GRAPH THEORY

[MTS6 B14 (E01)]

VI SEMESTER
ELECTIVE COURSE
B.Sc. MATHEMATICS
(2019 Admission onwards)
CBCSS

UNIVERSITY OF CALICUT
School of Distance Education,
Calicut University P.O.
Malappuram - 673 635, Kerala.

19573
UNIVERSITY OF CALICUT
School of Distance Education
Study Material
VI Semester
Elective Course (MTS6 B14(E01))
B.Sc. MATHEMATICS

GRAPH THEORY
Prepared by:
Sreehari. T,
Asst. Professor in Mathematics,
School of Distance Education,
University of Calicut.
Srutinized by:
Dr. Bijumon. R,
Associate. Professor & Hod, Dept. of Mathematics,
Mahathmagandhi College, Iritty.

DISCLAIMER
“The author(s) shall be solely responsible for the
content and views expressed in this book”
Contents

1 An Introduction to Graphs 1
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . 1
1.2 Graph as Models . . . . . . . . . . . . . . . . . . 3
1.3 More definitions . . . . . . . . . . . . . . . . . . . 5
1.4 Vertex degrees . . . . . . . . . . . . . . . . . . . . 8
1.5 Subgraphs . . . . . . . . . . . . . . . . . . . . . . 10
1.6 Paths and Cycles . . . . . . . . . . . . . . . . . . 14
1.7 Matrix representation of a graph . . . . . . . . . . 20

2 Trees and Connectivity 24

2.1 Definitions and Simple Properties . . . . . . . . . 24
2.2 Bridges . . . . . . . . . . . . . . . . . . . . . . . 29
2.3 Spanning Trees . . . . . . . . . . . . . . . . . . . 32
2.4 Cut Vertices and Connectivity . . . . . . . . . . . 34

3 Euler Tours and Hamiltonian Cycles 39

3.1 Euler Tours . . . . . . . . . . . . . . . . . . . . . 39
3.2 Hamiltonian Graphs . . . . . . . . . . . . . . . . . 41
3.3 Plane and Planar graphs . . . . . . . . . . . . . . . 45
3.4 Euler’s Formula . . . . . . . . . . . . . . . . . . . 47
Module 1

An Introduction to Graphs

(Reference text: ‘A First Look at Graph Theory’ by John Clark and

Derek Allan Holton)

1.1 Introduction
Graph theory is a branch of mathematics which deals the problems,
with the help of diagrams. There are may applications of graph
theory to a wide variety of subjects which include operations re-
search, physics, chemistry, computer science and other branches of
science. In this chapter we introduce some basic concepts of graph
theory and provide variety of examples. We also obtain some ele-
mentary results.

Definition 1.1.1. A graph G = (V (G), E(G)) consists of two finite

sets: V (G), the vertex set of the graph, often denoted by just V ,
which is a nonempty set of elements called vertices, and E(G), the
edge set of the graph, often denoted by just E, which is a possibly

1
empty set of elements called edges, such that each edge e in E is
assigned an unordered pair of vertices (u, v) called the end vertices
of e.

Vertices are also sometimes called points, nodes, or just dots. If

e is an edge with end vertices u and v then e is said to join u and v.
Note that the definition of a graph allows the possibility of the edge
e having identical end vertices, i.e., it is possible to have a vertex u
joined to itself by an edge - such an edge is called a loop.

Example 1.1.1. Let G = (V, E) where

V = {a, b, c, d, e}, E = {e1 , e2 , e3 , e4 , e5 , e6 , e7 , e8 }

and the ends of the edges are given by:

e1 ↔ (a, b), e2 ↔ (b, c), e3 ↔ (c, c), e4 ↔ (c, d)

e5 ↔ (b, d), e6 ↔ (d, e), e7 ↔ (b, e), e8 ↔ (b, e)

We can then represent G diagrammatically as in Figure 1.1.

Figure 1.1: A graph with 5 vertices and 8 edges

Example 1.1.2. V (G) = {v1 , v2 , v3 , v4 , v5 } and E(G) = {e1 , e2 , e3 , e4 , e5 , e6 }

and the ends of the edges are given by:

2
 e1 ↔ (v1 , v2 ), e2 ↔ (v2 , v3 ), e3 ↔ (v2 , v4 ),
e4 ↔ (v2 , v5 ), e5 ↔ (v2 , v5 ), e6 ↔ (v3 , v3 )

Figure 1.2: A graph with 5 vertices and 6 edges

1.2 Graph as Models

Problem 1.2.1. Suppose that the graph of Figure 1.3 represents a
network of telephone lines and poles. We are interested in the net-
work’s vulnerability to accidental disruption. We want to identify
those lines and poles that must stay in service to avoid disconnect-
ing the network. There is no single line whose disruption (removal)

Figure 1.3: A network of telephone lines

3
will disconnect the graph (network), but the graph will become dis-
connected if we remove the two lines represented by the edges t4
and t5 , for example. When it comes to poles, the network is more
vulnerable since there is a single vertex, vertex d, whose removal
disconnects the graph.
We may also want to find a smallest possible set of edges needed
to connect the six vertices. There are several examples of such
minimal sets. One is

{t1 , t3 , t5 , t6 , t7 } .

Problem 1.2.2. Suppose that we have five people A, B, C, D, E

and five jobs a, b, c, d, e and some of these people are qualified for
certain jobs. Is there a feasible way of allocating one job to each
person, or to show that no such matching up of jobs and people is
possible. We can represent this situation by a graph having a vertex
for each person and a vertex for each job, and edges joining people
up to jobs for which they are qualified. Does there exist a feasible
matching of people to jobs for the graph of Figure 1.4?

Figure 1.4: A job applications graph

4
 The answer is no. The reason can be found by considering peo-
ple A, B, and D. These three people as a set are collectively qual-
ified for only two jobs, c and d, hence there is no feasible matching
possible for these three people, much less all five people.

1.3 More definitions

Definition 1.3.1. Let G be a graph. If two (or more) edges of G
have the same end vertices then these edges are called parallel.

For example, the edges e4 and e5 of the graph of Figure 1.2 are
parallel.

Definition 1.3.2. A vertex of G which is not the end of any edge is

called isolated. Two vertices which are joined by an edge are said
to be adjacent or neighbours. The set of all neighbours of a fixed
vertex v of G is called the neighbourhood set of v and is denoted
by N(v).

Thus, in the graph of Figure 1.2, v1 and v5 are adjacent but v1

and v2 are not. The neighbourhood set N (v2 ) of v2 is {v5 , v3 , v4 }.

Definition 1.3.3. A graph is called simple if it has no loops and no

parallel edges.

Example 1.3.1. Let G be graph with 4 vertices and 3 edges given

below in Figure 1.5. Which is a simple graph since it has no loops
and parallel edges.

Definition 1.3.4. A graph G1 = (V1 , E1 ) is said to be isomorphic

to the graph G2 = (V2 , E2 ) if there is a one-to-one correspondence

5
 Figure 1.5: Simple graph

between the vertex sets V1 and V2 and a one-to-one correspondence

between the edge sets E1 and E2 in such a way that if e1 is an edge
with end vertices u1 and v1 in G1 then the corresponding edge e2 in
G2 has its end points the vertices u2 and v2 in G2 which correspond
to u1 and v1 respectively. Such a pair of correspondences is called
a graph isomorphism.

Example 1.3.2. Let G1 = (V1 , E1 ) and G2 = (v2 , E2 ) be two

graphs with vertex set V1 = {e1 , e2 , e3 , e4 , e5 } and
V2 = {c1 , c2 , c3 , c4 , c5 } respectively. Their Adjacency is given in
the Figure 1.6. We can find and one-one correspondence between
the vertices and edges of the graph G1 and G2 such a way that
e1 ↔ c1 e2 ↔ c3 e3 ↔ c5 e4 ↔ c2 e5 ↔ c4
This maps preserves adjacency also.

Definition 1.3.5. A simple graph G is said to be complete if every

pair of distinct vertices of G are adjacent in G: Any two complete
graphs each on a set of n vertices are isomorphic and each such
graph is denoted by Kn .

Example 1.3.3. Examples for complete graphs :

6
 Figure 1.6: Isomorphic graphs with 5 vertices

Figure 1.7: Complete Graph K3 and K4

Definition 1.3.6. An empty graph (or trivial) is a graph with no

edges. Let G be a graph. If the vertex set V of G can be partitioned
into two nonempty subsets X and Y (i.e., X ∪ Y = V and X ∩ Y =
∅) in such a way that each edge of G has one end in X and one end
in Y then G is called bipartite. The partition V = X ∪ Y is called
a bipartition of G
A Complete bipartite graph is a simple bipartite graph G, with
bipartition V = X ∪ Y , in which every vertex in X is joined to
every vertex of Y . If X has m vertices and Y has n vertices, such
a graph is denoted by Km,n

Example 1.3.4. Example of complete bipartite graph

7
1.4 Vertex degrees
Definition 1.4.1. An edge e of a graph G is said to be incident with
the vertex v if v is an end vertex of e. In this case we also say that
v is incident with e. Two edges e and f which are incident with a
common vertex v are said to be adjacent.
Let v be a vertex of the graph G. The degree d(v) (or dG (v) if
we want to emphasize G ) of v is the number of edges of G incident
with v, counting each loop twice, i.e., it is the number of times v is
an end vertex of an edge.

Example 1.4.1. Let G be a graph with 5 vertices 7 edges given

below.

Figure 1.8: The graph G

In the graph of Figure 1.8 we have d (v1 ) = 3, d (v2 ) = 4,

8
d (v3 ) = 3, d (v4 ) = 3, d (v5 ) = 1 and note that in this example

d (v1 ) + d (v2 ) + d (v3 ) + d (v4 ) + d (v5 )

= 14
= 2 × ( number of edges in G)

Theorem 1.4.1. Euler :The sum of the degrees of the vertices of a

graph is equal to twice the number of its edges.

Proof. Let the degrees of the vertices be d1 , d2 , . . . , dn . Let n =

Pn
i=1 di and m denote the total number of edges in the graph. No-
tice that each edge contributes 2 to the sum of degrees of the ver-
tices, thus n = 2m.

Corollary 1.4.1.1. In any graph G the number of vertices of odd

degree is even.

Proof. Let V1 and V2 be the subsets of vertices of graph G with ver-

tices of odd and even degrees respectively. Then from the previous
theorem, we have

2m = n
P
= v∈V dG (v) (1.1)
P P
= v∈V1 dG (v) + v∈V2 dG (v)
P P
Both 2m and v∈V2 dG (v) are even, which forces v∈V1 dG (v) to
be even.

Definition 1.4.2. If for some positive integer k, d(v) = k for every

vertex v of the graph G then G is called k-regular. A regular graph
is one that is k-regular for some k.

9
1.5 Subgraphs
Definition 1.5.1. Let H be a graph with vertex set V (H) and edge
set E(H) and, similarly, let G be a graph with vertex set V (G)
and edge set E(G). Then we say that H is a subgraph of G if
V (H) ⊆ V (G) and E(H) ⊆ E(G). In such a case, we also say
that G is a supergraph of H.

Example 1.5.1. For example in Figure 1.9, G1 is a subgraph of

both G2 and G3 but G3 is not a subgraph of G2 .

Figure 1.9: G1 ⊆ G2 , G1 ⊆ G3 and G3 not subgraph of G2

Definition 1.5.2. A subgraph H of a graph G is a proper subgraph

of G if either V (H) 6= V (G) or E(H) 6= E(G). A subgraph H of
G is a spanning subgraph of G if V (H) = V (G).

From the definition we can get that ant simple graph with n
vertices is a proper subgraph of complete graph Kn .
In Figure 1.7, G1 is a proper subgraph of G2 . G1 is a spanning
subgraph of G3 , since both the graph have same vertex set.

Definition 1.5.3. If G = (V, E) and V has at least two elements

(i.e., G has at least two vertices), then for any vertex v of G, G − v

10
denotes the subgraph of G with vertex set V −{v} and whose edges
are all those of G which are not incident with v, i.e., G − v is
obtained from G by removing v and all the edges of G which have
v as an end. G − v is referred to as a vertex deleted subgraph. If
G = (V, E) and e is an edge of G then G − e denotes the subgraph
of G having V as its vertex set and E−{e} as its edge set, i.e., G−e
is obtained from G by removing the edge e, (but not its endpoint(s)).
G − e is referred to as an edge deleted subgraph.

Example 1.5.2. Example of vertex deleted graph

Figure 1.10: A vertex deleted graph G − v6

We extend the above definition to cater for the deletion of sev-

eral vertices or edges.

Definition 1.5.4. If G = (V, E) and U is a proper subset of V then

G − U denotes the subgraph of G with vertex set V − U and whose
edges are all those of G which are not incident with any vertex in
U.
If F is a subset of the edge set E then G − F denotes the sub-
graph of G | with vertex set V and edge set E − F , i.e., obtained
by deleting all the edges in F , but not their endpoints. G − U and

11
G − F are also referred to as vertex deleted subgraph and edge
deleted subgraph(respectively).

Definition 1.5.5. By deleting from a graph G all loops and in each

collection of parallel edges all edges but one in the collection we
obtain a simple spanning subgraph of G, called the underlying
simple graph of G.

Definition 1.5.6. If U is a nonempty subset of the vertex set V of

the graph G then the subgraph G[U ] of G induced by U is defined
to be the graph having vertex set U and edge set consisting of those
edges of G that have both ends in U Similarly if F is a nonempty
subset of the edge set E of G then the subgraph G[F ] of G induced
by F is the graph whose vertex set is the set of ends of edges in F
and whose edge set is F .

Example 1.5.3. Let G = K4 with vertex set V (G) = {v1 , v2 , v3 , v4 }

and let U = {v1 , v2 , v3 }. Then the induced graph of G by U , that
is G[U ] is given in below figure.

v2 v3 v2 v3

v1 v4 v1

G = (V, E) G[U ]

Definition 1.5.7. Two subgraphs G1 and G2 of a graph G are said

to be disjoint if they have no vertex in common, and edge disjoint
if they have no edge in common.

12
Example 1.5.4. Let G = K4 with vertex set V (G) = {v1 , v2 , v3 , v4 }
and G1 and G2 are subgraphs of G with vertex set V1 (G) = {v1 , v2 }
and V2 = {v3 , v4 } respectively whose picture is given below.

v2 v3 v2 v3

v1 v4 v1 v4

G = (V, E) G1 G2
Here the graph G1 and G2 are disjoint graphs. Since they have
no common vertex and also they are edge disjoint also.

Definition 1.5.8. Given two subgraphs G1 and G2 of G, the union

G1 ∪ G2 is the subgraph of G with vertex set consisting of all those
vertices which are in either G1 or G2 (or both) and with edge set
consisting of all those edges which are in either G1 or G2 (or both);
symbolically

V (G1 ∪ G2 ) = V (G1 ) ∪ V (G2 )

E (G1 ∪ G2 ) = E (G1 ) ∪ E (G2 )

Definition 1.5.9. If G1 and G2 are two subgraphs of G with at least

one vertex in common then the intersection G1 ∩ G2 is given by

V (G1 ∩ G2 ) = V (G1 ) ∩ V (G2 )

E (G1 ∩ G2 ) = E (G1 ) ∩ E (G2 )

Example 1.5.5. Let G = K4 be a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 } and G1 and G2 are subgraphs of G given below.

13
 v2 v3 v2 v3 v2 v3

v1 v4 v1 v4 v1 v4

G = (V, E) G1 G2
Then the union and intersection of graphs G1 ∪ G2 and G1 ∩ G2
is given below.

v2 v3 v2 v3

v1 v4 v1 v4

G1 ∪ G2 G1 ∩ G2

1.6 Paths and Cycles

Definition 1.6.1. A walk in a graph G is an alternating sequence
W : v0 e1 v1 e2 v2 , . . . , en vp of vertices and edges, beginning and
ending with vertices in which vi−1 and vi are the ends of ei . v0
is the origin and vp is the terminus of W and the walk W is said
to join v0 and vp . If the graph G is simple, then a walk is deter-
mined by the sequence of its vertices. The walk is closed if v0 = vp
and otherwise it is open . A walk is called a trail if all the edges
appearing in the walk are distinct and it is called a path if all the
vertices are distinct. Thus, every path in G is automatically a trail
in G. When writing a path, we usually omit the edges. A cycle is a

14
closed trail in which the vertices are all distinct. A cycle is odd or
even depending on whether its length is odd or even. The length of
a walk is the number of edges in it. A walk of length 0 consists of
just a single vertex.

Clearly any two paths with the same number of vertices are
isomorphic. A path with n vertices will sometimes be denoted by
Pn . Note that Pn has length n − 1.

Theorem 1.6.1. Given any two vertices u and v of a graph G, every

u − v walk contains a u − v path, i.e., given any walk

W = ue1 v1 . . . vk−1 ek v

then, after some deletion of vertices and edges if necessary, we can

find a sub sequence P of W which is a u − v path.

Proof. If u = v, i.e., if W is closed, then the trivial path P = u

will do. Now suppose u 6= v, i.e., W is open and let the vertices of
W ke given, in order, by

u = u0 , u1 , u2 , . . . , uk−1 , uk = v

If none of the vertices of G occurs in W more than once then W is

already a u − v path and so we are finished by taking P = W
So now suppose that there are vertices of G that occur in W
twice or more. Then there are distinct i, j, with i < j, say, such that
ui = uj . If the terms ui , ui+1 , . . . , uj−1 (and the preceding edges)
are deleted from W then we obtain a u − v walk W1 having fewer
vertices than W. If there is no repetition of vertices in W1 , then W1
is a u − v path and setting P = W1 finishes the proof.

15
 If this is not the case, then we repeat the above deletion pro-
cedure until finally arriving at a u − v walk that is a path, as re-
quired.

Definition 1.6.2. Let G be a graph. Two vertices u and v in G are

said to be connected if there exits a u − v path in G.

Remark: The relation ‘connected’ forms a equivalence relation

on V (G).

Definition 1.6.3. A graph G is called connected if every two of

its vertices are connected. A graph that is not connected is called
disconnected. Given any vertex u of a graph G, let C(u) denote the
set of all vertices in G that are connected to u. Then the subgraph of
G induced by C(u) is called the connected component containing
u, or simply the component containing u.

Remark: If u and v are two connected vertices in the graph

G, i.e., if there is a path from u to v, then, by the remarks above,
C(u) = C(v) and so u and v have the same connected component.
Conversely, if u and v have the same component then v is in C(u)
so v and u must be connected.

Example 1.6.1. Let G = (V, E) be a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 , v5 , v6 , v7 , v8 } is given below.

16
 v1 v4 v5 v6

v2 v3 v8 v7

Figure 1.11: A graph with 3 connected components

The graph G has 3 components C(v1 ),C(v4 ) and C(v5 ).

Definition 1.6.4. The number of components of G is called ω(G).

Thus ω(G) = 3 of a graph G in the Fig 1.11.

Definition 1.6.5. A nontrivial closed trail in a graph G is called a

cycle if its origin and internal vertices are distinct. In detail, the
closed trail C = v1 v2 . . . vn v1 is a cycle if
(i) C has at least 1 edge and v1 , v2 , . . . , vn are n distinct vertices.
A cycle of length k, i.e., with k edges, is called a k-cycle. A k-cycle
is called odd or even depending on whether k is odd or even.
A 3-cycle is often called a triangle. Clearly any two cycles of the
same length are isomorphic. An n-cycle, i.e., a cycle with n ver-
tices, will sometimes be denoted by Cn .

Example 1.6.2. Let G = (V, E) be graph with the vertex set V (G) =
{v1 , v2 , v3 , v4 } is given in Fig 1.12.

17
 v1 v2

v4 v3

Figure 1.12: Cycle of length 4

Which is a cycle of length 4 and C = v1 v2 v3 v4 v1

Theorem 1.6.2. Let G be a nonempty graph with at least two ver-

tices. Then G is bipartite if and only if it has no odd cycles.

Proof. Suppose that G is bipartite with vertex set V and bipartition

V = X ∪ Y . Let C = v0 v1 . . . vk v0 be a cycle of G. For the sake
of argument, assume that v0 is in X. Then, because G is bipartite,
v1 must be in Y . Similarly v2 must be in X, v3 in Y , etc. In fact,
in general, the odd-indexed vertices v2i+1 must be in Y while the
even-indexed vertices v2i must be in X. Now, since v0 is in X, we
must have (at the ‘other end’ of the cycle) vk in Y . Hence k must
be an odd number. Thus the cycle C = v0 v1 . . . vk v0 is even. Since
C was any cycle in the graph G, G has no odd cycles.
Now, to prove the converse, we assume that G is a nonempty graph
which has no odd cycles. We wish to show that G is bipartite. Now
G will be bipartite if each of its nonempty connected components
is bipartite, since, if these components are C1 , C2 , C3 , . . . , Cn and
their vertex sets V1 , V2 , V3 , . . . , Vn have bipartitions V1 = X1 ∪
Y1 , . . . , Vn = Xn ∪ Yn , then the vertex set V of G has bipartition

18
V = X ∪ Y where

X = X0 ∪ X1 ∪ X2 ∪ . . . ∪ Xn and Y = Y1 ∪ Y2 ∪ . . . ∪ Yn

where X0 is the set of isolated vertices in G. As a result of this it

is enough to show that if G is a nonempty connected graph with no
odd cycles then G is bipartite.
With this assumption, let u be a fixed vertex of G. We define
two subsets of the vertex set V of G as follows: X is the set of all
vertices v of G with the property that any shortest u − v path of G
has even length, Y is the set of all vertices w of G with the property
that any shortest u − w path of G has odd length, i.e., X consists
of those vertices of G an ‘even distance’ from u, while Y consists
of those vertices of G an ‘odd distance’ from u.
Note that u itself is in X. Then, clearly, V = X ∪ Y and X
and Y have no element in common. We show that V = X ∪ Y is a
bipartition of G by showing that any edge of G must have one end
in X and the other in Y .
Let v and w be two vertices both in X and assume they are
adjacent. Let P and Q be a shortest u − v path and a shortest u − w
path respectively, say

P = u1 , u2 , . . . , u2n+1 and Q = w1 , w2 , . . . , w2m+1

(so that u = u1 = w1 , v = u2n+1 and w = w2m+1 .) Suppose that

w0 is a vertex that the two paths have in common, and further that
w0 is the last such vertex. (If v = w then of course w0 = v =
w. Moreover, there is always such a vertex w0 since the vertex u
is common to both paths.) Then that part of P from u to w0 is

19
a shortest path from u to w0 and that part of Q from u to w0 is
a shortest path from u to w0 also. In other words, we have two
shortest paths from u to w0 . It follows that, since these two paths
have the same length, there exists an i such that w0 = ui = wi .
However this produces an odd cycle in G :

C = ui ui+1 · · · u2n+1 w2m+1 w2m . . . wi

| {z }| {z }
∗ ∗∗

since if i is odd then the above parts ∗ and ∗∗ are both of even
length while if i is even then they are both of odd length, giving the
total length of C as odd +1+ odd or even +1+ even, in either case
odd. Since G has no odd cycles, the assumption that v and w are
adjacent is wrong.
Hence there are no edges in G joining vertices of X. A similar
argument shows that there are no edges of G joining vertices of Y.
Hence G is bipartite, as required.

1.7 Matrix representation of a graph

Definition 1.7.1. Let G be a graph with n vertices, listed as v1 ,v2 ,v3
. . ., vn . The adjacency matrix of G, with respect to this particular
listing of the n vertices of G, is the n×n matrix A(G) = (aij ) where
the (i, j) th entry aij is the number of edges joining the vertex vi to
the vertex vj .

Example 1.7.1. Let G = (V, G) be graph with vertex set V (G) =

{v1 , v2 , v3 , v4 } is given in Figure 1.13. The Adjacency matrix A(G)
of graph G also listed below.

20
 v1 v2

v4 v3

Figure 1.13: Cycle of length 4

 
0 1 0 1
 
1 0 1 0
 
0 1 0 1
 
1 0 1 0

Figure 1.14: A(G): a 4 × 4 matrix

Theorem 1.7.1. Let G be a graph with n vertices v1 , . . . , vn and

let A denote the adjacency matrix of G with respect to this listing
of the vertices. Let k be any positive integer and let Ak denote the
matrix multiplication of k copies of A. Then the (i, j) th entry of
Ak is the number of different vi − vj walks in G of length k.

Theorem 1.7.2. Let G be a graph with n vertices v1 , . . . , vn and

let A denote the adjacency matrix of G with respect to this listing
of the vertices. Let B = (bij ) be the matrix

B = A + A2 + · · · + An−1

Then G is a connected graph if and only if for every pair of distinct

indices i, j we have bij 6= 0, i.e., if and only if B has no zero entries

21
off the main diagonal.

(k)
Proof. Let aij denote the (i, j) th entry of the matrix Ak for each
k = 1, . . . , n − 1 Then

(1) (2) (n−1)

bij = aij + aij + · · · + aij

(k)
However, by Theorem 1.5, aij denotes the number of distinct walks
of length k from vi to vj and so i.e., bij is the number of different
vi − vj walks of length less than n. Now suppose that G is con-
nected. Then for every pair of distinct indices i, j there is a path
from vi to vj . Since G has only n vertices this path goes through at
most n vertices and so it has length less than n, i.e., there is at least
1 path from vi to vj of length less than n. Hence bij 6= 0 for each
i, j with i 6= j, as required.
Conversely, suppose that for each distinct pair i, j we have bij 6=
0. Then, from above, there is at least 1 walk (of length less than
n) from vi to vj . In particular, vi is connected to vj . Thus G is a
connected graph, as required, since i and j were an arbitrary pair
of distinct vertices.

Definition 1.7.2. Suppose that G has n vertices, listed as v1 , . . . , vn ,

and t edges, listed as e1 , . . . , et . The incidence matrix of G, with
respect to these particular listings of the vertices and edges of G, is
the n × t matrix M (G) = (mij ) where mij is the number of times
that the vertex vi is incident with the edge ej , i.e.,

22
 


0 if vi is not an end of ej

mij = 1 if vi is an end of the non-loop ej



2 if vi is an end of the loop ej

Example 1.7.2. Let G = (v, G) be graph with vertex set V (G) =

{v1 , v2 , v3 , v4 } is given in Figure 1.15. The incidence matrix M (G)
of graph G also listed below.

v1 v2

v4 v3

Figure 1.15: A graph with 4 vertices and 3 edges

 
1 1 0
 
1 0 0 
 
0 0 1 
 
0 1 1

Figure 1.16: M (G): a 4 × 3 matrix

23
Module 2

Trees and Connectivity

2.1 Definitions and Simple Properties

Definition 2.1.1. A Graph G is called acyclic if it contains no cy-
cles.

Definition 2.1.2. A Graph G is called a Tree if it is a connected

acyclic graph.

Example 2.1.1. Let G = (V, E) is a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 } given in Figure 2.1.

v1 v2

v4 v3

Figure 2.1: Acyclic graph

24
 The graph G do not contain any cycles. Hence G is a acyclic
graph.

Example 2.1.2. Let G = (V, E) is a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 } given in Figure 2.2.

v1 v2

v4 v3

Figure 2.2: A Tree with 4 vertices

The graph G is a connected graph also it do not contain a cycle.

Hence it is a tree.

Theorem 2.1.1. 1. Let u and v be distinct vertices of a tree T.

Then there is precisely one path from u to v.

2. Let G be a graph without any loops. If for every pair of

distinct vertices u and v of G there is precisely one path from
u to v, then G is a tree.

Proof. 1. Suppose that the result is false. Then there are two
different paths from u to v, say P = uu1 u2 . . . um v and P 0 =
uv1 v2 . . . vn v. Let w be the first vertex after u which belongs
to both P and P 0 . (The vertex w might well be v, but at least
there is such a vertex) Then w = ui = vj for some indices i
and j. (See Figure 2.3.)

25
 u1 u3 u5

u u2 w v

v1 v2 v4

Figure 2.3: A Tree with 4 vertices

This produces a cycle C = uu1 . . . ui vj−1 . . . v1 u, since no

two of the 0 u0 terms are repeated (since P is a path), no two
of the 0 v 0 terms are repeated (since P 0 is a path), and, by the
definition of w, no“ u ” term in C is the same as a vv 00 term.
Since T is a tree, T bas no cycles. This contradiction means
that our initial assumption (the first sentence of the proof)
must be false. Thus there is precisely one path from u to v.

2. Since, by assumption, there is a path between each pair of

vertices u and v, G must be connected. Thus, to show that
G is a tree it remains to show that G has no cycles. Firstly,
since G has no loops it has no cycles of length one. Now
suppose that G has a cycle of length greater than one, say
C = v1 v2 . . . vn v1 where n ≥ 2. Since, by definition, a cycle
is a trail, the edge vn v1 does not appear in the path v1 v2 . . . vn .
Thus P = v1 vn and P 0 = v1 v2 . . . vn are two different paths
from v1 to vn . This contradicts our assumptions. Hence G
has no cycles and so is a tree.

Theorem 2.1.2. Let T be a tree with at least two vertices and let
P = u0 u1 . . . un be a longest path in T (so that there is no path in

26
T of length greater than n ). Then both u0 and un have degree 1,
i.e., d (u0 ) = 1 = d (un )

Proof. Suppose that d (u0 ) > 1. The edge f = u0 u1 contributes

1 to the degree of u0 and so there must be another edge e from u0
to a vertex v of T (which is different: from f ). If this vertex v is
one of the vertices of the path P then we may set v = ui for some
i = 0, 1, . . . , n and this produces a cycle C = u0 u1 . . . ui u0 (the
last edge being e ). Since T is a tree it has no cycles and so this
is a contradiction. Thus the remaining possibility for v is that it is
not one of the vertices of the path P . But then P1 = vu0 u1 . . . un ,
where the first edge is e2 gives a path of length n + 1 in T , in
contradiction to our assumption that P is a longest path (and of
length n ). This final contradiction shows that there is no such
edge e and so d (u0 ) = 1, as required. Similarly d (un ) = 1, as
required.

Corollary 2.1.2.1. Any tree T with at least two vertices has more
than one vertex of degree 1.

Proof. In such a tree T there is a longest path P (of length greater

than 0 ) and so the Theorem produces at least two vertices of degree
1.

Theorem 2.1.3. If T is a tree with n vertices then it has precisely

n − 1 edges.

Proof. We use induction on n. When n = 1, i.e., T has only 1

vertex, then, since it has no loops, T can not have any edges, i.e.,
it has n − 1 = 0 edges. This establishes that the result is true for
n=1

27
 Now suppose that the result is true for n = k where k is some
positive integer. We use this to show that the result is true for n =
k + 1. Let T be a tree with k + 1 vertices and let u be a vertex of
degree 1 in T . (Note that such a vertex exists by Corollary 2.3.) Let
e = uv denote the unique edge of T which has u as an end. Then if
x and y are vertices in T both different from u, any path P joining x
to y does not go through the vertex u since if it did it would involve
the edge e twice. Thus the subgraph T − u, obtained from T by
deleting the vertex u (and the edge e ), is connected. Moreover if C
is a cycle in T − u then C would be a cycle in T - impossible, since
T is a tree. Thus the subgraph T − u is also acyclic. Hence T − u
is a tree. However T − u has k vertices (since T has k + 1vertices)
and so, by our induction assumption, T − u has k − 1 edges. Since
T − u has exactly 1 edge less than T (the edge e ), it follows that T
has k edges, as required. In other words, assuming the result is true
for k, we have shown that it is true for k + 1. Thus, by the principle
of mathematical induction, it is true for all positive integers k.

Note: An acyclic graph is also called Forest.

Theorem 2.1.4. Let G be an acyclic graph with n vertices and k

connected components, i.e., ω(G) = k. Then G has n − k edges.

Proof. Denote the k components of G by C1 , C2 , . . . , Ck and sup-

pose that for each i, 1 ≤ i ≤ k, the i th component Ci has ni
vertices. Then n = n1 + n2 + · · · + nk Also, since each Ci is a
tree, by Theorem 2.1.3 it has ni − 1 edges and so since each edge
of G belongs to precisely one component of G the total number of
edges in G is (n1 − 1) + (n2 − 1) + · · · + (nk − 1) . Thus G has
(n1 + n2 + · · · + nk ) − k edges, i.e., n − k edges, as required.

28
2.2 Bridges
Theorem 2.2.1. Let e be an edge of the graph G and, as usual,
let G − e be the subgraph obtained by deleting e. Then ω(G) ≤
ω(G − e) ≤ ω(G) + 1.

Definition 2.2.1. An edge e of a graph G is called a bridge (or a

cut edge or an isthmus) if the subgraph G − e has more connected
components than G has.

Example 2.2.1. Let G = (V, E) be a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 , v5 } is given in figure 2.4

v1 v1

v2 v3 v2 v3

v4 v5 v4 v5

Figure 2.4: A Graph G and G − e1

Here the edge e1 = v1 v2 is a bridge, since the graph G with

removal of the edge e1 has more components than G. That is, G−e1
has 2 components but G has only one component.

Theorem 2.2.2. An edge e of a graph G is a bridge if and only if e

is not part of any cycle in G.

29
Proof. Let e have end vertices u and v. If e is not a bridge then,
by the above remarks, it is either a loop or there is a path P =
uu1 . . . un v from u to v, different from the edge e. If it is a loop
then it forms a cycle (by itself). If there is such a path P then
C = uu1 . . . un vu, the concatenation of P with e, is a cycle in G.
This shows that if e is not a bridge then it is part of a cycle. This is
equivalent to saying that if e is not part of any cycle then e must be
a bridge.
Conversely, suppose that e is part of some cycle C = u0 u1 . . . um
in G. Let e = ui ui+1 . In the case where m = 1, C = u0 u1 and so C
is just the edge e and e is a loop. On the other hand, if m > 1 then
P = ui ui−1 . . . u0 um−1 . . . ui+1 is a path from u to v different from
e. (See Figure 2.5.) Thus, by the remarks preceding the proof, e is
not a bridge. This shows that if e is a bridge then it is not part of
any cycle in G, completing the proof.

Figure 2.5:

Theorem 2.2.3. Let G be a connected graph. Then G is a tree if

and only if every edge of G is a bridge, i.e., if and only if for every
edge e of G the subgraph G − e has two components.

Proof. Suppose that G is a tree. Then G is acyclic, i.e., it has no

30
cycles, and so no edge of G belongs to a cycle. In other words, if e
is any edge of G then, by Theorem 2.2.2, it is a bridge, as required.
Conversely suppose that G is connected and that every edge e of
G is a bridge. Then G can have no cycles since any edge belonging
to a cycle is not a bridge, by Theorem 2.2.2. Hence G is acyclic and
so is a tree as required.

Corollary 2.2.3.1. A connected graph G with n vertices has at

least n − 1 edges.

Theorem 2.2.4. Let G be a graph with n vertices. Then the follow-

ing three statements are equivalent:
(i) G is a tree,
(ii) G is an acyclic graph with n − 1 edges,
(iii) G is a connected graph with n − 1 edges.

Proof. We prove (i) ⇒ (ii) ⇒ (iii) ⇒ (i).

(i) ⇒ (ii): Suppose that G is a tree. Then by definition G is an
acyclic graph and by Theorem 2.1.3 it has n − 1 edges. Thus (ii)
holds.
(ii) ⇒ (iii): Assume that G is an acyclic graph with n−1 edges and,
as usual, let ω(G) denote the number of connected components of
G. Then, by Theorem 2.1.4, G has n−ω(G) edges. Thus ω(G) = 1,
in other words, G is connected. This establishes (iii).
(iii) ⇒ ( i): Assume that G is a connected graph with n − 1 edges.
To prove (i), i.e., that G is a tree, we must show that G is acyclic.
We do this by contradiction. Thus, assume that G is not acyclic.
Then G contains a cycle and every edge of this cycle can not be a
bridge, by Theorem 2.2.2. Choose such an edge e. Then, since e is

31
not a bridge, G − e is still connected. However G − e has n − 2
edges and n vertices, which is impossible by the above corollary.
This contradiction has arisen from our assumption that G is not
acyclic. Hence G is acyclic and so a tree as required.

2.3 Spanning Trees

Definition 2.3.1. A spanning tree of a graph G is a spanning sub-
graph of G that is a tree.

Theorem 2.3.1. A graph G is connected if and only if it has a

spanning tree.

Proof. Suppose that G is connected with n vertices and q edges.

Then, by Corollary 2.2.3.1, we have q ≥ n − 1. If q = n − 1 then,
by (iii) ⇒ ( i) of Theorem 2.2.4, G is a tree and so we can take
T = G as a spanning tree of G.
If q > n − 1 then, by Theorem 2.1.3 (or by (i) ⇒ (iii) of The-
orem 2.2.4), G is not a tree and so G must contain a cycle. Let e1
be an edge of such a cycle. Then the subgraph G − e1 is connected
(since e1 is not a bridge), has n vertices, and has q − 1 edges. If
q − 1 = −1 then, repeating the above argument gives T = G − e1
as a spanning tree of G.
If q − 1 > n − 1 then G − e1 is not a tree so, as before, there
is a cycle in G − e1 . Removing an edge e2 from such a cycle gives
a subgraph G − {e1 , e2 } = (G − e1 ) − e2 which is connected, has
n vertices and q − 2 edges. We keep on repeating this process,
deleting q−n+1 edges altogether, to eventually produce a subgraph
T which is connected, has n vertices and q − (q − n + 1) = n − 1

32
edges. Thus by Theorem 2.2,4, T is a tree and since it has the same
vertex set as G it is a spanning tree of G.
Conversely, if G has a spanning subtree T , then given any two
vertices u and v of G then u and v are also vertices of the connected
subgraph T . Thus u and v are connected by a path in T and so by
a path in G. This shows that G is connected.

Example 2.3.1. Let G = (V, E) be a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 , v5 } is given in Figure 2.6

v1 v1

v2 v3 v2 v3

v4 v5 v4 v5

Figure 2.6: A Graph G and a spanning subgraph H

The graph H is a spanning subgraph of G and also tree. Hence

it’s a spanning tree.

Example 2.3.2. Figure 2.7 illustrates a graph K4 and all it’s dif-
ferent spanning subgraphs.

Theorem 2.3.2. (Cayley, 1889) The complete graph Kn has nn−2

different spanning trees.

33
 Figure 2.7: K4 and it’s 16 different spanning graphs

2.4 Cut Vertices and Connectivity

Definition 2.4.1. A vertex v of a graph G is called a cut vertex (or
articulation point) of G if ω(G − v) > ω(G).

Example 2.4.1. Let G = (V, E) be a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 , v5 } is given in figure 2.8.
Here the vertex v4 is a cut vertex. Since ω(G − v4 ) = 2 and
ω(G) = 1. That is, G − v4 has more components than G.

Theorem 2.4.1. Let v be a vertex of the connected graph G. Then

v is a cut vertex of G if and only if there are two vertices u and w
of G, both different from v, such that v is on every u − w path in G.

Proof. First let v be a cut vertex of G. Then G − v is disconnected

and so there are vertices u and w of G which lie in different com-
ponents of G − v. Thus, although there is a path in G from u to w,

34
 v1 v1

v2 v3 v2 v3

v4 v5 v5

Figure 2.8: A Graph G and G − v4

there is no such path in G − v. This implies that every path in G

from u to w contains the vertex v, as required.
Conversely, suppose that u and w are two vertices of G, differ-
ent from v, such that every path in G from u to w contains v. Then
there can be no path from u to w in G − v. Thus G − v is discon-
nected (with u and w lying in different components). Hence v is a
cut vertex, as required.

Theorem 2.4.2. Let G be a graph with n vertices, where n ≥ 2.

Then G has at least two vertices which are not cut vertices.

Proof. Clearly we may suppose that G is a connected graph. We

proceed by assuming the result is false for our G and so the proof
will be complete if we derive a contradiction from this.
Thus we are assuming that there is at most one vertex in G
which is not a cut vertex. Now let u, v be vertices in G such that
the distance d(u, v) between them is the greatest of distances be-
tween pairs of vertices in G, i.e., d(u, v) = diam(G). Since G

35
is connected and has at least two vertices, u 6= v. Thus, by our
assumption, one of these two vertices must be a cut vertex, say v.
Then G − v is disconnected and so there is a vertex w in G which
does not belong to the same component as u does in G − v. This
implies that every uw path in G contains the vertex v.
It follows from this that the shortest path in G from u to w con-
tains the shortest path from u to v and this contradiction completes
our proof.

Definition 2.4.2. Let G be a simple graph. The vertex connectivity

of G, denoted by κ(G), is the smallest number of vertices in G
whose deletion from G leaves either a disconnected graph or K1 .

Example 2.4.2. Let G = (V, E) be a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 , v5 } is given in Figure 2.9.

v1

v2 v3

v4 v5

Figure 2.9: A Graph G with 5 vertices

Here vertex connectivity of G is one. Since the vertex v4 is a

cut vertex of the graph G deletion of that vertex results in a discon-
nected graph. Hence κ(G) = 1.

36
Example 2.4.3. Let G = (V, E) be a graph with vertex set V (G) =
{v1 , v2 , v3 , v4 , v5 } is given in figure 2.8.

v1

v2 v3

v4 v5

Figure 2.10: G = C5

Here vertex connectivity of C5 is two. Since it has no cut vertex

therefore κ(G) > 1. Also removal of the vertices v2 and v3 result’s
in a disconnected graph. Hence κ(G) = 2.

Definition 2.4.3. A simple graph G is called n-connected (where

n ≥ 1) if κ(G) ≥ n.

Note: It follows that G is 1 -connected if and only if G is con-

nected and has at least two vertices. Moreover G is 2 -connected
if and only if G is connected with at least three vertices but no cut
vertices.

Definition 2.4.4. Let u and v be two vertices of a graph G. A


collection P(1) , . . . , P(n) of u − v paths is said to be internally
disjoint if, given any distinct pair P(i) and P(j) in the collection, u
and v are the only vertices P(i) and P(j) have in common.

37
Theorem 2.4.3. (Whitney, 1932) Let G be a simple graph with
at least three vertices. Then G is 2-connected if and only if for
each pair of distinct vertices u and v of G there are two internally
disjoint u − v paths in G.

38
Module 3

Euler Tours and Hamiltonian

Cycles

3.1 Euler Tours

Definition 3.1.1. A trail in G is called an Euler trail if it includes
every edge of G.

Definition 3.1.2. A tour of G is a closed walk of G which includes

every edge of G at least once.
An Euler tour of G is a tour which includes each edge of G
exactly once.

Definition 3.1.3. A graph G is called Eulerian or Euler if it has an

Euler tour.

Example 3.1.1. Let G = (V, E) be a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 , v5 } is given in Figure 3.1.

39
 v1

v2 v3

v4 v5

Figure 3.1: Eulerian graph G

The trail v2 − v1 − v3 − v2 − v4 − v5 − v3 is a euler trail since it’s

contain all the edges of G exactly once. v2 −v1 −v3 −v2 −v4 −v5 −v3
is a euler tour in G. Hence G is a eulerian graph.

Theorem 3.1.1. Let G be a graph in which the degree of every

vertex is at least two. Then G contains a cycle.

Proof. If G is not simple then it contains a cycle since any loop is

a cycle of length 1 while a pair of parallel edges gives a cycle of
length 2 . We now suppose that G is simple. Let v0 be any vertex
of G. Since d (v0 ) ≥ 2, we can choose an edge e1 with one end v0
and the other v1 , say. Since d (v1 ) ≥ 2 we can choose an edge e2
with one end v1 and the other v2 , say, different from v0 . We repeat
this process so that (see Figure 3.2) at the (i + 1) th stage we have
an edge ei incident with vi and vi+1 and v1+1 6= vi−1 .
Since G has only finitely many vertices, we must eventually
choose a vertex which has been chosen before. If vk is the first
such vertex then the walk between the first two occurrences of vk is

40
 Figure 3.2:

a cycle (since the internal vertices of this walk are distinct and also
different from vk as vk is the first vertex to be repeated).(See Figure
3.3.)vk+2

Figure 3.3:

Theorem 3.1.2. A connected graph G is Euler if and only if the

degree of every vertex is even.

3.2 Hamiltonian Graphs

Definition 3.2.1. A Hamiltonian path in a graph G is a path which
contains every vertex of G

Definition 3.2.2. A Hamiltonian cycle (or Hamiltonian circuit) in

a graph G is a cycle which contains every vertex of G.

41
 Note: A hamiltonian cycle in G with initial vertex v contains
every other vertex of G precisely once and then ends back at v.

Definition 3.2.3. A graph G is called Hamiltonian if it has a Hamil-

tonian cycle.

Example 3.2.1. Let G = (V, E) be a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 , v5 , v6 , v7 , v8 } is given below. Consider the path v1 −

v1 v4 v5 v6

v2 v3 v8 v7

Figure 3.4: Hamiltonian graph G

v2 − v3 − v5 − v6 − v7 − v8 − v4 which contain all the vertices of

G. Hence it’s a hamiltonian path. v1 − v2 − v3 − v5 − v6 − v7 −
v8 − v4 − v1 which is a cycle contain all the vertices of G. Hence
G is a Hamiltonian graph.

Definition 3.2.4. A simple graph G is called maximal non-Hamiltonian

if it is not Hamiltonian but the addition to it of any edge connecting
two non-adjacent vertices forms a Hamiltonian graph.

Example 3.2.2. Let G = (V, E) is a graph with vertex set V (G) =

{v1 , v2 , v3 , v4 } given in figure 3.5.
The graph G is maximal non-hamiltonian graph since addition
of any edge become a hamiltonian graph.

42
 v1 v2

v4 v3

Figure 3.5: A maximal non-hamiltonian graph G

Theorem 3.2.1. (Dirac, 1952) If G is a simple graph with n ver-

tices, where n ≥ 3, and the degree d(v) ≥ n/2 for every vertex v
of G, then G is Hamiltonian.

Theorem 3.2.2. Let G be a simple graph with n vertices and let u

and v be non-adjacent vertices in G such that

d(u) + d(v) ≥ n

Let G + uv denote the supergraph of G obtained by joining u and

v by an edge. Then G is Hamiltonian if and only if G + uv is
Hamiltonian

Proof. Suppose that G is Hamiltonian. Then, as noted earlier, the

supergraph G + uv must also be Hamiltonian.
Conversely, suppose that G + uv is Hamiltonian. Then, if G is
not Hamiltonian, just as in the proof of Theorem 3.6 we obtain the
inequality d(u)+d(v) < n. However, by hypothesis, d(u)+d(v) ≥
n. Hence G must be Hamiltonian also, as required.

Definition 3.2.5. Let G be a simple graph. If there are two nonad-

jacent vertices u1 and v1 in G such that d (u1 ) + d (v1 ) ≥ n in G,

43
join u1 and v1 by an edge to form the supergraph G1 . Then, if there
are two nonadjacent vertices u2 and v2 such that d (u2 )+d (v2 ) ≥ n
in G1 , join u2 and v2 by an edge to form the supergraph G2 . Con-
tinue in this way, recursively joining pairs of nonadjacent vertices
whose degree sum is at least n until no such pair remains. The final
supergraph thus obtained is called the closure of G and is denoted
by c(G).

Example 3.2.3. We give an example of a closure operation in Fig-

ure 3.6. For this example c(G) = K7

Figure 3.6: Closure operation of graph G

Note:For a graph G on n vertices of d(u) + d(v) < n for any

pair u, v of nonadjacent vertices in G and so the closure operation
does not get off the ground, i.e., c(G) = G.

44
Theorem 3.2.3. (Bondy and Chvatal, 1976) A simple graph G is
Hamiltonian if and only if its closure c(G) is Hamiltonian.

Proof. Since c(G) is a supergraph of G, if G is Hamiltonian then

c(G) must be Hamiltonian. Conversely, suppose that c(G) is Hamil-
tonian. Let G, G1 , G2 , . . . , Gk−1 , Gk = c(G) be the sequence of
graphs obtained by performing the closure procedure on G. Since
c(G) = Gk is obtained from Gk−1 by setting Gk = Gk−1 + uv,
where u, v is a pair of nonadjacent vertices in Gk−1 with d(u) +
d(v) ≥ n, it follows by Theorem 3.2.2 that Gk−1 is Hamiltonian.
Similarly Gk−2 , so Gk−3 , . . ., so G1 , and so G must be Hamiltonian,
as required.

Corollary 3.2.3.1. Let G be a simple graph on n vertices, with n ≥

3. If c(G) is complete, i.e., if c(G) = Kn , then G is Hamiltonian.

Proof. This is immediate from the Theorem since any complete

graph is Hamiltonian.

3.3 Plane and Planar graphs

Definition 3.3.1. A plane graph is a graph drawn in the plane (of
the paper, blackboard, etc.) in such a way that any pair of edges
meet only at their end vertices (if they meet at all).
A planar graph is a graph which is isomorphic to a plane
graph, i.e., it can be (re)drawn as a plane graph.

Example 3.3.1. We give a example of a planar graphs in Figure

3.7
All 5 graphs are planar but G1 and G4 are not plane graphs.

45
 Figure 3.7: 5 planar graphs

Definition 3.3.2. A Jordan curve in the plane is a continuous non-

self-intersecting curve whose origin and terminus coincide.
Example 3.3.2. For example, in Figure 3.8 the curve C1 is not a
Jordan curve because it intersects itself, C2 is not a Jordan curve
since its origin and terminus do not coincide, i.e., its two end points
do not meet up, but C3 is a Jordan curve.

Figure 3.8: C1 and C2 are not Jordan curves but C3 is

Definition 3.3.3. If J is a Jordan curve in the plane then the part

of the plane enclosed by J is called the interior of J and denoted

46
by int J. We exclude from int J the points actually lying on J.
Similarly the part of the plane lying outside J is called the exterior
of J and denoted by ext J.
Theorem 3.3.1. The Jordan Curve Theorem states that if J is a
Jordan curve, if x is a point in int J and y is a point in ext J then
any (straight or curved) line joining x to y must meet J at some
point, i.e., must cross J.
Theorem 3.3.2. K5 , the complete graph on five vertices, is nonpla-
nar.
Theorem 3.3.3. The complete bipartite graph K3,3 is not planar

3.4 Euler’s Formula

Definition 3.4.1. A plane graph G partitions the plane into a num-
ber of regions called the faces of G. More precisely, if x is a point
on the plane which is not in G, i.e., is not a vertex of G or a point
on any edge of G, then we define the face of G containing x to be
the set of all points on the plane which can be reached from x by a
(straight or curved) line which does not cross any edge of G or go
through any vertex of G.
Example 3.4.1. For example, for the point x in the graph G1 of
Figure 3.9, the face containing x is shown as the interior of the
cycle v2 − v4 − v3 − v6 − v5 − v2 . In this example obviously the face
of G1 containing the point y is the same face as that containing x.
It is bounded by the cycle v2 v4 v3 v6 v5 v4 . The face of G1 containing
the point z is not bounded by any cycle. It is called the exterior
face. of G1 .

47
 Figure 3.9:

Note:Any plane graph has exactly one exterior face. Any other
face is bounded by a closed walk in the graph and is called an inte-
rior face.

Theorem 3.4.1. (Euler’s Formula) Let G be a connected plane

graph, and let n, e, and f denote the number of vertices, edges
and faces of G, respectively. Then

n−e+f =2

Corollary 3.4.1.1. Let G be a plane graph with n vertices,e edges,

f faces and k connected components. Then

n−e+f =k+1

Corollary 3.4.1.2. Let G1 and G2 be two plane graphs which are

both redrawings of the same planar graph G. Then f (G1 ) = f (G2 ),
i.e., G1 and G2 have the same number of faces.

48
Proof. Let n (G1 ) , n (G2 ) denote the number of vertices and e (G1 ),
e (G2 ) the number of edges in G1 , G2 respectively. Then, since G1
and G2 are both isomorphic to G we have n (G1 ) = n (G2 ) and
e (G1 ) = e (G2 ) . Using Euler’s Formula we get

f (G1 ) = e (G1 ) − n (G1 ) + 2 = e (G2 ) − n (G2 ) + 2 = f (G2 )

as required.

Definition 3.4.2. Let ϕ be a face of a plane graph G. We define

the degree of ϕ, denoted by d(ϕ), to be the number of edges on the
boundary of ϕ.

Theorem 3.4.2. Let G be a simple planar graph with n vertices

and e edges, where n ≥ 3. Then

e ≤ 3n − 6

Corollary 3.4.2.1. If G is a simple planar graph then G has a ver-

tex v of degree less than 6 , i.e., there is a v in V (G) with d(v) ≤ 5

Proof. If G has only one vertex this vertex must have degree 0. If
G has only two vertices then both must have degree at most 1. Thus
we can suppose that n ≥ 3, i.e., that G has at least three vertices.
Now if the degree of every vertex of G is at least six we have
X
d(v) ≥ 6n
v∈V (G)

P
However, by Theorem 1.4.1, v∈V (G) d(v) = 2e. Thus 2e ≥ 6n
and so e ≥ 3n. This is impossible since, by the above theorem,

49
e ≤ 3n − 6. This contradiction shows that G must have at least one
vertex of degree less than 6, as required.

Corollary 3.4.2.2. K5 is nonplanar.

Proof. Here n = 5 and e = (5 × 4)/2 = 10 so that 3n − 6 = 9.

Thus e > 3n − 6 and so, by the theorem, G = K5 can not be
planar.

Corollary 3.4.2.3. K3,3 is nonplanar.

Proof. Since K3,3 is bipartite it contains no odd cycles (by Theo-

rem 1.6.1) and so in particular no cycle of length three. It follows
that every face of a plane drawing of K3,3 , if such exists, must
have at least four boundary edges. Thus, using the argument of the
proof of Theorem 3.4.2, we get b ≥ 4f and then 4f ≤ 2e, i.e.,
2f ≤ e = 9. This gives f ≤ 9/2. However, by Euler’s Formula,
f = 2 − n + e = 2 − 6 + 9 = 5, a contradiction.

50