
N A
i
A
j
=
k

=0
P
i
()P
j
()
k

=0
Q

()A

,
so evidently p

ij
=
1
N

k
=0
P
i
()P
j
()Q

(). The second eigenma-

trix Q can be calculated by inverting P (also see Theorem 30.2).
Problem 30C. For the Latin square graphs,
P =
_
1 (n1)r (n1)(nr+1)
1 nr n+r1
1 r r1
_
.
Problem 30D. To evaluate the entries P

(0) of the top row of

P, take the inner product (as introduced in the proof of Theorem
30.2) of both sides of (30.3) with E
0
.
Problem 30E. This is the dual to Problem 30B.
Problem 30F. Try [A[ = [B[ = 3.
Problem 30G. Note that for any automorphism, [A(B)[ 1.
Count the number of automorphisms of G.
Problem 30H. Given a metric scheme arising from a distance
regular graph G, rst note that A
i
1
is a linear combination of
570 A Course in Combinatorics
A
0
, A
1
, . . . , A
i
where the coecient of A
i
is positive. Given a P-
polynomial scheme, dene a graph G by saying that x and y are
adjacent in G if and only if they are 1st associates in the scheme,
and show that two points are i-th associates if and only if they are
at distance i in G.
Problem 30I. Use the counting argument of Problem 20K for
words of weight 8 rst; then nd the other words satisfying the
condition. Use the subcodes with a xed j and the subcode of
Problem 20K.
Problem 30J. If such a code did exist, it would be equivalent to
a code with a generator matrix with (1111100 . . . 0) as rst row and
below that (AB), where B generates a code of length 8, dimension
5, and minimum distance at least 3. Now use the fact that the
[7,4,3] Hamming code is unique (up to equivalence).
Problem 30K. (i) Distance depends on the size of the intersec-
tion only; just count; (ii) The graph has degree 4 and two triples
corresponding to the Fano plane meet in one point.
Problem 30L. For J(k, v) show that a necessary condition is that
1 +k(v k) divide
_
v
k
_
.
Problem 31A. Consider the intersection of the eigenspaces of A
1
and A
2
corresponding to the eigenvalue 1.
Problem 31B. Since the adjacency matrix of the complement of
a graph G with adjacency matrix A is J I A, the eigenvalues
of G are easily determined from those of G. The rst examples in
Chapter 21 are instances in which equality can occur.
Problem 31C. (i) The Petersen graph has 3-claws, i.e. induced
subgraphs on four vertices consisting of three edges incident with
a common vertex (and no other edges). Explain why a line graph
cannot contain a 3-claw. (ii) The matrix N

N is positive semidef-
inite and is equal to 2I +A, where A is the adjacency matrix.
Problem 31D. From the equation following (31.1) with x the
vector of all 1s, it follows that the largest eigenvalue of a graph is
bounded below by the minimum degree of the graph. By Lemma
31.5, the largest eigenvalue of a graph G is bounded below by the
Hints and comments on problems 571
minimum degree of any induced subgraph H of G. If (H) = (G)
and the deletion of any vertex x of H decreases the chromatic
number, then deg(x) (H) 1.
Problem 31E. If there are no directed paths from x to y in a
digraph G with adjacency matrix A, let S be the set of vertices
z such that there does exist a directed path in G from x to z.
Then A(x, y) = 0 whenever a S and b V (G) S, so A is not
irreducible. The converse is even easier.
Problem 31F. (i) Show by straightforward calculation that u
j
A =

j
u
j
, where u
j
= (1,
j
,
2j
, . . . ,
(n1)j
). For (ii), dene, for each
character , a vector u

whose coordinates are indexed by G and

where u

(g) = (g), and show that u

A =

.
Problem 31G. To prove the lemma, show that if x is a row-vector
so that xS = 0, then xa = 0. Apply the lemma to S = A + I,
where A is the adjacency matrix of the graph. The lemma may
be found in B. Bagchi and N. S. Narasimha Sastry, Even order
inversive planes, generalized quadrangles and codes, Geometriae
Dedicata 22 (1987), 137147. The graph problem (phrased in terms
of acquaintances) is Problem 10851, by D. Beckworth, from the
American Mathematical Monthly 108.
Problem 31H. Most of the work has been done in the proof of
Theorem 31.10: If the distinct eigenvalues of G are its degree d and

i
, i = 1, 2, then (A
1
I)(A
2
I) =
1
v
(d
1
)(d
2
)J where A
is the adjacency matrix of G. This means that A
2
is in the algebra
generated by I, J, and A. So G is strongly regular.
Problem 31I. Let G be an srg(v, k, , ). The eigenvalues of
(x) interlace the eigenvalues
k, r, r, . . . , r, s, s, . . . , s (r > s, say)
of G. Any sequence that interlaces this can have at most one term
which is greater than r.
Except in the half-case, r and s are integers with rs = k; in
particular, r k (even in the half-case). The graph (x) is
572 A Course in Combinatorics
regular of degree k . Were it not connected, k would be an
eigenvalue of (x) of multiplicity greater than one.
Problem 31J. The eigenspace of A corresponding to eigenvalue 1,
and the eigenspace of B corresponding to 1, both have dimension
5, but are contained in the 9-dimensional space of vectors orthogo-
nal to the vector of all 1s, and so there is a nonzero vector u in the
intersection of the eigenspaces. Show that connectivity of K leads
to a contradiction to Theorem 31.11.
Problem 31K. (i) This follows from Theorem 31.13 or part of its
proof. (ii) Multiply (31.6) by J. (iii) The number of undirected
edges is one-half the trace of A
2
. (iv) Show that the undirected
edges do not have a common vertex and that they are not joined
by an edge. Up to reversing directions, this leaves only one way to
complete the graph.
Problem 32A. One of the graphs has two vertices, one has three
vertices, and two have four vertices.
Problem 32B. Let X be the set of vertices reachable by a directed
path from s with f(e) > 0 on all edges of the path, and let Y be
the remaining vertices, if any. If t / X, we have a contradiction to
Eq. (7.1) (why?). For part (ii), use induction on k.
Problem 32C. Assign capacity 1 to all edges of D. Use Theorem
7.1.
Problem 32D. If (H, K) is an -separation and the deletion of
the vertices S := V (H) V (K) does not disconnect the graph,
then at least one of H or K has vertex set contained in S.
Problem 33A. Briey explain why spanning trees of G that con-
tain the edge e are in one-to-one correspondence with the spanning
trees of G
//
e
and why spanning trees of G that do not contain the
edge e are in one-to-one correspondence with the spanning trees of
G
/
e
.
Problem 33B. Find an answer of the form f()+(1)
n
g(). Use
induction, Equation (32.1), and Example 32.1.
Problem 33C. More strongly, a subdivision of K
3,3
(where edges
are replaced by paths) occurs as a subgraph of the Petersen graph.
Hints and comments on problems 573
Problem 33D. Suppose a vertex of G
//
S
is incident with three dis-
tinct edges e
1
, e
2
, e
3
. Such a vertex is a connected component C of
G : S. So e
i
has an end x
i
in C, i = 1, 2, 3 (these ends are not
necessarily distinct). Explain why the connectivity of C implies
that there exists a vertex x and possibly degenerate but internally
disjoint paths in C from x to each of x
1
, x
2
, and x
3
. That is, we
have a subgraph of C isomorphic to a subdivision of the letter Y
with the x
i
s at the ends of the arms if the paths all have length
1, but otherwise some of the arms may be degenerate. This ob-
servation immediately leads to a solution of the rst part of the
problem.
Given four vertices x
i
, i = 1, 2, 3, 4, of a connected graph C, show
that there is a subgraph isomorphic to a subdivision of either the
letter X or the letter I, though with possibly degenerate arms, and
with the x
i
s at the ends of the arms.
Check that if a vertex, say 1, of K
5
is replaced by two vertices
1
L
and 1
R
joined by an edge, and the edges joining 1 to the other
vertices are split, say 1
L
is joined to 2 and 3, and 1
R
is joined to 4
and 5, then the resulting graph has K
3,3
as a subgraph. Combine
this with the observation of the previous paragraph to nish the
problem.
Problem 33E. Use Eulers formula. The only pairs (d
1
, d
2
) that
arise other than those (3, 3), (3, 4), (3, 5), (4, 3), (5, 3) that corre-
spond to the ve Platonic solids are those (2, n) and (n, 2), n 2,
that correspond to the polygons and bond-graphs.
Problem 34A. We have a linear mapping from the vector space
over F
2
of all subsets X of the vertex set V (G) onto the cutset space
of G, namely X (X, V (G) X). (The matrix of this mapping
is N, the incidence matrix.) What are the sets X in the kernel of
this mapping? The dimension of the cutset space is the number of
vertices minus the dimension of the kernel.
Problem 34B. Use induction on the cardinality of the support of
codewords.
Problem 34C. The circuits of G
/
e
are exactly those circuits of G
that do not contain e. What are the bonds of H
//
e
?
574 A Course in Combinatorics
Problem 34D. One way to do this would be to rst prove two
things about subsets S of the edge set of a connected graph K: (1)
S is the edge set of a spanning tree in K if and only if no circuit of
K is contained in S and S is maximal with respect to this property
and (2) S is the edge set of a spanning tree in K if and only if every
bond of K meets S nontrivially and S is minimal with respect to
this property.
Problem 34E. Show rst that the two components that result
from deleting an isthmus from a connected trivalent graph each
have an odd number of vertices.
Problem 34F. Here is an idea to construct bonds. Deleting any
edge of a spanning tree of G leaves a subgraph with two compo-
nents; the set of edges with one end in each component is a bond
of G.
Problem 34G. Deleting the edges of a bond S from a connected
graph G always gives a graph with two components, but recall that
we have chosen to delete isolated vertices when we dened G
/
S
. In
part (ii), an equivalence of codes gives a one-to-one correspondence
between the edge sets of H and G that takes circuits to circuits and
bonds to bonds. Use part (i) and the fact that a graph without
isolated vertices is nonseparable if and only if any two edges are
contained together in a circuit; cf. Theorem 32.2.
Problem 34H. Given a polygon P in G, one may consider the
subgraphs H and K where H = P and K is the spanning subgraph
of G containing all edges of G other than those of P.
Consider a bond B in G that consists of all edges with one end
in X and one end in Y , where X and Y partition V (G) and each
induces a connected subgraph. Suppose for contradiction that G is
Tutte-k-connected but that [B[ = < k. Let
1
be the number of
vertices in X incident with edges of B. If the subgraph induced by
X has at least
1
edges, we easily nd an
1
-separation of G. Deal
with the remaining situation.
Problem 34I. We give independent hints, even though one can
derive (i) from (ii), or, as we have mentioned, the rst two parts
from (iii).
Hints and comments on problems 575
(i) If A and B are edge sets of polygons P and Q, start traversing
P in both directions starting with the edge y, and stop as soon as
vertices in Q are reached.
(ii) If A and B are the supports of codewords a and b, the support
S of some linear combination of a and b will contain coordinate y
but not x. Show that any support S of a codeword is the union of
minimal supports of codewords, by induction on [S[.
(iii) Show that, in general, if z S, then there is a minimal
dependent set D so that z D S z. (Review Lemma 23.2.)
Then show that y is in the closure of (A B) x, y.
Problem 35A. Take the vertices of K
n
to be Z
n
. For n = 5, use
all cyclic shifts (translates) of [1, 2, 4, 3] (mod 5).
Problem 35B. Let us drop the last coordinate in the description
of Example 21.4 and describe the Clebsch graph as having vertex
set F
4
2
and where x and y are adjacent when x +y has weight 3 or
4. Try to choose one walk w and obtain the others as translates of
w by all elements of F
4
2
.
Problem 35C. Part (i) is similar to Theorem 33.5. For the proof
of uniqueness of the mesh, we may take V (K
7
) = 0, 1, . . . , 6 and,
without loss of generality, the walks traversing 0 are
[6, 0, 1], [1, 0, 2], [2, 0, 3], [3, 0, 4], [4, 0, 5], [5, 0, 6].
The walk traversing the edge (6, 1) has third vertex 3 or 4 (otherwise
the vertex condition is violated at 1 or 6) and in either case the
other walks are uniquely determined.
Problem 35D. The vertices of the graph /
x
are the edges inci-
dent with x. These edges form the edge set of a bond in G and
hence are the edges of a circuit in H.
Problem 35E. Let us use the term ag for a triple (x, e, F) con-
sisting of a vertex x, an edge e incident with x, and a face F incident
with e. There is exactly one other vertex x
/
so that (x
/
, e, F) is a
ag, exactly one other edge e
/
so that (x, e
/
, F) is a ag, and ex-
actly one other face F
/
so that (x, e, F
/
) is a ag; these objects
must hence be xed by any automorphism that xes the terms
of (x, e, F).
576 A Course in Combinatorics
Problem 35F. Use Theorem 35.1. The least value of n > 2 for
which K
n,n
embeds in an orientable surface so that all faces are
quadrilaterals is n = 6; n must be even, but K
4,4
does not so
embed in T
0
. Some ideas and/or analysis of cases is required.
Problem 35G. All faces will have the same size. Consider the
walk traversing the edge (0, 1). The next several edges will be
(1, 1 ), (1 , 1 +
2
), (1 +
2
, 1 +
2

3
), etc. If
m 2 (mod 4), for example, then the walk has length m/2.
Problem 36A. The sum of the determinants of all principal n1
by n 1 submatrices of M is an appropriate sign times the coe-
cient of x in the characteristic polynomial det(xI M) of M.
Problem 36B. There is one such arborescence for every mapping
f : 2, 3, . . . , n 1, 2, . . . , n 1 so that f(i) < i for each i.
Problem 36C. If a vector g with coordinates indexed by E(H)
has the property that for every closed walk w, the signed sum of the
values of g on the edges of w is zero, then dene a vector h indexed
by V (H) as follows. (We assume H is connected for convenience;
otherwise work separately with each component.) Fix a vertex x,
dene h(x) := 0, and for any vertex y, let h(y) be the signed sum
of the values of g on the edges of any walk from x to y. Check that
h is well dened and that g is the coboundary of h.
Problem 36D. Let Z denote the cycle space and B the cobound-
ary space. These are orthogonal complements, so their (direct) sum
is 1
m
(the space of all vectors with coordinates indexed by E(D)).
Show that the linear transformation R is one-to-one on Z and that
ZR B = 0, so that 1
m
is the direct sum of these subspaces.
Problem 36E. The sizes of all squares in Fig. 36.4 may be ex-
pressed as integral linear combinations of the sizes x, y, z of the
three indicated squares. For example, the smallest square has size
z x and the next smallest has size x + z y. It will be found
that the sizes of some squares can be so expressed in more than one
way, and we thus have linear relations between x, y, z. Eventually,
we can express the size of any square as a rational scalar multiple
of x, and the least common denominator of these rational numbers
is the value we should take for x.
Hints and comments on problems 577
Problem 36F. It may be quickest to proceed as in Problem 36E;
let x, y, z, for example, denote the values of the current on three
edges, etc. But one can evaluate determinants as in equation (36.3)
to nd the solutions.
Problem 37A. The rotations include: the identity (1), rotations
of 90, 180, and 270 degrees about an axis through the centers of
two opposite faces (3 + 3 + 3), rotations of 180 degrees about an
axis through the centers of two opposite edges (6), and rotations of
120 and 240 degrees about an axis through two opposite vertices
(2 +2 +2 +2). The latter eight rotations, for example, contribute
8X
2
3
to the cycle index.
Problem 37B. Use Theorem 37.2 and Problem 37A.
Problem 37C. Explain why the number of injections f xed by
(, ) depends only on the cycle structure of and (i.e. on the
numbers z
i
of cycles of length i for each i) and nd an expression
for the number of injections in terms of the z
i
()s and the z
i
()s.
Problem 37D. In Theorem 37.3, take G to be dihedral and H to
be the symmetric group S
2
.
Problem 37E. If a rotation of the cube has a cycles on the faces
and b cycles on the vertices, the number of colorings xed by is
t
a
s
b
. Use Burnsides Lemma.
Problem 38A. Let m := v/2. For each of the M := v 1 colors
i, we obtain an m-partition of V (K
v
) by taking the edges of color
i, all singletons x for x not incident with an edge of color i, and
enough copies of the empty set to make an m-partition. Check that
the condition (38.1) holds. Review the proof of Theorem 38.1.
Appendix 2
Formal power series
Before introducing the subject of this appendix, we observe that
many of the assertions that we shall make will not be proved. In
these cases the proof will be an easy exercise for the reader.
Consider the set
C
N
0
:= (a
0
, a
1
, a
2
, . . . ) :
iN
0
[a
i
C] .
On this set we introduce an addition operation and a multiplication
as follows:
(a
0
, a
1
, . . . ) + (b
0
, b
1
, . . . ) := (a
0
+b
0
, a
1
+b
1
, . . . ),
(a
0
, a
1
, . . . )(b
0
, b
1
, . . . ) := (c
0
, c
1
, . . . ),
where c
n
:=

n
i=0
a
i
b
ni
.
This denition produces a ring that we denote as C[[z]] and call
the ring of formal power series. This name is explained in the
following way. Let z := (0, 1, 0, 0, . . . ). Then z
n
is the sequence
(0, . . . , 0, 1, 0, . . . ) with a 1 in the n-th position. So, formally we
have
a = (a
0
, a
1
, . . . ) =

n=0
a
n
z
n
=: a(z).
We shall use both notations, i.e. a and a(z), for the sequence. We
say that a
n
is the coecient of z
n
in a (or a(z)). Notice that C[z],
the ring of polynomials with coecients in C, is a subring of C[[z]].
Some of the power series will be convergent in the sense of analysis.
For these, we can use results that we know from analysis. Quite
often these results can be proved in the formal sense, i.e. not using
convergence or other tools from analysis.
Formal power series 579
Example 1. Let f := (1, 1, 1, . . . ). Using the denition of multi-
plication and 1z = (1, 1, 0, . . . ), we nd (1z)f = (1, 0, 0, . . . ) =
1. So, in C[[z]], we have f = (1 z)
1
, i.e.
1
1 z
=

n=0
z
n
,
a result that we knew from analysis.
The regular elements in C[[z]] are the power series with a
0
,= 0.
This can be seen immediately from the denition of multiplication.
From the relation a(z)b(z) = 1 we can calculate the coecients b
n
,
since b
0
= a
1
0
and b
n
= a
1
0

n
i=1
a
i
b
ni
. From this we see that
the quotient eld of C[[z]] can be identied with the set of so-called
Laurent series

n=k
a
n
z
n
, where k Z. Since it plays a special
role further on, we give the coecient of z
1
a name familiar from
analysis. If a(z) is the series, we say that a
1
is the residue of a(z).
This will be written as Res a(z).
Let f
n
(z) =

i=0
c
ni
z
i
(n = 0, 1, 2, . . . ) be elements of C[[z]] with
the property

n
i
[n > n
i
c
ni
= 0].
Then we can formally dene
(1)

n=0
f
n
(z) =

i=0
_
n
i

n=0
c
ni
_
z
i
.
This denition allows us to introduce substitution of a power series
b(z) for the variable z of a power series a(z). If b
0
= 0, which we
also write as b(0) = 0, then the powers b
n
(z) := (b(z))
n
satisfy the
condition for formal addition, i.e.
a(b(z)) :=

n=0
a
n
b
n
(z)
makes sense.
580 A Course in Combinatorics
Example 2. Let f(z) := (1 z)
1
, g(z) = 2z z
2
. Then formally
h(z) := f(g(z)) = 1 + (2z z
2
) + (2z z
2
)
2
+
= 1 + 2z + 3z
2
+ 4z
3
+ .
From calculus we know that this is the power series expansion of
(1 z)
2
, so this must also be true in C[[z]]. Indeed we have
(1 z)h(z) =

n=0
z
n
= (1 z)
1
.
This also follows from (legitimate) algebraic manipulation:
f(g(z)) =
_
1 (2z z
2
)
_
1
= (1 z)
2
.
Many power series that we often use represent well known func-
tions, and in many cases the inverse function is also represented by
a power series. This can also be interpreted formally for series f(z)
with f
0
= 0 and f
1
,= 0. We solve the equation f(g(z)) = z by
substitution. This yields f
1
g
1
= 1, f
1
g
2
+ g
2
1
= 0, and in general
an expression for the coecient of z
n
starting with f
1
g
n
for which
the other terms only involve coecients f
i
and coecients g
k
with
k < n. Setting this 0 allows us to calculate g
n
.
Example 3. The reader who likes a combinatorial challenge can
give a proof by counting of the formula
n

k=0
_
2k
k
__
2n 2k
n k
_
= 4
n
.
This would be equivalent to proving that the formal power series
f(z) :=

n=0
_
2n
n
_
(z/4)
n
satises the relation f
2
(z) = (1z)
1
. By
similar arguments or by algebraic manipulation one can then nd
the formal power series that deserves the name (1 +z)
1
2
.
Now consider the formal power series f(z) := 2z + z
2
. The in-
verse function procedure that we described above will yield a power
series g(z) that satises 2g(z) +g
2
(z) = z, i.e. (1 +g(z))
2
= 1 +z.
So this should be the series that we just called (1 +z)
1
2
. It should
Formal power series 581
no longer be surprising that algebraic relations that hold for con-
vergent power series are also true within the theory of formal power
series.
Remark. We point out that substitutions that make perfect sense
in calculus could be forbidden within the present theory. The power
series

n=0
z
n
n!
will of course be given the name exp(z). In calculus
we could substitute z = 1+x and nd the power series for exp(1+x).
We do not allow that under formal addition.
We now introduce formal derivation of power series.
Denition. If f(z) C[[z]], then we dene the derivative
(Df)(z) = f
/
(z)
to be the power series

n=1
nf
n
z
n1
.
The reader should have no diculty proving the following rules
by using the denitions in this appendix:
(D1) (f(z) +g(z))
/
= f
/
(z) +g
/
(z);
(D2) (f(z)g(z))
/
= f
/
(z)g(z) +f(z)g
/
(z);
(D3) (f
k
(z))
/
= kf
k1
(z)f
/
(z);
(D4) (f(g(z)))
/
= f
/
(g(z))g
/
(z).
The chain rule (D4) is an example of a more general statement,
namely that
D
_

n=0
f
n
(z)
_
=

n=0
D(f
n
(z)).
If convergence plays a role, this is a dicult theorem with extra
conditions, but for formal power series it is trivial!
The familiar rule for dierentiation of a quotient is also easily
proved. So we can carry the theory over to Laurent series. We
shall need the following two facts. Again we leave the proof as an
exercise.
If w(z) is a Laurent series, then
(R1) Res(w
/
(z)) = 0;
(R2) the residue of w
/
(z)/w(z) is the least integer such that the
coecient of z

in w(z) is nonzero.
582 A Course in Combinatorics
We have already mentioned the idea of an inverse function and
have shown how to calculate its coecients recursively. The next
theorem gives an expression for the coecients.
Theorem 1. Let W(z) = w
1
z +w
2
z
2
+ be a power series with
w
1
,= 0. Let Z(w) = c
1
w +c
2
w
2
+ be a power series in w, such
that Z(W(z)) = z. Then
c
n
= Res
_
1
nW
n
(z)
_
.
Proof: Observe that c
1
= w
1
1
. We now use formal derivation and
apply it to the relation Z(W(z)) = z. This yields
(2) 1 =

k=1
kc
k
W
k1
(z)W
/
(z).
Consider the series obtained by dividing the right-hand side of (2)
by nW
n
(z). If n ,= k, then the term W
k1n
(z)W
/
(z) is a derivative
by (D3) and hence has residue 0 by (R1). By applying (R2) to the
term with n = k we nd the assertion of the theorem.
This theorem makes it possible for us to give a proof of the
Lagrange inversion formula (see Theorem 14.3) within the theory
of formal power series. Let f(z) be a power series with f
0
,= 0.
Then W(z) := z/f(z) is a power series with w
1
,= 0. Now apply
Theorem 1. We nd that z =

n=1
c
n
w
n
with
c
n
= Res
_
f
n
(z)
nz
n
_
=
1
n!
(D
n1
f
n
)(0),
which is (14.19). This approach is based on P. Henrici, An alge-
braic proof of the LagrangeB urmann formula, J. Math. Anal. and
Appl. 8 (1964), 218224. The elegant simplication is due to J. W.
Nienhuys.
As was observed above, we dene
(3) exp(z) :=

n=0
z
n
n!
.
Formal power series 583
From previous knowledge we expect that replacing z by z will
yield the inverse element in the ring C[[z]]. Formal multiplication
and the fact that

n
k=0
(1)
k
_
n
k
_
= 0 for n > 0 shows that this is
true.
It is now completely natural for us to also dene
(4) log(1 +z) :=

n=1
(1)
n
z
n
n
.
Again, calculus makes us expect a relation between the functions
log and exp. By our substitution rule, it makes sense to consider
the power series log(exp(z)). From (D4) we then nd
D(log(exp(z))) =
exp(z)
exp(z)
= 1,
i.e. log(exp(z)) = z. (Here we have used the fact that the formal
derivative of log(1 +z) is (1 +z)
1
.)
Of course, much more can be said about formal power series.
One could explore how many familiar results from analysis can be
carried over or given a formal proof. We hope that this sketchy
treatment will suce to make things clearer.
Name Index
AardenneEhrenfest, T. van 520
Alltop, W.O. 223, 241
Andre, D. 138, 151
Andre, J. 320, 324
Assmus, E.F. 249, 259, 260
Appel, K. 24, 34, 35, 466, 471
Baer, R. 255
Bagchi, N. 571
Baker, R.D. 300
Balinski, M. 457, 458
Bannai, E. 413, 431
Baranyai, Zs. 536, 537, 540, 541
Batten, L.M. 324, 363, 368
Baumert, L.D. 204, 214, 294,
295, 567
Beauregard Robinson, G. de
165, 167
Beckworth, D. 571
Beineke, L.W. 281
Belevitch, V. 200, 214, 265
Bermond, J.-C. 537
Bertrand, J.L.F. 151
Best, M.R. 207, 214
Beth, T. 301
Bhattacharya, K.N. 240, 241
Biggs, N. 450, 466, 471
Binet, A. 508, 510
Birkho, G. 48, 51, 52, 111
Block, R.E. 371, 382
Blokhuis, A. 367, 368
Bollob as, B. 38, 41, 57
Bondy, J.A. 9, 542
Bose, R.C. 240, 264, 271, 272,
279, 280, 284, 288, 300, 301,
365, 367, 368, 409
Bouwkamp, C.J. 520, 521
Bregman, L.M. 101, 108
Brooks, R.L. 24, 35
Brouwer, A.E. 85, 87, 88, 272,
279, 280, 355, 368, 407, 431,
537, 541
Brouwer, L. E. J. 458
Bruck, R.H. 230, 241, 280, 381,
382, 383, 398
Bruijn, N.G. de 55, 59, 71, 75,
76, 216, 241, 433, 513, 514,
520, 532, 535
Buekenhout, F. 281
B urmann, H. 582
Burnside, W. 94, 370
Cameron, P.J. 259, 260, 279,
280, 281
Catalan, E. 150, 151
Catherine the Great 284
Cauchy, A.L. 94, 108, 508, 510
Cayley, A. 12, 22, 504, 505, 508
Name Index 585
Chakravarti, I.M. 367, 368
Chandrasekharan, K. 167, 230,
241
Chang, L.C. 271, 280, 281
Chartrand, G. 462, 471, 491, 505
Chowla, S. 230, 241, 301, 383,
398
Chvatal, V. 70
Clebsch, A. 263, 279, 498, 505,
575
Cohen, A.M. 407, 431
Cohen, G.D. 241
Connor, W.S. 228, 241, 280, 281
Conway, J.H. 216, 560
Crapo, H. 324, 490, 564
Crawley, P. 304, 313, 319,324
Cruse, A. 541
Cvetkovic, D.M. 440, 450
Da Silva, D.A. 96
De Clerck, F. 280, 281
Dedekind, J.W.R. 167
Dehn, M. 520, 521
Delsarte, P. 279, 280, 281, 405,
412, 413, 419, 422, 424, 431
Dembowski, P. 324, 368
Denes, J. 197
Denniston, R.H.F. 223, 242, 252,
268
Desargues, G. 317, 323
Diaconis, P. 181
Diestel, R. 462, 463, 471
Dilworth, R.P. 53, 60, 304, 313,
319, 324
Dinitz, J.H. 294, 301, 469
Dirac, G.A. 40
Doob, M. 450
Dowling, T. 341, 342, 349
Duijvestijn, A.J.W. 517, 520,
521
Dulmage, A.L. 295, 302
Duval, A.M. 276, 281
Ebbenhorst Tengbergen, C. van
55, 59
Edmonds, J. 69, 503, 505
Egoritsjev, G.P. 110
Egecioglu,

O. 16, 23
Elspas, B. 446, 450
Erd os, P. Preface, 5, 10, 33, 35,
51, 59, 216, 241, 301, 331,
418, 430, 433, 520, 544
Euler, L. 6, 10, 136, 157, 166,
182, 284, 290, 300, 465, 466,
480, 491, 497, 513 542, 573
Evans, T. 189, 197
Falikman, D.I. 110
Fan, K. 491, 505
Fano, G. 225, 240, 242, 492
Fekete, M. 103, 109
Ferrers, N.M. 166, 327
Fibonacci 150
Fisher, R.A. 222, 240, 558
Flye Sainte-Marie, C. 75, 76
Ford, L.R. 64, 66, 70
Fourier, J.B. 212
Frame, J.S. 165, 167, 553
Frankl, P. 59, 60, 331, 332
Franklin, F. 157, 167
Frechet, M. 491, 505
Freeman, J.M. 553
Frobenius, G. 94, 167, 442, 568
Fulkerson, D.R. 64, 66, 69
Gale, D. 181
Galvin, F. 194, 197, 469
Gantmacher, F.R. 442, 450
Gardner, M. 521
Garey, M. 9, 10
Gauss, C.F. 97, 325, 332
Gelling, E.N. 537, 541
586 Name Index
Gewirtz, A. 267, 281
Godel, K. 458
Godlewski, P. 241
Godsil, C.D. 227, 281
Goethals, J.-M. 266, 279, 280,
281
Golay, M.J.E. 252, 259, 417, 429
Golomb, S.W. 204, 214
Gordon, B. 381, 382
Goulden, I.P. 150, 151
Graham, R.L. 32, 35, 78, 88,
332, 432, 521
Granell, M.J. 223, 242
Greene, C. 59, 60, 312, 334, 547
Griggs, T.S. 223, 242
Grinstead, C.M. 35
Grossman, J.W. 10
Gr unbaum, B. 349, 350, 456,
457, 458
Gustin, W. 504, 506
Guthrie, Francis 465
Guthrie, Frederick 465
Hadamard, J. 213, 268, 375
Haemers, W. 280, 435, 450
Haken, W. 24, 34, 35, 466, 471
Hall, M. 51, 60, 196, 204, 214,
281, 298, 302, 319, 324, 383,
394, 395
Hall, P. 43, 51, 52
Halmos, P.R. 51, 52
Hamilton, W.R. 8, 10, 465
Hamming, R.W. 259, 406, 417
Harary, F. 535
Hardy, G.H. 166, 167
Hautus, M.L.J. 51
Heawood, P.J. 433, 466, 493,
494, 498, 505
Henrici, P. 582
Hickerson, D. 150
Higman, D.G. 267, 279, 281
Hilbert, D. 470
Hirschfeld, J.W.P. 358, 359, 368,
566
Homan, A.J. 41, 280, 282, 434,
440, 450, 561
Homan, D. 197
Hsieh, W.N. 331, 332
Hu, T.C. 70
Hubaut, X. 279, 282
Hughes, D.R. 321, 324
Ito, T. 413, 431
Jackson, D.M. 150, 151, 181,
280, 431
Jacobi, C.G.J. 167
Jeans, J.H. 519, 521
Johnson, D. 295, 302
Johnson, D.S. 9, 10
Johnson, S. 35, 406
Jones, B.W. 80, 88
Joyal, A. 141, 151
Jungnickel, D. 227, 242, 301, 404
Karp, R.M. 69
Katona, G.O.H. 59, 60
Keedwell, A.D. 197
Kempe, A.B. 466, 467, 470
Kirchho, G.R. 500, 515
Kirkman, T.P. 240, 242
Klarner, D.A. 132, 151
Klein, F. 285, 499
Kleitman, D.J. 59, 60, 547
Knuth, D. 87, 163, 167, 181
Ko, Chao 51, 331, 418, 430
Koch, J. 466, 470, 471
Konig, D. 51, 52
Korner, J. 258
Koornwinder, T.H. 280, 282
Kramer, E.S. 557
Krause, M. 172, 181
Name Index 587
Kreher, D.L. 224, 242
Krein, M.G. 267, 279, 280
Kronecker, L. 201, 268
Kruyswijk, D. 55, 59
Kuratowski, K. 462, 470
Lagrange, J.L. 145, 151, 230,
240, 582
Lam, C.W.H. 242
Lander, E.S. 394, 398, 404
La Poutre, J.A. 86, 88
Laurent, P.M.H. 579
Leavitt, D.W. 224, 242
Leeb, K. 332
Lenz, H. 301
Leonardo of Pisa 150
Lesniak, L. 462, 491, 505
Lewin, M. 35, 36
Lint, J.H. van 51, 52, 60, 86, 88,
108, 109, 110, 118, 166, 167,
259, 260, 279, 280, 281, 282,
431, 543, 546, 551, 552, 558
Lloyd, E.K. 466, 471
Lloyd, S.P. 425, 427, 431
Locke, S.C. 553
London, D. 114, 118
Lorentz, H.A. 118
Lovasz, L. 41, 42, 436, 437, 438,
450, 546
Lubell, D. 54, 60
Lucas, F.E.A. 95, 97, 150
MacLane, S. 480, 490
MacMahon, P.A. 166, 167
MacNeish, H.F. 288, 290, 300,
302
MacWilliams, F.J. 248, 258, 259,
260
Magliveras, S.S. 224, 242
Maldeghem, H. Van 280, 282
Mann, H.B. 298, 302, 394, 395
Mantel, W. 37, 42
Marcus, M. 112, 114, 118
Mathieu,

E. 260
Mathon, R. 280
Mattson, H.F. 249, 259, 260
McFarland, R.L. 381, 382
McKay, B.D. 176, 181, 277, 281
Medema, P. 520, 521
Mendelsohn, E. 537, 541
Mendelsohn, N.S. 295, 302
Menger, K. 454, 455, 458
Merkx, F. 241
Mesner, D.M. 264, 279, 280, 409
Mills, W.H. 223, 242, 381, 382
Minc, H. 101, 108, 109, 118
Mirsky, L. 54, 60
Mobius, A.F. 97, 333, 362
Moivre, A. de 108
Montmort, P.R. de 97
Moon, J.W. 108, 109
Moore, E.H. 300, 302
Morgan, A. de 465
Muir, T. 108
Muller, D.E. 199, 213, 214, 261
Muzychuk, M. 448
Narasima Sasbry, N.S. 571
Nesetril, J. 521
Neumaier, A. 270, 272, 282, 407,
431
Neumann, P.M. 97
Newman, M. 112, 114, 118
Nichols, W. 553
Niderhausen, H. 553
Nienhuys, J.W. 582
Niven, I. 150, 151
Nordstrom, A.W. 419
Ohm, G.S. 515
Ostrand, P. 51, 52
588 Name Index
Paige, L.J. 298, 302
Paley, R.E.A.C. 203, 213, 214,
262, 375
Paola, J. di 368
Pappus of Alexandria 318, 323
Parker, E.T. 284, 288, 301
Pasch, M. 314, 323
Payne, S.E. 280, 282
Pedoe, D. 319, 324
Peltesohn, R. 537, 541
Perron, O. 442
Petersen, J.P.C. 261, 433, 462,
465,477, 483, 491, 508, 542
Petrenjuk, A.Ya. 222, 242
Pierce, J.R. 88
Piper, F.C. 321, 324
Pless, V. 252, 260, 404
Plotkin, M. 212
Pl ucker, J. 239
Pollak, H.O. 78, 88, 432
Polya, G. 522, 535
Posner, E.C. 213, 214, 557
Pr ufer, H. 13, 23
Pulver, E.D. 535
Pythagoras 166
Rademacher, H. 166, 167
Rado, R. 51, 331, 418, 430
Radziszowski, S.P. 35, 36, 224,
242
Ramanujan, S. 166, 167
Ramsey, F.P. 28, 35, 36
Raney, G.N. 151
Ray-Chaudhuri, D.K. 222, 242,
367, 368, 425, 431
Reed, I.S. 213, 214, 361
Rees, G.M.J., van 294, 302
Remmel, J.B. 16, 23
Riemann, G.F.B. 97
Ringel, G. 493, 494, 502, 506
Roberts, S.M. 35
Robertson, N. 463
Robinson, J.P. 419
Rogers, D.G.E.D. 181
Rosa, A. 537, 541
Rota, G.-C. 151, 324, 349, 350,
378, 490, 564
Rothschild, B.L. 32, 33, 35, 332
Ryser, H.J. Preface, 99, 108,
109, 181, 187, 197, 213, 229,
230, 241, 292, 383, 394, 395,
398
Sachs, H. 450
Schellenberg, P.J. 294, 302
Schl ai, L. 275
Schonheim, J. 59, 60
Schrijver, A. 65, 101, 107, 108,
109, 280, 282, 355, 368, 537,
541
Schur, I. 22, 34, 544
Schutzenberger, M.P. 240, 242
Schwenk, A.J. 450
Scott, L.L. 279
Seberry, J. 213
Segner, J.A. von 136
Seidel, J.J. 266, 279, 280, 281,
282, 558
Seymour, P. 463
Shannon, C.E. 259, 260, 436
Shimamoto, T. 279
Shrikhande, S.S. 284, 288, 301,
561
Simonyi, G. 258
Sims, C.C. 267, 279, 281
Singer, J. 377, 378, 382
Singleton, R.R. 41, 561
Skolem, Th. 241, 242
Sloane, N.J.A. 258, 259, 260, 560
Smetaniuk, B. 189, 197
Smith, C. A. B. 520
Spencer, J.L. 33, 35, 59, 60, 332,
544
Name Index 589
Sperner, E. 54, 59, 60, 326
Sprott, D.A. 376, 384
Stanley, R.P. 150, 151, 349, 350,
460, 471
Stanton, R.G. 376, 384
Staudt, K.G.C. von 240
Steiner, J. 217, 240, 242
Steinitz, E. 465, 470, 471, 486
Stinson, D.R. 284, 302
Stirling, J. 108, 128
Strauss, E.G. 301
Swiercz, S. 241
Sylvester, J.J. 96, 166, 432
Szego, G. 535
Szekeres, G. 33, 35
Tait, P.G. 481486, 490
Tarry, G. 284
Tarsy, M. 35
Teirlinck, L. 224, 242
Thas, J.A. 280, 287
Thiel, L. 242
Thomas, S. 330, 332
Thomassen, C. 469, 471
Thompson, J.G. 259, 260
Thompson, T.M. 258, 263
Thrall, R.M. 165, 167
Tits, J. 221, 242
Todd, J.A. 375
Tur an, P. 38, 41, 42
Turner, J. 446, 450
Tutte, W.T. 457, 458, 462, 471,
483, 489, 490, 512, 517, 520
Tverberg, H. 53, 60
Valiant, W.G. 107, 109
Vallee-Poussin, C.J. De La 213
Vandermonde, A.-T. 163
Vanstone, S.A. 181, 227, 242,
280, 294, 302, 431
Vaughan, H.E. 51
Veblen, O. 279, 290
Vedder, K. 362
Voigt, M. 470
Voorhoeve, M. 105, 109
Waerden, B.L. van der 32, 36,
76, 110, 118, 197
Wald, A. 458
Wallis, J.S. 213, 214
Watkins, J.J. 10, 11
Watson, G.N. 151
Wedderburn, J.H.M 319, 402
Weisner, L. 335, 339, 350
Welch, L.R. 381, 382
Welsh, D.J.A. 490
White, A.T. 506
Whitney, H. 454, 458, 472, 490
Whittaker, E.T. 151
Wilbrink, H.A. 396, 404
Wilf, H. 450
Williamson, J. 204, 214
Wilson, R.J. 10, 11, 281, 466,
471
Wilson, R.M. 222, 242, 279, 282,
301, 302, 331, 334, 341, 350,
412, 430
Winkler, P. 80, 88
Witt, E. 223, 242, 397
Woolhouse, W.S.B. 240, 243
Wright, E.M. 160, 166, 167, 168
Yamamoto, K. 395
Young, A. 167, 168
Young, J.W. 313, 324
Youngs, J.W.T. 493, 494, 502,
506
Subject Index
(0, 1)-matrices 48, 100, 169
1-skeleton of a polytope 456
1-factor in a graph 537
1-factorization of a graph 537
15 schoolgirls problem 240
2-cell embedding of a graph 463,
473, 494495
2-designs 224, 311
3-claw 570
3-connected graph 483484
36 ocers problem 284
5-designs 223, 242, 249, 252253,
255, 266
6-designs 224
absolute bound for strongly
regular graphs 270, 273
acyclic edge 482
acyclic orientations of a graph
461
addressing problem 77
adjacency matrix of a digraph
432
adjacency matrix of a graph
263, 391402, 449
adjacency matrix of a
multigraph 507
adjacency matrix of a scheme
409
adjacency matrix of a
tournament 432
adjacent vertices in a graph 4
ane geometry 304
ane hyperplane 211
ane plane 227, 273, 287
ane subspace 304
ane translation plane 320
algebraic methods 264, 267,
432
alphabet of a code 244
ancestor 20
Andres reection principle 118
antichain 53, 327
arborescence 15, 145
arc 351353, 367
arithmetic progression 32
AssmusMattson theorem 249
associates in a scheme 406
association matrices of a scheme
409
association scheme 279, 405
associative block design 83
atomic lattice 305
augmenting path; see also special
path 63, 69
automorphism of a code 260
automorphism of a graph 3, 422,
542, 569
Subject Index 591
automorphism of a map 499, 575
automorphism of a symmetric
design 316, 329, 379, 566
automorphism of a tree 13
axis of perspectivity 317
Baer subplane 255, 355, 394
balanced incomplete block
design (BIBD) 217
balanced orientation of a graph
68
ballot problem 150
Baranyais theorem 536
base blocks 236, 241
base of a Ferrers diagram 158
bases in a combinatorial
geometry 310
Bell numbers 125, 149
binary code 244
binary Golay code 243, 252255,
259, 417, 560
binary relation on a set 406
binary tree 139
bipartite graph 24, 43, 55, 69,
196, 324
biplane 267
Birkhos theorem 48, 65, 111
block design 217
block graph of a design 266
Blocks Lemma 371
block of cheese 41
blocking set in a projective plane
355
blocks of an incidence structure
215
block-size of a t-design 216
bond-graph 451, 473
bond of a graph 475
Boolean algebra 303
BoseMesner algebra of a
scheme 409, 419, 430
BoseMesner algebra of a
strongly regular graph 264
breadth-rst search 19, 81
bridge 20, 482
bridges of K onigsberg 6, 7, 10,
59
Brooks theorem 24, 34
BruckRyserChowla theorem
230, 241, 383, 398
De Bruijn graphs 71
De Bruijn sequences 71, 513
De BruijnErd os theorem 216,
433
Burnsides lemma 94, 97, 370,
524
capacity of a cut 62
capacity of an edge 62
Catalan numbers 136141, 150,
16
CauchyBinet theorem 508, 510
Cauchy inequality 115
Cayley graph 446, 504505
Cayleys theorem 12, 145, 508
cell 162
central collineation 367
chain 53, 305, 348
Chang graphs 271, 280
character 202, 248, 394
cheapest spanning tree 18, 543
Chinese Remainder Theorem
293
chromatic number 24, 327, 398,
466, 571
chromatic polynomial 341, 349,
459
circle geometries 365
circuits of a graph 475
circulant graph 446, 448
circular sequences 9495, 120
circulation on a digraph 66
592 Subject Index
claw 218, 500
claw bound 272273
Clebsch graph 263, 276, 279,
440, 447, 498, 505, 575
clique in a graph 271, 435
clique in a scheme 421
closed path 5
closed walk 5, 515
closure in a combinatorial
geometry 308
coboundary space of a digraph
514
coclique in a graph 434435, 440
coclique in a scheme 421
code 244, 396
code in a scheme 422
code of a design 256258, 396
code of a graph 475
codewords 244
coding theory 244
coline 309
collinear 272
collineation of a projective plane
353, 366
colorings of graphs 24, 306, 459
column-complete 196
column of a Latin square 182
combinatorial design 351
combinatorial geometry 306
combinatorial proofs of identities
121122, 125, 138, 147, 552
cometric scheme 422
complement of a design 220,
251, 569
complement of a dierence set
391
complement of a graph 393, 563,
570
complement of a symmetric
design 401
complement of an incidence
structure 215
complementing permutation 342
complete bipartite graph 6, 39,
432
complete cycle 7275
complete graph 3, 12, 24, 2728,
30, 33, 37, 78, 217, 261, 432,
491, 492, 494, 500, 502, 508,
540
complete mapping of a group
298
complete matching in a bipartite
graph 43, 56
complete multipartite graph 38
complete uniform hypergraph
540
complexity of a graph 460, 519
component 5
composition methods 288
composition of a number 154,
553
conference matrix 200, 233,
265, 402
conguration counting series 528
confoundable words 436
conjugate of a partition 156, 169
conjugates of a Latin square 183
connected combinatorial
geometry 313, 564
connected graph 56, 347, 451
connected strongly regular graph
261
connectivity of a graph 451
conservation of ow 62
contraction of edges 462
convex n-gon 33, 139, 146
convex combination of
permutation matrices 49, 111
convex polytope 336, 349, 456,
465, 486
cost of an edge 17
cotree 462, 473
Subject Index 593
counting in two ways 4, 207,
212, 219, 262
covering radius 245, 247
covering in a poset 303, 305, 335
cut in a transportation network
62
cutset space of a graph 476
cutsets of a graph 475
cycle index 524
cycle space of a digraph 514
cycle space of a graph 453
cycles of a graph 475
cycles of a permutation 123, 127,
524
cyclic dierence set 373, 386,
567
cyclotomic scheme 408
d-code in a metric scheme 422
De Bruijn graphs 71
De Bruijn sequences 71
De BruijnErd os theorem 216,
433
decode 209, 212
Dedekind -function 167
degree of a face 465
degree of a vertex 4
degrees of a scheme 406
deletion of edges 459
Delsartes inequalities 405, 416,
418
dependency graph 31
depth-rst search 19, 81
depth of an orthogonal array 183
derangements 90, 97, 129, 142,
149
derivation 146
derived design 228
Desargues conguration 317
Desargues theorem 318, 321,
322
Desarguesian planes 319, 352
descendant 20
design of experiments 240
design theory 215
designs in a scheme 422
determinant of the distance
matrix 80
dierence methods 234, 240,
288, 294, 498
dierence set 369, 389
digraph 2, 194, 454
Dilworths theorem 53, 55
dimension of a combinatorial
geometry 304
Dinitz conjecture 194, 469
directed Eulerian circuit 16, 71,
75
directed graph 2
directed path 16
discrepancy function 82
distance in a code 244
distance between vertices 5
distance matrix of a graph 7980
distance regular graph 407
distribution vector in a scheme
406, 416417
dodecahedron 8, 492
dominant eigenvalue 444
doubly stochastic matrix 49,
104, 108, 110
drawing of a graph 1, 461
dual arc in projective planes
352353
dual code 246
dual graph 466, 472
dual of a symmetric design 228
dual partial geometry 272
dual poset 307
echelon form 246, 327328
edge colorings 27
594 Subject Index
edges of a graph 1
eigenmatrices of a scheme 413,
415
eigenspaces 264, 410
eigenvalues of a graph 264, 433
electrical network 507, 514, 515
elementary cycle 514
elementary ow 64
elliptic quadratic form 359
elliptic quadric 359
embedding of a graph 461
endpoints of an edge 1
ends of an edge 1
equivalent codes 246
equivalent designs 221
equivalent dierence sets 374
equivalent Latin squares 183
Erd osKo-Rado theorem 56, 59,
331, 418, 430
Erd osnumber 5, 10
error-correcting code 208,
244246, 258-259
Euler characteristic 497
Euler function 92
Eulers conjecture 281, 288, 300
Eulers formula 349, 464, 465,
466, 480, 491, 542, 573
Eulers identity 157
Eulerian circuit 69, 71, 513
Eulerian graph 6
Eulerian poset 350
Evans conjecture 189193
excess of a Hadamard matrix
207
exchange axiom 309
exponential generating function
129
extended code 246
extended design 239
extremal graphs 3840
extremal set theory 56, 59
face of a convex polytope 456
face of a graph or an embedding
463
falling factorial 119
Fano conguration or plane 225,
237, 255, 258, 354, 394, 430,
434, 492
fast Fourier transform 212
feasible ow 61
feasible parameters for strongly
regular graphs 265
Feketes lemma 103, 106, 108,
130, 150, 436
Ferrers diagram 156, 170171,
327328
Fibonacci numbers or sequence
97, 149, 150, 550
Fibonacci recursion 150, 159,
551
nite dierences, calculus of 97
nite graph 4
rst order ReedMuller code
210212, 361
Fishers inequality 222, 226, 353,
364, 405, 430
Five Color Theorem 34, 466, 469
xed points of an automorphism
353, 369
ags of an incidence structure
215
ats 303, 309, 315
ow 61
Ford sequence 75
forest 17
formal derivative 145, 581
formal power series 130151,
578
formally dual schemes 420
formally dual theorems 424
four color problem 459
Subject Index 595
Four Color Theorem 24, 34, 427,
466, 482
Friendship Theorem 279
Frobenius automorphism 568
fully indecomposable matrix 111
g-torus 491
Gaussian coecients 325
Gaussian numbers 312, 325
Gaussian polynomials 326
generalized quadrangle 273274,
280, 363, 367
generating function 126, 129
generator matrix of a code 246,
570
Generalized Steiner System
238239, 304
genus of a surface 491
geometric graph 272
geometric lattice 305, 340, 342,
490, 564
Gewirtz graph 267
girth of a graph 9, 3940
Golay codes 241244, 252255,
259, 279, 417, 429, 560
graceful labeling 22
Graeco-Latin square 284
grand clique in a strongly
regular graph 271, 280, 561
graph of a polytope 456
graph 1
greedy algorithm 18
group ring 383
Hadamard 2-design 218
Hadamard 3-design 218, 220,
252
Hadamard dierence set
375377, 380, 569
Hadamard matrix 199, 218,
279, 411412, 424
Hadamard product of matrices
268269, 409, 415, 419
half-case of strongly regular
graphs 265, 277
Hall multiplier 386
Halls theorem 4347, 65
Hamiltonian circuit 8, 4041,
449, 483, 545
Hamming bound 245
Hamming code 246247, 376,
560
Hamming distance 77, 244
Hamming scheme 406, 411
handle of a representation 437
hash-coding 84
hash functon 84
head of a directed edge 2, 460
Heawood conjecture 493, 498
Heawood graph 433, 434, 492
hermitian form 364
hermitian variety 364
higher incidence matrices of a
t-design 222
HigmanSims graph 267, 279
HomanSingleton graph 41
homeomorphic graphs 453
homogeneous coordinates 317
homomorphisms of the group
ring 385
hooks in Young tableaux 165
hooklengths 165166
horizontally convex polyomino
132
hyperbolic quadratic form 359
hyperbolic quadric 359
hypercube 77
hyperoval 227, 248, 255, 258,
266267, 274
hyperplanes in combinatorial
geometries 309
596 Subject Index
idempotent quasigroup 28
idempotents in the BoseMesner
algebra 236, 410
idempotents in the group ring
402
incidence algebra of a poset 333
incidence in a graph 1
incidence matrix 221
incidence matrix of a design
221, 225
incidence matrix of a directed
graph 62, 509
incidence matrix of a graph 475
incidence matrix of a symmetric
design 228
incidence structure 215
inclusionexclusion 8997, 98,
122, 177, 337, 535, 549, 556
independence number of a graph
436
independent edges in a graph 18
independent set 37
independent subset of a
combinatorial geometry 309
independent vertices in a graph
37, 434435
index of a design 216
index of an orthogonal array
216, 423
induced drawing 463
induced subgraph 4, 77
information rate 209
inner (or dot or scalar) product
115116, 396, 415
Instant Insanity 7
integral ow 6466, 171, 540
integrality condition for strongly
regular graphs 265, 561
interlacing of eigenvalues
439440, 561, 571
interval of a poset 307
invariant factors 399
inversive plane 362
involution 323
irreducible matrix 442
isolated vertex 1
isomorphic designs 221
isomorphic dierence sets 380
isomorphic graphs 2, 411, 466
isomorphic Latin squares 183
isomorphic orthogonal arrays
183
isotropic vector 115
isthmus of a graph 20, 482
Jacobi triple product identity
160
Johnson scheme 406, 413
join (least upper bound) 305
joins (in a graph) 1
Jordan arcs 472
Jordan curve theorem 464
Joyal theory 141
k-class association scheme 406
k-connected graph 451
k-subspace 325
k-Tutte connected 457, 489
k-vertex connected 451, 454
Kempe chain 467
Kirchhos laws 500, 515
Kirkman schoolgirl problem 240
Klein bottle 499
Konigs theorem 48, 51
Krein condition 267, 269, 273,
279, 280
Krein parameters of a scheme
420
Kronecker product 201, 208,
213, 268, 287, 309, 437
Kuratowskis theorem 462
Subject Index 597
labeled graph 12, 149
labeled tree or forest 12, 141,
145
Lagrange inversion formula 145,
151, 582
Latin rectangle 186
Latin square 182, 226, 273, 358
Latin square graph 274, 276, 414
Latin squares, number of
186187, 192
lattice (poset) 305, 333
lattice graph 262
lattice of contractions 308, 340
Laurent series 579, 581
length of a walk 5
length of an edge 17
line graph 194, 435436, 570
line of a block design 323
line of a linear space 215
line of a matrix 48
line of a projective plane 225
line of perspectivity 317
linear code 244, 398
linear order 53
linear programming 69, 457
linear programming bound 406,
418, 419
linear recursion 130134
linear space 215, 289, 303, 313
line-sums of (0, 1)-matrices
100-101, 173181
link-graph 451, 471, 453
list assignment 194
list-colored graph 194
Lloyds theorem 405, 427
loop of a graph 1
loop switching problem 88
Lorentz space 115118
Lovasz Sieve 31
MacNeishs theorem and
conjecture 288, 290
MacWilliams theorem 248, 418
MacWilliams relations 249
magic square 300
majorize 169170
Mantels theorem 37
map on a surface 491
marriage theorem 43, 65
matching in a graph 43
Mathieu groups 259, 560
matrix-tree theorem 507
matroid 324, 489, 490
maxowmincut theorem 64,
171, 540
maximal arc in a projective
plane 367
maximal chain in a poset 5354,
306, 325
maximum distance separable
code 248, 259, 347348
maximum ow 62
maximum matching 69
meet (greatest lower bound) 305
Mengers theorem 454, 455
mesh in a graph 494
method of dierences; see
dierence methods
metric scheme 407, 422
Mincs conjecture 101103, 108
minimal counterexample 2426
minimizing matrix 110
minimum cut 64
minimum distance of a code 211,
244
minimum weight 245
minor of a graph 462
Mobius function of a poset 333,
344
Mobius function, number
theoretic 9294, 333
Mobius inversion 93, 136, 298,
333, 336, 337, 339, 346, 347,
370
598 Subject Index
Mobius plane 362
modular combinatorial geometry
313
modular law 313
modular complement 314
monochromatic triangles 2728
monotone subsequence 55
monovalent vertex 13
multigraph 2
multinomial coecients 16, 120
multiple edges 2
multiplicities of a scheme 411
multiplier of a dierence set
386
Multiplier Theorem 386, 390
mutually orthogonal Latin
squares 300
n-gon 6, 153, 460
near pencil 216, 313, 354
necklaces 9495, 522, 525530,
533
negative Latin square graphs 414
neighborhood regular graph 277
neighbors of a vertex 4
net (partial geometry) 273, 280,
286, 287, 563
nonassociative product operation
136
nondegenerate quadratic form
356
nonembeddable design 228229,
240
nonnegative matrices 111, 442
nonseparable graph 451, 483,
499
nontrivial dierence set 372
NordstromRobinson code 419
normalized dierence set 374,
388
normalized Hadamard matrix
199
NP-complete 9
numerical multiplier of a
dierence set 386
odd graph 430
Ohms law 515
optimal representation of a
graph 437
orbits of a group 95, 370, 407,
523, 528, 530, 532
order of a Hadamard matrix 199
order of an orthogonal array 183
order of a projective plane 225
ordinary generating function 129
orientable mesh 497
orientation of a graph 21, 460
orthogonal array 182, 294, 423
orthogonal idempotents of a
scheme 410
orthogonal Latin squares 283,
563
orthogonality relations 415
orthomorphism of an abelian
group 297298, 339
orthonormal representation of a
graph 437
oval in a projective plane 227,
381
oval in a symmetric design 367
ovoid in projective 3-space 362,
394
P-polynomial 422
pairwise orthogonal Latin
squares 285, 300
Paley graph 262, 265
Paley matrix 203, 252253, 256,
374
PaleyTodd dierence sets
375377
Pappian planes 319
Subject Index 599
Pappus theorem 318
parabolic quadratic form 359
parabolic quadric 359
parallel class in a Steiner system
353
parallel class in an ane plane
227, 273
parallel class of sets 537
parallel edges 2
parameters of a scheme 406
parent 20
parity check matrix of a code
246
parity check symbol 246
partial fractions 153, 553
partial geometry 271, 363
partial Latin square 185
partial match query 84
partially balanced design 430
partially ordered set (poset) 53,
333
partition function 155, 166
partition lattice 307308, 336,
344346
partitions of a number 152,
169, 327329
partitions of a set 49, 106107,
126, 537
partly decomposable matrix 111
Pasch axiom 314
path in a graph 5
paths in the X, Y plane 122,
130, 138, 149, 552
pentagonal numbers 157
perfect arc 351353, 367
perfect code 245, 248, 250, 254,
259, 405, 425, 427, 430
perfect matching 44, 546
permanent of a matrix 98, 549
permanents of (0,1)-matrices
100104
permanents of nonnegative
integral matrices 105
permutation matrices 48, 83,
114, 148, 205, 221, 193, 329
PerronFrobenius theorem
442443
perspective from a line 317
perspective from a point 317
Petersen graph 9, 40, 261, 263,
276, 430, 433434, 436, 449,
462, 465, 477, 483, 485, 486,
491, 492,508, 542, 570, 572
pigeonhole principle 55, 559
planar dierence set 372,
388389, 396, 403
planar graph 1, 461, 479, 480
plane at innity 367
plane tree 139
planted plane trees 139
Platonic solids 465, 499, 573
Plotkins bound 212
point graph of a partial
geometry 272
point of a combinatorial
geometry 303, 309
point of an incidence structure
215
point of a lattice 305
point of perspectivity 317
Polya theory 522
polygon 5, 83, 475
polynomial scheme 422
polyomino 132, 150
poset (partially ordered set) 53,
333
positive and negative vectors 115
positive semidenite or denite
223, 435, 439
probabilistic method 3032, 35,
544
probl`eme des menages 95, 549
600 Subject Index
probl`eme des rencontres 97
projective design 224
projective geometry 225, 303,
304, 313, 351
projective plane 225, 232, 256,
280, 287, 290, 313, 351355,
396
projectively equivalent hermitian
forms 364
projectively equivalent quadratic
forms 356
proper coloring of a graph 24,
341, 459
proper drawing of a graph 461
proper partial geometry 273, 280
Pr ufer code 13
pseudo-geometric graph 272, 280
q-analogues 326, 337, 408
q-ary code 244
Q-polynomial 422
quadratic form 7880, 231, 355,
558, 559, 566
quadric in projective space 356,
566
quasigroup 182, 288
quasiresidual design 228, 275
quasisymmetric design 266, 275
quotient set 372373
rain in Holland 97
Raleigh quotient 434
Ramseys theorem 27, 37, 332
rank in a combinatorial
geometry 309
rank of a connected graph 453
rank of a at 309
rank of a hermitian form 364
rank of a quadratic form 356
rational function 132
real projective plane 226, 492,
496
recoloring 24, 467
recurrence relation 124, 134,
159, 329
recursive method 201, 233
ReedMuller codes 210213,
361, 560
renement order on partitions
307
regions of an embedding 463,
472
regular bipartite graph 44
regular graph 4, 40, 149, 422,
445
regular Hadamard matrix 208,
218, 226, 375
repeated blocks 215, 223
repetition 177
repetition code 209, 245
replication number 219, 240, 430
residual design 228, 275
resolvable Steiner system 353
resolvable transversal designs
291
Riemann zeta function 93
RingelYoungs theorem 494
rooted trees and forests 15, 19,
145146
rotating drum problem 71
rotations of the cube 526, 577
row 182
row-complete 196
row-sum 169
saturated edge 63
scheme 405f
Schlaei graph 275
Schr oderBernstein Theorem 51
SDR theorem; see Halls
theorem
SDRs and permanents 100
self-conjugate partitions 162
Subject Index 601
self-dual code 246, 396
self-orthogonal codes 246, 396
semi-simple algebra 402
semimodular lattice 306, 342
semimodular law 311
semiregularity 44
Shannon capacity of a graph
436437
sieve methods 96
signless Stirling numbers 123,
534
simple answers 288
simple closed path 5
simple design 215, 223
simple dierence set 372
simple digraph 2
simple graph 2
simple path 5, 25
simplex code 247
Singer dierence sets 378380
Singers theorem 377, 381
Singleton bound 247
sink in a transportation network
61
Six Color Theorem 467
Skolem sequences 241
slope of a Ferrers diagram 158
Smith normal form 399
source in a transportation
network 61
spanning arborescence 16, 512,
543
spanning subgraph 4
spanning subset in combinatorial
geometries 310
spanning tree 12, 81, 479,
507, 519, 543, 548, 572, 574
special network 517
special path 63
Sperners theorem 54, 56, 59,
326, 484
sphere packing bound 245, 419,
425
spherical 2-distance set 269
spread 320, 330
square design 224
squared rectangle 516
squared square 507
standard tableaux 162
StantonSprott dierence sets
376
Steiner system 216, 221, 223,
303, 351, 362, 364365
Steiner triple system 233241,
276, 278
Stirling numbers 123127, 534
Stirlings formula 32, 107, 108,
187, 208
strength of a ow 62
strength of an orthogonal array
423
strong product of graphs 436
strongly connected digraph 442
strongly regular graph 261,
405, 407, 449
sub-Latin square 185
subdivision of a graph 452
subgeometry 304
subgraph 4
subplane of a projective plane
353
subsquare of a Latin square 185
subsquares of orthogonal Latin
squares 298
substitution of power series 579
substitution principle 114, 117
substructure of an incidence
structure 353
support of a codeword 249, 475
supporting hyperplane 446
surjections, number of 90, 125,
337
602 Subject Index
Sylvesters law 96, 432
symbol 182
symmetric chain 55, 57, 58
symmetric designs 224, 299, 367,
369, 396, 568
symmetric functions 535
symmetry code 252
system of distinct representatives
(SDR) 43, 100, 106108, 186
system of walks in a graph 487
t-designs 216, 223, 249253,
255, 266, 330, 405, 423
tail of a directed edge 2, 460
Tait coloring of a graph 481483
tangent 367
tennis matches 563
ternary code 244
ternary Golay code 256, 417
tight design 223, 425, 428
torus 492
total order 53
totally isotropic 397
totally unimodular matrices 510
tournament 33, 432
transitive tournament 33
translation plane 320, 331
transportation network 61,
170, 538
transversal design 273, 290, 563
tree 6, 1222, 7780, 459
triangles in a graph 27, 3739,
543, 544
triangular embedding of a graph
500, 505
triangular graph 262, 270, 275
triangular mesh 494
trierent code 258
trivalent graph 481482, 546,
551, 574
trivalent tree 139
Tur ans theorem 37
Tutte connectivity 451, 457, 458
type of a partition of a set 346,
526
type of a permutation 370,
525528
unipotent Latin square 288
unital 365
unordered partition 152
valency or valence of a vertex 4
Van der Waerden conjecture
104, 108, 186, 197
Vandermonde determinant 163
Vandermonde matrix 251
varieties 240
vertex condition 497
vertex connectivity 451
vertices of a graph 1
van der Waerden conjecture 104,
110118, 186, 197
walk in a graph 5
walks in the X, Y plane 122,
130, 138, 149, 552
weight of a codeword 244
weight enumerator of a code 248
weighted graph 17, 434
Weisners theorem 335, 339, 344
Whitney dual of a graph 472
Whitneys theorem 479, 480
Williamsons method 204, 206
WilsonPetrenjuk inequality 222
Witt design 223, 255, 256, 266
write-once memory 237
Young diagram 156
Young tableau 162167
zeta function of a poset 333
zeta function, Riemann 93, 97