Introduction to the Module

Dear learner, Welcome to this module on "Fundamental concepts of Geometry".

The module has five chapters. Chapter 1 is on Absolute Geometry, the geometry common to the
Euclidean Geometry and the Hyperbolic Geometry. Chapter 2 deals with the Euclidean Geometry
and chapter 3 on Hyperbolic Geometry chapter 4 discusses on consistency of The Hyperbolic
Geometry and chapter 5 is on the consistency of the Euclidean Geometry.
The approach of the module is axiomatic approach and so you need to remember how an Axiomatic
structure works.
> In an Axiomatic structure, you will have some undefined terms and some unproved assumption
(called Axioms or postulates).
We accept the axioms with out any proof and understand the undefined terms from their properties
explained in the axioms. Then, all the terms other than the undefined terms must be defined
unambiguously and every other assertion (statement) must be proved using the lows of logic.
The statements that need proofs are called Theorems. To prove a theorem you can use
 The previously stated axioms
 The undefined terms
 The previously proved theorems or
 The newly defined terms in the system
So, in this module also, we follow this procedure strictly. At each point, you must identify the
Axioms and the theorems that follow. You must also depend on the definitions given in this module.
You are also advised not to escape over a section and go to the next section because the sections are
logically ordered.
The undefined terms in this course are "Point" , " line" , "Plane" and "Space"
An effort is made to define each and every other terms in this module.
To make illustrations clear, we used diagrams. To represent points we used dots and represent them
by capital letters like:
¿Ρ

 To represent lines, we used straight edges and represent them by letters except in The
Hyperbolic Geometry where efforts are made what lines and planes look like in the usual
Euclidean plane.
L

1
 To represent planes we used parallelograms and represent them by the Greek letters
π , π 1 , π 2 ,⋯

 To represent line segments, rays, angles, triangles, and other polygons we used the
corresponding portions of lines or plane.
There are short (mathematical) ways of expressing some concepts and efforts are made to
explain each and every notation in the particular sections.
Other texts may use other notations and so you must be familiar with the notations in this
module.
.Objectives of the Module
At a successful accomplishment of the module, you will be able to:
 Understand the basic notions in absolute geometry and apply them in Euclidean and
Hyperbolic Geometries,
 Apply the distance function and related concepts to prove the congruence between triangles,
 Understand the basic axioms of Euclidean geometry and its consistency,
 Apply axioms and theorems to solve different problems,
 Understand the properties of congruence and similarity theorems and apply them to solve
problems,
 Understand the basic axioms and unique properties of hyperbolic geometry and its
consistency.
 Understand the Poincare Model
 Distinguish the differences between Euclidean and Hyperbolic Geometries.
 Develope skills in mathematical proofs.

2
3
 CHAPTER ONE
Absolute Geometry
Introduction to The Chapter
This chapter has 12 sections numbered from 1.1 up to 1.12. It discusses concepts that are used in
Absolute Geometry, the geometry that does not use the parallelity Axiom.
Based on their assumptions about parallel lines, there are three different geometries: The
Euclidean Geometry, The Hyperbolic Geometry and the Spherical Geometry. Each has a
different assumption about parallel lines leading to different results in different geometries. But
in Absolute Geometry we are unaware of these assumptions. It is independent of these
assumptions. Based on their assumptions the three geometries may or may not satisfy some of
the results in Absolute Geometry. But fortunately, the two geometries (Euclidean Geometry and
The Hyperbolic Geometry) satisfy almost all of the results of Absolute Geometry. In particular
the Euclidean Geometry satisfies all the axioms in Absolute Geometry.
The Hyperbolic Geometry also satisfies all the axioms of Absolute Geometry except a few of the
first group of axioms called the Axioms of Incidence.
1.1. Axioms of Incidence
These are the first group of axioms in the Absolute Geometry. There are five axioms represented by
I-1 , I-2 , I-3 , I-4 and I-5 together called Axioms of Incidence (or Axioms of connections). They
show the relationships among the undefined terms points, lines, planes and space. They also
guarantee us the existence of a system involving points, lines planes and space that obey all the
properties that will be discussed here after.

! It is important to remember the axioms with their sequential order.

After reading and understanding each axiom paraphrase the statement again.
Ready ! Let us start with I -1. I-0 is just to remind you that every geometric object is a set of
points.

I-0
All lines and planes are sets of points.

4
 I-0 tells us that all lines and planes are sets of points and so like any other sets we can work with
lines and planes as sets.

I-1
Given any two distinct points, there is one and only one line that
contains (or passes through) both of them.

I-1 can also be stated as:

If P and Q are any two distinct points in space, then there is exactly one line ℓ that contains both

P and Q. We represent a line through P & Q by PQ

⃗ or QΡ
⃗
If we have a single point, how many lines are there that pass through that point?
The answer to this is not immediate at this time because we don't know how many points there are in
space and how they are situated in space.
Wait! After studying all the axioms of incidence, we may have some answer to this question.
You can also ask yourself that, if we have three points, can we have a line containing them?
The answer to this is, well! If we have three points in space, then by I-1 there is exactly one line
containing the two points and so if the third point is also on that line, then we can get a line
containing all of them. But if the third point is not on that line we can not have any other line
containing all the three points. Such three points that are on the same line are called collinear points.
Practice the following.
Put any two dots on a piece of paper or on the black board and connect them by a straight edge. How
many straight lines can you draw?
Did you verify the statement in I-1? Good! Now, paraphrase I-1.
Next , let us go to I-2.

I-2
Given three distinct points not on the same line, there is one and
only one plane that contains all the three points.

I-2 can also be stated as:

5
 If P,Q & R are three distinct points in space that are not on the same line, then there is a unique plane
(call it π ) that passes through P , Q and R.
In I-2 , the condition that the three points are not on the same line is necessary to have a unique
plane.

! Ask yourself
If the three points are on the same line, how many planes are there that contain all of them?

I-3
If two points of a line lie on a plane, then every point of the line lies
on that plane.
I-3 says that if any two points of a line L lie in a particular plane π , then the whole line lies on
that plane.
Symbolically, if P and Q are two distinct points on a line L and if P and Q are on a plane π ,

then L is on π .
It is not possible that two points of a line lie on a plane and some other point of the line not on that
plane.
Check Your Progress
Now, can you state I-1 , I-2 and I-3 with out referring to them? Try it!
Next, let us see I -4.

I-4
If two planes have one point in common, then they have at least a
second point in common.
I-4 tell us that the intersection of two planes can not be a single point. It is either empty or more than
a single point.
Still now none of I-1 up to I-4 guarantee us the existence of points, lines and planes. They only tell
us that if they exist they satisfy the indicated properties.
The last axiom of incidence (I-5) however, guarantees us the existence of points, lines and planes in

I-5space.
i. On every line, there are at least two points.
ii. On every plane there are at least three points that are not on the same line6
iii. In space, there are at least four points that are not on the same plane.
What is the difference between I-1 and I-5 (i).
What is the difference between I-2 and I-5 (ii).
In statements of the form "there exist at least ---" you could have more objects than the
specified number satisfying the given condition.

In statements of the form "-----there exist at most -----" you can't have objects more than the
specified number that satisfy the given condition.
Now, let's come back to I - 5.
I-5(i) says that whenever we have a line, we are sure that there are at least two distinct points on the
line and so whenever you are given a line, you can always take two distinct points on the line. I - 5
(ii) says that whenever we have a plane, we are sure that there are at least three points that are not on
the same line on the plane and so whenever you are given a plane , you can pick any three distinct
points on the plane that are not on the same line. I-5 (iii) tells us that in space, there are at least four
points that are not on the same plane consequently, there are more than one plane and more than one
line in space.

! Check your Progress

At this point you must be able to state each of the axioms I - 1 up to I - 5. Try it !
Next, we will manipulate two or more of the axioms together to study more properties of space such
statements need logical proofs.

Do you know the different methods of proofs?

?
Here are some of the methods of proofs that we use in this module.

1. Direct Proof

7
 In a direct proof, to prove statements of the form "If p, then q", we assume that there is a
system satisfying the hypotheses p and by a step by step logical reasoning we arrive at the
conclusion q.
2. Indirect proof
In an indirect proof, to prove statements of the form "If p, then q " we assume that the
conclusion q is wrong and step - by - step we arrive that the hypotheses p is also wrong.
3. Proof by Contradiction
In proof by contradiction, to prove a statement of the form "If p, then q" we assume that the
hypotheses p is true and the conclusion q is wrong and so we assume the negation of q is
true. Then step - by - step we arrive at a conclusion that disagrees with one of the axioms or
one of a previously proved theorem. And so we arrive at the conclusion q is not true is wrong
there by we conclude that q is true.
4. Proof by contra positive
In proof by contrapositive, to prove a statement of the form "If p, then q" we prove its contra,
positive "If ¬¿ ¿ q, then ¬¿ ¿ p" by using any of the methods above.
Hence since an implication is equivalent to its contra positive we conclude that the statement
"If p, then q" is true.
5. Proof of existence
To prove statements of the form “--- there exists ---- “we have to construct at least one
system satisfying the conditions of the statement.
6. Proof of Uniqueness
To prove an object satisfying a property in the system is unique (the only one of its kind) we
take two objects satisfying the conditions of the statement and step - by - step we arrive at the
conclusion that the two objects are exactly the same.

7. Proof existence and uniqueness

To prove statements of the form "--- there exists a unique ----" we first show the existence
and then show the uniqueness of the object satisfying the conditions of the statement.
8. Proof of bi - implications

8
 To prove statements of the form "p if and only if q" we must prove both” If p, then q" and "If
q, then p" by using any of the convenient methods discussed above.
9. Proof by Cases
In proof by cases, we must prove the statement for each of the possible cases and so conclude
that the statement is true.
10. Disprove by counter example
To disprove a mathematical statement (To show that it is false), it is sufficient to give a
single case in which the statement becomes false. But to prove a statement thousands or even
infinite examples can not prove the statement.
Now, let us see some consequences of Axioms of Incidence.

Theorem 1.1.1
Given a line L and a point P not on L , there is exactly one plane
containing L and P.
Proof:
Suppose a line L and a point P not on L are given.
Then by I - 5 i, there are at least two points (say Q and R) on L .
Clearly , P , Q and R are distinct and not lie on the same line (Why?!)
Hence by I - 2 , through P , Q and R there is a unique plane (say π ) that contains P , Q and R.
Then, since Q and R are on π , the line through Q and R (i.e. L ) is on π .( by I - 3)

Hence there is a unique plane containing L and P.

Theorem 1.1 can also be restated as:
A line and a point not on the line determine a unique plane.
? What is the method of proof used in the above argument.
Convention:In this module, when we say “two,three,…objects ”,we mean “two,three,…distinct
objects.”

Theorem 1.1.2
If two planes intersect, their intersection is exactly a line.
Proof:

9
Let π 1 and π 2 be two intersecting planes. Let Ρ∈ π 1∩π 2 .

Then by I - 4 , there is at least a second point say Q≠Ρ in the intersection of π 1 and π 2 .

Then by I - 1 , through P & Q , there is unique line, call it L .

Now , since P and Q are on π 1 , by I - 3, the line through P and Q (i.e. L ) is on π 1 , Hence

L ⊆π 1 similarly, since P and Q are on π 2 , by I - 3 again, the line through P and Q (i.e. L )

is on π 2

Hence L ⊆π 2 .

⇒ L ⊆ π 1∩π 2 .

To show that L is the only intersection of π 1 and π 2 , assume that there is a point P that is not

on L but Ρ∈ π 1∩π 2 .

Then by Theorem 1.1, there is a unique plane that contains both L and P. But both π 1 and π 2

contain L and P.
⇒ π 1 =π 2 . contradicting that π 1 and π 2 are distinct. Hence it is not possible that there is a

point not on L but in the intersection of π1 ∧ π2 .

Hence the only intersection of π 1 and π 2 is the line L .

Theorem 1.1.3
If two lines intersect, then their intersection is a single point.

Proof:

Let L1 and L2 be two intersecting lines. Let Ρ∈ L1 ∩L2 .

If P is the only intersection of L1 and L2 then we are proved.

Assume that there is a point Q other than P that is in the intersection of L1 and L2 .
But by I - 1, there is one and only one line through P and Q.
⇒ L1 =L2 . Contradicting that L1 and L2 are distinct.

10
Hence P is the only intersection of L1 and L2 .
? Which methods of proof are used in the above argument?

Theorem 1.1.4
Two intersecting lines determine a unique plane.
Proof:

Let L1 and L2 be two intersecting lines. Then by Theorem 1.3, their intersection is a single point

(say P). Then on L1 there is a point Q other than P. (Why?)

Then Q is not on L2 (Why?)

Then by Theorem 1.1. there is a unique plane (say π ) containing L2 and Q .

Since Ρ∈ L2 and L2 ⊆π , we have Ρ∈ π .

Now P and Q are on π and so the line through P and Q (i.e. L1 ) is on π .

Hence π contains both L1 and L2 .

? Which methods of proof are used in the above argument?

? Can you give a different proof for Theorem 1.4?

Theorem 1.1.5
If a line L that is not on a plane π interests π , then L∩π is a single
point.
Proof:
Let L be a line intersecting a plane π
⇒ there exists a point Ρ∈ L∩π .

If P is the only intersection of L and π , then we are done. Assume that there is a point Q other

than P in the intersection of L and π

Then, since P and Q are on π , the line through P and Q (i.e. L ) is on π .
Hence if L is not on π , then L∩π is a single point.

11
 ? Which methods of proof are used in the above argument?

Theorem 1.1.6
On every plane there are at least three lines.
Proof:
Let π be any plane.
Then by I - 5(ii) there are at least three points (say P, Q and R) that are not on the same line.

L1 P π

Q
L3 R
L2

Then through P and Q , there is a unique line (call it L1 ), through P and R a unique line (call it

ℓ 2 ) and through Q and R a unique line (call it L3 ). To show L1 , L2 & L3 are distinct,

Since Q∈ L1 and Q∉ L2 , then L1≠L2

Since
Ρ∈ L1 and Ρ∉ L3 , then L1 ≠L3

Since
Ρ∈ L2 and Ρ∉ L3 , then L2 ≠L3 .

Since P and Q are on π , the line through P and Q (i.e. L1 ) is on π (by I - 3)

Since P and R are on π , the line through P and R (i.e. L2 ) is on π (by I - 3)

L
Since Q and R are on π , the line through Q and R (i.e. 3 ) is on π (by I - 3)
Hence there are at least three distinct lines on π .
Since π is arbitrary , on any plane there are at least three lines.

Theorem 1.7
Given any line L , there is at least one line that does not intersect L .
Proof:

12
Let L be any line. Then by I - 5 i there are at least two points on L (say P and Q).
But by I - 5 (iii) there are at least four points in space that are not on the same plane.
Hence in addition to P and Q there are at least two points R and T such that P,Q,R and T are not on
the same plane.

Now, by I - 1, there is a unique line through R and T (call it L1 ).

Next, we want to show that L and L1 do not intersect.

Assume that L and L1 intersect say at a point O.

Then by Theorem 1.4, there is a unique plane π containing L and L1

⇒ P , Q ,R and T are on π
But this contradicts that P,Q,R and T are not on the same plane.

Hence L1 and L2 intersect is not possible.

Hence for any line L there is at least one line L1 that does not intersect L .
Which methods of proof are used in the above argument?
Dear learner, these are not the only consequences of the Axioms of Incidence. You can have more.
Look ! In each of the proofs of the above theorems, we have never used any concept other than the
Axioms of Incidence.
In the following exercises use the Axioms of incidence only to prove each of the following
propositions.
Exercise 1.1
I. Use the Axioms of Incidence only and prove each of the following propositions
1. Proposition 1
Given any point P, there is at least one line that does not pass through P.

2. Proposition 2
Given any point P, there are at least two lines in a plane that pass through P.
3. Proposition 3
There are at least six lines and four planes in space.
4. Proposition 4

13
 Given any plane π , there is at least one line that intersects π at exactly one point.
5. Proposition 5
Every line lies on at least two distinct planes.
1.2. Distance functions and the Ruler postulate
Dear Learner, in the previous section we have worked with the Axioms of Incidence and its
consequences. But the Axioms of Incidence alone do not show all the properties of the general
geometric spaces.
For example, using the axioms of Incidence alone, we are not sure if there are only two points on
every line or there are more points on a line.
For this reason, we need more properties of the general geometric space. This section will provide
more points in space.
The distance function

Remember that for any set of points S, S x S = {( P ,Q ) : P ∈ S and Q ∈ S } .

Now , we define the distance function as follows.
Definition:
A function d : S x S → ℜ satisfying the following conditions is said to be a distance function on
S, where S is a set of points.

i. d ( P ,Q ) ≥0 for all Ρ∈ S and Q∈ S

ii. d ( P ,Q ) =0 if and only if P=Q

iii. d ( P ,Q ) =d (Q , P ) for all Ρ ∈ S , Q ∈ S

Notation
d ( P , Q ) is simply denoted as PQ ,and PQ is said to be the distance between P and Q.
Definition
Let L be a line considered as a set of points. A one - to - one correspondence g from L on to R

( g: L→ R ) is called a coordinate system for L .

Theorem 1.2.1.
If g : L→ℜ is a coordinate system for a line L , then the function

d : L x L →ℜ given by d ( P ,Q ) =|g ( P )−g ( Q ) | is a distance function.

14
Proof

Let g : L→ℜ be a coordinate system for L . i.e. g is one - to - one and onto.

i. d ( P,Q ) =|g ( P )−g ( Q )|≥0 for all P∈ L and Q ∈ L .

ii. d ( P ,Q ) =0 if and only if d ( P ,Q ) =|g ( P )−g ( Q ) |=0

if and only if g ( P )−g ( Q )=0

if and only if g ( P )=g ( Q )

But since g is one - to - one we must have P=Q .

iii. d ( P,Q )=|g ( P )−g ( Q ) |

=|g ( Q ) −g ( P ) |
= d (Q , P )
Hence d satisfies all the properties i - iii of a distance function and so d is a distance function.
Definition:
If g is a coordinate system for L , then

i) For each Ρ∈ L , g ( Ρ ) is called the coordinate of P.

ii) The Pre image of 0 ∈ R is called the origin of L .
Dear learner, do you know how to draw a number line?
> To draw a number line we draw a straight line L and choose any point on L and choose it
as the origin O and coordinatize it with the real number 0.
Next , we choose a unit u of measurement. Choose a point U to either side of O at a distance of 1
unit and coordinatize it with the real number 1.
Then we use the unit of measurement u and to each point P on L that is on the same side of O as
U we coordinatize it with the positive real number x where the distance from O to P is x units. On
the other hand to each point P in the opposite side of O as U, we coordinatize it with the negative
real number x where the distance from O to P is /x/ units.
Conversely, for any real number x, correspond x as follows

a/ if x=0 , correspond it to O.

15
 b/ if x >0 correspond it to a point P that is at a distance of x units from O and to the side of
U.

c/ if x <0 correspond it to a point P that is at a distance of /x/ units from O and to the side
opposite to U.
With this correspondence, there is a one - to - one correspondence between set of points on ℓ and
set of all real numbers.
This correspondence is the natural correspondence but it is not the only correspondence.
For an arbitrary line, you may have a different correspondence.
The following Axiom guarantees you that every line has a coordinate system.

D: (The Ruler Postulate)

Every line has a coordinate system.

Convention:
From now onward when we say the distance between two point P and Q denoted by PQ, we mean
the distance relative to some coordinate system.
Now, let us see some consequences of the Ruler postulate.
Theorem 1.2.2.
If f : L→ ℜ is a coordinate system for L then g : L→R given by
g ( Ρ )=− f ( Ρ ) for all Ρ∈ L is also a coordinate system for L .
Proof:
one - to - oneness

Let g ( Ρ )=g ( Q )
⇒− f ( Ρ )=− f ( Q )
⇒ f ( Ρ )=f ( Q ) sin ce
⇒ Ρ=Q (since f is a coordinate system)
Hence g is one - to - one .
Onto ness
Let x ∈ ℜ .Then −x∈ ℜ and since g is onto, there exists a point P on L such that

16
g(P)= −x .Then f (P )=−(−x )=x

Hence f is onto.
Theorem 1.2.3
If f is a coordinate system for L and k is a constant then g : L→R given
by
g ( Ρ )=f ( Ρ ) +k
Proof: Exercise
Hint: Show that g is one - to one and onto.

Theorem 1.2.4. (The Ruler placement Theorem)

If L is a line and P, Q ∈L , Ρ≠Q then L has a coordinate system f
such that f ( Ρ )=0 and f ( Q ) >0 .

Proof:
By the Ruler postulate L has a coordinate system say h and

h ( Ρ )=a and h ( Q )=b ,a , b ∈ R and a≠b .

Then g : L→ℜ given by g ( Τ ) =h ( Τ ) −a is also a coordinate system for L (by Theorem

1.2.3) and g ( Ρ )=h ( Ρ ) −a=a−a=0

If g (Q )=h ( Q )−a=b−a is positive then take f :=g if g ( Q )=h ( Q ) −a=b−a<0 , take

f :=−g
Then f is also a coordinate system for L and f(P) =0 and f(Q)>0.
>To understand the Ruler placement Theorem remember how you measure the length of an edge.
You put the 0 point of your ruler at one point and read the number on the ruler at the other point.
1.3. The Axioms of Betweenness
This section guarantees us the existence of more points, lines and planes in space we begin by the
definition of betweenness
Definition:
Let P, Q, R be three distinct points on a line we say Q is between P and R and write

17
 P - Q-R if PQ + QR = PR
P Q R

Note: P - Q - R if and only if R - Q - P.

Notation
If x, y, z ∈ ℜ we write x - y - z to mean either x < y < z or z < y < x

Theorem 1.3.1
Let f be a coordinate system for L . Suppose P,Q,R are three distinct
points on L such that f ( Ρ ) =x , f ( Q )= y , f ( R )=z.
Then Ρ − Q − R if and only if x−y−z .
Proof:
( ⇒ ) Forward proof,
Suppose P - Q - R
⇒ PQ + QR = PR

⇒ | f ( Ρ )−f (Q ) | +| f ( Q )−f ( R ) |=| f ( Ρ ) −f ( R ) |

= | ( f ( Ρ )−f ( Q ) ) + ( f ( Q )−f ( R ) )|

But |a+b|=|a|+|b| if and only if a and b have the same sign

⇒ f ( Ρ ) −f ( Q ) >0 and f ( Q )−f ( R ) >0 or f ( Ρ )−f ( Q ) <0 and f ( Q )−f ( R )<0 .

 f ( Ρ ) >f (Q ) and f ( Q ) >f ( R ) or f ( Ρ ) <f ( Q ) and f ( Q ) <f ( R )

 f ( Ρ ) > f ( Q ) >f ( R ) or f ( Ρ ) <f ( Q ) < f ( R )

 x> y> z or x < y < z

 x− y−z

( ⇐) Backward Proof:
Suppose x - y - z
 x< y< z or z < y < x

 ΡR=| z −x |=| z−y+ y−x|

18
But z - y and y - x have the same sign and so | z−y+ y−x |=|z−y|+| y−x|
 ΡR=|z−x|=|z− y+ y−x|
= |z− y|+|y−x|
= PQ + PR
Hence, P - Q - R
Theorem 1.3.2
Of any three distinct points on a line, exactly one of them is between the other two.

Proof:
Let P, Q, R are lines on a line L and f be a coordinate system for L and

f ( Ρ )=x , f ( Q )= y and f ( R ) =z .
Then by the order property of real numbers exactly one of x, y & z is between the other two. Hence
either
x is between y and z ⇒ Q-P-R
y is between x and z ⇒ P-Q-R
z is between x and y ⇒ P-R-Q
Consequently, exactly one of P, Q, R is between the other two.

Theorem 1.3.3
Given any two distinct points P and Q on a line L , there is at least one
point R on L such that P-R-Q.

Proof:
By the ruler postulate, L has a coordinate system say f .

Let f ( Ρ )=x and f ( Q ) = y

Since f is one-to-one , x≠ y . Then either x< y or y < x . With out loss of generality (WLOG)

x+ y x+ y
x< <y
say x< y .Then by property of real numbers 2 is a real number and 2 .

19
 x+ y
f ( R)=
Since f is on to, there is a point R on L such that 2 so by theorem 1.3.1 since

x+ y
x− − y we have Ρ−Q−R
2 .
Hence for any two points on a line, there is at least one point between them.
Dear learner: What do you understand from Theorem 1.2.7? That is how many points are there
between any two points? By applying this theorem repeatedly what do you understand?

Theorem 1.3.4.
If P and Q are any two points on a line L then there exists at least one
pointExercise
Proof: R on L such that P - Q - R.
(Hint: Let f ( Ρ ) =x , f ( Q )= y, x< y then take a point R such that f ( R )= y +1)
Corollary
If P and Q are two points on a line L, then the sets

1. { R ∈L |Ρ−R−Q } is infinite

2. { R ∈L |Ρ−Q−R } is infinite

3. { R ∈L|R−P−Q } is infinite
Remark
From the property of real numbers and Theorems 1.3.3 and 1.3.4 we conclude that every line
contains uncountably infinite points. Consequently, the results of this section and the Axioms of
Incidence provide more properties of space. For example we can prove the following proposition.

Proposition
Every plane contains uncountably infinite lines.

Proof
By I - 5 ii. every plane contains at least three points that are not on the same line. Let the points be
P,Q, and R.

20
Then through P and Q there is exactly one line say L .
Then by the results of this section there are uncountably infinite points on L .
Now for each point T on ℓ there is exactly one line call it L T passing through R and T.
Next, to show that any of these lines are distinct let M and N are distinct points on L . To show

MR and ⃗
⃗ NR are distinct, assume they meet at a point O.

But there is exactly one line OR )

(⃗ through O and R. But both MR and ⃗
⃗ NR pass through O and

R which implies that M and N are on OR

⃗ which intern shows the line through M and N (that is

the line L ) contains R

Contradicting that P,Q and R are not on the same line.
Hence any of the above lines are distinct.
Hence there are as many lines on the plane as there are points on L .

P Q
N M L

! You can also prove that there are uncountably infinite planes in space.

Check Your Progress

At this time you must be familiar with some of the methods of proof and see how argument goes
step by step. If you have any difficulty in understanding any of this section go back to the beginning
and see how the coordinate system is constructed.
1.4. Segments, Rays, Angles, Triangles and Congruency of Segments

21
In this section you will work with some Geometric objects. Such as line segments, Rays, Angles,
Triangles.

?Are Are you familiar with these objects in the Euclidean plane? Draw a line segment, a ray ,an angle
and a triangle.
These objects have their counter parts in Hyperbolic Geometry also. Their definitions in both the
Euclidean Geometry and the Hyperbolic Geometry are the same.
Definition
Let P and Q be any two distinct points. Then the set of all points between P and Q including P and

Q is called a line segment with end points P and Q and is denoted PQ or QP .

i .e
ΡQ={ Χ : Ρ− Χ−Q } U { P ,Q } .
In Euclidean Geometry a line segment is represented as

P Q

You can see that for any two points P and Q, the line segment ΡQ is exactly the same as

QΡ .

i.e. you can call a line segment with end points P and Q either line segment ΡQ or line

segment QΡ .
This is given as our next theorem.

Theorem 1.3.5
For any two points P and Q, ΡQ = QΡ
Proof:

ΡQ = { X : P− X−Q } ∪ { Ρ , Q } .

but { X : Ρ− X−Q }= { X :Q−X −P }

22
 because if X is between P and Q, then X is between Q and P and also { Ρ , Q }={ Q , Ρ }
because in a set, the order we list its elements does not change the set.
Hence,
ΡQ={ X : Ρ− X−Q } ∪ { Ρ , Q }
= { X:Q−X−Ρ } ∪ {Q , Ρ }
=QΡ
Hence ΡQ=QΡ .
Definition

Let P and Q be any two distinct points. Then the line segment PQ together with the set of all

points R such that P - Q - R is called a ray with end point P. and denoted ΡQ
⃗ .

i.e. ΡQ={ Χ :P − Q − Χ } U ΡQ
⃗
In Euclidean Geometry a ray is represented as

P Q

Definition
The union of two rays with a common end point is called an angle.
i.e.

Let P, Q, R be three distinct points. Then the union of ΡQ

⃗ and ΡR
⃗ is called an angle QPR

denoted as ∠ QPR or ∠ RPQ .Whenever there is no confusion we write the angle ∠ QPR by
∠P

P is called the vertex and ΡQ

⃗ and ΡR
⃗ are called the sides of ∠ QPR.

23
 P
R

Definition
Let P, Q, R be three distinct points that are not on the same line.

Then the union of ΡQ , PR and QR is called triangle PQR denoted as Δ PQR.

P, Q, & R are called the vertices of Δ PQR and ΡQ , ΡR and QR are called the sides of Δ
PQR.

Given a ray with end point P, we can use any other points on the ray to name the ray i.e.

give a ray ΡQ
⃗ and R is a point on ΡQ
⃗ other than P, then ΡQ
⃗ = ΡR
⃗ . This is given
as our next theorem.

Theorem 1.3.6
ΡQ other than P, then ⃗
If R is a point on ⃗ ΡQ = ⃗
ΡR .
Proof is left as an exercise
Another fact about line segments is that a line segment with end points P and Q is named as

either PQ or QΡ . i.e. if line segments ΡQ

⃗ and RS represent the same line

segment then the sets of end points { Ρ , Q }= { R , S } .

Next, we define what is meant by line segments are congruent.
Definition

24
 Line segments PQ and RS (on the same or different planes) are said to be congruent if
PQ = RS.i.e. they have the same length.

?
You can verify the following statements easily.

1/ Every line segment AΒ is congruent to itself i.e. AΒ≡AΒ Reflexive Property )

2/ If AΒ≡CD , then CD≡ AΒ . Symmetric Property )

3/ If AΒ≡CD and CD≡ΕF , then AΒ≡ΕF ( Transitive Property )

Hence congruence of segments is an Equivalence relation.

Next, we will see more properties of congruent segments.

Theorem 1.3. 7 (The segment construction Theorem)

ΡQ , there is a unique point T on
Given a line segment AΒ and a ray ⃗
ΡQ such that
⃗ ΡΤ≡ΑΒ .

A P Q T

Proof:

By the Ruler placement Theorem there is a coordinate system f for ΡQ

⃗ such that f ( Ρ )=0 and

f ( Q ) >0 .
Let AB = x > 0.

Since f is onto, there is a unique point T on ΡQ

⃗ such that f ( Τ )=x >0

25
Now since f ( Τ ) >0=f ( Ρ ) ∧f ( Q ) >0 , then either f ( Ρ ) −f ( Τ )−f ( Q ) or

f ( Ρ ) −f ( Q )−f ( Τ ) or f ( Τ )=f (Q ) .
Consequently, either P - T - Q or P - Q - T or T = Q.

In all the cases T is on ΡQ

⃗ ,
Now , PT = / f (P) - f (T) /
=/0-x/
=x
= AB

 ΡΤ≡AΒ

Hence there is a unique point T on ΡQ

⃗ such that ΡΤ≡AΒ

The above statement tells us that on a given line, we can always construct a line segment
congruent to a given line segment.

Theorem 1.3.8 - (Segment Addition Theorem)

Let AΒ and ΡQ be line segments and C and R be points on AΒ and ΡQ

respectively such that A - C - B and P - R - Q, AC≡ΡR and CΒ≡RQ . Then

AΒ≡ΡQ
B

C P R Q
A

Proof:
Since A - C - B and P - R - Q
AC + CB = AB and PR + RQ = PQ

26
Now AB = AC + CB = PR + RQ = PQ
(Since AC = PR and CB = RQ) .

Hence AΒ≡ΡQ .

One application of this is for example if A, B , C and D are points on a line such that

A - B - C - D and AΒ≡CD , then AC≡ΒD .

A B C D
because AC = AB + BC
= CD + BC
= BC + CD
= BD
As a result of the above theorem we have also the segment subtraction corollary.
Corollary: (Segment subtraction)

If AΒ and ΡQ are line segments and C and R are points on AΒ and ΡQ respectively such that

A - C - B and P - R - Q and if AΒ≡ΡQ and AC≡ΡR , then CΒ≡RQ .

Definition

Let AΒ be a line segment. Then a point P on AΒ such that AΡ≡ΡΒ is called the mid point

of AΒ .

Theorem 1.3.9
Every line segment has a unique mid point.
Proof:

Let AΒ be given.

Then AΒ
⃗ has a coordinate system f.

27
 x+ y
z=
. Then there is a unique point P on AΒ
Let f ( A )=x and f ( Β )= y . Take 2 ⃗ such

that f ( Ρ ) =z .
x+ y
x− −y , A − Ρ − Β
Since 2
x+ y
AΡ=| f ( Ρ )−f ( A ) | = | −x|
2
x+ y−2 x
=| |
2
y−x 1
=| |= |y−x|
Then 2 2
x+ y
| f ( Β ) −f ( Ρ ) |=|y− |
and P B = 2
2 y−( x+ y )
=| |
2
2 y−x− y
=| |
2
y−x
=| |
2
1
= |y−x|
2
= AΡ
 AΡ≡ΡΒ .

Hence, there is a unique point P on AΒ such that AΡ≡ΡΒ .

i.e. every line segment has a unique mid - point.
Exercise 1-3
Use the Axioms of Incidence and the results of sections 1.2 and 1.3 and prove each of the following.
1/ 5 Every line segment contains infinitely many points.
2/ 4 Given any line L, there are infinitely many planes containing L.

3/ 3 If P,Q and R are three collinear points, such that P - Q - R , then ΡQ=⃗
⃗ ΡR

28
 4/ 2 Given an angle < ABC , if D is a point on ΒA
⃗ \ {Β } and E is a point on ΒC
⃗ \ {Β } ,
then < ABC = < DBE.
5/ 1 If B is between A and C and D is between B and C, then D is between A and C

1.5. Plane Separation Postulate

This section deals with the third group of Axioms in Absolute Geometry, namely the plane
separation postulate.
In this section you can use the results found in both the Axioms of Incidence and the Axioms of
order discussed in the preceding sections whenever you need them.
We start with the first axiom known as the plane separation Axiom (PS -1).

PS -1: (Plane Separation Postulate)

Given a plane π and a line ℓ in π , ℓ divides the rest of points on π in to
two sets H1 and H2 in such a way that

i. If P and Q are on the same set (H1 or H2), then ΡQ does not intersect ℓ ,

while
π
ii. If R and S are on Pdifferent sets one is on H 1 and the other is on H2, then
Q L
RS intersects ℓ .
Each of these two sets is called a Half -Plane and ℓ is called the edge of H1 and

P Q ℓ

H1
H2 S

Then plane separation Axiom has some consequences that we see them as theorems.

Theorem 1.5.1
Each of the two half - planes contain at least three points that are not on
29
the same line.
 Proof:
Let π be a plane and L is a line on π .
Then by I - 5 i, there are at least two points (say P and Q) on L .
But by I - 5 ii. π Contains at least three points that are not on the same line and so there is at least

one point (say R) on the plane π but not on L .

Then R is on one of H1 or H2. With out loss of generality say R is on H1.

T π
S L
R
H1 Q
P
M
N O

Then by Theorem 1.3.4, there is a point T such that P - R - T and there is a point S such that Q - R -
S.

Now, verify that RΤ and RS do not intersect L !

And so R , T and S are on the same set H1.

Now again verify that R , T and S are not on the same line !
Hence H1 contains at least three points that are not on the same line.
Now, let us see H2.

By Theorem 1.2.8 again there is at least one point (say M) such that R - P - M. Then RM
intersects L at P and so R and M are on different sets and so M is on H2.
Now, construct points N and O such that Q - M - N and P - M - O and verify that M , N and O are
not on the same
Theorem line and are on H2.
1.5.2.
Given a triangle Δ AΒC and a line L on the plane of Δ AΒC
, if L does not pass through any of the vertices A, B and C,
30
then L can not intersect all the three sides of Δ AΒC .
 Proof:
L
M
C
A
C

B
Let L be a line on the plane determined by A, B , C not passing through any of A, B and C. If

L does not intersect any of ΑΒ , ΑC and ΒC , then the theorem is satisfied.

Suppose L intersects at least one of ΑΒ , ΑC or ΒC . Without loss of generality assume L

intersects ΑΒ . But by assumption L does not pass through A or B. Hence there exists a point M

on ΑΒ - { Α , Β} such that M is on L .
Then M is between A and B.
Then by the plane separation postulate L divides the plane in to two half planes H1 and H2. Since

L intersects ΑΒ at M A and B are on different half planes (Say A is on H1 and B is on H2).

Since L does not pass through C, then C is on H1 or C is on H2.

If C is on H1 , then C and A are on the same half plane and so L does not intersect AC .

If C is on H2, then C and B are on the same half plane and so L does not intersect ΒC .

Hence in both cases L does not intersect at least one of ΑC or ΒC .

Hence L can not intersect all the three sides of Δ AΒC .

31
 As you have seen in the above proof, if L is a line in the plane of Δ AΒC and does not pass

through any of A, B and C and intersects ΑΒ , then L also intersects exactly one of AC or

ΒC . This is given as the next theorem called Pasch's Axiom.

Theorem 1.5.3 (Pasch's Axiom)
Let L be in the plane of Δ AΒC
L does not pass through any of A , B and C and intersects ΑΒ , then
IfProof: L
Suppose L is on the plane of triangle ABC and does not pass through any of A , B and C.
intersects exactly of AB or ΒC .
If L intersects ΑΒ , then as the proof of the above theorem say A is in H1 and B is in H2. Then

since L does not pass through C, C is in H1 or C is in H2.

If C is in H2, then C and A are on different half planes and so L intersects ΒC but L does

not intersect AC
If C is in H1 , then C and B are on different half planes and so L intersects AC but L does

not intersect ΒC .

Hence in both cases, L intersects exactly one of AC or ΒC .

The above statement is a theorem but we called it Pasch's Axiom.

This is because by assuming that the Pasch's Axiom is true we can prove that the plane separation
Axiom is true. Try this first by yourself and compare it with the proof below.

Theorem 1.5.4
Assuming the Pasch’s Axiom is true the Plane separation Axiom is true.
Proof:
Suppose the Pasch's Axiom is true. Now let a line L on a plane π is given.
Then there are at least two points P and Q on L . ( Why ?)
And there is at least one point R on π that is not on L . (Why?)
Then define H1 by

32
 Η 1 := { Χ ∈ π : RT does not int er sect L } U { R } .

R ¿ L
Q
P

H1 \ { R} is non - empty because there is at least one point T such that P-R-T and so RΤ does not

intersect L . (Why?)

Now, let X and Y be any two points in H1 . If X = R or X = Y clearly XY does not intersect

L .So, suppose X and Y are different from R. Then RX and RY do not intersect L and so X

and Y are not on L .

Case 1
If R, X and Yare on the same line, then either X - R - Y or X - Y - R or R - X - Y.

In all these cases ΧΥ does not intersect L . (Why?)

Case 2
R , X and Y are not on the same line.

X Y

33
In this case L does not intersect RΧ and RΥ and so by Pasch's Axiom L can not

intersect ΧΥ . Otherwise, if L intersects ΧΥ , it also intersects either RΧ or RΥ .

In both cases, if X and Y are on H1, then ΧΥ does not intersect L .

Put H2 = T \ ( Η1 U L)
Then for any point X on H2 X is not on L and RΧ intersects L .
Now, Let X and Y be in H2.
Case 1
If X , Y and R are on the same line, then either R - X - Y or R - Y - X (X - R - Y is not possible
Why?)

R L
X
Y

In both these cases ΧΥ does not intersect L (Why?)

Case 2
If X , Y and R are not on the same line,

X Y

L intersects both RΧ and RΥ and so by Pasch's Axiom L can not intersect ΧΥ .

In both cases , if X and Y are on H2 , then ΧΥ does not intersect L .

On the other hand let X and Y be points in different sets.
Say X is on H1 and Y is on H2.
Case 1

34
X , Y and R are colinear.
Then either R - X - Y or X - R - Y ( X - Y - R is not possible. Why?)

X L
R
Y

In both of these cases L intersects ΧΥ (Why?)

Case 2
X , Y and R are not colinear.

R L

X Y
In this case L intersects RΥ but does not intersect RΧ and so by Pasch's Axiom L

intersects ΧΥ .

In both cases ΧΥ intersects L .

Hence we have found two sets H1 and H2 such that for any two points X and Y on the same set

ΧΥ does not intersect L while for any two points X & Y on different sets ΧΥ intersects
L .
Hence by assuming Pasch's Axiom we have proved the plane separation Axiom as a theorem.
Definition
Let L be a line and Ρ∈ L .

We say that Q and R on L are in opposite sides of P if Q - P - R and they are on the same side of P
if P - Q - R or P - R - Q.

Theorem 1.5.5

35
Let L be a line on a plane π . If P and Q are on opposite sides of L and
Q and T are on the same side of L , then P and T are on opposite sides of
L .
Proof:
Let L be a line on a plane π .
We say P and Q are on the same side of L if P and Q are on the same half plane and say P and Q

are on opposite sides of L if they are on different half planes.

Let P and Q be on opposite sides of L , say Ρ∈ H1 and Q ∈ H2.
If Q and T are on the same side of L , then T ∈ H2
Hence P is in H1 and T is in H2 and so P and T are in opposite sides of L .
Next , let us see one property of rays.
Theorem 1.5.6
Let L be a line on a plane π dividing π into two half planes H1 and
H . If P is a point on
2
Proof:
L and Q is a point on π but not on L then except
ΡQ lie on the same half
the end point P, all the other points of the ray ⃗
Q
L

WLOG assume Q is on H1

Then Let T be a point on ΡQ

⃗ other than P.
Then P - T - Q or P - Q - T.

To show that QΤ does not intersect L , assume QΤ intersects L at a point R, then R can
not be P.
From this we understand that the line through P & Q intersects L at two different points (P and
R).

36
This implies that the line ΡQ
⃗ is L itself and so Q is on L .
But we have already assumed that Q is not on L .

Hence QΤ can not cross L which means Q and T are on the same half plane. In other words

every point of ΡQ
⃗ other than P lies on the same half plane with Q.

Convention: If L is a line and P is a point not on L , when we say the side of L

containing P, we mean the half plane determined by L in the plane of L and P that
contains P.

? Look at the following diagram of an angle ∠ ABC which points are in the interior of
∠ ABC ? Which points are in the exterior of ∠ ABC and which points are on ∠

ABC ?

A
¿ H ¿ D

¿ G
¿ I B J
¿ E
¿ F C

Next, we define the interior and exterior of an angle in a formal language.

Definition:
Let A,B and C be three non-collinear points. Then the interior of ∠ ABC is the intersection of the

side of AΒ
⃗ containing C and the side of ΒC
⃗ containing A.
A

Interior of ∠ ABC

37
 -------------------------- - - - ------------- C
B
And the exterior of ∠ ABC is the complement of the interior to and the angle ∠ ABC itself.

---------------------------------------------
B C

Exterior of < ABC

Next, let us see some more properties of points due to separation.

D
Draw three non - collinear points A , B and C and take points D , F and G on the plane of
Δ ABC in such a way that
 C is between A and D
 F is between B and C and
 F is between A and G.
Can you see that G is in the interior of ∠ BCD?
This is given as Theorem 1.4.7.
Theorem 1.5.7.
Let A , B and C be three non - collinear points in a plane π .
If D, F and G are points in π such that i/ A - C - D ii. B - F - C and
iii. A - F - G , then G is in the interior of ∠ BCD.

38
 A

B F C

G D

To prove this, we have to show that G is in the side of ΒC

⃗ containing D and in the side of CD
⃗
containing B.

Look ! since A, B , C are non - collinear ,A is not on ΒC

⃗ .

Since B - F - C , F is on ΒC
⃗ .

Since A - F - G, ΑG intersects ΒC
⃗ at F and so A and G are on opposite sides of ΒC
⃗ .

Since A - C - D , AD intersects ΒC
⃗ at C and so A and D are on opposite sides of ΒC
⃗ .

Now, can you conclude that G and D are on the same side of ΒC
⃗ ? (Why ?)

So, G and D are on the same side of ΒC .

⃗ i.e. G is on the side of ΒC
⃗ containing D.

Next, to show that G is on the side of CD

⃗ containing B, again since A , B and C are not collinear

B is not on AC
⃗ .

But since A - C - D, AC
⃗ = CD
⃗ so, B is not on CD
⃗ .

Then by Theorem 1.4.6, the ray CΒ

⃗ (except the end point C lies on the same half plane and since

B - F - C , F is on the ray CΒ
⃗ and so B and F are on the same side of CD
⃗ and so F is not on

CD
⃗ .

Since F is not on CD
⃗ and A is on CD
⃗ , by Theorem 1.4.6. again the ray AF
⃗ (except the end

point A) lies on the same half plane of CD

⃗ .

39
Since A - F - G , G is on AF
⃗ and so F and G are on the same side of CD
⃗ . But we have already

shown that B and F are on the same side of CD

⃗ .

Hence B and G are on the same side of CD

⃗ . i.e. G is in the side of CD
⃗ containing B.

But we have already shown that G is on the side of ΒC

⃗ containing D. By combining these two we
see that G is in the interior of < BCD.
Now , it is simple to define the interior and exterior of a triangle.
Definition
The interior of Δ ABC is the intersection of the interiors of ∠ ABC , ∠ BCA and
∠ CAB.
A

Interior
of Δ ABC

B --------------------------------------C

And the exterior of Δ ABC is the complement of the interior and the triangle itself.

B C

Exterior of Δ ABC

Theorem 1.5.8.

If P lies in the interior of ∠ ABC, then the ray ΒΡ

⃗ (except the end
point B) lies in the interior of ∠ ABC. 40
Proof:

B
C

Let P be in the interior of < ABC i.e. P is in the side of AΒ

⃗ containing C and in the side of ΒC
⃗
containing A.
Since Ρ∉ ⃗
ΑΒ and B is on ΑΒ
⃗ ΑΒ (except B) lies on the same
⃗
by Theorem 1.4.6, the ray
side of ΑΒ containing P. but P and C are on the same side of ΑΒ .
⃗ ⃗
Hence every point of ΒΡ (except B) is in the side of ΑΒ containing C.
⃗ ⃗

Similarly, every point of ΒΡ

⃗ (except B) lies on the side of ΒC
⃗ containing A.

Hence every point of ΒΡ

⃗ (except B) is in the interior of ∠ ABC.

Next, draw an angle < ABC and take a point P in the interior of ∠ ABC. Draw ΒC
⃗ and draw

ΒΡ
⃗ what did you observe? Do you see that ΒΡ
⃗ intersects AC ?
This is given as the next theorem. called the crossbar Theorem)
Theorem 1.5.9
(The crossbar Theorem)

If P is in the interior of ∠ ABC, then AΡ

⃗ intersects AC .

41
 B P

Proof of Cross bar Theorem

Let P be an interior point of ∠ ABC

¿Ρ

B C

A'

Assume that ΒΡ
⃗ does not intersect AC
Let A' be such that A - B - A'
Then A , A' and C are not collinear (Why?)

Now, consider Δ AA ' C

Now, can the line ΒΡ

⃗ intersect AC ? assume it intersects AC at R.

Then R can not be on ΒΡ

⃗ . So R is such that R - B - P.

Since R intersects AC , R and C are on the same side of AΒ .

But since R - B - P , R and P are on opposite sides of AΒ

 P and C are on opposite sides of AΒ

⃗ .
 P is not in the interior of < ABC.

So the line ΒΡ
⃗ can not intersect AC
Then, ΒΡ
⃗ intersects AA ' but does not intersect AC . Hence by Pasch's Axiom ΒΡ
⃗

intersects A'C of Δ AA ' C say at Q

42
Then ΒΡ
⃗ = ΒQ
⃗

B
C

A' Q
P

Then since A' - Q - C , Q & C are on the same side of AΒ

⃗ . But since P is in the interior of <

ABC , P and C are on the same side of AΒ

⃗

 P and Q are on the same side of AΒ

⃗ . ⇒Ρ is on ΒQ
⃗ .

i.e. ΒΡ
⃗ intersects A'C at Q.

Then since A - B - A' , A and A' are on opposite side of ΒC

⃗ .

But since A' - Q - C , Q & A' are on the same side of ΒC

⃗ .

 A and Q are on opposite sides of ΒC

⃗ .
 Q is not an interior point of ∠ ABC which implies that P is not an interior point of ∠
ABC.
Which contradicts that P is in the interior of ∠ ABC

Hence ΒΡ
⃗ does not intersect AC is not possible.

i.e. ΒΡ
⃗ intersects AC .
1.6. Angular Measure
As defined in the preceding sections, an angle is a union of two rays with a common end point.
Next, for each angle in space we associate a real number called the measure of the angle. We need
this measure to satisfy some conditions.
Dear learner, do you know a protractor (instrument used to measure angles)? Do you know
?
how to measure a given angle and how to construct an angle that has a given measure using a
Protractor?

43
 It is advisable that you find a protractor and practice to measure given angles and also to
construct angles with a given degree measure.

90

4
3
2
1
180 0
0
Protractor

To measure a given angle you put the origin of the protractor on the vertex of the angle that you need
to measure and align the zero mark of the protractor with one side of the angle so as the protractor
covers the other side of the angle.
Then you read the protractor at the point where the other side of the angle passes through some times
you may need to extend the other side of your angle so as it is not smaller than the radius of the
protractor.

0 C

To construct an angle of a given measure, you construct one side of the angle and select your vertex.

44
Then you put the origin of the protractor at the vertex and align the zero mark of the protractor with
the side you have already drawn. Then at the point the protractor reads the measure you want you
put a dot and then you join that point with the vertex. Then the angle constructed has the given
measure.
In the next section, we will see angular measure in an axiomatic approach.

Axioms of Angular measure

A function m from the set of all angles to the set of real numbers is said to be an angular measure
iff it satisfies the following Axioms.

AM - 1 : The range of m is [ 0 , 180 ] .

AM - 2 : (Angle construction Axiom)
Let L be a line in a plane π dividing π in to two half planes H1 and H2. Let O and A be

points on L . Then for every real number r ∈ [ 0, 180 ] , on a given side of L there is a

unique ray OΒ
⃗ such that m( ∠ AOB) = r.
AM - 3: (The Angle Addition Postulate)
If D is in the interior of ∠ ABC then m ( ∠ ABD) + m ( ∠ DBC) = m( ∠ ABC).

B D

AM - 4: The supplement postulate

If two angles ∠ BAC and ∠ CAD in a plane are such that B - A - D, then
m ( ∠ BAC) + m ( ∠ CAD) = 180.

45
 C

B A D

Definition

If A - B - C then the union ΒA

⃗ U ΒC
⃗ is said to be a straight angle.

A B C

In this case we say ΒA

⃗ and ΒC
⃗ are opposite rays.
Hence by AM - 4 , the measure of a straight angle is 180.
Definition
A measure of the above type is called a degree measure

Notation
If m ( ∠ ABC) = r, we write m ( ∠ ABC) = ro where the superscript "0" stands for the term
“degree”.
Definitions
1. Two angles in a plane are said to be adjacent if they have the same vertex, a common side
and no interior points in common.

C
A
E

46
 In the above figure
∠ EBA and ∠ ABC are adjacent angles
∠ EBA and ∠ ABD are adjacent angles
Can you name other two pairs of adjacent angles?
?

But ∠ EBA and ∠ CBD are not adjacent angles because they don't have a common side ∠

EBC and ∠ ABD are also not adjacent angles because they have common interior points.

Actually any point in the interior of ∠ ABC is a common interior point of ∠ EBC and
∠ ABD.

Also ∠ ABC and ∠ BCD are not adjacent angles because they don't have the same verte

? Can you identify other pairs of angles that are not adjacent?

Two adjacent angles are said to form a linear pair if their non common sides are opposite rays.
i.e. ∠ ABC and ∠ CBD are in a plane are linear pair if ∠ ABD is a straight angle. (or
A - B - D).

F
C

A D
B

E G

∠ ABC and ∠ CBD are linear pair

∠ FBC and ∠ CBG are linear pair

? Can you name other two linear pairs ?

47
 But ∠ EBF and ∠ FBD are not linear pair.

∠ EBF and ∠ CBG are also not linear pair.

? Can you list other angles that are not linear pair?
3. If ∠ ABC and ∠ DEF are any two angles such that m( ∠ ABC) + m ( ∠ DEF) = 180o,

then we say ∠ ABC and ∠ DEF are supplementary angles.

Note that every linear pair are supplementary but every supplementary angles are not a linear
pair.

A D
C
F
120o
60o

B
E

∠ ABC and ∠ DEF are supplementary angles but they are not a linear pair.
? Can you see why?
Definition
Two angles are said to be congruent if and only if they have the same measure.
i.e. ∠ ABC and ∠ DEF are congruent iff m ( ∠ ABC) = m ( ∠ DEF).
Notation
If ∠ ABC and ∠ DEF are congruent we write ∠ ABC ¿ ∠ DEF.

Note:
1. Since m( ∠ ABC) = m ( ∠ ABC), ∠ ABC ¿ ∠ ABC.
i.e. every angle is congruent to it self.

48
 2. If m ( ∠ ABC) = m ( ∠ DEF) , then m ( ∠ DEF) = m ( ∠ ABC) and so if ∠ ABC

¿ ∠ DEF , then ∠ DEF ¿ ∠ ABC.

3. If m ( ∠ ABC) = m ( ∠ DEF) and m( ∠ DEF) = m ( ∠ GHI) , then m( ∠ ABC)
=m( ∠ GHI) and so if ∠ ABC ¿ ∠ DEF and ∠ DEF ¿ ∠ GHI then ∠

ABC ¿ ∠ GHI.i.e. congruence of angles is transitive.

Definition
An angle is said to be a right angle if it is congruent to its supplement.
? Can you justify that an angle is a right angle if and only if its measure is 90o?
Notation
In diagrams we represent a right angle as follows

B C

∠ ABC is a right angle.

Next, using the Angle Construction Axioms, we prove some consequences.

Theorem 1.5.1 (Angle construction Theorem)

Let ∠ ABC be an angle and Β 'C'

⃗ be a ray lying in a plane π where Β 'C'
⃗
divides π in to two half planes H1 and H2.

Then on a given side of Β 'C'

⃗ there is a unique ray Β ' A'
⃗ such that ∠ A' B' C'
is congruent to ∠ ABC

49
 You can see that this is an immediate consequence of the Angle construction Axiom and the
D
definition of congruent angles. The proof can be constructed as follows:
Proof:
Let ∠ BC be an angle.

Let m ( ∠ ABC) =r. Then by Angle Construction Axiom in a given side of Β 'C'
⃗ , there is a

unique ray Β ' A'

⃗ such that m ( ∠ A'B'C') = r.
Then m ( ∠ ABC) = m ( ∠ A'B'C') = r and so ∠ ABC ¿ ∠ A' B' C'.

D Next, also, we will see one consequence of the Angle Addition Axiom. This is given as
Angle Addition Theorem.

Theorem 1.5.2 (Angle Addition Theorem)

Given angle ∠ ABC and ∠ A'B'C' if D and D' are respectively in their

interiors such that ∠ ABD ¿ ∠ A' B' D' and ∠ DBC ¿ ∠ D'B'C'

then

A D A' D'

C
B' C'
B

Proof:
By Angle Addition Axiom,
m ( ∠ ABC) = m( ∠ ABD) + m( ∠ DBC) and

50
m ( ∠ A'B'C') = m ( ∠ A'B'D') + m ( ∠ D'B'C')
But since ∠ ABD ¿ ∠ A'B'D' and ∠ DBC ¿ ∠ D'B'C' , we have

m ( ∠ ABD) = m ( ∠ A'B'D') and m ( ∠ DBC) = m ( ∠ D'B'C')

Hence m ( ∠ ABC) = m ( ∠ ABD) + m ( ∠ DBC)
= m ( ∠ A'B'D') +m ( ∠ D'B'C')
= m ( ∠ A'B'C')
Hence ∠ ABC ¿ ∠ A'B'C'.
From the angle addition theorem we can derive the angle subtraction theorem.

Theorem 1.5.3 (Angle subtraction Theorem)

Given ∠ ABC and ∠ A'B'C', if D and D' are respectively in their interiors
such that
∠ ABC ∠ A'B'C' and ∠
ABD ∠ A'B'D' , then ∠ DBC
 The proof is similar to the angle addition Theorem.

Proof:
By angle addition Theorem again
m ( ∠ ABC) = m ( ∠ ABD) + m ( ∠ DBC)
⇒ m ( ∠ DBC) = m ( ∠ ABC) - m ( ∠ ABD)
and m ( ∠ A'B'C') = m ( ∠ A'B'D') +m ( ∠ D'B'C')
⇒ m ( ∠ D'B'C') = m ( ∠ A'B'C') - m ( ∠ A'B'D')
But since ∠ ABC ¿ ∠ A'B'C' and ∠ ABD ¿ ∠ A'B'D' we have

m( ∠ ABC) = m ( ∠ A'B'C') and m ( ∠ ABD) = m( ∠ A'B'D')

Hence m ( ∠ DBC) = m ( ∠ ABC) - m ( ∠ ABD)
= m ( ∠ A'B'C') - m ( ∠ A'B'D')
= m ( ∠ D'B'C')
Hence ∠ DBC ¿ ∠ D'B'C'

? Do you know what is means by two lines are perpendicular to each other?
Here you have the definition of perpendicular lines.
Definition

51
Two intersecting lines are said to be perpendicular to each other if one of the angles formed at the
intersection by taking one ray from each line form a right angle.

L1

L2

Notation
If two lines ℓ 1 and ℓ 2 are perpendicular to each other, we write ℓ 1
¿ ℓ 2 or ℓ 2
¿

ℓ 1.
Classifications of Angles.
Based on the measurements of angles, angles can be classified as Acute Angles, Right Angles,
obtuse Angles or straight Angles.
Definitions
An angle ∠ ABC is said to be
1/ Acute angle if 00 < m ( ∠ ABC) < 900
2/ Right Angle if m ( ∠ ABC) = 900
3/ Obtuse Angle if 90o < m ( ∠ ABC) < 180o
4/ Straight Angle if m ( ∠ ABC) = 180o
Definition
Two angles are said to be vertically opposite angles if the sides of one are opposite rays of the other.

If AΒ
⃗ and CD
⃗ are lines intersecting at O, then ∠ AOD and ∠ BOC are vertically
opposite pairs. Also ∠ AOC and ∠ BOD are vertically opposite pairs.
Next, we will see one property of vertically opposite Angles.

Theorem 1.5.5. (vertically opposite Angle Theorem)

If two angles are vertically opposite angles, then they are congruent to each other.

52
Proof:

A O
B

Let ∠ AOC and ∠ BOD vertically opposite Angles.

Then by the above definition A - O - B and C- O - D.
Then ∠ AOC is a supplement of ∠ AOD and ∠ BOD is also a supplement of ∠ AOD.
But by the supplement angles Axiom (AM - 4)
m ( ∠ AOC) + m ( ∠ AOD) = 180o and
m ( ∠ BOD) + m ( ∠ AOD) = 180o
 m ( ∠ AOC) + m ( ∠ AOD) = m ( ∠ BOD) + m ( ∠ AOD)
 m ( ∠ AOC) = m ( ∠ BOD)
Hence ∠ AOC ¿ ∠ BOD
Similarly, you can show that ∠ AOD ¿ ∠ BOC.

Exercise

Let ΑΒ , ⃗
⃗ CD and ⃗
ΕF be three lines intersecting at a common point O as indicated in the figure
below.
F
D

A O B

53
 C
E

Then List six pairs of vertically opposite angles.

As a consequence of the vertically opposite angles Theorem, you can see that if two lines are
perpendicular to each other then all the four angles formed at the intersection by taking one ray from
each line are all right angles.

Theorem 1.5.6.

If L1 and L2 be two perpendicular lines, then each of the four angles formed at

the intersection by taking one ray from each line are right angles.

Proof: L1

C O D L2

Let L1 and L2 be two perpendicular lines intersecting at O and A and B are points on L1 such

that A - O - B and C & D be points on L2 such that C - O - D .

Then since L1 ¿ L2 , at least one of the angles ∠ AOD or ∠ DOB or ∠ BOC or ∠

COA is a right angle. With out loss of generality, say ∠ AOD is a right angle.
Then since ∠ AOD and ∠ BOC are vertically opposite angles, by Theorem 1.5.6,
∠ AOD ¿ ∠ BOC.

 Hence ∠ BOC is also a right angle. But, by definition of right angle an angle is a right
angle if it is congruent to its supplement and so ∠ COA is a supplement of ∠ AOD

and so ∠ COA ¿ ∠ AOD.

 Hence ∠ COA is a right angle. Also ∠ DOB is vertically opposite angle to

54
 ∠ COA and so ∠ DOB ¿ ∠ COA.

 Hence ∠ DOB is also a right angle. Hence all the four angles are right angles.

Congruence Between Triangles

Dear Learner, now you are familiar with the concepts related to congruence of line segments and
congruence of angles.

> Remember that line segments ΑΒ and CD are congruent if and only if they have the same

length i.e. ΑΒ≡CD iff AΒ=CD

You can also think of this as if you copy ΑΒ and try to fit it on top of CD in such a way that

one end point of ΑΒ fits to one end point of CD , then the other end points also fit together.
Similarly, angles ∠ ABC and ∠ DEF are congruent if and only if
m( ∠ ABC) = m( ∠ DEF).
You can also think of this as if you copy ∠ ABC and try to fit it on top of ∠ DEF in such a

way that the vertex B concides with the vertex E and the ray ΒΑ
⃗ concides with one of the sides

ΕD
⃗ or ΕF
⃗ , then necessarily the ray ΒC
⃗ concides with the remaining side of ∠ DEF.

Here , the order in which you fit the sides of the angles is immaterial. That is to say if ∠ ABC

¿ ∠ DEF, you can also say ∠ ABC ¿ ∠ FED or ∠ CBA ¿ ∠ DEF or

∠ CBA ¿ ∠ FED.
In a similar way, you can think of congruence of triangles as if you copy one triangle and fit it on top
of the other in such a way that if you fit the vertices of the two triangles in some order, then the sides
and angles of one will fit exactly to the corresponding parts of the other.
Here the order in which you fit the vertices is important so as the triangles are congruent. So when
we are talking about congruence of triangles, in addition to telling the triangles are congruent, we

also indicate which vertices to be fitted together. That is to say if Δ ΑΒC is congruent to

Δ DEF , we mean A fits to D, B fits to E and C fits to F and as a result the corresponding parts
also fit.

55
So, if Δ ΑΒC is congruent to Δ DEF you should not conclude that Δ ΑΒC is congruent to
Δ FΕD .
Now, let us make the above discussion formal.
Definition

Any one - to - one function from the vertices { Α , Β ,C } of Δ ΑΒC on to the vertices

{ D ,Ε , F } of Δ DΕF is called a correspondence between Δ ΑΒC and Δ DEF .

eg. If the correspondence is Α ↔D, Β↔Ε , C↔ F, we write Δ ΑΒC ↔ Δ DEF or

ΔΒCΑ ↔ ΔΕ FD or ΔC ΑΒ↔ Δ FD Ε and so on.

This correspondence gives rise to other correspondences. Namely,

ΕF and ∠ AΒC ↔∠ DEF , ∠BCA ↔∠ EFD,∠CAB↔∠ FDE

ΑΒ↔ DΕ , ΑC ↔ DF , ΒC ↔ ⃗
Next, we will give the formal definition of congruent triangles.
Definition

Triangles Δ ΑΒC and Δ DEF are said to be congruent if there is a correspondence between
them in such a way that the corresponding sides are congruent and the corresponding angles are
congruent.

If the correspondence is A ↔ D , Β ↔ Ε and C ↔ F and the two triangles are congruent, we

write Δ ΑΒC ¿ Δ DEF or ΔΒCA ¿ΔΕ FD or ΔCA Β≡Δ FD Ε so on.

From the above definition and notation, you can see that Δ ΑΒC ¿ Δ DEF if and only if

AΒ≡DΕ , AC≡DF , ΒC≡ΕF ,

∠ ABC ¿ ∠ DEF , ∠ BCA ¿ ∠ EFD and ∠ CAB ¿ ∠ FDE.

In exercises and other figures different letters may be used to designate points and so you
need to find the correct correspondence between the vertices.

56
  In figures, congruent segments are indicated by putting similar marks on the segments and
congruent angles are also indicated by putting similar marks / symbols.
For example, in the following figure.

C E F

A B
D

The above figure tells us that AΒ≡DF , ΒC ≡ΕF , AC≡DΕ ,

∠CBA≡∠ DEF , ∠ ABC≡∠ DEF and ∠ ACB≡∠ DEF .
From the definition of congruent triangles, to show that two triangles (not necessarily distinct) are
congruent to each other, we need to show that all the corresponding sides are congruent and all the
corresponding angles are congruent.
But this is not always the case. Since there are some relations between the sides and angles of a
triangle, there are some sufficient conditions that involve only some of the sides and / or some of the
angles.
Before we go to that section, let us define some terminologies that we use in the next section.
Definition

In Δ AΒC , the angle formed by two sides is called the included angle of the two sides.

For example, ∠ B is the included angle of sides ΒΑ and ΒC .

Definition

In Δ AΒC , the common side of two angles is called the included side of the two angles.

For example, ΒC is the included side of ∠ B and ∠ C.

Definition

In Δ AΒC , an angle is opposite to a side if the side is not a side of the angle. In this case we also
say the side is opposite to the angle.

57
For example, ∠ A is opposite to ΒC and ΒC is opposite to ∠ A.

A
C
∠ A and ΒC are opposite

? Verify each of the following.

1/ For any triangle Δ AΒC ,

Δ AΒC ¿ Δ AΒC
2/ For any triangles Δ AΒC and Δ DΕF if Δ AΒC ¿ Δ DΕF , then
Δ DΕF ¿ Δ AΒC
3/ For any triangles Δ AΒC , Δ DΕF & ΔG ΗΙ if Δ AΒC ¿ Δ DΕF ¿

ΔG ΗΙ then Δ AΒC ¿ ΔG ΗΙ .
Next, let's see the axioms and theorem that help us to identify congruent triangles.
As usual, we will start with an axiom and use this axiom to prove the theorems that follow.
In this group, we have only one Axiom of congruence, namely, the side - Angle - side (SAS) Axiom.
It is stated as follows.

Side - Angle - Side (SAS) congruence Axiom

If two sides and the included angle between them of one triangle are congruent to
the Axiom
SAS corresponding
can alsosides and the
be stated included angle
symbolically as: of another triangle, then the triangles
are congruent to each other.

In Δ AΒC and Δ DΕF , if AΒ≡DΕ , ΒC ≡ΕF and ∠B≡∠ Ε, then Δ ΑΒC≡Δ DΕF .

C E
D

58
 B
F
A

? State the SAS Axioms symbolically by using the other pairs of sides and included angles of

Δ AΒC and Δ DΕF .

This is because if we copy one triangle and fit it on top of the other so that the corresponding parts
concide, then the remaining parts concide necessarily. Consequently, all the corresponding six
parts(three sides and three angles) of the two triangles become congruent and so the two triangles
become congruent.
This is not a proof for the SAS Axiom, but an illustration.
Since it is an axiom, we accept it with out a proof.
We can use this axiom to identify congruent triangles.
Example:

Let ΑΒ and CD bisect (divide into two equal segments) each other at a point O as shown in
the figure below.

A O B

Then prove that Δ AOC≡ Δ ΒΟ D .

Proof:
Step Argument Reason
1 ΑΟ≡ΒΟ Given
2 CΟ≡DΟ Given

59
 3 ∠ AΟC ∠ ΒΟ D are vertically opposite Why?
angles
4 ∠ AΟC≡∠ΒΟ D by vertically opposite
angles Theorem.
5 Δ AΟC≡ΔΒΟ D by steps 1, 2, 4 and SAS
Axiom

We can also use congruence of triangles to identify congruent segments and / or congruent
angles.

For example ; in the above example after we show that Δ ΑΟC≡ΔΒΟ D , we conclude that

AC≡ΒD ,
∠ AC Ο≡∠ ΒDΟ and ∠ CAΟ≡∠ D ΒΟ .
So, if you are asked to prove that two given line segments or two given angles are congruent, you
have to identify the triangle(s) containing the given parts and together with the given assumptions
you have first to prove the triangles you identified are congruent and conclude that the given parts
are congruent.

You can also use the SAS Axiom to prove the other congruence theorems.
The next is the Angle - side - Angle (ASA) congruence theorem and we will prove it using all
our knowledge so far discussed in this module including the SAS Axiom.

Theorem 1.6.1. (Angle - Side - Angle (ASA) Congruence Theorem.)

If two angles and the included side between them of one triangle are respectively congruent to
the corresponding angles and the included side of another triangle, then the two triangles are
congruent.

The ASA Theorem can be stated symbolically as: In Δ ΑΒC and Δ DΕF ,

Ι̇ f ∠ A≡∠ D , ∠ Β≡∠ Ε and AΒ≡DΕ, then Δ ΑΒC ¿ Δ DΕF .

C F

60
 A B E
D

? State the ASA Theorem using other angles and included sides
Proof of ASA Theorem

Suppose in Δ ΑΒC and Δ DΕF , we've ∠ A ¿ ∠ D, ∠ B ¿ ∠ E and

ΑΒ≡DΕ .

If in addition AC≡DF , then Δ ΑΒC ¿ Δ DΕF by the SAS Axiom.

Now assume AC is not congruent to DF .

Then by segment construction Axiom there is a unique point P on DF

⃗ such that AC≡DP
B E

C D P F
A

Then P can not be F (since AC≠DF )

Then either D - P - F or D - F - P.

Now construct ΕΡ . Since P is not on ΕF , ⃗

⃗ ΕΡ is different from ΕF
⃗ .

Now in Δ ΑΒC and Δ D ΕΡ ,

ΑΒ≡DΕ , ∠ A≡∠ D and AC≡DΡ and hence Δ ΑΒC ¿ Δ D ΕΡ

This implies that ∠ ABC ¿ ∠ DEP. But already ∠ ABC ¿ ∠ DEF .

This implies that on the same side of ΕD

⃗ , we have found two different rays DΡ and ⃗
⃗ DF such
that < ABC is congruent to both ∠ DEP and ∠ DEF. But this is not possible by Angle

61
construction Theorem. Hence AC not congruent to DF is not possible.Hence AC ¿ DF

and so by SAS Axiom Δ ΑΒC ¿ Δ DΕF .

Before we go to the next results of this section we need to define one classification of triangles.
Definitions
A triangle is said to be:
a/ Equilateral triangle if all of the three sides are congruent to each other.
b/ Isosceles triangle if at least two of its sides are congruent to each other.
c/ Scalen triangle if no two sides are congruent.
d/ Equiangular triangle if all of its angles are congruent to each other .
Definition
In an isosceles triangle ,the angles opposite to the congruent sides are called base angles and the
remaining angle is called the vertex angle.The side opposite the vertex angle is called the base.
Next, let us see one property of isosceles triangles.This is given as Isosceles Triangle Theorem

Theorem 1.6.2.
(Isosceles Triangle Theorem)
In an isosceles triangle, the angles opposite to the congruent sides are congruent to each
other.

The Isosceles Triangle Theorem can be stated symbolically as

In Δ ΑΒC , if ΑΒ≡ AC , then ∠ B ¿ ∠ C.

( AΒ≡AC )⇒(∠ Β≡∠C ) .

? State the Isosceles Triangle Theorem using the other sides of Δ ΑΒC ?

Proof of Isosceles Triangle Theorem.

Look the correspondence ABC ↔ ACB of the same triangle.

In Δ ΑΒC and Δ AC Β
No Argument Reason
1 ΑΒ≡ ΑC given (assumed)

62
 2 ¿ BAC ≡¿ CAB It is the same angle given
3 AC≡ AΒ ?
4 Δ AΒC≡ Δ AC Β by steps 1,2,3 and SAS
5 ∠ AΒC ≡∠ AC Β by step 4 and definition of congruent
triangles.

Hence in an isosceles triangle the angles opposite to the congruent sides are congruent.

You may remember that in the definition of an isosceles triangle we said that an isosceles
triangle has at least two congruent sides.
This means it is possible that an isosceles triangle can have all the three sides congruent, that we
already said an equilateral triangle. Hence we can consider an equilateral triangle to be isosceles
triangle. Consequently you can prove that all the three angles of an equilateral triangle are
congruent to each other. In other words an equilateral triangle is equiangular triangle.
The proof is left to you.
(Hint: use the Isosceles triangle Theorem twice).
As another consequence of the Isosceles Triangle Theorem, we have the following.

Theorem 1.6.3.
If two angles of a triangle are not congruent, then their opposite sides are not
congruent.

Proof: Here we use proof by contra positive. suppose the two angles of a triangle are congruent.
Then by Isosceles triangle theorem a sides opposite to these angles are also congruent.
By taking the contra positive of this statement, if two angles of a triangle are not congruent,
then the sides opposite these angles can not be congruent.

You may remember that the converse of an implication is obtained by interchanging the
hypothesis and conclusion of the statement. i.e. The converse of a statement of the form
"If p , then q" is "If q, then p"

? Can you write the converse of the Isosceles Triangle Theorem ?

Theorem 1.6.4 (Converse of the Isosceles Triangle Theorem).

If two angles of a triangle are congruent , then the sides opposite these angles are congruent
63
D For the proof, use the same procedure we used to prove the isosceles triangle theorem but in
this case use the ASA theorem.
As a consequence of this, you can also show that an equiangular triangle is equilateral.
The next congruence theorem is the side - side - side (SSS) congruence Theorem.

? Can you state this theorem by yourself?

Theorem 1.6.5.
(The side - side - side (SSS) Congruence Theorem).
If the three sides of a triangle are congruent to the corresponding sides of another triangle,
then the triangles are congruent.

Symbolically, the SSS Theorem can be stated as:

In Δ AΒC and Δ DΕF , if

ΑΒ≡DΕ
ΑC≡DF and
ΒC≡ΕF , then Δ AΒC ¿ Δ DΕF

C E D

A B F

Proof of SSS Theorem

Suppose in Δ AΒC and Δ DΕF , ΑΒ≡DΕ , ΑC≡DF and ΒC≡ΕF .

E

64
 B

D F

A C
G

Then by angle construction theorem, on the half plane of DF

⃗ not containing E there is a unique

ray DΡ
⃗ such that ∠ BAC ¿ ∠ DEP . Also by segment construction Theorem on DΡ
⃗ there

is a unique point G such that DG≡ΑΒ . The construct GF .

Then in Δ AΒC and Δ DGF ,

ΑΒ≡DG, AC≡DF and ∠ ΒΑC≡∠GDF .

So by SAS Δ AΒC ¿ Δ DGF .

This implies that ΒC≡GF .

Now, E and G are on opposite sides of DF

⃗ and so ΕG intersects DF
⃗ at a point E'. Then we
have different possibilities.
D - E' - F or D - F - E' or E' - D - F or E' = D or E' = F.

Case 1 D - P - F.
E

D
E' F

65
In this case, E' is an interior point of both ∠ DEF and ∠ DGF.

Then Δ DEG and Δ FEG are both isosceles triangles and so by isosceles triangle theorem

∠ DΕG≡∠ DG Ε and ∠ FΕG≡∠ FG Ε and so by Angle Addition Theorem ∠ DΕF≡∠ DGF .

But already we have shown that Δ ΑΒC≡Δ DGF and so by transitivity of congruence of

triangles Δ ΑΒC≡Δ DEF .

Case 2: E' - D - F.

E' D

In this case, D is an interior point of ∠ E' E F and ∠ E' G F. and Δ DΕG and Δ FΕG are
both isosceles triangles and so by isosceles triangle theorem, ∠ DEG ¿ ∠ DGE and
∠ FEG ¿ ∠ FGE .

Then by Angle subtraction,

∠ FED ¿ ∠ FGD or ∠ DEF ¿ ∠ DGF.

Case 3: D - F - E'

F E'
D

66
The proof is similar to case 2 (E' - D - F).
In this case also we get ∠ DEF ¿ ∠ DGF

Case 4: E' = D

E' =D
F

In this case Δ FΕG is isosceles and by isosceles triangle theorem again

∠ FED ¿ ∠ FGD i.e. ∠ DEF ¿ ∠ DGF.

Case 5: E' = F
The proof is similar to case 4 (E' = D)
In this case also ∠ DEF ¿ ∠ DGF.
Now, in all the possible cases , we have shown that ∠ DEF ¿ ∠ DGF.

Now, in Δ DΕF and Δ DGF ,

DΕ≡DG,∠ DΕF≡∠ DGF ,ΕF≡GF and so by SAS Axiom, we've Δ DEF≡Δ DGF .

But we have already shown that Δ AΒC≡ Δ DGF . Hence by transitivity we've
Δ AΒC≡ Δ DΕF .
The other congruence theorem in this group is the two angles and non - included side (AAS)
Theorem but its proof needs some concepts that we had never covered and so we will wait until
the necessary back ground is covered.
Dear learner, you must be careful in that you should not think that if any three parts of a triangle
are congruent to the corresponding parts of another, then the triangles are congruent.

67
 No , for example, you may immediately disprove that if the three angles of a triangle are
congruent to the corresponding angles of another, then the triangles are not necessarily congruent.

You may have a big triangle and a small triangle with all the three angles are congruent
consequently, we don't have an Angle - Angle - Angle congruence theorem.
Another common mistake is to think if two sides and a non- included angle of one triangle
are congruent to the corresponding sides and a non - included angle of another triangle, then
the two triangles are congruent. To disprove this look at the figure below.

B D C

Suppose ΔΑΒC is isosceles with ΑΒ≡ ΑC and D is a point on ΒC such that BD < CD

Now, in Δ ΑΒ D and Δ ΑCD , ΑΒ≡ ΑC , ∠Β≡∠ C and ΑD is a common side for the
two triangles.
i.e. in the two triangles we have two sides and a non - included angle of one congruent to the

corresponding parts of another but the two triangles are never congruent as ΒD < CD .
Hence, we don't have a two sides and a non - included angle theorem.

Exercise 1.6.
1/ Prove that , in an isosceles triangle, the line segment from the vertex angle to the mid point of the
base is perpendicular to the base and bisects the vertex angle.
2/ Prove that, in an isosceles triangle, the bisector of the vertex angle is perpendicular bisector of
the base.

68
3/ Prove that, in an isosceles triangle, the perpendicular bisector of the base passes through the
vertex and bisects it.

4/ Let Δ AΒC be isosceles with AΒ≡AC and D and E are points on AΒ and AC
respectively such that AD≡AΕ . Then prove that

a/ ΒΕ≡CD
b/ Δ ΒΕC≡Δ CD Β
c/ < BCD ¿ < CBE

1.7. Geometric Inequalities

The inequalities between length of segments, and angular inequalities are the basic determinant
factors for the inequalities formed within geometric figures.

Definition: i) Given line segments AΒ and CD , we say AB < CD if and only if there exist point
P such that C - P - D and AB = CP. (Fig1.7.1(a)) )
ii) Given two angles ∠ ABC and ∠ DEF , we say ∠ ABC < ∠ DEF if
and only if there exists a point Q in the interior of ∠ DEF such that ∠

ABC ¿ ∠ DEQ. (Fig. 1.7.1(b) )

A

69
 B C D
B
C
Q
A P
D E F

Figure 1.7.1(a) Figure1.7.1 (b)

iii) Given Δ AΒC , if point D is such that A - B - D, then ∠ CBD is said to be an exterior

angle of Δ AΒC .
C

A
B D

Figure1.7.2

Moreover, ∠ CAB and ∠ ACB are said to be the remote interior angles to the exterior angle
∠ CBD.

Theorem 1.7.1: Any exterior angle of a triangle is greater than each of its remote
interior angles.
Proof: Given any triangle ABC, and the exterior angle ∠ CBD , we need to show that
∠ CBD > ∠ ACB.
Let E be the mid point of BC and construct A-E-F such that AE = EF. (Fig1.7.3 ).
C
F
E

D
A B

70
 Figure 1.7.3

Then since ∠ AEC and ∠ BEF are vertically opposite angles, and thus congruent,

We have Δ AΕC≡ Δ F ΕΒ by SAS Axiom

From this, ∠ ACB = ∠ CBF
On the other hand, point F is interior of ∠ CBD so that ∠ CBF < ∠ CBD
Thus, we have ∠ ACB = ∠ CBF < ∠ CBD. ■

Theorem1.7.2 : ( The AAS Congruence Theorem)

If two angles and the non – included side of one triangle are congruent to the
corresponding two angles and the non-included side of another, then the two
triangles are congruent.
That is, If the correspondence is such that

∠ A≡∠ D , ∠C≡∠ F and AB≡DE then Δ ABC≡ Δ DEF

Proof: Let be two triangles such that

∠BAC≡∠ EDF , ∠BCA≡∠ EFD and AB≡DE

A D
P

B C E F
Figure 1.7.4

By segment construction, there is a point P on the ray DF

⃗ such that AC≡DP
If P = F , then AC = DP = DF and the theorem reduces to SAS Axiom.
Assume P≠F , say for instance , D-P-F
Then , Δ ABC≡ Δ DEP by SAS Axiom. This imply ∠BCA≡∠ EPD

71
But this is a contradiction to Theorem 1.7.1, since ∠ EPD is the exterior angle of Δ EPF

while ∠ EFP is one of its remote interior angles.

The proof for the case when P is such that D-F-P involves exactly the same reasoning , except

that ∠ EPD is the remote interior of the exterior angle ∠ EFP .

Now let us consider a triangle and the relations between its angles and sides. Inequality between
two of its angles yields an inequality between their opposite sides.
Indeed, taking the contra positive of isosceles triangle theorem and its converse, it follows that;
Two sides of a triangle are not congruent if and only if the corresponding opposite
angles are not congruent.

Theorem 1.7.3: In any triangle, a side is less than an other side if and only if the angle opposite to
the smaller side is smaller than the angle opposite to the larger side.

That is, in Δ AΒC , AC < ΒC iff ∠ ABC < ∠Β AC

Proof: Take any triangle ABC and suppose AC ∠ BC.
Then we need to show that ∠ CBA < ∠ BAC.

A
C

B
Figure 1.7.5

Let D be on BC (such that B - D - C ) and CD = CA.

Then it follows that ∠ CAD = ∠ CDA by isosceles triangle theorem.

Also , ∠ CDA > ∠ CBA since it is exterior angle of Δ AΒD and Theorem 1.7.1
∠ CDA = ∠ CAD < ∠ CAB because D is interior of ∠ CAB

72
Thus, we conclude that ∠ CBA < ∠ CDA < ∠ CAB.
Conversely, assume ∠ CBA < ∠ BAC and show that AC ∠ BC.
( This is left as an exercise).
Next is to identify the shortest distance from a point to a line (when the point is not on the given
line).
Theorem 1.7.4: The shortest segment joining a point to a line is the perpendicular segment.
Proof: Let ℓ be a line and P a point not on ℓ . (Fig.1.7.6 )
Draw the perpendicular PQ to ℓ , which is unique.(Why?)
P

ℓ
S Q R

Figure1.7.6

If R is any other point on ℓ , we show that PQ < PR

Let S be such that S - Q - R.
Then since ∠ PRQ is remote interior of the exterior angle ∠ SQP, we have
∠ PRQ < ∠ SQP = ∠ PQR (right angles)
Thus, ∠ PRQ < ∠ PQR and it follows from Theorem 1.7.2 that PQ < PR.
Therefore, since R is arbitrary, PQ is the shortest distance to ℓ .
Theorem 1.7.5: (The Triangle Inequality)
For any points A,B and C, we have
AB + BC ¿ AC
Proof: case I: Assume A , B and C are collinear, say for instance A - B - C.
Then AB + BC = AC, and this satisfy the inequality AB + BC ¿ AC

Case II: Suppose A, B, C are not collinear and they form Δ AΒC . Let D be such that C-B-
D and AB = BD. (Fig1.7.7 )
A

73
 C D
B

Figure 1.7.7

Then ∠ BAD = ∠ BDA. But ∠ BAD < ∠ CAD. (Why?)

This implies ∠ CAD > ∠ CDA so that CD >AC
 CB + AB = CB + BD = CD > AC
Thus, AB + BC > AC. Therefore, AB + BC ¿ AC
Remark:
1. The above theorem says that in any triangle, the sum of any two sides is larger than the third
side.
2. For any n points (n ¿ 3) , A1 , A2, A3, - - - An , we have
A1 An ¿ A1A2 + A2A3 + - - - + An-1 An.

Exercises

1. Suppose Δ AΒC and Δ DΕF are two triangles such that.

AΒ=DΕ , ΒC=ΕF and ∠ Β > ∠ Ε . Then how do you compare the sides AC and DF ?
Are the two triangles congruent ? (Justify your answer).

2. Let Δ ΡQR be an isosceles triangle with vertex at P.

If T is any point such that Q - T - R, drawing PT divides Δ ΡQR in to two triangles, Δ ΡQΤ

and Δ ΡRΤ .
Are these two triangles congruent? Explain the reason and indicate a possibility (if any).

1.8. Sufficient Conditions for Parallelism

In this module you will discover two different views regarding the consequences and properties
of parallel lines; which are to be presented in Chapters 2 and 3.

74
This is due to the differences of assumptions for the number of parallel lines to a given line that can
be drawn through a given point not on the given line.
In this section we consider the common results that hold for both because it is only the sufficient
conditions for parallelism that involved here.
Definitions: 1) Two lines are said to be parallel if they lie in the same plane and they do not
intersect each other.

2 If ℓ and m are lines in a plane and a third line, n intersects ℓ and m at distinct points,
then n is called a transversal line.

n
V
R P S ℓ

T Q W m

U Figure1.8.1

If n is a transversal to ℓ and m (as shown in fig1.8.1 ) , the angles formed by ℓ , m and n

are grouped as follows.
i) ∠ QPS and ∠ TQP , or , ∠ WQP and ∠ RPQ are called alternate
interior angles.
ii) ∠ RPV and ∠ UQW or , ∠ TQU and ∠ SPV are alternate
exterior angles.
iii) ∠ RPV and ∠ TQP; ∠ SPV and ∠ WQP ;
∠ RPQ and ∠ TQU; or, ∠ QPS and ∠ UQW are called
corresponding angles or interior exterior angles.
Now let us identify some conditions that are sufficient to say two lines are parallel, where
parallelism is in the sense as defined above.

75
Theorem 1.8.1: If two lines in a plane are cut by a transversal and alternative interior angles are
congruent, then the two lines are parallel.

That is, ∠ TQP ¿ ∠ SPQ ⇒ ℓ is parallel to m.

P S
ℓ

¿R

T Q m

Figure 1.8.2

Proof: Assume the lines intersect, say at R (to the right direction)

Then Δ ΡQR will be formed and for this triangle ∠SΡQ≡∠ΤQΡ by hypothesis. But this is
impossible since ∠ SPQ is a remote interior angle of the exterior angle ∠ TQP.
Therefore, the assumption is false and the two lines do not intersect or, they are parallel.

Corollary: If two lines in a plane are perpendicular to the same line, then they are parallel.

Theorem 1.8.2: (Existence of Parallels)

Given a line ℓ and a point P not on ℓ , there is at least one line m through P
parallel to ℓ .
Proof: It has been proved that there is a line through P which is perpendicular to ℓ , and is
unique, say line n.
n
P
m

76
 ℓ

Figure1.8.3

Also, there exist a line perpendicular to n at P (or, this is possible by angle construction postulate)

, say m . Accordingly, ℓ and m are both perpendicular to n, and thus ℓ and m are parallel, by
the above corollary.
Remark: The above theorem is proved for the existence; at least one parallel line through P exists.
But its uniqueness or the existence of other parallels is still not proved; rather, it forced
mathematicians to classify a geometry in to two; Euclidean and Hyperbolic.
Saccheri Quadrilaterals
From the elementary concepts of geometry, you may recall that a quadrilateral is a four sided
polygon. In particular, a rectangle is a quadrilateral in which all the four angles are right angles. If
two adjacent angles, angle P and angle Q of the quadrilateral PQTR are right angles and PR = QT,
then in Euclidean Geometry, it can be shown that PQTR is a rectangle. That is, angle T and angle R
are also right angles.
However, all these are possible only in the Euclidean geometry. The above results are consequences
of Euclid’s parallel postulate; through a point not on a given line, there is exactly one line parallel to

the given line. That is, if P is not on ℓ , only one line, say m can be drawn through P parallel to
ℓ (Fig. 1.9.1).

P R
m

ℓ
Q T

77
 Figure 1.9.1
And this parallel is the perpendicular to PQ
Moreover, since the distance between two parallel lines is constant at every point, if we take points

R and T on m and ℓ respectively such that PR = QT , then RT is perpendicular to both lines m

and ℓ ; and thus quadrilateral PQTR is a rectangle.

On the other hand, the geometry of our interest is not only the particular one, the Euclidean
Geometry. Absolute geometry also includes other geometries such as Hyperbolic Geometry, in
which the Euclid’s parallel postulate is denied. Under such conditions, if PQ is the common

perpendicular to ℓ and m , and PR = QT (as given in the above), then RT will not be a common
perpendicular. That is, angles PRT and QTR are not right angles so that the quadrilateral PQTR is
not a rectangle. This is due to the Hyperbolic Postulate, a contradicting statement to Euclid’s parallel
postulate. (Refer to chapter three).
Therefore, such quadrilateral will have its own definition in absolute geometry.
Definition: A quadrilateral PQTR with two adjacent right angles at P and Q and with sides PR =
QT is called Saccheri quadrilateral.
P R

Q T

Figure1.9.2 Saccheri Quadrilateral.

In the Saccheri quadrilateral PQTR (Fig. 1.9.2) , the side PQ is called the base and its opposite, RT
is called the summit of the quadrilateral.
- A Saccheri quadrilateral has important properties, some of which are given below and in the
exercises of this section.
Theorem 1.9.1 : In any Saccheri quadrilateral
i) the diagonals are congruent.
ii) the summit angles are congruent.
Proof: Let PQTR be a Saccheri quadrilateral with right angles at P and Q, and draw the diagonals
PT and QR. (Fig. 1.9.3 )

78
 P R

Q T

Fig 1.9.3
Then, since PR = QT, ∠ QPR ¿ ∠ PQT and PQ is a common side, for triangles RPQ and

TQP, we have Δ RΡQ≡Δ ΤQΡ by SAS Axiom.

Thus, RQ = TP follows from this congruence.

Further, Δ QΤR≡ Δ ΡRΤ by SSS

Therefore, ∠ QTR ¿ ∠ PRT.

Theorem 1.9.2 : In any Saccheri quadrilateral, the base and the summit are parallel.
Proof: Let M and N be the mid points of the base and the summit of the Saccheri quadrilateral
PQTR, respectively. (Fig 1.9.4)
Join M to the three points R, N and T.
P R

M N

Q T

Figure 1.9.4
Then, Δ ΜΡR≡Δ ΜQΤ by SAS; and thus MR = MT

This yields Δ ΜRΝ ≡Δ ΜΤΝ by SSS

From the two congruencies, we get
∠ PMR ¿ ∠ QMT and ∠ RMN ¿ ∠ TMN

 ∠ PMN ¿ ∠ QMN ( because sums of equals are equal)

Moreover, these angles are supplementary

79
Therefore, ∠ PMN is right angle or, ΜΝ is perpendicular to ΡQ
Again, since Δ ΜRΝ ≡Δ ΜΤΝ , then it follows
∠ MNR ¿ ∠ MNT and also they are supplementary. MN
 ∠ MNR is right or, MN is perpendicular to RT .

Therefore, since ΡQ and RΤ have a common perpendicular , ΜΝ , we conclude that ΡQ

and RΤ are parallel.

Theorem 1.9.3 : Given two Saccheri quadrilaterals PQTR and P'Q'T'R' with bases
PQ and P' Q', if PQ = P'Q' and PR = P'R', then prove that
i) RT = R'T'
ii) ∠ PRT ¿ ∠ P'R'T' and so ∠ QTR ¿ ∠ Q'T'R'
Proof: Draw the diagonals PT and P' T' and it is possible to show that
Δ PQT≡ Δ P ' Q' T ' so that PT = P' T'

Using this again we get Δ PRT ≡Δ P ' R ' T ' and thus ∠ PRT ¿ ∠ P'R'T'

Theorem1.9.4 :
In any Saccheri quadrilateral, the base is either equal or shorter than the summit.
Proof:
Let A1A2B2B1 be a Saccheri quadrilateral with base A1A2 and summit B1B2 (Fig. 1.9.5)

B1 B2 B3 B4 … B+1

A1 A2 A3 A4 … An+1

Figure 1.9.5

Let A3 , A4, … ,An+1 be points on ray A1A2 such that

80
 A1A2 = A2A3 = A3A4 = … AnAn+1 and let B3 , B4, … ,Bn+1 be points in the plane of the
quadrilateral such that BiAi is perpendicular to ray A1A2 and AiBi = A1B1 for all i = 3,4,.. n+1
Then by Theorem 1.9.3, B1B2 = B2B3 = B3B4 = … BnBn+1
Now by the last remark in section 1.7 ( Polygonal inequality) we have
B 1 B n+1 ≤B1 B2 +B2 B3 +. . .+B n B n+1 =nB 1 B2 and,
A 1 An+1 ≤ A1 B1 +B1 B2 +. ..+Bn B n+1 +B n+1 A n+1
= A1 B1 +nB1 B2 + A 1 B1
¿ 2 A 1 B1 +nB1 B2

But
A 1 An+1 = A1 A2 +A 2 A3 +.. .+ An A n+1 = nA 1 A2

Thus, nA 1 A2 ≤2 A 1 B 1 +nB 1 B2

⇒ n( A1 A 2−B1 B2 )≤2 A1 B1
2 A 1 B1
n≤ , ∀n∈N
If we assume that A1 A2 >B 1 B 2 , then it gives A1 A¿ B2 .
2 A1 B1
But A 1 A ¿ B2 is a fixed positive number while n is arbitrary natural number.
Thus this is a contradiction, since the set of natural numbers is not bounded above.

Then we conclude that A1 A2 can not be greater than B1 B 2

and therefore
A1 A2 ≤B1 B2 ■
Theorem1.9.5 : In any Saccheri quadrilateral PSRQ with base PS, we have

i) ∠ PQS ≤ ∠QSR
ii) ∠SRP ≤ ∠QPR

Proof: Let PSRQ be a Saccheri quadrilateral with base PS. (Fig. 1.9.6)
Q R

81
 P S
Figure 1.9.6

In Δ PQS and Δ RSQ we have PQ = SR and QS = PR, and

Angles ∠ PQS and ∠ QSR are the included angles between these two pair of congruent

sides. Also, ∠ PQS is the opposite angle to the side PS (the base) and ∠ QSR is opposite angle
to the side QR ( the summit).

Therefore, by Theorem 1.9.4 and Theorem 1.7.2 , we conclude: ∠ PQS ≤ ∠QSR

Exercise
2. Explain the difference between a rectangle and a Saccheri quadrilateral in absolute
geometry.
3. Repeat question 1 in Euclidean geometry.
4. Given two Saccheri quadrilaterals PQTR and P'Q'T'R' with bases PQ and P' Q', if PQ = P'Q'
and PR = P'R', then prove that
i) RT = R'T'
ii) ∠ PRT ¿ ∠ P'R'T' and so ∠ QTR ¿ ∠ Q'T'R'
5. For any Saccheri quadrilateral PQTR with base PQ and summit RT , prove that PQ ¿ RT
1.10 The Angle - Sum Inequality for Triangles
The sum of the measures of the three angles of a triangle is one of the special properties of
triangles which are taken as common properties for all triangles. As you might have been introduced
in elementary geometry courses, this sum is always equal to two right angles.
However, this is valid only in the Euclidean geometry where the five postulates of Euclid are
taken to hold. Indeed, the magnitude of this angle sum particularly depends on the Fifth postulate
and thus the change of the fifth postulate results in an inequality between angle sum of a triangle
and two right angles. Further, the differences between the Euclidean and non-Euclidean Geometries
are the consequences of the basic changes made to the fifth postulate.
Therefore, in this chapter the angle sum of triangles will be presented as a general case by
showing its direct dependence on the parallel postulate (fifth postulate) so that one can simply
reduce to the particular case (the geometry under consideration).

82
To make precise, we state the angle sum as:
The sum of the three angles of any triangle is less than or equal to two right angles.
Then it will become obvious to draw a conclusion that the equality in the statement holds in
Euclidean Geometry and the inequality hold in Hyperbolic Geometry, which are to be dealt with in
the next two consecutive chapters.
Theorem 1.10.1:
In a right angled triangle, the sum of its acute angles is either equal
to or smaller than a right angle.
o
That is, in Δ PQR with ∠ P right angle, m(∠Q)+m(∠R ) ≤90

Proof: Given right angled triangle Δ PQR with ∠ P right angle , let S be a point such
that PR is perpendicular to RS, and PQ = RS. ( Figure 1.10.1)

Q S

P R
Figure 1.10.1

Then PRSQ is a Saccheri quadrilateral of base PR and summit QS

Then by Theorem 1.9.5 , we have ∠ PQR ≤ ∠QRS

o
⇒ m( ∠ PQR+m(∠ PRQ ) ≤ m( ∠QRS)+m(∠PRQ )=90
o
Therefore, m( ∠ PQR+m(∠ PRQ ) ≤ 90 ■
Corollary: 1. Every right angled triangle has only one right angle and the other two
are always acute.
2 In a right angled triangle the hypotenuse is the longest side.
Theorem1.10.2: (Angle sum Inequality)
The sum of the three angles of any triangle is less than or equal to two right angles.

83
 That is, in any Δ PQR , m(∠P )+m(∠Q)+m(∠ R) ≤180o

Proof: Let Δ PQR be any triangle.

Q S R
Figure 1.10.2

Then it can be divided in to two right angled triangles, Δ PQS and Δ PRS .
But by Theorem 1.10.1 , we have

m(∠PQS )+m(∠QPS )≤90o and m(∠ RPS)+m(∠PRS ) ≤90o

Then adding this two inequalities

m(∠P )+m(∠Q)+m(∠ R) ≤180o ■

Proposition: Through a given point P not on a given line ℓ , there can always be drawn a line

making with ℓ an angle less than any given angle α , how ever small.
Proof: Let ℓ be any line and P any point not on ℓ . From P draw PA1, perpendicular to ℓ

(Fig 1.10.3) , and take A2 on ℓ such that A1A2 = PA1.

P B

ℓ θ
A1 A2 A3 An

Figure 1.10.3.

Then clearly angles A1PA2 and A1A2P are equal, let both be θ .

84
 π π
⇒ θ≤
The from Theorem 1.10.1, we have 2θ≤ 2 4
π
θ≤
That is , 22

π
2
Without loss of generality, assume θ = 2
Next take A3 to the right of A2 so that A2A3 = PA2
θ π
=
Then, ∠ A3 PA2 = ∠ A2A3P = 2 23

Repeated construction leads to a triangle PAn-1 An for which

π
∠ AnPAn-1 = ∠ An-1AnP = 2n
where n is any positive integer greater than unity.
Then given any angle measure (number) α , by the postulate of Archimedes, there exists a number

k such that kα >π .

Then, if a positive integer n is chosen sufficiently large such that
2n > k
π
<α
it follows that 2n , and the lemma is proved.
Now as n increases, the construction of line PAn continuous but in the limiting case leaves to

π
intersect the line ℓ and the angle 2n be comes zero.
This happens for the first non - intersecting line, say PB and the sum of the right angles of triangle
PA1An will be equal to the sum of the right angle at A1 and the angle A1PB.

Case I : If the first non - intersecting line PB is the parallel line to ℓ through P (in the Euclidean
sense) , then PB is perpendicular to PA1 and it is the only possible parallel through P.
Then, angle A1PB is right angle and together with the right angle at A 1 , yields the sum of the angles
of triangle PA1 An (as n tends to infinity) is equal to two right angles.
Case II : If the first non - intersecting line PB is the sensed parallel (that results from the hyperbolic
postulate) which is not unique as assumed in hyperbolic geometry, then the perpendicular

85
 to PA1 is another non- intersecting line that comes after PB, so that angle A 1PB is less
than a right angle.
That is, angle A1PB is acute. (see Hyperbolic parallels)
Thus, the sum of the angles of a triangle, being the sum of right angle at A 1 and acute angle A1PB, is
less than two right angles.
Note : The above two different cases arose due to the difference in ideas whether there is only one or

more than one lines through P that do not intersect line ℓ . Consequently, different results
are obtained regarding angle sum of a triangle.
But it is possible to bring the two cases and state the result using the general statement as follows.
The sum of the three angles of a right angled triangle is less than or equal to two right angles. This
statement holds for any triangle since a triangle can be divided into two right angled triangles.
NB: Further explanations will be given on each of the two cases in chapter two and three.

1.11 The Critical Function

As already discussed in the previous sections exactly one line can be drawn through a point P

perpendicular to line ℓ , and at least one line parallel to ℓ .

Let ℓ be a line and P a point not on ℓ , and A is the foot of the perpendicular from P to ℓ .
(Fig1.11.1 )
P

ℓ
A B

Figure 1.11.1
Take point B on ℓ , and different rays such as ΡD
⃗ can be drawn from P forming an angle ∠
APD with AP.
Suppose r = m ( ∠ APD) and; define a set S as

S := { r∈ [ 0,180 ] :r=m ( ¿ APD ) and ⃗

ΡD int er sect ⃗
AΒ }

86
Note i) ΡΒ intersects ⃗
S≠φ since ⃗ AΒ and r=m ( ∠ A ΡΒ ) < 180 is in [ 0 ,180 ] .

i) If r=90o , then ΡD
⃗ can not intersect ℓ , and thus r  90  S . Also , if

r >90 o , ΡD
⃗ will not intersect the ray AΒ and r ∉S.
⃗
Thus, S is bounded from above as r = 90 is an upper bound.
Then the set S has the least upper bound or the Supremum (Sup S).
o
Observe that Sup S ¿ 90 .

Let this supremum to be

r o , that is, r o : = Sup S.

Any point P, therefore, determines the number

ro .

Definition: i) This number

r o is said to be the critical number of ⃗
AΒ for the point P.
ii) The angle ∠ APD with m ( ∠ APD) =
ro is called the angle of

parallelism of AΒ
⃗ and P.

Remark : 1) If m (∠ AΡD )=r o , then ΡD∩⃗

⃗ AΒ=φ
2) If
m (∠ AΡD )<r o , then ⃗
ΡD ∩⃗
AΒ≠φ

For part (1) of the remark , assume ΡD

⃗ AΒ≠φ
¿⃗ so that ΡD
⃗ intersect AΒ
⃗ , say at point

R. Take point Q on AΒ
⃗ such that A - R - Q and draw ΡQ
⃗ .

It follows that ΡQ
⃗ intersect AΒ
⃗ and r 1 =m ( ∠ AΡQ ) ∈ S

This implies
r 1 >r o , r 1 ∈ S and thus, r is not Sup S.
o

A contradiction is reached. Then, the assumption is false and remark (1) holds true.
Similarly, prove remark (2)
Theorem 1.11.1 : Suppose P, A, B and P', A', B' are points as in the definition with critical numbers

r o and r o ' respectively. (Fig 1.11.2 ) . If AP = A' P', then r o=r o ' .
P P'
D
ro ' D'
ro

87
 A B Q A' B' Q'

Figure 1.11.2

Proof: Let S := { r : m ( ∠ AΡD )=r and ⃗

ΡD∩⃗
AΒ≠φ }
S ' : = { r ':m ( ∠ A ' Ρ ' D' )=r ' and ⃗
Ρ ' D' ∩ ⃗
A ' Β '≠φ

Then, Sup S =
r o and Sup S' =r o '
Claim : S = S'
Suppose r ∈ S .

Then the exist ΡD

⃗ such that ΡD
⃗ AΒ≠φ and m ( ∠ AΡD )=r , say ⃗
¿⃗ ΡD∩⃗
AΒ={ Q } . Then

take point Q' on A 'Β'

⃗ such that A' Q' = AQ.

Consequently, by SAS Axiom, Δ Ρ AQ≡Δ Ρ ' A ' Q '

Thus, ∠ APQ ¿∠ A' P' Q ' so that m (∠ APQ )=r=m ( ∠ A ' P ' Q' ) ⇒ r∈ S ' .

This proves that S ⊆S' .

By similar argument, it can be shown that S ' ⊆S.

Therefore, S = S' and so are the supremums.

i.e.
r 0 =r 0 ' .

Note: - For each point P not on ℓ , there corresponds a positive real number, its
distance h = PA from the line ℓ .
- To such point P there corresponds also a critical number
- A rule (function) can be defined that relates the distance h of point P from ℓ to the
corresponding critical number ro.
Definition: Given a righ angle ∠ CAB (right at A), with a function ϕ where
ϕ : ( 0 , h )→ ( 0 , ∞ ) given by ϕ ( h )=r 0
This function is called a critical function.
C

88
 h3

ro3
h2
ro2
h1
ro1

A B

Figure1.11.3

For different points on segment AC at different distances from A, h1, h2, h3 for instance as shown in
figure1.11.3, there corresponds critical numbers ro1 , ro2 and ro3 which are values of the critical
function ϕ .
ϕ ( h1 ) =r 01 ; ϕ ( h2 ) =r 02 , ϕ ( h3 ) =r 03
i.e.
Theorem 1.11.2: The critical function ϕ is decreasing.
That is, for any two points P and P' on AC such that A - P - P' - C (with

distances h and h') , we have ϕ (h) ¿ϕ ( h' ) .

Proof: Let P, P' be on AC

⃗ such that A - P - P' and h, h' are their respective distances from A.
(Fig.1.11.4)

Then we need to show that ϕ (h) ¿ ϕ ( h ' ) or , r o ≥r o ' .

P' D'

A
Figure1.11.4

Let D be the point such that m (∠ AΡD )=r o =ϕ ( h )

By angle construction, let D' be such that m (∠ AΡ ' D' )=m ( ∠ AΡD ) .

Then ΡD
⃗ is parallel to Ρ' D'
⃗ . (by alternate angles)

89
Now, ΡD
⃗ AΒ=φ.
¿⃗
Since
⃗ ΡD ' we have P'D' AΒ=φ
¿⃗ .

assume Ρ' D' intersects AΒ ΡD

Because if we
⃗ ⃗ , say at Q', then
⃗ , cutting side AP' of

Δ AΡ' Q' and being parallel to Ρ 'Q',

⃗ must cut side AB (by Pasch's Axiom)

Thus, m (∠ AΡ ' D' ) is an upper bound of the set S' so that m (∠ AΡ ' D' )≥r o ' .
⇒ ϕ ( h )=r o =m ( ∠ AΡD )=m (∠ AΡ' D' )≥r o '=ϕ ( h ' )

Therefore, ϕ ( h )≥ϕ ( h' ) showing that ϕ is decreasing.

☺ Note that in Euclidean Geometry, ϕ is a constant function

That is, ϕ ( h )=90o , ∀ h>0 .

But for the general case (in Absolute Geometry), the following conclusions can be made
(i). If ϕ (h) < 90, then ϕ (h/2) < 90 ⇒ ϕ (h/4)< 90, etc), and this yields;
(ii). If ϕ (ho) < 90 for some ho, then ϕ (h) < 90 for all h > 0.
Finally, we can reasonably express parallel lines as:
If Parallels are unique for one given line and one point not on the line, then
parallels are unique for all lines and external points.

1.12 Open triangles and Critically Parallel rays

Consider two parallel lines ℓ and m , where parallelism is in the sense as defined in section 1.8.

If P and A are points on m and ℓ respectively, an important figure will be formed, which
consists the segment AP and the two parallel rays (Fig. 1.12.1 ).

Definition 1) Suppose AΒ and ⃗

⃗ ΡD are parallel rays of the two parallel lines ℓ and m . Then,

the union AΡ∪⃗

AΒ∪⃗
ΡD is called an open triangle, and denoted as Δ DΡAΒ .
D
P

90
 A B

Figure 1.12.1

2) An open triangle, Δ DΡAΒ is said to be isosceles if and only if ∠ PAB ¿

∠ APD.

3) In an open triangle DOAB, if every interior ray ΡD

⃗ of ∠ APD intersects

AΒ
⃗ , then we say ΡD
⃗ is critically parallel to AΒ
⃗ , and write ΡD
⃗ / AΒ
⃗ .
4) Two rays R1 and R2 are said to be equivalent if and only if one of them contains
the other, and written as R1 ~ R2.

A A' B AA'
⃗ ~ AΒ
⃗ ~ A 'Β
⃗

Figure 1.12.2
5) Two open triangles are said to be equivalent if and only if the rays forming their
sides are equivalent.
P D

Δ DΡAΒ ~ Δ DΡCΒ
A C B
Figure 1.12.3
The above definitions are followed by several properties of open triangles and that of critically
parallel lines.
Theorem 1.12.1: Suppose that R1 , R2 , R'1, R2' are rays in a plane such that R 1 is critically parallel
to R2.
i) If R'1 ~ R1 , then R'1 is critically parallel to R2
ii) If R'2 ~ R2 , then R1 is critically parallel to R'2
iii) If R'1 ~ R1 and R2' ~ R2, then R1' is critically parallel to R'2.

Proof: Suppose ΡD
⃗ / AΒ
⃗ and AΒ
⃗ ~ A 'Β
⃗ (Fig.1.12.4 ).

Then to show that ΡD

⃗ / A 'Β
⃗ ,

91
 P D

D'

A A' B

Figure1.12.4

Let ΡD'
⃗ be any interior ray of ∠ A' PD. Then ΡD'
⃗ is an interior ray of ∠ APD.

 ΡD'
⃗ intersects AΒ
⃗ because ΡD
⃗ / AΒ
⃗

Thus, ΡD'
⃗ intersects A 'Β
⃗ , since A 'Β
⃗ ~ AΒ
⃗

Therefore, every interior ray ΡD'

⃗ of ∠ A'PD intersects A 'Β
⃗ so that A 'Β
⃗ / ΡD
⃗ .
The other assertions of the theorem can be proved in a similar way.

Corollary: Critical parallel to a given ray, through a fixed external point, is unique.

That is, given AΒ

⃗ and point P not on AΒ
⃗ , there exist a unique ray ΡD
⃗ through P such

that ΡD
⃗ / AΒ
⃗ .

Theorem 1.12.2: If ΡD
⃗ is critically parallel to AΒ
⃗ , then the open triangle, Δ DΡAΒ is
equivalent to an isosceles open triangle with P as one vertex.
P D

A Q B
Figure 1.12.5
The proof requires to find some point Q such that A-Q-B and ∠ DPQ ¿ ∠ BQP.
For the existence of such point Q, as visible from the figure, we observe

92
 ∠ PAB < ∠ APD. If we let a point X to move from A to B, then ∠ XPD is continuously

decreasing and the smaller ∠ PXB is increasing continuously so that they must be equal at some
point X.
That is, there is some point X = Q such that ∠ QPD ¿ ∠ PQB.
Finally, we complete this section by stating the following properties of critical parallels the proof
of which is left as an exercises for you.

Theorem 1.12.3 : i) Critical parallelism is symmetric. ( i.e ΡD

⃗ / AΒ
⃗ ⇒ AΒ
⃗ / ΡD
⃗ )

ii) If AΒ
⃗ / CD
⃗ and ΕF , ⃗
⃗ CD , then ⃗
AΒ / ⃗
ΕF
Exercises
1 Define what is meant by two lines are critically parallel; and explain the difference
between parallelism and critical parallelism in absolute geometry.
2. Repeat question number 1 for lines in a Euclidean plane.
3. Prove the corollary following theorem 1.12.1

CHAPTER TWO
The Euclidean Geometry
Introduction To The Chapter
D Dear learner, Welcome to this Chapter on the Euclidean Geometry.
Based on their assumptions about parallel lines, there are three different geometries. These
are:
1/ The Euclidean Geometry
2/ The Lobacheviskian Geometry / The Hyperbolic Geometry/
3/ The Riemannian Geometry /The spherical Geometry/
D The Euclidean Geometry assumes that:
Given a line L and a point P not on L, on the plane determined by L and P, there is one
and only one line that passes through P and parallel to L.

P L'

93
 L
The Hyperbolic Geometry assumes that :
Given a line L and a point P in a plane π with P not on L, there are at least two lines L' and L'' in
π through P parallel to L.

L''
L
L'

And the Riemannian Geometry assumes that:

No two lines in a plane are parallel.

All these three Geometries have their own models and their own meanings to planes, lines, distance
functions and angular measures.
Among these geometries, the Euclidean geometry satisfies all the Axioms, theorems, remarks and so
on discussed in chapter 1. The Hyperbolic Geometry satisfies all the results discussed in chapter 1
except some of the Axioms of Incidence.
And the Riemannian Geometry (spherical Geometry) fails to satisfy many of the results discussed in
chapter 1.
So, our interest is to work with the two geometries (The Euclidean and the Hyperbolic Geometries)
and see some consequences of the parallel postulate in each of the two geometries.
Due to the differences of these assumptions, there are many differences in the two geometries
.Consequently, we will work with the Euclidean Geometry in this chapter and the Hyperbolic
Geometry in the coming chapter.
You must be careful that the results obtained in this chapter apply only to the Euclidean plane .
2.1. The Euclidean parallel Postulate and some consequences
The Euclidean parallel postulate states that:

EPP: Given a line L and a point P in a plane π such that P is not on L, there is one
and only one line L' on the plane π through P and parallel to L.

L'
P
94
L
Notation
If a line L is parallel to L' we write L // L'
In this section, you need to remember what you studied in section 1.8 of chapter 1.

Before you proceed with this section you need to revise section 1.8 in particular.
You may remember from section 1.8 that if two lines are cut by a transversal and alternate interior
angles are congruent, then the two lines are parallel.

The converse of this statement is also true in the Euclidean Geometry. We will state it next and
prove. This statement is commonly called The Alternate Interior Angle Theorem

Theorem 2.1.1. (The Alternate Interior Angles Theorem)

If two parallel lines are crossed by a transversal, then the alternate interior angles
formed are congruent.

A B

F
C D

If ΑΒ
⃗ // CD
⃗ , then ∠ AΕF≡∠ Ε FD and ∠ΒΕ F≡∠ Ε FC .
The proof makes use of the result you we found in section 1.8. in the Euclidean plane.
D

95
Proof:
Let L1 // L2 and L is a transversal cutting L1 and L2 at E and F respectively as shown in the figure
below.

L
A
L1
A' E B

F L2
C D

Now assume that ∠ AEF is not congruent to ∠ EFD.

Then by Angle construction Theorem in the half plane of ΕF

⃗ not containing D, there is a unique

ray ΕA'
⃗ such that ∠ A ' ΕF≡∠ Ε FD .

Then ∠ A' EF and ∠ EFD are alternate interior angles and are congruent and so by Theorem

1.8. 1 the line A'Ε

⃗ is parallel to L2. But A'Ε
⃗ is different from AΕ
⃗ .

Hence we have found two lines AΕ

⃗ and A'Ε
⃗ both through E and parallel to L 2. Which is a
contradiction to EPP.
Hence the assumption that ∠ AEF not congruent to ∠ EFD is not possible.
i.e, ∠ AEF ¿ ∠ EFD.

Also ∠ BEF and ∠ EFC are alternate interior angles and are supplements of ∠ AEF and
∠ EFD respectively

Since ∠ AEF ¿ ∠ EFD , we've ∠ BEF ¿ ∠ EFC.

Hence alternate interior pairs are congruent.

Corollary:

96
When two parallel lines are crossed by a transversal, then
1/ The corresponding pairs are congruent.
2/ The alternate exterior pairs are congruent

L
A E G B L1

L2
C F D
H

Then by the Alternate Interior Angles Theorem , we have

∠ AΕF≡∠ Ε FD and ∠ ΒΕF≡∠ Ε FC .
Now, ∠ G E B and ∠ A E F are vertically opposite angles and so by vertically opposite Angles
Theorem
∠ GEB ¿ ∠ AEF . But ∠ AEF ¿ ∠ EFD and so ∠ GEB ¿ ∠ EFD

Similarly, show that

∠ΒΕ F≡∠ DF Η
∠G ΕΑ≡∠ Ε FC
∠ΑΕ F≡ ∠ CF Η }
Hence, the corresponding pairs are congruent.

Next 1/ ∠ G ΕΒ≡∠ ΑΕ F (by vertically opposite Angles Theorem).

2/ ∠ Ε FD≡∠CF Η
3/ ∠ΑΕ F≡∠ Ε FD (by Alternate Interior Angles Theorem)
4/ ∠G ΕΒ≡∠CF Η (by steps 1,2,3)
Similarly, you can show that ∠G ΕΑ≡∠ DFH .
Hence, Alternate Exterior pairs are congruent.

97
D Another important consequence of the Euclidean parallel postulate is the
Angle - sum Theorem.
In Absolute Geometry, you have seen that the sum of the three angles of a triangle is less
than or equal to 180o. But due to the parallel postulate in Euclidean plane it is exactly equal
to 180o. This is stated next. It may be less than 180o in the hyperbolic geometry.

Theorem 2.2.2. (Angle - Sum theorem).

Δ AΒC , m (∠ A )+m ( ∠ Β ) +m ( ∠C )=180 .

∘
In any triangle

Proof of Angle - Sum Theorem

Let Δ AΒC be any triangle. L'

A
E
D

C
B

Since A is not on ΒC
⃗ , by EPP, there is a unique line L' through A parallel to ΒC
⃗ as in the
figure above.

Now , DΕ // ⃗
⃗ ΒC and ⃗
AΒ is a transversal to them. ∠ DAB and ∠ ABC are alternate

interior angles and so by Alternate Interior Angles Theorem ∠ DA Β≡∠ AΒC .

Similarly AC
⃗ is a transversal to DΕ
⃗ and ΒC
⃗ , ∠ Ε AC and ∠ AC Β are alternate interior

pairs and
⃗ ΒC and so ∠ ΕΑC≡∠ ΑCΒ
DΕ // ⃗ .

But m (∠ DA Β ) +m ( ∠Β AC ) +m (∠ Ε AC ) =180∘ (Straight angle)

and so
m (∠ AΒC ) +m ( ∠Β AC ) +m ( ∠ AC Β )=180∘ .

98
Due to this Angle - sum Theorem, we have many results.
Theorem 2.1.3
Each exterior angle measures the same as the sum of the two remote interior angles.

Proof:
A

C D

1. m (∠ AΒC ) +m ( ∠Β AC ) +m ( ∠ AC Β )=180∘ (by Angle - Sum Theorem)

2. m (∠ DCA ) +m ( ∠ AC Β )=180∘ ( Straight Angle).

 m (∠ AΒC ) +m ( ∠Β AC ) +m ( ∠ AC Β )=m ( DCA )+m ( ∠ AC Β )

 m (∠ DCA )=m ( ∠ AΒC ) +m ( ∠ Β AC ) .

Similarly, you can show for the other exterior angle. Hence, in any triangle, the measure of each
exterior angle is the sum of the two remote interior angles.
We have also the next result.

Theorem 2.1.4
The sum of exterior angles of a triangle is 3600.

Proof:

99
Remember that we take one exterior angle at each vertex.

D
A

C
F

m ( ¿ DA Β )=m ( ¿ AΒC )+m ( ¿ ACB )

m ( ¿ ΕΒC ) =m ( ¿ BAC ) +m ( ¿ ACB )
m ( ¿ FCA ) =m ( BAC ) +m ( ¿ AΒC )

m ( ¿ DA Β ) +m ( ¿ ΕΒC ) +m ( ¿ FCA )=m ( ¿ AΒC ) +m ( ¿ ACB ) +m ( ¿ BAC ) +

Hence m ( ¿ ACB )+m ( ¿ BAC ) +m ( ¿ AΒC )

=2 ( m ( ¿ BAC )+m ( ¿ AΒC )+m ( ¿ ACB ) )

=2 ( 180∘)
¿ 360∘
You can use the results obtained so far together with the results of this section to solve many
problems in geometry.
Example

Let Δ AΒC be isosceles with AC≡ AΒ and AC>BC and D is a point on AC such that

CD≡ΒC . If m ( ¿ Β AC )=40∘ then find m ( ¿ AΒD )

A

40o
D

100
 B C

Solution

1. m (∠ Β AC ) +m (∠ AΒC ) +m ( ∠ AC Β )=180∘ (by Angle - Sum Theorem)

2. AΒ≡AC ⇒ ∠ AΒC≡∠ AC Β (by Isosceles Triangle Theorem)

So, 50o + 2 m (∠ ABC )=180∘

 2 m ( ∠ ABC )=180∘−40∘ =140∘

 m (∠ ABC )=m ( ∠ ACB ) =70∘

Again Δ Β CD is isosceles with ΒC≡CD

 ∠CBD≡∠CDB

and m ( ¿CΒD ) +m (¿ CD Β )+m ( ΒCD )=180∘

But m ( ¿ Β CD )=70∘

 m (∠CΒD ) +m (∠CD Β )+70∘=180∘

 2 m ( ∠CΒD )=110∘ .

 m (∠CΒD )=55∘

But m ( ¿CΒA ) =70∘ and D is in the interior of ∠ AΒC

Then by Angle addition
m (∠ AΒD ) +m ( ∠CΒD )=70∘
 m (∠ AΒD )+55∘=70∘

 m (∠ AΒD )=70∘−55∘

 m (∠ AΒD )=15∘ .
The study of triangles is very basic in geometry because every other polygon can be
subdivided into triangles.

101
 In the next section, we will see some properties of polygons in general and quadrilaterals in
particular.
Definition

1/ Let
Ρ1 , Ρ2 ,⋯,Ρ n+1 be points. Then a sequence of points Ρ1 Ρ2 ∪Ρ 2 Ρ 3 ∪⋯∪ Ρ n Ρ n+

is called a broken line denoted by

Ρ1 Ρ2⋯Ρ n+1

P2 Pn
P5
Pn+1
P1

P4
P3 P6

2/ In a plane, if P1 P2 --- Pn+1 is a broken line such that P1 concides with Pn+1 , then the broken line is
called a polygon and denoted P1 P2 ---- Pn.

P1 P5
P8 P4

P2
P3

P7 P6

P1 , P2 , ----- , Pn are called its vertices

Ρ1 Ρ2 , Ρ2 Ρ3 , ⋯ , Ρn−1 Ρ n are called its sides.

If n = 3 , the polygon is called a trianle

If n = 4 , the polygon is called a quadrilateral
If n = 5 , the polygon is called a Pentagon
If n = 6 , the polygon is called a Hexagon

102
 If n = 7 , the polygon is called an Heptagon.
If n = 8 , the polygon is called an Octagon
If n = 9 , the polygon is called a Nonagon
If n = 10, the polygon is called a Decagon
for the corresponding Greek words showing the number of sides.
In general, a polygon with n sides is called an n - g on.
3/ A polygon is called a simple polygon if its vertices are all distinct, no three vertices lie on a side
and no two sides intersect.

not simple polygon

simple polygon.

4/ A Polygon is said to be convex if its interior is convex , otherwise it is concave.

Convex Polygon Concave Polygon

5/ A polygon is said to be regular if all its sides are congruent and all its angles are congruent to
each other.

103
 Regular quadrilateral

not regular

6/ In a polygon P1 P2 - - - Pn , the line segment joining two non consecutive vertices is called a
diagonal

Next, let us see some properties of simple polygons

Property 1
In a simple polygon of n sides, there are( n – 3) diagonals from one vertex

Proof

104
For a given convex polygon we do not draw a diagonal to itself and to the two consequitive vertices.
So, there are n - 3 diagonals.
Property 2
In a convex polygon of n sides, the diagonals drawn from one vertex partition the polygon into (n -
2) triangle.
Quadrilaterals
There are different classifications of quadrilaterals.
? Are you familiar with rectangles, trapezoid, parallelograms, rhombus, and square?
In this section, first we will define these quadrilaterals and then see their properties. In the proofs of
these properties we will apply many of the results so far.
Definitions
A quadrilateral is said to be:
1/Trapezium / Trapezoid if it has at least one pair of opposite sides parallel.

2/ Parallelogram if it has two pairs of opposite sides parallel.

Definitions
A parallelogram is said to be:
1/ a Rectangle if it has at least one right angle.

2/ a Rhombus if all its sides are congruent

105
Definition
A rectangle is said to be a Square if all its sides are congruent.

You must be careful of these definitions because one is defined in terms of the others.

According to the definitions, you can see that a parallelogram can also be considered as a
trapezium, a square can be considered as a rectangle or as a parallelogram or as a rhombus.
D Next, let us see more properties of these quadrilaterals

Theorem 2.1.5
Opposite sides of a parallelogram are congruent.

Proof:

A B

D C

Let ABCD be a parallelogram.

Then AΒ // ⃗
⃗ CD and AD // ΒC
Now construct the diagonal AC .

Then in Δ ADC and Δ CΒA

Argument Reason

106
 1 ∠Β AC≡∠ ACD b/c AC is a transversal and by alternate interior
angle. Theorem
2 ∠Β CA≡∠CAD b/c AC is a transversal AD and ΒC and
by Alternate interior angle theorem.
3 Ac is common side
4 Δ AΒC≡ Δ CDA by step 1,2,3 and ASA congruence theorem
5 AΒ≡CD and ΒC≡AD by step 4 and the definition of congruent triangles

In the proof of the above statement you may have seen that a diagonal of a parallelogram
divides the parallelogram into two congruent triangles.

Theorem 2.1.6
The diagonals of a parallelogram bisect each other.

Proof:
A B

D C

Let ABCD be a parallelogram and the diagonals AC and ΒD intersect at O.

Then in Δ AOB and Δ COD .

∠ A ΒΟ≡∠CΟD (Alternate interior angles) ∠ΒΑΟ≡∠ DC Ο

AΒ≡CD (by Theorem 2.7)

So Δ A ΟΒ≡Δ CΟD (by ASA)

 AΟ≡ΟC and ΒΟ≡ΟD

Hence the diagonals bisect each other

Theorem 2.1.7
If the opposite sides of a quadrilateral are both parallel and congruent, then the
quadrilateral is a parallelogram. 107
Proof:
A
B

D C
Let ABCD be a quadrilateral such that AΒ //CD and AΒ≡CD .

Then In Δ AΒC and Δ CDA ,

1. ∠Β AC≡∠ ACD (Alternate interior angle)

2. AΒ≡CD (given)

3. AC is a common side

4. Δ AΒC≡ Δ CDA (by SAS)

5. ∠Β CA≡∠ DAC (by step 4 and definition)

6. AD // ΒC (because AC is a transversal to AD and ΒC and the alternate interior

angles ∠ BCA and ∠ DAC are congruent
Hence the quadrilateral has two pairs of parallel sides and so it is a parallelogram.

Theorem 2.1.8.
The opposite angles of a parallelogram are congruent.

Proof:

A B

108
 D C

As in the proof of theorem 2.1.7, Δ AΒC≡ Δ CDA and Δ AΒD≡Δ CDB which imply
∠ ABC≡∠CDA and ∠ Β AD≡∠ DC Β .
Hence the opposite angles are congruent.
Theorem 2.1.9.
The consecutive angles of a parallelogram are supplementary

Proof:

A B

D C

1. ∠ ABD ¿ ∠ BDC (Why?)

o
2. m (∠ AΒD )+m ( ∠ Β DA )+m ( ∠ DAB )=180 (Why?)
o
3. m (∠ Β DC ) +m ( ∠Β DA ) +m ( ∠DA Β )=180 (Substitute step 1 into step 2).
o
4. m (∠ ADC ) +m (∠ DA Β )=180 (because m (∠ BDC ) +m (∠ BDA )=m ( ∠ ADC ) .
Similarly, for the other pairs.

From the properties of parallelograms discussed so far, you can easily see that a rectangle satisfies
1/ The opposite sides of a rectangle are congruent.
2/ All the angles of a rectangle are right angles.

Theorem 2.1.10. In a Plane,

Two lines that are parallel to the same line are parallel to each other.

109
Proof:
Let L, L1 and L2 be lines in a plane and L1 // L and L2 //L.

T
L2
3
L1
2
1 L

Let T be a transversal
Since L1 // L , ∠ 1 ¿ ∠ 2
Since L2 // L, ∠ 1 ¿ ∠ 3

Hence ∠ 2 and ∠ 3 are congruent and they are alternate interior angle formed by L1 , L2 and T
Then by the converse of Alternate interior angles Theorem L1 // L2

Theorem 2.1.11
In a plane, two lines that are perpendicular to the same line are parallel to each other.

Proof:
Let L1 , L2 and L be lines in a plane and L1 ¿ L and L2 ¿ L.

110
 L1 L2

L
A B D

Then the angles ∠ ABC and ∠ BDE indicated above are right angles and so are congruent.
Then by converse of Alternate Interior Angles Theorem L1 // L2

Theorem 2.1.12.
Proof:
Every Saccheri quadrilateral is a rectangle.
A B

D C

Let ABCD be a Saccheri quadrilateral .i.e AD≡ΒC and <D and <C are right angles. Then

AD and ΒC are perpendicular to CD and so by theorem 2.1.11 AD // ΒC

Then AD and ΒC are both parallel and congruent and so by Theorem 2.1.7 ABCD is a
parallelogram and has at least one right angle and so it is a rectangle.
Hence, in the Euclidean Geometry every Saccheri quadrilateral is a rectangle.

2.2. Parallel Projections

111
 Remember that given a line L and a point P on a plane, there is a unique line L' through P
perpendicular to L. Consequently given any two lines L and L' in a plane, through each point
P on L', there is a unique line through P perpendicular to L intersecting L at a point P'

L'

P'

This gives rise to the notion of perpendicular projection / vertical projection.

Definition
Let L and L' be lines in a plane.

Then the function f : L→ L' such that PP' ⊥ L' for all Ρ∈ L where f ( Ρ )=Ρ ' is called a
vertical projection.
L
P

L'
P'

Similarly, if L and L' are two lines in a plane and T is a transversal to L and L', then for any point P
on L, there is a unique line through P parallel to T intersecting L' at a point P.

112
 T

L'

P'

If we define a function f from L into L' in this way, we get a one - to one Correspondence called the
parallel projection.

With this function if P = T ¿ L, then P' = T ¿ L' and f (P) = P.

Theorem 2.2.1
Every parallel projection is a bijective

Proof: T L
Q
P

L'
P' Q'

Let f : L→ L' be parallel projection.

To show one - to - one ness

Let P and Q be two distinct points on L and f ( Ρ ) =Ρ ' , f ( Q ) =Q '

Then ΡΡ' // Τ and ⃗

⃗ QQ' // Τ
Then by Theorem 2.1.10, two lines in a plane parallel to the same line are parallel to each other.

113
Hence ΡΡ' // ⃗
⃗ QQ
 Ρ'≠Q '

Hence for Ρ≠Q , f ( Ρ )≠f (Q ) .

To Show onto ness

Let A be any point on L'. If A = L' ¿ T then B = T ¿ L is its pre image.

If 0 A≠Τ ∩L' , there is a unique line through A parallel to T intersecting L at a point B. Then

f ( Β )= A .

Hence for any point A on L' there is a point B on L such that f ( Β )= A .

Hence f is a bijection.
f is one - to - one and onto and so f is a bijection.
Theorem 2.2.2
Parallel Projection preserves
i. betweenness ii. Congruence of segments
i.e. if P, Q, R, S are points on L and P', Q', R', S' are their images respectively under
a parallel projection, then
i. if P - Q - R , then P' - Q' - R'

ii. if ΡQ≡RS , then Ρ'Q'≡R'S '

Proof: S L
P Q R
B
A

P' Q' R' S' L'

114
 i. By parallel projection , ΡΡ' , QQ' , RR' and SS' are all parallel to T and so by Theorem
2.1.12, they are parallel to each other
Let P - Q - R

Since QQ' // RR' , RR' does not intersect QQ'

⃗ . Hence R and R' are on the same side

of QQ'
⃗ .

But since P - Q - R , ΡR intersects QQ'

⃗ and so P and R are on opposite sides of

QQ'
⃗ .

 P' and R' are on opposite sides of QQ'

⃗

 Ρ' R' intersects QQ'

⃗ at Q'
 Q' is between P' and R'
i.e. P' - Q' - R'

Hence if P - Q - R, then f ( Ρ )−f ( Q )−f ( R ) \

i.e. a parallel projection preserves betweenness.

ii. Let ΡQ≡RS

Draw lines ΡA and ⃗
⃗ RΒ such that ΡA // L' and ⃗
⃗ RΒ // L' intersecting QQ' and

SS' at A and B respectively.

Then in Δ Ρ QA and Δ RS Β ,

ΡQ≡RS , ∠QΡA≡∠ SRB , ∠ΡQA≡∠RSB and so by ASA, Δ Ρ QA≡ Δ RS Β .

 ΡA≡RΒ .
But by construction,

PAQ'P' and RBS'R' are parallelograms and so ΡA≡Ρ' Q ' and RΒ≡R' S' .

but ΡA≡RΒ and so Ρ'Q'≡R'S ' .

Hence a parallel projection preserves congruence of segments

115
 2.3. The Basic Similarity Theorem
Before we see the basic similarity theorem, let us make some preparations.
We first define what is meant by two sequences of numbers are said to be proportional.
Definition

Let
a1 , a2 ,a3 ,⋯and b1 b 2 , b3 ⋯ be two sequences of numbers. Then we say these sequences are
proportional and write
a1 ,a 2 , a3 ,⋯ ~ b1 ,b 2 , b ,…
3

Where ai≠0 , bi≠0, i=1,2⋯

a1 a2 a3
= = =⋯
If b1 b2 b3

a1 a2
k= = =⋯
The constant b1 b2 is called the constant of proportionality

In this case, a1 =kb 1 , a 2=kb 2 ,⋯

Observe that if a, b, ~ c, d , then a, c ~ b,d
Proof:
a b
⇒ = ⇒ ad=bc
a,b~c,d c d
ad bc a c
⇒ = ⇒ = ⇒a,c
bd bd b d ~ b d.
 Recall that between any two real numbers x and y, there is at least one rational number

p
r=
q where p and q are integers and q≠0.
 Also remember one property of real numbers known as the comparison Theorem stated below.

Theorem 2.3.1. (The comparison Theorem)

Let x and y be real numbers such that every rational number r less than x is also less than y
and every rational number r less than y is also less than x. Then x and y are equal.

Proof:

116
Assume x≠ y . Then either x < y or y < x. WLOG assume x < y.
Then there is at least one rational number r between x & y
i.e. x < r < y.
Hence if every rational number less than x is also less than y and every rational number less than y is
also less than x, then x = y.
 Now we are ready to state the Basic Similarity Theorem. It is basic because it gives way to the
similarity Theorems

Theorem 2.3.2. (Basic Similarity Theorem).

If three or more parallel lines are cut by two , transversals, then the segments on one
transversal are proportional to the segments on the other taken in the same order.

Proof:

A A' L1

B B' L2

C C'
L3

T2 T1

117
Let L1 , L2 & L3 are three parallel lines intercepted by the transversals T 1 and T2 as indicated in the
figure above.
Let A, B, C, A' , B' , C' be as in the figure above.
Let r = n/m be any positive rational number. i.e. n and m are any positive integers.
ΒC Β'C'
x= and y=
Let AΒ A'Β'
AΒ
1/ Divide AΒ in to m congruent segments of length m such that
A= A 0 , A1 , ⋯, A m=Β .
AΒ
A 0 A 1= A 1 A 2=−−−¿ A m−1 Am=
i.e. m .

2/ Let B = B o, B1, B2 --- , Bn be point on the ray ΒC

⃗ such that

AΒ
Β o Β1 =Β 1 Β 2=⋯=Β n−1 Βn =
m
3/ Project each Ai and Bi onto T2 in the direction of L1 to get points Ai 's and Bi 's
Β Βn n Β0 Β1 n AΒ/m n
= = =
Then AΒ m A 0 A1 m AΒ/m m . and since by Theorem 2.2.2, parallel projection

Β ' Β' n n
=
preserve congruence and so A ' Β ' m .
ΒΒ nn Β' Β'n
= =
i.e. AΒ m A ' Β '

n ΒΒ n ΒC
r= <x ⇒ < x=
Now, m AΒ AΒ
⇒ ΒΒn <ΒC
⇒ Β−Β n−C
⇒ Β '−Β n '−C ' by Theorem2 .2.2
⇒ Β ' Β n '<Β ' C '
Β ' Βn ' Β ' C '
⇒ <
A' Β' A' Β'
n Β'C'
⇒ < =y
m A' Β'
n
⇒ <y
m

118
i.e. every rational number less than x is also less than y.
on the other hand,
n Β' Β n ' Β' C '
r= < y⇒ < y=
m A ' Β' A 'Β'
⇒ Β ' Βn'< Β ' C '
⇒ Β '−Β n '−C '
⇒ Β−Β n−C
⇒ Β Βn < ΒC
ΒΒ ΒC
⇒ n< =x
AΒ AΒ
⇒r<x

i.e. every rational number less than y is also Less than x.

Hence by the comparison theorem x = y
ΒC Β ' C '
=
i.e. AΒ A ' Β '
As a result of this, we can see that parallel projections preserve ratio.

As a result of the Basic Similarity theorem, in Δ AΒC , if P and Q are points on AΒ and AC

respectively such that ΡQ // ⃗

⃗ ΒC, then
AΡ AQ
=
ΡΒ QC

P Q

B C

Draw a line through A parallel to ΒC

⃗ . Then by above theorem.

119
 AΡ AQ
=
ΡΒ QC
AΡ AQ AΡ AQ
= ⇒ = ⇒ AΡ QC =AQ ΡΒ
This also yields AΒ AC ΡΒ QC
⇒ ΑΡ ( AC−AQ )= AQ ( AΒ− AΡ )
⇒ AΡ AC− AΡ/ AQ= AQ AΒ− AΡ AQ
⇒ AΡ AC= AQ AΒ

AΡ AQ
=
AΒ AC

Theorem 2.3.4.
AΡ AQ
=
In Δ AΒC , if P and Q are points on AΒ and AC respectively such that ΡΒ QC ,

then ΡQ// ⃗
⃗ ΒC A

P Q

B C

Proof:

Assume that ΡQ
⃗ not parallel to ΒC
⃗

Then there is a unique line through P parallel to ΒC

⃗ intersecting AC at a point R.
Then by the Basic Similarity Theorem

120
 AΡ AR
= ⇒ AΡ RC=AR ΡΒ
ΡΒ RC
⇒ AΡ ( AC− AR )= AR ( AΒ− AΡ )
⇒ AΡ AC − AΡ AR=AR AΒ−AΡ AR
⇒ AΡ AC = AR AΒ
AΡ AR
⇒ =
AΒ AC

AΡ AQ AΡ AQ
= ⇒ =
also ΡΒ QC AΒ AC
AΡ AR AQ
⇒ = =
AΒ AC AC

⇒ AR= AQ

⇒ R=Q⇒ ΡR=ΡQ
⇒⃗ΡQ // ⃗ ΡR // ⃗
ΒC Since ⃗ ΒC
2.4. Similarity Between Triangles
? Do you know how map makers construct a map of a large portion of the earth?
Actually, they construct a map that resembles the actual shape but diminished in size
proportionally with a proportionality constant called the scale of the map in addition
they preserve direction. This gives you a clear meaning of what is meant by two
triangles are similar.

For example if the scale of the map is 1cm. 2km , then a triangular farm of sides 3 km, 4 km , 5 km
will be represented by a triangle of sides 1.5 cm, 2 cm, 2.5 cm. Since they also preserve directions,
the angles of the farm and the angles of the triangle in the map are congruent.
This gives rise to the formal definition of similarity.
Definition

Given a correspondence Δ AΒC ↔ Δ DΕF , we say the sides of the triangles are proportional if

AΒ AC ΒC
= =
DΕ DF ΕF

121
Definition

Given a correspondence Δ AΒC ↔ Δ DΕF we say Δ AΒC is similar to Δ DΕF written

Δ AΒC ~ Δ DΕF , if
i. The corresponding sides are proportional and
ii. The corresponding angles are congruent
Symbolically D
A

B C

E F

Δ AΒC ~ Δ DΕF if ∠ A ¿ ∠ D , ∠ B ¿ ∠ E , ∠ C ¿ ∠ F and

AΒ AC ΒC
= =
DΕ DF ΕF
As in the case of congruence of triangles, to show that two triangles are similar, there are some
sufficient conditions for similarity of triangles that we call Similarity Theorems.

According to the definition, if Δ AΒC ¿ Δ DΕF , then Δ AΒC ~ Δ DΕF

Also , we have the following

1. Δ AΒC ~ Δ AΒC for any Δ AΒC .

2. If Δ AΒC ~ Δ DΕF , then Δ DΕF ~ Δ AΒC .

3. If Δ AΒC ~ Δ DΕF and Δ DΕF ~ Δ GΗΙ then Δ AΒC ~ Δ G ΗΙ .
Theorem 2.4.1. (AAA Similarity Theorem).If the three angles of a triangle are
congruent to the corresponding three angles of another , then the triangles are
similar

Symbolically,in Δ AΒC and Δ DΕF , if ∠ A ¿ ∠ D, ∠ B ¿ ∠ E and ∠ C

¿ ∠ F , then Δ AΒC ~ Δ DΕF .

122
Proof:

Let Δ AΒC and Δ DΕF be such that ∠ A ¿ ∠ D, ∠ B ¿ ∠ E and ∠ C ¿

∠ F.

A D

C E' F'
B

E P F

On DΕ and ⃗
⃗ DF there exist unique points E' and F' respectively such that

AΒ ≡DΕ' and AC≡DF '

Then in Δ AΒC and Δ DΕ ' F ',

∠ A ¿∠ D , AΒ≡DΕ ' and AC≡DF ' and so by SAS congruence axiom

Δ AΒC ¿ Δ DΕ ' F ',
⇒∠ AΒC≡∠ DΕ ' F ' and ∠ AC Β≡∠ DF ' Ε '
⇒∠ DΕ ' F '≡∠ DΕF
Ε' F' // ⃗
⇒⃗ ΕF (Why ?)
Then by The Basic similarity Theorem
DΕ' Ε Ε '
=
DF ' F ' F
⇒
DΕ ' DF '
= - (1) -------- (1)
DΕ- - DF

Now, construct F' Ρ so that F' Ρ // ⃗

Ε' Ε
FΡ FF '
=
Then FΕ FD

123
 FΕ−ΕΡ FD−DF '
⇒ =
FΕ FD

ΕΡ DF '
⇒1− =1−
FΕ FD
ΕΡ DF '
⇒ =
FΕ FD
but ΕΡ=Ε ' F '

Ε ' F ' DF '

Ε ' F ' DF ' =
⇒ = ΕF DF
FΕ FD or - - - (2)
Then by (1) and (2) above
DΕ ' DF ' Ε' F '
= =
DΕ DF ΕF
Now in Δ DΕF and Δ DΕ ' F'
∠ D is common angle,
DΕ ' DF ' Ε' F '
∠ DΕ' F '≡∠ DΕF , ∠ DF ' Ε '≡∠DFE and = =
DΕ DF ΕF
Hence Δ DΕF ~ Δ DΕ ' F' .

But since Δ DΕ ' F '≡Δ AΒC , Δ DΕ' F ~ Δ AΒC .

⇒ Δ AΒC ~2.4.2.
Theorem Δ DΕF(SAS
. Similarity Theorem)
If two sides of a triangle are proportional to two sides of another and the included angles
between them are congruent, then the two triangles are similar.

Symbolically
AΒ AC
Δ AΒC and Δ DΕF , if ∠ A≡∠ D and = , then Δ AΒC
In DΕ DF ~ Δ DΕF .

Proof:
A D

124
 E' F'

C
B

E F

AΒ AC
Δ AΒC and Δ DΕF , ∠ A≡∠ D , and =
Let in DΕ DF .

Take points E' and F' on DΕ

⃗ and DF
⃗ respectively such that AΒ≡DΕ' and AC≡DF ' .

Then Δ AΒC≡ Δ DΕ' F ' and so Δ AΒC ~ Δ DΕ ' F '

DΕ' AΒ DF ' AC
= and =
DΕ DΕ DF DF

DΕ ' DF'
⇒ =
DΕ DF

DΕ ' DF '
⇒ =
Ε' F F' F

Then by Theorem 2.3.4, we've Ε' F' // ΕF

⇒∠ DΕ ' F '≡∠ DΕF and ∠ DF ' Ε '≡∠ DF Ε .
Then by AAA Similarity Theorem
Δ DΕ ' F ' ~ Δ DΕF
⇒ Δ ABC ~ Δ DEF
Theorem 2.4.3 (SSS Similarity Theorem)
If the three sides of one triangle are proportional to the corresponding sides of
another, then the two triangles are similar.

AΒ AC ΒC
Δ AΒC and Δ DEF if = = , then Δ AΒC
Symbolically, in DΕ DF ΕF ~ Δ DΕF .
Proof:
Let in Δ AΒC and Δ DΕF ,
AΒ AC ΒC
= =
DΕ DF ΕF

125
 D
A

E' F'
B C

E F

Let E' be a point on DΕ

⃗ such that AΒ≡DΕ' . Then through E' draw a line parallel to ΕF

intersecting DF at F'
⇒< DΕ' F ' ≡¿ DΕF , <DF ' Ε '≡¿ DF Ε and < D is common angle.
⇒ Δ DΕ' F ' ~ Δ DΕF
DΕ' DF ' Ε ' F '
⇒ = =
DΕ DF ΕF
Since DΕ'= AΒ ,
AΒ DF ' Ε' F '
= =
DΕ DF ΕF
AΒ AC ΒC
= =
but DΕ DF ΕF
DF ' AC Ε ' F ' ΒC
⇒ = and =
DF DF ΕF ΕF
⇒ DF '= AC and Ε ' F '=ΒC
Then in Δ AΒC and Δ DΕ ' F '
⇒ Δ AΒC≡Δ DΕ ' F ' (by SSS congruence Theorem
⇒ Δ AΒC ~ D' F '
Also, we have shown that DΕ' F ' ~ Δ DΕF

Hence Δ AΒC ~ Δ DΕF .

You can use similarity to solve problems
Example 1

Let in the figure below DΕ // ΒC

A

126
 D E

B C

If AE = 7 cm , EC = 3 cm , DB = 2 cm DE = 8 cm. Then find AD , BC , AB

Solution

Since ΒC, ∠ AD Ε≡∠ AΒC and ∠ AΕD≡∠ AC Β, ∠ A

DΕ // ⃗
⃗ is common so Δ AD Ε ~
Δ AΒC
AD AΕ DE
⇒ = =
1/ AΒ AC BC
AΕ 7 cm
⇒ AD= AB= ( AD+2 cm )
AC ( 7+3 ) cm

7
⇒ AD= ( AD+2 )
10
10 AD=7 AD+14
3 AD=14
14
AD= cm
3
AΕ DΕ
=
Also AC ΒC
7 8
=
10 BC
10 ( 8 )
BC=
7
80
BC= cm
7
Example 2
Prove that the line segment joining the mid points of two sides of a triangle is parallel to the third
and half of it.

127
Proof:
A

D E

B C

AD AΕ AD 1
= = =
In Δ AD Ε and Δ AΒC , ∠ A is common and AΒ AC 2 AD 2
Hence by SAS Similarity Theorem
Δ ADE ~ Δ AΒC
⇒∠ AD Ε≡∠ AΒC⇒ DΕ // ΒC
AD DΕ 1
= =
and AΒ ΒC 2
DΕ 1 1
⇒ = ⇒ DΕ= ΒC
ΒC 2 2

2.5. The Pythagorean Theorem

? Are you familiar with the famous theorem in geometry known as the Pythagoras'
Theorem ?
In this section, we will see how to prove this famous theorem using our knowledge so far.
Before we go to the details of this theorem, let us see some properties of right - angled triangles.
Definition

In a right - angled triangle Δ AΒC , if the right angle is at C, then we call AC and ΒC the legs

and AΒ the hypotenuse of Δ AΒC .

? Can you justify that in a right angled triangle, the hypotenuse is the longest side?

128
? Can you also justify that the altitude drawn to the hypotenuse of a right angled triangle
intersects the hypotenuse?
Next let us see one property of right angled triangles.
Definition
A number b is said to be a mean proportional between two numbers a and c iff the ratio a:b = b:c
a b
=
i.e. b c

Theorem 2.5.1.
Suppose the altitude to the hypotenuse of a right triangle is drawn. Then
i) The triangles formed are each similar to the given right triangle.
ii) The altitude is mean proportional between the segments of hypotenuse.
iii) Each of the legs of the triangle is mean proportional between the
hypotenuse and the segment of the hypotenuse adjacent to it.

Symbolically, let Δ AΒC be right angled at C. Let D be the foot of the altitude to the hypotenuse
drawn from C.

A D

C
B

Then i. say Δ ADC ~ Δ AC Β ~ Δ CD Β

AD CD
=
ii. CD BD
AD AC ΒD ΒC
= , =
iii. AC AΒ ΒC AΒ
Proof:

In Δ AΒC and Δ ADC ,

129
1/ ∠ A is common angle
2/ ∠ ADC ¿ ∠ ACB (both are right angles)
o o
3/ m (∠ A )+m ( ∠ ADC )+m ( ∠ ACD )=180 and m ( ¿ A )+m ( ¿ AC Β )+m ( ¿ Β )=180

4/ m (∠ A )+m ( ∠ ADC )+m ( ∠ ACD ) =m ( ∠ A ) +m ( ∠ AC Β )+m ( ∠ Β )

 m (∠ ACD )=m (∠ Β )

 ∠ ACD ≡∠ Β

5/ Δ AC Β ~ Δ ADC (by steps 1,2,4 and AAA similarity Theorem)

Similarly Δ ADC ~ Δ Β CA ⇒ Δ CD Β ~ Δ AC Β
Hence Δ ADC ~ Δ AC Β ~ Δ CD Β .
AD CD
=
ii. From Δ ADC ~ Δ CD Β , we get CD ΒD
Hence the altitude is mean Proportional between AD and BD.

iii. From Δ ADC ~ Δ AC Β , we get

AD AC AΒ AC
= ⇒ =
AC AΒ AC AD
From Δ CD Β ~ Δ AC Β , we get
AΒ CΒ AΒ ΒC
= or =
CΒ DΒ ΒC ΒD
Hence each leg is mean proportional to the hypotenuse and the segment of the hypotenuse adjacent
to it.
 Now we are ready to state and prove the famous Pythagoras' Theorem.

Theorem 2.5.2 (Pythagoras' Theorem)

If Δ AΒC is right angled at C, then AC2 + BC2 = AB2

In the usual way, if the hypotenuse is c units long and the legs are a and b units
respectively, them we've c2 = a2+b2.

The proof makes use of the above result in Theorem 2.5.1

130
Proof:
A

c
b

C a B

Let Δ AΒC be right angled at C.

Then draw the altitude to AΒ intersecting AΒ at D.
A
D

C B

Then by Theorem 2.5.1 iii. we've

AΒ AC
= ⇒ AC 2 = AΒ . AD
AC AD

AΒ ΒC
= ⇒ ΒC 2 =AΒ ΒD
ΒC ΒD
Adding the two, AC2 + BC2 = AB.AD + AB . BD
= AΒ ( AD+ΒD )
= AΒ ( AΒ )
= AΒ2
2 2 2
Hence AΒ = AC + ΒC
i.e. c2 = a2 + b2

You can use this theorem to solve many problems in Euclidean Geometry.

Exercise

131
In the figure below ABCD is a square of side 10 units and P is a point on the diagonal AC such

that AΡ≡AD and ⃗

DΡ inter sects ΒC at Q. Then find DQ .
A B

PP

D C

132
 CHAPTER THREE
HYPERBOLIC GEOMETRY
By Nasir Asfaw

Introduction
So far, the basic concepts of Euclidean Plane Geometry has been introduced in detail. The
Axioms of Euclid were the central facts on which the relations, theorems, and other results obtained
in this geometry are greatly based. The development of Euclidean Geometry was gradual and it is
not due to a single person or society; but it consists the contribution of several ancient
Mathematicians such as Euclid, Pythagoras, etc.

However, different views to a space and the attempts made to prove the Euclid's fifth postulate
were resulted strange ideas and led mathematicians to think of and develop a geometry of non-
Euclidean type, which is consistent under its own assumptions. Although it happened independently
and at about the same time, the discovery of a logically consistent geometry, in which the fifth
postulate of Euclid was denied, was made by Carl Friedrich Gauss (1777 - 1855) in Germany,
Wolfgang Bolyai (1775 - 1856) in Hungary and Nikolai Ivanovich Lobachewsky (1793-1856) in
Russia

The Non-Euclidean Geometry consists not only the Hyperbolic Geometry, but also others such as
Elliptic geometry. But only Hyperbolic Geometry is to be introduced in this chapter and the next.

Objectives
Up on the completion of this chapter, you will be able to:
 Upgrade your view of Plane Geometry to that of the Non - Euclidean thoughts.
 Explain what is meant by parallel line in Hyperbolic Geometry .
 Indentify elementary properties of parallels in Hyperbolic Geometry.
 Define ideal points and ultra - ideal points; and relate to ordinary points of a line.
 Identify different properties of a triangle whose vertex is an ideal point.
 Define Saccheri and Lambert quadrilaterals and explain their special properties
 Prove basic theorems in Hyperbolic Geometry.

133
  Determine the angle sum theorem for triangles and appreciate the result as the consequence
of the change made to the fifth postulate.
 Identify and list the basic differences of Euclidean and Hyperbolic Geometries.
 Explain the concepts of similarity in hyperbolic geometry and the reason for its collapse to
congruence.

3.1 Hyperbolic Parallel Postulate and Definition of parallels

☺ Dear student:
Before you read this section, turn your attention back to chapter two for a moment and try
to restate the Postulates of Euclid. Hope fully, you will state the five postulates of Euclid based on
which the Euclidean Geometry is developed.

These postulates can be stated (using equivalent statements) as follows.

1. Through two distinct points, there exist exactly one straight line.
2. A line segment can be produced or extended in a straight line
3. Given a point and a distance, there always exists a circle.
4. All right angles are equal to one an other.
5. Through a point not on a given line, there exists one and only one line that is parallel to the
given line.

☺ Once again consider each postulate one by one and analyze its geometric interpretation. In
comparison with what you have read in Absolute Geometry, can you make any classification among
these five postulates?
The following activity helps you to imagine the existence of a Plane Geometry, which is different
from Euclidean Geometry. Please, do the activity before you pass to what follows.

134
 Activity 3.1
1. Take any line ℓ and a point P which is not on ℓ .

2. Draw lines through P, as many as you can, including the perpendicular to ℓ .

3. Classify these lines to two sets as: those that intersect and those that do not intersect ℓ .
4. Consider the lines through P that do not really intersected ℓ on your paper but that you

thought they will intersect ℓ somewhere, since the lines are to be extended. To which
set of lines, in (3), do you Categorize such lines? Why?

An attempt to prove the fifth postulate, the difficulties and unexpected results led mathematicians to
develop a new kind of geometry. By assuming the negation of the fifth postulate to reach a
contradiction, a new idea has been created rather; that is, using this assumption together with the
postulates 1 - 4 , a new geometry, which is consistent as well, can be developed.

A geometry, which is consistently developed by taking the negation of any of the Euclid's
postulates as a postulate, is called non-Euclidean geometry. One of such geometries, which is based
on postulates 1 - 4 and the negation of the fifth postulate, is the hyperbolic geometry, which is to be
presented in this chapter.
The hyperbolic axiom (hyperbolic parallel postulate) is that stated opposing the Euclid's fifth
postulate as follows.

☺ Now it is time to state an Axiom which substitute the fifth postulate. Just Ready!

Hyperbolic Axiom: Through a given point, not on a given line, at least two lines can be
drawn that do not intersect the given line.

☺ Dear student; take few minutes and try to interpret this Axiom in terms of what you have
done in Activity 3.1.
The geometry, which is developed based on the first four Euclid 's postulates and the hyperbolic
axiom, is called Hyperbolic Geometry.

135
It can be observed that, if there is more than one line through the given point not intersecting the
given line, there is an infinite number of such lines.
If P (Fig. 3.1) is the given point, ℓ the given line and AB and CD two lines through P which do

not intersect ℓ , then no line, such as EF, lying with in the vertical angles APC and DPB, will cut
ℓ .

A D
P
E F
C B

S
R I

J Q K G H ℓ

Figure 3.1

If one starts with the perpendicular PQ from P to ℓ and allows PQ to rotate about P in either

direction, it will intersect ℓ a while and then cease to intersect it. Let the first line not to intersect
ℓ is PS. Assume same from the left to be PR. Eventually, a situation is reached in which the lines

through P are divided into two, those which cut ℓ and those which do not.

Definition3.1: Through the point P not on ℓ the first line that do not intersect ℓ , as PS

136
 above, is called sensed parallel or simply parallel to ℓ in the right
direction.
- The line PR is also parallel to ℓ in the left direction.
Note that in Hyperbolic Geometry, parallelism is sensed in either of the directions, left or right
(sensed) parallels.
- Through a point not on a line, there are two parallel lines to the given line.

Figure 3.2

It can be observed that those lines that lie with in angle RPS (fig. 3.1) intersects ℓ and the
others do not intersects and are non - intersecting lines or ultra parallel lines.

☺ Well student! It might surprised you. Don’t worry! This is what others also thought as such,
including the discoverers of Hyperbolic Geometry. But at this time you have to identify the
parallels. Moreover, you have to interpret the term parallel in this sense whenever it appeared.

Theorem 3.1 : The parallels PR and PS form equal and acute angles with PQ.
That is, angle RPQ is equal to angle SPQ. Also these two angles are
acute angles.
Proof:
Let PR and PS be the parallels to ℓ (Fig.3.1). Then we need to show that angle
RPQ is equal to angle SPQ.
Assume that these two angles are not equal, let one of them is greater, say RPQ.
Draw PI within angle RPQ so that angle IPQ is equal to angle SPQ .

137
 Then PI intersects ℓ , say at J.
Then measure from point Q to H on ℓ so that JQ is equal to QH. It follows that triangles PQJ
and PQH are congruent, and thus angles JPQ and HPQ are equal. Since angle JPQ (or, IPQ) is equal
to angle SPQ, it follows that angle HPQ is equal to angle SPQ . But this is a contradiction since PH

intersects ℓ while PS does not. Therefore, the assumption is false and we conclude that angle
RPQ is equal to angle SPQ.
Moreover, each of these angles is acute, for otherwise if right angle for instance, then PR and PS lie
on the same straight line, the perpendicular through P to PQ, which leads to the conclusion that this

line is the only line through P not intersecting ℓ . This is a contradiction to the hyperbolic parallel
postulate. ■
Now we continue the study of Hyperbolic Geometry by taking all the above concepts in to
consideration

Definition 3.2 : Given line m parallel to ℓ . From point P on m draw a perpendicular to ℓ

(Fig. 3.3). Then angle QPR is called angle of parallelism for the

distance h and will be denoted by π (h) .

P
R
h m

Q T ℓ

Figure 3.3
-Recall that angle of Parallelism is an a cute angle.

3.2 Elementary properties of Parallels in Hyperbolic Geometry.

138
Certain properties of Euclidean parallels hold also for parallels in Hyperbolic Geometry. Some of
these are described in the following theorems.

Theorem 3.2: If a straight line is the parallel through a given point in a given sense to a
given line, it is, at each of its points, the parallel in the given sense to the
given line.

Proof :
If AB (Fig.3.4 ) is the parallel to ℓ through P, we wish to prove that it is parallel to ℓ
through any point R on AB.

A P R B

T
ℓ
Q S M

Figure 3.4

Draw PQ and RS perpendicular to ℓ . We have to show that every line through R passing with

in the angle SRB intersect ℓ .

Let RT be any such line and select on it any point U. Draw PU and RQ. Then PU cut ℓ , say at M

since AB is parallel to ℓ at P.
Also, PU cut RQ at N by Pasch's axiom. Again resorting to Patch’s axiom, since RU cut PM of

triangle PQM, we conclude that RU also intersects QM. Thus, RT intersects ℓ showing that AB is

parallel to ℓ through point R as well. ■

139
Theorem 3.3 : If one line is parallel to a second, then the second is parallel to the first.

This property of parallels reveals symmetry of parallelness in Hyperbolic Geometry.

That is, if m is parallel to ℓ , then ℓ is parallel to m.

Theorem 3.4 : If two lines are both parallel to a third line in the same direction, then
they are parallel to one an other.
Proof:
Case I. Assume the third line lies between the other two.
Let AB and CD , both be parallel to EF in the same direction and let AC cut EF at G.

A
B

E G I F

C D

Figure 3.5

Draw any line AH through A and passing within the angle CAB. This line will cut EF in a point I.
Draw CI. Since EF is parallel to CD, AI produced will intersect CD. Since AB does not cut CD, but
every line through A lying within angle CAB does cut, it follows that AB and CD are parallels.
Case II: Assume the two lines are on the same side of the third line.
Let AB and CD (Fig 3.6 ) both be parallel to EF in the same direction.

140
 A

G
B
C H

E F
Figure 3. 6

Assume that AB is not parallel to CD in the designated direction. Then through any point of AB
draw the parallel GH to CD in that direction. It follows from the first case that GH is parallel to EF.
But only one parallel in that direction can be drawn to EF through G. Therefore, GH must coincide
with AB, and AB is parallel to CD. ■

Ideal Points
☺ You know a common point of two lines when they intersect each other at a point which is
visible to us. Think now for if two parallel lines intersect each other at a point which is too far
from us to visualize!

We wish to introduce at this point an important concept in connection with parallel lines. Two
intersecting lines have a point in common, but two parallel lines do not, since they do not intersect.
However, two parallel lines have something in common. It is convenient to recognize this
relationship by saying two parallel lines intersect at an ideal point.
Thus all of the lines parallel in the same sense to any line, and consequently parallel to one an other,
will be thought of as being concurrent in an ideal point.

Every line contains thus, in addition to its ordinary or actual points, two ideal points through
which all lines parallel to it pass. The postulate that two points determine a line will be extended to
the case when these points are both ideal points or one ordinary and one ideal point.
Notation: Ideal points are denoted by capital Greek letters, such as Ω

141
 ℓ P

m Q

} Ω
R
N

Figure 3.7

In figure 3.7, points P , Q and R are ordinary points of the lines ℓ , m and n, respectively. The
three lines, being parallel, have common ideal point Ω .

☺ Dear student:
Have you created the imaginary point Ω in your mind?
To convince you more, consider any pair of rope or wire stretched horizontally (such as electric
wires). When you look across farther distances, the wires seem approaching each other. Do you
agree? Even continue to look these wires farthest again, as far as possibly visible to you. What do
you think of the pair of parallel wires? Write your suggestion and expectation on the space
reserved below.
_____________________________________________________________________
_____________________________________________________________________
______________________________________________________________________

3.3 Triangles and angle sum

3.3.1 Asymptotic Triangles

142
 ☺ At this time you have to test yourself whether you visualized the ideal point
Ω because in this section you will consider the triangle for which Ω is one of
the vertices.

A triangle has three points which are its vertices. In this section we include an ideal point to be a
vertex of a triangle together with ordinary points, and name it asymptotic triangle or simply a
triangle. For instance, in figure 3.8, we call triangle AB Ω , in which two of the vertices are
ordinary and one vertex is ideal point. Such triangles have many common properties with ordinary
triangles.

B Q
} Ω

Figure 3.8

Theorem 3.5: If a line passes within the triangle AB Ω through one of the vertices, it
will intersect the opposite side.
Proof: Let P be any point with in triangle AB Ω (fig. 3.8)
The line through vertex A, line AP intersects B Ω since A Ω and B Ω are parallel lines. For
the same reason, line BP must intersect A Ω .
Now draw P Ω so that line P Ω is an arbitrary line through the vertex Ω . Since this line
cut side AQ of triangle ABQ , by Pasch's Axiom, it intersects the side AB. ■
Theorem 3.6: If a straight line intersects one of the sides of triangle AB Ω , but does not
pass through a vertex, it will intersect one and only one of the other two
sides.
If a line intersects A Ω or B Ω , the above theorem is easily proved. It intersects AB, say at R,
draw R Ω and use the preceding theorem.

143
The details are left as an exercise. ■

The familiar exterior angle theorem also holds for such triangles.
Theorem 3.7 The exterior angles of triangle AB Ω at A and B, made by producing AB,
are greater than their respective opposite interior angles.
Proof: Let AB be produced through B to C (Fig 3.9 ) we need to prove that angle CB Ω is
greater than angle BA Ω .
Through B draw BD, making angle CBD equal to angle BA Ω .

L A

B N
} Ω

C D

Figure 3. 9

Clearly, BD can not intersect A Ω (i.e. D is not above B Ω ) otherwise it forms triangle ABD
with exterior angle CBD equal to opposite interior angle BAD.
Furthermore, BD can not coincide with B Ω (i.e. BD is not parallel to A Ω ). To show this,
assume BD coincides with B Ω . Draw MN from M, the mid point of AB, perpendicular to B Ω
(BD). Produce A Ω to L so that AL is equal to BN, and draw ML. Then , points N,M and L are
collinear since triangles MLA and MNB are congruent ; and angle MLA is right.
This implies LN is perpendicular to both lines A Ω and B Ω (BD); which is a contradiction
since A Ω and B Ω are parallel: Thus, BD doesn't coincide B Ω , but lie below B Ω

144
showing that angle CB Ω is greater than angle BA Ω because angle CB Ω is the sum of
angles CBD and DB Ω while angle BA Ω is equal to angle CBD. ■

Theorem 3.8: If AB and A'B' are equal, and angle BA Ω is equal to angle B'A' Ω ', then
angle AB Ω is equal to angle A'B' Ω ' and triangle AB Ω is congruent to
triangle A'B' Ω '.

Proof: Assume angles AB Ω and A'B' Ω ' is not equal under the given conditions so that one of
them is the greater, say AB Ω . Construct, angle ABC equal to angle A'B' Ω ' as shown in figure
3.10.

A
D A'

C D'
} Ω

B
} Ω '
B'
Figure 3.10

Let BC cut A Ω in point D. Measure A'D' on A' Ω ' equal to AD and draw B'D'. Then
triangles ABD and A'B'D' are congruent. Thus, angle A'B'D' is equal to angle ABD and hence equal
to angle A'B' Ω ', which is a contradiction.
Thus, the assumption is false, and angles AB Ω and A'B' Ω ' are congruent. ■
Theorem 3.9: If angles AB Ω and BA Ω are, respectively, congruent to angles A'B' Ω '
and B'A' Ω ' , then segments AB and A'B' are equal and the triangles are
congruent.

Proof: Assume AB and A'B' are not equal, one of them, say AB, is the greater (Fig. 11). Take point
C on segment AB such that AC is equal to A'B' and draw C Ω .

145
 A A'

B Ω '

} Ω B'

Figure 3.11

Then angles AC Ω and A'B' Ω are equal, by theorem3.8. But from the hypothesis of the
theorem, angles AB Ω and A'B' Ω ' are equal implying that angle AC Ω is equal to angle AB
Ω . This is a contradiction to theorem 3.7.

Therefore, segments AB and A'B' are equal. ■

Theorem 3.10 : If segments AB and A'B', angles AB Ω and A'B' Ω ', and angles A'B' Ω '
and B'A' Ω ' are equal, then all the four angles AB Ω , BA Ω , A'B' Ω ' and
B'A' Ω ' are equal to one another and the triangles are congruent.

Proof: Assume one of angles AB Ω and BA Ω , say AB Ω is not equal to the equal
angles A'B' Ω ' and B'A' Ω '.
Draw BC so that angle ABC is equal to angle A'B' Ω '.
Similarly, draw AD so that angles BAD and B'A' Ω ' are equal
A

D A'
E'
C E

B
} Ω } Ω '
B'

Figure 3.12

Then AD and BC intersect, say at E. Take point E' on A' Ω ' such that AE is equal to A'E' and
draw B'E'.

146
Then triangles ABE and A'B'E' are congruent, and thus angle ABE is equal to angle A'B'E'. But by
construction, angle ABE is equal to angle A'B' Ω '. Thus, we conclude that angle A'B'E' is equal to
angle A'B' Ω ', which is a contradiction. Therefore, the assumption is false and angle AB Ω is
equal to angles A'B' Ω ' and B'A' Ω '. Same is true for angle BA Ω , and thus, the four angles
are equal.

Exercises 3.3.1

Prove the following statements

1. In triangle AB Ω , the sum of angles AB Ω and BA Ω is always less than two right
angles.
2. If , for two triangles AB Ω and A'B' Ω ', angles AB Ω and A'B' Ω ' are equal but
segment AB is greater than segment A'B', then angle BA Ω is smaller than angle B'A' Ω '.

3. Two points P and Q on the two lines ℓ and m, respectively are said to be corresponding

points if PQ forms equal angles with ℓ and m on the same side. If P and Q are

corresponding points of the two parallel lines ℓ and m (Fig. 3.13), then

ℓ
P

} Ω
m Q

Figure 3.13

a) Compare the angle that these corresponding points formed (in the direction of
parallelism) with the angle of parallelism.

147
 b) Relate triangle ΡQΩ to ordinary isosceles triangles by illustrating some of their
common properties.

4. Discuss some properties of those corresponding points on intersecting and non - intersecting
lines by indicating results which are similar to (different from)that of parallels.
5. For two parallel lines, is it possible to have a common perpendicular line? Explain in detail and
answer the same question for two intersecting and non - intersecting lines.
6. Given line ℓ and point P not on ℓ , let be the angle of parallelism and h the corresponding

distance (i.e. h = PQ, where PQ is perpendicular to ℓ ). If P moves on segment PQ towards Q,

what will happen to α ? What if P moves in the opposite direction? In both cases explain the

α =π ( h ) , find lim π ( h ) and lim π ( h )

result for the limiting case. That is, with convention h→0 h→∞ .

3.3.2 The Saccheri Quadrilateral

☺ Dear student, you are requested to do the activity given below before you pass to
the definition. Also remember what you have seen in section 1.9 of chapter one.

Activity 3.2 A special

1. Draw a line segment AB, then draw the perpendicular lines

ℓ and m to AB at the end points A and B, respectively.

2. Take points D and C on ℓ and m respectively such that

AD = BC and join DC to obtain quadrilateral ABCD.
3. What type of quadrilateral is it? How can you relate segments
AB and CD? Angles ADC and BCD?
4. Please, do not forget that you are in the world of Hyperbolic
Geometry!!

quadrilateral which is formed by drawing equal perpendiculars at the end points of a line segment is

148
one that is used systematically by the Italian Mathematician Geralamo Saccheri (1667-1733); and it
is named Saccheri quadrilateral.

Definition 3.3 A Saccheri quadrilateral is a quadrilateral ABCD with two adjacent

right angles at A and B and with sides AD equal to BC.
D C

A B

Figure 3.14 Saccheri quadrilateral

- The side adjacent to the two right angles, side AB is called the base and its opposite, DC is
called summit of the Saccheri quadrilateral.
- The angles adjacent to the summit (i.e. angles ADC and BCD) are called summit angles.
-
Next, some important properties of this bi rectangular quadrilateral will be introduced.
Theorem 3.11 : The line joining the mid - points of the base and summit of a Saccheri
Quadrilateral is perpendicular to both of them, the summit angles are
equal and a cute.

Proof: Let ABCD be a Saccheri quadrilateral with right angles at A and B. (Fig 3.15) and let M
and H be the mid points of AB and DC, respectively.

D H C

149
 A M B

Figure 3.15

Join DM and CM. Then triangles DAM and CBM are congruent, and then triangles DMH and CMH
are congruent. Then, angles DHM and CHM are equal, and thus are both right showing that MH is
perpendicular to DC. Again, since angles AMD and CMD, angles DMH and CMH are congruent,
we have angle AMH congruent to BMH, and thus are both right angles showing that HM is
perpendicular to AB.
The two pair of congruent triangles above also yields the congruence of the summit angles ADC
and BCD.
To prove that the summit angles are acute, draw D Ω and C Ω , the parallels to AB of the
Saccheri quadrilateral ABCD. (Fig 3.16)
Since lines DC and AB are non-intersecting (i.e. AB and DC have common perpendicular HM), the
parallels D Ω and C Ω lie within the angles ADC and BCE, respectively. Angles of parallelism
AD Ω and BC Ω are equal because AD is equal to BC. But, angle EC Ω is greater than angle
CD Ω (exterior angle of triangle CD Ω ).

D C E

} Ω

A B

Figure 3.16

Consequently, angle BCE is greater than angle ADC

Thus, angle BCE is greater than angle BCD since angles ADC and BCD are equal summit angles.

150
Therefore, angle BCD is acute since it is less than its supplement, angle BCE. ■

☺ Let you return to activity 3.2 once again! After procedure 1, draw a perpendicular line n to one
of the lines ℓ or m; say to m at any point different from B so that n intersects ℓ and m at points
E and F, respectively.
Here again you have no right to say line n is also perpendicular to ℓ because there is no article
that enables you to say so.
[The article that was your guarantee to say so in Euclidean Geometry, the fifth postulate of Euclid,
doesn’t appear here in the new constitution (Hyperbolic Geometry), rather it is substituted by an
opposite article, the Hyperbolic axiom]
However, a quadrilateral ABEF of special kind, as defined below, will be formed.

Definition 3.4 : A quadrilateral with three of its angles right angles is called Lambert
quadrilateral.

- This trirectangular quadrilateral has got the name Lambert quadrilateral because Lambert
used such quadrilateral as a fundamental figure in his researches.
Theorem3.12 : In a Lambert quadrilateral, the fourth angle is acute.
Proof: Let ABCD be a Lambert quadrilateral with the angles at A, B and D right
angles (Fig. 3.17)
Extend BA to E so that EA is equal to BA; and at E draw EF perpendicular to BE and equal to BC.
Then, triangles FEA and CBA are congruent and thus triangle FAD is congruent to triangle CAD.

F D C

151
 E A B

Figure 3.17

This implies that angles FDA and CDA are equal, and points F,D and C are collinear.
Consequently, EBCF is a Saccheri quadrilateral and thus angle BCD, which is one of the summit
angles, is acute. ■

Theorem 3.13: In a quadrilateral ABCD with right angles at A and B (Fig 3.18), if AD
is greater than BC, then angle ADC is less than angle BCD.

Proof: Since AD is greater than BC, take point E on AD such that AE is equal to
BC.
D C

A B
Figure 3.18
Then, ABCE is Saccheri quadrilateral so that angles AEC and BCE are equal.
Thus, angle AEC is less than angle BCD while it is greater than angle ADC.
Consequently, angle ADC is less than angle BCD. ■

3.3.3 The Sum of the angles of a triangle

☺ Dear student! You are not expected to be confused when you obtain a completely different
result from now on; because you have experienced some of such results. You will discover other
strange results too; through out the study of Hyperbolic Geometry.

152
So far, we have seen several concepts which are out of the facts in Euclidean geometry. In fact, it is
due to the negation made to the fifth postulate of Euclid. One of the major surprising results in
hyperbolic geometry is that the sum of the angles of a triangle is less than two right angles.

Theorem 3.14 : The sum of the angles of every right - angled triangle is less than two
right angles.

Proof: Let ABC be right triangle with right at C (Fig 3.19).

At A construct angle BAD so that it is equal to angle ABC; and from the mid - point M of AB, draw
a perpendicular to CB. Also take point Q on AD such that AQ is equal to PB.

A Q D

C
B
P

Figure 3.19

Then, triangles MBP and MAQ are congruent, and angles AQM and BPM are equal. It follows that
angle AQM is right, and points A,Q and D are collinear.
Thus, ACPQ is a Lambert quadrilateral with right at C, P and Q while angle CAQ, the fourth angle,
is acute. That is, the sum of angles CAB and QAM is acute.
Then, since angles QAM and ABC are equal, we conclude that the sum of angles CAB and ABC
is also acute (or , less than a right angle). Therefore, the sum of angles CAB, ABC and ACB is less
than two right angles. ■

Theorem3.15 : The sum of the angles of every triangle is less than two right angles.
Proof: Assume triangle ABC is not right, and let angles at B and C be acute (Fig 3.20).

153
 (This is since at least two of the angles of every triangle are acute)

B C
D
Figure 3.20

Draw the altitude AD from A to BC. Thus, triangle ABC is divided in to two right triangles ADB
and BAD is acute, and the sum of angles ACD and CAD is also acute; which is shown in the proof
of the previous theorem. Therefore, the sum of these four angles ABD, BAD , CAD and ACD,
which is the sum of the three angles of triangle ABC, becomes the sum of two acute angles (less than
two right angles). ■

Corollary: The sum of the angles of every quadrilateral is less than four right angles.
Note that we do not have a rectangle or a square in hyperbolic geometry.

3.4 The Defect of a triangle and the Collapse of the Similarity Theorem

In the previous section, the sum of angles of a triangle has been introduced to be less than two right
angles. Based on this, one can think the difference between two right angles and the sum of the
angles of a triangle, which is a non-zero number.

Definition3.5:- The difference between two right angles and the angle-sum of a triangle
is called the defect of the triangle.

154
 That is, given triangle ABC, using the Euclidean angle measure,

Defect ( Δ ABC) = π −[ m ( ∠ A ) +m ( ∠ B ) +m (∠C ) ]

Theorem 3.20: If a triangle is divided into two triangles by a line from a vertex to a
point on the opposite side, the defect of the original triangle is equal to
the sum of the defects of the two smaller triangles.

Proof: Let ABC be a triangle with measure of angles at A,B and C labeled
α , β and γ , respectively (Fig. 3.25).

B β

α γ

A D C

Figure 3.25

Clearly, the defect of triangle ABC is π −( α + β+γ ) . Now, let D be any on AC, and draw BD,

which divides angle β into two smaller angles, say β 1 and β2 , and which forms two

supplement angles at D, say θ1 and θ2 .

π −( α +θ1 + β1 )
Then accordingly, defect of triangle ABD = , and
=π−( γ +θ2 + β 2 )
defect of triangle CBD .

Thus, the sum of the two defects will be equal to

2 π −( α +γ + θ1 +θ 2 + β 1 + β 2 )

This is equal to
π −( α +γ + β ) , the defect of triangle ABC, since from the construction we have
θ1 +θ 2=π and β 1 + β 2 =β . ■

155
 Next, the concept of similarity of triangles will be introduced. However, as a consequence of the
change in the fifth postulate, surprising result about similar triangles is to be discovered.
In Euclidean geometry, you have had the conclusion:
If two triangles are congruent, then they are similar, but not conversely. But here in hyperbolic
geometry, the converse is also true so that the concept of similarity and congruence of triangles are
one and the same.

Theorem3.21: If the three angles of one triangle are equal, respectively, to the three angles of a
second triangle, then the two triangles are congruent.

Proof: Let angles A, B, C of triangle ABC be equal, respectively, to angles A', B', C'
of triangle A'B'C' (Fig 3.25)

A A'

D E
B' C'
B C

Figure 3.25

If any pair of corresponding sides, say AB and A'B', are equal, the triangles are of course congruent
(by ASA).
Assume that AB and A'B' are not equal; then one of them, say AB is greater.
Take points D and E on the sides AB and AC, respectively such that
AD = A'B' and AE = A'C'.
Then, triangles ADE and A'B'C' are congruent and thus
angle ADE = angle A'B'C' and,
angle AED = angle A'C'B'.
This, together with the equality of the corresponding angles given in the hypothesis, yields:
Angle ADE = angle ABC , and angle AED = angle ACB

156
 And consequently, DBCE is a quadrilateral with the sum of its angles equal to four right angles.

(i.e. let angle ADE = angle ABC = θ , angle AED = angle ACB = β so that the sum of angles

of the quadrilateral BCED is θ+β +π−β +π−θ=2 π , four right angles).

But this is a contradiction and thus sides AB and A'B' must be equal and the triangles are congruent.
If AE were equal to AC, the congruence is obvious as AE = AC = A'B'.
If E is so that C lies between A and E, a situation will be encountered in which an exterior angle
of a triangle is equal to one of the opposite interior angles, which is again a contradiction. ■

- Note that we reached the remarkable conclusion that in hyperbolic geometry similar
triangles of different sizes do not exist.
That is, every two similar triangles are congruent !

Exercises
Prove the following based on the basic assumptions of Hyperbolic Geometry.
1. If angles AB Ω and BA Ω are equal, the triangle is in the nature of isosceles triangle
with vertex an ideal point. Prove that, if M is the midpoint of AB, then M Ω is
perpendicular to AB. Show also that the perpendicular to AB at M is parallel to A Ω and B
Ω and that all points on it are equally distant from those two lines.
2. Define each figure precisely and explain the differences and similarities between them i)
Saccheri Quadrilateral
ii) Lambert Quadrilateral
3. Can the base and summit of a Saccheri quadrilateral be equal? If not which one is greater?
Explain the reason in detail.

3.5 Summary

The main points of this Chapter are to be summarized as follows.

♦ Hyperbolic parallel postulate is stated as : Through a given point, not on a given line, at least
two lines can be drawn that do not intersect the given line.

157
 ♦The geometry, which is developed based on the first four Euclid 's postulates and the hyperbolic
axiom, is called hyperbolic geometry.
♦ Given a line ℓ and a point P not on ℓ , point Q on ℓ such that PQ perpendicular to ℓ .
Rotate line PQ about P in either directions. The first non-intersecting lines during the rotation are

called the Parallels to ℓ .

♦ The parallels through P to ℓ form equal and acute angles with the perpendicular PQ each of
which is called Angle of Parallelism.
♦ Two parallel lines are assumed to intersect at an imaginary point called ideal point; denoted by
Ω.
♦ There is no line that is perpendicular to both of the two parallel lines; and two non - intersecting
lines have one and only one common perpendicular.
♦ Two non - intersecting lines have one and only one common perpendicular.
♦ Given two points P and R on two parallel lines ℓ and m respectively, an important figure or
triangle PR Ω is formed, called asymptotic triangle. Asymptotic triangles share many properties of
the ordinary triangles. For example,
- If a line passes within the triangle AB Ω through one of the vertices, it will intersect the
opposite side.
- If a straight line intersects one of the sides of triangle AB Ω , but does not pass through a
vertex, it will intersect one and only one of the other two sides.
- The exterior angles of triangle AB Ω at A and B, made by producing AB, are
greater than their respective opposite interior angles
♦ A Saccheri quadrilateral is a quadrilateral ABCD with two adjacent right angles at A and B and
with sides AD equal to BC. The side adjacent to the two right angles, side AB is called the base and
its opposite, DC is called summit of the Saccheri quadrilateral.
- The angles adjacent to the summit (i.e. angles ADC and BCD) are called summit angles.
♦ A quadrilateral with three of its angles right angles is called Lambert quadrilateral.
♦ In hyperbolic geometry;
- The sum of the angles of every triangle is less than two right angles.
- The sum of the angles of every quadrilateral is less than four right angles
- Every two similar triangles are congruent

158
♦ The difference between two right angles and the angle-sum of a triangle is called the defect of the
triangle.

i.e defect ( Δ ABC) = π −[ m ( ∠ A ) +m ( ∠ B ) +m (∠C ) ]

3.6 Self Test Exercises

I) True or False for questions.

1. A line can be parallel to both of the two intersecting lines.
2. A line can be parallel to both of the two non-intersecting lines.
3. Twice of any angle of parallelism is obtuse.
4. A line has exactly two ideal points.
5. If P and R are two points on two parallel lines such that ∠ PRΩ≡∠ RPΩ , then the
angle of parallelism at P is less than each of these equal angles.
6. A line has exactly two ultra-ideal points.
7. The base of a Saccheri quadrilateral is shorter than its summit.
8. There is a Lambert quadrilateral which is also a Saccheri quadrilateral.
II) Work out problems
9. State the hyperbolic parallel postulate and identify the parallels guaranteed by the postulate.
10. How many parallels can be drawn to a line through a point not on the given line?
11. Show how two parallel lines may be thought of as intersecting at a zero angle.
12. If two lines BA and BC are both parallel to line ℓ , show that the bisector
of angle ABC is perpendicular to ℓ .
13. If in the figure below (Fig.3.37) , the angles at A and B are right angles and the angles at C and
D are equal,
A D

B C
Figure 3.27

159
 then prove the figure is a Saccheri quadrilateral.
14. Prove that, in a Lambert Quadrilateral, the sides adjacent to the acute angle are greater
than their respective opposite sides.
15. Prove that the line joining the mid points of the equal sides of a Saccheri quadrilateral is
perpendicular to the line joining the midpoints of the base and summit and that it bisects
the diagonals.

CHAPTER FOUR
THE CONSISTENCY OF THE HYPERBOLIC GEOMETRY

By Nasir Asfaw

Introduction
In this chapter we present the fundamental concepts of hyperbolic geometry so that the
consistency of the required geometric principles can be observed.
A remarkable attention will be given on illustrations of how to realize the imaginary ideas of
hyperbolic geometry by Euclidean models, how a transformation can be applied to the non -

160
Euclidean points, lines and how to extend several axioms of Euclidean geometry to their
corresponding statements in hyperbolic geometry.

As discussed at the beginning of the previous chapter the concepts in hyperbolic geometry are
based on the four postulates of Euclid and the negation of the fifth postulate. The results obtained are
to be sought as an independent theory which is developed on the basis of its own axioms. No
contradiction has ever been discovered by anyone with in the development of hyperbolic geometry.
That is, hyperbolic geometry is consistent; and its consistency has been proved comparatively, for if
inconsistency in hyperbolic geometry yields inconsistent concepts in Euclidean geometry.

In addition, the results of hyperbolic geometry, the defined terms or points, such as ideal points
are consistently used in this chapter.

Objectives
After completing the chapter successfully, you will be able to:
Appreciate the consistency of Hyperbolic Geometry on its own axioms without any
contradicting ideas.
Illustrate the concepts of hyperbolic geometry using Poincare’s Euclidean model.
Define inversion of points and explain the effects of inversion on lines, circles, angles and
line segments.
Identify the relations between inversion of points and reflection along ps- lines in the
Poincare’s model
Elaborate basic similarities and differences between Euclidean and hyperbolic geometry
regarding axioms, principles and the consistency of ideas with in the two geometries
Define length of a segment in hyperbolic geometry and show its invariance under reflection
or displacement.

4.1 The Poincare’s Model

The general principles of Hyperbolic Geometry, being based on imaginary ideas, are difficult
to visualize and illustrate on the plane of our perceptions in which most events of the physical world

161
can be expressed. But, the properties of geometric figures and relations of points, lines and their role
in geometry could be visualized by drawings in a plane.

Different models have been made so as to present the concepts of Non-Euclidean Geometry, and
for further illustrations of its consistency. The two frequently used geometric models that can aid our
visualization of hyperbolic plane geometry are the Poincare’s Model and Klein model. In this
module only Poincare’s model will be presented.

Although Lobachevsky, Bolyai and Gauss had developed the new theory of hyperbolic geometry
with confidence, still mathematicians did not feel as sure of its validity as they felt of Euclid’s
geometry. However a few decades after mathematicians began to find models (realizations) for
hyperbolic geometry in side Euclidean geometry.

The French mathematician J.Henri Poincare had developed a model which enables the concepts
of Non-Euclidean Geometry to be represented by using Euclidean geometry.
The model is named Poincare's Euclidean model for non-Euclidean geometry or simply Poincare's
model. Poincare used terms of Euclidean geometry to represent that of Hyperbolic Geometry.

To indicate terms of Non-Euclidean Geometry or Pseudo Geometry, he used ps in short. For

instance, ps-line, ps-point, ps-angle, etc for lines, points and angles of the pseudo geometry (non-
Euclidean geometry). Further, he designated, as ps-line, any circumference which cuts the circle C
perpendicularly; as ps-angle between two intersecting curves is their geometric angle.
In general, the following table can be used as dictionary for Poincare's terms, which translates
geometric features in the hyperbolic plane into corresponding features in the ordinary Euclidean
plane in which the model is embedded.

Terms in Poincare's disk model of the Euclidean Plane

hyperbolic plane
a ps- point a Euclidean point inside the unit disk
a ps-line the portion inside the unit disk of any
Euclidean circle meeting C
perpendicularly, or a diameter of the circle

162
 C
a ps-angle between ps- lines the Euclidean angle between the two
Euclidean curves that are the ps-lines
a ps-distance between ps-points given by a particular formula that involves
the distance of the points from the circle C.

Table 4.1 Dictionary for Poincare's disk model.

The above dictionary will be used to explore properties of non-Euclidean geometry by working
simply with their Euclidean counter parts. Some of the features of Poincare’s model are illustrated
below (Figure 4.4).

A B
I C
O⋅¿ ¿ P
R
Q D
H
G E
F

Figure 4.4

The model involves a circle of center O which is the plane of the geometry (or, ps-plane); and five
ps-lines, namely ps-lines AB , BE, CF, DG and HI. Three of them intersect in three ps-points P, Q
and R and thus forming a ps-triangle PQR, and the other two lines do not intersect any of the other
lines shown.

Note that points A, B, C, etc are not ps-points since they are not inside circle O (the ps-plane); and
thus ps- lines AB and BE do not intersect.

Since the model is developed to realize the imaginary concepts of hyperbolic geometry on the
Euclidean disk, points on the circle O are assumed to correspond the ideal points where parallel lines

163
intersect. For instance , the ps-lines AB and BE are parallel. Moreover, ps-line HI is non-intersecting
(or, ultiraparallel) to any of the other ps-lines.
The ps-angles are simply the angles between two intersecting ps-lines such as angle PQR, ERD, etc
and ps-distance is not as simply defined as angles, but it consists ratio of straight segments which
will be given later.

Using the model, one can verify that all of Euclid’s postulates except the parallel postulate are
satisfied. Moreover, the basic principles in the theory of Hyperbolic Geometry hold within this
model.

4.2 Inversion of a Punctured Plane

Inversion is a kind of transformation of a plane through which points are taken to other points,
called their inverse choose any point P in the plane of a fixed circle with center at O and radius r. On
ray OP construct the point P' such that the product of OP and OP' is the square of the radius r.
That is take P' on ray OP such that
(OP)( OP') = r2

R = R'
O Q' P P' Q

Figure 4.5
Then, point P' is called the inverse of P and, the circle is circle of inversion while its center point O,
is called the center of inversion and r2 is the constant of inversion.

Suppose that Q is out side the circle of inversion. Then its inverse, Q' lies in side the circle. If R is on
the circle of inversion, the inverse R' is also on the circle.

164
 In general, points inside the circle of inversion are inverted to points outside the circle and vice
versa; and the points on the circle of inversion are fixed.

Note: If point P moves closer to the center O, then its inverse P' moves farther away so that no point
corresponds to the inverse of O. Thus, the above definition defines an inverse for every point
in a plane, except the center of inversion. This is why the inversion is for the punctured
plane.

But, since an ideal point is already introduced in hyperbolic geometry as an imaginary point at
infinity, we may extend the concept of inversion to that of a plane by assigning the ideal point on ray
OP to be the inverse of the center O.

Theorem 4.1 : If ℓ is a line through the center of inversion, then ℓ is invariant under the

inversion. That is , for any point P on ℓ , its inverse P' is also on ℓ .

Proof: Given line ℓ through the center of inversion O, if P is any point on ℓ , it is the definition

that P' lies on the ray OP, and thus on ℓ .

Similarly if Q is any point on ℓ to the side on O opposite to P, then Q' lies on ray OQ

which is also on the line ℓ .

Now inversion can be defined for lines or curves in general. Since a curve is the trace of a moving
point, the corresponding inverse points traces the inverse of the curve.
That is, if a moving point traces any curve continuously, its inverse will trace continuously a curve
called the inverse of the first (Fig. 4.6).
When point P moves through R to point Q tracing the curve PRQ, its inverse, P' moves through R' to
point Q' forming the curve P'R'Q', inverse of the curve PQR.
- The image of plane figures such as triangles, circles, etc under an inversion will be similarly
defined since plane figures are composed of several curves.

P'
P

165
 P
O R R'
Q

Q'

Figure 4.6

Exercise 4.2

1. If A and A' are inverse points, prove that any circle through them is orthogonal to the circle
of inversion.
2. If A, A' and B, B' are pairs of inverse points, O the center and r 2 the constant of inversion,
show that
2
r
AΒ= Α' Β'
( ΟΑ' ) ( ΟΒ' )
3. Prove that if two intersecting circles are orthogonal to a third circle, their points of
intersection are inverse points with regard to the third circle.
4. Show that measures of angles between any two lines (curves) are preserved under inversion
with respect to any circle of inversion.

4.3 Cross ratio and inversion

In hyperbolic Geometry the ratio of ratios may have a remarkable importance than ratios of
lengths. In this section the meaning of cross ratio of four points will be given and then the effect of
inversion on cross ratio of points is to be introduced.

Definition 4.3 : Let A, B, C and D be any four points in a plane. The cross ratio of these four
points, in the given order, is defined to be

166
 AC ΒC
ℜ ( A , Β , C , D)= ÷
AD ΒD
where AC, AD, BC and BD are the distances between the respective points.
From the definition we can observe that cross ratio is a ratio of two ratios and it can be expressed
alternatively by equivalent ratios.
AC ΒC AC ΒD AC AD
ℜ ( A , Β , C , D)= ÷ = . = ÷
That is, AD ΒD AD ΒC ΒC ΒD

B
A

Figure 4.7

☺ Note that the definition of cross ratio includes the particular case when the four
points are collinear.
Theorem 4.2: If A, B, C, D are four distinct points in a plane, then the cross ratio,
R (A, B, C, D) remains unchanged when any two pairs of the points are
interchanged.

That is, ℜ (A, B, C, D) = R(B, A, D, C) = ℜ (C, D, A, B) = R(D, C, B, A)

Proof: In each case, not a pair, but two pairs of points are interchanged. For instance , the pair A, B
and C, D are interchanged to change the order from A, B, C, D in to B, A, D, C.
It can be verified simply by using the definition as;
ΒD AD
= ÷
ℜ (B, A, D, C) ΒC AC
ΒD AC AC ΒD AC BC
= . = . = ÷
ΒC AD AD ΒC AD ΒD
=ℜ ( A , Β , C , D )

167
 The others can be shown in the same way .
Theorem 4.3 : If the cross ratio of four points in a given order is ρ , the interchanging either the

1
first pair or the second pair of points changes the cross ratio to its reciprocal ρ .

1
ℜ ( A ,B , C , D )= ρ , then ℜ ( Β , A ,C , D )=ℜ ( A ,B , C , D )=
i.e. If ρ
AC ΒD
ℜ ( A , Β , C , D )= . =ρ
Proof: Let AD ΒC . Then,
ΒC AD AD ΒC 1 1 1 1
ℜ ( Β , Α , C , D )= . = . = . = =
ΒD AC AC ΒD AC BD AC ΒD ρ
.
AD ΒC AD ΒC .
1
ℜ ( A , Β , D , C )=
Similarly, we have ρ .

Remark:
The order in which points A,B,C,D are taken is so important because it changes their cross
ratio.
Interchanging two points may change the cross ratio depending on which pair of points is
interchanged.
i.e. Interchanging the first pair or the second pair
- Interchanging the inner pair or the outer pair.

Now, let us consider any four points A,B,C, and D in a plane and their inverses A',B',C' and D' under
any center or circle of inversion. An important conclusion will be made regarding the cross ratio of
the four points and that of the inverse points.

Theorem 4.4: Inversion preserves the cross ratio ℜ (A,B,C,D) of any four distinct points
A, B,C and D in the plane.

Proof: Let A,B,C and D be any four points, and A',B',C' and D' their inverses with respect to a
circle of inversion centered at O.
Then, we need to show that

168
 AC ΒD A ' C ' Β ' D'
(
ℜ ( A , Β ,C , D )=ℜ ( A ', Β ', C ', D' ) ; or , . = .
AD ΒC A ' D' Β ' C ' )
For two points A and C, and their inverses A' and C' under circle of center O(Fig. 4.8) , by
2 2
definition we have ( ΟΑ ) ( ΟΑ' )=r and ( ΟC ) ( ΟC' )=r , where r is the constant of inversion.

A'

C C'

Figure 4.8

ΟA ΟC '
=
Then ( ΟΑ ) ( ΟΑ' )=( ΟC ) ( ΟC ' ) so that ΟC ΟA'

 two sides of ΔΟ AC are proportional to the corresponding two sides of ΔΟC ' A ' .
Moreover, since angle at O is common for both triangles, and thus,
Δ Ο AC ~ Δ ΟC ' A'

AC ΟA
= ⋯⋯⋯⋯ ( i )
From this we obtain A ' C ' ΟC '

Similarly, for two points A and D , we get Δ Ο AD ~ Δ ΟD ' A' so that

AD ΟA
= ⋯⋯⋯⋯⋯( ii )
A ' D ' ΟD '
Then, solve for AC in (i) and for AD in (ii), and then taking the ratio of the two,
we get

169
 AC A ' C ' ΟD '
= ⋯⋯⋯⋯( iii )
AD A ' D ' ΟC '
Again, by taking points B, D and B, C together with their respective inverses, we can obtain two
equal ratios;
ΒC ΟΒ ΒD ΟΒ
= and =
Β ' C ' ΟC ' Β ' D' ΟD '
Further , solving for BC and BD, and taking their ratio, we get

ΒC Β ' C ' ΟD '

= . ⋯⋯⋯⋯⋯( iv )
ΒD Β ' D ' ΟC '
Finally, taking the ratio of ratios (iii) and (iv), it follows
AC C A' C ' ' C '
/  /
AD D A' D ' B ' D '

Therefore, ℜ (A, B, C, D) = ℜ (A', B', C', D')

This proves that cross ratio of points is always preserved under inversion with respect to any
circle.

Exercises 4.3

1. Given the points A(1,1) , B(3, 2) C(1,0) and D (-1,2) , find ℜ (A,B,C,D) and
ℜ (C,A,B,D).
2. Prove that , if A,B,C,D and E are five distinct collinear points, then
ℜ (A,B,C,D) . ℜ (A,B,D,E) = ℜ (A,B,C,E)
3. If A,B and D are on a circle of inversion and C is any other point different from the center of
inversion , then show that
ℜ (A,B,C,D) = ℜ (B,A,D,C' )
4. Given four points A, B, C, D and their inverses, under a certain circle are A',B',C', and D',
respectively. Verify the following
i) ℜ (A, B, D, C) = ℜ (B',A'.C',D')

ii) ℜ (B,A,C,D) = ℜ (A', B',D,' C' )

iii) ℜ (D,C,B,A) = ℜ (C', D', A', B')

170
 4.4 Angular Measure and Inversion

Given points A and B (not collinear with the center of inversion O) and their respective inverses A'
and B' , then different angles will be formed at points A,B, A' and B' when segments AB and A'B'
are constructed. These angles do have an important relation which is stated in the following theorem.

Theorem 4.5 : If A and B are any two points (not collinear with the center of
inversion O) and A' and B' are their respective inverses, then angle OAB
is equal to angle OB'A'.
Proof: Let A,B and A',B' be the four points as given in the theorem.

A'
A

O
B
B'
Figure 4.9

We have (OA) (OA') = r2 and (OB) (OB') = r2 , where r is the constant of inversion (the radius of
circle of inversion).
 (OA) (OA') = (OB) (OB') so that we obtain
 

' '
This gives that the ratio of two sides of triangle OAB is equal to that of the two sides of triangle
OB'A'.
The (common) included angle O, thus yields: triangle OAB similar to triangle OB'A'
Consequently, angle OAB is equal to angle OB'A', and angle OBA is equal to angle OA'B'.
Note that we are considering the finite case (inversion of punctured plane ) so that similarity
is that of Euclidean plane.

171
Corollary :- The four points A,B, A' and B' given in Theorem 4.5 lie on the same circle.

From Theorem 4.5 it is easy to obtain that angles BAA' and BB'A' are supplementary and so are
angles ABB' and AA'B; which yields quadrilateral ABB'A' of opposite angles supplementary.
Therefore, there exists a circle circumscribing the quadrilateral (i.e. containing points A,B, A' and
B').

Recall that inversion fixes every line through the center of inversion. But the image of a line
which does not pass through the center of inversion is different, even not a line as shown below.
Theorem 4.6 : Every straight line which does not pass through the center of inversion
inverts in to a circle passing through the center of inversion, and
conversely.
Proof: Let ℓ be any straight line not passing through the center of inversion O
(Fig. 4.10). Draw from O the perpendicular OA to ℓ .

B'

O A
C A'
ℓ

Figure 4.10

Let A' be the inverse of A and take any point B on ℓ .

Then B' is its inverse and we have angle OB'A' = angle OAB = right angle.

172
 Furthermore this is true no matter where on line ℓ the point B is located. As B moves along
ℓ , B' traces out the locus of all points where angle OB'A' is right angle. Point C, being mid point
of OA' is equidistant from the vertices O, A and B' of the right triangle OB'A'.
That is for every point B on ℓ , B' is a fixed distance CA' waits far from C forming a circle
centered at C.
Therefore, line ℓ is inverted to the circle of diameter OA'. Similarly, for any circle through O,
draw a secant from O through the center of the circle cutting at P. Take P', the inverse of P. Then
show that the line through P' perpendicular to OP' is the required image of the circle.

Now inverse of lines are already determined for both that pass through the center of inversion and
that do not as well. But image of a circle is known only for circles which pass through the center of
inversion. What would the inverse be for circles which do not pass through the center of inversion?
Theorem 4.7 : Circles which do not pass through the center of inversion invert in to
circles.
Proof: Let C be the center of any circle not passing through the center of inversion O. (Fig. 4.11).

P
P' R
C
Q
O Q' C'

Figure 4.11

Draw 0C , cutting the circle in points Q and R. Let P be any other point on the circle. Then P' , Q'
and R' are the respective inverses of points P, Q and R. Consequently, angle OR'P' = angle OPR and
angle OQ'P' = angle OPQ
Then we conclude that;
angle R'P'Q' = angle QPR = right angle. (show!)
This is true for every point P on the circle so that P' traces a circle of diameter R'Q'.
Therefore, the inverse of circle PQR is circle P'Q'R'.

173
 The Effect of Inversion on Angles
Let us first consider the effect of inversion on angles between two straight lines. Then, it is possible
to extend this to the angles between any two curves.

Suppose two straight lines PR and QR intersect at R (Figure 4.12). What will happen to the size
of angle PRQ if inverted, under the center of inversion O?

R R'

0
P

Q
Q' P'

Figure 4.12

Draw OP so that it contain Q, and let P' and Q' be the inverses of P and Q. Similarly, draw OR and
let the inverse of R be R'.
Claim: Angles PRQ and P'R'Q' are equal.
Now from the previous theorems,
we have angle ORP = angle OP'R', angle ORQ = angle OQ'R'.
Then, taking appropriate similarities, it can be shown that: Angle PRQ = Angle P'R'Q'
Now recall that the angle between two curves (or, precisely nominal lines) is the angle between
their tangent lines drawn at the point of intersection.
Consider the two intersecting curves at point R; then it is obvious for their inverses to intersect at
R' as shown in the figure below.

174
 R'

0
P Q Q' P'

Figure 4.13

To show that the angles between the curves PR and QR are equal, first draw the straight segments
PR, QR , Q'R' and P'R'.
Clearly, angles PRQ and P'R'Q' are equal.
If P moves on the curve PR to wards R, the secant line PR moves and approaches to the tangent at R.
Similarly, secant QR approaches the tangent at R when Q moves on the curve RQ and approaches
point R.
These points, while moving, forces their images to move in the same way, keeping the equality of
the angle and its image unchanged. (Why?)
In the limiting case (when P,Q approach R resulting in P; Q' approach R') , we get the angles
between tangents at R equal to that of the tangents at R'.
Therefore, the angle between the curves PR and QR is equal to the inverse angle under the required
inversion.
Note that in general, inversion preserves the measure (size) of angles formed by any two
intersecting curves.

4.5 Reflection across ps- lines in the Poincare Model

In Euclidean geometry, a point P is reflected along the line  means it is taken to the point P'

such that  is the perpendicular bisector of the segment PP'.

175
 Q 

P Q'

P'

Figure 4.14

Under the same reflection point Q is taken to Q' so that segment P'Q' is the image of PQ. Different
plane figures have also their own images under a given reflection. This special transformation,
reflection, has many properties, such as preserving length, angles, congruence, etc.

Now in hyperbolic geometry we do have reflection through a line. Of course, the reflection to be
considered here will be along the ps-lines of the plane (as viewed in the Poincare model).

However, reflection involves equality of segments and it requires to preserve the length of line
segments. But in hyperbolic geometry length of a segment (ps- length of the ps- segment) is still not
defined. It will be defined now so that hyperbolic geometry is consistent. That is the length of a
segment which is preserved under reflection is to be defined as follows.

4.5.1 The ps- length of segments of ps- lines

There are three requisites for ps-lines which must be kept in mind in selecting a definition for the
ps- length of a ps-segment, if the analogy to hyperbolic geometry is to be preserved.

1. A ps-line must be infinitely long.

176
 2. If A, B and C are any three points at all on a ps-line, then ps-length AB plus
ps-length BC equals ps- length AC, sense being taken in to account.
3. The ps- length of a ps- segment must be unchanged by displacement.
The following definition satisfies requirements.
Consider the ps-line TS and two points on it, namely A and B on the fundamental circle of center
at O (Fig. 4.15).

 B

O  A

S
Figure 4.15

It is to say the circle with which the ps-line coincides cut the fundamental circle at points T and S.
Then we define the ps- length of the ps- segment AB as:
A 
log e (  )
AS S
where AT , AS , BT , BS designate chords of the circle.
A segment AB will be a unit segment when this logarithm is unity.
A S
. e
That is, when AS 

The ps- length of the ps- segment AB may be defined more generally
A 
k log e ( / )
AS S
where k is a parameter, the choice of which determines the unit of length. But, for simplicity, we
take k = 1.
? Question: Show that, the ps length of AB becomes infinite as A approaches S or B approaches T.
From the statement of this question is that the first condition for length of ps- segment satisfied.

177
 It is also clear that the given definition of length satisfy the second condition. But to show that the
third condition is satisfied by the given definition, first we have to explain what we mean by a
displacement (or, reflection) in the geometry we are dealing with.

4.5.2 Reflection along ps- lines

Let the circle with center O be the fundamental circle and AB any nominal line (Fig.4.16)

P  B
O M

A T
P'
C

Figure 4.16

Then AB coincides with an arc of a circle which is orthogonal to the fundamental circle, let the
center of that circle be C. If P is any point and P' its inverse with regard to the circle of center C, then
the ps- line PP' coincides with a circle which is orthogonal to the circle of inversion and to the
fundamental circle as well. That is, curve PM is inverted to curve P'M under the circle of inversion
with center C.
Thus, the ps- line PP' is perpendicular to the nominal line AB.
We have next to show that the ps- segment PP' is bisected by ps- line AB. For this, let the ps- line
PP' cut the fundamental circle at S and T. Clearly, S and T are inverses with regard to the circle of
center C.
 C  '  C '
 
Then, ' S C' , and S CS

178
 which are taken from the similarities of triangles CPT and CSP' , and triangles CP'T and CSP.
(Justify this ! use the effect of inversion on angles when P and S are inverses of P' and T,
respectively).
Taking the product of each side of the two, we get
 '  C
. 
' S S CS
Again since M is on circle of inversion, we have M = M' and thus, triangles CMT and CSM are
similar so that
 C C
 
 S CS C
   2 
C
This implies  S  2 CS

 C  C

(i.e, take  S C and multiply  S to the left and CS to the right)
C
,
Returning to the value of CS we get
 '    
2
, 
 ' S S   S  2
 ' 
.
Then dividing both sides by  S ' S , we get
  ' 
/ /
S  S  ' S
   ' 
log /  log l
Therefore S  S  S ' S
and the ps- lengths of ps- segments PM and MP' are equal; this then enables us to say:
ps- line AB is perpendicular bisector of the ps- segment PP'.
Which enables us to equivalently say that P' is the image of P when reflected along the ps- line AB
We conclude that the inversion of a point, with regard to the circle up on which a ps- line lies, may
be regarded as a reflection of the point with the ps- line used as axis of reflection. That is, reflection
along ps-line AB is the same as inversion with circle of inversion the circle coinciding the ps- line
AB.

179
 Now once we deduced such a good relation between inversion and reflection, we can use the
concept of inversion to show that the ps- length is invariant under reflection.

That is, if ps-segment PQ is inverted to the ps- segment P'Q', under circle of inversion with center
C (Fig 4.17),

O Q T
 P
S M
A B P' Q'

T'
S

Figure 4.17

We need to show that the ps- length of PQ is preserved.

 Q ' ' Q' '
log /  log /
That is, S QS ' S ' Q' S '

Let the circle with center 0 be the fundamental circle and AB any ps- line coinciding circle of center
C. If PQ is any ps- segment and P'Q' is it inverse under circle of inversion of center C.
Further, if ps- line PQ meets AB at M, then its inverse P'Q' also meets AB at M, and makes the
same angle with it. (Why?).
Then since S' and T' are inverses of S and T , respectively, the following equality of ratios holds.
 C

 ' ' C  '
 ' S ' C '

S CS

180
 Q C

Q' ' CQ'
Q' S ' CQ'

and QS CS

(All of the above ratios are due to the fact that P', Q', T', S' are inverse points of P , Q , T, S under
center of inversion C ).
Then, rearranging the above equations, we get
 ' ' CS C' Q Q' ' CS CQ' '
 ,  ,  and 
C C' S ' S ' C CQ' QS Q' S
And multiplying the first two and dividing this product by the product of the third and fourth
equations, we obtain.
 Q ' ' Q' '
/  /
S QS ' S ' Q' S '
Therefore, taking the logarithms proves the equality of the ps- segments PQ and P'Q' showing that
ps-length is preserved under the required inversion, and thus under reflection.

In general, the following conclusions can be made for the consistency of concepts in hyperbolic
geometry.
1. Reflection about a given ps- line and inversion with respect to the circle that coincides this
ps- line are the same transformations.
2. Reflection about a given ps-line preserves measure of angles and length of segments, where
length is as defined recently; and thus figures will remain unchanged in size and shape.
3. With ps-length and displacement (reflection) defined, every proposition of hyperbolic
geometry will have its counter part in the geometry of ps- lines.

Exercises 4.5

1. Let A and B be two ps-points within the Poincare model.

If Q is any other ps- point:
i) draw two ps- lines through Q which are parallel to the ps- line AB.
ii) draw a ps- line through Q which is perpendicular to the ps- line AB

181
 iii) Prove that the ps- line drawn in (ii) bisects the ps- angle between the two ps- lines
drawn in (i). then relate this property to that of angle of parallelism in hyperbolic
geometry.
iv) draw ps-lines AQ and BQ , to form the ps- triangle ABQ and describe some of its
properties in relation to that of triangles in hyperbolic geometry.
2. If length of ps-segment DE were defined simply as the arc length of arc DE, as it was in
Euclidean circle, then list some of the major properties of length which are inconsistent with
the principles and assumptions of hyperbolic geometry as viewed in the Poincare’s model.

4.6 Uniqueness of a ps- line through two points

Among the postulates of Euclid, one is that it is possible to draw a straight line from any point to
any other point. That is, through two points exactly one line can be drawn. Also this postulate has
been extended to hyperbolic geometry. Indeed it is one of the first four postulates of Euclid which
are also assumed to hold in hyperbolic geometry.

As already indicated in chapter three, this postulate is stated as: through two points, whether both
ordinary or both ideal points or one ordinary and the other ideal, there is exactly one line.

The lines are to be thought as arcs of orthogonal circles or simply ps-lines in hyperbolic geometry
as introduced in the Poincare model. Therefore, it is to be shown that two points determine a unique
ps-line and this in turn shows that hyperbolic geometry is consistent with its axioms.

Let A and B be two points in a plane (fundamental circle of center O) (Fig. 4.18)
Then we need to show that exactly one ps- line passes through the two points A and B.

P
Q

A P' 
O 

182
 B

Figure 4.18

That is, there is exactly one circle through A and B which is orthogonal to the fundamental circle.
If we assume there are two ps- lines through A and B, intersecting the fundamental circle at points
P and Q respectively, then these two ps- lines intersect each other at two points A and B, which is a
contradiction to the fact that two ps- lines can intersect at most at one point.
Alternatively, if these ps- lines can be drawn through A and B, say ps-line BAP and BAQ as in the
figure above, then regarding circle of inversion circle BAQ, points B, A and P will be inverted to
points B,A and P' respectively (since A and B are on the circle of inversion).
That is, B = B', A= A' and P  P'
But since inversion preserve measure of angles, it holds true that:
angle BAP = angle B'A'P'
 angle BAP = angle BAP'
 P = P' , which is only when P lies on circle of inversion circle BAQ.
This implies circles BAP and BAQ are not distinct
Therefore, only one ps- line can be draw through two distinct points.

Exercise 4.6
i. In the above conclusions, does it include the case when one of the two points
A and B, is the center of the fundamental circle? Explain!
ii. Draw two parallel ps-lines, two non- intersecting ps-lines, and the common
perpendicular ps- line to the two non-intersecting nominal lines.

4.7 The Ruler Postulate; betweenness; Plane Separation and

angular Measure

183
 In Hyperbolic Geometry all the results obtained are based on the assumptions taken which are
stated as axioms or postulates. These are the first four postulates of Euclid and the hyperbolic
parallel postulate. Also it has been made clear that the differences between the Euclidean and
hyperbolic geometries are due to the fifth postulate of Euclid, which is substituted by the hyperbolic
parallel postulate.

Thus, the geometric relations, principles postulates, which are consequences of or related to the
first four postulates of Euclid, need to hold consistently in both geometries. Some of these common
properties will be given here with their particular meaning in hyperbolic geometry.
The Ruler Postulate
Euclid’s first postulate enables us to draw a straight line from one point P to another point Q; and
the second postulate to extend PQ infinitely from both directions. The Ruler postulate can be
interpreted from this point of view.
Postulate :- (The Ruler Postulate)
The points of a line can be placed in correspondence with the real numbers in such a way
that:
1. To every point of a line there corresponds exactly one real number.
2. To every real number there corresponds exactly one point of the line.

This can be illustrated using the Poincare’s model.

The ps-lines, the arcs of orthogonal circles to the fundamental circle, are of infinite length; because
the length (ps- length) of ps-segment AB is given by the cross ratio

B∙ T
A  S
log ( . )
 AS O∙ A∙
S
Figure 4.19

That is, as B approaches T, BT approaches to zero so that the ps- distance between T and B
approaches to zero so that the ps- distance between A and B increases indefinitely.

184
Therefore, the ruler postulate can be stated as:
The points of a ps- line can be placed in correspondence with the real numbers in such a way
that
1. To every ps- point on a ps- line there corresponds exactly one real number.
2. To every real number there corresponds exactly one ps- point on the ps- line.

Betweenness
In absolute geometry, point C is said to be between points A and B if A, B and C are collinear such
that A - C - B and AB = AC + CB

For the Poincare unit disk, let C be the center of the circle that coincides with the

ps- line containing ps- points P,Q and R. Then, R is said to be between P and Q if the ray CR lies

between C and CQ (Fig. 4.20 )

Q
R
P C

Figure 4.20

In this case, we write P - R - Q or Q - R - P.

Also, the ps- lengths PR, RQ and PQ satisfy
PQ = PR + RQ
To verify, this if PQ is the ps- length PQ, we have
  QS 
Q  log  . 
 Q  S 

185
   RS   R QS  
 log  .   .  
 R S   Q RS  
  RS   R QS 
 log  .   log  . 
 R S   Q RS 
 PR  RQ
Therefore , if P , Q , and R are three ps- points lying on the same ps- line , then exactly one of them
lies between the other two ; for instance R is between P and Q, and in this case P - R - Q and PQ =
PR + RQ.

The Plane Separation Postulate

Again recall the plane separation axiom in Absolute geometry:
Given s line and a plane containing it, the points of the plane that do not lie on
the line form two sets such that:
i) each of the sets is convex
ii) if P is in one set and Q is in the other, then the segment PQ intersects the line.

Consequently, points P and Q lie in the same set if they are on the same side of the line and Q does

not intersect the line. But if P and Q are on opposite sides of the line, Q intersects the line.

Now consider the fundamental circle or the Poincare’s unit disk which is a plane according to the
model.
A ps- line divides it into two set of ps- points; those which lie on one side of the ps- line and those
lying on the other side.

Q
R


Figure 4.21

186
For instance, P and Q are on the same side, and P and R are on opposite sides of
the ps- line.
The following statement is the consequence of the plane separation postulate.
For every ps- line and for any three ps- points P,Q and R not lying on the ps- line:
a. If P and Q are on the same side of the line and Q and R on the same side of
the ps- line, then P and R are on the same side of the ps- line.
b. If P and Q are on opposite sides of the ps- line and Q and R are on opposite sides of the ps-
line, then P and R are on the same side of the ps- line.

Note: To determine whether the two points A and B lie on the same side or opposite side
of the given ps- line, we draw the ps- segment AB and check whether it intersects
the given ps- line or not.
Drawing the straight segment AB is not appropriate! (Why?)
- It has to be remarked also that this ps- segment AB is the one that is uniquely determined

Angular Measure
In hyperbolic geometry angles and angle measurements are defined to be the same as that of the
Euclidean case.
The Angle measurement postulate and the angle construction postulate, therefore, can be simply
extended from their statement in Absolute geometry. That is,
1. Angle measurement postulate:
To every angle there corresponds a real number between 0 and 180
2. The Angle construction postulate

Let A be a ray on the edge of the half - plane H. For every number x between 0 and

180 there is exactly one ray A , with P in H, such that m  A   x.

Now when we turn our concerns to the Poincare’s model, angles and angular measures will be
viewed in the same sense . Of course, we need to recall the angle between two
ps-lines is defined by the angle between their tangent lines at their point of intersection.

Q R

P
187
 Q

Figure 4.22

When the two ps- lines intersect each other at ps- point P (Fig4.22) , one of the angles formed

between them is that of the rays Q and R ;( i.e ∠ QPR).

This angle is of Euclidean type so that there exist a real number x between 0 and 180 that
corresponds to it, which is its degree measure.
That is , m ( ∠ QPR) = xo.
The angle addition postulate is similarly stated as:
m( ∠ QPT ) + m ( ∠ TPR) = m ( ∠ QPR). If ∠ QPR is a straight angle and S is any
point not on QR, then m ( ∠ QPS) + m ( ∠ SPR) = 180

Exercises
1. State the ruler postulate and illustrate its relevance to the hyperbolic terms with in the
Poincare’s unit disk model.
2. When do we say a ps- point is between the other two ps- points? Explain by taking three ps-
points A, B, C.
3. Within the Poincare unit disk take a ps- line containing the two ps- points A and B. Draw a

straight line  passing through A and B. Let P and Q are points inside the circle and on the

line  such that P - A - B - Q

Then, i) are P and Q on the same side or opposite side of the ps- line AB?
ii) the straight segment PQ intersects the ps- line at A and B. Does it affect your
answer to (i) regarding the plane separation postulate? Explain in detail!
4. In the Poincare’s unit disk model, the straight segments through the center of the disk are
also taken as ps- lines of the plane. Concerning such ps- lines, examine the ruler postulate,
betweenness, plane separation.

188
 5. Define the angle formed between two intersecting ps- lines of the Poincare’s unit disk model;
when
ii) both the two ps- lines are arcs of orthogonal circles to the disk.
iii) both ps- lines are diameters of the disk.
iv) one of the ps- lines is an arc of the orthogonal circle and the other is a diameter of the
disk.

4.8 Summary

We can summarize the main concepts of the chapter as follows.

- When two circles intersect one another in such a way that their tangent
lines at a point of intersection are perpendicular, they are said to cut orthogonally and
each is orthogonal to the other.
- The arcs of such orthogonal circles to a fixed circle ( called, fundamental circle ) are of
great importance for that they act as lines in a Poincare model.
- Poincare’s model is the model which uses the interior of a unit disk as a plane and
orthogonal arcs and diameters of the disk as lines; called ps-lines . The ps-lines satisfy
the properties of lines in hyperbolic geometry.
- Given a circle C of center O and radius r, inversion with respect to C is the transformation that
takes any point P to P' such that
(OP)(O P') = r2
The point P' is called the inverse of P and, the circle C is circle of inversion, 0 is called the center of
inversion and r2 is the constant of inversion.
- Points inside the circle of inversion are inverted to points outside the circle and vice versa; and the
points on the circle of inversion are fixed.
- If ℓ is a line through the center of inversion, then ℓ is invariant under the inversion. That is ,

for any point P on ℓ , its inverse P' is also on ℓ .

189
- Every straight line which does not pass through the center of inversion inverts in to a circle
passing through the center of inversion, and conversely.
- Circles which do not pass through the center of inversion invert in to circles
- Inversion preserves the measure of angles formed by any two intersecting curves.
- Suppose A and B are two points on the ps-line that intersects the circle at S and T as shown in the
figure.

 B

O  A

S
Figure 4.23
Then the distance between A and B, called a ps-distance (or, ps-length of
the ps-segment AB) is given by:
A 
log e 
AS S
where AT , AS , BT , BS designate chords of the circle.
- The ps -length of AB becomes infinite as A approaches S or B approaches T. This
shows that ps- line is of infinite length
- Reflection about a given ps- line and inversion with respect to the circle that coincides this ps- line
are the same transformations.
- Reflection about a given ps-line preserves measure of angles and length of segments, where length
is the ps-length; and thus figures will remain unchanged in size and shape under this reflection.
- Through two ps- points in the Poincare’s unit disk, there is exactly one ps-line.

190
- The Ruler postulate, betweenness, plane separation angular measure postulates also hold in
hyperbolic geometry and can be illustrated in the Poincare’s model.

4.9 Self - Test Exercises

1. Determine the inverse of a line with respect to a circle if the line doesn't contain
the center of inversion and
i) intersects the circle at two points
ii) is tangent to the circle
iii) doesn't intersect the circle at all.
2 If two circles are orthogonal, show that the inverse of the center of the first, with respect to
the second as circle of inversion, coincides with the inverse of the center of the second with
respect to the first.
3. Let A, B, C and D be any four collinear points in a plane and assume distances as signed (i.e.
AB = -BA, for instance). If the cross ratio R (A,B,C,D) =  , then R (A,C,B,D) = R
(D,B,C,A) = 1-  .
4. Assume that points A, B, C and D are collinear such that R (A,B,C,D) = -2. Then find all the
cross ratios
i) R (C,D,A,B) ii) R (C,D,B,A)
iii) R (C,B,D,A) iv) R (C,B,A,D)
5. Define precisely a ps-length of a line segment (the ps-distance between two points) and
show that it is preserved under reflection along ps-lines.
6. Do ps-lines contain the center of the fundamental circle? Is a diameter of the fundamental
circle a ps-line? Justify your response!
7. Show that given a ps- line and a point not on it, only one ps- line can be drawn from the point
perpendicular to the given ps- line.
8. Suppose A,B and D are on the same ps- line such that A - D - B. Define what is meant by
the ps- point D is the midpoint of the ps- segment AB.

191
 9. Refer Figure 4.20 If CR bisects ∠ PCQ, then can we say ps- point R is the midpoint of
the ps- segment PQ? Justify your answer by considering the length of the arcs PR and RQ, or
the length of the straight segments PR and RQ; and
the ps-length of the ps- segments PR and RQ.

Chapter 5
Concistency of Euclidean Geometry Chapter Five

Introduction to the chapter

This chapter has three sections numbered 5.1, 5.2, 5.3. It deals with concepts that make use of the
Euclidean Geometry. One application of the Euclidean Geometry is in coordinate Geometry that
uses ordered pair of numbers to represent points and equations in the variables x and y to represent
other geometric objects like lines, line segments, triangles, and etc. Your knowledge of the
Euclidean Geometry is often helpful to solve many problems in coordinate Geometry. The other use
of the Euclidean geometry is in working with Transformation Geometry that deals with isometries
(Congruence transformations) and other similarity transformations.
In this module, only the congruence transformations (isometries) such as translation, reflection and
rotation are discussed. For more details of isometries refer to Transformation Geometry.
5.1The Coordinate plane and Isometries
5.1.1The Coordinate plane
Definition

192
Let L1 and L2 be two coordinatized lines that are perpendicular to each other at their origins. Then
the plane determined by L1 and L2 is called a coordinate plane / The cartesian coordinate plane. The
point of intersection of L1 and L2 is called the origin of the coordinate plae.
We usually call one of L1 or L2 , the X - axis (the first axis) and the other one the Y – axis(the second
axis). We usually call the cartesian coordinate plane the XY - Plane.
 In the cartesian coordinate plane, to each point P on the plane, there is a unique line Lx
through P perpendicular to the X - axis at a point Q with coordinate x relative to the
coordinate of X - axis and also there is a unique line Ly through P perpendicular to the Y -
axis intersecting the Y - axis at a point R with coordinate y relative to the coordinate of Y -
axis.Then correspond the ordered pair (x,y) of real numbers to P.
Conversely, to each ordered pair (x,y) of real numbers, let Q and R be the points on the X - axis and
Y - axis respectively such that Q has the coordinate x relative to the coordinate of the X - axis and R
has coordinate y relative to the coordinate of the Y - axis. Then through Q there is a unique line L x
perpendicular to the X - axis and through R there is a unique line Ly perpendicular to the Y - axis
Then Lx and Ly can not be parallel (Why?)
Let Lx and Ly intersect at a point P.
Then for (x,y) correspond the unique point P determined this way.
With this correspondence, to each point P on the xy - plane, there is a unique ordered pair (x,y) of
real numbers associated to P and to each ordered pair (x,y) of real numbers. there is a unique point P
on the xy - plane associated with (x,,y). i.e. there is a one - to - one correspondence between points
on the xy - plane and set of all ordered pair of real numbers.

We call (x,y) the coordinate of P.

We call x the first coordinate /the x - coordinate / abscissa of P and y the second coordinate / the y -
coordinate / ordinate of P

Y- axis Lx
y P Ly
R (x,y)

193
 X - axis
O x Q

Note:
The coordinate of the origin is ( 0,0 ).
Definition
Let P = (x,y) , Q = (x2 , y2) be points in the xy - plane. Then the distance PQ between P and Q is
given by

Q   x1  x2  2   y1  y 2  2

  x1 , y1 
y1

 y1  y 2 
R
 x1  x2  Q  x2 , y2 

x1 x2

 QR is right angled at R and

RQ  x1  x 2 , R  y1  y 2

Then by pythagoras' Theorem.

194
Q 2  R 2  RQ 2
2 2
Q 2  x1  x 2  y1  y 2
Q 2   x1  x 2    y1  y 2 
2 2

 Q   x1  x 2  2   y1  y 2  2

5.1.2. Isometries
Definitions
1/ A one - to - one correspondence of the coordinate plane on to itself is called a transformation of
the plane.
If T is a transformation, then the image of a point P with coordinate (x,y) is a point P' with
coordinate (x',y'). In this case, we write T (P) = P' or T ((x,y)) = (x' , y').
2/ A transformation T is said to be an Isometry if it preserves distance
i.e. for any two points P and Q if T (P) = P' and T (Q) = Q' , then T is an isometry if and only if PQ
= P'Q'
Theorem 5.1.1.
An isometry preserves betweenness.
Proof:
Let P,Q and R be points on the plane such that P - Q - R.
Then PQ + QR = PR and let T be an isometry and T (P) = P' , T (Q) = Q' , T (R) = R'.
Since T is an isometry, PQ = P'Q', QR = Q'R' and PR = P'R'.
Since P - Q - R , PQ + QR = PR
 P' Q'  Q' R'  P' R'
 'Q' R '.
Hence T preserves betweenness.
Corollary:
The image of a line segment under an isometry is a line segment congruent to it and the image of a
ray with end point P is a ray with end point P' where P' is the image of P.
Theorem 5.1.2.
The image of a triangle under an isometry is a triangle congruent to it.
Proof:

195
Let  AC be given and T is an isometry . Let T(A) = A' , T(B) = B', T (C) = C'

T  A  A' ' ,   AC   A' C ',   C   ' C ' and A  A' ' , AC  A' C ' , C  ' C ' by above
corollary.

Then  AC   A' ' C ' by SSS congruence theorem.

Corollary
The image of an angle under an isometry is an angle congruent to it.
Example

Show that T:  x    x  given by    x, y     x,1  y  is an isometry.

Solution:

Let    x1 , y1  , Q   x 2 , y 2  .

Then        x1 , y1     x1 ,1  y1   '
  Q     x 2 , y 2     x 2 ,1  y 2   Q'

Q   x1  x 2  2   y1  y 2  2 and

' Q'   x1  x 2  2   1  y1   1  y 2   2
  x1  x2  2   y1  y 2  2
 Q
So T is an isometry.

Definition
Let T be a transformation (one - to - one and onto). The transformation T -1 such that T-1 (Q) = P if
and only if T (P) = Q is called the inverse of T.

Theorem 5.1.3
The inverse of an isometry is also an isometry.
Proof:
Let T be an isometry and T-1 is its inverse.
Let P' and Q' be points on the plane and T-1 (P') = P and T-1 (Q') = Q.
 T (P) = P' and T (Q) = Q'

196
 Since T is an isometry PQ = P'Q'
 P'Q' = PQ
 T-1 is an isometry.
Theorem 5.1.4
The images of two parallel lines under an isometry are parallel.
Proof:
Let L1 and L2 be parallel lines and T be an isometry and T (L1) =L'1 , T (L2) = L'2.
To show that L'1 // L'2 Assume that L'1 and L'2 intersects say at R'. Since T-1 is an isometry and R' is
on L'1 , T-1(R') is on L1 and T-1(R') is on L2
 L1 and L2 intersect.
Hence if L1 // L2, then L'1 // L'2.

Definition
Let T1 and T2 be transformations.
Then the transformation T1T2 given by (T1T2) (P) = T1 (T2(P)) is called the composition of T2 by T1.
Theorem 5.1.5
The composition of two isometries is itself an isometry.
Proof:
Let T1 and T2 be isometries and T2 (P) = P' , T2 (Q) = Q' , T1(P') = P" and (T1T2) (Q) = T1 (T2(Q)) =
T1 (Q') = Q"
Since T2 is an isometry, PQ = P'Q'
Since T1 is an isometry, P'Q' = P"Q"
 PQ = P"Q"
 T1T2 is an isometry.
Notation
  
  
n n factors
For a transformation T and natural number n, T =
Definition
The transformation I given by I (P) = P for all P in the plane is called the identity transformation
Definition

197
A transformation T is said to be involution if T2 = I.

5.2 .Translation, Reflection and Rotation

1/ Translation
Definition
Let A and B be two distinct points on the plane. Then the directed line segment with initial point A

and terminal point B denoted A in the specified direction is called a directed vector.

Definition

Let A be a directed vector. A translation of the plane is the transformation A defined as follows.
   
A P' where AB P'P is a parallelogram
B

P'
A

Theorem 5.2.1. _

A translation A is an isometry.
Proof:

Let P and Q be points and  A (P) = P'  A (Q) = Q'

Then ABP'P is a parallelogram and ABQ'Q is a parallelogram.

 A // ' and A // QQ and A  '  QQ'

 '  QQ' and ' // QQ'

 PP'Q'Q is a parallelogram.

198
  Q  'Q'

 A is an isometry.
2.Rotation
Definition

A directed angle < PQR is an angle with initial ray Q and terminal ray R .
Definition
Two directed angles are said to be congruent if they have the same orientation and equal measure.

Definition

A rotation  of the plane at an angle  about a point 0 is the transformation 0, defined by

0 if   0
 0,     
' where   ' and the directed angle  ' has measure , if   
Theorem 5.2.2
A rotation is an isometry.
Proof:
 0,  0,
Let be a rotation. Let P and  be points on the plane and (P) = P' and
 0,
(Q) = Q'.

P
Q Q

199
 P'

In  Q and  P ' Q' ,   ', Q  Q' and  QQ'  '

m  Q   m  Q'  m   QQ '
m   ' Q'  m  Q'  m  '
 Q   ' Q'
  Q   ' Q'
 Q   ' Q '
3.Reflection
Definition
Let L be any line on the plane.

Then for any point P on the plane define the transformation  L by

 P if P is on L

 L (P) =  P ' , where L is a perpendicu lar bi sec tor of ' if   L.

Then the transformation defined this way is called a reflection about the line L.

P'

Next, we will show that a reflection is an isometry.

Theorem 5.2.3
A reflection about a line is an isometry.
Proof:
Let L be a line and P and Q are points on the plane.

200
Case 1, if both P and Q are on L, then  L (P)=P and  L (Q) = Q and PQ = PQ

  L is an isometry.
Case 2, if one is on L and the other not on L, say P is on L and Q not on L.

Q
R
Q'

Then construct  PQ Q' intersecting L at R. Then by definition of reflection

QR  Q' R and QQ'  L.

Then in  QR and  Q' R, R is a common side, QR  Q' R, and  RQ  RQ'

  QR   RQ'

 Q  Q '

But  L (P) =P and  L (Q) = Q'

Hence PQ = PQ' implies  L is an isometry.

Case 3, both P and Q are not on L,
Sub case 1, P and Q are on the same side of L.
L
Q
A Q'

201
 B
P R R' P'

Let  L (P) = P' ,  L (Q) = Q'

Let L bisects QQ' and ' at A and B respectively. Then by definition of reflection,
QA  Q' A' ,   '  and QQ'  L, '  L.

Then QQ' // ' . If in addition QQ'   ' , then QQ'P'P is a parallelogram and so Q  'Q' is not

congruent to ' then AQ    either AQ > BP or BP > AQ

WLOG, assume BP > AQ. Then there exist unique points R and R' such that

AQ  R and AQ'  R '

Then QABR and Q'R'B A are rectangles. (Why?)

and so QR  A  Q' R ' and QR   and Q' R'  ' .

 < QRP and < Q'R'P' are both rectangles QR  Q' R' and R  ' R' (Why?)

Hence  RQ   P ' R ' Q' (by SAS)

 Q  'Q'
Sub case 2, if P and Q are on opposite sides of L

P' p

202
 Q Q'

Let Q intersects L at O.

Then  L ( 0 ) = 0 and since Q - O - P

we have Q' - O - P ( Since  L is a transformation it preserves betweenness).

Then by case 2 above ,   ' and Q  Q '

 Q  'Q' (by segment addition)

Hence in all the cases,
Q  'Q' and so  L is an isometry.

5.3. Incidence and Parallelism

In this section, we will apply the parallelity Axiom to Axioms of Incidence and see more properties
of the Euclidean space.
As an immediate consequence of the parallelity Axiom, given any line L on a plane  , there is at
least one line on  parallel to L.
Theorem 5.31
Given a line L on a plane  , there is at least one line L' on  that is parallel to L.
Proof:
Let L be a line on a plane  . Then there is at least one point P on  but not on L.
Then by the parallelity Axiom. there is exactly one line L' on the plane of L and P that is parallel to
L.
But L and P determine a unique plane that is  .  L' is on  .
Hence there is at least one line L' on the plane  parallel to L.

203
Next, we will define what is meant by a line and a parallel and what is meant by two planes are
parallel and their existence.
Definition
1/ A line L and a plane  are said to be parallel if they do not intersect. at all. In this case we write
L //  .

2/ Two planes  1 and  2 are said to be parallel if they do not intersect at all. In this case we write

 1 //  2 .

Theorem 5.3.2
Given a plane  and a point P not on  , there is at least one line L through P parallel to  .
Proof:
Let a plane  and a point P not on  are given. Let L1 be any line on the plane  . Then P is not on
L1 and so on the plane determined by L1 and P there is exactly one line L through P parallel to L1.

Now, to show L is parallel to  , assume L intersects  at a point Q.

Then Q can not be on L1 because L // L1. Hence Q is not on L1

Then by the parallelity Axiom, on the plane determined by L 1 and Q, there is exactly one line (Call it
L') through Q and parallel to L1. But since L1 and Q are on  , L' is also on  .
But L is also a line through Q and parallel to  and so L' = L.
 L is on 
 P is on 
Contradicting that P is not  .
Hence L intersects  is not possible and so L is parallel to  .
Hence there is at least one line through P parallel to  .
Theorem 5.3.3
If a line L is parallel to a plane  and P is any point on  , then there is a unique line L 1 on the
plane  through P and parallel to L.
Proof:
Let L //  and P is a point on  .

204
Then P is not on L and so there is a unique plane  1 containing L and P. Then  1 is distinct from 

(because  1 continuous L but  does not contain L).

Then  1 intersects  at P and so the intersection of  1 and  is a line (call it L 1) . Then L1 and L
are on the same plane  , and since L //  , and L1 is on  , L and L1 can not intersect.

Hence there is a line L1 on  through P and parallel to L.

But through P there is a unique line parallel to L and so L1 is unique.

Theorem 5.3.4
If a plane  is parallel to each of two intersecting lines L1 & L2, then  is parallel to the plane
determined by L1 and L2.
Proof:
Let L1 and L2 are two lines intersecting at P and both are parallel to a plane  .
Assume that the plane determined by L1 and L2 (call it  1) intersects  at a point Q. Clearly Q can
not be on L1 or L2 (Why?)
Then through Q there is a unique line L'1 parallel to L1. Then L'1 can not be parallel to L2 and so L'1 is
through Q and parallel to L1 intersecting L2 at a point say R.
Clearly L'1 is on  1. (Why?)

Since Q is on  and L1 //  , by theorem 5.4.3. there is a unique line on the plane  through Q and
parallel to L1.
 L'1 is on  .

 R  L2  
Contradicting that L2 //  .
Hence  , and  intersect is not possible.
  1 //  .
Theorem 5.3.5
A plane containing one and only one of two parallel lines is parallel to the other.
Proof:

205
Let L1 // L2 and L1 is on  but L2 is not on  .
L2 is not on  means either L2 //  or L2   is a single point.
We show that L2   is a single point is impossible.

Let   L2   . Clearly P is not on L1 (Since L1 // L2).

Then trough P there is a unique line L1 parallel to L1 and L'1 is on  .
But already L2 // L1
 L'1 = L2
 L2 is on 
Hence L2   is a single point is not possible
 L2 //  .
Theorem 5.3.6
Given a plane  and a point P not on  , there is a unique plane  , through P parallel to  .
Proof:
Let a plane  and a point P not on  are given.
Then let Q be any point on  ,
Then through Q there at least two lines on the plane  both passing through Q (call them L1 and L2).

Since P is not on  and L1 and L2 are on  , P is not on L1 and not on L2.

Then by parallelity Axiom, there is a unique line L 1 through P parallel to L1. Then  contains L1 but
 does not contain L'1 because P is on L'1 but P is not on  .

Hence by theorem 5.4.5. a plane containing one and only one of two parallel lines is parallel to the
other and so L'1 //  .

Similarly, there is a unique line L'2 through P parallel to L2. Then L'2 is also parallel to  . Clearly ,
L'1 and L'2 are distinct (Why?) lines intersecting at P and both L'1 and L'2 are parallel to  . Then by
Theorem 5.4.4. a plane parallel to each of two intersecting lines is parallel to the plane determined
by the lines and so the plane determined by L' 1 and L'2 (call it  1) is parallel to  and passes
through P.
To show uniqueness, let  1 and  2 be two planes both through P and parallel to  . Then their
intersection is a single line (call it L)

206
Then L is parallel to  . Let R be any point on  . Then there is a unique line L'1 on  through R
and parallel to L. Let also be any other line on the plane  , through R other than L'.

Then since both  1 and  2 are parallel to  and h' is on  , h' is parallel to both  1 and  2.
Then by theorem 5.4.3 there is a unique line h1 on the plane  1 through P parallel to h' and a unique
line h2 on the plane  2 through P and parallel to h'  h1 = h2
But both h1 and h2 are distinct from L.
  1 and  2 intersect at more than a line.
  1 = 2
Hence there is a unique plane  1 through P parallel to  .
Theorem 5.3.7.
Two lines in space that are parallel to the same line are parallel to each other.
Proof:
Let L1 and L2 be lines both parallel to L.
Case 1
If L1 , L2 and L are all in the same plane, we have already shown that L1 // L2.
Case 2
L1 , L2 and L are not on the same plane.
Assume L1 // L, L2 // L. Then by definition of parallel lines L1 and L are on the same plane (call it
 ) and L2 is not on  . Then since L2 // L and  does not contain L2, by Theorem 5.4.5.  is
parallel to L2.
Now let P be any point on L1.
Then P is on  and so by Theorem 5.4.3. there is a unique line L' 1 on the plane  through P and
parallel to L2.
To show L'1 is parallel to L, assume L'1 intersects L at a point R.
Then we have found two lines L' 1 and L both through R and parallel to L 2. But this is not possible.
Hence L'1 and L can not intersect. Since they are on the same plane  , L'1 // L.

But through P there is a unique line parallel to L. But already P is on L1 and L1 // L and so L'1 = L1.
Since L'1 // L2 and L'1 = L1 , we have L1 // L2.

207
 APPENDIX

Answers/Key/ to Self-Test Exercises

Chapter Three ( Section 3.6)

1. True 2. False 3. False 4. True 5. True 6. False 7. True 8. False

9. Refer to the first non-intersecting lines (Figure 3.1)

10. Two. (The right and left sensed parallels)
11. Let  and m be parallel. Take point P on  and points A1 , A2 ,A3, … ,An on
m such that PA1 is perpendicular to m. Draw PA2 , PA3 , …, PAn and determine the
PA3 A1 PA3 A1 PA n A1
angles PA 2 A1 , ,
Observe the n tends to the line PAn approaches to the line 
12. First make a construction. The correct drawing is that lines BA and BC are the two
parallels of  through B. Draw the bisector of ABC and then show that this angle
bisector is the perpendicular from B to  .
13. It is required to prove AD = BC. To do this assume not! (i.e. AD  BC ) and show that
this assumption leads to a contradiction to the hypothesis .
Drawing the diagonals AC and BD is necessary.
14. Draw a Lambert Quadrilateral ABCD with acute angle at D; and show that AD > BC
and DC > AB by referring to Theorem 3.13.
15. Let ABCD be Saccheri Quadrilateral with base AB; and let P and Q be mid points of
AD and BC , respectively. Also, let M and N be mid points of AB and DC. Then, to

show PQ  MN , observe that ABQP is also a Saccheri quadrilateral and refer Theorem
3.11 to proceed.

Chapter Four ( Section 4.8)

1. i) The inverse is a circle through the center of inversion that intersects the circle of
inversion at two points.
ii) The inverse is a circle through the center of inversion and internally tangent to the
circle of inversion.
iii) The inverse is a circle through the center of inversion that is inside the circle of
inversion.

208
2. Suppose two circles C1 and C2 with centers O1 and O2 are orthogonally intersecting
each other at points P and Q. Draw O1P , O2P, and O1O2 . Then by definition
O1 P  O2 P and observe that r = O P and r = O P
1 1 2 2

Then, O1 PO2 is right angle with hypotenuse O O 1 2

Clearly, inverse of O1 with respect to C2, O'1 and that of O2 with respect to C1 , O'2 both
lie on the segment O1O2 .

Then, (O1O2 )(O1O'2 )  r1 and (O2O1 )(O2O'1 )  r2

2 2
(Why?)

 O1O2 [O1O '2  O2O'1 ]  r12  r22  (O1O2 ) 2 (Why?)

 O1O'2  O2O'1  O1O2  O '2  O'1 (Why ?)
3. Since A, B, C, D are collinear points you can use addition of segments by assuming any
one of the orders, for instance A-B-C-D.

DC BA (CD) ( AB) CD AB
( A, B, C , D)  .  .   .
DA BC ( AD) BC AD BC
Then, use
CD  AD  AC and AB  AD  BD
4. With ( A, B, C , D)   , where    2 , apply the result of problem (3) above and that of
Theorem 4.3 .
5. Refer to Section 4.5.2
6. Yes, of course. In this case ps-lines are the diameters.
7. If the ps-line is the diameter, it simply holds from the Euclidean concept of uniqueness of
a perpendicular from a point to a given line. If ps-lines are the orthogonal arcs, use the
concepts in Euclidean circles `How many circles can be drawn orthogonal to both of the
two orthogonal circles?`
8. The ps-point D is the midpoint of the ps-segment AB if
ps-length AD = ps-length DB
AT DS DT BS
.  .
i.e. if AS DT DS BT

9. Ray CR bisects PCQ implies that it bisects the arc PQ in Euclidean sense.
( i.e. R is such that arc PR = arc RQ ) . But, the ps-midpoint is not as such since ps-
length is not the same as arc length.

209
 References

1. Harold E. Wolfe ( ). Introduction to Non-Euclidean Geometry.

New York: The Dryden Press.
2. Walter Meyer (1999). Geometry and Its Applications.
New York: Harcourt Academic Press.
3. Cederberg, Judith N. (1989). A Course in Modern Geometry. New York: Springer Verlang.

210