IMOMATH - Projective Geometry

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10
At a glance
Powered by AI
The passage discusses various concepts in projective geometry including the cross ratio, harmonic conjugates, perspectivity, projectivity, Desargues' theorem, Pappus' theorem, and Pascal's theorem.

The cross ratio is defined for four collinear or cocyclic points. It is a projective invariant that is preserved under transformations like inversions. Harmonic conjugates are pairs of points with a cross ratio of -1.

Desargues' theorem states that two triangles are perspective with respect to a center if and only if they are perspective with respect to an axis.

c

2007 The Author(s) and The IMO Compendium Group

Projective Geometry
Milivoje Lukić

Contents
1 Cross Ratio. Harmonic Conjugates. Perspectivity. Projectivity . . . . . . . . . . . . 1
2 Desargue’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3 Theorems of Pappus and Pascal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
4 Pole. Polar. Theorems of Brianchon and Brokard . . . . . . . . . . . . . . . . . . . 3
5 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
6 Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

1 Cross Ratio. Harmonic Conjugates. Perspectivity. Projectivity


Definition 1. Let A, B, C, and D be colinear points. The cross ratio of the pairs of points (A, B) and
(C, D) is

→ −→
AC AD
R(A, B;C, D) = −
→ : −→ . (1)
CB DB

Let a, b, c, d be four concurrent lines. For the O


given lines p1 and p2 let us denote Ai = a ∩ pi , D2
C2
Bi = b ∩ pi , Ci = c ∩ pi , Di = d ∩ pi , for i = 1, 2. A2 B2
Then
A1 C1 B1 D1
R(A1 , B1 ;C1 , D1 ) = R(A2 , B2 ;C2 , D2 ). (2)

Thus it is meaningful to define the cross ratio


of the pairs of concurrent points as O2

R(a, b; c, d) = R(A1 , B1 ;C1 , D1 ). (3)


O1
Assume that points O1 , O2 , A, B, C, D belong
to a circle. Then
D
R(O1 A, O1 B; O1C, O1 D)
A
= R(O2 A, O2 B; O2C, O2 D). (4) B
C

Hence it is meaningful to define the cross-ratio for cocyclic points as

R(A, B;C, D) = R(O1 A, O1 B; O1C, O1 D). (5)


2 Olympiad Training Materials, www.imomath.com

Assume that the points A, B, C, D are colin- D


ear or cocyclic. Let an inversion with center O
maps A, B, C, D into A∗ , B∗ , C∗ , D∗ . Then B
D∗
B∗
R(A, B;C, D) = R(A∗ , B∗ ;C∗ , D∗ ). (6) C∗ C
O A∗ A

Definition 2. Assume that A, B, C, and D are cocyclic or colinear points. Pairs of points (A, B) and
(C, D) are harmonic conjugates if R(A, B;C, D) = −1. We also write H (A, B;C, D) when we want
to say that (A, B) and (C, D) are harmonic conjugates to each other.
Definition 3. Let each of l1 and l2 be either line or circle. Perspectivity with respect to the point S
S
∧ , is the mapping of l1 → l2 , such that
=

(i) If either l1 or l2 is a circle than it contains S;


(ii) every point A1 ∈ l1 is mapped to the point A2 = OA1 ∩ l2 .
According to the previous statements perspectivity preserves the cross ratio and hence the har-
monic conjugates.
Definition 4. Let each of l1 and l2 be either line or circle. Projectivity is any mapping from l1 to l2
that can be represented as a finite composition of perspectivities.
Theorem 1. Assume that the points A, B, C, D1 , and D2 are either colinear or cocyclic. If the
equation R(A, B;C, D1 ) = R(A, B;C, D2 ) is satisfied, then D1 = D2 . In other words, a projectivity
with three fixed points is the identity.
Theorem 2. If the points A, B, C, D are mutually discjoint and R(A, B;C, D) = R(B, A;C, D) then
H (A, B;C, D).

2 Desargue’s Theorem
The triangles A1 B1C1 and A2 B2C2 are perspective with respect to a center if the lines A1 A2 , B1 B2 ,
and C1C2 are concurrent. They are perspective with respect to an axis if the points K = B1C1 ∩ B2C2 ,
L = A1C1 ∩ A2C2 , M = A1 B1 ∩ A2 B2 are colinear.

Theorem 3 (Desargue). Two triangles are perspective with respect to a center if and only if they
are perspective with respect to a point.

3 Theorems of Pappus and Pascal


Theorem 4 (Pappus). The points A1 , A2 , A3 belong to the line a, and the points B1 , B2 , B3 belong
to the line b. Assume that A1 B2 ∩ A2 B1 = C3 , A1 B3 ∩ A3 B1 = C2 , A2 B3 ∩ A3 B2 = C1 . Then C1 , C2 , C3
are colinear.
Proof. Denote C2′ = C1C3 ∩ A3 B1 , D = A1 B2 ∩ A3 B1 , E = A2 B1 ∩ A3 B2 , F = a ∩ b. Our goal is to
prove that the points C2 and C2′ are identical. Consider the sequence of projectivities:

A3 B1 DC2′ .
A1 A2 C3
A3 B1 DC2 =
∧ FB1 B2 B3 =
∧ A3 EB2C1 =

We have got the projective transformation of the line A3 B1 that fixes the points A3 , B1 , D, and maps
C2 to C2′ . Since the projective mapping with three fixed points is the identity we have C2 = C2′ . 
Milivoje Lukić: Projective Geometry 3

B1 B3
B1 B2
B2 C3 D
D C2
C3
C1
B3
C2 C1 A1

A1 A2 A3 F A2 A3

Theorem 5 (Pascal). Assume that the points A1 , A2 , A3 , B1 , B2 , B3 belong to a circle. The point in
intersections of A1 B2 with A2 B1 , A1 B3 with A3 B1 , A2 B3 with A3 B2 lie on a line.
Proof. The points C2′ , D, and E as in the proof of the Pappus theorem. Consider the sequence of
perspectivities
A1 A2 C3 ′
A3 B1 DC2 =
∧ A3 B1 B2 B3 = ∧ A3 B1 DC2 .
∧ A3 EB2C1 =

In the same way as above we conclude that C2 = C2′ . 

4 Pole. Polar. Theorems of Brianchon and Brokard


Definition 5. Given a circle k(O, r), let A∗ be the image of the point A 6= O under the inversion with
respect to k. The line a passing through A∗ and perpendicular to OA is called the polar of A with
respect to k. Conversely A is called the pole of a with respect to k.
Theorem 6. Given a circle k(O, r), let and a and b be the polars of A and B with respect to k. The
A ∈ b if and only if B ∈ a.
Proof. A ∈ b if and only if ∠AB∗ O = 90◦ . Analogously B ∈ a if and only if ∠BA∗ O = 90◦ , and it
reamins to notice that according to the basic properties of inversion we have ∠AB∗ O = ∠BA∗ O. 
Definition 6. Points A and B are called conjugated with respect to the circle k if one of them lies on
a polar of the other.
Theorem 7. If the line determined by two conjugated points A and B intersects k(O, r) at C and D,
then H (A, B;C, D). Conversely if H (A, B;C, D), where C, D ∈ k then A and B are conjugated with
respect to k.
Proof. Let C1 and D1 be the intersection points of OA with k. Since the inversion preserves the
cross-ratio and R(C1 , D1 ; A, A∗ ) = R(C1 , D1 ; A∗ , A) we have

H (C1 , D1 ; A, A∗ ). (7)

Let p be the line that contains A and intersects k at C and D. Let E = CC1 ∩ DD1 , F = CD1 ∩ DC1 .
Since C1 D1 is the diameter of k we have C1 F⊥D1 E and D1 F⊥C1 E, hence F is the orthocenter of
the triangle C1 D1 E. Let B = EF ∩CD and Ā∗ = EF ∩C1 D1 . Since

C1 D1 AĀ∗ = ∗
E F
∧ D1C1 AĀ
∧ CDAB =

have H (C1 , D1 ; A, Ā∗ ) and H (C, D; A, B). (7) now implies two facts:
1◦ From H (C1 , D1 ; A, Ā∗ ) and H (C1 , D1 ; A, A∗ ) we get A∗ = Ā∗ , hence A∗ ∈ EF. However,
since EF⊥C1 D1 , the line EF = a is the polar of A.
2◦ For the point B which belongs to the polar of A we have H (C, D; A, B). This completes the
proof. 
Theorem 8 (Brianchon’s theorem). Assume that the hexagon A1 A2 A3 A4 A5 A6 is circumscribed
about the circle k. The lines A1 A4 , A2 A5 , and A3 A6 intersect at a point.
4 Olympiad Training Materials, www.imomath.com

Proof. We will use the convention in which the points will be denoted by capital latin letters, and
their repsective polars with the corresponding lowercase letters.
Denote by Mi , i = 1, 2, . . . , 6, the points of tangency of Ai Ai+1 with k. Since mi = Ai Ai+1 , we
have Mi ∈ ai , Mi ∈ ai+1 , hence ai = Mi−1 Mi .
Let b j = A j A j+3 , j = 1, 2, 3. Then B j = a j ∩ a j+3 = M j−1 M j ∩ M j+3 M j+4 . We have to prove
that there exists a point P such that P ∈ b1 , b2 , b3 , or analogously, that there is a line p such that
B1 , B2 , B3 ∈ p. In other words we have to prove that the points B1 , B2 , B3 are colinear. However this
immediately follows from the Pascal’s theorem applied to M1 M3 M5 M4 M6 M2 . 2
From the previous proof we see that the Brianchon’s theorem is obtained from the Pascal’s by
replacing all the points with their polars and all lines by theirs poles.

Theorem 9 (Brokard). The quadrilateral ABCD is inscribed in the circle k with center O. Let
E = AB ∩CD, F = AD ∩ BC, G = AC ∩ BD. Then O is the orthocenter of the triangle EFG.

Proof. We will prove that EG is a polar of F. Let E


X = EG ∩ BC and Y = EG ∩ AD. Then we also
have
E G
ADY F =∧ BCXF = ∧ DAY F,
A

which implies the relations H (A, D;Y, F) and Y D


H (B,C; X, F). According to the properties of G
polar we have that the points X and Y lie on a O F
polar of the point F, hence EG is a polar of the B X C
point F.

Since EG is a polar of F, we have EG⊥OF. Analogously we have FG⊥OE, thus O is the


orthocenter of △EFG. 

5 Problems
1. Given a quadrilateral ABCD, let P = AB ∩ CD, Q = AD ∩ BC, R = AC ∩ PQ, S = BD ∩ PQ.
Prove that H (P, Q; R, S).

2. Given a triangle ABC and a point M on BC, let N be the point of the line BC such that ∠MAN =
90◦ . Prove that H (B,C; M, N) if and only if AM is the bisector of the angle ∠BAC.

3. Let A and B be two points and let C be the point of the line AB. Using just a ruler find a point
D on the line AB such that H (A, B;C, D).

4. Let A, B, C be the diagonal points of the quadrilateral PQRS, or equivalently A = PQ ∩ RS,


B = QR ∩ SP, C = PR ∩ QS. If only the points A, B, C, S, are given using just a ruler construct
the points P, Q, R.

5. Assume that the incircle of △ABC touches the sides BC, AC, and AB at D, E, and F. Let M be
the point such that the circle k1 incscibed in △BCM touches BC at D, and the sides BM and
CM at P and Q. Prove that the lines EF, PQ, BC are concurrent.

6. Given a triangle ABC, let D and E be the points on BC such that BD = DE = EC. The line p
intersects AB, AD, AE, AC at K, L, M, N, respectively. Prove that KN ≥ 3LM.

7. The point M1 belongs to the side AB of the quadrilateral ABCD. Let M2 be the projection of
M1 to the line BC from D, M3 projection of M2 to CD from A, M4 projection of M3 to DA from
B, M5 projection of M4 to AB from C, etc. Prove that M13 = M1 .
Milivoje Lukić: Projective Geometry 5

8. (butterfly theorem) Points M and N belong to the circle k. Let P be the midpoint of the chord
MN, and let AB and CD (A and C are on the same side of MN) be arbitrary chords of k passing
through P. Prove that lines AD and BC intersect MN at points that are equidistant from P.

9. Given a triangle ABC, let D and E be the points of the sides AB and AC respectively such that
DEkBC. Let P be an interior point of the triangle ADE. Assume that the lines BP and CP
intersect DE at F and G respectively. The circumcircles of △PDG and △PFE intersect at P
and Q. Prove that the points A, P, and Q are colinear.

10. (IMO 1997 shortlist) Let A1 A2 A3 be a non-isosceles triangle with the incenter I. Let Ci , i = 1,
2, 3, be the smaller circle through I tangent to both Ai Ai+1 and Ai Ai+2 (summation of indeces
is done modulus 3). Let Bi , i = 1, 2, 3, be the other intersection point of Ci+1 and Ci+2 . Prove
that the circumcenters of the triangles A1 B1 I, A2 B2 I, A3 B3 I are colinear.

11. Given a triangle ABC and a point T , let P and Q be the feet of perpendiculars from T to the
lines AB and AC, respectively. Let R and S be the feet of perpendiculars from A to TC and T B,
respectively. Prove that the intersection of PR and QS belongs to BC.

12. Given a triangle ABC and a point M, a line passing through M intersects AB, BC, and CA at
C1 , A1 , and B1 , respectively. The lines AM, BM, and CM intersect the circumcircle of △ABC
repsectively at A2 , B2 , and C2 . Prove that the lines A1 A2 , B1 B2 , and C1C2 intersect in a point
that belongs to the circumcircle of △ABC.

13. Let P and Q isogonaly conjugated points and assume that △P1 P2 P3 and △Q1 Q2 Q3 are their
pedal triangles, respectively. Let X1 = P2 Q3 ∩ P3 Q2 , X2 = P1 Q3 ∩ P3 Q1 , X3 = P1 Q2 ∩ P2 Q1 .
Prove that the points X1 , X2 , X3 belong to the line PQ.

14. If the points A and M are conjugated with respect to k, then the circle with diameter AM is
orthogonal to k.

15. From a point A in the exterior of a circle k two tangents AM and AN are drawn. Assume that
K and L are two points of k such that A, K, L are colinear. Prove that MN bisects the segment
PQ.

16. The point isogonaly conjugated to the centroid is called the Lemuan point. The lines connected
the vertices with the Lemuan point are called symmedians. Assume that the tangents from B
and C to the circumcircle Γ of △ABC intersect at the point P. Prove that AP is a symmedian
of △ABC.

17. Given a triangle ABC, assume that the incircle touches the sides BC, CA, AB at the points M,
N, P, respectively. Prove that AM, BN, and CP intersect in a point.

18. Let ABCD be a quadrilateral circumscribed about a circle. Let M, N, P, and Q be the points of
tangency of the incircle with the sides AB, BC, CD, and DA respectively. Prove that the lines
AC, BD, MP, and NQ intersect in a point.

19. Let ABCD be a cyclic quadrilateral whose diagonals AC and BD intersect at O; extensions of
the sides AB and CD at E; the tangents to the circumcircle from A and D at K; and the tangents
to the circumcircle at B and C at L. Prove that the points E, K, O, and L lie on a line.

20. Let ABCD be a cyclic quadrilateral. The lines AB and CD intersect at the point E, and the
diagonals AC and BD at the point F. The circumcircle of the triangles △AFD and △BFC
intersect again at H. Prove that ∠EHF = 90◦ .
6 Olympiad Training Materials, www.imomath.com

6 Solutions
1. Let T = AC ∩ BD. Consider the sequence of the perspectivities
A C
PQRS = ∧ QPRS.
∧ BDT S =

Since the perspectivity preserves the cross-ratio R(P, Q; R, S) = R(Q, P; R, S) we obtain that
H (P, Q; R, S).
2. Let α = ∠BAC, β = ∠CBA, γ = ∠ACB and ϕ = ∠BAM. Using the sine theorem on △ABM
and △ACM we get
BM BM AM sin ϕ sin γ
= = .
MC AM CM sin β sin(α − ϕ )
Similarly using the sine theorem on △ABN and △ACN we get
BN BN AN sin(90◦ − ϕ ) sin γ
= = .
NC AN CN sin(180◦ − β ) sin(90◦ + α − ϕ )
Combining the previous two equations we get
BM BN tan ϕ
: = .
MC NC tan(α − ϕ )
Hence, |R(B,C; M, N)| = 1 is equivalent to tan ϕ = tan(α − ϕ ), i.e. to ϕ = α /2. Since B 6= C
and M 6= N, the relation |R(B,C; M, N)| = 1 is equivalent to R(B,C; M, N) = −1, and the
statement is now shown.
3. The motivation is the problem 1. Choose a point K outside AB and point L on AK different
from A and K. Let M = BL ∩ CK and N = BK ∩ AM. Now let us construct a point D as
D = AB ∩ LN. From the problem 1 we indeed have H (A, B;C, D).
4. Let us denote D = AS ∩ BC. According to the problem 1 we have H (R, S; A, D). Now we
construct the point D = AS ∩ BC. We have the points A, D, and S, hence according to the
previous problem we can construct a point R such that H (A, D; S, R). Now we construct
P = BS ∩CR and Q = CS ∩ BR, which solves the problem.
5. It is well known (and is easy to prove using Ceva’s theorem) that the lines AD, BE, and CF
intersect at a point G (called a Gergonne point of △ABC) Let X = BC ∩ EF. As in the problem
1 we have H (B,C; D, X). If we denote X ′ = BC ∩ PQ we analogously have H (B,C; D, X ′ ),
hence X = X ′ .
6. Let us denote x = KL, y = LM, z = MN. We have to prove that x + y + z ≥ 3y, or equivalently
x + z ≥ 2y. Since R(K, N; L, M) = R(B,C; D, E), we have
−→ −−→ −→ −→
x x + y KL KM BD BE 1 1
: = −→ : −−→ = −→ : −→ = : ,
y+z z LN MN DC EC 2 2
implying 4xz = (x + y)(y + z).
If it were y > (x + z)/2 we would have
3 1 1 √
x + y > x + z = 2 (x + x + x + z) ≥ 2 4 xxxz,
2 2 4

and analogously y+ z > 2 4 xzzz as well as (x + y)(y+ z) > 4xz which is a contradiction. Hence
the assumption y > (x + z)/2 was false so we have y ≤ (x + z)/2.
Let us analyze the case of equality. If y = (x + z)/2, then 4xz = (x + y)(x + z) = (3x + z)(x +
3z)/4, which is equivalent to (x − z)2 = 0. Hence the equality holds if x = y = z. We leave to
the reader to prove that x = y = z is satisfied if and only if p k BC.
Milivoje Lukić: Projective Geometry 7

7. Let E = AB ∩CD, F = AD ∩ BC. Consider the sequence of perspectivities


D A B C
ABEM1 ∧ FBCM2 =
= ∧ DECM3 = ∧ EABM5 .
∧ DAFM4 = (8)

According to the conditions given in the problem this sequence of perspectivites has two
be applied three more times to arrive to the point M13 . Notice that the given sequence of
perspectivities maps A to E, E to B, and B to A. Clearly if we apply (8) three times the points
A, B, and E will be fixed while M1 will be mapped to M13 . Thus M1 = M13 .
8. Let X ′ be the point symmetric to Y with respect to P. Notice that
D B
R(M, N; X, P) = R(M, N; P,Y ) (from MNXP = ∧ MNAC = ∧ MNPY )

= R(N, M; P, X ) ( the reflection with the center P preserves
the ratio, hence it preserves the cross-ratio)
1
= = R(M, N; X ′ , P),
R(N, M; X ′ , P)

where the last equality follows from the basic properties of the cross ratio. It follows that
X = X ′.
9. Let J = DQ ∩ BP, K = EQ ∩CP. If we prove that JK k DE this would imply that the triangles
BDJ and CEK are perspective with the respect to a center, hence with repsect to an axis as
well (according to Desargue’s theorem) which immediately implies that A, P, Q are colinear
(we encourage the reader to verify this fact).
Now we will prove that JKkDE. Let us denote T = DE ∩PQ. Applying the Menelaus theorem
on the triangle DT Q and the line PF we get
→ −→ −→

DJ QP T F
→ −→ −→ = −1.

JQ PT FD
Similarly from the triangle ET Q and the line PG:
−→ −→ −→
EK QP T G
−→ −→ −→ = −1.
KQ PT GE
Dividing the last two equalities and using DT · T G = FT · T E (T is on the radical axis of the
circumcircles of △DPG and △FPE), we get
→ −→

DJ EK
→ = −→ .

JQ KQ

Thus JK k DE, q.e.d.


10. Apply the inversion with the respect to I. We leave to the reader to draw the inverse picture.
Notice that the condition that I is the incentar now reads that the circumcircles A∗i A∗i+1 I are of
the same radii. Indeed if R is the radius of the circle of inversion and r the distance between
I and XY then the radius of the circumcircle of △IX ∗Y ∗ is equal to R2 /r. Now we use the
following statement that is very easy to prove: ”Let k1 , k2 , k3 be three circles such that all
pass through the same point I, but no two of them are mutually tangent. Then the centers of
these circles are colinear if and only if there exists another common point J 6= I of these three
circles.”
In the inverse picture this transforms into proving that the lines A∗1 B∗1 , A∗2 B∗2 , and A∗3 B∗3 intersect
at a point.
8 Olympiad Training Materials, www.imomath.com

In order to prove this it is enough to show that the corresponding sides of the triangles A∗1 A∗2 A∗3
and B∗1 B∗2 B∗3 are parallel (then these triangles would be perspective with respect to the infinitely
far line). Afterwards the Desargue’s theorem would imply that the triangles are perspective
with respect to a center. Let Pi∗ be the incenter of A∗i+1 A∗i+2 I, and let Q∗i be the foot of the
perpendicular from I to Pi+1 ∗ P∗ . It is easy to prove that
i+2

−− → −−−→ −−→
A∗1 A∗2 = 2Q∗1 Q∗2 = −P1∗ P2∗ .

Also since the circles A∗i A∗i+1 I are of the same radii, we have P1∗ P2∗ k B∗1 B∗2 , hence A∗1 A∗2 k B∗1 B∗2 .

11. We will prove that the intersection X of PR and QS lies on the line BC. Notice that the points
P, Q, R, S belong to the circle with center AT . Consider the six points A, S, R, T , P, Q that lie
on a circle. Using Pascal’s theorem with respect to the diagram

T P Q

B X
C

A S R

we get that the points B, C, and X = PR ∩ QS are colinear.

12. First solution, using projective mappings. Let A3 = AM ∩ BC and B3 = BM ∩ AC. Let X be
the other intersection point of the line A1 A2 with the circumcircle k of △ABC. Let X ′ be the
other intersection point of the line B1 B2 with k. Consider the sequence of perspectivities

ABCX ′
A2 M 2 B
ABCX =
∧ A3 BCA1 ∧ AB3CB1 =
= ∧

which has three fixed points A, B, C, hence X = X ′ . Analogously the line C1C2 contains X and
the problem is completely solved.
Second solution, using Pascal’s theorem. Assume that the line A1 A2 intersect the circumcircle
of the trianlge ABC at A2 and X. Let XB2 ∩ AC = B′1 . Let us apply the Pascal’s theorem on the
points A, B, C, A2 , B2 , X according the diagram:

B2 A2 C

M A1
B′1

A B X

It follows that the points A1 , B′1 , and M are colinear. Hence B′1 ∈ A1 M. According to the
definition of the point B′1 we have B′1 ∈ AC hence B′1 = A1 M ∩ AC = B1 . The conclusion is that
the points X, B1 , B2 are colinear. Analogously we prove that the points X, C1 , C2 are colinear,
hence the lines A1 A2 , B1 B2 , C1C2 intersect at X that belongs to the circumcircle of the triangle
ABC.

13. It is well known (from the theory of pedal triangles) that pedal triangles corresponding to
the isogonally conjugated points have the common circumcircle, so called pedal circle of the
points P and Q. The center of that circle which is at the same time the midpoint of PQ will
be denoted by R. Let P1′ = PP1 ∩ Q1 R and P2′ = PP2 ∩ Q2 R (the points P1′ and P2′ belong to the
pedal circle of the point P, as point on the same diameters as Q1 and Q2 respectively). Using
the Pascal’s theorem on the points Q1 , P2 , P2′ , Q2 , P1 , P1′ in the order shown by the diagram
Milivoje Lukić: Projective Geometry 9

P2 P1′ Q2

P R
X3

P1 P2′ Q1

we get that the points P, R, X1 are colinear or X1 ∈ PQ. Analogously the points X2 , X3 belong
to the line PQ.
14. Let us recall the statement according to which the circle l is invariant under the inversion with
respect to the circle k if and only if l = k or l ⊥ k.
Since the point M belongs to the polar of the point A with respect to k we have ∠MA∗ A = 90◦
where A∗ = ψl (A). Therefore A∗ ∈ l where l is the circle with the radius AM. Analogously
M ∗ ∈ l. However from A ∈ l we get A∗ ∈ l ∗ ; A∗ ∈ l yields A ∈ l ∗ (the inversion is inverse to
itself) hence ψl (A∗ ) = A). Similarly we get M ∈ l ∗ and M ∗ ∈ l ∗ . Notice that the circles l and
l ∗ have the four common points A, A∗ , M, M ∗ , which is exactly two too much. Hence l = l ∗
and according to the statement mentioned at the beginning we conclude l = k or l ⊥ k. The
case l = k can be easily eliminated, because the circle l has the diameter AM, and AM can’t be
the diameter of k because A and M are conjugated to each other.
Thus l ⊥ k, q.e.d.
15. Let J = KL ∩ MN, R = l ∩ MN, X∞ = l ∩ AM. Since MN is the polar of A from J ∈ MN we get
M
∧ PQRX∞ we also have H (P, Q; R, X∞ ). This implies that R is the
H (K, L; J, A). From KLJA =
midpoint of PQ.
16. Let Q be the intersection point of the lines AP and BC. Let Q′ be the point of BC such that the
ray AQ′ is isogonal to the ray AQ in the triangle ABC. This exactly means that ∠Q′ AC = ∠BAQ
i ∠BAQ′ = ∠QAC.
For an arbitrary point X of the segment BC, the sine theorem applied to triangles BAX and
XAC yields
BX BX AX sin ∠BAX sin ∠ACX sin ∠ACX sin ∠BAX AB sin ∠BAX
= = = = .
XC AX XC sin ∠ABX sin ∠XAC sin ∠ABX sin ∠XAC AC sin ∠XAC
Applying this to X = Q and X = Q′ and multiplying together afterwards we get
BQ BQ′ AB sin ∠BAQ AB sin ∠BAQ′ AB2

= ′
= . (9)
QC Q C AC sin ∠QAC AC sin ∠Q AC AC2
Hence if we prove BQ/QC = AB2 /AC2 we would immediately have BQ′ /Q′C = 1, making
Q′ the midpoint of BC. Then the line AQ is isogonaly conjugated to the median, implying the
required statement.
Since P belongs to the polars of B and C, then the points B and C belong to the polar of the
point P, and we conclude that the polar of P is precisely BC. Consider the intersection D of
the line BC with the tangent to the circumcircle at A. Since the point D belongs to the polars
of A and P, AP has to be the polar of D. Hence H (B,C; D, Q). Let us now calculate the ratio
BD/DC. Since the triangles ABD and CAD are similar we have BD/AD = AD/CD = AB/AC.
This implies BD/CD = (BD/AD)(AD/CD) = AB2 /AC2 . The relation H (B,C; D, Q) implies
BQ/QC = BD/DC = AB2 /AC2 , which proves the statement.
17. The statement follows from the Brianchon’s theorem applied to APBMCN.
18. Applying the Brianchon’s theorem to the hexagon AMBCPD we get that the line MP contains
the intersection of AB and CD. Analogously, applying the Brianchon’s theorem to ABNCDQ
we get that NQ contains the same point.
10 Olympiad Training Materials, www.imomath.com

19. The Brokard’s theorem claims that the polar of F = AD ∩ BC is the line f = EO. Since the
polar of the point on the circle is equal to the tangent at that point we know that K = a ∩ d,
where a and d are polars of the points A and D. Thus k = AD. Since F ∈ AD = k, we have
K ∈ f as well. Analogously we can prove that L ∈ f , hence the points E, O, K, L all belong to
f.
20. Let G = AD ∩ BC. Let k be the circumcircle of ABCD. Denote by k1 and k2 respectively the
circumcircles of △ADF and △BCF. Notice that AD is the radical axis of the circles k and k1 ;
BC the radical axis of k and k2 ; and FH the radical axis of k1 and k2 . According to the famous
theorem these three radical axes intersect at one point G. In other words we have shown that
the points F, G, H are colinear.
Without loss of generality assume that F is between G and H (alternatively, we could use
the oriented angles). Using the inscribed quadrilaterals ADFH and BCFH, we get ∠DHF =
∠DAF = ∠DAC and ∠FHC = ∠FBC = ∠DBC, hence ∠DHC = ∠DHF + ∠FHC = ∠DAC +
∠DBC = 2∠DAC = ∠DOC. Thus the points D, C, H, and O lie on a circle. Similarly we prove
that the points A, B, H, O lie on a circle.
Denote by k3 and k4 respectively the circles circumscribed about the quadrilaterals ABHO and
DCHO. Notice that the line AB is the radical axis of the circles k and k3 . Simlarly CD and
OH, respectively, are those of the pairs of circles (k, k2 ) and (k3 , k4 ). Thus these lines have
to intersect at one point, and that has to be E. This proves that the points O, H, and E are
colinear.
According to the Brocard’s theorem we have FH⊥OE, which according to FH = GH and
OE = HE in turn implies that GH⊥HE, q.e.d.

You might also like