Assignment 2

MATH3900  Geometry and Topology

Topology
1. In lectures we have seen that the projective plane can be visualised by drawing a 2-sphere S 2
in three-dimensional space and identifying antipodal points. Which surface do you get when
you identify antipodal points on the torus? (Here, the torus is regarded as a surface, and not a
three-dimensional solid, as we have done in classes.)
(a) Draw a suitable PIE which represents your surface
Solution. To see this, imagine a at torus sitting inside three-dimensional space. We can
parametrize the torus using two coordinates, say u and v , so that each point on the
torus is given by (cos u, sin u, cos v, sin v) . Now, identify opposite points on the torus, which
corresponds to the identication (u, v) (u + π, v + π) . This gives us a new surface which
can be visualized by taking the quotient of the torus by this equivalence relation. To draw
a suitable PIE for this surface, we can start with a rectangle representing the fundamental
domain of the torus, and then identify opposite sides of the rectangle to get a torus. Finally,
we identify opposite points on the torus to get the projective plane, which can be represented
as a rectangle with opposite sides identied as shown below:

(b) Use the techniques that we have encountered in this unit to identify this surface, up to
homeomorphism as usual.
Solution. To see that the resulting surface is indeed the projective plane, we can use the
fact that the projective plane can be obtained by taking the quotient of the unit sphere in
three dimensions by the equivalence relation (x, y, z) (−x, −y, −z) , where (x, y, z) are non-
zero vectors in three dimensions. To see this, imagine drawing a sphere in three-dimensional
space and considering pairs of antipodal points. Each pair of antipodal points corresponds
to a line passing through the origin, and every line passing through the origin intersects the
sphere in exactly two antipodal points. Hence, we can identify antipodal points on the sphere
to get a new space which we call the projective space. Now, consider the torus sitting inside
three-dimensional space, which we can parametrize using (cos u, sin u, cos v, sin v) . We can
embed the torus inside the unit sphere by considering the map, which sends each point on
the torus to a point on the unit sphere. Note that this map is injective since the only points
on the torus that get mapped to the north pole of the sphere are those with v = 0 , and
these correspond to the same point on the torus. Now, if we identify opposite points on the
torus as described above, this induces an identication of points on the sphere. Specically,
we have (cos u cos v, sin u cos v, sin v) (− cos u cos v, − sin u cos v, − sin v) , which corresponds
to the equivalence relation (x, y, z) (−x, −y, −z) on the unit sphere. Hence, we see that
identifying antipodal points on the torus gives us the same space as identifying antipodal
points on the unit sphere, which is the projective plane. Therefore, the resulting surface is
homeomorphic to the projective plane.

1
2. Consider the following PIE representing a surface:

Carry out the following tasks:

(a) Determine the number of holes in the surface
(b) Determine whether or not the surface is orientable.
(c) Determine the surface's Euler characteristic
(d) Express the surface as a sum of disks and projective planes, or as a sum of disks and toruses.
Solution.

(a) To determine the number of holes in the surface, we count the number of handles, which
are the regions bounded by circles in the PIE. In this case, there are four handles, so the
surface has four holes.
(b) To determine if the surface is orientable, we can look at the way the bands twist around
each other. If we can continuously deform the surface so that the bands no longer twist
around each other, then the surface is orientable. Otherwise, the surface is non-orientable.
In this case, we see that the bands cannot be deformed to remove the twisting, so the surface
is non-orientable.
(c) To nd the Euler characteristic of the surface, we use the formula V − E + F = χ ,
where V is the number of vertices, E is the number of edges, F is the number of faces, and
χ is the Euler characteristic.
Counting in the PIE, we see that there are 4 edges, and 1 face. We can use the following
picture to nd the number of vertices:

There are 4 vertices.

Therefore, the Euler characteristic is:
χ=V −E+F =4−4+1=1
(d) To express the surface as a sum of disks and toruses (or projective planes), we can see
that χ = 1 = 2 − 1 , so we have 1 torus and
Surface = Torus + 2 Projective planes
2
3. Show that the following graph is planar:

Without changing the vertex-set, what is the maximum number of edges that we can add while
still retaining planarity? Give justications for your answer.
Solution. A planar graph is a graph that can be drawn on a plane in such a way that its edges
intersect only at their endpoints. In other words, no two edges cross each other in the plane.
We can establish a one-to-one correspondence of the given graph with the following planar graph

We can add only two edge, since when adding already three edges, this engrave will turn into
the complete bipartite graph K3,3 .
Geometry
1. In the 2-dimensional plane R2 , consider the points A(1, 0) , B(2, 0) , C(0, 3) , D(2, 2 + m) ,
E(3, 4 + m) and F (−2, 9 + m) where m is a real number.
(a) Find the ane transformation T which maps A to D, B to E and C to F.
Solution. The general formulas of ane transformation T are
x′ = a1 x + b1 y + c1 , y ′ = a2 x + b2 y + c2

If the transformation T maps A to D , B to E and C to F we have the next system




a1 + c1 = 2,
a2 + c2 = 2 + m,


(1)

2a1 + c1 = 3,

3b1 + c1 = −2,




3b + c = 9 + m.
2 2

The solution of the system (1) is a1 = 1, a2 = 2, b1 = −1, b2 = 3, c1 = 1, c2 = m . As a

result we have:
3
 (
x′ = x − y + 1,
y ′ = 2x + 3y + m
(2)
(b) Find the inverse transformation of T if it exists.
Solution. The transformation T has the inverse one because the determinant formed by
the coecients near variables isn't equal to zero ( it equals to 5).We can use the system 2
to express x and y and we have
(
y ′ − 2x′ = 5y + m − 2,
3x′ + y ′ = 5x + 3 + m,
(3)
so the inverse transformation is
(
x = 53 x′ + 51 y ′ − m 3
5 − 5,
2 ′ 1 ′
y = − 5 x + 5 y − 5 + 52 .
m

2. Given that A, B, C and D are four collinear points in the projective plane, show that the two
cross ratios R(A, B; C, D) and R(C, D; A, B) are equal.
Solution. For mutually dierent collinear points of projective plane are the points of some
projective line. Let the points as points of a projective line have, correspondingly, projective
coordinates A(a1 ; a2 ) , B(b1 ; b2 ) , C(c1 ; c2 ) , D(d1 ; d2 ) according some projective coordinate
system of the line. As it known ,
a1 a2 a1 a2
c1 c2 d1 d2
R(A, B; C, D) = : =
c1 c2 d1 d2
b1 b2 b1 b2

c1 c2 d d2 c1 c2 d1 d2
− − 1
a1 a2 a1 a2 a1 a2 a1 a2
= : = : = R(C, D; A, B)
b b2 b b2 b1 b2 b1 b2
− 1 − 1
c1 c2 d1 d2 c1 c2 d1 d2

3. Consider a triangle ABC , with a line m intersecting the segments AB, AC at R, Q, and
intersecting the extension of BC at P . Assume that AR = 3 , RB = p , BC = 8 , CP = 4 ,
CQ = q and QA = 4 as in the picture below.

(a) Show that 4p = 9q .

4
Solution. Menelaus' theorem shows the ratio of the sides of a triangle, which is obtained when
a line intersects two sides of the triangle and the extension of the third. More formally:
BP/P C ∗ CQ/QA ∗ AR/RB = 1
or 12 q 3
∗ ∗ =1
4 4 p
So 4p = 9q .
(b) If the circumference of the triangle ABC is 28 , nd p and q .
Solution. As 4p = 9q so PABC = AB + AC + BC = p + 3 + 8 + 4 + q = 15 + p + q = 28
p + q = 28 − 15, 4p = 9q

9q
+ q = 13
4

13q
= 13, q = 4
4
So q = 49 p = 4
9 ∗4= 16
9 .