Lecture 1, version April 6, 2023 page 1

Math 530 Spring 2022, Lecture 1: Simple graphs

website: https://www.cip.ifi.lmu.de/~grinberg/t/22s

0.1. Plan
This is a course on graphs – a rather elementary concept (actually a cluster of
closely related concepts) that can be seen all over mathematics. We will discuss
several kinds of graphs (simple graphs, multigraphs, directed graphs, etc.) and
study their features and properties. In particular, we will encounter walks on
graphs, matchings of graphs, flows on networks (networks are graphs with
extra data), and take a closer look at certain types of graphs such as trees and
tournaments.
The theory of graphs goes back at least to Leonhard Euler, who in a 1736
paper [Euler36] (see [Euler53] for an English translation) solved a puzzle about
an optimal tour of the town of Königsberg. It saw some more developments in
the 19th century and straight-up exploded in the 20th; now it is one of the most
active fields of mathematics.
We won’t follow any book. But we will follow (to an extent) my lecture notes
from Spring 2017: https://www.cip.ifi.lmu.de/~grinberg/t/17s (but keep
in mind that they were written for a semester, not a quarter). The first few
lectures will occasionally follow a stub of a text [17s] that I started writing back
then. A long list of books appears on the course website; you don’t strictly
need any of them, but it’s worth skimming them to get a feel for the topic (far
beyond what we can do in this course) and learn more about directions you
care about.
A few administrativa:

• See the website ( https://www.cip.ifi.lmu.de/~grinberg/t/22s ) for

any info you might be looking for.
We will use gradescope for HW. Blackboard will only be used for archiv-
ing announcement emails.

• Please interrupt whenever something is unclear!

Don’t hesitate to email questions either.

0.2. Notations
Notations:

• We let N = {0, 1, 2, . . .}. Thus, 0 ∈ N.

• The size (i.e., cardinality) of a finite set S is denoted by |S|.

Lecture 1, version April 6, 2023 page 2

• If S is a set, then the powerset of S means the set of all subsets of S. This
powerset will be denoted by P (S).
Moreover, if S is a set, and k is an integer, then Pk (S) will mean the set of
all k-element subsets of S. For instance,
P2 ({1, 2, 3}) = {{1, 2} , {1, 3} , {2, 3}} .

 any number n and any k ∈ N, we define the binomial coefficient

• For
n
to be the number
k
k −1
∏ (n − i )
n ( n − 1) ( n − 2) · · · ( n − k + 1) i =0
= .
k! k!
These binomial coefficients have many interesting properties, which can
often be found in textbooks on enumerative combinatorics. Some of the
most important ones are the following:
 
n n!
– The factorial formula: If n, k ∈ N and n ≥ k, then = .
k k! · (n − k )!
– The combinatorial
  interpretation: If n, k ∈ N, and if S is an n-element
n
set, then is the number of all k-element subsets of S (in other
k  
n
words, |Pk (S)| = ).
k
– Pascal’s recursion: For any number n and any positive integer k, we
have
n−1 n−1
     
n
= + .
k k−1 k

1. Simple graphs
1.1. Definitions
The first type of graphs that we will consider are the “simple graphs”, named
so because of their very simple definition:
Definition 1.1.1. A simple graph is a pair (V, E), where V is a finite set, and
where E is a subset of P2 (V ).
To remind, P2 (V ) is the set of all 2-element subsets of V. Thus, a simple
graph is a pair (V, E), where V is a finite set, and E is a set consisting of 2-
element subsets of V. We will abbreviate the word “simple graph” as “graph”
in the next few lectures, but afterwards we will learn some more advanced and
general notions of “graphs”.
Lecture 1, version April 6, 2023 page 3

Example 1.1.2. Here is a simple graph:

({1, 2, 3, 4} , {{1, 3} , {1, 4} , {3, 4}}) .

Example 1.1.3. For any n ∈ N, we can define a simple graph Copn to be the
pair (V, E), where V = {1, 2, . . . , n} and

E = {{u, v} ∈ P2 (V ) | gcd (u, v) = 1} .

We call this the n-th coprimality graph.

(Some authors do not require V to be finite in Definition 1.1.1; this leads to

infinite graphs. But I shall leave this can of worms closed for this quarter.)
The purpose of simple graphs is to encode relations on a finite set – specif-
ically the kind of relations that are binary (i.e., relate pairs of elements), sym-
metric (i.e., mutual) and irreflexive (i.e., an element cannot be related to itself).
For example, the graph Copn in Example 1.1.3 encodes the coprimality (aka
coprimeness) relation on the set {1, 2, . . . , n}, except that the latter relation is
not irreflexive (1 is coprime to 1, but {1, 1} is not in E; thus, the graph Copn
“forgets” that 1 is coprime to 1). For another example, if V is a set of people,
and E is the set of {u, v} ∈ P2 (V ) such that u has been married to v at some
point, then (V, E) is a simple graph. Even in 2022, marriage to oneself is not a
thing, so all marriages can be encoded as 2-element subsets.1
The following notations provide a quick way to reference the elements of V
and E when given a graph (V, E):

Definition 1.1.4. Let G = (V, E) be a simple graph.

(a) The set V is called the vertex set of G; it is denoted by V ( G ). (Notice

that the letter “V” in “V ( G )” is upright, as opposed to the letter “V”
in “(V, E)”, which is italic. These are two different symbols, and have
different meanings: The letter V stands for the specific set V which is
the first component of the pair G, whereas the letter V is part of the
notation V ( G ) for the vertex set of any graph. Thus, if H = (W, F ) is
another graph, then V ( H ) is W, not V.)
The elements of V are called the vertices (or the nodes) of G.

(b) The set E is called the edge set of G; it is denoted by E ( G ). (Again, the
letter “E” in “E ( G )” is upright, and stands for a different thing than
the “E”.)
1 The more standard example for a social graph would be a “friendship graph”; here, V is
again a set of people, but E is now the set of {u, v} ∈ P2 (V ) such that u and v are friends.
Of course, this only works if you think of friendship as being automatically mutual (true
for facebook friendship, questionable for the actual thing).
Lecture 1, version April 6, 2023 page 4

The elements of E are called the edges of G. When u and v are two
elements of V, we shall often use the notation uv for {u, v}; thus, each
edge of G has the form uv for two distinct elements u and v of V. Of
course, we always have uv = vu.
Notice that each simple graph G satisfies G = (V ( G ) , E ( G )).

(c) Two vertices u and v of G are said to be adjacent (to each other) if
uv ∈ E (that is, if uv is an edge of G). In this case, the edge uv is said
to join u with v (or connect u and v); the vertices u and v are called
the endpoints of this edge. When the graph G is not obvious from the
context, we shall often say “adjacent in G” instead of just “adjacent”.
Two vertices u and v of G are said to be non-adjacent (to each other) if
they are not adjacent (i.e., if uv ∈
/ E).

(d) Let v be a vertex of G (that is, v ∈ V). Then, the neighbors of v (in
G) are the vertices u of G that satisfy vu ∈ E. In other words, the
neighbors of v are the vertices of G that are adjacent to v.

Example 1.1.5. Let G be the simple graph

({1, 2, 3, 4} , {{1, 3} , {1, 4} , {3, 4}})

from Example 1.1.2. Then, its vertex set and its edge set are

V ( G ) = {1, 2, 3, 4} and E ( G ) = {{1, 3} , {1, 4} , {3, 4}} = {13, 14, 34}

(using our notation uv for {u, v}). The vertices 1 and 3 are adjacent (since
13 ∈ E ( G )), but the vertices 1 and 2 are not (since 12 ∈
/ E ( G )). The neighbors
of 1 are 3 and 4. The endpoints of the edge 34 are 3 and 4.

1.2. Drawing graphs

There is a common method to represent graphs visually: Namely, a graph can
be drawn as a set of points in the plane and a set of curves connecting some of
these points with each other.
More precisely:

Definition 1.2.1. A simple graph G can be visually represented by drawing

it on the plane. To do so, we represent each vertex of G by a point (at which
we put the name of the vertex), and then, for each edge uv of G, we draw a
curve that connects the point representing u with the point representing v.
The positions of the points and the shapes of the curves can be chosen freely,
as long as they allow the reader to unambiguously reconstruct the graph G
Lecture 1, version April 6, 2023 page 5

from the picture. (Thus, for example, the curves should not pass through
any points other than the ones they mean to connect.)

Example 1.2.2. Let us draw some simple graphs.

(a) The simple graph ({1, 2, 3} , {12, 23}) (where we are again using the
shorthand notation uv for {u, v}) can be drawn as follows:

1 2 3
.

This is (in a sense) the simplest way to draw this graph: The edges are
represented by straight lines. But we can draw it in several other ways as
well – e.g., as follows:

1 3 2

Here, we have placed the points representing the vertices 1, 2, 3 differently.

As a consequence, we were not able to draw the edge 12 as a straight line,
because it would then have overlapped with the vertex 3, which would make
the graph ambiguous (the edge 12 could be mistaken for two edges 13 and
32).
Here are three further drawings of the same graph ({1, 2, 3} , {12, 23}):

2 2
1 3 2 .
1 3 1 3

(b) Consider the 5-th coprimality graph Cop5 defined in Example 1.1.3.
Here is one way to draw it:

2
3

4
5
.
Lecture 1, version April 6, 2023 page 6

Here is another way to draw the same graph Cop5 , with fewer intersections
between edges:

2
3

4
5
.

By appropriately repositioning the points corresponding to the five vertices

of Cop5 , we can actually get rid of all intersections and make all the edges
straight (as opposed to curved). Can you find out how?

(c) Let us draw one further graph: the simple graph

({1, 2, 3, 4, 5} , P2 ({1, 2, 3, 4, 5})). This is the simple graph whose ver-
tices are 1, 2, 3, 4, 5, and whose edges are all possible two-element sets
consisting of its vertices (i.e., each pair of two distinct vertices is adjacent).
We shall later call this graph the “complete graph K5 ”. Here is a simple way
to draw this graph:

2
3

4
5
.

This drawing is useful for many purposes; for example, it makes the ab-
stract symmetry of this graph (i.e., the fact that, roughly speaking, its vertices
1, 2, 3, 4, 5 are “equal in rights”) obvious. But sometimes, you might want to
draw it differently, to minimize the number of intersecting curves. Here is a
Lecture 1, version April 6, 2023 page 7

drawing with fewer intersections:

3
1

4
5
.

In this drawing, we have only one intersection between two curves left. Can
we get rid of all intersections?
This is a question of topology, not of combinatorics, since it really is about
curves in the plane rather than about finite sets and graphs. The answer is
“no”. (That is, no matter how you draw this graph in the plane, you will
always have at least one pair of curves intersect.) This is a classical result
(one of the first theorems in the theory of planar graphs), and proofs of it
can be found in various textbooks (e.g., [FriFri98, Theorem 4.1.2], which is
generally a good introduction to planar graph theory even if it uses termi-
nology somewhat different from ours). Note that any proof must use some
analysis or topology, since the result relies on the notion of a (continuous)
curve in the plane (if curves were allowed to be non-continuous, then they
could “jump over” one another, so they could easily avoid intersecting!).

1.3. A first fact: The Ramsey number R (3, 3) = 6

Enough definitions; let’s state a first result:

Proposition 1.3.1. Let G be a simple graph with |V ( G )| ≥ 6 (that is, G has at

least 6 vertices). Then, at least one of the following two statements holds:

• Statement 1: There exist three distinct vertices a, b and c of G such that

ab, bc and ca are edges of G.

• Statement 2: There exist three distinct vertices a, b and c of G such that

none of ab, bc and ca is an edge of G.

In other words, Proposition 1.3.1 says that if a graph G has at least 6 vertices,
then we can either find three distinct vertices that are mutually adjacent2 or find
2 bywhich we mean (of course) that any two distinct ones among these three vertices are
adjacent
Lecture 1, version April 6, 2023 page 8

three distinct vertices that are mutually non-adjacent (i.e., no two of them are
adjacent), or both. Often, this is restated as follows: “In any group of at least
six people, you can always find three that are (pairwise) friends to each other,
or three no two of whom are friends” (provided that friendship is a symmetric
relation).
We will give some examples in a moment, but first let us introduce some
convenient terminology:

Definition 1.3.2. Let G be a simple graph.

(a) A set { a, b, c} of three distinct vertices of G is said to be a triangle (of

G) if every two distinct vertices in this set are adjacent (i.e., if ab, bc and
ca are edges of G).

(b) A set { a, b, c} of three distinct vertices of G is said to be an anti-triangle

(of G) if no two distinct vertices in this set are adjacent (i.e., if none of
ab, bc and ca is an edge of G).

Thus, Proposition 1.3.1 says that every simple graph with at least 6 vertices
contains a triangle or an anti-triangle (or both).

Example 1.3.3. Let us show two examples of graphs G to which Proposi-

tion 1.3.1 applies, as well as an example to which it does not:

(a) Let G be the graph (V, E), where

V = {1, 2, 3, 4, 5, 6} and
E = {{1, 2} , {2, 3} , {3, 4} , {4, 5} , {5, 6} , {6, 1}} .

(This graph can be drawn in such a way as to look like a hexagon:

3 2

4 1

5 6
.

) This graph satisfies Proposition 1.3.1, since {1, 3, 5} is an anti-triangle

(or since {2, 4, 6} is an anti-triangle).
Lecture 1, version April 6, 2023 page 9

(b) Let G be the graph (V, E), where

V = {1, 2, 3, 4, 5, 6} and
E = {{1, 2} , {2, 3} , {3, 4} , {4, 5} , {5, 6} , {6, 1} , {1, 3} , {4, 6}} .

(This graph can be drawn in such a way as to look like a hexagon with
two extra diagonals:

3 2

4 1

5 6
.

) This graph satisfies Proposition 1.3.1, since {1, 2, 3} is a triangle.

(c) Let G be the graph (V, E), where

V = {1, 2, 3, 4, 5} and
E = {{1, 2} , {2, 3} , {3, 4} , {4, 5} , {5, 1}} .

(This graph can be drawn to look like a pentagon:

2
3

4
5
.

) Proposition 1.3.1 says nothing about this graph, since this graph does
not satisfy the assumption of Proposition 1.3.1 (in fact, its number of
vertices |V ( G )| fails to be ≥ 6). By itself, this does not yield that the
claim of Proposition 1.3.1 is false for this graph. However, it is easy
to check that the claim actually is false for this graph: It has neither a
triangle nor an anti-triangle.

Proof of Proposition 1.3.1. We need to prove that G has a triangle or an anti-

Lecture 1, version April 6, 2023 page 10

triangle (or both).

Choose any vertex u ∈ V ( G ). (This is clearly possible, since |V ( G )| ≥ 6 ≥ 1.)
Then, there are at least 5 vertices distinct from u (since G has at least 6 vertices).
We are in one of the following two cases:
Case 1: The vertex u has at least 3 neighbors.
Case 2: The vertex u has at most 2 neighbors.
Let us consider Case 1 first. In this case, the vertex u has at least 3 neighbors.
Hence, we can find three distinct neighbors p, q and r of u. Consider these p, q
and r. If one (or more) of pq, qr and rp is an edge of G, then G has a triangle
(for example, if pq is an edge of G, then {u, p, q} is a triangle). If not, then G has
an anti-triangle (namely, { p, q, r }). Thus, in either case, our proof is complete
in Case 1.
Let us now consider Case 2. In this case, the vertex u has at most 2 neighbors.
Hence, the vertex u has at least 3 non-neighbors3 (since there are at least 5
vertices distinct from u in total). Thus, we can find three distinct non-neighbors
p, q and r of u. Consider these p, q and r. If all of pq, qr and rp are edges of G,
then G has a triangle (namely, { p, q, r }). If not, then G has an anti-triangle (for
example, if pq is not an edge of G, then {u, p, q} is an anti-triangle). In either
case, we are thus done with the proof in Case 2. Thus, both cases are resolved,
and the proof is complete.
Notice the symmetry between Case 1 and Case 2 in our above proof: the ar-
guments used were almost the same, except that neighbors and non-neighbors
swapped roles.

Remark 1.3.4. Proposition 1.3.1 could also be proved by brute force as well
(using a computer). Indeed, it clearly suffices to prove it for all simple graphs
with 6 vertices (as opposed to ≥ 6 vertices), because if a graph has more than
6 vertices, then we can just throw away some of them until we have only 6
left. However, there are only finitely many simple graphs with 6 vertices (up
to relabeling of their vertices), and the validity of Proposition 1.3.1 can be
checked for each of them. This is, of course, cumbersome (even a computer
would take a moment checking all the 215 possible graphs for triangles and
anti-triangles) and unenlightening.

Proposition 1.3.1 is the first result in a field of graph theory known as Ramsey
theory. I shall not dwell on this field in this course, but let me make a few more
remarks. The first step beyond Proposition 1.3.1 is the following generalization:

r and s betwo positive integers. Let G be a simple

Proposition 1.3.5. Let 
r+s−2
graph with |V ( G )| ≥ . Then, at least one of the following two
r−1
statements holds:
3 Theword “non-neighbor” shall here mean a vertex that is not adjacent to u and distinct from
u. Thus, u does not count as a non-neighbor of u.
Lecture 1, version April 6, 2023 page 11

• Statement 1: There exist r distinct vertices of G that are mutually adja-

cent (i.e., each two distinct ones among these r vertices are adjacent).

• Statement 2: There exist s distinct vertices of G that are mutually non-

adjacent (i.e., no two distinct ones among these s vertices are adjacent).

Applying Proposition 1.3.5 to r = 3 and s = 3, we can recover Proposi-

tion 1.3.1.
r+s−2
 
One might wonder whether the number in Proposition 1.3.5 can
r−1
be improved – i.e., whether we can replace it by a smaller number without
making Proposition 1.3.5 false. In the case of r = 3 and s = 3, this is im-
possible, because the number 6 in Proposition 1.3.1 cannot be made 4
 smaller .
r+s−2

However, for some other values of r and s, the value can be im-
r−1
proved.
 (For example, for r = 4 and s = 4, the best possible value is 18 rather
4+4−2

than = 20.) The smallest possible value that could stand in place
4 − 1
r+s−2

of in Proposition 1.3.5 is called the Ramsey number R (r, s); thus,
r−1
we have just showed that R (3, 3) = 6. Finding R (r, s) for higher values of r
and s is a hard computational challenge; here are some values that have been
found with the help of computers:

R (3, 4) = 9; R (3, 5) = 14; R (3, 6) = 18; R (3, 7) = 23;

R (3, 8) = 28; R (3, 9) = 36; R (4, 4) = 18; R (4, 5) = 25.

(We are only considering the cases r ≤ s, since it is easy to see that R (r, s) =
R (s, r ) for all r and s. Also, the trivial values R (1, s) = 1 and R (2, s) = s + 1
for s ≥ 2 are omitted.) The Ramsey number R (5, 5) is still unknown (although
it is known that 43 ≤ R (5, 5) ≤ 48).
Proposition 1.3.5 can be further generalized to a result called Ramsey’s the-
orem. The idea behind the generalization is to slightly change the point of
view, and replace the simple graph G by a complete graph (i.e., a simple graph
in which every two distinct vertices are adjacent) whose edges are colored in
two colors (say, blue and red). This is a completely equivalent concept, be-
cause the concepts of “adjacent” and “non-adjacent” in G can be identified
with the concepts of “adjacent through a blue edge” (i.e., the edge connecting
them is colored blue) and “adjacent through a red edge”, respectively. State-
ments 1 and 2 then turn into “there exist r distinct vertices that are mutu-
ally adjacent through blue edges” and “there exist s distinct vertices that are
mutually adjacent through red edges”, respectively. From this point of view,

4 Indeed, we saw in Example 1.3.3 (c) that 5 vertices would not suffice.
Lecture 1, version April 6, 2023 page 12

it is only logical to generalize Proposition 1.3.5 further to the case when the
edges of a complete graph are colored in k (rather than two) colors. The corre-
sponding generalization is known as Ramsey’s theorem. We refer to the well-
written Wikipedia page https://en.wikipedia.org/wiki/Ramsey’s_theorem
for a treatment of this generalization with proof, as well as a table of known
Ramsey numbers R (r, s) and a self-contained (if somewhat terse) proof of
Proposition 1.3.5. Ramsey’s theorem can be generalized and varied further;
this usually goes under the name “Ramsey theory”. For elementary introduc-
tions, see the Cut-the-knot page http://www.cut-the-knot.org/Curriculum/
Combinatorics/ThreeOrThree.shtml , the above-mentioned Wikipedia article,
as well as the texts by Harju [Harju14], Bollobas [Bollob98] and West [West01].
There is one more direction in which Proposition 1.3.1 can be improved a bit:
A graph G with at least 6 vertices has not only one triangle or anti-triangle, but
at least two of them (this can include having one triangle and one anti-triangle).
I posed this as a homework exercise (homework set #1, Exercise 1 (a)) in my
Spring 2017 course; see the course page for solutions.

References
[17s] Darij Grinberg, Notes on graph theory, draft of two chapters, 4th April
2022.
https://www.cip.ifi.lmu.de/~grinberg/t/17s/nogra.pdf

[Bollob98] Béla Bollobás, Modern Graph Theory, Graduate Texts in Mathematics

184, Springer 1998.

[Euler36] Leonhard Euler, Solutio problematis ad geometriam situs pertinentis,

Euler Archive - All Works 53, 1741.

[Euler53] Leonhard Euler, James R. Newman, Leonhard Euler and the Koenigs-
berg bridges (English translation of “Solutio problematis ad geometriam
situs pertinentis”), Scientific American 1953.

[FriFri98] Rudolf Fritsch, Gerda Fritsch, The Four-Color Theorem, translated by

Julie Peschke, Springer 1998.

[Harju14] Tero Harju, Lecture notes on Graph Theory, 24 April 2014.

[West01] Douglas Brent West, Introduction to Graph Theory, Pearson 2002.