HUE UNIVERSITY

HUE UNIVERSITY OF EDUCATION

Notes on Projective Geometry

Tran Dao Dong

Department of Mathematics

September 01, 2022

 CONTENTS

Preface .......................................................................................................................... .......3

1. Projective spaces .......................................................................................................................................................... 5

1.1. Projective spaces and planes ........................................................................................ 5

1.2. Projective frames and homogenous coordinates ........................................................... 6

1.3. Equations of projective planes ................................................................................... 10

1.4. The relation position of two planes............................................................................. 12

1.5. Duality......................................................................................................................... 15

1.6. The cross ratio ............................................................................................................. 16

Exercises ............................................................................................................................ 27

2. Projective maps and projective geometry ................................................................. 29

2.1. Projective maps ........................................................................................................... 29

2.2. Projective transformations .......................................................................................... 33

2.3. Projective geometry .................................................................................................... 35

2.4. Projective homologies and radial projections ............................................................. 36

2.5. Projective model of the affine space ........................................................................... 42

2.6. Some applications of projective geometry and affine geometry ............................... 49

Exercises ............................................................................................................................ 60

3. Quadratic hypersurfaces ............................................................................................. 62

n
3.1. Quadratic hypersurfaces in .................................................................................... 62

3.2. Projective classification of quadrics ........................................................................... 64

1
 n n n1
3.3. Quadrics in the model = \ .......................................................................... 66

3.4. Pole and dual of pole................................................................................................... 69

3.5. Tangent hyperplanes .................................................................................................. 72

Exercises ............................................................................................................................ 75

4. Projective plane geometry ........................................................................................... 77

4.1. Theorem of Desargues and Theorem of Pappus. ........................................................ 77

2
4.2. Quadratic curves and conic in . ............................................................................. 79

4.3. Projective maps between two lines and two pencil of lines. ...................................... 81

4.4. Theorem of Steiner ..................................................................................................... 84

4.5. Theorem of Pascal and Theorem of Brianchon .......................................................... 87

4.6. Involution transformations of a conic and a line ........................................................ 95

Exercises .......................................................................................................................... 100

Final exam review ............................................................................................................ 103

References ....................................................................................................................... 106

2
 PREFACE
Projective geometry has its origins in Renaissance Italy, in the development of
perspective in painting: the problem of capturing a 3-dimensional image on a 2-dimensional
canvas. It is a familiar fact that objects appear smaller as the they get farther away, and that
the apparent angle between straight lines depends on the vantage point of the observer. The
more familiar Euclidean geometry is not well equipped to make sense of this, because in
Euclidean geometry length and angle are well-defined, measurable quantities independent of
the observer. Projective geometry provides a better framework for understanding how shapes
change as perspective shifts. The projective geometry most relevant to painting is called the
2 2
real projective plane, and is denoted ( ). By definition, the real projective plane ( ) is
3
the set of 1-dimensional sub-spaces of .
One of the characteristic features of projective geometry is that every distinct pair of
projective lines in the projective plane intersect. This runs contrary to the parallel postulate in
Euclidean geometry, which says that lines in the plane intersect except when they are parallel.
We will see that two lines that appear parallel in a Euclidean plane will intersect at a point at
infinity when they are considered as projective lines. As a general rule, theorems about
intersections between geometric sets are easier to prove, and require fewer exceptions when
considered in projective geometry.
This notes covers both multi-dimensional projective geometry and plane projective
geometry. Based on this, the notes will concentrates to projective properties, invariant properties
under projective transformations. The properties studied in projective geometry is poor (properties
related to measurements, parallelism will not be considered) but the results in projective geometry
are more general than other geometries. The main property in the projective geometry is the
contained relationship. In a certain sense, it can be said that Euclid's geometry is the geometry of
ruler and compass and the projective geometry is the geometry of the ruler only.
The strength of the subject is to help students solve problems concerning to concurrency and
collinearity (especially in plane geometry) in a general way. The theorems related to conic sections
will be very interesting for students when looking back at similar exercises in high school. The
subject also helps students to create new elementary problems through the dual principle, the
projective model of affine spaces,...
The method of study is still the coordinate method. Practices with projective problems will
be caried out on the basis of equations. Students are required to be capable of using the projective
approach in solving and creating problems at elementary levels, thereby having a new outlook
on a number of problems taught at high schools such as problems on properties of the three-conics
based on the projective point of view, problems on colinear points, concurrent lines,...

The notes is organized as follows. Firstly, chapter 1 introduce to projective spaces and
planes, projective frames and homogenous coordinates, consider parametric equation and general
equation of an m-plane, principle of duality, cross-ratio of the four collinear points and of the

3
pencil of four hyperrplanes, complete quadrangles and complete quadrilaterals. Chapter 2 introduce
to projective maps and projective transformations, equation of projective transformations,
projective geometry, fixed points of an projective transformation, projective homologies and radial
projections, projective model of affine spaces, some applications of projective geometry and affine
geometry.

In chapter 3 we introduce to quadratic hypersurfaces, intersetion of a quadratic hypersurface

and a m-plane,canonical form of equation, projective classification of quadric hypersurfaces, apply
for classification of quadratic curves in 2 , quadratic surfaces in 3 and their names, pole and dual
of pole, tangent hyperplanes, dual of quadratic hypersurfaces and Theorem of Mac-Laurin. The last
chapter introduce to Theorem of Desargues and Theorem of Pappus, quadratic curves, projective
maps between lines and concurrent lines, apply for considering the theorem of Steiner, Theorem of
Pascal and Theorem of Bianchon, involution transformations of a conic and a line.

The notes is aimed at providing students with an amount of knowledge relating to concepts
of projection and invariant properties under projective tranformations. Students are expected to
obtain a general view on problems of plane geometry in relation to theorems on conic, line,
collinear points, concurrent lines,... Moreover, the notes will help students have a general point of
view on the exercises in high school levels. This subject is designed in the belief that it helps
develop in students the ability of thinking logically and abstractly, which facilitates the teaching of
geometrical section at high schools in a creative way.

4
 Chapter 1: PROJECTIVE SPACES

This chapter introduce to projective spaces and planes, projective frames and
homogenous coordinates, consider parametric equation and general equation of an
m-plane, principle of duality, cross-ratio of the four collinear points and of the
pencil of four hyperrplanes, complete quadrangles and complete quadrilaterals.

1.1. Projective spaces and planes

1.1.1. Basic definitions
Let n1
be a vector space of dimension n +1 (n ≥ 0) over a field and  be an
arbitrary non-empty set, where is the real numbers or the complex numbers . We
denote P( n1
) the set of 1-dimensional vector subspaces of Vn+1, that is, every element of
P( n1 ) is a subspace 1 of n1 . Then, the triad ( , h, n1 ) is said to be a n–dimensional
n1 n
projective space associated to the vector space and denoted shortly by if there exists
a bijective map h: P( n1 ) →  .

By definition, each point of the projective space n is image of a subspace 1

generated by a vector x that differs vector 0 of 1 through the bijective map h. In the case of
the field is the real numbers , or the complex numbers , the projective space n is
called the real projective space or complex projective space, respectively.

If m1 is a vector subspace of n1 (0 ≤ m ≤ n) then subset h(P( m1 )) of n will be

called a m - dimensional projective plane of the projective space n and denoted by m .
From now, m is called shortly a m-plane. By definition, we have the following:

 m = 0: 0 - plane is an point.
 m = 1: 1 - plane is called a projective line.
 m = 2: 2 - plane is called a projective plane.
 ..........................................................
 m = n-1: (n-1)-plane is called a projective hyperplane.
 m = n: n - plane is just n .
m m1
Suppose that = h(P( )) is an m-plane. Then, there is a restricted bijective map

h': P( m1
) → m

that is induced from the bijective map h: P( n1

)→ n
.
m m1
Hence, ( , h', ) is also a projective space of dimension m.

5
1.1.2. Remark

Let n
be a n–dimensional projective space associated to a vector space n1
and 
is an arbitrary non-empty set. Then, we see that  will become a n–dimensional
projective space associated to n1 if there exists a bijection of n onto  .
Now we consider some concretely "models" of the projective space.
1.1.3. Models of the projective space
a. Let n1 be a vector space. Then the set P( n1 ) of 1-dimensional vector subspaces of
n1
is a projective space associated to n1 and called the "vector model" of the projective
space. In this model, each point of P( n1 ) is a subspace 1 of n1 . Our intuition is best
served by thinking of the real case. In this case, the projective space P ( n1 ) is the set
of rays of n1 .
b. Let V = n1
and  = n1
/~, where ~ is the equivalent relation on n1
defined by x ~ y
iff for all x, y  n1 , there exists a non-zero scalar   such that y =  x. Then, it follows
that the quotient space  = n1 /~ is a projective space associated to n1 and called the
"arithmetic model" of the projective space. In this model, each point of  is an equivalent
class [x], where x  n1 .
c. Let n1 be the affine space of dimension n+1 and suppose that O  n1 . Denote by B(O)
the set of lines passing through O in the space n1 . Then B(O) is a projective space of
dimension n and called the "sheaf model" of the projective space.
d. Let n1 be the affine space of dimension n+1 associated to the vector space n1 and
n
( n ) be a hyperplane in n1 . Denote by P( n ) the set of 1-dimensional vector
subspaces of n . Then n = n  P( n ) is a projective space of dimension n and called the
"extended affine model" of the projective space.

1.2. Projective frames and homogenous coordinates

1.2.1. Representative vectors
Let n
= ( , h, n1
) be a projective space of dimension n. Suppose that X = h([x]) is a
point in defined by a 1-dimensional subspace [x], where x  n1 is a non-zero vector.
n

Then, vector x is called a representative vector for the point X. It follows from the definition
that if x and y are representative vectors for a same point X, then there is a non-zero scalar k
such that y = kx. Hence the representative vectors of a point X are defined up a scale factor
and the correspondence between points and representative vectors is not one-to-one. In other
words, every point X in n defined by a class of representative vectors in which two
arbitrary vectors are collinear.

6
1.2.2. Projective independent points

Let A1, A2,..., Am (m  1) be points in n and a1, a2,..., am  n1 be their corresponding
representative vectors, recspectively, in n1 . We say that points A1, A2,...,Am are projective
independent, or shortly independent iff representative vectors a1, a2,..., am are linearly
independent in n1 . In the case of A1, A2,..., Am are non projective independent, we say that
they are projective dependent. By definition, we see that:
n
+ A point in is independent.
n
+ Two points in are independent if and only if they are distinct points.
n
+ Three points in are independent if and only if they are not collinear.

+ In the case of A1, A2,..., Am are independent, we get that m  n+1. Inversely, every
system of m points, where m > n +1, are always dependent.

Moreover, we have the following assertion:

1.2.3. Theorem: A system of m points A1, A2,..., Am (m  2) in n

is projective independent
if and only if they do not lie in the same (m-2)-plane.

Proof: Suppose that points A1, A2,..., Am are projective independent in n and a1, a2,...,am are
corresponding representative vectors, recspectively, in n1 . Then, by definition, points
A1, A2,..., Am are projective independent in n if and only if representative vectors
a1, a2,..., am are linearly independent in n1 . This means that vectors a1, a2,..., am are not in
the same vector subspace of dimension m-1. In other words, points A1, A2,..., Am are
projective independent if and only if they do not lie in the same (m-2)-plane.
n
1.2.4. Theorem: Given m independent points A1, A2,..., Am in . Then, there exists a unique
projective plane of dimension m-1 that is containing them.

Proof: Suppose that points A1, A2,..., Am are projective independent in n . Then, their
corresponding representative vectors a1, a2,..., am are linearly independent and they define a
unique vector subspace m of dimension m. This means that m1 = h(P( m )) is the unique
(m-1)-plane that is containing A1, A2,..., Am.

1.2.5. Projective frames

Let {A1, A2,..., An+1} be a system of ordered points in n and E  n . Then the ordered set
{A1, A2,..., An+1 ; E} is called a projective frame (or projective basis) of n iff every n +1
points of the ordered set {A1, A2,..., An+1 ; E} are projective independent. Then, the points

7
 n
A1, A2,..., An+1 is called vertices and E is called the unit point. A projective frame of
often denote shortly by R = {Ai ; E}.
Consider m +1 vertices of the frame R = {A1, A2,..., An+1 ; E}. The m-plane (0 < m < n)
passing through given vertices is called the coordinates m-plane. Moreover, suppose that m
is the plane passing through m +1 vertices An-m+1, An-m+2,..., An, An+1.
Then, the following system of equations

x1 = 0, x2 = 0,..., xn-m = 0
m
is called the general equation of . Similarly, we can determine the general equation of
the another coordinates m-planes.
1.2.6. Theorem: For a given projective frame R = {A1, A2,..., An+1 ; E} of n , there exists a
basis {e1, e2, ... , en+1} of n1 such that ei is the representative vector for the vertice Ai, for
all i = 1, 2,..., n +1 and the sum vector e = e1+ e2 + ... + en+1 is the representative vector for
the unit point E.
Proof: Suppose that a1, a2,..., an+1 are representative vectors of vertices A1, A2,..., An+1,
respectively, in n1 . Since points A1, A2,..., Am are projective independent, it follows that
representative vectors a1, a2,...,an+1 are linearly independent in n1 . This means that vectors
a1, a2,..., an+1 constitute a basis of n1 . Let e be a representative vector of the unit point E.
Then we have e =  1a1+  2a2 + ... +  n+1an+1, where  1,...,  n+1  .
Since every n+1 points of the given projective frame R = {A1, A2,..., An+1 ; E} are projective
independent, we obtain that the scalers  1,...,  n+1 are not zero.
Denote e1 =  1a1, e2 =  2a2 ,..., en +1 =  n+1an+1. Then, we can prove that {e1, e2, ... , en+1} is
a basis of n1
such that ei is the representative vector for the vertice Ai,  i = 1, 2,..., n +1
and e = e1+ e2 + ... + en+1.
1.2.7. Remark: Let R = {A1, A2, ... , An+1 ; E} be a projective frame of n . Suppose that
{e1, e2, ... , en+1} and {e'1, e'2, ... , e'n+1} are bases determined by the frame R based on
Theorem 1.2.6. Then, there exists a non-zero scalar  such that e'i =  ei ,  i =1,..., n+1. In
other words, the bases {e1, e2,..., en+1} and {e'1, e'2,..., e'n+1} are different by a homothetic
transformation of n1 .
1.2.8. Homogenous coordinates
Let R = {A1, A2,..., An+1 ; E} be a projective frame of n and {e1, e2, ... , en+1} be a basis of
n1
determined by R. Consider X  n and suppose that x  n1 is a representative vector
for X. Then the coordinates (x1,..., xn+1) of x for the basis {e1, e2, ... , en+1} is called the
homogenous coordinates (or projective coordinates) of X for the projective frame R. We
denote by X (x1,..., xn+1) or X = [x1,..., xn+1]. By definition, it follows that:

8
- The projective coordinates (x1,..., xn+1) of X differs from zero.
- If (x1,..., xn+1) and (x'1,..., x'n+1) are projective coordinates of the same point X in n , then
they are proportional, that is, there exists a non-zero scalar  such that x'i =  xi , for all
i = 1,..., n+1. Based on this property, we often denote X (x1:...: xn+1).
- For a given projective frame R = {A1, A2,..., An+1 ; E}, we can determine projective
coordinates of vertices and the unit point as follows:
A1(1: 0: ...: 0), A2 (0: 1:...: 0), ... , An+1 (0:....: 0: 1) ; E (1: 1: ...: 1).
1.2.9. Fomula of changing homogenous coordinates
Let R = {A1, A2,..., An+1 ; E} and R' = {A'1, A'2,..., A'n+1 ; E'} be projective frames of n .
Denote by {e1, e2, ... , en+1} and {e'1, e'2, ... , e'n+1} for the corresponding bases of R and R',
respectively.
Consider a point X  n and suppose that X(x1:...: xn+1), X(x'1:...: x'n+1) for R and R',
respectively. Denote S = [sij](n+1) for the changing matrix of the bases {e1, e2, ... , en+1} and
{e'1, e'2, ... , e'n+1}. Then we have the following system of equations:
n 1
kxi = s
j 1
ij x 'j , k  0,  i = 1, 2,..., n +1.

This system of equations is called the formula of changing projective coordinates and the
matrix S is called the changing matrix of projective frames R and R'.
Note that the changing matrix S is determined modulo a non-zero scalar, that is, if T is an
another changing matrix of projective frames R and R', there exists a non-zero scalar 
such that S =  T.
1.2.10. Example: Let R = {A1, A2, A3 ; E} and R' = {A'1, A'2, A'3 ; E'} be projective
frames of 2 . Suppose that A'1 (0: 1: 1), A'2 (2: 0: 1) , A'3 (1: 1: 0) ; E' (1: 1: 1) for the
projective frame R. Find the changing matrix of projective frames from R to R'.
Proof: Suppose that {e'1, e'2, e'3 } is the basis in 3 determined by the frame R based on
Theorem 1.2.6. Then, we see that e'1, e'2, e'3 are representation vectors for A'1, A'2, A'3,
respectively, and the sum e' = e'1 + e'2 + e'3 is a representative vector for E'.
0 2 1  1
Hence there exists real numbers a, b, c  such that a  1  + b  0  + c  1  =  1  .
     
1  1  0 1
       
2 1 1
Solve this equation to get a = , b = , c = .
3 3 3
Then, we have the following changing matrix
 2 1
 0
3 3
S =  
 2 2.
0
 3 3
 
 2 1
 0 

 3 3 

9
 We can choose
 0 2 1
T = 3S =  2 0 1.

 2 0 
 1

Based on this, we get the formula of changing homogenous coordinates

kx1 = 2x'2 + x'3 , kx2 = 2x'1 + x'3 , kx3 = 2x'1 + x'2 .

1.3. Equations of projective planes

Let m be a plane of dimension m. We have known that the m-plane m
is well-defined by
m+1 independent points in n , that is, in the space n1 associated to n
, m+1 vectors that
represent for m+1 given independent points are linearly independent.

1.3.1. Parametric equation

Let R = {A1, A2,..., An+1 ; E} be a given projective frame of n . Suppose that m is defined
by m +1 independent points P1 ,..., Pm+1, where Pj (a1j: a2j:...: am+1,j ), j = 1, 2,..., m +1.
Denote by m1 = L(a1, a2,..., am+1), the vector subspace of n1 spanning by a1, a2,....,am+1,
where a1, a2,....,am+1 are the representative vectors for P1,..., Pm+1, respectively. Then, we see
that m is defined by the vector subspace m1 .

Consider X(x1: x2:...: xn+1)  m

and denote by x(x1, x2,..., xn+1)  m1
, the representative
m1
vector for X. Since {a1, a2,....,am+1} is a basis of , there exists scalars t1, ...., tm+1 in
such that

x = t1 a1 +...............+ tm+1 am+1. (1.4.1)

Then, the equation (1.4.1) is equivalent with

x1 = t1 a11 + t2a12 + ...... + tm+1 a1, m+1

x2 = t1 a21 + t2a22 + ...... + tm+1 a2, m+1 (1.4.2)

..... .. .. ........................ .....................

xn+1 = t1 a n+1,1 + t2a n+1, 2 + ...... + tm+1 an+1, m+1.

The above system of equations is said to be the parametric equation of the m - projective
plane m that is defined by m +1 points P1 ,..., Pm+1 and t1, ....., tm+1 is said to be the
parametric variables. Since m +1 vectors a1, a2,..., am+1 are linearly independent, we see that
the rank of matrix A = [aij](n+1) is m +1.

10
Inversely, it follows from a parametic equation of the form (1.4.2) in which the rank of
matrix A = [aij](n+1) is m +1 that there exists a unique m-plane m in n so that its parametric
equation is just the system of equations (1.4.2).

1.3.2. General equation

Now, suppose that m has the parametric equation of the form (1.4.2) for the given projective
frame R of n . Then, we see that the system of equations (1.4.2) contains n+1 equations
with m+1 parametric variables. Since the rank of the matrix A = [aij](n+1) equals to m+1, we
can eliminate m+1 parametric variables from (1.4.2).
Namely, we choose m+1 independent equations in (1.4.2) and solve the chosen system of
equations to find parametric variables t1 , t2 ,..., tm+1 that depend on aij.
Substitute the value of t1 , t2 ,..., tm+1 to the n-m rest equations of (1.4.2) to get the system of
n - m homogeneous linear equations:

c11 x1 + ............................. + c1, n+1 xn+1 = 0

c21 x1 + .............................+ c2,n+1 xn+1 =0 (1.4.3)

.... ..... ..... ..... ...................

cn – m, 1 x1 + ........................+ cn – m, n+1 xn+1 = 0.

m
The equation (1.4.3) is said to be the general equation of the projective plane . It
follows that the rank of matrix C = [cij](n+1) is n - m.

Inversely, assume that one has a system of homogeneous linear equations as (1.4.3)
and the rank of matrix C = [cij](n+1) is n - m. Since this system of equations has n+1
variables and n-m independent equations, it has m+1 linearly independent roots and another
roots are linearly expressed through m+1 those roots. Let m+1 roots be (bi1, bi2,..., bi n+1),
where i = 1, ..., m+1.

Denote by bi (bi1, bi2,......., bi, n+1), forall i = 1, 2, ..., m+1. Then, we get that m+1
vectors b1,..., bm+1 are linearly independent and they span a vector subspace m1 of
dimension m+1. Let m be the m - projective plane associated to the vector subspace m1 .
Then, m is the unique m - projective plane such that its general equation is just (1.4.3).

1.3.3. General equation of a projective hyperplane

Consider a projective hyperplane n1 in n . Then the general equation of n1

consists of one equation and has the following form:

u1x1 + u2x2 + ...........+ un+1xn+1 = 0, (1.4.4)

11
where u1, u2,..., un+1 are not simultaneously zero. The (n+1)-tuple (u1, u2, ...., un+1) is called
the projective coordinates of n1 and we denote n1 = [u1, u2, ....., un+1]. It follows that:

 An (n+1)-tuple (u1, u2,....., un+1), where (u1)2 +.....+ (un+1)2  0 defines a unique
projective hyperplane.
 Two (n+1)-tuples (u1 ,u2 , ......., un+1) and (v1 ,v2 , ....., vn+1) define the same
hyperplane if and only if there is a number k  0 such that ui = [Link] for all i = 1, 2, ...., n+1.
n1
Based on this, we often denote = [u1: u2: .....: un+1].

1.4. The relative position of two planes

In a projective space n , if we want to consider the relative position of two planes then
we need to consider whether they are intersecting or non- intersecting.
n
1.4.1. Definition: Let and be two planes in . Then, we say that

a. and are intersecting iff they have common points, that is, they have non-empty
intersection.

b. and are skew iff they have no intersection.

n
1.4.2. Remark: Let be a projective space of dimnsion n.
n
a. If and are planes in which is intersecting, then the intersection of and is a
plane that defined by the intersection of linear subspaces and , where and are
subspaces defining and , respectively. Moreover, the intersection of two planes is the
plane that has the greatest dimension containing in given two planes.

b. The intersection of all planes in n which containing both planes and is a plane and
it is said to be the sum of two planes and . This implies that the sum of two planes is the
plane that has the least dimension among the planes containing those given planes.

Let and be two planes in n . Suppose that and are the corresponding direction
subspace of and , respectively.

Then, we see that

dim ( + ) = dim + dim - dim ( ∩ ).

Based on this formula, we can determine the dimension of the sum of two planes. In other
words, for two planes and in the projective space n , we have the following assertion:

12
 n
1.4.3. Theorem: Let and be two planes in .

a. If planes and are intersecting, then

dim ( + ) = dim + dim - dim ( ∩ ).
b. If planes and are skew, then
dim ( + ) = dim + dim + 1.

1.4.4. Definition: (The family of independent hyperplanes)

n
Let 1 , 2 , ....., m be projective hyperplanes in defined by
i = [ ui1: ui2: ......: ui, n+1], i = 1, 2,...., m and 0 < m ≤ n +1.
Then the projective hyperplanes 1 , 2 , ....., m is called a family of independent hyperplanes
iff rank [uij]m×(n+1) = m.
n
1.4.5. Theorem: Let 1 , 2 , ....., m be a family of independent hyperplanes in . Then, the
intersection of hyperplanes 1 , 2 , ....., m is a plane of dimension n-m (0 < m < n ).

Proof:
Suppose that i = [ ui1: ui2: ......: ui, n+1] ( i = 1, 2,..., m), where rank [uij]m×(n+1) = m.
Then, we see that X(x1: x2:...: xn+1)  i for all i = 1, 2,..., m if and only if

u11 x1 + .............................+ u1, n+1 xn+1 = 0

u21 x1 + .............................+ u2, n+1 xn+1 = 0 (1.4.5)

.... ..... ..... ..... ...........

u m1 x1 + ...........................+ u m, n+1 xn+1 = 0.

Since rank [uij]m×(n+1) = m, the system of equations (1.4.5) defines a plane of dimension
n-m. It means that the intersection of hyperplanes 1 , 2 , ....., m is a (n-m)-plane.
n
1.4.6. Theorem: Let be a hyperplane and be a plane of dimension m in such that
is not contained in . Then the intersection ∩ has the dimension m - 1.

Proof:

Suppose that the hyperplane has the general equation

u1x1 + u2x2 + ...........+ un+1xn+1 = 0,

and the m - plane has the general equation

13
 a11 x1 + .............................+ a1, n+1 xn+1 = 0

a21 x1 + .............................+ a2, n+1 xn+1 = 0

.... ..... ..... ..... ...........

a n - m,1 x1 + ........................+ a n -m, n+1 xn+1 = 0.

Then, the system of equations for determining ∩ consists of n - m +1 equations in which

the rank of its coefficent matrix equals n – m +1.

Indeed, if the above system of equations has rank n - m then any points of belong to .
This is a contradiction! Since the above system of equations has rank n - m + 1, we see that
it defines a (m-1) - plane.

Based on this theorem, we obtain the following corollary:

1.4.7. Corollary:
a. The intersection of n independent hyperplanes in n is just a point. Especially, two distinct
lines in 2 always intersect at a unique point.
b. Two distinct hyperplanes of the projective space n always intersect at a plane of
dimension n-2.
2
It follows from the general equation and the relative position of lines in , we obtain the

following:
2
1) Let A(a1: a2: a3) and B(b1: b2: b3) be two distinct points in . Then the general equation
of the line d passing through A and B defined by
d = [a2 b3 - a3 b2: a3 b1 - a1 b3: a1 b2 - a2 b1 ].
2
2) Let a = [u1: u2: u3] and b = [v1: v2: v3] be two distinct lines in . Then the intersection
point D of a and b defined by
D (u2 v3 - u3 v2: u3 v1 - u1 v3: u1 v2 - u2 v1).

1.5. Duality
1.5.1. Definition: Let R = {Ai;; E} be a given projective frame in n . Denote by ( n )* the
set of all projective hyperplanes in n and consider a map π : n  ( n )* that is defined by
π (U) = [u1: u2: ...: un+1],
for every U (u1: u2: ...: un+1) in n . Then, we see that the map π is a bijection.
So ( n )* becomes a model of n-dimensional projective spaces. By definition, a point in

14
( n )* is just a hyperlane in n . By this way, we say that "point" and "hyperplane" are two
dual version each other.
In general, we have the following theorem:
n
1.5.2. Theorem: The duality of a m-plane in is just a (n - m -1)-plane.

Proof:
m n
Suppose that is a m-plane in defined by the following general equation:

a11 x1 + .............................+ a1, n+1 xn+1 =0

a21 x1 + .............................+ a2, n+1 xn+1 =0

.... ..... ..... ..... ...........

a n - m, 1 x1 + ........................+ a n -m, n+1 xn+1 = 0.

For each point U(u1: u2: ...: un+1) in n

, denote π (U) = [u1: u2: ...: un+1] the corresponding
hyperplane through the bijection π .
m
Based on the general equation of , we can obtain (n - m)-independent points
P1 (a11:...: a1, n+1), P2 (a21:...: a2, n+1),..., Pn-m (an-m,1:...: an-m, n+1).
n  m 1 n
Denote by the (n- m -1)-plane in defined by (n - m)-independent points P1,..., Pn-m.
n  m 1
Then, we can prove that the (n-m-1)-plane is just the intersection plane of all
hyperplanes π (U), where π (U) mapped by U  m
through the bijection π.
n
In other words, the dual version of a m-plane in is just a (n- m -1)-plane.

1.5.3. Remark:
Let r and s be r-plane and s-plane in n
. Denote by ( r
)* and ( s
)* the dual planes in
( n )* of r and s , respectively.
Then, we see that r  s if and only if ( s
)*  ( r
)*.

1.5.4. The principle of duality

Consider a concept (α) in n that tells about planes and containing relations between them. If
we replace the words "m-planes" of the concept (α) with the words "(n - m -1)-planes", we
obtain a new concept called the dual concept of (α) and denoted by (α)*.
By a similar way, we can obtain the dual version of a given proposition (P). Moreover, we
see that the proposition (P) is true if and only if the dual proposition (P)* is true. Especially,
we get (P) = (P)**. Now, we consider dual versions of some concepts and propositions.

15
1.5.5. Examples:
n
a. The dual version of "collinear points" in is "concurrent hyperplanes".
n
b. The dual version of "independent points" in is "independent hyperplanes".
c. Consider the proposition "Two distinct hyperplanes in the projective space n intersect
in a unique (n - 2)-plane". Then, the dual proposition is "Through any two distinct points in
the projective space n there passes a unique projective line".
1.6. The cross-ratio
1.6.1. The cross ratio of four collinear points
a. Definition:
In the projective space n with a given projective frame R, consider a line d passing
through two distinct points A, B. Choose two points C and D on d such that they are not
coincident with A, B. Let a, b, c, d be representative vectors of four points A, B, C, D,
respectively.
Suppose that the line d has an parametric equation as follows
[X] = λ [A] + μ[B].
Since C, D lie on the line d, there exists non-zero scalars 1 , 1 , 2 , 2 such that:
[C] = λ1[A] + μ1[B];
[D] = λ2[A] + μ2[B].
 
Then, the ratio 1 : 2 is said to be a cross ratio (or double ratio) of the four collinear
 1 2
1 2
points A, B, C, D and denoted by (ABCD) = : .
1 2
In the case of (ABCD) = -1, we say that the two points A, B are harmonic division two
points C, D. Then, the pair C, D is often said to be harmonically conjugate with the pair A,
B and vice versa.
b. Remark:
i) The coefficients i , i can be replaced by k i , k i , where k , i , i  0 , i = 1, 2.
n
ii) The cross ratio does not depend on the choice of projective frames in .

Proof:

i) It follows from the definition of the cross ratio.

ii) Consider a new projective frame and let S be the changing matrix from the old frame
to the new one. Then, we get:

16
 [A'] = S[A] ; [B'] = S[B] ; [C'] = S[C] ; [D'] = S[D] ;

Substituting to the expression of definition, we get

[C'] = S[C] = λ1 S[A] + μ1S[B] = λ1[A'] + μ1[B']

[D'] = S[D] = λ2 S[A] + μ2S[B] = λ2[A'] + μ2[B'].

1 2
Therefore, we get (A'B'C'D') = : with respect to the new projective frame, that is
1 2
n
the value of cross ratio does not depend on the way how to choose projective frames in .

c. Properties: (Permutation of cross ratio)

By definition, we get the following expressions, where A, B, C, D are the four collinear
points in n .

1. (ABCD) = (BADC) = (CDAB) = (DCBA).

2. (ABCD) = (BACD)-1 = (ABDC)-1.
3. (ABCD) = 1 - (ACBD) = 1 - (DBCA).
4. (ABCC) = 1 ; (AACD) = 1.
5. (ABAD) = (ABCB) = 0 ; (ABBD) = (ABCA) =  .

Proof.
1. (ABCD) = (BADC) = (CDAB) = (DCBA):
1 2
One has, [C] = λ1[A] + μ1[B] ; [D] = λ2[A] + μ2[B]. Then (ABCD) = :  a:b .
 1 2
It follows that, [D] = μ2[B] + λ2[A] ; [C] = μ1[B] + λ1[A].
  1 1
Then, (BADC) = 2 : 1  :  a : b = (ABCD).
2 1 b a

Moreover, we get [C] = λ1[A] + μ1[B] = λ1[A] + 1 ([D] - λ2[A]), or equivalently,
2
 12  1    
[C] =  1   [A] + [D] = 1 2 1 2 [A] + 1 [D].
 2  2 2 2
2 1
So, [A] = [C] + [D].
12  2 1 12  2 1
2
In other hands, [D] = λ2[A] + μ2[B] = ([C] - μ1[B]) + μ2[B], or equivalently,
1
2        
 D  C     2  2 1   B   2 C   2 1 1 2  B  .
1  1  1 1

17
 2 1
So,  B    C    D .
21  12 21  12
1 1

12  12 21  12     1
Then, (CDAB) = :  1 : 1  1  2  a   a : b = (ABCD).
2 2 2 2 2 1 b

12  12 21  121
1 2
Finally,  B    D  C  ;
21  12 21  12
1 2
 A    D  C  .
12  12 12  12
2 2

21  12 12  12    
Then, (DCBA) = :  2 : 2  2  1  a : b = (ABCD).
1 1 1 1 1 2

21  12 12  12

Hence, (ABCD) = (BADC) = (CDAB) = (DCBA).

2. (ABCD) = (BACD)-1 = (ABDC)-1:

One has, C   1  A  1  B   1  B   1  A ;

 D  2  A  2  B   2  B   2  A.
1 2 1 1 b 1
Then, the cross ratio (BACD) = :  :    (ABCD)-1 or equivalently,
1 2 a b a a : b
(ABCD) = (BACD)-1.

Similarly, we get (ABCD) = (ABDC)-1.

3. (ABCD) = 1 - (ACBD) = 1 - (DBCA):

1   1
One has,  B   C   1  A   1  A  C .
1 1 1 1
 1 1  2 1  12 
So,  D   2  A  2  B   2  A  2    A   C     A  2  C  .
 1 1  1 1
1 2
  
Then, 1 : 1

1
:
2

2 1  12
 1   1 : 2   1   a : b.
1 2 1  12 1 2 1  12 12  1 2 

1 1
Thus, (ABCD) = 1 - (ACBD). Similarly, we get (ABCD) = 1 - (DBCA).

18
4. (ABCC) =1 and (AACD) = 1:

The truth that D is coincident with C implies that [C] = λ1[A] + μ1[B] = λ2[A] + μ2[B],
 
so 2  1 and 2  1 , or equivalently, 1 : 2  1 =(ABCC). Similarly, we get (AACD)= 1.
 1 2

5. (ABAD) = (ABCB) = 0 ; (ABBD) = (ABCA) =  :

One has A is coincident with C, so we get [C] = λ1[A] + μ1[B] = [A]. Thus 1  0, 1  1 ,
that is, (ABAD) = 0. Similarly, we obtain (ABCB) = 0.

One has C is coincident with B, we have [C] = λ1[A] + μ1[B] = [B], so 1  1, 1  0 . Thus
(ABBD)=  . Similarly, we get (ABCA)=  .
It indicates that as C approaches A, also, as C approaches B, the numerical value of
(ABCD) increases without limit.
Conversely, if you see that the value of cross ratio gets one of 0, 1,  , then some two of
points are coincident.
1.6.2. The cross ratio of a pencil of four hyperplanes
a. Definition:
In the projective space n with a given projective frame R, a set of hyperpalnes
hyperplanes passing through (n-2)-plane is called a pencil of hyperplanes, and the
corresponding (n-2)-plane is called the support of this pencil.
A pencil of hyperplanes is determined by two distinct hyperplanes of this pencil.
Let and be two arbitrary distinct hyperplanes of the pencil of hyperplanes, and their
coordinates with respect to a given projective frame be, respectively:
  p1 : p2 :...: pn1 
  q1 : q2 :...: qn 1  .
Then, the support of the pencil of hyperplanes is just the intersection  so that it is
defined by the system of following equations:
 p1 x1  p2 x2  ...  pn 1 xn 1  0
 .
 q1 x1  q2 x2  ...  qn 1 xn 1  0
Suppose that the equation of an arbitrary hyperplane of this pencil is
r1 x1  r2 x2  ...  rn 1 xn 1  0 .
Then, the truth that passes through the intersection  follows that ri  pi  qi ,
where i  1, 2,..., n  1 and (λ, µ) ≠ (0, 0).
Therefore, for each (λ, µ) ≠ (0, 0), we will determine a hyperplane of the pencil, and
obviously, the pairs (λ, µ) and (kλ, kµ), where k ≠ 0, determine the same hyperplane of the

19
pencil of hyperplanes. Let denote by [ ], [ ], [ ] the coordinates column matrices of the
hyperplanes , , , respectively, we get that [ ]  [ ]  [ ] .
b. Definition:
n
In the projective space with a given projective frame, consider a pencil of hyperplanes
defined by two given distinct hyperplanes and . Suppose that and are two
hyperplanes of the pencil such that they are not coincident with hyperplanes , . Since
and belong to the pencil defined by and , there exists non-zero scalars 1 , 1 , 2 , 2
such that:
[ ]  1[ ]  1[ ]

[ ]  2 [ ]  2 [ ] .
1 2
Then, the ratio : is called the cross ratio of the pencil of four hyperplanes
 1 2
1 2
, , , and denoted by ( ) : .
 1 2
In the case of ( ) = -1, we say that the two hyperplanes , are harmonic
division two hyperplanes , . Then, hyperplanes , , , is often said to constitute a
harmonic pencil.
By definition, the cross ratio of the pencil of four hyperplanes has the same properties as
the case of cross ratio of the four collinear points.

Moreover, there is a nice relationship between the cross ratios of collinear points and
pencil of hyperplanes as follows:

d. Theorem:
In the projective space n , consider a pencil of four hyperplanes , , , and a line d
which does not intersect the support of the pencil of hyperplanes. Let A, B, C, D be the
intersection points of the line d and the hyperplanes
, , , , respectively. Then, we get that

( )   ABCD  .
Proof:

We choose a projective frame R = {Ai; E} such

that A1  A , A2  B , and A3 , A4 ,..., An1 belong to

20
the support of the pencil of hyperplanes. Then, consider the hyperplane passes through all
vertices of the projective frame except A2 , we see that has the equation x2  0 . In other
words, has the coordinates [0: 1: 0: …: 0].

Similarly, the hyperplane has the equation x1  0 , and the coordinates [1: 0: 0:…: 0].

Suppose is another hyperplane of the pencil ( , ) determined by the pair (λ1, µ1), that is
[ ]  1[ ]  1[ ] .

It follows that has the coordinates  1 : 1 : 0 : ... : 0 , and the equation 1 x1  1 x2  0.

Consider the intersection point C of the line AB and the hyperplane of the pencil
( , ). Suppose that the parametric equation of the line AB is [ X ]  t1[ A]  t2 [ B ] , this
implies that X (t1 : t2 : 0 : ...: 0) .

Then the coordinates of the point C satisfy the following equation:

1t1  1t2  0 .

We can choose t1  1 , t 2  1 . Then, it follows that C  1 : 1 : 0 : ... : 0  .

Thus, C  AB  , or equivalently, [C] = -λ1[A] + µ1[B]. (1.6.1)

Let D be the intersection point of the line AB and the hyperplane .

Similarly, we get

[D] = - λ2[A] + μ2[B], where [ ]  2 [ ]  2 [ ] (1.6.2)

From (1.6.1) and (1.6.2), we obtain:

1 2
(ABCD)  : ( ).
 1 2

A special case of the cross ratio of a pencil of hyperplanes is the cross ratio of a pencil of
lines in 2 .
Let a,b,c,d be the four concurrent lines in 2 , and O be their common point.
Suppose that [a] = λ1[c] + µ1[d], and [d] = λ2[c] + µ2[d], where c and d are distinct.
 
By definition, we have (abcd)  1 : 2 is the cross ratio of the lines a,b,c,d.
 1 2

21
 Note that the cross ratio of the four concurrent lines has the same properties as the case
of cross ratio of the four collinear points. Moreover, based on the above theorem, we have
the following corollary:
h. Corollary:
Let O be a point in 2 . Suppose that d and d' are two lines in 2 not passing through O.
If four points A, B, C, D on d are projected by the eye point O to four points A', B', C', D' on
d', then the corresponding cross ratios satisfy (ABCD) = (A'B'C'D').

This corollary justifies another concept. We can assign to any quadruple of lines which pass
through one point O just one cross ratio. We can assign this cross ratio by the following
way: We cut the four lines by an arbitrary line l. The four points of intersection define a
cross ratio. We know that the value of this cross ratio is independent on the specific choice
of l. Thus, we can consider the cross ratio defined above as the cross ratio of the lines. This
fact is nothing else, but a consequence of the fact that every concept in projective geometry
must have a reasonable dual. So, if one can assign a cross ratio to four points on a line, one
must also be able to assign a cross ratio to four lines through a point.

This corollary demonstrates how cross ratios stay invariant under geometric
projections.

Example:
2
Let a,b,c,d be four distinct lines in and r, s be two lines intersecting a,b,c,d at points

A, B, C, D, and M, N, P, Q, [Link] we have (ABCD) = (MNPQ).

.
In other words, sets of four collinear points present the same cross ratios by perspective
projections.

22
 1.6.3. Complete quadrilaterals and complete quadrangles
a. Definition:
A complete quadrilateral or a complete four-line in 2 is a set of any four lines such that
no three of which are concurrent and the six points of the intersection of these lines.
Each line is a side or an edge (there is 4 sides); each intersetion point of two sides is a
vertex (there is 6 vertices). Two points which are not in a same side are called opposite
points (there is 3 pairs of opposite points). Each line connecting two opposite points is a
diagonal (there is 3 diagonals). Each intersection point of two diagonals is a diagonal point
(there is 3 diagonal points).

b. Example:

ACBDEF is a complete quadrilateral

AB, BC, CD, DA: edges

A, B, C, D, E, F: vertices

BD, AC, EF: diagonal lines (diagonals)

I, H, K: diagonal points

c. Theorem:
Consider a complete quadrilateral in 2 . Then, on each its diagonal line, two opposite
vertices along with corresponding two diagonal points are harmonically conjugate.

Proof.

Consider a complete quadrilateral AA’BB’CC’, where the diagonals are AA’, BB’, CC’. Let
E and F be the diagonal points which lie on the diagonal CC’, we need to prove that
 CC ' EF   1. Firstly, we choose a projective frame {A, B, B'; A'}, where A (1: 0: 0),
B (0: 1: 0), B'(0: 0: 1) and A' (1: 1: 1). One has, C  AB  A'B' , so C (1: 1: 0).

Similarly, C'  AB' A' B , so C' (1: 0: 1). Therefore,

CC '  [1:1:1]  CC '  [1:1:1] 

  CC ' AA '  E (2 :1:1) ;   CC ' BB '  F (0 :1: 1) .
AA '  [0 : 1:1]  BB '  [1: 0 : 0] 

23
 To calculate the cross ratio (CC’EF), we suppose that:

[E] = λ1[C] + μ1[C’],

[F] = λ2[C] + μ2[C’] .

Substituting the coordinates of the points C, C', E, F into above expressions, we get:

2 1  1 
1     0    1    1  1 .
  1   1     1
1  1  0   1

0 1 1 
 1    0   1  2  1
  2  2   .
 2  1
 1 1 0

1 2 1 1
Hence, (CC’EF)  :  : = -1. Similarly, we can prove (AA’DE)= -1, (BB’DF)= -1,
1 2 1 1
where D is the diagonal point that lies on the diagonal BB'.

d. Definition:

A complete quadrangle or complete four-point is

a set of any four points such that no three of which
are collinear and the six lines connecting each pair
of these points.

The four points are called vertices, and the six

lines are called sides of the quadrangle. Two sides
of a complete quadrangle which is not passing
through the same vertex is said to be opposite sides.

24
The diagonal point is a point at which pairs of opposite sides intersect. The diagonal line is a
line connecting each pair of diagonal points. The set of diagonal points is a triangle, it is
called a diagonal triangle.

e. Theorem: Consider a complete quadrangle in 2 . Then at each its diagonal point, two
opposite sides along with corresponding two diagonal lines are harmonically conjugate.
Proof:

Consider a complete quadrangle ABCD, where P, E, F are the diagonal points. Firstly, we
prove that (FA FC FP FE) = -1.

Choose a projective frame D, A, C ; B , where D (1: 0: 0), A (0: 1: 0), C (0: 0: 1), B (1; 1: 1).

One has, AB = [1: 0: -1], DC = [0: -1: 0], AC = [1: 0: 0],

AD = [0: 0: -1], BC = [1: -1: 0], BD = [0: 1: -1].

Since F is the intersection point of the lines AD and BC, the coordinates of F satisfy the
following equation:

  x3  0 x 0
  3 . We choose F (1: 1: 0).
 x1  x2  0  x1  x2

Similarly, the lines DC and AB intersect at the point E, so the coordinates of E satisfy the
following equation:

  x2  0  x2  0
   . We choose E (1: 0: 1).
 x1  x3  0  x1  x3

The point P is the intersection point of the lines AC and BD, so the coordinates of P
satisfy the following equation:

 x1  0  x1  0
   . We choose P (0: 1: 1).
 x2  x3  0  x2  x3

Thus, we obtain: FA = [0: 0: 1], FC = [1: -1: 0], FP = [1: -1: 1], FE = [1: -1: -1].

We observe that FP = [1: -1: 1] = 1.[0: 0: 1] + 1.[1: -1: 0];

FE = [1: -1: -1] = -1.[0: 0: 1] + 1.[1: -1: 0].

1 1
Then, (FA FC FC FE) = .  1. Similarly, (PD PB PA PC) = (EA EC EP EF) = -1.
1 1

25
Exercises:
Ptroblem 1. Prove that the set consists of affine hyperplanes passing through a given
fixed point I in the affine space n can be constructed to become a projective space of
dimension n.
Problem 2. In the Euclidean space n with a given orthonormal frame, consider a
hypersphere (S) defined by the following equation

x12  x22  x32 ....  xn2  1.

Prove that the set consists of antipodal points of (S) can be constructed to become a
project space of dimension n.

Problem 3.
a. Given two distinct points [u], [v]  P( n1
). Prove that there is a unique projective
line containing them both.
3
b. Prove that two distinct projective lines in a projective plane P( ) intersect at a unique
point.
Problem 4. Let A, B, C, A', B', C' be distinct points in a projective plane 2 such that t h e
lines AA', BB', C C ' are distinct and the three points of intersection X = AB ∩ A'B',
Y = BC ∩ B'C' and Z = AC ∩ A'C' are collinear. Prove that the lines AA', BB', C C '
are concurrent.
Problem 5. Let α, β, γ, α', β', γ' be distinct lines in a projective plane 2 such that the
intersection points α ∩ α', β ∩ β' and γ ∩ γ' are distinct and the lines joining α ∩ β to α'
∩ β', α ∩ γ to α' ∩ γ' and β ∩ γ to β' ∩ γ' are concurrent. Prove that the intersection
points α ∩ α', β ∩ β' and γ ∩ γ' are collinear.
Problem 6. In the projective plane 2 , let a, b, c be three lines that are concurrent at a point
O and a’, b’, c’ be three lines that are concurrent at a point O’ such that the intersections of
pairs of lines a and a’, b and b’, c and c’ are collinear. Denote A = a  b’, A’ = a  b’,
B = b  c’, B’ = b’  c. Prove that the lines OO’, AA’, BB’ are concurrent. State the dual
problem of the given problem.
Problem 7. In the projective plane 2 , let A, B, C be three points that lie on a line d and A’,
B’, C’ be three points that lie on a line d’ such that the lines AA’, BB’, CC’ are concurrent.
Prove that the intersections of pairs of lines d and d’, AB’ and A’B, BC’ and B’C are
collinear. State the dual problem of the given problem.

26
Problem 8. In the projective plane 2 with a given projective frame R = {A1 A2, A3; E},
consider points B1 (0: 1: -1), B2 (-1: 0: 1), B3 (1: 1: 0), E’(1: 1: 1).
a. Prove that R’ = {B1, B2, B3 ; E’} is a projective frame in 2
.
b. State the changing formular of coordinates from the frame R to the frame R'.

Problem 9. In the projective plane 3 with a given projective frame R = {A1 A2, A3; E}, let
A (1 : 1: 0 : -3), B (1 : 0 : 0 : -1) và C (2 : 0 : 1 : 1). Prove that points A, B, C are
projective independent and state the equation of a plane () passing through A, B, C.

Problem 10. In the projective plane 3 with a given projective frame, consider a line
(d): x1 = t1, x2 = t2, x3 = 0, x4 = - t1 - 2t2 and a point P (2 : 0 : 1 : 1).
a. Determine the general equation of (d).
b. State the equation of a plane () passing through P and containing the line (d).

Problem 11. In the projective plane 3 with a given projective frame, consider points
A (1 : 1 : 2 : 1), B (1 : 0 : 0 : -1), C (2 : 0 : -1 : 0), D (3 : 1 : 0 : 0).
a. Prove that A, B, C, D is a system of projective independent points.
b. Consider the relative position of lines AB and CD.
Problem 12. In n with a given projective frame R = {Ai ; E}, let be the m-plane passing
through m +1 vertices A 1 ,…, A m1 and be the (n-m-1)- plane passing through vertices
A m 2 ,…, A n1. Determine the general equations of and and calculate dim( + ).
Problem 13. In the projective plane 2 with a given projective frame, consider four lines d1,
d2, d3, d4 defined by the following equations, respectively:
2x1 – x2 + x3 = 0; 3x1 + 2x2 + x3 = 0; x1 + 3x2 = 0; 7x1 + 3x3 = 0.
a. Prove that lines d1, d2, d3, d4 are concurrent and determine their common point.
b. Calculate the cross - ratio (d1 d2 d3 d4).

Problem 14. In the projective plane 2 with a given projective frame, consider four points
A (1: 2: 0), B (2: 1: 1), C (5: 4: 2) and D (4: 5: 1).
a. Prove that four points A, B, C, D are collinear and calculate the cross-ratio (ABCD).
b. Fine a point M such that the cross-ratio (ABCM) = -1.
Problem 15. In the projective plane n consider distinct five points A, B, C, D, E that are
collinear. Prove that (ABCD).(ABDE).(ABEC) = 1.

27