3.

3
Connectivity
Many problems can be modeled with paths formed by travelling along the edges
of graphs. Problems of efficiently planning routes for mail delivery, garbage
pickup, diagnostics in computer networks can be solved using models that involve
paths in graphs.

Paths
A path is a sequence of edges that begins at a vertex of a graph and travels from
vertex to vertex along edges of the graph.

The following is a formal definition of a path in an undirected graph:

A path in an undirected graph: Let be an undirected graph and be a

nonnegative integer. A path of length from a vertex to a vertex in is a
sequence of edges of such that is associated with ,
is associated with , and so on, finally is associated with
where and . Such a path is said to pass through the vertices
or traverse the edges . If the graph is simple,
then we denote this path by its sequence of vertices .

Note that a path of length zero consists of a single vertex

Circuit: A path is a circuit if it begins and ends at the same vertex, (i.e., ) and
the length .

Simple path: A path or a circuit is said to be simple if it does not contain the
same edge more than once.
Example 1: In the following simple undirected graph (figure 1), is a
simple path from to of length 4, because the edges
are all distinct.

Figure 1

Notice that is not a path, because is not an edge.

The path is a circuit of length 4, because it is a path of length 4 that

begins and ends at .

Note that is a path of length 7 and it is not simple because it has

the edge occurs twice in the path.

The following is the definition of a path in a digraph.

A path in a digraph: Let be a digraph and be a nonnegative integer .A path of

length from to in is a sequence of edges of such that is
associated with the directed edge , with , and so on, is
associated with , where and . When there are no
multiple edges in the digraph , this path is denoted by its vertex sequence
.

A path of length greater than zero that begins and ends at the same vertex is
called a circuit.

A path or circuit is said to be simple if it does not contain the same edge more
than once.
Note:

(1) The alternate terminology for path, circuit and simple path are walk,
closed walk and trail respectively.
(2) There may be more than one path from the initial vertex to the terminal
vertex .
(3) Paths represent useful information in many graph models.

Example 2: In the following digraph

(i) is a path of length 4 and it is simple.

(ii) is a circuit of length 4 and it is simple.
(iii) is a circuit of length 6 but it is not simple, because it
contains the edge twice.
(iv) is not a path, because is not an edge in the digraph

Connectedness in undirected graphs

When does a computer network have the property that every pair of computers
can share information, if messages can be sent through one or more intermediate
computers? If computers are represented by vertices and communication links
represent edges then the computer network represents an undirected graph and
the question now becomes: Is there a path between every pair of vertices in the
graph?

Undirected connected graph: An undirected graph is said to be connected if

there is a path between every pair of distinct vertices of .
Thus, any two computers in the network can communicate if and only if the graph
of the network is connected.

Example 3: The graph given in figure 1 is connected, because there is a path

between every pair of distinct vertices.

The following is a useful result:

Theorem 1: In a connected undirected graph, there is a simple path between

every pair of vertices.

Connected component: A connected component of a graph is a connected

subgraph of that is not a subgraph of another connected subgraph of . That is,
a connected component of a graph is a maximal connected subgraph of .

If a graph is not connected then it has two or more connected components that
are disjoint and have as their union. That is, if is not connected then is
partitioned into connected components.

Example 4: The following graph is not connected, because there is no path

between the vertices and .

Its connected subgraphs are the subgraphs and , where

, , and
, , .
Notice that and are maximal connected sub graphs. Therefore, are
components and is partitioned into and .

Cut vertex: The removal of a vertex and all edges incident with it in a graph
produces a subgraph of with more connected components than in . Then such
a vertex is called a cut vertex or a articulation point.

Note that the removal of a cut vertex from a connected graph produces a
subgraph that is not connected.

Cut edge: An edge whose removal produces a graph with more connected
components than the original graph is called a cut edge or a bridge.

Example 5: Find all cut vertices and all cut edges in the following graph.

Solution: The removal of the vertex and all edges that incident with , i.e.,
, , disconnects the graph. (Note that remove the edges that
incident with only but not the other end vertices namely and ). Therefore
is a cut vertex. Similarly and are also cut vertices.

Removal of the edge (but not the end points ) disconnects the graph.
Therefore is a cut edge. Similarly is also a cut edge.

Connectedness in Digraphs

The following are two notions of connectedness in digraphs:

Weakly connectedness: A digraph is weakly connected if there is a path

between every pair of distinct vertices in the underlying undirected graph. That is,
a digraph is weakly connected if the undirected graph of obtained by ignoring
the directions of the edges is connected.

Strongly connectedness: A digraph is strongly connected if for every two

distinct vertices and in , there is a path from to as well as a path from
to .

Note: Every strongly connected digraph is also weakly connected, but not
conversely.

Example 6: Are the following digraphs strongly connected? Are they weakly
connected?

Solution: Notice the following in :

Vertex to vertex Path

to
to
to
to
to
to
to
to
to
to
to
to
Thus for every two distinct vertices ; has a path from to as well as a path
from to . Thus is strongly connected. Therefore, is also weakly connected.

The digraph is weakly connected, because the undirected graph derived from
by ignoring the directions of the edges of is connected. Further, it is not
strongly connected because there is no directed path from to

Strongly connected component: A subgraph of a digraph that is strongly

connected but not contained in a larger strongly connected subgraph (
maximal strongly connected subgraph) is called a strongly connected component
or strong component of

Example 7: The following graph has three strongly connected components,

consisting the vertex ; the vertex ; and the subgraph consisting of the vertices
and edges , and

Paths and Isomorphism:

Paths and circuits can help to determine whether two graphs are isomorphic .
Paths can be used to construct mapping that may be isomorphisms. A useful
isomorphic invariant for simple graphs is the existence of a simple circuit of length
, where is a natural number greater than .
Example 8: Determine whether the following graphs and are isomorphic.

Solution: Both the graphs and are undirected graphs with six vertices and
eight edges. Both have the degree sequence . Notice that has a
simple circuit of length . Observe that has no simple circuit of
length . It may be noted that all simple circuits in have length atleast
Therefore is not isomorphic to , because the existence of a simple circuit of
length is an isomorphic invariant.

Example 9: Determine whether the following graphs and are isomorphic.

Solution: Both the graphs and are undirected graphs with five vertices and six
edges .Both have the degree sequence . Further ,both have a simple
circuit of length , a simple circuit of length and a simple circuit of length .
Because all these isomorphic invariants agree, the graphs and may be
isomorphic. Observe that the circuit of length in ; in which
the vertex of degree is trapped between two vertices of degree and
. Notice that the circuit of length in ; in a similar circuit
with the same characteristics .These paths guide us to set up the following
bijection from the vertex set of to the vertex set of .

Let and be the adjacency matrices of and respectively, w.r.t. the

ordering of the vertices.

in and in

We notice that

0 1 1 0 1 
1 0 1 0 0 

AG  1 1 0 1 0   AH
 
0 0 1 0 1 
1 0 0 1 0 

Thus

Counting Paths Between Vertices

The number of paths between two vertices in a graph can be determined using its
adjacency matrix.

Theorem 2: Let be any graph with . Let be the adjacency

matrix of w.r.t. the ordering of vertices . The number of
different paths of length from to equals the th entry of (where
is a natural number)

Proof: Follows by mathematical induction.

In the above theorem, directed or undirected edges, multiple edges and loops are
allowed in
Note:

1. In the above theorem denotes the matrix multiplication

. It is not Boolean product.
2. The above theorem can be used to find the length of the shortest path
between two vertices and it can be used to determine whether a graph is
connected.

Example 10: How many paths of length four are there from to in the
following simple graph

Solution: The adjacency matrix of , w.r.t. the ordering of vertices as is

To determine the number of paths of length between two vertices of , we

have to compute . Now

2 0 0 2 0 4 4 0 8 0 0 8
  4  0 8 8 0 
2 0 2 2 0 0 0 4
A  3
, A  4
and A  
0 2 2 0 4 0 0 4 0 8 8 0
     
2 0 0 2 0 4 4 0 8 0 0 8

The number of paths of length from to in entry in

By inspection, we see the following paths from to

P1:

Determine whether the following graphs are connected.

Solution:

It is not connected ,because there is no path between the vertices and .

. It is connected ; because there is a path between every pair of distinct vertices

It is not connected ,because there is no path between the vertices and

2. How many connected components does each of the above graphs have? For
each graph ,find its connected components.

Solution:

There are components .They are

where ,

where ,

where ,

There is one connected component the whole graph , because it is

connected

There are two connected components . They are

where ,

where ,
P2:

Find all cut vertices and cut edges of the following graphs

Solution:

. The removal of the vertex and all edges that incident with disconnects the
graph. Therefore, is the cut vertex. Further , it is the only cut vertex. Note that
the removal of any edge in this graph is not producing a graph with more
connected components than the original graph. Therefore , there are no cut
edges(bridges) in this graph.

The cut vertices are .The cut edge is . The removal of the edge
(not the vertices and ) disconnects the graph into two connected
components.

The cut vertices are : and the cut edges are:

P3:

Determine whether each of these graphs is strongly connected and if not ,

whether it is weakly connected.

Solution:

a) The undirected graph derived from this graph by ignoring the directions of the
edges is connected. Therefore it is a weakly connected graph. Notice that
there is a path from to but there is no path from to . Therefore it is not
strongly connected. Thus, it is a weakly connected graph, but not strongly
connected.
b) It is weakly connected, because the undirected graph derived from it by
ignoring the directions of edges is connected. Notice that there is path from
to , but there is no path from to . Therefore, this graph is not
strongly connected. Thus, it is a weakly connected graph, but not strongly
connected.
c) Notice that the undirected graph derived from this graph by ignoring the
directions of the edges is not connected (because it is portioned into two
connected components, one is a rectangle and the other is a triangle
. Therefore, it is not weakly connected. Consequently it is not strongly
connected (If it is strongly connected, then it is weakly connected
a contradiction(because every strongly connected graph is a weakly
connected).).It is neither weakly connected nor strongly connected.
P4:

Find the strongly connected components of each of these graphs.

Solution:

a) First note that there is a path from to ,but there is no path from to .
Therefore it is not strongly connected.
Recall that a strongly connected component of a graph is a maximal
strongly connected subgraph of
Notice that there are two strongly connected components in this
graph, where
and
and

b) Notice that there is no path from to any vertex. Therefore,

where and is a strongly connected component. Similarly
,where and is a strongly connected
component.
Notice that is a simple circuit and is another simple
circuit; there is a interconnecting path between these circuits. Thus
every two vertices in these circuits are connected. This shows that the
subgraph is a strongly connected component where
and
P5:

Use paths either to show that these graphs are not isomorphic or to find an
isomorphism between them.

Solution:

Both the graphs and are undirected graphs with vertices and edges.
Both have the degree sequence . Notice that has two simple
circuits of length and . Observe that has no
simple circuit of length . Therefore, is not isomorphic to ,because the
existence of a simple circuit of length is an isomorphic invariant.
P6:

Use paths either to show that these graphs are not isomorphic or to find an
isomorphism between them.

Solution:

Both the graphs and are undirected graphs with vertices and 12 edges.
Both have the degree sequence . Observe that the graphs and
has the following circuit of length respectively,

These circuits guide us to set up the following bijection from the vertex set of
to the vertex set of :

Let and be the adjacency matrices of and respectively, w.r.t. the

ordering of the vertices.

in

in

We notice that
 0 1 0 0 1
0 1 0
1 0 1 0 0 
0 0 1

0 1 0 1 1
0 0 0
 
0 0 1 0 1
0 0 1
AG    AH
0 0 0 1 1
0 1 0
 
1 0 0 0 1 0 1 0
0 1 0 1 0 1 0 0
 
1 0 1 0 1 0 0 0 

Thus ,
P7:

Find the number of paths of length between two different vertices in

Solution:

We have

The adjacency matrix of w.r.t. the ordering of vertices is

0 1 1 1
1 0 1 1 
A
1 1 0 1
 
1 1 1 0

To determine the number of paths of length between two vertices of , we

have to compute . Now

3 2 2 2 6 7 7 7
 3 2 2  7 6 7 7 
2 2
A  , 3
A 
2 2 3 2 7 7 6 7
   
2 2 2 3 7 7 7 6

The number of paths of length between two different vertices in is

The number of circuits of length , from to is .They are

P8:

Find the number of paths of length in

i. between any two adjacent vertices

ii. between any two nonadjacent vertices

Solution:

We have

The adjacency matrix of , w.r.t. the ordering of vertices .

0 0 0 1 1 1
0 0 0 1 1 1 

0 0 0 1 1 1
A 
1 1 1 0 0 0
1 1 1 0 0 0
 
1 1 1 0 0 0

To determine the number of paths of length between two vertices of , we

have to compute .Now

3 3 3 0 0 0 0 0 0 9 9 9
3 3 3 0 0 
0 0 0 0 9 9 9 
 
3 3 3 0 0 0 3 0 0 0 9 9 9
A2    , A  
0 0 0 3 3 3 9 9 9 0 0 0
0 0 0 3 3 3 9 9 9 0 0 0
   
0 0 0 3 3 3 9 9 9 0 0 0
Each vertex of is adjacent to each vertex of and no two vertices
of are adjacent and no two vertices of are adjacent.

Now the number of paths of length in , between any two adjacent vertices
is and the number of paths of length in between any two non adjacent
vertices is