Dutch MO & TST 2008-20 With Solutions

The 46th
Dutch
Mathematical
Olympiad
2007
DSE
W
IS
N
K
RLA
U
N
D
E
E
NED
P I AD E
OLYM
and the team selection
for IMO 2008 Madrid

We thank our main sponsors:
Contents
Introduction i
First Round, January 2007 1
Solutions 3
Second Round, September 2007 5
Solutions 6
Team Selection Test, June 2008 8
Solutions 9
Solutions 15
Board of the Training committee

Dutch Mathematical 2007–2008
Olympiad Foundation
prof.dr. Frits Beukers, dr. Quintijn Puite,

president chair
drs. Melanie Steentjes, Birgit van Dalen MSc,
secretary vice chair
dr.ir. Tom Verhoeff, Johan Konter
treasurer Julian Lyczak
drs. Alex van den Brandhof dr. René Sitters
drs. Jaap de Jonge Arne Smeets
dr. Quintijn Puite Sietske Tacoma
Introduction
The Dutch Mathematical Olympiad consists of two rounds. The first round is held on the
participating schools and consists of eight multiple choice questions and four open questions
(see page 1–4). Students get two hours to work on the paper. On January 26, 2007, in total
2742 students of 185 secondary schools participated in this first round.
Those students that scored 20 points or more on the first round (out of a maximum of
36 points) were invited to the second round, which took place at University of Technology
Eindhoven after the summer holidays. Also some outstanding participants in the Kangaroo
math contest or the Pythagoras Olympiad were invited.
In total 132 students were invited, of which 119 participated in the second round on September
14th, 2007. The second or final round contains five problems for which the students have to
give extensive solutions and proofs. They have three hours for the paper (see page 5–7).
After the prices had been awarded in the beginning of November, the Dutch Mathematical
Olympiad concluded its 46th edition 2007. In 2011 we will have our 50th edition.
The 24 most outstanding candidates of the Dutch Mathematical Olympiad 2007 were invited
to an intensive seven-month training programme, consisting of weekly problem sets. Also, the
students met twice for a three-day training camp, three times for a day at the university, and
finally for a six-day training camp in the beginning of June. At this time, only 14 candidates
were left. Out of those, the team was selected by a final selection test on June 7, 2008 (see
page 8–12). The team will have a training camp in Spain from July 7 until July 14, together
with the team from New Zealand.
In the meantime, a new edition of the Dutch Mathematical Olympiad has started at the
participating schools on January 25, 2008 (see page 13–16). We are happy to see that the
number of participating schools as well as the number of participants has increased to 201
respectively 3004. As a new initiative, we have made some training material and we have
organized training sessions at six universities in the country for the 144 students who have
been invited for the second round in September 2008.
i
The Dutch team for IMO 2008 Madrid consists of
• Raymond van Bommel (16 y.o., participated in IMO 2007 as well)
• Remy van Dobben de Bruyn (17 y.o.)
• Floris van Doorn (17 y.o., observer C in IMO 2007)
• Alexander van Hoorn (18 y.o.)
• Milan Lopuhaä (18 y.o., participated to in IMO 2007 as well)
• Maarten Roelofsma (17 y.o.)
The Dutch delegation for IMO 2008 Madrid further consists of
• dr. Quintijn Puite, team leader

University of Technology Eindhoven
• Birgit van Dalen MSc, deputy team leader

Leiden University
• drs. Wim Berkelmans, project manager IMO 2011, observer A

VU University Amsterdam
• prof.dr. Gerhard Wöginger, observer A

University of Technology Eindhoven
• Teunis van Wijngaarden, observer C

CWI Amsterdam
We are grateful to Jinbi Jin for the translation into English of both the problems and the solutions.
ii
SE
W
AND
IS
K
U
ERL
N
D
E
NED Dutch Mathematical Olympiad
P IADE
OLYM Round 1
Friday, January 26, 2007
Please read the following before you start: Available time: 2 hours
• The A-problems are multiple choice questions. Only one of the five given options is correct. Please state
clearly behind which letter the correct solution is stated. You get 2 points for each correct answer.
• The B-problems are open questions, which have a number as answer. You get 5 points for each correct
answer. Please work accurately, since an error in your calculations
√ can cause your solution to be considered
wrong. Please give exact answers, for example 1181
or 2 + 3 or π + 1.
• This is a competition, not an exam. So it is to be expected that only few people will get all the answers
right, and you don’t need to worry if you have only solved a part of the problems.
• The use of formula sheets, calculators and similar tools is not allowed.
• The point is that you have fun solving unusual mathematical problems. Good luck!
A-problems
A1. The number M has 2007 digits, each of which is equal to 1, M = 111 . . . 111. What is the
sum of the digits of 2007 × M ?
A) 2007 B) 18036 C) 18063 D) 18084 E) 4028049
A2. The next five sequences all contain the same numbers. Which sequence is ordered correctly?
117 513
A) 0.16 < <
101 < 33
7 97 <
17 1 5 13
B) 101 < 0.16 < 7 < 33 < 97
1 5 17 13
C) 7 < 0.16 < 33 < 101 < 97
5 1 13 17
D) 33 < 7 < 97 < 101 < 0.16
13 1 5 17
E) 97 < 7 < 33 < 0.16 < 101
A3. In the figure, nine grid points are drawn. In how many ways can we
draw a triangle, the vertices of which lie on three of these nine grid
points?
A) 76 B) 84 C) 92 D) 496 E) 504
1
A4. How many pairs of positive integers (a, b) with a + b < 100 satisfy the equation a + b =
13 × (b + a1 )?
A) 5 B) 7 C) 9 D) 13 E) 28
A5. We use five different paths to connect five grid points.

Which of these paths is the shortest one?
A) B) C) D) E)
1
A6. We construct a sequence of numbers as following:
The first number is equal to 2, and so is the second number. Every following number
is the product of the two previous ones. So the first few numbers of the sequence are:
2, 2, 4, 8, 32, . . .
What is the final digit of the 2007th number in this sequence?
A) 0 B) 2 C) 4 D) 6 E) 8
A7. Does the equation 9n + 9n + 9n = 32007 have any integral solutions? If so, which of the
following is a solution?
A) 667 B) 669 C) 1003 D) 2006 E) No solutions exist
A8. Behind a long table, nine chairs are arranged in a row, on which six students and three
teachers are going to sit. The three teachers arrive first and they decide to pick their chairs in
such a way that every teacher sits between two students. In how many ways can the teachers
pick their chairs, given this constraint?
A) 12 B) 36 C) 60 D) 84 E) 630
B-problems
B1. Mark one card with a ‘1’, two cards with a ‘2’, . . ., fifty cards with a ‘50’. Put these 1 + 2 +
· · · + 50 = 1275 cards into a box and shuffle them. How many cards do you need to take from
the box to be certain that you will have taken at least 10 cards with the same mark?
D C
B2. We have a quadrilateral ABCD with side lengths |AB| = 16, |BC| =
21, |CD| = 2 and |DA| = 28. The sides AB and CD are parallel. Two
lines, both parallel to AB and CD, divide ABCD into three similar
quadrilaterals.
Calculate the perimeter of the smallest of these quadrilaterals.
A B
B3. For every pair of integers a, b, we define an operator a ⊗ b with the following three properties.
1. a ⊗ a = a + 2;
2. a ⊗ b = b ⊗ a;
a⊗(a+b) a+b
3. a⊗b = b .
Calculate 8 ⊗ 5.
B4. A flag with the shape of an equilateral triangle is hung up on the tops
of two vertical poles by two of its vertices. One of these poles has
length 4, while the other has length 3. Also, the third vertex of the
flag touches the ground. Calculate the length of the flag’s side.
2
SE
W
AND
IS
K
U
ERL
N
D
E
NED
P IADE Solutions
OLYM
of the problems of Round 1 of the
Dutch Mathematical Olympiad 2007
A-problems
A1. C) 18063 If M had 4 ones, then 2007 × M = 2229777, and the sum of the digits equals
3 × (2 + 7) + 9 = 4 × 9.
If M had 5 ones, then 2007×M = 22299777, and the sum of the digits equals 3×(2+7)+2×9 =
5 × 9.
Continuing like this, we see that if M has 2007 ones, then the sum of the digits equals
2007 × 9 = 18063.
A2. E) We don’t need to give all the fractions the same denominator; we can choose
16
the denominators in such a way that they are near 100. So 0, 16 = 100 , 71 = 14 5 15
98 , 33 = 99 .
Since, for k < n, nk < n+1
k+1 k+1
, (because of n+1 − nk = n(n+1)
n−k
> 0,) we now can see that
13 14 15 16 17
97 < 98 < 99 < 100 < 101 .
A3. A) 76 There are 9×8×7

3×2×1 = 84 ways to pick three points from the nine grid points. But
if those three points lie on a single line, then those three points do not form a triangle. This
occurs eight times; three times horizontally, three times vertically, and two times diagonally.
So there are 84 − 8 = 76 ways to pick three points from the grid points such that they do
form a triangle.
A4. B) 7 First we rewrite the equation as ab+1b = 13 × ( ba+1

a ). So a = 13b, and we see
that the only pairs satisfying the conditions are the seven pairs (13, 1), (26, 2), . . . , (91, 7).
A5. B) First, we see that path A is longer than path B, since

√
5 > 2. So path A √ cannot be the shortest one. Also, path E is longer
than path A, since 5 > 1; path E cannot√ be
√ the shortest one either.
And path D is longer than path B; 13 > 10. So path D isn’t the C
shortest path. B F
Now paths B and C remain. They have common segments of lengths D E
√
5 and 2, so it remains to compare A
√ the lengths of the other segments.
√
Path B has a remaining
√ length of 10 + 1, path C has√one of 2 5. As
in the figure, 10 + 1 = ADE < ADF < ABC = 2 5, from which
follows that path B has the shortest length.
A6. C) 4 We see that the sequence of final digits is equal to
2, 2, 4, 8, 2, 6, 2, 2, 4, 8, 2, 6, 2, . . . ,
and that we have a repeating part of length 6, namely 2, 2, 4, 8, 2, 6. Since 2007 = 334 × 6 + 3,
the final digit of the 2007th number must be a 4.
A7. C) 1003 Since 9n + 9n + 9n = 3 × 9n = 32n+1 , n = 1003 is a solution of our equation.
3
A8. C) 60 Imagine that the six students sit on their chairs in a row. The teachers then can
pick two neighbouring students, between which they are going to sit. For the first teacher,
there are exactly five ways to pick a chair that is between the chairs of two students. For
the second teacher, there are four, and for the third, there are three. So the total number of
ways for the teachers to pick their chairs in such a way that every teacher sits between two
students, is 5 × 4 × 3 = 60.
B-problems
B1. 415 Considering the worst case scenario, we see that we can get all the cards marked
‘1’ to ‘9’, and 9 of each of the cards marked ‘10’ to ‘50’, without having 10 cards of one kind.
In that case, we have 45 + 9 × 41 = 414. Now picking one card more (so 415 in total) ensures
that we have 10 cards of one kind.
B2. 13 Let P, Q, R, S be the intersections of the two lines where
P and R are on DA, and Q and S are on CB, see the figure. Now the
quadrilaterals ABSR, RSQP and P QCD are similar. So
D 2 C
|DC| |P Q| |RS| 4 3
= = . P
4
Q
|P Q| |RS| |AB|
R S
8
Hence |P Q|2 = 2 × |RS|, and |RS|2 = |P Q| × 16, so |P Q|3 = 64,
and |P Q| = 4, |RS| = 8. So the sides of RSQP are twice as large
as the corresponding sides of P QCD, and the sides of ABSR are A B
16
twice as large as the corresponding sides of RSP Q. Thus |RP | =
2 × |P D|, |AR| = 2 × |RP | = 4 × |P D|, and |AD| = 7 × |P D|, so we
conclude that |P D| = 4. Similarly, we see that |QC| = 3.
Thus the perimeter of P QCD equals 13.
B3. 120 Recursively, we compute
5 ⊗ (5 + 3) 8 8
8 ⊗ 5 = (5 + 3) ⊗ 5 = 8⊗5= × (5 ⊗ 3)
5⊗3 3 3
3 ⊗ (3 + 2) 5 5
5 ⊗ 3 = (3 + 2) ⊗ 3 = 5 ⊗ 3 = × (3 ⊗ 2)
3⊗2 2 2
2 ⊗ (2 + 1) 3
3 ⊗ 2 = (2 + 1) ⊗ 2 = 3 ⊗ 2 = 3 × (2 ⊗ 1)
2⊗1 1
1 ⊗ (1 + 1) 2
2 ⊗ 1 = (1 + 1) ⊗ 1 = 2 ⊗ 1 = 2 × (1 ⊗ 1)
1⊗1 1
Since 1 ⊗ 1 = 1 + 2 = 3, we have 2 ⊗ 1 = 2 × 3 = 6, then 3 ⊗ 2 = 3 × 6 = 18, and
5 ⊗ 3 = 52 × 18 = 45, and finally, 8 ⊗ 5 = 38 × 45 = 120.
2
√
B4. 3 39 Let x denote the side of the equilateral triangle. By re- C
2 2 x
peated application of Pythagoras’
√ Theorem, √ we get 2x = |BD|2 +
1
D B
2 2 2 2 2
√ + |AF
1 = (|EA| √ |) + 1 = ( x − 4 + x − 3 ) + 1 = 2x −
25√ + 2 x√− 4 x2 − 32 + 1. Rewriting this equation gives us
2 2 x
3 3
2 x2 − 42 x2 − 32 = −x2 + 24. Then we square both sides: 4(x4 − x
25x2 + 144) = x4 − 48xq 2 + 576. Thus 3x4 − 52x2 = 0. Now the only
√ E A F
positive solution is x = 523 = 2
3 39.
4
SE
W
AND
IS
K
U
ERL
N
D
E
P IADE
OLYM Round 2
Friday, September 14, 2007
Available time: 3 hours
• Writing down just the answer itself is not sufficient; you also need to describe the way you solved the problem.
• Calculators and formula sheets are not allowed; you are only allowed to use a pen, a compass and a ruler or
set square. And of course your common sense.
• Please write the solutions of each problem on a different sheet of paper.
• Good luck!
1. Consider the equilateral triangle ABC with |BC| = |CA| = |AB| = 1.

A2 B1
On the extension of side BC, we define points A1 (on the
same side as B) and A2 (on the same side as C) such that
|A1 B| = |BC| = |CA2 | = 1. Similarly, we define B1 and C
B2 on the extension of side CA such that |B1 C| = |CA| =
|AB2 | = 1, and C1 and C2 on the extension of side AB
such that |C1 A| = |AB| = |BC2 | = 1. C1 C2
A B
Now the circumcentre of 4ABC is also the centre of the
circle that passes through the points A1 , B2 , C1 , A2 , B1 and
C2 . B2 A1
Calculate the radius of the circle through A1 , B2 , C1 , A2 , B1 and C2 .
2. Is it possible to partition the set A = {1, 2, 3, . . . , 32, 33} into eleven subsets that contain
three integers each, such that for every one of these eleven subsets, one of the integers
is equal to the sum of the other two? If so, give such a partition; if not, prove that such
a partition cannot exist.
3. Does there exist an integer having the form 444 · · · 4443 (all fours, and ending with a
three) that is divisible by 13? If so, give an integer having that form that is divisible
by 13; if not, prove that such an integer cannot exist.
4. Determine the number of integers a satisfying 1 6 a 6 100 such that aa is a perfect

square. (And prove that your answer is correct.)
5. A triangle ABC and a point P inside this triangle are given.

C
Define D, E and F as the midpoints of AP , BP and CP ,
respectively. Furthermore, let R be the intersection of AE F
and BD, S the intersection of BF and CE, and T the
intersection of CD and AF . T S
Prove that the area of hexagon DRESF T is independent P
of the position of P inside the triangle. D E

R
A B
5
Solutions of Round 2 of the Dutch Mathematical Olympiad 2007.
Problem 1:
Let M be the circumcentre of 4ABC and D be the midpoint of BC.
A2 B1
Draw AD through M , and√draw M A1 . Since AD is a me-
dian, |M D| = 13 |AD| = 61 3 and |BD| = 21 . Now, apply-
C
ing Pythagoras’ Theorem
√ to 4M DA1 , we find |M A1√ |2 =
D
2 2
|M D| + |DA1 | = ( 16 3)2 + ( 32 )2 = 37 . So |M A1 | = 13 21.
M
(Similarly, one can compute
√ the distance from M to the A B
C1 C2
other points, yielding 13 21 every time.) So the radius of
the circle passing
√ through the points A1 , B2 , C1 , A2 , B1 , C2
is equal to 31 21.
B2 A1
Problem 2:
Suppose that we have such a partition into eleven subsets. Then for each of those eleven
subsets {a, b, c} we have that, say, a + b = c. So a + b + c = 2c, and the sum of the integers
in each subset is even. Hence also the sum of 1, 2, . . . , 32, 33 must be even.
But since 1 + 2 + 3 + · · · + 33 = 21 × 33 × (33 + 1) = 33 × 17 is an odd number, we have a
contradiction. So there doesn’t exist a partition with the desired properties.
Problem 3:
| · {z
Suppose that such an integer 444 · · 4443} exists and that it is divisible by 13, then substracting
k
13 from that integer gives another integer that is divisible by 13. So 444 · · · 4430 is divisible
by 13. Since 444 · · · 4430 = 444 · · · 443 × 10, and since 10 is not divisible by the prime 13,
| ·{z
444 · · 443}, the integer with one 4 less, must also be divisible by 13. We can repeat this
k−1
argument until we get the integer 43, from which we conclude that 43 must also be divisible
by 13. But 43 is not a multiple of 13, hence the integer we started with cannot be divisible
by 13 either. So there is no integer having that form that is divisible by 13.
(When the amount of 4’s equals 0, we have the number 3, which obviously isn’t a multiple of
13 either.)
6
Problem 4:
In the prime factorisation of a perfect square, every prime factor occurs an even number of
times. Also, if in the prime factorisation of an integer every prime factor occurs an even
number of times, then that integer must be a perfect square.
We consider two cases; namely a being even, or a being odd.
- Suppose a is even, and write a = 2c, then aa = (2c)2c = (2c cc )2 . So in this case, aa
always is a perfect square.
- Now suppose that a is odd, and write a = 2c + 1. Then

2
aa = (2c + 1)2c+1 = (2c + 1)2c (2c + 1) = (2c + 1)c (2c + 1).
2
Now (2c + 1)c is a perfect square, so aa is a perfect square if and only if every prime
factor of (2c + 1) occurs an even number of times, so if and only if (2c + 1) itself is a
perfect square.
So we conclude that every even integer and every odd perfect square no larger than 100 satisfy
the condition, and there are 50 + 5 = 55 of those.
Problem 5:
We denote the area of 4ABC as [ABC]. Similarly,
C
we denote the area of a quadrilateral ABCD as [ABCD].
Let us first consider 4ABP . The segments AE and BD F

are two medians of this triangle, so R is the centroid of
this triangle. Drawing the third median from P through T S
R, we see that the triangle is divided into 6 triangles with P
R
equal area. So [DREP ] = 62 [ABP ]. D E
A B
In a similar way, we find for 4BCP and 4CAP respectively that [ESF P ] = 31 [BCP ] and
[F T DP ] = 13 [CAP ]. Hence
[DRESF T ] = 13 ([ABP ] + [BCP ] + [CAP ]) = 31 [ABC],
which is independent of the location of P .
7
Team Selection Test
Valkenswaard, June 7, 2008
1. Find all funtions f : Z>0 → Z>0 that satisfy
f (f (f (n))) + f (f (n)) + f (n) = 3n
for all n ∈ Z>0 .
2. Julian and Johan are playing a game with an even number of cards, say 2n cards,
(n ∈ Z>0 ). Every card is marked with a positive integer. The cards are shuffled and
are arranged in a row, in such a way that the numbers are visible. The two players
take turns picking cards. During a turn, a player can pick either the rightmost or the
leftmost card.
Johan is the first player to pick a card (meaning Julian will have to take the last card).
Now, a player’s score is the sum of the numbers on the cards that player acquired during
the game.
Prove that Johan can always get a score that is at least as high as Julian’s.
3. Let m, n be positive integers. Consider a sequence of positive integers a1 , a2 , . . . , an

that satisfies m = a1 ≥ a2 ≥ · · · ≥ an ≥ 1. Then define, for 1 ≤ i ≤ m,
bi = #{j ∈ {1, 2, . . . , n} : aj ≥ i},
so bi is the number of terms aj of the given sequence for which aj ≥ i. Similarly, we

define, for 1 ≤ j ≤ n,
cj = #{i ∈ {1, 2, . . . , m} : bi ≥ j},
thus cj is the number of terms bi in the given sequence for which bi ≥ j.
E.g.: If a is the sequence 5, 3, 3, 2, 1, 1 then b is the sequence 6, 4, 3, 1, 1.
(a) Prove that aj = cj for 1 ≤ j ≤ n.
(b) Prove that for 1 ≤ k ≤ m: ki=1 bi = k · bk + nj=bk +1 aj .
P P
√
4. Let n be positive√integer such that 1 + 12n2 is an integer.
Prove that 2 + 2 1 + 12n2 is the square of an integer.
5. Let 4ABC be a right triangle with ∠B = 90◦ and |AB| > |BC|; and let Γ be the
semicircle with diameter AB that lies on the same side as C. Let P be a point on Γ
such that |BP | = |BC| and let Q be on AB such that |AP | = |AQ|.
Prove that the midpoint of CQ lies on Γ.
8
Solutions of the Team Selection Test 2008
Problem 1. If f (m) = f (n), then 3m = 3n, thus m = n. So f is injective. Now we use

induction on n to show that f (n) = n for all n.
Since f : Z>0 → Z>0 , f (1) ≥ 1, f (f (1)) ≥ 1 and f (f (f (1))) ≥ 1. Substituting n = 1 in the
functional equation yields
f (f (f (1))) + f (f (1)) + f (1) = 3,
so equality must hold everywhere, meaning that f (1) = f (f (1)) = f (f (f (1))) = 1.
Let k ≥ 2, and for any n < k, suppose that f (n) = n. Then injectivity implies for all m ≥ k
that f (m) ≥ k. So in particular, f (k) ≥ k, which implies f (f (k)) ≥ k, which again implies
f (f (f (k))) ≥ k. Substituting n = k in the functional equation now yields
f (f (f (k))) + f (f (k)) + f (k) = 3k.
Thus equality must hold everywhere; f (k) = f (f (k)) = f (f (f (k))) = k. This completes our
induction.
The above proves that the only function that can satisfy the functional equation is f (n) = n.
Checking this,
f (f (f (n))) + f (f (n)) + f (n) = n + n + n = 3n,
also proves that f also satisfies the functional equation.
Problem 2. Let 2n be the number of cards, and assume that the cards are marked with
(from left to right) a1 , a2 , . . . , a2n . We’ll prove by induction on n that Johan can always pick
either the odd cards a1 , a3 , . . . , a2n−1 , or the even cards a2 , a4 , . . . , a2n . This is clear for n = 1;
Johan picks the first card if he wants the odd cards, and he picks the second if he wants the
even cards.
So now suppose that the above statement is true for n = k, and consider the situation with
2k + 2 cards, a1 , a2 , . . . , a2k+2 . If Johan wants the odd cards, he first picks a1 , and Julian
will have to either pick a2 or a2k+2 in that case. In the first case, the sequence of cards after
2 turns becomes {bi } where bi = ai+2 , and according to the induction hypthesis, Johan can
get the odd bi , which in combination with a1 , will get him all the odd ai . In the second case,
the sequence of cards becomes {bi } where bi = ai+1 , and again, according to the induction
hypothesis, Johan can get the even bi , which in combination with a1 , will get him all odd ai .
So Johan can pick the odd cards. In a similar way we can prove that Johan can pick the even
cards. This completes the induction.
Now if the sum of the numbers on the odd cards is larger than or equal to the sum of the
numbers on the even numbers, Johan picks the odd cards. Otherwise he picks the even cards.
This ensures that Johan will get a score at least as high as Julian’s.
9
Problem 3.
(a) Solution 1. Note that, for 1 ≤ i ≤ m, 1 ≤ j ≤ n:
aj ≥ i ⇐⇒
a1 , a2 , . . . , aj ≥ i ⇐⇒
at least j terms of a are greater than or equal to i ⇐⇒
bi ≥ j ⇐⇒
b1 , b 2 , . . . , b i ≥ j ⇐⇒
at least i terms of b are greater than or equal to j ⇐⇒
cj ≥ i.
Now let 1 ≤ j ≤ n, and i = aj . Then aj ≥ i, from which follows that cj ≥ i = aj .

Similarly, if i = cj , then similarly, we get aj ≥ i = cj . Thus aj = cj .
Solution 2. Since the sequences are non-increasing, bi = max{l : al ≥ i} and cj =

max{i : bi ≥ j}. Thus, for 1 ≤ j ≤ n:
cj = max{i : bi ≥ j} = max{i : max{l : al ≥ i} ≥ j}.
For fixed i, we have

max{l : al ≥ i} ≥ j ⇐⇒ aj ≥ i,
so cj = max{i : aj ≥ i} = aj .
(b) Solution 1. For 1 ≤ k ≤ m,

k
X k
X k
X
(bi − bk ) = (#{l : al ≥ i} − #{l : al ≥ k}) = #{l : k > al ≥ i}.
i=1 i=1 i=1
Any term al with k > al is counted exactly al times in this sum (for every i ≤ al ), so
this sum must equal the sum of all such al . By solution 1 of a), we now have k ≤ al if
and only if l ≤ bk , so k > al if and only if l ≤ bk . Thus,
k
X X n
X
(bi − bk ) = al = al ,
i=1 l:k>al l=bk +1
which is what we needed to prove.
10
Solution 2. By induction on k. If k = 1, then we need to prove
n
X
b1 = b1 + aj ,
j=b1 +1
which is obviously true since b1 = #{j : aj ≥ 1} = n, so the sum on the right hand side
is empty. So suppose that what we need to prove is true for k = s, 1 ≤ s ≤ m − 1. Then
s+1
X s
X
bi = bi + bs+1
i=1 i=1
n
IH
X
= s · bs + aj + bs+1
j=bs +1
n
X bs
X
= (s + 1) · bs+1 + aj + s(bs − bs+1 ) − aj
j=bs+1 +1 j=bs+1 +1
n
X bs
X
= (s + 1) · bs+1 + aj + s(bs − bs+1 ) − s
j=bs+1 +1 j=bs+1 +1
X n
= (s + 1) · bs+1 + aj .
j=bs+1 +1
Problem 4. Let a be such that 1 + 12n2 = a2 . Rewrite this as
12n2 = a2 − 1 = (a + 1)(a − 1).
Note that the left hand side is even, hence so is the right hand side, and a has to be odd.
Since the left hand side has an even number of factors 2, both a + 1, a − 1 have an odd number
of factors 2, since one of these two has exactly one factor 2. Also, (a + 1, a − 1) = 2, from
which follows that for any odd prime divisor p of the left hand side, p divides exactly one
of a + 1, a − 1. Thus, if p ≥ 5, the right hand side has an even number of factors p, and if
p = 3, the right hand side has an odd number of factors p. Hence we have the following two
possibilities:
a + 1 = 6b2 and a − 1 = 2c2
for certain integers b, c such that bc = n, or
a + 1 = 2b2 and a − 1 = 6c2
for certain integers b, c such that bc = n.

Consider the first case. Then 3|a + 1, so a − 1 ≡ 1 mod 3. Thus c2 ≡ 2 mod 3, which is a
contradiction. Hence the second case must hold. In that case,
p
2 + 2 1 + 12n2 = 2 + 2a = 2(a + 1) = 4b2 = (2b)2 ,
which was what we needed to prove.
11
Problem 5. Solution 1. Let S be the intersection of Γ en CQ. We need to prove that
|QS| = |SC|.
Note that BC is tangent to Γ, so ∠CBP = ∠BAP = ∠QAP . Since both 4CBP and 4QAP
are both isosceles, they’re similar. Let α = ∠BCP , then
α = ∠BCP = ∠CP B = ∠QP A = ∠AQP.
Now note that
∠CP Q = ∠CP B + ∠BP Q = ∠QP A + ∠BP Q = ∠BP A = 90◦ .
So P , just like B, lies on the circle with diameter CQ. S lies on this diameter, and all that’s
left to show, is
∠BSP = 2α = 2∠BCP.
Considering the cyclic quadrilateral QBCP , we see that ∠CP B = ∠CQB, so
2α = ∠CQB+∠AQP = 180◦ −∠P QC = 180◦ −∠P BC = 90◦ +∠QBC−∠P BC = 90◦ +∠QBP.
Considering the cyclic quadrilateral ABSP , with diameter AB, we see that
90◦ + ∠QBP = 90◦ + ∠ABP = ∠BSA + ∠ASP = ∠BSP,
from which follows that

2α = ∠BSP,
which is what we needed to prove.
Solution 2. First of all, as in solution 1, the two isosceles triangles are similar. Since BC is
orthogonal to AB, one of the triangles is rotated by 90◦ with respect to the other. Now let
l1 be the angle bisector of ∠P BC and l2 be the angle bisector ∠P AQ, and note that these
angle bisectors are perpendicular. Let T be the intersection of l1 and l2 . Then ∠AT B = 90◦ ,
so T lies on Γ.
Note that P is the image of C under reflection in l1 , and Q is the image of P under reflection
in l2 . The composition of these two reflections is the rotation about T by 2 · 90 = 180 degrees.
This rotation sends C to Q, so CT Q is a straight line. Thus T = S. Also, |T C| = |T Q|, so
S is the midpoint of CQ.
12
SE
W
AND
IS
K
U
ERL
N
D
E
P IADE
OLYM Round 1
Available time: 2 hours
Please read the following before you start:
• The A-problems are multiple choice questions. Only one of the five given options is correct. Please state
clearly behind which letter the correct solution is stated. You get 2 points for each correct answer.
• The B-problems are open questions, which have a number as answer. You get 5 points for each correct
answer. Please work accurately, since an error in your calculations
√ can cause your solution to be considered
wrong. Please give exact answers, for example 1181
or 2 + 3 or π + 1.
• This is a competition, not an exam. So it is to be expected that only few people will get all the answers
right, and you don’t need to worry if you have only solved a part of the problems.
• The use of formula sheets, calculators and similar tools is not allowed.
• The point is that you have fun solving unusual mathematical problems. Good luck!
A-problems
A1. Alex, Birgit, Cedric, Dion and Ersin all write their names on a sheet of paper, and they put
those five sheets into a large box. They each take one sheet out of the box at random. Now
it turns out that Birgit has Alex’ sheet, Cedric has Dion’s and Dion has Ersin’s. Also, Ersin
doesn’t have Cedric’s sheet. Whose sheet does Alex have?
A) Alex’ B) Birgit’s C) Cedric’s D) Dion’s E) Ersin’s
A2. In a magic 3 × 3 square, the three row sums, the three column sums
and the two diagonal sums are all equal to each other. (A row sum 7
being the sum of the numbers on a certain row, etc.) In the magic
3 × 3 square shown here three numbers have already been filled in. ?
What number must be filled in instead of the question mark?
A) 2 B) 4 C) 6 D) 8 E) 9
10 3
A3. Calculating 6 × 5 × 4 × 3 × 2 × 1 yields 720. How many divisors does
720 have? (A divisor of an integer n is a positive integer by which n is
divisible. For example: the divisors of 6 are 1, 2, 3 and 6; the divisors
of 11 are 1 and 11.)
A) 6 B) 8 C) 20 D) 30 E) 36
D 5
A4. Of a quadrilateral ABCD, we know that |AB| = 3, |BC| = 4, |CD| = C
5, |DA| = 6 en ∠ABC = 90◦ . (|AB| stands for the length of segment
AB, etc.) What is the area of quadrilateral ABCD? 6
√ 4
A) 16 B) 18 C) 18 21 D) 20 E) 6 + 5 11
A 3 B
13
A5. How many five-digit numbers (like 12345 or 78000; the first digit must
be non-zero) are there that end on a 4 and that are divisible by 6?
A) 1500 B) 2000 C) 3000 D) 7500 E) 8998
D C
A6. We have a square ABCD with |AB| = 3. On AB, there is a point E
such that |AE| = 1 and |EB| = 2. AC and DE intersect in H. What
is the area of triangle CDH?
9 21 27
H
A) 8 B) 2 C) 8 D) 3 E) 8
A E B
A7. The seven blocks S E T T E E S are shuffled. For example, you can
get E E E S S T T or T E S E T E S .
How many different “words” of length 7 can we get this way? (Any
combination of the 7 letters counts as word.)
A) 210 B) 420 C) 840 D) 1260 E) 5040

2
A8. How many distinct real solutions does the equation (x2 − 2)2 − 5 = 1 have?
A) 4 B) 5 C) 6 D) 7 E) 8
B-problems
B1. We number both the rows and the columns of an 8×8 chessboard with
the numbers 1 to 8. A number of grains is placed onto each square, in
such a way that the number of grains on a certain square equals the
product of its row and column numbers. How many grains are there
on the entire chessboard?
B2. We take 50 distinct integers from the set {1, 2, 3, . . . , 100}, such that
their sum equals 2900. What is the minimal number of even integers
amongst these 50 numbers?
B3. For a certain x, we have x + x1 = 5. Define n = x3 + x13 . It turns out

D C
that n is an integer.
5
Calculate n. (Give your answer using decimal notation.)
P
B4. Inside a rectangle ABCD, there is a point P with |AP | = 6, |BP | = 7

and |CP | = 5. What is the length of segment DP ?
6 7
A B
14
SE
W
AND
IS
K
U
ERL
N
D
E
NED
P IADE Solutions
OLYM
of the problems of Round 1 of the
Dutch Mathematical Olympiad 2008
A-problems
A1. C) Cedric When we put the data into a table, we see that Birgit’s
and Cedric’s sheets haven’t been picked yet. Since Ersin didn’t pick A B C D E
Cedric’s sheet, he must have picked Birgit’s. So Alex must have picked ? A D E ?
Cedric’s sheet.
A2. B) 4 See the figure. From F + 10 + 3 = F + D + 7, we get

D = 6. Then from 7 + E + 3 = C + D + E = C + 6 + E, we can deduce
A B 7
that C = 7 + 3 − 6 = 4.
C D E
A3. D) 30 The number 720 only has the prime factors 2, 3 and 5.
The prime factor 2 occurs four times (once in 2, twice in 4 and once
F 10 3
in 6), the prime factor 3 twice (once in 3 and 6), and 5 just once. The
divisors without any factors 3 or 5 are 1, 2, 4, 8 and 16. The divisors
having one factor 3 and no factors 5 are 3, 6, 12, 24, 48. And the
divisors having two factors 3 and no factors 5 are 9, 18, 36, 72 and
144. So 720 has 15 divisors that do not have factors 5. Multiplying
all of these divisors by 5 gives us the other 15 divisors, which makes
30 in total.
Alternative solution: Every divisor of 720 = 24 ×32 ×51 can be written
as 2a × 3b × 5c with 5 possibilities for a (being 0 to 4), 3 possibilities
for b (being 0 to 2) and 2 for c (being 0 and 1). So we conclude that
720 has 5 × 3 × 2 = 30 divisors.
D 5
A4. B) 18 According to Pythagoras’ Theorem, we have |AC| = C
5. So triangle ACD is isosceles with base AD. In this triangle, the 3
altitude from C divides the triangle into two triangles with sides 3, 4
4
and 5, and we can divide quadrilateral ABCD in three triangles with 3
sides 3, 4 and 5. So its area must be equal to 3 × 6 = 18.
A 3 B
A5. C) 3000 If x is a positive multiple of 6 that ends with a 4, then the next multiples of
6 end with a 0 (x + 6), a 6 (x + 12), a 2 (x + 18), an 8 (x + 24), a 4 (x + 30), so the next
multiple of 6 that ends with a 4 is x+30. So any 30 consecutive positive integers must contain
exactly one integer with the desired properties. How many such integers lie between 10000
and 99999? Since we have 90000 consecutive positive integers, we find 90000 ÷ 30 = 3000
such integers amongst them.
D Q C
27
A6. E) 8 Draw a line through H parallel to AD, and let P and Q
be the intersections of this line with AB and CD, respectively. Now
we have |HP | : |HQ| = |AE| : |CD| = 1 : 3, so |HQ| = 34 × |P Q| =
3 9 1 9 27 H
4 × 3 = 4 . So the area of triangle CDH equals 2 × 3 × 4 = 8 .
A PE B
15
S S . . . . .
We have 6 + 5 + 4 + 3 + 2 + 1 = 21 (or 72 ) possibilities

A7. A) 210 S . S . . . .
to arrange the S-blocks onto the 7 places (see figure). For each choice, S . . S . . .
we have 4 + 3 + 2 + 1 = 10 (or 52 ) possibilities to arrange the T-blocks ..
.
on the remaining 5 places; after which the positions of the E-blocks . S S . . . .
are determined. So we have 21 × 10 = 210 possibilities.
. S . S . . .
Alternative solution: If all the blocks were different, we would have got
. S . . S . .
7! possibilities. But the three E-blocks aren’t different, so we end up ..
counting each word 3! times this way. Similarly for the two S-blocks .
7!
and the two T-blocks. So we find 3!×2!×2! = 7×6×5×4×6 3!
= 7×6×5 = . . S S . . .
63!×4
210 different words. . . S . S . .
..
.
A8. B) 5 This equation is equivalent to
(x2 − 2)2 − 5 = 1 or (x2 − 2)2 − 5 = −1.
√ √
The first is equivalent to x2 − √
2 = 6 of x2 − 2 = − 6, with 2 and 0
solutions respectively (since − 6 + 2 < 0).
The latter is equivalent to x2 − 2 = 2 of x2 − 2 = −2, with 2 and 1
solution(s) respectively. So we have 2 + 0 + 2 + 1 = 5 solutions in total.
B-problems
B1. 1296 In the first column, we have, successively, 1 × 1, 1 × 2, 1 × 3, . . . , 1 × 8 grains.
So, in the first column: 1 × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
In the second column: 2 × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
In the third column is: 3 × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
..
.
Finally, in the eighth column: 8 × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
So, in total: (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8) × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
Since 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 = 21 × 8 × (1 + 8) = 36, there are 362 = 1296 grains on the
board.
B2. 6 The 50 odd integers from the set {1, 2, 3, . . . , 100} sum up to 12 × 50 × (1 + 99) =
2500, which is still 400 short of 2900. Now exchange the smallest odd integers for the largest
even integers, in pairs, since 400 is even. First exchanging 1 and 3 for 100 and 98 makes the
sum equal to 2694. The next step gives us 2694 − 5 − 7 + 96 + 94 = 2872. Which is still less
than 2900, so we require another exchange. Now exchanging 9 and 11 for 20 and 28 works,
making the sum 2900 with 6 even integers, showing along the way that that is the minimal
number of even integers we need to do so.
1
B3. 110 We know that x must satisfy x + = 5, so x2 − 5x + 1 =
√x
5± 21
0, from which follows that x = x1,2 = 2 . Note that x1 x2 = 1.
√ 3 √ 3
Now we have x3 + x−3 = x31 + x32 = 1
8 5 + 21 + 5 − 21 =
1
√ √ 2 √ 3
√ √ 2 √ 3

8 53 + 3 · 52 · 21 + 3 · 5 · 21 + 21 + 53 − 3 · 52 · 21 + 3 · 5 · 21 − 21 =
2 3

8 5 + 3 · 5 · 21 = 110.
Alternative solution: From (x + x1 )3 = x3 + 3x2 ( x1 ) + 3x( x1 )2 + ( x1 )3 = x3 + 3x + x3 + x13 we
can deduce that x3 + x13 = (x + x1 )3 − 3(x + x1 ) = 53 − 3 × 5 = 110.
S
√ D C
B4. 2 3 Let Q, R, S, T be the orthogonal projections of P on
AB, BC, CD, DA, respectively. Then we have P
|AQ|2 + |QP |2 = 36 and |BQ|2 + |SP |2 = 25 (since |BQ| = |CS|), T R
2 2 2 2 2
so |AQ| + |QP | + |BQ| + |SP | = 61. We also have |BQ| + |QP | = 2
49. √
So |DS| 2 + |SP |2 = |AQ|2 + |SP |2 = 61 − 49 = 12 and |DP | = 12 =
√
2 3. A B
Q
16
The 47th
Dutch
Mathematical
Olympiad
2008
and the team selection
for IMO 2009 Bremen

SE
W
AND
IS
K
U
ERL
N
D
E
NED
PIADE
OLYM
Contents
Introduction i

Solutions 3
Second Round, September 2008 5

Solutions 6
Benelux Mathematical Olympiad, May 2009 8

Solutions 9
Team Selection Test, June 2009 13

Solutions 14
Junior Mathematical Olympiad part 1, October 2008 17

part 2 19

Solutions 23
Board of the Training committee

Dutch Mathematical 2008–2009
Olympiad Foundation
Frits Beukers, president Quintijn Puite, chair

Melanie Steentjes, secretary Birgit van Dalen, vice chair
Tom Verhoeff, treasurer Adriana Gabor
Alex van den Brandhof Johan Konter
Jaap de Jonge Julian Lyczak
Quintijn Puite René Sitters
Sietske Tacoma
Lenny Taelman
We thank our main sponsors:

Introduction
The Dutch Mathematical Olympiad consists of two rounds. The first round is held on the
participating schools and consists of eight multiple choice questions and four open questions
(see page 1–4). Students get two hours to work on the paper. On January 25, 2008, in total
3004 students of 201 secondary schools participated in this first round.
The best students are invited for the second round. For the first time we set different thresh-
olds for students from different grades, in order to stimulate young students to participate.
Those students from grade 5 (4, ≤ 3) that scored 26 (22, 18) points or more on the first round
(out of a maximum of 36 points) were invited to the second round. Also some outstanding
participants in the Kangaroo math contest or the Pythagoras Olympiad were invited.
As a new initiative, we organized training sessions at six universities in the country for the
143 students who had been invited for the second round in September 2009. Former Dutch
IMO-participants are involved at each of the universities.
From those 143, in total 123 participated in the second round on September 12th, 2008 at
the Eindhoven University of Technology. This final round contains five problems for which
the students have to give extensive solutions and proofs. They have three hours for the paper
(see page 5–7). After the prices had been awarded in the beginning of November, the Dutch
Mathematical Olympiad concluded its 47th edition 2008. In 2011 we will have our 50th
edition.
The 31 most outstanding candidates of the Dutch Mathematical Olympiad 2008 were invited
to an intensive seven-month training programme, consisting of weekly problem sets. Also,
the students met twice for a three-day training camp, three times for a day at the university,
and finally for a six-day training camp in the beginning of June.
In May, ten of them were invited to participate to the first Benelux Mathematical Olympiad,
held in Bergen op Zoom, the Netherlands (see page 8–12). We think this has been a very
stimulating experience for our students.
In June, 24 out of these 31 candidates were left. Out of those, the team was selected by a
final selection test on June 13, 2009 (see page 13–16). The team will have a training camp
from July 6 until July 13, together with the team from New Zealand.
In the meantime, a new edition of the Dutch Mathematical Olympiad had started. In October
2008, at the VU University Amsterdam we organized the first Junior Mathematical Olympiad
for the winners of the popular Kangaroo math contest (see page 17–20). The 30 best students
of grade 2, grade 3 and grade 4 were invited. We hope all of them will enjoy those problems
so much, that they will participate to the first round in the years that come.
The new first round took place at participating schools on January 30, 2009 (see page 21–24).
We are very happy to see that the number of schools as well as the number of participants
has increased again: 230 schools and 4379 participants, a new record since ages!
i
The Dutch team for IMO 2009 Bremen consists of
• Wouter Berkelmans (18 y.o., participated in IMO 2006, 2007 as well)
• Raymond van Bommel (17 y.o., participated in IMO 2007, 2008 as well)
• Harm Campmans (17 y.o.)
• Saskia Chambille (18 y.o.)
• David Kok (16 y.o.)
• Maarten Roelofsma (18 y.o., participated in IMO 2008 as well)
As a promising young student, we bring
• Merlijn Staps (14 y.o.)
The team is coached by
• Quintijn Puite (team leader), Eindhoven University of Technology
• Birgit van Dalen (deputy leader), Leiden University
The Dutch delegation for IMO 2009 Bremen further consists of
• Wim Berkelmans (member AB, observer A), VU University Amsterdam
• Hans van Duijn (observer A), Eindhoven University of Technology
• Tom Verhoeff (observer A), Eindhoven University of Technology
• Gerhard Wöginger (observer A), Eindhoven University of Technology
• Ronald van Luijk (observer A), Leiden University
• Jelle Loois (observer B), Ortec
• Anick van de Craats (observer C), Netherlands Forensic Institute
• Lidy and Theo Wesker (observer C), University of Amsterdam
• Karst Koymans, University of Amsterdam
• Freek van Schagen, VU University Amsterdam
• Rob Wieleman, Movisie Utrecht
• Wendoline Timmerman, Ministry of Education, Culture and Science
We are grateful to Jinbi Jin for the translation into English of most of the problems and the solutions.
We also thank Lieneke Notenboom-Kronemeijer for her useful remarks concerning the formulation of
the First Round 2009 paper.
ii
First Round
SE
W
AND
IS
Dutch Mathematical Olympiad K
U
ERL
N
D
E
NED
PIADE
Problems OLYM
• Time available: 2 hours.
• The A-problems are multiple choice questions. Only one of the five options given is correct.
Please state clearly which letter precedes the correct solution. Each correct answer is worth 2
points.
• The B-problems are open questions; the answers to these questions are a number, or numbers.
Each correct answer is worth 5 points. Please work accurately, since an error in your calculations
may cause your solution to be considered incorrect and then you won’t√get points at all for
11
that question. Please give your answers exactly, for example 81 or 2 + 12 5 or 14 π + 1.
• You are not allowed to use calculators and formula sheets; you can only use a pen, a compass
and a ruler or set square. And your head, of course.
• This is a competition, not an exam. The main thing is that you have fun solving unusual
mathematical problems. Good luck!
A-problems
A1. Alex, Birgit, Cedric, Dion and Ersin all write their names on a sheet of paper, and they put
those five sheets into a large box. They each take one sheet out of the box at random. Now
it turns out that Birgit has Alex’ sheet, Cedric has Dion’s and Dion has Ersin’s. Also, Ersin
doesn’t have Cedric’s sheet. Whose sheet does Alex have?
A) Alex’ B) Birgit’s C) Cedric’s D) Dion’s E) Ersin’s
A2. In a magic 3 × 3 square, the three row sums, the three column sums
and the two diagonal sums are all equal to each other. (A row sum 7
being the sum of the numbers on a certain row, etc.) In the magic
3 × 3 square shown here three numbers have already been filled in. ?
What number must be filled in instead of the question mark?
A) 2 B) 4 C) 6 D) 8 E) 9
10 3
A3. Calculating 6 × 5 × 4 × 3 × 2 × 1 yields 720. How many divisors does
720 have? (A divisor of an integer n is a positive integer by which n is
divisible. For example: the divisors of 6 are 1, 2, 3 and 6; the divisors
of 11 are 1 and 11.)
A) 6 B) 8 C) 20 D) 30 E) 36
D 5
A4. Of a quadrilateral ABCD, we know that |AB| = 3, |BC| = 4, |CD| = C
5, |DA| = 6 en ∠ABC = 90◦ . (|AB| stands for the length of segment
AB, etc.) What is the area of quadrilateral ABCD? 6
√ 4
A) 16 B) 18 C) 18 21 D) 20 E) 6 + 5 11
A 3 B
1
A5. How many five-digit numbers (like 12345 or 78000; the first digit must
be non-zero) are there that end on a 4 and that are divisible by 6?
A) 1500 B) 2000 C) 3000 D) 7500 E) 8998
D C
A6. We have a square ABCD with |AB| = 3. On AB, there is a point E
such that |AE| = 1 and |EB| = 2. AC and DE intersect in H. What
is the area of triangle CDH?
9 21 27
H
A) 8 B) 2 C) 8 D) 3 E) 8
A E B
A7. The seven blocks S E T T E E S are shuffled. For example, you can
get E E E S S T T or T E S E T E S .
How many different “words” of length 7 can we get this way? (Any
combination of the 7 letters counts as word.)
A) 210 B) 420 C) 840 D) 1260 E) 5040

2
A8. How many distinct real solutions does the equation (x2 − 2)2 − 5 = 1 have?
A) 4 B) 5 C) 6 D) 7 E) 8
B-problems
B1. We number both the rows and the columns of an 8×8 chessboard with
the numbers 1 to 8. A number of grains is placed onto each square, in
such a way that the number of grains on a certain square equals the
product of its row and column numbers. How many grains are there
on the entire chessboard?
B2. We take 50 distinct integers from the set {1, 2, 3, . . . , 100}, such that
their sum equals 2900. What is the minimal number of even integers
amongst these 50 numbers?
B3. For a certain x, we have x + x1 = 5. Define n = x3 + x13 . It turns out

D C
that n is an integer.
5
Calculate n. (Give your answer using decimal notation.)
P
B4. Inside a rectangle ABCD, there is a point P with |AP | = 6, |BP | = 7

and |CP | = 5. What is the length of segment DP ?
6 7
A B
2
First Round
SE
W
AND
IS
U
ERL
N
D
E
NED
PIADE
Solutions OLYM
A1. C) Cedric When we put the data into a table, we see that Birgit’s
and Cedric’s sheets haven’t been picked yet. Since Ersin didn’t pick A B C D E
Cedric’s sheet, he must have picked Birgit’s. So Alex must have picked ? A D E ?
Cedric’s sheet.
A2. B) 4 See the figure. From F + 10 + 3 = F + D + 7, we get

D = 6. Then from 7 + E + 3 = C + D + E = C + 6 + E, we can deduce
A B 7
that C = 7 + 3 − 6 = 4.
C D E
A3. D) 30 The number 720 only has the prime factors 2, 3 and 5.
The prime factor 2 occurs four times (once in 2, twice in 4 and once
F 10 3
in 6), the prime factor 3 twice (once in 3 and 6), and 5 just once. The
divisors without any factors 3 or 5 are 1, 2, 4, 8 and 16. The divisors
having one factor 3 and no factors 5 are 3, 6, 12, 24, 48. And the
divisors having two factors 3 and no factors 5 are 9, 18, 36, 72 and
144. So 720 has 15 divisors that do not have factors 5. Multiplying
all of these divisors by 5 gives us the other 15 divisors, which makes
30 in total.
Alternative solution: Every divisor of 720 = 24 ×32 ×51 can be written
as 2a × 3b × 5c with 5 possibilities for a (being 0 to 4), 3 possibilities
for b (being 0 to 2) and 2 for c (being 0 and 1). So we conclude that
720 has 5 × 3 × 2 = 30 divisors.
D 5
A4. B) 18 According to Pythagoras’ Theorem, we have |AC| = C
5. So triangle ACD is isosceles with base AD. In this triangle, the 3
altitude from C divides the triangle into two triangles with sides 3, 4
4
and 5, and we can divide quadrilateral ABCD in three triangles with 3
sides 3, 4 and 5. So its area must be equal to 3 × 6 = 18.
A 3 B
A5. C) 3000 If x is a positive multiple of 6 that ends with a 4, then the next multiples of
6 end with a 0 (x + 6), a 6 (x + 12), a 2 (x + 18), an 8 (x + 24), a 4 (x + 30), so the next
multiple of 6 that ends with a 4 is x+30. So any 30 consecutive positive integers must contain
exactly one integer with the desired properties. How many such integers lie between 10000
and 99999? Since we have 90000 consecutive positive integers, we find 90000 ÷ 30 = 3000
such integers amongst them.
D Q C
27
A6. E) 8 Draw a line through H parallel to AD, and let P and Q
be the intersections of this line with AB and CD, respectively. Now
we have |HP | : |HQ| = |AE| : |CD| = 1 : 3, so |HQ| = 34 × |P Q| =
3 9 1 9 27 H
4 × 3 = 4 . So the area of triangle CDH equals 2 × 3 × 4 = 8 .
A PE B
3
S S . . . . .
We have 6 + 5 + 4 + 3 + 2 + 1 = 21 (or 72 ) possibilities

A7. A) 210 S . S . . . .
to arrange the S-blocks onto the 7 places (see figure). For each choice, S . . S . . .
we have 4 + 3 + 2 + 1 = 10 (or 52 ) possibilities to arrange the T-blocks ..
.
on the remaining 5 places; after which the positions of the E-blocks . S S . . . .
are determined. So we have 21 × 10 = 210 possibilities.
. S . S . . .
Alternative solution: If all the blocks were different, we would have got
. S . . S . .
7! possibilities. But the three E-blocks aren’t different, so we end up ..
counting each word 3! times this way. Similarly for the two S-blocks .
7!
and the two T-blocks. So we find 3!×2!×2! = 7×6×5×4×6 3!
= 7×6×5 = . . S S . . .
63!×4
210 different words. . . S . S . .
..
.
A8. B) 5 This equation is equivalent to
(x2 − 2)2 − 5 = 1 or (x2 − 2)2 − 5 = −1.
√ √
The first is equivalent to x2 − √
2 = 6 of x2 − 2 = − 6, with 2 and 0
solutions respectively (since − 6 + 2 < 0).
The latter is equivalent to x2 − 2 = 2 of x2 − 2 = −2, with 2 and 1
solution(s) respectively. So we have 2 + 0 + 2 + 1 = 5 solutions in total.
B1. 1296 In the first column, we have, successively, 1 × 1, 1 × 2, 1 × 3, . . . , 1 × 8 grains.
So, in the first column: 1 × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
In the second column: 2 × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
In the third column is: 3 × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
..
.
Finally, in the eighth column: 8 × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
So, in total: (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8) × (1 + 2 + 3 + 4 + 5 + 6 + 7 + 8).
Since 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 = 21 × 8 × (1 + 8) = 36, there are 362 = 1296 grains on the
board.
B2. 6 The 50 odd integers from the set {1, 2, 3, . . . , 100} sum up to 12 × 50 × (1 + 99) =
2500, which is still 400 short of 2900. Now exchange the smallest odd integers for the largest
even integers, in pairs, since 400 is even. First exchanging 1 and 3 for 100 and 98 makes the
sum equal to 2694. The next step gives us 2694 − 5 − 7 + 96 + 94 = 2872. Which is still less
than 2900, so we require another exchange. Now exchanging 9 and 11 for 20 and 28 works,
making the sum 2900 with 6 even integers, showing along the way that that is the minimal
number of even integers we need to do so.
1
B3. 110 We know that x must satisfy x + = 5, so x2 − 5x + 1 =
√x
5± 21
0, from which follows that x = x1,2 = 2 . Note that x1 x2 = 1.
√ 3 √ 3
Now we have x3 + x−3 = x1 + x32
3 = 1
85 + 21 + 5 − 21 =
1
√ √ 2 √ 3 3 √ √ 2 √ 3
8 53 + 3 · 52 · 21 + 3 · 5 · 21 + 21 + 5 − 3 · 52 · 21 + 3 · 5 · 21 − 21 =
2 3

8 5 + 3 · 5 · 21 = 110.
Alternative solution: From (x + x1 )3 = x3 + 3x2 ( x1 ) + 3x( x1 )2 + ( x1 )3 = x3 + 3x + 3
x + 1
x3
we
can deduce that x3 + x13 = (x + x1 )3 − 3(x + x1 ) = 53 − 3 × 5 = 110.
S
√ D C
B4. 2 3 Let Q, R, S, T be the orthogonal projections of P on
AB, BC, CD, DA, respectively. Then we have P
|AQ|2 + |QP |2 = 36 and |BQ|2 + |SP |2 = 25 (since |BQ| = |CS|), T R
so |AQ|2 + |QP |2 + |BQ|2 + |SP |2 = 61. We also have |BQ|2 + |QP |2 =
49. √
2 2 2 2
So√ |DS| + |SP | = |AQ| + |SP | = 61 − 49 = 12 and |DP | = 12 =
2 3. A B
Q
4
Second Round
SE
W
AND
Dutch Mathematical Olympiad IS
K
U
ERL
N
D
Friday, September 12, 2008 E
NED
Eindhoven University of Technology
PIADE
OLYM
Problems
• Available time: 3 hours.
• Writing down just the answer itself is not sufficient; you also need to describe the way you
solved the problem.
• Usage of calculators and formula sheets are not allowed; you are only allowed to use a pen, a
compass and a ruler or set square. And your common sense of course.
• Please write the solutions of each problem on a different sheet of paper. Good luck!
1. Suppose we have a square ABCD and a point S in the interior of this square. A D
A D
Under homothety with centre S and ratio of magnification k > 1, this square
becomes another square A0 B 0 C 0 D0 . S
Prove that the sum of the areas of the two quadrilaterals A0 ABB 0 and
C 0 CDD0 are equal to the sum of the areas of the two quadrilaterals B 0 BCC 0 B C
and D0 DAA0 . B C
2. Find all positive integers (m, n) such that
3 · 2n + 1 = m2 .
3. Suppose that we have a set S of 756 arbitrary integers between 1 and 2008
(1 and 2008 included).
Prove that there are two distinct integers a and b in S such that their sum
a + b is divisible by 8.
C2
4. Three circles C1 , C2 , C3 , with radii 1, 2, 3 respectively, are externally tangent. C1
In the area enclosed by these circles, there is a circle C4 which is externally
tangent to all three circles.
Find the radius of C4 .
C3
5. We’re playing a game with a sequence of 2008 non-negative integers. A move

consists of picking a integer b from that sequence, of which the neighbours
a and c are positive. We then replace a, b and c by a − 1, b + 7 and c − 1
respectively. It is not allowed to pick the first or the last integer in the
sequence, since they only have one neighbour.
If there is no integer left such that both of its neighbours are positive, then
there is no move left, and the game ends.
Prove that the game always ends, regardless of the sequence of integers we
begin with, and regardless of the moves we make.
5
Second Round
SE
W
AND
Dutch Mathematical Olympiad IS
K
U
ERL
N
D
Friday, September 12, 2008 E
NED
Eindhoven University of Technology
PIADE
OLYM
Solutions
1. Let p = |AB| and q = |A0 B 0 |, so q = k · p. Note that the sides AB and

A0 B 0 are parallel, because of the homothety. Hence quadrilateral A0 ABB 0 is
a trapezium, so its area equals the sum of the area of the triangles 4ABB 0 A h4 D
A D
and 4A0 B 0 A, so it is equal to 12 · |AB| · h1 + 12 · |A0 B 0 | · h1 = p+q
2 h1 , where q p
h1 is the distance between the parallel lines AB and A0 B 0 . Similarly, we see h1 S h3
that the area of quadrilateral C 0 CDD0 is equal to p+q 2 h3 , where h3 is the
distance between CD and C 0 D0 . Hence the area of the two trapezia together B h2 C
is p+q
2 (h1 + h3 ). In the same way, we see that the area of the two trapezia B C
A0 ADD0 and B 0 BCC 0 together is equal to p+q 2 (h2 + h4 ). Now note that
h1 + h3 = q − p = h2 + h4 , so the two areas are equal.
2. We can rewrite the equation as 3 · 2n = (m − 1)(m + 1). Since n > 0, we see that 3 · 2n is an even
number, so at least one of m − 1, m + 1 is even as well. Hence they’re both even. As the factors
m − 1 and m + 1 differ by 2, they can’t both contain multiple factors of 2. Hence one of these
factors contains exactly one factor 2. This factor either contains the factor 3 as well, or it doesn’t.
So it must be equal to either 2 or 6. The other factor then must differ from this one by exactly 2.
If this factor is equal to 2, then the other one has to be either 0 or 4. Since none of these contain
a factor 3, none of these solve the equation. If this factor is equal to 6, then the other one has to
be either 4 or 8, both of which yield a solution, as they are powers of 2. Hence there are exactly
two solutions, given by (m, n) = (5, 3) and (m, n) = (7, 4).
3. We divide the integers 1 up to 2008 amongst eight distinct subsets V1 to V8 , where Vi is the subset
consisting of the 251 integers of the form 8k + i with 0 ≤ k ≤ 250. So we have
V1 = {1, 9, . . . , 2001}, ..., V8 = {8, 16, . . . , 2008}.
The union of these subsets consists of the positive integers up to 2008.

Now suppose for a contradiction that there are no two integers a, b such that a + b is divisible by
8. Then let us consider the distribution of S among the Vi . Note that the sum of two multiples of
8 is again a multiple of 8, so we see that V8 then cannot contain more than 1 element of S. Since
the sum of two integers from V4 is a multiple of 8, V4 cannot contain more than 1 element of S
either. Next, we see that the sum of a integer from V1 and one from V7 , is a multiple of 8, so at
least one of them contains no elements of S. Similarly, we see that at least one of V2 and V6 , and
at least one of V3 and V5 contain no elements of S at all. Now note that every Vi contains at most
251 elements of S. Hence V1 , V2 , V3 , V5 , V6 , V7 together contain at most 3 · 251 = 753 elements
of S. So the sets V1 to V8 together contain at most 755 elements of S, which is a contradiction,
since S consists of 756 elements, and each element is contained in a certain Vi .
6
4. Let A, B, C be the centres of the circles C1 , C2 , C3 respectively. The triangle
formed by these three centres, has sides |AB| = 3, |AC| = 4 and |BC| = 5,
hence it is a right-angled triangle. Now choose the x- and the y-axis in such C
a way that A = (0, 0), B = (3, 0) and C = (0, 4). Let r and M = (x, y) be
the radius and centre of C4 , respectively. Then we see that |AM | = r + 1,
y
|BM | = r + 2 and |CM | = r + 3. This implies the three following equations A x B
with three unknowns.
r2 + 2r + 1 = |AM |2 = x2 + y 2 (1)
2 2 2 2 2 2
r + 4r + 4 = |BM | = (3 − x) + y = x − 6x + 9 + y (2)
2 2 2 2 2 2
r + 6r + 9 = |CM | = x + (4 − y) = x + y − 8y + 16 (3)
3−r
Taking the difference of (1) and (2), we see that 6x − 9 = −2r − 3, so 6x = 6 − 2r, hence x = 3 .
2−r
Taking the difference of (1) and (3), we see that 8y − 16 = −4r − 8, so 8y = 8 − 4r, hence y = 2 .
2 2 2 2
Substituting this in (1) yields r2 + 2r + 1 = x2 + y 2 = (3−r)
9 + (2−r)
4 = 9−6r+r
9 + 4−4r+r
4 , thus
23 2 11 r
2 r
r 2
36 r + 3 r −1 = 23 6 +22 6 −1 = 0. Substituting p = 6 , we then see that 0 = 23p +22p−1 =
(23p − 1)(p + 1). This gives the two possible solutions 6r = p = −1 and 23 · 6r = 23p = 1. But
6
since r has to be positive, we see that r has to be equal to 23 . (We can deduce from this that
21 20
M = (x, y) = ( 23 , 23 ).)
5. Consider an arbitrary sequence n1 , n2 , . . . , n2008 and an arbitrary sequence of moves. The first
integer n1 is reduced by 1 every time we pick b = n2 , since n1 is one of the neighbours of n2 .
If we pick b = nk , where k > 2, then we see that n1 remains unchanged. Hence we can pick
b = n2 at most n1 times. So in our sequence of moves, either there is a last time that the
move b = n2 occurs, or that move doesn’t occur at all. In the former case, we consider the
sequence of moves following that move, and in the latter case, we simply consider the sequence
of all moves. In this new sequence of moves, we never pick b = n2 . So all we do in this new
sequence, is picking b = n3 up to b = n2007 . Let n2 now be the value of the second integer at
the beginning of this new sequence (this integer need not be the same as before, since every
time we’ve picked b = n2 until then, it is increased by 7, and every time we’ve picked b = n3
until then, it is reduced by 1). In the new sequence, n2 is reduced by 1 every time we pick
b = n3 , and remains unchanged if we pick any other integer (i.e. b = n4 up to b = n2007 ).
Hence we can conclude that in the new sequence, either there is a last time that the move
b = n3 occurs, or this move doesn’t occur at all. So from that point onward, we only pick the
integers b = n4 up to b = n2007 . We can repeat this argument to see that, from some point
onward, only b = n2007 is picked, and that there either is a last time that the move b = n2007 oc-
curs, or this move doesn’t occur at all, indicating that our sequence of moves ends at some point.
Variant Suppose that there exists an infinite sequence of moves. By the pigeonhole
principle, there is a move b = nk that occurs infinitely often. Now let k be the smallest
integer for which the move b = nk occurs infinitely often. Then we see that the integer
b = nk−1 is only picked finitely often, hence is increased only finitely often. But this integer
is decreased infinitely often; namely for every time we pick b = nk . This yields our contradiction.
Alternative solution For each sequence of 2008 integers n1 , n2 , . . . , n2008 , we compute

the weighted sum S = 7 · n1 + 72 · n2 + 73 · n3 + · · · + 72007 · n2007 + 72008 · n2008 . If we replace a, b, c
in this sequence by a − 1, b + 7, c − 1, where b = nk for a certain integer k such that 2 ≤ k ≤ 2007,
then S becomes equal to S − 7k−1 + 7 · 7k − 7k+1 = S − 7k−1 , so every move reduces the value of
S. On the other hand, we note that S is a sum of non-negative integers, hence is a non-negative
integer itself. Writing down the new value of S after each move, we get a decreasing sequence of
non-negative integers. Such a sequence can not be infinite. Hence the game ends after a finite
number of moves.
7
1st BENELUX MATHEMATICAL OLYMPIAD
Bergen op Zoom (Netherlands)
May 9, 2009
Language: English
Problem 1. Find all functions f : Z>0 → Z>0 that satisfy the following two conditions:
• f (n) is a perfect square for all n ∈ Z>0 ;
• f (m + n) = f (m) + f (n) + 2mn for all m, n ∈ Z>0 .
Problem 2. Let n be a positive integer and let k be an odd positive integer. Moreover, let
a, b and c be integers (not necessarily positive) satisfying the equations
an + kb = bn + kc = cn + ka.
Prove that a = b = c.
Problem 3. Let n ≥ 1 be an integer. In town X there are n girls and n boys, and each girl
knows each boy. In town Y there are n girls, g1 , g2 , . . . , gn , and 2n − 1 boys, b1 , b2 , . . . ,
b2n−1 . For i = 1, 2, . . . , n, girl gi knows boys b1 , b2 , . . . , b2i−1 and no other boys. Let r be an
integer with 1 ≤ r ≤ n. In each of the towns a party will be held where r girls from that town
and r boys from the same town are supposed to dance with each other in r dancing pairs.
However, every girl only wants to dance with a boy she knows. Denote by X(r) the number
of ways in which we can choose r dancing pairs from town X, and by Y (r) the number of
ways in which we can choose r dancing pairs from town Y .
Prove that X(r) = Y (r) for r = 1, 2, . . . , n.
Problem 4. Given trapezoid ABCD with parallel sides AB and CD, let E be a point on
line BC outside segment BC, such that segment AE intersects segment CD. Assume that
there exists a point F inside segment AD such that ∠EAD = ∠CBF . Denote by I the point
of intersection of CD and EF , and by J the point of intersection of AB and EF . Let K be
the midpoint of segment EF , and assume that K is different from I and J.
Prove that K belongs to the circumcircle of 4ABI if and only if K belongs to the circumcircle
of 4CDJ.
Time allowed: 4 hours and 30 minutes

Each problem is worth 7 points
We are grateful to the authors of these problems for making them available for the Benelux Olympiad. We have
agreed upon not publishing them on any forum or by any other means before August 1st, 2009. Participating
to the BxMO entails that you respect this agreement.
8
1st BENELUX MATHEMATICAL OLYMPIAD
Bergen op Zoom (Netherlands)
May 9, 2009
Language: English
Solutions
Problem 1.
Solution 1. Let a be a positive integer satisfying f (1) = a2 . We will prove that f (n) =
na2 + n(n − 1) for all n by induction on n. For n = 1 it follows from the definition of a. Now
suppose we have f (n) = na2 + n(n − 1) for a certain positive integer n. Then by the second
condition with m = 1, we have
f (n + 1) = f (n) + f (1) + 2n = na2 + n(n − 1) + a2 + 2n = (n + 1)a2 + n(n + 1).
This completes the induction.
Suppose a > 1 and let p be a prime divisor of a. We know that f (p) = pa2 + p(p − 1) is a
square, and as it is obviously divisible by p, it must be divisible by p2 . Hence a2 + p − 1 is
divisible by p. But this is a contradiction, as a and p are both divisible by p.
We conclude a = 1 and f (n) = n2 for all n ∈ Z>0 . This function indeed satisfies the condi-
tions.
Comment. Proving that a = 1 can be done in various ways. Here is another possibility. We
have f (n) = na2 + n(n − 1) = n(a2 − 1) + n2 for all n > 0. Suppose a > 1, then n = a2 − 1 > 0
and f (a2 − 1) = 2(a2 − 1)2 . As all prime factors in (a2 − 1)2 occur an even number of times,
the prime factor 2 occurs an odd number of times in f (a2 − 1) = 2(a2 − 1)2 . Hence this is
not a square, which contradicts the first condition.
Solution 2. Define g(n) = f (n)−n2 for all n ∈ Z>0 . Then we can rewrite the second condition
as
g(m + n) + (m + n)2 = g(m) + m2 + g(n) + n2 + 2mn,
hence g(n) satisfies the functional equation
g(m + n) = g(m) + g(n) for all m, n ∈ Z. (4)
Let b = g(1) = f (1) − 1 ≥ 0. By setting n = 1 in (4) we find g(m + 1) = g(m) + b, and hence
by induction we have g(n) = nb for all n. Therefore f (n) = nb + n2 for all n. Suppose b > 0,
then taking n = b yields f (b) = 2b2 , which is not a square, contradicting the first condition.
We conclude b = 0 and f (n) = n2 for all n ∈ Z>0 . This function indeed satisfies the condi-
tions.
9
Problem 2.
Solution 1. First suppose a = b. From an + kb = bn + kc it then follows that b = c, which

means we are done. Similarly, we are done if b = c or c = a. Now suppose a 6= b, b 6= c en
c 6= a. We will derive a contradiction.
From an + kb = bn + kc we find an − bn = k(c − b). In a similar way we find two more such
equations, so we have
an − bn = k(c − b), bn − cn = k(a − c), cn − an = k(b − a). (5)
Multiply these three equations and divide by (a − b)(b − c)(c − a):
an − bn bn − cn cn − an an − bn bn − cn cn − an
· · = · · = −k 3 . (6)
a−b b−c c−a b−c c−a a−b
Note that the left-hand side is the product of three integers.
Suppose n is odd. Then xn is a monotonically increasing function of x. So if a < b, then

n −bn
an < bn , and so on. Hence a a−b is positive and the same holds for the other two factors at
the left of (6). Contradiction with −k 3 < 0. We conclude that n is even.
Consider a, b and c modulo 2. According to the box principle, at least two of those are
n −bn
congruent, say a ≡ b mod 2. The integer a a−b = an−1 + an−2 b + · · · + abn−2 + bn−1 is the
sum of n terms, and as a ≡ b mod 2, all of these terms are even, or all of these terms are
odd. In any case, the sum of the n terms is even. So the left-hand side of (6) is even, while
the right-hand side is odd. This is a contradiction, which finishes the proof.
Solution 2. As in the first solution, we assume a 6= b, b 6= c en c 6= a and we find
an − bn = k(c − b), bn − cn = k(a − c), cn − an = k(b − a). (5)
First consider the case that n is odd. Then xn is a monotonically increasing function of x. So
if a < b, then an < bn , and so on. So suppose a > b, then we have an > bn , hence an − bn > 0.
Hence by (5) we have c − b > 0. So bn − cn < 0 and therefore a − c < 0. From this we have
cn − an > 0, so b − a > 0, which contradicts the assumption. Similarly we get a contradiction
if a < b.
Now assume n is even. Consider the equalities in (5) modulo 2. As k is odd, the first
n −an
equality gives a − b ≡ c − b mod 2, hence a ≡ c mod 2. Now the integer c c−a =
cn−1 + cn−2 a + · · · + can−2 + an−1 is the sum of n terms, and as a ≡ c mod 2, all of these
terms are even, or all of these terms are odd. In any case, the sum of the n terms is even. Let
i be the exponent of the prime factor 2 in c − a. Then the exponent of the prime factor 2 in
cn − an is at least i+ 1. From the third equality in (5) we derive that the exponent of 2 in b − a
is at least i + 1 as well. Similarly to the previous argument, the exponent of 2 in an − bn must
be at least i + 2, and the same holds for the exponent of 2 in c − b. Finally the exponent of
2 in bn −cn and hence also in a−c must be at least i+3. This contradicts the definition of i.
10
Problem 3.
As X(r) and Y (r) are dependent on n, we will from now on denote them by Xn (r) and Yn (r).
There are nr ways to pick r girls from town X, and nr ways to pick r boys from town X,

and r! ways to make r pairs of these boys and girls. As all girls in town X know all boys in
town X, each girl is then with a boy she knows. Hence
2
n (n!)2
Xn (r) = · r! = .
r r!((n − r)!)2
Let An (r) be the number of different ways in which r girls from town Y can dance with r
boys from town Y , forming r pairs, each girl with a boy she knows, such that gn is one of the
girls in the pairs. Let Bn (r) = Yn (r) − An (r).
If girl gn is not in one of the pairs, then boys b2n−2 and b2n−1 are not in one of the pairs
either. So for n ≥ 2 and r ≤ n − 1 we have Bn (r) = Yn−1 (r).
On the other hand, if girl gn is in one of the pairs, then we can delete that pair to find a way
in which r −1 girls from a town with n−1 girls can dance with r −1 boys from the same town.
Given such a set of r − 1 pairs, we can extend this set to a set of r pairs from a town with
n girls by adding the pair (gn , bi ) for some i. For bi we can choose from {b1 , b2 , . . . , b2n−1 }
except that the r − 1 boys that are already in one of the pairs are not allowed. So there are
(2n − 1) − (r − 1) = 2n − r possibilities for bi . We conclude that for n ≥ 2 and r ≥ 2 we have
An (r) = (2n − r)Yn−1 (r − 1).
We will now prove by induction on n that for r = 1, 2, . . . , n we have
(n!)2
Yn (r) = , (7)
r!((n − r)!)2
which finishes the proof.
For n = 1 we just need to prove that Y1 (1) = 1. As there is only one girl and one boy, this is
trivial. Now let k ≥ 1 and suppose (7) is true for n = k. Then for r = 2, 3, . . . , k we have
Yk+1 (r) = Ak+1 (r) + Bk+1 (r)

= (2(k + 1) − r)Yk (r − 1) + Yk (r)
(k!)2 (k!)2
= (2k + 2 − r) +
(r − 1)!((k − r + 1)!)2 r!((k − r)!)2
r(2k + 2 − r)(k!)2 + (k − r + 1)2 (k!)2
=
r!((k − r + 1)!)2
(k!) (2kr + 2r − r2 ) + (k 2 + r2 + 1 − 2kr + 2k − 2r)
2

=
r!((k + 1 − r)!)2
2 2
(k!) (k + 1 + 2k)
=
r!((k + 1 − r)!)2
((k + 1)!)2
= ,
r!((k + 1 − r)!)2
which is what we wanted to prove. Furthermore, for r = 1 we have
Yk+1 (1) = Ak+1 (1) + Bk+1 (1) = (2k + 1) + Yk (1) = (2k + 1) + k 2 = (k + 1)2 .
11
Finally, for r = k + 1 we have
(k!)2 ((k + 1)!)2
Yk+1 (k + 1) = Ak+1 (k + 1) = (k + 1)Yk (k) = (k + 1) = .
k! (k + 1)!
Problem 4.
We use signed distances throughout the proof. Assume that B is inside segment CE; in the
other case a similar proof can be used. We have
∠F AE = ∠DAE = ∠CBF = ∠EBF.
As B and F are on different sides of the line AE, we conclude that ABEF is a cyclic
quadrilateral. Hence JA · JB = JE · JF . Furthermore, K belongs to the circumcircle of
ABI if and only if JA · JB = JK · JI. Therefore K belongs to the circumcircle of ABI
if and only if JE · JF = JK · JI. Expressing JI = JF + F I, JE = JF + F E and
JK = 12 (JE + JF ) = 12 F E + JF , we find that K belongs to the circumcircle of ABI if and
only if
(JF + F E)JF = ( 21 F E + JF )(JF + F I) ⇔
1
2 · F E · JF = ( 21 F E
+ JF )F I ⇔
FE · FI
JF = .
F E − 2F I
Since ABEF is cyclic and AB is parallel to CD, we have
∠F EC = ∠F EB = 180◦ − ∠F AB = ∠F DC.
Hence CEDF is cyclic as well, yielding IC · ID = IE · IF . Furthermore, K belongs to the
circumcircle of CDJ if and only if IC · ID = IK · IJ. Therefore K belongs to the circumcircle
of CDJ if and only if IE · IF = IK · IJ. Expressing IJ = IF + F J, IE = IF + F E and
IK = 12 (IE + IF ) = 12 F E + IF , we find that K belongs to the circumcircle of CDJ if and
only if
(IF + F E)IF = ( 12 F E + IF )(IF + F J) ⇔
1
2 · F E · IF = ( 21 F E
+ IF )F J ⇔
F E · IF
FJ = ⇔
F E + 2IF
FE · FI
JF = .
F E − 2F I
We conclude that K belongs to the circumcircle of ABI if and only if K belongs to the cir-
cumcircle of CDJ.
Comment 1. After deriving that what needs to be proved is: JE · JF = JK · JI if and only if
IE · IF = IK · IJ, the solution comes down to eliminating all but three well-chosen distances
and then manipulating the equalities until it is clear that they are equivalent. The above
solution is just one way of doing this.
Comment 2. The conditions JE · JF = JK · JI and IE · IF = IK · IJ are two of the

many equivalent ways of expressing that the points E, F and J, I are in harmonic division,
i.e. (EF JI) = −1, where K is the midpoint of segment EF . Observing this fact would also
suffice to finish the solution.
12
Team Selection Test
Valkenswaard, June 13, 2009
Problem 1. Let n ≥ 10 be an integer, and consider n in base 10. Let S(n) be the sum of
the digits of n. A stump of n is a positive integer obtained by removing a number of digits
(at least one, but not all) from the right side of n. E.g.: 23 is a stump of 2351. Let T (n) be
the sum of all the stumps of n.
Prove that n = S(n) + 9 · T (n).
Problem 2. Let ABC be a triangle, P the midpoint of BC, and Q a point on segment CA
such that |CQ| = 2|QA|. Let S be the intersection of BQ and AP . Prove that |AS| = |SP |.
Problem 3. Let a, b and c be positive reals such that a + b + c ≥ abc. Prove that
√
a2 + b2 + c2 ≥ 3 abc.
Problem 4. Find all functions f : Z → Z satisfying
f (m + n) + f (mn − 1) = f (m)f (n) + 2
for all m, n ∈ Z.
Problem 5. Suppose that we are given an n-gon of which all sides have the same length,
and of which all the vertices have rational coordinates. Prove that n is even.
13
Solutions of the Team Selection Test 2009
Problem 1.
Let us denote the digits of n from right to left by a0 , a1 , . . . , ak . We have
n = a0 + 10a1 + · · · + 10k ak .
A stump of n consists of (from right to left) the digits ai , ai+1 , . . . , ak , where 1 ≤ i ≤ k.

We see that such a stump is equal to ai + 10ai+1 + · · · + 10k−i ak . Summation over i then
yields T (n). Now write T (n) in a different way, by taking all the terms involving a1 , then
those involving a2 , and so on (effectively, as we will see below, we are changing the order of
summation; it does not matter whether we first sum over i from 1 up to k, and then, for each
i, sum over j from i up to k, or we first sum over the j from 1 up to k, and then, for each j,
sum over i from j up to k):
k
X Xk X
k
T (n) = ai + 10ai+1 + · · · + 10k−i ak = 10j−i aj
i=1 i=1 j=i
j
k X k k
X X X 10j − 1
10j−i aj = 1 + 10 + · · · + 10j−1 aj =

= aj ,
10 − 1
j=1 i=1 j=1 j=1
where we used the summation formula for the geometric series in the final step. So we get
k
X k
X
9 · T (n) = (10j − 1)aj = (10j − 1)aj .
j=1 j=0
Pk
Recall that S(n) = j=0 aj . Hence we have
k
X k
X
S(n) + 9 · T (n) = (10j − 1 + 1)aj = 10j aj = n.
j=0 j=0
Problem 2.
Solution 1. Let T be a point on BQ such that P T is parallel to AC. Then P T joins the
midpoints of BC and BQ, so |P T | = 21 |CQ| = |QA|. So we see that AT P Q is a quadrilateral
of which two sides are parallel and of the same length. This implies that it is a parallelogram.
Since any parallelogram has the property that its diagonals bisect each other, we see that
|AS| = |SP |.
14
Solution 2. We apply Menelaos’ Theorem to triangle P CA. Since the points B, S and Q are
collinear, we have
P B CQ AS −1 2 AS AS
−1 = · · = · · =− .
BC QA SP 2 1 SP SP
AS
Hence SP = 1, from which we can deduce that S is the midpoint of the segment AP .
Solution 3. Let M be the midpoint of QC and let x = [AQS] = [QM S] = [M CS]

and y = [CP S] = [P BS]. Since [CP A] = [P BA], we see that [ASB] = 3x. But then
we have [AQB] = x + 3x, whereas on the other hand, we have 2[AQB] = [QCB], so
2x + 2y = [QCB] = 2[AQB] = 8x. Hence y = 3x. But then [ASB] = 3x = y = [SP B], from
which we conclude that |AS| = |SP |.
Solution 4. Let R be the intersection of CS and AB. According to Ceva’s Theorem, we have
AR BP CQ
· · = 1,
RB P C QA
from which we deduce that 2|AR| = |RB|. Let a = [P SC], b = [QSC], c = [QSA], d = [RSA],
e = [RSB] and f = [P SB]. Then we see that
b + a + f = 2(c + d + e) en a + e + f = 2(b + c + d),
which implies that
b − e = 2e − 2b,
hence b = e. Furthermore, we have 2d = e and 2c = b, hence c = d and c + d = e. Now
a + e + f = 2(b + c + d) implies that
a + f = b + 2c + 2d = b + c + d + e,
so
2(a + f ) = a + b + c + d + e + f = [ABC].
Hence 2|P S| = |P A|, implying that |P S| = |AS|.
Problem 3.
First of all, note that a2 + b2 + c2 ≥ ab + bc + ca, and hence that 3(a2 + b2 + c2 ) ≥ (a + b + c)2 .
1
Applying AM-GM we see that a+b+c ≥ 3(abc) 3 . On the other hand, note that a+b+c ≥ abc.
Now we have the following two inequalities:
2
a2 + b2 + c2 ≥ 31 (a + b + c)2 ≥ 3(abc) 3 ,
a2 + b2 + c2 ≥ 13 (a + b + c)2 ≥ 13 (abc)2 .
Raising the former equation to the 34 -th power, and the latter one to the 1
4 -th (which is
allowed, since every expression is positive):
3 3 1
(a2 + b2 + c2 ) 4 ≥ 3 4 (abc) 2 ,
1 1 1
(a2 + b2 + c2 ) 4 ≥ 3− 4 (abc) 2 .
Taking the product of these two equations, we then see that
1
a2 + b2 + c2 ≥ 3 2 (abc),
which is as desired.
15
Problem 4.
Suppose that there exists a c ∈ Z such that f (n) = c for all n. Then 2c = c2 + 2, so we
get the equation c2 − 2c + 2 = 0, which has no solutions in Z. Hence f cannot be constant.
Now substitute m = 0. This yields the equation f (n) + f (−1) = f (n)f (0) + 2, from which
we deduce that f (n)(1 − f (0)) is constant. Since f (n) isn’t a constant function, we see that
f (0) = 1. Using the same equation, we then get f (−1) = 2. Now substitute m = n = −1.
This yields f (−2) + f (0) = f (−1)2 + 2, from which follows that f (−2) = 5. Substituting
m = 1 and n = −1 now yields f (0) + f (−2) = f (1)f (−1) + 2, which implies that f (1) = 2.
Now substitute m = 1, then we obtain f (n + 1) + f (n − 1) = f (1)f (n) + 2, or, equivalently
f (n + 1) = 2f (n) + 2 − f (n − 1).
By induction, it then follows that f (n) = n2 +1 for all non-negative n, and, using the equation
f (n − 1) = 2f (n) + 2 − f (n + 1)
also for all non-positive n. Hence f (n) = n2 + 1 for all n, and we can easily check that this
function satisfies the given equation.
Problem 5.
Let (x1 , y1 ), . . . , (xn , yn ) be the coordinates of the vertices of the given n-gon. Define ai =
xi+1 − xi , bi = yi+1 P − yi for i = P1, 2, . . . , n, where xn+1 = x1 and yn+1 = y1 . Then we note
that ai , bi ∈ Q and ni=1 ai = ni=1 bi = 0; and that the sum a2i + b2i does not depend on i.
By multiplying with a suitable factor we can get rid of denominators and common divisors of
the ai and bi , so we may assume that ai , bi ∈ Z and that gcd(a1 , . . . , an , b1 , . . . , bn ) = 1. Let
c be the integer such that a2i + b2i = c for all i.
Suppose that c is odd. Then for every i, exactly

Pn one of ai , bi is odd. Hence of the 2n integers
ai , bi , exactly n are odd. Then we get 0 = i=1 (ai + bi ) ≡ n mod 2, implying that n is even.
Now suppose that c is even. Then, for all i, we have ai ≡ bi mod 2. If there exists an i
such that ai and bi are both odd, then c = a2i + b2i ≡ 1 + 1 ≡ 2 mod 4. If there exists an
i such that ai and bi are both even, then c = a2i + b2i ≡ 0 + 0 ≡ 0 mod 4. These two cases
cannot occur together. Hence either all of the ai and bi are odd, or all of them are even.
The latter statement contradicts our assumption about the greatest common divisor of the ai
and bi . The former statement yields 0 = ni=1 ai ≡ n mod 2, implying again that n is even.
P
16
Junior Mathematical Olympiad October 3, 2008
Vrije Universiteit Amsterdam
Problems part 1
• The problems in part 1 are five-choice questions. At each question exactly one of the given
five answers is correct. Indicate clearly on the answer sheet which letter corresponds to the
right answer.
• Each correctly given answer will get you 2 points. For wrong answers no points are sub-
tracted.
• You are allowed scrap paper, as well as a compass and a ruler or protractor. Calculators and
other electronic devices are not allowed.
• The time allowed for this part is 60 minutes.
• Good luck!
1. A rectangle has been divided into four smaller rectangles. The areas of
three of the small rectangles are 6, 8 and 9 (see the figure). Determine 6 9
the area of the fourth small rectangle.
? 8
A) 4 32 B) 5 C) 5 31 D) 5 23 E) 7
2. We multiply all odd numbers between 0 and 100. What is the last digit of the result?
A) 1 B) 3 C) 5 D) 7 E) 9
3. Six students are sitting next to each other on six chairs, numbered from 1 to 6. Now they all
get up at the same time, and then they sit down on a chair again, according to the table below.
standing up from chair number 1 2 3 4 5 6

↓ ↓ ↓ ↓ ↓ ↓
sitting down on chair number 4 3 1 6 5 2
This process of standing up and sitting down again happens 642 times. Which chair
is now occupied by the student who was at the start sitting on chair number 1?
A) 1 B) 2 C) 3 D) 4 E) 6
4. Consider the sequence of positive integers 1, 2, 2, 3, 3, 3, 4, 4, 4, 4, . . . in which the n-th positive

integer occurs exactly n times. We divide the hundredth number in the sequence by 5. What
is the remainder?
A) 0 B) 1 C) 2 D) 3 E) 4
D
5. In a quadrilateral ABCD the sides AB and CD have equal length.
A 65
Moreover, three angles are given: ∠A1 = 65◦ , ∠A2 = 80◦ and ∠B =
1
2
80 ?
50◦ . Determine ∠C2 . 2
1
C
A) 30◦ B) 40◦ C) 50◦ D) 60◦ E) 65◦
50
B
12
6. How many (positive or negative) integers n exist such that n+5 is an integer?
A) 2 B) 6 C) 8 D) 10 E) 12
17
7. What is

1 1 1 1 1 1
1− × 1− × 1− × 1− × 1− × 1− ?
4 9 16 25 36 49
1 2 3 4 5
A) 7 B) 7 C) 7 D) 7 E) 7
8. How many integers of the form bbcac exist, where a, b and c are digits (0, 1, . . . , 9) with c > a
and with b equal to the mean of a and c?
A) 20 B) 21 C) 22 D) 23 E) 24
9. Two identical rhombuses are lying on top of each other, one of them
rotated 90 degrees compared to the other. The area where the two
rhombuses overlap, happens to be a regular octagon: all eight sides
have equal length and all eight angles have equal sizes. Determine the α
smallest angle of the rhombus, indicated in the figure by α.
A) 22,5◦ B) 30◦ C) 40◦ D) 45◦ E) 60◦
10. How many of the integers 1, 2, 3, . . . , 2008 are not divisible by 2 and not divisible by 5 either?
A) 403 B) 603 C) 803 D) 1205 E) 1405
11. A large pond is being emptied by means of three pumps. Using just the first pump, emptying
the pond would take four days. Using just the second pump, it would take three days, and
using just the third pump, it would take two days. How many days does it take to empty the
pond using all three pumps at the same time?
1 12 13
A) 9 day B) 13 day C) 1 day D) 12 day E) 3 days
B
12. Through the vertices A, B and C of a triangle ABC passes a circle 2
1
with midpoint M . Of the three angles at the midpoint, two are equal 139 A
123
to 123◦ and 139◦ (see figure). Determine ∠B12 . M
A) 45◦ B) 49◦ C) 50◦ D) 51◦ E) 59◦

C
13. If we divide the number 22008 − 22007 + 22006 − 22005 + 22004 − 22003 + 22002 − 22001 by the
number 22000 , then the result is an integer (that is, there is no remainder). Determine this
integer.
A) 4 B) 36 C) 72 D) 170 E) 200
14. Consider an integer abcd consisting of four distinct digits a, b, c and d (where a is not allowed
to be 0). Of this integer you know that abcd × 11 = ac9bd. The digit b can be equal to only
one of the following five options. Which one?
A) 2 B) 3 C) 4 D) 5 E) 6
15. A block of cheese consisting of 3 × 4 × 5 small cubes of cheese, has

been perforated by a thin pin along a space diagonal. How many of
the small cubes have been perforated by the pin?
A) 5 B) 8 C) 10 D) 11 E) 12
18
Junior Mathematical Olympiad October 3, 2008
Vrije Universiteit Amsterdam
Problems part 2
• The problems in part 2 are open questions. At each question your answer should be a
number or an expression (for example 2 23 or a2 ). Write this answer on your answer sheet
at the indicated place.
• Each correctly given answer will get you 2 points. For wrong answers no points are sub-
tracted.
• You are allowed scrap paper, as well as a compass and a ruler or a protractor. Calculators
and other electronic devices are not allowed.
• The time allowed for this part is 60 minutes.
• Good luck!
1. What is the maximum number of points of intersection between a circle and a triangle?
2. During a tournament with six players, each player plays a match against each other player.
At each match there is a winner; ties do not occur. A journalist asks five of the six players
how many matches each of them has won. The answers given are 4, 3, 2, 2 and 2. How many
matches have been won by the sixth player?
3. A 3 by 5 rectangle has been coloured like a chess board. What is the

total area of the black parts inside the triangle drawn in the figure?
4. I have written four numbers on a piece of paper. In six different ways I can choose two and
add them up. The resulting sums are 11, 15, 16, 16, 17 and 21. Now I multiply all of the four
numbers on my piece of paper. What is the result?
5. What is the sum of the positive integers smaller than 100,000 (a hundred thousand) in which
only the digits 0 and 1 occur?
Turn over for the remaining questions. www.wiskundeolympiade.nl/junior
19
6. Find the smallest positive integer with the property that if you multiply its digits, the result
is 1890.
1
2
3
4
5
7. In this stretched chess board the small rectangles have sides with 6
length 1 up to 8. Determine the total area of all black rectangles. 7
12 3 4 5 6 7 8
8. Write the following expression as one fraction in its simplest terms, given that for a, b and c
we have a × b × c = 1.
1 1 1
+ +
1 + a + (a × b) 1 + b + (b × c) 1 + c + (c × a)
9. The two smallest circles in the figure have radiuses 2 and 3. Find the
radius of the largest circle.
3
2008
10. A hexagon has six angles all equal to 120 degrees. The lengths of 2006 2009
four consecutive sides are 2000, 2006, 2008 and 2009. Determine the
perimeter of the hexagon.
2000
www.wiskundeolympiade.nl/junior
20
First Round
SE
W
AND
IS
U
ERL
N
D
E
NED
PIADE
Problems OLYM
• Time available: 2 hours.
• The A-problems are multiple choice questions. Only one of the five options given is correct.
Please state clearly which letter precedes the correct solution. Each correct answer is worth 2
points.
• The B-problems are open questions; the answers to these questions are a number, or numbers.
Each correct answer is worth 5 points. Please work accurately, since an error in your calculations
may cause your solution to be considered incorrect and then you won’t√get points at all for
11
that question. Please give your answers exactly, for example 81 or 2 + 12 5 or 14 π + 1.
• You are not allowed to use calculators and formula sheets; you can only use a pen, a compass
and a ruler or set square. And your head, of course.
• This is a competition, not an exam. The main thing is that you have fun solving unusual
mathematical problems. Good luck!
A-problems
A1. Ella has answered three sets of questions. Of the first set, consisting of 25
questions, she answered 60% correctly. Of the second set, consisting of 30
questions, she answered 70% correctly, and of the third set, consisting of 45
questions, she answered 80% correctly. Now, if we combine the three sets to
form one set of 100 questions, what percentage of these 100 questions did
Ella answer correctly?
(A) 68% (B) 70% (C) 72% (D) 74% (E) 76%
A2. How many of the integers from 10 to 99 (10 and 99 included) have the
property that the sum of their digits is equal to the square of an integer?
(An example: The sum of the digits of 27 is equal to 2 + 7 = 9 = 32 .)
(A) 13 (B) 14 (C) 15 (D) 16 (E) 17
A3. Ronald rolls three dice. These dice look like normal dice, but the numbers
on their sides are unusual.
On the sides of the first dice are the numbers 1, 1, 2, 2, 3, 3.
On the sides of the second dice are the numbers 2, 2, 4, 4, 6, 6.
On the sides of the third dice are the numbers 1, 1, 3, 3, 5, 5.
He then adds up the three numbers he gets from rolling the three dice.
What is the probability that the resulting number is odd?
1 1 1 2 3
(A) 4 (B) 3 (C) 2 (D) 3 (E) 4
A4. Three distinct numbers from the set {1, 2, 3, 4, 5, 6, 7, 8, 9} are placed in the
three squares at the top of the figure to the right, after which the numbers
are added as indicated in said figure. We call Max the highest number that + +
can appear in the bottom square, and Min the lowest number that can appear
there. What is the value of Max − Min? +
(A) 16 (B) 24 (C) 25 (D) 26 (E) 32
21
A5. The ratio between the lengths of the diagonals of a rhombus is 3 to 4. (A
rhombus is an equilateral quadrilateral.) The sum of the lengths of the
diagonals is 56. What is the perimeter of this rhombus?
(A) 80 (B) 96 (C) 100 (D) 108 (E) 160
A6. Wouter walks from his home to the fitness center. He could also have chosen
to go by bike, in which case he would have covered the distance between his
home and the fitness center 7 times as fast. However, he has left his bike at
home. After having walked 1 km he reaches a bridge. Continuing on foot
will take Wouter just as long as walking back home to get his bike and then
cycle to the fitness center. What is the distance between the bridge and the
fitness center in kilometers?
(A) 87 (B) 67 (C) 65 (D) 54 (E) 43
A7. On the sides of an equilateral triangle, we draw three squares. The sides of
these squares that are parallel to the sides of the triangle are extended until
they intersect. These three intersections form another equilateral triangle.
Suppose that the length of a side of the original triangle is equal to 1. What
is the length of a side of the large equilateral triangle? 1 1
√ √ √ √ √
(A) 1 + 2 2 (B) 5 − 12 3 (C) 3 2 (D) 1 + 2 3 (E) 2 6 1
A8. Consider all four-digit numbers in which each of the digits 3, 4, 6 and 7
occurs exactly once. How many of these numbers are divisible by 44?
(A) 2 (B) 4 (C) 6 (D) 8 (E) 12
B-problems
B1. A sheet of paper shows a grid of 101 by 101 white squares. A chain is formed
by coloring squares grey as shown in the figure to the right. The chain starts
in the upper left-hand corner and goes on until it cannot go on any further.
Only part of the grid is shown. In total, how many squares are colored grey
in the original grid of 101 by 101 squares?
B2. The integer N consists of 2009 consecutive nines. A computer calculates

N 3 = (99999 . . . 99999)3 . How many nines does the number N 3 contain in |99999 .{z
. . 99999}
2009×
total?
B3. Using a wide brush, we paint the diagonals of a square tile, as in the figure. 1
Exactly half of the surface of this tile is covered with paint. Given that the
width of the brush is 1, as indicated in the figure, what is the length of the
side of the tile?
B4. Determine a triplet of integers (a, b, c) satisfying:

a + b + c = 18
a2 + b2 + c2 = 756
a2 = bc
22
First Round
SE
W
AND
IS
U
ERL
N
D
E
NED
PIADE
Solutions OLYM
A1. (C) 72% 60% of 25 is 15; 70% of 30 is 21; and 80% of 45 is 36. So in total, Ella answered
15 + 21 + 36 = 72 of the 100 questions correctly.
A2. (E) 17 We check how many of these numbers have sum of digits equal to 1, 2,
etc. There is 1 number with sum 1 (being 10); there are 2 with sum 2 (being 20 and
11); etc.; 9 with sum 9 (being 90, 81, . . . , 18); also, 9 with sum 10 (being 91, 82, . . . , 19);
etc.; and finally, 1 with sum 18 (being 99); see the table below. Then the sum of dig-
its is a square of an integer (i.e. 1, 4, 9 or 16) in 1 + 4 + 9 + 3 = 17 of the 90 cases.
sum: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
number: 1 2 3 4 5 6 7 8 9 9 8 7 6 5 4 3 2 1
A3. (B) 13 Note that the second die only has even numbers on it, and that the third die
only has odd numbers on it. So essentially the question is to find the probability that rolling the
first die gives an even number. Since 2 of the 6 numbers on this die are even, this probability is
equal to 62 = 13 .
A4. (D) 26 Let’s say we put a, b, and c in the top three squares. Then a b c
the result in the bottom square is a + 2b + c. So we can maximize this result + +
by making first b, then a and c as large as possible. Taking b = 9, a = 8 and
a+b b+c
c = 7 then yields 33 as result. In the same way, we can minimize the result
by making first b, then a and c as small as possible. Taking b = 1, a = 2, +
c = 3 yields 7 as result. The difference between these numbers is 33 − 7 = 26. a+2b+c
3 4
A5. (A) 80 Note that the diagonals have lengths 7 · 56 = 24 and 7 · 56 = 5 5
32. So the halves of diagonals have lengths 12 = 3 · 4 and 16 = 4 · 4. So the 3
4 4
rhombus is 4 times larger than the rhombus in the figure, which consists of
5 3
four triangles with sides 3, 4 and 5. Hence the sides of the original rhombus
5
have length 4 · 5 = 20, and thus the perimeter has length 4 · 20 = 80.
A6. (E) 34 Let x be said distance and let us suppose that he has spent a quarter of an hour
walking by then. Then continuing walking will take him x quarters of an hour. On the other
hand, if he decides to walk back home to pick up his bike, he’ll first have to spend one quarter
of an hour to get back, and then 1+x 7 quarters of an hour by bike; since he travels 7 times faster
that way. Then we have x = 1 + 7 , so 7x = 7 + (1 + x), or 6x = 8. We deduce that x = 86 = 43 .
1+x
Taking for ‘quarter of an hour’ any other time unit will give us the same result.
√
A7. (D) 1 + 2 3 In 4ABC, ∠A is half of 60◦ , so 30◦ . Also, ∠C is a right
angle, so 4ABC is a 30◦ -60◦ -90◦ -triangle, where |BC| = 1. So it’s half of B
an equilateral triangle with sides 2: |AB|
√ = 2. Now
√ we calculate |AC| with 2
1
the√Theorem√of Pythagoras: |AC| = 22 − 12 = 3. So the required length A 3 C
is 3 + 1 + 3.
A8. (A) 2 Suppose that n, having digits a, b, c and d (so n = 1000a + 100b + 10c + d) is
divisible by 44. Then it is also divisible by 11. Since the number m = 1001a + 99b + 11c is also
divisible by 11, so is m − n. Hence m − n = a − b + c − d is a multiple of 11. But this number
is at most the sum of the two highest digits, minus the sum of the lowest two, so 13 − 7 = 6,
and in the same way, we see that this number is at least −6. So it has to be equal to 0. Thus
a + c = b + d, and since the sum of the digits is 20, we have a + c = b + d = 10. First suppose that
d = 4. Then b = 6 so we get two possibilities for n, namely 3674 and 7634. But neither of these
is divisible by 4, let alone by 44. Now suppose that d = 6, then we have b = 4, and in this case
we get 3476 and 7436, both of which are divisible by 44. Finally, note that since n is divisible by
4, d must be even. We deduce that we have only 2 such numbers that are divisible by 44.
23
B1. 5201 We can subdivide this grid of 1012 squares as follows. In
the upper left corner, we have one (grey) square, then two L-shaped pieces,
one having 3 squares (one of which grey), the other having 5 squares (all
of which grey). Then we have two more L-shaped pieces, one having 7
squares (one of which grey), the other having 9 squares (all of which grey),
etc. Of the last two L-shaped pieces, the first one has 199 squares (one of
which grey), and the second one has 201 squares (all of which grey). We
have 50 pairs of L-shapes in total, so the total number of grey squares is
1 + (1 + 5) + (1 + 9) + (1 + 13) + . . . + (1 + 201) = 1 + (6 + 10 + 14 + . . . + 202) =
1 + 12 · 50 · (6 + 202) = 5201.
Alternative solution In each pair of these L-shapes, there are 4 more
grey squares than white squares. So there are 50 · 4 = 200 more grey squares
than white squares in the 1012 − 1 = 10200 squares contained in the 50 pairs
of L-shapes, so we have 5000 white squares and 5200 grey ones. Since the
upper left square is grey, in total, we have 5201 grey squares.
B2. 4017 93 = 729; 993 = 970299; 9993 = 997002999. It seems to
be the case that in general, the third power of a number n consisting of k
consecutive nines takes the following form: first k − 1 nines; then a 7; then
k−1 zeroes; then a 2; and finally k nines. To prove this, we write n = 10k −1.
Indeed: (10k − 1)3 = 103k − 3 · 102k + 3 · 10k − 1 = 102k (10k − 3) + (3 · 10k − 1). |99999 .{z
. . 99999}
The number 10k −3 can be written as 999 . . . 997 with k −1 nines. Multiplied 2009×
with 102k this gives a number that ends in 2k zeroes. Adding 3 · 10k − 1 to
this number, the last k + 1 zeroes are replaced with 2999 . . . 999 with k nines.
So in total, we have (k − 1) + k nines; in our case, k = 2009, so we have 4017
nines.
√
B3. 2 + 2 2 We only need to look at a quarter of the tile: 4ABC. The
B Q
area of 4P QR is half of the area 1
2 2
of 4ABC.
√ The triangles are√similar, so corresponding sides have a ratio of Q
1 : 2, so |QR| : |BC| = 1 : 2. x P
Now we calculate |BQ| using the Theorem q of Pythagoras in 4BQQ0 : A
√ R
2|BQ|2 = |BQ0 |2 + |BQ|2 = 12 , so |BQ| = 12 = 12 2. Now let us write x 1
2
√ √ √ √ 2
C
for |QR|. Then √
we√
find x + 2 = 2 · x, so x( 2 − 1) = 2 or equivalently,
√
2 2( 2+1) √ √ √
x = √2−1 = 2−1 = 2 + 2. Hence |BC| = x + 2 = 2 + 2 2 (or
√ √ √ √
|BC| = 2 · x = 2 · (2 + 2) = 2 2 + 2).
B4. (a, b, c) = (−12, 6, 24) of (a, b, c) = (−12, 24, 6) (one answer is enough)
We calculate (b + c)2 in two different ways. (b + c)2 = (18 − a)2 = 324 − 36a + a2 and (b + c)2 =
b2 + 2bc + c2 = (756 − a2 ) + 2a2 . So a2 − 36a + 324 = a2 + 756, or −36a = 756 − 324 = 432,
so a = −12. Substituting this in the first and in the last equation, we obtain the equations
b + c = 30 and bc = 144. Trying some divisors of 144 = 122 , we then should be able to find a
solution. Or we can just substitute c = 30−b in the last equation, yielding the quadratic equation
b(30 − b) = 144, or equivalently b2 − 30b + 144 = 0. We can factorize this as (b − 6)(b − 24) = 0
(or we can use the abc-formula) to see that we have two solutions b = 6 (and c = 24) or b = 24
(and c = 6).
Alternative solution Just as above, we see that a = −12. Then by substituting this in all three
equations, we see that b + c = 30, b2 + c2 = 756 − 144 = 612 and bc = 144. Combining√the last two
equations yields (b−c)2 = b2 +c2 −2bc = 612−2·144 = 324, or equivalently b−c = ± 324 = ±18.
Adding the equations b + c = 30 and b − c = −18, we get 2b = (b + c) + (b − c) = 12 so b = 6
(and c = 24). Adding the equations b + c = 30 and b − c = 18, we get 2b = (b + c) + (b − c) = 48
so b = 24 (and c = 6).
24
48th Dutch Mathematical Olympiad 2009
and the team selection for IMO 2010 Astana
First Round, January 2009
Final Round, September 2009
BxMO Team Selection Test, March 2010
Benelux Mathematical Olympiad, April 2010 C
IMO Team Selection Tests, June 2010 CM
MY
CY
We eat problems
CMY
Junior Mathematical Olympiad, October 2009 K
for breakfast.
Preferably unsolved ones...
In juli 2011 wordt de internationale wiskunde olympiade

48 Dutch Mathematical
in Nederlandthgehouden: IMO2011
In de opmaat naar IMO2011 wordt op 3 oktober 2008 op
Olympiad 2009
de VU de eerste Junior Wiskunde Olympiade gehouden
voor de 100 beste deelnemers aan de Kangoeroewedstrijd.
International
De JWO wordt een jaarlijks terugkerend evenement. Mathematical
Zie ook: www.wiskundeolympiade.nl/junior Olympiad Am
WISKUNDE
Hoofdsponsors van de OLYMPIADE
We thank our sponsors
NEDERLANDSE
WISKUNDE 2010
OLYMPIADE
Overige sponsors
en donateurs
Contents
1 Introduction
3 First Round, January 2009
9 Final Round, September 2009
14 BxMO Team Selection Test, March 2010
18 Benelux Mathematical Olympiad, April 2010
24 IMO Team Selection Test 1, June 2010
29 IMO Team Selection Test 2, June 2010 Centraal Bureau voor de Statistiek
33 Junior Mathematical Olympiad, October 2009
© Stichting Nederlandse Wiskunde Olympiade, 2010

Introduction
In 2009 the Dutch Mathematical Olympiad consisted of two rounds. The
first round was held on 30 January 2009 at the participating schools. The
paper consisted of eight multiple choice questions and four open-answer
questions, and students got two hours to work on it. In total 4379 students
of 230 secondary schools participated in this first round.
The best students were invited for the final round. In order to stimulate
young students to participate, we set different thresholds for students from
different grades. Those students from grade 5 (4, ≤ 3) that scored 26 (23,
20) points or more on the first round (out of a maximum of 36 points)
were invited to the final round. Also some outstanding participants in the
Kangaroo math contest or the Pythagoras Olympiad were invited.
For the second consecutive year we organised training sessions at six uni-
versities in the country for the 155 students who had been invited for the
final round. Former Dutch IMO-participants were involved in the training
sessions at each of the universities.
Out of those 155, in total 131 participated in the final round on 18 Septem-
ber 2009 at Eindhoven University of Technology. This final round contained
five problems for which the students had to give extensive solutions and
proofs. They had three hours for the paper. After the prizes had been
awarded in the beginning of November, the Dutch Mathematical Olympiad
concluded its 48th edition 2009. In 2011 we will have our 50th edition.
The 31 most outstanding candidates of the Dutch Mathematical Olympiad

2009 were invited to an intensive seven-month training programme, con-
sisting of weekly problem sets. Also, the students met twice for a three-day
training camp, three times for a day at the university, and finally for a
six-day training camp in the beginning of June.
On 5 March 2010 the first selection test was held. The best ten students
participated in the second Benelux Mathematical Olympiad (BxMO), held
in Amsterdam, the Netherlands.
In June, out of those 10 students and 3 reserve candidates, the team for
the International Mathematical Olympiad 2010 was selected by two team
selection tests on 9 and 12 June 2010. A seventh, young, promising student
was selected to accompany the team to the IMO. The team had a training
camp in Astana from 28 June until 5 July, together with the team from
1
New Zealand.
For younger students we organised the second Junior Mathematical Olym-

piad in October 2009 at the VU University Amsterdam. The students
invited to participate in this event were the 30 best students of grade 1,
grade 2 and grade 3 of the popular Kangaroo math contest. The competi-
tion consisted of two one-hour parts, one with 15 multiple choice questions
and one with 10 open-answer questions. The goal of this Junior Mathe-
matical Olympiad is to scout talent and stimulate them to participate in
the first round of the Dutch Mathematical Olympiad.
The Dutch team for IMO 2010 Kazakhstan consists of

• Guus Berkelmans (16 y.o.)
• Harm Campmans (18 y.o., participated in IMO 2009 as well)
• Madelon de Kemp (17 y.o.)
• David Kok (17 y.o., participated in IMO 2009 as well)
• Daniël Kroes (16 y.o.)
• Merlijn Staps (15 y.o., observer C at IMO 2009)
We bring as observer C the promising young student

• Jeroen Huijben (14 y.o.)

• Johan Konter (team leader), Utrecht University
• Quintijn Puite (observer A), Eindhoven University of Technology
• Sietske Tacoma (observer B), Utrecht University
The Dutch delegation for IMO 2010 Kazakhstan further consists of

• Wim Berkelmans (member AB, observer A), VU University Amster-
dam
• Karst Koymans (observer A), University of Amsterdam
• Aad Loois (observer C)
• Jelle Loois (observer A), ORTEC
• Ronald van Luijk (observer A), Leiden University
• Rozemarijn Schalkx (observer A), Eindhoven University of Technol-
ogy
We are grateful to Jinbi Jin and Raymond van Bommel for the composition
of this booklet and the translation into English of most of the problems and
the solutions.
2
Problems
A-problems
A1. Ella does three tests. During the first one, she answered 60% of
the 25 questions correctly, during the second one, she answered 70%
of the 30 questions correctly, and during the last one, she answered
80% of the 45 questions correctly. Now if we merge these three tests
together to form one of 100 questions, what is the percentage of these
100 questions that Ella answered correctly?
A) 68% B) 70% C) 72% D) 74% E) 76%
A2. How many of the integers from 10 to 99 (10 and 99 included) have the
property that the sum of its digits is equal to the square of an integer?
(An example: The sum of the digits of 27 is equal to 2 + 7 = 9 = 32 .)
A) 13 B) 14 C) 15 D) 16 E) 17
A3. Ronald throws three dice. These dice look just like ordinary dice, but
their faces are numbered differently. The first die has the numbers 1,
1, 2, 2, 3 and 3 on it. The second die has the numbers 2, 2, 4, 4, 6
and 6 on it. And the third die has the numbers 1, 1, 3, 3, 5 and 5 on
it. He then adds up the three numbers he gets from rolling the three
dice. What is the probability that the resulting number is odd?
1 1 1 2 3
A) 4 B) 3 C) 2 D) 3 E) 4
A4. Three distinct numbers from the set

{1, 2, 3, 4, 5, 6, 7, 8, 9} are placed in the three squares
in the top of the figure to the right, after which the + +
numbers are added as described in said figure. We
call Max the highest number that can appear in the
bottom square, and Min the lowest number that
+
can appear there. What is the value of Max − Min?
A) 16 B) 24 C) 25 D) 26 E) 32
3
A5. The lengths of the diagonals of a rhombus have a ratio of 3 : 4. (A
rhombus is an equilateral quadrilateral.) The sum of the lengths of
the diagonals is 56. What is the diameter of this rhombus?
A) 80 B) 96 C) 100 D) 108 E) 160
A6. Wouter is traveling by foot from his home to the fitness center. He
also could have chosen to travel by bike, in which case he would travel
7 times as fast. But he left his bike at home. After walking for 1 km,
continuing to walk would take just as long as walking back to get his
bike, and then travel further by bike. By then, what is the distance
in km to the fitness center?
8 7 6 5 4
A) 7 B) 6 C) 5 D) 4 E) 3
A7. On the sides of an equilateral triangle, we draw

three squares. The sides of these squares that
are parallel to the sides of the triangle are ex-
tended until they intersect. These three intersec-
1 1
tions form another equilateral triangle. Suppose
1
that the length of a side of the original triangle is
equal to 1. What is the length of a side of the large
equilateral triangle?
√ √ √
A) 1 + 2 2 B) 5 − 12 3 C) 3 2
√ √
D) 1 + 2 3 E) 2 6
A8. Consider all four-digit numbers where each of the digit 3, 4, 6 and 7
occurs exactly once. How many of these numbers are divisible by 44?
A) 2 B) 4 C) 6 D) 8 E) 12
B-problems
The answer to each B-problem is a number.
B1. On a sheet of paper, a grid of 101 by 101 white

squares is drawn. A chain is formed by coloring
squares grey as shown in the figure to the right.
The chain starts in the upper left corner and goes
on until it cannot go on any further. A large piece
of the sheet is torn off. How many squares were col-
ored grey in the original grid of 101 by 101 squares?
4
B2. The integer N consists of 2009 consecutive nines.
A computer calculates N 3 = (99999 . . . 99999)3 . 99999
| .{z
. . 99999}
3 2009×
How many nines does the number N contain in
total?
B3. Using, a wide brush, we paint the diagonals of a

square tile, as in the figure. Exactly half of the
1
area of this tile is covered with paint. Knowing
that as the width of the brush is 1, as indicated
in the figure, what is the length of the side of the
tile?
B4. Determine a triplet of integers (a, b, c) satisfying:

a + b + c = 18
a + b2 + c2 = 756
2
a2 = bc
Solutions
A-problems
A1. C) 72% 60% of 25 is 15; 70% of 30 is 21; and 80% of 45 is 36.

So in total, Ella answered 15 + 21 + 36 = 72 of the 100 questions
correctly.
A2. E) 17 We check how many of these numbers have sum of

digits equal to 1, 2, etc. There is 1 number with sum 1 (being 10);
there are 2 with sum 2 (being 20 and 11); etc.; 9 with sum 9 (being
90, 81, . . . , 18); also, 9 with sum 10 (being 91, 82, . . . , 19); etc.; and
finally, 1 with sum 18 (being 99); see the table below. Then the sum
of digits is a square of an integer (i.e. 1, 4, 9 or 16) in 1+4+9+3 = 17
of the 90 cases.
A3. B) 13 Note that the second die only has even numbers on it,
and that the third die only has odd numbers on it. So essentially
the question is to find the probability that rolling the first die gives
an even number. Since 2 of the 6 numbers on this die are even, this
probability is equal to 26 = 13 .
5
A4. D) 26 Let’s say we put a, b, and c in the
top three squares. Then the result in the bottom a b c
square is a+2b+c. So we can maximize this result
+ +
by making first b, then a and c as large as possible. b+c
a+b
Taking b = 9, a = 8 and c = 7 then yields 33 as
result. In the same way, we can minimize the result +
a+2b+c
by making first b, then a and c as small as possible.
Taking b = 1, a = 2, c = 3 yields 7 as result. The
difference between these numbers is 33 − 7 = 26.
A5. A) 80 Note that the diagonals have lengths

5 5
3 4 3
7 · 56 = 24 and 7 · 56 = 32. So the halves of diag- 4 4
onals have lengths 12 = 3 · 4 and 16 = 4 · 4. So the
rhombus is 4 times larger than the rhombus in the 3
5 5
figure, which consists of four triangles with sides
3,4 and 5. Hence the sides of the original rhombus
have length 4 · 5 = 20, and thus the diameter has
length 4 · 20 = 80.
A6. E) 43 Let x be said distance and let us suppose that he has

spent a quarter of an hour walking by then. Then continuing walking
will take him x quarters of an hour. On the other hand, if he decides
to walk back home to pick up his bike, he’ll first have to spend one
quarter of an hour to get back, and then 1+x 7 quarters of an hour
by bike; since he travels 7 times faster that way. Then we have
x = 1 + 1+x 7 , so 7x = 7 + (1 + x), or 6x = 8. We deduce that
x = 68 = 34 . Taking for ‘quarter of an hour’ any other time unit will
give us the same result.
√
A7. D) 1 + 2 3 In 4ABC, ∠A is half of 60◦ , so
30◦ . Also, ∠C is a right angle, so 4ABC is
a 30◦ -60◦ -90◦ -triangle, where |BC| = 1. So
it’s half of an equilateral triangle with sides
2: |AB| = 2. Now we calculate |AC| √ with the B
|AC| 2 12 =
Theorem
√ of Pythagoras: √ 2 −√
= 2
1
3. So the required length is 3 + 1 + 3.
A 3 C
6
A8. A) 2 Suppose that n, having digits a, b, c and d (so n =
1000a + 100b + 10c + d) is divisible by 44. Then it is also divisible by
11. Since the number m = 1001a + 99b + 11c is also divisible by 11,
so is m − n. Hence m − n = a − b + c − d is a multiple of 11. But this
number is at most the sum of the two highest digits, minus the sum
of the lowest two, so 13 − 7 = 6, and in the same way, we see that this
number is at least −6. So it has to be equal to 0. Thus a + c = b + d,
and since the sum of the digits is 20, we have a + c = b + d = 10.
First suppose that d = 4. Then b = 6 so we get two possibilities for
n, namely 3674 and 7634. But neither of them is divisible by 4, let
alone by 44. Now suppose that d = 6, then we have b = 4, and in this
case we get 3476 and 7436, both of which are divisible by 44. Finally,
note that since n is divisible by 4, d must be even. We deduce that
we have only 2 such numbers that are divisible by 44.
Alternative solution Check all 24 possibilities, or just the 12 even
possibilities, of even only the 6 multiples of 4.
B-problems
B1. 5201 We can subdivide this grid of 1012 squares as follows. In

the upper left corner, we have one (grey) square, then two L-shaped
pieces, one having 3 squares (one of which grey), the other having 5
squares (all of which grey). Then we have two more L-shaped pieces,
one having 7 squares (one of which grey), the other having 9 squares
(all of which grey), etc. Of the last two L-shaped pieces, the first
one has 199 squares (one of which grey), and the second one has 201
squares (all of which grey). We have 50 pairs of L-shapes in total, so
the total number of grey squares is 1+(1+5)+(1+9)+(1+13)+. . .+
(1 + 201) = 1 + (6 + 10 + 14 + . . . + 202) = 1 + 21 · 50 · (6 + 202) = 5201.
Alternative solution In each pair of these L-shapes, there are 4

more grey squares than white squares. So there are 50 · 4 = 200 more
grey squares than white squares in the 1012 − 1 = 10200 squares
contained in the 50 pairs of L-shapes, so we have 5000 white squares
and 5200 grey ones. Since the upper left square is grey, in total, we
have 5201 grey squares.
7
B2. 4017 93 = 729; 993 = 970299; 9993 = 997002999. It seems to
be the case that in general, the third power of a number n consisting
of k consecutive nines takes the following form: first k − 1 nines; then
a 7; then k − 1 zeroes; then a 2; and finally k nines. To prove this, we
write n = 10k − 1. Indeed: (10k − 1)3 = 103k − 3 · 102k + 3 · 10k − 1 =
102k (10k − 3) + (3 · 10k − 1). The number 10k − 3 can be written as
999 . . . 997 with k − 1 nines. Multiplied with 102k this gives a number
that ends in 2k zeroes. Adding 3 · 10k − 1 to this number, the last
k + 1 zeroes are replaced with 2999 . . . 999 with k nines. So in total,
we have (k − 1) + k nines; in our case, k = 2009, so we have 4017
nines.
√
B3. 2 + 2 2 We only need to look at a
quarter of the tile: 4ABC. The area of
4P QR is half of the area of 4ABC. The B Q
1
triangles are similar,
√ so corresponding sides 2 2
have
√ a ratio of 1 : 2, so |QR| : |BC| = 1 : Q
2. Now we calculate |BQ| using the The-
orem of Pythagoras in 4BQQ0 : 2|BQ| q =
2 x P
A
0 2 2 2 1
|BQ | + |BQ| = 1 , so |BQ| = 2 = R
1
√
2 2. Now √ let us write x for |QR|. Then 1
2
√ √ √ 2
we find x + 2 = 2 · x, so x( √ 2 −√1) = 2 C
√
or equivalently, x = √2−12
= 2(2−1 2+1)
=
√ √ √
2+ 2. Hence
√ |BC|
√ = x+ √ 2 = 2+2
√ 2 (or
|BC| = 2 · x = 2 · (2 + 2) = 2 2 + 2).
B4. (a, b, c) = (−12, 6, 24) of (a, b, c) = (−12, 24, 6) (one answer is enough)
We calculate (b + c)2 in two different ways. (b + c)2 = (18 − a)2 =
324 − 36a + a2 and (b + c)2 = b2 + 2bc + c2 = (756 − a2 ) + 2a2 .
So a2 − 36a + 324 = a2 + 756, or −36a = 756 − 324 = 432, so
a = −12. Substituting this in the first and in the last equation, we
obtain the equations b + c = 30 and bc = 144. Trying some divisors
of 144 = 122 , we then should be able to find a solution. Or we can
just substitute c = 30 − b in the last equation, yielding the quadratic
equation b(30 − b) = 144, or equivalently b2 − 30b + 144 = 0. We can
factorize this as (b − 6)(b − 24) = 0 (or we can use the abc-formula)
to see that we have two solutions b = 6 (and c = 24) or b = 24 (and
c = 6).
8
Problems
For these problems not only the answer is important; you also have to describe the way you
solved the problem.
1. In this problem, we consider integers consisting of 5 digits, of which

the first and last one are nonzero. We say that such an integer is a
palindromic product if it satisfies the following two conditions:
• the integer is a palindrome, (i.e. it doesn’t matter if you read it
from left to right, or the other way around);
• the integer is a product of two positive integers, of which the
first, when read from left to right, is equal to the second, when
read from right to left, like 4831 and 1384.
For example, 20502 is a palindromic product, since 102 · 201 = 20502,
and 20502 itself is a palindrome.
Determine all palindromic products of 5 digits.
2. Consider the sequence of integers 0, 1, 2, 4, 6, 9, 12, . . . obtained by

starting with zero, adding 1, then adding 1 again, then adding 2, and
adding 2 again, then adding 3, and adding 3 again, and so on. If we
call the subsequent terms of this sequence a0 , a1 , a2 , . . . , then we
have a0 = 0, and
a2n−1 = a2n−2 + n, a2n = a2n−1 + n
for all integers n > 1.

Find all integers k > 0 for which ak is the square of an integer.
3. A tennis tournament has at least three participants. Every partici-

pant plays exactly one match against every other participant. More-
over, every participant wins at least one of the matches he plays.
(Draws do not occur in tennis matches.)
Show that there are three participants A, B and C for which the
following holds: A wins against B, B wins against C, and C wins
against A.
9
4. Let ABC be an arbitrary triangle. On the perpendicular bisector of
AB, there is a point P inside of triangle ABC. On the sides BC and
CA, triangles BQC and CRA are placed externally. These triangles
satisfy 4BP A ∼ 4BQC ∼ 4CRA. (So Q and A lie on opposite
sides of BC, and R and B lie on opposite sides of AC.)
Show that the points P , Q, C and R form a parallelogram.
5. We number a hundred blank cards on both sides with the numbers 1

to 100. The cards are then stacked in order, with the card with the
number 1 on top.
The order of the cards is changed step by step as follows: at the 1st
step the top card is turned around, and is put back on top of the stack
(nothing changes, of course), at the 2nd step the topmost 2 cards are
turned around, and put back on top of the stack, up to the 100th
step, in which the entire stack of 100 cards is turned around. At the
101st step, again only the top card is turned around, at the 102nd
step, the topmost 2 cards are turned around, and so on.
Show that after a finite number of steps, the cards return to their
original positions.
Solutions
1. Note that the two integers mentioned in the second condition cannot
end in a 0; since otherwise the product would also have to end in
a 0. Thus, the two numbers have an equal number of digits. But
a product of two integers of four or more digits consists of at least
seven digits, hence is too large, and a product of two integers of at
most two digits consists of at most four digits, hence is too small.
So to satisfy the second condition, we have to consider products of
integers consisting of three digits, say abc and cba.
If we work this out, we obtain: abc · cba = (100a + 10b + c)(100c +
10b + a) = 10000ac + 1000b(a + c) + 100(a2 + b2 + c2 ) + 10b(a + c) + ac.
If ac > 9, then this integer consists of more than 5 digits. Hence
ac 6 9. Now if b(a + c) > 9, then the first digit will be greater
than ac, hence it will no longer be equal to the last digit. Hence we
also have b(a + c) 6 9. We can use a similar argument to show that
a2 +b2 +c2 6 9. We deduce from this that 1+b2 +1 6 a2 +b2 +c2 6 9,
and hence that b is equal to one of 0,1,2.
If b = 0, then a2 + c2 6 9, so we have 1 6 a 6 2 and 1 6 c 6 2.
10
If b = 1, the conditions become a + c 6 9 and a2 + c2 6 8, and we
must also have 1 6 ac 6 9. From the second condition, we obtain
1 6 a 6 2 and 1 6 c 6 2.
If b = 2, the conditions become a + c 6 4 and a2 + c2 6 5. Hence we
have three possibilities for (a, b).
In the above table, we have worked out all the possibilities. It turns
out that there are eight possible palindromic products: 10201, 12321,
14641, 20502, 23632, 26962, 40804 and 44944.
2. For n > 1, we have a2n = a2n−1 + n = a2n−2 + 2n, so a2n = 0 + 2 +

4 + · · · + 2n = 2(0 + 1 + 2 + · · · + n) = n(n + 1). Since n > 0, we have
n2 < n2 +n < n2 +2n+1 = (n+1)2 . Hence a2n = n(n+1) cannot be a
square; it lies between two subsequent squares. So for the even k > 2,
ak is not a square. From a2n−1 = a2n − n = n(n + 1) − n = n2 , we
deduce that a2n−1 is always a square, hence all odd k are as desired.
Finally, we note that a0 = 0 is a square as well. So ak is a square if
k is odd or k = 0, and it’s not a square otherwise.
3. We first show that there is a cycle, i.e. m distinct participants

A1 , . . . , Am such that A1 won against A2 , A2 won against A3 , and
so on, up to Am , who won against A1 . Pick an arbitrary participant
B1 . He won against another participant B2 , who won against a cer-
tain B3 , and so on. Since we have only finitely many players, we will
have a repetition in our sequence of participants, i.e., B1 won against
B2 , B2 won against B3 , . . . , Bu won against Bu+1 , where Bu+1 is a
participant that occurred earlier in our sequence, say Bt . Then Bt
up to Bu indeed form a cycle. Note that the length m of a cycle is at
least 3. In fact, we now need to show that there is a cycle of length
3.
Since there exists a cycle, we can take one of minimal length M , say
C1 , . . . , CM . If M > 3, then we consider the match between C1 and
C3 . If C3 won against C1 , then we have obtained a cycle of length
3, which is smaller than M , yielding a contradiction. On the other
hand, if C1 won against C3 , then, by removing C2 , we obtain a cycle
of length M − 1, which is again smaller than M . This is a contra-
diction as well. We deduce that M = 3 and that C1 , C2 , C3 is the
desired cycle of length 3.
11
Alternative solution 1: Pick a participant A who has won the least
number of matches, and a participant B who lost to A. Consider
the set X of participants who lost to B. Then the set X does not
contain A. If there is no triple of participants as desired, then A
must have won against all participants in X. But A also won against
B. But then A has won at least one match more than B, yielding a
contradiction.
4. Note that |AP | = |P B|, and hence that |BQ| = |QC| and |CR| =
|RA|. Since 4BP A ∼ 4BQC, it follows that |P B| |QB|
|AB| = |CB| , and hence
|P B| |AB|
that |QB| = |CB| . Moreover, since Q lies outside triangle ABC and
P lies inside of it, we have
∠QBP = ∠QBC + ∠CBP = ∠P BA + ∠CBP = ∠CBA,
from which follows that 4ABC ∼ 4P BQ. Note that 4ABC is

|AB|
|P B| times larger than 4P BQ. Similarly, it follows that 4ABC ∼
|AB|
4AP R, and note that 4ABC is |P A| times larger than 4AP R. Since
|P A| = |P B|, from which follows that |AB| |AB|
|P A| = |P B| , we deduce that
4P BQ ∼ = 4AP R. Hence |QP | = |RA| = |CR| and |P R| = |BQ| =
|QC|.
So the two pairs of opposite sides of quadrilateral P QCR have equal
length, hence quadrilateral P QCR is a parallelogram.
12
5. We first consider the effect of two consecutive steps. If before step
2k −1, the stack of cards, from top to bottom, consists of a1 , . . . , a100 ,
the first 2k − 1 cards are reversed in order, so we obtain
a2k−1 , a2k−2 . . . , a2 , a1 , a2k , a2k+1 . . . , a100 . Then, at the 2k th step,
the top 2k cards are reversed in order, yielding
a2k , a1 , a2 , . . . , a2k−2 , a2k−1 , a2k+1 , . . . , a100 . Hence the effect of these
two steps is that the 2k th card is moved up to the top, and the rest
is shifted down. So if we consider the first 100 steps, then first 2 is
put on top, then 4 is put on top of that, then 6, and so on, so we ob-
tain 100, 98, 96, . . . , 4, 2, 1, 3, 5, . . . , 97, 99. We call this a megastep.
In general, a megastep changes the order a1 , a2 , . . . , a99 , a100 into
a100 , a98 , a96 , . . . , a4 , a2 , a1 , a3 , a5 , . . . , a97 , a99 . This megastep is in-
vertible; the order after a megastep can only come from one unique
order. Now we there are only finitely many possible ways (namely
100!) to order the 100 cards, so after applying megasteps for a while,
we’ll obtain an order that has occurred before. Consider the first
megastep that yields an order that has occurred before. If this is not
the starting position, then one megastep before, the position should
have occurred earlier as well; a contradiction. Hence we return to the
starting position after a finite number of (mega)steps.
Alternative solution: We again consider a megastep. The first

(topmost) card goes to the 51st place; the 51st card goes to the 76th
place, etc. This yields a so-called cycle, which we denote as 1 → 51 →
76 → 13 → · · · → 1. From this notation, we deduce for example that
after two megasteps, the first card moves to the 76th place. If the first
card returns to its original place after k megasteps, we say that the
length of the cycle is k. It’s now clear that the 51st card also returns
to the 51st place after k megasteps. In this way, we can describe the
behavior of the cards in a megastep, with a number of cycles. Now if
we take a common multiple of the lengths of the cycles, all cards will
return to their original places.
13
Problems
1 Let ABCD be a trapezoid with AB k CD, 2|AB| = |CD| and BD ⊥

BC. Let M be the midpoint of CD and let E be the intersection BC
and AD. Let C be the intersection of AM and BD. Let N be the
intersection of OE and AB.
(a) Prove that ABM D is a rhombus.
(b) Prove that the line DN passes through the midpoint of the line
segment BE.
2 Find all functions f : R → R satisfying
f (x)f (y) = f (x + y) + xy
for all x, y ∈ R.
3 Let N be the number of ordered 5-tuples (a1 , a2 , a3 , a4 , a5 ) of positive

integers satisfying
1 1 1 1 1
+ + + + = 1.
a1 a2 a3 a4 a5
Is N even or odd?
4 The two circles Γ1 and Γ2 intersect at P and Q. The common tan-

gent that’s on the same side as P , intersects the circles at A and B,
respectively. Let C be the second intersection with Γ2 of the tan-
gent to Γ1 at P , and let D be the second intersection with Γ1 of the
tangent to Γ2 at Q. Let E be the intersection of AP and BC, and
let F be the intersection of BP and AD. Let M be the image of
P under point reflection with respect to the midpoint of AB. Prove
that AM BEQF is a cyclic hexagon.
5 For any non-negative integer n, we say that a permutation

(a0 , a1 , . . . , an ) of {0, 1, . . . , n} is quadratic if k + ak is a square for
k = 0, 1, . . . , n. Show that for any non-negative integer n, there exists
a quadratic permutation of {0, 1, . . . , n}.
14
Solutions
1 From 2|AB| = |CD| and AB k CD, we deduce that AB is a mid-

segment in triangle CDE. Hence A is the midpoint of DE. Since
∠DBE = 90◦ , by Thales’ Theorem, A is the center of the circle pass-
ing through D, B and E. Hence |AD| = |AE| = |AB|. We already
know that |AB| = |DM | = |M C|. In a similar way, using Thales’
Theorem, we obtain |BM | = |CM | = |DM |. Hence all sides of the
quadrilateral ABM D have the same length; ABM D is a rhombus
(a). The intersection of the diagonals of a rhombus is the midpoint
of both diagonals, hence O is the midpoint of BD. Since A is the
midpoint of DE, it follows that N is the centroid of triangle BDE.
Hence DN passes through the midpoint of BE (b).
2 First, note that the function given by f (x) = 0 for all x ∈ R, does
not satisfy the functional equation. Hence there exists an x0 ∈ R for
which f (x0 ) 6= 0. Substituting x = x0 and y = 0 gives f (x0 )f (0) =
f (x0 ). Since f (x0 ) 6= 0, we obtain f (0) = 1. Now substituting x = 1
and y = −1 gives f (1)f (−1) = f (0) − 1 = 1 − 1 = 0. Hence f (1) = 0
or f (−1) = 0.
We now distinguish the two cases. First, suppose that f (1) = 0, and
substitute x = 1. We then get 0 = f (1)f (y) = f (1 + y) + y for all
y ∈ R, so f (1 + y) = −y for all y ∈ R. Substituting y = t − 1 now
gives f (t) = −t + 1 for all t ∈ R. This is our first candidate solution.
Now suppose that f (−1) = 0 and substitute x = −1. We obtain
0 = f (−1)f (y) = f (−1 + y) − y for all y ∈ R, hence f (−1 + y) = y
for all y ∈ R. Substituting y = t + 1, we obtain f (t) = t + 1 for all
t ∈ R. This is our second candidate solution.
Since we have covered all the cases, these are the only two possible
solutions. Let’s check them.
If f (t) = −t + 1 for all t ∈ R, then for all x, y ∈ R:
f (x)f (y) = (−x + 1)(−y + 1) = (−x − y + 1) + xy = f (x + y) + xy.
If f (t) = t + 1 for all t ∈ R, then for all x, y ∈ R:
f (x)f (y) = (x + 1)(y + 1) = (x + y + 1) + xy = f (x + y) + xy.
Hence both candidate solutions are indeed solutions of the functional

equation.
15
3 Solution I. Consider an unordered 5-tuple that satisfies the given
equation, and suppose that it consists of the distinct integers b1 , . . . ,
bk , where bi occurs exactly ti times. Hence t1 +· · ·+tk = 5. Note that
5!
this unordered 5-tuple can be ordered in t1 !···t k!
ways. This is odd
if and only if three factors 2 occur in the denominator, which holds
if and only if 4! or 5! occurs in the denominator. Hence the only
unordered 5-tuples that can be ordered in a odd number of ways, are
those that consist of at least 4 numbers that are the same.
Such a 5-tuple has the form (a, a, a, a, b), where b can be equal to a,
and where a, b are positive integers. We obtain the equation a4 + 1b = 1,
or, equivalently, 4b + a = ab. We deduce that b has to be a divisor
of a and that a is a divisor of 4b. Hence there are three possibilities:
a = b, a = 2b and a = 4b. In the first case, we obtain 5b = b2 ,
hence b = 5, yielding the solution (5, 5, 5, 5, 5). In the second case,
we obtain 6b = 2b2 , hence b = 3, yielding the solution (6, 6, 6, 6, 3).
In the third and final case, we obtain 8b = 4b2 , so b = 2, yielding the
solution (8, 8, 8, 8, 2). Hence there are three unordered 5-tuples which
can be ordered in an odd number of ways. Hence N is odd.
Solution II. Let (a1 , a2 , a3 , a4 , a5 ) be an ordered 5-tuple that satisfies
the given equation. Then also the 5-tuple (a2 , a1 , a3 , a4 , a5 ). If a1 6=
a2 , this is a different ordered 5-tuple. Hence there is an even number
of solutions (a1 , a2 , a3 , a4 , a5 ) with a1 6= a2 . So we may only consider
the solutions with a1 = a2 . In the same way, we show that there is
an even number of solutions (a1 , a1 , a3 , a4 , a5 ) with a3 6= a4 . Hence
we may only consider the solutions with a1 = a2 , a3 = a4 . For
each solution (a1 , a1 , a3 , a3 , a5 ) with a1 6= a3 , there is yet another
solution (a3 , a3 , a1 , a1 , a5 ), hence there is an even number of solutions
(a1 , a1 , a3 , a3 , a5 ) with a1 6= a3 . Thus we may only consider the
solution of the form (a, a, a, a, b), where b can be equal to a.
We obtain the equation a4 + 1b = 1, or equivalently, 4b + a = ab. We
can rewrite this last equation as (a − 4)(b − 1) = 4. Since b is a
positive integer, we have b − 1 ≥ 0. Hence b − 1 is a positive divisor
of 4, namely 1, 2 or 4. This yields the three solutions (a, b) = (8, 2),
(a, b) = (6, 3) and (a, b) = (5, 5). Since the number of solutions that
we covered previously, is even, we deduce that N is odd.
16
4 We express all relevant angles in terms of α = ∠BAP and β =
∠P BA. The midpoint of AB is, by definition, also the midpoint of
P M , hence the diagonals of quadrilateral divide each other into equal
segments. Hence AP BM is a parallelogram and we have ∠AM B =
∠AP B. Note that (using the sum of the angles of a triangle) ∠AP B+
α + β = 180◦ . Hence 180◦ − ∠AM B = α + β.
By the inscribed angle theorem, applied on the tangent AB to Γ1 ,
we have α = ∠ADP . Then, by the inscribed angle theorem, applied
on the chord AP of Γ1 , it follows that ∠ADP = ∠AQP . Hence
α = ∠AQP . In a similar way, we deduce that β = ∠P CB = ∠P QB.
Hence ∠AQB = ∠AQP + ∠P QB = α + β = 180◦ − ∠AM B, so
AQBM is a cyclic quadrilateral.
Now let S be the intersection of DP and AB. Using the inscribed an-
gle theorem, we deduce that ∠SP B = ∠P CB = ∠P BS = ∠P BA =
β. Hence ∠DP F = ∠SP B = β, by opposite angles. Now we apply
the exterior angle theorem to 4DF P , to obtain ∠AF B = ∠AF P =
∠F DP + ∠DP F = α + β = ∠AQB. Hence AF QB is a cyclic quadri-
lateral. Hence F lies on the circumcircle of the cyclic quadrilateral
AQBM . Similarly, we see that E also lies on that circle. Hence
AM BEQF is a cyclic hexagon.
5 We use induction on n. For n = 0 the permutation (0) is quadratic,

since 0 + 0 is a square.
Now let l ≥ 0 and assume that for every n ≤ l, there exists a quadratic
permutation (induction hypothesis). Consider n = l + 1. Let m be
such that m2 is the least square greater than or equal to l + 1. Now
we have l ≥ (m − 1)2 = m2 − 2m + 1, so 2(l + 1) ≥ 2m2 − 4m + 4 =
m2 + (m − 2)2 ≥ m2 . We deduce that there exists an integer p with
0 ≤ p ≤ l + 1 such that (l + 1) + p = m2 .
Now define a permutation as follows. If p ≥ 1, then we take
(a0 , a1 , . . . , ap−1 ) a quadratic permutation of {0, 1, . . . , p − 1}; which
exists, by our induction hypothesis, since p − 1 ≤ l. For p ≤ i ≤ l + 1,
we define ai = m2 − i. (Note that for p = 0, we now also have define
a permutation of (a0 , a1 , . . . , al+1 ).)
Now the sequence (ap , ap+1 , . . . , al+1 ) is exactly (l +1, l, . . . , p), which
implies that, together with the initial part, we have used all val-
ues from 0 up to l + 1. Also, ai + i is a square for all i. Hence
(a0 , a1 , . . . , al+1 ) is a quadratic permutation of {0, 1, . . . , l + 1}.
17
Benelux Mathematical Olympiad, April 2010
Problems
1. A finite set of integers is called bad if its elements add up to 2010. A fi-
nite set of integers is a Benelux-set if none of its subsets is bad. Deter-
mine the smallest integer n such that the set {502, 503, 504, . . . , 2009}
can be partitioned into n Benelux-sets.
(A partition of a set S into n subsets is a collection of n pairwise
disjoint subsets of S, the union of which equals S.)
2. Find all polynomials p(x) with real coefficients such that
p(a+b−2c)+p(b+c−2a)+p(c+a−2b) = 3p(a−b)+3p(b−c)+3p(c−a)
for all a, b, c ∈ R.
3. On a line l there are three different points A, B and P in that order.

Let a be the line through A perpendicular to l, and let b be the line
through B perpendicular to l. A line through P , not coinciding with l,
intersects a in Q and b in R. The line through A perpendicular to BQ
intersects BQ in L and BR in T . The line through B perpendicular
to AR intersects AR in K and AQ in S.
(a) Prove that P , T , S are collinear.

(b) Prove that P , K, L are collinear.
4. Find all quadruples (a, b, p, n) of positive integers, such that p is a

prime and
a3 + b3 = pn .
18
Solutions
1. As 502 + 1508 = 2010, the set S = {502, 503, . . . , 2009} is not a

Benelux-set, so n = 1 does not work. We will prove that n = 2 does
work, i.e. that S can be partitioned into 2 Benelux-sets.
Define the following subsets of S:
A = {502, 503, . . . , 670},

B = {671, 672, . . . , 1005},
C = {1006, 1007, . . . , 1339},
D = {1340, 1341, . . . , 1508},
E = {1509, 1510, . . . , 2009}.
We will show that A ∪ C ∪ E and B ∪ D are both Benelux-sets.

Note that there does not exist a bad subset of S of one element, since
that element would have to be 2010. Also, there does not exist a bad
subset of S of more than three elements, since the sum of four or more
elements would be at least 502 + 503 + 504 + 505 = 2014 > 2010. So
any possible bad subset of S contains two or three elements.
Consider a bad subset of two elements a and b. As a, b ≥ 502 and
a + b = 2010, we have a, b ≤ 2010 − 502 = 1508. Furthermore, exactly
one of a and b is smaller than 1005 and one is larger than 1005. So
one of them, say a, is an element of A∪B, and the other is an element
of C ∪ D. Suppose a ∈ A, then b ≥ 2010 − 670 = 1340, so b ∈ D.
On the other hand, suppose a ∈ B, then b ≤ 2010 − 671 = 1339, so
b ∈ C. Hence {a, b} cannot be a subset of A ∪ C ∪ E, nor of B ∪ D.
Now consider a bad subset of three elements a, b and c. As a, b, c ≥
502, a + b + c = 2010, and the three elements are pairwise distinct,
we have a, b, c ≤ 2010 − 502 − 503 = 1005. So a, b, c ∈ A ∪ B. At least
one of the elements, say a, is smaller than 2010
3 = 670, and at least
one of the elements, say b, is larger than 670. So a ∈ A and b ∈ B.
We conclude that {a, b, c} cannot be a subset of A ∪ C ∪ E, nor of
B ∪ D.
This proves that A ∪ C ∪ E and B ∪ D are Benelux-sets, and therefore
the smallest n for which S can be partitioned into n Benelux-sets is
n = 2.
Remark. Observe that A ∪ C ∪ E1 and B ∪ D ∪ E2 are also Benelux-
sets, where {E1 , E2 } is any partition of E.)
19
2. For a = b = c, we have 3p(0) = 9p(0), hence p(0) = 0. Now set
b = c = 0, then we have
p(a) + p(−2a) + p(a) = 3p(a) + 3p(−a)
for all a ∈ R. So we find a polynomial equation
p(−2x) = p(x) + 3p(−x). (1)
Note that the zero polynomial is a solution to this equation. Now

suppose that p is not the zero polynomial, and let n ≥ 0 be the
degree of p. Let an 6= 0 be the coefficient of xn in p(x). At the
left-hand side of (1), the coefficient of xn is (−2)n · an , while at the
right-hand side the coefficient of xn is an + 3 · (−1)n · an . Hence
(−2)n = 1 + 3 · (−1)n . For n even, we find 2n = 4, so n = 2, and
for n odd, we find −2n = −2, so n = 1. As we already know that
p(0) = 0, we must have p(x) = a2 x2 + a1 x, where a1 and a2 are real
numbers (possibly zero).
The polynomial p(x) = x is a solution to our problem, as
(a+b−2c)+(b+c−2a)+(c+a−2b) = 0 = 3(a−b)+3(b−c)+3(c−a)
for all a, b, c ∈ R. Also, p(x) = x2 is a solution, since
(a + b − 2c)2 + (b + c − 2a)2 + (c + a − 2b)2

= 6(a2 + b2 + c2 ) − 6(ab + bc + ca)
= 3(a − b)2 + 3(b − c)2 + 3(c − a)2
for all a, b, c ∈ R.
Now note that if p(x) is a solution to our problem, then so is λp(x)
for all λ ∈ R. Also, if p(x) and q(x) are both solutions, then so is
p(x) + q(x). We conclude that for all real numbers a2 and a1 the
polynomial a2 x2 + a1 x is a solution. Since we have already shown
that there can be no other solutions, these are the only solutions.
20
3. Solution 1.
(a) Since P , R and Q are collinear, we have 4P AQ ∼ 4P BR,
hence
|AQ| |AP |
= .
|BR| |BP |
Conversely, P , T and S are collinear if it holds that
|AS| |AP |
= .
|BT | |BP |
So it suffices to prove
|BT | |AS|
= .
|BR| |AQ|
Since ∠ABT = 90◦ = ∠ALB and ∠T AB = ∠BAL, we have

4ABT ∼ 4ALB. And since ∠ALB = 90◦ = ∠QAB and
∠LBA = ∠ABQ, we have 4ALB ∼ 4QAB. Hence 4ABT ∼
4QAB, so
|BT | |AB|
= .
|BA| |AQ|
Similarly, we have 4ABR ∼ 4AKB ∼ 4SAB, so
|BR| |AB|
= .
|BA| |AS|
Combining both results, we get

|BT | |BT |/|BA| |AB|/|AQ| |AS|
= = = ,
|BR| |BR|/|BA| |AB|/|AS| |AQ|
which had to be proved.
(b) Let the line P K intersect BR in B1 and AQ in A1 and let the
line P L intersect BR in B2 and AQ in A2 . Consider the points
A1 , A and S on the line AQ, and the points B1 , B and T on the
line BR. As AQ k BR and the three lines A1 B1 , AB and ST
are concurrent (in P ), we have
A1 A : AS = B1 B : BT,
where all lengths are directed. Similarly, as A1 B1 , AR and SB

are concurrent (in K), we have
A1 A : AS = B1 R : RB.
21
This gives
BB1 RB1 RB + BB1 BB1 BB1
= = =1+ =1− ,
BT RB RB RB BR
so
1
BB1 = 1 1 .
BT + BR
Similary, using the lines A2 B2 , AB and QR (concurrent in P )

and the lines A2 B2 , AT and QB (concurrent in L), we find
B2 B : BR = A2 A : AQ = B2 T : T B.
This gives
BB2 T B2 T B + BB2 BB2 BB2
= = =1+ =1− ,
BR TB TB TB BT
so
1
BB2 = 1 1 .
BR + BT
We conclude that B1 = B2 , which implies that P , K and L are
collinear.

Solution 2. As ∠AKB = ∠ALB = 90◦ , the points K and L belong

to the circle with diameter AB. Since ∠QAB = ∠ABR = 90◦ , the
lines AQ and BR are tangents to this circle.
Apply Pascal’s theorem to the points A, A, K, L, B and B, all on the
same circle. This yields that the intersection Q of the tangent in A
and the line BL, the intersection R of the tangent in B and the line
AK, and the intersection of KL and AB are collinear. So KL passes
through the intersection of AB and QR, which is point P . Hence P ,
K and L are collinear. This proves part b.
Now apply Pascal’s theorem to the points A, A, L, K, B and B.
This yields that the intersection S of the tangent in A and the line
BK, the intersection T of the tangent in B and the line AL, and the
intersection P of KL and AB are collinear. This proves part a.
22
4. Let (a, b, p, n) be a solution. Note that we can write the given equation
as
(a + b)(a2 − ab + b2 ) = pn .
As a and b are positive integers, we have a + b ≥ 2, so p | a + b.
Furthermore, a2 − ab + b2 = (a − b)2 + ab, so either a = b = 1 or
a2 −ab+b2 ≥ 2. Assume that the latter is the case. Then p is a divisor
of both a+b and a2 −ab+b2 , hence also of (a+b)2 −(a2 −ab+b2 ) = 3ab.
This means that p either is equal to 3 or is a divisor of ab. Since p
is a divisor of a + b, we have p | a ⇔ p | b, hence either p = 3, or
p | a and p | b. If p | a and p | b, then we can write a = pa0 , b = pb0
with a0 and b0 positive integers, and we have (a0 )3 + (b0 )3 = pn−3 , so
(a0 , b0 , p, n − 3) then is another solution (note that (a0 )3 + (b0 )3 is a
positive integer greater than 1, so n − 3 is positive).
Now assume that (a0 , b0 , p0 , n0 ) is a solution such that p - a. From the
reasoning above it follows that either a0 = b0 = 1, or p0 = 3. After
all, if we do not have a0 = b0 = 1 and we have p0 6= 3, then p | a.
Also, given an arbitrary solution (a, b, p, n), we can divide everything
by p repeatedly until there are no factors p left in a.
Suppose a0 = b0 = 1. Then the solution is (1, 1, 2, 1).
Suppose p0 = 3. Assume that 32 | (a20 − a0 b0 + b20 ). As 32 | (a0 + b0 )2 ,
we then have 32 | (a0 + b0 )2 − (a20 − a0 b0 + b20 ) = 3a0 b0 , so 3 | a0 b0 .
But 3 - a0 by assumption, and 3 | a0 + b0 , so 3 - b0 , which contradicts
3 | a0 b0 . We conclude that 32 - (a20 − a0 b0 + b20 ). As both a0 + b0 and
a20 − a0 b0 + b20 must be powers of 3, we have a20 − a0 b0 + b20 = 3. Hence
(a0 − b0 )2 + a0 b0 = 3. We must have (a0 − b0 )2 = 0 or (a0 − b0 )2 = 1.
The former does not give a solution; the latter gives a0 = 2 and b0 = 1
or a0 = 1 and b0 = 2.
So all solutions with p - a are (1, 1, 2, 1), (2, 1, 3, 2) and (1, 2, 3, 2).
From the above it follows that all other solutions are of the form
(pk0 a0 , pk0 b0 , p0 , n0 + 3k), where (a0 , b0 , p0 , n0 ) is one of these three
solutions. Hence we find three families of solutions:
• (2k , 2k , 2, 3k + 1) with k ∈ Z≥0 ,
• (2 · 3k , 3k , 3, 3k + 2) with k ∈ Z≥0 ,
• (3k , 2 · 3k , 3, 3k + 2) with k ∈ Z≥0 .
It is easy to check that all these quadruples are indeed solutions.
23
IMO Team Selection Test 1, June 2010
Problems
1 Let ABC be an acute triangle such that ∠BAC = 45◦ . Let D a

point on AB such that CD ⊥ AB. Let P be an internal point of the
segment CD. Prove that AP ⊥ BC if and only if |AP | = |BC|.
2 Let A and B be positive integers. Define the arithmetic sequence

a0 , a1 , a2 , . . . by an = An + B. Suppose that there exists an n ≥ 0
2
such that an is a square. Let M be a positive integer such that √ M
is the smallest square in the sequence. Prove that M < A + B.
3 Let n ≥ 2 be a positive integer and p a prime such that n | p − 1 and

p | n3 − 1. Show that 4p − 3 is a square.
4 Let ABCD be a cyclic quadrilateral satisfying ∠ABD = ∠DBC. Let

E be the intersection of the diagonals AC and BD. Let M be the
midpoint of AE, and N be the midpoint of DC. Show that M BCN
is a cyclic quadrilateral.
5 Find all triples (x, y, z) of real (but not necessarily positive) numbers
satisfying
3(x2 + y 2 + z 2 ) = 1,
x2 y 2 + y 2 z 2 + z 2 x2 = xyz(x + y + z)3 .
Solutions
1 Let E be the intersection of AP and BC. Note that ∠DCA = 90◦ −

∠CAD = 90◦ − ∠CAB = 45◦ , so 4ACD is isosceles: |AD| = |CD|.
Now suppose that |AP | = |BC|. Since ∠ADP = 90◦ = ∠CDB, we
have 4ADP ∼ = 4CDB, by (SSR). Hence ∠AP D = ∠CBD, from
which follows that
∠CEA = ∠CEP = 180◦ − ∠EP C − ∠P CE
= 180◦ − ∠AP D − ∠DCB = 180◦ − ∠CBD − ∠DCB
= ∠BDC = 90◦ ,
24
hence AP ⊥ BC.
Conversely, suppose that AP ⊥ BC, or equivalently, ∠CEP = 90◦ .
Then we have
∠AP D = ∠EP C = 90◦ − ∠P CE = 90◦ − ∠DCB = ∠CBD.
Since we also have ∠ADP = 90◦ = ∠CDB, from (SAA), it follows

that 4ADP ∼ = 4CDB. Hence |AP | = |BC|.
√
2 If M ≤ A, then automatically, M < A + B, so we’re done in this
case.
Hence suppose that M > A. Let k be such that ak = M 2 . So
Ak + B = M 2 . Since 0 < M − A < M , the square (M − A)2 is smaller
than M 2 . We have (M − A)2 = M 2 − 2M A + A2 = M 2 − A(2M − A).
So if k − (2M − A) ≥ 0, then
ak−(2M −A) = A(k − (2M − A)) + B = (Ak + B) − 2M A + A2

= M 2 − 2M A + A2 = (M − A)2 ,
and in that case, M 2 is not the smallest square in the sequence,

contrary to our assumption. Hence k − (2M − A) < 0. From this, it
follows that A(k−(2M −A))+B < B, or equivalently,√(M −A)2 < B.
Since M −√A is positive, we deduce that M − A < B, hence that
M < A + B.
3 Solution I. From n | p − 1, we deduce that n < p. Since p is prime,

p divides one of the two factors of n3 − 1 = (n − 1)(n2 + n + 1), but p
is too large to be a divisor of n − 1 > 0. Hence p | n2 + n + 1. Since
n | p − 1, we can write p as kn + 1, where k is a positive integer. Since
kn + 1 | n2 + n + 1, we have
kn + 1 | k(n2 + n + 1) − (n + 1)(kn + 1) = k − n − 1.
We now distinguish three cases.

Case 1: k > n + 1. In this case, k − n − 1 is positive, and we must
have kn + 1 ≤ k − n − 1. It follows that (k + 1)n ≤ k − 2. But the
left hand side is clearly larger than the right hand side, yielding a
contradiction.
Case 2: k < n + 1. In this case, k − n − 1 is negative, and we must
have kn + 1 ≤ n + 1 − k. It follows that (k − 1)n ≤ −k. But the
25
left hand side is non-negative, and the right hand side is negative,
yielding a contradiction.
Case 3: k = n + 1. Now we have p = (n + 1)n + 1 = n2 + n + 1.
Hence 4p − 3 = 4n2 + 4n + 1 = (2n + 1)2 is a square.
We deduce that in the only possible case, 4p − 3 is a square.
Solution II. As in the first solution, we show that p | n2 + n + 1.

If p = n2 + n + 1, then we have 4p − 3 = 4n2 + 4n + 1 = (2n + 1)2 ,
which is a square. Now suppose that p 6= n2 + n + 1. Then there is
an integer m > 1 such that pm = n2 + n + 1. Reducing modulo n, we
get pm ≡ 1 mod n. From n | p − 1, we deduce that p ≡ 1 mod n,
hence m ≡ 1 mod n. So p and m are at least n + 1. However,
(n + 1)2 = n2 + 2n + 1 > n2 + n + 1 = pm, yielding a contradiction.

4 Solution I. Since ABCD is a cyclic quadrilateral, we have ∠BDC =

∠BAC = ∠BAE. Also, from the given properties, it follows that
∠CBD = ∠EBA. Hence we have, by (aa), 4DCB ∼ 4AEB. We
now show that this implies that 4N CB ∼ 4M EB.
First of all, we have ∠N CB = ∠DCB = ∠AEB = ∠M EB. Also,
|N C| 2|N C| |DC| |CB|
we have |M E| = 2|M E| = |AE| = |EB| . By (sas), we deduce that
indeed 4N CB ∼ 4M EB. Hence ∠BM C = ∠BM E = ∠BN C, and
it follows that M BCN is a cyclic quadrilateral.
Solution II. As in the first solution, we show that 4DCB ∼ 4AEB.

Now note that the lines BM and BN are median in these two similar
triangles. Hence the corresponding angles ∠BM E and ∠BN C are
equal. Hence ∠BM C = ∠BM E = ∠BN C, from which we deduce
that M BCN is a cyclic quadrilateral.
5 We first show that for reals x, y and z satisfying the first condition,
we have
x2 y 2 + y 2 z 2 + z 2 x2 ≥ xyz(x + y + z)3 . (2)
So the solutions that we are looking for, are the cases in which equality
holds in 2.
2 2
Let a, b and c be reals. Then (a−b)2 ≥ 0, so a +b
2 ≥ ab with equality
if and only if a = b. We repeat this for the pairs (b, c) and (c, a), and
after summing, it follows that
a2 + b2 + c2 ≥ ab + bc + ca
26
with equality if and only if a = b = c.
We apply this inequality to the triple (xy, yz, zx), and obtain
x2 y 2 + y 2 z 2 + z 2 x2 ≥ xy 2 z + yz 2 x + zx2 y = xyz(x + y + z)
with equality if and only if xy = yz = zx. We also apply it to (x, y, z)

to obtain x2 + y 2 + z 2 ≥ xy + yz + zx. Hence from the first condition,
it follows that
1 = 3(x2 + y 2 + z 2 ) = (x2 + y 2 + z 2 ) + 2(x2 + y 2 + z 2 )

≥ x2 + y 2 + z 2 + 2xy + 2yz + 2zx = (x + y + z)2 ,
with equality if and only if x = y = z. Now we’re going to combine

these two inequalities,
x2 y 2 + y 2 z 2 + z 2 x2 ≥ xyz(x + y + z) (3)
and
1 ≥ (x + y + z)2 (4)
2 2
To this end, we first multiply (4) by the non-negative factor x y +
y 2 z 2 + z 2 x2 ; we obtain
(x2 y 2 + y 2 z 2 + z 2 x2 ) · 1 ≥ (x2 y 2 + y 2 z 2 + z 2 x2 ) · (x + y + z)2
with equality if and only if x = y = z or x2 y 2 + y 2 z 2 + z 2 x2 = 0.

Next, we multiply (3) by the non-negative factor (x + y + z)2 , from
which follows that
(x2 y 2 + y 2 z 2 + z 2 x2 ) · (x + y + z)2 ≥ xyz(x + y + z) · (x + y + z)2
with equality if and only if xy = yz = zx or x + y + z = 0. Combining

these yields
(x2 y 2 + y 2 z 2 + z 2 x2 ) · 1 ≥ (x2 y 2 + y 2 z 2 + z 2 x2 ) · (x + y + z)2

≥ xyz(x + y + z) · (x + y + z)2 ,
with equality if and only if
x = y = z or x2 y 2 + y 2 z 2 + z 2 x2 = 0

and (xy = yz = zx or x + y + z = 0) .
From the second condition, it follows that indeed, equality must hold.
Hence (x, y, z) must satisfy the conditions above. Furthermore, we
27
still have the first condition, which says that 3(x2 + y 2 + z 2 ) = 1:
3(x2 + y 2 + z 2 ) = 1,
x=y=z or x2 y 2 + y 2 z 2 + z 2 x2 = 0,
xy = yz = zx or x + y + z = 0.
We distinguish two cases. First, suppose that x = y = z. Then

we also have xy = yz = xz. Using the first condition, we obtain
1 = 3(x2 + y 2 + z 2 ) = 9x2 = (3x)2 , hence x = y = z = ± 31 .
Next, suppose that x2 y 2 + y 2 z 2 + z 2 x2 = 0; then each term must be
zero, hence xy = yz = zx = 0. From xy = 0, we deduce without loss
of generality that x = 0, and also, from yz = 0, we deduce without
loss of generality
√ that y = 0. In this way, √ we find the√solutions
(0, 0, ± 13 3), and also the solutions (0, ± 31 3, 0) and (± 13 3, 0, 0).
Remark. Once you’ve shown (3) and (4), you can also obtain (2)
by multiplying (4) by xyz(x + y + z), but you’ll need to show that
xyz(x + y + z) is non-negative. From the second condition, we deduce
that xyz(x + y + z)3 is a sum of squares, hence non-negative. If
x + y + z 6= 0, it follows that xyz(x + y + z) ≥ 0, and otherwise,
we have xyz(x + y + z) = 0. Hence indeed, in both cases we have
xyz(x + y + z) ≥ 0.
Using this method, the equality case becomes:
3(x2 + y 2 + z 2 ) = 1,
xy = yz = zx,
x=y=z or xyz(x + y + z) = 0.
28
Problems
1. Consider sequences a1 , a2 , a3 , . . . of positive integers. Determine the

smallest possible value of a2010 if
(i) an < an+1 for all n ≥ 1,
(ii) ai + al > aj + ak for all quadruples (i, j, k, l) which satisfy 1 ≤
i < j ≤ k < l.
2. Find all functions f : R → R which satisfy
f (x) = max (2xy − f (y))

y∈R
for all x ∈ R.
(In general the expression a = max g(s) means: a ≥ g(s) for all s ∈ S
s∈S
and furthermore there is an s ∈ S for which a = g(s).)
1
b be positive integers such that M (a, b) = a −
3. (a) Let a and b +
b b + a3 is an integer. Proof that M (a, b) is a square.
(b) Find nonzero integers a and b such that M (a, b) is a positive
integer, but not a square.
4. Let ABCD be a square with circumcircle Γ1 . Let P be a point on

the arch AC that also contains B. A circle Γ2 touches Γ1 in P and
also touches the diagonal AC in Q. Let R be a point on Γ2 such that
the line DR touches Γ2 . Proof that |DR| = |DA|.
5. The polynomial A(x) = x2 + ax + b with integer coefficients has the

following property: for each prime p there is an integer k such that
A(k) and A(k + 1) are both divisible by p. Proof that there is an
integer m such that A(m) = A(m + 1) = 0.
29
Solutions
1. By induction we prove that an − a1 ≥ 2n−1 − 1 for all n ≥ 2. For

n = 2 this reduces to a2 − a1 ≥ 1 and this follows from condition (i).
Let m ≥ 2 and suppose that am − a1 ≥ 2m−1 − 1 (IH). We apply
condition (ii) with i = 1, j = k = m and l = m + 1. This yields
a1 + am+1 > 2am . So
IH
am+1 − a1 > 2am − 2a1 ≥ 2(2m−1 − 1) = 2m − 2,
and since am+1 is a positive integer, this yields am+1 − a2 ≥ 2m − 1.

We now know that for n ≥ 1:
an ≥ 2n−1 − 1 + a1 ≥ 2n−1
and in particular a2010 ≥ 22009 .

On the other hand we prove that a2010 is possible by showing that the
sequence given by an = 2n−1 satisfies the conditions. This sequence
consists of positive integers and is strictly increasing (condition (i)).
Let (i, j, k, l) be a quadruple that satisfies 1 ≤ i < j ≤ k < l. It is
true that
aj + ak = 2j−1 + 2k−1 ≤ 2k−1 + 2k−1 = 2k ≤ 2l−1 < 2i−1 + 2l−1 ,
and thus the sequence also satisfies condition (ii).

We conclude that the smallest possible value of a2010 is equal to 22009 .

2. For all x ∈ R it is true that
f (x) = max (2xy − f (y)) ≥ 2x2 − f (x),

y∈R
so 2f (x) ≥ 2x2 , that is f (x) ≥ x2 .

Because (x − y)2 ≥ 0, it is true that x2 ≥ 2xy − y 2 for all x, y ∈
R. Because we already have showed that f (y) ≥ y 2 , we know that
2xy − f (y) ≤ 2xy − y 2 ≤ x2 and consequently
f (x) = max (2xy − f (y)) ≤ x2 .

y∈R
We conclude f (x) = x2 .
30
We check if this function satisfies. Let x ∈ R be given. Because
(x − y)2 ≥ 0, it is true that x2 ≥ 2xy − y 2 for all y ∈ R with equality
iff x = y, therefore
x2 = max (2xy − y 2 ).
y∈R
3. (a) Because a + b2 is an integer, − 1b + 3b a is also an integer. We may

−a+3b2
write this as ab . We know that ab is a divisor of 3b2 − a. In
particular b is a divisor of 3b2 − a and therefore b | a. However,
this means that b2 is a divisor of ab and consequently it is also
a divisor of 3b2 − a, which yields b2 | a. We now write a = mb2
with m a positive integer. Then is is true that mb3 is a divisor
of 3b2 − mb2 , so mb is a divisor of 3 − m. This yields that m is
a divisor of 3 (that is m = 1 or m = 3) and that b is a divisor of
3 − m.
First suppose that m = 3. Then we have a = 3b2 . Filling this in
yields M (3b2 , b) = 3b2 − 1b + b2 + 1b = 4b2 and that is the square
of 2b.
Now suppose that m = 1. From b | 3 − m follows b = 1 or b = 2.
In the first case we have a = 1 and in the second case a = 4.
Filling in the first possibility yields M (1, 1) = 1 − 1 + 1 + 3 =
4, which is a square. Filling in the second possibility yields
M (4, 2) = 4 − 21 + 4 + 32 = 0, which also is a square.
We conclude that M (a, b) is a square in all cases.
(b) Take a = 4 and b = −2. Then we have M (4, −2) = 7, which is
a positive integer, but not a square.
After doing part (a) this answer is not hard to find. You know
that a has to be equal to mb2 , but now for some m ∈ Z, and that
m has to be a divisor of 3. Further more m = 3 will not work,
because it yields a square. You can try the other possibilities of
m yourself.

4. Let M be the intersection of AC and BD (that is the midpoint of

Γ1 ) and let N be the midpoint of Γ2 . First we prove that P , Q and
D are collinear.
If P = B, then Q = M and it is trivial. If not, let S be the in-
tersection of P Q and BD. We will first prove that S = D. Notice
31
that M , N and P are collinear and that QN is parallel to DB. This
yields because of F-angles that ∠P SB = ∠P QN = ∠N P Q, because
|N P | = |N Q|. By Z-angles we see that ∠P M B = ∠M N Q and be-
cause of the exterior angle theorem inside triangle P QN that is equal
to ∠P QN + ∠N P Q = 2∠P SB. So ∠P M B = ∠P SB, together with
the inscribed angle theorem this yields that S lies on Γ1 . So S = D,
from which follows that P , Q and D are collinear.
Because ∠DP B = ∠DM Q = 90◦ and ∠P DB = ∠QDM , it is true
|DP |
that 4DP B ∼ 4DM Q. This yields |DB| = |DM |
|DQ| . Since |DB| =
2|DM |, this is equivalent to |DP ||DQ| = 2|DM |2 .√The power theo-
rem on Γ2 yields |DP ||DQ| = |DR|2 , so |DR| = 2|DM | = |DA|.

5. Let p be a prime and let k be such that A(k) and A(k + 1) are both
divisible by p. The difference between A(k) and A(k + 1) is also
divisible by p and this is equal to
A(k + 1) − A(k) = (k + 1)2 + a(k + 1) + b − (k 2 + ak + b) = 2k + 1 + a,
so 2k ≡ −1 − a mod p. Because A(k) is divisible by p, the integer

4A(k) is also divisible by p, so we have modulo p that
4A(k) ≡ 4k 2 + 4ak + 4b ≡ (−1 − a)2 + 2(−1 − a)a + 4b

≡ −a2 + 4b + 1 mod p.
The right hand side is independent of k. We now see that for each
prime p the number −a2 + 4b + 1 has to be divisible by p. This is
only possible if −a2 + 4b + 1 = 0. So we have a2 = 4b + 1. We now
see that a has to be odd and we write a = 2c + 1 with c ∈ Z. Then
it is true that 4b + 1 = a2 = 4c2 + 4c + 1, so b = c2 + c. Therefore,
the polynomial can be written as A(x) = x2 + (2c + 1)x + (c2 + c)
for an integer c. We can factor this as A(x) = (x + c)(x + c + 1),
consequently x = −c and x = −c − 1 are zeroes of the polynomial.
These are both integer points. We conclude that m = −c − 1 satisfies
A(m) = A(m + 1) = 0.
32
Junior Mathematical Olympiad, October 2009
Problems
Part 1
1. The station hall of Nijmegen is tiled in a repeating

pattern with white, speckled and black tiles (see
figure). Which part is speckled?
1 1 1 1 1
A) 10 B) 9 C) 8 D) 6 E) 4
2. There are 120 chairs in a row. A number of chairs is already occupied

in such a way that if someone comes in, this person can only sit on a
chair next to someone else. What is the smallest amount of occupied
chairs such that this may happen?
A) 40 B) 41 C) 59 D) 60 E) 80
3. Anne has got a lot of tiles like in the figure. She

can lay down four of these tiles such that the white
lines form a closed circuit. What is the smallest
number of tiles with which she can lay down a
larger circuit?
A) 6 B) 8 C) 9 D) 10 E) 12
4. Bram has got coins of 5 cent, 10 cent, 20 cent and 50 cent, of each
type at least one. In total he has got 9 coins and together they are
worth 2,10 euro. How many coins of 20 cent does Bram have?
A) 1 B) 2 C) 3 D) 4 E) 5
5. Exactly in the middle of a square with side length

7 a grey square is drawn. The area of the white ?
part of the large square is three times as large as
the area of the grey square. What is the width of 7
the white ring?
A) 1 B) 1 13 C) 1 12 D) 1 23 E) 1 34
33
6. Alice, Bas, Chris, Daan and Eva know each other very well. Each
of them always tells the truth or always lies. Chris says: “Alice is
honest”, on which Eve replies: “Chris lies! ” Chris says: “Bas is a
liar.” Eva claims: “Daan is honest.” Which two persons could both
be telling the truth?
A) Alice and Bas B) Bas and Chris C) Chris and Daan
D) Daan and Eva E) Eva and Alice
7. Jan is looking for a triangle which has got circumference 21 and in-
teger side lengths. Moreover, for each pair of sides the length of one
of them has to divide the length of the other. How many different
triangle with these properties exist?
A) 0 B) 1 C) 3 D) 5 E) 6
8. How many triangles are in this figure?

A) 20 B) 25 C) 30 D) 35 E) 40
9. Vincent wants to tile a 1,30 by 1,70 meter terrace with 50 by 20

centimeter tiles. Because it is not possible to exactly tile the terrace
he has to let some pieces of some tiles outside of the terrace. How
many tiles does Vincent need to cover his terrace?
A) 22 B) 23 C) 24 D) 27 E) 28
10. A large bag contains a number of balls: more then 100, but less than
150. If you divide the balls (evenly) over 7 children, then 2 balls will
remain. If you divide the balls over 6 children, 3 balls will remain.
How many balls will remain if you divide the balls over 5 children?
A) 0 B) 1 C) 2 D) 3 E) 4
11. A pyramid with a square base is cut into two pieces

by a straight cut. Which form cannot be the cut?
A) triangle B) square C) pentagon
D) hexagon E) All four forms are possible.
34
12. Mister Jansen departs every morning at 8:00 to his work place by
car. If he (constantly) drives 40 km/h, he would arrive three minutes
late at his work place. If he drives 60 km/h, he would arrive three
minutes too early. How fast does he need to drive to be exactly on
time?
A) 48 B) 49 C) 50 D) 51 E) 52
13. The 2 × 2 × 2-cube in the figure has got four trans-

parent cubes and four nontransparent cubes. You
cannot look through the cube when looking from
the front, side and upper face. A 3 × 3 × 3-cube
has got 27 cubes. What is the smallest amount of
cubes you have to make nontransparent such that
you can’t look through the cube from the front,
side and upper face?
A) 9 B) 10 C) 12 D) 13 E) 14
14. Sir Tuinder has got a ‘half-open’ fence around his garden made of
shelves. The shelves are alternating at the front and the rear side,
exactly in front of the middle of the gap between the two shelves on
the other side. Here you can see the fence from above. The width of
the shelves is 21 cm and the thickness is 3 cm. Between the front- and
the rear side is a distance of 2 cm. How many cm has the distance
between two shelves next to each other be such that people you can’t
look into the garden from the outside?
A) 7 B) 8 C) 9 D) 10 E) 11
21 ?
3
2
15. Rectangle ABCD has got area 1. The points P , D R C

Q, R and S are the middles of the sides and T T
is the middle of RS. What is the area of triangle S Q
AQT ?
1 3 5 9 17
A) 4 B) 8 C) 16 D) 32 E) 64 A P B
35
Part 2
The answer to each problem is a number.
1. The numbers on the faces of this cube are six con- 11

secutive integers. If we add up the numbers on
two opposing faces, we will always get the same 15
answer. What is the sum of all six numbers? 14
2. The entrance tickets of a amusement park normally cost 38 euro, but

when it is a rainy day, you can go in for half of that price. Last week
there 800 tickets were sold, together costing 19057 euro. How many
tickets were sold for half the price?
3. In a 4 × 4 × 4-cube each small cube has got 3, 4,

5 or 6 neighbors (that touch in a face to the small
cube). What is the average number of neighbors
that a small cube has got?
4. Jaap has got eight integers. Of each integer the first and last dig-
its are 1 and the other digits are 0. The eight integers have got
2, 3, 5, 9, 17, 33, 65 and 129 digits. If Jaap multiplies these eight digits
and adds up the digit of the result, which number will he get?
5. A rectangular table has got three rows and two

columns. On how many ways can you fill in the
numbers 1 to 6 in the six positions in the table
such that in each row the first number is greater
than the second?
36
6. A square with area 54 is divided into four equal
squares. The upper left square is colored grey;
the lower right square is again divides into four
equal squares, and so on. The pattern continues
infinitely long. What is the area of the grey region?
7. Start with a number of two digits (the first digit is nonzero) and do
the following: multiply the two digits and add the same two digits
to the product. In this way 27 will yield 14 + 2 + 7 = 23. For how
many two digit numbers the result is equal the the number, you began
with?
8. A line segment is drawn in between the points (36,22)
(1, 1) and (36, 22). How many points with inte-

ger coordinates lie on this line segment, the end
(1,1)
points included?
9. Anneke can drive from A to B on the high way or via a shorter road.
On the high way Anneke can drive 120 km/h and on the shorter road
she can drive 60 km/h. The high way is 8 km longer, but the trip
takes 8 minutes less. How many kilometers long is the high way from
A to B?
32
10. In the figure you see a square. The square is di-
vided into four rectangles. Of three rectangles the
circumference is given and written inside the rect- 36 42
angle. What is the area of the fourth rectangle?
37
Solutions
Part 1
1
1. C) 8 6. D) Daan and Eva 11. D) hexagon
2. A) 40 7. C) 3 12. A) 48
3. E) 12 8. D) 35 13. A) 9
4. A) 1 9. B) 23 14. C) 9
5. E) 1 34 10. A) 0 15. C) 5
16
Part 2
1. 81 5. 90 8. 8
2. 597 6. 18 9. 32
3. 4 12 7. 9 10. 42
4. 111
| .{z
. . 111}
256 ones
38
WISKUNDE
Hoofdsponsors van de OLYMPIADE
NEDERLANDSE
WISKUNDE 2010
OLYMPIADE
Overige sponsors
en donateurs
Contents
1 Introduction
18 Benelux Mathematical Olympiad, April 2010
29 IMO Team Selection Test 2, June 2010 Centraal Bureau voor de Statistiek

Introduction
In 2010 the Dutch Mathematical Olympiad consisted of three rounds. The
first round was held on 29 January 2010 at the participating schools. The
paper consisted of eight multiple choice questions and four open-answer
questions, to be solved within 2 hours. In total 4150 students of 226 sec-
ondary schools participated in this first round.
In March we organised a new round at ten universities in the country. This

round contained five open-answer questions, and two problems for which
the students had to give extensive solutions and proofs. The contest lasted
2,5 hours. Those students from grade 5 (4, 6 3) that scored 14 (12, 10)
points or more on the first round (out of a maximum of 36 points) were
invited to this new second round.
From those 599 participants to the second round, the best students were
invited for the final round. Those students from grade 5 (4, 6 3) that scored
28 (18, 14) points or more on the second round (out of a maximum of 40
points) were invited to the final round. Also some outstanding participants
in the Kangaroo math contest or the Pythagoras Olympiad were invited.
We organised training sessions at the ten universities for the 154 students
who had been invited for the final round. Former Dutch IMO-participants
were involved in the training sessions at each of the universities.
proofs. They were allowed 3 hours for this round. After the prizes had been
concluded its 49th edition 2010.

On 18 March 2011 the first selection test was held. The best ten students
participated in the third Benelux Mathematical Olympiad (BxMO), held
in Mersch, Luxembourg.
1
In June, out of those 10 students and 1 reserve candidate, the team for
the International Mathematical Olympiad 2011 was selected by two team
selection tests on 8 and 11 June 2011. A seventh, young, promising student
was selected to accompany the team to the IMO. The team had a training
camp on Texel, one of the Dutch Frisian Islands, from 9 until 16 July,
together with the team from New Zealand.
For younger students we organised the third Junior Mathematical Olympiad

in October 2010 at the VU University Amsterdam. The students invited
to participate in this event were the 30 best students of grade 1, grade 2
and grade 3 of the popular Kangaroo math contest. The competition con-
sisted of two one-hour parts, one with fifteen multiple choice questions and
one with ten open-answer questions. The goal of this Junior Mathematical
Olympiad is to scout talent and to stimulate them to participate in the
first round of the Dutch Mathematical Olympiad.
The Dutch team for IMO 2011 Amsterdam consists of
• Ragnar Groot Koerkamp (16 y.o.)

• Jeroen Huijben (15 y.o., observer C at IMO 2010)
• Madelon de Kemp (18 y.o., bronze medal in IMO 2010)
• Daniël Kroes (17 y.o., honourable mention in IMO 2010)
• Merlijn Staps (16 y.o., bronze medal in IMO 2010)
• Jetze Zoethout (16 y.o.)
• Jeroen Winkel (14 y.o.)
• Johan Konter (team leader), Utrecht University

• Sietske Tacoma (deputy leader), Utrecht University
the solutions.
2
Problems
A-problems
A1. Consider figures consisting of three circles and two lines. What is the
maximal number of intersection points in such figures?
A) 15 B) 16 C) 18 D) 19 E) 20
A2. A test consists of six questions worth successively 1 to 6 points. If your

answer to a question is correct, the number of points that that question
is worth will be added to your score. If your answer is incorrect, that
number of points will be deducted from your score. So, if you only answered
questions 1, 3 and 4 correctly, your score will be 1 − 2 + 3 + 4 − 5 − 6 = −5.
What is the number of possible scores you can get for this test?
A) 20 B) 22 C) 41 D) 43 E) 64
A3. A regular hexagon ABCDEF has area 1. What is

the area of the kite ACDE?
√ √
A) 32 B) 12 3 C) 65 D) 34 E) 14 6
A4. Three players play a game with coins. Each round the player (or one of
the players) who has the largest amount of coins will put one coin in a vase
and then he will give each of the other players one coin. The vase is empty
when the game starts and the three players possess respectively 13, 14 and
15 coins. The game ends when one of the players has lost all his coins.
How many coins will be in the vase when the game ends?
A) 36 B) 37 C) 38 D) 39 E) 40
A5. What is the last digit of ((((76 )5 )4 )3 )2 ?

A) 1 B) 3 C) 5 D) 7 E) 9
√ 7 √ 7 2 √ 7 √ 7 2
A6. Calculate 2+1 + 2−1 − 2+1 − 2−1 .
√
A) 2 B) 4 C) 8 2 D) 128 E) 512
3
A7. An odometer indicates that an car has driven 2010 km. The odometer
consists of six gears and there are no decimals; so the odometer displays
002010. However, each of the gears misses the digit 4 and will hop from
3 to 5 directly. What is the actual number of kilometres that the car has
driven?
A) 1409 B) 1467 C) 1647 D) 1787 E) 1809
A8. Thirty people with different lengths are positioned in a rectangle of six rows
each containing five persons. From each row we select the shortest person
and from these six shortest persons we select the tallest; that is Piet. We
also select the tallest person from each row and select the shortest from
these six tallest persons; that is Jan. Then we put all thirty people in one
line ordered by their length; the shortest person is standing on the left end
and the tallest on the right end of the line. On which position can Jan not
stand?
A) 21 positions left of Piet D) 19 positions right of Piet
B) 19 positions left of Piet E) 21 positions right of Piet
C) next to Piet
B-problems
?
B1. Seven equally long matches are situated on a table
as illustrated. How many degrees is the indicated
angle?
B2. How many positive integers a exist such that dividing 2216 by a gives
remainder 29?
B3. A figure consists of a square ABCD and a semicir-

cle with diameter AD outside of the square. The
square has side length 1. What is the radius of the
circumscribed circle of the figure?
4
B4. On a board containing 28 rows and 37 columns a number will be written in
red in each of the squares in the following way: in the top row the numbers
1 to 37 will be written from left to right, in the second row the numbers
38 to 74, etcetera.
In green a number will be written in each of the squares in the following
way: in the leftmost column the numbers 1 to 28 will be written from top
to bottom, in the column next to that column the numbers 29 to 56 will
be written, etcetera.
In the square in the top left corner the number 1 is written in red and
green. Add the red numbers of the squares in which the numbers written
in red and green are the same. What is the sum of these numbers?
Solutions
A-problems
A1. D) 19 Any two circles intersect in at most 2

points, a circle and a line intersect in at most 2 points
and two lines intersect in at most 1 point. Therefore,
the number of intersections cannot be more than 6 +
6 + 6 + 1 = 19. You can easily check that this number of intersections can
be realized, see for example the figure on the right.
A2. B) 22 The possible scores are exactly the odd numbers

−21, −19, . . . , 19, 21. To see that only odd numbers can occur, first con-
sider the perfect score: 21 points. Now for every wrong answer, an even
number must be subtracted from it, leaving an odd number.
Conversely, all odd numbers between −21 and 21 are possible scores. In-
deed, if none or exactly one question is answered incorrectly, the resulting
scores are 21, 19, 17, 15, 13, 11 and 9. Answering question 6 and one of the
first four questions incorrectly results in scores of 7, 5, 3 and 1. Similarly,
answering at most one question correctly, or answering question 6 and one
of the first four questions correctly, result in scores of −21, −19, . . . , −1.
5
A
A3. B) 2
3 Connect the centre M to A, C and E. B F
Also connect C and E. Then the hexagon is split

M
into six equal triangles, four of which form the kite.
C E
A4. B) 37 After three rounds, every player has lost 1 chip, and the pool
has gained 3. After 12 · 3 rounds, the players have 1, 2 and 3 chips left
respectively and 36 chips are in the pool. In the next and final round, one
more chip is added to the pool.
A5. A) 1 If two numbers have a 1 as their last digit, then so does

their product. Since the last digit of 74 = 2401 equals 1, the same holds
for every power of 74 . In particular, the last digit of ((((76 )5 )4 )3 )2 =
76·5·4·3·2 = (74 )180 is equal to 1.
√ √
A6. B) 4 Let a = ( 2 + 1)7 and b = ( 2 − 1)7 . Then the given
expression is equal to
√ √
(a + b)2 − (a − b)2 = 4ab = 4( 2 + 1)7 ( 2 − 1)7
√ √
= 4(( 2 + 1)( 2 − 1))7 = 4 · 17 = 4.
A7. B) 1467 After 9 miles, the second digit from the right increases. After
nine increments, so after 9 · 9 miles, the third digit from the right increases.
After 9 · 9 · 9 = 729 miles the fourth digit from the right increases. As
the display shows 002010, the total distance travelled must be equal to
2 · 729 + 1 · 9 = 1467 miles.
A8. E) 21 places to the right of Paul Among 6 selected people, Paul is tallest.
Therefore at least 5 people are shorter than Paul. James is shortest among
6 selected people, so at least 5 people are taller than James. James can
therefore not be 21 places to the right of Paul, since then at least 5 + 1 +
20 + 1 + 5 = 32 people would be standing in line.
The other positions are indeed possible. Here we only show this for an-
swer C: James and Paul stand next to each other. Number the people in
increasing height from 1 to 30. Put numbers 1, 2, 3, 4 and 10 in one row
and put one of the numbers 5, 6, 7, 8 en 9 in each of the other rows. The
rest may be distributed freely over the remaining positions. The shortest
6
people in the six rows are numbers 1, 5, 6, 7, 8 and 9. Therefore Paul has
number 9. James has number 10, since he is the tallest of his row and in
every other row, there is someone with a number bigger than 10.
B-problems
B1. 100 The outer four matches form a paral- 180−α 120−α
α α
lelogram. The four angles at the bases of the two 60
isosceles triangles are therefore all equal. This an- 120−α

gle, denoted by α, equals 180 degrees minus the 60
angle we are looking for. The inner three matches α
α
form an equilateral triangle (angles equal to 60 de-
grees). As the angles of any triangle add up to 180 degrees, we have that
(180 − α) + 2(120 − α) = 180. It follows that α = 80 and therefore the
angle in the question measures 100 degrees.
B2. 4 The statement that dividing 2216 by a leaves a remainder

of 29, is the same as saying that a divides 2216 − 29 = 2187 and is larger
than 29 (the remainder is always smaller than the divisor a). The divisors
of 2187 = 37 larger than 29 are 81, 243, 729 and 2187. There are 4 choices
for a.
5
B3. 6 Let O be the centre of the circum-
scribed circle, E the point where it touches the semi-
circle and let M be the midpoint of BC. Applying
the Pythagorean theorem gives |OC|2 = |M C|2 +
|OM |2 . Since |OM | = |EM | − |OE| = 32 − |OC| this
implies that |OC|2 = ( 12 )2 + ( 32 − |OC|)2 . Solving
this equation, we obtain |OC| = 65 .
B4. 5185 Number the rows of the array from top to bottom using 0 to
27 and the columns from left to right using 0 to 36. Consider the position
in row r and column k. The red number written equals 1 + k + 37r and
the green number equals 1 + r + 28k. These two numbers are equal exactly
when 36r = 27k, that is, if and only if 4r = 3k. As solutions we obtain for
r the multiples of three: 0, 3, . . . , 27 and for k the matching multiples of
four 0, 4, . . . , 36. The coloured numbers in the corresponding ten positions
are 1, 1 + 115, 1 + 2 · 115, . . . , 1 + 9 · 115. Adding these numbers we find the
solution: (1 + (1 + 9 · 115)) · 5 = 5185.
7
Second Round, March 2010
Problems
B-problems
B1. Alice has got five real numbers a < b < c < d < e. She takes the sum
of each pair of numbers and writes down the ten sums. The three small-
est sums are 32, 36 and 37, while the two largest sums are 48 and 51.
Determine e.
B2. Let AB be a diameter of a circle. Point C is the point on segment AB

such that
2 · |AC| = |BC|. The points D and E lie on the circle such that CD is
perpendicular to AB and such that DE is also a diameter of the circle.
Write the areas of the triangles ABD and CDE as O(ABD) and O(CDE).
O(ABD)
Determine the value of O(CDE) .
B3. A 24-hour digital clock displays the times from 00:00:00 till 23:59:59 during
the day. You can add the digits of the time on every second of the day;
this will give you an integer. For example, at 13:07:14 you will get 1 + 3 +
0 + 7 + 1 + 4 = 16. When you write down this sum for every possible state
of the clock and then take the average of all these numbers, what will be
the result?
B4. For the infinite sequence of numbers
0, 1, 2, 2, 1, −1, −2, −1, 1, 3, . . .
the following rule holds. For each four consecutive numbers

. . . , a, b, c, d, . . . of the sequence the number d is equal to c minus the small-
est of the two numbers a and b. For example, the ninth number of the
sequence is equal to −1 − (−2) = 1 and the tenth number is equal to
1 − (−2) = 3. Calculate the 100th number of this sequence.
8
B5. Raymond has got five coins. On the heads side of each coin is the number 1.
On the tails sides of the coins are the fractions 12 , 31 , 14 , 15 and 16 respectively.
Because every coin has got either its heads side or its tails side facing up,
there are 32 ways to put the five coins on the table. Raymond multiplies
the five numbers facing up for each of these 32 situations and writes down
all results.
If Raymond adds up these 32 numbers, what will be the final result?
C-problems
For the C-problems not only the answer is important; you also have to describe the way you
solved the problem.
C1. Determine all positive integers n consisting of four digits for which it holds
that n plus the sum of the digits of n is equal to 2010.
C2. Line segment AB has length 10. A point C lies on line segment AB such
that |AC| = 6 and |CB| = 4. Two points X and Y lie on the same side
of the line AB, such that |Y B| = |Y C| = 3, |XA| = 8 and |XC| = 6.
Determine the length of line segment XY .
Solutions
B-problems
B1. Out of the ten sums 32, 36, 37, . . . , 48, 51 the largest one is d + e and the
second largest one is c+e. Therefore d+e = 51 and c+e = 48. Furthermore
a + b is the smallest sum and a + c the second smallest, so a + b = 32 and
a + c = 36.
The third smallest sum could be either a + d or b + c. However, we know
that
a + d = (a + c) + (d + e) − (c + e) = 36 + 51 − 48 = 39.
Apparently a + d is not the third smallest sum, so b + c = 37.

Combining the things we have found so far yields
2e = 2(c + e) − (a + c) − (b + c) + (a + b) = 2 · 48 − 36 − 37 + 32 = 55.
55
Thus, the answer is e = 2 .
9
B2. Let M be the midpoint of the circle. Tri-
angles CDM and CEM have equal areas,
because they have bases of the same length
|DM | = |EM | and the same height. This
yields
O(CDE) = 2 · O(CDM ).
Since |AC| = 31 |AB| and |AM | = 12 |AB|

are true, |CM | = |AM | − |AC| = 21 |AB| −
1 1
3 |AB| = 6 |AB|. Because triangles ABD
and CDM have the same height with respect to the bases AB and CM ,
O(ABD) = 6 · O(CDM ).
O(ABD) 6
Combining the two equalities yields O(CDE) = 2 = 3.
B3. Let S be the number of possible states of the clock (for this problem it is
not necessary to calculate S). If we would write down only the last digit for
1
each state, then we would write down each of the digits 10 S times. Thus,
1 45
the total sum of these last digits is 10 S (0 + 1 + 2 + . . . + 9) = 10 S = 92 S.
If we would do the same for the second last digits, then we would only
write down the digits 0 to 5 each exactly 16 S times. Thus, the total sum
of these digits is 16 S (0 + 1 + . . . + 5) = 15 5
6 S = 2 S.
Analogously the sum of the digits in the middle two positions is equal to
9 5
2 S + 2 S.
We have to pay special attention to the first two digits, representing the
hours of the time, because every digit does not appear evenly frequent.
1
However, each of the digit pairs 00, 01, . . . , 23 appears exactly 24 S times.
Thus, the total sum of the first to digits is
1
24 S ((0 + 0) + (0 + 1) + (0 + 2) + . . . + (2 + 3))
1
= 24 S (2(0 + 1 + . . . + 9) + 10 · 1 + 4 · 2 + (0 + 1 + 2 + 3))
= 114 19
24 S = 4 S.
The total sum of all digits of all possible states is 2· 92 S +2· 52 S + 19 75

4 S = 4 S.
75 3
The average sum is the total sum divided by S and that is 4 (= 18 4 ).
10
B4. It is easy to calculate some more numbers of the sequence:
0, 1, 2, 2, 1, −1, −2, −1, 1, 3, 4, 3, 0, −3, −3,

0, 3, 6, 6, 3, −3, . . .
We now see a clear pattern: after fifteen terms the sequence repeats itself,
but all the terms are three times as large. By considering the following
argument, we can verify that the sequence indeed does follow this pattern.
Each number of the sequence is uniquely determined by its three prede-
cessors. Assume that three consecutive numbers in the sequence a, b, c will
repeat fifteen positions further, multiplied by 3, however:
. . . , a, b, c, d, . . . , 3a, 3b, 3c, . . .
The successor of 3c is (by definition) equal to 3c minus the smallest of 3a

and 3b. That is exactly three times as much as: c minus the smallest of a
and b. So this is exactly three times d. If three consecutive numbers in the
sequence repeat themselves after fifteen positions (with an extra factor 3),
this also applies to the next number and therefore the number after the
next number, etcetera.
Because of this pattern we see that the hundredth number is equal to the
tenth number (3), but then multiplied 90 15 = 6 times by 3. This yields:
3 · 36 = 2187.
B5. The results that Raymond will get

correspond exactly to the terms
of the
expansion of the product 1 + 21 1 + 13 1 + 14 1 + 15 1 + 16 .

Therefore, the total sum is
1 + 12 1 + 13 1 + 41 1 + 15 1 + 61 = 32 · 43 · 54 · 56 · 67 = 27 .

C-problems
C1. Suppose that n is an integer consisting of four digits. We use the notation
n = 1000a+100b+10c+d, where a, b, c and d are the digits of n. Therefore,
0 6 a, b, c, d 6 9 and because a number does not start with a zero, a > 0.
We now want to know for which n it is true that n + (a + b + c + d) =
1001a+101b+11c+2d = 2010. Because 1001·3 is already larger than 2010,
a has to be smaller than 3, so a = 1 or a = 2. We try both possibilities.
1. If a = 1, then 101b + 11c + 2d = 2010 − 1001 · 1 = 1009. We will
determine the possible values of b.
11
It is true that 0 6 11c + 2d 6 13 · 9 = 117. Because 101b =
1009 − (11c + 2d), this yields 1009 > 101b > 1009 − 117 = 892.
We are looking for multiples of 101 between 892 and 1009, because
101b is a multiple of 101. The only multiple of 101 between 892 and
1009 is 909 and therefore b must be equal to 9.
Now 11c + 2d = 1009 − 909 = 100 has to be true. We will determine
the possible values of b.
It is true that 0 6 2d 6 2 · 9 = 18. Because 11c = 100 − 2d, this also
yields 100 > 11c > 100 − 18 = 82. Since 11c = 100 − 2d is even, 11c
has to be equal to an even multiple of 11 between 82 and 100. The
only even multiple of 11 between 82 and 100 is 88. Therefore, c = 8
and d = 6 have to be true.
We find n = 1986 as a possible solution.
2. If a = 2, then 101b + 11c + 2d = 2010 − 1001 · 2 = 8.
Now we see that b = c = 0 and d = 4.
We find n = 2004 as possible solution.
Finally, we check if the solutions 1986 and 2004 really suffice. Indeed,
1986 + 1 + 9 + 8 + 6 = 2010 and 2004 + 2 + 0 + 0 + 4 = 2010.
So, n = 1986 and n = 2004 are the only two solutions.
C2. Let D be the midpoint of BC and M

the midpoint of AX.
Since |AM | = |M X|, |AC| = |CX|
and |M C| = |M C|, the triangles AM C
and XM C are congruent (SSS). This
yields ∠CM A = ∠CM X. Because
∠CM A + ∠CM X = 180◦ , this yields
∠CM A = ∠CM X = 90◦ .
Analogously, the triangles CDY and
BDY are congruent. This yields ∠CDY = ∠BDY = 90◦ .
|CD| |CY | 1 ◦
It is true that |AM | = |AC| = 2 . Combined with ∠AM C = ∠CDY = 90
this yields that triangles AM C and CDY are similar (ssr).
We now know that ∠ACM = ∠CY D and because of that ∠ACM +
∠DCY = ∠CY D + ∠DCY = 180◦ − ∠CDY = 90◦ . This yields that
∠M CY = 180◦ − ∠ACM − ∠DCY = 90◦ .
Draw the line through Y perpendicular to AX. Let F be the intersection
point of this line with AX. We see that M CY F is a rectangle.
The Pythagorean theorem yields |M C|2 = |XC|2 − |M X|2 = 62 − 42 = 20.
It is also true that |XF | = |M X| − |M F | = |M X| − |CY | = 1 and
2
|F Y | = |M C|. Applying the Pythagorean theorem
√ again yields |XY | =
2 2 2
|F Y | + |F X| = |M C| + 1 = 21 and |XY | = 21.
12
Problems
solved the problem.
1. Consider a triangle ABC such that ∠A = 90◦ , ∠C =

60◦ and |AC| = 6. Three circles with centers A, B B
and C are pairwise tangent in points on the three
sides of the triangle.
Determine the area of the region enclosed by the
three circles (the grey area in the figure).
C A
2. A number is called polite if it can be written as m+(m+1)+· · ·+(n−1)+n,

for certain positive integers m < n. For example: 18 is polite, since 18 =
5 + 6 + 7. A number is called a power of two if it can be written as 2` for
some integer ` > 0.
(a) Show that no number is both polite and a power of two.
(b) Show that every positive integer is polite or a power of two.
3. Consider a triangle XY Z and a point O in its X
interior. Three lines through O are drawn, par- F

A f E
allel to the respective sides of the triangle. The a
e
intersections with the sides of the triangle de- B b O
d Z
c
termine six line segments from O to the sides of D
the triangle. C
The lengths of these segments are integer num- Y
bers a, b, c, d, e and f (see figure).
Prove that the product a · b · c · d · e · f is a perfect square.
4. (a) Determine all pairs (x, y) of (real) numbers with 0 < x < 1 en 0 <
y < 1 for which x + 3y and 3x + y are both integer. An example is
(x, y) = 83 , 78 , because x + 3y = 83 + 21 24 9 7
8 = 8 = 3 en 3x + y = 8 + 8 =
16
8 = 2.
(b) Determine the integer m > 2 for which there are exactly 119 pairs
(x, y) with 0 < x < 1 en 0 < y < 1 such that x + my en mx + y are
integer.
Remark: if u 6= v, the pairs (u, v) and (v, u) are different.
13
5. Amber and Brian are playing a game using 2010 coins. Throughout the
game, the coins are divided into a number of piles of at least 1 coin each. A
move consists of choosing one or more piles and dividing each of them into
two smaller piles. (So piles consisting of only 1 coin cannot be chosen.)
Initially, there is only one pile containing all 2010 coins. Amber and Brian
alternatingly take turns to make a move, starting with Amber. The winner
is the one achieving the situation where all piles have only one coin.
Show that Amber can win the game, no matter which moves Brian makes.
Solutions
1. We recognize triangle ABC to be half an equilat-

eral triangle. This implies that |BC| = 2|AC| =
12.
p The Pythagorean √ theorem√ yields: |AB| =
|BC|2 − |AC|2 = 108 = 6 3. B
Denote the pairwise tangent points of the three
circles by D, E and F (see figure) and the radii of
D
the three circles by rA , rB and rC . The strategy F
will be to determine the area of the three circu-
lar sectors and subtract them from the area of C E A
triangle ABC.
We √ |AC|+|AB|−|BC|
√ see that 2rA = (r√A +rC )+(rA +rB )−(rB +rC ) = √ =
6 3 − 6, so rA = 3 √3 − 3. It follows that rB = 6 3 − rA = 3 3 + 3 en
rC = 6 − rA = 9 − 3 3.
2
The area of a circle of radius r equals πr√ . Therefore,√the area of circular
sector AF E equals 360 ·πrA , or 4 π(36−18 3) = 9π− 92 3π. For the area of
90 2 1
30
√
circular sectors BDF and CED we find, respectively, 360 2
πrB = 3π + 32 3π
60 2
√
and 360 πrC = 18π − 9 3π.
√
Since ABC has an area of 21 · |AB| · |AC| = 18 3, we obtain a value of
√ √ √ √ √ √
18 3 − (9π − 92 3π) − (3π + 32 3π) − (18π − 9 3π) = 18 3 − 30π + 12 3π
for the area of the grey region.
2. (a) Suppose that k = m+(m+1)+· · ·+(n−1)+n is a polite number. The

sum formula for arithmetic sequences gives k = 12 (m + n)(n − m + 1).
As m and n are different positive numbers, m+n > 3 and (n−m)+1 >
2 must hold.
Since (m + n) + (n − m + 1) = 2n + 1 is odd, one of the numbers
m + n and n − m + 1 is odd. Hence 2k = (m + n)(n − m + 1) has
14
an odd divisor (greater than 1) and is therefore not a power of two.
This implies that k is not a power of two either.
We conclude that no number can be both polite and a power of two.
(b) Suppose that k is a positive integer, not a power of two. We will
show k to be a polite number. Collecting all factors 2, we can write
k = c · 2d , where c is odd and d > 0 is a nonnegative integer. The
assumption that k is not a power of two, means that c > 1. We need
to find n > m such that m + · · · + n = 12 (m + n)(n − m + 1) = c · 2d , or
(m + n) · (n − m + 1) = c · 2d+1 . We can achieve this by choosing m en
n in such a way that m + n = c and n − m + 1 = 2d+1 , or conversely:
m + n = 2d+1 and n − m + 1 = c. To ensure that m will be positive,
we consider two cases.
For c > 2d+1 we solve: m + n = c, n − m + 1 = 2d+1 . This gives
m = (c − 2d+1 + 1)/2 and n = (c + 2d+1 − 1)/2. Obviously, n > m
(since 2d+1 > 2). Both m and n are integers (the numerators are even
since c is odd) and positive by the assumption c > 2d+1 .
For c < 2d+1 we solve: m + n = 2d+1 , n − m + 1 = c. This gives
m = (2d+1 − c + 1)/2 and n = (2d+1 + c − 1)/2. Clearly, n > m holds
(since c > 1) and both m and n are positive integers.
3. Since AO and XZ are parallel, ∠OAB = ∠ZXY X

are corresponding angles. Similarly, since BO F
A f E
and Y Z are parallel, ∠ABO = ∠XY Z holds. a
e
We deduce that 4OAB ∼ 4ZXY (equal an- B b O
d Z
gles). Hence there is a scaling factor u such c
that a = u|XZ| and b = u|Y Z|. Using simi- D
lar arguments we find that 4OCD ∼ 4XY Z Y C
and 4OEF ∼ 4Y ZX. So there are scal-
ing factors v and w such that c = v|XY |, d = v|XZ|, e = w|Y Z| and
f = w|XY |.
We now see that a · c · e = uvw · |XY | · |Y Z| · |ZX| = b · d · f . This implies
that a · b · c · d · e · f = (a · c · e)2 , which is a perfect square since a, c and e
are integers.
4. (a) Suppose that (x, y) is such a pair and consider the integers a = x + 3y
and b = 3x + y. From 0 < x, y < 1 it follows that 0 < a, b < 4, or:
1 6 a, b 6 3.
Conversely, let a and b be integers such that 1 6 a, b 6 3. There is a
unique pair of numbers (x, y) that satisfies a = x + 3y en b = 3x + y.
Indeed, combining the two equations, we get 3b − a = 3(3x + y) −
15
(x + 3y) = 8x and 3a − b = 8y. In other words x = (3b − a)/8 and
y = (3a − b)/8 (and these x and y do satisfy the two equations). If we
substitute 1, 2, 3 for a and b, we obtain the following nine pairs (x, y):
( 82 , 28 ), ( 58 , 18 ), ( 88 , 08 ), ( 18 , 58 ), ( 48 , 84 ), ( 78 , 38 ), ( 08 , 88 ), ( 38 , 78 ), ( 86 , 68 ).
The condition 0 < x, y < 1 rules out the two candidates (x, y) = ( 88 , 80 )
and (x, y) = ( 08 , 88 ). This leaves the 7 pairs we were asked to find.
(b) Suppose that 0 < x, y < 1 holds and that a = x + my and b = mx + y
are integers. Then 1 6 a, b 6 m holds.
Given integers a and b with 1 6 a, b 6 m, there is a unique pair (x, y)
for which x + my = a and mx + y = b hold. Indeed, combining the
two equalities gives : mb − a = (m2 − 1)x and ma − b = (m2 − 1)y,
or: x = (mb − a)/(m2 − 1) and y = (ma − b)/(m2 − 1). These x and
y indeed satisfy the two equations.
For given a and b, we determine whether the corresponding numbers
x and y satisfy 0 < x, y < 1. From 1 6 a, b 6 m it follows that
x > (m · 1 − m)/(m2 − 1) = 0 and x 6 (m · m − 1)/(m2 − 1) = 1.
The cases x = 0 and x = 1 exactly correspond to (a, b) = (m, 1)
and (a, b) = (1, m) respectively. Similarly, 0 < y < 1 holds, unless
(a, b) = (1, m) or (a, b) = (m, 1). Among the m2 possible pairs (a, b),
there are exactly two for which (x, y) is not a solution. In total there
are m2 − 2 solutions (x, y).
From m2 − 2 = 119, we see that m = 11.
5. A strategy that guarantees a win for Amber is as follows. In Amber’s turn,

she splits every pile with an even number of coins (say 2k) in two piles with
an odd number of coins: 1 coin and 2k − 1 coin respectively. The piles
having an odd number of coins, she leaves untouched. So in her first turn,
she created one pile of 1 coin and one of 2009 coins.
When Brian gets to make a move, all piles will have an odd number of
coins. He is therefore forced to split an odd pile, creating a new pile with
an even number of coins. This implies that Amber, in the next turn, can
continue her strategy, since there will be at least one even pile.
With each turn, the number of piles increases, so after at most 2009 turns,
the game is over. Since Brian always creates an even pile, the game cannot
end during his turn. Therefore, it will be Amber who wins the game.
16
Problems
1. All positive integers are coloured either red or green, such that the following
conditions are satisfied:
• There are equally many red as green integers.
• The sum of three (not necessarily distinct) red integers is red.
• The sum of three (not necessarily distinct) green integers is green.
Find all colourings that satisfy these conditions.
2. In an acute triangle ABC the angle ∠C is greater than ∠A. Let E be such
that AE is a diameter of the circumscribed circle Γ of 4ABC. Let K be
the intersection of AC and the tangent line at B to Γ. Let L be the the
orthogonal projection of K on AE and let D be the intersection of KL
and AB.
Prove that CE is the bisector of ∠BCD.
3. Find all triples (x, y, z) of real numbers that satisfy
x2 + y 2 + z 2 + 1 = xy + yz + zx + |x − 2y + z|.
4. Let n > 2 be an integer. Let a be the greatest positive integer such that
2a | 5n − 3n . Let b be the greatest positive integer such that 2b 6 n.
Prove that a 6 b + 3.
5. A trapezoid ABCD is given with BC k AD. Assume that the bisectors of

the angles BAD and CDA intersect on the perpendicular bisector of the
line segment BC.
Prove that |AB| = |CD| or |AB| + |CD| = |AD|.
17
Solutions
1. Suppose that we colour the integer k red. We prove by induction to n that

(2n + 1)k then also is coloured red for all n > 0. For n = 0 this is trivial.
Suppose that (2n − 1)k is red for a certain n, then k + k + (2n − 1)k =
(2n + 1)k is also red. This concludes the induction. Analogously if k is
coloured green, then also (2n + 1)k is coloured green for all n > 0.
Without loss of generality we may assume that 1 is red. Then all odd
numbers are red. Now suppose that there is also an even number 2m that
is red. Because both 1 and 2m are red, we can easily show by induction
that 2m + 2n is red for all n > 0. Then only finitely many numbers
are left that could be green, namely the even numbers smaller than 2m.
However, if one of these numbers is green, then also all odd multiples of
that number are green and these are infinitely many numbers, yielding a
contradiction. If on the other hand none of these numbers is green, then
we have a contradiction with the first condition. We conclude that no even
number is red. Therefore all even numbers are green.
This colouring also satisfies all conditions: the sum of three odd numbers
is always odd and the sum of three even numbers is always even, which
shows that the second and third condition are satisfied. The first condition
is also clearly satisfied.
Hence the only possible colourings are: all even numbers are coloured green
and all odd numbers red, or all odd numbers green and all even numbers
red.
2. The conditions in the problem fix the configuration. Because ∠LAD =

∠EAB = ∠ECB by the inscribed angle theorem on chord EB of circle Γ:
∠BDK = ∠ADL = 180◦ − ∠DLA − ∠LAD = 90◦ − ∠LAD

= 90◦ − ∠ECB = ∠ECA − ∠ECB = ∠BCA = 180◦ − ∠BCK.
This yields that BDKC is a cyclic quadrilateral. Therefore ∠BCD =

∠BKD. Now we will prove that ∠BKD = 2∠BCE, from which the
problem follows.
By the inscribed angle theorem, ∠EBK = ∠BCE, hence ∠ABK =
∠ABE + ∠EBK = 90◦ + ∠BCE. Together with the sum of the angles of
triangle ABC this yields
∠AKB = 180◦ − (90◦ + ∠BCE) − ∠BAK = 90◦ − ∠BCE − ∠BAK.
18
The inscribed angle theorem gives ∠BAK = ∠EAK + ∠BAE = ∠EAK +
∠BCE, therefore we find
∠AKB = 90◦ − 2∠BCE − ∠EAK = 90◦ − 2∠BCE − ∠LAK.
From the sum of the angles of triangle AKL follows ∠AKL = 90◦ −∠LAK,
hence
∠AKB = ∠AKL − 2∠BCE.
This yields
∠BKD = ∠AKL − ∠AKB = 2∠BCE,
what we wanted to prove.
3. We rewrite the given equation as

1 2
2x − xy + 21 y 2 + 12 y 2 − yz + 12 z 2 + 12 z 2 − zx + 12 x2 + 1 = |x − 2y + z|,
or, equivalently,
1
2 (x − y)2 + 12 (y − z)2 + 12 (z − x)2 + 1 = |(x − y) + (z − y)|. (1)
Now substitute a = x − y and b = z − y. Then x − z = a − b, thus we get

1 2
2a + 12 b2 + 12 (a − b)2 + 1 = |a + b|. (2)
From (a − b)2 > 0, it follows that a2 − 2ab + b2 > 0, hence 2a2 + 2b2 >
2
a2 + b2 + 2ab, which means that a2 + b2 > (a+b)
2 with equality if and only
if a = b. Furthermore, also (a − b)2 > 0. Hence
(a + b)2
|a + b| = 12 a2 + 12 b2 + 12 (a − b)2 + 1 > + 1.
4
Now write c = |a + b|, then the expression becomes
c2
c> + 1.
4
We can rewrite this to c2 − 4c + 4 6 0, or, equivalently (c − 2)2 6 0.
Because the left hand side is a square, equality must hold, so c = 2.
Furthermore, in our previous inequalities equality also has to hold, so a = b.
Substituting this in (2) gives a2 + 1 = 2, so a = ±1. Thus we find the
triples (y + 1, y, y + 1) and (y − 1, y, y − 1) for arbitrary y ∈ R. Substituting
this in the equation (1) (that is equivalent to the original equation) shows
that these triples are indeed solutions for all y ∈ R. Hence, all solutions
are given by (y + 1, y, y + 1) and (y − 1, y, y − 1) with y ∈ R.
19
4. First we prove the statement for all odd numbers n. In this case modulo
4 we have 5n ≡ 1n = 1 and 3n ≡ (−1)n ≡ −1, hence 5n − 3n ≡ 2 mod 4.
Therefore, if n is odd, then a = 1. Because b > 1, the inequality a 6 b + 3
is satisfied.
Now suppose that n ≡ 2 mod 4. Write n = 2k with k an odd positive
integer. Notice that 52k − 32k = (5k − 3k )(5k + 3k ). We just showed that
5k − 3k has precisely one factor 2, since k is odd. Now consider 5k + 3k
modulo 16. For m = 1, 2, 3, 4 we have that 5m modulo 16 is congruent
to respectively 5, 9, 13, 1. Because 54 ≡ 1 mod 16, we have 5k ≡ 5 for all
k ≡ 1 mod 4 and 5k ≡ 13 for all k ≡ 3 mod 4. For m = 1, 2, 3, 4 we
have that 3m modulo 16 is congruent to respectively 3, 9, 11, 1. Because
34 ≡ 1 mod 16, we have 3k ≡ 3 for all k ≡ 1 mod 4 and 3k ≡ 11 for all
k ≡ 3 mod 4. Altogether 5k + 3k ≡ 5 + 3 ≡ 8 mod 16 if k ≡ 1 mod 4
and 5k + 3k ≡ 13 + 11 = 24 ≡ 8 mod 16 if k ≡ 3 mod 4. In both cases
5k + 3k contains precisely 3 factors 2.
We conclude that for n ≡ 2 mod 4 we have: a = 4. Because b > 1, the
inequality a 6 b + 3 is now satisfied.
We will prove by induction to m that a 6 b + 3 for all positive numbers n
with precisely m > 1 factors 2. The induction basis is m = 1, that is the
case n ≡ 2 mod 4, which we just solved.
Now let m > 1 and assume as induction hypothesis that we already showed
that a 6 b + 3 for all numbers n with precisely m factors 2. Now consider
a number n with m + 1 factors 2. We write n = 2k, where k has precisely
m factors 2. For clarity let a(k) and b(k) be the a and the b corresponding
to k, and a(n) and b(n) the a and b corresponding to n. The induction
hypothesis yields a(k) 6 b(k) + 3. Now we will prove that a(n) 6 b(n) + 3.
We have 5n − 3n = 52k − 32k = (5k − 3k )(5k + 3k ). Because k is even (it
contains m > 1 factors 2) modulo 4 we have 5k + 3k ≡ 1k + (−1)k ≡ 2
mod 4. Hence 5k +3k contains precisely one factor 2. Furthermore, 5k −3k
contains precisely a(k) factors 2. Thus a(n) = a(k) + 1. We already know
that 2b(k) 6 k and 2b(k)+1 > k, which yields 2b(k)+1 6 2k and 2b(k)+2 > 2k.
Hence b(n) = b(k) + 1. Now we conclude: a(n) = a(k) + 1 6 b(k) + 3 + 1 =
b(n) + 3, which concludes the induction.
This proves that a 6 b + 3 for all integers n > 2.
20
5. Let M be the midpoint of BC and let P be the intersection of the per-
pendicular bisector of BC and AD. Let K be the intersection of M P and
the two angle bisectors. Let L and N be the feet of the lines through K
perpendicular to sides AB and DC, respectively. Because AK and DK
are bisectors, |KL| = |KP | = |KN |. Furthermore K also lies on the per-
pendicular bisector of BC, thus |KB| = |KC|. Because triangles BLK
and CN K both have a right angle, they are congruent by (RHS).
Now we distinguish between four cases. First consider the case that L
and N lie on the interior of respectively sides AB and DC. Notice that
triangle KBC is isosceles, yielding ∠KBC = ∠BCK. Hence by 4BLK ∼ =
4CN K:
∠ABC = ∠LBK + ∠KBC = ∠N CK + ∠BCK = ∠BCD.
From this follows that ABCD is a isosceles trapezoid and hence |AB| =
|CD|.
In the case that L and N both lie on the exterior of sides AB and CD we
can show analogously that |AB| = |CD|.
Now consider the case that L lies on the interior of side AB, but N on the
exterior of side DC. Because AK and DK are bisectors, |AL| = |AP | and
|DN | = |DP |. Therefore by 4BLK ∼ = 4CN K:
|AB| + |CD| = (|AL| + |LB|) + (|DN | − |N C|)

= |AP | + |LB| + |DP | − |LB| = |AD|.
In the case that L lies on the exterior of AB and N on the interior of DC,
we analogously show that |AB| + |CD| = |AD|.
21
Benelux Mathematical Olympiad, May 2011
Mersch, Luxembourg
Problems
1. An ordered pair of integers (m, n) with 1 < m < n is said to be a Benelux

couple if the following two conditions hold: m has the same prime divisors
as n, and m + 1 has the same prime divisors as n + 1.
(a) Find three Benelux couples (m, n) with m 6 14.

(b) Prove that there exist infinitely many Benelux couples.
2. Let ABC be a triangle with incentre I. The angle bisectors AI, BI and CI
meet [BC], [CA] and [AB] at D, E and F , respectively. The perpendicular
bisector of [AD] intersects the lines BI and CI at M and N , respectively.
Show that A, I, M and N lie on a circle.
3. If k is an integer, let c(k) denote the largest cube that is less than or
equal to k. Find all positive integers p for which the following sequence is
bounded:
a0 = p and an+1 = 3an − 2c(an ) for n > 0.
(A sequence a0 , a1 , . . . of reals is said to be bounded if there exists an

M ∈ R such that, for all n > 0, |an | 6 M .)
4. Abby and Brian play the following game: They first choose a positive
integer N . Then they write numbers on a blackboard in turn. Abby starts
by writing a 1. Thereafter, when one of them has written the number n,
the other writes down either n + 1 or 2n, provided that the number is not
greater than N . The player who writes N on the blackboard wins.
(a) Determine which player has a winning strategy if N = 2011.
(b) Find the number of positive integers N 6 2011 for which Brian has a
winning strategy.
22
Solutions
1. (a) It is possible to see that (2, 8), (6, 48) and (14, 224) are Benelux cou-
ples.
(b) Let k > 2 be an integer and m = 2k − 2. Define n = m(m + 2) =
2k (2k − 2). Since m is even, m and n have the same prime factors.
Also, n + 1 = m(m + 2) + 1 = (m + 1)2 , so m + 1 and n + 1 have the
same prime factors as well. We have thus obtained a Benelux couple
2k − 2, 2k (2k − 2) for each k > 2.
2. The quadrilateral AM DB is cyclic. Indeed, M is the intersection of the

line BI, which bisects the angle ∠ABD in ABD and the perpendicular
bisector of [AD]. By uniqueness of this intersection point, it follows that M
lies on the circumcircle of ABD, and hence AM DB is cyclic. Analogously,
AN DC is cyclic.
Since AM DB and AN DC are cyclic, ∠AM I+∠AN I = ∠AM B+∠AN C =
◦
∠ADB + ∠ADC = 180 , because B and M , and C and N lie on either
side of AD. Hence AM IN is cyclic, for M and N lie on opposite sides of
AD.
3. Since c(an ) 6 an for all n ∈ N, an+1 > an , where equality holds if and only
if c(an ) = an . Hence the sequence is bounded if and only if it is eventually
constant, which is if and only if an is a perfect cube, for some n > 0. In
particular, the sequence is bounded if p is a perfect cube.
We now claim that, if an is not a cube for some n, then neither is an+1 .
Indeed, if an is not a cube, q 3 < an < (q + 1)3 for some q ∈ N, so that
c(an ) = q 3 . Suppose to the contrary that an+1 is a cube. Then
an+1 = 3an − 2c(an ) < 3(q + 1)3 − 2q 3 = q 3 + 9q 2 + 9q + 3

< q 3 + 9q 2 + 27q + 27 = (q + 3)3 .
Also, since c(an ) < an , an+1 > an > q 3 , so q 3 < an+1 < (q + 3)3 . It follows
that the only possible values of an+1 are (q + 1)3 and (q + 2)3 . However,
in both of these cases,
3an − 2q 3 = an+1 = (q + 1)3 ⇐⇒ 3an = 3(q 3 + q 2 + 1) + 1

3an − 2q 3 = an+1 = (q + 2)3 ⇐⇒ 3an = 3(q 3 + 2q 2 + 4q) + 8
a contradiction modulo 3. This proves that, if an is not a cube, then

neither is an+1 . Hence, if p is not a perfect cube, an is not a cube for any
23
n ∈ N, and the sequence is not bounded. We conclude that the sequence
is bounded if and only if p is a perfect cube.
4. (a) Abby has a winning strategy for odd N : Observe that, whenever
any player writes down an odd number, the other player has to write
down an even number. By adding 1 to that number, the first player
can write down another odd number. Since Abby starts the game by
writing down an odd number, she can force Brian to write down even
numbers only. Since N is odd, Abby will win the game. In particular,
Abby has a winning strategy if N = 2011.
(b) - Let N = 4k. If any player is forced to write down a number
m ∈ {k + 1, k + 2, ..., 2k}, the other player wins the game by
writing down 2m ∈ {2k + 2, 2k + 4, ..., 4k}, for the players will
have to write down the remaining numbers one after the other.
Since there is an even number of numbers remaining, the latter
player wins. This implies that the player who can write down k,
i.e. has a winning strategy for N = k, wins the game for N = 4k.
- Similarly, let N = 4k + 2. If any player is forced to write down a
number m ∈ {k + 1, k + 2, ..., 2k + 1}, the other player wins the
game by writing down 2m ∈ {2k + 2, 2k + 4, ..., 4k + 2}, as in the
previous case. Analogously, this implies that the player who has
a winning strategy for N = k wins the game for N = 4k + 2.
Since Abby wins the game for N = 1, 3, while Brian wins the game
for N = 2, Brian wins the game for N = 8, 10 as well, and thus for
N = 32, 34, 40, 42 too. Then Brian wins the game for a further 8
values of N between 128 and 170, and thence for a further 16 values
between 512 and 682, and for no other values with N 6 2011. Hence
Brian has a winning strategy for precisely 31 values of N with N 6
2011.
24
Problems
1. Find all pairs (x, y) of integers that satisfy

√
x2 + y 2 + 33 = 456 x − y.
2. We consider tilings of a rectangular m × n-board with 1 × 2-tiles. The tiles

can be placed either horizontally, or vertically, but they aren’t allowed to
overlap and to be placed partially outside of the board. All squares on the
board must be covered by a tile.
(a) Prove that for every tiling of a 4 × 2010-board with 1 × 2-tiles there is
a straight line cutting the board into two pieces such that every tile
completely lies within one of the pieces.
(b) Prove that there exists a tiling of a 5 × 2010-board with 1 × 2-tiles
such that there is no straight line cutting the board into two pieces
such that every tile completely lies within one of the pieces.
3. The circles Γ1 and Γ2 intersect at D and P . The common tangent line of

the two circles closest to point D touches Γ1 in A and Γ2 in B. The line
AD intersects Γ2 for the second time in C. Let M be the middle of line
segment BC.
Prove that ∠DP M = ∠BDC.
4. Determine all integers n for which the polynomial P (x) = 3x3 − nx − n − 2

can be written as the product of two non-constant polynomials with integer
coefficients.
5. Let ABC be a triangle with |AB| > |BC|. Let D be the midpoint of AC.
Let E be the intersection of the angular bisector of ∠ABC and the line
AC. Let F be the point on BE such that CF is perpendicular to BE.
Finally, let G be the intersection of CF and BD.
Prove that DF divides the line segment EG into two equal parts.
25
Solutions
1. Since the left hand side is an integer, the right hand side must also be an
integer. The square root of an integer is either an integer or irrational (but
√
never a rational number that is not an integer). Hence x − y must be an
integer. The right hand side is divisible by 3, so the left hand side must
also be divisible by 3. We deduce that 3 | x2 + y 2 . Since squares are always
congruent to 0 or 1 modulo 3, it follows that x2 ≡ y 2 ≡ 0 mod 3, hence
both x and y are divisible by 3. Writing and substituting x = 3a, y = 3b:
√
9a2 + 9b2 + 33 = 456 3a − 3b.
√
Square roots of integers are either integer or irrational. But since 3a − 3b =
2 2 3
9a +9b +3
456 is rational, it must be an integer as well. Hence 3a − 3b is a
square, and a multiple of 3, so it must also be a multiple of 9. Now we can
divide both sides by 9, which yields
r
2 2 a−b
a + b + 3 = 152 .
3
Writing a − b = 3c2 with c > 0, and substituting a = b + 3c2 gives
9c4 + 6c2 b + 2b2 + 3 = 152c.
Considering this as a quadratic equation in b and knowing that we are

looking for real solutions, we can deduce that the discrimant must be non-
negative. Hence 36c4 −8(9c4 −152c+3) > 0 and thus 36c4 +24 6 8·152c. If
c > 4 we have 36c4 +24 > 36·64c > 8·152c, this is a contradiction. Hence we
may conclude c 6 3. Furthermore, 152c is even, and so is 6c2 b + 2b2 , hence
9c4 + 3 must be even as well. We deduce that c is odd, and consequently
that the only possibilities are c = 1 and c = 3. For c = 3 the discriminant
is 36 · 34 − 8(9 · 34 − 152 · 3 + 3) ≡ 1 · 3(1 − 152) ≡ −1 · 3 · 151 ≡ 6 mod 9.
Hence it is not a square and the solutions of the quadratic equation will
not be integers. Substituting c = 1 gives
9 + 6b + 2b2 + 3 = 152,
or, equivalently,
b2 + 3b − 70 = 0.
This can also be written as (b − 7)(b + 10) = 0. Hence either b = 7 or
b = −10. In the first case, we have a = b + 3c2 = 10, so x = 30 and
y = 21. In the second case, we have a = b + 3c2 = −7, hence x = −21
and y = −30. Note that both pairs do satisfy the equation. Hence the
solutions are (x, y) = (30, 21) and (x, y) = (−21, −30).
26
2a. Let a dividing line be a straight line that divides the board in two parts in
such a way that every tile lies entirely in one of these parts. Suppose that
there exists a tiling without dividing lines. Consider the columns k and
k+1, with 1 6 k 6 2009. Then there is a tile lying horizontally in these two
columns; otherwise, the vertical line between these two columns would have
been a dividing one. There are 4k squares in the first k columns, an even
number. Since every tile lying entirely inside the first k columns, covers
exactly 2 squares, it follows that the number of tiles lying horizontally in
the columns k and k + 1 is even. We’ve shown earlier that this number is
at least 1, so it must be at least 2.
So for every k with 1 6 k 6 2009, there are two tiles lying horizontally
in columns k and k + 1. These tiles together cover 2 · 2009 · 2 squares.
Furthermore, for every i with 1 6 i 6 3, there must be a tile lying vertically
in columns i and i + 1. These tiles together cover 3 · 2 squares. So the total
number of squares covered by these tiles is (2 · 2009 + 3) · 2 > 2 · 2010 · 2.
But the board only contains 4 · 2010 squares. This is a contradiction.
2b. We use induction on n to show that for all n > 3, a

5 × 2n-board can be tiled without dividing line. For
n = 3, this can be done as indicated in the picture.
Suppose that we have a tiling of a 5×2n-board with-
out dividing lines. Then there are at least 4 tiles that
lie vertically. Hence there is a k with 1 6 k 6 2n − 1
such that there is a tile lying vertically in column k. Now add two columns,
called k1 and k2 , between columns k and k+1. Every tile lying horizontally
in columns k and k + 1, is replaced by two horizontal tiles, one in columns
k and k1 , one in columns k2 and k + 1. Since column k contains a vertical
tile, not all squares in columns k1 and k2 are covered yet. Also note that if
in a certain row, the square in k1 is not covered, then neither is the square
in k2 in the same row. Hence we can put a horizontal tile there. In that
way, all the squares of the new columns are covered.
Now it’s clear that no horizontal dividing lines are created in this process,
and no vertical ones between columns that are unchanged. Moreover, there
is at least one tile lying horizontally in columns k1 and k2 , so there is no
dividing line between these two columns. In the original board, there was
no dividing line between columns k and k + 1. That implies that there are
tiles lying horizontally in columns k and k1 , and in columns k2 and k + 1.
So there are no dividing lines between these pairs of columns. Hence the
board constructed contains no dividing lines at all.
This completes the induction. We conclude that there is a tiling of a
5 × 2010-board without dividing line.
27
3. Let S be the intersection of P D and AB. Then S lies on the radical axis
of the two circles, so |SA| = |SB|. So P S is a median in triangle P AB.
By the inscribed angle theorem on Γ1 with chord AP , we have ∠BAP =
180◦ − ∠ADP = ∠CDP = ∠CBP . By the inscribed angle theorem on
Γ2 with chord BP , we have ∠ABP = ∠BCP . Hence 4P AB ∼ 4P BC
(aa). Since P M is a median in triangle P BC, we have ∠SP B = ∠M P C.
It follows that
∠DP M = ∠DP B + ∠BP M = ∠SP B + ∠BP M = ∠M P C + ∠BP M

= ∠BP C = ∠BDC.
4. Suppose that P (x) can be written as P (x) = A(x)B(x) with A and B are
non-constant polynomials with integer coefficients. Since A and B are non-
constant, they both have degree at least 1. The sum of the degrees is equal
to the degree of P , hence equal to 3. This implies that the two degrees are 1
and 2. So we can write, without loss of generality, that A(x) = ax2 + bx + c
and B(x) = dx + e, where a, b, c, d and e are integers. The product of the
leading coefficients a and d is equal to the leading coefficient of P , hence
equal to 3. By multiplying both A and B if necessary, we may assume that
both a and d are positive, so that they are equal to 1 and 3 in some order.
First suppose that d = 1. Substituting x = −1 yields
P (−1) = 3 · (−1)3 + n − n − 2 = −5,
so
−5 = P (−1) = A(−1)B(−1) = A(−1) · (−1 + e).
Note that −1 + e is a divisor of −5, hence equal to −5, −1, 1 or 5, giving
four possible values for e, namely −4, 0, 2 or 6. Moreover, x = −e is a
zero of B, hence also of P .
If e = −4, then
0 = P (4) = 3 · 43 − 4n − n − 2 = 190 − 5n,
so n = 38. So we can indeed factor P (x):
3x3 − 38x − 40 = (3x2 + 12x + 10)(x − 4).
If e = 0, then
0 = P (0) = −n − 2,
so n = −2. So we can indeed factor P (x):
3x3 + 2x = (3x2 + 2)x.
28
If e = 2, then
0 = P (−2) = 3 · (−2)3 + 2n − n − 2 = −26 + n,
3x3 − 26x − 28 = (3x2 − 6x − 14)(x + 2).
If e = 6, then
0 = P (−6) = 3 · (−6)3 + 6n − n − 2 = −650 + 5n,
3x3 − 130x − 132 = (3x2 − 18x − 22)(x + 6).
Now suppose that d = 3. Then we have:
−5 = P (−1) = A(−1)B(−1) = A(−1) · (−3 + e).
Note that −3 + e is a divisor of −5, hence equal to −5, −1, 1 or 5. This

gives four possible values for e, namely −2, 2, 4 or 8. Furthermore, x = −e
3
is a zero of B, hence also of P . Note that e is never divisible by 3. We
have
e3
0 = P ( −e −e 3 e e−3
3 ) = 3 · ( 3 ) + 3 n − n − 2 = − 9 + 3 n − 2,
3 3
so e−3 e e
3 n = 9 + 2, hence (e − 3)n = 3 + 6. But this is a contradiction,
since the left hand side is an integer, whereas the right hand side is not,
since 3 does not divide e.
Finally, we deduce that the solutions are: n = 38, n = −2, n = 26 and
n = 130.
5. By |AB| > |BC|, the points D and E are distinct, and the order of the
points on the line AC is: A, D, E, C. Moreover, ∠BCA > ∠CAB, so
∠BCA + 12 ∠ABC > 90◦ , which implies that F lies on the interior of BE.
Consequently, G lies on the interior of BD.
Now let K be the intersection of CF and AB. Since BE is the angle
bisector of ∠ABC, we have ∠KBF = ∠F BC. Furthermore, ∠BF K =
90◦ = ∠CF B and |BF | = |BF |, hence by (ASA), we have 4KBF ∼ =
4CBF . This implies that |BK| = |BC| and |KF | = |CF |. In particular,
F is the midpoint of CK. Since D is the midpoint of AC, it follows that
DF is a midparallel in 4AKC, so DF k AK and |DF | = 12 |AK|.
29
From DF k AB, it follows that 4KGB ∼ 4F GD (aa), hence
1 1 1
|DG| |F D| |AK| 2 (|AB| − |KB|) 2 (|AB| − |BC|)
= = 2 = = . (3)
|BG| |KB| |KB| |KB| |BC|
|AE| |AB|
The angle bisector theorem now implies that |CE| = |CB| , or, equivalently
|AC|−|CE| |AB|
|CE| = |BC| .This yields |BC| · (|AC| − |CE|) = |AB| · |CE|, so
|BC| · |AC| = |CE| · (|AB| + |BC|), hence
|BC| · |AC|
|CE| = .
|AB| + |BC|
Continuing the calculation:

1
|DE| = |DC| − |CE| = |AC| − |CE|
2
1 1
|AC| · (|AB| + |BC|) − |AC| · |BC| 2 |AC| · (|AB| − |BC|)
= 2 = .
|AB| + |BC| |AB| + |BC|
Thus 1
2 |AC|·(|AB|−|BC|) 1
|DE| |AB|+|BC| 2 (|AB|− |BC|)
= |BC|·|AC|
= .
|CE| |BC|
|AB|+|BC|
Combined with (3) this gives
|DE| |DG|
= .
|CE| |BG|
In triangle DBC this implies that EG k BC. Now let S be the intersection
of DF and EG. Then we have
∠SGF = ∠EGF = ∠F CB = ∠F KB = ∠GF D = ∠GF S,
so 4SF G is isosceles, with |SF | = |SG|. Moreover,
∠SEF = ∠GEF = 90◦ − ∠EGF = 90◦ − ∠GF S = ∠SF E,
so 4SF E is also isosceles, with |SF | = |SE|. We deduce that |SG| = |SE|,
hence that S is the midpoint of EG.
30
Problems
1. Let n > 2 and k > 1 be positive integers. In a country there are n cities and
between each pair of cities there is a bus connection in both directions. Let
A and B be two different cities. Prove that the number of ways in which
you can travel from A to B by using exactly k buses is equal to
(n − 1)k − (−1)k
.
n
2. Find all functions f : R → R satisfying

xf (x + xy) = xf (x) + f (x2 )f (y)
for all x, y ∈ R.
3. Let Γ1 and Γ2 be two intersecting circles with midpoints respectively O1

and O2 , such that Γ2 intersects the line segment O1 O2 in a point A. The
intersection points of Γ1 and Γ2 are C and D. The line AD intersects Γ1
a second time in S. The line CS intersects O1 O2 in F . Let Γ3 be the
circumcircle of triangle ADF . Let E be the second intersection point of
Γ1 and Γ3 .
Prove that O1 E is tangent to Γ3 .
4. Prove that there exists no infinite sequence of prime numbers p0 , p1 , p2 , . . .

such that for all positive integers k:
pk = 2pk−1 + 1 or pk = 2pk−1 − 1.
5. Find all triples (a, b, c) of positive integers with a+b+c = 10 such that there
are a red, b blue and c green points (all different) in the plane satisfying
the following properties:
• for each red point and each blue point we consider the distance be-
tween these two points; the sum of these distances is 37;
• for each green point and each red point we consider the distance
between these two points; the sum of these distances is 30;
• for each blue point and each green point we consider the distance
between these two points; the sum of these distances is 1.
31
Solutions
1. Let α(k) be the number of ways to travel from city A to city B 6= A by

using k buses. Let β(k) be the number of ways to travel from city A to
city A by using k buses. If we start in city A and then take a bus k times,
this can be done in (n − 1)k ways. In β(k) of the cases we end at city A
and in (n − 1)α(k) of these cases we end at a city different from A. Hence
(n − 1)α(k) + β(k) = (n − 1)k . (4)
Now suppose that k > 2. To travel from city A to city A by using exactly
k buses, we take a bus from A to an arbitrary city (this can be done in
(n − 1) ways); then we must travel from a city different from A to city A
by using k − 1 buses, which can be done in α(k − 1) ways. Therefore
β(k) = (n − 1)α(k − 1) for k > 2. (5)
Now we substitute this expression for β(k) in (4). This yields for k > 2
that
(n − 1)α(k) + (n − 1)α(k − 1) = (n − 1)k
and hence
α(k) = (n − 1)k−1 − α(k − 1). (6)
Now we prove by induction to k that for n > 2 and k > 1:
(n − 1)k − (−1)k
α(k) = .
n
For k = 1 this is equivalent to α(1) = (n−1)+1

n = 1 and that is true, because
there is exactly one way to travel from city A to city B 6= A by using one
bus. Now let m > 1 be an integer and suppose that the expression for α(k)
has been proven for k = m. Then, using (6), for k = m + 1 > 2:
(n − 1)m − (−1)m
α(m + 1) = (n − 1)m − α(m) = (n − 1)m −
n
n(n − 1)m − (n − 1)m + (−1)m (n − 1)m+1 − (−1)m+1
= =
n n
and that is exactly the expression we wanted to prove for k = m + 1. This
completes the induction.
32
2. Substituting x = 0 and y = 0 yields 0 = f (0)2 , hence f (0) = 0. Sub-
stituting x = 1 and y = −1 yields f (0) = f (1) + f (1)f (−1), hence
0 = f (1) 1 + f (−1) , therefore f (1) = 0 or f (−1) = −1. Substituting
x = −1 yields
−f (−1 − y) = −f (−1) + f (1)f (y) for all y ∈ R. (7)
Suppose that f (1) = 0, then this is equivalent to −f (−1 − y) = −f (−1)
and because −1 − y takes all values in R, this means that f is constant.
Because f (0) = 0, it must be true that f (x) = 0 for all x. It is clear that
this function satisfies the original equation. We found our first solution:
f (x) = 0 for all x ∈ R.
Now suppose that f (1) 6= 0, then f (−1) = −1. Now substitute y = −1
in (7), then we get −f (0) = −f (−1) + f (1)f (−1), hence 0 = 1 − f (1),
therefore f (1) = 1. Substituting x = 1 in the original equation yields
f (1 + y) = 1 + f (y) for all y ∈ R. (8)
Furthermore substituting y = −1 in the original equation yields xf (0) =
xf (x) − f (x2 ), hence
xf (x) = f (x2 ) for all x ∈ R. (9)
The original equation now can be written as
xf (x + xy) = xf (x) + xf (x)f (y) = xf (x)(1 + f (y)).
If x 6= 0, we may divide the left and right hand side by x and this yields
together with (8):
f (x + xy) = f (x)f (1 + y) for x 6= 0.
Notice that this is also true for x = 0. Now define z = 1 + y. Because this
takes all values in R, we get
f (xz) = f (x)f (z) for all x, z ∈ R.
Applying this to (9), we find
xf (x) = f (x2 ) = f (x)f (x),
hence for all x we have f (x) = 0 or f (x) = x. Now suppose there is a
x 6= 0 with f (x) = 0, then
1 = f (1) = f (x · x1 ) = f (x)f ( x1 ) = 0,
which is a contradiction. Hence for all x 6= 0 we have f (x) = x. This is
also true for x = 0. Substituting it in the original equation shows that
this is indeed a solution. The two functions satisfying the condition are
f (x) = 0 for all x ∈ R and f (x) = x for all x ∈ R.
33
3. The intersection point of O1 O2 and the arch SD of Γ1 containing C, is
named T . Because A lies in the interior of Γ1 , we now know
∠O1 AS = ∠AT S + ∠T SA exterior angle theorem in 4AT S

= ∠O1 T S + ∠T SD
= ∠T SO1 + ∠T SD 4O1 ST is isosceles (|O1 S| = |O1 T |).
The line O1 O2 is perpendicular to CD and splits the line segment CD into

two equal parts, hence it is the perpendicular bisector of CD. Therefore
T lies on the perpendicular bisector of CD, which yields that arches T C
and T D are equally long. Hence by the inscribed angle theorem ∠T SD =
∠CST . Thus
∠O1 AS = ∠T SO1 + ∠T SD = ∠T SO1 + ∠CST = ∠CSO1 = ∠F SO1 .
This means that 4O1 AS ∼ 4O1 SF (hh). This yields
|O1 A| |O1 S|
= ,
|O1 S| |O1 F |
hence |O1 A| · |O1 F | = |O1 S|2 = |O1 E|2 . Because A and F lie on the
same side of O1 , it is even true that O1 A · O1 F = O1 E 2 . Using the power
theorem we now see that O1 E is tangent to the circumcircle of 4AF E and
that is Γ3 .
4. Suppose that such an infinite sequence exists. By eventually leaving out

the first two elements, we can make sure that the first prime number of
the sequence is at least 5. Now we assume without loss of generality that
p0 > 5. Then we know that p0 6≡ 0 mod 3.
Suppose that p0 ≡ 1 mod 3. We prove by induction to k that for all
positive integers k: pk ≡ 1 mod 3 and pk = 2pk−1 − 1. Namely, suppose
that k > 1 and pk−1 ≡ 1 mod 3. Then 2pk−1 ≡ 2 mod 3, hence 2pk−1 + 1
is divisible by 3. Because pk is a prime number greater than 3, we have
pk = 2pk−1 − 1 and pk ≡ 2 − 1 ≡ 1 mod 3. This completes the induction.
In the case p0 ≡ 2 mod 3, we can prove analogously that pk ≡ 2 mod 3
and pk = 2pk−1 + 1 for all positive integers k.
Now we can compose a direct formula for the sequence. If p0 ≡ 1 mod 3,
then we get pk = (p0 − 1)2k + 1 for all k > 0. If p0 ≡ 2 mod 3, then we
get pk = (p0 + 1)2k − 1. We can prove these formulas by induction again.
By Fermat’s little theorem 2p0 −1 ≡ 1 mod p0 , hence also (−p0 +1)2p0 −1 ≡
1 mod p0 and (p0 +1)2p0 −1 ≡ 1 mod p0 . This yields p0 | (p0 −1)2p0 −1 +1
34
and p0 | (p0 + 1)2p0 −1 − 1. We see that p0 always is a divisor of pp0 −1 .
Because it is also clear that pp0 −1 is greater than p0 , this yields that pp0 −1
is not a prime number. This is a contradiction.
5. We construct triangles consisting of a blue, a red and a green point. These

triangles may also be degenerated. In each of these triangles the (non-
strict) triangle inequality holds: the distance between the blue and the
red point is at most the sum of the distances between the blue and the
green and between the red and the green point. We add all these triangle
inequalities (one for each triangle we can form with a blue, a red and a
green point). We now count each distance between a red and a blue point c
times (because with a fixed blue and red point you may choose c points as
third green point), each distance between a green and a red point b times
and each distance between a blue and a green point a times. Thus, we get
37c 6 30b + a.
Because a + b + c = 10, this yields 37c 6 30b + (10 − b − c) = 10 + 29b − c,

hence 38c 6 10 + 29b. Otherwise stated
38c − 10
6 b. (10)
29
We can also apply the triangle inequality differently: the distance between
the green and the red point is at most the sum of the distances between
the red and the blue and between the blue and the green point. If we add
these inequalities, we get
30b 6 37c + a.
This yields 30b 6 37c + (10 − b − c) = 10 + 36c − b, so 31b 6 10 + 36c,
otherwise stated
10 + 36c
b6 . (11)
31
Combining (10) and (11) yields
38c − 10 10 + 36c
6 ,
29 31
hence 31(38c − 10) 6 29(10 + 36c). Expanding yields 134c 6 600, therefore
c 6 4. Now we consider one by one all possibilities for c.
46
Suppose c = 1. Then (11) yields b 6 31 < 2, which yields b = 1. Now we
get (a, b, c) = (8, 1, 1).
66 82
Suppose c = 2. Then (10) and (11) yield 2 < 29 6b6 31 < 3, hence there
is no integer b that satisfies the conditions.
35
Suppose c = 3. Then (10) and (11) yield 3 < 104 29 6 b 6
118
31 < 4, hence
there is no integer b that satisfies the conditions.
Suppose c = 4. Then (10) and (11) yield 4 < 142 29 6 b 6
154
31 < 5, hence
there is no integer b that satisfied the conditions.
The only triple that possibly could satisfy the conditions is (8, 1, 1). Now
we show that there are indeed 8 red points, 1 blue point and 1 green point in
the plane that satisfy all conditions. We use a standard coordinate system.
Choose for the blue point (0, 0) and for the green point (1, 0). Choose red
points (i, 0) with 2 6 i 6 8. Also choose a red point such that that point
together with the green and blue point forms an isosceles triangle with two
sides of length 2. The sum of the distances between the red points and
the blue point now is 2 + 2 + 3 + · · · + 8 = 37. The sum of the distances
between the red points and the green point is 2 + 1 + 2 + · · · + 7 = 30. The
distance between the blue point and the green point is 1.
We conclude that the only possible solution is: (8, 1, 1).
36
Problems
Part 1
1. The letters A, B, C and D represent digits. The following holds:
A B A B
C A + and C A −
D A A
What is D?
A) 5 B) 6 C) 7 D) 8 E) 9
2. Peter is constructing a sequence of seven integers (they can be negative or

zero as well), such that the sum of four consecutive integers is always 1.
He wants his sequence to contain as many integers as possible, which are
greater than 13. What is the maximal number of such integers?
A) 0 B) 2 C) 3 D) 5 E) 6
3. An ant is walking on the surface of a cuboid with edges of length 3, 4

and 5. It starts walking on a vertex, and it wants to visit all the other
seven vertices. It doesn’t need to return to its starting vertex. What is
the length of the shortest possible route that accomplishes this?
A) 24 B) 25 C) 26 D) 27 E) 28
1
4. The integers 1 up to 5 are placed to form a circle.
We add every pair of neighbours together. The five ? ?
sums that we obtain, turn out to be consecutive in-
tegers. What are the neighbours of 1?
A) 2 and 4 B) 2 and 5 C) 3 and 4 D) 3 and 5 E) 4 and 5
5. Bert and Ernie both have 64 candies. Every day, one of them gives half of
his candies to the other. After six days, Bert has 61 candies, so Ernie has
67 candies. How many of the six days has Ernie shared his candies?
A) 1 B) 2 C) 3 D) 4 E) 5
37
6. If you add 36 to 37, you obtain 73. When its digits are written in reverse
order, this becomes 37 again. How many two-digit integers are there with
the property that if you add 36 to it, then write its digits in reverse order,
you obtain the integer you started with?
A) 4 B) 5 C) 6 D) 9 E) 10
7. One hundred students participate in a mathematical olympiad. Problem

1 was solved by 90 participants. Problem 2 was solved by 80 participants
and problem 3 has been solved by 75 participants. What is the minimal
number of participants that solved all of the problems?
A) 35 B) 45 C) 54 D) 55 E) 60
8. From 125 small cubes, a 5 × 5 × 5-cube is made. In every direction, the

cubes are coloured white and black alternatingly; the cubes on the vertices
are black. We only consider the cubes that are visible from the outside.
What is the difference between the number of black cubes and the number
of white ones?
A) 2 more black cubes B) 1 more black cube C) no difference
D) 1 more white cube E) 2 more white cubes
9. We start by drawing an equilateral triangle, and then

its circumcircle. Around this circle, we draw a per-
fectly fitting square. We then draw its circumcircle,
and draw a perfectly fitting pentagon around it, and
so on, up to a regular 16-gon. In the figure, you can
see that the area inside of the pentagon is divided in
17 pieces. In how many pieces is the area inside the
regular 16-gon divided?
A) 134 B) 136 C) 248 D) 264 E) 267
10. How many integers between 1 and 1000 do not contain the digit 1?
A) 700 B) 728 C) 729 D) 880 E) 900
38
11. A square with edges of length 2010 is divided into
nine rectangles by four lines which are parallel to the
edges. You can see an example in the figure. You can
divide this square in such a way that the perimeter of
the rectangles obtained are nine consecutive integers.
What is the perimeter of the largest rectangle?
A) 671 B) 1340 C) 1790 D) 2684 E) 3577
12. A farmer has a stack of hay to feed his horse, cow and goat. With it,
he can feed his cow and horse for 12 months, or his cow and goat for 15
months, or his horse and goat for 20 months. For how many months can
he feed all three of his animals?
A) 7 56 B) 9 C) 10 D) 15 23 E) 47
13. There are twenty-five towers: five of height 1, five of 2

height 2, five of height 3, five of height 4 and five of 1
height 5. The towers have to be places on a 5 × 5- 3
board with on each square a tower, such that in every
row and every column, every height occurs exactly 2
once. Moreover, in the direction of each of the ar- 4
rows in the figure, you have to be able to see exactly
the number of towers mentioned at that arrow. You 3 5 ? 2 1
can’t see a tower if it’s behind a higher tower. What number should be
filled in the place of the question mark?
A) 1 B) 2 C) 3 D) 4 E) 5
14. Tania has played 10 basketball matches this season. In the sixth up to the
ninth match, she scored 23, 14, 11 and 20 points, respectively. As a result,
her average score per match became higher after nine matches, than it was
after five matches. After ten matches, she has an average score of more
than 18 points. What is the minimal number of points that Tania scored
in the tenth match?
A) 19 B) 27 C) 28 D) 29 E) 31
39
15. Lucas’ sequence starts with the integers 1 and 3. After that, the next
integer in the sequence is found by adding the two preceding integers. So
you obtain 1, 3, 4, 7, 11, . . .. What is the last digit of the hundredth
integer in Lucas’ sequence?
A) 1 B) 3 C) 4 D) 7 E) 8
Part 2
1. The length of the side of the small square is 5, and

the length of the side of the large square is 10. What
is the total area of all the black pieces?
2. The following sum is incorrect:
7 4 2 5 8 6
8 2 9 4 3 0 +
1 2 1 2 0 1 6
This sum can made correct by picking two distinct digits and replace every
occurrence of one of them with the other one. Which two digits should
you pick?
3. A rhombus is cut from a pattern of grey and white

squares. What fraction of the rhombus is grey?
4. Consider all four-digit integer in which each of the digits 1, 2, 3 and 4

occur exactly once. What’s the average of all these integers?
5. Three equilateral triangles with edges of length 21

are placed on a white table in such a way that we
obtain an equilateral triangle with edges of length
36. In the centre, a small triangular piece of table is
visible. How long are the edges of that piece?
40
6. A rectangle is divided into eight squares. The area
of the grey square is 1. What is the area of the
rectangle?
7. Anton and Ben are competing in a race. Ben runs three times as fast
as Anton, but Anton gets a 30 meter lead. The arrive simultaneously at
the finish line. What is the distance in meters that Ben has run when he
crosses the finish line?
8. A square is divided into seven pieces by its diago- 160

nal and by two lines parallel to the edges. Of three
pieces, the area is given. What is the area of the 50
square? 90
9. Anne answered 33 questions, most of them well and the rest even very well.
Both grades are rewarded with a fixed number of points per questions,
very well answered questions being rewarded with more points than well
answered questions. Both number of points are integers 1, 2, 3, . . ., or
10, but we don’t know exactly which two integers they are. When Anne
calculated her average number of points per question, it turned out to be
an integer. How many questions did Anne answer very well?
10. On a circle, there are eleven points, numbered from 1 9

1 up to 11. From each point except point 11, a 5
6
line segment is drawn between that point, and the
point with the next number. Another line segment 10
is drawn between point 11 and point 1. What is the 2
maximal number of intersections of these 11 line seg- 4
ments? In the figure, there’s an example with only 7 8
16 intersections. 3 11
41
Solutions
Part 1
1. E) 9 6. B) 5 11. D) 2684
2. E) 6 7. B) 45 12. C) 10
3. B) 25 8. A) 2 more black 13. B) 2
4. C) 3 and 4 9. C) 248 14. D) 29
5. D) 4 10. B) 728 15. D) 7
Part 2
595
1. 25 6. 8 = 74 83
2. 2 and 6 7. 45
2
3. 3 8. 484
4. 2777 12 9. 11
5. 9 10. 44
42
and the team selection for IMO 2012 Argentina
First Round, February 2011
BxMO/EGMO Team Selection Test, March 2012 C
IMO Team Selection Test 1, June 2012 CM
MY

CY
We eat problems
CMY

for breakfast.

in Nederland thgehouden: IMO2011
Olympiad 2011
International
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
4 First Round, February 2011
10 Second Round, March 2011
19 BxMO/EGMO Team Selection Test, March 2012

Introduction
In 2011 the Dutch Mathematical Olympiad celebrated its 50th anniversary.
The most special fact in this anniversary year was that we hosted the
International Mathematical Olympiad (IMO) for the first time. From 16
until 24 July, 564 contestants from 101 countries came to Amsterdam to
test their mathematical talents and to enjoy the social events. More than
300 guides, invigilators, coordinators and many other organisers worked
together to make the event a big success.
In the meantime the entire selection process for IMO 2012 started with the
first round on 4 February 2011, held at the participating schools. The paper
consisted of eight multiple choice questions and four open-answer questions,
to be solved within 2 hours. In total 5258 students of 245 secondary schools
participated in this first round.
Those 799 students from grade 5 (4, 6 3) that scored 15 (13, 11) points or
more on the first round (out of a maximum of 36 points) were invited to the
second round, which was held in March at ten universities in the country.
This round contained five open-answer questions, and two problems for
which the students had to give extensive solutions and proofs. The contest
lasted 2.5 hours.
Those students from grade 5 (4, 6 3) that scored 32 (28, 22) points or
more on the second round (out of a maximum of 40 points) were invited to
the final round. Also some outstanding participants in the Kangaroo math
contest or the Pythagoras Olympiad were invited. In total 149 students
were invited. They also received an invitation to some training sessions at
the ten universities, in order to prepare them for their participation in the
final round.
concluded its 50th edition 2011.

1
Among the participants of the training programme, there were some extra
girls, as this year we would participate in the first European Girls’ Mathe-
matical Olympiad (EGMO). In total there were eight girls competing to be
in the EGMO team. The team of four girls was selected by a selection test,
held on 16 March 2012. They attended the EGMO in Cambridge from 10
until 16 April, and the team returned with two honourable mentions and
a bronze medal.
The same selection test was used to determine the ten students partici-
pating in the Benelux Mathematical Olympiad (BxMO), held in Namur,
Belgium, from 20 until 22 April. The Dutch team managed to come first
in the country ranking, and received two honourable mentions, two bronze
medals, three silver medals and two gold medals.
In June the team for the International Mathematical Olympiad 2012 was
selected by two team selection tests on 6 and 9 June 2012. A seventh,
young, promising student was selected to accompany the team to the IMO
as an observer C. The team had a training camp in Buenos Aires, from 29
June until 8 July, together with the team from New Zealand.
For younger students the Junior Mathematical Olympiad was held in Oc-
tober 2011 at the VU University Amsterdam. The students invited to
participate in this event were the 30 best students of grade 1, grade 2 and
grade 3 of the popular Kangaroo math contest. The competition consisted
of two one-hour parts, one with fifteen multiple choice questions and one
with ten open-answer questions. The goal of this Junior Mathematical
the solutions.
2
Dutch delegation
The Dutch team for IMO 2012 in Argentina consists of
• Guus Berkelmans (18 y.o., bronze medal at IMO 2010)

• Jeroen Huijben (16 y.o., observer C at IMO 2010, bronze medal at
IMO 2011)
• Matthijs Lip (16 y.o.)
• Michelle Sweering (15 y.o.)
• Jeroen Winkel (15 y.o., observer C at IMO 2011)
• Jetze Zoethout (17 y.o., bronze medal at IMO 2011)
• Peter Gerlagh (15 y.o.)

• Birgit van Dalen (deputy leader), Aloysius College The Hague
3
First Round, February 2011
Problems
A-problems
A1. The squares of a 4×4-field are colored black or white. 2

The number next to each row and below each column
indicates how many squares in that row or column 0
have to be black. 1
In how many ways can the field be colored?
1
A) 0 B) 1 C) 4 D) 5 E) 8
2 0 1 1
A2. Today is 4 February 2011. This date is written down as 04-02-2011. In

this problem we consider the first day from now on, of which the date is
written using eight different digits.
What is the month of that date?
A) January B) March C) June
D) October E) December
B A
A3. A heptagon ABCDEF G is given, all sides of which
have length 2. Moreover: ∠E = 120◦ , ∠C = C G
∠G = 90◦ and ∠A = ∠B = ∠D = ∠F .
What is the area of the heptagon?
√ √ D F
A) 10 + 2 √2 B) 8 + 3√ 3 C) 14
D) 10 + 2 6 E) 8 + 3 6 E
A4. Alice, Brian and Carl have participated in a math contest consisting of 12
problems. Before the contest they were pessimistic and made the following
statements.
Alice: “Brian will answer at least two more problems correctly than I will.”
Brian: “I will not answer more than five problems correctly.”
Carl: “I will at the most answer correctly as many problems as Alice.”
Their teacher tried to encourage them by saying: “Together, you will an-
swer more than 18 problems correctly.” Afterwards, it transpired that all
three students and their teacher had made a wrong prediction.
4
Who has/have answered the least number of problems correctly?
A) only Alice D) both Alice and Brian
B) only Brian E) you cannot be sure of that
C) only Carl
A5. Jack wants to write down some of the numbers from 1 to 100 on a piece of
paper. He wants to do it in such a way, that no two numbers on the piece
of paper will add up to 125.
How many numbers, at most, can Jack write down on his piece op paper?
A) 50 B) 61 C) 62 D) 63 E) 64
A6. The number a = 11 . . . 111 consists of 2011 digits ‘1’.

What is the remainder of a when divided by 37?
A) 0 B) 1 C) 3 D) 7 E) 11
A7. Ann and Bob are sitting in a fairground attraction.

They move in circles around the same center and in
the same direction. Ann moves around once every
Ann
20 seconds, Bob once every 28 seconds. At a certain Bob
moment they are at minimum distance from each
other (see figure).
How many seconds does it take, from that moment
on, until Ann and Bob are at maximum distance from each other?
A) 22.5 B) 35 C) 40 D) 49 E) 70
B
C
A8. The vertices of a regular 15-gon are connected as in A
the figure. (Mind you: the sizes in the figure are not D
entirely correct!)
What is the size of the angle, indicated by the arc,
between AC and BD?
A) 130◦ B) 132◦ C) 135◦ D) 136◦ E) 137.5◦
B-problems
1 1
B1. A number x satisfies: x = . Determine x − . Simplify your answer
1+x x
as much as possible.
5
B2. An escalator goes up from the first to the second floor of a department
store. Dion, while going up the escalator, also walks at a constant pace.
Raymond, going in the opposite direction, tries to walk downwards, from
the second to the first floor, on the same escalator. He walks at the same
pace as Dion. They both take one step of the escalator at a time. Dion
arrives at the second floor after exactly 12 steps; Raymond arrives at the
first floor after exactly 60 steps.
How many steps would it take Dion to get upstairs if the escalator would
stand still?
B3. Six scouts are on a scouting expedition. They are going to the woods on
Saturday, and into the mountains on Sunday. On both days, they have
to go in pairs. The scoutmaster wants to group them into pairs for both
expeditions in such a way that nobody has the same partner on the second
day as on the first day.
In how many ways can he do that?
C
B4. In the figure you see a pointed arch ABC and its
inscribed circle. The pointed arch consists of a line
segment AB of length 1, a circular arc BC with cen-
ter A and a circular arc AC with center B.
What is the radius of the inscribed circle of this
pointed arch?
A B
Solutions
A-problems
2
A1. D) 5 0
Notice that all squares in the second
1
row and second column must be white. We consider 1
two cases, depending on the color of the upper left 2 0 1 1
square. If this square is white, then the last two
squares in the first row and column must be black.
This determines the coloring. See the top figure.
If this square is black, both the first row and first
column require exactly one more black square. For
each of the resulting 2×2 = 4 choices, there is exactly
one solution. Indeed, one row and one column are
left that need an additional black square.
6
Therefore the square at the intersection of this row
and column must be colored black, and the remain-
ing squares must be colored white, see bottom four figures.
A2. C) June The year of the date we are looking for, starts with a digit
2 or higher. We will look for the first date of which the year starts with
digit 2, and all eight digits are different. If such a date exists, we are done.
For the month, both 11 (two equal digits) and 12 (digit 2 is already used)
can be rejected. Therefore, the month (01 to 10) contains digit 0. This
implies that the day starts with digit 1 or 3. In the second case, it’s the
31st, since digit 0 is already taken. In both cases, the day contains digit
1. Both digit 0 and 1 being taken, the smallest possible year is 2345. The
smallest number we can use for the month is then 06, that is, June. Finally,
the day will be the 17th. Observe that the constructed date 17-06-2345
consists of eight different digits, as required.
√
A3. B) 8 + 3 3 The heptagon can be partitioned into B A
two squares and three equilateral triangles, all with

C G
sides of length 2. We know that each of the squares
M
has an area of 4. Using the Pythagorean theorem,
we can compute
√ √ the height of triangle ABM to be D F
22 − √ 1 = √ 3. Hence, the area of the triangle equals
1
· 2 · 3 = 3. Summing up the areas of the squares E
2 √ √
and triangles, we arrive at 2 · 4 + 3 · 3 = 8 + 3 3
for the area of the heptagon.
A4. A) only Alice As Brian’s prediction was wrong, he has at least six correct
answers. Alice’s prediction was wrong as well, which implies that Brian
answered at the most one more question correctly than Alice. Hence, Alice
has at least five correct answers. Since Carl made a wrong prediction, he
answered more questions correctly than Alice, hence at least six. Alice
cannot have more than five correct answers. Indeed, then Carl would have
seven correct answers, leading to a total of at least 6 + 6 + 7 = 19 correct
answers, as the teacher (incorrectly) predicted. We can conclude that Alice
answered five questions correctly. Since the others have at least six correct
answers, Alice is the only one with the smallest number of correct answers.
7
A5. C) 62 Jack can certainly write down 62 numbers, for example: the
numbers from 1 to 62 (since 62 + 61 < 125). More than 62 numbers will
not be possible. Indeed, the numbers from 25 to 100 can be partitioned
into pairs of sum 125: 25 + 100 = 125, 26 + 99 = 125, and so on up to
62 + 63 = 125. Jack must skip at least one number from each of the 38
pairs. In total, therefore, he can write down no more than 100 − 38 = 62
of the numbers.
A6. B) 1 Using long division to divide a = 11 · · · 11 (2011 digits) by

37, you will quickly notice that 111 is divisible by 37. This is the fact
that we will be using. It implies that the number 1110 · · · 0 is divisible by
37, regardless of the number of trailing zeros. In particular, the following
numbers are divisible by 37: 1110 · · · 0 (2008 zeros), 1110 · · · 0 (2005 zeros),
1110 · · · 0 (2002 zeros), and so on up to 1110 (1 zero). The sum of these
numbers equals 1 . . . 10 (2010 digits 1), which is again divisible by 37. In
conclusion, the remainder of a when divided by 37 is 1, since a − 1 is
divisible by 37.
A7. C) 35 After 140 seconds, Ann has made 7 rounds and Bob only 5.
At that moment, Ann leads by two full rounds. Hence after only 1404 = 35
seconds, Anne leads by half a round. That is exactly the first moment she
and Bob are at maximal distance from each other.
A8. B) 132◦ Denote the center of the 15- B C
gon by M and the intersection of AC and S
BD by S (see figure). In quadrilateral

M RST , we see that ∠M RS = 90◦ and A
R T
D
∠ST M = 90◦ . Furthermore, we see that
∠T M R = 15 2
· 360◦ = 48◦ . As the an-
M
gles of a quadrilateral sum to 360◦ , we find:
∠RST = 360◦ − 2 · 90◦ − 48◦ = 132◦ . Ob-
serve that ∠RST and ∠BSC are opposite
angles. Hence the sought-after angle also
equals 132◦ .
8
B-problems
1
B1. −1 We are given that x = 1+x . Clearly, x 6= 0, since 0 6= 11 .
Hence, on both sides of the equation, we may flip the numerator and
denominator of the fraction. This results in: x1 = 1 + x. Combining both
formulas, we obtain x − x1 = x − (1 + x) = −1.
B2. 20 Consider three escalators in a row: the first one going up,
the second standing still, and the third going down. If Dion walks up
the first escalator, he arrives at the top after exactly 12 steps. Raymond,
walking up the third escalator, takes 60 steps to reach the top and reaches
only 15 of the escalator after 12 steps. A third person, say Julian, takes
the second escalator and walks at the same pace as Dion and Raymond.
After 12 steps, he will be positioned exactly in between Dion and Julian,
at ( 55 + 51 )/2 = 35 of the escalator. Therefore, he will need 53 · 12 = 20 steps
to reach the top.
B3. 120 Let’s call the eldest scout A. There are 5 possibilities for
finding him a partner B for the first day. Then, there are 4 possible
partners C for B on the second day, because he cannot be paired with A
twice. Now for C, there are 3 possible partners D on the first day, since he
cannot go with B again, and A is already paired. For D, there are now 2
possible partners E on the second day, since B and C are already paired,
and A cannot be his partner because that would leave two scouts that are
forced to form a pair on both days. Finally, there is one scout left. He
has no choice but to team op with E on the first day, and with A on the
second. In total there are 5 × 4 × 3 × 2 = 120 possibilities.
3
B4. 8 Consider the inscribed circle. We de-
C
note its center by O and its radius by r. The points
where the circle is tangent to AB and BC are de- R
noted by M and R, respectively. O
Since A, O and R are on a line, we have: |AO| =
|AR| − |OR| = 1 − r. We also know that |OM | = r
and |AM | = 12 |AB| = 21 . Using the Pythagorean
A M B
theorem, we find |AM |2 +|OM |2 = |AO|2 , and hence
1 2 2 2
4 + r = (1 − r) = r − 2r + 1. This implies that
2r = 4 and therefore r = 38 .
3
9
Problems
B-problems
B1. At a gala, a number of pairs (consisting of one man and one woman) are
dancing, in such a way that 23 of the women present is dancing with 35 of
the men present.
What part of those present at the gala is dancing?
B2. A square with edges of length 2 is inside of a square

with edges of length 7. The edges of the smaller
square are parallel to the ones of the larger one.
What is the area of the black-coloured part?
B3. In a classroom, there are 23 students who chose to learn a foreign language,
namely German and French. Of those 23 students, 10 are girls, and 11
of those 23 students have chosen French as their foreign language. The
number of girls that have chosen French, plus the number of boys that
have chosen German, is equal to 16.
What is the number of girls that have chosen French?
B4. We have a deck of 10.000 cards, numbered from 1 to 10.000. A step consists
of removing all the cards that has a square on it, and then renumbering
the remaining cards, starting from 1, in a successive way.
What is the number of steps needed to remove all but one card?
B5. We put a grey ribbon over a cylindrical white pole,

under an angle of 45 degrees. The ribbon is then
45 o
wound tightly around the pole (without creases). In
this way, we get a grey spiral around the pole. Be-
tween the grey spiral, a white spiral runs around the
pole; that is the part of the pole that is not covered
by the ribbon. The radius of the pole is 2 cm.
It turns out that the white and grey spirals have the same widths. What
is the width of the ribbon?
10
C-problems
For the C-problems not only the answer is important; you also have to describe the way you
solved the problem.
C1. Determine all triples (a, b, c) of positive integers with the following prop-
erties:
• we have a < b < c, and a, b and c are three successive odd integers;
• the number a2 + b2 + c2 consists of four equal digits.
C2. Thirty students participate in a mathematical competition with sixteen

questions. They have to answer each question with a number. If a student
answers a question correctly within a minute, he gets 10 points for that
question. If a student answers a question correctly, but not within one
minute, then he gets 5 points for that question. And if a student answers
a question incorrectly, he gets no points at all for that question.
After the competition has ended, it turns out that from all the 480 answers
that were given, more than half was correct and given within a minute.
The number of answers that were correct, but not given within a minute,
turns out to be equal to the number of incorrect answers.
Show that there are two students with the same total score.
Solutions
B-problems
12
B1. 19 Let w be the number of women present, and let m be the
number of men present. The problem tells us that 32 w = 35 m and hence
9
that w = 10 m. The number of people dancing, is exactly twice the number
of men dancing, namely 65 m.
9
The number of people present is of course m + w = m + 10 m = 1910 m. So
it follows that the part of those present that is dancing, is equal to
6
5m 6 10 12
19 = 5 · 19 = 19 .
10 m
11
B2. 10 We split the black part into four tri-
angles, of which we coloured two of them gray. The
two gray triangles both have base 2, and their com-
bined height is 7 − 2 = 5, namely the height of the
larger square minus the height of the smaller square.
Hence the area of the two grey triangles together is
equal to 21 · 2 · 5 = 5. The same holds for the two
black triangles. It follows that the combined area is 5 + 5 = 10.
B3. 7 There are 23 students in total. From what’s given, it follows

that:
16 + 11 + 10 = (girls with French + boys with German)

+ everyone with French + all girls
= (girls with French + boys with German)
+ (girls with French + boys with French)
+ (girls with French + girls with German)
= 3 × girls with French + boys with German
+ boys with French + girls with German
= 2 × girls with French + 23.
16+11+10−23
So the total number of girls that have chosen French is equal to 2 =
14
2 = 7.
B4. 198 On the first step, we remove the cards numbered by 12 , 22 , 32 ,

. . . , 1002 . Then 9900 cards remain. Since 992 6 9900 < 1002 , we remove
12 , 22 , . . . , 992 in the second step. After that, 9900 − 99 = 9801 = 992
cards are left, which is a square.
In general, if we start with n2 cards, with n > 2, we remove n cards in the
first step, after which n2 − n cards remain. Since (n − 1)2 = n2 − 2n + 1 6
n2 − n < n2 , we remove n − 1 cards in the second step. Then exactly
(n2 − n) − (n − 1) = (n − 1)2 are left. So in two steps we can reduce the
number of cards from n2 to (n − 1)2 . It follows that we need 2 · 99 = 198
steps to remove all but one of the cards when we start with 1002 cards.
12
√
B5. π 2 cm Imagine the pole as a paper cylin-
der. Cut it open along its length, then un-
roll it, to get a rectangular strip of paper. So
points A and D correspond to the same point A D
on the cylinder, just like points B and C. The
width of the strip is equal to the perimeter of 45o
the cylinder, so |AD| = |BC| = 2π · 2 cm =
4π cm.
Note that the grey ribbon forms a 45◦ angle
with the cutting line, ABCD is a square. √The
length of the diagonal BD is equal to 2 · B C
4π cm and also equal to four times the width
of the grey ribbon, since the white and grey√stripes have the same width.
It follows that the grey ribbon has width π 2 cm.
C-problems
C1. Since a, b and c are three successive positive odd integers, we can write:
a = 2n − 1, b = 2n + 1 and c = 2n + 3, with n a positive integer.
A calculation then gives:
a2 + b2 + c2 = (2n − 1)2 + (2n + 1)2 + (2n + 3)2

= (4n2 − 4n + 1) + (4n2 + 4n + 1) + (4n2 + 12n + 9)
= 12n2 + 12n + 11.
This needs to be equal to an integer that consists of four digits p. Hence the
integer 12n2 + 12n consists of four digits, of which the first two are equal
to p, and the last two are equal to p − 1. Since 12n2 + 12n is divisible by 2,
p − 1 has to be even. So we have the following possibilities for 12n2 + 12n;
1100, 3322, 5544, 7766 and 9988. This integer also must be divisible by
3, so the only integer remaining is 5544, so n2 + n = 5544 12 = 462. We
can rewrite this as n2 + n − 462 = 0. Factorizing this quadratic equation
then gives: (n − 21)(n + 22) = 0. Since n is a positive integer, the only
solution is n = 21. So the only triple satisfying the given properties is
(a, b, c) = (41, 43, 45).
13
C2. Note that the possible scores are multiples of 5. The lowest score a student
can get is 0, and the highest score is 16 · 10 = 160. Now suppose that there
are no two students with the same score. Then the combined score of the
students is at most 160 + 155 + 150 + · · · + 15 = 12 · 175 · 30 = 2625. We’ll
derive a contradiction from this.
Let A be the combined number of correct answers that were given within
one minute, B be the combined number of correct answers that were not
given within a minute, and C be the combined number of incorrect answers.
The students answered 16·30 = 480 questions together, so A+B+C = 480.
More than half of the questions was answered correctly within one minute,
so A > 240. Also note that B = C, so B = C = 480−A 2 . We now can
express the combined score in A. This is equal to:
480 − A 15
10 · A + 5 · B + 0 · C = 10 · A + 5 · = 2 A + 1200.
2
Since A > 240, the combined scores of the students is greater than 15
2 ·
240 + 1200 = 3000. But from the assumption that no two students have
the same score, we deduced that the combined score was at most 2625.
This is a contradiction. We deduce that this assumption was wrong, so
that there are two students with the same score.
14
Problems
solved the problem.
1. Determine all triples of positive integers (a, b, n) that satisfy the following
equation:
a! + b! = 2n .
Notation: k! = 1×2×· · ·×k, for example: 1! = 1, and 4! = 1×2×3×4 = 24.
2. Let ABC be a triangle. Points P and Q lie on side BC and satisfy |BP | =
|P Q| = |QC| = 31 |BC|. Points R and S lie on side CA and satisfy |CR| =
|RS| = |SA| = 13 |CA|. Finally, points T and U lie on side AB and satisfy
|AT | = |T U | = |U B| = 13 |AB|. Points P, Q, R, S, T and U turn out to lie
on a common circle.
Prove that ABC is an equilateral triangle.
3. In a tournament among six teams, every team plays against each other
team exactly once. When a team wins, it receives 3 points and the losing
team receives 0 points. If the game is a draw, the two teams receive 1
point each.
Can the final scores of the six teams be six consecutive numbers a, a +
1, . . . , a + 5? If so, determine all values of a for which this is possible.
4. Determine all pairs of positive real numbers (a, b) with a > b that satisfy
the following equations:
√ √ √ √
a a + b b = 134 and a b + b a = 126.
5. The number devil has coloured the integer numbers: every integer is
coloured either black or white. The number 1 is coloured white. For
every two white numbers a and b (a and b are allowed to be equal) the
numbers a − b and a + b have different colours.
Prove that 2011 is coloured white.
15
Solutions
1. Since a and b play the same role in the equation a! + b! = 2n , we will

assume for simplicity that a 6 b. The solutions for which a > b are found
by interchanging a and b. We will consider the possible values of a.
Case a > 3: Since 3 6 a 6 b, both a! and b! are divisible by 3. Hence

a! + b! is divisible by 3 as well. Because 2n is not divisible by 3 for
any value of n, we find no solutions in this case.
Case a = 1: The number b must satisfy b! = 2n − 1. This implies that
b! is odd, because 2n is even (recall that n > 1). Since b! is divisible
by 2 for all b > 2, we must have b = 1. We find that 1! = 2n − 1,
which implies that n = 1. The single solution in the case is therefore
(a, b, n) = (1, 1, 1).
Case a = 2: There are no solutions for b > 4. Indeed, since b! would
then be divisible by 4, 2n = b! + 2 would not be divisible by 4,
which implies that 2n = 2. However, this contradicts the fact that
2n = b! + 2 > 24 + 2.
For b = 2, we find 2n = 2 + 2 = 4. Hence n = 2 and (a, b, n) = (2, 2, 2)
is the only solution.
For b = 3, we find 2n = 2 + 6 = 8. Hence n = 3 and (a, b, n) =
(2, 3, 3) is the only solution. By interchanging a and b, we obtain the
additional solution (a, b, n) = (3, 2, 3).
In all, there are four solutions: (1, 1, 1), (2, 2, 2), (2, 3, 3) and (3, 2, 3).
2. Denote by K the midpoint of P Q. Then K C

is also the midpoint of BC, and AK is a me-
dian of triangle ABC. We denote by L the
R Q
intersection of AK and ST .
Triangles AST and ACB are similar (sas), be- K P
S
cause ∠CAB = ∠SAT and |CA| |BA|
|SA| = 3 = |T A| .
L
This implies that ST and CB are parallel lines
(equal corresponding angles). A T U B
Triangles ASL and ACK are similar (aa), because ∠SAL = ∠CAK and
∠LSA = ∠T SA = ∠BCA = ∠KCA. Hence |CK| |CA|
|SL| = |SA| = 3. This
|SL| 3·|SL| |CK|
implies that L is the midpoint of ST , because |ST | = 3·|ST | = |CB| = 12 .
Consider the center M of the circle through P , Q, R, S, T and U . It
is incident to both the perpendicular bisector of P Q, and that of ST .
16
However, since P Q and ST are parallel, the two perpendicular bisectors
must coincide: they are the same line. This line is incident to L and K,
and is therefore equal to line AK, which shows that AK ⊥ BC.
It follows that |AC| = |AB|, because AK is the perpendicular bisector of
BC.
In a similar fashion, one can show that |AC| = |BC|, concluding the proof
that triangle ABC is equilateral.
3. In all, 15 matches are played. In each match, the two teams together earn
2 or 3 points. The sum of the final scores is therefore an integer between
15 · 2 = 30 (all matches end in a draw) and 15 · 3 = 45 (no match is a
draw).
On the other hand, the sum of the six scores equals a+(a+1)+· · ·+(a+5) =
15 + 6a. Hence 30 6 15 + 6a 6 45, which shows that 3 6 a 6 5. We will
prove that a = 4 is the only possibility.
First consider the case a = 5. The sum of the scores equals 15 + 30 = 45,
so no match ends in a draw. Because in every match the teams earn either
0 or 3 points, every team’s score is divisible by 3. Therefore, the scores
cannot be six consecutive numbers.
Next, consider the case a = 3. The scores sum up to 3+4+5+6+7+8 = 33.
The two teams scoring 6 and 7 points must both have won at least one out
of the five matches they played.
The team scoring 8 points must have won at least two matches, because
3 + 1 + 1 + 1 + 1 = 7 < 8. Hence at least 4 matches did not end in a draw,
which implies
A B C D E F
that the sum of the scores is at least 4 · 3 +
A - 3 1 0 0 0 4
11 · 2 = 34. But we have already see that B 0 - 1 0 3 1 5
this sum equals 33, a contradiction. C 1 1 - 3 0 1 6
Finally, we will show that a = 4 is possi- D 3 3 0 - 1 0 7
E 3 0 3 1 - 1 8
ble. The table depicts a possible outcome
F 3 1 1 3 1 - 9
in which teams A to F have scores 4 to
9. The rightmost column shows the total
scores of the six teams.
17
√ √
4. For convenience, write y = a and z = b. The equations transform to
y 3 + z 3 = 134 and y 2 z + yz 2 = 126.
Combining these two equations in a handy way, we find
(y + z)3 = (y 3 + z 3 ) + 3(y 2 z + yz 2 ) = 134 + 3 · 126 = 512 = 83 .
This immediately implies that y + z = 8.
Rewrite the first equation as follows: (y + z)yz = y 2 z + yz 2 = 126. Since
y + z = 8, we see that yz = 126 63
8 = 4 .
From y + z = 8 and yz = 63 4 , we can determine y and z by solving a
2
√ the roots of the equation x −
quadratic equation: y and z are precisely
63 8± 64−4· 63 9
8x + 4 = 0. The two solutions are 2
4
, that is 2 and 72 .
9
Since a > b, also y > z holds. Hence y = 2 and z = 72 . We therefore find
that (a, b) = ( 81 49
4 , 4 ).
Because (a, b) = ( 81 49
4 , 4 ) satisfies the given equations, as required, we
conclude that this is the (only) solution.
5. We are give that 1 is white. Hence 0 is black, because otherwise 1 = 1 − 0

and 1 = 1+0 would have different colours. The number 2 is white, because
0 = 1 − 1 (black) and 2 = 1 + 1 have different colours.
By induction on k, we show that the following claim holds for every k > 0:
3k is black, 3k + 1 and 3k + 2 are white.
We have just shown the base case k = 0. Assume that the claim holds true
for k = `.
Since 1 is white, and 3` + 2 is white by the induction hypothesis, the
numbers (3` + 2) − 1 = 3` + 1 and (3` + 2) + 1 = 3(` + 1) have different
colours. As 3` + 1 is white by the induction hypothesis, 3(` + 1) must be
black.
Since 2 and 3` + 2 are both white, the numbers (3` + 2) + 2 = 3(` + 1) + 1
and (3` + 2) − 2 = 3` must have different colours. As 3` is black by the
induction hypothesis, 3(` + 1) + 1 must be white.
Since 3(` + 1) + 1 and 1 are both white, the numbers 3(` + 1) + 1 + 1 =
3(` + 1) + 2 and 3(` + 1) have different colours. We already know that
3(` + 1) is black, so 3(` + 1) + 2 must be white.
This proves the claim for k = ` + 1.
Because 2011 = 3 · 670 + 1, this shows that 2011 is white.
18
BxMO/EGMO Team Selection Test, March 2012
Problems
1. Do there exist quadratic polynomials P (x) and Q(x) with real coefficients
such that the polynomial P (Q(x)) has precisely the zeros x = 2, x = 3, x =
5 and x = 7?
2. Let 4ABC be a triangle and let X be a point in the interior of the triangle.
The second intersection points of the lines XA, XB and XC with the
circumcircle of 4ABC are P, Q and R. Let U be a point on the ray XP
(these are the points on the line XP such that P and U lie on the same
side of X). The line through U parallel to AB intersects BQ in V . The
line through U parallel to AC intersects CR in W .
Prove that Q, R, V , and W lie on a circle.
3. Find all pairs of positive integers (x, y) for which
x3 + y 3 = 4(x2 y + xy 2 − 5).
4. Let ABCD a convex quadrilateral (this means that all interior angles are
smaller than 180◦ ), such that there exist a point M on line segment AB
and a point N on line segment BC having the property that AN cuts the
quadrilateral in two parts of equal area, and such that the same property
holds for CM .
Prove that M N cuts the diagonal BD in two segments of equal length.
5. Let A be a set of positive integers having the following property: for each
positive integer n exactly one of the three numbers n, 2n and 3n is an
element of A. Furthermore, it is given that 2 ∈ A. Prove that 13824 ∈ / A.
19
Solutions
1. Suppose that such polynomials exist and write Q(x) = ax2 + bx + c. If

we evaluate Q in 2, 3, 5, and 7, we must get exactly the (at most) two
zeros of P . Since Q(x) attains each value at most two times (because Q is
quadratic) we get two different values exactly two times.
Now suppose that Q(n) = Q(m) for different numbers m and n. Then
an2 + bn + c = am2 + bm + c, hence a(n2 − m2 ) = b(m − n), hence
a(n + m)(n − m) = −b(n − m). As m − n 6= 0, this yields a(n + m) = −b,
or equivalently, n + m = −ba .
We know that 2, 3, 5 and 7 split in two pairs (m, n) and (k, l) such that
Q(m) = Q(n) and Q(k) = Q(l). Therefore, it holds that m + n = −b a =
k + l. We have to split the four numbers 2, 3, 5 and 7 in two pairs having
the same sum. However, this is impossible, since 2 + 3 + 5 + 7 = 17 is odd.
We conclude that there are no polynomials P and Q having the required
properties.
2. As AB and U V are parallel, it holds that ∆ABC ∼ ∆U V X (AA). Anal-

ogously, it holds that ∆ACX ∼ ∆U W X. These similarities yield that
|XA| |XU | |XA| |XU |
= and = .
|XB| |XV | |XC| |XW |
This yields that
|XB| |XB| |XA| |XV | |XU | |XV |
= = = .
|XC| |XA| |XC| |XU | |XW | |XW |
The power of a point theorem for the point X and the cyclic quadrilateral
BCQR gives that
|XB| |XV |
|XC| · |XR| = |XB| · |XQ| = · |XC| · |XQ| = · |XC| · |XQ|,
|XC| |XW |
hence
|XW | · |XR| = |XV | · |XQ|.
From the given configuration (X in the interior of the triangle, U and P
on the line XP on the same side of X) follows that R and W lie on the
same side of X on the line XC, and also that V and Q lie on the same
side of X on the line XB. Hence, the following equality also holds for the
directed distance: XW · XR = XV · XQ. The power of a point theorem
then yields that W, R, V and Q lie on a circle.
20
3. It holds that (x + y)3 = x3 + 3x2 y + 3xy 2 + y 3 , hence the given equality
yields
(x + y)3 = x3 + y 3 + 3x2 y + 3xy 2

= 4(x2 y + xy 2 − 5) + 3x2 y + 3xy 2
= 7x2 y + 7xy 2 − 20
= 7xy(x + y) − 20.
Now x + y is a divisor of 20, because x + y is a divisor of (x + y)3 and

7xy(x + y). As x + y > 2, the possible values for x + y are 2, 4, 5, 10, 20.
Considering the equality (x + y)3 − 7xy(x + y) − 20 modulo 7 yields
(x + y)3 ≡ −20 ≡ 1 mod 7.
For each of the possible values for x + y we check whether its third power
is congruent to 1 modulo 7. It holds that 53 ≡ −1 mod 7, hence x + y = 5
is impossible. It holds that 103 ≡ −1 mod 7, hence x + y = 10 is also
impossible. It holds that 203 ≡ −1 mod 7, hence x + y = 20 is also
impossible. We are left with the possibilities x + y = 2 and x + y = 4.
If x + y = 2, then x = y = 1 and then the left hand site is positive and the
right hand side is negative, hence also x + y = 2 is impossible. If x + y = 4,
then (x, y) is equal to (1, 3), (2, 2) or (3, 1). By substituting this into the
equality we find that only (1, 3) and (3, 1) are solutions.
4. For a polygon P, denote by O(P) the area of P. Let T be the midpoint

of BD. Now triangles CDT and CBT have equal length bases, namely
|DT | = |BT |, and their altitudes have the same length, hence their areas
are equal. In the same way it holds that O(ADT ) = O(ABT ). Hence,
O(AT CD) = 12 O(ABCD) = O(AN CD). Remark that T cannot lie in the
interior of triangle ACD, because then O(AT CD) < O(ACD) would hold,
while it holds that O(AN CD) > O(ACD). Therefore,
O(AT C) = O(AT CD) − O(ACD) = O(AN CD) − O(ACD) = O(AN C).
Triangles AT C and AN C have the same base AC, hence their altitudes
have equal lengths. This means that the line N T is parallel to the base
AC.
Analogously, we show that M T is parallel to AC. Hence, N T k M T and
this yields that M , N and T lie on a line. Hence, the midpoint of BD lies
on the line M N .
21
5. We will prove the following two assertions:
(i) If m ∈ A and 2 | m, then 6m ∈ A.
(ii) If m ∈ A and 3 | m, then 43 m ∈ A.
First we prove assertion (i). Suppose that m ∈ A and 2 | m. By choosing

n= m m 3
2 , we find that 2 and 2 m do not lie in A. By choosing n = m, we
find that 2m and 3m do niet lie in A. If we now consider n = 23 m, then
as 2n = 3m and n = 32 m do not lie in A, hence 3n = 92 m lies in A. Using
n = 92 m, we find that 9m ∈ / A. Finally, we consider n = 3m: we know that
3m ∈ / A and 9m ∈ / A, hence 6m ∈ A.
Now we prove assertion (ii). Suppose that m ∈ A and 3 | m. By choosing
n= m m 2
3 , we find that 3 and 3 m do not lie in A. By choosing n = m, we
find that 2m does not lie in A. If we now choose n = 23 m, we know that n
and 3n do not lie in A, hence 2n = 43 m does lie in A. This proves assertion
(ii).
We know that 2 ∈ A. By repeatedly applying assertions (i) and (ii), we
show that the following numbers all lie in A:
(i) (i) (i) (i) (ii) (ii)
2 → 22 · 3 → 23 · 32 → 24 · 33 → 25 · 34 → 27 · 33 → 29 · 32 .
As 29 · 32 ∈ A, it holds that 29 · 33 ∈
/ A. Since 13824 = 29 · 33 , this is what
we had to prove.
22
Problems
1. Let I be the incentre of triangle ABC. A line through I intersects the

interior of line segment AB in M and the interior of line segment BC in
N . We assume that BM N is an acute triangle. Let K and L be points on
line segment AC such that ∠BM I = ∠ILA and ∠BN I = ∠IKC.
Prove that |AM | + |KL| + |CN | = |AC|.
2. Let a, b, c and d be positive real numbers. Prove that

a−b b−c c−d d−a
+ + + > 0.
b+c c+d d+a a+b
a a+1
3. Determine all positive integers that cannot be written as b + b+1 where a
and b are positive integers.
4. Let n be a positive integer divisible by 4. We consider the permutations

(a1 , a2 , . . . , an ) of (1, 2, . . . , n) having the following property: for each j we
( 1 n)!
have ai + j = n + 1 where i = aj . Prove that there are exactly ( 21 n)! such
4
permutations.
5. Let Γ be the circumcircle of the acute triangle ABC. The angle bisector
of angle ABC intersects AC in the point B1 and the short arc AC of Γ in
the point P . The line through B1 perpendicular to BC intersects the short
arc BC of Γ in K. The line through B perpendicular to AK intersects AC
in L.
Prove that K, L and P lie on a line.
23
Solutions
1. Let D, E and F be the orthogonal projections of I on respectively BC,

CA and AB. It holds that N lies between C and D: since if N would
lie between D and B, then ∠BN I would be greater than ∠BDI = 90◦ ,
but it is given that 4BM N is acute. Hence, N lies between C and D.
Analogously, M lies between A and F . Furthermore, L cannot lie between
A and E, because in that case, ∠ILA > 90◦ would hold, while we have
∠ILA < ∠BM I < 90◦ . Hence, L lies between E and C. Analogously, K
lies between A and E. Altogether, E lies between K and L.
It holds that
|AC| = |AE| + |CE| = |AF | + |CD| = |AM | + |M F | + |CN | + |N D|,
where the second equality holds because the segments on the tangent lines
to the incircle have equal lengths.
Furthermore, we have ∠IKE = ∠IKC = ∠BN I = ∠DN I and ∠KEI =
90◦ = ∠IDN , hence 4IKE ∼ 4IN D (AA). Since line segments EI and
DI both are radii of the incircle, they have equal lengths, hence we even
have 4IKE ∼ = 4IN D. This yields that |EK| = |N D|.
Analogously, we find that |EL| = |M F |. Altogether, we have
|AC| = |AM | + |M F | + |N D| + |CN |

= |AM | + |EL| + |EK| + |CN | = |AM | + |KL| + |CN |.
2. It holds that
a−b a−b+b+c a+c
= −1= − 1.
b+c b+c b+c
By applying the equality also to the other three fractions and adding 4 to
both sides, we find that it is sufficient to prove that:
a+c b+d c+a d+b
LHS := + + + > 4. (1)
b+c c+d d+a a+b
Next, we apply the inequality of the harmonic and geometric mean to the
two positive numbers b + c and d + a:
2 (b + c) + (d + a)
1 1 6 ,
b+c + d+a
2
hence,
1 1 4
+ > .
b+c d+a a+b+c+d
24
Analogously, it holds that
1 1 4
+ > .
c+d a+b a+b+c+d
Using this, we find the following inequality for the left-hand side of (1):

1 1 1 1
LHS = (a + c) + + (b + d) +
b+c d+a c+d a+b
4 4
> (a + c) · + (b + d) ·
a+b+c+d a+b+c+d
(a + c) + (b + d)
=4·
a+b+c+d
= 4.
This proves (1).
3. It holds that
a a+1 2ab + a + b
+ = .
b b+1 b(b + 1)
Next, suppose that this is equal to an integer n. Then we have b | 2ab+a+b
and b + 1 | 2ab + a + b. The former yields that b | a and hence b | a − b.
The latter yields b + 1 | (2ab + a + b) − (b + 1) · 2a = −a + b and hence also
b + 1 | a − b. Since the gcd of b and b + 1 is equal to 1, we may conclude
that b(b + 1) | a − b. Hence, we can write a as a = b(b + 1) · k + b. Since a
is positive, k must be a non-negative integer. Substitution yields
2ab + a + b
n=
b(b + 1)

2 · b(b + 1) · k + b · b + b(b + 1) · k + b + b
=
b(b + 1)
b(b + 1) · (2kb + k) + 2b2 + 2b
=
b(b + 1)
= (2b + 1)k + 2.
Hence, n is of the form n = (2b + 1)k + 2. This yields that n > 2 (because
k > 0) and that n − 2 is divisible by an odd integer greater than 1 (namely
2b + 1 > 3).
For the converse, suppose that for a number n we have that n > 2 and that
n − 2 is divisible by an odd integer greater than 1, say 2b + 1 where b > 1
is an integer. Then there is a k > 0 for which we have n = (2b + 1)k + 2.
25
Now choose a = b(b + 1) · k + b, then a is a positive integer and it holds
that
a a+1
+ = (b + 1)k + 1 + bk + 1 = (2b + 1)k + 2 = n.
b b+1
We conclude that the numbers that can be written as ab + a+1 b+1 , where a
and b are positive integers, are exactly the numbers n > 2 for which n − 2
is divisible by an odd integer greater than 1. Hence, the numbers that
cannot be written like this, are exactly 1 and the numbers n > 2 for which
n − 2 has no odd divisor greater than 1, otherwise stated for which n − 2
is a power of two, say 2m where m > 0.
We conclude that the numbers that cannot be written as ab + a+1 b+1 , are
exactly 1 and the numbers of the form 2m + 2 for m > 0.
4. Let t ∈ {1, 2, . . . , n}. Suppose that at = t, then we may choose i = j = t

to find that at + t = n + 1, hence 2t = n + 1. However, n is divisible by 4,
hence n+1 is odd. This is a contradiction. Now suppose that at = n+1−t.
Then we may choose i = n + 1 − t and j = t to find that an+1−t + t = n + 1,
hence an+1−t = n + 1 − t. We have just seen that this cannot occur.
Now suppose that at = u where u 6= t, u 6= n + 1 − t. Then we may choose
i = u and j = t to find that au + t = n + 1, hence au = n + 1 − t. Next,
we may choose i = n + 1 − t and j = u to find that an+1−t = n + 1 − u.
Next, we may choose i = n + 1 − u and j = n + 1 − t to find that
an+1−u = n + 1 − (n + 1 − y) = y. Altogether, we have found that:
at = u,
au = n + 1 − t,
an+1−t = n + 1 − u,
an+1−u = t.
Since u 6= t and u 6= n + 1 − t, the four numbers on the right hand side are
pairwise distinct. Furthermore, we can divide the four numbers into two
pairs of the form (v, n + 1 − v). We have found four numbers for which
it holds that on the same positions in the permutation are the same four
numbers, but in a different order. Now we may choose a t0 not equal to
one of the four mentioned numbers and a u0 with at0 = u0 and in the same
way we find a quadruple containing t0 . Note that n + 1 − t0 and n + 1 − u0
cannot be a number of the first quadruple, since then also u0 and t0 would
be a number of the first quadruple. We can continue this way until all n
numbers have been divided into quadruples.
We can generate all permutations by using the following procedure:
26
• Choose the smallest number k for which ak is not yet determined.
Choose ak = u for a certain u for which au is not yet determined and
for which u 6= k, u 6= n + 1 − k. This also determines the values of
au , an+1−u and an+1−k .
• Repeat the last step until all values of ak are determined.
For the first k we have n − 2 possible choices of u. During the next step
we have n − 6 possible choices left. One step later, we have n − 10, and so
on. Hence, the total number of permutations having the property, is equal
to
2 · 6 · 10 · . . . · (n − 10) · (n − 6) · (n − 2).
Write n = 4m, then we can write this as
(2m)!
2m · 1 · 3 · 5 · . . . ·(2m − 5) · (2m − 3) · (2m − 1) = 2m ·
2 · 4 · . . . · (2m)
(2m)! (2m)! ( 1 n)!
= = = 21 .
1 · 2 · 3 · ... · m m! ( 4 n)!
5. The fact that the angle bisector of angle ABC intersects the short arc AC
in P , implies that P lies exactly on the middle of this arc AC. We have
to prove that KL also goes through P , hence that KL intersects the arc
AC in the middle. Because K lies on Γ, this means that we have to prove
that KL is the angle bisector of ∠AKC.
Let S be the intersection point of B1 K and BC and let T be the intersection
point of BL and AK. Then we have ∠BSK = 90◦ and ∠BT K = 90◦ ,
hence BT SK is a cyclic quadrilateral. This yields that
∠CBL = ∠SBT = ∠SKT = ∠B1 KA. (2)
Because ABKC is a cyclic quadrilateral, we have ∠B1 AK = ∠CAK =
∠CBK. By the exterior angle theorem, we have ∠LB1 K = ∠B1 AK +
∠B1 KA, hence because of (2) we get
∠LB1 K = ∠B1 AK + ∠B1 KA = ∠CBK + ∠CBL = ∠LBK.
This yields that LKBB1 is a cyclic quadrilateral, which means that ∠LBB1 =
∠LKB1 . If we add (2) to this, we get
∠CBB1 = ∠CBL + ∠LBB1 = ∠B1 KA + ∠LKB1 = ∠LKA.
Hence,
∠LKA = ∠CBB1 = 12 ∠CBA = 21 ∠CKA,
where we used that ABKC is a cyclic quadrilateral. This yields that KL
is the angle bisector of ∠AKC, as desired.
27
Problems
a−b
1. For all positive integers a and b, we define a b = .
gcd(a, b)
Show that for every integer n > 1, the following holds: n is a prime power
if and only if for all positive integers m such that m < n, it holds that
gcd(n, n m) = 1.
2. There are two boxes containing balls. One of them contains m balls, and
the other contains n balls, where m, n > 0. Two actions are permitted:
(i) Remove an equal number of balls from both boxes.

(ii) Increase the number of balls in one of the boxes by a factor k.
Is it possible to remove all of the balls from both boxes with just these two
actions,
1. if k = 2?
2. if k = 3?
3. Determine all pairs (x, y) of positive integers satisfying
x + y + 1 | 2xy and x + y − 1 | x2 + y 2 − 1.
4. Let 4ABC be a triangle. The angle bisector of ∠CAB intersects BC at

L. On the interior of line segments AC and AB, two points, M and N ,
respectively, are chosen in such a way that the lines AL, BM and CN are
concurrent, and such that ∠AM N = ∠ALB. Prove that ∠N M L = 90◦ .
f x + xy + f (y) = f (x) + 12 f (y) + 12

for all x, y ∈ R.
28
Solutions
1. First suppose that n = pk with p prime and k > 0. Then we need to show
that for all positive integers m with m < n, we have gcd(n, n m) = 1.
So let m be a positive integer such that m < n. Write m = pl s with p - s
and 0 6 l < k. Then gcd(n, m) = gcd(pk , pl s) = pl , so
pk − pl s
n m= = pk−l − s.
pl
Since k −l > 1, we have gcd(pk−l −s, p) = gcd(−s, p) = 1, hence gcd(n, n

m) = gcd(pk , pk−l − s) = 1.
Now suppose that n is not a prime power. Then we need to show that there
exists some positive integer m with m < n such that gcd(n, n m) 6= 1. Let
q be the smallest prime dividing n, and let t be the unique positive integer
such that q t | n and q t+1 - n. As n is not a power of q, there exists a prime p
with p > q dividing n. Hence qnt > p > q, from which follows that n > q t+1 .
Now let m = n − q t+1 . Then we have gcd(n, m) = gcd(n, q t+1 ) = q t , so
n − (n − q t+1 ) q t+1
n m= = = q.
qt qt
As q | n, it follows that gcd(n, n m) = q > 1.
2. First suppose that k = 2. Then we can remove all of the balls from both
boxes as follows.
If m = n, we remove m balls from both boxes, and we are done. If m 6= n,
we may assume without loss of generality that m < n. If furthermore,
2m < n holds, then we double the number of balls in the first box, until
there are m0 balls in the first box, where m0 satisfies both m0 < n and
2m0 > n. Hence we may assume without loss of generality that 2m > n.
Now remove 2n − m > 0 balls from both boxes. In the first box, m − (2m −
n) = n−m > 0 balls remain; in the second box, n−(2m−n) = 2(n−m) > 0
balls remain. Doubling the number of balls in the first box, then removing
2(n − m) balls from both boxes now makes both boxes empty.
Now consider the case k = 3. Let S denote the number of balls in the
first box, minus the number of balls in the second one. The first action
does not change S. Applying the second action to some box with t balls,
does not change the parity of S either, since we have 3t ≡ t (mod 2).
Hence the parity of S does not change. We conclude that the parity of
S is an invariant under both actions. Now suppose that we start with
29
(m, n) = (1, 2), then S = −1 ≡ 1 (mod 2). Then it is impossible to reach
the situation in which both boxes are empty, since then one would have
S ≡ 0 (mod 2). Hence we cannot empty both boxes if k = 3.
3. Note that
(x2 +y 2 −1)−(x+y+1)(x+y−1) = (x2 +y 2 −1)−(x2 +y 2 +2xy−1) = −2xy.
As x + y + 1 divides x2 + y 2 − 1 and (x + y + 1)(x + y − 1), it follows that

x + y − 1 divides 2xy. It was given that x + y + 1 divides 2xy as well. But
the difference of x + y − 1 and x + y + 1 is 2, so their greatest common
divisor is either 1 or 2. So either 2xy is divisible by (x + y + 1)(x + y − 1)
(if the gcd is 1), or 2xy is divisible by 12 (x + y + 1)(x + y − 1) (if the gcd is
2). In both cases, we have (x + y − 1)(x + y + 1) | 4xy, so for some k > 1,
we have
4xy = k(x + y − 1)(x + y + 1) = k(x2 + y 2 + 2xy − 1) > k(4xy − 1).
Note that the last inequality holds, as for all real numbers x and y, we
have x2 + y 2 > 2xy.
If k > 2, then 4xy > 2·(4xy−1), so 4xy 6 2. This contradicts x and y being
positive integers. Hence k = 1, so we deduce that 4xy = x2 + y 2 + 2xy − 1.
This implies that x2 + y 2 − 1 − 2xy = 0, or equivalently, (x − y)2 = 1.
Hence either x = y − 1, or x = y + 1.
Hence the only pairs that can satisfy the conditions are (x, x + 1) with
x > 1, and (x, x − 1) with x > 2. We check that they indeed satisfy the
conditions. Indeed, for the first family of pairs, we note that 2x + 2 divides
2x(x + 1), and that 2x divides x2 + (x + 1)2 − 1 = 2x2 + 2x, and similarly
for the second family of pairs. We conclude that the solutions are precisely
the pairs (x, x + 1) with x > 1 and the pairs (x, x − 1) with x > 2.
4. Let T be the intersection of M N and BC. As ∠ACB = ∠ALB −∠LAC =

∠AM N −∠LAC < ∠AM N , it follows that T and B lie on the same side of
C, and that T and N lie on the same side of N . Since ∠AM T = ∠AM N =
∠ALB = ∠ALT , we note that AM LT is a cyclic quadrilateral. Hence
∠N M L = ∠T M L = ∠T AL. So it suffices to show that ∠T AL = 90◦ . This
is the case if and only if AT is the external angle bisector of ∠CAB, because
the internal and external angle bisectors of any angle are perpendicular.
As AL, BM and CN are concurrent, Ceva’s Theorem gives
BL CM AN
· · = 1.
LC M A N B
30
As M , N and T are collinear, Menelaus’ Theorem gives
BT CM AN
· · = −1.
TC MA NB
From these two identities, it follows that
BL BT
=− .
LC TC
By the angle bisector theorem, we have |BL| |BA| |BT | |BA|

|LC| = |CA| , hence |T C| = |CA| .
So, again by the angle bisector theorem, it follows that AT is the external
angle bisector of ∠CAB, as desired.
5. Substituting y = −1 gives
f f (−1) = f (x) + 12 f (−1) + 12 .

If f (−1) 6= − 21 , then we can divide this equation by f (−1) + 21 , which gives

1 f f (−1)
f (x) + 2 = ,
f (−1) + 12
so f is constant. Then there is some c ∈ R such that f (x) = c for all

x ∈ R. Substituting this in the functional equation, gives
c = c + 12 c + 12 ,

or equivalently, 0 = c2 + 41 , but this equation has no real solutions. We

deduce that f cannot be constant.
Hence it follows that f (−1) = − 12 . This also implies that f f (−1) = 0,

so f (− 12 ) = 0. Substituting x = 0 and y = − 12 then gives
f f (− 12 ) = f (0) + 12 f (− 12 ) + 12 ,

hence
1
· 21 ,

f (0) = f (0) + 2
so f (0) = 12 .
Now suppose that there exists some a 6= −1 such that f (a) = − 21 . Substi-
tuting y = a now gives
f x(1 + a) − 12 = 0.

31
As 1 + a 6= 0, the function mapping x to x(1 + a) − 12 attains all values
in R. Hence f must be identically zero, hence constant, contradicting our
earlier findings. So the unique x ∈ R such that f (x) = − 12 is x = −1.
Now let b be such that f (b) = 0. Substitute x = b − 21 , and y = 0:
f b − 12 + 21 = f b − 12 + 12 12 + 12 ,

or equivalently,
1 1

f (b) = f b − 2 + 2 .
1
Since f (b) = 0, we have f (b − 2) = − 21 . Hence it follows by an earlier
result that b − 12 = −1, hence b = − 12 . Hence the unique x ∈ R with
f (x) = 0 is x = − 21 .
Now substitute x = −1. This gives

f −1 − y + f (y) = 0.
From this it follows that −1 − y + f (y) = − 12 , so f (y) = y + 21 . Hence if a

function f satisfies the functional equation, it must be the function given
by f (x) = x + 21 .
We check that this function indeed satisfies the functional equation. In-
deed, the left hand side becomes x + xy + y + 1, which is equal to the right
hand side (x + 1)(y + 1). Hence the function f given by f (x) = x + 21 is
the unique function satisfying the functional equation.
32
Problems
Part 1
1. A chord of a circle is a line segment whose endpoints

lie on the circle.
What is the maximal number of intersections that
six chords of a circle can have?
A) 10 B) 12 C) 13 D) 14 E) 15
2. There are five buttons on the vertices of the

Pentagon. Each of these are either black or
white. Pressing one of these buttons will cause
this button, together with the buttons on the
edge opposite to it, to change colours; from voor na
black to white, and vice versa.
Suppose that all of the buttons are white. What is the minimal number of
button presses needed to turn all of the buttons black?
A) 3 B) 4 C) 5 D) 7 E) 10
3. Let a, b, c, d, e, f , g, h be a sequence of numbers with the property that

any three consecutive numbers sum to 30 (e.g. b + c + d = 30). Suppose
that c = 5.
What is a + h?
A) 10 B) 14 C) 15 D) 20 E) 25
4. In a cube with edges of length 3 cm, some holes are made. In the center of
each face, a square hole with sides of length 2 cm is made, such that the
sides of this square are parallel to the edges of the cube. This hole runs
through the entire cube.
What is (in cm3 ) the volume of the object that remains after making these
three holes?
A) 6 B) 7 C) 8 D) 9 E) 10
33
5. In a classroom, one or more pupils always speak the truth. The other
pupils sometimes do speak the truth, and sometimes do not. The pupils
were asked how many of them always speak the truth. The answers were:
5, 6, 2, 3, 4, 6, 3, 6, 3, 4, 6, 5, 4, 3, and 6.
How many pupils do always speak the truth?
A) 2 B) 3 C) 4 D) 5 E) 6
6. Keith owns a machine in which he can put two numbers, A and B. He

can choose A and B to be any of the numbers 0, 1, 2, . . . , 1000 (A and
B are allowed to be equal). The machine will then return the number
100×A + 3×B. Keith tries to get the machine to return as many of the
numbers from 1 up to 1000 as possible. However, for some numbers, it is
impossible for the machine to return it, no matter what numbers Keith
puts into the machine.
How many such numbers from 1 up to 1000 are there?
A) 66 B) 67 C) 99 D) 100 E) 363
7. A square with sides of length 1 is divided into two

pieces by a line segment of length 1, parallel to a
diagonal.
What is the area of the smallest piece?
1 2 5 1 3
A) 4 B) 7 C) 16 D) 3 E) 8
8. In the addition to the right, the question mark and ∗ ∗ ∗

the asterisks all represent a digit. Every digit from
1 up to 9 occurs exactly once. ∗ ∗ ∗ +
What is the digit represented by the question mark? 4 ? 9
A) 2 B) 3 C) 5 D) 7 E) 8
9. Adrian puts stones in the nine squares of a 3×3-board. Into each of the
squares, he is allowed to put any number of stones. He is also allowed to
leave squares empty. When Adrian is finished, he counts the numbers of
stones lying in each of the columns and rows. He wants these six numbers
to be pairwise distinct.
At least how many stones does Adrian need to achieve this?
A) 7 B) 8 C) 9 D) 10 E) 11
34
10. Peter received 100 euro for his birthday. He uses all of it to buy exactly
100 objects. What he buys, consists of liquorice wheels of 10 cents a piece,
bouncing balls of 2 euro a piece, and decks of playing cards of 5 euro a
piece.
How many decks of playing cards does Peter buy?
A) 10 B) 11 C) 15 D) 16 E) 19
11. How many integers have the following properties; the digits are pairwise
distinct, they are non-zero, and the sum of the digits is 10?
A) 32 B) 48 C) 56 D) 511 E) 512
12. Eight children together enumerate all of the numbers from 1 up to 2011,
in the following manner.
• Angie enumerates all of the numbers from 1 up to 2011 in groups
of three, skipping the middle number of each triple. So she says:
“1, 3, 4, 6, 7, 9, . . . , 2005, 2007, 2008, 2010, 2011”.
• Ben enumerates all of the numbers Angie skipped, in groups of three,
skipping the middle number of each triple.
• Catherine enumerates all of the numbers that were skipped by both
Angie and Ben, again in groups of three, and skipping the middle
number of each triple.
• Dorothy, Eve, Francis, and Gerald continue in the same way.
• In the end, Henry enumerates the only remaining number.
Which number is enumerated by Henry?
A) 712 B) 1094 C) 1123 D) 1265 E) 1387
13. In this magic square, the three rows, the three columns,
2
and the two diagonals all have the same sum.
Which number is represented by the question mark?
?
A) 5 B) 6 C) 7 D) 10 E) 16 9
35
14. Ten ants walk upwards on a long and thin blade of grass. When an ant
reaches the tip, it turns around. The ants cannot pass each other, so when
they ‘collide’, they both turn around as well. In the end, all ten ants reach
the bottom of the blade of grass safely.
What is the total number of times that an ant turned around?
A) 45 B) 50 C) 90 D) 95 E) 100
15. On a large sheet of squared paper, the natural num- 17 16 15 14 13

bers 1, 2, 3, 4, . . . are written along a spiral, as shown 18 5 4 3 12
in the figure. Somewhere on this sheet, the number
1000 is written with its eight neighbours around it.
19 6 1 2 11
20 7 8 9 10
Among these eight neighbouring numbers, which one
is the smallest? 21 22 ... ... ...
A) 873 B) 874 C) 875 D) 876 E) 877
Part 2
1. Use the digits 4, 5, 6, and 7 to make two numbers (every digit may only be
used once in total), such that the first number is a multiple of the second
one.
What is the largest possible ratio between these two numbers?
2. The three sides of a triangle have three consecutive integers as length.

The length of the shortest side is 30% of that of the circumference of the
triangle.
What is the length of the longest side?
3. A group of 10 boys and 9 girls goes to watch a movie. They all buy a ticket
for 6 euro. Fourteen of these 19 children buy a bag of popcorn. After the
movie finished, it turns out that the combined amount of money that the
boys spent (on the tickets and bags of popcorn), is equal to the combined
amount of money that the girls spent.
Determine all of the possible prices of a bag of popcorn.
36
4. Quintijn writes down four pairwise distinct numbers that sum up to 44.
Of every pair of numbers, he calculates their difference. These are: 1, 3,
4, 5, 6 and 9.
What possibilities are there for the value of the largest number of the four
numbers that were written down? Write down all possibilities.
5. Julian has a big jar of liquorice wheels. Every day, he eats exactly a quarter
of the liquorice wheels that are in the jar at the beginning of the day. At
the end of the third day, less than 100 liquorice wheels remain in the jar.
At most how many liquorice wheels were in the jar at the beginning of the
first day?
6. The number 1782379 has seven digits. Subtract this number from 9999999,
and you get another seven digit number. Put this number at the end of
1782379, giving a fourteen digit number. Divide this number by 9999999.
What is the result?
7. At the beginning of class, all boys are present, but only two girls are
present. After a while, three more girls enter class. This doubles the
percentage of girls in class.
How many boys are in class?
8. We denote by max(a, b) the largest of the numbers a and b. E.g. max(4, −7) =
4.

What is the smallest possible value of max 5 − max(a, 3), a + 3 ?
9. Divide a 3-by-5 rectangle with one straight cut in

such a way, that a rhombus can be formed with the
two pieces. A rhombus is a quadrilateral with four
sides of equal length.
What is the length of the cut?
10. Given a three digit number, we do the following. We multiply all of its
digits, and subtract the three digits one by one from the result.
Find the smallest three digit number such that this procedure gives the
number 221.
37
Solutions
Part 1
1. E) 15 6. C) 99 11. C) 56
1
2. C) 5 7. A) 4 12. B) 1094
3. E) 25 8. C) 5 13. E) 16
4. B) 7 9. B) 8 14. E) 100
5. B) 3 10. B) 11 15. D) 876
Part 2
1. 189 6. 1782380
2. 11 7. 10
3. 1.50 and 3 euro 8. 2
4. 15 and 16 9. 5
5. 192 10. 568
38
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
4 First Round, February 2011

51st Dutch Mathematical Olympiad 2012
and the team selection for IMO 2013 Colombia
MY

CY
We eat problems
CMY

for breakfast.

in Nederland stgehouden: IMO2011
Olympiad 2012
International
sponsored by: Centrum Wiskunde & Informatica sterdam 2011
OmslagNWO2013-v1.indd 1 14-6-2013 14:56:52
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
OmslagNWO2013-v1.indd 2 14-6-2013 14:56:56

Introduction
The selection process for IMO 2013 started with the first round on 27
January 2012, held at the participating schools. The paper consisted of
eight multiple choice questions and four open-answer questions, to be solved
within 2 hours. In total 5612 students of 270 secondary schools participated
in this first round.
Those 811 students from grade 5 (4, ≤ 3) that scored 20 (16, 12) points
or more on the first round (out of a maximum of 36 points) were invited
to the second round, which was held in March at twelve universities in
the country. This round contained five open-answer questions, and two
problems for which the students had to give extensive solutions and proofs.
The contest lasted 2.5 hours.
Those students from grade 5 (4, ≤ 3) that scored 32 (27, 22) points or
the universities, in order to prepare them for their participation in the final
round.
concluded its 51st edition 2012.

1
girls, as this year we participated for the second time in the European Girls’
Mathematical Olympiad (EGMO). In total there were nine girls competing
to be in the EGMO team. The team of four girls was selected by a selection
test, held on 8 March 2013. They attended the EGMO in Luxembourg from
8 until 14 April, and the team returned with an honourable mention and a
silver medal. For more information about the EGMO (including the 2013
paper), see www.egmo.org.
The same selection test was used to determine the ten students particip-
ating in the Benelux Mathematical Olympiad (BxMO), held in Dordrecht,
the Netherlands, from 26 until 28 April. The Dutch team managed to
come first in the country ranking, and received five bronze medals, two
silver medals and two gold medals. For more information about the BxMO
(including the 2013 paper), see www.bxmo.org.
as an observer C. The team had a training camp in Santa Marta, from 14
July until 21 July.
of two one-hour parts, one with fifteen multiple choice questions and one
with ten open-answer questions. The goal of this Junior Mathematical
the solutions.
2
Dutch delegation
The Dutch team for IMO 2013 in Colombia consists of
• Peter Gerlagh (16 years old)

– bronze medal at BxMO 2011, honourable mention at BxMO
2012, gold medal at BxMO 2013
– observer C at IMO 2012
• Ragnar Groot Koerkamp (18 years old)
– honourable mention at BxMO 2010, gold medal at BxMO 2012,
gold medal at BxMO 2013
– honourable mention at IMO 2011
• Jeroen Huijben (17 years old)
– bronze medal at BxMO 2010, bronze medal at BxMO 2011
– observer C at IMO 2010, bronze medal at IMO 2011, gold medal
at IMO 2012
• Michelle Sweering (16 years old)
– bronze medal at EGMO 2012, silver medal at EGMO 2013
• Djurre Tijsma (18 years old)
– honourable mention at BxMO 2012, silver medal at BxMO 2013
• Jeroen Winkel (16 years old)
– bronze medal at BxMO 2011, silver medal at BxMO 2012
– observer C at IMO 2011, bronze medal at IMO 2012
• Bas Verseveldt (16 years old)

– silver medal at BxMO 2012, bronze medal at BxMO 2013

• Julian Lyczak (observer B), Utrecht University
3
Love, fire and mathematics
Meet four passionate fireflies. Anne, Bob, Casey and David.

Their love for each other represents one of the many beautiful mathematical figures that nature can
show us. Our fireflies are buzzing around in the spatial shape of a tetrahedron ABCD. Each firefly
represents one of the four vertices.
Silent and attractive Anne is head over heels in love with Bob. She is drawn directly towards him. If
Bob moves to the right, she moves to the right. If Bob flies up, she flies up. Sporty Bob has a crush
too, but not on Anne. If he would have been in love with her, they would fly towards each other and
meet in the middle of the edge. That would have been easy! But no, unfortunately Bob is chasing the
playful firefly Casey. Casey is a dancer and if she flies to the right, Bob will follow. If she moves down,
he moves down. But Casey won’t fly towards Bob. She will do everything she can to reach the one
she loves: the adventurous David. However, David is crazy about Anne. He flies directly towards her.
All four fireflies are equally in love. Therefore each will fly at the same speed. As time passes the flies
will eventually meet in the centre of the tetrahedron. An outburst of passion, lovesickness and
heartbreaks follows.
Love is for fools. If they were smart, they all would directly fly to the centre of the figure. But they are
not. They each long for their own centre of the world: their loved one.
Question:
What is the distance each firefly will travel before he or she reaches the centre of our geometric
form?
Please give the answer in units of the length of an edge of the tetrahedron.
Enjoy and good luck!
If you have found the solution to this puzzle, please send your answer to wiskunde@transtrend.com.
Problems
A-problems
A1. In the multiplication table on the right, the stars rep- × ∗ ∗ ∗ 7

resent positive integer numbers such that the table ∗ 24 ∗ ∗ 56
is correct. ∗ ∗ 36 8 ∗
What is the largest number that occurs more than ∗ ∗ 27 6 ∗
once among the entries of the full 5 × 5-table? 6 18 ∗ ∗ 42
A) 6 B) 8 C) 9 D) 12 E) 18
A2. A palindromic number is a number that does not change when the order of
its digits is reversed, like 707 and 154451. Leon lists all 5-digit palindromic
numbers (numbers do not start with the digit 0) in increasing order.
Which number is 12th on Leon’s list?
A) 11111 B) 11211 C) 12221 D) 12321 E) 12421
A3. The figure shows a regular nonagon along with all

its diagonals. Consider all triangles whose three dif-
ferent vertices are vertices of the nonagon.
How many of these triangles are isosceles?
(An isosceles triangle is a triangle having two or three
equal sides.)
A) 27 B) 30 C) 33 D) 36 E) 39
A4.
F
G H I J
K L M
5
A4. On the previous page is a paper model of a dipiramid. The unfolded version
is depicted on the left, and the resulting folded dipiramid is depicted on
the right.
Which three points of the unfolded model correspond to vertices of the
grey triangle?
A) F , G and L B) H, I and M C) G, H and I
D) H, K and L E) F , I and K
A5. Frank has a drawer containing single socks. There are 10 red socks, the
other socks are blue. Without looking, he picks a number of socks from
the drawer. The number of socks he needs to pick to be sure of getting
at least two red socks, is twice the number he needs to pick for getting at
least two blue socks.
How many socks does the drawer contain?
A) 14 B) 18 C) 26 D) 32 E) 40
D E C
A6. Point E lies on side CD of a square ABCD. Line
segment AE is divided into four equal parts by points T U
P , R, and T . Line segment BE is divided into four
R S
equal parts by points Q, S, and U . The length of
P Q equals 3. P Q
What is the area of quadrilateral P QU T ? 3
15 17 9
A) 4 B) 4 C) 4 D) 2 E) 5
A B
A7. Carry has six cards. On each card a positive integer number is written.
Carry chooses three cards and calculates the sum of the three numbers on
these cards. Doing this for each of the 20 possible selections of three cards,
she finds the following sums: 16 (ten times) and 18 (also ten times).
What is the smallest number occurring on the cards?
A) 1 B) 2 C) 3 D) 4 E) 5
A8. A sequence of numbers starts as 27, 1, 2012, . . . The sequence has the fol-
lowing property: the numbers in positions 1, 2, and 3 add up to 2040. The
numbers in positions 2, 3, and 4 add up to 2039, the numbers in positions
3, 4, and 5 add up to 2038, and so on. More generally: the numbers in
positions k, k + 1, and k + 2 add up to 2041 − k.
What number is in position 2013?
A) −670 B) −669 C) 670 D) 1341 E) 1342
6
B-problems
B1. Consider all 5-digit numbers. Of these numbers, a numbers have the prop-
erty that the product of their digits is 25, and b numbers have the property
that the product of their digits is 15.
Determine ab .
C
B2. An equilateral triangle ABC has sides of length 12.

A semicircle with diameter AB intersects sides AC
and BC. The two circular segments outside the tri-
angle, and the part of the triangle outside of the
circle are coloured grey.
Determine the total grey area.
A B
B3. A rectangle consists of a number of cells of a regular graphing paper. The

number of these cells that touch the boundary is the same as the number
of cells that do not touch the boundary.
How many cells does the rectangle contain? Give all possible values.
B4. For all positive numbers x and y, the operation ¯ obeys the following
three rules:
1
Rule 1: (2x) ¯ y = 2 + (x ¯ y).
2 2
Rule 2: y ¯ x = x ¯ y.
Rule 3: 2 ¯ 2 = 32 .
Compute 32 ¯ 8.
7
Solutions
A-problems
× ∗∗ 7
A1. D) 12 Consider the top figure. The cell 24 56
marked by the star must contain a number that di- 36 8
vides both 27 and 6. Hence this number is either 1 ∗ 27 6
or 3. The first option implies that the cell marked 6 18 42
by the double star contains the number 27. But 36
is not divisible by 27, so this option is off. × 3 9 2 7
Using the second option, we can complete the full 8 24 72 16 56
table and arrive at the bottom figure. Clearly, 12 is 4 12 36 8 28
the largest number occurring more than once. 3 9 27 6 21
6 18 54 12 42
A2. A) 11111 A 5-digit palindromic number is precisely determined by spe-

cifying the first three digits: the last digit then equals the first digit, and
the second last digit equals the second digit. Therefore, the first twelve
5-digit palindromic numbers are:
10001, 10101, 10201, . . . , 10901, 11011, 11111.
A3. B) 30 We start by counting the number of

equilateral triangles: there are 3 of those. Next, we
count the number of isosceles triangles that are not
equilateral. Such a triangle has 9 possibilities for
its apex. For each choice of the apex, there are 3
possibilities for its base. This gives a total of 9 × 3 =
27 possibilities. Hence, in total we have 3 + 27 = 30
isosceles triangles.
A4. B) H, I and M Different letters may represent the same vertex of the
dipiramid. The five vertices are G = F = J, K = M , H, I and L. Since
H and I both have four neighbours, they represent vertices of the grey
triangle. Also K = M has four neighbours: I, H, L and G = F = J, so that
must be the third vertex of the grey triangle.
8
A5. D) 32 Denote the number of blue socks in the drawer by b. To be
sure of getting two blue socks, Frank must take at least 12 socks. Indeed,
in the worst case he will start by picking the ten red socks. To be sure of
getting at least two red socks, he must take at least b + 2 socks, because
in the worst case he will first pick all b blue socks.
Since we know that this second number is twice the first number, we find
that b + 2 = 2 × 12 = 24. This implies that b = 22. The total number of
socks is therefore 22 + 10 = 32.
A6. B) 4 Triangles EAB, EP Q, and ET U are D E C

similar since |AE| : |P E| : |T E| = 4 : 3 : 1 = |BE| :
T U
|QE| : |U E| and ∠AEB = ∠P EQ = ∠T EU (SAS).
This implies that |AB| : |P Q| : |T U | = 4 : 3 : 1. It R S
follows that |AB| = 4 and hence the area of triangle
ABE equals 12 × 4 × 4 = 8. P Q
Triangles P QE and T U E must therefore have area 3
( 34 )2 × 8 = 92 and ( 14 )2 × 8 = 12 , respectively, by sim-
A B
ilarity. It follows that the area of the quadrilateral
P QU T is equal to 92 − 12 = 4.
A7. D) 4 Because only two different outcomes occur, there can be no

more than two different numbers on the cards. Indeed, if there would be
three cards with different numbers, they would give three different out-
comes when combined with a pair of the remaining three cards.
Let us denote the two numbers on the cards by a and b. Without loss of
generality, we can assume that a occurs on at least three cards. Because
a + a + a, a + a + b, and a + b + b are different, the number b can occur
only once.
There are two cases: the case a + a + a = 16, a + a + b = 18 and the case
a + a + a = 18, a + a + b = 16. The first case is off because 16 is not a
multiple of 3 (a is an integer). Hence a = 18
3 = 6 and b = 16 − 12 = 4. We
conclude that 4 is the smallest number occurring on the cards.
A8. E) 1342 For any four consecutive terms a, b, c, d in the sequence, we

have d = a − 1. This follows from the fact that b + c + d = (a + b + c) − 1
by the given property of the sequence. In other words: skipping forward
in the sequence by three positions, decreases the current number by one.
Hence in positions 3, 6, 9, . . . , 2013 (= 3 + 3 × 670) we find the numbers
2012, 2011, . . ., (2012 − 670 =) 1342.
9
B-problems
1
B1. 2 The 5-digit numbers with product of the digits equal to 25,
consist of two fives and three ones. There are a of these numbers, where
a is the number of ways to place the three ones (because the positions of
the fives is then fixed as well).
The 5-digit numbers with product of the digits equal to 15, consist of one
three, one five, and three ones. There are b of those numbers, where b
equals the number of ways to first place the three ones, and then place the
digits 3 and 5 in one of two possible ways.
We see that b = 2a, which implies that ab = 21 .
B2. 6π Point F is the midpoint of AB (and C

the center of the semicircle). Points D and E are the
midpoints of BC and AC. They lie on the semicircle
because triangles BDF and AEF are equilateral.
Draw the arc between D and E of the circle with
center C and radius |CE| = 6. Because ∠AF E =
∠EF D = 60◦ , the circular segments on top of AE
and on top of DE are congruent. The circular seg- A B
ment below DE is congruent to the other two, because the circles around
C and F have the same radius. It follows that the three grey regions to-
gether have the same area as the circle sector CED, which has an area of
1 2
6 (π · 6 ) = 6π.
B3. 48 and 60 Let a and b denote the number of cells in the length and
width of the rectangle. We may assume that a ≥ b. Therefore, the total
number of cells in the rectangle is ab and the number of cells at the edges
equals 2a + 2b − 4. Given the fact that half of the cells are at the edge of
the rectangle, we know that ab = 2(2a + 2b − 4). Rearranging terms gives
ab−4a−4b+16 = 8. Factoring the left-hand side, we get (a−4)(b−4) = 8.
Since a and b are positive integers and a ≥ b, the only possibilities are
a − 4 = 8, b − 4 = 1 and a − 4 = 4, b − 4 = 2 (note that a − 4 and b − 4
cannot be negative because in that case b − 4 is at most −4, contradicting
the fact that b is positive). We find a = 12, b = 5 or a = 8, b = 6. This gives
two possible rectangles with 12 × 5 = 60 and 8 × 6 = 48 cells respectively.
10
11 1 2
B4. 2 Using Rule 1 we find: 32 ¯ 8 = 2 + 16 ¯ 8 = 2 + 8¯8 =
3
2 + 4 ¯ 8.
From Rule 2 with y = 2 and x = 8 it follows that : 4 ¯ 8 = 64 ¯ 2.
Repeated application of Rule 1 gives: 64 ¯ 2 = 12 +32 ¯ 2 = 22 +16 ¯ 2 =
. . . = 25 + 2 ¯ 2.
Collecting these results and using Rule 3, we obtain:
3 3 8 11
32 ¯ 8 = 2 + 4¯8 = 2 + 64 ¯ 2 = 2 + 2¯2 = 2 .
11
Faculty of Science
Nederlandstalige bachelor Wiskunde www.mastersinleiden.nl
• Fundamentele èn toegepaste wiskunde • International atmosphere

• Informele en kleine opleiding ALGANT exchange programme
• Veel keuzevrijheid • Top researchers

• Dubbele bachelors teaching advanced courses
met Natuurkunde,
• Subjects of your choice
Sterrenkunde of
no fixed programme
Informatica
www.unileidenbachelors.nl English Mathematics master
Bij ons leer je de wereld kennen

Problems
B-problems
B1. In this addition, each letter represents a digit (0 TW E E DE

to 9). Different letters represent different digits. R ONDE+
Determine the value of W × R. 2 3 0 3 1 2
B2. All 2012 camels in the Netherlands are to be distributed among 40 pastures.
No two pastures are allowed to get the same number of camels. The pasture
in the city centre of Amsterdam has to get the largest number of camels.
At least how many camels have to be placed in that pasture?
B3. One of the four dwarfs Anne, Bert, Chris and Dirk stole the king’s gold.
The dwarfs, who know each other very well, each make two statements. If
a dwarf is a liar, then at least one of these two statements is a lie. If the
dwarf is not a liar, then both statements are true.
Anne says: “Bert is a liar.” and “Chris or Dirk stole the gold.”
Bert says: “Chris is a liar.” and “Dirk or Anne stole the gold.”
Chris says: “Dirk is a liar.” and “Anne or Bert stole the gold.”
Dirk says: “Anne is a liar.” and “Bert or Chris stole the gold.”
How many of these eight statements are true?
B4. On each of the 10,000 squares of a 100×100-chess board a number is written.

Along the top row the numbers 0 to 99 are written from left to right. In
the left column the numbers 0 to 99 are written from top to bottom. The
sum of the four numbers in a 2×2-block always equals 20.
What number is written in the bottom right square of the board?
13
B5. A square ABCD with side length 8 is fol- D A C
ded in such a way that vertex A becomes

the midpoint of CD (see figure).
Find the area of the grey triangle.
C-problems
For the C-problems not only the answer is important; you also
have to describe the way you solved the problem. B
C1. You have one card with the number 12 on it. You are allowed to add new
cards to your collection according to the following rules.
• If you have a card with the number a on it, then you are allowed to
make a new card with the number 2a + 1 on it.
• If you have a card with the number b on it and b is divisible by 3,
then you are allowed to make a new card with the number 3b on it.
(a) Show that you can make a card with the number 29 on it.
(b) Show that you can make make a card with the number 22012 − 1 on
it.
(c) Show that you can never make a card with the number 100 on it.
C2. Given is a triangle ABC, a point D on line segment AC and a point E

on line segment AB. The intersection of BD and CE is called S. The
midpoint of line segment CS is called M . The line BM intersects line
segment CD in point T . Finally we are given that |BE| = |ES| = 1 and
|CD| = |DS| = 2.
Prove that |AB| = |AT |.
D
2
E 1 S 2
1 M T
B C
14
Solutions
B-problems
B1. 32 We will reason through the addition TW E E DE

from right to left. Either 2E = 12 or 2E = 2. The R ONDE+
second case is excluded because then 2D would equal 2 3 0 3 1 2
1 or 11. Therefore E = 6.
Either D = 0 or D = 5. The second case is excluded because then E + N +
1 = 13 or N = 6 would hold, but 6 is already taken by E. So D = 0 and
E + N = 13, which implies that N = 7.
From E + O + 1 = 10 it follows that O = 3. Hence W + R = 12 or
W + R = 2. The second case is excluded: W + R ≥ 1 + 2 = 3 because 0 is
already taken. Hence, W + R = 12 (and T = 1).
The pair {W, R} must be one of the following: {3, 9}, {4, 8}, and {5, 7}.
The first and last possibility are excluded because 3 and 7 are already
taken. We conclude that W · R = 8 · 4 = 32.
B2. 70 We may put 23 camels in one pasture, and put 32, 33, 34, . . . ,
70 camels in the remaining 39 pastures, respectively. The last pasture is
the one in the centre of Amsterdam. The total number of assigned camels
indeed equals 23+(32+33+· · ·+70) = 23+39· 32+70
2 = 2012. Hence, a valid
distribution having 70 camels in the pasture in the centre of Amsterdam
exists.
It is not possible to do it with less than 70. Indeed, suppose that at most
69 camels are put in the pasture in the centre of Amsterdam. Because the
number of camels is different for each pasture, the second most populated
pasture has at most 68 camels, the third most populated pasture has at
most 67 camels, and so on. In total the pastures have at most 69 + 68 +
. . . + 30 = 40 · 30+69
2 = 1980 camels, leaving at least 32 unassigned camels.
Therefore, no solution exists with fewer than 70 camels in the pasture in
the centre of Amsterdam. The minimum is therefore 70.
B3. 5 First consider the case that Anne has stolen the king’s gold.
The last statement of both Bert and Chris is true in this case, and the last
statement of both Anne and Dirk is false. Hence, Anne and Dirk are liars.
Since both of Chris’ statements are true, Chris is no liar. This implies
that Bert is a liar, because his first statement was a lie. Now that we know
who is a liar and who is not, it is easy to see that exactly five of the eight
15
statements are true.
In the cases where Bert, Chris or Dirk is the thief, a similar reasoning
holds. By the symmetry of the problem (cyclically permuting the names
‘Anne’, ‘Bert’, ‘Chris’, and ‘Dirk’ does not change the problem), exactly
five of the statements are true in each case.
B4. −178 Colour the 10,000 squares in a chess board pattern: the upper
left square and bottom right square will be white and the squares in the
upper right and bottom left will be black. Consider the 99 · 99 = 9801 2×2-
blocks. The 4901 blocks having a white square in the upper left corner are
called the white blocks and the 4900 blocks having a black square in the
upper left corner are called the black blocks. For each white block, add
the four numbers it contains. Let W be the result of adding these 4901
outcomes (some squares are counted twice!). Let Z be the result when
applying the same procedure to the 4900 black blocks. Since the number
of white blocks is one more than the number of black blocks, we have
W − Z = 20.
Now consider for each square how many times it is counted in total. Each
of the four corner squares is counted only in one white block. Each of the
other squares at the edge of the board are counted in exactly one white
block and one black block. Each of the remaining squares is counted in
exactly two white blocks and two black blocks. Considering the difference
W − Z, only the four corner squares are counted, each exactly once. For
the number x in the bottom right corner, we find that x + 0 + 99 + 99 =
W − Z = 20, hence x = −178.
B5. 2
3 Introduce points E, F , and D A C
G as in the figure. Suppose that |DE| = x.
Then |AE| = 8−x, because the square has
sides of length 8. Using the Pythagorean
theorem, we get (8−x)2 = |AE|2 = |DE|2 + E
|AD|2 = x2 + 16. Solving this quadratic
equation gives x = 3. Observe that we
F
have ∠CAF = 180◦ − ∠DAE − ∠EAB =
90◦ − ∠DAE = ∠DEA. Also, ∠ADE = B
90◦ = ∠F CA. Therefore, triangles DEA G
|AF | |EA|
and CAF are similar (AA). This implies that 14 |AF | = 5
|AC| = |DE| = 3 ,
|CF | |AD|
and hence |AF | = 20 1
3 . We also see that 4 |CF | =
4
|AC| = |DE| = 3 and so
|CF | = 3 . It follows that |BF | = 8 − |AF | = 34 . Since ∠CF A = ∠BF G
16
(vertical angles), and ∠ACF = ∠GBF = 90◦ , triangles CF A and BF G
16
are similar (AA). This implies that 34 |BG| = |BG| |AC| 4
|BF | = |CF | = 16 = 43 and
3
therefore |BG| = 1. It follows that the area of the grey triangle equals
1 1 4 2
2 · |BG| · |BF | = 2 · 1 · 3 = 3 .
C-problems
C1. (a) Applying the two rules, we can make the following sequence of cards:
12 → 4 → 9 → 3 → 7 → 15 → 31 → 63 → 21 → 43 → 87 → 29.
(b) We have seen in part (a) how to make a card with the number 3 =
22 − 1. Iterating the first rule, we can sequentially construct cards
with numbers 2 · (22 − 1) + 1 = 23 − 1, 2 · (23 − 1) + 1 = 24 − 1, and
so on. In particular, we can make the number 22012 − 1.
(c) Applying rule 1 to any card produces a card with an odd number.
Applying rule 2 to a card with an odd number (assuming it is a
multiple of three), produces a card with an odd number. As soon as
we apply rule one, the resulting card will only give rise to cards with
an odd number. To get even numbers, we must therefore restrict to
using rule 2 only (repeatedly). The only even numbers that can be
made are therefore 12 and 4.
C2. Observe that ∠ESB = ∠DSC (vertical angles). Since triangles BES
and SDC are isosceles, ∠EBS = ∠ESB = ∠DSC = ∠DCS. Hence
|BS|
triangles BES and SDC are similar (AA). In particular, |BS| = |BE| =
|SC|
|SD|= 12 |SC| = |SM |. Therefore, triangle BSM is isosceles and ∠SBM =
∠SM B = ∠T M C (vertical angles). Using that the angles of a triangle sum
to 180◦ , we find that ∠T M C = 180◦ − ∠M T C − ∠T CM and hence also
∠AT B = 180◦ − ∠M T C = ∠T M C + ∠T CM . We had already shown
that ∠T M C = ∠SBM and ∠T CM = ∠DCS = ∠ABS. We therefore see
that ∠AT B = ∠SBM + ∠ABS = ∠ABT . It follows that triangle BAT is
isosceles with top A, and therefore |AB| = |AT | holds.
17
// I DREAM OF
SOFTWARE EVEN
MY GRANDAD
UNDERSTANDS//
Make your dreams come true

Eindhoven University of Technology (TU/e) believes in people with
dreams because that’s where every innovation begins.
Create your ideal study at TU/e’s Bachelor College and make your
dreams come true!
www.tue.nl
Problems
solved the problem.
1. Let a, b, c, and d be four distinct integers.

Prove that (a − b)(a − c)(a − d)(b − c)(b − d)(c − d) is divisible by 12.
2. We number the columns of an n×n-board from 1 to n. In each cell, we place

a number. This is done in such a way that each row precisely contains the
numbers 1 to n (in some order), and also each column contains the numbers
1 to n (in some order). Next, each cell that contains a number greater than
the cell’s column number, is coloured grey. In the figure below you can see
an example for the case n = 3.
1 2 3
3 1 2
1 2 3
2 3 1
(a) Suppose that n = 5. Can the numbers be placed in such a way that
each row contains the same number of grey cells?
(b) Suppose that n = 10. Can the numbers be placed in such a way that
each row contains the same number of grey cells?
3. Determine all pairs (p, m) consisting of a prime number p and a positive

integer m, for which
p3 + m(p + 2) = m2 + p + 1
holds.
19
4. We are given an acute triangle ABC and points D on BC and E on AC
such that AD is perpendicular to BC and BE is perpendicular to AC.
The intersection of AD and BE is called H. A line through H intersects
line segment BC in P , and intersects line segment AC in Q. Furthermore,
K is a point on BE such that P K is perpendicular to BE, and L is a
point on AD such that QL is perpendicular to AD.
E
D
Q H
P
L K
A B
Prove that DK and EL are parallel.
5. The numbers 1 to 12 are arranged in a sequence. The number of ways this

can be done equals 12×11×10×· · ·×1. We impose the condition that in
the sequence there should be exactly one number that is smaller than the
number directly preceding it.
How many of the 12×11×10×· · ·×1 sequences satisfy this condition?
20
Solutions
1. In order to show that the product
n = (a − b)(a − c)(a − d)(b − c)(b − d)(c − d)
is divisible by 12, it suffices to show that it is divisible by 3 and by 4. We

consider the remainder upon division by 3 for the numbers a, b, c, and d.
Two of these numbers must have the same remainder since there are four
numbers and only three possible remainders. The difference of these two
numbers must then be divisible by 3, which implies that also n is divisible
by 3.
If among the numbers a, b, c, and d there are at least three numbers of the
same parity (all three even or all three odd ), this implies that the three
pairwise differences between these numbers are even. Hence n is divisible
by 2×2×2 = 8.
Otherwise, two of the numbers are even and the other two are odd. Both
pairwise differences are even, which shows that n is divisible by 4.
2. (a) Yes, it can be done. A possible filling of the 5×5-board containing

exactly two grey cells in each row can be seen in the figure below.
1 2 3 4 5
5 4 3 2 1
1 5 4 3 2
2 1 5 4 3
3 2 1 5 4
4 3 2 1 5
(b) Consider a 10 × 10-board that is filled in accordance to the given
rules. Every column now contains the numbers 1 to 10. In column
1, there are 9 grey cells (those containing numbers ‘2’ to ‘10’), in
column 2 there are 8 grey cells (containing ‘3’ to ‘10’), in column 3
there are 7 grey cells (containing ‘4’ to ‘10’), etc. In total, there are
9 + 8 + 7 + · · · + 0 = 45 grey cells. Because 45 is not divisible by 10,
it is impossible for all rows to have the same number of grey cells.

21
3. All terms containing a factor p are brought to the left-hand side of the
equation. In this way we obtain p3 + mp − p = m2 − 2m + 1, or
p(p2 + m − 1) = (m − 1)2 .
Note that m − 1 is non-negative by assumption. Observe that p is a divisor

of (m − 1)2 . Since p is a prime number, p must also divide m − 1. We may
write m − 1 = kp, with k a non-negative integer. Substituting this into the
previous equation gives: p(p2 + kp) = k 2 p2 . Dividing by p2 on both sides,
we find p + k = k 2 , or
p = k(k − 1).
As p is prime, one of the factors k and k − 1 must equal 1 (the case
k − 1 = −1 is excluded). The case k = 1 does not lead to a solution
because then k − 1 = 0. Hence, we must have k − 1 = 1, which gives
the only candidate for a solution k = 2, p = 2 en m = 5, and hence
(p, m) = (2, 5). It is clear that this is indeed a solution.
4. We will use similarity of certain triangles. Observe that ∠DHP = ∠LHQ

(opposite angles), and that ∠P DH = 90◦ = ∠QLH. It follows that
triangles DHP and LHQ are similar (AA). This implies that |DH| |HP |
|LH| = |HQ| .
In the same way, we can see that triangles KHP and EHQ are similar,
which implies that |KH| |HP |
|EH| = |HQ| .
Combining these two inequalities shows that |DH| |KH|

|LH| = |EH| . Because we
have ∠DHK = ∠LHE (opposite angles), this implies that triangles DHK
and LHE are similar (SAS). Similarity of these triangles implies that
∠HKD = ∠HEL, from which we may conclude (corresponding angles)
that DK and EL are parallel.
C
E
D
Q H
P
L K
A B
22
5. A sequence satisfying the condition is called ‘good’. We will determine the
number of good sequences in an indirect way, by first counting something
else.
First consider the number of ways in which we can colour the numbers
1 to 12, each number being coloured either red or blue. As there are
two options (red or blue) for each of the 12 numbers independently, there
are 212 possible colourings. We call a colouring ‘good’ if there is at least
one number of each colour, and the largest red number is larger than the
smallest blue number. There are exactly 13 colourings that are not good:
the colouring having only blue numbers, and the twelve colourings in which
the red numbers are precisely the numbers 1 to k, for some k = 1, 2, . . . , 12
(in the case k = 12, there are no blue numbers.) In total, there are
212 − 13 = 4083 good colourings.
We will now show that the number of good sequences is equal to the number
of good colourings, by perfectly matching colourings and sequences. Take a
good sequence and suppose that a is the number smaller than its immediate
predecessor in the sequence. To this sequence we associate the following
colouring: the numbers preceding a in the sequence are coloured red, and
the other numbers are coloured blue. This will be a good colouring.
As an example, consider the good sequence 1, 3, 4, 5, 8, 9, 2, 6, 7, 10, 11, 12.
Hence, a = 2. The numbers 1, 3, 4, 5, 8, and 9 will be coloured red, the
numbers 2, 6, 7, 10, 11, and 12 will be blue. This is a good colouring,
because 9 > 2.
In this way, each good colouring is obtained precisely once from a good
sequence. Indeed, for a given good colouring, we can find the unique
corresponding good sequence as follows: write down the red numbers in
increasing order, followed by the blue numbers in increasing order.
We conclude that there are 4083 sequences satisfying the condition.
23
TTENT
O RK ! ATTENT ION ! MATH A
IO ! M
NA! TMHATH AT W
ATTENTNIO AT WO
RK! AT
TENTI
ON! M
ATH A
T WO
IMC FINANCIAL MARKETS WISHES ALL

INTERNATIONAL MATHEMATICAL OLYMPIAD
PARTICIPANTS GOOD LUCK AND LOADS OF FUN!
IMC FINANCIAL MARKETS IS A SUCCESSFUL PROPRIETARY TRADING GROUP, ACTIVE ACROSS ALL
SIGNIFICANT MARKETS, EXCHANGES AND INSTRUMENTS. IT WAS FOUNDED IN AMSTERDAM IN 1989
WITH A PHILOSOPHY THAT HAS LED TO TWO DECADES OF GROWTH. AT THE CORE OF THAT PHILOSOPHY
IS A BELIEF THAT WE WILL ONLY SUCCEED BY APPLYING UNCOMPROMISING LEVELS OF EXCELLENCE AND
INNOVATION TO EVERY ASPECT OF THINKING, PLANNING, AND EXECUTION. INTEGRAL TO THIS PROCESS
HAS BEEN THE WAY IN WHICH WE HAVE EMBRACED AND PIONEERED THE TECHNOLOGY THAT HAS
TRANSFORMED THE FINANCIAL SECTOR. FOR INSTANCE, WE WERE ONE OF THE FIRST TO BE ON THE
TRADING FLOOR OF THE AMSTERDAM STOCK EXCHANGE WITH A HAND HELD COMPUTER TERMINAL.
WWW.IMC.NL
Problems
1. In quadrilateral ABCD the sides AB and CD are parallel. Let M be the

midpoint of diagonal AC. Suppose that triangles ABM and ACD have
equal area. Prove that DM k BC.
2. Consider a triple (a, b, c) of pairwise distinct positive integers satisfying

a + b + c = 2013. A step consists of replacing the triple (x, y, z) by the
triple (y + z − x, z + x − y, x + y − z). Prove that, starting from the given
triple (a, b, c), after 10 steps we obtain a triple containing at least one
negative number.
3. Find all triples (x, n, p) of positive integers x and n and primes p for which
the following holds:
x3 + 3x + 14 = 2 · pn .
4. Determine all functions f : R → R satisfying

f x + yf (x) = f xf (y) − x + f y + f (x)
for all x, y ∈ R.
5. Let ABCD be a cyclic quadrilateral for which |AD| = |BD|. Let M be the
intersection of AC and BD. Let I be the incentre (centre of the inscribed
circle) of 4BCM . Let N be the second point of intersection of AC and
the circumscribed circle of 4BM I. Prove that |AN | · |N C| = |CD| · |BN |.
25
Solutions
1. Because M is the midpoint of AC, the area of triangle ABM is equal to

the area of triangle BCM . Hence the area of triangle ABC is two times
the area of triangle ABM and hence it is also two times the area of triangle
ACD. The height of triangle ACD with respect to the basis CD is the
distance between the parallel lines AB and CD. That is also the height
of triangle ABC with respect to the basis AB. Because these heights are
equal, we have |AB| = 2 · |CD|.
Let K be the midpoint of AB. Then KM is the line segment connecting
the midpoints of AB and AC and hence KM k BC. Furthermore we have
|KB| = 21 |AB| = |CD|, hence quadrilateral KBCD has a pair of parallel
sides and a pair of sides that have the same length, which proves that it is
a parallelogram. Hence DK k BC. But this means that DK and KM are
the same line (because they both pass through K and are parallel to BC)
and this line is the line DM . Hence, DM is parallel to BC.
2. Because the sum of a, b and c is equal to 2013, we can write
(a, b, c) = (671 + u, 671 + v, 671 + w)
with u + v + w = 0. Let (ai , bi , ci ) be the triple that one obtains after

applying i steps. After one step, the first number is replaced by
a1 = b + c − a = 671 − 2u.
The difference with 671 is multiplied by -2 after one step. The same holds
for the second and third number of the triple. By using induction we get
(ai , bi , ci ) = (671 + (−2)i · u, 671 + (−2)i · v, 671 + (−2)i · w).
Because u + v + w = 0 and u, v and w must be pairwise distinct, one of

them must be negative, say that u is negative. Then
a10 = 671 + (−2)10 · u ≤ 671 − 1024 < 0.
3. The left hand side can be factorised as (x + 2)(x2 − 2x + 7). The equation
becomes
(x + 2)(x2 − 2x + 7) = 2pn .
26
First consider the case in which x is even. Then x + 2 = 2pa for a certain
integer a ≥ 0 and x2 − 2x + 7 = pn−a . (Notice that this also holds in the
case in which p = 2.) Because x is a positive integer, we have x+2 ≥ 3 and
we can rule out a = 0. Furthermore we have x2 − 3x + 5 > (x − 23 )2 ≥ 0
for all x, hence x2 − 2x + 7 > x + 2. This yields that n − a > a. We now
substitute x = 2pa − 2 in x2 − 2x + 7 = pn−a :
4p2a − 12pa + 15 = (2pa − 2)2 − 2(2pa − 2) + 7 = pn−a .
Because n − a > a, we have pa | pn−a . Furthermore pa | pa and pa |

p2a hold. Hence pa is a divisor of 15. This yields p = 3 or p = 5 and
furthermore a = 1. If p = 3, then the left hand side is 15 and the right
hand side is 3n−1 ; that equation has no solution. If p = 5, the left hand side
is 55 and the right hand side is 5n−1 ; this equation also has no solution.
Now consider the case in which x is odd. Then x + 2 = pa holds for a
certain integer a ≥ 0 and x2 − 2x + 7 = 2pn−a . Because x is a positive
integer, we have x + 2 ≥ 3 and we can rule out a = 0 again. Furthermore
x 6= 2 holds, hence also x2 − 4x + 3 = (x − 2)2 − 1 ≥ 0, hence x2 − 2x + 7 ≥
2(x + 2). Therefore we have n − a ≥ a. We substitute x = pa − 2 in
x2 − 2x + 7 = 2pn−a :
p2a − 6pa + 15 = 2pn−a = (pa − 2)2 − 2(pa − 2) + 7 = 2pn−a .
Because n − a ≥ a, we have pa | pn−a . Hence, also in this case we find

pa | 15, which yields p = 3 or p = 5 and furthermore a = 1. If p = 3, the
left hand side is 6 and the right hand side is 2 · 3n−1 , hence n = 2. We find
x = 3 − 2 = 1. Indeed, (1, 2, 3) is a solution. If p = 5, the left hand side is
10 and the right hand side is 2 · 5n−1 , hence n = 2. We find x = 5 − 2 = 3.
Indeed, (3, 2, 5) is a solution.
There are two solutions, namely (1, 2, 3) and (3, 2, 5).
4. Substituting x = y = 0 gives

f (0) = f (0) − 0 + f f (0) ,

hence f f (0) = 0. Substituting x = y = 1 gives

f 1 + f (1) = f f (1) − 1 + f 1 + f (1) ,

hence f f (1) = 1. Now substitute x = 1 and y = 0 to find

f (1) = f f (0) − 1 + f f (1) .
27

We know that f f (0) = 0 and f f (1) = 1, hence we now find f (1) = 0.
Because f f (1) = 1, this also yields that f (0) = 1. Now substitute y = 0
to get
f (x) = f (x) − x + f f (x) for all x ∈ R,

hence f f (x) = x for all x ∈ R. Substitute x = 1 to get

f (1) = f f (y) − 1 + f (y) for all x ∈ R.

Together with f (1) = 0 and f f (y) = y we now find 0 = y − 1 + f (y),
hence f (y) = 1 − y for all y ∈ R. Now we know for sure that f (x) = 1 − x
for all x ∈ R is the only possible solution to the equation. If we substitute
this in the original equation both the left and right hand side are equal to
1 − x − y + xy.
Hence this function is a solution and it is the only solution.
5. Let α = ∠DAB. Because |AD| = |BD|, we also have ∠ABD = α. By the

inscribed angle theorem we find ∠ACD = α, while the cyclic quadrilateral
theorem yields that ∠BCD = 180◦ − α. Hence ∠BCA = 180◦ − 2α. The
angle sum theorem in triangle BIM yields, together with the fact that I
is the intersection point of the angular bisectors of 4BCM , that
∠BIM = 180◦ − ∠IM B − ∠M BI = 90◦ + 90◦ − 12 ∠CM B − 21 ∠M BC
= 90◦ + 12 ∠BCM = 90◦ + 12 ∠BCA = 90◦ + 90◦ − α = 180◦ − α.
Because BIM N is a cyclic quadrilateral, this yields ∠BN M = α. The
angle sum theorem in 4BN C now yields
∠N BC = 180◦ − ∠BCN − ∠CN B = 180◦ − ∠BCA − ∠M N B
= 180◦ − (180◦ − 2α) − α = α.
This means that
∠ABN = ∠ABC − ∠N BC = ∠ABC − α = ∠ABC − ∠ABD = ∠CBD.
This fact combined with ∠N AB = ∠CAB = ∠CDB, which holds by the
inscribed angle theorem, yields 4ABN ∼ 4DBC (AA). Hence
|AN | |BN |
= ,
|CD| |CB|
or equivalently |CD| · |BN | = |AN | · |CB|. We know that ∠N BC =
α = ∠BN M = ∠BN C, hence 4BN C is isosceles with apex C, hence
|CB| = |CN |. Hence we have |CD| · |BN | = |AN | · |CN | and this is what
we wanted to prove.
28
Problems
1. Determine all 4-tuples (a, b, c, d) of real numbers satisfying the following

four equations.
ab + c + d = 3,
bc + d + a = 5,
cd + a + b = 2,
da + b + c = 6.
4n−2
2. Determine all integers n for which n+5 is the square of a rational number.
3. Fix a triangle 4ABC. Let Γ1 be the circle through B tangent to edge

AC in A. Let Γ2 be the circle through C tangent to edge AB in A.
The second intersection of Γ1 and Γ2 is denoted by D. The line AD has
second intersection E with the circumcircle of 4ABC. Show that D is the
midpoint of the segment AE.
4. Let n ≥ 3 be an integer, and consider a n × n-board, divided into n2

unit squares. For all m ≥ 1, arbitrarily many 1 × m-rectangles (type I)
and arbitrarily many m × 1-rectangles (type II) are available. We cover
the board with N such rectangles, without overlaps, and such that every
rectangle lies entirely inside the board. We require that the number of
type I rectangles used is equal to the number of type II rectangles used.
(Note that a 1 × 1-rectangle has both types.) What is the minimal value
of N for which this is possible?
5. Let a, b, and c be positive real numbers satisfying abc = 1. Show that

q
a + b + c ≥ 13 (a + 2)(b + 2)(c + 2).
29
Solutions
1. Subtracting the first equation from the second one yields

2 = 5 − 3 = (bc + d + a) − (ab + c + d) = b(c − a) + a − c = (b − 1)(c − a).
Subtracting the third equation from the fourth one yields
4 = 6 − 2 = (da + b + c) − (cd + a + b) = d(a − c) + c − a = (1 − d)(c − a).
We deduce that c − a 6= 0, and hence that 1 − d = 2(b − 1), or equivalently,
3 = 2b + d.
In the same way as above, we obtain 3 = (c−1)(b−d) from subtracting the
third equation from the second one, and 3 = (1−a)(b−d) from subtracting
the first equation from the fourth one. So c − 1 = 1 − a, or equivalently,
a + c = 2.
Now we add the first two equations. This gives
8 = ab + c + d + bc + d + a = b(a + c) + (a + c) + 2d = 2b + 2 + 2d = 5 + d,
hence d = 3. Since 2b + d = 3, it immediately follows that b = 0. Now
the first equation becomes 0 + c + 3 = 3, from which we deduce that
c = 0. Substituting this in the second equation gives 0 + 3 + a = 5, hence
a = 2. Hence the only possible solution is (a, b, c, d) = (2, 0, 0, 3), and
by substituting this in the equations, we see that this 4-tuple is indeed a
solution of the system of equations.
4n−2
2. Suppose that n+5 is the square of a rational number. Then we can
2
write this number as pq2 , where p, q are non-negative integers, q 6= 0, and
gcd(p, q) = 1. As gcd(p, q) = 1, we also have gcd(p2 , q 2 ) = 1, so there
exists a non-zero integer c 6= 0 such that 4n − 2 = cp2 and n + 5 = cq 2 .
This implies that
22 = 4(n + 5) − (4n − 2) = 4cq 2 − cp2 = c (2q)2 − p2 = c(2q − p)(2q + p).

Hence c is a divisor of 22. Note that 2 divides 22, so the right hand side
is divisible by 2. Also note that 2q − p contains a factor 2 if and only if
2q + p does; their difference 2p is even. As 22 contains exactly one factor
2, if follows that 2q − p and 2q + p cannot both be even. Hence they must
be odd, and c must be even.
As p ≥ 0 and q ≥ 1, we have 2q + p ≥ 2, which implies that the factor
11 of 22 must be factor of 2q + p. We deduce that there are only two
possibilities.
30
• c = 2, 2q + p = 11, 2q − p = 1;
• c = −2, 2q + p = 11, 2q − p = −1.
In the first case, we find 2p = 11 − 1 = 10, so p = 5 and q = 3. Hence
52
4n−2 = 50, so n = 13. Indeed, we see that 4n−2 50 25
n+5 = 18 = 9 = 32 , so n = 13
has the desired property. In the second case, we find 2p = 11 − (−1) = 12,
so p = 6, but then q is not an integer. Hence this case cannot occur. We
conclude that n = 13 is the unique integer having the desired property.
3. We consider the configuration in which D lies in the interior of 4ABC;

the proof for the other configuration is analogous. By the inscribed angle
theorem for Γ1 , we have ∠DAC = ∠DBA. By the inscribed angle theorem
for Γ2 , we have ∠DAB = ∠DCA. Hence 4ABD ∼ 4CAD (AA), from
which we deduce that
|BD| |AD|
|AD| = |CD| , (1)
or equivalently, |AD|2 = |BD||CD|.

The inscribed angle theorem now gives ∠ECB = ∠EAB = ∠DAB =
∠DCA. Hence ∠ECD = ∠ECB + ∠BCD = ∠DCA + ∠BCD = ∠BCA.
Now ∠DEC = ∠AEC = ∠ABC by the inscribed angle theorem. Hence
4CDE ∼ 4CAB (AA). Analogously, 4BDE ∼ 4BAC. We deduce that
|CD| |ED|
4CDE ∼ 4EDB, from which follows that |DE| = |DB| , or equivalently,
|DE| = |BD||CD|. Combining this with (1), we obtain |DE|2 = |AD|2 ,
2
so |DE| = |AD|, from which we deduce that D is the midpoint of AE.
4. We show that the minimal value of N is 2n − 1. First, we construct an

example by induction.
For n = 3 we can cover a 3 × 3-board by 5 rectangles, by putting a 1 × 1-
rectangle in the centre, and covering the remaining squares with four 1 × 2-
and 2 × 1-rectangles. Note that we used 3 rectangles of type I and 3 of
type II, as required.
Now let k ≥ 3, and assume that we can cover a k × k-board with 2k − 1
rectangles in a way satisfying all the conditions. Consider a (k+1)×(k+1)-
board, and cover the lower right k × k square with 2k − 1 rectangles in a
way satisfying all the conditions. Cover the top row with a 1 × (k + 1)-
rectangle, and the remaining squares with a k × 1-rectangle. We used one
more rectangle of type I, and one more of type II, so the covering obtained
still satisfies the conditions. Moreover, we used 2k − 1 + 2 = 2(k + 1) − 1
rectangles.
31
Hence we can cover a n × n-board by 2n − 1 rectangles. We now show that
any covering with N rectangles must have N ≥ 2n − 1. So fix a covering
of the n × n-board satisfying the conditions, and let N be the number of
rectangles used. Let k be the number of type I rectangles that are not of
type II. Then the number of type II rectangles that are not of type I is also
k. Furthermore, let l be the number of 1 × 1-rectangles. Then N = 2k + l.
If k ≥ n, then N ≥ 2n, in which case there is nothing to prove. Hence
assume that k < n. It then suffices to show l ≥ 2n − 2k − 1.
Every rectangle of type I can only cover squares in a single row. Hence
there are n − k rows in which no square is covered by a rectangle that is
of type I, but not of type II. Similarly, every rectangle of type II can only
cover squares in a single column. Hence there are n − k columns in which
no square is covered by a rectangle that is of type II, but not of type I.
Consider the (n − k)2 squares that lie on such a row and such a column.
They must be covered by 1 × 1-rectangles, hence l ≥ (n − k)2 .
Since (k − n + 1)2 ≥ 0, we have k 2 + n2 + 1 − 2kn + 2k − 2n ≥ 0, so
n2 − 2kn + k 2 ≥ 2n − 2k − 1. Hence l ≥ (n − k)2 ≥ 2n − 2k − 1, as desired.
We conclude that the minimal value of N is 2n − 1.
5. Note that a, b, and c are positive, so the conditions of AM-GM are satisfied
in each of the cases below.
First, apply AM-GM to a2 and 1 to get
√
a2 + 1 ≥ 2 a2 = 2a.
We repeat this for b and c, and add the resulting inequalities. This yields
a2 + b2 + c2 + 3 ≥ 2a + 2b + 2c. (2)
Next, we apply AM-GM to bc, ca, and ab, yielding

√3
bc + ca + ab ≥ 3 a2 b2 c2 = 3, (3)
using abc = 1. Moreover, again by AM-GM on a2 , b2 , and c2 , we have

√
3
a2 + b2 + c2 ≥ 3 a2 b2 c2 = 3. (4)
Now we add 2 times (2), 4 times (3), and 1 time (4) to deduce that
2a2 + 2b2 + 2c2 + 6 + 4bc + 4ca + 4ab + a2 + b2 + c2 ≥ 4a + 4b + 4c + 12 + 3.
Adding 2bc + 2ca + 2ab − 6 to this inequality then gives
3a2 + 3b2 + 3c2 + 6bc + 6ca + 6ab ≥ 2bc + 2ca + 2ab + 4a + 4b + 4c + 9.
32
Note that the left hand side is equal to 3(a2 + b2 + c2 + 2bc + 2ca + 2ab) =
3(a + b + c)2 . For the right hand side, note that 9 = 8 + abc. Hence the
right hand side is equal to (a + 2)(b + 2)(c + 2). Both sides are positive, so
dividing by 3 and taking square roots gives us
q
a + b + c ≥ 13 (a + 2)(b + 2)(c + 2).
33
Optimize your career
ORTEC addresses complex optimization issues in various sectors worldwide. Our employees
assist customers in making well-founded decisions based on mathematical models and the use of
simulation and optimization techniques.
ORTEC is a young, professional organization with great career opportunities. You can work
for ORTEC during or after your studies. You will be involved in projects immediately and given
considerable personal responsibility. We provide a working environment with plenty of room to
develop your talents within your own line of interest, both in the Netherlands and internationally.
Do you like the sound of this? Are you currently pursuing, or have you previously completed a
degree in Econometrics, Operational Research, Computer Science or Mathematics? Are you good
with statistical models and do you feel at home in the world of logistics or finance? Then ORTEC
is the place to be!
ORTEC Logistics
Houtsingel 5
2719 EA Zoetermeer Tell us how your talents will improve our products and services, and how you can help ORTEC
Tel.: 088-678 32 65 continue its international growth. Please visit our website www.ortec.com/carriere for more
088-ORTECNL
recruitment@ortec.com information about a career at ORTEC and an up-to-date overview of our vacancies, internships and
graduation projects. If you do not find your ideal job or subject, please send an open application.
www.ortec.com
Problems
1. Show that
2013
X 4026!
2
n=0 n!(2013 − n)!
is the square of an integer.
2. Let P be the intersection of the diagonals of a convex quadrilateral ABCD.

Let X, Y , and Z be points on the interior of AB, BC, and CD, respect-
ively, such that
|AX| |BY | |CZ|
|XB| = |Y C| = |ZD| = 2.
Suppose moreover that XY is tangent to the circumcircle of 4CY Z and

that Y Z is tangent to the circumcircle of 4BXY . Show that ∠AP D =
∠XY Z.
3. Fix a sequence a1 , a2 , a3 , . . . of integers satisfying the following condition:

for all prime numbers p and all positive integers k, we have
apk+1 = pak − 3ap + 13.
Determine all possible values of a2013 .
4. Determine all positive integers n ≥ 2 satisfying
i + j ≡ ni + nj

(mod 2)
for all i and j such that 0 ≤ i ≤ j ≤ n.
5. Let ABCDEF be a cyclic hexagon satisfying AB ⊥ BD and |BC| = |EF |.

Let P be the intersection of BC and AD, and let Q be the intersection of
EF and AD. Assume that P and Q are on the same side of D, and that
A is on the opposite side. Let S be the midpoint of AD. Let K and L be
the centres of the incircles of 4BP S and 4EQS, respectively. Prove that
∠KDL = 90◦ .
35
Solutions
1. We prove the following more general statement.

m
X (2m)! 2m 2

2 = m . (5)
n=0 n!(m − n)!
We have
(2m)! (m!)2 (2m)! m 2
2m

2 = 2 · = n · m .
n!(m − n)! n!(m − n)! (m!)2
Hence it suffices to show that

m
m 2
X
2m

n = m .
n=0
We will do this combinatorially. Consider 2m balls, numbered from 1 up

to 2m. Balls 1 up to m are coloured blue, and balls m + 1 up to 2m are
coloured red. We can choose m balls from these 2m balls in 2m m ways.
On the other hand, we can also first choose n blue balls, with 0 ≤ n ≤ m,
and then choose m − n red balls. Equivalently, we can choose n blue balls
to include, and n red balls to not include. Hence the number of ways in
which one can choose m balls is also equal to
m
m 2
X
n .
n=0
2m

Hence this sum is equal to n . This proves (5).
2. By the inscribed angle theorem, we have ∠CZY = ∠BY X and ∠BXY =

∠CY Z. It follows from this that
∠XY Z = 180◦ − ∠BY X − ∠CY Z = 180◦ − ∠BY X − ∠BXY = ∠ABC.
Moreover, we see that 4XBY ∼ 4Y CZ (AA). This implies that
|XB| |Y C|
|BY | = |CZ| .
From the assumptions, it immediately follows that |XB| = 13 |AB|, |BY | =

2 1 2
3 |BC|, |Y C| = 3 |BC|, and |CZ| = 3 |CD|. Substituting this yields
1 1
3 |AB| 3 |BC|
2 = 2 ,
3 |BC| 3 |CD|
36
|AB| |BC|
hence |BC| = |CD| . As 4XBY ∼ 4Y CZ, we also have ∠ABC =
∠XBY = ∠Y CZ = ∠BCD. Hence 4ABC ∼ 4BCD (SAS). We de-
duce that ∠CAB = ∠DBC. From this, it follows that
∠P AB + ∠ABP = ∠CAB + ∠ABD = ∠DBC + ∠ABD = ∠ABC.
We established before that ∠ABC = ∠XY Z, so by the external angle

theorem applied to 4ABP , we have
∠BP C = ∠P AB + ∠ABP = ∠ABC = ∠XY Z.
As ∠BP C = ∠AP D, it follows that ∠AP D = ∠XY Z.
3. Let q and t be primes. Substituting k = q and p = t yields
aqt+1 = taq − 3at + 13.
Substituting k = t and p = q yields
aqt+1 = qat − 3aq + 13.
Hence the right hand sides are equal, so
taq − 3at = qat − 3aq ,
or equivalently,
(t + 3)aq = (q + 3)at .
In particular, we have 5a3 = 6a2 and 5a7 = 10a2 . Substituting k = 3 and
p = 2 now gives
6
a7 = 2a3 − 3a2 + 13 = 2 · a2 − 3a2 + 13.
5
10
As a7 = 5 a2 , this implies that
13
a2 = 13,
5
(p+3)a2
hence a2 = 5. We deduce that for all primes p, we have ap = 5 =
p + 3.
Substituting k = 4 and p = 3 gives
a13 = 3a4 − 3a3 + 13.
As a13 = 16 and a3 = 6, it follows that 3a4 = 21, or equivalently, a4 = 7.
37
Finally, substitute k = 4 and p = 503 to get
a2013 = a4·503+1 = 503 · a4 − 3a503 + 13 = 503 · 7 − 3 · (503 + 3) + 13 = 2016.
Hence a2013 = 2016, so this is the only possible value.

It remains to check that this value is attained for some sequence of integers,
i.e. there exists a sequence satisfying the condition. Define the sequence
a1 , a2 , a3 , . . . by an = n + 3 for all n ≥ 1. Then for all primes p and all
positive integers k, we have
apk+1 = pk+4 = pk+3p−3p+4 = (pk+3p)−3p−9+9+4 = pak −3ap +13,
as desired.
4. We first show that n satisfies the condition if and only if ni ≡ i+1 mod 2

for all i such that0 ≤ i ≤ n. Suppose that ni ≡ i+1 mod 2 for all i, then
we have ni + nj ≡ i + 1 + j + 1 ≡ i + j mod 2 for all i and j. Hence n

satisfies
the condition. Conversely,
suppose that n satisfies the condition.
As n0 = 1, we have i ≡ 1 + ni mod 2 for all i, so ni ≡ i − 1 ≡ i + 1
mod 2 for all i.
Write n as n = 2k + m with 0 ≤ m < 2k . Since n ≥ 2, we may assume
that k ≥ 1. Consider
k
(2k + m)(2k + m − 1) · · · (m + 4)(m + 3)

n 2 +m
k
= k
= .
2 −2 2 −2 (2k − 2)(2k − 3) · · · 2 · 1
k k
The product in the denominator has b 2 2−2 c factors divisible by 2, b 2 4−2 c
k
−2
factors divisible by 4, . . . , b 22k−1 c factors divisible by 2k−1 and no factors
divisible by 2 . The product in the numerator consists of 2k −2 consecutive
k
k k
factors, hence has at least b 2 2−2 c factors divisible by 2, at least b 2 4−2 c
k
−2
factors divisible by 4, . . . , at least b 22k−1 c factors divisible by 2k−1 . We
deduce that the number of factors 2 of the product in the numerator is
at least that of the one in the denominator. If 2k occurs as factor of the
numerator, then the number of factors 2 of the product in the numerator
is greater than that of the one in the denominator. This is the case if
m + 3 ≤ 2k , i.e. if m ≤ 2k − 3. So if m ≤ 2k − 3, then 2kn−2 is even,
whereas 2k − 2 is even as well. Hence n = 2k + m does not satisfy the
condition.
We deduce that n can only satisfy the condition if there exists a k ≥ 2
such that 2k − 2 ≤ n ≤ 2k − 1. If n is odd, then n0 + n1 = 1 + n ≡ 0

mod 2, so n does not satisfy the condition. This implies that n can only
38
satisfy the condition if n is of the form n = 2k − 2 with k ≥ 2 is. Suppose
that n is of that form. We show that n satisfies the condition. We have
the following.
k
(2k − 2)(2k − 3) · · · (2k − c)(2k − c − 1)

2 −2
= .
c c · (c − 1) · · · 2 · 1
Note that the number of factors 2 in 2k −i is equal to the number of factors

2 in i for 1 ≤ i ≤ 2k − 1. Hence the number of factors 2 in the numerator
is equal to the number of factors 2 in the product (c + 1) · c · (c − 1) · · · 2, or
equivalently, the number of factors 2 in the denominator plus the number
k
of factors 2 in c + 1. We deduce that 2 c−2 is even if and only if c is odd.

Hence n = 2k − 2 satisfies the condition.

We conclude that n satisfies the condition if and only if n is of the form
2k − 2 with k ≥ 2.
5. The configuration is fixed by the order in which the points A up to F occur

on the circle, and by the condition on the points P and Q on AD. So there
is no need to distinguish between different configurations. (The positions
of P and Q with respect to one another are irrelevant.)
First note that S is the centre of the circumcircle of ABCDEF , as AD is its
diameter, since ∠ABD = 90◦ , and S is the midpoint of AD. We will show
that ∠KDS = ∠KBS. As KS is the angle bisector of ∠BSD, we have
∠BSK = ∠KSD. Furthermore, |SD| = |SB|, as the segments SD and
SB are both radii of the circumcircle of ABCDEF . Since |SK| = |SK|,
we have 4DSK ∼ = 4BSK (SAS). Hence ∠KDS = ∠KBS.
As BK is the angle bisector of ∠CBS, we have ∠KBS = ∠CBK =
1 1
2 ∠CBS. Hence ∠KDS = 2 ∠CBS.
Analogously, we find ∠LDS = 12 ∠QES, which implies that ∠LDS = 21 ·
(180◦ −∠F ES) = 90◦ − 12 ∠F ES. Hence ∠LDS+∠KDS = 90◦ − 12 ∠F ES+
1 ∼
2 ∠CBS. Now note that 4SBC = 4SEF (SSS), so ∠F ES = ∠CBS. We
conclude that ∠KDL = ∠LDS + ∠KDS = 90◦ .
39
Problems
Part 1
1. A bag contains red and blue marbles. Of these marbles 53 is blue, the rest
is red.
If we double the number of red marbles in the bag, what portion of the
marbles will be blue?
1 3 4 3 4
A) 5 B) 7 C) 7 D) 5 E) 5
2. Alyssa sums odd numbers. First she takes 1, then she calculates 1 + 3,
next she calculates 1 + 3 + 5, then 1 + 3 + 5 + 7, etc. In this way she obtains
a very long list of numbers. The largest number that she calculates is
1 + 3 + 5 + 7 + · · · + 197 + 199.
How many of the numbers on Alyssa’s list end with a 4?
A) 9 B) 10 C) 19 D) 20 E) 50
3. We have got a dice and we will colour its pips. For each of the six faces we
choose one of the colours red, white and blue and colour all pips on that
face in that colour. There is one rule: in each vertex we must choose three
different colours for the three adjacent faces.
In how many ways can we colour the dice?
A) 3 B) 6 C) 12 D) 27 E) 36
4. The grey rectangle in the figure has width 8. Around

it is a ring consisting of squares of two different sizes;
see the figure.
What is the height of the grey rectangle?
28 19 48
A) 9 B) 3 C) 2 D) 5 E) 10
40
5. Ans, Ben, Carla and Dirk participate to a lottery. In a hat there are eight
cards with the numbers 1 to 8. One by one each person takes two cards
out of the hat such that Dirk takes the last two cards. To make it even
more exciting, everybody adds his or her two numbers and tells the result
to the others. For Ans this is 10, for Ben 14 and for Carla 5.
What are the numbers that are on Dirk’s cards?
A) 1 and 6 B) 1 and 7 C) 2 and 5 D) 2 and 6 E) 3 and 4
6. On a 3×4-board there are eight pieces with the num-

bers 1 to 8, see the figure. Pieces may be captured
horizontally and vertically. I.e.: if a piece in the
8 7 6 5
horizontal or vertical direction borders on an empty
square on the one side and on an occupied square on
1 2 3 4
the other side, then the piece on the occupied square
may jump to the empty square. The piece on the middle square is cap-
tured and removed from the board. Jan keeps capturing pieces until there
is only one piece left on the board.
The only possibilities for this last piece are:
A) 1 and 2 B) 3 and 4 C) 2 and 3 D) 1 and 4 E) 1, 2, 3 and 4
7. How many zeros does the number that is the result of the division
1010101010101010101 : 101
contain?
A) 8 B) 9 C) 10 D) 11 E) 12
8. Jan and Katrijn live on a straight dike, exactly 11,7 kilometres from each
other. Jan cycles to Katrijn. On each place on the road he can see how
far he is from his own house and how far is he from Katrijn’s house. Both
distances we round to whole kilometres. On some part of the dike this
rounded distances are equal.
How many kilometres long is this part?
A) 0,1 B) 0,3 C) 0,5 D) 0,7 E) 0,9
41
9. Anne draws a triangle of which the sides have have different lengths and
the shortest side is 6 cm long. Bert redraws Anne’s triangle, but he draws it
twice as big (all sides twice as long). Christa also redraws Anne’s triangle,
but she draws it three times as big. It appears that Anne’s triangle has
a side that has the same length as one of the sides of Bert’s triangle and
that Bert’s triangle has a side that has the same length as one of the sides
of Christa’s triangle.
How many centimetres long is the longest side of Anne’s triangle?
A) 12 B) 18 C) 24 D) 30 E) 36
10. We calculate a recurrent average of the numbers 1, 2, 3, 4 and 5 in the

following way. First take the average of two of these numbers. Take the
average of this result and a third numbers. Then take the average of this
next result with a fourth number. Finally take the average of this result
with the last number.
What is the difference between the largest and smallest result we may
obtain in this way?
3 31 17 65
A) 0 B) 2 C) 16 D) 8 E) 16
a b c
11. Of a 3×3-board with squares a1 till c3 three squares
are painted black. This is done in such a way that 1
no two black fields are neighbouring fields (i.e. shar-
2
ing an edge). In the figure there is one way to do
this.
3
In how many ways can this be done?
A) 14 B) 18 C) 20 D) 22 E) 30
12. While Snow White is cooking, the seven dwarfs play chess. Doc plays
against all other six dwarfs. Happy plays against five others, Grumpy
against four others, Sneezy against three others, Bashful again two others
and Sleepy plays against one other dwarf, after which he falls asleep.
Against how many dwarfs does Dopey, the seventh dwarf, play?
A) 0 B) 1 C) 3 D) 5 E) 6
42
13. In a regular hexagon ABCDEF the diagonals are C B
drawn as in the figure. The area of the hexagon
ABCDEF is 1.
What is the area of the grey hexagon?
D A
1 1 1 3 5
A) 2 B) 3 C) 4 D) 8 E) 16
E F
14. Peter has some numbers of which the average is 20. If he leaves out the
smallest number, the average of the other numbers will be 22. If he leaves
out the biggest number, then the average of the others is 13. If he leaves
out both the smallest and the biggest number, then the average becomes
14.
How many numbers does Peter have?
A) 4 B) 5 C) 6 D) 7 E) 8
15. Ann, Bo, Cas, Dex and Eva cycle in the same direction on a long road,
each with a constant speed. Their speeds are different and for each pair
there is a moment when they cycle exactly next to each other and the one
takes over the other. After the trip Ann tells: “First I took over Bob, then
I got taken over by Cas. Then I took over Dex and finally I got taken over
by Eva.”
Bo tells: “I got taken over by Eva. That was after the moment when I
cycled next to Ann, but before the moment I cycled next to Cas or Dex.”
Put the five in order from fastest to slowest.
A) Eva, Cas, Ann, Bo, Dex D) Cas, Eva, Ann, Dex, Bo
B) Cas, Eva, Ann, Bo, Dex E) This cannot be determined
C) Eva, Cas, Ann, Dex, Bo
43
Part 2
1. A long chain hangs over the pulley of a crane, see

the figure. On one side of the chain a weight is at-
tached. The length of the part of the chain between
the pulley and the weight is exactly one fourth of the
length of the chain between the pulley and the other
end (the size of the pulley itself is negligible). The
chain is hanging exactly in balance. The chain and
the weight together weigh 320 kilogram.
How many kilograms does the chain weigh?
2. Huey, Dewey and Louie are going to run. They start at the same time
and each of them walks at constant speed. When Huey finished three laps,
Dewey is exactly halfway his third lap. When Dewey finished his third
lap, Louie is exactly halfway his third lap. After a while Huey, Dewey and
Louie are together at the start at the same moment.
How many laps did Louie finish at this moment?
3. What is the largest number of 2×2×1-bricks that one can put in a box of
7×7×6? The bricks must be aligned parallel to the sides of the box.
4. Anne is completely broke, but today (day 1) she got a new job. She works
every day (even during the weekends) and after every shift, she is always
paid 13 euro. After every shift, she spends the money on groceries. On
the first day Anne spends 1 euro on groceries, on the second day 2 euro,
on the third day 3 euro, etc.
What is the first day on which Anne cannot afford her groceries anymore?
5. Ronald has bought a large barrel of water. The first day of the year, 1
January 2012, he uses half of it. The second day he uses a third of the
remaining water, and on the third day a fourth of the then remaining
water, and so on until the 365th day of the year. On the last day of the
year (2012 has 366 days) he has only one litre of water left and he drinks
it.
How many litres of water did the barrel contain when Ronald bought it?
44
6. To a three digit number you add its three digits. For example, the number
216 becomes 216 + 2 + 1 + 6 = 255.
What is the largest three digit number that you cannot make in this way?
7. Four students made a multiple choice test consisting of four questions.

For each question they could choose from the answers A, B, C, D and E.
The first student answered DDAE, the second CBAD, the third CDAC
and the fourth answered BBCC. Unfortunately each of them had only two
questions right.
What are the four right answers?
8. To a two digit number we apply the following recipe:
• multiply the number with itself,

• then subtract 6 times the initial number from this result and
• finally add 9 to that result.
For example with 41 you get: 41×41 = 1681, 1681 − 6×41 = 1681 − 246 =
1435, 1435 + 9 = 1444. If the result is equal to the initial number, but
then with the two digits interchanged, followed by two zeros, then we call
the initial number great. For example, if the result for 41 would have been
equal to 1400, then 41 would have been great. There is exactly one great
two digit number.
What is that number?
9. In each of the nine circles a number

must be placed. It must be done in 7 10
such a way that when you add the three
numbers on a line, you would get the
same answer for each line. In two circles
a number already has been placed.
What is the number that must be put ?
on the place of the question mark?
10. A mathematical landscape gardener makes a geometrical design for a big

garden. On the big land he places two poles at a distance of 20 metres
from each other. On each place where the distance to each of the poles is
an integer number of metres that is at most 14, he plants a box-tree.
How many box-trees does he plant?
45
Solutions
Part 1
3
1. B) 7 6. E) 1, 2, 3 and 4 11. D) 22
2. D) 20 7. E) 12 12. C) 3
1
3. B) 6 8. D) 0,7 13. B) 3
48
4. D) 5 9. A) 12 14. D) 7
17
5. C) 2 and 5 10. D) 8 15. B) C, E, A, B, D
Part 2
1. 200 6. 995
2. 25 7. DBAC
3. 72 8. 63
4. 26 9. 10
5. 366 10. 81
46
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
OmslagNWO2013-v1.indd 2 14-6-2013 14:56:56

52nd Dutch Mathematical Olympiad 2013
and the team selection for IMO 2014 South Africa
MY

CY
We eat problems
CMY

for breakfast.

in Nederland nd
gehouden: IMO2011
Olympiad 2013
International
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

Introduction
The selection process for IMO 2014 started with the first round in January
2013, held at the participating schools. The paper consisted of eight mul-
tiple choice questions and four open questions, to be solved within 2 hours.
In total 32% more students than in 2012 participated in this first round:
to be precise: 7424 students of 283 secondary schools.
Those 800 students from grade 5 (4, ≤ 3) that scored 24 (21, 18) points or
more on the first round (out of a maximum of 36 points) were invited to
the second round, which was held in March at twelve universities in the
country. This round contained five open questions, and two problems for
which the students had to give extensive solutions and proofs. The contest
lasted 2.5 hours.
Those students from grade 5 (4, ≤ 3) that scored 26 (24, 21) points or
the universities, in order to prepare them for their participation in the final
round.
concluded its 52nd edition 2013.

2013 were invited to an intensive seven-month training programme. The
students met twice for a three-day training camp, three times for a single
day, and finally for a six-day training camp in the beginning of June. Also,
they worked on weekly problem sets under supervision of a personal trainer.
1
girls, as this year we participated for the third time in the European Girls’
Mathematical Olympiad (EGMO). In total there were eight girls competing
to be in the EGMO team. The team of four girls was selected by a selection
test, held on 21 March 2014. They attended the EGMO in Antalya, Turkey
from 10 until 16 April, and the team returned with a gold and a bronze
medal. For more information about the EGMO (including the 2014 paper),
see www.egmo.org.
The same selection test was used to determine the ten students particip-
ating in the Benelux Mathematical Olympiad (BxMO), held in Brugge,
Belgium, from 2 until 4 May. The Dutch team managed to come first in
the country ranking, and received two bronze medals, two silver medals
and two gold medals. For more information about the BxMO (including
the 2014 paper), see www.bxmo.org.
as an observer C. The team had a training camp in Cape Town, from 28
June until 6 July.
of two one-hour parts, one with eight multiple choice questions and one with
eight open questions. The goal of this Junior Mathematical Olympiad is
to scout talent and to stimulate them to participate in the first round of
the Dutch Mathematical Olympiad.
the solutions.
2
Dutch delegation
The Dutch team for IMO 2014 in South Africa consists of
Tysger Boelens (18 years old)

– bronze medal at BxMO 2013, gold medal at BxMO 2014
Peter Gerlagh (17 years old)
– bronze medal at BxMO 2011, honourable mention at BxMO
2012, gold medal at BxMO 2013
Matthew Maat (14 years old)
– bronze medal at BxMO 2014
Michelle Sweering (17 years old)
– bronze medal at EGMO 2012, silver medal at EGMO 2013, gold
medal at EGMO 2014
– honourable mention at IMO 2012, bronze medal at IMO 2013
Bas Verseveldt (17 years old)
– silver medal at BxMO 2012, bronze medal at BxMO 2013, gold
medal at BxMO 2014
Jeroen Winkel (17 years old)
– observer C at IMO 2011, bronze medal at IMO 2012, silver medal
at IMO 2013
Bob Zwetsloot (16 years old)

Quintijn Puite (team leader), Eindhoven University of Technology

Birgit van Dalen (deputy leader), Leiden University
Julian Lyczak (observer B), Utrecht University
3
Problems
A-problems
A1. A traffic light is alternately green and red. The periods green and red
are equally long and always of the same length: either 1, 2, or 3 minutes.
There are four combinations for the colour of the light at 12:08 pm and at
12:09 pm: red–red, red–green, green–red, and green–green.
How many of these four combinations are possible, given that the light is
red at 12:05 pm and also red at 12:12 pm?
A) 1 B) 2 C) 3
D) 4 E) The light cannot be red at both times.
D C
A2. The rectangle ABCD is divided into five equal rect-
angles. The perimeter of each of these small rect-
angles is 20.
What is the area of rectangle ABCD?
A) 72 B) 112 C) 120 D) 140 E) 150
A B
A3. The numbers a, b, c, d and e satisfy:
a + b + 1 = b + c − 2 = c + d + 3 = d + e − 4 = e + a + 5.
Which is the largest of these five numbers?

A) a B) b C) c D) d E) e
A4. Nine light bulbs are put in a square formation. Each

bulb can be either on or off. We can make a move
by pressing a bulb. Then, the pressed bulb and the
bulbs in the same row or column change their state
from on to off or vice versa. Initially, all light bulbs
are on.
What is the minimum number of moves needed to
turn off all the light bulbs?
A) 3 B) 4 C) 5 D) 9 E) This is impossible.
4
A5. Out of a shipment of boxes, one fourth is empty. We open one fourth of
all boxes and notice that one fifth of them is non-empty.
Which part of the unopened boxes is empty?
4 1 1 1 1
A) 15
B) 4
C) 15
D) 16
E) 20
A6. A regular hexagon and an equilateral triangle have the same perimeter.
What is the ratio area hexagon : area triangle?
A) 2 ∶ 3 B) 1 ∶ 1 C) 4 ∶ 3 D) 3 ∶ 2 E) 2 ∶ 1
A7. What are the last four digits of 52013 ?

A) 0625 B) 2525 C) 3125 D) 5625 E) 8125
A8. Twenty students did a test. No two students answered the same number
of questions correctly. Each question was answered correctly by at most
three students.
What is the smallest number of questions that the test could have had?
A) 63 B) 64 C) 67 D) 70 E) 71
5
B-problems
B1. What is the smallest positive integer consisting of the digits 2, 4 and 8,
such that each digit occurs at least twice and the number is not divisible
by 4?
D C
B2. A rectangle ABCD has sides of length a and b, where 4
a < b. The lines through A and C perpendicular
5
to the diagonal BD divide the diagonal into three a
segments of lengths 4, 5, and 4. 4
Calculate ab .
A b B
B3. A bus calls at three stops. The middle bus stop is equally far from the
first stop as from the last stop. Fred, standing at the middle bus stop, has
to wait for 15 minutes for the bus to arrive. If he cycles to the first stop,
he will arrive there at the same time as the bus. If instead he runs to the
last stop, he will also arrive there at the same moment as the bus.
How long would it take Fred to cycle to the last stop and then run back
to the middle stop?
B4. We write down the numbers from 1 to 30 000 one after the other to form
a long string of digits:
123456789101112 . . . 30000.
How many times does 2013 occur in this sequence?
6
Solutions
A-problems
A1. B) 2 It is given that the light is red at 12:05 pm. The colours at
times 12:05 pm to 12:12 pm are now fixed for a traffic light of period 1:
they are alternately red and green. When the period is 2, there are two
possibilities and for period 3 there are three possibilities:
period 12:05 12:06 12:07 12:08 12:09 12:10 12:11 12:12
1 min red green red green red green red green
2 min red red green green red red green green
2 min red green green red red green green red
3 min red red red green green green red red
3 min red red green green green red red red
3 min red green green green red red red green
In three of the six cases, the light is red at 12:12 pm, as required. This
gives us two colour combinations for the light at 12:08 pm and at 12:09
pm: red–red and green-green.
A2. C) 120 Twice the length of a small rectangle equals three times its
width. Therefore, the ratio between length and width equals 3 ∶ 2. As the
perimeter is 20, the length must be 6 and the width must be 4. We see
that the area of each small rectangle equals 6 × 4 = 24, hence the area of
rectangle ABCD is 5 × 24 = 120.
A3. B) b Comparing sums of pairs of numbers, we find:
e + a < c + d < a + b < b + c < d + e.
Every sum of four of the numbers can be obtained by adding two of the
pairs. For example, a+b+c+e is equal to (e+a)+(b+c). Of all four-tuples,
a, c, d, e has the smallest sum, because a + c + d + e = (e + a) + (c + d) is the
sum of the two smallest pairs. The remaining number b must be the largest
among the five numbers. Indeed, the largest number is the one for which
the remaining four numbers have the smallest sum.
A4. A) 3 The order in which the moves are made is irrelevant for the
final result. By pressing the three bulbs in the top row, all bulbs will change
7
from being on to being off. Indeed, the bulbs in the top row change their
state three times, and the other bulbs change their state exactly once.
It is not possible to turn off all light bulbs by pressing two or fewer bulbs,
because some bulb will not be in the same row or column as the chosen
bulbs, and hence remain on.
1
A5. C) 15 Without changing the problem, we may assume that there
are 20 boxes. Hence, a total of 5 boxes is empty. Out of the 5 boxes
that are opened, one fifth turns out to be non-empty, exactly 1 box. As
4 of the opened boxes are empty, there is exactly 1 empty box among the
remaining 15 unopened boxes.
A6. D) 3 ∶ 2 We divide the hexagon into 6 equal equilat-

eral triangles and divide the triangle into 4 equal equilateral
triangles, see the figure. Since the hexagon and the triangle
have the same perimeter, the sides of the triangle are twice
as long as the hexagon’s sides. Therefore, the triangles in
both divisions have the same size. It follows that the ra-
tio between the area of the hexagon and the area of the
triangle equals 6 ∶ 4, or: 3 ∶ 2.
A7. C) 3125 We consider the last four digits of the powers of 5:
51 = 0005 55 = 3125
52 = 0025 56 = 1 5625
53 = 0125 57 = 7 8125
54 = 0625 58 = 39 0625
The last four digits of a power of 5 are already determined by the last
four digits of the previous power of 5. For example: the last four digits of
5 × 390625 and 5 × 0625 are both 3125. Because the last four digits of 54
and 58 are the same, the last four digits of powers of 5 will repeat every
four steps. The last four digits of 52013 will be the same as those of 52009
and those of 52005 , continuing all the way down to the last four digits of
55 = 3125. We conclude that 3125 are the last four digits of 52013 .
A8. B) 64 Since every student answered correctly a different number of

questions, at least 0 + 1 + 2 + ⋯ + 19 = 190 correct answers were given in
8
total. Because every question was answered correctly at most three times,
there must be at least 190
3
= 63 31 questions. That is, there were at least 64
questions.
B-problems
3
B2. 2
Triangles ADF , BDA en BAF are
similar (AA). Hence
D C
b ∣BA∣ ∣AF ∣ ∣BF ∣
= = = .
a ∣DA∣ ∣DF ∣ ∣AF ∣ F
a
Therefore
b 2 ∣AF ∣ ∣BF ∣ ∣BF ∣ 9
( ) = ⋅ = = .
a ∣DF ∣ ∣AF ∣ ∣DF ∣ 4 A b B
We conclude that b
a
= 32 .
B3. 30 minutes The time needed by Fred to bike from the middle stop to the
last stop and run back to the middle stop, is equal to the time he needs to
bike from the middle stop to the first stop plus the time he needs to run
from the middle stop to the last stop. This is because the distance to both
stops is the same.
It is given that this amount of time equals the time the bus needs to get to
the first stop plus the time it needs to get to the last stop. This is exactly
twice the time the bus needs to get to the middle stop: 2 × 15 = 30 minutes.
B4. 25 times The combination of digits “2013” occurs 13 times as part

of the following numbers: 2013, 12013, 22013, and 20130 to 20139. In
addition, “2013” also occurs as the end of one number followed by the
beginning of the next number. The different possibilities are:
2∣013 does not occur, since no number starts with digit ‘0’.
20∣13 occurs 11 times: 1320∣1321 and 13020∣13021 to 13920∣13921.
201∣3 occurs only once: 3201∣3202, because no numbers larger than 30 000
were written down.
It is easy to verify that “2013” does not occur as a combination of three
consecutive numbers. Therefore, “2013” occurs a total of 13 + 11 + 1 = 25
times in the sequence of digits.
9
Problems
B-problems
B1. A number of students took a test for which the maximum possible score
was 100 points. Everyone had a score of at least 60 points. Exactly five
students scored the maximum of 100 points. The average score among the
students was 76 points.
What is the minimum number of students that could have taken the test?
D C
B2. In the figure, a square ABCD of side length 4 is

given. Inside the square, two semicircles with dia-
meters AB and BC are drawn.
Determine the combined area of the two grey shapes.
A B
B3. Consider two clocks, like the ones in the figure
12 12
on the right, whose hands move at a constant 11 1 11 1
10 2 10 2
speed. Both clocks are defective; the hands
9 3 9 3
of the first clock turn at a pace that is 1%
8 4 8 4
faster than it should be, while the hands of 7 5 7 5
6 6
the second clock turn at a pace that is 5% too
fast. At a certain moment, both clocks show a time of exactly 2 o’clock.
Some time passes until both clocks again show exactly the same time.
At that moment, what time do the clocks show?
B4. The number square on the right is filled with positive 1

2 32 A B
numbers. The product of the numbers in each row,
C 2 8 2
in each column, and in each of the two diagonals is
always the same. 4 1 D E
What number is H? F G H 16
10
B5. A regular hexagon is divided into seven parts by lines
parallel to its sides, see the figure. Four of those 11
pieces are equilateral triangles, whose side lengths

are indicated in the figure.
What is the side length of the regular hexagon? 16
9
C-problems
For the C-problems not only the answer is important; you also 5
have to describe the way you solved the problem.
C1. We say a positive n-digit number (n ≥ 3 and n ≤ 9) is above average if it

has the following two properties:
the number contains each digit from 1 to n exactly once;
for each digit, except the first two, the following holds: twice the digit
is at least the sum of the two preceding digits.
For example, 31254 is above average because it consists of the digits 1 to

5 (each exactly once) and also
2 ⋅ 2 ≥ 3 + 1, 2 ⋅ 5 ≥ 1 + 2, and 2 ⋅ 4 ≥ 2 + 5.
(a) Give a 4-digit number that is above average and has ‘4’ as its first
digit.
(b) Show that no 4-digit number that is above average has ‘4’ as its second
digit.
(c) For 7-digit numbers that are above average, determine all possible
positions of the digit ‘7’.
C2. We will call a triple (x, y, z) good if x, y, and z are positive integers such
that y ≥ 2 and the equation x2 − 3y 2 = z 2 − 3 holds.
An example of a good triple is (19, 6, 16), because 6 ≥ 2 and 192 − 3 ⋅ 62 =
162 − 3.
(a) Show that for every odd number x ≥ 5 there are at least two good
triples (x, y, z).
(b) Find a good triple (x, y, z) with x being even.
11
Solutions
B-problems
B1. 13 A solution with thirteen students is possible. If five students

scored 100 points and the remaining eights students scored 61 points, the
average score equals 5⋅100+8⋅61
13
= 988
13
= 76 points, as required.
It is not possible that the number of students taking the test was twelve
or less. Indeed, suppose that n ≤ 12 students took the test. Five of them
scored 100 points and the remaining n−5 students scored at least 60 points.
Their total score would be at least 500 + (n − 5) ⋅ 60 = 60n + 200. Their
average score would then be at least
60n + 200 200 200
= 60 + ≥ 60 + = 76 32 ,
n n 12
since n ≤ 12. This, however, contradicts the fact that their average score
was 76.
B2. 8 Note that both circles go through the D C
middle of the square. Therefore, the four circle seg- p

ments indicated by p, q, r, and s all belong to one
fourth of a circle with radius 2, hence the four seg-
ments have equal areas. Therefore, the combined r s
area of the grey shapes equals the area of triangle q
ACD which is 21 ⋅ 4 ⋅ 4 = 8.
A B
B3. 5 o’clock In 12 hours, the amounts by which the hour hands of the
1 5
first and the second clock are ahead increase by 100 and 100 of a turn
respectively. Therefore, in 12 hours, the second clock increases its lead
compared to the first clock by 5−1
100
= 25
1
of a full turn. Hence after 25 ⋅ 12
hours, the hour hand of the second clock has made exactly one extra full
turn compared to that of the first clock. This is the first time the two
clocks again display the same time.
During those 25⋅12 hours, the hour hand of the first clock has made exactly
101
100
⋅ 25 = 25 41 full turns. Both clocks will then display a time of 2 + 3 = 5
o’clock.
12
1 1
B4. 4
The product of the eight numbers in 2 32 8 1
the second and fourth row equals the product of the 4 2 8 2
eights numbers in the first and second column. Writ- 4 1 8 4
ing this out, we get:
16 2 14 16
C ⋅ 2 ⋅ 8 ⋅ 2 ⋅ F ⋅ G ⋅ H ⋅ 16 = 1
2
⋅ C ⋅ 4 ⋅ F ⋅ 32 ⋅ 2 ⋅ 1 ⋅ G.
Since C, F , and G are nonzero, we may divide both sides of the equation
by C, F , and G. The resulting equation is 512 ⋅ H = 128, which implies
that H = 41 . The figure shows one solution.
B5. 19 By dividing the regular hexagon into

six equilateral triangles, we deduce that the length of
the long diagonal AC of the hexagon equals twice the
side length of the hexagon. We will compute three
times the side length, namely ∣AB∣ + ∣BC∣ + ∣CD∣.
Observe that AB is a side of a parallelogram with
the parallel side having length 11 + 16 = 27. Thus we
have ∣AB∣ = 27.
As triangle BCE is equilateral, we have ∣BC∣ = ∣EB∣. A
As BCDF ia a parallelogram, we have ∣CD∣ = ∣BF ∣.

11
From the figure we see that ∣EB∣ + ∣BF ∣ = ∣EF ∣ =
5 + 16 + 9 = 30. 16 9 F
If we combine these facts, we find that three times

B D
the side length of the regular hexagon equals 5
E
∣AB∣ + ∣BC∣ + ∣CD∣ = 27 + ∣EB∣ + ∣BF ∣ = 27 + 30 = 57. C
The side length therefore equals 57

3
= 19.
13
C-problems
C1. (a) The number 4132 starts with a ‘4’ and is above average because 2 ⋅ 3 ≥
4 + 1 and 2 ⋅ 2 ≥ 1 + 3.
(b) Suppose that a4bc is a 4-digit number that is above average, where a,
b, and c are the digits ‘1’, ‘2’, and ‘3’ (possibly in a different order).
Then 2 ⋅ b ≥ a + 4 ≥ 5. So b ≥ 3.
Similarly, we find that 2 ⋅ c ≥ 4 + b ≥ 7, hence c ≥ 4. However, this is
impossible because c was at most 3.
(c) The numbers 1243756, 1234576, and 1234567 are above average and
have digit ‘7’ in the fifth, sixth, and seventh position, respectively.
Digit ‘7’ cannot be in the first position. Indeed, suppose that 7abcdef
would be above average. Then 2 ⋅ b ≥ 7 + a ≥ 8, hence b ≥ 4. Then we
must have 2 ⋅ c ≥ a + b ≥ 5, hence c ≥ 3. Now we find (in turn) that
also d, e, f ≥ 4. It follows that both digit ‘1’ and digit ‘2’ must be in
the position of a, which is impossible.
Digit ‘7’ cannot be in the second or third position. Indeed, otherwise
the digit following ‘7’ must be at least 4, which implies that also the
digits following it must be at least 4. Digits ‘1’, ‘2’, and ‘3’ must
therefore all be in the first two positions, which is impossible.
Finally, digit ‘7’ cannot be in the fourth position. Digit ‘1’ cannot be
in the third position since 2 ⋅ 1 < 2 + 3. Because the digit in the third
position must be at least 2, the digit in the fifth position must be at
least 5. The next digit must therefore be at least 6, as must be the
digit following it. The digits ‘1’ to ‘4’ must therefore all be in the first
three positions, which is impossible.
14
C2. (a) Since x ≥ 5 is odd, we can write x = 2n + 1 for an integer n ≥ 2. Now
(x, y, z) = (2n + 1, n, n + 2) and (x, y, z) = (2n + 1, n + 1, n − 1)
are two different good triples. In both cases it is clear that y and
z are indeed positive integers and that y ≥ 2. Substitution into the
equation shows that they are indeed solutions:
(2n + 1)2 − 3n2 = n2 + 4n + 1 = (n + 2)2 − 3 and

(2n + 1) − 3(n + 1) = n − 2n − 2 = (n − 1) − 3.
2 2 2 2
Thic concludes the solution.
Remark. One way to arrive at the idea of considering these triples

is the following. First substitute x = 5. It is then easy to see that z
can be no more than 5. The case z = 5 is not possible, because then
y = 1, which is not allowed. Hence z is at most 4. For z = 1, . . . , 4
compute the corresponding value of y, if it exists. This way, you will
find two good triples with x = 5. Repeating this for x = 7 and x = 9, you
will find good triples as well. The triples found show a clear pattern:
when x increases by 2, y and z both increase by 1. This holds for both
series of triples. Using this, you can guess a general expression for y
and z when x = 2n + 1. Checking that the found triples are good by
substitution in the equation will then suffice for a complete solution.
(b) An example is triple (16, 9, 4). This triple is good because 162 −3⋅92 =
13 = 42 − 3.
Remark. Stating a suitable triple and showing that it is a good triple

suffices for a complete solution. To find such a triple, one possibility
is to take the following approach. Rewrite the equation as x2 − z 2 =
3y 3 − 3. Both sides of the equation can be factored, which gives you
: (x − z)(x + z) = 3(y − 1)(y + 1). This will help in finding triples by
substituting different values for y. For example, you can try y = 4.
Then the right-hand side becomes 3 ⋅ 3 ⋅ 5, hence the left-hand side
becomes 5 ⋅ 9, 3 ⋅ 15, or 1 ⋅ 45. The value of x will always be the average
of the two factors, so x = 7, x = 9, and x = 23 in these three cases.
There are no even values for x when y = 4. If you try further values
of y, you will find even values of x for y = 7 and y = 9.
15
Problems
solved the problem.
1. In a table consisting of n by n small squares some squares are coloured

black and the other squares are coloured white. For each pair of columns
and each pair of rows the four squares on the intersections of these rows
and columns must not all be of the same colour.
What is the largest possible value of n?
2. Find all triples (x, y, z) of real numbers satisfying

x + y − z = −1, x2 − y 2 + z 2 = 1 and − x3 + y 3 + z 3 = −1.
3. The sides BC and AD of a quadrilateral A D

ABCD are parallel and the diagonals in-
tersect in O. For this quadrilateral ∣CD∣ =
∣AO∣ and ∣BC∣ = ∣OD∣ hold. Furthermore
CA is the angular bisector of angle BCD.
Determine the size of angle ABC.
O
Attention: the figure is not drawn to scale.
You have to write down your reasoning step
by step in text and formulas. No points will
be awarded for annotations in a picture alone. B C
4. For a positive integer n the number P (n) is the product of the positive
divisors of n. For example, P (20) = 8000, as the positive divisors of 20 are
1, 2, 4, 5, 10 and 20, whose product is 1 ⋅ 2 ⋅ 4 ⋅ 5 ⋅ 10 ⋅ 20 = 8000.
(a) Find all positive integers n satisfying P (n) = 15n.
(b) Show that there exists no positive integer n such that P (n) = 15n2 .
5. The number S is the result of the following sum:

1 + 10 + 19 + 28 + 37 + ⋯ + 102013 .
If one writes down the number S, how often does the digit ‘5’ occur in the
result?
16
Solutions
1. We will prove that n = 4 is the largest possible n for which an n × n-table

can be coloured according to the rules. The following figure shows a valid
colouring for n = 4.
Now we prove that there is no colouring of the squares in a 5 × 5-table

satisfying the requirements. Suppose, for contradiction, that such a col-
ouring exists. Of the squares in each row either the majority is black, or
the majority is white. We may suppose that there are at least three rows
for which the majority of the squares is black (the case where there are at
least three rows for which the majority of the squares is white is treated
in an analogous way). We now consider the squares in these three rows.
Of these 15 squares at least 9 are black.
If there is a column in which each of the three rows has a black square,
then each other column can contain at most one black square in these three
rows. The total number of black squares in the three rows will therefore
be no more than 3 + 1 + 1 + 1 + 1 = 7, contradicting the fact that the number
should be at least 9.
Hence, in each column at most two of the three rows have a black square.
We consider the number of columns with two black squares in the three
rows. If there are more then three, then there are two columns in which the
same two rows have a black square, which is impossible. It follows that the
number of black squares in the three rows is no more than 2+2+2+1+1 = 8,
again contradicting the fact that this number should be at least 9.
Hence, it is impossible to colour a 5 × 5-table according to the rules. It is
clear that it also will be impossible to colour an n × n-table according to
the rules if n > 5. ◻
17
2. The first equation yields z = x+y +1. Substitution into the second equation
gives x2 − y 2 + (x + y + 1)2 = 1. Expanding gives 2x2 + 2xy + 2x + 2y = 0, or
2(x + y)(x + 1) = 0. We deduce that x + y = 0 or x + 1 = 0. We consider the
two cases.
If x + y = 0, then y = −x holds. The first equation becomes z = 1.

Substitution into the third equation yields −x3 + (−x)3 + 13 = −1, or
x3 = 1. We deduce that (x, y, z) = (1, −1, 1).
If x + 1 = 0, then x = −1 holds. The first equation becomes z = y.
Substitution into the third equation yields −(−1)3 + y 3 + y 3 = −1, or
y 3 = −1. Hence, we have (x, y, z) = (−1, −1, −1).
In total we have found two possible solutions. Substitution into the original
equations shows that these are indeed both solutions to the system. ◻
3. First, we prove that some triangles in the figure are isosceles (the top angle
coincides with the middle letter).
(1) Triangle ADC is isosceles, because ∠DAC = ∠ACB = ∠ACD. The
first equality holds because AD and BC are parallel and the second
equality follows from the fact that AC is the interior angle bisector
of angle BCD.
(2) Triangle DAO is isosceles, because ∣AD∣ = ∣CD∣ = ∣AO∣. The first
equality follows from (1) and the second equality is given in the prob-
lem statement.
(3) Triangle BCO is isosceles, because it is similar to triangle DAO (hour-
glass shape).
(4) Triangle COD is isosceles, because ∣DO∣ = ∣BC∣ = ∣CO∣. Here the first
equality is given in the problem statement and the second one follows
from (3).
(5) Triangle BDC is isosceles, because it is similar to triangle BCO as
two pairs of corresponding angles are equal: ∠DBC = ∠OBC and
∠BDC = ∠DCO = ∠OCB. Here the last equality follows from (4).
(6) Triangle ADB is isosceles, because ∣AD∣ = ∣CD∣ = ∣BD∣ because of (1)
and (5).
Denote ∠ACB = α and ∠CBD = β (in degrees). From (5) it follows that
2α = β. From (3) it follows that 180○ = 2β + α = 5α, hence α = 180 = 36○
○
5
and β = 72 . In the isosceles triangle ADB ○the ○ top angle is equal to
○
∠ADB = β, hence its equal base angles are 180 2−72 = 54○ . The requested
angle therefore equals ∠ABC = ∠ABD + ∠DBC = 54○ + 72○ = 126○ .
18
A D
α β
α
β
O
β
α
β α
B C
◻
4. a) Because P (n) = 15n is the product of the positive divisors of n, the

prime divisors 3 and 5 of P (n) must also be divisors of n. It follows
that n is a multiple of 15. If n > 15, then 3, 5, 15 and n are distinct
divisors of n, yielding P (n) ≥ 3 ⋅ 5 ⋅ 15 ⋅ n = 225n. This contradicts the
fact that P (n) = 15n. The only remaining possibility is n = 15. This
is indeed a solution, because P (15) = 1 ⋅ 3 ⋅ 5 ⋅ 15 = 15 ⋅ 15.
b) Suppose that P (n) = 15n2 holds. Again, we find that n is a multiple
of 15. It is clear that n = 15 is not a sulution, hence n ≥ 30. We
observe that n5 > 5. It follows that 1 < 3 < 5 < n5 < n3 < n are six
distinct divisors of n. Thus P (n) ≥ 1 ⋅ 3 ⋅ 5 ⋅ n5 ⋅ n3 ⋅ n = n3 . Because
n > 15 holds, we have P (n) ≥ n ⋅ n2 > 15n2 , which contradicts the
assumption of this problem. We conclude that no n exists for which
P (n) = 15n2 . ◻
19
5. To illustrate the idea, we first calculate the number of fives in the result s
of the sum 1 + 10 + 19 + ⋯ + 100000. First notice that each term in the sum
is a multiple of nine plus 1:
s = 1 + (1 + 9) + (1 + 2 ⋅ 9) + ⋯ + (1 + 11111 ⋅ 9).
The number of terms in the sum is 11111 + 1 = 11112 and the average value
of a term is 1+100000
2
. It follows that s = 11112 ⋅ 100001
2
= 11112
2
⋅ 100001.
Because 2 = 5⋅2 = 10 = 5556, we find that s = 5556 + 555600000 =
11112 5⋅11112 55560
555605556. Hence, the number of fives is equal to 6 in this case.
Now we will solve the actual problem. Remark that 102013 = 1 + 9 ⋅ 11 . . . 1

(2013 ones). For simplicity let n = 11 . . . 1 be the number consisting of 2013
ones. We see that the sum
S = 1 + (1 + 9) + (1 + 2 ⋅ 9) + ⋯ + (1 + n ⋅ 9)
1+101023
has exactly n + 1 terms, with an average value of 2
. Hence S =
n+1
2
⋅ (1 + 102013 ).
n+1
Calculating the fraction 2
gives:
n + 1 5n + 5 555 . . . 560
= = = 555 . . . 56,
2 10 10
a number with 2011 fives followed by a 6. Because the last 2013 digits of
the number 102013 ⋅ n+1
2
are all zeroes, there is no ‘overlap’ between the
non-zero digits of 2 and 102013 ⋅ n+1
n+1
2
. We deduce that
S = n+1
2
⋅ (1 + 102013 )
= n+1
2
+ 102013 ⋅ n+1
2
= 55 . . . 56055 . . . 56.
Hence S is a number that contains exactly 2011 + 2011 = 4022 fives. ◻
20
Problems
1. Find all non-negative integers n for which there exist integers a and b such
that n2 = a + b and n3 = a2 + b2 .
2. Find all functions f ∶ R ∖ {0} → R for which
xf (xy) + f (−y) = xf (x)
for all non-zero real numbers x, y.
3. In triangle ABC, I is the centre of the incircle. There is a circle tangent

to AI at I which passes through B. This circle intersects AB once more
in P and intersects BC once more in Q. The line QI intersects AC in R.
Prove that ∣AR∣ ⋅ ∣BQ∣ = ∣P I∣2 .
4. Let m ≥ 3 and n be positive integers such that n > m(m − 2). Find the
largest positive integer d such that d ∣ n! and k ∤ d for all k ∈ {m, m +
1, . . . , n}.
5. Let n be a positive integer. Daniël and Merlijn are playing a game. Daniël
has k sheets of paper lying next to each other on a table, where k is a
positive integer. On each of the sheets, he writes some of the numbers
from 1 up to n (he is allowed to write no number at all, or all numbers).
On the back of each of the sheets, he writes down the remaining numbers.
Once Daniël is finished, Merlijn can flip some of the sheets of paper (he is
allowed to flip no sheet at all, or all sheets). If Merlijn succeeds in making
all of the numbers from 1 up to n visible at least once, then he wins.
Determine the smallest k for which Merlijn can always win, regardless of
Daniël’s actions.
21
Solutions
1. By AM-GM applied to a2 and b2 , we find that a2 + b2 ≥ 2ab. As 2ab =

(a + b)2 − (a2 + b2 ), it follows that n3 ≥ (n2 )2 − n3 , i.e. 2n3 ≥ n4 . Hence
either n = 0 or 2 ≥ n, so n = 0, n = 1, and n = 2 are the only possibilities.
For n = 0 we find a = b = 0 as solution, for n = 1 we find a = 0, b = 1 as
solution, and for n = 2 we find a = b = 2 as solution. Thus the non-negative
integers n with the desired property are precisely n = 0, n = 1, and n = 2.◻
2. Substituting x = 1 gives f (y) + f (−y) = f (1) for all y, or equivalently,

f (−y) = f (1) − f (y) for all y. Substituting y = −1 then gives xf (−x) +
f (1) = xf (x) for all x. We now substitute f (−x) = f (1) − f (x) to obtain
x(f (1) − f (x)) + f (1) = xf (x), so xf (1) + f (1) = 2xf (x). We see that f is
of the form f (x) = c + xc for a certain c ∈ R.
We check whether this family of functions satisfies the equation. The left
hand side now reads xf (xy) + f (−y) = x(c + xyc
) + c + −y
c
= xc + c, and the
right hand side reads xf (x) = x(c+ x ) = xc+c. Hence this function satisfies
c
the given equation for all c ∈ R. ◻
3. There is only one configuration. We have

∠AIP = ∠IBP (inscribed angle theorem)
= ∠IBQ (IB is angle bisector)
= ∠IP Q (quadrilateral P BQI is cyclic)
hence AI ∥ P Q. This implies ∠IAB = ∠QP B = ∠QIB, using the cyclic
quadrilateral for the latter equality. We have already seen that ∠AIP =
∠IBQ, so △IAP ∼ △BIQ (aa). Hence
∣AP ∣ ∣QI∣
= . (1)
∣P I∣ ∣BQ∣
Moreover, ∠RIA is the angle opposite to an inscribed angle, hence equal
to ∠IP Q, which we already know to be equal to ∠AIP . Hence ∠RIA =
∠AIP . As AI is an angle bisector, we have ∠RAI = ∠P AI, so △RAI ≅
△P AI (ASA). Hence ∣AR∣ = ∣AP ∣. Moreover, I is the centre of the arc P Q
as BI is an angle bisector, so ∣P I∣ = ∣QI∣. Now (1) gives
∣AR∣ ∣P I∣
= ,
∣P I∣ ∣BQ∣
implying that ∣AR∣ ⋅ ∣BQ∣ = ∣P I∣2 , as desired. ◻
22
4. We prove that d = m − 1 is the largest integer satisfying the conditions.
First note that m − 1 ∣ n! and that for k ≥ m we have k ∤ m − 1, so d = m − 1
indeed satisfies the conditions.
Now suppose that for some d we have d ∣ n! and k ∤ d for all k ∈ {m, m +
1, . . . , n}. We prove that d ≤ m − 1. Write d = p1 p2 ⋯pt , where the pi
are prime for all i (but not necessarily pairwise distinct). If t = 0, then
d = 1 ≤ m − 1, so we may assume that t ≥ 1. From the first condition on d,
it follows that pi ≤ n for all i. From the second condition on d, it follows
that pi ∈/ {m, m + 1, . . . , n} for all i. Hence pi ≤ m − 1 for all i. Now consider
the integers p1 , p1 p2 , . . . , p1 p2 ⋯pt . These are divisors of d and hence all are
not in {m, m + 1, . . . , n}. Moreover, we know that p1 ≤ m − 1. Consider the
largest j ≤ t such that p1 p2 ⋯pj ≤ m − 1. If j < t, then
p1 p2 ⋯pj pj+1 ≤ (m − 1)pj+1 ≤ (m − 1)(m − 1) = m(m − 2) + 1 ≤ n.
But then p1 p2 ⋯pj pj+1 ≤ m − 1, contradicting the maximality of j. Hence

j = t, so d = p1 p2 ⋯pt ≤ m − 1.
We conclude that d = m − 1 is indeed the largest integer satisfying the
conditions. ◻
5. We give each of Daniël’s sheets of paper a different colour. Moreover, we

have n boxes with the numbers from 1 up to n on them. We make sure we
have enough chips in the colours of Daniël’s sheets. For each sheet, Merlijn
puts a chip with the colour of this sheet in every box of which the number
is on the front of this sheet. So every box will contain precisely those chips
of which the colours are those of the sheets on which this number is on the
front.
For each sheet of paper Merlijn flips, he takes a chip of the same colour
from the supply. A number is not visible on the table if and only if the
sheets on which this number was on the front, are precisely the sheets that
Merlijn flipped, i.e. if and only if the set of chips that he took is equal to
the set of chips in the box with this number. Hence Merlijn wins if and
only if the set of chips he took does not occur in any of the boxes, since
then all of the numbers are visible. Now we claim that Merlijn can win
if and only if 2k > n. So the smallest k for which Merlijn can win, is the
smallest k such that 2k > n.
Suppose 2k > n. The number of possible sets of colours is 2k , hence larger
than n. There are n boxes, so not every set of colours can occur in the
boxes. Hence Merlijn can pick a set of colours not occurring in any of the
boxes, flip the corresponding sheets, and win.
23
Now suppose that 2k ≤ n. Then Daniël first fills the boxes with sets of
chips, in such a way that every possible set of colours occurs in some box.
This is possible, since there are 2k possible sets of colours, and we have at
least 2k boxes. Now Daniël writes on the front of each sheet of coloured
paper precisely those numbers whose boxes contain the colour of that sheet,
and on the back the remainder of the numbers. In this way, the chips in
the boxes correspond to the numbers on the fronts of the sheets. Now
Merlijn cannot choose a set of colours not occurring in any of the boxes;
he cannot win. ◻
24
Problems
1. Determine all pairs (a, b) of positive integers satisfying
a2 + b ∣ a2 b + a and b2 − a ∣ ab2 + b.
2. Let △ABC be a triangle. Let M be the midpoint of BC and let D be a

point on the interior of side AB. The intersection of AM and CD is called
E. Suppose that ∣AD∣ = ∣DE∣. Prove that ∣AB∣ = ∣CE∣.
3. Let a, b and c be rational numbers for which a + bc, b + ac and a + b are all
non-zero and for which we have
1 1 1
+ = .
a + bc b + ac a + b
√
Prove that (c − 3)(c + 1) is rational.
4. Let △ABC be a triangle with ∣AC∣ = 2∣AB∣ and let O be its circumcentre.
Let D be the intersection of the angle bisector of ∠A and BC. Let E be
the orthogonal projection of O on AD and let F ≠ D be a point on AD
satisfying ∣CD∣ = ∣CF ∣. Prove that ∠EBF = ∠ECF .
5. On each of the 20142 squares of a 2014 × 2014-board a light bulb is put.

Light bulbs can be either on or off. In the starting situation a number
of the light bulbs is on. A move consists of choosing a row or column in
which at least 1007 light bulbs are on and changing the state of all 2014
light bulbs in this row or column (from on to off or from off to on). Find
the smallest non-negative integer k such that from each starting situation
there is a finite sequence of moves to a situation in which at most k light
bulbs are on.
25
Solutions
1. From a2 + b ∣ a2 b + a it follows that
a2 + b ∣ (a2 b + a) − b(a2 + b) = a − b2 .
From b2 − a ∣ ab2 + b it follows that
b2 − a ∣ (ab2 + b) − a(b2 − a) = b + a2 .
Hence we have a2 +b ∣ a−b2 ∣ a2 +b. This means that a2 +b is equal to a−b2 ,

up to sign. We distinguish two cases: a2 + b = b2 − a and a2 + b = a − b2 . In
the latter case we have a2 + b2 = a − b. But a2 ≥ a and b2 ≥ b > −b, hence
this is impossible. Therefore we must be in the former case: a2 + b = b2 − a.
This yields a2 − b2 = −a − b, hence (a + b)(a − b) = −(a + b). As a + b is
positive, we may divide by it and we get a − b = −1, hence b = a + 1. All
pairs that could possibly satisfy the conditions are of the form (a, a + 1)
for a positive integer a.
We consider these pairs. We have a2 +b = a2 +a+1 and a2 b+a = a2 (a+1)+a =
a3 + a2 + a = a(a2 + a + 1), hence the first divisibility condition is satisfied.
Furthermore, we have b2 − a = (a + 1)2 − a = a2 + a + 1 and ab2 + b =
a(a+1)2 +(a+1) = a3 +2a2 +2a+1 = a(a2 +a+1)+a2 +a+1 = (a+1)(a2 +a+1),
hence also the second divisibility condition is satisfied. Hence the pairs
(a, a + 1) satisfy the conditions and they are exactly the pairs satisfying
the conditions. ◻
2. We apply Menelaos’s theorem to the line through A, E and M inside

triangle BCD. This yields
∣BM ∣ ∣CE∣ ∣DA∣

⋅ ⋅r = 1.
∣M C∣ ∣ED∣ ∣AB∣
Because M is the midpoint of BC, we have ∣BM ∣

∣M C∣
= 1. Furthermore, it is
given that ∣AD∣ = ∣DE∣. Altogether this yields ∣CE∣ = ∣AB∣. ◻
26
3. We have
1 1 (b + ac) + (a + bc) a + b + ac + bc
+ = = .
a + bc b + ac (a + bc)(b + ac) ab + a2 c + b2 c + abc2
Hence, the problem statement’s equality yields
(a + b)(a + b + ac + bc) = ab + a2 c + b2 c + abc2 ,
or equivalently,
a2 + ab + a2 c + abc + ab + b2 + abc + b2 c = ab + a2 c + b2 c + abc2 ,
or equivalently,
a2 + 2abc + ab − abc2 + b2 = 0.
We can consider this as a quadratic equation in a, of which we know that
it has a rational solution. Hence, the discriminant of this equation must
be the square of a rational number. This discriminant equals
D = (2bc + b − bc2 )2 − 4b2

= (2bc + b − bc2 − 2b)(2bc + b − bc2 + 2b)
= b2 (2c − 1 − c2 )(2c + 3 − c2 )
= b2 (c2 − 2c + 1)(c2 − 2c − 3)
= b2 (c − 1)2 (c + 1)(c − 3).
If c = 1, then the original equation becomes

1 1 1
+ = ,
a+b a+b a+b
which has no solution. If a = 0, then the original equation becomes
1 1 1
+ = ,
bc b b
which also has no solution. Hence, c ≠ 1 and a ≠ 0. In particular,
D
(c − 3)(c + 1) =
b2 (c − 1)2
must be the square of a rational number. ◻
4. Let G ≠ A be the intersection of AD with the circumcircle of △ABC.

Then we have ∣AG∣ = 2∣AE∣ because the projection of the centre of a circle
27
on a chord of this circle is the midpoint of that chord. Let M be the
midpoint of AC. Because ∣AB∣ = ∣AC∣
2
= ∣AM ∣ and because AD is the angle
bisector of ∠BAM , we have that M is the image of B under reflection
in AD. Now we have ∠DGC = ∠AGC = ∠ABC = ∠ABD = ∠DM A =
180○ − ∠DM C, hence DM CG is a cyclic quadrilateral. Then we have
AM 2 = AM2⋅AC = AD⋅AG2
= AD ⋅ AE, hence △AM E ∼ △ADM (SAS).
Now we have 180○ − ∠EM C = ∠EM A = ∠M DA = ∠BDA = ∠CDF =
∠DF C = ∠EF C, hence EM CF is a cyclic quadrilateral. Hence, ∠EBF =
∠EM F = ∠ECF , which is exactly what we needed to prove. ◻
5. Number the rows from 1 up to 2014 and also number the columns. Consider
the following beginning situation: in row i the light bulbs in columns i,
i + 1, . . . , i + 1005 are on and the rest are off, in which we take the column
numbers modulo 2014. In each row and in each column there are exactly
1006 light bulbs that are on. Hence, there is no move possible. It is not
always possible to get to a situation in which less than 2014 ⋅ 1006 light
bulbs are on.
Now we shall show that it is always possible to get to a situation in which
at most 2014 ⋅ 1006 are on. Suppose, from the contrary, that in a certain
situation at least 2014 ⋅ 1006 + 1 light bulbs are on and it is not possible to
get to a situation in which less bulbs are on. If there is a row or column in
which at least 1008 bulbs are on, then we can change the states of these
bulbs in this row or column and there will be less bulbs that are on. This
is a contradiction, hence in each row and column at most 1007 bulbs are
on.
Let I be the set of rows in which exactly 1007 bulbs are on and let J1
be the set of columns in which exactly 1007 bulbs are in. We will try to
change the state of the bulbs in all rows in I. This we call the big plan.
If after executing the big plan there is a column in which at least 1008
bulbs are on, then we get a contradiction. Hence we shall assume that this
does not happen. Let J2 be the set of columns that, after executing the
big plan, contain exactly 1007 bulbs that are on. If there exists a square
(i, j) with i ∈ I and j ∈ J1 containing a bulb that is off, we can switch row
i and then column j gets more than 1007 bulbs that are on, which is a
contradiction. Hence every bulb on (i, j) with i ∈ I and j ∈ J1 is on. If
there exists a square (i, j) with i ∈ I and j ∈ J2 containing a bulb that is
off, after executing the big plan, then we get a contradiction in the same
way. Hence every bulb on (i, j) with i ∈ I and j ∈ J2 is off after executing
the big plan. Because the columns in J2 contain exactly 1007 bulbs that
are on after executing the big plan, there are 1007 − ∣I∣ bulbs that are on
before executing the big plan. (For a set X, we denote by ∣X∣ its number
28
of elements.) In the columns of J1 there are exactly 1007 that are on, and
this also means that J1 and J2 are disjoint. In the other columns at most
1006 bulbs are on. The total number of bulbs that are on before the big
plan is at most
(1007−∣I∣)∣J2 ∣+1007∣J1 ∣+1006(2014−∣J1 ∣−∣J2 ∣) = 1006⋅2014+∣J1 ∣+∣J2 ∣−∣I∣⋅∣J2 ∣.
This has to be at least 1006 ⋅ 2014 + 1, hence ∣J1 ∣ + ∣J2 ∣ − ∣I∣ ⋅ ∣J2 ∣ ≥ 1. This
yields ∣J1 ∣ > (∣I∣ − 1)∣J2 ∣. If ∣I∣ ≥ 2, then ∣J1 ∣ > ∣J2 ∣. By executing the big
plan, the number of columns in which 1007 bulbs are on decreases. But
after that, we again have a situation with ∣I∣ rows containing 1007 bulbs
that are on, and in which J1 and J2 have been interchanged. Then we can
again apply the big plan, to decrease the number of columns that contain
1007 bulbs that are on again. This is a contradiction, because now we are
back in the beginning situation. We conclude that we must have ∣I∣ = 1.
We have a situation in which there is exactly one row with exactly 1007
bulbs that are on. Analogously, we can show that there must be exactly one
column in which exactly 1007 bulbs are on. Because there are 1006⋅2014+1
bulbs that are on, each other row and column must contain exactly 1006
bulbs that are on. Now change the state of the bulbs in the row with 1007
bulbs that are on. In 1007 columns there will be 1006 + 1 = 1007 bulbs that
are on. We have already seen that in this situation we can decrease the
number of bulbs that are on.
We conclude that if there are more than 1006 ⋅ 2014 bulbs that are on,
it is always possible to decrease this number. Hence, the smallest k is
1006 ⋅ 2014. ◻
29
Problems
1. Let f ∶ Z>0 → R be a function such that for all n > 1 there is a prime divisor
p of n such that
n
f (n) = f ( ) − f (p).
p
Furthermore, it is given that f (22014 ) + f (32015 ) + f (52016 ) = 2013.
Determine f (20142 ) + f (20153 ) + f (20165 ).
2. The sets A and B are subsets of the positive integers. The sum of any two
distinct elements of A is an element of B. The quotient of any two distinct
elements of B (where we divide the largest by the smallest of the two) is
an element of A. Determine the maximum number of elements in A ∪ B.
3. Let H be the orthocentre of an acute triangle ABC. The line through A

perpendicular to AC and the line through B perpendicular to BC intersect
in D. The circle with centre C through H intersects the circumcircle of
triangle ABC in the points E and F . Prove that ∣DE∣ = ∣DF ∣ = ∣AB∣.
4. Determine all pairs (p, q) of prime numbers for which pq+1 +q p+1 is a square.
5. Let P (x) be a polynomial of degree n ≤ 10 with integral coefficients such

that for every k ∈ {1, 2, . . . , 10} there is an integer m with P (m) = k.
Furthermore, it is given that ∣P (10) − P (0)∣ < 1000. Prove that for every
integer k there is an integer m such that P (m) = k.
30
Solutions
1. If n = q with q prime, then there is only one prime divisor of n, namely

q, hence we must have that f (q) = f (1) − f (q), hence f (q) = 21 f (1). If
n = q 2 with q prime, then n also has only one prime divisor, hence we have
f (q 2 ) = f (q) − f (q) = 0. We will prove by induction to k that f (q k ) =
2−k
2
f (1) if q is a prime number and k a positive integer. For k = 1 and
k = 2 we have already shown this. Now suppose that f (q k ) = 2−k2
f (1) for
certain k ≥ 2 and substitute n = q k+1 . Then we have
2−k 1 2 − (k + 1)
f (q k+1 ) = f (q k ) − f (q) = f (1) − f (1) = f (1).
2 2 2
This completes the induction argument.
Now we will use the second equality. We have
2013 = f (22014 ) + f (32015 ) + f (52016 )
2 − 2014 2 − 2015 2 − 2016
= f (1) + f (1) + f (1)
2 2 2
6039
=− f (1),
2
hence f (1) = 2013⋅2
−6039
= − 23 . Then we have for each prime number q that
f (q) = 21 f (1) = − 13 .
We will prove the following statement: if n = p1 p2 ⋯pm with p1 , p2 , . . . , pm
not necessarily distinct prime numbers and m ≥ 0, then we have f (n) =
m−2
3
. This we will do by induction to m. For m = 0 we have n = 1 and
f (1) = − 23 = 0−2
3
, hence in this case it is true. Now suppose that we have
proved the induction hypothesis for certain m ≥ 0. Consider an arbitrary
n of the form n = p1 p2 ⋯pm+1 . Then n > 1, hence there is a prime factor
p ∣ n for which we have f (n) = f ( np ) − f (p); without loss of generality this
is p = pm+1 . Now it follows that
m−2 1 (m + 1) − 2
f (n) = f (p1 p2 ⋯pm ) − f (pm+1 ) = −− = .
3 3 3
Now we can calculate the answer. The prime factorisations of 2014, 2015
and 2016 are 2014 = 2 ⋅ 19 ⋅ 53, 2015 = 5 ⋅ 13 ⋅ 31 and 2016 = 25 ⋅ 32 ⋅ 7, hence
6 − 2 9 − 2 40 − 2 49
f (20142 ) + f (20153 ) + f (20165 ) = + + = .
3 3 3 3
◻
31
2. Suppose that A contains at least three elements, say a < b < c. Then B
contains the three distinct elements a + b < a + c < b + c. Hence, A certainly
b+c
contains the element a+c . Apparently this fraction is an integer, hence
a + c ∣ b + c. But then it follows that a + c ∣ (b + c) − (a + c) = b − a. We know
that b > a, hence b − a is positive, hence we must have a + c ≤ b − a. This
yields c ≤ b − 2a < b, which is in contradiction with c > b. Hence, A contains
at most two elements.
Suppose that B contains at least four elements, say a < b < c < d. Then
A contains the three distinct elements ad , db and dc . But A cannot contain
three distinct elements, contradiction. Hence, B contains at most three
elements.
In total A ∪ B contains at most 5 elements. This number can be attained,
take for example A = {2, 4} and B = {3, 6, 12}. Now 2 + 4 = 6 ∈ B and
12
6
= 63 = 2 ∈ A and 12
3
= 4 ∈ A, hence this pair of sets satisfies the conditions.
We conclude that the maximum number of elements of A ∪ B is 5. ◻
3. The triangle is acute, hence H lies inside the triangle. This means that E
and F lie on the short arcs AC and BC. Suppose that E lies on the short
arc AC and that F lies on the short arc BC.
If we reflect H in AC, then the reflection H ′ lies on the circumcircle of
△ABC. (This is a well-known fact, which can be proved by angle chasing
to prove that ∠AHC = 180○ − ∠ABC.) On the other hand, this reflection
also lies on the circle with centre C through H, because ∣CH ′ ∣ = ∣CH∣.
Hence, H ′ is the intersection point of the two circles and that is E. We
conclude that E is the image of H under the reflection in AC.
This means that EH is perpendicular to AC and hence it is the same
line as BH. Because AD is also perpendicular to AC, the lines BE are
AD parallel. Furthermore, D lies on the circumcircle of △ABC because
∠CAD + ∠CBD = 90○ + 90○ = 180○ . We have already seen that E lies on
the short arc AC, hence EADB is a cyclic quadrilateral (in this order).
Now we have ∠BEA + ∠EAD = 180○ since BE and AD are parallel, but
also ∠EBD + ∠EAD = 180○ because of the cyclic quadrilateral. Hence,
∠BEA = ∠EBD, hence the corresponding chords BA and ED have the
same length.
Analogously we can prove that ∣AB∣ = ∣DF ∣, which solves the problem. ◻
32
4. First suppose that both p and q are odd. Then the exponents in the
sum pq+1 + q p+1 are both even, from which it follows that both terms are
congruent to 1 mod 4. Hence, the sum is congruent to 2 mod 4, but this
is never a square.
Now suppose that both p and q are even. Then, they are both equal to
2. That yields pq+1 + q p+1 = 23 + 23 = 16 = 42 , hence this pair satisfies the
conditions.
Finally, suppose that one of both, say p, is even and the other one is odd.
We then have p = 2 and 2q+1 +q 3 = a2 for a certain positive integer a. Write
q + 1 = 2b with b a positive integer, then the equality becomes 22b + q 3 = a2 ,
or equivalently
q 3 = a2 − 22b = (a − 2b )(a + 2b ).
Both factors on the right hand side now have to be a power of q, say
a − 2b = q k and a + 2b = q l with l > k ≥ 0. Both factors are divisible by
q k , hence also the difference is divisible by it. Hence, q k ∣ 2 ⋅ 2b = 2b+1 .
However, q is an odd prime, hence the only power of q that is a divisor of
a power of two, is 1. Hence, k = 0. Now we get q 3 = a + 2b and a − 2b = 1,
hence q 3 = (2b + 1) + 2b = 2b+1 + 1. This yields
2b+1 = q 3 − 1 = (q − 1)(q 2 + q + 1).
However, q 2 + q + 1 ≡ 1 mod 2 and furthermore q 2 + q + 1 > 1, hence this

can never be a power of two. Contradiction.
We conclude that (p, q) = (2, 2) is the only solution. ◻
33
5. For i = 1, 2, . . . , 10 let ci be an integer such that P (ci ) = i. For i ∈
{1, 2, . . . , 9} we have that
ci+1 − ci ∣ P (ci+1 ) − P (ci ) = (i + 1) − i = 1,
hence ci+1 −ci = ±1 for all i ∈ {1, 2, . . . , 9}. Furthermore, it holds that ci ≠ cj
for i ≠ j, because P (ci ) = i ≠ j = P (cj ). We conclude that c1 , c2 , . . . , c10 are
ten consecutive integers, either in ascending or descending order. Hence,
we will consider the following two cases:
(A) ci = c1 − 1 + i for i = 1, 2, . . . , 10 (i.e., c1 , c2 , . . . , c10 is an ascending

sequence of consecutive integers),
(B) ci = c1 + 1 − i for i = 1, 2, . . . , 10 (i.e., c1 , c2 , . . . , c10 is a descending
sequence of consecutive integers).
First consider case (A). Define Q(x) = 1 + x − c1 . Then for 1 ≤ i ≤ 10 we
have that
Q(ci ) = Q(c1 − 1 + i) = 1 + (c1 − 1 + i) − c1 = i = P (ci ),
hence P (ci ) − Q(ci ) = 0. Hence, we can also write

10
P (x) − Q(x) = R(x) ⋅ ∏(x − ci ),
i=1
or equivalently,
10
P (x) = 1 + x − c1 + R(x) ⋅ ∏(x − ci ).
i=1
Because the degree of P is at most 10, the degree of R cannot be greater

than 0. Hence R(x) is a constant, say R(x) = a with a ∈ Z. We then get
that
10
P (x) = 1 + x − c1 + a ⋅ ∏(x − ci ).
i=1
Now we substitute x = 10 and x = 0:

10 10
P (10) − P (0) = 1 + 10 − c1 + a ⋅ ∏(10 − ci ) − (1 + 0 − c1 ) − a ⋅ ∏(0 − ci )
i=1 i=1
10 10
= 10 + a ⋅ (∏(10 − ci ) − ∏(0 − ci )) .
i=1 i=1
34
The numbers 10 − c1 , 10 − c2 , . . . , 10 − c10 are ten consecutive numbers and
the numbers 0−c1 , 0−c2 , . . . , 0−c10 are the next ten consecutive numbers.
Hence, there is an N such that
10 10
∏(10−ci )−∏(0−ci ) = (N +20)(N +19)⋯(N +11)−(N +10)(N +9)⋯(N +1).
i=1 i=1
We will find a bound for this quantity. First we suppose that N + 1 > 0.
Then we have
(N + 20)(N + 19)⋯(N + 11) − (N + 10)(N + 9)⋯(N + 1)

> (N + 20)(N + 9)⋯(N + 1) − (N + 10)(N + 9)⋯(N + 1)
= 10 ⋅ (N + 9)(N + 8)⋯(N + 1)
≥ 10!.
If N + 20 < 0, then all factors are negative. Completely analogously the

absolute difference is again much greater than 10!. If N + 20 ≥ 0 and
N + 1 ≤ 0, then one of the factors is 0. Hence, exactly one of the two
terms is equal to zero and the other one is at least 10! in absolute value.
We conclude that the absolute difference always is at least 10!. Hence,
if a ≠ 0, then ∣P (10) − P (0)∣ ≥ 10! − 10 > 1000. Given is, however, that
∣P (10) − P (0)∣ < 1000. Apparently we must have that a = 0. Now we find
that
P (x) = 1 + x − c1 .
Let k ∈ Z be arbitrary and pick m = k − 1 + c1 . Then it holds that P (m) =
1 + (k − 1 + c1 ) − c1 = k. Hence for any integer k there is an integer m with
P (m) = k.
Now consider case (B). We can use the exact same reasoning, in which we
now define Q(x) = 1 − x + c1 and eventually get that
10
P (x) = 1 − x + c1 + a ⋅ ∏(x − ci ).
i=1
In the same way, we deduce that a = 0, yielding
P (x) = 1 − x + c1 .
And now it follows again that for any integer k there is an integer m with
P (m) = k. ◻
35
Problems
Part 1
1. The four symbols ◯, ◁, ☆, and ◻ represent distinct digits. Suppose

that
◯ × ◯ = ◁◯ and ☆ + ☆ = ◻◯.
Which digit is ☆?
A) 5 B) 6 C) 7 D) 8 E) 9
2. The Free family is driving on the German highway to a faraway resort.

Their fuel tank is full at the beginning of their journey. A third of the way
through the journey, 75% of the fuel is remaining in the fuel tank. How
much fuel is remaining in the fuel tank halfway through the journey?
A) 25% B) 33 13 % C) 50% D) 60% E) 62,5%
3. A football tournament with five teams is held, in which every pair of teams
plays one match against each other. Two points are awarded for winning
a match, one point is awarded for a draw, and zero points are awarded for
losing a match. After the tournament, every team has a number of points
different from all of the other teams. What are the possibilities for the
number of points of the winning team?
A) 8 B) 8, 7 C) 8, 6 D) 8, 7, 6 E) 8, 7, 6, 5
4
4. A small 4 by 4 square lies partially on a larger 5 by
5 square, in such a way that one of the vertices of
the larger square lies directly beneath the centre of
the small square. What part of the large square is
covered by the small square? 5
A) 15% B) 16% C) 17,5%

D) 18% E) This cannot be determined.
36
5. A large sheet of paper can be divided into at most 6 1 6
pieces with two horizontal lines and one other line,
see figure. With five horizontal lines and five other
lines, what is the maximal number of pieces that you
2 5
can obtain?
3 4
A) 43 B) 44 C) 45 D) 46 E) 47
6. Five children, Ahmed, Bob, Celine, Dan and Eve, are standing in a queue
to buy ice cream at an ice cream van. They all have an integer number of
euros. The one or more children in front of Ahmed in the queue, together
have 4 euros. The children between Celine and Dan together have 7 euros.
The children in front of Eve together have 6 euros, and Eve has 2 euros
herself. How many euros does Ahmed have?
A) 1 B) 2 C) 3 D) 4 E) 5
7. We split the numbers from 1 up to 9 into a group of four and a group

of five. In both groups, we multiply the numbers. Then we divide the
larger of the two results by the smaller one. We require the result to be
an integer. In how many ways can we split the numbers 1 up to 9 into a
group of four and a group of five so that this is the case?
A) 1 B) 2 C) 3 D) 4 E) 5
8. John has six squares of equal size: two red ones, two grey ones, and two
blue ones. He makes a cube out of them by gluing them together. How
many different cubes can John make? Two cubes are different if they
cannot be transformed into one another by a rotation of the cube.
A) 3 B) 4 C) 5 D) 6 E) 8
37
Part 2
1. What is the smallest number that can be obtained by adding three consec-
utive positive even integers, but also by adding four consecutive positive
even integers? (A number is positive if it is larger than 0.)
2. Nick has red and blue marbles. All red marbles have the same weight, and
so do all blue marbles. Nick weighs a red marble and a blue one: together
they weigh 30 grams. Then he weighs a number of red marbles and a blue
one: they weigh 180 grams. Exactly the same total number of marbles,
but now with one red instead of one blue marble, together weigh 60 grams.
What is the weight of a red marble in grams?
3. You have square tiles with edges of length 1 metre, 2 metres, 3 metres, and
so on. You have more than enough tiles of each kind. What is the length
of the side of the smallest square you can make using precisely 11 of these
tiles?
4. How many numbers between 1 and 1000 are not divisible by 2, divisible
by 3, not divisible by 5, and divisible by 7?
5. A collection of distinct positive integers satisfies the following. Every triple

of integers in this collection adds up to less than 37. Every quadruple of
integers in this collection adds up to more than 37. What is the maximum
number of integers this collection can contain?
6. In this problem, a, b, c, and d represent digits not a b 2 0

equal to 0, in such a way that the calculation to the 1 3 c d −
right is correct. Determine all possibilities for cdab. c d a b
7. In the figure below there is a triangle ABC. The segment AB is horizontal,

and the segment CD is vertical. The segment DB is thrice as long as AD.
Moreover, E and F divide CD into three parts of equal length. The two
horizontal lines through E and F divide, together with CD, the triangle
ABC into six parts. Triangle ABC has area 1. What is the combined area
of the three grey parts?
38
C
A D B
8. Ionica and Jeanine cross out four numbers each in

the figure to the right, so that exactly one number
4 20 18
remains. They each add up the numbers they crossed 7 11 9
out. Ionica’s sum is thrice as large as Jeanine’s sum.
Which numbers could be the remaining one?
25 2 6
39
Solutions
Part 1
1. D) 8 5. D) 46
2. E) 62, 5% 6. B) 2
3. D) 8, 7, 6 7. D) 4
4. B) 16% 8. D) 6
Part 2
1. 36 5. 6
2. 25 grams 6. 5367
5
3. 5 metres 7. 12
4. 19 8. 6 and 18
40
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

53rd Dutch Mathematical Olympiad 2014
and the team selection for IMO 2015 Thailand
MY

CY
We eat problems
CMY

for breakfast.

in Nederlandrdgehouden: IMO2011
Olympiad 2014
International
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
© Stichting Nederlandse Wiskunde Olympiade, 2015 FOUNDATION COMPOSITIO MATHEMATICA

Introduction
2014, held at the participating schools. The paper consisted of eight multi-
ple choice questions and four open questions, to be solved within 2 hours.
In total 23% more students than in 2013 participated in this first round:
to be precise: 9161 students of 323 secondary schools.
The 1000 best students were invited to the second round, which was held in
March at twelve universities in the country. This round contained five open
questions, and two problems for which the students had to give extensive
solutions and proofs. The contest lasted 2.5 hours.
The 130 best students were invited to the final round. Also some outstand-
ing participants in the Kangaroo math contest or the Pythagoras Olympiad
were invited. In total about 150 students were invited. They also received
an invitation to some training sessions at the universities, in order to pre-
pare them for their participation in the final round.
The final round in September contained five problems for which the stu-
dents had to give extensive solutions and proofs. They were allowed 3
hours for this round. After the prizes had been awarded in the beginning
of November, the Dutch Mathematical Olympiad concluded its 53rd edition
2014.

1
girls, as this year we participated again in the European Girls’ Mathemat-
ical Olympiad (EGMO). In total there were eight girls competing to be in
the EGMO team. The team of four girls was selected by a selection test,
held on 6 March 2015. They attended the EGMO in Minsk, Belarus from
14 until 20 April, and the team returned with two honourable mentions.
For more information about the EGMO (including the 2015 paper), see
www.egmo.org.
The same selection test was used to determine the ten students partici-
pating in the Benelux Mathematical Olympiad (BxMO), held in Mersch,
Luxembourg, from 8 until 10 May. The Dutch team received four bronze
medals and three silver medals. For more information about the BxMO
as an observer C. The team had a training camp in Chiang Mai, from 30
June until 8 July.
participate in this event were the 100 best students of grade 2 and grade 3
of the popular Kangaroo math contest. The competition consisted of two
one-hour parts, one with eight multiple choice questions and one with eight
open questions. The goal of this Junior Mathematical Olympiad is to scout
talent and to stimulate them to participate in the first round of the Dutch
Mathematical Olympiad.
the solutions.
2
Dutch delegation
The Dutch team for IMO 2015 in Thailand consists of
• Eva van Ammers (17 years old)

– participated in EGMO 2014, honourable mention at EGMO
2015
• Dirk van Bree (17 years old)
• Tim Brouwer (18 years old)
• Yuhui Cheng (19 years old)
– participated in BxMO 2014, bronze medal at BxMO 2015
• Mike Daas (17 years old)
• Bob Zwetsloot (17 years old)
– bronze medal at BxMO 2013, bronze medal at BxMO 2014,
silver medal at BxMO 2015
• Levi van de Pol (13 years old)

– silver medal at BxMO 2015

• Merlijn Staps (observer B), Utrecht University
3
Problems
A-problems
1. We are given a 4 × 4 table and want to colour four 4

of the 16 cells black. This should be done in such 3
a way that every row and every column has exactly
2
one black cell, and no two black cells are diagonally
adjacent (share a corner point). 1
In how many ways can we choose the four black cells? A B C D
A) 1 B) 2 C) 3 D) 4 E) It is impossible.
2. A pond contains both red and yellow carp. Two fifths of the carp are
yellow, the rest of the carp are red. Three quarters of the yellow carp are
female. In total, there are an equal number of male and female carp.
Which fraction of the total carp population are red males?
1 1 3 2 1
A) 5 B) 4 C) 10 D) 5 E) 2
3. Seven lily pads are numbered 1 through 7 from left to right. A frog jumps
along these pads. It can jump to the left and to the right, but only by
leaps of three or five pads at once. For example, starting from pad 2, it
can only leap to pads 5 and 7. The frog wants to make a journey in which
he visits each pad exactly once (so the first and last pad on his journey
will be different).
1 2 3 4 5 6 7
Which pads can be the starting point of such a journey?

A) pads 1 to 7 B) pads 1, 3, 5, and 7 C) pads 3 and 5
D) pad 4 E) none of the pads
4
4. A square paper ring has height 1, and the sides have length 4. The ring is
depicted in the left hand figure. By folding it flat on the tabletop, we get
the right hand figure, where ABCD is a square.
1 A D
A D
4 B C
B C
What is the length of side AB?
5 7 9
A) 2 B) 3 C) 2 D) 4 E) 2
5. Tom and Jerry were running a race. The number of runners finishing before
Tom was equal to the number of runners finishing after him. The number
of runners finishing before Jerry was three times the number of runners
finishing after him. In the final ranking, there are precisely 10 runners in
between Tom and Jerry. All runners finished the race, and no two runners
finished at the same time.
How many runners participated in the race?
A) 22 B) 23 C) 41 D) 43 E) 45
6. A garden with a pond (the black hexagon) will

be tiled using hexagonal tiles as in the figure.
The tiles come in three colours: red, green and
blue. No two tiles that share a side can be of
the same colour.
In how many ways can the garden be tiled?
A) 3 B) 6 C) 12 D) 18 E) 24
5
7. In the figure, a quadrilateral ABCD is drawn. The midpoint of side AB
is called M . The four line segments AM , BM , BC, and AD each have
length 8, and the line segments DM and CM both have length 5.
What is the length of line segment CD?
D C
?
8 8
5 5
A 8 M 8 B
Beware: the figure is not drawn to scale.

40 25 16 13
A) 3 B) 13 C) 8 D) 5 E) 4
8. A motorboat is moving with a speed of 25 kilometres per hour, relative to

the water. It is going from Arnhem to Zwolle, moving with the constant
current. At a certain moment, it has travelled 42% of the total distance.
From that point on, it takes the same amount of time to reach Zwolle as
it would to travel back to Arnhem.
What is the speed of the current (in kilometres per hour)?
9
A) 3 B) 4 C) 2 D) 5 E) 6
6
B-problems C
A B
1. A square is divided into six rectangles, all of the same

area. The length of side AB equals 5.
What is the length of side BC?
2. Carl has a large number of apples and pears. He wants to pick ten pieces
of fruit and place them in a row. He wants to do it in such a way that there
is no pear anywhere between two apples. For example, the fruit sequences
AAAAAAAAAA and AAPPPPPPPP are allowed, but AAPPPPPPPA
and APPPPPPPAA are not.
How many sequences can Carl make?
3. If you were to compute
999
| {z. . . 99} × 444
| {z. . . 44}
2014 nines 2014 fours
and then add up all digits of the resulting number, what number would
the final outcome be?
4. We consider 5×5-tables containing a number in each of the 25 cells. The

same number may occur in different cells, but no row or column contains
five equal numbers. Such a table is called pretty if in each row the cell in
the middle contains the average of the numbers in that row, and in each
column the cell in the middle contains the average of the numbers in that
column. The score of a pretty table is the number of cells that contain a
number that is smaller than the number in the cell in the very middle of
the table.
What is the smallest possible score of a pretty table?
7
Solutions
A-problems
A1. B) 2 Suppose that we colour B2 black. Then the surrounding 8

cells cannot be coloured black. Indeed, the cells above, below, to the left,
and to the right of B2 are in the same row or column as B2, while the
other four cells are diagonally adjacent to B2. This way, only row 4 and
column D remain and in each we can colour only one cell black. In total
we can colour no more than 3 cells black. We conclude that B2 cannot be
coloured black.
Similarly, we may deduce that cells B3, C2, and C3 cannot be coloured
black. It follows that in row 2, we can only colour A2 or D2 black. If we
colour A2 black, then in row 3 cell D3 must be coloured black because A3
is in the same column as A2. In rows 1 and 4 we now have no choice but
to colour cells C1 and B4 black. This gives us one solution.
If instead of A2 we colour cell D2 black, then we find a solution where
cells D2, A3, B1, and C4 are coloured black. In total, we have two ways
of choosing the black squares.
2 2 3 3
A2. D) 5 From the given data, we deduce that 5 · 4 = 10 of the carp
3
are yellow females. Since half the carp are female, we find that 12 − 10 = 15
of the carp are red females. Finally, using the fact that three fifths of the
carp are red, we see that 53 − 15 = 52 of the carp are red males.
A3. C) 3 and 5 The journey that visits the pads in the order 3, 6, 1, 4, 7, 2,
and 5 (or in the opposite order), shows that pads 3 and 5 are the starting
pad of a possible journey. We will see that these are the only possible
starting points.
We say that two pads are neighbours if the frog can jump from one pad
to the other (and hence also the other way around). Every intermediate
pad in the frog’s journey must have at least two neighbours: the pad the
frog came from and the pad it will go next. Since pads 3 and 5 have only
one neighbour (pad 6 and pad 2 respectively), these must be the starting
point and end point of the frog’s journey. The other pads must therefore
be intermediate pads.
8
A4. B) 3 Consider the top rim of the paper ring (indicated in bold).
The rim has length 4 × 4 = 16. In the folded state, the rim becomes
rectangle EF GH. Since |AE| = |BF | = |CG| = |DH| = 1, we find that
|AB| + |F G| + |CD| + |EH| = 16 − 4 = 12. These four lengths equal the
length of the sides of square ABCD. It follows that the square has sides
of length 12
4 = 3.
E H
A D
1
B C
4
F G
A5. E) 45 Let n be the number of runners. The number of runners

that finished before Tom equals 21 (n − 1) (half of all runners besides Tom).
The number of runners that finished before Jerry equals 34 (n − 1). Since
exactly 10 runners finished between Tom and Jerry, it follows that 34 (n −
1) − 12 (n − 1) equals 11 (Tom and the 10 runners between Tom and Jerry).
We find that 14 (n − 1) = 11, hence n = 4 × 11 + 1 = 45. There were 45
runners participating in the race.
A6. D) 18 We start by colouring the two indi-

cated tiles at the bottom. This can be done in
six ways: there are three options for the first tile
and for each option there are two possible colours
for the second tile. In the figure, the colours red
(r) and green (g) are chosen.
g
Now that these two tiles are coloured, the colours r
of most of the other tiles are determined as well.
The tile above the red tile can only be blue. The
tile above the green tile must be red and therefore the tile left of the green
tile must be blue. In this way the colours of all tiles, except the two on
the right (white in the figure), are fixed. For these last two tiles, there are
three possible colourings. The upper and lower tile can be coloured either
green and red, or blue and red, or blue and green.
Since each of the six allowed colourings of the first two tiles can be extended
in three ways to a complete colouring, we find a total of 6 × 3 = 18 possible
colourings.
9
A7. C) 25
8 Observe that AM B is a straight angle. This implies that
∠AM D + ∠DM C + ∠CM B = 180◦ . Since triangles AM D and BM C are
equal (three equal sides), we see that ∠CM B = ∠M DA. Hence ∠DM C =
180◦ − ∠AM D − ∠M DA = ∠DAM , because the angles of triangle AM D
sum to 180 degrees. It follows that DM C and DAM are isosceles triangles
with equal apex angles. Hence these two triangles are equal up to scaling.
|CD| |DM | 5
This means that |DM | = |AD| . Therefore, the length of CD equals 8 · 5 =
25
8 .
A8. B) 4 From the mentioned point, it takes the same time to go

42% of the distance upstream and to go 58% of the distance downstream.
This means that the boat is 58
42 times as fast going downstream as going
upstream. If the water flows at a speed of v kilometres per hour, then we
25+v
find 25−v = 58
42 . Hence 58·(25−v) = 42·(25+v), or 1450−58v = 1050+42v.
We find 400 = 100v, hence v = 4.
B-problems
C
B1. 24
5 The six rectangles have equal areas.
f d c a
Rectangles c and d are twice as tall as rectangle
a and therefore also twice as thin. Hence they 5
B
have width 25 . Rectangle e thus has a width of
5 5 b
2 + 2 +5 = 10 and must be half as tall as rectangle
a. This means that rectangle f is precisely 25
times as tall as rectangle a and therefore has a e
5
width of 5/2 = 2. It follows that the square has
sides of length 5 + 52 + 25 + 2 = 12. Because
the square has a height of 25 times the height of
12 24
rectangle a, the height of rectangle a equals |BC| = 5/2 = 5 .
B2. 56 One sequence consists of pears alone. Next, we count se-

quences containing at least one apple. In such a sequence, all apples occur
consecutively, because there can be no pear anywhere between two apples.
If we want to have 8 apples, we can place them in positions 1 through 8,
2 through 9, or 3 through 10. This gives three possible sequences. In this
way we find 1 sequence containing 10 apples, 2 sequences containing 9 ap-
ples, 3 sequences containing 8 apples, and so on, ending with 10 sequences
10
containing 1 apple. In total there are 1 + 2 + 3 + · · · + 10 = 55 sequences
containing at least one apple. The total number of sequences is therefore
55 + 1 = 56.
B3. 18126 A good strategy is to first consider smaller examples. We

find
9×4 = 40 − 4 = 36,
99 × 44 = 4400 − 44 = 4356,
999 × 444 = 444000 − 444 = 443556,
9999 × 4444 = 44440000 − 4444 = 44435556.
The pattern should be clear. To solve the problem, observe that 999 . . . 99 =
1000 . . . 00 − 1, where the first number has 2014 zeroes. The product is
therefore equal to
444
| {z . . . 00} − |444 {z
. . . 44} |000 {z . . . 44} = 444
| {z. . . 44} 3 |555 {z
. . . 55} 6.
2014 fours 2014 zeroes 2014 fours 2013 fours 2013 fives
Adding these digits, we obtain 2013·4+3+2013·5+6 = 2013·9+9 = 18126.
B4. 3 We first show that every pretty ta- 4 4 3 4 0

ble has a score of at least 3. Consider such a table 4 4 3 4 0
and let a be the number at the very middle. The 3 3 0 3 −9
five numbers in the middle row have an average 4 4 3 4 0
of a and are not all equal to a. Hence at least 0 0 −9 0 −36
one of these numbers must be smaller than a. Similarly, at least one of the
numbers in the middle column must be smaller than a. Let this number
be b. Since b is the average of the numbers in its row, one of the numbers
in that row must be smaller than b, and hence also smaller than a. Thus
the table contains at least three different cells that have a number smaller
than the number in the very middle. Its score is therefore at least 3.
In the figure on the right you can find a pretty table with a score equal to
3. It follows that 3 is the smallest possible score.
11
Problems
B-problems
B1. Brenda is filling pouches from an unlimited supply of red and blue marbles.
In each pouch she puts more red than blue marbles, and each pouch can
contain at most 25 marbles. For example, she can make a pouch containing
6 red marbles and 2 blue marbles, or a pouch containing 1 red marble and
0 blue marbles.
How many differently filled pouches can she make in total?
B2. In the figure an equilateral triangle ABC C

is drawn with points D and E on sides
BC and AC. When folding the triangle
along the line DE, the vertex C is folded D
onto point C 0 on line AB. Furthermore,
∠DC 0 B = 90◦ holds. E
?
What is the size of ∠DEC 0 ?
A C0 B
B3. For how many of the integers n from 1 up to and including 100 is the
number 8n + 1 a perfect square?
B4. Evan and nine other people are standing in a circle. All ten of them think
of an integer (that may be negative) and whisper their number to both
of their neighbours. Afterwards, they all state the average of the two
numbers that were whispered in their ear. Evan states the number 10,
his right neighbour states the number 9, the next person along the circle
states the number 8, and so on, finishing with Evan’s left neighbour who
states the number 1.
What number did Evan have in mind?
12
B5. The numbers of dots on two opposite faces of a die
always sum to 7. Nine identical dice are glued in
a 3×3-array. This is done in such a way that when
?
two faces are glued together, they must contain
the same number of dots. In the figure you can
see the top view of the array. For five of the dice
the number of dots is not shown.
What number of dots must be on the place of the
question mark?
C-problems For the C-problems not only the answer is important; you also have to
describe the way you solved the problem.
C1. We are given a quadrilateral ABCD. The midpoint of AB is denoted by

E and the midpoint of CD is denoted by F . The segments AF , BF , CE,
DE, and EF divide the C
F
quadrilateral into eight tri- D
angles. The areas of these r
c
triangles are denoted by the q s
b d
letters from a to d and p to s,
see the figure.
p
a
(a) Prove that a + d = p + q.
A E B
(b) Prove that a + r = c + p.
(c) Prove that b + s = d + q.
C2. A positive integer n is called a jackpot number if it has the following prop-
erty: there exists a positive integer k consisting of two or more digits, all of
which are equal (such as 11111 or 888), for which the product n · k is again
a number consisting of equal digits. For example, 3 is a jackpot number
because 3 · 222 = 666.
(a) Find a jackpot number consisting of 10 digits and prove that it is a
jackpot number.
(b) Prove that 11 is not a jackpot number.
(c) Determine whether 143 is a jackpot number and prove that your an-
swer is correct.
13
Solutions
B-problems
B1. 169 Brenda can make 25 differently filled pouches without blue
marbles as she can put 1 to 25 red marbles in a pouch. There are 23
differently filled pouches possible containing 1 blue marble because 2 to 24
red marbles may be added. Using 2 blue marbles there are 21 possibilities,
namely by adding 3 to 23 red marbles. In total, there are 25+23+. . .+1 =
169 differently filled pouches that Brenda can make.
B2. 45◦ First, we notice that triangles C

0
CDE and C DE are each other’s mirror im-
age and hence have equally sized angles. In
particular, we have ∠DC 0 E = ∠DCE =
D
60◦ . Furthermore, we have ∠CED = ∠DEC 0,
see the figure.
E
From ∠AC 0 B = 180◦ it follows that ∠AC 0 E =
180◦ − 90◦ − 60◦ = 30◦ . The sum of the an- 60◦
gles in triangle AC 0 E is 180◦ and hence we
find ∠AEC 0 = 180◦ − 60◦ − 30◦ = 90◦ . A C0 B
◦ 0 ◦ ◦ ◦
From ∠AEC = 180 it follows that ∠CEC = 180 − 90 = 90 .
We conclude that ∠DEC 0 = 12 ∠CEC 0 = 45◦ .
B3. 13 If 8n + 1 is the square of an integer, then this integer must

be odd. Conversely, the square of an odd integer is always a multiple of
8 plus 1. Indeed, suppose that k is an odd integer, then we may write
k = 2` + 1 for an integer `. We see that
k 2 = (2` + 1)2 = 4`2 + 4` + 1 = 4`(` + 1) + 1.
Because either ` or ` + 1 is even, we deduce that 4`(` + 1) is divisible by

8. Hence, k 2 is a multiple of 8 plus 1.
As a result, we only need to determine the number of odd squares x for
which 8 · 1 + 1 ≤ x ≤ 8 · 100 + 1. These are the squares 32 = 9, 52 = 25 up
to and including 272 = 729, because 292 = 841 is greater than 801. Thus,
the number of squares of the desired form is 13.
14
B4. 5 The number that Evan came up with is denoted by c10 , the
number of his right neighbour is denoted by c9 , continuing in this way until
the left neighbour of Evan, whose number is denoted by c1 . From the data
we deduce that
c10 + c8 = 2 · 9 = 18,
c8 + c6 = 2 · 7 = 14,
c6 + c4 = 2 · 5 = 10,
c4 + c2 = 2 · 3 = 6,
c2 + c10 = 2 · 1 = 2.
Adding up these equations yields 2(c2 + c4 + c6 + c8 + c10 ) = 50, hence
c2 + c4 + c6 + c8 + c10 = 25.
Finally, we find c10 = (c2 + c4 + c6 + c8 + c10 ) − (c2 + c4 ) − (c6 + c8 ) =
25 − 6 − 14 = 5.
B5. 3 The numbers of dots on opposite faces of a die will be called

complementary. Together, they always add to 7. Consider a pair of dice
that touch in faces with equal numbers of dots. We still allow them to
rotate with respect to each other. When rotating the dice, their faces
show the same numbers of dots, but in reverse cyclic order.
Consider the situation where the numbers of dots on the top faces of the
dice are complementary, say a and 7 − a. This is depicted in the figure on
the left. On the four faces around the gluing axis, the left hand die will
have a, b, 7 − a, and 7 − b dots in this order, for some b. Hence, the right
hand die has these numers in reverse cyclic order: 7 − a, b, a, and 7 − b
dots on the corresponding faces. It follows that the dice have the same
number of dots on the front face, namely b, and the same number of dots
on the back face, namely 7 − b.
Conversely, if two glued dice have the same number of dots on the front
faces (or back faces), then the numbers of dots on the top faces must be
complementary.
?
g h
a 7_ a
7_ b e f
7_ a b b c d
15
We will now apply this to the six dice in the first two columns of the
3 × 3 array, see the figure on the right. The two dice in the third row
show complementary numbers of dots on their top faces, namely 2 and 5.
Therefore, the numbers of dots c and d, on the faces where the dice are
glued to dice in the second row, must be equal. This in turn implies that
the numbers of dots e and f on the top faces must be complementary.
Therefore, the numbers g and h are equal. Finally, the top faces of the
dice in the first row must show complementary numbers of dots. Hence,
there must be 3 dots on the place of the question mark.
C-problems
C1. (a) First, we notice that a + d is the area of triangle AEF and p + q is the
area of triangle BEF . The base AE of triangle AEF has the same
length as the base BE of triangle BEF . Because the two triangles
have the same height, they must also have equal areas.
(b) Here we use that triangles DEF and CEF have bases of the same
length (|DF | = |CF |), and equal corresponding heights. Hence, they
have equal areas. This implies that c + d = q + r. Subtracting the
equation of part (a) yields c − a = r − p, or a + r = c + p, as required.
(c) The heights of triangles AED, BEC and ABF with respect to the
bases AE, BE, and AB are denoted by x, y, and z. Because F is the
midpoint of CD, the height z is the average of the heights x and y,
in formulas: x+y 1
2 = z. The area of triangle AED is 2 · x · |AE|, the
1
area of triangle BEC is 2 · y · |BE|, and the area of triangle ABF is
1
2 · z · |AB|. Because E is the midpoint of AB we have |AE| = |BE|
and |AB| = 2 · |AE|. The sum of the areas of triangles AED and
BEC is thus equal to 12 · (x + y) · |AE| = z · |AE|, while the area of
triangle ABF is equal to 21 · z · 2|AE| = z · |AE|. Hence, these areas
are equal and we find a + b + p + s = a + d + p + q. By subtracting
a + p on both sides of the equation, we find b + s = d + q, as required.
16
C2. (a) Consider the ten digit number 1001001001. If we multiply it with 111,
we get the number 111111111111 which consists of ones only. Hence,
the number 1001001001 is a jackpot number.
(b) Let k be a number of at least two digits, all of which are equal, say
equal to a. Remark that a 6= 0. We have to prove that the digits of
the number
11k = k + 10k = a · · · a + a · · · a0
are not all equal.

The last digit of 11k is a. We will show that the second last digit of
11k is unequal to a. There are two cases. If a ≤ 4, then the second
last digit of 11k is equal to a + a. This is unequal to a as a 6= 0. If
a ≥ 5, then the second last digit of 11k is equal to a + a − 10. This is
unequal to a as a 6= 10. We conclude that 11 is not a jackpot number.
(c) That 143 is a jackpot number follows directly from the fact that 143 ·
777 = 111111.
17
Problems
1. Determine all triples (a, b, c), where a, b, and c are positive integers that
satisfy a ≤ b ≤ c and abc = 2(a + b + c).
D C
2. Version for junior students

Let ABCD be a parallelogram with an
acute angle at A. Let G be a point on G
the line AB, distinct from B, such that A B
|CG| = |CB|. Let H be a point on the line
BC, distinct from B, such that |AB| =
|AH|. H
Prove that triangle DGH is isosceles.
C
2. Version senior students
W
On the sides of triangle ABC, isosce-
V
les right-angled triangles AU B, CV B,
U
and AW C are placed. These three tri-
angles have their right angles at ver-
tices U , V , and W , respectively. Tri-
angle AU B lies completely inside tri-
angle ABC and triangles CV B and A B
AW C lie completely outside ABC. See
the figure.
Prove that quadrilateral U V CW is a parallelogram.
3. At a volleyball tournament, each team plays exactly once against each

other team. Each game has a winning team, which gets 1 point. The
losing team gets 0 points. Draws do not occur. In the final ranking, only
one team turns out to have the least number of points (so there is no
shared last place). Moreover, each team, except for the team having the
least number of points, lost exactly one game against a team that got less
points in the final ranking.
a) Prove that the number of teams cannot be equal to 6.

b) Show, by providing an example, that the number of teams could be
equal to 7.
18
4. A quadruple (p, a, b, c) of positive integers is called a Leiden quadruple if
• p is an odd prime number,
• a, b, and c are distinct and
• ab + 1, bc + 1 and ca + 1 are divisible by p.
a) Prove that for every Leiden quadruple (p, a, b, c) we have p + 2 ≤

a+b+c
3 .
b) Determine all numbers p for which a Leiden quadruple (p, a, b, c) exists
with p + 2 = a+b+c
3 .
5. We consider the ways to divide a 1 by 1 square into rectangles (of which

the sides are parallel to those of the square). All rectangles must have the
same circumference, but not necessarily the same shape.
a) Is it possible to divide the square into 20 rectangles, each having a

circumference of 2.5?
b) Is it possible to divide the square into 30 rectangles, each having a
circumference of 2?
19
Solutions
1. Suppose that (a, b, c) is a solution. From a ≤ b ≤ c it follows that abc =

2(a + b + c) ≤ 6c. Dividing by c yields ab ≤ 6. We see that a = 1 or a = 2,
because from a ≥ 3 it would follow that ab ≥ a2 ≥ 9.
We first consider the case a = 2.
From ab ≤ 6 it follows that b = 2 or b = 3. In the former case, the equation
abc = 2(a + b + c) yields 4c = 8 + 2c and hence c = 4. It is easy to check
that the triple (2, 2, 4) we got is indeed a solution. In the latter case, we
have 6c = 10 + 2c, hence c = 52 . Because c has to be an integer, this does
not give rise to a solution.
Now we consider the case a = 1.
We get that bc = 2(1 + b + c). We can rewrite this equation to obtain
(b − 2)(c − 2) = 6. Remark that b − 2 cannot be negative (and hence
also c − 2 cannot be negative). Otherwise, we would have b = 1, yielding
(1 − 2)(c − 2) = 6, from which it would follow that c = −4. However, c has
to be positive.
There are only two ways to write 6 as a product of two non-negative
integers, namely 6 = 1×6 and 6 = 2×3. This gives rise to two possibilities:
b − 2 = 1 and c − 2 = 6, or b − 2 = 2 and c − 2 = 3. It is easy to check
that the corresponding triples (1, 3, 8) and (1, 4, 5) are indeed solutions.
Thus, the only solutions are (2, 2, 4), (1, 3, 8), and (1, 4, 5).
D C
We know that ∠ABH = ∠CBG, because
these are opposite angles. Because trian-
gles ABH and CBG are isosceles, we have
G
∠AHB = ∠ABH and ∠CBG = ∠CGB.
A B
Triangles ABH and CBG are similar (AA)
and hence we have ∠BAH = ∠BCG. Be-
cause ABCD is a parallelogram, we have
H
∠DAB = ∠DCB and hence ∠DAH =
∠DAB + ∠BAH = ∠DCB + ∠BCG = ∠DCG holds. Because ABCD is
a parallelogram, we have |CD| = |AB| = |AH| and |AD| = |BC| = |CG|.
Therefore, triangles DAH and GCD are congruent (SAS) and we have
|DH| = |DG|. In other words, triangle DGH is isosceles.
20
2. Version for senior students C
Because triangle AU B is isosceles with top
W
angle ∠AU B = 90◦ , we have ∠U AB =
45◦ . In the same way, we have ∠CAW = V
U
45◦ . Combining these two equalities, we
find ∠W AU = 45◦ + ∠CAU = ∠CAB.
By the Pythagorean theorem, we find 2|AW |2 =
2 2 2
|AW
√ | +|CW | = |AC| and hence |AW | =
1
2 √2·|AC|. In the same way we find |AU | = A B
1
2 2 · |AB|. Hence, √
triangles W AU and CAB are similar (SAS) with scale
√
factor |AW | 1 |AU | 1
|AC| = 2 2 = |AB| . In particular, we find |W U | = 2 2 · |BC| =
|CV |.
In the same√ way, we see that triangles V BU and CBA are similar and that
|V U | = 21 2 · |AC| = |CW |. It follows that in quadrilateral U V CW the
opposite sides have equal lengths, hence U V CW is a parallelogram.
3. a) Suppose that the number of teams is 6. We shall derive a contradic-

tion.
First remark that the number of games equals 6×5 2 = 15. Hence, the
total number of points also equals 15.
Let team A be the (only) team with the lowest score. Team A has
at most 1 point, because if team A had 2 or more points, then each
of the other five teams would have at least 3 points, giving a total
number of points that is at least 2 + 3 + 3 + 3 + 3 + 3 = 17. Each team
on the second last place in the ranking has lost to team A, because
this is the only team with a lower score. Hence, team A also has at
least 1 point. We deduce that A has exactly 1 point and that there is
exactly one team, say team B, in the second last place in the ranking.
Team B has at least 2 points and the remaining four teams, teams
C, D, E and F , each have at least 3 points. The six teams together
have at least 1 + 2 + 3 + 3 + 3 + 3 = 15 points. If team B had more
than 2 points, or if any of the teams C through F had more than
3 points, then the total number of points would be greater than 15,
which is impossible. Hence, team B has exactly 2 points and teams C
through F each have exactly 3 points. The four teams C through F
each lost to a team having a lower score (team A or team B). Hence,
together, team A and team B must have won at least 4 games. This
contradicts the fact that together they have only 1 + 2 = 3 points.

21
b) In the table below there is a possible outcome for 7 teams called A
through G. In the row corresponding to a team, crosses indicate wins
against other teams. Row 2, for example, indicates that team B won
against teams C and D and obtained a total score of 2 points. Each
team (except A) has indeed lost exactly one match against a team
with a lower score. These matches are indicated in bold.
A B C D E F G Score
A - X 1
B - X X 2
C X - X X 3
D X X - X 3
E X X X - X 4
F X X X - X 4
G X X X X - 4

4. a) Without loss of generality, we may assume that a < b < c. The inte-
gers a and c are not divisible by p because that would imply that ac+1
is a multiple of p plus 1, hence not divisible by p. Since bc+1 and ac+1
are both divisible by p, their difference (bc+1)−(ac+1) = (b−a)c is di-
visible by p as well. Hence, since c is not divisible by p, it must be the
case that b − a is divisible by p. Similarly, (ac + 1) − (ab + 1) = a(c − b)
is divisible by p and since a is not divisible by p, this implies that c−b
is divisible by p.
Thus, we find that b = a + (b − a) ≥ a + p and c = b + (c − b) ≥ a + 2p.
We have a ≥ 2. Indeed, suppose that a = 1. Then, both integers

b + 1 = ab + 1 and b − 1 = b − a are divisible by p, which implies that
their difference (b + 1) − (b − 1) = 2 is divisible by p as well. However,
p is an odd prime and can therefore not divide 2.
Using a ≥ 2, b ≥ a + p, and c ≥ a + 2p, we conclude that
a+b+c a + (a + p) + (a + 2p)
≥ = p + a ≥ p + 2.
3 3

22
b) Again, we may assume that a < b < c. In part a) we have seen
that a+b+c
3 ≥ a+(a+p)+(a+2p)
3 = p + a ≥ p + 2. We can only have
a+b+c
3 = p + 2 if b = a + p, c = a + 2p, and a = 2. Since ab +
1 = 2(2 + p) + 1 = 2p + 5 must be divisible by p, it follows that
5 is divisible by p. We conclude that p = 5, b = 7, and c = 12.
The quadruple (p, a, b, c) = (5, 2, 7, 12) is indeed a Leiden quadruple,
because ab + 1 = 15, ac + 1 = 25, and bc + 1 = 85 are all divisible by
p.
We conclude that p = 5 is the only number for which there is a Leiden
quadruple (p, a, b, c) that satisfies a+b+c
3 = p + 2.
5. a) Consider a rectangle with sides of length a ≤ b inside the square.

Since b ≤ 1 and 2a + 2b = 25 hold, we see that a ≥ 14 . The area of
the rectangle equals ab and is therefore at least 41 × 14 = 16
1
. Hence,
we can have no more than 16 rectangles inside the square without
creating overlaps.
b) A solution is sketched in the figure below. The four outer rectangles,
A through D, are equal with the shorter side having length x, and
the longer side having length 1 − x. Together they leave uncovered
a square area with sides of length 1 − 2x. This area is then tiled by
26 equal rectangles. These have sides of length 1 − 2x and 1−2x26 , and
54
therefore have a circumference of 26 (1 − 2x). To obtain a circumfer-
1
ence of length 2, we take x = 54 .
1−x
x A
1
2
3
.
B . D
.
24
25
26
C
23
Problems
1. Let m and n be positive integers such that 5m + n is a divisor of 5n + m.

Prove that m is a divisor of n.
2. Given are positive integers r and k and an infinite sequence of positive

integers a1 ≤ a2 ≤ . . . such that arr = k + 1. Prove that there is a t
satisfying att = k.
3. Let n ≥ 2 be a positive integer. Each square of an n × n-board is coloured

red or blue. We put dominoes on the board, each covering two squares
of the board. A domino is called even if it lies on two red or two blue
squares and colourful if it lies on a red and a blue square. Find the largest
positive integer k having the following property: regardless of how the
red/blue-colouring of the board is done, it is always possible to put k non-
overlapping dominoes on the board that are either all even or all colourful.
4. In a triangle ABC the point D is the intersection of the interior angle

bisector of ∠BAC and side BC. Let P be the second intersection point
of the exterior angle bisector of ∠BAC with the circumcircle of 4ABC.
A circle through A and P intersects line segment BP internally in E and
line segment CP internally in F . Prove that ∠DEP = ∠DF P .
(x2 + y 2 )f (xy) = f (x)f (y)f (x2 + y 2 )
for all real numbers x and y.
24
Solutions
1. There is a positive integer k with (5m + n)k = 5n + m. Hence, 5km − m =
5n − kn, or (5k − 1)m = (5 − k)n. The left hand side is positive, hence also
the right hand side is positive, which yields k < 5. If k = 1, then 4m = 4n,
hence m = n, hence m | n. If k = 2, then 9m = 3n, hence 3m = n, hence
m | n. If k = 3, then 14m = 2n, hence 7m = n, hence m | n. If k = 4,
then 19m = n, hence m | n. We conclude that in all cases we have that
m | n.
2. We will prove this by contradiction. Suppose that such a t does not exist.
If ak = 1, then akk = k would hold, contradicting our assumption. Hence,
ak ≥ 2. We will now prove by induction to i that aik ≥ i + 1. We just
proved the base case. Now suppose that for certain i ≥ 1 we have that
aik ≥ i + 1. Then we also have that a(i+1)k ≥ i + 1. If a(i+1)k = i + 1, then
(i+1)k
a(i+1)k = k, which is a contradiction. Hence, a(i+1)k ≥ i + 2. This finishes
the induction. Now take i = ar , then we have aar k ≥ ar + 1. Moreover,
because r = ar (k + 1) we have ar = aar (k+1) ≥ aar k ≥ ar + 1, which is a
contradiction.
2
3. We will prove that k = b n4 c is the largest possible integer.
2
Suppose that n is even. Then it is possible to cover the board with n2
2
dominoes (without considering the colours). Because there are n2 domi-
2 2
noes, each of which is either colourful or even, there are at least b n4 c = n4
2 2
colourful or at least b n4 c = n4 even dominoes.
2
When n is odd, we can cover the board with n 2−1 dominoes. (Notice
that this number is an even integer.) Of these dominoes either at least
n2 −1 2 2 2
4 = b n4 c are colourful, or at least n 4−1 = b n4 c are even. This proves

2 2
that it is always possible to put at least b n4 c colourful or at least b n4 c
even dominoes on the board.
We will now prove that it is possible to create a colouring of the board
2
with blue and red squares, such that no more than b n4 c even and no more
2
than b n4 c colourful dominoes can be put on the board.
Colour the squares of the board in the colours white and black like the
squares on a chess board, such that the lowerleft square is white. If n is
2
even, there are equally many white as black square, namely n2 . If n is
odd, there is one black square less and the number of black squares equals
n2 −1 2
2 = b n2 c. In both cases this is an even number of squares, as for odd n

we have that n2 ≡ 1 mod 4. Now colour half of the black squares red and
25
2
all the other squares blue. Then there are b n4 c red squares, hence we can
2
put at most b n4 c non-overlapping colourful dominoes on the board as each
of these dominoes covers one red square. An even domino cannot cover
two red squares, because there are no pairs of adjacent squares coloured
red. Hence, it must cover two blue squares. One of these blue squares must
have been black, hence the number of even dominoes is at most the number
2
of black-blue squares and that is b n4 c. Hence of both the colourful as the
2
even dominoes we can put at most b n4 c simultaneously on the board.
2
We conclude that the maximum k is indeed k = b n4 c.
4. We consider the configuration in which the points A, C, B, and P lie in

that order on the circumcircle. The other case is analogous.
By the inscribed angle theorem for the circumcircle of 4ABC, we have
∠ABE = ∠ABP = ∠ACP = ∠ACF.
Moreover, by the inscribed angle theorem for the circle through A, P , E,

and F :
∠AEB = 180◦ − ∠AEP = 180◦ − ∠AF P = ∠AF C.
We therefore see (AA) that 4ABE ∼ 4ACF . From this, it follows that
|AB| |BE|
= .
|AC| |CF |
The angle bisector theorem then implies that
|AB| |DB| |BE| |DB|

= , so therefore = . (1)
|AC| |DC| |CF | |DC|
Choose Z on P A such that A lies between P and Z. As AP is the external

angle bisector of ∠BAC, we have ∠P AB = ∠ZAC = 180◦ − ∠P AC. So
using the inscribed angle theorem and the fact that ACBP is a cyclic
quadrilateral, we see that
∠DCF = ∠P CB = ∠P AB = 180◦ − ∠P AC = ∠P BC = ∠EBD.
If we combine this with (1), we obtain 4BED ∼ 4CF D (SAS). Therefore

∠BED = ∠CF D, hence we have ∠DEP = 180◦ − ∠BED = 180◦ −
∠CF D = ∠DF P .
26
5. Substituting x = y = 0 gives 0 = f (0)3 , so f (0) = 0. We consider two
more cases; either f has another zero, or f has no other zeroes. In the first
case there is some a 6= 0 such that f (a) = 0. Substituting x = a then gives
(a2 + y 2 )f (ay) = 0. As a2 + y 2 > 0 (since a 6= 0) we find that f (ay) = 0
for all y. As ay can attain all values in R, it follows that f (x) = 0 for all x;
i.e. f is the constant function with value 0. Note that this function indeed
satisfies the given equation.
So now consider the second case, in which f (x) 6= 0 for all x 6= 0. Substi-
tuting x 6= 0 and y = 1 then gives (x2 + 1)f (x) = f (x)f (1)f (x2 + 1) and
f (x) 6= 0, so we can divide this equation by f (x). Therefore (x2 + 1) =
2
f (1)f (x2 + 1). Let c = f (1). Note that c 6= 0, so f (x2 + 1) = x c+1 . Since
x
x2 + 1 attains all reals that are at least 1, we find f (x) = c for all x > 1.
Substituting x = y = 2 now gives (4 + 4)f (4) = f (2)f (2)f (4 + 4). Since
we know the values f (x) takes for x > 1, we know the values of f (2), f (4),
and f (8). Therefore 8 · 4c = 2c · 2c · 8c , or equivalently, 1c = c13 . We deduce
that c2 = 1, so c = 1 or c = −1. As c2 = 1, it follows that f (x) = cx for
all x > 1.
Let x > 1, and substitute y = x1 . This gives (x2 + x12 )f (1) = f (x)f ( x1 )f (x2 +
1 2 1 2 1
x2 ). Since x > 1, we have x + x2 > 1, so we deduce that (x + x2 )c = xc ·
1 1 1 1 1
2
f ( x )·c(x + x2 ), or equivalently, 1 = xc·f ( x ). Therefore f ( x ) = xc = c· x1 .
We deduce that f (x) = cx for all x > 0 with x 6= 1. Since we also have
f (1) = c, it follows that f (x) = cx for all x > 0.
Substituting x = y = −1 gives 2f (1) = f (−1)2 f (2). Since f (1) = c
and f (2) = 2c, it follows that 2c = f (−1)2 · 2c, so since c 6= 0 we have
f (−1)2 = 1. Therefore either f (−1) = 1 or f (−1) = −1.
Let x > 0. Substituting y = −1 then gives (x2 +1)f (−x) = f (x)f (−1)f (x2 +
1), so (x2 + 1)f (−x) = cx · f (−1) · c(x2 + 1), or equivalently, f (−x) =
c2 xf (−1) = xf (−1). Let d = f (−1). Then f (x) = −dx for all x < 0,
where d2 = 1.
So aside from f (x) = 0 there are four more possible solutions, namely
f (x) = x, f (x) = −x, f (x) = |x|, and f (x) = −|x|. We first check
f (x) = tx with t = ±1. The left hand side then reads (x2 + y 2 ) · txy, and
the right hand side then reads tx · ty · t(x2 + y 2 ). As t2 = 1, the left hand
side and the right hand side are equal, so these two functions are indeed
solutions of the functional equation.
Next, we check the two functions f (x) = t|x|, with t = ±1. Now the left
hand side reads (x2 + y 2 ) · t|xy|, and the right hand side reads t|x| · t|y| ·
t|x2 + y 2 |. Since x2 + y 2 = |x2 + y 2 |, |xy| = |x||y|, and t2 = 1, the left hand
side and the right hand side are equal. So these two functions are indeed
solutions of the functional equations. Therefore there are five solutions;
f (x) = 0, f (x) = x, f (x) = −x, f (x) = |x| and f (x) = −|x|.
27
Problems
1. In a quadrilateral ABCD we have ∠A = ∠C = 90◦ . Let E be a point

in the interior of ABCD. Let M be the midpoint of BE. Prove that
∠ADB = ∠EDC if and only if |M A| = |M C|.
2. Find all polynomials P (x) with real coefficients such that the polynomial
Q(x) = (x + 1)P (x − 1) − (x − 1)P (x)
is constant.
3. Let n be a positive integer. Consider sequences a0 , a1 , . . . , ak and b0 , b1 , . . . , bk

such that a0 = b0 = 1 and ak = bk = n and such that for all i such that
1 ≤ i ≤ k, we have that (ai , bi ) is either equal to (1 + ai−1 , bi−1 ) or
(ai−1 , 1 + bi−1 ). Consider for 1 ≤ i ≤ k the number
(
ai if ai = ai−1 ,
ci =
bi if bi = bi−1 .
Show that c1 + c2 + · · · + cn = n2 − 1.
4. Let Γ1 and Γ2 be circles – with respective centres O1 and O2 – that intersect

each other in A and B. The line O1 A intersects Γ2 in A and C and the
line O2 A intersects Γ1 in A and D. The line through B parallel to AD
intersects Γ1 in B and E. Suppose that O1 A is parallel to DE. Show that
CD is perpendicular to O2 C.
5. For a positive integer n, we define Dn as the largest integer that is a divisor

of an + (a + 1)n + (a + 2)n for all positive integers a.
1. Show that for all positive integers n, the number Dn is of the form
3k with k ≥ 0 an integer.
2. Show that for all integers k ≥ 0 there exists a positive integer n such
that Dn = 3k .
28
Solutions
1. Let N be the midpoint of BD. By Thales’s Theorem the circle with

diameter BD also passes through A and C, and N is the centre of this
circle. Moreover, we have M N k DE; if E doesn’t lie on BD, then M N
is a midparallel in triangle BDE, and if E does lie on BD, then M N and
DE are equal.
The claim that |AM | = |CM | is equivalent to that of M being on the
perpendicular bisector of AC. As said bisector passes through N , the
above claim is equivalent to M N ⊥ AC. Note that this is the case if and
only if DE ⊥ AC. Let T be the intersection point of DE and AC; then
|AM | = |CM | if and only if ∠DT C = 90◦ .
Using the sum of angles of a triangle, we see that ∠DT C = 180◦ −∠T DC −
∠DCT = 180◦ − ∠EDC − ∠DCA. We have ∠DCA = 90◦ − ∠ACB =
90◦ − ∠ADB, where we used in the last equality that ABCD is a cyclic
quadrilateral. Therefore ∠DT C = 180◦ − ∠EDC − (90◦ − ∠ADB) =
90◦ − ∠EDC + ∠ADB. Now it immediately follows that ∠ADB = ∠EDC
if and only if ∠DT C = 90◦ , which we already know to be equivalent to
|AM | = |CM |.
Note that this proof does not depend on the configuration.
2. Suppose that P (x) is a constant polynomial, say P (x) = a with a ∈ R.

Then
Q(x) = (x + 1)a − (x − 1)a = ax + a − ax + a = 2a,
which is constant. So every constant P (x) satisfies the condition.
Now we assume that P is not constant. We can then write P (x) = an xn +
an−1 xn−1 +· · ·+a1 x+a0 with n ≥ 1 and an 6= 0. Consider the xn -coefficient
in Q(x). It is the sum of the xn -coefficients of xP (x−1), P (x−1), −xP (x),
and P (x). In the first of these, this coefficient equals an−1 − nan , in the
second one it equals an , in the third one it equals −an−1 and the fourth
one it equals an . Summing these gives
an−1 − nan + an − an−1 + an = (2 − n)an .
But in Q(x) this coefficient must be equal to 0. Since an 6= 0, it follows

that n = 2. Therefore P (x) = a2 x2 + a1 x + a0 with a2 6= 0.
Now consider the constant coefficients of Q(x). It is the sum of the con-
stant coefficients of xP (x − 1), P (x − 1), −xP (x), and P (x). These are
29
respectively 0, a2 − a1 + a0 , 0, and a0 . These sum up to a2 − a1 + 2a0 . On
the other hand, we can compute Q(1);
Q(1) = 2P (0) − 0 = 2a0 .
As Q(x) is constant, it follows that Q(1) is the constant coefficient of Q,

so 2a0 = a2 − a1 + 2a0 . Therefore a2 = a1 . So P (x) is of the form
bx2 + bx + a = bx(x + 1) + a with a, b ∈ R and b 6= 0. To see whether these
polynomials indeed satisfy the condition, we substitute this form for P (x).

Q(x) = (x + 1) b(x − 1)x + a − (x − 1)(bx(x + 1) + a)
= (x − 1)x(x + 1)b + (x + 1)a − (x − 1)x(x + 1)b − (x − 1)a
= 2a.
This is indeed constant, so all such P (x) satisfy the condition. In fact,
every constant polynomial is also of this form, with b = 0. Therefore
the polynomials satisfying the condition are precisely those of the form
P (x) = bx2 + bx + a with a, b ∈ R.
3. We prove by induction on j that c1 + . . . + cj = aj bj − 1. For j = 1 this

reads c1 = a1 b1 − 1, which is true since (a1 , b1 ) ∈ {(1, 2), (2, 1)}. Now
suppose that c1 + · · · + ci−1 = ai−1 bi − 1 − 1. We assume without loss of
generality that (ai , bi ) = (ai−1 , 1 + bi−1 ), so that ai = ai−1 and therefore
ci = ai−1 . We see that
(c1 + . . . + ci−1 ) + ci = (ai−1 bi−1 − 1) + ai−1 = ai−1 (bi−1 + 1) − 1 = ai bi − 1,
which completes the induction argument. Substituting j = k, it now fol-

lows that c1 + · · · + ck = ak bk − 1 = n2 − 1.
30
4. We consider only the configuration in which A, B, E, and D lie in that
order on a circle, in which O1 , A, and C lie in that order on a line, and O2 ,
A, and D lie in that order on a line; the proof is analogous for the other
configurations.
As ABED is a cyclic quadrilateral, we have ∠BED = 180◦ − ∠DAB.
Moreover, using the parallel lines, we see that ∠BED = ∠DAO1 and as
|O1 A| = |O1 D|, we have ∠DAO1 = ∠ADO1 . We deduce that 180◦ −
∠DAB = ∠ADO1 . Therefore DO1 and AB are parallel.
We already know that ∠ADO1 = ∠DAO1 . Since |O2 A| = |O2 C|, it follows
that ∠DAO1 = ∠O2 AC = ∠O2 CA. So ∠O2 DO1 = ∠ADO1 = ∠O2 CA =
∠O2 CO1 , so O1 DCO2 is a cyclic quadrilateral.
The line O1 O2 is the perpendicular bisector AB, therefore is also perpen-
dicular to DO1 , as this line is parallel to AB. Therefore ∠O2 O1 D = 90◦ .
As O1 DCO2 is a cyclic quadrilateral, we now also have ∠O2 CD = 90◦ .
5. 1. Let p be a prime and suppose that p divides Dn . Then p divides
(a+1)n +(a+2)n +(a+3)n − an +(a+1)n +(a+2)n = (a+3)n −an

for all positive integers a.

Substituting a = p, then it follows that p | (p + 3)n − pn , i.e. we
have (p + 3)n − pn ≡ 0 mod p. This simply reads 3n ≡ 0 mod p, so
p = 3. We deduce that Dn only contains prime factors equal to 3,
and therefore is of the form 3k with k ≥ 0 an integer.
2. For k = 0 we take n = 2. We have 12 +22 +32 = 14 and 22 +32 +42 =
29 and these two numbers are coprime, so D2 = 1. Now assume that
k ≥ 1. We show that Dn = 3k for n = 3k−1 .
We first show that 1n + 2n + 3n for n = 3k−1 is divisible by 3k , but
not by 3k+1 . For k = 1 we have n = 1, and indeed, we see that
1 + 2 + 3 = 6 is divisible by 3 but not by 32 . For k ≥ 2 we have n > k,
so that 3n is divisible by 3k+1 . So we are reduced to showing that
1 + 2n for n = 3k−1 is divisible by 3k but not by 3k+1 . We show this
by induction on k. For k = 2 we have n = 3, so indeed 1 + 8 = 9 is
divisible by 9, but not by 27. Let m ≥ 2, and suppose we have proved
our claim for k = m. Let n = 3m−1 . Then 1 + 2n is divisible by 3m ,
but not by 3m+1 . It suffices to show that 1 + 23m is divisible by 3m+1 ,
but not by 3m+2 . Write 1 + 2n = 3m c with 3 - c. Then 2n = 3m c − 1,
so
3
1 + 23n = 1 + (3m c − 1) = 33m c3 − 3 · 32m c2 + 3 · 3m c.
31
Modulo 3m+2 , this is congruent to 3m+1 c, and since 3 - c, it follows
that this is divisible by 3m+1 , but not by 3m+2 , as desired. This
completes our inductive argument.
Next we show that for n = 3k−1 , we have that (a+3)n −an is divisible
by 3k for all positive integers a. Again, we prove this by induction on
k. For k = 1 we have n = 1, so indeed we see that (a + 3) − a = 3
is divisible by 3 for all positive integers a. Now suppose that m ≥ 1,
and suppose that we proved our claim for k = m. Let n = 3m−1 .
Then (a + 3)n − an is divisible by 3m for all positive integers a, so
we can write (a + 3)n = an + 3m c for some integer c. Taking third
powers of both sides then yields
(a + 3)3n = a3n + 3a2n · 3m c + 3an · 32m c2 + 33m c3 ,
so
(a + 3)3n − a3n = a2n · 3m+1 c + an · 32m+1 c2 + 33m c3 ,
which is divisible by 3m+1 . This completes our inductive argument.
We have now shown for n = 3k−1 that 3k | 1n + 2n + 3n and 3k |
(a + 3)n − an for all positive integers a, from which we, by induction
on a, immediately deduce that 3k | an + (a + 1)n + (a + 2)n for all
positive integers a. Therefore 3k | Dn . As 3k+1 - 1n + 2n + 3n , we
also have 3k+1 - Dn . Therefore Dn = 3k , as desired.
32
Problems
1. Let a and b be two positive integers satifying gcd(a, b) = 1. Consider a

pawn standing on the grid point (x, y). A step of type A consists of moving
the pawn to one of the following grid points: (x+a, y+a), (x+a, y−a), (x−
a, y + a) or (x − a, y − a). A step of type B consists of moving the pawn
to (x + b, y + b), (x + b, y − b), (x − b, y + b) or (x − b, y − b).
Now put a pawn on (0, 0). You can make a (finite) number of steps,
alternatingly of type A and type B, starting with a step of type A. You
can make an even or odd number of steps, i.e., the last step could be of
either type A or type B. Determine the set of all grid points (x, y) that
you can reach with such a series of steps.
2. Determine all positive integers n for which there exist positive integers a1 ,
a2 , . . . , an with
a1 + 2a2 + 3a3 + . . . + nan = 6n
and
1 2 3 n 1
+ + + ... + =2+ .
a1 a2 a3 an n
3. An equilateral triangle ABC is given. On the line through B parallel to

AC there is a point D, such that D and C are on the same side of the line
AB. The perpendicular bisector of CD intersects the line AB in E. Prove
that triangle CDE is equilateral.
4. Each of the numbers 1 up to and including 2014 has to be coloured; half of

them have to be coloured red the other half blue. Then you consider the
number k of positive integer that are expressible as the sum of a red and
a blue number. Determine the maximum value of k that can be obtained.
5. Find all functions f : Z>0 → Z>0 such that f (1) = 2 and such that for
all m, n ∈ Z>0 we have that min(2m + 2n, f (m + n) + 1) is divisible by
max(f (m) + f (n), m + n).
33
Solutions
1. We will prove that the grid point (x, y) is reachable if and only if x + y ≡ 0
mod 2.
If we move the pawn from (x, y) to (x ± a, y ± a), then the sum of the new
coordinates equals x + y + 2a, x + y or x + y − 2a, hence it is congruent to
the sum of the old coordinates modulo 2. The same holds when making a
step of type B. Because the pawn starts at (0, 0), after executing any finite
number of steps, we will have that the sum of the coordinates of the grid
point where the pawn is standing, is even. Hence, the points (x, y) with
x + y ≡ 1 mod 2 are not reachable.
Now we will show that all other points are reachable. Because gcd(a, b) =
1, there exist integers m, n with ma+nb = 1. Then we have 2ma+2nb = 2.
Of the numbers m and n one must be positive and the other negative. We
assume that m is positive. The other case is treated analogously. We will
now first make 2m steps of type A and 2m steps of type B. For the steps
of type A we choose to do m times the step (x, y) 7→ (x + a, y + a) and m
times (x, y) 7→ (x + a, y − a). For the steps of type B we choose m times
(x, y) 7→ (x + b, y + b) and m times (x, y) 7→ (x − b, y − b). The effect of
all these steps together is that the x-coordinate increased by 2ma and the
y-coordinate did not change (in face, the B-steps cancel each other). After
this we do 2|n| steps of type A and 2|n| steps of type B. For the steps of
type A we choose to do |n| times the step (x, y) 7→ (x + a, y + a) and |n|
times (x, y) 7→ (x − a, y − a). For the steps of type B we choose to do |n|
times (x, y) 7→ (x − b, y + b) and |n| times (x, y) 7→ (x − b, y − b). The
effect of all these steps together is that the x-coordinate decreased by 2|n|
and the y-coordinate did not change (in fact, now the A-steps cancel each
other). Altogether, after these 4m + 4|n| steps, we moved the pawn from
a starting point (x, y) to the point (x + 2ma − 2|n|b, y) = (x + 2, y). We
can construct an analogous series of steps that move the pawn from (x, y)
to (x, y + 2), and also analogous series of steps that move the pawn from
(x, y) to (x − 2, y) and (x, y − 2). Each of these series starts with a step of
type A and ends with a step of type B.
By combining series like these, we can move the pawn from (0, 0) to any
point (x, y) with x ≡ y ≡ 0 mod 2. Now consider a point (x, y) with
x ≡ y ≡ 1 mod 2. Because gcd(a, b) = 1, at least one of a and b is odd.
Suppose that a is odd. Then the point (x − a, y − a) is a point with two
even coordinates, to which we can construct a series of steps, ending with
a step of type B. After that, we can make a step of type A that moves the
pawn from (x − a, y − a) to (x, y). Next, suppose that a is even. Then b is
odd. The point (x − b, y − b) now has two even coordinates, hence we can
34
reach this point with a series of steps ending with a step of type B. Now
we execute three more steps after this:
A B A
(x − b, y − b) 7−→ (x − b − a, y − b − a) 7−→ (x − a, y − a) 7−→ (x, y).
In this way we can also reach the point (x, y).

We conclude that we can reach all points (x, y) with x + y ≡ 0 mod 2 and
we cannot reach any other point.
2. If we apply the inequality of the arithmetic and harmonic mean to a1 , two

copies of a2 , three copies of a3 , . . . , n copies of an , then we find that
1 1
6n a1 + 2a2 + . . . + nan 2 n(n + 1) 2 n(n + 1)
1 = 1 ≥ 1 2 n = 1 .
2 n(n + 1) 2 n(n + 1) a1 + a2 + . . . + an 2+ n
We have
6n 12 12
1 = <
2 n(n + 1) n+1 n
and
1 1 2 1 2
2 n(n + 1) 2 n (n+ 1) 2 n (n+ 1)
1 = > = 41 n2 .
2+ n
2n + 1 2n + 2
12
Altogether we find that n > 14 n2 , or 48 > n3 , which yields n ≤ 3.
1
For n = 1 we get a1 = 6 and a1 = 3, which is a contradiction. Hence,
n = 1 is not possible.
For n = 2 we get a1 + 2a2 = 12 and a11 + a22 = 2 + 12 . If a2 ≥ 2 we have
1 2
a1 + a2 ≤ 1 + 1 and this is too small. Hence, we have a2 = 1, but then
we find that a1 = 12 − 2 = 10 and hence a11 + a22 = 10 1
+ 2, which is a
contradiction. Hence, n = 2 is impossible.
For n = 3 there is a solution, namely a1 = 6, a2 = 3 and a3 = 2. Hence,
n = 3 is possible and we conlude that n = 3 is the only solution.
3. We consider the configuration in which E lies between A and B. The case

in which B lies between A and E is treated analogously. (Because of the
condition that D and C lie on the same side of AB, it is impossible that
A lies between B and E, hence we have treated all cases.)
As E lies on the perpendicular bisector of CD, we have |EC| = |ED|.
Hence, it is sufficient to prove that ∠CED = 60◦ . First suppose that
E = B. Then we have ∠CED = ∠CBD = ∠ACB = 60◦ because of
alternating (Z) angles, hence we are done. Now suppose that E 6= B.
35
As BD is parallel to AC, we have ∠CBD = ∠ACB = 60◦ = ∠CBA.
Hence, the point E is the intersection point of the perpendicular bisector
of CD and the exterior angle bisector of ∠CBD. This means that E lies
on the circumcircle of triangle CDB. (This is a known fact, but it is
also possible to prove it as follows. Let E 0 be the intersection point of the
exterior angle bisector of ∠CBD and the circumcircle of 4CBD. Because
BE 0 is the exterior angle bisector, we have ∠CBE 0 = 180◦ − ∠DBE 0 .
Hence, chords CE 0 and DE 0 have the same length, which means that E 0
lies on the perpendicular bisector of CD.) We conclude that CEBD is a
cyclic quadrilateral. Hence, ∠CED = ∠CBD = 60◦ .
4. Let n = 2014. We shall prove that the maximum k equals 2n − 5. The

smallest number that you could possibly write as the sum of a red and a
blue number is 1 + 2 = 3 and the largest number is (n − 1) + n = 2n − 1.
Hence, there are at most 2n − 3 numbers expressible as the sum of a red
and a blue number.
Suppose that the numbers can be coloured in such a way that 2n − 3 or
2n − 4 of numbers are expressible as the sum of a red and a blue number.
Now at most one of the numbers from 4 up to and including 2n − 1 is not
expressible in such a way. We will now show that we may assume without
loss of generallity that this number is at least n + 1. Indeed, we could
make a second colouring in which a number i is blue if and only if n + 1 − i
is blue in the initial colouring. Then in the case of the second colouring
a number m is expressible as the sum of a red and a blue number if and
only if 2n + 2 − m was expressible as the sum of a red and a blue number
in the initial colouring. Hence, if in the initial colouring a number smaller
than n + 1 is not expressible as the sum of red and blue, then in the second
colouring a number greater than 2n + 2 − (n + 1) = n + 1 is not expressible
as the sum of red and blue.
Hence, we may assume that the numbers 3 up to and including n are
all expressible as the sum of red and blue. Because red and blue are
interchangable, we may also assume without loss of generality that 1 is
coloured blue. Because 3 is expressible as the sum of red and blue and this
can only be 3 = 1 + 2, the number 2 must be red. Now suppose that we
know that 2 up to and including l are red, for certain l with 2 ≤ l ≤ n − 2.
Then in all the possible sums a + b = l + 2 with a, b ≥ 2 both numbers are
colored red. However, we know that we can express l + 2 as the sum of
red and blue (because l + 2 ≤ n), hence that must be 1 + (l + 1). Hence,
l + 1 is also coloured red. By induction, we now see that the numbers 2
up to and including n − 1 are all red. These are n − 2 = 2012 numbers.
However, only 21 n = 1007 numbers are red, which is a contradiction.
36
We conclude that at least two numbers of 3 up to and including 2n − 1
are not expressible as the sum of a red and a blue number. We shall now
show that we can colour the numbers in such a way that all number from
4 up to and inlcuding 2n − 2 are expressible as the sum of a red and a blue
number, implying that the maximum k equals 2n − 5.
To obtain this, colour all the even numbers, except n, and also the number
1 blue. All odd numbers, except 1, and also the number n we colour red.
By adding 1 to an odd number (unequal to 1) we can obtain all even
numbers from 4 up to and inlcuding n as the sum of a red and a blue
number. By adding 2 to an odd number (unequal to 1), we can obtain all
odd numbers from 5 up to and including n + 1 as the sum of a red and a
blue number. By adding n − 1 to an even number (unequal to n), we can
obtain all odd numbers from n + 1 up to and including 2n − 3 as the sum
of a red and a blue number. By adding n to an even number (unequal to
n), we can obtain all even numbers from n + 2 up to and including 2n − 2
as the sum of a red and a blue number. Altogether, we can express all
numbers from 4 up to and including 2n − 2 as the sum of a red and a blue
number.
We conclude that the maximum k equals 2n − 5 = 4023.
5. By substituting m = n we get that min(4n, f (2n) + 1) is divisible by

max(2f (n), 2n). That is, a number that is at most 4n is divisible by
another number, which is at least 2f (n). This yields that 4n ≥ 2f (n),
hence f (n) ≤ 2n for all n.
By substituting m = n = 1 we get that min(4, f (2) + 1) is divisible by
max(2f (1), 2) = max(4, 2) = 4. Hence, min(4, f (2) + 1) cannot by smaller
than 4, hence f (2) + 1 ≥ 4. But we have already deduced that f (2) ≤
2 · 2 = 4, hence either f (2) = 3 or f (2) = 4 holds.
First suppose that f (2) = 3. We will prove by induction to n that f (n) =
n + 1 for all n. Namely, suppose that this holds for n = r − 1 for some
r ≥ 3 and subtitute m = 1 and n = r − 1. Then we get that min(2r, f (r) +
1) is divisible by max(f (1) + f (r − 1), r) = max(r + 2, r) = r + 2. As
min(2r, f (r) + 1) ≤ 2r < 2(r + 2), it holds that min(2r, f (r) + 1) = r + 2.
Because r ≥ 3, we have 2r > r + 2, hence we have f (r) + 1 = r + 2, or
f (r) = r + 1. This finishes the induction. Hence, we find the candidate
function f (n)+1. We will check immediately whether this function satisfies
the conditions. For all m, n ∈ N we have that min(2m+2n, f (m+n)+1) =
min(2m+2n, m+n+2) = m+n+2 as m, n ≥ 1, and max(f (m)+f (n), m+
n) = max(m + n + 2, m + n) = m + n + 2. The former is divisible by the
latter (as it equals the latter), hence this function satisfies.
37
Now suppose that f (2) = 4. We will prove by induction to n that f (n) =
2n for all n. Namely, suppose that this holds for n = r − 1 for some
r ≥ 4. We will prove that also f (r) = 2r holds. First substitute m =
1 and n = r − 1. Then we find that min(2r, f (r) + 1) is divisible by
max(f (1) + f (r − 1), 1 + r − 1) = max(2r, r) = 2r. Hence, min(2r, f (r) + 1)
cannot be smaller than 2r, hence f (r) ≥ 2r − 1. However, we already knew
that f (r) ≤ 2r, hence f (r) ∈ {2r−1, 2r}. Suppose that f (r) = 2r−1. Then
substitute m = 1 and n = r. Then we find that min(2(r + 1), f (r + 1) + 1)
is divisible by max(f (1) + f (r), 1 + r) = max(2 + 2r − 1, r + 1) = 2r + 1.
Because 2r + 1 - 2(r + 1), the minimum does not equal 2(r + 1), hence we
have f (r +1)+1 < 2r +2 and, moreover, it must also be divisible by 2r +1,
hence f (r + 1) = 2r. Now also substitute m = 2 and n = r − 1. Then we
find that min(2(r+1), f (r+1)+1) = min(2r+2, 2r+1) = 2r+1 is divisible
by max(f (2) + f (r − 1), 1 + r) = max(4 + 2r − 2, r + 1) = 2r + 2, which is a
contradiction,. We conclude that f (r) = 2r, which finishes the induction.
Hence, we find the candidate function f (n) = 2n for all n. We will check
whether this function satisfies the conditions. In this case, for all m, n we
have that min(2m+2n, f (m+n)+1) = min(2m+2n, 2m+2n+1) = 2m+2n
and max(f (m) + f (n), m + n) = max(2m + 2n, m + n) = 2m + 2n. The
former is divisible by the latter (because it equals the latter), hence this
function satisfies the conditions.
We conclude that there are exactly two solutions: the function given by
f (n) = n + 1 for all n and the function given by f (n) = 2n for all n.
38
Problems
Part 1
1. Note that 555555 : 7 = 79365. Consider the number 55 · · · 55 consisting of

1000 fives.
What is the remainder of this number on division by 7?
A) 2 B) 3 C) 4 D) 5 E) 6
2. A pawn is placed on a board consisting of ten squares, numbered from 1

up to 10. The pawn is allowed to move from the square it is on to a square
that either has a number that is two less, or a number that is twice as
large. The pawn wants to make a sequence of moves that visits as many
squares as possible. It may freely choose its starting point. It may visit
squares multiple times. How many squares can the pawn visit in a single
sequence of moves?
A) 6 B) 7 C) 8 D) 9 E) 10
3. Jan has huge square table of which the cells are 11

numbered as in the figure. Which of the follow- 10 12
ing five numbers does not occur in the leftmost 4 9
column? 3 5 8
A) 55 B) 105 C) 172 D) 212 E) 300 1 2 6 7
4. Birgit has a combination lock that consists of

three rings next to one another, each having the
2 9 7
digits 0 up to 9 in order. She turns the three rings
until her secret combination is visible. Aside from 3 0 8
this combination, there are 9 more combinations 4 1 9
visible on the three rings. Coincidentally, one of
these numbers is three times the secret combination.
What is Birgit’s secret combination?
A) 106 B) 123 C) 272 D) 318 E) 328
39
5. In a triangle ABC, we have ∠A = 84◦ . Moreover, D is a point on the line
segment AB such that ∠D1 = 3 · ∠C2 and such that the line segments DC
and DB have equal lengths.
What is ∠C1 ?
A) 27◦ B) 28◦ C) 30◦ D) 32◦ E) 36◦
1 D
2
1
2 84◦
C A
6. A piece of apple pie had been stolen, and five children are being questioned
on this. They all know who the culprit is, but not all of them are speaking
the truth. Whenever one of the children lies, the next one will feel so
guilty about this that he or she will tell the truth. The children make the
following claims in the order shown:
• Asim: “Coen and I both didn’t do it.”
• Bob: “Either Coen or Dilan is the culprit.”
• Coen: “Eva and I both didn’t do it.”
• Dilan: “Asim is the culprit.”
• Eva: “At least two of Asim, Bob, Coen, and Dilan lied.”
Who stole the apple pie?
A) Asim B) Bob C) Coen D) Dilan E) Eva
7. Consider the numbers a = (34 )5 , b = (44 )4 , and c = (54 )3 . If you sort a, b

and c from smallest to largest, you obtain:
A) a < b < c B) a < c < b C) b < a < c
D) c < a < b E) c < b < a
40
8. Max has a lot of white and red paint. He starts with a 2-litre bucket in
which one litre of red paint and one litre of white paint. Max now repeats
the following step a number of times.
Step. Max pours precisely one litre out of the bucket, into a large con-
tainer. Next, he fills the bucket back up to 2 litres of paint, using
either the white paint, or the red paint. After this, he mixes the
paint in the bucket.
After a number of steps, the percentage of red paint in the bucket must
be between 83 and 84 percent. What is the smallest number of steps Max
needs to attain this?
A) 5 B) 6 C) 7
D) 8 E) Max cannot obtain such a percentage.
Part 2
1. A member of a group of ten friends buys a bag of candy to share among

the group. First he himself, who likes candy more than the rest of the
group, takes a quarter of the candy. Another member grabs 30 pieces of
candy. A third member grabs 10% of what is left. The remainder of the
group distributes the remainder of the candy evenly. The total number of
pieces of candy was less than 500 and everyone got at least one piece of
candy.
How many pieces of candy were there in the bag?
2. There are 36 balls, numbered from 1 up to 36. We want to put these into
boxes in such a way that the following two conditions are satisfied:
(1) Every box contains at least 2 balls.
(2) Whenever you pick up two balls from a box, the sum of the two
numbers of these balls is always a multiple of 3.
What is the smallest number of boxes for which this is possible?
41
3. We are given a square ABCD. A line is drawn through A that intersects
the segment BC in E, and the line through C and D in F . The ratio of
the lengths of the segments BE and EC is 1 : 2. The area of the grey area
is 60.
What is the area of the square?
D C F
A B
4. We want to exchange a 200-euro bill for bills of 5, 10, and 20 euros. One
possibility is to exchange it for 5 bills of 20 euros, 6 bills of 10 euros, and
8 bills of 5 euros. Another possibility is to exchange it for 20 bills of 10
euros.
How many possibilities are there to exchange a 200-euro bill for bills of 5,
10, and 20 euros?
5. By stacking small cubes (all of the same size) neatly, a larger cube is
formed. Two small cubes with faces placed against one another are called
neighbours. So a cube can have at most six neighbours. The number of
cubes having precisely four neighbours is 96.
How many small cubes are there having precisely five neighbours?
6. Michelle colours each of the numbers from 1 up to 2014. The first two
numbers (1 and 2) are coloured red, the next two (3 and 4) are coloured
white, the following two (5 and 6) are coloured blue, the two following
those are coloured red, the two after those are coloured white, the two
after that are coloured blue, and so on and so forth. Michelle then sums
all blue numbers, and subtracts from that the sum of all red numbers.
What is the result?
42
7. The figure below represents a puzzle. The goal is to fill each of the 16 cells
with a number from 1 up to 4. This has to be done in such a way that in
each column and in each row, the four numbers are distinct. Moreover, in
each of the four 2×2-squares, the four numbers have to be distinct as well.
Finally, the four numbers in the grey squares also need to be distinct.
How many solutions does this puzzle have?
4
3
2
1
A B C D
8. Mies has drawn a regular hexagon with area 1. She notices that the mid-
points of the six sides also form a regular hexagon.
What is the area of this small hexagon?
43
Solutions
Part 1
1. C) 4 5. A) 27◦
2. D) 9 6. B) Bob
3. D) 212 7. D) c < a < b
4. E) 328 8. B) 6
Part 2
1. 320 5. 384
2. 13 6. −1343
3. 36 7. 168
3
4. 121 8. 4
44
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

and the team selection for IMO 2016 Hong Kong
BxMO Team Selection Test, March 2016 C
MY

CY
We eat problems
CMY

for breakfast.

Olympiad 2015
International
OmslagNWO2016-v2wit.indd 1 06-06-16 20:49
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

Introduction
2015, held at the participating schools. The paper consisted of eight mul-
tiple choice questions and four open questions, to be solved within 2 hours.
In this first round 10277 students from 354 secondary schools participated.
The 130 best students were invited to the final round. Also some outstand-
ing participants in the Kangaroo math contest or the Pythagoras Olympiad
an invitation to some training sessions at the universities, in order to pre-
pare them for their participation in the final round.
The final round in September contained five problems for which the stu-
dents had to give extensive solutions and proofs. They were allowed 3
hours for this round. After the prizes had been awarded in the beginning
of November, the Dutch Mathematical Olympiad concluded its 54rd edition
2015.

In February a team of four girls was chosen from the training group to
represent the Netherlands at the EGMO in Buşteni, Romania, from 10
until 16 April. The team brought home a silver medal and two bronze
medals, a very nice achievement. For more information about the EGMO
(including the 2016 paper), see www.egmo.org.
In March a selection test of three and a half hours was held to determ-
ine the ten students participating in the Benelux Mathematical Olympiad
(BxMO), held in Soest, the Netherlands, from 29 April until 1 May. The
Dutch team received six bronze medals and three silver medals, and man-
aged to get the highest total score. For more information about the BxMO
1
selected by three team selection tests on 2, 3 and 4 June 2016, each lasting
four hours. A seventh, young, promising student was selected to accompany
the team to the IMO as an observer C. The team had a training camp in
Hong Kong, from 30 June until 9 July.
the solutions.
2
Dutch delegation
The Dutch team for IMO 2016 in Hong Kong consists of
• Erik van Cappellen (17 years old)

– participated in BxMO 2016
• Wietze Koops (15 years old)
– silver medal at BxMO 2015, silver medal at BxMO 2016
• Reinier Schmiermann (14 years old)
• Pim Spelier (16 years old)
• Gabriel Visser (18 years old)
• Matthijs van der Poel (15 years old)

• Julian Lyczak (team leader), Leiden University

• Merlijn Staps (observer B), Utrecht University
3
Problems
A-problems
1. A square is divided into two rectangular pieces by a

straight line. The sum of the circumferences of the two
rectangles is 30 centimetres.
What is the side length of the square in centimetres?
15
A) 5 B) 6 C) 2 D) 8 E) 12
2. Five suspects are being questioned about the order of arrival at a crime
scene. They make the following statements.
Aad: “I arrived first.”

Bas: “I arrived second.”
Carl: “I arrived third.”
Dave: “Of Aad and Bas, one arrived before me and the other after me.”
Erik: “Of Bas and Carl, one arrived before me and the other after me.”
It is known that exactly one of the suspects lied.

Who was the fourth to arrive at the crime scene?
A) Aad B) Bas C) Carl D) Dave E) Erik
3. A big square consists of 2015 times 2015 small squares.

The small squares on the two main diagonals and on
the four adjacent diagonals are coloured grey, and
the rest is coloured white (see the figure).
How many small squares are coloured grey?
A) 12081 B) 12082 C) 12085
D) 12086 E) 12090
4. The difference of two integers is 10. If you multiply the two integers, you
will get one of the following five numbers.
Which number do you get?
A) 22398 B) 22420 C) 22442 D) 22453 E) 22475
4
5. Jan has got a wooden cube. He divides each of the
faces into 2×2 squares that he subsequently paints in a
black-white pattern: two diagonally opposite squares
are painted black, the other two are painted white. In
each vertex of the cube three squares meet. If two or
three of these squares are black, we call the vertex a
dark vertex.
What is the smallest number of dark vertices that the cube can have?
A) 0 B) 1 C) 2 D) 3 E) 4
6. In how many ways can you get the number 100 by adding some consecutive
integers between 1 and 99 inclusive?
A) 1 B) 2 C) 3 D) 4 E) 5
7. In the figure, you see two circles and two lines

together with the nine nodes in which they
intersect. Jaap wants to colour exactly four
of the nodes red, in such a way that no three
red nodes are on the same line or on the same
circle.
How many such colourings can Jaap make?
A) 6 B) 12 C) 18 D) 24 E) 36
8. A tree grows in the following manner. On day 1, one branch grows out of
the ground. On day 2, a leaf grows on the branch and the branch tip splits
up into two new branches. On each subsequent day, a new leaf grows on
every existing branch and each branch tip splits up into two new branches.
See the figure below.
How many leaves does the tree have at the end of the tenth day?
A) 172 B) 503 C) 920 D) 1013 E) 2047
day 1 day 2 day 3
5
B-problems
1. Julia constructs a sequence of numbers. She starts with two integers she
chooses herself. Then, she calculates the next numbers in the sequence as
follows: if the last number she wrote down is b and the number before that
is a, then the next number will be 2b − a. The second number in Julia’s
sequence is 55 and the hundredth number is 2015.
What is the first number in Julia’s sequence?
2. Two points A and B and two circles are given, one having A as centre and
going through B and the other one having B as centre and going through
A. Point C lies on the second circle and on line AB. Point D also lies on
the second circle. Point E lies on the first circle and on line BD. See the
figure below. Moreover, ∠D = 57◦ .
What is the value of ∠A in degrees?
57◦
A C
? B
3. A positive integer is called alternating if its digits alternate between even

and odd. For example, 2381 and 3218 are alternating, but 2318 is not.
An integer is called super alternating if the number itself is alternating
and twice that number is alternating as well. For example, 505 is super
alternating, because both 505 and 1010 are alternating.
How many super alternating integers consisting of four digits exist?
Pay attention: a four digit integer cannot start with a 0.
6
4. On a school trip, twenty students will be abseiling. In each round, one stu-
dent will descend the mountain. Hence, after twenty rounds, all students
will have gone down the mountain safely. In the first round, cards bearing
the numbers 1 to 20 are distributed among the students. The student get-
ting number 1 will go down first. In round 2, cards bearing the numbers 1
to 19 are distributed among the remaining students. The student receiving
the number 1 is next to descend. They continue in this way, until there is
only one student left in round 20, who automatically gets a card bearing
the number 1. By an amazing coincidence, no student gets the same num-
ber twice. In the first round, Sara gets a card with number 11.
What is the sum of the numbers on the cards received by Sara?
7
Solutions
A-problems
1. A) 5 5. C) 2
2. E) Erik 6. B) 2
3. A) 12081 7. C) 18
4. E) 22475 8. D) 1013
B-problems
1. 35
2. 48◦
3. 70
4. 66
8
Problems
B-problems
1. We consider numbers consisting of two or more digits with no digit being

0. Such a number is called thirteenish if every two consecutive digits form
a multiple of 13. For example: 139 is thirteenish because 13 = 1 × 13 and
39 = 3 × 13.
How many five digit numbers are thirteenish?
2. A quadrilateral ABCD has right angles at A and B. Also, |AB| = 5 and

|AD| = |CD| = 6.
Determine all possible values of |BC|.
3. Berry has picked 756 raspberries. He divides the raspberries among himself
and his friends in such a way that everyone gets the same number of
raspberries. However, three of his friends are not feeling hungry and they
each return a number of raspberries: exactly one quarter of their share.
Berry has a healthy appetite and eats not only his own share, but the
returned raspberries as well. Berry has lost count, but does know for a
fact that he has eaten more than 150 raspberries.
How many raspberries did Berry eat?
4. Four line segments divide a rectangle into eight pieces as indicated in the
figure. For three of the pieces, the area is indicated as well: 3, 5, and 8.
8
5
What is the area of the grey quadrilateral?
9
5. In the cells of a 5×5-table, the numbers 1 to 5 are placed in such a way
that in every row and in every column, each of the five numbers occurs
exactly once. A number in a given row and column is well-placed if the
following conditions are met.
• In that row, all smaller numbers are to the left of the number and all
larger numbers are to the right of it, or conversely.
• In that column, all smaller numbers are below the number and all
larger numbers are above it, or conversely.
What is the maximum number of well-placed numbers in such a table?
1. A set of different numbers are evenly spread if after sorting them from
small to large, all pairs of consecutive numbers have the same difference.
For example: 3, 11 and 7 are evenly spread, because after sorting them,
both differences are 4.
a) Kees starts out with three different numbers. He adds each pair of
these numbers to obtain three outcomes. According to Jan, these
three outcomes can be evenly spread only if the three starting numbers
were evenly spread.
Is Jan right? If so, prove this; if not, use an example to prove that
Jan is wrong.
b) Jan starts out with four different numbers. He also adds each pair of
them to obtain six outcomes. He wants to choose his four numbers in
such a way that the six resulting numbers are evenly spread.
Is this possible? If so, give an example; if not, prove that it is im-
possible.
10
2. We consider rectangular boards consisting of m × n cells that are arranged
in m (horizontal) rows and n (vertical) columns. We want to colour each
cell of the board black or white in such a way that the following rules are
obeyed.
• For every row, the number of white cells equals the number of black
cells.
• If a row and a column meet in a black cell, the row and column contain
equal numbers of black cells.
• If a row and a column meet in a white cell, the row and column
contain equal numbers of white cells.
Determine all pairs (m, n) for which such a colouring is possible.
11
Solutions
B-problems
1. 6
√
2. 6± 11
3. 189
4. 16
5. 5
C-problems
1. a) Kees starts with three numbers a < b < c. The three sums are then
ordered as follows: a + b < a + c < b + c. If these are evenly spread,
then the difference (b+c)−(a+c) = b−a equals (a+c)−(a+b) = c−b.
This is exactly the condition for the three original numbers a, b and
c to be evenly spread. Hence, Jan was right.
b) Jan can accomplish this by taking the four numbers 0, 1, 2 and 4 to
start with. The six results then are 0 + 1 = 1, 0 + 2 = 2, 1 + 2 = 3,
0 + 4 = 4, 1 + 4 = 5, and 2 + 4 = 6. These six numbers are evenly
spread.
2. In the following cases, a colouring meeting all the requirements exists.

• m = n is even
We colour the board as in a chessboard pattern. That is: in each
row and column the squares are alternately black and white. This
colouring meets all requirements.
• n = 2m
We colour all the squares in the left half of the board white, and colour
all the squares in the right half of the board black. This colouring
meets all requirements.
12
Now we shall show that these are the only possible board sizes. Consider a
coloured board that meets all requirements. Because each row has equally
many black and white squares, the total number of squares in a row must
be divisible by 2. Write n = 2k. Each row has exactly k white and k black
squares. Now consider the left column. If all its squares are white, then
the column has k white squares because of the second requirement. Hence,
we have m = k in this case. The same happens when all squares in the
left column are black. If there are both black and white squares in the left
column, then there must be exacly k white and k black squares because of
the second requirement. Hence, we find m = 2k = n in this case.
We conclude that for a pair (m, n) there exists a colouring if and only if
n = 2m or if m = n and n is even.
13
Problems
1. We make groups of numbers. Each group consists of five distinct numbers.

A number may occur in multiple groups. For any two groups, there are
exactly four numbers that occur in both groups.
(a) Determine whether it is possible to make 2015 groups.

(b) If all groups together must contain exactly six distinct numbers, what
is the greatest number of groups that you can make?
(c) If all groups together must contain exactly seven distinct numbers,
what is the greatest number of groups that you can make?
2. On a 1000×1000-board we put dominoes, in such a way that each domino

covers exactly two squares on the board. Moreover, two dominoes are not
allowed to be adjacent, but are allowed to touch in a vertex.
Determine the maximum number of dominoes that we can put on the board
in this way.
Attention: you have to really prove that a greater number of dominoes is
impossible.
3. Version for junior students B C

S
In quadrilateral ABCD sides BC and AD are R
P
parallel. In each of the four vertices we draw an
angular bisector. The angular bisectors of angles Q
A and B intersect in point P , those of angles B A D
and C intersect in point Q, those of angles C and D intersect in point R,

and those of angles D and A intersect in point S. Suppose that P S is
parallel to QR.
Prove that |AB| = |CD|.
14
3. Version for senior students E
Points A, B, and C are on a line in this order.
D
Points D and E lie on the same side of this line, S
in such a way that triangles ABD and BCE are
equilateral. The segments AE and CD intersect
in point S. A B C
Prove that ∠ASD = 60◦ .
4. Find all pairs of prime numbers (p, q) for which
7pq 2 + p = q 3 + 43p3 + 1.
5. Given are (not necessarily positive) real numbers a, b, and c for which
|a − b| ≥ |c|, |b − c| ≥ |a|, and |c − a| ≥ |b|.
Here |x| is the absolute value of x, i.e. |x| = x if x ≥ 0 and |x| = −x if

x < 0.
Prove that one of the numbers a, b, and c is the sum of the other two.
15
Solutions
1. (a) It is possible to make 2015 groups. For example, take the 2015 groups
{−4, −3, −2, −1, i}, where i runs from 1 to 2015. Each group consists
of five distinct numbers, as required, and any two groups have exactly
four numbers in common: −4, −3, −2, and −1.
(b) Using six available numbers, there are only six possible groups of
five numbers (each obtained by leaving out one of the six numbers).
Those six groups do satisfy the requirement that any two of them
have exactly four numbers in common. We conclude that six is the
greatest number of groups we can make in this case.
(c) A way to make three groups is to take {1, 2, 3, 4, 5}, {1, 2, 3, 4, 6}, and
{1, 2, 3, 4, 7}.
More than three groups is not possible. Indeed, suppose we have
four or more groups. The first two groups are A = {a, b, c, d, e} and
B = {a, b, c, d, f }, where a, b, c, d, e, and f are distinct numbers.
Then there must be a third group C containing a seventh number g.
The remaining four numbers in C must be in both A and B, hence
C = {a, b, c, d, g}.
Now consider a hypothetical fourth group D. This group cannot
contain the number g since otherwise, using a similar reasoning as
for C, we would have D = {a, b, c, d, g}. Because D does not contain
the number g, it must contain the remaining four numbers a, b, c,
and d from C. Comparison with groups A and B then shows that D
can contain neither e nor f . It follows that besides a, b, c, and d, D
cannot contain a fifth number, contradicting the requirements.
We conclude that the greatest number of groups we can make is three.
2. A maximum of 250,000 dominoes can fit on the board. We first show to

place this number of dominoes on the board. In each row we put 250
dominoes with two empty squares in between consecutive dominoes. In
the odd numbered rows we start with a domino and end with two empty
squares (since the number of squares in a row is a multiple of four). In
the even numbered rows we start with two empty squares and end with
a domino. Thus, we place a total of 1000 · 250 = 250,000 dominoes, see
the figure. Clearly, no two dominoes in the same row are adjacent, and
dominoes in adjacent rows touch in a vertex, at most.
16
To complete the proof, we need to show that it is not possible to place more
than 250,000 dominoes on the board. Partition the board into 500×500
patches consisting of 2×2 squares each. Of each patch, at most two out of
the four squares can be covered by dominoes since otherwise two dominoes
would be adjacent. Hence, no more than 2 · 500 · 500 = 500,000 squares can
be covered by dominoes. This shows that no more than 250,000 dominoes
can fit on the board.
3. Version for junior students B C

The intersection of CQ and AD is called
S
H. We have 21 ∠BAD = ∠SAD = ∠CHD
P R
(corresponding angles). Also, we have
1
∠CHD = ∠HCB = 2 ∠DCB (altern-
ate interior angles). It follows that the Q
A H D
two angles ∠BAD and ∠DCB of quad-
rilateral ABCD are equal. This implies that
∠BAD + ∠ADC = ∠DCB + ∠ADC = 180◦ , because AD and BC are
parallel. From ∠BAD+∠ADC = 180◦ it follows that AB and CD are par-
allel. Hence, ABCD is a parallelogram. We conclude that |AB| = |CD|.
3. Version for senior students E

Observe that ∠ABE = 180◦ − ∠EBC =
D
120◦ and ∠DBC = 180◦ − ∠ABD = 120◦ . S
Furthermore, |AB| = |DB| and |BE| = |BC|.
It follows that triangles ABE and DBC are
congruent (SAS). In particular, ∠EAB = A B C
∠CDB.
Observe that ∠ASD = 180◦ − ∠SDA − ∠DAS =
180◦ − (60◦ + ∠CDB) − ∠DAE. Substituting ∠CDB = ∠EAB shows
that ∠ASD = 120◦ − ∠EAB − ∠DAE = 120◦ − 60◦ = 60◦ .
17
4. We start by observing that in the equation 7pq 2 + p = q 3 + 43p3 + 1 the
numbers p and q cannot both be odd. Otherwise, 7pq 2 + p would be even,
while q 3 + 43p3 + 1 would be odd. Since 2 is the only even prime number,
we conclude that p = 2 or q = 2.
In the case p = 2, we obtain the equation 14q 2 +2 = q 3 +344+1, which can
be rewritten as q 3 − 14q 2 = −343. This shows that q must be a divisor of
343 = 73 , hence q = 7. Substitution confirms that (p, q) = (2, 7) is indeed
a solution since 14q 2 + 2 = 2 · 7 · 72 + 2 and q 3 + 344 + 1 = 73 + (73 + 1) + 1 =
2 · 73 + 2 are equal.
Next, we consider the case that q = 2 and p is odd. This results in the
equation 28p + p = 8 + 43p3 + 1. Since p is odd, we see that 28p + p is odd,
while 8 + 43p3 + 1 is even. Hence, no solutions exist in this case.
We conclude that (p, q) = (2, 7) is the only solution.
5. The system of inequalities is symmetric in the variables a, b, and c: if

we exchange two of these variables, the system remains unchanged (up to
rewriting it). For example, if we exchange variables a and b, we obtain
|b − a| ≥ |c|, |a − c| ≥ |b|, |c − b| ≥ |a|.
Since |b − a| = |a − b|, |a − c| = |c − a|, and |c − b| = |b − c|, this can be

rewritten as
|a − b| ≥ |c|, |c − a| ≥ |b|, |b − c| ≥ |a|,
obtaining the original system of inequalities. Due to this symmetry, we

may assume without loss of generality that a ≥ b ≥ c.
First observe that c ≤ 0. Indeed, if a ≥ b ≥ c > 0 would hold, then
|b − c| ≥ |a| would imply that b − c ≥ a. Rewriting gives b ≥ a + c > a,
which contradicts the fact that b ≤ a.
Next, consider the following series of inequalities.
|a| + |c| = |a| − c ≥ a − c = (a − b) + (b − c) = |a − b| + |b − c| ≥ |c| + |a|.
Since the first and the last term are equal, we can conclude that all of the
above inequalities must be equalities. In particular, a − b = |a − b| = |c|.
Since c ≤ 0, this implies that a + c = b. This shows that one of the three
numbers equals the sum of the other two.
18
Problems
1. For a positive integer n that is not a power of two, we define t(n) as the
greatest odd divisor of n and r(n) as the smallest positive odd divisor of
n unequal to 1. Determine all positive integers n that are not a power of
two and for which we have
n = 3t(n) + 5r(n).
2. Determine all triples (x, y, z) of non-positive real numbers that satisfy the
following system of equations
x2 − y = (z − 1)2 ,
y 2 − z = (x − 1)2 ,
z 2 − x = (y − 1)2 .
3. Let 4ABC be a right-angled triangle with ∠A = 90◦ and circumcircle Γ.

The inscribed circle is tangent to BC in point D. Let E be the midpoint
of the arc AB of Γ not containing C and let F be the midpoint of the arc
AC of Γ not containing B.
(a) Prove that 4ABC ∼ 4DEF .
(b) Prove that EF goes through the points of tangency of the incircle to
AB and AC.
4. The Facebook group Olympiad training has at least five members. There
is a certain integer k with following property: for each k-tuple of members
there is at least one member of this k-tuple friends with each of the other
k − 1. (Friendship is mutual: if A is friends with B, then also B is friends
with A.)
(a) Suppose k = 4. Can you say with certainty that the Facebook group
has a member that is friends with each of the other members?
(b) Suppose k = 5. Can you say with certainty that the Facebook group
has a member that is friends with each of the other members?
5. Determine all pairs (m, n) of positive integers for which

(m + n)3 | 2n(3m2 + n2 ) + 8.
19
Solutions
1. Let p be the smallest odd prime divisor of n. Then r(n) = p holds. Now
we can write n = 2t mp with m odd and t ≥ 0. Then we have t(n) = pm,
hence the given equality becomes 2t mp = 3pm + 5p, or (2t − 3)mp = 5p,
hence (2t − 3)m = 5. We see that m must be a divisor of 5, hence m = 1
or m = 5. If m = 1 holds, then 2t = 8, hence t = 3. We get that n = 8p
with p an odd prime number. This is indeed a solution for all odd prime
numbers p. If m = 5 holds, then 2t = 4, hence t = 2. We get that n = 4·5·p
with p an odd prime. This only gives a solution if p is the smallest odd
prime divisor, which is if p = 3 or p = 5. In this way, we find two more
solutions: n = 60 and n = 100.
2. Expanding the right hand sides and adding up all equations gives
x2 + y 2 + z 2 − (x + y + z) = x2 + y 2 + z 2 − 2(x + y + z) + 3,
hence x + y + z = 3. Without loss of generality we assume that x ≤ y, z.

Then 0 ≤ x ≤ 1 holds. Therefore, x2 ≤ x, hence x2 − y ≤ x − y ≤ 0. On
the other hand we have x2 − y = (z − 1)2 ≥ 0. Thus, equality has to hold
in x2 ≤ x and x − y ≤ 0. From the first equality we obtain x = 0 or x = 1
and the second one yields x = y. Suppose x = y = 0, then x + y + z = 3
implies that z = 3. But then we do not have x2 − y = (z − 1)2 , which is
a contradiction. We are only left with the case x = y = 1. Then we have
z = 3 − 1 − 1 = 1. This triple indeed satisfies all equations and hence it is
the only solution.
3. (a) The midpoint E of the arc AB not containing C, lies on the angu-
lar bisector CI. In the same way, F lies on BI. We have ∠IF C =
∠BF C = ∠BAC = 90◦ , because ABCF is a cyclic quadrilateral
and ∠IDC = 90◦ because D is the point of tangency of the in-
circle to BC. Hence, ∠IF C + ∠IDC = 180◦ , which yields that
F IDC is a cyclic quadrilateral. Now we have ∠DF I = ∠DCI =
1 1
2 ∠ACB, while also ∠IF E = ∠BF E = ∠BCE = 2 ∠ACB. Hence,
1 1
∠DF E = 2 ∠ACB + 2 ∠ACB = ∠ACB. Analogously, we have
∠DEF = ∠ABC. Altogether, this yields 4DEF ∼ 4ABC.
20
(b) Let S be the intersection of EF and AB. In the previous part we
already saw that BF is the angular bisector of ∠DF E = ∠DF S.
Hence, BF is also the angular bisector of ∠ABC = ∠SBD. There-
fore, 4BDF ∼ = 4BSF because of (ASA). This means that |BD| =
|BS|. On the other hand, the distances of B to the points of tangency
of the incircle to BC and BA are equal and one of these points of
tangency is D, hence the other point of tangency must be S. Hence,
EF goes through the point of tangency of the incircle to AB. Ana-
logously, it also goes through the point of tangency of the incircle to
AC.
4. (a) Yes, you can. If everybody is friends with everyone else, then we are
done. Hence, suppose that there are two members, say A and B, who
are not friends with each other. If we consider a group of four with A,
B, and two other members, then one of the other two must be friends
with the other and with A and B. In particular, the two others are
friends with each other. This holds for any two members (unequal to
A and B) that we choose, hence each pair not containing A and B is
friends with each other. Now take A, B, C, and D and suppose that
C is friends with A, B, and D. Moreover, he is also friends with all
other members of the group, hence C is someone who is friends with
everybody else.
(b) No, this is not possible. We give a counterexample. Suppose that the
Facebook group has six members, called A, B, C, D, E, and F . They
are all friends with each other, except for the pair (A, B), the pair
(C, D) and the pair (E, F ). This means that no member is friends
with every other member. If we take a group of five then without loss
of generality this is A, B, C, D, and E. Here we can find someone
who is friends with the other four, namely E. Hence, the condition is
met.
21
5. Suppose that the quotient of 2n(3m2 + n2 ) + 8 and (m + n)3 is unequal to
1. Then it is at least 2, hence we have
(m + n)3 ≤ n(3m2 + n2 ) + 4,
or
m3 + 3m2 n + 3mn2 + n3 ≤ 3m2 n + n3 + 4,
or
m3 + 3mn2 ≤ 4.
This yields m < 2, and therefore m = 1. Then we have 1 + 3n2 ≤ 4, and
hence also n = 1. The pair (m, n) = (1, 1) is indeed a solution, because
23 | 2 · 4 + 8 holds.
The other possibility is that the quotient does equal 1. Then we have
(m + n)3 = 2n(3m2 + n2 ) + 8,
or
m3 + 3m2 n + 3mn2 + n3 = 6m2 n + 2n3 + 8,
or
m3 − 3m2 n + 3mn2 − n3 = 8.
The left hand side we can factor as (m − n)3 . Hence, we have m − n = 2,
or m = n + 2. From the previous calculations it also follows that (m, n) =
(n + 2, n) is indeed a solution for all positive integers n.
We conclude that the solutions are: (m, n) = (1, 1) and (m, n) = (n + 2, n)
for n ≥ 1.
22
Problems
1. Let ABC be an acute triangle. Let H be the foot on AB of the altitude

through C. Suppose that |AH| = 3|BH|. Let M and N be the midpoints
of the segments AB and AC, respectively. Let P be a point such that
|N P | = |N C| and |CP | = |CB| and such that B and P lie on opposite
sides of the line AC.
Show that ∠AP M = ∠P BA.
2. Let n be a positive integer, and consider a square of dimensions 2n ×2n . We

cover this square by a number of (at least 2) rectangles, without overlaps,
and in such a way that every rectangle has integer dimensions and a power
of two as area. Show that two of the rectangles used must have the same
dimensions. (Two rectangles are said to have the same dimensions if they
have the same height and the same width, without rotating them.)
3. Find all positive integers k for which the equation
lcm(m, n) − gcd(m, n) = k(m − n)
does not have any solutions (m, n) in positive integers with m 6= n.
f (xy − 1) + f (x)f (y) = 2xy − 1
for all x, y ∈ R.
23
Solutions
1. The configuration is unique. As N is the midpoint of AC, we have |N C| =

|N A|. Since |N P | = |N C| by assumption, it follows that N is the centre
of a circle passing through A, C, and P . Using Thales’s theorem, we find
∠AP C = 90◦ .
Since |AH| = 3|BH| and M is the midpoint of AB, we have |M H| =
|BH|. Moreover, because ∠CHB = 90◦ = ∠CHM , the triangles CHB
and CHM are congruent, and therefore |CM | = |CB|. By assumption,
|CP | = |CB|, so C is the centre of a circle through P , M , and B. As
∠AP C = 90◦ , the line AP is tangent to this circle. By the tangent-chord
theorem we now have ∠AP M = ∠P BM = ∠P BA.
2. First note that a rectangle with integer dimensions has a power of two as
area if and only if the dimensions are powers of two.
Consider a covering in which no two rectangles have the same dimensions.
We first show that then there are no rectangles of width 1. Suppose to
the contrary that such a rectangle occurs in the covering. Colour every
square covered by a rectangle of width 1 with the colour blue. Let M be
the number of blue squares. Then M is the sum of all rectangles of width
1, so it is the sum of (at least one) distinct powers of 2. Let 2k be the
largest one. As 2k − 1 = 2k−1 + 2k−2 + · · · + 2 + 1, there are fewer than
2k blue squares not covered by this rectangle of width 1 and height 2k , so
there is at least one row which contains a blue square of this rectangle of
height 2k and no other blue square. But the remaining squares in this row,
of which there are an odd number, must then be covered by rectangles of
even width, which is a contradiction. So there are no rectangles of width
1. Analogously, there are no rectangles of height 1, either. Therefore all
rectangles have an even width and an even height.
Now consider the smallest n for which a covering as in the problem exists
in which no two rectangles have the same dimension. As all rectangles
have an even width and an even height, we can divide all dimensions (of
both the square and the rectangles) by 2 to obtain a square with smaller
dimensions covered by rectangles with integer dimensions, each having a
power of two as area. Since no two rectangles had the same dimensions,
this contradicts the minimality of n.
24
3. Let d = gcd(m, n) and write m = da and n = db. Then lcm(m, n) ·
gcd(m, n) = mn, so we can rewrite the given equation as
da·db
d − d = k(da − db),
or equivalently,
ab − 1 = k(a − b). (1)
So from now on we consider the following equivalent problem: find all
positive integers k for which (1) has no solutions (a, b) in positive integers
with a 6= b and gcd(a, b) = 1. Note that if a pair (a, b) satisfies this
equation, then it automatically follows that gcd(a, b) = 1; suppose that
t | a and t | b, then we have t | ab and t | a − b, so t | 1, and we deduce that
a and b have no common divisor greater than 1.
First suppose that k ≥ 3. We claim that (a, b) = (k 2 − k − 1, k − 1) is a
solution. Indeed, we have
ab − 1 = a(k − 1) − 1 = ka − a − 1 = ka − k 2 + k = k(a − k + 1) = k(a − b).
Moreover, as noted earlier, it follows that gcd(a, b) = 1, so it remains to

check that a and b are positive and distinct. As k ≥ 3, we have b = k − 1 ≥
2, and a = k 2 − k − 1 ≥ 2k − k − 1 = k − 1 ≥ 2, so a and b are both positive.
If a = b, then k 2 − k − 1 = k − 1, so k 2 = 2k, and therefore k = 2, which
is a contradiction. Therefore (a, b) is as required. Hence (1) has a solution
(a, b) in positive integers with a 6= b and gcd(a, b) = 1.
Now suppose that k = 1. We claim that (a, b) = (2, 1) is a solution.
Clearly, a and b are positive and distinct, and gcd(a, b) = 1. Moreover, we
have
ab − 1 = 2 − 1 = 1 = 1 · (2 − 1) = k(a − b).
Hence (1) has a solution (a, b) in positive integers with a 6= b and gcd(a, b) =
1.
Finally, suppose that k = 2. Then the equation (1) becomes
ab − 1 = 2(a − b).
The right hand side is at most 2a − 2 as b is a positive integer, so ab − 1 ≤

2a−2, hence ab < 2a, and therefore b < 2. We deduce from this that b = 1.
The equation then becomes a − 1 = 2(a − 1), which implies a − 1 = 0. We
therefore must have a = 1 and b = 1, so a = b. Therefore there are no
solutions (a, b) for (1) in positive integers with a 6= b and gcd(a, b) = 1.
We conclude that the unique k for which the given equation has no solu-
tions (m, n) in positive integers with m 6= n, is k = 2.
25
4. If f were constant, then the left hand side is constant, whereas the right
hand side is not, this is a contradiction. Therefore f is not constant.
Substitute x = 0. This gives f (−1) + f (0)f (y) = −1, so f (0)f (y) is a
constant function in y. As f is not constant, it follows that f (0) = 0, and
therefore it also follows that f (−1) = −1. Substituting x = y = 1 gives:
f (0) + f (1)2 = 1, so we have either f (1) = 1 or f (1) = −1.
Now substituting y = 1 + x1 with x 6= 0 gives f (x + 1 − 1) = f (x)f (1 + x1 ) =
2x + 2 − 1, so we have
f (x)f 1 + x1 = 2x + 1 − f (x)

for all x 6= 0. (2)
1
Substituting y = x with x 6= 0 gives f (1 − 1) + f (x)f ( x1 ) = 2 − 1, so we
have
1

f (x)f x =1 for all x 6= 0. (3)
Substituting y = 1, x = z + 1 gives f (z + 1 − 1) + f (z + 1)f (1) = 2z + 2 − 1,
so we have
f (z) + f (z + 1)f (1) = 2z + 1 for all z.
In this last equation, substitute z = x1 with x 6= 0, and then multiply both

sides with f (x); we obtain
f (x)f x1 + f x1 + 1 f (1)f (x) = x2 f (x) + f (x)

for all x 6= 0.
Using (3) and (2), we can rewrite the first and second term respectively as
follows:
1 + 2xf (1) + f (1) − f (x)f (1) = x2 f (x) + f (x) for all x 6= 0.
This we can rewrite as

2

f (x) · x + 1 + f (1) = 1 + 2xf (1) + f (1) for all x 6= 0.
If the second factor of the left hand side is non-zero, then we can divide
by this, so we obtain
1 + 2xf (1) + f (1) 2

f (x) = 2 if x 6= 0 and x + 1 + f (1) 6= 0.
x + 1 + f (1)
Recall that we had two possible values for f (1). First suppose that f (1) =
1. Then we have
2 + 2x 2
f (x) = 2 =x if x 6= 0 and x + 2 6= 0.
x +2
26
Note that x2 + 2 = 0 only if x = −1. As we already know that f (−1) = −1
and f (0) = 0, it follows that f (x) = x for all x. We check that this f
satisfies the original equation: the left hand side then is xy − 1 + xy =
2xy − 1, so this f is indeed a solution.
Suppose on the other hand that f (1) = −1. Then we have
−2x
f (x) = 2 = −x2 if x 6= 0 and 2
x 6= 0.
x
Note that x2 = 0 is impossible. Hence, as f (0) = 0, we deduce that

f (x) = −x2 for all x. We check that this f satisfies the original equation:
the left hand side then is −x2 y 2 − 1 + 2xy + x2 y 2 = 2xy − 1, so this f is
indeed a solution.
We conclude that there are two solutions, namely f (x) = x for all x and
f (x) = −x2 for all x.
27
Problems
1. Prove that for all positive reals a, b, c we have:

√ √
3
a + ab + abc ≤ 43 (a + b + c).
2. Determine all pairs (a, b) of integers having the following property: there
is an integer d ≥ 2 such that an + bn + 1 is divisible by d for all positive
integers n.
3. Let 4ABC be an isosceles triangle with |AB| = |AC|. Let D, E and

F be points on line segments BC, CA and AB, respectively, such that
|BF | = |BE| and such that ED is the exterior angle bisector of ∠BEC.
Prove that |BD| = |EF | if and only if |AF | = |EC|.
4. Determine the number of sets A = {a1 , a2 , . . . , a1000 } of positive integers

satisfying a1 < a2 < . . . < a1000 ≤ 2014, for which we have that the set
S = {ai + aj | 1 ≤ i, j ≤ 1000 en i + j ∈ A}
is a subset of A.
28
Solutions
√
1. We can write 3 abc as 3 a4 · b · 4c. Applying the inequality of the arithmetic
p
and geometric mean on the positive reals a4 , b and 4c yields
√ a
r
3 a + b + 4c a b 4c
abc = 3 · b · 4c ≤ 4 = + + .
4 3 12 3 3
Next we apply AM-GM on a2 and 2b:
√ a
r
a 2 + 2b a
ab = · 2b ≤ = + b.
2 2 4
We add up these two inequalities and we also add a to both sides to obtain:
√ √
3 a a b 4c 4
a+ ab + abc ≤ a + +b+ + + = (a + b + c).
4 12 3 3 3

2. Consider a pair (a, b) that has the property, with the corresponding d. Let
p be a prime divisor of d (which exists as d ≥ 2). Because d | an + bn + 1
for all n, we also have p | an + bn + 1 for all n. Consider n = p − 1. Then
an ≡ 0 mod p holds if p | a and an ≡ 1 mod p holds if p - a, because of
the Fermat’s little theorem. Similarly, this holds for b. Hence, an + bn + 1
can attain the values 1, 2 and 3 modulo p. On the other hand, it must be
congruent to 0 modulo p, hence p = 2 or p = 3. We consider the two cases.
Suppose that p = 3. Then we must have 3 - a, 3 - b. The case n = 1
moreover yields that 3 | a + b + 1, hence a + b ≡ 2 mod 3. These two
requirements together are equivalent to a ≡ b ≡ 1 mod 3. In this case, it
is true that for all positive integers n we have an + bn + 1 ≡ 1 + 1 + 1 ≡ 0
mod 3, hence such a pair has the property, with d = 3.
Now suppose that p = 2. Then 2 must be a divisor of exactly one of a and b.
In this case we have for all positive integers n that an +bn +1 ≡ 0+1+1 ≡ 0
mod 2, hence each such pair has the property, with d = 2.
We conclude that these are the pairs that have the property: (a, b) with
a ≡ b ≡ 1 mod 3, (a, b) with a ≡ 1, b ≡ 0 mod 2, and (a, b) with a ≡ 0,
b ≡ 1 mod 2.
3. From the data and the angle bisector theorem it follows that
|BF | |BE| |CE|
= = .
|BD| |BD| |CD|
29
As 4ABC is isosceles, we have ∠F BD = ∠ECD, which yields together
with the first equality that 4BF D ∼ 4CED. This yields ∠BF D =
∠CED = ∠BED, hence BDEF is a cyclic quadrilateral. It is well-known
that a cyclic quadrilateral BDEF is a trapezoid with DE k BF if and
only if |BD| = |EF |. We have |AF | = |EC| if and only if |BF | = |AE| (as
|AB| = |AC|), that is to say, if and only if |EA| = |EB|, which holds if and
only if ∠BAE = ∠ABE. Because we have 2∠BED = ∠BEC = ∠BAE +
∠ABE by the exterior angle theorem, ∠BAE = ∠ABE is equivalent to
∠BED = ∠ABE which is again equivalent to DE k BF . To summarise,
|BD| = |EF | holds if and only if DE k BF which holds if and only if
|AF | = |EC|.
4. We will prove that there are 214 such sets. In particular, we will prove that
the sets A that satisfy these conditions are of the form B ∪ C, with C a
subset of {2001, . . . , 2014} and B = {1, 2, . . . , 1000 − |C|}. The sets of this
form will be called “nice”. As there are 214 subsets of {2001, . . . , 2014},
there are 214 nice sets.
First we will show that each nice set A satisfies the conditions. Suppose
that i, j ≤ 1000 with i+j ∈ A. Then we have i+j ≤ 2000, hence i+j ∈ B.
Therefore, there exists a k with k ≤ 1000 − |C| such that i + j = ak (= k).
Because ak ≤ 1000 − |C| holds, we also have i, j ≤ 1000 − |C|, hence we
have ai = i and aj = j. That means that ai + aj = i + j = ak is an element
of A. Hence each element of S is an element of A, from which we deduce
that A satisfies the conditions.
We will now show that each set A satisfying the conditions is nice. First
suppose that there exists an integer k with 1 ≤ k ≤ 1000 satisfying ak ∈
{1001, . . . , 2000}. Then we have ak = 1000+i for a certain i with i ≤ 1000,
hence a1000 + ai is a an element of S and hence it must also be an element
of A. However, we have a1000 + ai > a1000 , which yields a contradiction.
Hence such an ak cannot occur. This means that A can be written as a
disjoint union B ∪ C, with C ⊆ {2001, . . . , 2014} and B ⊆ {1, 2, . . . , 1000}.
Let b be the number of elements of B. Then b ≥ 986 holds, because
C has at most 14 elements. In order to prove that A is nice, we must
prove that B = {1, 2, . . . , b}. To prove this it is sufficient to prove that
ab , the maximum of B, equals b. Therefore, suppose the contrary, i.e.
that ab > b. For i with i = ab − b we then have b + i = ab ≤ 1000,
hence i ≤ 1000 − b ≤ 14 < b. Therefore we have ai ≤ 1000 and hence
ab + ai ≤ 2000. Because i + b = ab ∈ A, we have ab + ai ∈ S ⊂ A, but then
ab +ai is an element of B greater than ab , which was the maximum. This is
a contradiction. Hence, ab = b, from which it follows that B = {1, 2, . . . , b}.
Therefore, A is nice.
30
Problems
1. Let n be a positive integer. In a village, n boys and n girls are living.

For the yearly ball, n dancing couples need to be formed, each of which
consists of one boy and one girl. Every girl submits a list, which consists
of the name of the boy with whom she wants to dance the most, together
with zero or more names of other boys with whom she wants to dance. It
turns out that n dancing couples can be formed in such a way that every
girl is paired with a boy who is on her list.
Show that it is possible to form n dancing couples in such a way that every
girl is paired with a boy who is on her list, and at least one girl is paired
with the boy with whom she wants to dance the most.
2. For distinct real numbers a1 , a2 , . . . , an , we calculate the n(n−1)

2 sums ai +
aj with 1 ≤ i < j ≤ n, and sort them in ascending order. Find all
integers n ≥ 3 for which there exist a1 , a2 , . . . , an for which this sequence
of n(n−1)
2 sums form an arithmetic progression (i.e. the difference between
consecutive terms is constant).
3. Let k be a positive integer, and let s(n) denote the sum of the digits
of n. Show that among the positive integers with k digits, there are as
many numbers n satisfying s(n) < s(2n) as there are numbers n satisfying
s(n) > s(2n).
4. Let Γ1 be a circle with centre A and Γ2 be a circle with centre B, with

A lying on Γ2 . On Γ2 there is a (variable) point P not lying on AB. A
line through P is a tangent of Γ1 at S, and it intersects Γ2 again in Q,
with P and Q lying on the same side of AB. A different line through Q is
tangent to Γ1 at T . Moreover, let M be the foot of the perpendicular to
AB through P . Let N be the intersection of AQ and M T .
Show that N lies on a line independent of the position of P on Γ2 .
31
Solutions
1. For each girl, call the boy with whom she wants to dance the most her
favourite.
We solve the problem by induction on n. If n = 1, the only girl will form a
couple with the only boy, who is therefore her favourite. So suppose that
k ≥ 1, and assume that the problem has been solved for n = k.
Consider the case n = k + 1. We distinguish two cases. First suppose
that every boy occurs exactly once as a favourite. In this case we can just
couple every girl to her favourite, and form n dancing couples that way.
In the remaining case, not every boy occurs exactly once as a favourite.
Since there are n favourites and n boys, once of the boys, say X, is not the
favourite of any girl (and someone else is the favourite of more than one
girl). Choose a pairing as in the problem; this exists by assumption. Let
Y be the girl coupled with boy X, and remove X and Y from the village.
There are k boys and k girls left. Note that the pairing chosen still has the
property that every girl is paired with a boy on her list. Moreover, every
girls still has a favourite among the k remaining boys, as boy X is not the
favourite of any girl. Therefore, by the induction hypothesis, we can form
k dancing couples, in such a way that every girl is paired with a boy on
her list, and at least one of the girls is paired with her favourite. Adding
the couple X − Y back in completes the induction.
2. For n = 3 we consider (a1 , a2 , a3 ) = (1, 2, 3). The sums of pairs are in this
case 3, 4, and 5, and these form an arithmetic progression. For n = 4 we
consider (a1 , a2 , a3 , a4 ) = (1, 3, 4, 5). The sums of pairs are in this case 4,
5, 6, 7, 8, and 9, and these form an arithmetic progression.
Now suppose that n ≥ 5, and suppose that a1 , a2 , . . . , an satisfies the
condition. Without loss of generality, we assume that a1 < a2 < · · · < an .
Let d be the difference between two consecutive terms of the corresponding
arithmetic progression. Note that the smallest sum is a1 + a2 , and the
second smallest is a1 + a3 . As the difference between these sums is d, we
have a3 − a2 = d. The largest sum is an + an−1 and the second largest is
an + an−1 , therefore we have an−1 − an−2 = d as well. Hence
a2 + an−1 = (a3 − d) + (an−2 + d) = a3 + an−2 .
If n ≥ 6, then the left hand side and the right hand side are sums of distinct
pairs, but the difference between such sums must be at least d. This is a
contradiction. Therefore there are no solutions for n ≥ 6.
32
For n = 5, we have a3 − a2 = d and a4 − a3 = d. Therefore the third
smallest sum must be a1 + a4 (as this one is d larger than a1 + a3 ), and
the third largest sum must be a5 + a2 . Between these, we have a2 + a3 <
a2 + a4 < a3 + a4 , and the difference between these consecutive sums is d,
and we also have a1 + a5 . Therefore a1 + a5 is either the fourth smallest
sum or the fourth largest sum.
Without loss of generality, assume that a1 + a5 is the fourth smallest sum.
Then we have
a1 + a2 < a1 + a3 < a1 + a4 < a1 + a5 < a2 + a3

< a2 + a4 < a3 + a4 < a2 + a5 < a3 + a5 < a4 + a5 .
Then (a5 + a2 ) − (a3 + a4 ) = d, so a5 − a4 = d + a3 − a2 = d + d. On the

other hand, (a1 + a5 ) − (a1 + a4 ) = d. This is a contradiction. Therefore
there are no solutions for n = 5 either.
We conclude that there exist a1 , a2 , . . . , an satisfying the condition if and
only if n = 3 or n = 4.
3. We show that among the positive integers with at most k digits there are
as many numbers n satisfying s(n) < s(2n) as there are numbers satisfying
s(n) > s(2n). The required result then follows by combining this result for
k and for k − 1.
We pair each number n with at most k digits to another number m with
at most k digits. Let m = 999 · · · 999 − n, where the first number consists
of k nines, so that m has at most k digits. We show that s(m) − s(2m) =
s(2n) − s(n).
To calculate m, we subtract n from 999 · · · 999, in which every digit of n
is of course at most 9. Therefore every digit of m is equal to 9 minus
the corresponding digit of n. Here we consider both m and n as numbers
having precisely k digits by adding zeroes to the left if necessary. Hence
s(m) + s(n) = s(999 · · · 999) = 9k.
Next, consider 2m and 2n. We have 2m = 1999 · · · 998 − 2n. Consider 2m
and 2n as numbers having exactly k + 1 digits, where the first digit of 2n
is either a 0 or a 1. Subtracting that from 1999 · · · 998, we find that the
first digit of 2m is either 1 − 1 = 0 or 1 − 0 = 1. The last digit of 2m is
8 minus the last digit of 2n, which cannot be a 9 as 2n is even. All other
digits of 2m are equal to 9 minus the corresponding digit of 2n. Therefore,
we have s(2m) + s(2n) = s(1999 · · · 998) = 1 + 9(k − 1) + 8 = 9k. Hence
s(m) + s(n) = s(2m) + s(2n), so s(m) − s(2m) = s(2n) − s(n).
33
Now we see that s(m) > s(2m) if and only if s(n) < s(2n). Moreover, no
number is paired to itself as 999 · · · 999 is odd. Hence there are as many
numbers with s(n) < s(2n) as there are numbers with s(n) > s(2n).
4. Point P lies outside Γ1 , since otherwise there is no tangent P S to Γ1 . Since

P and Q lie on the same side of AB, we see that S lies on the part of Γ1
on that same side of AB, and that S lies outside Γ2 . (In the extremal case
in which P lies on AB, we see that S an intersection point of Γ1 and Γ2 by
Thales’s theorem.) Consider the configuration in which Q lies between P
and S; then Q lies on the short arc AP . The other configuration is treated
analogously. (Note that by assumption P 6= Q.)
We show that N lies on the radical line of Γ1 and Γ2 . We have ∠ASP =
90◦ = ∠AM P , so ASP M is a cyclic quadrilateral by the Thales’s theorem.
Hence we have
∠P SM = ∠P AM = ∠P AB = 90◦ − 21 ∠ABP
= 90◦ − (180◦ − ∠AQP ) = 90◦ − ∠AQS,
where in the second to last step, we applied the inscribed angle theorem.
Moreover, using the sum of angles in 4AQS, we find that 90◦ − ∠AQS =
∠QAS. As ASQT is a cyclic quadrilateral with |QT | = |QS| (since QS
and QT are both tangent to Γ1 ), we have ∠QAS = ∠QT S = ∠QST . To
summarise, we have ∠P SM = 90◦ − ∠AQS = ∠QAS = ∠QST = ∠P ST .
Hence S, T , and M are collinear.
From this, it follows that N is the intersection of ST and AQ. In the
cyclic quadrilateral ASQT we find, using the power of a point theorem,
that N T · N S = N A · N Q. Note that the left hand side of N is the power
of N with respect to Γ1 and that the right hand side is the power of N
with respect to Γ2 . Therefore N lies on the radical line of Γ1 and Γ2 .
34
Problems
Part 1
1. A booklet is made by forming a stack of 11 sheets of paper and then folding

the stack in half. The pages of the booklet are numbered, like in a book,
from 1 to 44, where the front cover gets the number 1 and the back cover
gets number 44. Now the booklet is opened up and from the stack of 11
sheets we take the one in the middle.
Adding up the four numbers on this sheet, what outcome do we get?
A) 82 B) 84 C) 86 D) 88 E) 90
2. We draw a circle through the four vertices of a square

of area 1. Then, we draw a square around this circle
in such a way that that all the sides are tangent to the
circle.
What is the area of this square?
10 3 5 100
A) 7 B) 2 C) 3 D) 2 E) 49
3. Quintijn has three equally big and equally filled bottles of wine. Bottles
1 and 3 contain the same kind of white wine, while bottle 2 contains red
wine. Quintijn now pours a small amount of wine from bottle 1 into bottle
2. Next, after mixing the content of bottle 2 really well, he pours the same
amount from bottle 2 into bottle 3. In the same way, he pours the same
amount of wine from bottle 3 into bottle 1. Now all bottles contain the
same amount of wine as they did at the start. However, the content of
each bottle is polluted with wine of the other type.
Which bottle is polluted most?
A) bottle 1 B) bottle 2 C) bottle 3
D) all three equally E) you cannot determine this
35
4. Aad, Bep, Cor, Dirk, Eva, and Fenna are sitting in this order in a circle
around the campfire. Aad has a torch. He gives it to Bep who is sitting
one place to his right. She gives the torch to Dirk, who is sitting two places
to her right. He gives the torch to Aad, who is sitting three places to his
right, et cetera. It happens that someone must give the torch to the person
sitting, for instance, six or twelve positions to the right. Then this person
gives the torch to him- or herself.
When Dirk is given the torch for the hundredth time, who does he pass
the torch to?
A) Aad B) Bep C) Cor D) Dirk E) Eva
5. When a cube is cut with a plane, a cross section is created. This figure is
formed by the lines where the plane cuts the facets of the cube. In the left
figure, you can see an example in which the cross section is a triangle.
In the right figure, a flattened cube is drawn with the cut lines drawn on
the facets.
What is the cross section of the cube corresponding to this figure?
A) a triangle B) a square C) a rectangle, but not a square
D) a hexagon E) a parallelogram, but not a rectangle
6. On a machine there are three buttons. The first button can be used to
add 20 marbles to a tray in the machine. The second button can be used
to increase the number of marbles in the tray by 20%, after which 15
additional marbles are added. The third button can be used to increase
the number of marbles in the tray by 50%. If pressing a button would
cause the number of marbles to be non-integral, the pressing of the button
is not allowed. In the beginning the tray is empty. After some button
presses, there are 91 marbles in the tray.
How often has the first button been pressed?
A) 0 B) 1 C) 2 D) 3 E) 4
36
7. Ria, Sophie, and Tine are sitting around a round table in clockwise order
and are playing a game with chips. Ria starts with 3 chips, Sophie with 4
chips, and Tine with 5 chips. In each round they simultaneously give chips
to one of their neighbours. Each player can choose to give 2 chips to her
right neighbour or 1 chip to her left neighbour. If someone has no more
chips, the game ends.
Is it possible for the players to have the same number of chips after some
number of rounds, and if this is the case, how many rounds have to be
played at least to accomplish this?
A) no, it is impossible. B) yes, 3 rounds. C) yes, 6 rounds.
D) yes, 7 rounds. E) yes, 8 rounds.
8. Six people are sitting around a round table. Each of them is either a
knight or a knave. Knights always speak the truth, while knaves always
lie. Each of them has a card containing a number. All numbers are different
and everyone knows the numbers of their two neighbours. When asked:
“Is your number greater than the numbers of both your neighbours?”,
everyone answers with “Yes”. When asked: “Is your number smaller than
the numbers of both your neighbours?”, at least one person answers “Yes”
and at least one answers “No”.
What are the possible numbers of people answering “Yes” to this second
question?
A) 1 or 2 B) 1 or 3 C) 2 or 3 D) 2 or 4 E) 2, 3, or 4
37
Part 2
1. In the figure on the right there is a square with side

length 3. The square is divided into nine equal squares.
Then, another line is drawn that creates a pentagon
inside the middle square (coloured grey).
What is the area of this pentagon?
2. A palindromic number is a number that is the same when read from left
to right as when read from right to left. A number does not start with the
digit 0. To a six-digit palindromic number the palindromic number 21312
is added. The result is a seven-digit palindromic number.
What is this resulting number?
3. Using exactly six zeros and six ones we create two or more numbers which
we then multiply. For instance, we could get 10 × 10 × 10 × 10 × 10 × 10 =
1,000,000, or 10,011 × 100 × 1,100 = 1,101,210,000.
What is the largest possible result we could get in this way?
4. Pieter is staying at a hotel. The hotel has a ground floor (numbered 0) and
seven additional floors numbered 1 to 7. Pieter wants to make a trip by
elevator. He starts on one of the floors 1 to 7 and ends at the ground floor.
In between, he travels from floor to floor, never stopping at a previously
visited floor and never stopping at the ground floor (except for the last
stop).
The distance between any two consecutive floors is 3 metres. If, for ex-
ample, Pieter starts at floor 3, then goes to floor 6, then to floor 4 and
finally to the ground floor, he travels a total of (3 + 2 + 4) × 3 = 27 metres.
What is the maximum length of his trip in metres?
5. How many 3-digit numbers (the first digit cannot be 0) have the property
that adding all the digits gives a strictly greater result than multiplying
all the digits?
38
6. The mayor of a town wants to build a network of express trams. She wants
it to meet the following conditions:
• There are at least two distinct tram lines.
• Each tram line serves exactly three stops (also counting the start and
terminus).
• For each two tram stops in the town there is exactly one tram line
that serves both stops.
What is the minimum number of stops that the mayor’s tram network can
have?
E
7. A rectangle ABCD and a point E are given. D C
Line segments BE and BA have the same
length. Line segments CE and CB also have
the same length. Moreover, the area of rect-
angle ABCD is four times as large as the area
A B
of triangle BCE. Side AD has length 10.
What is the length of diagonal AC?
8. We consider ways to fill a 5×5-board by writing a 1 or a 3 in each square.

Such a filling is called balanced if the following holds:
• If you take an arbitrary 3 × 3-square of the board and multiply all

the numbers that it contains, and after that you do the same for an
arbitrary 4×4-square, then the second result is always three times as
large as the first result.
In the figure on the right, you see a filling that is 1 1 3 3 1
not balanced. For example, when multiplying the
numbers in the indicated 4×4-square, the result 3 1 1 1 3
is nine times as large as the result obtained by 1 3 1 3 1
multiplying the numbers in the indicated 3 × 3-
3 1 1 3 1
square.
Give (on the answer form) a balanced filling of 1 1 3 1 1
the board containing a maximum number of 3-s.
39
Solutions
Part 1
1. E) 90 5. C) a rectangle, but not a square
2. D) 2 6. D) 3
3. B) bottle 2 7. A) no, this is impossible.
4. B) Bep 8. D) 2 or 4
Part 2
11
1. 12 5. 199
2. 1008001 6. 7
3. 12,321,000,000 7. 20
4. 84 8. There are two possibilities:

1 1 1 1 1 1 1 3 1 1
1 1 1 1 1 1 1 3 1 1
3 3 3 3 3 1 1 3 1 1
1 1 1 1 1 1 1 3 1 1
1 1 1 1 1 1 1 3 1 1
40
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

AA01.pdf 22-05-2008 14:01:01

and the team selection for IMO 2017 Brazil
MY

CY
We eat problems
CMY

for breakfast.

Olympiad 2016
International
OmslagNWO2017-v3.indd 1 10-06-17 17:15
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
FOUNDATION COMPOSITIO MATHEMATICA

OmslagNWO2017-v3.indd 2 10-06-17 17:15

Introduction
2016, held at the participating schools. The paper consisted of eight multiple
choice questions and four open questions, to be solved within 2 hours. In
this first round 11101 students from 347 secondary schools participated.
The 122 best students were invited to the final round. Also some outstanding
participants in the Kangaroo math contest or the Pythagoras Olympiad
an invitation to some training sessions at the universities, in order to prepare
them for their participation in the final round.
The final round in September contained five problems for which the students
had to give extensive solutions and proofs. They were allowed 3 hours for
this round. After the prizes had been awarded in the beginning of November,
the Dutch Mathematical Olympiad concluded its 55th edition 2016.

represent the Netherlands at the EGMO in Zürich, Switzerland, from 6
until 12 April. The team brought home a silver medal, a bronze medal,
and a honourable mention; a very nice achievement. For more information
about the EGMO (including the 2017 paper), see www.egmo.org.
In March a selection test of three and a half hours was held to determine the
ten students participating in the Benelux Mathematical Olympiad (BxMO),
held in Namur, Belgium, from 5 until 7 May. The Dutch team received
two gold medals, three silver medals and four bronze medals, and managed
to get the highest total score. For more information about the BxMO
1
Rio de Janeiro, from 8 until 15 July.
the solutions.
2
Dutch delegation
The Dutch team for IMO 2017 in Brazil consists of
• Nils van de Berg (17 years old)

• Wietze Koops (16 years old)
– bronze medal at BxMO 2016, bronze medal at BxMO 2017
– silver medal at BxMO 2015, silver medal at BxMO 2016
• Ward van der Schoot (18 years old)
• Gabriel Visser (19 years old)
– bronze medal at IMO 2016
• Lammert Westerdijk (16 years old)

– participated in BxMO 2017

• Jetze Zoethout (observer B), Utrecht University
3
Problems
A-problems
1. Frank has two integers that add up to 26. Kees adds two more integers to
it and gets 41. Pieter adds another two integers and gets 58.
At least how many of the six integers that were added up are even?
A) 0 B) 1 C) 2 D) 3 E) 4
2. In a square with side length 12, line segments are drawn

between the vertices and the midpoints of the sides and
between the midpoints of opposite sides (see the figure).
In this way, a star shaped figure is created.
What is the area of this figure?
A) 12 B) 16 C) 20 D) 36 E) 48
3. A positive integer is called fully divisible if it is divisible by each of its digits.

Moreover, these digits must all be distinct (and nonzero). For example, 162
is fully divisible, because it is divisible by 1, 6, and 2.
How many fully divisible two-digit integers are there?
A) 4 B) 5 C) 6 D) 7 E) 8
4. An eight is a figure consisting of two equal circles touch-

ing each other, like , or . In the figure you see 66
circles stacked in the shape of a triangle.
How many eights can you find in this stack?
A) 99 B) 108 C) 120 D) 135 E) 165
5. Five integers are written around a circle. Two neighbouring numbers never
add up to a multiple of three. Also, a number and its two neighbours never
add up to a multiple of three.
How many of the five integers are multiples of three?
A) 1 B) 2 C) 3 D) 4 E) 2 and 3 are both possible
4
6. In the figure you see a wire-frame model of a 2×2×2-cube
consisting of 8 small cubes with side length 1 dm. This
figure uses 54 dm of wire.
How many dm of wire are needed for a wire-frame model
of a 10×10×10-cube consisting of one thousand small
cubes with side length 1 dm?
A) 121 B) 1000 C) 1210 D) 3000 E) 3630
7. A square board is divided into 4×4 squares. At the start, all squares are
white. Now, we want to colour some of the squares blue, in such a way
that each blue square will be adjacent to exactly one white square (two
squares are called adjacent if they have a side in common).
What is the maximum number of squares that we can colour blue?
A) 6 B) 8 C) 10 D) 12 E) 14
8. For three distinct positive integers a, b, and c we have a + 2b + 3c < 12.

Which of the following inequalities is certainly satisfied?
A) 3a + 2b + c < 17 B) a + b + c < 7 C) a − b + c < 4
D) b + c − a < 3 E) 3b + 3c − a < 6
B-problems
1. We construct a list of all positive integers that divide 707070. The numbers
are listed in decreasing order. The first number in the list is therefore
707070 and the last one is 1.
What is the seventh number in the list?
2. In the AO-language all words consist of only A’s and O’s and every possible
sequence of A’s and O’s is a word. There are, for example, eight three letter
words: ‘OOO’, ‘OOA’, ‘OAO’,. . . , ‘AAO’, and ‘AAA’. Words that contain
the letter combinations ‘AO’ and ‘OA’ equally often are called special. For
example, ‘AOAAOOOAA’ is special, because the word contains both letter
combinations ‘AO’ and ‘OA’ twice.
Find a special word consisting of four A’s and four O’s with the additional
property that after removing any of its letters, the resulting seven letter
word is again special.
5
D C
3. In the square ABCD lies a point U such that BU U
and AB have the same length. Point V is the
intersection of BU and the diagonal AC. The size V
?
of angle DAU is 28 degrees.
What is the size of the angle at V in triangle BV C?
28◦
A B
4. Seven people are suspects of a theft:
- Alex, a brown-haired man with blue eyes;
- Boris, a blond man with green eyes;
- Chris, a blond man with brown eyes;
- Denise, a blond woman with brown eyes;
- Eva, a brown-haired woman with blue eyes;
- Felix, a brown-haired man with brown eyes;
- Gaby, a blond woman with blue eyes.
Detectives Helga, Ingrid, and Julius know that one of the suspects is the
thief. After conducting some investigations they share their information.
Helga: “I know the eye and hair colour of the thief, but I do not know
who the thief is.”
Ingrid did not hear Helga and says:
“I know the hair colour and the gender, but I do not know who
the thief is.”
At last, Julius says:
“First I knew only the gender of the thief, but after hearing you
I know who the thief is.”
The detectives spoke the truth. Who is the thief?
6
Solutions
A-problems
1. C) 2 5. B) 2
2. D) 36 6. E) 3630
3. B) 5 7. D) 12
4. E) 165 8. D) b + c − a < 3
B-problems
1. 70707
2. ‘AAOOOOAA’ (or ‘OOAAAAOO’)
3. 101◦
4. Denise
7
Problems
B-problems
B1. How many of the integers from 10 to 99 have the property that the number
equals four times the sum of its two digits?
B2. In a box there are 100 cards that are numbered from 1 to 100. The numbers
are written on the cards. While being blindfolded, Lisa is going to draw
one or more cards from the box. After that, she will multiply together the
numbers on these cards.
Lisa wants the outcome of the multiplication to be divisible by 6. How
many cards does she need to draw to make sure that this will happen?
D C
B3. In the trapezium ABCD the sides AB and CD are

parallel and we have |BC| = |CD| = |DA| = 12 |AB|. On
the exterior of side AB there is an equilateral triangle A
R Q
B
BAP . The point Q is the intersection of P C and AB,
and R is the intersection of P D and AB (see the figure).
The area of triangle BAP is 12.
Determine the area of quadrilateral QCDR.
P
B4. At championships of ‘The Settlers of Catan’, three participants play against
each other in each game. At a certain championship, three of the parti-
cipants were girls and they played against each other in the first game.
Each pair of participants met each other in exactly one game and in each
game at least one girl was playing.
What is the maximum number of participants that could have competed in
the championship?
8
B5. Triangle ABC has a right angle at B. Moreover,
the side length of AB is 1 and the side length of
A
BC is 2. On the side BC there are two points
D and E such that E lies between C and D and
1 G F
DEF G is a square, where F lies on AC and G
lies on the circle through B with centre A.
D E C
Determine the length of DE. B
2
C1. A positive integer is called 2016-invariant if the sum of its digits does
not change when you add 2016 to the integer. For example, the integer
8312 is 2016-invariant: the sum of the digits of 8312 is 8 + 3 + 1 + 2 = 14,
and this equals the sum of the digits of 8312 + 2016 = 10328, which is
1 + 0 + 3 + 2 + 8 = 14.
(a) Determine the largest four-digit number that is 2016-invariant.
(b) There are 9999 positive integers having at most four digits.
Determine how many of these are 2016-invariant.
C2. For the upcoming exam, the desks in a hall are arranged in n rows containing
m desks each. We know that m ≥ 3 and n ≥ 3. Each of the desks is occupied
by a student. Students who are seated directly next to each other, in front
of each other, or diagonally from each other, are called neighbours. Thus,
students in the middle of the hall have 8 neighbours. Before the start of
the exam, each student shakes hands once with each of their neighbours.
In total, there are 1020 handshakes.
Determine the number of students.
9
Solutions
B-problems
1. 1
2. 68
3. 5
4. 7
2
5. 5
C-problems
C1. When adding two integers, we write one on top of the other and then add
the digits from right to left. If at a certain position the sum of the two
digits is greater than 9, a carry occurs at this position: we must carry a 1
to the addition of the two digits in the next position (the position to the
left).
Let n be a number whose sum of digits equals s. Now consider the
number n + 2016. We can derive the sum of its digits from the addition
procedure. If no carry occurs, then the sum of the digits of n + 2016 equals
s + 2 + 1 + 6 = s + 9. If carries do occur, then with each carry the sum
of the digits decreases by 9. After all, if two digits x and y add up to a
number greater than 9, the digit below these two will be x + y − 10 instead
of x + y, while we carry a 1 causing the digit to the left to increase by 1.
In total, the sum of the digits thus decreases by 9 for each carry.
For example, taking n = 1015 (and s = 1 + 0 + 1 + 5 = 7), then there will
be one carry in the addition n + 2016 = 3031. For the sum of the digits we
have 3 + 0 + 3 + 1 = (s + 9) − 1 · 9.
As another example, take n = 3084 (where s = 15). Then, there will be two
carries in the addition n + 2016 = 5100 and indeed we have 5 + 1 + 0 + 0 =
(s + 9) − 2 · 9.
Hence, a number is 2016-invariant if and only if exactly one carry occurs
when adding 2016 to it.
10
(a) After all observations in the previous paragraph, we conclude that
we are looking for the greatest four-digit number having the property
that you need to carry exactly one 1 when adding 2016 to it. This
number is 9983. In the addition 9983 + 2016 exactly one carry occurs
and if you raise one of the digits 8 or 3, a second carry occurs.
(b) To count how many of the numbers n from 1 to 9999 give exactly
one carry when adding 2016 to it, we consider four cases according to
the position in which the carry occurs. In each case we consider the
possible values of the digits of n.
• First, consider the case that the only carry occurs in the thousands
place. The thousands digit must then be either 8 or 9. The unit
digit can only be 0, 1, 2, or 3. The tens digit can be any digit
except for 9 (because 1 + 9 would give a carry in the tens place).
The hundreds digit can be any value. Hence, in total there are
2 · 10 · 9 · 4 = 720 numbers in this case.
• There are no numbers for which a carry occurs only in the hundreds
place (a carry there can only occur if we also carry a 1 from the
tens place).
• Next, consider the case that the only carry occurs in the tens
place. For the units digit, the only possible values are 0, 1, 2,
and 3. The tens digit must be a 9. Because you now need to
carry a 1 from the tens place to the hundreds place, the digit in
the hundreds place cannot be a 9 anymore since this would cause
another carry (for example, in the addition 990 + 2016). The digit
in the thousands place can take all values except 8 and 9. Hence,
in total there are 8 · 9 · 1 · 4 = 288 numbers in this case.
• Finally, consider the case that the only carry occurs in the units
place. For this digit, the possible values are 4, 5, 6, 7, 8, and 9.
The digit in the tens place cannot be 9 or 8, because then another
carry would occur (for example in 84 + 2016). The hundreds digit
can be any value and the thousands digit can be any value except
8 and 9. Hence, in total we find 8 · 10 · 8 · 6 = 3840 numbers in
this case.
Altogether, there are 720 + 288 + 3840 = 4848 numbers from 1 to 9999
that are 2016-invariant.
11
C2. There are four students seated in the corners, since n, m ≥ 2. They each
shook the hand of three other people. Not counting the students in the
corners, there are m − 2 students who are seated completely in the front,
m − 2 who are seated in the back, n − 2 students who are seated in the
leftmost row and n−2 in the rightmost row. These 2n+2m−8 students each
shook the hand of five other people. The remaining mn−4−(2n+2m−8) =
mn − 2n − 2m + 4 students each shook the hand of eight other people.
Adding the numbers of handshakes for each of the students gives a total of
4 · 3 + (2n + 2m − 8) · 5 + (mn − 2n − 2m + 4) · 8 = 8mn − 6n − 6m + 4.
This number is exactly twice the number of handshakes, because each

handshake involves two students. Hence, we see that 8mn − 6n − 6m + 4 =
2040, or 16mn − 12n − 12m = 4072. We can rewrite this equation as
(4m − 3) · (4n − 3) = 4081.
The integers 4m − 3 and 4n − 3 both have remainder 1 when dividing by 4.

Also, 4m − 3 and 4n − 3 must both be greater than 1. If we look at the
prime factorisation 4081 = 7 · 11 · 53, then we see that there is only one
way to write 4081 as the product of two integers greater than 1 that each
have remainder 1 when dividing by 4: namely as 4081 = 77 · 53. Hence, we
find 4m − 3 = 77 and 4n − 3 = 53 (or the other way around). It follows
that m = 20 and n = 14 (or the other way around). In both cases we see
that the total number of students equals 20 · 14 = 280.
12
Problems
1. (a) On a long pavement, a sequence of 999 integers is written in chalk.

The numbers need not be in increasing order and need not be distinct.
Merlijn encircles 500 of the numbers with red chalk. From left to right,
the numbers circled in red are precisely the numbers 1, 2, 3, . . . , 499, 500.
Next, Jeroen encircles 500 of the numbers with blue chalk. From
left to right, the numbers circled in blue are precisely the numbers
500, 499, 498, . . . , 2, 1.
Prove that the middle number in the sequence of 999 numbers is
circled both in red and in blue.
(b) Merlijn and Jeroen cross the street and find another sequence of 999
integers on the pavement. Again Merlijn circles 500 of the numbers
with red chalk. Again the numbers circled in red are precisely the
numbers 1, 2, 3, . . . , 499, 500 from left to right. Now Jeroen circles
500 of the numbers, not necessarily the same as Merlijn, with green
chalk. The numbers circled in green are also precisely the numbers
1, 2, 3, . . . , 499, 500 from left to right.
Prove: there is a number that is circled both in red and in green that
is not the middle number of the sequence of 999 numbers.
2. For an integer n ≥ 1 we consider sequences of 2n numbers, each equal to 0,

−1 or 1. The sum product value of such a sequence is calculated by first
multiplying each pair of numbers from the sequence, and then adding all
the results together.
For example, if we take n = 2 and the sequence 0, 1, 1, −1, then we find the
products 0·1, 0·1, 0·−1, 1·1, 1·−1, 1·−1. Adding these six results
gives the sum product value of this sequence: 0+0+0+1+(−1)+(−1) = −1.
The sum product value of this sequence is therefore smaller than the sum
product value of the sequence 0, 0, 0, 0, which equals 0.
Determine for each integer n ≥ 1 the smallest sum product value that such
a sequence of 2n numbers could have.
Attention: you are required to prove that a smaller sum product value is
impossible.
13
3. Find all possible triples (a, b, c) of positive integers with the following
properties:
• gcd(a, b) = gcd(a, c) = gcd(b, c) = 1;
• a is a divisor of a + b + c;
• b is a divisor of a + b + c;
• c is a divisor of a + b + c.
(Here gcd(x, y) is the greatest common divisor of x and y.)
4. Version for junior students D

In a quadrilateral ABCD the intersection of the
diagonals is called P . Point X is the orthocentre C
of triangle P AB. (The orthocentre of a triangle is Y
the point where the three altitudes of the triangle

intersect.) Point Y is the orthocentre of triangle P
P CD. Suppose that X lies inside triangle P AB

X
and Y lies inside triangle P CD. Moreover, sup- A
pose that P is the midpoint of line segment XY .
Prove that ABCD is a parallelogram.
B
4. Version for senior students C

In the acute triangle ABC, the midpoint of side M
BC is called M . Point X lies on the angle bi- Y
sector of ∠AM B such that ∠BXM = 90◦ . Point X
B
Z
Y lies on the angle bisector of ∠AM C such that
∠CY M = 90◦ . Line segments AM and XY in-
tersect in point Z.
A
Prove that Z is the midpoint of XY .
5. Bas has coloured each of the positive integers. He had several colours at
his disposal. His colouring satisfies the following requirements:
• each odd integer is coloured blue;
• each integer n has the same colour as 4n;
• each integer n has the same colour as at least one of the integers n + 2
and n + 4.
Prove that Bas has coloured all integers blue.
14
Solutions
1. (a) For brevity, we will say that a number encircled in red is a red number,
and similarly for blue. So some numbers could be both red and blue.
Since there are 999 numbers written on the pavement, of which 500
are red and 500 are blue, we have at least one bicoloured number
by the pigeonhole principle. Consider such a bicoloured number and
suppose it is the number k. From left to right, the red numbers form
the sequence 1, 2, . . . , 500. Hence, to the left of the bicoloured number
we have the red numbers 1 to k − 1, and to the right we have the red
numbers k + 1 to 500. The blue numbers are written in the opposite
order: from left to right they form the sequence 500, 499, . . . , 1. To
the left of the bicoloured number we therefore have the blue numbers
k + 1 to 500, and to the right we have the blue numbers 1 to k − 1.
We count how many numbers there are on each side of the bicoloured
number. On the left we have the red numbers 1 to k − 1 and the blue
numbers k + 1 to 500. Hence, there are at least 499 distinct numbers
on the left. On the right we have the red numbers k + 1 to 500 and
the blue numbers 1 to k − 1. Again at least 499 distinct numbers.
Since there are only 999 = 499 + 1 + 499 numbers on the pavement, we
have already considered all numbers on the pavement. We conclude
that there are precisely 499 numbers on each side of the bicoloured
number. The bicoloured number is therefore precisely in the middle
of the sequence.
(b) Just as in part (a), the pigeonhole principle yields that at least one
number is bicoloured, i.e. both red and green. If more than one number
is bicoloured, one of them is not exactly in the middle of the sequence
and we are done. Therefore, it suffices to examine the case where
there is exactly one bicoloured number. Let this number be k.
Again, we count how many numbers there are on each side of the
bicoloured number. On the left we have the red numbers 1 to k − 1
and the green numbers 1 to k − 1. Since none of these numbers is
bicoloured, there are at least 2 · (k − 1) distinct numbers on the left.
On the right we have the red numbers k + 1 to 500 and the green
numbers k + 1 to 500. Since none of these numbers is bicoloured,
we have at least 2 · (500 − k) distinct numbers on the right. Since
2 · (k − 1) + 1 + 2 · (500 − k) = 999, we have already counted all numbers.
On the left of the bicoloured number we therefore have exactly 2·(k−1)
numbers, and on the right we have exactly 2 · (500 − k) numbers. Since
2 · (k − 1) is an even number, it is unequal to 499. We conclude that
the bicoloured number is not exactly in the middle of the sequence.
15
2. Suppose that our sequence has x ones, y minus ones (and hence 2n − x −
y zeroes). We calculate the sum product value of the sequence (as an
expression in x and y).
In the sum product value, six different types of terms occur: 1 · 1, 1 · −1,
−1 · −1, 1 · 0, −1 · 0, and 0 · 0. Only the first three types contribute since
the other types are equal to 0.
The number of terms of the type 1 · 1 = 1 equals the number of ways to
select two out of x ones. This can be done in x(x−1) 2 ways: there are x
options for the first 1, and then x − 1 options for the second 1. Since the
order in which we select the two ones does not matter, we effectively count
each possible pair twice.
y(y−1)
Similarly, the number of terms of the type −1 · −1 = 1 is equal to 2 .
The number of terms of the type 1 · −1 = −1 is equal to xy, since there
are x options for choosing a 1 and, independently, there are y options for
choosing a −1.
In total, we obtain a sum product value of
x(x − 1) y(y − 1) (x − y)2 − (x + y)

S= ·1+ · 1 + xy · −1 = .
2 2 2
Since (x − y)2 ≥ 0 (squares are non-negative) and −(x + y) ≥ −2n (there

are only 2n numbers in the sequence), we see that S ≥ 0−2n
2 = −n. So the
sum product value cannot be smaller than −n. If we now choose x = y = n,
then (x − y)2 = 0 and −x − y = −2n, which imply a sum product value of
exactly 0−2n
2 = −n. Hence, the smallest possible sum product value is −n.
3. The problem is symmetric in a, b, and c. That is, if we consistently swap a

and b, or a and c, or b and c, then the conditions on (a, b, c) do not change.
We will therefore consider solutions for which a ≤ b ≤ c. The remaining
solutions are then found by permuting the values of a, b, and c.
Since a and b are positive, we see that a + b + c > c. Since c is largest among
the three numbers, we also have a + b + c ≤ 3c. Since we are given that
a + b + c is a multiple of c, we are left with two possibilities: a + b + c = 2c
or a + b + c = 3c. We consider both cases separately. If a + b + c = 3c,
then a, b, and c must all be equal, because otherwise, the fact that a, b ≤ c
implies that a + b + c < 3c. This means that gcd(b, c) = gcd(c, c) = c. Since
gcd(b, c) must be equal to 1, we find (a, b, c) = (1, 1, 1). This is indeed a
solution, since gcd(1, 1) = 1 and 1 is a divisor of 1 + 1 + 1.
16
If a + b + c = 2c, then c = a + b. We know that b must be a divisor of
a + b + c = 2a + 2b. Since a > 0, we have 2a + 2b > 2b. Since b ≥ a, we
also have 2a + 2b ≤ 4b. Therefore, since 2a + 2b must be a multiple of b,
there are only two possibilities: 2a + 2b = 3b or 2a + 2b = 4b. Again, we
consider these cases separately.
If 2a + 2b = 3b, then b = 2a. Similarly to the first case, we find that
a = gcd(a, 2a) = gcd(a, b) = 1. Therefore, b = 2 and c = a + b = 3. The
resulting triple (a, b, c) = (1, 2, 3) is indeed a solution, since gcd(1, 2) =
gcd(1, 3) = gcd(2, 3) = 1 and 1 + 2 + 3 = 6 is divisible by 1, 2, and 3.
If 2a + 2b = 4b, then a = b. Again we see that a = gcd(a, a) = gcd(a, b) = 1.
From b = a = 1 it follows that c = a + b = 2 and hence (a, b, c) = (1, 1, 2).
This is indeed a solution since gcd(1, 1) = gcd(1, 2) = 1 and 1 + 1 + 2 = 4
is divisible by 1 and 2.
We conclude that the solutions for which a ≤ b ≤ c holds are: (a, b, c) =
(1, 1, 1), (a, b, c) = (1, 1, 2), and (a, b, c) = (1, 2, 3). Permuting the values of
a, b and c, we obtain a total of ten solutions (a, b, c):
(1, 1, 1), (1, 1, 2), (1, 2, 1), (2, 1, 1),
(1, 2, 3), (1, 3, 2), (2, 1, 3), (2, 3, 1), (3, 1, 2), (3, 2, 1).
D
Let K be the intersection of AX and BD, C
L
and let L be the intersection of CY and BD. Y
Consider the triangles P LY and P KX.
The angles ∠P LY and ∠P KX are both P
right angles. The angles ∠Y P L and ∠XP K
are opposite angles and therefore equal. X K
A
Since |P X| = |P Y |, we see that the tri-
angles P KX and P LY are congruent (SAA).
Hence, we have |P K| = |P L|. B
Now consider triangles P AK and P CL. Angles ∠AKP and ∠CLP are both
right angles. Angles ∠KP A and ∠LP C are opposite angles, hence equal.
We have already shown that |P K| = |P L|. Therefore, triangles P AK and
P CL are congruent (ASA). From this, we conclude that |AP | = |P C|.
In a similar fashion, we may deduce that |BP | = |DP |. The two diagonals
of ABCD bisect each other, hence ABCD is a parallelogram.
17
4. Version for senior students
We start by observing that we have that ∠CM Y = 12 ∠CM A = 12 (180◦ −
∠AM B) = 90◦ − ∠XM B. Since ∠M XB = 90◦ , and the angles of triangle
BM X sum to 180◦ , we see that ∠CM Y = 90◦ − ∠XM B = ∠M BX.
Looking at triangles CM Y and M BX, we observe that ∠CM Y = ∠M BX,
∠M Y C = 90◦ = ∠BXM , and |CM | = |M B|. The two triangles are
therefore congruent (SAA). In particular, we obtain the equalities |M X| =
|CY | and |M Y | = |BX|.
Now consider triangle XY M . We already know that |M Y | = |BX| and
that ∠Y M X = ∠Y M A + ∠AM X = 12 ∠CM A + 12 ∠AM B = 12 · 180◦ =
90◦ . Since triangles XY M and M BX also share the side M X, they are
congruent (SAS).
In particular, we see that ∠M XY = ∠XM B. Since M X is the angle
bisector of ∠AM B, we have ∠XM B = ∠AM X. This implies that triangle
M XZ has two equal angles and is therefore an isosceles triangle with vertex
angle Z. We conclude that |M Z| = |XZ|.
In a similar manner, we see that triangles XY M and CM Y are congruent,
and find that ∠XY M = ∠CM Y = ∠Y M A. Triangle M Y Z is therefore
isosceles with vertex angle Z. This implies that |Y Z| = |M Z|. Together
with |M Z| = |XZ|, this concludes the proof.
5. Suppose that not all numbers are coloured blue. Then, there must be a
number k that is not blue. We will use this to derive a contradiction.
Without loss of generality, we may assume that k is coloured red. Since all
odd numbers are blue, k must be even, say k = 2m for some integer m ≥ 1.
From the second requirement, it follows that 8m is red as well. From the
third requirement, it now follows that at least one of the numbers 8m + 2
and 8m + 4 is red. However, 2m + 1 is odd and therefore blue, which by
the second requirement implies that 8m + 4 = 4 · (2m + 1) is blue as well.
So 8m + 2 must be red.
By the third requirement, 8m − 2 must be the same colour as 8m or 8m + 2.
Since both 8m and 8m + 2 are red, this implies that 8m − 2 must be red
as well. Since 8m and 8m − 2 are red, this implies (again by the third
requirement) that 8m − 4 is also red. The second requirement now implies
that (8m − 4)/4 = 2m − 1 is also red. But that is impossible since 2m − 1
is odd, and therefore blue.
We conclude that the assumption that not all numbers are blue leads to a
contradiction. Therefore, all numbers must be blue.
18
Problems
1. Let n be an even positive integer. A sequence of n real numbers is called

complete if for every integer m with 1 ≤ m ≤ n either the sum of the first
m terms of the sum or the sum of the last m terms is integral. Determine
the minimum number of integers in a complete sequence of n numbers.
2. A function f : Z>0 → Z is given, which has the following properties:

(i) f (p) = 1 for all prime numbers p,
(ii) f (xy) = yf (x) + xf (y) for all x, y ∈ Z>0 .
Determine the smallest n ≥ 2016 satisfying f (n) = n.
3. Let ABC be a triangle with ∠A = 90◦ and let D be the orthogonal

projection of A onto BC. The midpoints of AD and AC are called E
and F , respectively. Let M be the circumcentre of 4BEF . Prove that
AC k BM .
4. A quadruple (a, b, c, d) of positive integers with a ≤ b ≤ c ≤ d is called good

if we can colour each integer red, blue, green or purple, in such a way that
• of each a consecutive integers at least one is coloured red;
• of each b consecutive integers at least one is coloured blue;
• of each c consecutive integers at least one is coloured green;
• of each d consecutive integers at least one is coloured purple.
Determine all good quadruples with a = 2.
5. Determine all pairs of prime numbers (p, q) such that p2 + 5pq + 4q 2 is the
square of an integer.
19
Solutions
1. We will prove that the minimum number of integers in a complete sequence
is 2. First consider the case n = 2. Let a1 and a2 be the numbers in
the sequence. Then either a1 or a2 is integral. Without loss of generality
assume a1 is integral. Moreover, a1 + a2 is integral, hence also a2 is integral.
Therefore, the sequence contains at least two integers.
Now consider the case n > 2. Write n = 2k (because n is even) with
k ≥ 2. Then either a1 + a2 + . . . + ak or ak+1 + ak+2 + . . . + a2k is integral.
But, as the sum of both expressions is also integral, they are both integral.
Moreover, either a1 + a2 + . . . + ak−1 or ak+2 + ak+3 + . . . + a2k is integral.
This yields that either ak or ak+1 is integral. Moreover, we know that a1
or a2k is integral, and these do not coincide with ak or ak+1 because k ≥ 2.
Hence, at least two different numbers are integers.
Finally, we will show for each even integer n that it is possible to write a
complete sequence with exactly two integers. Again write n = 2k. If k is
odd, we take a1 = ak+1 = 1 and all other terms equal to 12 . The sum of
all numbers in the sequence is integral, hence it is sufficient to show that
the sum of the first or last m terms is integral for 1 ≤ m ≤ k; the cases in
which m > k follow directly. For odd m ≤ k, the sum of the first m terms
is integral, for even m < k, the sum of the last m terms is integral.
If k is even, we take a1 = ak = 1 and all other terms equal to 12 . For
odd m < k, the sum of the first m terms is integral, and for even m ≤ k,
the sum of the last m terms is integral. Moreover, the sum of all terms is
integral, hence the requirement is also met for m > k.
We conclude that the minimum number of integers in a complete sequence
of n numbers is 2.
2. We will first prove that for prime numbers p and positive integers k we
have f (pk ) = kpk−1 . We will prove this using mathematical induction to k.
For k = 1, the statement becomes f (p) = 1, which is known to hold. Now,
let l ≥ 1 and suppose that we proved the statement for k = l. Consider
k = l + 1. Then we apply the second property with x = p and y = pl :
f pl+1 = f p · pl = pl · f (p) + p · f pl = pl + p · lpl−1 = (l + 1)pl .

This finished the induction. Now we will prove that for distinct prime
numbers p1 , p2 , . . . , pt and positive integers a1 , a2 , . . . , at we have

f (pa1 1 pa2 2 · · · pat t ) = pa1 1 pa2 2 · · · pat t · ap11 + ap22 + · · · + aptt .
20
We will prove this by induction to t. For t = 1, this is exactly the formula
that we have just proven. Now, let r ≥ 1 and suppose that the statement
is proved for t = r. Then we apply the second property again:
ar+1
f pa1 1 pa2 2 · · · par r · pr+1
ar+1 ar+1
= pa1 1 pa2 2 · · · par r · f pr+1 + pr+1 · f (pa1 1 pa2 2 · · · par r )

ar+1 −1 ar+1
= pa1 1 p2a2 · · · par r · ar+1 pr+1 + pr+1 · pa1 1 pa2 2 · · · par r · ap11 + ap22 + · · · + aprr

ar+1 ar+1 ar+1
= pa1 1 pa2 2 · · · par r pr+1 · pr+1 + pa1 1 pa2 2 · · · par r pr+1 · ap11 + ap22 + · · · + aprr

ar+1
= pa1 1 pa2 2 · · · par r pr+1 · ap11 + ap22 + · · · + aprr + apr+1 r+1
.
This finishes the proof by induction.

For an integer n > 1 of the shape n = pa1 1 pa2 2 · · · pat t with pi distinct and
prime, the equality f (n) = n is equivalent to

pa1 1 pa2 2 · · · pat t · ap11 + ap22 + · · · + aptt = pa1 1 pa2 2 · · · pat t ,
or
a1 a2 at
p1 + p2 + ··· + pt = 1.
Multiplying this with p1 p2 · · · pt yields
a1 p2 p3 · · · pt + a2 p1 p3 · · · pt + . . . + at p1 p2 · · · pt−1 = p1 p2 · · · pt .
Assume without loss of generality that p1 is the smallest prime divisor of n.

In the expression above, p1 is a divisor of the right hand side and of each
term in the left hand side except for the first term possibly. But then p1
must also be a divisor of the first term. As p2 , . . . , pt are all prime numbers
unequal to p1 , this is only possible when p1 | a1 . In particular, we have
a1 ≥ p1 , yielding ap11 ≥ 1. We now see that equality must hold here, and
hence t = 1, because the sum of the apii would otherwise be greater than 1.
Hence, n = pp for a certain prime number p.
Now we are searching for the smallest n ≥ 2016 of the shape n = pp . As
33 = 27 and 55 = 3125, this smallest n is 3125.
3. Because of the right angles at A and D we have 4ADB ∼ 4CAB (AA).

This yields that |AD| |CA| 1 1
|AB| = |CB| . Because |AE| = 2 |AD| and |CF | = 2 |CA|,
|AE| |CF |
we also have that |AB| = |CB| . From the previous similarity, we also obtain
∠BAE = ∠BAD = ∠BCA = ∠BCF . Hence, using the (SAS) criterion,
we get 4AEB ∼ 4CF B, from which we obtain ∠ABE = ∠CBF .
21
Moreover, EF is a mid-parallel in triangle ADC, hence EF k BC. There-
fore, we have ∠CBF = ∠BF E. Using the inscribed angle theorem and
the sum of the angles in the isosceles triangle EBM , we obtain ∠BF E =
1 ◦ ◦
2 ∠BM E = 90 − ∠EBM . Therefore, ∠ABE = ∠CBF = 90 − ∠EBM .
◦
We conclude that ∠ABM = ∠ABE + ∠EBM = 90 , and AC k BM .
4. Each time, we only consider quadruples (a, b, c, d) of positive integers

satisfying a ≤ b ≤ c ≤ d. Each quadruple with b ≥ 6 meets the requirements:
in this case, we can colour the even numbers red, and the odd numbers
consecutively blue, green, purple, blue, green, purple, et cetera. Moreover,
each quadruple with b ≥ 4 and c ≥ 8 meets the requirements: in this
case, we can colour the even numbers red, the numbers 1 mod 4 blue, the
numbers 3 mod 8 green, and the numbers 7 mod 8 purple. We will prove
that these are the only possibilities.
Suppose that (2, b, c, d) is good. We consider a purple number. Its neigh-
bours must be red, so then we have 3 consecutive numbers of which none is
coloured blue. Hence, b ≥ 4. If b ≥ 6, then the quadruple is good, as we’ve
just seen. Now suppose that either b = 4 or b = 5. We were considering a
purple number with its two red neighbours. On at least one of both ends
there must be a blue number, say on the left side. Right left of this blue
number, there is another red number. So we got RBRP R. Of the next two
numbers in the sequence, there is at least one red one and at least one blue
one, so there is certainly no green one. In this way, we found 7 consecutive
numbers none of which is coloured green, hence c ≥ 8. We have already
seen that the quadruple is good in this case.
We conclude that the following quadruples (2, b, c, d) are good quadruples:
these with either b ≥ 6, or with b ≥ 4 and c ≥ 8.
5. Write p2 + 5pq + 4q 2 = a2 , for an integer a ≥ 0. The left hand side

equals (p + 2q)2 + pq, hence we can rewrite this to pq = a2 − (p + 2q)2 , or
pq = (a − p − 2q)(a + p + 2q). The second factor on the right hand side
is greater than p and greater than q, but it is a divisor of pq. Because p
and q are prime, it must be equal to pq, hence a + p + 2q = pq. Then,
we have a − p − 2q = 1. Subtracting these two equalities, we obtain
pq − 1 = (a + p + 2q) − (a − p − 2q) = 2(p + 2q), or pq − 2p − 4q − 1 = 0.
This is equivalent to (p − 4)(q − 2) = 9. The factor q − 2 cannot be negative,
hence also p − 4 is not negative. The factors must be equal to 1 and 9, or
to 9 and 1, or to 3 and 3. For (p, q), this yields the possibilities (5, 11),
(13, 3) and (7, 5), respectively. We can check that p2 + 5pq + 4q 2 is a square
in each of these cases, namely 282 , 202 and 182 , respectively. Therefore,
these three pairs are the solutions.
22
Problems
1. Let n be a positive integer. Suppose that we have disks of radii 1, 2, . . . , n.

Of each size there are two disks: a transparent one and an opaque one. In
every disk there is a small hole in the centre, with which we can stack the
disks using a vertical stick. We want to make stacks of disks that satisfy
the following conditions:
• Of each size exactly one disk lies in the stack.
• If we look at the stack from directly above, we can see the edges of all
of the n disks in the stack. (So if there is an opaque disk in the stack,
no smaller disks may lie beneath it.)
Determine the number of distinct stacks of disks satisfying these conditions.
(Two stacks are distinct if they do not use the same set of disks, or, if they
do use the same set of disks and the orders in which the disks occur are
different.)
2. Let n ≥ 4 be an integer. Consider a regular 2n-gon for which to every

vertex, an integer is assigned, which we call the value of said vertex. If four
distinct vertices of this 2n-gon form a rectangle, we say that the sum of
the values of these vertices is a rectangular sum.
Determine for which (not necessarily positive) integers m the integers
m + 1, m + 2, . . . , m + 2n can be assigned to the vertices (in some order) in
such a way that every rectangular sum is a prime number. (Prime numbers
are positive by definition.)
3. Determine all possible values of x1 + 1

y if x and y are non-zero real numbers
satisfying x3 + y 3 + 3x2 y 2 = x3 y 3 .
4. Let ABC be a triangle, let M be the midpoint of AB, and let N be the
midpoint of CM . Let X be a point satisfying both ∠XM C = ∠M BC and
∠XCM = ∠M CB such that X and B lie on opposite sides of CM . Let Ω
be the circumcircle of triangle AM X.
(a) Show that CM is tangent to Ω.
(b) Show that the lines N X and AC intersect on Ω.
23
Solutions
1. We say that a stack of disks is valid if it satisfies the conditions. Let an

denote the number of valid stacks with n disks (of radii 1, 2, . . . , n). We
show by induction on n that an = (n + 1)!. For n = 1, we note that we can
make two distinct stacks, namely one with the transparent disk, and one
with the opaque one. So suppose that for some n ≥ 1, we have shown that
an = (n + 1)!. Consider a valid stack with n + 1 disks. If we remove the
disk with radius n + 1, the edge of every disk is still visible from directly
above, so we are left with a valid stack of n disks. Therefore, every valid
stack of n + 1 disks can be made by adding a disk of radius n + 1 to a valid
stack of n disks at some location. This can in principle be done in n + 1
ways: above the top disk, above the second highest disk, . . . , above the
bottom disk, and under the bottom disk. The disk of radius n + 1 is always
visible, regardless of where in the stack it is inserted. If it is inserted under
the bottom disk, it may both be transparent or opaque, so there are 2an
valid stacks of n + 1 disks in which the disk of radius n + 1 is the bottom
disk. If an opaque disk of radius n + 1 is inserted anywhere but under the
bottom disk, the disks below this opaque disk will now longer be visible.
So in the remaining n locations, we can only insert the transparant disk
of radius n + 1. Therefore there are nan valid stacks in which the disk of
radius n + 1 is not the bottom disk. Hence
an+1 = 2an + nan = (n + 2)an = (n + 2)(n + 1)! = (n + 2)!,
which completes the induction.
2. Number the vertices of the 2n-gon from 1 up to 2n, clockwise, and let ai be
the value of vertex i. Since the number of vertices of the polygon is even,
we can pair each vertex to the one directly opposite to it. Sum the values
of each pair of opposite vertices to get the numbers si = ai + ai+n .
If four vertices A, B, C, and D lie in that order on a rectangle, then
∠ABC = 90◦ and ∠ADC = 90◦ , so B and D lie on the circle with
diameter AC by Thales’s theorem. As all vertices of the polygon lie on its
circumcircle, AC is the diameter of this circumcircle. Therefore A and C
are opposite vertices, and the same holds for B and D. Conversely, if A and
C are opposite vertices, and so are B and D, then ABCD is a rectangle,
again by Thales’s theorem. In short, four vertices form a rectangle if and
only if they form two pairs of opposite vertices.
So the condition in the problem is equivalent to si + sj being prime for
all 1 ≤ i < j ≤ n. Suppose that at least three of the si have the same
24
parity, say sj , sk , and sl . Then the sum of each pair of these three numbers
is even, but prime as well, so this sum must be 2 in all cases. Hence
sj + sk = sk + sl = sj + sl , from which follows that sj = sk = sl . Since
sj + sk = 2, we have sj = sk = sk = 1. Therefore at most two of the si are
even, and if three or more of the si are odd, then the odd si all equal 1.
The sum of all si is equal to the sum of the values of the vertices, which is
(m+1)+(m+2)+· · ·+(m+2n) = 2mn+ 12 ·2n·(2n+1) = 2mn+n(2n+1).
On the one hand, this sum is congruent to n modulo 2. On the other hand,
modulo 2 this sum must be congruent tot the number of odd si . Hence
the number of odd si has the same parity as n. As there are n of the si , it
follows that the number of even si is even. Since we have already seen that
the number of even si is at most 2, it follows this number is either 0 or 2.
Suppose that n = 4 and and that there are two even si . Then there are
also two odd si . The sum of the even si , as well as that of the odd si , is
equal to 2, so the sum of the si is equal to 4 = n. Now suppose that n ≥ 5
and that there are two even si . Then n − 2 ≥ 3 of the si are odd, so all odd
si are equal to 1. The sum of the two even si is 2, so the sum of the si is
equal to (n − 2) + 2 = n. Now suppose that there are no even si . Then all
si are odd, and therefore equal to 1, so again, the sum of the si equals n.
So in all of the cases, the sum of the si is n. We have also seen that this
sum is equal to 2mn + n(2n + 1). Therefore 2mn + n(2n + 1) = n, so
dividing by the non-zero integer n gives the equality 2m + 2n + 1 = 1.
Hence m = −n.
Finally, we show that if m = −n, there indeed exists a solution. Let
ai = i and an+i = 1 − i for 1 ≤ i ≤ n. Then the integers 1, 2, . . . , n and
the integers 0, −1, . . . , −n + 1 are values of vertices; these are precisely
the integers m + 1, m + 2, . . . , m + 2n for m = −n. Moreover, we have
si = i + (1 − i) = 1 for all i. Hence all rectangular sums are equal to 2,
which is a prime number.
Therefore m = −n is the only value of m for which the there exists a
solution.
3. We rewrite the equation as x3 + y 3 − x3 y 3 = −3x2 y 2 , which gives
(x + y)3 − x3 y 3 = x3 + 3x2 y + 3xy 2 + y 3 − x3 y 3

= −3x2 y 2 + 3x2 y + 3xy 2 = 3xy(−xy + x + y).
25
Using the identity a3 − b3 = (a − b)(a2 + ab + b2 ) we can also write this as
(x + y)3 − x3 y 3 = (x + y − xy) (x + y)2 + xy(x + y) + (xy)2

= (x + y − xy)(x2 + y 2 + 2xy + x2 y + xy 2 + x2 y 2 ).
Therefore the two right hand sides are equal, so we see that either x+y−xy =
0 or x2 +y 2 +2xy +x2 y +xy 2 +x2 y 2 = 3xy. The latter case can be rewritten
as
0 = x2 + y 2 − xy + x2 y + xy 2 + x2 y 2
= 12 (x − y)2 + 12 x2 (y 2 + 2y + 1) + 12 y 2 (x2 + 2x + 1)
= 21 (x − y)2 + 12 x2 (y + 1)2 + 12 y 2 (x + 1)2 .
So since the sum of three squares must be 0, each of the squares must be
0 as well. Hence x = y = −1. This indeed gives a solution of the given
equation, and we have x1 + y1 = 2.
In the remaining case we have x + y − xy = 0, so y1 + x1 − 1 = 0, from which
we deduce that x1 + y1 = 1. This is attained for example for x = y = 2.
Therefore the possible values are −2 and 1.
C
X
S
4. We consider the configuration in the figure; N
the proof is similar for the other configur-
ations. A B
M
(a) First, we have 4XM C ∼ 4M BC.
We have ∠AM X = 180◦ − ∠XM C − ∠BM C = 180◦ − ∠XM C −
|AM | |BM | |M C|
∠M XC = ∠M CX, and |M X| = |M X| = |CX| , so we have 4AM X ∼
4M CX. Hence ∠XAM = ∠XM C, so using the converse of the
tangent chord angle theorem we see that CM is tangent to Ω.
(b) Let S be the second intersection point of AC and Ω (or the point at
which AC and Ω are tangent; which then would be A). We need to
show that S lies on N X. By the tangent chord angle theorem we have
∠SM C = ∠SAM = ∠CAM . (If S = A, then ∠SM C = ∠CAM
follows from CA and CM both being tangents, as then 4CAM is
isosceles.) Therefore 4CSM ∼ 4CM A (AA), so |CM | |SM |
|CA| = |M A| .
1 |M N |
Since |CM | = 2|M N | and |M A| = 2 |AB|, it follows that |CA| =
|SM |
|AB| .
Togther with ∠SM N = ∠SM C = ∠CAM = ∠CAB, we get
4SN M ∼ 4BCA. Hence ∠M SN = ∠ABC. Moreover, we have
∠XSM = 180◦ − ∠XAM . In (a), we have seen that ∠XAM =
∠XM C = ∠M BC = ∠ABC, so ∠XSM = 180◦ − ∠ABC = 180◦ −
∠M SN . Therefore S lies on XN .
26
Problems
1. Let a, b, and c be distinct positive integers, and suppose that p = ab+bc+ca

is a prime number.
(a) Show that a2 , b2 , and c2 give distinct remainders after division by p.
(b) Show that a3 , b3 , and c3 give distinct remainders after division by p.
2. The incircle of a non-isosceles triangle 4ABC has centre I and is tangent to

BC and CA in D and E, respectively. Let H be the orthocentre of 4ABI,
let K be the intersection of AI and BH and let L be the intersection of
BI and AH. Show that the circumcircles of 4DKH and 4ELH intersect
on the incircle of 4ABC.
3. Let k > 2 be an integer. A positive integer ` is said to be k-pable if the

numbers 1, 3, 5, . . . , 2k − 1 can be partitioned into two subsets A and B in
such a way that the sum of the elements of A is exactly ` times as large as
the sum of the elements of B.
Show that the smallest k-pable integer is coprime to k.
4. Find all functions f : R → R such that

(y + 1)f (x) + f xf (y) + f (x + y) = y
for all x, y ∈ R.
27
Solutions
1. (a) Suppose for a contradiction that two of a2 , b2 , and c2 have the same
remainder after division by p, say a2 ≡ b2 mod p. Then p | a2 − b2 =
(a − b)(a + b), so p | a − b or p | a + b. In the latter case, we have
p ≤ a + b ≤ c(a + b) < ab + bc + ca = p, which is a contradiction. In
the former case, we see that by a 6= b, we have p ≤ |a − b| ≤ a + b < p,
which is a contradiction.
(b) Suppose for a contradiction that two of a3 , b3 , and c3 have the same
remainder after division by p, say a3 ≡ b3 mod p. Then p | a3 − b3 =
(a−b)(a2 +ab+b2 ), so p | a−b or p | a2 +ab+b2 . The former case leads
to a contradiction, as we have seen in (a). Consider the remaining case:
p | a2 +ab+b2 . Then p | a2 +ab+b2 +(ab+bc+ca) = (a+b)(a+b+c), so
p | a+b or p | a+b+c. However, note that a+b < a+b+c < ab+bc+ca,
since a, b, and c cannot all be equal to 1. Therefore p is neither a
divisor of a + b, nor a + b + c, which is a contradiction.

2. Consider the configuration in the fig-

ure; the solution is similar for the H
other configurations.
C
We have ∠IDB = 90◦ = ∠IKB, K
D
so BKDI is a cyclic quadrilateral. S
Moreover, we have ∠ALB = 90◦ = E L I
∠AKB, so BKLA is also a cyclic
quadrilateral. So A F B
∠BKD = 180◦ − ∠BID = 180◦ − (90◦ − 12 ∠ABC)

= 180◦ − ∠BAL = ∠BKL.
Hence K, D, and L are collinear. Analogously, K, E, and L are collinear,

from which we deduce that all four of K, D, E, and L are collinear.
Let S be the second intersection point of the circumcircles of 4DKH and
4ELH. Then we have
∠DSE = 360◦ − ∠DSH − ∠HSE = ∠DKH + 180◦ − ∠HLE

= ∠LKH + ∠HLK = 180◦ − ∠KHL.
28
Since HLIK is also a cyclic quadrilateral (as it has two opposite right
angles) we have
180◦ − ∠KHL = ∠KIL = ∠AIB = 180◦ − ∠IBA − ∠IAB

= 180◦ − 12 ∠CBA − 12 ∠CAB.
So ∠DSE = 180◦ − 12 ∠CBA − 12 ∠CAB. Let F be the point at which

the incircle is tangent to AB. Then AF IE and BF ID are both cyclic
quadrilaterals (in both cases because of two opposite right angles). Therefore
∠DF E = ∠DF I + ∠IF E = ∠DBI + ∠IAE = 21 ∠CBA + 12 ∠CAB.
We deduce that ∠DF E + ∠DSE = 180◦ , and it follows that S lies on the
circumcircle of 4DEF , which is the incircle of triangle 4ABC.
3. We show that if p is the smallest prime divisor of k, then p − 1 is the

smallest k-pable integer. The claim follows from this as gcd(p − 1, k) = 1.
Note that 1 + 3 + 5 + · · · + (2k − 1) = k 2 . If ` is a k-pable integer, and s is
the corresponding sum of the elements of B, then the sum of the elements
of A is `s. So the sum 1 + 3 + 5 + · · · + (2k − 1) is also equal to (` + 1)s.
Hence (` + 1)s = k 2 .
Since ` + 1 ≥ 2 we have ` + 1 ≥ p, with p the smallest prime divisor of k.
Hence ` ≥ p − 1. We now show that ` = p − 1 is k-pable, from which it will
immediately follow that that is the smallest k-pable integer.
First suppose that k is even, so that p = 2. Then we need to show that
we can partition the set {1, 3, 5, . . . , 2k − 1} into two sets with equal sum
of elements. We do this by induction on k. If k = 4 and k = 6, we have
the partitions {1, 7}, {3, 5} and {1, 3, 5, 9}, {7, 11}. If we can partition the
set {1, 3, . . . , 2k − 1} in such a way, then we can also partition the set
{1, 3, . . . , 2(k + 4) − 1} by taking a partition for the set {1, 3, . . . , 2k − 1},
and add the elements 2k + 1, 2k + 7 to one of the subsets, and 2k + 3, 2k + 5
to the other. This completes the induction.
Now suppose that k is odd. Write k = pm. Then it suffices to find a subset
B of {1, 3, 5, . . . , 2k − 1} of which the sum of elements is pm2 , since then
the sum of the elements of A = {1, 3, 5, . . . , 2k − 1} \ B is k 2 − pm2 =
p2 m2 − pm2 = (p − 1)pm2 , which is exactly p − 1 times as large as the
sum of the elements of B. Consider the subset B = {p, 3p, . . . , (2m − 1)p}.
Then the sum of elements of B is
p + 3p + · · · + (2m − 1)p = p 1 + 3 + · · · + (2m − 1) = pm2 ,

as desired.
29
4. Substituting x = 0 gives (y + 1)f (0) + f (f (y)) = y, so f (f (y)) = y · (1 −
f (0)) − f (0). If f (0) 6= 1, the right hand side is a bijective function of y,
hence so is the left hand side. So in this case, f is bijective.
We will next show that in case f (0) = 1, the function f is bijective as well.
So suppose that f (0) = 1. Then f (f (y)) = −1 for all y ∈ R. Substituting
y = 0 gives f (x) + f (x + f (x)) = 0, so f (x + f (x)) = −f (x). Substituting
x = f (z) and y = z and replacing each occurrence of f (f (z)) by −1, we
see that
(z + 1) · −1 + f f (z)f (z) + f (z + f (z)) = z,
so using that f (z + f (z)) = −f (z), we see that
f f (z)2 − f (z) = 2z + 1.

Hence f is surjective. If there exist a and b with f (a) = f (b), then

substituting z = a and z = b in the last equation above gives equal left
hand sides, therefore 2a + 1 = 2b + 1, from which it follows that a = b.
Hence f is injective, and therefore bijective.
So from now on, we may assume that f is bijective, dropping the assumption
that f (0) = 1. Note that f (f (y)) = y · (1 − f (0)) − f (0) so substituting
y = −1 gives f (f (−1)) = −1. Substituting y = −1 in the original equation
gives
f xf (−1) + f (x − 1) = −1 = f (f (−1)).
As f is injective, it follows that xf (−1) + f (x − 1) = f (−1), so f (x − 1) =
f (−1) · (1 − x). Substituting x = z + 1 then gives f (z) = −f (−1)z for all
z ∈ R. So the function f must be of the form f (x) = cx for all x ∈ R, with
c ∈ R a constant. Let us check functions of this form.
Note that

(y + 1)f (x) + f xf (y) + f (x + y) = (y + 1)cx + c(xcy + cx + cy)
= cxy + cx + c2 xy + c2 x + c2 y.
This must equal y for all x, y ∈ R. Substituting y = 0 and x = 1 gives

c + c2 = 0, so either c = 0 or c = −1. Substituting x = 0 and y = 1 gives
c2 = 1, so c = 1 or c = −1. We deduce that c = −1, and we see that the
function given by f (x) = −x for all x ∈ R indeed satisfies the required
equation. So the only solution of the equation is the function given by
f (x) = −x for all x ∈ R.
30
Problems
1. A circle ω with diameter AK is given. The point M lies in the interior of

the circle, but not on AK. The line AM intersects ω in A and Q. The
tangent to ω at Q intersects the line through M perpendicular to AK, at
6 Q.
P . The point L lies on ω, and is such that P L is tangent to ω and L =
Show that K, L, and M are collinear.
2. Let a1 , a2 , . . . , an be a sequence of real numbers such that a1 + · · · + an = 0

and define bi = a1 + · · · + ai for 1 ≤ i ≤ n. Suppose that bi (aj+1 − ai+1 ) ≥ 0
for all 1 ≤ i ≤ j ≤ n − 1. Show that
max |a` | ≥ max |bm |.

1≤`≤n 1≤m≤n
3. Compute the product of all positive integers n for which 3(n!+1) is divisible
by 2n − 5.
4. Let n ≥ 2 be an integer. Find the smallest positive integer m for which

the following holds: given n points in the plane, no three on a line, there
are m lines such that no line passes through any of the given points, and
for all points X 6= Y there is a line with respect to which X and Y lie on
opposite sides.
31
Solutions
1. Let O be the centre of ω, and let V be the intersection of M P and AK.

We first show that ∠P V L = ∠P OL. If V and O coincide, then there
is nothing to prove. If they don’t, then ∠OV P = 90◦ = ∠OLP , so
OV P L or V OP L is a cyclic quadrilateral. (In fact, Q also lies on the
corresponding circumcircle.) It follows that ∠P V L = ∠P OL. So in all
cases, ∠M V L = ∠P V L = ∠P OL. Since P L and P Q are tangent to ω,
we have 4OQP ∼ = 4OLP , so ∠P OL = 12 ∠QOL. By the interior angle
theorem applied to ω this angle is equal to ∠QAL. Therefore
∠M V L = ∠P OL = ∠QAL = ∠M AL,
and it follows that M V AL is a cyclic quadrilateral. So ∠ALM = 180◦ −

∠AV M = 90◦ . Moreover, by Thales’s theorem we have ∠ALK = 90◦ , so
∠ALM = ∠ALK, in other words, L, M , and K are collinear.
2. We have bn = 0. Suppose that there exists an i ≤ n−1 such that bi > 0 and
ai+1 ≥ 0. Then from bi (aj+1 − ai+1 ) ≥ 0 it follows that aj+1 ≥ ai+1 ≥ 0,
for all i ≤ j ≤ n − 1. So bn = bi + ai+1 + ai+2 + · · · + an ≥ bi > 0, which
is a contradiction. Therefore bi > 0 implies that ai+1 < 0. Analogously, if
bi < 0 then ai+1 > 0.
Let k now be such that |bk | = max1≤m≤n |bm |. We may assume without
loss of generality that bk > 0 (by multiplying all ai by −1 if necessary).
If k = 1, then bk = a1 , so |bk | = |a1 | ≤ max1≤`≤n |a` |, in which case we
are done. Now suppose k > 1. If bk−1 > 0, then by the above, we have
ak < 0. On the other hand, ak = bk − bk−1 ≥ 0 since bk was maximal,
which is a contradiction. Hence ak = bk − bk−1 = bk + |bk−1 | ≥ bk , so
|bk | ≤ |ak | ≤ max1≤`≤n |a` |, which is what we needed to prove.
3. The integers n = 1, 2, 3, 4 satisfy the given condition, since in these cases,

we have that 2n − 5 is equal to −3, −1, 1, 3, respectively, which divides
3(n! + 1). So suppose from now on that n > 4 so that 2n − 5 > 3.
We first show that if n satisfies the given condition, then 2n − 5 must be a
prime. We distinguish two cases. First suppose that 2n − 5 is not prime,
and has a prime divisor p > 3. Since p 6= 2n − 5 and since 2n − 5 is odd, it
follows that p ≤ 2n−5
3 < n. So p | n!, but then p - n! + 1, so p - 3(n! + 1)
since p 6= 3. Therefore 2n − 5 - 3(n! + 1), so n does not satisfy the given
condition. Now suppose that 2n − 5 is not prime, and only has 3 as prime
divisor; so 2n − 5 is a power of 3 which is greater than 3. However, for
32
n > 4 we have 3 - n! + 1, so 3(n! + 1) is divisible by at most one factor 3.
So in this case, n cannot satisfy the given condition either.
So for n > 4 satisfying the given condition, we must have that 2n − 5
is a prime number greater than 3. Write q = 2n − 5. Then q | n! + 1,
or equivalently, n! ≡ −1 mod q. Moreover, Wilson’s theorem states that
(q − 1)! ≡ −1 mod q. Therefore
−1 ≡ (2n − 6)! ≡ (2n − 6)(2n − 7) · · · (n + 1) · n!

≡ (−1) · (−2) · · · · · (−n + 6) · n! ≡ (−1)n−6 · (n − 6)! · n!
≡ (−1)n · (n − 6)! · −1 mod q.
So (n − 6)! ≡ (−1)n mod q. Since n! ≡ −1 mod q, we see that n · (n −

1) · · · · · (n − 5) ≡ (−1)n−1 mod q. Multiplying this by 26 gives
2n · (2n − 2) · (2n − 4) · (2n − 6) · (2n − 8) · (2n − 10) ≡ (−1)n−1 · 64 mod q.
Modulo q = 2n − 5, the left hand side is 5 · 3 · 1 · −1 · −3 · −5 = −225.

Suppose n is odd. Then −225 ≡ 64 mod q, so q | −225 − 64 = −289 =
−172 . Hence q = 17 and therefore n = 17+5
2 = 11. Now suppose n is even.
Then −225 ≡ −64 mod q, so q | −225 + 64 = −161 = −7 · 23. Hence q = 7
or q = 23, which gives n = 6 or n = 14, respectively.
We check these three possibilities. For n = 11 and 2n − 5 = 17, we have
11! = 1 · (2 · 9) · (3 · 6) · (5 · 7) · 4 · 8 · 10 · 11
≡ 4 · 8 · 10 · 11 = 88 · 40 ≡ 3 · 6 ≡ 1 mod 17,
so n = 11 doesn’t satisfy the condition. For n = 14 and 2n − 5 = 23,
14! = 1 · (2 · 12) · (3 · 8) · (4 · 6) · (5 · 14) · (7 · 10) · 9 · 11 · 13

≡ 9 · 11 · 13 = 117 · 11 ≡ 2 · 11 ≡ −1 mod 23,
so n = 14 does satisfy the condition. Finally, for n = 6 and 2n − 5 = 7

we have 6! ≡ −1 mod 7 by Wilson’s theorem, so n = 6 also satisfies the
condition.
So the integers n that satisfy the condition are 1, 2, 3, 4, 6, and 14. Their
product is 2016.
n n+1
4. We prove the smallest such m equals 2 if n is even and 2 if n is odd.
Choose the n points in such a way that they all lie on a circle, and denote
them by P1 , P2 , . . . , Pn , in the order in which they lie on the circle. The
n segments P1 P2 , P2 P3 , . . . , Pn P1 must all be intersected by at least one
33
line. As every line intersects the circle at most twice, every line intersects
at most two of these segments. This shows that the number of lines is at
least n2 if n is even and at least n+1
2 if n is odd.
We now show this number of lines suffices. First we show that given four
distinct points P1 , P2 , and Q1 , Q2 , there always exists a line with respect
to which P1 and P2 lie on opposite sides, as well as Q1 and Q2 . Take the
line through the midpoints of the segments P1 P2 and Q1 Q2 . Suppose this
line passes through P1 . Then this line passes through P2 as well, and as no
three of the given points lie on a line, this line does not pass through Q1 ,
nor through Q2 . Hence one can rotate this line a bit around the midpoint of
P1 P2 in such a way that it still intersects the segment Q1 Q2 , and such that
neither P1 nor P2 lie on it. The three remaining cases are done analogously.
Now if this line still passes through one of the other given points, one can
move this line around a tiny bit for it to no longer pass through any of the
given points; this is always possible as there are only finitely many points.
Now P1 and P2 , as well as Q1 and Q2 , lie on opposite sides of this line.
Assume that n is even; if n is odd, we add an arbitrary point that is not on
any line through any pair of given points. The lines that we are about to
construct, still give a correct example if we remove the extra point. So we
need to construct n2 lines. Fix an arbitrary line such that no line through
two of the given points is parallel to it; this is possible as there are only
finitely many pairs of points, and infinitely many directions to choose from.
Translate this line across the plane. This line will meet the given points
one by one. So at some point, there are no points on the line, and on both
sides of the line there are n2 points. This will be our first line.
The plane is now split into two regions, say the left region, and the right
region. We now add lines in such a way that every line creates a new region
in the left region, as well as the right one. (We will only consider regions
containing at least one of the given points.) To this end, we pick two points
in the left region that aren’t separated yet by a line, and two points in
the right region that aren’t separated yet by a line, and pick a line that
separates the two points in the left region, as well as the points in the right
region; we have seen before that this is possible. This creates a new region
in both the left region and the right region, as the line separates pairs of
points in both regions that weren’t separated before. If at some point,
either the left or the right region no longer contains pairs of points that
are not separated, then we will only use the points in the remaining region.
After adding n2 − 1 lines, we will have, in both the left and the right region,
at least n2 regions, each containing at least one given point; so each region
has to contain exactly one given point. In total we have used n2 lines. Hence,
it is always possible to satisfy the given condition with n2 lines.
34
Problems
Part 1
1. The sum of the digits of a number is obtained by adding the digits of this
number. For example, the sum of the digits of 76 equals 7 + 6 = 13. The
sum of the digits of the double of 76 is 1 + 5 + 2 = 8.
How many numbers consisting of two digits are there for which the sum of
the digits equals the sum of the digits of the double of the number?
Attention: the first digit cannot be a zero. Thus, the number 09, for example,
is ruled out.
A) 0 B) 8 C) 9 D) 10 E) 11
2. Birgit, Dion, Huub, Jaap, Peter, and Thijs are

Peter Jaap
standing in this order along a circle. They are
playing a ball game, in which, at every turn,
they pass the ball to a person directly next
to them or exactly opposite to them. Initially, Thijs Huub
Birgit has the ball. After five passes, everybody
has had the ball exactly once and the game
ends.
Who can have the ball at the end of the game? Birgit Dion
A) Only Dion and Thijs D) Only Dion, Huub, Peter, and Thijs
B) Only Dion, Jaap, and Thijs E) Everybody except Birgit
C) Only Huub, Jaap, and Peter
3. Four distinct straight lines are drawn on a (infin-

itely big) piece of paper. The number of points
in which two or more lines intersect is counted.
In the figure on the right, you see an example in
which four lines intersect each other in 6 points.
This number of intersection points does not al-
ways have to be 6.
What number of intersection points is not pos-
sible?
A) 1 B) 2 C) 3 D) 4 E) 5
35
4. A baker’s helper is filling cream puffs for one and a half hours. He is not
in a hurry and he is filling two cream puffs every minute. At some point,
the baker comes in and supervises the helper for a while. This motivates
the helper to work a bit faster: during this time he is filling three cream
puffs per minute. As soon as the baker has left, the helper reverts to filling
the cream puffs at the initial, slow, pace. Afterwards, it turns out that
during any continuous one hour period within these one and a half hours,
the helper has filled exactly 140 cream puffs.
How many cream puffs are filled by the helper during the full one and a
half hours?
A) 180 B) 200 C) 210 D) 230 E) 270
5. Start with the number 60. Then, keep repeating the following two steps:
(1) Throw a die and look at the number that comes up.
(2) If your number is divisible by the number on the die, then you divide
your number by the number on the die. If not, then you multiply your
number by the number on the die.
In this way, you obtain a sequence of numbers. If your first three rolls are
5, 6, and 3, consecutively, then the first four numbers in your sequence are
60, 60/5 = 12, 12/6 = 2, and 2 × 3 = 6.
What is the greatest number you can obtain in this way?
A) 60 B) 120 C) 240
D) 360 E) You can obtain arbitrarily large numbers
6. Harry and Hermione are trapped in a room in which 6 bottles are put next
to each other. From left to right, the bottles are numbered 1 to 6. One
of the bottles contains a potion that helps them to escape. On a piece of
paper there are four clues to help them:
• 3 of the bottles contain poison, 2 of the bottles contain sleeping
draught; the remaining bottle contains the potion to escape.
• Immediately left to the sleeping draught there is poison.
• The smallest bottle contains poison.
• The second bottle from the left and the second bottle from the right
have the same content.
36
Hermione now has enough information to identify the bottle which contains
the potion to escape. Which bottle is the smallest bottle?
A) Bottle 1 B) Bottle 2 or 5 C) Bottle 3
D) Botttle 4 E) Bottle 6
7. A net of a cube is made by cutting a cube along some

of the ribs until you can flatten it out (after cutting,
you must still have one connected whole). By doing
this in different ways, you can create different nets.
The figure on the right consists of 8 squares. The 6
grey squares together form a net.
In how many other ways can you choose 6 squares in the figure that together
form a net?
A) 3 B) 4 C) 5 D) 7 E) 9
8. Between the digits of the number 2016, we put one or more symbols from
×, +, and − (you are allowed to use a symbol multiple times). In this
way, we can create different numbers, such as 20 + 1 × 6, which is 26, and
201 × 6, which is 1206.
How many of the numbers from 1 to 10 are, just like 26 and 1026, the result
of such a calculation? (Attention: you cannot put a − before the 2!)
A) 6 B) 7 C) 8 D) 9 E) 10
37
Part 2
1. All vehicle registration plate numbers in the country Wissewis consist of

three two-digit numbers. A plate number is considered beautiful if it has
the following two properties:
• it consists of six distinct digits;
• the first number is smaller than the second number and the second
number is smaller than the third number.
An example of a beautiful plate number is 03-29-64.
How many beautiful plate numbers are there that have 61 as the first
number?
2. Alice, Bob, Carla, Daan, and Eva are standing in this order along a circle
(Bob is standing to the left of Alice). Each of them has a number of sweets,
they have 100 sweets in total. All at the same time, they give part of their
sweets to their left neighbour: Alice gives away 13 of her sweets, Bob 14 ,
Carla 15 , Daan 16 , and Eva 17 . After this, everybody has the same number
of sweets as before.
How many sweets does Eva have? E
D C
3. In the figure on the right, rectangles ABCD and
F
BDEF are shown. The length of AB is 8 and the
length of BC is 5.
What is the area of pentagon ABF ED?
A B
4. In this problem we consider three-digit numbers of which no digit is a zero.

Such a number is called a lucky number if:
• the number is divisible by 4, and
• if you change the order of the three digits, you will still always get a
number divisible by 4.
For example, the number 132 is not a lucky number, because 132 is divisible
by 4, but 231 is not.
How many lucky numbers are there?
5. How many times a day (which is 24 hours) are the small hand and the big
hand of the clock perpendicular?
38
6. Janneke, Karin, Lies, Marieke, and Nadine participated in a running race.
They all finished at distinct times except for two of them; they finished at
the same time. Moreover, we know that:
• at least three runners finished before Janneke;

• after Karin finished but before Lies finished, exactly two others crossed
the finish line;
• Marieke was not the first to finish;
• shortly after Nadine finished, Janneke crossed the finish line; nobody
else was in-between.
Which two runners finished at the same time?
7. For all positive integers a and b we make the number a ♥ b. The following
rules hold:
• rule 1: 1 ♥ 1 = 1;
• rule 2: a ♥ b = b ♥ a;
• rule 3: a ♥(b + c) = a + (a ♥ b) + (a ♥ c).
From these rules it follows, for example, that
2 ♥ 1 = 1 ♥ 2 = 1 ♥(1 + 1) = 1 + 1 ♥ 1 + 1 ♥ 1 = 1 + 1 + 1 = 3.
Calculate 20 ♥ 16.
8. We create a sequence of numbers. To get the next number in the sequence,

we repeatedly do the following:
• if the previous number is odd: multiply this number by itself and add
3;
• if the previous number is even: divide this number by 2.
For example, when we start with 5, we obtain 5 × 5 + 3 = 28 as second

number and 282 = 14 as third number in the sequence. As starting number
we are allowed to choose any of the numbers from 1 to 1000.
For how many of these starting numbers will the tenth number in the
sequence be smaller than 10?
39
Solutions
Part 1
1. D) 10
2. B) Only Dion, Jaap, and Thijs
3. B) 2
4. B) 200
5. E) You can obtain arbitrarily large numbers
6. E) Bottle 6
7. C) 5
8. B) 7
Part 2
1. 90 5. 44
2. 28 6. Janneke and Lies
3. 60 7. 924
4. 8 8. 17
40
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

OmslagNWO2017-v3.indd 2 10-06-17 17:15

AA01.pdf 22-05-2008 14:01:01

and the team selection for IMO 2018 Romania
MY

CY
We eat problems
CMY
Junior Mathematical Olympiad, September 2017

for breakfast.

Olympiad 2017
International
OmslagNWO2018-v2.indd 1 17-06-18 13:42
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
36 Junior Mathematical Olympiad, September 2017
OmslagNWO2018-v2.indd 2 17-06-18 13:42

Introduction

represent the Netherlands at the EGMO in Florence, Italy, from 9 until 15
April. The team brought home a silver medal, two bronze medals, and a
honourable mention; a very nice achievement. For more information about
the EGMO (including the 2018 paper), see www.egmo.org.
held in Mersch, Luxemburg, from 27 until 29 April. The Dutch team
received a gold medal, four silver medals and three bronze medals, and
managed to get the highest total score of the Benelux countries, beaten only
by guest country France. For more information about the BxMO (including
the 2018 paper), see www.bxmo.org.
1
Cluj-Napoca, from 30 June until 6 July.
the solutions.
Dutch delegation
The Dutch team for IMO 2018 in Romania consists of
• Nils van de Berg (18 years old) • Jovan Gerbscheid (15 years old)
– bronze medal at BxMO 2017 – silver medal at BxMO 2018
– gold medal at BxMO 2018 • Jippe Hoogeveen (15 years old)
– hon. mention at IMO 2017
• Szabi Buzogany (18 years old)
– silver medal at BxMO 2018 – bronze medal at BxMO 2017
• Thomas Chen (17 years old) – observer C at IMO 2016
– gold medal at BxMO 2017 – silver medal at IMO 2017

• Richard Wols (15 years old)

• Jetze Zoethout (observer B), Utrecht University
2
Problems
A-problems
1. In a certain year, August has only 4 Mondays and 4 Fridays.

Which day of the week was 31 August that year?
A) Tuesday B) Wednesday C) Thursday
D) Saturday E) Sunday
2. We consider dotted hexagons with 1, 2, 3, . . . dots on each side,

see also the picture. The number of dots in such a hexagon is
called a hexagonal number. The first hexagonal number is 1,
the second is 7, and the third is 19.
Which of the following numbers is also a hexagonal number?
A) 81 B) 128 C) 144 D) 169 E) 187
3. Five suspects are arrested in a criminal investigation. Each of them makes

one statement:
Eva: “We are all innocent.”
Fatima: “Exactly one of us is innocent.”
Kees: “Exactly one of us is guilty.”
Manon: “At least two of us are innocent.”
Mustafa: “At least two of us are guilty.”
It turns out that those who are guilty lied, while those who are innocent
told the truth. How many of the five suspects are guilty?
A) 1 B) 2 C) 3 D) 4 E) 5
4. Two regular hexagons share a side and are situated

inside a parallelogram as indicated in the figure. The
area of the parallelogram equals 1.
What is the area of the two grey areas combined?
1 2 5 3 1
A) 3 B) 5 C) 12 D) 7 E) 2
3
5. In the expression below, the ten dots are replaced by ten distinct digits
(0 to 9) in such a way that none of the resulting two-digit numbers starts
with 0:
·· + ·· + ··
·· − ··
What is the largest possible outcome we can obtain?
255 219
A) 7 B) 2 C) 116 D) 222 E) 255
6. A 100×100 table is filled with numbers. The bottom .. .. .. .

left cell contains the number 0. For every other cell . . . ..
V , we consider a route from the bottom left cell to V , 3 7 15 · · ·
where in each step we go one cell to the right or one 1 3 7 ···
cell up (not diagonally). If we take the number of
0 1 3 ···
steps and add the numbers from the cells along the
route, we obtain the number in cell V . In the figure, you see a partially filled
table. The number 15, for example, is obtained as 4 + (0 + 1 + 3 + 7) = 15.
What is the last digit of the number in the upper right cell of the 100×100
table?
A) 1 B) 3 C) 5 D) 7 E) 9
7. Rectangle ABCD is divided into squares. The length of side AB is 16.
D C
A B
What is the length of side AD?

27 29
A) 13 B) 2 C) 14 D) 2 E) 15
8. Joep assigns the numbers 1 to 8 to the vertices of a cube (each vertex

receiving a number different from the other vertices). For each face of the
cube he adds the four numbers assigned to the vertices of that face and
writes the resulting number on the face. Then, he cuts the cube open along
some of the sides and flattens it out to obtain one of the five figures given
below. Only one of these figures could represent Joep’s cube.
4
17 18 17 27 18
16 18 20 17 18 18 18 20 20 26 16 16 12 20 24 18 18 18
19 16 10 19 9 18 18
Which figure could represent Joep’s cube?

A) the first B) the second C) the third
D) the fourth E) the fifth
B-problems
1. Isaac writes down a three digit number. None of its digits is a zero. Isaac
gives his sheet with the number to Dilara, and below Isaac’s number she
writes down all three digit numbers that one can obtain by putting the
digits of Isaac’s number in a different order. Then she adds up all numbers
on the sheet. The outcome is 1221.
What is the greatest number that Isaac could have written down?
2. There are two triples (a, b, c) of positive integers that satisfy the equations
ab + c = 34,
a + bc = 29.
Which two triples are these?
B
3. Triangle ABC is an isosceles right angled triangle whose

right angle is at C, with |AC| = |BC| = 12. Point M
M
is the midpoint of side AB. A point D lies on side AC.
Finally, point E is the intersection point of line segments
CM and BD, see the figure. E
If |CD| = 3, what is the area of quadrilateral AM ED?
C D A
4. At a quiz you have to answer 10 questions. Each question is either difficult

or easy. For a difficult question 5 points are being awarded for a correct
answer and −1 point for an incorrect answer; for an easy question 3 points
are being awarded for a correct answer and −1 point for an incorrect answer.
Moreover, if you answer a question correctly, then the next question will be
a difficult one; if you answer a question incorrectly, then the next question
will be an easy one. You start with a difficult question.
How many distinct final scores are possible after 10 questions?
5
Solutions
A-problems
1. C) Thursday 5. D) 222
2. D) 169 6. B) 3
29
3. C) 3 7. D) 2
1
4. E) 2 8. E) the fifth
B-problems
1. 911
2. (13, 2, 8) and (5, 6, 4)
162
3. 5
4. 27
6
Problems
B-problems
1. A finite sequence of consecutive positive integers is called balanced if it

contains equally many multiples of three and multiples of five. An example
of a sequence of length 7 that is not balanced is 30, 31, 32, 33, 34, 35, 36,
because this sequence contains 3 multiples of three (namely 30, 33, and 36)
and just 2 multiples of five (namely 30 and 35).
What is the maximal length of a balanced sequence of consecutive positive
integers?
C
2. The area of a given triangle ABC equals 40. F

Point D on side AB satisfies |BD| = 3 · |AD|. E
Point E on side BC satisfies |CE | = 3 · |BE |.
B
Point F on side CA satisfies |AF | = 3 · |CF |.
Determine the area of triangle DEF .
A D
3. In math class, a student has written down a sequence of 16 numbers on

the blackboard. Below each number, a second student writes down how
many times that number occurs in the sequence. This results in a second
sequence of 16 numbers. Below each number of the second sequence, a third
student writes down how many times that number occurs in the second
sequence. This results in a third sequence of numbers. In the same way,
a fourth, fifth, sixth, and seventh student each construct a sequence from
the previous one. Afterwards, it turns out that the first six sequences are
all different. The seventh sequence, however, turns out to be equal to the
sixth sequence.
Give one sequence that could have been the sequence written down by the
first student.
7
4. A parallelogram ABCD is intersected by a line D C
m. From each of the four vertices A, B, C, and ?
D we draw a perpendicular to m. The four 25 Q
feet are P , Q, R, and S, respectively. Point S R
m
A 7
is also the intersection of line m and AB . The
lengths of line segments AP , BQ, and DS are 6 S B
P
6, 7, and 25, respectively.
What is the length of CR?
Be careful: the figure is not drawn to scale.
5. Simon has 2017 blue blocks that are numbered from 1 up to and including
2017. He also has 2017 yellow blocks that are numbered from 1 up to and
including 2017. Simon wants to arrange his 4034 blocks in a row, in such a
way that, for every k = 1, 2, . . . , 2017, the following conditions are met:
• to the left of blue block number k there are k or more yellow blocks;
• to the right of yellow block number k there are k or fewer blue blocks.
Determine all possible numbers for the 1000th block from the left in the
row.
1. You have 1000 tiles of each of the following five types:
1 0 1 1 1 0 1 1 0 0 0 1 0
A B C D E
0 1 0 1 1 1 0 1 0 1 0 0 1
You want to form a row of tiles such that the same sequence of zeroes and
ones is formed on the top and the bottom. We will call this a matching
combination. Consider, for example, the row ‘DDC’ consisting of three tiles
of types D, D, and C, in that order. The top sequence is 1001001, while
the bottom sequence is 010010101. Since the two sequences are not the
same, the row of tiles is not a matching combination.
(a) Construct a matching combination using only tiles of type A, B, and
C.
8
(b) Show that no matching combination using only tiles of types B, C,
and D exists.
(c) Does a matching combination using only tiles of types B, C, D, and E
exist?
If so, give an example. If not, prove that such a combination does not
exist.
2. A multi-square is a number obtained by concatenating two or more square

two-digit numbers. (A two-digit number is not allowed to start with digit
0). For example, since 16 and 25 are squares, 1625 is a multi-square.
(a) Determine all four-digit multi-squares whose first and last digit are
equal.
(b) Determine all six-digit multi-squares that are themselves squares.
9
Solutions
B-problems
1. 11
35
2. 2
3. Multiple solutions, e.g.: 0, 1, 2, 2, 4, 4, 4, 4, 8, 8, 8, 8, 8, 8, 8, 8
4. 12
5. 500
C-problems
C1. (a) A matching combination is ‘BABC’ having a top row and bottom row
equal to 1101011101.
All other solutions are of the form ‘BABC BABC . . . BABC’.
(b) Suppose that we have a matching combination using only tiles of types
B, C, and D.
The top row of these tiles always starts with a 1. In a matching
combination, the bottom row must therefore start with a 1 as well.
This rules out the first tile being of type D. Type C is ruled out as
well, since otherwise the second tile must have a top row starting with
0. The first tile of a matching combination must therefore be of type
B.
The next tile must have a bottom row starting with 0. Hence, it must
be of type D. After that, we again require a tile with a bottom row
starting with a 0. That is, we again require a tile of type D. This
continues indefinitely. After depleting our supply of 1000 tiles of type
D, we still need another tile having a bottom row starting with a 0. It
follows that we cannot complete our row to a matching combination:
a contradiction.
We conclude that there is no matching combination using only tiles of
types B, C, and D.
The proof of part (c) will give an alternative proof for part (b).
(c) First, consider the number of 1’s in the top row and bottom row of
each type of tile. Tiles of type B, D, and E have equal numbers of
10
1’s in the top and bottom row. Tiles of type C have more 1’s in the
bottom row than in the top row. Since a matching combination must
have equal numbers of 1’s in both rows, it cannot contain tiles of type
C.
A matching combination can therefore only contain tiles of types B,
D, and E. Tiles of type D have an equal number of digits in the top
row and bottom row (three digits). In contrast, tiles of type B and
E have more digits in their top row than in their bottom row. Since
any matching combination has equal numbers of digits in both rows,
it can contain tiles of neither type B nor type E.
A matching combination can therefore only contain tiles of type D.
But that is also not possible since tiles of type D have a top row
starting with 1 and a bottom row starting with 0. We conclude that
no matching combination exists containing only tiles of types B, C, D,
and E.
C2. (a) No two-digit square ends in a 2, 3, 7, or 8. Also, no two-digit square

starts with a 5 or 9. Hence, we only need to consider multi-squares
with first and last digit equal to 1, 4, or 6.
In the case that the first and last digit are 1, the first square must be 16
and the second square must be 81. This yields the multi-square 1681.
In the case that the first and last digit equal 4, the first square must
be 49 and the second square must be 64. This yields the multi-square
4964. In the case that the first and last digit equal 6, the first square
must be 64 and the second square must be 16 or 36. This yields two
multi-squares: 6416 and 6436.
In total, there are four multi-squares having the same first and last
digit: 1681, 4964, 6416, and 6436.
(b) Let K be a six-digit multi-square, say K = abcdef , where a, b, c, d,
e, and f are the six digits of K. The fact that K is a multi-square
means that ab, cd , and ef are two-digit squares, hence equal to 16,
25, 36, 49, 64, or 81. We want K to be a square, say K = n2 . To
determine all solutions, we consider the different cases for ab.
ab = 16 Since K > 160000, we must have n > 400. Write n =
400 + x, where x is a positive integer. We have K = (400 + x)2 =
160000+800x+x2 . Since K < 170000, it follows that 800x < 10000
and therefore x ≤ 12.
Observe that the last two digits ef of K = 160000 + 800x + x2
are equal to the last two digits of x2 . Since the last two digits
of 102 = 100, 112 = 121, and 122 = 144 do not form a square,
11
the candidates x = 10, 11, 12 are ruled out. Since 12 = 1, 22 = 4,
and 32 = 9 have only one digit, also the candidates x = 1, 2, 3 are
ruled out.
We consider the remaining candidates x = 4, 5, 6, 7, 8, 9. In these
cases, x2 is a two-digit number, which implies that cd = 8 · x. This
is a square only in the case x = 8. Hence, we find one solution:
4082 = 166464.
ab = 25 Similarly to the previous case, we can write n = 500 + x,
where x is a positive integer. We have K = (500 + x)2 = 250000 +
1000x + x2 . Since K < 260000, it follows that 1000x < 10000 and
hence x ≤ 9.
We observe that for every possible choice of x, the digit d of K
will be equal to 0. Hence, there are no solutions.
ab = 36 We write n = 600 + x, where x is a positive integer. We
have K = (600 + x)2 = 360000 + 1200x + x2 . Since K < 370000,
it follows that 1200x < 10000 and hence x ≤ 8. Also, we have
x ≥ 4 as otherwise digit e will be equal to 0.
For the remaining candidates x = 4, 5, 6, 7, 8 we obtain cd = 12 · x.
However, this is not a square for any of the possible choices for x.
Hence, there are no solutions.
have K = (700 + x)2 = 490000 + 1400x + x2 . Since K < 500000, it
follows that 1400x < 10000 and hence x ≤ 7. As in the previous
case, we also have x ≥ 4 because digit e cannot be 0.
For the remaining candidates x = 4, 5, 6, 7 we see that cd = 14 · x
is not a square. Hence, there are no solutions.
have K = (800 + x)2 = 640000 + 1600x + x2 . Since K < 650000,
it follows that 1600x < 10000 and hence x ≤ 6. Again, we also
have x ≥ 4.
Of the remaining candidates x = 4, 5, 6, the number cd = 16 · x is
a square only for x = 4. Thus we find one solution 8042 = 646416.
have K = (900 + x)2 = 810000 + 1800x + x2 . Since K < 820000,
it follows that 1800x < 10000 and hence x ≤ 5. Again we also
have x ≥ 4.
For the remaining candidates x = 4, 5, the number cd = 18 · x is
not a square. Hence, there are no solutions.
We conclude that, in total, there are two six-digit multi-squares that
are themselves a square: 166464 and 646416.
12
Problems
1. We consider positive integers written down in the (usual) decimal system.

Within such an integer, we number the positions of the digits from left to
right, so the leftmost digit (which is never a 0) is at position 1.
An integer is called even-steven if each digit at an even position (if there is
one) is greater than or equal to its neighbouring digits (if these exist).
An integer is called oddball if each digit at an odd position is greater than
or equal to its neighbouring digits (if these exist).
For example, 3122 is oddball but not even-steven, 7 is both even-steven
and oddball, and 123 is neither even-steven nor oddball.
(a) Prove: every oddball integer greater than 9 can be obtained by adding
two oddball integers.
(b) Prove: there exists an oddball integer greater than 9 that cannot be
obtained by adding two even-steven integers.
2. A parallelogram ABCD with |AD| = |BD| has been D C

given. A point E lies on line segment BD in such
a way that |AE| = |DE|. The (extended) line AE
intersects line segment BC in F . Line DF is the F
angle bisector of angle CDE. E
Determine the size of angle ABD.
A B
3. Six teams participate in a hockey tournament. Each team plays exactly

once against each other team. A team is awarded 3 points for each game
they win, 1 point for each draw, and 0 points for each game they lose. After
the tournament, a ranking is made. There are no ties in the list. Moreover,
it turns out that each team (except the very last team) has exactly 2 points
more than the team ranking one place lower.
Prove that the team that finished fourth won exactly two games.
4. If we divide the number 13 by the three numbers 5, 7, and 9, then these

divisions leave remainders: when dividing by 5 the remainder is 3, when
dividing by 7 the remainder is 6, and when dividing by 9 the remainder is 4.
If we add these remainders, we obtain 3 + 6 + 4 = 13, the original number.
13
(a) Let n be a positive integer and let a and b be two positive integers
smaller than n. Prove: if you divide n by a and b, then the sum of
the two remainders never equals n.
(b) Determine all integers n > 229 having the property that if you divide
n by 99, 132, and 229, the sum of the three remainders is n.
5. The eight points below are the vertices and the midpoints of the sides of a
square. We would like to draw a number of circles through the points, in
such a way that each pair of points lie on (at least) one of the circles.
Determine the smallest number of circles needed to do this.
14
Solutions
1. An integer for which the digits (from left to right) are c1 , c2 , . . . , ck will be
denoted by c1 c2 . . . ck .
(a) Let n = c1 c2 . . . ck be an oddball integer greater than 9 (hence k ≥ 2).
We will show that n is indeed the sum of two oddball integers.
If c2 ≥ 1, then we can write n as the sum of the following two oddball
integers: 10 . . . 0 (k − 2 zeros) and c1 (c2 −1)c3 . . . ck .
If c2 = 0 and c1 ≥ 2, then we can write n as the sum of the following
two oddball integers: 10 . . . 0 (k − 1 zeros) and (c1 −1)c2 c3 . . . ck .
If n = 10 . . . 0, then we can write n as the sum of the following two
oddball integers: 1 and 9 . . . 9 (k − 1 nines).
The last case is the case in which c1 = 1, c2 = 0 and not all digits
c3 , . . . , ck are equal to 0. Let ct be a digit unequal to 0, with t ≥ 3 as
small as possible. Hence, n = 10 . . . 0ct . . . ck with ct ≥ 1.
Because n is oddball and ct−1 = 0 < 1 ≤ ct , we find that t must be
odd. We can now write n as the sum of the integers 10 . . . 0 (k − 1
zeros) and m = ct ct+1 . . . ck . Because t is odd, the digits at the odd
positions of m are also at odd positions of n. Therefore, these digits
are greater than or equal to their neighbouring digits (because n is
oddball), which yields that m is oddball.
(b) The integer n = 109 is oddball, but it is not the sum of two even-steven
integers. We will prove this by contradiction. Suppose that n = p + q
for some even-steven integers p and q. We will show that this leads to
a contradiction.
First observe that the integers 100 to 108 are not even-steven. Hence,
both p and q must be smaller than 100, and hence both are also greater
than 9. In other words, p and q have exactly two digits. Suppose
p = ab and q = cd. The equation p + q = 109 now yields b + d = 9
(because b + d < 19) and a + c = 10. Hence, b + d < a + c, which
implies that either b < a or d < c (or both). In the first case p is
not even-steven and in the second case q is not even-steven. This
contradicts the assumption that p and q are even-steven.
We conclude that the oddball integer 109 cannot be written as the
sum of two even-steven integers.
15
2. Since AED is an isosceles triangle, angles D C
∠EDA and ∠DAE are equal. In turn, these
angles are equal to angles ∠EBF and ∠BF E
(alternate interior angles). This implies that
triangle BF E is isosceles as well, with |BE| =
|EF |.
F
Comparing triangles ABE and DF E, we see
that |AE| = |DE| and |BE| = |F E|. Since E
∠BEA and ∠F ED are a pair of opposite
angles, they have the same size. It follows
that triangles ABE and DF E are congruent A B
(SAS).
From the fact that ABE and DF E are congruent it follows that |DF | =
|AB|. Since ABCD is a parallelogram, we also have |AB| = |CD|. It
follows that triangle CDF is isosceles as well (with apex D).
On the one hand, this implies that ∠F CD = ∠DF C. On the other hand,
we know that triangle DBC is isosceles (since |BD| = |AD| = |BC|), which
implies that ∠F CD = ∠CDB = 2 · ∠CDF since DF is the angle bisector
of ∠CDB.
Altogether, we have ∠DF C = ∠F CD = 2 · ∠CDF . Since the angles in any
triangle sum to 180 degrees, we also know that
180◦ = ∠DF C + ∠F CD + ∠CDF = 5 · ∠CDF.
From this, it follows that ∠CDF = 15 · 180◦ = 36◦ , and hence ∠F CD =

2 · ∠CDF = 72◦ . Since triangle DBC is isosceles, also ∠CDB = 72◦ holds.
Using alternating interior angles, we now find that ∠ABD = ∠CDB = 72◦ .
3. Let the scores of the six teams be s, s + 2, s + 4, s + 6, s + 8, and s + 10. Let

T be the total number of awarded points, so that T = 6s + 30. It follows
that the total number of points is a multiple of six.
The number of games played equals 6·52 = 15. Let g be the number of
games that ended in a draw. A game that ends in a draw results in
1 + 1 = 2 awarded points and every other game results in 3 + 0 = 3
awarded points. Therefore, the total number of awarded points equals
T = g · 2 + (15 − g) · 3 = 45 − g.
From T = 45 − g it follows that 30 ≤ T ≤ 45 because the number of draws
satisfies 0 ≤ g ≤ 15. Since T is a multiple of six, this leaves the following
possibilities: T = 30, T = 36, and T = 42.
16
If T = 30, we have g = 45 − 30 = 15. But then all games must have ended
in a draw and all teams must have the same score. Hence, the case T = 30
is ruled out.
If T = 36, then g = 45 − 36 = 9 and s = T −30 6 = 1. The six scores are
therefore 1, 3, 5, 7, 9, 11. The team that scored 1 point must have lost 4
games (and played one draw). The team that scored 3 points must have
lost at least 2 games (at most 3 games were not lost). The team that scored
11 points must have won at least 3 games (otherwise the score is at most
3 + 3 + 1 + 1 + 1 = 9), so apart from the teams with scores 1 and 3 at least
one other team has lost a game. In total, at least 4 + 2 + 1 = 7 games
ended in a loss for some team, contradicting the fact that 15 − 9 = 6 games
did not end in a draw. This rules out the case T = 36.
Finally, we consider that case T = 42. The six scores are 2, 4, 6, 8, 10, 12 and
we have g = 3. Since the total number of points obtained from won games
is a multiple of three, the six teams must have received at least 2, 1, 0, 2, 1, 0
points from draws, respectively. In total, exactly 2·3 = 6 points are awarded
in games that ended in a draw. Hence, since 2 + 1 + 0 + 2 + 1 + 0 = 6,
the six teams have received exactly the mentioned numbers of points from
draws. In particular, the team ending in the fourth place (with 6 points),
was involved in 0 draws and must have won exactly two games.
4. (a) Let r be the remainder upon dividing n by a. We will first prove that
r < n2 . If 2a ≤ n, this follows from the fact that r < a. If 2a > n, we
have r = n − a (since we already knew that a < n), which implies that
r = n − a < n − n2 = n2 .
For the same reasons, the remainder upon dividing n by b is smaller
than n2 .
It follows that the two remainders obtained by dividing n by a and b
add up to a number smaller than n.
(b) Let r, s, and t be the remainders upon dividing n by 99, 132, and 229,
respectively. The number n − t is a multiple of 229 and nonzero since
n > 229 > t. We know that r + s + t = n, and hence n − t = r + s.
We can conclude that r + s is a positive multiple of 229. Since
99 + 132 < 2 · 229, we have r + s < 2 · 229, which implies that we must
have r + s = 229.
Since r ≤ 98 and s ≤ 131, the fact that r + s = 229 implies that
r = 98 and s = 131. Therefore, the number n + 1 is divisible by both
99 and 132, and hence by their least common multiple lcm(99, 132) =
lcm(9 · 11, 3 · 4 · 11) = 4 · 9 · 11 = 396. Also, from n = 229 + t and
t < 229 we deduce that n + 1 ≤ 458. It follows that the only possibility
is n + 1 = 396, hence n = 395.
17
When n = 395 the three remainders are r = 98, s = 131, and t = 166,
and indeed satisfy the equation n = r + s + t.
5. Four of the eight points are coloured black and the other four points are
coloured white in the way indicated in the figure on the left. The circle
through the four black points is denoted C1 and the circle through the four
white points is denoted C2 . If two points lie on a circle C, we say that C
covers that pair of points.
Circle C1 covers all pairs of black points and circle C2 covers all pairs of
white points. It is easy to check that each of the 4 · 4 = 16 pairs consisting
of a white point and a black point is covered by one of the four circles in
the figure on the right. It follows that the six circles form a solution.
We will now prove that five or fewer circles do not suffice. First observe
that any circle passing through more than two black points must be equal
to C1 and that any circle passing through more than two white points must
be equal to C2 . Indeed, a circle is already determined by three points.
A circle passing through 2 or fewer black points covers at most one of the
4·3
2 = 6 pairs of black points. A solution consisting of only five circles must
therefore contain circle C1 (since otherwise at most 5 pairs of black points
are covered). In the same way we see that such a solution must contain
circle C2 .
Each of the remaining three circles in the (hypothetical) solution contains
at most 2 black points and at most 2 white points. Such a circle covers
at most 2 · 2 = 4 pairs consisting of a white and a black point. In total,
the five circles can therefore cover at most 0 + 0 + 3 · 4 = 12 such pairs,
while there are 16 to be covered. The five circles can therefore not form a
correct solution after all. We conclude that the smallest number of circles
in a solution is 6.
18
Problems
1. We have 1000 balls in 40 different colours, 25 balls of each colour. Determine

the smallest n for which the following holds: if you place the 1000 balls in
a circle, in any arbitrary way, then there are always n adjacent balls which
have at least 20 different colours.
2. Let 4ABC be a triangle of which the side lengths are positive integers
which are pairwise coprime. The tangent in A to the circumcircle intersects
line BC in D. Prove that |BD| is not an integer.
3. Let p be a prime number. Prove that it is possible to choose a permutation

a1 , a2 , . . . , ap of 1, 2, . . . , p such that the numbers a1 , a1 a2 , a1 a2 a3 , . . . ,
a1 a2 a3 · · · ap all have different remainder upon division by p.
4. In a non-isosceles triangle 4ABC we have ∠BAC = 60◦ . Let D be the

intersection of the angular bisector of ∠BAC with side BC, O the centre
of the circumcircle of 4ABC and E the intersection of AO and BC. Prove
that ∠AED + ∠ADO = 90◦ .
5. Let n be a positive integer. Determine all positive real numbers x satisfying
22 32 (n + 1)2 n(n + 3)
nx2 + + + ... + = nx + .
x+1 x+2 x+n 2
19
Solutions
1. Consider the circle in which the 25 balls of one colour are all next to each
other. To get at least 20 different colours, you have to take at least 18 of
these groups plus one ball one one side and one ball on the other side of
these groups. In total, you need at least 18 · 25 + 2 = 452 adjacent balls.
Hence, n ≥ 452.
Now we will prove that 452 balls is enough. Consider an arbitrary circle
of balls and all possible sets of consecutive balls having exactly 20 colours.
(There exists at least one such a set: take an arbitrary ball and add balls
on the left one by one, until you have exactly 20 colours.) Take such a set
having a minimum number of balls. Suppose that the first ball is white. If
there is another white ball in the set, then we could have removed the first
ball to get a smaller set with the same number of colours, but less balls in
total. This contradicts the minimality of our set. Therefore, no other ball
is white. In particular, also the last ball of the set is not white; suppose it
is black. In the same way, we find that no other ball in the set is black. So
there are only one white ball, one black ball, and balls in 18 other colours,
at most 25 of each colour. In total, there are at most 18 · 25 + 2 = 452 balls.
Indeed, it is always possible to find a set of 452 consecutive balls in at least
20 different colours. We conclude that the minimum n is 452.
2. There are two different configurations.

Without loss of generality, assume that
B lies between D and C. Let a = |BC|,
b = |CA|, c = |AB|, x = |BD|, and y =
|AD|. Due to the alternate segment theo-
rem we have ∠BAD = ∠ACB = ∠ACD,
hence 4ABD ∼ 4CAD (AA), therefore
|BD| |AB| |AD| x c y
|AD| = |CA| = |CD| , or y = b = a+x .
This yields yc = bx and ac + xc = by,
and hence byc = b2 x and ac2 + xc2 = byc.
Combining this, we obtain b2 x = ac2 +
xc2 , or x(b2 − c2 ) = ac2 .
Suppose on the contrary that x is an
integer. Then b2 − c2 is a divisor of ac2 .
But we know gcd(b, c) = 1, and hence gcd(b2 − c2 , c) = gcd(b2 , c) = 1.
Therefore, b2 − c2 is a divisor of a. This yields b2 − c2 ≤ a. Note that
b2 − c2 > 0, because x(b2 − c2 ) = ac2 ; hence, b − c > 0 also holds. Therefore,
b2 − c2 = (b − c)(b + c) ≥ 1 · (b + c), because b and c are positive integers.
Hence, a ≥ b + c, which is contradicting the triangle inequality. Therefore,
x = |BD| cannot be an integer.
20
3. Let bi = a1 a2 · · · ai , for 1 ≤ i ≤ p. We will prove that it is possible to
choose a permutation such that bi ≡ i mod p for all i. For i ≥ 2, the
congruence ai ≡ bi · b−1 i−1 mod p holds, if bi−1 6≡ 0 mod p. Therefore, we
now choose a1 = 1 and ai ≡ i · (i − 1)−1 mod p for 2 ≤ i ≤ p. Now it is
sufficient to prove that ai 6≡ 1 mod p for all 2 ≤ i ≤ p and ai 6≡ aj mod p
for all 2 ≤ j < i ≤ p.
Suppose the contrary. Then ai ≡ 1 mod p for certain 2 ≤ i ≤ p. Then
i · (i − 1)−1 ≡ 1 mod p holds, hence i ≡ i − 1 mod p, or 0 ≡ −1 mod p.
Because p ≥ 2, this is a contradiction. Now suppose that ai ≡ aj mod p
for certain 2 ≤ j < i ≤ p. Then we have i · (i − 1)−1 ≡ j · (j − 1)−1 mod p,
hence i(j − 1) ≡ j(i − 1) mod p, or ij − i ≡ ij − j mod p, or −i ≡ −j
mod p. However, as 2 ≤ j < i ≤ p, this is impossible.
We conclude that if we choose the ai as above, all ai are distinct, so that
they indeed form a permutation of 1, 2, . . . , p. Moreover, by definition
we have that a1 a2 · · · ai ≡ i mod p, therefore also the second condition is
satisfied.
4. Let M be the other intersection point

of AD with the circumcircle of 4ABC.
Then M is the midpoint of the circle arc
BC on which A does not lie. Now we have
∠COM = 12 ∠COB = ∠CAB = 60◦ .
Moreover, we have |OC| = |OM |, hence
4OCM is isosceles with apex angle equal
to 60◦ . This means that it is equilateral,
hence |CM | = |CO|. Because OM is
perpendicular to BC, this yields that M
is the reflection of O in BC, and hence
∠DOM = ∠DM O.
Moreover, we have ∠DM O = ∠AM O = ∠M AO, because |OA| = |OM |.
We now obtain ∠ODE = 90◦ − ∠DOM = 90◦ − ∠DM O = 90◦ − ∠M AO =
90◦ − ∠DAE. Therefore, ∠ODE + ∠DAE = 90◦ . In triangle ADE we
have 180◦ = ∠DAE + ∠AED + ∠ODE + ∠ADO, hence we conclude that
∠AED + ∠ADO = 90◦ .
5. For 1 ≤ i ≤ n we have
(i+1)2 (i+1)2 −(i+1)(x+i) (i+1)(1−x)
x+i =i+1+ x+i =i+1+ x+i ,
so we can rewrite the left hand side of the equation to
2(1−x) 3(1−x) (n+1)(1−x)
nx2 + 2 + 3 + . . . + (n + 1) + x+1 + x+2 + ... + x+n .
21
We have 2 + 3 + . . . + (n + 1) = 12 n(n + 3), hence this sum cancels against
n(n+3)
2 on the right hand side of the original equation. Moreover, we can
move nx2 to the other side, and write a separate factor 1 − x in all fractions.
The equation then becomes

(1 − x) · x+1 2 3
+ x+2 n+1
+ . . . + x+n = nx − nx2 .
The right hand side, we can factor as nx(1 − x). Now, we observe that
x = 1 is a solution to this equation. If there were another solutions x 6= 1,
then it would satisfy
2 3 n+1
x+1 + x+2 + ... + x+n = nx.
However, for 0 < x < 1 we have

2 3 n+1 2 3 n+1
x+1 + x+2 + ... + x+n > 2 + 3 + ... + n+1 = n > nx,
while for x > 1 we have

2 3 n+1 2 3 n+1
x+1 + x+2 + ... + x+n < 2 + 3 + ... + n+1 = n < nx.
6 1. We conclude that x = 1 is the

Hence, there are no solutions with x =
only solution, for all n.
22
Problems
1. Suppose a grid with 2m rows and 2n columns is given, where m and n

are positive integers. You may place one pawn on any square of this grid,
except the bottom left one or the top right one. After placing the pawn, a
snail wants to undertake a journey on the grid. Starting from the bottom
left square, it wants to visit every square exactly once, except the one with
the pawn on it, which the snail wants to avoid. Moreover, it wants to finish
in the top right square. It can only move horizontally or vertically on the
grid.
On which squares can you put the pawn for the snail to be able to finish
its journey?
2. Suppose a triangle 4ABC with ∠C = 90◦ is given. Let D be the midpoint

of AC, and let E be the foot of the altitude through C on BD. Show that
the tangent in C of the circumcircle of 4AEC is perpendicular to AB.
3. Let n ≥ 0 be an integer. A sequence a0 , a1 , a2 , . . . of integers is defined as

follows: we have a0 = n and for k ≥ 1, ak is the smallest integer greater
than ak−1 for√which ak + ak−1 is the square of an integer. Prove that there
are exactly b 2nc positive integers that cannot be written in the form
ak − a` with k > ` ≥ 0.
4. Let A be a set of functions f : R → R. For all f1 , f2 ∈ A there exists a

f3 ∈ A such that
f1 f2 (y) − x + 2x = f3 (x + y)
for all x, y ∈ R. Prove that for all f ∈ A, we have

f x − f (x) = 0
for all x ∈ R.
23
Solutions
1. Number the rows from bottom to top by 1, 2, . . . , 2m, and the columns
from left to right by 1, 2, . . . , 2n, so that the snail starts in square (1, 1) and
finishes in (2m, 2n). Colour the squares of the grid like a chessboard, where
(i, j) is coloured black if i + j is even and white if i + j is odd. Since 2m
and 2n are even, the number of black squares equals that of white squares.
The snail starts and finishes on a black square, and alternates between
black and white squares on its journey. Therefore, the number of black
squares it visits is one larger than that of white squares, so the pawn must
be on a white square (i.e. a square (i, j) with i + j odd) for the snail to be
able to finish its journey.
Let us show that if the pawn is on such a square, then the snail is always
able to undertake its journey. Let 1 ≤ k ≤ m and 1 ≤ l ≤ n be such the
pawn is either on row 2k − 1 or row 2k, and either on column 2l − 1 or
column 2l. Then have the snail move all the way to the right on odd rows
with a number less than 2k − 1 (and then move one square up, which puts
the snail on an even row), and all the way to the left on even rows with
a number less than 2k − 1 (and then move one square up, which puts the
snail on an odd row again). This puts the snail on square (2k − 1, 1). From
here, while the snail is in a column with number less than 2l − 1, have the
snail move through each 2 × 2-block in the order: bottom left, top left,
top right, bottom right. After moving one square to the right, this puts
the snail on the bottom left square of the next 2 × 2-block. This puts the
snail on square (2k − 1, 2l − 1), which cannot be occupied by the pawn as
2k − 1 + 2l − 1 is even. In this 2 × 2-block, the pawn is either on the top
left or the bottom right square; in the former case, move the snail in the
order: bottom left, bottom right, top right; in the latter case, move the
snail in the order: bottom left, top left, top right. So in both cases, the
snail ends up in the top right square of the 2 × 2-block. The remainder of
the 2 × 2-blocks in rows 2k − 1 and 2k can now be covered by moving in
each of them in the order: top left, bottom left, bottom right, top right.
This puts the snail on square (2k, 2n), with all required squares to the
bottom and left visited exactly once. The snail can now finish its journey
by repeatedly doing the following: move one square up, move all the way
to the left, move one square up, move all the way to the right. This puts
the snail on square (2m, 2n), as required.
Therefore, the snail is able to finish its journey if and only if the pawn is
put on a square (i, j) with i + j odd.
24
2. Let S be the intersection of the
tangent with AB. Then we need
to show that ∠BSC = 90◦ .
Since ∠BCD = 90◦ = ∠CED
we have 4BCD ∼ 4CED, so
|DC| |DE|
|DB| = |DC| . As |DA| = |DC|
|DA|
it follows that |DB| = |DE|
|DA| . To-
gether with ∠BDA = ∠EDA
this gives 4ADE ∼ 4BDA (SAS). Therefore ∠ABD = ∠EAD. Since
SC is a tangent to the circumcircle of 4AEC we also have that (by the
tangent chord angle theorem) ∠EAD = ∠EAC = ∠SCE. So ∠SCE =
∠ABD = ∠SBE. Hence SBCE is a cyclic quadrilateral. Therefore
∠BSC = ∠BEC = 90◦ .
√
3. Let m = b 2nc. We first show that the sequence of squares a0 + a1 , a1 +
a2 , . . . is the sequence (m + 1)2 , (m + 2)2 , . . . . After that, we will show that
the differences ai − ai−1 form a sequence of consecutive even numbers and
a sequence of consecutive odd numbers. We will use this to prove the given
statement.
Note that a0 + a1 is the smallest square greater than 2n. Therefore
a0 + a1 = (m + 1)2 . We show by induction on i that ai−1 + ai = (m + i)2 .
For i = 1, this is what we have just shown. Suppose that aj−1 +aj = (m+j)2 .
2
Then aj−1 ≥ (m+j−1) 2 (since otherwise aj could have been chosen in such
a way that aj−1 + aj = (m + j − 1)2 ), so
(m+j−1)2
aj = (m + j)2 − aj−1 ≤ (m + j)2 − 2
2m2 +4mj+2j 2 −(m2 +2mj+j 2 +1−2m−2j)
= 2
m2 +2mj+j 2 −1+2m+2j m2 +2mj+j 2 +1+2m+2j (m+j+1)2
= 2 < 2 = 2 .
We deduce that aj + aj+1 ≤ (m + j + 1)2 . Moreover, we have aj + aj+1 >

aj−1 + aj = (m + j)2 , so aj + aj+1 = (m + j + 1)2 , which completes the
induction.
Define bi = ai − ai−1 for all i ≥ 1. Then we have
bi+2 − bi = ai+2 − ai+1 − ai + ai−1
= (ai+2 + ai+1 ) + (ai + ai−1 ) − 2(ai+1 + ai )
= (m + i + 2)2 + (m + i)2 − 2(m + i + 1)2
= (m + i)2 + 4(m + i) + 4 + (m + i)2 − 2(m + i)2 − 4(m + i) − 2
=2
25
for all i ≥ 1. Hence
(b1 , b3 , b5 , . . .) = (b1 , b1 + 2, b1 + 4, . . .),

(b2 , b4 , b6 , . . .) = (b2 , b2 + 2, b2 + 4, . . .).
We have b1 + b2 = (a2 − a1 ) + (a1 − a0 ) = (a2 + a1 ) − (a1 + a0 ) =

(m + 2)2 − (m + 1)2 = 2m + 3. In particular, b1 and b2 have distinct parities.
So we can write every integer that is at least b1 and has the same parity
as b1 as ak − ak−1 for some k. The same is true for every integer that is
at least b2 and has the same parity as b2 . All integers of the form ak − a`
with k ≥ ` + 2 are at least bk + bk−1 ≥ b1 + b2 and therefore both greater
than b1 and greater than b2 . Hence this does not give us any new numbers
of the form ak − a` with k > ` ≥ 0.
We deduce that the numbers not of the form ak − a` with k > ` ≥ 0 are
precisely those that either are less than b1 and have the same
parity
as b1 , or
are less than b2 and have the same parity as b2 . There are b12−1 + b22−1
of those. Note that the argument of exactly one of the floors is an integer.
Hence we can rewrite the above as b12−1 + b22−1 − 12 = b1 +b22 −3 = 2m2 = m.
√
We conclude that there are exactly m = b 2nc positive integer which
cannot be written in the required form.

4. Substituting x = 0 gives f1 f2 (y) = f3 (y) therefore the f3 corresponding
to f1 , f2 ∈ A is the composition f3 (x) = f1 (f2 (x)). Substituting x = −y
now gives that for all f1 , f2 ∈ A we have

f1 f2 (y) + y − 2y = f3 (0) = f1 (f2 (0))
for all y ∈ R.
Substituting x = f2 (y) now gives that for all f1 , f2 ∈ A, there exists f3 ∈ A
with
f1 (0) + 2f2 (y) = f3 f2 (y) + y
for all y ∈ R. We have already seen that we can write f3 (f2 (y) + y) as
2y + f3 (f2 (0)). Hence
2f2 (y) = 2y + f3 (f2 (0)) − f1 (0)
for all y ∈ R. We deduce that there exists a d ∈ R, independent of y, such

that f2 (y) = y + d for all y ∈ R.
Hence f ∈ A is of
the form f (x) = x + d with d a constant. Therefore we
have f x − f (x) = f x − (x + d) = f (−d) = −d + d = 0.
26
Problems
1. (a) If c(a3 + b3 ) = a(b3 + c3 ) = b(c3 + a3 ) with a, b, c positive real numbers,

does a = b = c necessarily hold?
(b) If a(a3 + b3 ) = b(b3 + c3 ) = c(c3 + a3 ) with a, b, c positive real numbers,
does a = b = c necessarily hold?
2. Find all positive integers n for which there exists a positive integer k such
that for every positive divisor d of n, the number d − k is also a (not
necessarily positive) divisor of n.
3. Let ABC be an acute triangle, and let D be the foot of the altitude through
A. On AD, there are distinct points E and F such that |AE| = |BE| and
|AF | = |CF |. A point T 6= D satisfies ∠BT E = ∠CT F = 90◦ . Show that
|T A|2 = |T B| · |T C|.
4. In the classroom of at least four students the following holds: no matter

which four of them take seats around a round table, there is always someone
who either knows both of his neighbours, or does not know either of his
neighbours. Prove that it is possible to divide the students into two groups
such that in one of them, all students know one another, and in the other,
none of the students know each other.
(Note: if student A knows student B, then student B knows student A as
well.)
27
Solutions
√
1. (a) We claim that (a, b, c) = (2, 2, −1 + 5) satisfies the given equalities.
√
As all real numbers in this triple are positive, and 2 6= −1 + 5, the
answer to the first question is ”no”.
√ √ √ √
Note that c3 = (−1 + 5)3 = −1 + 3 √ · 5 − 3 · 5 + 5 5 = −16√+ 8 5.
Then we have c(a3 + b3 ) = (−1 + √5) · 2 · 8 = −16 + √ 16 5 and
a(b3 + c3 ) = b(c3 + a3 ) = 2 · (−16 + 8 5 + 8) = −16 + 16 5, so the
given equalities are indeed satisfied.
(b) We show that if the given equalities are satisfied, that then a = b = c.
Without loss of generality, we assume that a ≥ b, c. Then we have
a(a3 + b3 ) ≥ b(a3 + b3 ) ≥ b(c3 + b3 ) = a(a3 + b3 ),
so equality holds everywhere. Now a = b follows from a3 + b3 being

positive, and a = c follows from b being positive. Therefore a = b = c.

2. If n is either 1 or a prime number, then the positive divisors of n are 1 and

n (which coincide if n = 1). In this case we can take k = n + 1 and note
that 1 − (n + 1) = −n and n − (n + 1) = −1 are also divisors of n. Hence
if n = 1 or if n is a prime number, the given property is satisfied. If n = 4,
the positive divisors are 1, 2, and 4. Taking k = 3, we see that −2, −1, 1
are divisors of 4, so n = 4 also has the given property. If n = 6, the positive
divisors are 1, 2, 3, and 6. Taking k = 4, we see that −3, −2, −1, and 2
are divisors of 6, so n = 6 has the given property. Hence if n ≤ 6 or if n is
a prime number, n has the given property.
Now suppose that n > 6 is composite. Suppose that k is a positive integer
such that for every positive divisor d of n, the number d − k is also a divisor
of n. Since n is a positive divisor of n, it follows that n − k is a divisor of n.
As the next largest divisor of n is at most 12 n, it follows that n − k ≤ 21 n,
and therefore that k ≥ 12 n. Moreover, 1 is a positive divisor of n, so 1 − k
is a divisor of n. Note that 1 − k ≤ 1 − 12 n. As n > 6, we have 16 n > 1, so
1 1 1 1
2 n − 3 n > 1, and therefore − 3 n > 1 − 2 n. The only divisors that are at
most 1 − 2 n are therefore −n and (if n is even) − 12 n. Therefore 1 − k = −n
1
or 1 − k = − 12 n.
In the latter case, we have k = 12 n + 1, so n − k = 12 n − 1. However, in
the same way as before, we see that for n > 6 no divisor of n is equal to
1 1 1 1
2 n − 1, as the next largest divisor after 2 n is at most 3 n < 2 n − 1. This
contradicts n − k being a divisor of n.
28
In the former case, we have 1 − k = −n, so k = n + 1. As n is composite,
there exists a divisor d with 1 < d < n. Then d − k = d − n − 1 is also
a divisor of n. However, d ≤ 12 n, so d − n − 1 ≤ − 12 n − 1, but the only
divisor that is at most − 12 n − 1 is −n. We deduce that d − n − 1 = −n, so
d = 1, which is a contradiction.
Therefore the n that have the given property are precisely those n for which
either n ≤ 6 or n is prime.
A
3. We consider the configuration
in the figure.
Let M be the midpoint of AB,
and let N be the midpoint of
AC. From what is given in the E
problem, it follows that E lies
on the perpendicular bisector of
T
AB, so ∠BM E = 90◦ . Since
we also have ∠BT E = 90◦ and B D
C
◦
∠BDE = 90 , by Thales’s theo-
rem it follows that BDT EM is
F
a cyclic pentagon. In the same
way, it follows that CF DT N is a cyclic pentagon. Now ∠N T M =
360◦ − ∠DT N − ∠M T D = (180◦ − ∠DT N ) + (180◦ − ∠M T D) = ∠DCN +
∠M BD = ∠BCA + ∠ABC = 180◦ − ∠CAB = 180◦ − ∠N AM . Therefore
AM T N is a cyclic quadrilateral (which also follows directly from Miquel’s
theorem).
Now let us chase some angles in order to show that 4T BA ∼ 4T AC. We
have ∠N T C = ∠N F C = 90◦ − ∠F CN since ∠F N C = 90◦ . Moreover,
we have ∠F CN = ∠F CA = ∠CAF since |AF | = |CF |, so ∠N T C =
90◦ − ∠CAF = 90◦ − ∠CAD = ∠DCA = ∠BCA. Therefore ∠ACT =
∠N CT = 180◦ − ∠N T C − ∠CN T = 180◦ − ∠BCA − ∠CN T . As M N
is a mid-parallel in triangle ABC, we have M N k BC and it follows
that 180◦ − ∠BCA = ∠CN M . Hence ∠ACT = ∠CN M − ∠CN T =
∠T N M = ∠T AM = ∠T AB (using the cyclicity of the quadrilateral
AM T N ). Similarly, we have ∠ABT = ∠T AC, so 4T BA ∼ 4T AC. It
follows that |T B| |T A| 2
|T A| = |T C| , so |T A| = |T B| · |T C|, as required.
4. Consider a group of students in which all students know each other, and
which has the maximal number of students possible. (Such a group exists,
as there are only finitely many students in the classroom, and there exists
a group of students in which all students know one another, e.g. a group
29
consisting of only one student.) Denote this group of students by X. We
show that by dividing the students in X and the group of students not in
X, we may achieve a division with the required properties.
Note that it suffices to show that in the group of students not in X, none
of the students know each other. Suppose for a contradiction that A and
B are students not in X who know each other. As X is chosen to have
the maximal number of students possible, there is a student A0 in X who
doesn’t know A, and a student B 0 who doesn’t know B. We first show that
we can take A0 and B 0 to be different. If not, then there is a unique student
C in X who doesn’t know A, and a unique student in X who doesn’t know
B (and they are the same student). In other words all other students
in X know both A and B, so by replacing C by A and B, we obtain a
group of students in which all students know one another; all students in
X know each other, A and B know everyone in X (aside from C), and A
and B know each other. However this group is larger than X, which is a
contradiction.
Therefore we may assume that A0 and B 0 are distinct. Note that A0 and B 0
are in X, so they know each other. Now if A, B, B 0 , A0 , in that order, take
seats around a round table, everyone knows precisely one of his neighbours;
A and B, and A0 and B 0 know each other, but A and A0 , and B and B 0
do not know each other. This contradicts the given property. Therefore
dividing the students into X and the group of students not in X gives a
division with the required properties.
30
Problems
1. A set of lines in the plan is called nice if every line in the set intersects an
odd number of other lines in the set.
Determine the smallest integer k ≥ 0 having the following property: for
each 2018 distinct lines `1 , `2 , . . . , `2018 in the plane, there exist lines
`2018+1 , `2018+2 , . . . , `2018+k such that the lines `1 , `2 , . . . , `2018+k are
distinct and form a nice set.
2. Find all functions f : R → R such that
f (x2 ) − f (y 2 ) ≤ f (x) + y x − f (y)

for all x, y ∈ R.
3. Determine all pairs (a, b) of positive integers such that (a + b)3 − 2a3 − 2b3
is a power of two.
4. In a non-isosceles triangle ABC the centre of the incircle is denoted by

I. The other intersection point of the angle bisector of ∠BAC and the
circumcircle of 4ABC is D. The line through I perpendicular to AD
intersects BC in F . The midpoint of the circle arc BC on which A lies, is
denoted by M . The other intersection point of the line M I and the circle
through B, I and C, is denoted by N . Prove that F N is tangent to the
circle through B, I and C.
31
Solutions
1. First we prove that the number of lines in a nice set is even. Suppose on
the contrary that the number of lines is odd. Then on each of the odd
number of lines, there is an odd number of intersection points, so the total
number of intersection points is odd. However, each intersection point is
counted twice (once for each of the lines on which is is lying), so the total
should be even, which is a contradiction. So the total number of lines in a
nice set has to be even. In particular, 2018 + k must be even, so k must be
even.
Now suppose that there are 1009 directions, such that in each direction
there are two of the original lines. Then each line is parallel to exactly
one other line. If k < 1010, then k ≤ 1008, so there must be a direction
in which we do not add a line. Consider one of the original lines ` in this
direction. Within the final set of all lines, it intersects all lines except itself
and the line it is parallel to. This is an even number. Hence, the resulting
set is not nice. Therefore, there is an example in which at least 1010 are
needed.
We will now show that it is possible to add exactly 1010 lines such that the
resulting set is nice. We consider all directions in which there is an even
number (greater than 0) of the original lines. There are at most 1009 such
directions. For each of these directions, we add one line. Each line is then
parallel to an even number (possibly 0) of other lines. First suppose that
the total number of lines is even. Then each lines intersects an odd number
of lines (the total set minus itself and an even number of other lines). Now
suppose that the total number of lines is odd. Then each line intersects
an even number of lines. We ass a line in a new direction, which therefore
intersects all lines (an odd number), so that each line after that intersects
an odd number of other lines.
Now we have a nice set, and we have added at most 1009 + 1 = 1010
lines. Possibly, there are less than 1010. In this case, the number of lines
added is even, because the total number of lines in a nice set is always
even. We choose a direction in which there is at least one line. We then
keep adding two lines in this direction. These lines intersect all lines in
the other direction, which is an odd number in total. These other lines
each got two new intersection points, so they still have an odd number
of intersection points. This also holds for the existing lines in the chosen
direction, because they get zero new intersection points. The set is still
nice. We keep adding pairs of lines until we have 1010 lines in total.
We conclude that the minimum k satisfying the conditions is k = 1010.
32
2. Substituting x = y = 0 yields 0 ≤ f (0) · −f (0). However, squares are
non-negative, so this yields f (0)2 = 0 and hence f (0) = 0. Now taking
x = 0 and y = t, we obtain −f (t2 ) ≤ t · −f (t), while taking x = t and y = 0
yields f (t2 ) ≤ f (t) · t. We find tf (t) ≤ f (t2 ) ≤ tf (t), hence f (t2 ) = tf (t)
for all t ∈ R. On the left hand side of the function inequality we can replace
f (x2 ) − f (y 2 ) by xf (x) − yf (y). By expanding on the right hand side, we
obtain xf (x) − yf (y) + xy − f (x)f (y), hence
f (x)f (y) ≤ xy.
Then f (t2 ) = tf (t) yields f (1) = −f (−1). By first substituting y = 1 and

then y = −1, we obtain, for all x ∈ R:
x ≥ f (x)f (1) = −f (x)f (−1) ≥ −x · −1 = x.
Hence, equality must hold, which means that f (x)f (1) = x. In particular,
this yields f (1)2 = 1, hence f (1) = 1 or f (1) = −1. In the former case, we
get f (x) = x for all x and in the latter case, we get f (x) = −x for all x.
When we check f (x) = x with the original function inequality, we get
x2 − y 2 on the left hand side and (x + y)(x − y) = x2 − y 2 on the right
hand side, hence this function satisfies the inequality. When we check
f (x) = −x, the left hand side becomes −x2 + y 2 and the right hand side
becomes (−x + y)(x + y) = −x2 + y 2 , so also this function is a solution.
We conclude that the solutions are: f (x) = x for all x ∈ R and f (x) = −x
for all x ∈ R.
3. First we determine the pairs (a, b) with gcd(a, b) = 1. We have
(a + b)3 − 2a3 − 2b3 = a3 + 3a2 b + 3ab2 + b3 − 2a3 − 2b3

= −a3 − b3 + 3ab(a + b)
= −(a + b)(a2 − ab + b2 ) + 3ab(a + b)
= (a + b)(−a2 + ab − b2 + 3ab)
= (a + b)(−a2 − b2 + 4ab).
This has to be a power of two. Because a + b is positive, the other factor

must also be positive, and they are both powers of two. As gcd(a, b) = 1,
at least one of a and b is odd. Then a2 + b2 is congruent to 1 or 2 modulo
4, hence −a2 − b2 + 4ab is congruent to 3 or 2 modulo 4. However, it is also
a power of two, so 3 modulo 4 is impossible, and in the case of 2 modulo 4
it has to equal 2. We conclude that −a2 − b2 + 4ab = 2.
33
Suppose that a + b is at least 8. Because it is a power of two, it is also
divisible by 8, hence we can write b = 8m − a for a certain positive integer
m. Then we have
2 = −a2 − b2 + 4ab
= −a2 − (8m − a)2 + 4a(8m − a)
= −a2 − 64m2 + 16ma − a2 + 32ma − 4a2
= −6a2 + 48ma − 64m2 ,
hence
1 = −3a2 + 24ma − 32m2 .
Modulo 8, this becomes 1 ≡ −3a2 . If a is even, then the right hand side
is even, and if a is odd, then the right hand side is −3 modulo 8. In both
cases, the equality does not hold modulo 8. We conclude that a + b is not
greater or equal to 8. Because a and b are both at least 1, a + b is equal to
2 or 4.
• If a + b = 2, then a = b = 1. Now we have −a2 − b2 + 4ab = 2, hence
the product of the two factors is a power of two (namely 4). Hence
(1, 1) is a solution.
• If a + b = 4, then because of gcd(a, b) = 1 we have either a = 1 and
b = 3 or the other way around. Now −a2 − b2 + 4ab = 2, so also the
product of the two factors is a power of two (namely 8). Hence (1, 3)
and (3, 1) are solutions.
Now consider a solution (a, b) with gcd(a, b) = d > 1. Then d is a divisor

of (a + b)3 − 2a3 − 2b3 , hence d has to be a power of two itself. We now
remark that ( ad , db ) is another solution, because d3 is divided out from the
power of two (a + b)3 − 2a3 − 2b3 , leaving another power of two. Hence,
( ad , db ) has to be one of the three previously found pairs. On the other hand,
each of these three pairs multiplied by an arbitrary power of two is another
solution.
We conclude that all possible solutions are given by
(a, b) = (2k , 2k ), (a, b) = (2k , 3 · 2k ), and (a, b) = (3 · 2k , 2k ),
where k runs through all non-negative integers.
34
4. We consider the configuration in which |AB| > |AC|, causing C to lie
between B and F . For other configurations, the proof is analogous.
Because ∠DAC = ∠BAD holds, D is the midpoint of the circle arc BC on
which A does not lie. Hence, the line DM is a diameter of the circle. Thales’s
theorem yields ∠DBM = 90◦ . Let K be the intersection point of BC and
DM . As DM is the segment bisector of BC, we have ∠BKM = 90◦ =
|M B| |M D|
∠DBM . Therefore, 4M KB ∼ 4M BD (AA). This yields |M K| = |M B| ,
hence |M D| · |M K| = |M B|2 . Because K is on the interior of M D, this
equality also holds using directed distances: M D · M K = M B 2 .
Because D is the midpoint of arc BC, we have |DB| = |DC|. Moreover,
∠CDI = ∠CDA = ∠CBA and ∠DCI = ∠DCB + ∠BCI = ∠DAB +
∠BCI = 12 ∠CAB + 12 ∠BCA. Using the sum of the angles in triangle
DCI, we now get that ∠DIC = 180◦ − ∠CBA − 21 ∠CAB − 12 ∠BCA =
1 1
2 ∠CAB + 2 ∠BCA = ∠DCI. Hence, triangle DCI is isosceles with
|DC| = |DI|. We conclude that D is the centre of the circle through B,
C and I. Now DB is the radius of this circle, and DB is perpendicular
to M B, which yields that M B is a tangent line. Using the secant-tangent
theorem, we get M B 2 = M I · M N . Together with the previous paragraph
we conclude that M D · M K = M I · M N . Hence, N DKI is a cyclic
quadrilateral because of the same theorem.
We also know that ∠F ID = 90◦ = ∠F KD, hence DKIF is a cyclic
quadrilateral. We conclude that N DKIF is a cyclic pentagon. This yields
∠DN F = 180◦ − ∠DIF = 90◦ . Hence, N F is perpendicular to the radius
DN of the circle through B, I and C; this means that F N is tangent to
this circle.
35
Problems
Part 1
1. A positive three-digit number is called nice if the sum of the last two digits
equals the first digit. For example, 123 is not nice, because 1 is not equal
to 2 + 3. How many three-digit numbers are nice?
Note that a three-digit number cannot start with digit 0.
A) 45 B) 48 C) 50 D) 54 E) 55
2. The faces of a cube have different colours. In the figure you can see a net
for the cube. The points A and B in the net correspond to two vertices of
the same face of the cube.
What colour does that face have?
A) red B) blue C) green D) black E) yellow
B
red white blue
A
green black yellow
3. We consider sequences of 20 integers. The integers can be positive or

negative, but cannot be zero. Also, we impose the following conditions on
the sequences: of any two adjacent numbers at least one is positive; the
sum of any three adjacent numbers is negative; the product of any four
adjacent numbers is positive.
Consider the following four statements about such sequences:
• There can never be two adjacent positive numbers.
• There may be more positive than negative numbers.
• The sum of all 20 numbers is always negative.
• The number −1 can never occur.
How many of these statements are true?
A) 0 B) 1 C) 2 D) 3 E) 4
36
4. The number n2 + 21 is the square of an integer.
For how many positive integers n does this hold?
A) 0 B) 1 C) 2 D) 3 E) 4
5. Sanne is building a 9×9×9 cube by gluing 1×1×1 blocks together. She

doesn’t have quite enough blocks to complete the task. Therefore, she
decides to leave out some of the blocks from the large cube. In order to still
get a nice rigid cube, she makes sure that no two holes (left out blocks) are
adjacent. In fact, two holes should not even touch in an edge or a single
vertex. Also, she does not leave out any of the blocks on the outside of the
cube.
What is the minimum number of blocks that Sanne needs to build the
cube?
A) 365 B) 604 C) 665 D) 673 E) 702
6. Peter starts out with the numbers 1, 2, 3, and 4. He may take two of his
numbers and replace them by their sum, their product, or their difference.
He performs this replacement step three times, after which a single number
remains.
Example. He could replace the 2 and the 4 by 2 + 4 = 6, then replace the 1 and
the 3 by 3 − 1 = 2, and finally replace the 6 and the 2 by 8 = 2 + 6. Then, the
remaining number would be 8.
Which of the following five numbers cannot be the number that remains?
A) 28 B) 30 C) 32 D) 34 E) 36
7. Four circles together enclose ten regions in the

plane, as in the figure. We want to place the
numbers 1 to 10 inside the regions (one number ?
per region). This must be done in such a way
that adding the numbers inside a circle gives
the same answer for all four circles.
Which number should be placed in the region
with the question mark?
A) 1 B) 2 C) 4 D) 6 E) 7
37
8. You have a collection of hats. Each hat has three attributes: the colour
(red or blue), the shape (top hat or pointed hat), and the pattern (spots
or stripes). You put a number of gnomes in a room and put a hat on
each of them. For any two gnomes their hats must be different, yet share
at least one attribute (for example: both hats are blue). The gnomes
can see everyone’s hat, except their own. The gnomes are not allowed to
communicate with one another.
What is the minimum number of gnomes you have to put in the room in
order to be be sure that one of them can determine one of the attributes of
his own hat?
A) 3 B) 4 C) 5
D) 8 E) That is impossible for any number of gnomes.
Part 2
1. Stef has 18 coins of which 17 are identical, but one is slightly lighter than
the other coins. Together, the 18 coins weigh 214 grams. Stef removes
two of the coins and weighs the remaining 16 coins. Together the 16 coins
weigh 190 grams.
How much does the lighter coin weigh?
2. We say that a positive integer is balanced if the average of the first two
digits is 2, the average of the first three digits (if they all exist) is 3, the
average of the first four digits (if they all exist) is 4, et cetera.
What is the largest balanced number?
3. What is the area of the crown-shaped area? 5

Note that the figure is not drawn to scale.
5
4. A clock has the numbers 1 to 12 for indicating the hours. Ernie has
interchanged these twelve numbers in such a way that any two adjacent
numbers on the clock differ by either 2 or 3. Fortunately, the number 12 is
still in the right place, but the 9 is where the 1 is supposed to be.
What number is in the place where the 9 is supposed to be?
38
5. The numbers 1 to 8 are assigned to the eight faces
of an octahedron. For each vertex, we compute
the sum of the four numbers on the faces meeting
in that vertex. For four of the vertices we get the
same outcome. For a fifth vertex the outcome is
16.
What is the outcome for the sixth vertex?
6. When a, b, c, and d are digits, we denote by abcd the number composed

of those four digits. The numbers abcd and cbad are both perfect squares.
The number bad is the cube of a positive integer.
Determine the number abcd.
7. An employee at the supermarket is stacking crisps canisters. There are two

types of canisters: small ones and large ones. Three small canisters stack
to the same height as one large canister. The employee makes a stack of 12
small canisters. Next to it, he makes more stacks of the same height, but
all stacks are different. (If one stack starts with a small canister followed
by a large one, and the other starts with a large one followed by a small
one, the two stacks are different.)
How many different stacks can he make, including the first stack?
8. You may choose any number consisting of five different digits. The digit 0
cannot be used. The next step is to choose two adjacent digits and switch
their positions. You may perform this step five times in total. Finally, you
compute the difference between the initially chosen number and the final
number obtained after switching.
What is the largest possible difference that can be obtained?
Example. Suppose you choose the number 47632. Then you could switch
digits to obtain, in this order, the numbers 46732, 46372, 46327, 43627, and
34627. Then, the difference between the initial number and the final number is
47632 − 34627 = 13005.
39
Answers
Part 1
1. D) 54
2. D) black
3. C) 2 (statements 1 and 3)
4. C) 2
5. C) 665
6. D) 34
7. E) 7
8. E) That is impossible for any number of gnomes.
Part 2
1. 10 grams 5. 20
2. 40579 6. 1296
3. 60 7. 60
4. 5 8. 85230
40
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
OmslagNWO2018-v2.indd 2 17-06-18 13:42

and the team selection for IMO 2019 United Kingdom
IMO Team Selection Test 1, May 2019 CM
MY
IMO Team Selection Test 2, May 2019

CY
We eat problems
CMY

for breakfast.

Olympiad 2018
International
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction
23 IMO Team Selection Test 1, May 2019

Introduction

represent the Netherlands at the EGMO in Kyiv, Ukraine, from 7 until 13
April. The team brought home a silver medal, two bronze medals, and a
honourable mention; a very nice achievement. For more information about
the EGMO (including the 2019 paper), see www.egmo.org.
held in Valkenswaard, Netherlands, from 26 until 28 April. The Dutch team
received three silver medals, four bronze medals and a honourable mention.
For more information about the BxMO (including the 2019 paper), see
www.bxmo.org.
1
In May the team for the International Mathematical Olympiad 2019 was
selected by three team selection tests on 29, 30 and 31 May, each lasting
Bristol, United Kingdom, from 6 until 14 July.
the solutions.
Dutch delegation
The Dutch team for IMO 2019 in the United Kingdom consists of
• Szabi Buzogány (19 years old) • Matthijs van der Poel (18 years old)
– silver medal at BxMO 2018, – bronze medal at BxMO 2016,
silver medal at BxMO 2019 bronze medal at BxMO 2017
– honourable mention at IMO – observer C at IMO 2016,
2018 silver medal at IMO 2017,
• Jesse Fitié (17 years old) silver medal at IMO 2018
• Jovan Gerbscheid (16 years old) • Richard Wols (16 years old)
– silver medal at BxMO 2018 – bronze medal at BxMO 2018,
– bronze medal at IMO 2018 silver medal at BxMO 2019
• Jippe Hoogeveen (16 years old) – observer C at IMO 2018

• Tjeerd Morsch (17 years old)

• Jeroen Huijben (observer B), University of Amsterdam
2
Problems
A-problems
1. In a classroom there are chairs and stools. On each chair and on each stool
one child is seated. Each chair has 4 legs, each stool has 3 legs and each
child has 2 legs. Together, we have a total of 39 legs.
How many chairs are there in the classroom?
A) 3 B) 4 C) 5 D) 6 E) 9
2. On an island, there are knights and knaves. Knights always speak the truth
and knaves always lie. On the island you meet five people. You know that
four of them are knights and one of them is a knave, but you do not know
who is the knave. They make the following statements about the island
inhabitants:
• A: “All knaves have shoe size 40.”
• B: “All people with shoe size 40 have a goldfish.”
• C: “All people with a goldfish are knaves.”
• D: “I have shoe size 40.”
• E: “I have a goldfish.”
Which of them is the knave?
A) A B) B C) C D) D E) E
3. If you continue the chain of squares and regular pentagons

in the same way, does it connect to itself after going
around? If so, how many pentagons do you need?
A) 9 B) 10 C) 11 D) 12 E) It does not connect.
4. Julian wants to compose a list of integers. He wants the list to be as long

as possible. Each integer on the list must consist of one or more of the
digits 1 to 9. Moreover,
• each of the digits 1 to 9 is used exactly once;
• no integer in the list is divisble by another integer in the list.
What is the maximum number of integers in Julian’s list?
A) 4 B) 5 C) 6 D) 7 E) 8
3
5. Nine people are at a party. While entering, some of them shook hands.
Quintijn is at the party and asks each of the others how many hands they
shook. He gets eight different answers.
How many hands did Quintijn shake?
A) 0 B) 1 C) 2 D) 3 E) 4
6. Birgit is studying positive integers n for which n is divisible by 4, n + 1 is

divisible by 5, and n + 2 is divisible by 6.
How many of such integers n are smaller than 2018?
A) 16 B) 17 C) 18 D) 33 E) 34
7. A frog starts in the point at coordinates (0, 0) in the plane. He can make
three kinds of jumps:
• from (x, y) to (x, y − 5);
• from (x, y) to (x − 2, y + 3);
• from (x, y) to (x + 4, y + 9).
Ahead, there are three juicy snacks that the frog would like to eat: a worm
at (2013, 2018), a beetle at (2018, 2019), and a snail at (2018, 2023).
Which of these snacks can the frog reach?
A) worm and snail B) beetle and snail C) worm and beetle
D) only the beetle E) only the snail
8. Harold draws a trapezium with parallel top and A

bottom sides. The length of the top side is smal- C
ler than the length of the bottom side. The two D
diagonals divide the trapezium into four triangles. B
The area of the upper triangle is called A, of the
lower B, of the left C, and of the right D. An example of such a trapezium
is depicted on the right.
Which of the following equalities holds for any such trapezium?
A) A + C = B + D B) A + D = B + C C) A + B = C + D
D) A : B = D : C E) A : C = D : B
4
B-problems
1. Three years ago, Rosa’s mother was exactly five times as old as Rosa was
at that time. At that moment, Rosa’s mother was just as old as Rosa’s
grandmother was when Rosa’s mother was born. Now, Rosa’s grandmother
is exactly seven times as old as Rosa is.
How old is Rosa’s mother now?
2. Nanda and Mike both have a note containing the same five-digit number.
Nanda puts the digit 4 in front and the digit 8 at the end of her number to
obtain a seven-digit number. Mike puts one digit in front of his number.
Comparing their new numbers, Nanda’s number turns out to be exactly 6
times as large as Mike’s.
What was their starting number?
3. We consider a square, the circle through the vertices of the square and the
circle touching the four sides of the square (see the left-hand figure). The
ring-shaped area between the two circles is divided into four dark pieces
(inside the square) and four light pieces (outside the square). The area of
the square is 60.
What is the total area of two dark pieces and one light piece together as
depicted in the right-hand figure?
4. Elisa is making so-called dubious dice. Each face of a dubious die contains
one of the numbers 1 to 6, but not all these numbers need to occur and
some may occur more than once. However, from every direction it must
look like a real die. This means that in each corner three different numbers
must meet, no two of which add up to 7 (on a real die such pairs are always
on opposite faces). For example, the numbers 1, 2, and 4 may meet in a
corner, but 1, 2, and 5 may not as 2 + 5 = 7. Of course, a normal die is an
example of a dubious die as well.
Elisa is interested in the sum of the six numbers on a dubious die.
How many possible values are there for this sum?
5
Solutions
A-problems
1. B) 4 5. E) 4
2. C) C 6. E) 34
3. B) 10 7. E) only the snail
4. D) 7 8. E) A : C = D : B
B-problems
1. 33 years
2. 49998
3. 15
4. 19
6
Problems
B-problems
B1. Anouk, Bart, Celine, and Daan have participated in a math competition.
Each of their scores is a positive integer. The sum of the scores of Bart
and Daan is the same as the sum of the scores of Anouk and Celine. The
sum of the scores of Anouk and Bart is higher than the sum of the scores
of Celine and Daan. Daan’s score is higher than the sum of the scores of
Bart and Celine.
Write down the names of the four students in decreasing order of their
scores.
B2. In the figure, you can see a triangle ABC. A

The angle at B is equal to 63 degrees. Side
AC contains a point D, and side BC con- D
tains a point E. Points B, A, and D lie on 63◦ E C
B
a circle with centre E. Points E and C lie
on a circle with centre D.
What is the angle at point C?
B3. A palindromic number is a positive integer (consisting of one or more digits)

that remains the same when the digits are reversed. For example: 1245421
and 333 are palindromic numbers, but 345 and 100 are not. There is exactly
one palindromic number n with the following property: if you subtract
2018 from n, the result is again a palindromic number.
What number is n?
C
B4. Triangle ABC is isosceles with apex C. The midpoint of

E
AB is point M . On segment CM there is a point D such
|CD| 3 D
that |DM | = 2 . Line BD intersects segment AC in point E.
|CE|
Determine |EA| .
A M B
7
B5. A sawtooth number is a positive integer with the following property: for any
three adjacent digits, the one in the middle is either greater than its two
neighbours or smaller than its two neighbours. For example, the numbers
352723 and 314 are sawtooth numbers, but 3422 and 1243 are not.
How many 8-digit sawtooth numbers exist, for which each of the digits is
equal to 1, 2, or 3?
C1. You have n balls that are numbered from 1 to n. You need to distribute
the balls over two boxes. The value of a box is the sum of the numbers of
the balls in that box. Your distribution of balls must obey the following
rules:
• Each box has at least one ball.
• The two boxes do not have the same number of balls.
• The value of the box with the least number of balls must be at least 2
more than the value of the box with the most balls.
Determine for which positive integers n this is possible.
(Prove that for those values of n it is indeed possible, and prove that it is
not possible for other values of n.)
C2. In this problem we consider squares: numbers of the form m2 where m is

an integer.
(a) Does there exist an integer a such that 16 + a, 3 + a, and 16 · 3 + a
are squares?
If so, give such a number a and show that the three numbers are
indeed squares.
If not, prove that such a number a does not exist.
(b) Does there exist an integer a such that 20 + a, 18 + a, and 20 · 18 + a
are squares?
If so, give such a number a and show that the three numbers are
indeed squares.
If not, prove that such a number a does not exist.
(c) Prove that for every odd integer n there exists an integer a such that
2018 + a, n + a, and 2018 · n + a are squares.
8
Solutions
B-problems
1. A, D, B, C
2. 18 degrees
3. 2442
3
4. 4
5. 110
C-problems
C1. A distribution of the balls that follows all three rules will be called a correct
distribution. The box containing the larger number of balls will be called
the fullest box. It is clear that at least 1 + 2 = 3 balls are needed to obey
the first two rules.
We first consider the case that n is odd.
If n = 3, then the fullest box must have two balls. It therefore has a value
of at least 1 + 2 = 3, while the other box has one ball and hence a value of
at most 3. The third rule is broken so there is no correct distribution.
For n = 5 there is a correct distribution: put balls 1, 2, and 3 in one box
and put balls 4 and 5 in the other box. Since 4 + 5 is at least two more
than 1 + 2 + 3, this is indeed a correct distribution.
If there is a correct distribution for n balls, there is also one for n + 2 balls.
Indeed, we can simply add ball n + 1 to the fullest box and add ball n + 2
to the other box. The value of the fullest box increases by less than the
other box, hence we still follow rule 3.
Since we have a correct distribution for n = 5, we also have a correct
distribution for n = 7. Then, we also find a correct distribution for n = 9,
n = 11, et cetera.
Now we consider the case that n is even.
If n = 4, the fullest box must have at least three balls and hence has a
value of at least 1 + 2 + 3 = 6. The other box has only one ball and has a
9
value of at most 4. This means that the third rule is broken. There is no
correct distribution.
If n = 6, the fullest box must have at least four balls. It therefore has a
value of at least 1 + 2 + 3 + 4 = 10. The other box has at most two balls
and therefore has a value of at most 5 + 6 = 11. Since 11 < 2 + 10, there is
no correct distribution.
For n = 8, there is a correct distribution: put balls 1 to 5 into one box
and put balls 6 to 8 in the other box. The fullest box then has a value of
1 + 2 + 3 + 4 + 5 = 15, and the other box has a value of 6 + 7 + 8 = 21.
Since 21 ≥ 2 + 15, this is indeed a correct distribution.
Again, we see that a correct distribution of n balls gives a correct distribution
with n + 2 balls by adding ball n + 1 to the fullest box and adding ball
n + 2 to the other box. Since we have a correct distribution for n = 8, we
thus find correct distributions for n = 10, 12, 14, . . ..
We conclude: for n = 1, 2, 3, 4, 6 there is no correct distribution, but for all
other positive integers n a correct distribution does exist.
C2. (a) Yes, such a number a exists. For example, take a = 33. Then we have:
16 + a = 72 , 3 + a = 62 , and 16 · 3 + a = 92 .
A suitable a can be easily found by trying 48 + a = 49, 64, 81.
(b) No, such a number a does not exist. The numbers 20 + a and 18 + a
cannot both be squares since the difference of two squares is never
equal to 2. Indeed, suppose that m2 − n2 = 2. Then we would have
(m + n)(m − n) = 2, where m > n. Since 1 and 2 are the only positive
divisors of 2, we would have m + n = 2 and m − n = 1. But this would
imply that 2m = (m + n) + (m − n) = 3, which is impossible since
2m is even whereas 3 is odd.
(c) Given an odd integer n, we will show that there exist integers a, x, y,
and z such that
2018 + a = x2 , (1)
2
n+a = y , (2)
2018n + a = z2. (3)
If we subtract equation (2) from equation (1), we obtain
2018 − n = x2 − y 2 = (x − y)(x + y).
10
We therefore choose x and y such that x + y = 2018 − n and x − y = 1.
Addition of these two equations gives us 2x = 2019 − n, subtraction
gives us 2y = 2017 − n. We therefore choose
2019 − n 2017 − n
x= and y= .
2 2
Since n is odd, these are integers. Indeed, we now have x − y = 1 and
x + y = 2018 − n, and hence that
2018 − n = x2 − y 2 . (4)
We let a = y 2 − n. Then equation (2) certainly holds. Equation (1)

holds as well, since 2018 + a = 2018 + y 2 − n = x2 (the second equation
follows from (4)).
Finally, we show that 2018n + a is a square. We successively obtain
2018n + a = 2018n + y 2 − n
(2017 − n)2
= 2017n +
4
4 · 2017n + 20172 − 2 · 2017n + n2
=
4
20172 + 2 · 2017n + n2
=
4
2
(2017 + n)2

2017 + n
= = .
4 2
Since n is odd, the number 2017+n 2 is an integer. By choosing z =

2017+n
2 , we have found integers a, x, y, and z for which equations (1),
(2), and (3) are true.
11
Problems
1. We call a positive integer a shuffle number if the following hold:

(1) All digits are nonzero.
(2) The number is divisible by 11.
(3) The number is divisible by 12. If you put the digits in any other order,
you again have a number that is divisible by 12.
How many 10-digit shuffle numbers are there?
2. The numbers 1 to 15 are each coloured blue or red. Determine all possible
colourings that satisfy the following rules:
• The number 15 is red.

• If numbers x and y have different colours and x + y ≤ 15, then x + y
is blue.
• If numbers x and y have different colours and x · y ≤ 15, then x · y is
red.
3. Determine all triples (x, y, z) consisting of three distinct real numbers, that
satisfy the following system of equations:
x2 + y 2 = −x + 3y + z,
2 2
y +z = x + 3y − z,
2 2
x +z = 2x + 2y − z.
12
4. In triangle ABC, ∠A is smaller than ∠C. Point D lies on the (extended)
line BC (with B between C and D) such that |BD| = |AB|. Point E lies
on the bisector of ∠ABC such that ∠BAE = ∠ACB. Line segment BE
intersects line segment AC in point F . Point G lies on line segment AD
such that EG and BC are parallel.
Prove that |AG| = |BF |.
C
E
F
5. At a quiz show there are three doors. Behind exactly one of the doors, a
prize is hidden. You may ask the quizmaster whether the prize is behind
the left-hand door. You may also ask whether the prize is behind the
right-hand door. You may ask each of these two questions multiple times,
in any order that you like. Each time, the quizmaster will answer ‘yes’
or ‘no’. The quizmaster is allowed to lie at most 10 times. You have to
announce in advance how many questions you will be asking (but which
questions you will ask may depend on the answers of the quizmaster).
What is the smallest number you can announce, such that you can still
determine with absolute certainty the door behind which the prize is hidden?
13
Solutions
1. First observe that a shuffle number can only contain the digits 2, 4, 6, and
8. Indeed, if we place the digits in any order, we obtain an even number
(since it is divisible by 12) because of property (3). Since the last digit of
an even number is also even, the digit must be 2, 4, 6, or 8 since it cannot
be 0 because of property (1). Since we can put any of the digits in the last
position, this holds for each digit of a shuffle number.
Next, observe that a shuffle number can only contain the digits 4 and 8.
Indeed, suppose that we had a shuffle number containing the digit 2. We
could reorder the digits so that the last digit is 2. The last two digits would
then be 22, 42, 62, or 82. But then the number would not be divisible by
4 (and hence also not divisible by 12), contradicting property (3). In the
same way we see that a shuffle number cannot contain digit 6 because a
number ending in digits 26, 46, 66, or 86 is not divisible by 4.
A shuffle number is divisible by 3 (since it is divisible by 12), hence
the sum of its digits is divisible by 3 as well. If our 10-digit number
has k eights and 10 − k fours, then the sum of its digits is equal to
8k + 4(10 − k) = 40 + 4k = 36 + 4(k + 1). This is divisible by 3 if and only
if k + 1 is divisible by 3. That is, if k = 2, k = 5, or k = 8. We see that a
shuffle number must have 2 eights and 8 fours, or 5 eights and 5 fours, or 8
eights and 2 fours. Note that each of those numbers satisfy (1) and (3). It
remains to examine which of these numbers are divisible by 11 (prop. (2)).
For this we use the 11-criterion: a number is divisible by 11 if the alternating
sum of the digits is divisible by 11. By alternating sum we mean that
instead of adding them, we alternately add and subtract. Since in our
case all digits are equal to 4 or to 8, the alternating sum must even be
divisible by 4 × 11 = 44. Since the alternating sum cannot be greater than
8 + 8 + 8 + 8 + 8 − 4 − 4 − 4 − 4 − 4 = 20 (and cannot be smaller than −20),
it must be equal to 0. In other words: the sum of the five digits in the odd
positions must equal the sum of the five digits in the even positions. This
means that the number of digits 8 in the odd positions must be equal to
the number of digits 8 in the even positions. We examine the three cases
that we found before:
• Suppose exactly 2 digits are eights. The only requirement is now that
there is exactly 1 eight in the odd positions and exactly 1 eight in the
even positions. There are 5 × 5 = 25 ways to do this.
• Suppose exactly 5 digits are eights. Since we cannot divide the eights
into two equal groups, there are no solutions in this case.
14
• Suppose exactly 8 digits are eights. The only requirement is now that
we have 4 eights in the odd positions and 4 eights in the even positions.
In other words: 1 odd position must be a four and 1 even position
must be a four. Again we find 5 × 5 = 25 possibilities.
We conclude that the total number of 10-digit shuffle numbers is 25+25 = 50.
2. We first consider the case that 1 is red. Then all numbers from 1 to 15
are red. Indeed, suppose that some number k is blue. Then 1 and k have
different colours, hence by the third rule the number 1 · k = k must be
red. But this contradicts the assumption that it was blue. Observe that
colouring all numbers red indeed gives a correct colouring.
Now we consider the case that 1 is blue. Observe that when two numbers
sum to 15, those numbers must have the same colour by the second rule.
We get the following pairs of numbers of the same colour: 1 and 14, 2 and
13, 3 and 12, 4 and 11, 5 and 10, 6 and 9, 7 and 8.
The number 2 is blue. Indeed, suppose that 2 is red. From the second rule,
we derive that 3 = 1 + 2 is blue. Repeatedly applying the same rule we find
that 5 = 3 + 2 is blue, that 7 = 5 + 2 is blue, and finally that 15 is blue.
Since 15 is in fact not blue, 2 cannot be red.
The number 7 is blue. Indeed, suppose that 7 is red. It then follows from
the second rule that 8 = 1 + 7 is blue. However, 7 and 8 must have the
same colour, so this cannot be the case.
The number 4 is blue. Indeed, suppose that 4 is red. It then follows from
the second rule that 11 = 4 + 7 is blue. But 4 and 11 have the same colour,
so this cannot be the case.
Recall, that 3 and 12 have the same colour, and so do 6 and 9. In fact,
all four numbers must have the same colour. Indeed, otherwise 9 = 3 + 6
would be blue by the second rule and also 12 = 3 + 9 would be blue by the
second rule.
So far, we know that 15 is red, that 1, 2, 4, 7, 8, 11, 13, 14 are blue, that
3, 6, 9, 12 have the same colour, and that 5, 10 have the same colour. The
numbers 3 and 5 cannot both be red, since if 3 is red, 5 = 2 + 3 is blue by
the second rule. The three remaining colour combinations for the numbers
3 and 5 result in the following three colourings:
(1) Only 15 is red.
(2) Only 5, 10, 15 (the numbers divisible by 5) are red.
(3) Only 3, 6, 9, 12, 15 (the numbers divisible by 3) are red.
15
It is easy to check that all three colourings are indeed correct. We write this
out for the third colouring, the other two can be checked in a similar way.
That the sum of a red and a blue number is always blue follows from the
fact that the sum of a number divisible by 3 and a number not divisible by
3 is itself not divisible by 3. That the product of a blue and a red number
is always red follows from the fact that the product of a number divisible
by 3 and a number not divisible by 3 is itself divisible by 3.
We conclude that there are 4 correct colourings in total: the colouring in
which all numbers are red, and colourings (1), (2), and (3).
3. If we subtract the second equation from the first, we obtain
x2 − z 2 = −x + z − (x − z),
which can be rewritten as
(x − z)(x + z) = −2(x − z),
and then simplified further to get
(x − z)(x + z + 2) = 0.
Since x and z must be different, x − z is nonzero and we can conclude that

x + z + 2 = 0, hence z = −x − 2. If we subtract the third equation from
the second, we obtain
y 2 − x2 = x + 3y − (2x + 2y),
which can be rewritten as
(y − x)(y + x) = 1(y − x),
and then simplified further to get
(y − x)(y + x − 1) = 0.
Since x and y must be different, y − x is nonzero and we can conclude that

y + x − 1 = 0, hence y = 1 − x.
If we now substitute y = 1 − x and z = −x − 2 in the first equation, we get
x2 + (1 − x)2 = −x + 3(1 − x) + (−x − 2),
which we can rewrite as
2x2 − 2x + 1 = −5x + 1.
16
This is a quadratic equation 2x2 + 3x = 0, whose roots are x = 0 and
x = − 32 . In both cases we can deduce the values of y and z from the formulas
y = 1 − x and z = −x − 2. This gives the solutions (x, y, z) = (− 32 , 25 , − 21 )
and (x, y, z) = (0, 1, −2).
For these solutions we know that they satisfy the first equation. Since the
difference between the first and the second equation gives zero on both
sides (since we chose z = −x − 2), the second equation is satisfied as well.
Because of the choice y = 1 − x, also the third equation is satisfied. We
conclude that both solutions indeed satisfy all three equations.
4. Because BE is the bisector of ∠ABC, it follows that ∠ABE = ∠CBF .

It was given that ∠F CB = ∠EAB. Since the angles in a triangle sum
to 180◦ , both in triangle ABE and in triangle CBF , we must also have
∠BF C = ∠BEA. Since ∠BF C and ∠AF E are opposite angles, we see
that ∠AF E = ∠F EA, and hence that triangle AEF is isosceles.
C
E
F
B
A
G
We will show that 4ABF and 4EGA are congruent.

We see that ∠EGA = ∠CDA, since EG and BC are parallel. In the
isosceles triangle ABD, we see that ∠CDA = 12 (180◦ − ∠ABD), which is
equal to 12 ∠ABC (because of the straight angle). This in turn, is equal
to ∠ABF , because BF is the bisector of ∠ABC. Altogether, we find that
∠EGA = ∠ABF .
Triangle ABD is isosceles, so ∠BAG = ∠CDA. As we just observed, this
last angle is equal to ∠ABF . We have also seen that ∠EGA = ∠ABF .
Because the three angles in triangle GEA sum to 180◦ , we get
∠GEA = 180◦ −∠EGA−∠EAB−∠BAG = 180◦ −∠ABF −∠F CB−∠ABF.
Since 2 · ∠ABF = ∠ABC (since BF is the bisector) we thus find that
∠GEA = 180◦ − ∠F CB − ∠ABC = ∠BAC (because the angles in triangle
ABC sum to 180◦ ), hence ∠GEA = ∠BAF .
17
Using the fact that |AE| = |AF |, which follows from the fact that 4AEF
is isosceles, we can now conclude that 4ABF and 4EGA are congruent.
This immediately implies |AG| = |BF | as required.
5. The smallest number of questions that suffices is 32. First we will show a
strategy to locate the prize in only 32 questions.
Start by asking the quizmaster if the prize is behind the left-hand door.
Repeat this until you are sure whether the prize is there or not. You
can only be 100% sure when you have received the same answer 11 times,
because the quizmaster can lie a maximum of 10 times. After doing this,
suppose that the quizmaster has lied n times. This means that you have
thus far asked 11 + n questions and the quizmaster is entitled to 10 − n
more lies.
Now ask the quizmaster whether the prize is behind the right-hand door,
and keep asking this until you are 100% sure of the true answer. Since the
quizmaster can lie only 10 − n more times, you need at most 2(10 − n) + 1 =
20 − 2n + 1 questions for this. In total you have now asked 32 − n questions.
We conclude that with this strategy, 32 questions always suffice.
Having only 31 questions, it is not always possible to locate the prize. We
will show how the quizmaster can make sure of that. At the beginning,
as long as you have asked about both doors at most 10 times, he always
answers ‘no’. To keep it simple, we assume that the left-hand door is the
first door that you query for the eleventh time (the other case is completely
analogous). We consider the situation that the prize is not behind the left-
hand door (which may happen). In this case, we show that the quizmaster
can make sure that after 31 questions, you cannot tell whether the prize is
behind the middle door or behind the right-hand door.
So far, the quizmaster has been speaking the truth about the left-hand
door, and he will continue to do so: when asked about the left-hand door,
he will always answer ‘no’. For the right-hand door, the quizmaster will
keep saying ‘no’ up to and including the tenth time he is asked about this
door. The next ten times he is asked about this door, he will answer ‘yes’.
Since you ask at most 20 questions about the right-hand door (since you
already queried the left-hand door at least 11 times), the quizmaster needs
to lie at most 10 times. Whether the prize is behind the middle door or
behind the right-hand door, in both cases the quizmaster gives the same 31
answers. Therefore, you cannot determine the door behind which the prize
is located.
18
Problems
1. Prove that for each positive integer n there are at most two pairs (a, b) of
positive integers with following two properties:
(i) a2 + b = n,
(ii) a+b is a power of two, i.e. there is an integer k ≥ 0 such that a+b = 2k .
2. Let ABC be a triangle and let I be the incentre of this triangle. The line
through I perpendicular to AI intersects the circumcircle of 4ABC in
points P and Q, where P lies on the same side of AI as B. Let S be the
second intersection point of the circumcircles of 4BIP and 4CIQ. Prove
that SI is the angle bisector of ∠P SQ.
3. Let x and y be positive real numbers.
1. Prove: if x3 − y 3 ≥ 4x, then x2 > 2y.

2. Prove: if x5 − y 3 ≥ 2x, then x3 ≥ 2y.
4. Do there exist a positive integer k and a non-constant sequence a1 , a2 , a3 , . . .

of positive integers such that an = gcd(an+k , an+k+1 ) for all positive
integers n?
5. In a country, there are 2018 cities, some of which are connected by roads.
Each city is connected to at least three other cities. It is possible to travel
from any city to any other city using one or more roads. For each pair of
cities, consider the shortest route between these two cities. What is the
greatest number of roads that can be on such a shortest route?
19
Solutions
1. Suppose there are three or more of such pairs for one particular n. Then
by the pigeonhole principle, there are at least two such pairs, say (a, b) and
(c, d), with a ≡ c mod 2. Write a + b = 2k and c + d = 2` where k and `
are positive integers. Without loss of generality, assume that ` ≥ k. We
see that
2` − 2k = (c + d) − (a + b)
= (c + n − c2 ) − (a + n − a2 ) = c − a − c2 + a2
= (a + c − 1)(a − c).
As ` ≥ k, we have that 2k is a divisor of 2` − 2k . Because a ≡ c mod 2,
the integer a + c − 1 is odd, hence 2k is a divisor of a − c. As 2` − 2k ≥ 0
and a + c − 1 > 0, we have a − c ≥ 0. On the other hand, we have
a − c < a + b = 2k , hence 0 ≤ a − c < 2k , while 2k | a − c. We conclude
that a − c = 0, hence a = c. Now we have b = n − a2 = n − c2 = d, hence
the two pairs (a, b) and (c, d) are identical, which gives a contradiction.
Therefore, there at most two pairs (a, b) having the two properties.
2. Write ∠CAB = 2α, ∠ABC = 2β, and ∠BCA = 2γ. As the angles in
triangle 4AIB add up to 180◦ , we have
∠BIA = 180◦ − ∠IAB − ∠ABI = 180◦ − α − β = 90◦ + γ,
hence
∠BIP = ∠BIA − ∠AIP = 90◦ + γ − 90◦ = γ = ∠BCI.
Moreover, as BP QC is a cyclic quadrilateral, we have
180◦ − ∠BP I = 180◦ − ∠BP Q = ∠BCQ = ∠BCI + ∠ICQ.
If we now consider the sum of the angles in triangle BP I, and use both
previous results, we find
∠P BI = 180◦ − ∠BP I − ∠BIP = ∠BCI + ∠ICQ − ∠BCI = ∠ICQ.
Using the fact that BSIP and CQIS are cyclic quadrilaterals, we get
∠P SI = ∠P BI = ∠ICQ = ∠ISQ,
hence SI is the angle bisector of angle P SQ.
3. 1. We have x3 − 4x ≥ y 3 > 0, hence x(x2 − 4) > 0. As x is positive, this

yields x2 − 4 > 0, hence x2 > 4. That means that x > 2. Moreover,
we have x3 − y 3 ≥ 4x > 0, hence x > y. The combination of these
two results (which is allowed because x and y are both positive) gives
x2 = x · x > 2 · y = 2y.
2. We have (x4 − 4)2 ≥ 0. Expanding yields x8 − 8x4 + 16 ≥ 0. Because x
is positive, we can multiply this with x without changing the inequality
sign, hence we have x9 ≥ 8x5 − 16x. The inequality in the assumption
gives x5 − 2x ≥ y 3 . If we combine this with the preceding inequality,
we get x9 ≥ 8(x5 − 2x) ≥ 8y 3 . It follows that x3 ≥ 2y.
4. Such a k and a sequence do not exist. We prove this by contradiction,

so suppose they do exist. Note that an | an+k and an | an+k+1 for all
n ≥ 1. Using simple induction, it follows that an | an+`k and an | an+`k+`
for all ` ≥ 0. We will prove by induction to m that an | an+mk+(m+1)
for all 0 ≤ m ≤ k − 1. For m = k − 1, this follows from an | an+k·k =
an+(k−1)k+k . Now suppose that for a certain m with 1 ≤ m ≤ k −
1 we have that an | an+mk+(m+1) . We also know that an | an+mk+m .
Therefore, as an+(m−1)k+m = gcd(an+mk+m , an+mk+m+1 ), we also have
an | an+(m−1)k+m . This finishes the induction argument. Substituting
m = 0 yields an | an+1 .
Because an | an+1 , we also have gcd(an , an+1 ) = an for all n. Hence,
an = an−k for all n ≥ k + 1. Now we have an−k | an−k+1 | an−k+2 | . . . |
an = an−k . Because these are all positive integers, an−k , an−k+1 , . . . , an
must all be equal. This must be true for all n ≥ k + 1, hence the sequence
is constant, which gives a contradiction.
21
5. The greatest number of roads that can be in such a shortest route, is
1511. We first describe a country for which this number is attained. Divide
the cities in 504 groups: two groups of five cities (A0 , B0 , C0 , D0 , E0 ) and
(A503 , B503 , C503 , D503 , E503 ), and groups of four cities (Ai , Bi , Ci , Di ) for
1 ≤ i ≤ 502. For all i with 0 ≤ i ≤ 503: connect Ai to Bi , connect Ai to
Ci , connect Bi to Di , and connect Ci to Di . Moreover, connect E0 to A0 ,
B0 , and C0 . Connect E503 to B503 , C503 , and D503 . For each 1 ≤ i ≤ 502:
connect Bi to Ci . Finally, for each 0 ≤ i ≤ 502: connect Di to Ai+1 . Now
every city is connect two exactly three other cities.
If we now want to travel from A0 to D503 , then we have to travel through
all Ai and Di , because the only connections between the different groups
are there, and each group is only connected to the previous and the next
group. Moreover, for 0 ≤ i ≤ 503, the city Ai is not connected to Di , which
causes the route within group i to go through either Bi or Ci . At least
three of the four or five cities of each group must lie on the route. In total,
this route visits at least 3 · 504 = 1512 cities and has at least 1511 roads.
We will now prove that the shortest route between two cities cannot have
more than 1511 roads. Consider such a shortest route which visits the cities
A0 , A1 , . . . , Ak consecutively. Each of the cities Ai has one more neighbour
besides Ai−1 and Ai+1 , which we will call Bi (note that the cities B0 , B1 ,
. . . , Bk do not have to be distinct). Moreover, A0 and Ak are connected
to a third city, say C0 6= B0 and Ck 6= Bk respectively. If one of the cities
Bi or Ci equals one of the cities Aj , then we could have found a shorter
route by going directly from Ai to Aj (or vice versa), which would be a
contradiction. Hence, the cities Bi and Ci are not equal to any of the cities
Aj .
If one of the cities Bi is connected to four cities Aj , say Bi is connected to
Ai , Am , An , and Ap with i < m < n < p, then we could shorten the route
by going from Ai to Bi and then to Ap . This makes us skip at least two
cities of the original route (Am and An ) and in their place we only visit Bi ,
hence this new route is shorter, which would be a contradiction. Therefore,
within the cities Bi with 3 ≤ i ≤ k − 3 there are at least k−5 3 distinct cities.
For B0 and C0 , the route can only be shortened if one of these cities equals
Bi for certain i ≥ 3. That would be a contradiction, hence B0 and C0 are
two distinct cities, not equal to Bi for i ≥ 3. In the same way, Bk and
Ck are two distinct cities, not equal to Bi for i ≤ k − 3. Altogether, we
have k + 1 cities on the route itself and at least k−5 3 + 2 + 2 other cities.
Therefore, k−53 + k + 5 ≤ 2018, hence 4k − 5 + 15 ≤ 3 · 2018 = 6054, hence
4k ≤ 6044, hence k ≤ 1511.
We conclude that the greatest number of roads occurring in a shortest route
is 1511.
22
Problems
1. Let P (x) be a quadratic polynomial with two distinct real roots. For all
real numbers a and b satisfying |a|, |b| ≥ 2017, we have P (a2 + b2 ) ≥ P (2ab).
Show that at least one of the roots of P is negative.
2. Write Sn for the set {1, 2, . . . , n}. Determine all positive integers n for
which there exist functions f : Sn → Sn and g : Sn → Sn such that for
every x exactly one of the equalities f (g(x)) = x and g(f (x)) = x holds.
3. Let n be a positive integer. Determine the maximum value of gcd(a, b) +

gcd(b, c) + gcd(c, a) for positive integers a, b, c such that a + b + c = 5n.
4. We are given a triangle ABC. On edge AC there are points D and E such
that the order of points on this line is A, E, D, C. The line through E
parallel to BC intersects the circumcircle of 4ABD in a point F , with E
and F lying on opposite sides of AB. The line through E parallel to AB
intersects the circumcircle of 4BCD in a point G, with E and G lying on
opposite sides of BC. Prove that DEF G is a cyclic quadrilateral.
23
Solutions
1. Write P (x) = c(x − d)(x − e), where d and e are the distinct roots of P ,
so d =6 e. We also have c 6= 0, since otherwise P is not quadratic. We
distinguish two cases. First suppose that c > 0. Substituting b = −a =
−2017 in P (a2 + b2 ) ≥ P (2ab), we obtain P (2a2 ) ≥ P (−2a2 ), and therefore
c(2a2 − d)(2a2 − e) ≥ c(−2a2 − d)(−2a2 − 2).
Dividing both sides by the positive real number c, expanding, and then
canceling the terms 4a4 and de, we get
−(d + e) · 2a2 ≥ (d + e) · 2a2 ,
or equivalently,
4a2 (d + e) ≤ 0.
Dividing by 4a2 = 4 · 20172 , we see that d + e ≤ 0. As the roots d and e
are distinct, one of them must be negative.
Now consider the remaining case c < 0. Then the graph of P opens up
to the bottom, and therefore is descending on the right of the apex of the
parabola. Choose any a 6= b with a, b ≥ 2017 and a, b both to the right
side of the apex. Then by AM-GM we have a2 + b2 > 2ab. Since a, b are
positive, at least 1, and to the right of the apex, we have 2ab > a, so 2ab
(and therefore a2 + b2 as well) also lies to the right of the apex. Since P
is descending on the right side of the apex, we have P (a2 + b2 ) < P (2ab),
contrary to the given condition. We deduce that the case c < 0 is impossible
and therefore that c > 0, from which we conclude that at least one of the
roots of P is negative.
2. We first show that if n = 2m for some positive integer m, such functions

exist. Define
( (
x if 1 ≤ x ≤ m, x + m if 1 ≤ x ≤ m,
f (x) = g(x) =
x − m if m + 1 ≤ x ≤ 2m, x if m + 1 ≤ x ≤ 2m.
First note that all values lie in Sn , so these indeed define functions Sn →
Sn . The range of f is equal to {1, 2, . . . , m} and that of g is equal to
{m + 1, m + 2, . . . , 2m}. So f (g(x)) 6= x if x ≥ m + 1 and g(f (x)) 6= x if
x ≤ m. Moreover, for x ≤ m we have f (g(x)) = f (x + m) = x + m − m = x,
and for x ≥ m + 1 we have g(f (x)) = g(x − m) = x − m + m = x. Therefore
for every x ∈ Sn , exactly one of f (g(x)) = x and g(f (x)) = x holds.
24
Now suppose that n is odd, say equal to 2m + 1, and suppose that f and
g are functions satisfying the given conditions. Without loss of generality,
we assume that f (g(x)) = x for (at least) m + 1 elements of Sn , say for
x = x1 , . . . , xm+1 . Suppose that for some i, j with i ≤ i, j ≤ m + 1 we
have g(xi ) = xj . Then f (xj ) = f (g(xi )) = xi , so g(f (xj )) = g(xi ) = xj .
But now we have both f (g(x)) = x and g(f (x)) = x for x = xj , which
gives a contradiction. So we see that for all i with 1 ≤ i ≤ m + 1 the
number g(xi ) is not equal to any of the xj with 1 ≤ j ≤ m + 1. Since
there are m numbers in Sn not equal to any of the xj with 1 ≤ j ≤ m + 1,
and since there are m + 1 possible values of i, it follows that two of the
g(xi ) must be equal, say g(xk ) = g(xl ) with 1 ≤ k < l ≤ m + 1. But then
xk = f (g(xk )) = f (g(xl )) = xl , which gives a contradiction. Therefore no
such f and g can exist if n is odd.
Hence n satisfies the given conditions if and only of n is even.
3. Write G = gcd(a, b) + gcd(b, c) + gcd(c, a). Without loss of generality we

assume that a ≤ b ≤ c. Then we have gcd(a, b) ≤ a, gcd(b, c) ≤ b, and
gcd(c, a) ≤ a. Hence
G ≤ a + b + a ≤ a + b + c ≤ 5n.
If 3 | n, then we can achieve G = 5n via a = b = c = 53 n. Since all three

gcd’s equal 53 n, we see that G = 3 · 53 n = 5n.
So suppose that 3 - n. Then a, b, and c cannot all be equal to each other.
We distinguish a number of cases.
Case 1a. b = c and b ≤ 2n. Since in this case we have a = 6 b, it follows that
gcd(a, b) = gcd(a, b − a) ≤ b − a. Hence G ≤ (b − a) + b + a = 2b ≤ 4n.
Case 1b. b = c and b > 2n. We have a + 2b = 5n, so a = 5n − 2b, from which we
deduce that G ≤ a + b + a = 10n − 4b + b = 10n − 3b < 10n − 6n = 4n.
Case 2a. b 6= c and c−a ≥ n. Now we have G ≤ a+b+a = (a+b+c)−(c−a) =
5n − (c − a) ≤ 5n − n = 4n.
6 c and c − a < n. Since a ≤ b ≤ c we now also have c − b < n.
Case 2b. b =
Moreover, we have b 6= c and therefore a 6= c. So gcd(c, a) = gcd(c −
a, a) ≤ c − a < n and gcd(b, c) = gcd(b, c − b) ≤ c − b < n. Also note
that a ≤ 53 n. Therefore G ≤ 53 n + n + n < 4n.
So in all cases we have G ≤ 4n. We can achieve G = 4n via a = n and
b = c = 2n, since then we have gcd(a, b) = gcd(c, a) = n and gcd(b, c) = 2n.
Therefore the maximum value of G is 5n if 3 | n and 4n if 3 - n.
25
4. Let G0 be the intersection of the line BF
and the circumcircle 4BCD. We show that
DEF G0 is a cyclic quadrilateral and that G =
G0 .
Since EF k BC we have 180◦ − ∠DEF =
∠AEF = ∠ACB, and since BDCG0 is a cyc-
lic quadrilateral we have ∠ABC = ∠DCB =
∠DG0 B = ∠DG0 F , therefore 180◦ −∠DEF =
∠DG0 F . Therefore DEF G0 is a cyclic quad-
rilateral. It follows that we have ∠DEG0 =
∠DF G0 = ∠DF B. Since ADBF is a cyclic quadrilateral we have ∠DF B =
∠DAB, so ∠DEG0 = ∠DAB, and therefore EG0 k AB. As G0 lies on the
circumcircle of 4BCD, we deduce that G = G0 .
26
Problems
1. In each of the different grades of a high school there are an odd number of
pupils. Each pupil has a best friend (who possibly is in a different grade).
Everyone is the best friend of their best friend. In the upcoming school
trip, every pupil goes to either Rome or Paris. Show that the pupils can be
distributed over the two destinations in such a way that
(i) every student goes to the same destination as their best friend;
(ii) for each grade the absolute difference between the number of pupils
that are going to Rome and that of those who are going to Paris is
equal to 1.
2. Determine all 4-tuples (a, b, c, d) of positive real numbers satisfying a + b +

c + d = 1 and
2 2 2 2
a b c d
max , · max , = (min(a + b, c + d))4 .
b a d c
3. Let ABC be an acute triangle with circumcentre O. Point Q lies on the

circumcircle of 4BOC and OQ is a diameter of this circle. Point M lies
on CQ and point N lies on the interior of the line segment BC in such a
way that AN CM is a parallelogram. Show that the circumcircle of 4BOC
and the lines AQ and N M are concurrent.
4. Find all functions f : Z → Z satisfying

• f (p) > 0 for all prime numbers p,
f (p)
• p | f (x) + f (p) − x for all x ∈ Z and all prime numbers p.
27
Solutions
1. We call a pupil a singleton if their best friend is in a different grade.
First give all singletons a destination as follows. Start by choosing any
singleton A1 and send them to Paris together with their best friend A2 .
Suppose the destinations for pupils A2m+1 , A2m+2 , . . . , A2k , with m ≤ 0
and k ≥ 1 are chosen. If possible, first choose in the grade of A2k a
singleton A2k+1 who doesn’t have a destination yet, and send them to the
city different from the destination of A2k , together with their best friend
A2k+2 . If possible, then choose in the grade of A2m+1 a singleton A2m
who doesn’t have a destination yet, and send them to the city different
from the destination of A2m+1 , together with their best friend A2m−1 .
Continue this process until pupils A2m+1 , A2m+2 , . . . , A2k with m ≤ 0 and
k ≥ 1 have destinations assigned to them and there are no singletons in the
grades of A2m+1 and A2k who don’t have a destination yet. These pupils
A2m+1 , A2m+2 , . . . , A2k are by construction such that A2i−1 and A2i are
best friends and A2i and A2i+1 have different destinations.
In every grade except that of A2m+1 and A2k the same number of pupils
is assigned to each destination. In the grades of A2m+1 and A2k (which
are different: otherwise no other singletons are in this grade, and all other
pupils form pairs of best friends, making the number of pupils in that class
even; contradiction!) one destination is assigned one more pupil than the
other, and these grade no longer have any singletons without destination.
Repeat this process until every singleton has a destination. All remaining
pupils now are in the same grade as their best friends, so every grade has
an even number of pupils without destination and therefore an odd number
of pupils with destination. So every grade contains a pupil on an end of
exactly one of the sequences of pupils (which then must be the last time a
pupil of this grade occurs in such a sequence). Therefore for each grade,
one destination is assigned exactly one more pupil of that grade than the
other.
Now consider any grade, and suppose that the number of singletons assigned
to Paris is one more than that assigned to Rome. Give the pupils in this
grade whose best friends are also in this grade a destination as follows. If
the number of pairs of best friends is odd, first send one pair to Rome;
otherwise do nothing. From the remaining even number of pairs of best
friends, send half of them to Paris and half of them to Rome.
Now note that in the first step, if the number of pairs was odd, then Rome
is assigned one more pupil than Paris. Therefore in any grade, the absolute
difference of the number of students going to Paris and the number going
to Rome is equal to 1.
28
2. We assume that a + b ≤ c + d. As we can simultaneously interchange a, b
and c, d without changing the problem, this assumption will not give any
loss of generality. We also assume that a ≤ b and c ≤ d; no generality is
lost here as we can interchange a and b (resp. c and d) without changing
2 2
the problem. Now we have a3 ≤ b3 , so ab ≤ ba , and analogously we have
2 2
c d
d ≤ c . The given equation is now equivalent to
b2 d2
a · c = (a + b)4 .
2 2
As ba ≥ b and dc ≥ d, the left hand side is at least bd. We now show that
bd ≤ (a + b)4 , from which follows that equality must hold everywhere.
Since a ≤ b we have b ≥ 12 (a + b), and analogously we have d ≥ 12 (c + d).
Since c+d ≥ a+b, it follows that d ≥ 21 (a+b). Moreover, from a+b+c+d = 1
and a + b ≤ c + d we deduce that a + b ≤ 12 , so
bd ≤ 14 (a + b)2 ≤ (a + b)2 (a + b)2 = (a + b)4 .
Therefore, each inequality used in the above must actually be an equality.

2
Since we have used ba ≥ b it follows that a = b, and analogously that c = d,
and since we have used a + b ≤ 12 it follows that a + b = 12 . Therefore
a = b = c = d = 14 .
So the only possible solutions is (a, b, c, d) = ( 14 , 14 , 14 , 14 ). Substitution of
1
this candidate solution gives 16 on both sides of the equation, and therefore
actually is a solution.
Q
3. Let T be the intersection of the cir-
cumcircle of 4BOC with AQ. We
show that T lies on N M . Write α = C
∠BAC. Then we have ∠BOC =
M
2α by the inscribed angle theorem.
N
Since |OB| = |OC| and ∠OCQ = O
90◦ = ∠OBQ by Thales’s theorem, B
we have 4OBQ ∼ = 4OCQ. Hence
A
we have ∠COQ = ∠BOQ = α.

As AN CM is a parallelogram, we see that ∠AM C = 180◦ − ∠M CN =
∠QCN = ∠QCB = ∠QOB = α. Hence we have ∠CN A = ∠AM C = α.
Now we see that ∠QT B = ∠QOB = α = ∠CN A, from which follows that
∠AT B = 180◦ − ∠QT B = 180◦ − ∠CN A = ∠AN B. Hence AT N B is a
cyclic quadrilateral. Also note that AT CM is a cyclic quadrilateral since
∠AM C = α = ∠COQ = ∠CT Q = 180◦ − ∠AT C.
29
As AT CM is a cyclic quadrilateral, we have ∠AT M = ∠ACM = 180◦ −
∠QCB − ∠BCA, using supplementary angles at C. Recall that ∠QCB =
∠QOB = α = ∠CAB, so using the sum of the angles in triangle ABC,
we obtain ∠AT M = 180◦ − ∠CAB − ∠BCA = ∠ABC. Since AT N B is a
cyclic quadrilateral, we see that ∠ABC = ∠ABN = 180◦ − ∠AT N , so we
have ∠AT M = 180◦ − ∠AT N . Therefore M , T , and N are collinear, as we
have set out to prove.
4. Suppose that f is a function satisfying both relations.

f (p)
Substituting x = p gives p | 2f (p) − p and as p is prime, we have
p | 2f (p). So either p = 2 or p | f (p).
f (p)
Substituting x = 0 gives p | f (0) + f (p) − 0 and as p is prime, we
have p | f (0) + f (p). Since p | f (p) if p 6= 2, it follows that p | f (0) if p 6= 2.
So f (0) is divisible by infinitely many prime numbers and therefore must
be equal to 0. From 2 | f (0) + f (2) we now see that 2 | f (2), so p | f (p) for
all prime numbers p.
Now the second of the given relations translates to f (x)f (p) ≡ p mod p for
all integers x and prime numbers p. It follows that p | f (x) if and only if
p | x. Applying this observation to the case in which x is a prime number
q 6= p, then we see that f (q) is not divisible by any prime number p 6= q.
As f (q) > 0, we deduce that f (q) is a power of q. Fermat’s Little Theorem
states that for all prime numbers p and integers n we have np ≡ n mod p,
t
so we also have np ≡ n mod p for all non-negative integers t. As f (p) is
t
of the form p with t a non-negative integer for all prime numbers p, we
deduce from the second given relation that f (x) ≡ x mod p for all integers
x and prime numbers p. Thus p | f (x) − x for all integers x and all prime
numbers p.
Therefore, for any fixed x ∈ Z the integer f (x) − x is divisible by infinitely
many prime numbers and therefore equal to 0. Hence f (x) must be equal
to x for all integers x.
Now suppose that f (x) = x for all integers x. Then f (p) > 0 for all prime
numbers p and
f (p)
f (x) + f (p) − x = (x + p)p − x ≡ xp − x ≡ 0 mod p
for all integers x and all prime numbers p by Fermat’s Little Theorem.
It follows that f (x) = x is the unique function Z → Z satisfying both
relations.
30
Problems
1. In a cyclic quadrilateral ABCD, the intersection of the diagonals is called

E. A line through E, not equal to AC or BD, intersects AB in P and
BC in Q. The circle that is tangent to P Q in E and also goes through D,
intersects the circumcircle of ABCD another time in the point R. Prove
that B, P , R, and Q lie on a circle.
2. Let n be a positive integer. Prove that n2 + n + 1 cannot be written as

the
√ product of two positive integers of which the difference is smaller than
2 n.
3. Find all functions f : Z → Z satisfying the following two conditions:

(i) for all integers x we have f (f (x)) = x;
(ii) for all integers x and y such that x + y is odd, we have f (x) + f (y) ≥
x + y.
4. There are 300 participants to a mathematics competition. After the com-

petition some of the contestants play some games of chess. Each two
contestants play at most one game against each other. There are no three
contestants, such that each of them plays against each other. Determine
the maximum value of n for which it is possible to satisfy the following
conditions at the same time: each contestant plays at most n games of
chess, and for each m with 1 ≤ m ≤ n, there is a contestant playing exactly
m games of chess.
31
Solutions
1. We consider the configuration in which P lies on the interior of AB and

Q does not lie on the interior of BC; other configurations can be solved
analogously.
Because DBCR is cyclic, we have ∠QCR = 180◦ − ∠BCR = ∠BDR.

As EQ is tangent to the circle through E, D, and R we have ∠BDR =
∠EDR = ∠QER, hence ∠QCR = ∠QER. This implies that ERQC is a
cyclic quadrilateral. Analogously, EP AR is also cyclic.
We now get ∠P RQ = ∠P RE + ∠ERQ = ∠P AE + 180◦ − ∠ECQ =
∠BAE + ∠ECB. Because of the sum of the angles in 4ABC this equals
180◦ − ∠ABC = 180◦ − ∠P BQ. Hence ∠P RQ = 180◦ − ∠P BQ, and hence
BP RQ is a cyclic quadrilateral.
2. Suppose that a and b are √ positive integers satisfying ab = n2 + n + 1. We

will prove that |a − b| ≥ 2 n. Note that (a − b)2 ≥ 0. Adding 4ab to both
sides yields
(a + b)2 = (a − b)2 + 4ab ≥ 4ab = 4n2 + 4n + 4 > 4n2 + 4n + 1 = (2n + 1)2 .
As (a + b)2 is a square, we have (a + b)2 ≥ (2n + 2)2 . Then it follows that
(a − b)2 = (a + b)2 − 4ab ≥ (2n + 2)2 − 4(n2 + n + 1) = 4n
√
hence |a − b| ≥ 2 n.
32
3. The function f (x) = x for all x, satisfies the conditions. Now suppose that
f (x) = x does not hold for all x. As of (i), there is both a value of x for
which f (x) > x and a value of x for which f (x) < x. Let a ∈ Z be such
that f (a) < a and consider an arbitrary x with x 6≡ a mod 2. Then we
have f (x) + f (a) ≥ x + a, hence f (x) ≥ x + a − f (a) > x. Write w = f (x),
then we have f (w) = f (f (x)) = x < f (x) = w. If w 6≡ a mod 2, then we
would have f (w) > w, which gives a contradiction. Hence f (x) = w ≡ a
mod 2 and also f (x) 6≡ x mod 2.
Let x1 and x2 both be of different parity from a. Then they both do
not have the same parity as their function values, hence x1 + f (x2 ) and
x2 + f (x1 ) are odd. By substituting x = x1 , y = f (x2 ), and also x = x2 ,
y = f (x1 ), we find
f (x1 )+x2 = f (x1 )+f (f (x2 )) ≥ x1 +f (x2 ) = f (f (x1 ))+f (x2 ) ≥ f (x1 )+x2 .
Hence, equality must hold everywhere and f (x1 ) + x2 = x1 + f (x2 ). Fix

x1 and write c = f (x1 ) − x1 , then we have f (x2 ) = x2 + c for all x2 not
having the same parity as a. Moreover, we know that c must be odd. If
z = x2 + c, then z can take all possible values not having the same parity
as x2 (and thus having the same parity as a). We have
f (z) = f (x2 + c) = f (f (x2 )) = x2 = z − c.
Now write d = c if a is odd and d = −c if a is even, then the function

satisfies (
x + d if x is even,
f (x) =
x − d if x is odd.
We check functions of this shape for any arbitrary odd d. We see that f (x)
and x never have the same parity, hence f (f (x)) = x + d − d = x, which
means (i) is satisfied. Moreover, if x + y is odd, then x and y have different
parities, hence f (x) + f (y) = x + y + d − d = x + y ≥ x + y, which means
(ii) is satisfied as well.
We conclude that the functions of this form together with the function
f (x) = x for all x, are the solutions.
33
4. We will prove that the maximum value of n is 200. We will first give an
example with n = 200. Consider players A1 , A2 , . . . , A200 and B1 , B2 ,
. . . , B100 . These are 300 players in total. Let Bi play a game against the
players Aj with 1 ≤ j ≤ i + 100, and assume no other contestants play a
game against each other. Then Bi has exactly 100 + i opponents, so for
101 ≤ m ≤ 200 there is a contestant playing exactly m games of chess.
Moreover, for j > 100, the player Aj is playing against the contestants Bi
with i ≥ j − 100; these are 100 − (j − 101) = 201 − j players. Because
j can vary from 101 up to and including 200, the number of opponents
varies from 100 up to and including 1. Hence, for 1 ≤ m ≤ 100 there is
also a contestant playing exactly m games of chess. Lastly, for j ≤ 100,
contestant Aj plays against all Bi ; these are 100 opponents. Hence, there
is no contestant playing more than 200 games of chess. We see that this
example meets the requirements for n = 200.
Now we will prove that n > 200 cannot be realised. We prove this by
contradiction, so assume that n > 200. Then in any case there is a
contestant A playing exactly 201 games of chess, against players B1 , B2 ,
. . . , B201 . In total, there are 300 contestants, so besides them and A,
there are another 300 − 1 − 201 = 98 contestants, who we will call C1 , . . . ,
C98 . If a contestant Bi is playing a game against another contestant Bj ,
then together with A they form a triple that plays three games among
each other; this is not allowed. Hence Bi does not play against any of the
other Bj . Therefore, they play at most 1 + 98 = 99 games. This means
that the contestants who play against exactly m other contestants with
100 ≤ m ≤ 200, must all be among the contestants Ci . However, these are
only 98 distinct contestants, which gives a contradiction. We conclude that
n > 200 is impossible and that n = 200 is the maximum.
34
Problems
Part 1
1. When writing down the date 12 August 2018 using eight digits, each digit
occurs exactly twice: 12-08-2018. There are more dates in 2018 having the
same property. How many dates in 2018 have this property, including 12
August?
A) 5 B) 6 C) 7 D) 8 E) 9
2. In the crossword puzzle on the right, a b

each square has to be filled with
one of the digits 1 to 9. A digit
c
may occur multiple times. For the
2-digit numbers formed in the rows
and columns we are given the following four hints:
Across Down
a. An odd number a. A square
c. A square b. An odd number
The puzzle has more than one solu-
tion, but the digit in the top-left
corner is always the same. Which
digit is in the top-left corner?
A) 1 B) 2 C) 3 D) 4 E) 6
3. Sophie likes to wear red and blue T-shirts. She decided to wear either a
red or a blue T-shirt each day, starting from 1 January 2019. She does not
want to say which colour she will be wearing on 1 and 2 January. From 3
January on, she will choose the colour of her T-shirt each day according to
the following rule: she chooses red if she wore two different colours the last
two days, and she chooses blue if she wore the same colour the last two
days.
By following this rule, she will wear a blue T-shirt on her birthday, 14
January. Is it possible to determine with certainty the colours of the T-
shirts she will be wearing on 28 and 29 January?
35
A) 28th: red, 29th: blue D) 28th: could be either 29th: blue
B) 28th: blue 29th: blue E) 28th: blue 29th: could be either
C) 28th: blue 29th: red
4. We have an isosceles triangle with two angles of 45 de-

grees; the long side has length 1. Within the triangle we
put two squares of the same size. We can do this in two
ways depicted in the two figures.
What is the total area of the two squares in the top
figure minus the total area of the two squares in the
bottom figure?
1 1 1 1 1
A) 72 B) 48 C) 36 D) 24 E) 18
5. On a glass table, we arrange 100 dice tightly in a 10 by

10 square. This is done in such a way that if two dice are
touching each other along a face, these faces have the
same number of pips. Both the top and bottom faces of
the 100 dice are visible. Altogether, on the front, rear,
left, and right side there are 40 dice faces visible. We add the pips on all of
these 240 visible faces.
What is the largest outcome that we can get?
In the figure on the right, you can see a net representing a dice.
A) 840 B) 880 C) 920 D) 1240 E) 1440
6. A large square is subdivided into 16 squares. We put crosses in two of these

squares (see the figure). This can be done in different ways. Sometimes
two such ways only differ by a rotation, for example the left two squares
below. In this case, we consider the two ways as the same one and only
count it once.
Pay attention: the right two squares are not considered to be the same. You
could obtain the right square from the middle one by reflection, but not by
using a rotation!
In how many different ways can you put the two crosses?
A) 21 B) 30 C) 32 D) 34 E) 36
36
7. Using 27 small cubes, each coloured black or white, we build a 3×3×3 cube.
This large cube has six views: a front, rear, left, right, top, and bottom
view. In the figure, five of the views of the large cube are depicted.
Which could be the sixth view of the large cube?
A) B) C) D) E)
8. How many distinct pairs of digits a and b are there such that 5a68 × 865b
is divisible by 824?
Pay attention: we are counting pairs, so for example if for a = 0 the values
b = 0, b = 1, and b = 2 are all valid, then these are counted as three
different pairs.
A) 10 B) 11 C) 15 D) 19 E) 21
Part 2
1. Anne, Bert, Christiaan, Dirk, and Eveline are particpating in a chess

tournament. They find out that their average age is exactly 28 years.
Exactly one year later Anne, Bert, Christiaan, and Dirk participate together
with Freek in the tournament. This time, their average age is exactly 30
years.
How many years older is Freek compared to Eveline?
x x
2. What is the smallest positive integer x for which the outcome of 2 + 3 +
x x x
4 + 5 + 6 is an integer?
37
3. For each of the two fractions 2018 2054
2011 and 2019 we subtract the same integer a
from both the numerator and the denominator. The two fractions we get,
are equal.
What integer is a?
4. The sides of the square ABCD have length 10. The points P and Q lie on
the line connecting the midpoints of AD and BC. If we connect P with
A and C, and also Q with A and C, then the square is divided into three
parts having equal area.
What is the length of P Q? (The figure is not drawn to scale.)
D C
P
Q
A B
5. Jan has written the number A in two different ways as a fraction. Unfortu-
nately, two numbers have become invisible due to spilled ink spots:
+3 15
A= = .
12 26 −
Lia has found Jan’s note and wants to find all possibilities for the number
A. She knows that the numbers underneath the ink spots must be positive
integers, but the number A does not have to be an integer. Because she
does not know which numbers are hidden below the ink spots, she searches
for all combinations for which the equality holds. In this way, she finds
multiple possible values for A. The largest one she calls Amax and the
smallest one she calls Amin .
Determine Amax − Amin .
38
6. Tim draws five straight lines in a plane. The lines are continuing indefinitely,
and there are no three lines going through the same point. For each
intersection point of two lines he gets one sweet, and for each set of two or
more parallel lines he gets one sweet as well. For instance, in the example
below, he gets 8 sweets because firstly there are 7 intersection points, and
secondly there is 1 set of three parallel lines, which is worth another sweet.
What are all possible numbers of sweets that Tim can get?
7. Six equilateral and equally sized triangles are glued to form a parallelogram
ABCD, see the figure.
D C
Y
X
A B
The length of AC is 10. What is the length of XY ?

8. In a class room there are a number of students. They find out that for each
triplet of students, the following two statements are both true:
• Two of them never wrote a report together.
• Two of them did once write a report together.
What is the maximum possible number of students in the class room?
39
Answers
Part 1
1. B) 6
2. E) 6
3. D) 28th: could be either, 29th: blue
1
4. A) 72
5. C) 920
6. C) 32
7. D)
8. D) 19
Part 2
57
1. 5 years 5. 4 (= 14 14 )
2. 20 6. 1, 5, 8, 10
5 10
3. 2009 7. 6 (= 12 )
20
4. 3 (= 6 23 ) 8. 5
40
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

We eat problems
for breakfast.

Olympiad 2019
International
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

Introduction

represent the Netherlands at the EGMO, which was supposed to be held in
the Netherlands. Being the host country, a second team of another four girls
was allowed to participate. Unfortunately, due to the Covid-19 pandemic
the EGMO could not take place in Egmond aan Zee. Instead, a virtual
event was held, where the Dutch team managed to win three bronze medals.
For more information about the EGMO (including the 2020 paper), see
www.egmo.org.
ten students participating in the Benelux Mathematical Olympiad (BxMO).
This contest was turned into a virtual event as well, held on 2 May. The
Dutch team achieved an outstanding result: one gold medal, three silver
1
medals and five bronze medals. For more information about the BxMO
selected by three team selection tests on 10, 11 and 12 June, each lasting
four hours. Since the IMO had been postponed to September, we decided to
participate in the new virtual event Cyberspace Mathematical Competition
(CMC) in July with the same six students. Three young, promising students
were selected to accompany the team to the IMO/CMC training camp,
which was held from 4 until 11 July in Egmond aan Zee.
the solutions.
2
Dutch delegation
The Dutch team for the virtual IMO 2020 consists of
• Jesse Fitié (18 years old)

• Jovan Gerbscheid (17 years old)
– bronze medal at IMO 2018, bronze medal at IMO 2019
• Jippe Hoogeveen (17 years old)
– bronze medal at IMO 2018, bronze medal at IMO 2019
• Rafaël Houkes (18 years old)
– gold medal at BxMO 2020
• Tjeerd Morsch (18 years old)
• Hanne Snijders (18 years old)
– bronze medal at EGMO 2019, bronze medal at EGMO 2020
Also part of the IMO/CMC selection, but not officially part of the IMO
team, are:
• Jelle Bloemendaal (16 years old)

• Kevin van Dijk (16 years old)
• Casper Madlener (15 years old)

• Jeroen Huijben (observer B), University of Amsterdam
3
Problems
A-problems
1. Arthur has written down five distinct positive integers smaller than 10. If
you add any two of these five numbers, then the result will never be equal
to 10.
Which number did Arthur write down for sure?
A) 1 B) 2 C) 3 D) 4 E) 5
2. On a 2019×2019 chess board, there is a contagious disease. Each day some

of the squares on the chess board are sick and the rest are healthy. A
healthy square bordering a sick square (along a side), becomes sick itself
the next day. A sick square will always be healthy the next day. A healthy
square that has been sick before, can become sick again (if it is infected by
one of the adjacent squares). On day 1, only the middle square is sick.
How many squares are sick on day 100?
A) 200 B) 298 C) 396 D) 9999 E) 10000
3. Out of a circular disk of radius 3, we cut three small disks of radius 1 in

the way depicted in the figure on the left. This causes the remainder of
the big disk to fall apart into two pieces. The bottom part is rotated 90
degrees and is put on top of the upper part as shown in the figure on the
right. The part where the two pieces overlap is coloured a bit darker.
What is the total area of the figure on the right (i.e. both the light and
dark grey parts together)?
A) 4π B) 92 π C) 19
4 π D) 5π E) 21
4 π
4
4. There are 13 distinct multiples of 7 that consist of two digits. You want
to create a longest possible chain consisting of these multiples, where two
multiples can only be adjacent if the last digit of the left multiple equals
the first digit of the right multiple. You can use each multiple at most once.
For example, 21 – 14 – 49 is an admissible chain of length 3.
What is the maximum length of an admissible chain?
A) 6 B) 7 C) 8 D) 9 E) 10
5. In a table with two rows and five columns, each of the squares is coloured
black or white according to the following rules:
• Two adjacent columns may never have the same number of black
squares.
• Two 2×2-squares that overlap in one column may never have the same
number of black squares.
How many possible colourings of the table comply with these rules?
A) 6 B) 8 C) 12 D) 20 E) 24
6. Which of the following numbers is the largest number you can get by
separating the numbers 1, 2, 3, 4, and 5 by using each of the operations
+, −, :, and × exactly once, where you may use parentheses to indicate
the order in which the operations should be executed? For example:
(5 − 3)×(4 + 1) : 2 = 5.
53 69
A) 21 B) 2 C) 33 D) 2 E) 35
7. Agatha, Isa and Nick each have a different kind of bike. One of them has
an electric bike, one has a racing bike, and one has a mountain bike. The
bikes have different colours: green, blue and black. The three owners make
two statements each, of which one is true and the other is false:
• Agatha says: “I have an electric bike. Isa has a blue bike.”
• Isa says: “I have a mountain bike. Nick has an electric bike.”
• Nick says: “I have a blue bike. The racing bike is black.”
Exactly one of the following statements is certainly true. Which one?
A) Agatha has a green bike. D) Isa has a mountain bike.
B) Agatha has a mountain bike. E) Nick has an electric bike.
C) Isa has a green bike.
5
8. Quadrilateral ABCD has right angles at A and D. A D C
circle of radius 10 fits neatly inside the quadrilateral
and touches all four sides. The length of edge BC
is 24. The midpoint of edge AD is called E and the E F
midpoint of edge BC is called F .
What is the length of EF ?
√ √
A) 43 B) 13 11 C) 33 11 D) 22 E) 45 A B
2 2 5 2
B-problems
1. Every day, Maurits bikes to school. He can choose between two different
routes. Route B is 1.5 km longer than route A. However, because he
encounters fewer traffic lights, his average speed along route B is 2 km/h
higher than along route A. This makes that travelling along the two routes
takes exactly the same amount of time.
How long does it take for Maurits to bike to school?
2. Starting with a positive integer, a fragment of that number is any positive

number obtained by removing one or more digits from the beginning and/or
end of that number. For example: the numbers 2, 1, 9, 20, 19, and 201 are
the fragments of 2019.
What is the smallest positive integer n such that the following holds: there
is a fragment of n such that when you add this fragment to n itself, you
get 2019?
3. Inside an equilateral triangle, a circle is drawn that

touches all three sides. The radius of the circle is
10. A second, smaller, circle touches the first circle
and two sides of the triangle. A third, even smaller,
circle touches the second circle and two sides of the
triangle (see the figure). What is the radius of the
third circle?
4. Alice has a number of cards. Each card contains three of the letters A
to I. For any choice of two of those letters, there is at least one card that
contains both letters.
What is the smallest number of cards that Alice can have?
6
Solutions
A-problems
1. E) 5 5. D) 20
2. E) 10000 6. E) 35
3. D) 5π 7. D) Isa has a mountain bike.
4. B) 7 8. D) 22
B-problems
1. 45 minutes
2. 1836
10
3. 9
4. 12
7
Problems
B-problems
B1. After breakfast, the sisters Anna and Birgit depart for school, each going
to a different school. Their house is next to a bicycle path running between
the two schools. Anna is cycling with a constant speed of 12 km per hour
and Birgit is walking in the opposite direction with a constant speed of 4
km per hour. They depart at the same time. Shortly after their departure,
mother notes that the girls have forgotten their lunch and decides to go
after them. Exactly 10 minutes after Anna and Birgit have left, mother
departs on her electric bike. First, she catches up with Anna. She hands
her a lunch box, immediately turns around, and goes after Birgit. When
she catches up with Birgit, she hands her a lunch box and immediately
rides back home. Mother always rides at a constant speed of 24 km per
hour.
How many minutes after the departure of Anna and Birgit does mother
return home?
B2. In a tall hat there are one hundred notes, numbered from 1 to 100. You
want to have three notes with the property that each of the three numbers
is smaller than the sum of the other two. For example, the three notes
numbered 10, 15, and 20 would be suitable (as 10 < 15 + 20, 15 < 10 + 20,
and 20 < 10 + 15), but the notes numbered 3, 4, and 7 would not (as 7 is
not smaller than 3 + 4). You may (without looking at the numbers on the
notes) take some notes from the hat.
What is the smallest number of notes you have to take to be sure to have
three notes that meet your wish?
B3. On each of the twelve edges of a cube we write the −1

−1
number 1 or −1. For each face of the cube, we multiply −1
−1
1
the four numbers on the edges of this face and write −1
1 1
the outcome on this face. Finally, we add the eighteen −1
1
numbers that we wrote down. 1
−1
What is the smallest (most negative) result we can get?
In the figure you see an example of such a cube. You cannot see the numbers
on the back of the cube.
8
B4. If you try to divide the number 19 by 5, you will get a remainder. The
number 5 fits 3 times in 19 and you will be left with 4 as remainder. There
are two positive integers n having the following property: if you divide n2
by 2n + 1, you will get a remainder of 1000.
What are these two integers?
D C
B5. In a square ABCD of side length 2 we draw lines from
each vertex to the midpoints of the two opposite sides.
For example, we connect A to the midpoint of BC and to 2
the midpoint of CD. The eight resulting lines together

bound an octagon inside the square (see figure).
A 2 B
What is the area of this octagon?
C1. We consider sequences a1 , a2 , . . . , an consisting of n integers. For given

k ≤ n, we can partition the numbers of the sequence into k groups as
follows: a1 goes in the first group, a2 in the second group, and so on until
ak which goes in the k-th group. Then ak+1 goes in the first group again,
ak+2 in the second group, and so on. The sequence is called k-composite
if this partition has the property that the sums of the numbers in the k
groups are equal.
The sequence 1, 2, 3, 4, −2, 6, 13, 12, 17, 8, for instance, is 4-composite as
1 + (−2) + 17 = 2 + 6 + 8 = 3 + 13 = 4 + 12.
However, this sequence is not 3-composite, as the sums 1 + 4 + 13 + 8,

2 + (−2) + 12, and 3 + 6 + 17 do not give equal outcomes.
(a) Give a sequence of 6 distinct integers that is both 2-composite and

3-composite.
(b) Give a sequence of 7 distinct integers that is 2-composite, 3-composite,
and 4-composite.
(c) Find the largest k ≤ 99 for which there exists a sequence of 99 distinct
integers that is k-composite. (Give an example of such a sequence and
prove that such a sequence does not exist for greater values of k.)
9
C2. A year is called interesting if it consists of four distinct digits. For example,
the year 2019 is interesting. It is even true that all years from 2013 up
to and including 2019 are interesting: a sequence of seven consecutive
interesting years.
(a) Determine the next sequence of seven consecutive interesting years

and prove that this is indeed the next such sequence.
(b) Prove that there is no sequence of eight consecutive interesting years
within the years from 1000 to 9999.
10
Solutions
B-problems
1. 42
2. 11
3. −12
4. 666 and 1999
2
5. 3
C-problems
C1. (a) An example of a correct sequence is 5, 7, 6, 3, 1, 2. This sequence

consists of six distinct numbers and is 2-composite since 5 + 6 + 1 =
7 + 3 + 2. It is also 3-composite since 5 + 3 = 7 + 1 = 6 + 2.
This is just one example out of many possible correct solutions. Below
we describe how we found this solution.
We are looking for a sequence a1 , a2 , a3 , a4 , a5 , a6 that is 2-composite
and 3-composite. Hence, we need that
a1 + a4 = a2 + a5 = a3 + a6 and a1 + a3 + a5 = a2 + a4 + a6 .
If we choose a4 = −a1 , a5 = −a2 , and a6 = −a3 , then the first two

equations hold. The third equation gives us a1 + a3 − a2 = a2 − a1 − a3 ,
and therefore a1 + a3 = a2 . We choose a1 = 1, a3 = 2 (and therefore
a2 = 3). We obtain the sequence 1, 3, 2, −1, −3, −2 consisting of six
distinct integers. If we wish to do so, we can increase all six numbers
by 4 to get a solution with only positive numbers: 5, 7, 6, 3, 1, 2.
(b) A possible solution is 8, 17, 26, 27, 19, 10, 1. This sequence consists of
seven distinct integers and is 2-composite since 8 + 26 + 19 + 1 =
17 + 27 + 10. It is 3-composite since 8 + 27 + 1 = 17 + 19 = 26 + 10.
It is also 4-composite since 8 + 19 = 17 + 10 = 26 + 1 = 27.
This is just one example out of many possible correct solutions. Below
we describe how we found this solution.
11
We are looking for a sequence a1 , a2 , a3 , a4 , a5 , a6 , a7 that is 2-, 3-, and
4-composite. Hence, we need that
a1 + a3 + a5 + a7 = a2 + a4 + a6 , (1)
a1 + a4 + a7 = a2 + a5 = a3 + a6 , (2)
a1 + a5 = a2 + a6 = a3 + a7 = a4 . (3)
We notice that in equation (1) we have a1 + a5 and a3 + a7 on the left,

and a2 + a6 and a4 on the right. If the sequence is 4-composite, these
four numbers are equal. Hence, we find that a 4-composite sequence is
automatically 2-composite as well.
From the equations in (3) it follows that
a1 = a4 − a5 , a2 = a4 − a6 , a3 = a4 − a7 .
Substituting this in the equations (2), we obtain
2a4 + a7 − a5 = a4 + a5 − a6 = a4 + a6 − a7 .
Subtracting a4 from each part, we get
a4 + a7 − a5 = a5 − a6 = a6 − a7 .
Hence, we obtain
a4 = (a5 − a6 ) − (a7 − a5 ) = 2a5 − a6 − a7 ,

a5 = (a6 − a7 ) + a6 = 2a6 − a7 .
We have thus expressed a1 , a2 , a3 , a4 , and a5 in terms of a6 and a7 .

Every solution is obtained by a suitable choice of a6 and a7 for which
the seven numbers become distinct. We try a6 = 10 and a7 = 1, and
find:
a5 = 2·10−1 = 19, a4 = 2·19−10−1 = 27, a3 = 27−1 = 26,
a2 = 27 − 10 = 17, and a1 = 27 − 19 = 8.
Hence, we have found a solution.
(c) The largest k for which a k-composite sequence of 99 distinct integers
exists, is k = 50. An example of such a sequence is
1, 2, . . . , 48, 49, 100, 99, 98, . . . , 52, 51.
The 99 integers in the sequence are indeed distinct and we see that
1 + 99 = 2 + 98 = . . . = 48 + 52 = 49 + 51 = 100, so this sequence is
50-composite.
12
Now suppose that k > 50 and that we have a k-composite sequence
a1 , a2 , . . . , a99 . Consider the group that contains the number a49 .
Since 49 − k < 0 and 49 + k > 99, this group cannot contain any other
number beside a49 . Next, consider the group containing the number
a50 . Since 50 − k < 0 and 50 + k > 99, this group cannot contain any
other number beside a50 . Hence, the numbers a49 and a50 each form
a group by themselves and must therefore have the same value. But
this is not allowed since the 99 numbers in the sequence had to be
distinct.
C2. (a) The years 2103 to 2109 are seven consecutive interesting years. If
there is an earlier sequence of seven, then it must start before 2100.
We shall now prove that this is not possible.
Because an interesting year cannot end with the digits 99, the first two
digits are the same for all years in a sequence of consecutive interesting
years (of four digits). Now suppose we have seven consecutive years
starting with digits 20. The seven final digits are consecutive and
unequal to 0 and 2, and therefore also unequal to 1. The seven final
digits must be the digits 3 to 9, in this exact order. Hence, the third
digit must be the only remaining digit, namely digit 1. We conclude
that 2013 to 2019 is the only sequence of seven consecutive interesting
years between 2000 and 2100.
(b) Suppose that there is a sequence of eight consecutive interesting years
between 1000 and 9999. Because an interesting year cannot end with
99, all eight years have the same first two digits. If also the third digit
does not change, then there are only 7 possibilities for the last digit,
which is not enough. Therefore, there are two consecutive years in our
sequence of the shape abc9 and abd0 with d = c + 1. Because there
are eight possible final digits, these must be the eight digits unequal
to a and b. Hence, both c and d = c + 1 must occur as final digit.
Because the numbers abcc and abdd cannot occur, this means that in
our sequence both abcd and abdc must occur. The difference between
these two numbers is 9, and our sequence consists of eight consecutive
numbers. This is also not possible. We have obtained a contradiction,
and conclude that the assumption that there exists a sequence of eight
consecutive interesting years between 1000 and 9999 is wrong.
13
Problems
1. A complete number is a 9 digit number that contains each of the digits 1 to

9 exactly once. The difference number of a number N is the number you
get by taking the differences of consecutive digits in N and then stringing
these digits together. For instance, the difference number of 25143 is equal
to 3431. The complete number 124356879 has the additional property that
its difference number, 12121212, consists of digits alternating between 1
and 2.
Determine all a with 3 ≤ a ≤ 9 for which there exists a complete number
N with the additional property that the digits of its difference number
alternate between 1 and a.
2. There are n guests at a party. Any two guests are either friends or not friends.
Every guest is friends with exactly four of the other guests. Whenever a
guest is not friends with two other guests, those two other guests cannot
be friends with each other either.
What are the possible values of n?
3. Points A, B, and C lie on a circle with centre M . The reflection of point

M in the line AB lies inside triangle ABC and is the intersection of the
angular bisectors of angles A and B. (The angular bisector of an angle is
the line that divides the angle into two equal angles.) Line AM intersects
the circle again in point D.
Show that |CA| · |CD| = |AB| · |AM |.
4. The sequence of Fibonacci numbers F0 , F1 , F2 , . . . is defined by F0 = F1 = 1

and Fn+2 = Fn + Fn+1 for all n ≥ 0. For example, we have
F2 = F0 + F1 = 2, F3 = F1 + F2 = 3, F4 = F2 + F3 = 5, F5 = 8.
The sequence a0 , a1 , a2 , . . . is defined by

1
an = for all n ≥ 0.
Fn Fn+2
Prove that for all m ≥ 0 we have:
a0 + a1 + a2 + · · · + am < 1.
14
5. Thomas and Nils are playing a game. They have a number of cards,
numbered 1, 2, 3, et cetera. At the start, all cards are lying face up on
the table. They take alternate turns. The person whose turn it is, chooses
a card that is still lying on the table and decides to either keep the card
himself or to give it to the other player. When all cards are gone, each of
them calculates the sum of the numbers on his own cards. If the difference
between these two outcomes is divisible by 3, then Thomas wins. If not,
then Nils wins.
(a) Suppose they are playing with 2018 cards (numbered from 1 to 2018)
and that Thomas starts. Prove that Nils can play in such a way that
he will win the game with certainty.
(b) Suppose they are playing with 2020 cards (numbered from 1 to 2020)
and that Nils starts. Which of the two players can play in such a way
that he wins with certainty?
Solutions
1. For a = 4, an example of such a number is 126734895. For a = 5, an

example is the number 549832761. (There are other solutions as well.)
We will show that for a = 3, 6, 7, 8, 9 there is no complete number with a
difference number equal to 1a1a1a1a. It then immediately follows that there
is also no complete number N with difference number equal to a1a1a1a1
(otherwise, we could write the digits of N in reverse order and obtain a
complete number with difference number 1a1a1a1a).
For a equal to 6, 7, 8, and 9, no such number N exists for the following
reason. For the digits 4, 5, and 6, there is no digit that differs by a from
that digit. Since the difference number of the complete number N is equal
to 1a1a1a1a, every digit of N , except the first, must be next to a digit that
differs from it by a. Hence, the digits 4, 5, and 6 can only occur in the first
position of N , which is impossible.
For a = 3 the argument is different. If we consider the digits that differ by
3, we find the triples 1–4–7, 2–5–8, and 3–6–9. If the 1 is next to the 4 in
N , the 7 cannot be next to the 4 and so the 7 must be the first digit of
N . If the 1 is not next to the 4, the 1 must be the first digit of N . In the
same way, either the 2 or the 8 must be the first digit of N as well. This is
impossible.
15
2. We first consider the friends of one guest, say Marieke. We know that
Marieke has exactly four friends at the party, say Aad, Bob, Carla, and
Demi. The other guests (if there are any other guests) are not friends
with Marieke. Hence, they cannot have any friendships among themselves
and can therefore only be friends with Aad, Bob, Carla, and Demi. Since
everyone has exactly four friends at the party, each of them must be friends
with Aad, Bob, Carla, and Demi (and with no one else).
Since Aad also has exactly four friends (including Marieke), the group of
guests that are not friends with Marieke can consist of no more than three
people. If the group consists of zero, one, or three people, we have the
following solutions (two guests are connected by a line if they are friends):
D C D C B A
M
A D
M
B C A B M
Solutions with five, six, and eight guests in total.
Now we will show that it is not possible for this group to consist of two
people. In that case, Aad would have exactly one friend among Bob, Carla,
and Demi. Assume, without loss of generality, that Aad and Bob are friends.
In the same way, Carla must be friends with one of Aad, Bob, and Demi.
Since Aad and Bob already have four friends, Carla and Demi must be
friends. However, since they are both not friends with Aad, this contradicts
the requirement in the problem statement.
We conclude that there can be five, six, or eight guests at the party. Hence,
the possible values for n are 5, 6, and 8.
3. Let I be the reflection of point M in the line AB. We define α = ∠CAI

and β = ∠CBI. Since AI is the angular bisector of ∠CAB, we find that
∠IAB = α. Since I is the reflection of M in the line AB, we find that
∠BAM = α. Triangle AM C is isosceles with apex M , because |AM | =
|CM |. Hence, we see that ∠M CA = ∠CAM = 3α. In the same way, we
see that ∠IBA = ∠ABM = β and ∠M CB = 3β. The sum of the angles of
triangle ABC is therefore 2α+(3α+3β)+2β = 180◦ . From this, we conclude
◦
◦ ◦ ◦
that α + β = 180
5 = 36 , and hence that ∠ACB = 3α + 3β = 3 · 36 = 108 .
16
C
I
B
A
Since M AB is an isosceles triangle (as |AM | = |BM |), we see that α =

β = 18◦ . It follows from this that ∠CAB = 2α = ∠ABC and therefore
that triangle ACB is isosceles. By considering the sum of the angles in
triangle AM C, we find that ∠AM C = 180◦ − 6α = 72◦ . Hence we also
find that ∠CM D = 180◦ − ∠AM C = 108◦ . We have already seen that
∠ACB = 108◦ . It follows that triangles ACB and CM D are both isosceles
triangles with an angle of 108◦ at the apex. Hence, they are similar triangles.
This implies that |CM | |AC|
|CD| = |AB| . By multiplying by both denominators and
observing that |CM | = |AM |, we obtain the required result.
4. Note that for all n ≥ 0 the number an can be rewritten as follows:

1 Fn+1 1 Fn+2 − Fn 1
an = = · = ·
Fn Fn+2 Fn Fn+2 Fn+1 Fn Fn+2 Fn+1

1 1 1 1 1
= − · = − .
Fn Fn+2 Fn+1 Fn Fn+1 Fn+1 Fn+2
We now get that for each m ≥ 0 the sum a0 + a1 + a2 + · · · + am equals

1 1 1 1 1 1
− + − + ··· + − .
F0 F1 F1 F2 F1 F2 F2 F3 Fm Fm+1 Fm+1 Fm+2
In this sum all terms cancel, except the first and last. In this way, we get
1 1 1
a0 + a1 + a2 + · · · + am = − =1− < 1.
F0 F1 Fm+1 Fm+2 Fm+1 Fm+2
5. (a) Thomas and Nils both make 1009 moves and Nils makes the last move.
Nils can make sure that the last card on the table contains a number
that is not divisible by 3. Indeed, he could start taking cards with
numbers that are divisible by 3, until all these cards are gone. Because
there are only 672 such cards, he has enough turns to achieve that.
17
We now consider the situation before the last move of Nils. Let k
be the number on the last card, and let the sums of the numbers
of Thomas and Nils at that very moment be a and b. Nils has two
options. If he gives away the last card, the difference between the
outcomes becomes (a + k) − b, and if he keeps the card, the difference
becomes a − (b + k). Nils is able to win, unless both numbers are
divisible by 3. But in that case (a + k − b) − (a − b − k) = 2k would
also be divisible by 3. Because k is not divisible by 3, the number 2k
is also not divisible by 3 and hence Nils can win with certainty.
(b) Nils can win. We distinguish three types of cards, depending on
the number on the card: type 1 (the number has remainder 1 when
dividing by 3), type 2 (the number has remainder 2 when dividing by
3), and type 3 (the number is divisible by 3). Because 2019 = 3 · 673
and the card 2020 is of type 1, there are 674 cards of type 1, 673 cards
of type 2, and 673 cards of type 3.
In order to win, Nils chooses a card of type 3 in his first turn (and
gives it to Thomas). Then there are 674 cards of type 1 left, 673 of
type 2, and 672 of type 3. In the next turns he responds to Thomas’s
move in the following way (as long as he is able to).
(i) If Thomas chooses a card of type 1, then Nils chooses a card of
type 2 and gives it to the same person that got Thomas’s card.
(ii) If Thomas chooses a card of type 2, then Nils chooses a card of
type 1 and gives it to the same person that got Thomas’s card.
(iii) If Thomas chooses a card of type 3, then Nils does the same (and
gives the card to Thomas).
As long as Nils keeps this up, the sum of each player’s cards is divisible
by 3 after his turn (because a number of type 1 and a number of type
2 add up to a number which is divisible by 3).
Because the number of cards of type 3 is always even after Nils’s turn,
Nils can always execute his planned move in case (iii). Because the
number of cards of type 1 is always 1 greater than that of type 2 after
Nils’s turn, he can also always execute his planned move in case (ii).
Only at the moment when all cards of type 2 are gone and Thomas
takes the last card of type 1 (case (i)), Nils cannot execute his planned
move. However, in that case Nils cannot lose anymore. Indeed, after
Thomas’s turn the sum of the cards of one player is still divisible by
3, but the sum of the cards of the other player is not divisible by 3
anymore. Because there are only cards of type 3 left now, this will stay
the same until all cards are gone. At the end, the difference between
the sums of both players is not divisible by 3 and Nils wins.
18
Problems
1. For an integer n ≥ 3 we consider a circle containing n vertices. To each

vertex we assign a positive integer, and these integers do not necessarily
have to be distinct. Such an assignment of integers is called stable if the
product of any three adjacent integers is n. For how many values of n with
3 ≤ n ≤ 2020 does there exist a stable assignment?
2. In an acute triangle ABC the foot of the altitude from A is called D. Let D1
and D2 be reflections of D in AB and AC, respectively. The intersection of
BC and the line through D1 parallel to AB, is called E1 . The intersection
of BC and the line through D2 parallel to AC, is called E2 . Prove that
D1 , D2 , E1 , and E2 lie on a circle whose centre lies on the circumcircle of
4ABC.
f (x2 y) + 2f (y 2 ) = x2 + f (y) · f (y)

for all x, y ∈ R.
4. On a circle with centre M there are three distinct points A, B, and C such
that |AB| = |BC|. The point D lies inside the circle in such a way that
4BCD is isosceles. The second intersection point of AD and the circle is
called F . Prove that |F D| = |F M |.
5. A set S consisting of 2019 (distinct) positive integers has the following

property: the product of any 100 elements of S is a divisor of the product of
the other 1919 elements. What is the maximum number of prime numbers
that S could contain?
19
Solutions
1. Suppose n is not a multiple of 3 and that we have a stable assignment

of the numbers a1 , a2 , . . . , an , in that order on the circle. Then we have
ai ai+1 ai+2 = 3 for all i, where the indices are considered modulo n. Hence,
ai+1 ai+2 ai+3 = n = ai ai+1 ai+2 ,
which yields ai+3 = ai (as all numbers are positive). Through induction,
we find that a3k+1 = a1 or all integers k ≥ 0. Because n is not a multiple of
3, the numbers 3k + 1 for k ≥ 0 take on all values modulo n: indeed, 3 has
a multiplicative inverse modulo n, hence k ≡ 3−1 · (b − 1) implies 3k + 1 ≡ b
mod n for all b. We conclude that all numbers on the circle must equal a1 .
Hence, we have a31 = n, where a1 is a positive integer. Hence, if n is not a
multiple of 3, then n must be a cube.
If n is a multiple of 3, then we put the numbers 1, 1, n, 1, 1, n, . . . in that
order on the circle. In that case, the product of three adjacent numbers
always equals 1 · 1 · n = n. If n is a cube, say n = m3 , then we put
the numbers m, m, m, . . . on the circle. In that case, the product of three
adjacent numbers always equals m3 = n.
We conclude that a stable assignment exists if and only if n is a multiple of 3,
or a cube. Now we have the count the number of such n. The multiples of 3
with 3 ≤ n ≤ 2020 are 3, 6, 9, . . . , 2019; these are 2019
3 = 673 numbers. The
cubes with 3 ≤ n ≤ 2020 are 23 , 33 , . . . , 123 , because 123 = 1728 ≤ 2020
and 133 = 2197 > 2020. These are 11 cubes, of which 4 are divisible by 3,
hence there are 7 cubes which are not a multiple of 3. Altogether, there
are 673 + 7 = 680 values of n satisfying the conditions.
A
D2
D1
E2
E1 B D C
2. Let K be the midpoint of DD1 , and let L be the midpoint of DD2 . Then
K lies on AB and L lies on AC. Because ∠AKD = 90◦ = ∠ALD, the
20
quadrilateral AKDL is cyclic. Hence, ∠DLK = ∠DAK = ∠DAB =
90◦ − ∠ABC. Moreover, KL is a midsegment in triangle DD1 D2 , hence
∠DLK = ∠DD2 D1 . We conclude that ∠DD2 D1 = 90◦ − ∠ABC.
Because AC ⊥ DD2 en D2 E2 k AC, we have ∠DD2 E2 = 90◦ . Hence,
∠D1 D2 E2 = ∠D1 D2 D + ∠DD2 E2 = 90◦ − ∠ABC + 90◦ = 180◦ − ∠ABC.
On the other hand, as D1 E1 k AB, we have ∠D1 E1 E2 = ∠ABC, hence
we get ∠D1 D2 E2 = 180◦ − ∠D1 E1 E2 . We conclude that D1 E1 E2 D2 is a
cyclic quadrilateral.
Let M be the point such that AM is a diameter of the circumcircle of
4ABC. Thales’ theorem yields ∠ACM = 90◦ . Hence, CM ⊥ AC, which
yields CM ⊥ D2 E2 and CM k DD2 . Moreover, L is the midpoint of
DD2 and LC k D2 E2 , hence LC is a midsegment in triangle DD2 E2 .
This means that C is the midpoint of DE2 . Because CM k DD2 , we
get that CM is also a midsegment, hence CM intersects D2 E2 in the
middle. As CM ⊥ D2 E2 , the line CM is the perpendicular bisector of
D2 E2 . Analogously, we get that BM is the perpendicular bisector of D1 E1 .
Hence, M is the intersection point of the perpendicular bisectors of two
of the chords of the circle through D1 , D2 , E1 , and E2 . Hence, M is the
centre of this circle.
3. Substituting x = 1 gives f (y) + 2f (y 2 ) = (1 + f (y))f (y), hence
2f (y 2 ) = f (y)2 . (4)
Using this, we can cancel the 2f (y 2 ) on the left hand side of the original
functional equation against the f (y)2 on the right hand side:
f (x2 y) = x2 f (y).
Substituting y = 1 in this equations yields f (x2 ) = x2 f (1), and substituting

y = −1 yields f (−x2 ) = x2 f (−1). Because x2 takes on all non-negative
numbers as value when x ∈ R, we get
(
cx als x ≥ 0,
f (x) =
dx als x < 0,
with c = f (1) and d = −f (−1). Now substitute y = 1 in equation (4),

which yields 2f (1) = f (1)2 , hence 2c = c2 . It follows that c = 0 or
c = 2. If we actually substitute y = −1 in equation (4), then we find that
2f (1) = f (−1)2 , hence 2c = (−d)2 . For c = 0, we get d = 0, and for c = 2,
we get d = 2 or d = −2. So, there are three cases:
• c = 0, d = 0: then f (x) = 0 for all x;
21
• c = 2, d = 2: then f (x) = 2x for all x;
• c = 2, d = −2: then f (x) = 2x for x ≥ 0, and f (x) = −2x for x < 0,
or, in other words, f (x) = 2|x| for all x.
Using the first function, both sides of the functional equation become 0, so
this function is a solution. Using the second function, we get 2x2 y + 4y 2
on the left hand side, and (x2 + 2y) · 2y = 2x2 y + 4y 2 on the right hand
side, so this function is a solution as well. Using the third function, we get
2|x2 y| + 4|y 2 | = 2x2 |y| + 4y 2 on the left hand side, and (x2 + 2|y|) · 2|y| =
2x2 |y| + 4|y|2 = 2x2 |y| + 4y 2 on the right hand side, so also this function is
a solution.
Altogether, we found the three solutions: f (x) = 0, f (x) = 2x, and
f (x) = 2|x|.
C
A
M
4. We will prove that |F D| = |F C| and |F C| = |F M |, which proves the

statement.
In the cyclic quadrilateral ABCF , we have ∠BCF = 180◦ − ∠BAF . As
|AB| = |BC| = |BD| we also have ∠BAF = ∠BAD = ∠ADB, hence
∠BDF = 180◦ − ∠ADB = 180◦ − ∠BAF . We see that ∠BCF = ∠BDF .
Moreover, ∠DF B = ∠AF B and ∠CF B are inscribed angles on chords
AB and BC of the same length, hence ∠DF B = ∠CF B. Triangles BCF
and BDF have two pairs of equal angles; because they also have the side
BF in common, they are congruent. We conclude that |F C| = |F D| and
∠DBF = ∠CBF .
The inscribed angle theorem yields ∠CM F = 2∠CBF . From the equality
∠DBF = ∠CBF we just found, we get that 2∠CBF = ∠CBD = 60◦ ,
hence ∠CM F = 60◦ . Moreover, |M C| = |M F | (radius of the circle), hence
22
4CM F is isosceles with an angle of 60◦ , which yields that the triangle is
equilateral. This means that |F C| = |F M |.
This concludes the proof that |F D| = |F C| = |F M |.
5. The maximum number of prime numbers is 1819.

We start with the construction. Choose distinct primes p1 , p2 , . . . , p1819 ,
and let P = p1 p2 · · · p1819 . Let
S = {p1 , p2 , . . . , p1819 , P, P · p1 , . . . , P · p199 }.
For each pi , there are 201 numbers in S that are divisible by pi (namely,
pi and all multiples of P ). Of these, at most one has two factors pi ; the
rest has only one factor pi . If we now take 100 numbers from S, then their
product has at most 101 factors pi . The other numbers contain at least
101 numbers which are divisible by pi , hence their product has at least 101
factors pi . Because this holds for any pi , and the numbers in S do not have
any other prime factors, this implies that S has the desired property.
We now prove that S cannot contain more than 1819 primes. Consider a
prime divisor q of a number in S. Suppose that at most 199 numbers S
are divisible by q. Then we take the 100 elements of S having the most
factors q; these always have more factors q in total than the other elements,
which contradicts the condition in the problem statement. Hence, there
are at least 200 numbers in S which are divisible by q. If there are exactly
200, then we also get that the number of factors q in all of these numbers
must be equal, otherwise we get a contradiction again by taking the 100
elements having the most factors q.
We see that S contains at least 199 non-primes, because a prime p in S
divides at least 199 other elements of S. Suppose that S contains exactly
199 non-primes. Then the prime factor p in each of these 199 non-primes
occurs exactly once (namely, equally often as in the prime number p).
Moreover, the numbers in S cannot be divisible by a prime r that is not
contained in S, because then there would be at least 200 multiples of r
inside S, and these would be 200 non-primes, which is a contradiction.
We get that each of the 199 non-primes in S must be the product of the
primes in S. In particular, these 199 numbers are not distinct. This is a
contradiction, hence S must contain at least 200 non-primes, and hence at
most 1819 primes.
23
Problems
1. In an acute triangle ABC, the centre of the incircle is I, and |AC| + |AI| =
|BC|. Prove that ∠BAC = 2∠ABC.
2. Determine all polynomials P (x) with real coefficients for which

P (x2 ) + 2P (x) = P (x)2 + 2.
3. For a positive integer n, we consider an n × n-board and tiles with sizes

1 × 1, 1 × 2, . . . , 1 × n. In how many ways, can exactly 12 n(n + 1) squares of
the board be coloured red, so that the red squares can be covered by placing
the n tiles horizontally on the board, as well by placing the n vertically
on the board? Two colourings which are not identical, but which can be
obtained from one another by rotation or reflection, are counted as different
colourings.
4. Let a, b ≥ 2 be positive integers with gcd(a, b) = 1. Let r be the smallest

positive value that ab − dc can take, where c and d are positive integers
satisfying c ≤ a and d ≤ b. Prove that 1r is an integer.
Solutions
1. Let D be a point on BC such that C
|CD| = |AC|. Because |BC| =
|AC| + |AI|, the point D lies on
the interior of side BC, and we have
|BD| = |AI|. Because triangle ACD
I
is isosceles, the angle bisector CI
is also the perpendicular bisector
of AD, hence A is the reflection of B
D in CI. Hence, we get ∠CDI = A
∠CAI = ∠IAB, hence 180◦ −∠BDI =
∠IAB, which means that quadrilateral ABDI is cyclic. In this cyclic quad-
rilateral BD and AI have the same length. Therefore, AB and ID are
parallel. Hence, ABDI is an isosceles trapezium, which has equal angles at
the base. Hence, ∠CBA = ∠DBA = ∠BAI = 12 ∠BAC, which proves the
statement.
24
2. We rewrite the equation as
2
P (x2 ) − 1 = P (x) − 1 .
Let Q(x) = P (x) − 1, then Q is a polynomial with real coefficients satisfying
Q(x2 ) = Q(x)2 .
Suppose that Q is constant, say Q(x) = c with c ∈ R. Then we have

c = c2 , hence c = 0 or c = 1. Both possibilities give rise to solutions.
Form now on, we may assume that Q is non-constant, hence we can write
Q(x) = bxn + R(x) with n ≥ 1, b 6= 0, and R(x) a polynomial with real
coefficients of degree at most n − 1. The polynomial equation then becomes
bx2n + R(x2 ) = b2 x2n + 2bxn · R(x) + R(x)2 .
By comparing the coefficients in front of x2n on the left and right hand
side, we get b = b2 . As b =
6 0, we must have b = 1. Subtracting x2n on
both sides, yields
R(x2 ) = 2xn · R(x) + R(x)2 .
If R is non-zero, then it has degree m ≥ 0. We have m < n. Then the
left hand side of this equation has degree 2m, and the right hand side has
degree m + n, as m + n > 2m. This is a contradiction. Hence, R must
be the zero polynomial, which yields Q(x) = xn . This polynomial indeed
satisfies the polynomial equation for Q.
Hence, we find the following solutions: P (x) = 1, P (x) = 2, and P (x) =
xn + 1 with n ≥ 1.
3. The number of red squares must equal the total number of squares covered
by the n tiles, hence the tile are only put on top of red squares. Consider a
colouring of the board and the corresponding horizontal covering by the
tiles (where all tiles are places horizontally) and the vertical covering. We
will deduce a number of properties for the colouring, and then count how
many colourings there are. The tile whose size is 1 × k is called the k-tile.
Because the horizontal covering contains an n-tile, each column has at
least one red square. In the vertical covering, each column must therefore
contain at least one tile; because there are exactly n tiles, this means that
there must be exactly one tile in each column. In the same way, each row
must contain exactly one tile in the horizontal covering. Now number the
rows and columns depending on the number of the tile that has been put
there: so row i is the row containing the i-tile in the horizontal covering,
and analogously for the columns.
25
We will now prove that the square in row i and column j (which will be
called (i, j)) is red if and only if i + j ≥ n + 1. We prove this using induction
on i. In row 1, there is only one red square, so that must be in the column
containing the n-tile in the vertical covering, i.e. column n. Hence, the
square (1, j) is red if and only if j = n, or if and only if 1 + j ≥ n + 1. Now
let k ≥ 1 and suppose the statement has been proved for all i ≤ k. We want
to prove the statement for i = k + 1, i.e. that the square (k + 1, j) is red if
and only if k + 1 + j ≥ n + 1, or j ≥ n − k. Consider a column j ≥ n − k.
Because of the induction hypothesis, we know exactly how many red squares
this column has in rows 1, 2, . . . , k: namely, the square (i, j) is red if and
only if i+j ≥ n+1, or i ≥ n+1−j; these are k −(n−j) = j +k −n squares.
In the other n − k rows, this column needs another j − (j + k − n) = n − k
red squares. Hence, this column has a red square in each of these rows, in
particular in the row i = k + 1. In row i = k + 1, the squares (i, j) with
j ≥ n − k are all red, and these are k + 1 squares. Hence, these are exactly
all red squares in row i = k + 1, hence the square (i, j) is red if and only if
j ≥ n − k, or if and only if i + j ≥ n − k + k + 1 = n + 1. This finishes the
proof by induction.
Now consider two adjacent rows with row numbers a and b, with a > b.
In column n − b, there is a red square row a (because a + n − b > n), but
not in row b. In the row directly on the other side of row b (if this row
exists), there cannot be a red square in column n − b, because the red
squares in column n − b would otherwise not be consecutive, and then the
tile with number n − b cannot lie there. The row number of this row must
therefore be smaller than b. We conclude that the row numbers cannot
decrease first and then increase. Above and below row n, there must be a
row with a smaller number (or no row at all), and the row numbers must
descend in both directions from there. We see that the row numbers from
top to bottom must first ascend until we get to row n, and then they must
descend. The same can be proved for the column numbers.
Vice versa, we have to prove that if the row and columns numbers are first
ascending and then descending, then the horizontal and vertical tiles can be
put. To prove this, we colour the square (i, j) red if and only if i + j ≥ n + 1.
For a fixed i, the red squares are the squares (i, j) with j ≥ n + 1 − i;
because of the order of the column numbers, these columns are adjacent.
Hence, in each row, the red squares are adjacent. The horizontal tiles can
be put exactly on top of the red squares. In the same way, this can be done
for the vertical tiles. For these row and column numbers, there was not
other way to choose the colouring, because we already know that for each
suitable colouring the square (i, j) is red if and only if i + j ≥ n + 1.
Altogether, we are looking for the number of ways to choose the row
26
and column numbers such that the numbers are first ascending and then
descending; corresponding to each of these choices, there is exactly one
way to colour the squares so that they satisfy the conditions. The number
of ways to put the numbers 1 to n in an order that is first ascending and
then descending, equals the number of subsets of {1, 2, . . . , n − 1}. Namely,
each ordering corresponds to the subset of numbers that appear before the
number n; these can be sorted in a unique way (ascending), and the rest of
the numbers must be sorted descending and put after the n. The number
of subsets is 2n−1 . Hence, the total number of colourings satisfying the
conditions, is (2n−1 )2 = 22n−2 .
a
4. We will first show that it is possible to choose c and d such that b − dc = 1
bd .
Because gcd(a, b) = 1, there exists a multiplicative inverse b−1 of b modulo
a. Now let c with 1 ≤ c ≤ a be such that c ≡ −b−1 mod a. Then we
have bc ≡ −1 mod a, hence a | bc + 1. Define d = bc+1 a , which is a
bc+1 ba+1 1
positive integer. We have d = a ≤ a = b + a . Because a ≥ 2 and
d is an integer, we get d ≤ b. All conditions are met. Hence, we have
a c ad−bc
b − d = bd = bc+1−bc
bd = bd1
.
If this is the smallest possible outcome, then we are done, because 1r = bd
would be an integer. We will show that no smaller positive outcome is
achievable. Let c and d be as above, and suppose there are positive integers
0
c0 ≤ a and d0 ≤ b such that 0 < ab − dc0 < bd
1
. We will derive a contradiction.
0
Let x = ad0 − bc0 , then we have ab − dc0 = bdx0 , hence xbd < bd0 , which yields
xd < d0 . We also know that x > 0. Hence, 0 < xd < d0 ≤ b, which means
that xd and d0 are two distinct numbers whose difference is less than b.
Moreover, from x = ad0 − bc0 we get that x ≡ ad0 mod b. On the other
hand, we know that ad − bc = 1, hence ad ≡ 1 mod b, hence xad ≡ x
mod b. Combining this, we get ad0 ≡ xad mod b. Because gcd(a, b) = 1
we may divide by a, hence d0 ≡ xd mod b. However, we already saw that
d0 and xd are distinct numbers whose difference is smaller than b, hence
this is impossible.
We conclude that the c and d we found indeed give the smallest possible
outcome, and hence 1r is an integer.
27
Problems
1. Let a1 , a2 , . . . , a2020 be real numbers, not necessarily distinct. For all

n ≥ 2020, let an+1 be the minimal real root of the polynomial
Pn (x) = x2n + a1 x2n−2 + a2 x2n−4 + . . . + an−1 x2 + an ,
if it exists. Assume that an+1 exists for all n ≥ 2020. Prove that an+1 ≤ an
for all n ≥ 2021.
2. Ward and Gabrielle are playing a game on a large sheet of paper. At

the start of the game, there are 999 ones on the sheet of paper. Ward
and Gabrielle each take turns alternatingly, and Ward has the first turn.
During their turn, a player must pick two numbers a and b on the sheet
such that gcd(a, b) = 1, erase these numbers from the sheet, and write the
number a + b on the sheet. The first player who is not able to do so, loses.
Determine which player can always win this game.
3. Determine all pairs (a, b) of positive integers for which
a + b = ϕ(a) + ϕ(b) + gcd(a, b).
Here, ϕ(n) is the number of integers k ∈ {1, 2, . . . , n} satisfying gcd(n, k) =

1.
4. Let ABC be an acute triangle and let P be the intersection of the tangents
in B and C to the circumcircle of 4ABC. The line through A perpendicular
to AB and the line through C perpendicular to AC intersect in a point
X. The line through A perpendicular to AC and the line through B
perpendicular to AB intersect in a point Y . Prove that AP ⊥ XY .
28
Solutions
1. If x = α is a root of Pn , then x = −α is root of Pn as well, as all terms of
Pn have even degree. The minimal root of Pn therefore cannot be positive.
Therefore an ≤ 0 for all n > 2020. We have Pn+1 (x) = x2 · Pn (x) + an+1 .
Substitute x = an+1 ; as that is a root of Pn , we have Pn+1 (an+1 ) =
0 + an+1 ≤ 0.
As the maximal degree term in Pn (x) is x2n , there exists an N < 0 such
that Pn (x) > 0 for all x < N . Taking for example −N = max(2, |a1 | +
|a2 | + . . . + |an |), we see for x < N that x2i−2 ≤ x2n−2 for all 1 ≤ i ≤ n
and therefore that
a1 x2n−2 + a2 x2n−4 + . . . +an−1 x2 + an

≤ a1 x2n−2 + a2 x2n−4 + . . . + an−1 x2 + |an |

≤ |a1 |x2n−2 + |a2 |x2n−2 + . . . + |an−1 |x2n−2 + |an |x2n−2

≤ (|a1 | + |a2 | + . . . + |an |)x2n−2
≤ −N · x2n−2
< x2n ,
so x2n + a1 x2n−2 + a2 x2n−4 + . . . + an−1 x2 + an > 0. Hence for n ≥ 2021
there exists an N < 0 with Pn (x) > 0 for all x < N , whereas Pn (an ) ≤ 0.
Therefore Pn (x) has a root smaller than an . As an+1 is the minimal root,
we have an+1 ≤ an .
2. Gabrielle can always win using the following strategy: during each of her
turns, she picks the largest two numbers on the sheet as a and b. Using
induction on k, we will prove that she is always allowed to do so, and that
after her k-th turn, the sheet contains the number 2k + 1 and 998 − 2k
ones.
In his first turn, Ward can only pick a = b = 1, after which the sheet
contains the number 2 and 997 ones. Gabrielle then picks the two largest
numbers, a = 2 and b = 1, after which the sheet contains 3 and 996 ones.
This finishes the basis k = 1 of the induction.
Now suppose that for some m ≥ 1 after Gabrielle’s m-th turn the sheet
contains the number 2m + 1 and 998 − 2m ones. If 998 − 2m = 0, then
Ward cannot make a move. If not, then Ward can do one of two things,
either pick a = b = 1 or pick a = 2m + 1 and b = 1. We consider these two
cases separately:
• If Ward picks a = b = 1, then the sheet contains the number 2m + 1,

the number 2, and 996 − 2m ones. Gabrielle then picks the two largest
29
numbers, so a = 2m + 1 and b = 2 (which is allowed since their gcd is
1). After her turn the sheet contains the numbers 2m+3 = 2(m+1)+1
and 996 − 2m = 998 − 2(m + 1) ones.
• If Ward picks a = 2m + 1 and b = 1, then the sheet contains the
number 2m+2 and 997−2m ones. Gabrielle then picks the two largest
numbers, so a = 2m + 2 and b = 1 (which is allowed since their gcd is
1). Note that there is a one left, as 997 − 2m is odd, so not equal to 0.
After her turn the sheet contains the numbers 2m + 3 = 2(m + 1) + 1
and 996 − 2m = 998 − 2(m + 1) ones.

Therefore Gabrielle can always make a move. After Gabrielle’s turn 499
the only number left on the sheet is 999, so Ward can no longer make a
move, and Gabrielle wins.
3. First suppose that a = 1. Then ϕ(1) = 1. For all positive integers b we have
gcd(a, b) = 1. Therefore in this case the equation is 1 + b = 1 + ϕ(b) + 1, or
equivalently, ϕ(b) = b − 1. This is equivalent to the statement that there
exists a unique integer from {1, 2, . . . , b}; which then has to be b itself (since
unless b = 1, we have gcd(b, b) > 1, but if b = 1 we have ϕ(b) = b). In other
words, this is equivalent to b being a prime number. Hence the solutions
for a = 1 are precisely the pairs (1, p) with p a prime number. Similarly,
the solutions for b = 1 are precisely the pairs (p, 1) with p a prime number.
Now assume that a, b ≥ 2. As gcd(b, b) > 1 we have ϕ(b) ≤ b − 1. Therefore
gcd(a, b) = a + b − ϕ(a) − ϕ(b) ≥ a − ϕ(a) + 1.
Let p be the minimal prime divisor of a (which exists as a ≥ 2). Since for all
multiples tp ≤ a of p, we have gcd(tp, a) > 1, it follows that a − ϕ(a) ≥ p1 · a.
Therefore we have
a
gcd(a, b) ≥ a − ϕ(a) + 1 ≥ + 1.
p
The two largest divisors of a are a and ap . Since gcd(a, b) is a divisor of a

that is at least ap + 1, it must equal a. Hence gcd(a, b) = a. In the same
way we prove that gcd(a, b) = b. So a = b.
The equation now is equivalent to 2a = 2ϕ(a) + a, so also to a = 2ϕ(a).
Note that 2 | a. Therefore write a = 2k · m with k ≥ 1 and m odd. By
a well-known property of the ϕ-function, we have ϕ(a) = ϕ(2k ) · ϕ(m) =
2k−1 ·ϕ(m), and the equation becomes 2k ·m = 2·2k−1 ·ϕ(m), or equivalently
30
m = ϕ(m). Therefore m = 1, and a = b = 2k . Indeed, the equation holds
for all pairs (2k , 2k ) with k ≥ 1.
Therefore the solutions of the equation are: the pairs (1, p) and (p, 1) for
prime numbers p, and (2k , 2k ) for all positive integers k.
Y
B C
4. Let M the circumcentre of 4ABC and let α = ∠BAC. We first show that
4BY A ∼ 4BM P and then that 4Y BM ∼ 4ABP . By the inscribed
angle theorem we have ∠BM C = 2∠BAC = 2α. Quadrilateral P BM C is a
kite with axis of symmetry P M (by the equality of radii |M B| = |M C| and
equality of tangent segments |P B| = |P C|), so M P bisects angle ∠BM C.
Therefore ∠BM P = 12 ∠BM C = α. Moreover, we have ∠P BM = 90◦
(tangent to a circle is perpendicular to its radius), so by the sum of angles
of a triangle we have ∠M P B = 90◦ − α.
On the other hand, we are given that ∠ABY = 90◦ and we also have
∠Y AB = ∠Y AC − ∠BAC = 90◦ − α. Therefore ∠ABY = ∠P BM and
∠Y AB = ∠M P B, from which follows that 4BY A ∼ 4BM P . From this
similarity it follows that |Y B| |M B|
|AB| = |P B| . Combining this with the equality of
angles
∠Y BM = ∠Y BA+∠ABM = 90◦ +∠ABM = ∠ABM +∠M BP = ∠ABP,
we see that 4Y BM ∼ 4ABP .
Let T now be the intersection of AP and Y M , then we have
∠BY T = ∠BY M = ∠BAP = ∠BAT,
from which it follows that BY AT is a cyclic quadrilateral. Therefore
∠AT Y = ∠ABY = 90◦ , so AT ⊥ Y M . Analogously, AP ⊥ XM . But
from this it now follows that Y M and XM coincide and we get that
AP ⊥ XY .
31
1. For a positive integer n, let d(n) be the number of positive divisors of n.
Determine the positive integers k for which there exist positive integers a
and b satisfying
k = d(a) = d(b) = d(2a + 3b).
2. Let a triangle ABC such that |AC| < |AB| be given, together with its
circumcircle. Let D be a varying point on the short arc AC. Let E be the
reflection of A in the internal angular bisector of ∠BDC. Prove that the
line DE passes through a fixed point, independent of where D lies.
3. Find all functions f : Z → Z satisfying

f −f (x) − f (y) = 1 − x − y
for all x, y ∈ Z.
4. Suppose k and n are positive integers such that k ≤ n ≤ 2k − 1. Julian has

a large pile of rectangular k × 1-tiles. Merlijn picks a positive integer m, and
receives from Julian m tiles to place on an n × n-board. On each tile, Julian
writes whether this tile should be placed horizontally or vertically. Tiles
may not overlap on the board, and they must fit entirely inside the board.
What is the largest number m that Merlijn can pick while still guaranteeing
he can put all tiles on the board according to Julian’s instructions?
32
Solutions
1. For i ≥ 0, let a = 2 · 5i and b = 3 · 5i . Then both a and b have 2(i + 1)
divisors. Moreover, we have 2a + 3b = 4 · 5i + 9 · 5i = 13 · 5i , which also has
2(i + 1) divisors. Therefore all even values of k satisfy the condition in the
problem.
Now suppose that k is odd. Then a has an odd number of divisors and
therefore is a square, say a = x2 . The same reasoning shows that b is also a
square, say b = y 2 , and 2a + 3b is also a square, say 2a + 3b = z 2 . Therefore
we have
2x2 + 3y 2 = z 2 .
We show that this equation has no positive integer solutions.
Suppose for a contradiction that this equation does have a positive integer
solution. Let (x, y, z) = (u, v, w) be the solution with minimal x + y + z. So
2u2 + 3v 2 = w2 . Modulo 3 this equation is 2u2 ≡ w2 . If u is not divisible
by 3, then u2 ≡ 1 mod 3, so w2 = 2u2 ≡ 2 mod 3, however, that isn’t
possible. Therefore u is divisible by 3, from which it follows that w is
divisible by 3 as well. Now 2u2 and w2 are both divisible by 9, so 3v 2 is
divisible by 9. It follows that v is divisible by 3 as well. However, now
(x, y, z) = ( u3 , v3 , w3 ) also satisfies the equation, and it is a solution with
smaller x + y + z than the supposed minimal one. This is a contradiction.
Therefore the equation 2x2 + 3y 2 = z 2 has no positive integer solutions.
It follows that no odd k satisfies the condition in the problem. The positive
integers k that satisfy that condition are therefore the even integers.
2. Let M be the intersection of the in-

E
ternal angular bisector of ∠BDC with
the circumcircle of 4ABC. As D C
lies on the short arc AC, we see that

M lies on the arc BC not containing D
A. We have ∠BDM = ∠M DC as
DM is the internal angular bisector
of ∠BDC, so arcs BM and CM have
B
the equal lengths. Hence M is inde- A
pendent of D.
Let S be the intersection of DE and
the circumcircle of 4ABC. We show
that S is independent of D. As S and M lie on the circumcircle of 4ABC,
we have ∠AM D = ∠ASD = ∠ASE. As E is the reflection of A in DM ,
we now see that ∠AM E = 2∠AM D = 2∠ASE.
33
Consider the circle with centre M passing through A. As E is the reflection
of A in DM , we have |M A| = |M E|, so this circle also passes through E.
By the inscribed angle theorem, from ∠AM E = 2∠ASE it follows that S
is also on this circle. Therefore S is the second intersection point of the
circumcircle of 4ABC and the circle with centre M passing through A.
This is a description of S independent of D. As DE passes through S, the
point S is the point required.
3. Substituting x = y = 1 yields f (−2f (1)) = −1. Substituting x = n and

y = 1 yields f (−f (n) − f (1)) = −n. Substituting x = −f (n) − f (1) and
y = −2f (1) then yields

f −f (−f (n) − f (1)) − f (−2f (1)) = 1 − (−f (n) − f (1)) − (−2f (1))
in which the left hand side expands as f (−(−n) − (−1)) = f (n + 1) and
the right hand as 1 + f (n) + f (1) + 2f (1) = f (n) + 3f (1) + 1. Writing
c = 3f (1) + 1, we then get f (n + 1) = f (n) + c.
Applying induction in both directions, we see that f (n + k) = f (n) + ck
for all k ∈ Z. Substituting n = 0 then yields f (k) = f (0) + ck for all k ∈ Z,
so f is a linear function.
Now let a, b ∈ Z be such that f (x) = ax + b for all x ∈ Z. Then the left
hand side of the functional equation evaluates as
f −f (x) − f (y) = a(−ax − b − ay − b) + b = −a2 x − a2 y − 2ab + b.

For all x and y this must be equal to 1 − x − y. Therefore the coefficient

for x on both sides must be equal (for fixed y both sides must give the
same function in x), so −a2 = −1, and therefore a = 1 or a = −1. If
a = −1, substituting x = y = 0 yields 2b + b = 1, which contradicts b
being an integer. If a = 1, substituting x = y = 0 yields −2b + b = 1, so
b = −1. Indeed, if a = 1 and b = −1, then the left hand side also expands
to 1 − x − y. Therefore the only solution to the functional equation is the
function f (x) = x − 1.
4. We show that the largest m Merlijn can pick is min(n, 3(n − k) + 1). First
we show that m ≤ min(n, 3(n − k) + 1). If Merlijn asks for n + 1 tiles,
Julian can instruct Merlijn to place them all horizontally. As n ≤ 2k − 1,
it is impossible to place more than one tile horizontally on a single row, so
Merlijn would need at least n + 1 rows, this is a contradiction. Therefore
m ≤ n.
Now suppose that Merlijn asks for 3(n − k)+2 tiles. Julian can now instruct
Merlijn to place n − k + 1 tiles vertically and 2n − 2k + 1 tiles horizontally.
34
Note that vertical tiles always cover the k − (n − k) = 2k − n ≥ 1 rows in
the middle of the board. Therefore the vertical tiles together cover at least
n − k + 1 squares in each of these rows in the middle of the board, leaving at
most k − 1 squares for the horizontal tiles. Therefore no horizontal tiles fit
in these middle rows, and the 2n − 2k + 1 horizontal tiles need to fit in the
remaining n − (2k − n) = 2n − 2k rows. This is a contradiction. Therefore
we must have m ≤ 3(n − k) + 1. It follows that m ≤ min(n, 3(n − k) + 1).
Now we show that it is always possible to place a pile of min(n, 3(n + k) + 1)
tiles on the board. We first consider two special configurations. Put n − k
horizontal tiles into a (n − k) × k-rectangle at the bottom left. To the right
of that, we can fit n − k more vertical tiles into a k × (n − k)-rectangle in
the bottom right. Above that rectangle, we can fit n − k more horizontal
tiles into a (n − k) × k-rectangle in the top right. Finally, to the left of
that, we can fit n − k more vertical tiles into a k × (n − k)-rectangle in the
top left. In this way, we can fit 2(n − k) horizontal and 2(n − k) vertical
tiles on the board.
One other way to cover the board is as follows. Place n vertical tiles in
the top left, covering a k × n-rectangle. Below that there is room for n − k
horizontal tiles to fit in an (n − k) × k-rectangle. In this way, we can fit n
vertical and n − k horizontal tiles on the board.
n−k n−k
k k
Suppose that Merlijn receives A horizontal tiles and B vertical tiles from
Julian. Then we have A + B ≤ n and A + B ≤ 3(n − k) + 1. Without
loss of generality we assume that A ≤ B. If A ≤ n − k, then Merlijn
uses the second special configuration, omitting tiles he doesn’t have. As
B ≤ n, this works. Else, A ≥ n − k + 1, so B ≤ 3(n − k) + 1 − A ≤
3(n − k) + 1 − (n − k + 1) = 2(n − k). As A ≤ B, we also have A ≤ 2(n − k),
and Merlijn can use the first configuration, omitting tiles he doesn’t have.
Therefore it is always possible for Merlijn to place min(n, 3(n − k) + 1) tiles
on the board.
35
Problems
Part 1
1. At a conference, there were participants from four countries: the Nether-

lands, Belgium, Germany, and France. There were three times as many
participants from the Netherlands as there were Belgians, and three times
as many Germans as French. Five of the participants counted the total
number of participants (including themselves). They counted 366, 367, 368,
369, and 370 participants, respectively. Only one of them got the right
answer.
What is the the correct number of participants?
A) 366 B) 367 C) 368 D) 369 E) 370
2. We completely cover a big isosceles triangle

with triangles that are similar to the big tri-
angle as in the figure on the right.
What part of the area of the big triangle is
covered by the top triangle (indicated in grey)?
1 2 5 16 1
A) 4 B) 7 C) 16 D) 49 E) 3
3. In the puzzle below, a, b, c, d, and e are nonzero digits such that the two
calculations are correct. The digits need not be distinct.
How many solutions are there for which a < b?
ab × ab = cde and ba × ba = edc.
A) 1 B) 2 C) 3 D) 4 E) 5
4. We say that a number is a child of another number if we can get it by

placing between any two digits of the other number either nothing, a +, or
a ×. For example, 145 and 5 are children of 12121 because 145 = 12 × 12 + 1
and 5 = 1 + 2 × 1 + 2 × 1. The number 15 is both a child of 12121 and of
33333, because 12 + 1 + 2 × 1 = 15 = 3 + 3 + 3 + 3 + 3.
Which of the following numbers is also a child of both 12121 and 33333?
A) 18 B) 34 C) 39 D) 42 E) 45
36
5. In a class, 30 students did a test. Every student got a grade that was an
integer from 1 to 10. The grade 8 was given more often than any of the
other grades.
What is the smallest possible average grade of the students?
8
A) 3 15 B) 3 23 C) 3 56 D) 4 11
30
8
E) 4 15
6. We want to colour the 36 squares of a 6 × 6 board. Every square must be

coloured white, grey, or black, and the following requirement must be met:
Three adjacent squares in the same row or column, must always have
three different colours.
We say that two colourings are truly different if you cannot get one from
the other by rotating the board. Below, you can see three colourings that
meet the requirement. The first and second colouring are truly different,
but the third is the same as the second after rotating.
How many truly different colourings meet the requirement (including the
two from the figure)?
A) 2 B) 3 C) 4 D) 6 E) 12
7. Point D lies on side BC of triangle ABC. Angle A in triangle ABD is

equal to angle C in triangle ABC, and angle A in triangle ACD is equal
to angle B in triangle ABC.
The given information is not enough to derive the exact shape of triangle
ABC. However, you can still derive that one of the given statements below
is always false. Which statement is it?
By |AB| we denote the length of line segment AB.
A) |AD| < |AC| D) |AD|×|CD| < |AB|×|AC|
B) |AC| < |AB| E) |AB|×|AC| < |AD|×|BC|
C) |AB| < |BC|
8. Five smart students are sitting in a circle. The teacher gives one or more
marbles to each of them. He explains that he has handed out a total of
37
18 marbles, and that everyone got a different number of marbles. Each
student is allowed to see his own number of marbles, as well as the number
of marbles of his neighbour on the left and his neighbour on the right.
Using only this information, each student must try to logically deduce the
difference between the numbers of marbles of the two students opposite to
him. The teacher has distributed the marbles in such a way as to minimise
the number of students that are able to do this. How many students can
do it?
A) 0 B) 1 C) 2 D) 3 E) 5
Part 2
1. We compute the square of each of the numbers from 1 to 2019. We take

the last digit from each of the resulting squares, and then we add those
2019 digits together.
What number do we get?
2. The numbers abcd and dcba consist of the same four digits a, b, c, and d,
but in opposite orders. When we add the two numbers, we get 13552.
Determine a + b + c + d.
3. One hundred students wear shirts numbered from 1 to 100. The students
are arranged in a square of ten rows by ten columns. It turns out that
adding the ten shirt numbers of the students in any row or any column
always yields the same outcome.
Determine that outcome.
4. The four points A, B, C, and D lie on a common line (in this order). There
is a point T not on the line such that |AB| = |BT | and |CD| = |CT |. Also,
angle T of triangle BT C is 54 degrees. In the figure you can see a sketch
of the situation; angles and sizes are not necessarily accurate.
Determine angle T of triangle AT D.
T
54◦
A B C D
38
5. Mieke has a stack of 21 cards. Mieke repeats the following operation:
She takes the top two cards from the stack, changes their order, and
then puts them at the bottom of the stack (so the top card becomes
the bottom card).
Mieke repeats this operation until the cards are back in their original order.
How many times does Mieke perform the operation?
6. The number
· · · 93} 919
2222 |9393{z
100 times 93
is divided by 2019.
Determine the sum of the digits of the resulting number.
7. On a white strip that is 100 mm long and 10 mm wide, ten black squares
are drawn that, from left to right, have sides of length 1, 2, . . . , 10 mm. The
centre of each black square is in the middle of the strip and 5, 15, . . . , 95
mm from the start (left edge) of the strip.
A transparent square is moving along the strip from left to right (indicated
in grey). In the figure, two possible positions of the transparent square are
depicted: in the first, its left edge is 21 mm from the start of the strip, and
in the second, it is 54 mm from the start of the strip. In both cases, less
than half of the part underneath the square is coloured black (only 9% in
the first case). There is one position in which exactly half of the part of
the strip underneath the square is coloured black.
Determine, for that position, how far the left edge of the square is from
the start of the strip.
39
8. We are given a triangle with an additional two points on each side. So in
total, there are nine points (see figure).
We want to choose three of the nine points that are not on one line. For
example, we could choose (1) the three vertices of the triangle, or (2) the
left vertex and the two additional points on the opposite side.
How many possible choices are there in total, including the two examples
given?
Answers
Part 1
1. C) 368 5. B) 3 23
16
2. D) 49 6. B) 3
3. B) 2 7. E) |AB|×|AC| < |AD|×|BC|
4. D) 42 8. D) 3
Part 2
1. 9090 5. 110
2. 26 6. 103
3. 505 7. 61 18 mm
4. 117 graden 8. 72

and the team selection for IMO 2020 Russia

We eat problems
for breakfast.

Olympiad 2020
International
NEDERLANDSE
WISKUNDE
OLYMPIADE
Contents
1 Introduction

Introduction

In February a team of four girls was chosen from the training group to repre-
sent the Netherlands at the EGMO in Georgia, from 9 until 15 April. At this
virutal event the Dutch team won one bronze medal. For more information
about the EGMO (including the 2021 paper), see www.egmo.org.
In March a selection test of 3.5 hours was held to determine the ten students
participating in the Benelux Mathematical Olympiad (BxMO), also a virtual
event held on 1 and 2 May. The Dutch team achieved an outstanding result:
three gold medals, two silver medals and three bronze medals. For more
information about the BxMO (including the 2021 paper), see www.bxmo.org.
selected by three team selection tests on 2, 3 and 4 June, each lasting 4
1
hours. A seventh, young, promising student was selected to accompany
Egmond aan Zee from 10 until 18 July.
the solutions.
2
Dutch delegation
The Dutch team for the virtual IMO 2021 consists of
Jelle Bloemendaal (17 years old)

– bronze medal at BxMO 2019, silver medal at BxMO 2020 and
2021
– (virtual) observer C at IMO 2020
Kevin van Dijk (17 years old)
Hylke Hoogeveen (16 years old)
– bronze medal at BxMO 2020, honourable mention at BxMO 2021
Casper Madlener (16 years old)
Kees den Tex (17 years old)
– gold medal at BxMO 2021
Thian Tromp (18 years old)
Also part of the IMO selection, but not officially part of the IMO team, is:
Lars Pos (17 years old)

Quintijn Puite (team leader), Eindhoven University of Technology

Johan Konter (deputy leader), Stockholm University
Ward van der Schoot (observer A), University of Cambridge
3
Problems
A-problems
1. Francisca has a square piece of paper whose sides

have length 10 cm. She also has a rectangular piece
of paper having the exact same area as the square
piece of paper. She puts the rectangle right on top
of the square, putting the left bottom corner of both
pieces of paper in the same spot. Exactly one quarter
of the square remains uncovered by the rectangle.
What is the length in centimetres of the long side of
the rectangle?
A) 12 B) 12 14 C) 12 12 D) 12 34 E) 13 13
2. Each of Kwik, Kwek, and Kwak is lying on two consecutive days of the week
and is telling the truth on the other five days. No two of them are lying
on the same day. Uncle Donald wants to know who of his nephews ate his
sweets. The three nephews know all too well who did it. On Sunday, Kwik
says that Kwek ate the sweets. On Monday, Kwik says that it actually was
not Kwek who ate the sweets, while Kwak claims that Kwik is innocent.
On Tuesday, however, Kwak says that it was Kwik who ate the sweets.
Who ate the sweets?
A) It was Kwik.
B) It was Kwek.
C) It was Kwak.
D) It was either Kwik or Kwek, but you cannot determine who of the
two.
E) It was either Kwik or Kwak, but you cannot determine who of the
two.
3. We consider numbers with two digits (the first digit cannot be 0). Such a
number is called vain if the sum of the two digits is greater than or equal
to the product of the two digits. For example, the number 36 is not vain,
as 3 + 6 is smaller than 3 · 6.
How many numbers with two digits are vain?
A) 17 B) 18 C) 26 D) 27 E) 37
4
4. A box measuring 4 dm by 15 dm is shoved against
6
the wall. On top of it, a second box, measuring 12
dm by 6 dm, is placed. A ladder exactly touches the 12
ground, the two boxes and the wall. See the figure 4
(which is not drawn to scale). 15
What is the length of the ladder in dm?
√ √ √
A) 30 B) 8 15 C) 31 D) 22 2 E) 18 3
5. On a 4×4 board, there are 16 grass hoppers, each on its own square. At a
certain time, each grass hopper jumps to an adjacent square: to the square
above, below, left, or right of its current square, but not diagonally and
not leaving the board.
What is the maximum number of squares that can be empty after the grass
hoppers have jumped?
A) 8 B) 9 C) 10 D) 11 E) 12
6. In the table below each of the three rows is a correct calculation (the symbol
÷ denotes division). Also each of the three columns (read from top to
bottom) is a correct calculation. However, the digits in the table have been
replaced by letters. Different letters represent different digits and no digits
are 0.
ABC − ADF = F
+ − −
ADD ÷ GC = C
= = =
CEF ÷ GD = D
Which digit does E represent?

A) 1 B) 3 C) 5 D) 7 E) 9
5
7. We consider figures consisting of six squares whose
sides have length 1. The radius of such a figure is
the radius of the smallest circle containing the whole
figure. On the
√ right, there is an example of a figure
with radius 5.
Which of the following five figures has the smallest radius?
A) A B) B C) C D) D E) E
A B C D E
8. Lieneke is making bracelets with beads. Each bracelet has six beads: two
white, two grey, and two black beads. Some bracelets look different on first
sight, but are actually not different: by turning or flipping the first one
over, it looks the same as the other one. For example, the following three
bracelets are the same.
How many really different bracelets can Lieneke make?

A) 10 B) 11 C) 12 D) 14 E) 15
6
B-problems
1. By replacing each ∗ in the expression 1 ∗ 2 ∗ 3 ∗ 4 ∗ 5 ∗ · · · ∗ 2019 ∗ 2020 by a

+ or a − sign, we get a long calculation. Put the + and − signs in such
a way that the outcome is a positive number (greater than 0) which is as
small as possible.
What is this outcome?
2. Triangle ABC is subdivided into three isosceles triangles and a rhombus.

Note: the figure is not drawn to scale.
B C
What is the size of angle C in degrees?
3. Annemiek and Bart each have a note on which they have written three
different positive integers. It appears that there is exactly one number that
is on both their notes. Moreover, if you add any two different numbers
from Annemiek’s note, you get one of the numbers on Bart’s note. One
of the numbers on Annemiek’s note is her favourite number, and if you
multiply it by 3, you get one of the numbers on Bart’s note. Bart’s note
contains the number 25, his favourite number.
What is Annemiek’s favourite number?
4. We consider rows of 2020 coins. Each coin is of denomination 1, 2, or

3. Between two coins of denomination 1, there is at least one other coin.
Between two coins of denomination 2, there are at least two other coins.
Between two coins of denomination 3, there are at least three other coins.
How many different rows of 2020 coins satisfy these conditions?
7
Solutions
A-problems
1. E) 13 31 5. C) 10
2. C) It was Kwak. 6. E) 9
3. D) 27 7. B) B
4. A) 30 8. B) 11
B-problems
1. 2
2. 36
3. 5
4. 10
8
Problems
B-problems
B1. The digit sum of a number is obtained by adding all digits of the number.
For example, the digit sum of 1303 is 1 + 3 + 0 + 3 = 7.
Find the smallest positive integer n for which both the digit sum of n and
the digit sum of n + 1 are divisible by 5.
J
D F C
B2. Rectangle ABCD is subdivided into four rectangles as in
the figure. The area of rectangle AEIG is 3, the area of G
I
H
rectangle EBHI is 5, and the area of rectangle IHCF A E B
is 12.
What is the area of the parallelogram AHJF ?
B3. A square sheet of paper lying on the table is divided into 8×8 = 64 equal
squares. These squares are numbered from a1 to h8 as on a chess board
(see fig. 1). We now start folding, in such a way that square a1 always stays
in the same spot on the table. First we fold along the horizontal midline
(fig. 1). This will cause square a8 to fold on top of square a1. Then we fold
along the vertical midline (fig. 2). Next, we fold along the new horizontal
midline (fig. 3), et cetera. After folding six times, we have a small package
of paper in front of us (fig. 7) that we can consider as a stack of 64 square
pieces of paper.
8
7
6
5
4 4 4
3 3 3
2 2 2 ···
1 1 1 1
a b c d e f g h a b c d e f g h a b c d a
fig. 1 fig. 2 fig. 3 fig. 7
The squares in this stack are numbered from bottom to top from 1 to 64.
So square a1 gets number 1.
Which number does square f6 get?
9
B4. One hundred brownies (girl scouts) are sitting in a big circle around the
camp fire. Each brownie has one or more chestnuts and no two brownies
have the same number of chestnuts. Each brownie divides her number of
chestnuts by the number of chestnuts of her right neighbour and writes
down the remainder on a green piece of paper. Each brownie also divides
her number by the number of chestnuts of her left neighbour and writes
down the remainder on a red piece of paper. For example, if Anja has 23
chestnuts and her right neighbour Bregje has 5, then Anja writes 3 on her
green piece of paper and Bregje writes 5 on her red piece of paper.
If the number of distinct remainders on the 100 green pieces of paper equals
2, what is the smallest possible number of distinct remainders on the 100
red pieces of paper?
B5. Given is the sequence of numbers a0 , a1 , a2 , . . . , a2020 with a0 = 0. Further-

more, the following holds for every k = 1, 2, . . . , 2020:
(
ak−1 · k if k is divisible by 8,
ak =
ak−1 + k if k is not divisible by 8.
What are the last two digits of a2020 ?
10
C1. Given a positive integer n, we denote by n! (‘n factorial’) the number we get
if we multiply all integers from 1 to n. For example: 5! = 1 · 2 · 3 · 4 · 5 = 120.
(a) Determine all integers n with 1 ≤ n ≤ 100 for which n! · (n + 1)! is a
perfect square. Also, prove that you have found all solutions n.
(b) Prove that no positive integer n exists such that n! · (n + 1)! · (n + 2)! ·
(n + 3)! is a perfect square.
C2. Three consecutive vertices A, B, and C of a regular octagon (8-gon) are the
centres of circles that pass through neighbouring vertices of the octagon.
The intersection points P , Q, and R of the three circles form a triangle
(see figure).
B
A C A C
P
Q R
P
Q R
Prove that triangle P QR is equilateral.
11
Solutions
B-problems
1. 49999
2. 24 51
3. 43
4. 100
5. 02
C-problems
C1. (a) We observe that (n + 1)! = (n + 1) · n!, and therefore that n! · (n + 1)! =
(n!)2 · (n + 1). That product is a perfect square if and only if n + 1 is
a perfect square, since (n!)2 is a perfect square. For 1 ≤ n ≤ 100 this
is the case for n = 3, 8, 15, 24, 35, 48, 63, 80, 99 (perfect squares minus
one that are below 100).
(b) We rewrite the product n! · (n + 1)! · (n + 2)! · (n + 3)! as follows:
(n!)2 · (n + 1) · (n + 2)! · (n + 3)! = (n!)2 · (n + 1) · ((n + 2)!)2 · (n + 3).
Since (n!)2 and ((n + 2)!)2 are both perfect squares, the above product
is a perfect square if and only if (n + 1)(n + 3) is a perfect square.
However, (n + 1)(n + 3) cannot be a perfect square. Indeed, suppose
that (n + 1)(n + 3) = k 2 were a perfect square. Since (n + 1)2 <
(n + 1)(n + 3) < (n + 3)2 we would have n + 1 < k < n + 3, so k = n + 2.
This is impossible because (n + 1)(n + 3) = (n + 2)2 − 1, which is not
equal to (n + 2)2 .
C2. An octagon can be subdivided into six triangles (see figure on the left).
Together, the angles of those six triangles add up to the same number of
degrees as the eight angles of the octagon. Since the angles of any triangle
add up to 180 degrees, this means that the eight angles of the octagon add
up to 6 · 180◦ = 1080◦ . Hence, each of the angles of the regular octagon is
1 ◦ ◦
8 · 1080 = 135 .
12
We now consider the figure from the problem statement (see figure on the
◦
right). Line segment BP bisects angle ABC, so ∠ABP = ∠P BC = 67 12 .
Since triangles ABP and BCP are isosceles (as |AB| = |AP | and |BC| =
◦
|CP |), we also have ∠AP B = ∠BP C = 67 12 and ∠BAP = ∠BCP =
180◦ − 135◦ = 45◦ .
In triangles ABQ and BCR all sides have the same length. These triangles
are therefore equilateral and all angles are 60◦ . From this, we deduce that
∠P AQ = ∠BAQ − ∠BAP = 15◦ . In the same way, we find ∠P CR = 15◦ .
Furthermore, triangles P AQ and P CR are isosceles (since |AP | = |AQ|
◦ ◦
and |CP | = |CR|), so ∠AP Q = 12 (180◦ − 15◦ ) = 82 12 and ∠CP R = 82 12 .
By mirror symmetry, P Q and P R have the same length, so P QR is an
isosceles triangle with apex P . We have already determined all angles at
P , except ∠QP R. We deduce that
∠QP R = 360◦ − ∠AP Q − ∠AP B − ∠BP C − ∠CP R

◦ ◦
= 360◦ − 2 · 67 12 − 2 · 82 12 = 60◦ .
From this and the fact that P QR is isosceles, we directly conclude that
P QR is equilateral.
A C
P
Q R
13
Problems
1. Daan distributes the numbers 1 to 9 over the nine squares

8 1 2
of a 3×3-table (each square receives exactly one number).
Then, in each row, Daan circles the median number (the 7 6 3
number that is neither the smallest nor the largest of 9 5 4
the three). For example, if the numbers 8, 1, and 2 are
in one row, he circles the number 2. He does the same for each column and
each of the two diagonals. If a number is already circled, he does not circle
it again.
He calls the result of this process a median table. Above, you can see a
median table that has 5 circled numbers.
(a) What is the smallest possible number of circled numbers in a median

table?
Prove that a smaller number is not possible and give an example in
which a minimum number of numbers is circled.
(b) What is the largest possible number of circled numbers in a median
table?
Prove that a larger number is not possible and give an example in
which a maximum number of numbers is circled.
2. For a given value t, we consider number sequences a1 , a2 , a3 , . . . such that

an + t
an+1 = for all n ≥ 1.
an + 1
(a) Suppose that t = 2. Determine all starting values a1 > 0 such that
4 3
3 ≤ an ≤ 2 holds for all n ≥ 2.
(b) Suppose that t = −3. Investigate whether a2020 = a1 for all starting
values a1 different from −1 and 1.
14
3. Given is a parallelogram ABCD with ∠A < 90◦ and |AB| < |BC|. The
angular bisector of angle A intersects side BC in M and intersects the
extension of DC in N . Point O is the centre of the circle through M , C,
and N .
Prove that ∠OBC = ∠ODC.
D C N
O
M
A B
4. Determine all pairs of integers (x, y) such that 2xy is a perfect square and
x2 + y 2 is a prime number.
5. Sabine has a very large collection of shells. She decides to give part of her
collection to her sister.
On the first day, she lines up all her shells. She takes the shells that are in
a position that is a perfect square (the first, fourth, ninth, sixteenth, etc.
shell), and gives them to her sister. On the second day, she lines up her
remaining shells. Again, she takes the shells that are in a position that is a
perfect square, and gives them to her sister. She repeats this process every
day.
The 27th day is the first day that she ends up with fewer than 1000 shells.
The 28th day she ends up with a number of shells that is a perfect square
for the tenth time.
What are the possible numbers of shells that Sabine could have had in the
very beginning?
15
Solutions
1. (a) The smallest possible number of circled numbers is 3. Fewer than 3

is not possible since in each row at least one number is circled (and
these are three different numbers).
On the right, a median table is shown in which only 3
numbers are circled. In the rows, the numbers 7, 5, 3 are 4 9 7
circled, in the columns the numbers 3, 5, 7, and on the 2 5 8
diagonals the numbers 5 and 5. Together, these are three 3 1 6
different numbers: 3, 5, and 7.
(b) The largest possible number of circled numbers is 7. More than 7 is
not possible, since the numbers 9 and 1 are never circled, hence no
more than 9 − 2 = 7 numbers are circled.
On the right, a median table is shown in which 7 numbers
are circled. In the rows, the numbers 2, 6, 8 are circled, 4 1 2
in the columns the numbers 7, 5, 3, and on the diagonals 7 5 6
the numbers 4 and 5. Together, these are the numbers 8 9 3
2, 3, 4, 5, 6, 7, 8.
2. (a) First, we determine for what starting values a1 > 0 the inequalities
4 3
3 ≤ a2 ≤ 2 hold. Then, we will prove that for those starting values,
the inequalities 43 ≤ an ≤ 32 are also valid for all n ≥ 2.
First, we observe that a2 = aa11 +2
+1 and that the denominator, a1 + 1, is
positive (since a1 > 0). The inequality
4 a1 + 2 3
≤ a2 = ≤ ,
3 a1 + 1 2
is therefore equivalent to the inequality
4
3 (a1 + 1) ≤ a1 + 2 ≤ 32 (a1 + 1),
as we can multiply all parts in the inequality by the positive number

a1 + 1. Subtracting a1 + 2 from all parts of the inequality, we see that
this is equivalent to
1 2
3 a1 − 3 ≤ 0 ≤ 12 a1 − 12 .
We therefore need to have 13 a1 ≤ 23 (i.e. a1 ≤ 2), and 12 ≤ 12 a1 (i.e.

1 ≤ a1 ). The starting value a1 must therefore satisfy 1 ≤ a1 ≤ 2.
Now suppose that 1 ≤ a1 ≤ 2, so that a2 satisfies 43 ≤ a2 ≤ 23 . Looking
at a3 , we see that a3 = aa22 +2
+1 . That is the same expression as for a2 ,
16
only with a1 replaced by a2 . Since a2 also satisfies 1 ≤ a2 ≤ 2, the
same argument now shows that 43 ≤ a3 ≤ 32 .
We can repeat the same argument to show this for a4 , a5 , etcetera.
Hence, we find that 43 ≤ an ≤ 32 holds for all n ≥ 2. The formal proof
is done using induction: the induction basis n = 2 has been shown
above. For the induction step, see the solution of part (b) of the
version for klas 5 & klas 4 and below. The result is that all inequalities
hold if and only if 1 ≤ a1 ≤ 2.
(b) Let’s start by computing the first few numbers of the sequence in
terms of a1 . We see that
a1 − 3
a2 =
a1 + 1
and
a1 −3
a2 − 3 a1 +1 −3 a1 − 3 − 3(a1 + 1)
a3 = = a1 −3 =
a2 + 1 +1
a1 +1
a1 − 3 + (a1 + 1)
−2a1 − 6 −a1 − 3
= = .
2a1 − 2 a1 − 1
Here, it is important that we do not divide by zero, that is, a1 = 6 −1
and a2 6= −1. The first inequality follows directly from the assumption.
For the second inequality we consider when a2 = −1 holds. This is
the case if and only if a1 − 3 = −(a1 + 1), if and only if a1 = 1. Since
we assumed that a1 6= 1, we see that a2 6= −1. The next number in
the sequence is
−a1 −3
a3 − 3 a1 −1 −3 −a1 − 3 − 3(a1 − 1) −4a1
a4 = = −a1 −3 = = = a1 .
a3 + 1 a1 −1 +1 −a1 − 3 + (a1 − 1) −4
Again, we are not dividing by zero since a3 = −1 only holds when

−a1 − 3 = −a1 + 1, which is never the case.
We see that a4 = a1 . Since an+1 only depends on an , we see that
a5 = a2 , a6 = a3 , a7 = a4 , et cetera. In other words: the sequence is
periodic with period 3, and we see that
a2020 = a2017 = a2014 = · · · = a4 = a1 .
To conclude: indeed we have a2020 = a1 for all starting values a1
unequal to 1 and −1.
3. As an intermediate step, we first show that triangles OCM and OCN are
congruent. Since AD and BC are parallel, we have (F angles): ∠CM N =
17
∠DAM = 12 ∠DAB. Since DN and AB are parallel, we have (Z angles):
∠CN M = ∠N AB = 12 ∠DAB. It follows that ∠CM N = ∠CN M , so
triangle CM N is isosceles with apex C. We obtain |CM | = |CN |. Line
segments OC, ON , and OM are radii of the same circle, and therefore of
equal length. Triangles OCM and OCN are therefore congruent (three
pairs of equal sides).
To show that ∠OBC = ∠ODC, we will show that triangles OBC and
ODN are congruent. We will do this using the ZHZ-criterion. We will
show that ∠ON D = ∠OCB, and |ON | = |OC|, and |DN | = |BC|.
The equality |ON | = |OC| follows since ON and OC are radii of the
same circle. In part (a), we saw triangles OCM and OCN are congruent.
Furthermore, these two triangles are isosceles (|OC| = |OM | and |OC| =
|ON |). Hence, the four base angles ∠ON C, ∠OCN , ∠OM C, and ∠OCM
are equal. We see that ∠ON D = ∠OCB. The only thing we still need to
show is that |DN | = |BC|.
Observe that ∠BM A = ∠DAM (Z angles) and ∠DAM = ∠M AB (as AM
is the angular bisector of A). We find that ∠BM A = ∠M AB. Triangle
AM B is therefore isosceles and we have |AB| = |BM |. We previously
saw that |CM | = |CN |, and we also have |AB| = |CD| as ABCD is a
parallelogram. We therefore obtain
|DN | = |CD| + |CN | = |AB| + |CM | = |BM | + |CM | = |BC|,
which concludes the proof.
4. We have 2xy = a2 for some nonnegative integer a, and x2 + y 2 = p for

some prime number p.
Since a prime number is never a perfect square, we see that x, y =6 0. Since
2xy is a perfect square, it follows that x and y must both be positive, or
both be negative. If (x, y) is a solution, then so is (−x, −y). Therefore, we
may for now assume that x and y are positive, and at the end, add for each
solution (x, y) the pair (−x, −y) to the list of solutions.
Combining 2xy = a2 and x2 +y 2 = p yields (x+y)2 = x2 +y 2 +2xy = p+a2 .
By bringing a2 to the other side, we find
p = (x + y)2 − a2 = (x + y + a)(x + y − a).
Since x + y + a is positive, also x + y − a must be positive. The prime

number p can be written as a product of two positive integers in only two
ways: 1 · p and p · 1. Since x + y + a ≥ x + y − a, we obtain x + y + a = p
and x + y − a = 1.
18
Adding these two equations, we get 2x + 2y = p + 1. We also know that
x2 + y 2 + 1 = p + 1, so 2x + 2y = x2 + y 2 + 1. By bringing all terms to the
right-hand side and adding 1 to both sides, we obtain
1 = x2 + y 2 − 2x − 2y + 2 = (x − 1)2 + (y − 1)2 .
We now have two perfect squares that add up to 1. This implies that one
of the squares is 0 and the other is 1. So (x − 1)2 = 0 and (y − 1)2 = 1, or
(x − 1)2 = 1 and (y − 1)2 = 0. As x and y are positive, we find two possible
solutions: x = 1 and y = 2, or x = 2 and y = 1. In both cases 2xy = 4 is a
perfect square and x2 + y 2 = 5 is a prime number. It follows that both are
indeed solutions.
Adding the solutions obtained by replacing (x, y) by (−x, −y), we obtain a
total of four solutions (x, y), namely
(1, 2), (2, 1), (−1, −2), (−2, −1).
5. Suppose that on a given day, Sabine is left with n2 shells, where n > 1.
Then the next day, she will give n shells to her sister and will be left with
n2 − n shells. This is more than (n − 1)2 , since
(n − 1)2 = n2 − 2n + 1 = (n2 − n) − (n − 1) < n2 − n
as n > 1. The next day, she therefore gives n − 1 shells to her sister and is
left with n2 − n − (n − 1) = (n − 1)2 shells, again a perfect square. We see
that the numbers of shells that Sabine is left with are alternately a perfect
square and a number that is not a perfect square.
Let d be the first day that Sabine is left with a number of shells that is
a perfect square, say n2 shells. Then days d + 2, d + 4, . . . , d + 18 are the
second to tenth day that the remaining number of shells is a perfect square
(namely (n−1)2 , (n−2)2 , . . . , (n−9)2 shells). We conclude that d+18 = 28,
and hence d = 10.
On day 26 the number of remaining shells is at least 1000, but on days 27
and 28 this number is less than 1000. We see that (n−9)2 < 1000 ≤ (n−8)2 .
As 312 < 1000 ≤ 322 , we see that n − 8 = 32, and hence n = 40. We find
that day 10 is the first day that the number of remaining shells is a perfect
square, and that this number is 402 .
In the remainder of the proof, we will use the following observation.
Observation. On any day, starting with more shells, means that Sabine
will have more (or just as many) shells left after giving shells to her sister.
19
Indeed, suppose that Sabine starts the day with x shells, say n2 ≤ x <
(n + 1)2 . After giving away shells, she will be left with x − n shells. If she
had started with x + 1 shells instead of x, she would have been left with
x + 1 − n > x − n or x + 1 − (n + 1) = x − n shells.
Let x be the number of shells remaining on day 8. The obvious guess

x = 412 = 1681 is incorrect as x cannot be a perfect square. We therefore
try x = 412 − 2, x = 412 − 1, and x = 412 + 1. The table shows the number
of shells remaining on day 8, 9, and 10.
day 8 day 9 day 10

412 − 2 = 1679 1679 − 40 = 1639 1639 − 40 = 1599
412 − 1 = 1680 1680 − 40 = 1640 1640 − 40 = 1600
412 + 1 = 1682 1682 − 41 = 1641 1641 − 40 = 1601
We see that the case x = 1679 is ruled out because it would imply that
fewer than 402 = 1600 shells are left on day 10. By the above observation,
this also rules out the case x < 1679. The case x = 1682 is ruled out
because it would imply that more than 402 shells will be left on day 10.
Hence, also x > 1682 is ruled out. The number of shells left on day 8 must
therefore be 412 − 1.
To follow the pattern back in time, we consider the case that the number
of remaining shells is just shy of a perfect square. Suppose that on a
given day the number of remaining shells is n2 − a, where 1 ≤ a < n.
Then the following day, the number of remaining shells is n2 − a − (n − 1).
Since a < n, we have n2 − a − (n − 1) > n2 − n − (n − 1) = (n − 1)2 .
The day after that, the number of remaining shells must therefore be
n2 − a − (n − 1) − (n − 1) = (n − 1)2 − (a − 1).
So if Sabine originally had 452 − 5 shells, then the number of remaining
shells on days 2, 4, 6, and 8 are 442 − 4, 432 − 3, 422 − 2, and 412 − 1,
respectively. This gives us a solution.
If Sabine originally had 452 − 4 shells, then she would be left with too
many shells on day 8, namely 412 − 0. The original number of shells could
therefore not have been 452 − 4 or more.
If Sabine originally had 452 − 6 shells, then she would be left with too
few shells on day 8, namely 412 − 2. The original number of shells could
therefore not have been 452 − 6 or fewer.
We conclude that the only possibility is that Sabine started with a collection
of 452 − 5 = 2020 shells.
20
Problems
1. Let ABCD be a cyclic quadrilateral with |AB| = |BC|. Point E lies on the
arc CD which does not contain A and B. The intersection of BE and CD
is denoted by P , the intersection of AE and BD is denoted by Q. Prove
that P Q k AC.
2. Determine all triples (x, y, z) of real numbers satisfying:
x2 − yz = |y − z| + 1,
y 2 − zx = |z − x| + 1,
z 2 − xy = |x − y| + 1.
3. Let p be a prime number greater than 2. Patricia wants to assign 7 not

necessarily distinct numbers of {1, 2, . . . , p} to the black dots in the figure
below, in such a way that the product of three numbers on a line or circle
always gives the same remainder upon divison by p.
(a) Suppose Patricia is using the number p at least once. How often does
she need to use the number p at least?
(b) Suppose that Patricia does not use the number p. In how many ways
can she assign the numbers? (Two ways are considered to be different
if to at least one black dot distinct numbers have been assigned. The
figure is not being turned or mirrored.)
4. Jesse and Tjeerd are playing a game. Jesse has n ≥ 2 stones. There are
two boxes: in the black box there is space for half of the stones (rounded
21
down) and in the white box there is space for half of the stones (rounded
up). Jesse and Tjeerd alternate turns, with Jesse as first player. In his
turn, Jesse takes one new stone, writes a positive real number on the stone
and puts it in one of the boxes which is not full yet. Tjeerd can see all
the numbers on the stones in each of the boxes and is allowed to move one
stone of his choice to the other box, if that other box is not full yet, but he
is also allowed to choose to do nothing. The game stops when both boxes
are full. If the total value of the stones in the black box is greater than
the total value of the stones in the white box, Jesse wins; otherwise Tjeerd
wins. Determine for each n ≥ 2 who can always win this game (and give a
winning strategy).
5. A triangle ABC has the property that |AB| + |AC| = 3|BC|. Let T be
the point on line segment AC satisfying |AC| = 4|AT |. Let K and L be
points on the interior of line segments AB and AC, respectively, such that
KL k BC, and KL is tangent to the incircle of 4ABC. Let S be the
|SL|
intersection of BT and KL. Determine the ratio |KL| .
22
Solutions
1. Because |AB| = |BC|, we have ∠AEB = ∠BDC, hence ∠QEP =
∠AEB = ∠BDC = ∠QDP , which yields that QP ED is a cyclic quadri-
lateral. Therefore, ∠QP D = ∠QED = ∠AED = ∠ACD. From this, we
get that QP and AC are parallel.
E
Q
P
A C
2. The system of equations is symmetric: if you swap x and y, for example,

then the third equation stays the same and the first two equations are
swapped. Hence, we can assume without loss of generality that x ≥ y ≥ z.
Then the system of equations becomes:
x2 − yz = y − z + 1,
y 2 − zx = x − z + 1,
z 2 − xy = x − y + 1.
Subtracting the second equation from the first, we obtain x2 −y 2 +z(x−y) =

y − x, or (x − y)(x + y + z + 1) = 0. This yields x = y or x + y + z = −1.
Subtracting the third equation from the second, we obtain y 2 −z 2 +x(y−z) =
y − z, or (y − z)(y + z + x − 1) = 0. This yields y = z or x + y + z = 1.
We now distinguish two cases: x = y and x 6= y. In the first case, we have
y 6= z, as otherwise we would have x = y = z for which the first equation
becomes 0 = 1, a contradiction. Now it follows that x + y + z = 1, or
2x + z = 1. Substituting y = x and z = 1 − 2x in the first equation yields
x2 − x(1 − 2x) = x − (1 − 2x) + 1, which can be simplified to 3x2 − x = 3x,
or 3x2 = 4x. We get x = 0 or x = 43 . With x = 0, we find y = 0, z = 1,
23
but does not satisfy our assumption x ≥ y ≥ z. Thus, the only remaining
possibility is x = 43 , which gives the triple ( 43 , 43 , − 53 ). We verify that this
is indeed a solution.
Now consider the case x 6= y. Then we have x + y + z = −1, hence we
cannot have x + y + z = 1, and we see that y = z. Now x + y + z = −1
yields x + 2z = −1, hence x = −1 − 2z. Now the first equality becomes
(−1 − 2z)2 − z 2 = 1, which can be simplified to 3z 2 + 4z = 0. From this, we
conclude that z = 0 or z = − 43 . With z = 0, we find y = 0, x = −1, which
does not satisfy our assumption x ≥ y ≥ z. Hence, the only remaining
possibility is z = − 43 , and this gives rise to the triple ( 53 , − 34 , − 43 ). We
verify that this is indeed a solution.
By also considering the permutations of these two solutions, we find all
six solutions: ( 43 , 34 , − 53 ), ( 43 , − 53 , 43 ), (− 35 , 43 , 43 ), ( 53 , − 43 , − 43 ), (− 43 , 53 , − 34 ),
and (− 34 , − 43 , 53 ).
3. (a) From now on, with ‘lines’ we mean the six lines and the circle. As soon
as the number p is used somewhere, there is a ‘line’ whose product
is divisible by p, and hence has remainder 0 upon division by p. All
‘lines’ must give remainder 0 in that case, hence all ‘lines’ contain at
least one p. This can be achieved by assigning the number p to the
bottom three dots. Suppose that it is already possible with at most
two times the number p. Each dot is lying on exactly three ‘lines’,
hence there are at most six ‘lines’ containing a p. There are seven
‘lines’ in total, however, hence this is impossible. We conclude that
she needs the number p at least three times in total.
(b) Denote the bottom three numbers from left to right by a, b, and c.
Denote the number in the middle on the left side of the triangle by d.
Now we can compute all numbers modulo p; note that all numbers are
invertible modulo p because p itself does not appear, and p is prime.
The bottom line has product abc. The left side also needs to have this
product, hence the top number in the triangle must be congruent to
abc(ad)−1 = bcd−1 modulo p. If we consider the line from the right
bottom to the left middle, we see that the middle number must be
congruent to abc(cd)−1 = abd−1 . By considering the circle, we find
the number in the middle on the right side: abc(bd)−1 = acd−1 .
On the right side, we now get the equation bcd−1 · acd−1 · c ≡ abc
mod p, hence c2 ≡ d2 mod p. Similarly, the vertical line yields b2 ≡
d2 , and the line from the left bottom to the right middle yields a2 ≡ d2
mod p. We conclude that the numbers a, b, c, and d all must have
the same square. From x2 ≡ y 2 we obtain (x − y)(x + y) ≡ 0 mod p,
hence p | x − y or p | x + y, because p is prime. Therefore, we have
24
x ≡ y or x ≡ −y. These are two possibilities, because if y ≡ −y
were to hold, then 2y ≡ 0, hence p | 2y, hence p | y since p > 2; a
contradiction. Thus the numbers b, c, and d are all congruent to a or
−a. If these conditions are met, then we do have a2 ≡ b2 ≡ c2 ≡ d2
and from the preceding argument, we get that the product of the three
numbers on all six lines and the circle is congruent to abc.
There are p − 1 possibilities for the number a, and after choosing a,
there are 2 possibilities for each of the numbers b, c, and d. In total,
there are 8(p − 1) ways to assign the numbers.
4. We will show that the capicity of the two boxes does not matter, as long
as the total capacity is n (and at least 1 for each box). Jesse can always
win this game, and can do that by first playing the power 20 = 1 of two,
and then in each following turn the next power of two that is smaller or
greater. That means: if he played the numbers
2−i , 2−(i−1) , . . . , 2−1 , 20 , 21 , . . . , 2j−1 , 2j
at a certain moment, he will play either 2−(i+1) or 2j+1 in his next turn.
By playing cleverly, Jesse can make sure that the greatest power of two
among the stones played so far is always contained in the black box. We
will prove this by induction. In his first move, he puts the stone with
value 20 in the black box and the claim is true; this is the base case of the
induction. When it is his turn again, and Tjeerd moved the greatest power
of two so far, which according to the induction hypothesis was contained
in the black box, to the white box, then the black box actually has a free
space, and Jesse can put a new greater power of two in there, and the claim
is true. If Tjeerd moved some other stone or did nothing, then the greatest
power of two so far is still in the black box, and Jesse can play a smaller
power of two; it does not matter where he puts it. Also in this case, the
claim is true. This proves the induction step, and the claim is proved.
Therefore, after playing the last stone, the greatest power of two is in the
black box. It is greater than the sum of all smaller powers of two played
(2j > 2j − 2−i = 2j−1 + 2j−2 + · · · + 2−(i−1) + 2−i ), hence it is certainly
greater than the sum of the powers of two in the white box. Therefore, the
total value inside the black box is greater than the total value in the white
box.
25
5. Denote the radius of the incricle of 4ABC by r. Then the area of triangle
ABC is
1 1 1 1
|AB| · r + |BC| · r + |AC| = r · (3|BC| + |BC|) = 2r|BC|.
2 2 2 2
On the other hand, the area of ABC equals 12 h|BC|, where h is the altidude
from A. Hence, h = 4r. Because the distance from KL to BC is exactly
2r, the distance from A to KL is also 2r. Triangles AKL and ABC are
similar, because KL k BC, and the altitudes from A have lengths 2r and
4r, respectively, giving a multiplication factor of exactly 2. Hence, K is the
midpoint AB, and L is the midpoint of AC.
For the point T , we have |AC| = 4|AT |, hence |AT | = 14 |AC| = 12 |AL|,
hence T is the midpoint of AL. Now consider triangle ABL. In this triangle,
the segment BT is a median, because T is the midpoint of AL. Also LK
is a median as K is the midpoint AB. Their intersection point S is the
|SL|
centroid, from which we get that |KL| = 32 .
A K
B
26
Problems
1. The sequence a0 , a1 , a2 , . . . of integers is defined by a0 = 3 and
an+1 − an = n(an − 1)
for all n ≥ 0. Determine all integers m ≥ 2 for which gcd(m, an ) = 1 for

all n ≥ 0.
2. Find all quadruples (x1 , x2 , x3 , x4 ) of real numbers which are solutions of

the following system of six equations:
x1 + x2 = x23 + x24 + 6x3 x4 ,

x1 + x3 = x22 + x24 + 6x2 x4 ,
x1 + x4 = x22 + x23 + 6x2 x3 ,
x2 + x3 = x21 + x24 + 6x1 x4 ,
x2 + x4 = x21 + x23 + 6x1 x3 ,
x3 + x4 = x21 + x22 + 6x1 x2 .
3. Let ABC be an acute non-isosceles triangle with orthocentre H. Let O

be the circumcentre of triangle ABC, and let K be the circumcentre of
triangle AHO. Prove that the reflection of K in OH lies on BC.
4. On a rectangular board consisting of m × n squares (m, n ≥ 3), dominos

have been placed (2 × 1- or 1 × 2-tiles), not overlapping each other. Each
domino covers exactly two squares of the board. Suppose that the placement
of the dominos has the property that no extra domino can be placed on
the board, and the four corners of the board are not all empty. Prove that
at least 23 of the squares of the board is covered by dominos.
27
Solutions
1. The sequence is given by the formula an = 2 · n! + 1 for n ≥ 0. (We use the
usual definition 0! = 1, which satisfies 1! = 1 · 0!, in the same way we have
n! = n · (n − 1)! for other positive integers n.) We will prove the equality
by induction. We have a0 = 3, which equals 2 · 0! + 1. Now suppose for
certain k ≥ 0 that ak = 2 · k! + 1, then
ak+1 = ak +k(ak −1) = 2·k!+1+k ·2·k! = 2·k!·(1+k)+1 = 2·(k +1)!+1.
This finishes the induction.

We see that an is always odd, hence gcd(2, an ) = 1 for all n. It follows also
that gcd(2i , an ) = 1 for all i ≥ 1. Hence, m = 2i with i ≥ 1 satisfies the
condition. Now consider an m ≥ 2 which is not a power of two. Then m
has an odd prime divisor, say p. We will show that p is a divisor of ap−3 .
By Wilson’s theorem, we have (p − 1)! ≡ −1 mod p. Hence,
−1
2 · (p − 3)! ≡ 2 · (p − 1)! · (p − 2)(p − 1)
≡ 2 · −1 · (−2 · −1)−1 ≡ 2 · −1 · 2−1 ≡ −1 mod p.
So indeed we have ap−3 = 2 · (p − 3)! + 1 ≡ 0 mod p. We conclude that m

does not satisfy the condition. Hence, the only values of m satisfying the
condition are powers of two.
2. Subtracting the second equation from the first yields x2 − x3 = x23 − x22 +
6x4 (x3 − x2 ), which we can factor as 0 = (x3 − x2 )(x3 + x2 + 1 + 6x4 ). We
see that x2 = x3 or x2 + x3 + 1 + 6x4 = 0. Similarly, we also have either
x2 = x3 or x2 + x3 + 1 + 6x1 = 0. Hence, if x2 6= x3 , the second equality
must hold in both cases; subtracting one from the other, we obtain x1 = x4 .
We conlude that either x2 = x3 or x1 = x4 . Analogously, we get for each
permutation (i, j, k, l) of (1, 2, 3, 4) that either xi = xj or xk = xl .
We will prove that at least three of the xi must be equal. If all four are equal,
this is true of course. Otherwise, there are two unequal ones, say x1 6= x2
without loss of generality. Then we have x3 = x4 . If also x1 = x3 holds,
then there are three equal elements. Otherwise, we have x1 = 6 x3 , hence
x2 = x4 and we also get three equal elements. Up to order, the quadruple
(x1 , x2 , x3 , x4 ) is thus equal to a quadruple of the shape (x, x, x, y), where
we could have that x = y.
Substituting this in the equations gives x + y = 8x2 and 2x = x2 + y 2 + 6xy.
Adding these two equations: 3x + y = 9x2 + y 2 + 6xy. The right hand
side can be factored as (3x + y)2 . Defining s = 3x + y, the equation
28
becomes s = s2 , from which we get either s = 0 or s = 1. We have
s = 3x + y = 2x + (x + y) = 2x + 8x2 . Hence, 8x2 + 2x = 0 or 8x2 + 2x = 1.
In the first case, we have x = 0 or x = − 14 . We find y = 0 − 3x = 0 and
y = 0 − 3x = 34 , respectively. In the second case, we get the factorisation
(4x − 1)(2x + 1) = 0, hence x = 14 or x = − 12 . We find y = 1 − 3x = 14 or
y = 1 − 3x = 52 , respectively.
Altogether, we found the following quadruples: (0, 0, 0, 0), (− 14 , − 14 , − 14 , 34 ),
( 14 , 14 , 41 , 14 ) and (− 12 , − 12 , − 12 , 25 ), and permutations thereof. It is a simple
computation to verify that all these quadruples are indeed solutions to the
equations.
O H
A B
3. We consider the configuration as in the figure. Other configurations are

treated analogously. Denote by D the second intersection of AH with
the circumcircle of 4ABC. Denote by S the second intersection of the
circumcircles of ABC and AHO. (Because 4ABC is acute, both O and
H lie in the interior of ABC and also in the interior of the circumcircle,
hence D and S both exist.)
We have
∠OSH = ∠OAH = ∠OAD = ∠ODA = ∠ODH,
where we use that |OA| = |OD|. Moreover, we have
∠OHD = 180◦ − ∠OHA = 180◦ − ∠OSA = 180◦ − ∠OAS = ∠OHS,
where we use that |OA| = |OS|. Now we conclude that 4OHS ∼ = 4OHD
(SAA). This yields that D and S are each others reflection images in OH.
29
Therefore, if we reflect the circumcentre K of 4OHS in OH, we get the
circumcentre L of 4OHD. Now we must prove that L lies on BC.
Point D is the reflection of H in BC. This is a known fact, which we can
prove as follows: ∠DBC = ∠DAC = ∠HAC = 90◦ − ∠ACB = ∠HBC
and analogously ∠DCB = ∠HCB, hence 4DBC ∼ = 4HBC (ASA). Hence,
D is indeed the reflection of H in BC, from which we get that BC is the
perpendicular bisector of HD. Becuase L lies on the perpendicular bisector
of HD, we get that L lies on BC, which is what we wanted to prove.
4. Assign each empty square to the domino directly right of this square (unless
the square is on the right edge of the board). Now suppose that two empty
squares are assigned to the same domino, then this domino must be placed
vertically and both squares left of this domino are empty. However, that
would mean that another domino could fit, which is a contradiction. Hence,
no two empty squares are assigned to the same domino.
The empty squares on the right edge of the board have not been assigned a
domino yet. We try to assign these squares to dominos that do not have
an empty square directly left of them (i.e. dominos which have not been
assigned yet). First suppose that we succeed in assigning all empty squares
on the right edge of the board in this way to different dominos. In that case,
we assigned each empty square to a domino, where no domino has been
assigned more than one square. Because each domino covers two squares of
the board, there are two covered squares for each empty square, and hence
at most 13 of the squares is uncovered. In this case, we are done.
Now we will show that this assignment always works. Let k be the number
of empty squares on the right edge, and ` the number of empty squares on
the left edge. The empty squares on the left edge cannot be adjacent, hence
there are at least ` − 1 dominos on the left edge and these all do not have
an empty square left of them. If ` > k, then there are enough dominos on
the left edge to assign to all empty squares on the right edge. If ` < k, we
could turn everything around and assign all empty squares to the domino
left of them and we could also prove that at most 13 of the squares on the
board are uncovered. The only remaining situation is when ` = k and both
on the left and right exactly k − 1 dominos are on the edge. For both edges,
we have that there must be an empty square between each two dominos,
and there must also be empty squares in the corners. This, however, is
in contradiction with the condition that not all corners are empty. Hence,
this situation cannot occur.
30
Problems
1. Let Γ be the circumcircle of a triangle ABC and let D be a point on

segment BC. The circle that passes through B and D and is tangent to Γ
and the circle that passes through C and D and is tangent to Γ, intersect
at a point E 6= D. The line DE intersects Γ at two points, X and Y . Prove
that |EX| = |EY |.
2. Prickle and Sting are playing a game on an m × n-board, where m and

n are positive integers. They alternatingly take turns, and Prickle goes
first. Prickle must, during his turn, place a pawn on a square which doesn’t
contain a pawn yet. Sting must, during his turn, also place a pawn on a
square which doesn’t contain a pawn yet, but additionally, his pawn must
be placed in a square that is adjacent to the square in which Prickle placed
his pawn the previous turn.
Sting wins once the entire board is completely filled with pawns. Prickle
wins if Sting cannot place a pawn in his turn, while there is at least one
empty square on the board.
Determine for all pairs (m, n) of positive integers which of Prickle and Sting
has a winning strategy.
3. Show that for every positive integer n there exist positive integers a and b
with
n | 4a2 + 9b2 − 1.
4. Determine all positive integers n with the following property: for every
triple (a, b, c) of positive real numbers there exists a triple (k, `, m) of
non-negative integers such that ank , bn` , cnm are the lengths of sides of a
(non-degenerate) triangle.
31
Solutions
D
B C
1. We consider the configuration as in the figure, where E is at least as close

to B as it is to C. The proof in the case of the configuration in which this
is the other way around, is analogous.
Let O be the centre of Γ. The angle between the line BC and the com-
mon tangent in B is on the one hand, by the inscribed angle theorem
(tangent case), equal to ∠BED, and on the other hand equal to ∠BAC.
So ∠BED = ∠BAC. Analogously, we show that ∠CED = ∠BAC, so
∠BEC = ∠BED + ∠CED = 2∠BAC = ∠BOC, where we use the in-
scribed angle theorem to derive the last step. Therefore E lies on the circle
that passes through B, O, and C. If E = O, then we’re done, as |EX| and
|EY | then both are the radius of the circle.
So suppose that E 6= O, then in the configuration considered, BEOC
is a cyclic quadrilateral. Then ∠BEO = 180◦ − ∠BCO. In the isosceles
triangle BOC, we have ∠BCO = 90◦ − 12 ∠BOC = 90◦ ∠BAC, so ∠BEO =
180◦ −(90◦ −∠BAC) = 90◦ +∠BAC. Hence ∠DEO = ∠BEO −∠BED =
90◦ + ∠BAC − ∠BAC = 90◦ . Therefore EO is perpendicular to DE and
therefore also perpendicular to chord XY , from which follows that E is the
midpoint of XY . We conclude that |EX| = |EY |.
2. We use the convention that m is the number of rows, and that n is the
32
number of columns. If m is even, then we pair the squares of the board as
follows: in every column we pair the top two squares, then squares 3 and 4,
etc. As the number of rows is even, this pairs the squares of every column
completely. Sting can use the following strategy: whenever Prickle places
a pawn into one of a pair of squares, Sting places a pawn into the other,
which is necessary adjacent to the pawn that Prickle just placed. After
each move of Sting, all pairs contain either zero or two pawns, so Sting
can always place a pawn according to this strategy. Therefore Sting can
make sure that the entire board is eventually completely filled with pawns,
so Sting wins. Analogously, if n is even, then Sting has a similar winning
strategy.
If m = n = 1, Sting wins after Prickle’s first move. If m = 1 and n = 3 (or
the other way around), Sting can place his first pawn on a square adjacent
to the square in which Prickle has just placed his; such an adjacent (empty)
square always exists. The board will then be completely filled after Prickle’s
next move, so Sting has a winning strategy in these cases as well.
Now consider the case m = n = 3. Prickle can follow the following strategy.
He places his first pawn in the centre square. Sting must place his pawn
either in the same row or in the same column. Without loss of generality,
we assume that Sting’s pawn is placed in the same column as that of Prickle.
Prickle then places a pawn in the remaining square in the middle column.
At that point, the left and right columns are completely empty. Sting must
place his pawn into one of these columns. Prickle then chooses any square
in the other column, all three squares of which are still empty. This forces
Sting to place a pawn in that column as well, as the middle column was
already completely filled. Now Prickle places a pawn in the only remaining
square of that column, after which Sting can no longer place a pawn, while
there are two empty squares remaining. Therefore Prickle has a winning
strategy if m = n = 3.
The remaining case is that m and n are both odd, and that at least one
of m and n is at least 5. We consider the case that n (the number of
columns) is at least 5; the other case is analogous. Prickle can follow the
following strategy. Prickle places his pawns into the centre column until it
is completely filled with pawns. When it is Prickle’s turn again, there is an
even number of pawns on the board, all in the three columns in the centre
of the board, of which at least the middle one is completely filled. The
first and the last column are completely empty. As there is an odd number
of squares remaining, either the area to the left of the middle column has
an odd number of empty squares remaining, or the area to the right has.
Prickle chooses the area which contains an odd number of empty squares,
and places his pawns there until that area is full. As the centre column is
33
completely filled, Sting must also place all of his pawns in that area. As
this area contained an odd number of empty squares, Prickle is the last
player (to be able) to place a pawn in this area. Sting can no longer place a
pawn, while the other area contains a column of empty squares, so Prickle
wins. (Of course, if Sting runs out of moves earlier, Prickle also wins.)
We conclude that Prickle wins if m and n are odd and m ≥ 5, if m and n
are odd and n ≥ 5, and if m = n = 3. Sting wins in all other cases: if m
and n are both even, if m = n = 1, if m = 1 and n = 3, and if m = 3 and
n = 1.
3. If n = 1, all choices of a and b are solutions. Now suppose that n > 1

and let p be a prime divisor of n. Let k be the number of factors of p
in n. We give a condition for a and b modulo pk which guarantees that
pk |4a2 + 9b2 − 1. By doing this for every prime divisor of n, we get a system
of conditions for a and b modulo the various prime powers. Then, by the
Chinese remainder theorem, there exist a and b satisfying all conditions
simultaneously.
If p 6= 2, then we consider the condition that 2a ≡ 1 mod pk and b ≡ 0
mod pk . As 2 has a multiplicative inverse module pk , this condition can be
satisfied. We then have
4a2 + 9b2 − 1 = (2a)2 + 9b2 − 1 ≡ 12 + 9 · 0 − 1 = 0 mod pk .
Therefore all a and b satisfying this condition are solutions.

If p = 2, then we consider the condition that a ≡ 0 mod 2k and 3b ≡ 1
mod 2k . As 3 has a multiplicative inverse modulo 2k , this condition can be
satisfied. We then have
4a2 + 9b2 − 1 = 4a2 + (3b)2 − 1 ≡ 4 · 0 + 12 − 1 = 0 mod 2k .
Therefore all a and b satisfying this condition are solutions.
4. It is clear that n = 1 does not satisfy the property, as not every three
positive real numbers a, b, and c are the lengths of the sides of a triangle.
We first show that any n ≥ 5 cannot satisfy the property by considering
the triple (a, b, c) = (1, 2, 3). Suppose that there exist k, `, m such that
nk , 2n` , 3nm are the lengths of the sides of a triangle. As n 6= 2, 3, no three
of these are equal. By repeatedly removing common factors n if they exist,
we can and do assume that one of k, `, m is equal to 0. Suppose that the
other two of those three integers are positive, then their corresponding
34
side lengths both are multiples of n. Their difference then is at least n,
while the third side has length of at most 3. This contradicts the triangle
inequality. Hence of k, `, m, at least two must be zero. The corresponding
two side lengths then sum to at most 5, so the third side must have length
less than 5. As n ≥ 5, this third side cannot have a factor n either, so
k, `, m are all equal to 0. However, then 1, 2, 3 should be the lengths of
the sides of a triangle, while 3 = 2 + 1. This is a contradiction. We deduce
that n ≥ 5 does not satisfy the property.
Now consider n = 2, 3, 4. We construct (k, `, m) as follows. Take a triple
(a, b, c). If its entries already are the lengths of the sides of a triangle, we
take k = ` = m = 0. Otherwise there exists a triangle inequality that is not
satisfied. Without loss of generality, we assume that a ≥ b + c. Multiply
the lesser of b and c with n. If the right hand side still isn’t larger than a,
then take the new summands, and again multiply the lesser of the two by
n. So: if a ≥ ni b + nj c, then we multiply the lesser of ni b and nj c by n,
and repeat if the inequality still holds. This process always stops at some
point, as there exists i such that ni > a.
Consider the i and j such a ≥ ni b + nj c, such that applying the process
makes the right hand side larger than a. We assume without loss of
generality that ni b ≤ ni c, so we have a < ni+1 b + nj c. We claim that we
can take (k, `, m) to be equal to either (0, i + 1, j) or (0, i + 1, j + 1).
By definition, we have a < ni+1 b + nj c, and we have nj c < ni b + nj c ≤ a <
a+ni+1 b. Therefore, if (0, i+1, j) doesn’t satisfy the property, then we must
have ni+1 b ≥ a + nj c. Moreover ni b ≤ nj c, so ni+1 b ≤ nj+1 c < a + nj+1 c.
We also have a < ni+1 b + nj+1 c, so if (0, i + 1, j + 1) does not satisfy
the property, then we must have nj+1 c ≥ a + ni+1 b. We will derive a
contradiction in case both triples do not satisfy the property, i.e. in case
both ni+1 b ≥ a + nj c and nj+1 c ≥ a + ni+1 b.
Adding these inequalities, and subtracting ni+1 b on both sides, yields
nj+1 c ≥ 2a + nj c, or equivalently (n − 1)nj c ≥ 2a. Therefore
2a
ni+1 b ≥ a + nj c ≥ a + = 1 + n−12
a,
n−1
from which follows that
(n − 1) + 2 + 2n 3n + 1
a ≥ ni b + nj c ≥ n1 1 + n−1
2
a+ 2
n−1 a = a= a.
n(n − 1) n(n − 1)
So 3n + 1 ≤ n(n − 1). For n = 2, 3, 4, this inequality reads 7 ≤ 2, 10 ≤ 6,
and 13 ≤ 12, all of which are false. This is a contradiction.
Hence n = 2, 3, 4 are precisely the values which satisfy the property.
35
Problems
1. Let m and n be positive integers with mn even. Jetze is going to cover

an m × n-board (with m rows and n columns) with domino tiles, in such
a way that every domino tile covers exactly two squares, domino tiles do
not protrude out of the board or overlap one another, and every square is
covered by a domino tile. Merlijn then is going to colour all domino tiles on
the board either red or blue. Determine the smallest non-negative integer
V (depending on m and n) such that Merlijn can always make sure that
in each row, the number of squares covered by a red domino tile and the
number of squares covered by a blue domino tile differ by at most V , no
matter in what way Jetze covers the board.
2. Let ABC be a right angled triangle with ∠C = 90◦ and let D be the foot of
the altitude from C. Let E be the centroid of triangle ACD and let F be
the centroid of triangle BCD. Let P be the point satisfying ∠CEP = 90◦
and |CP | = |AP |, and let Q be the point satisfying ∠CF Q = 90◦ and
|CQ| = |BQ|. Show that P Q passes through the centroid of triangle ABC.
f x + yf (x + y) = y 2 + f (x)f (y)

for all x, y ∈ R.
4. Let p > 10 be a prime number. Show that there exist positive integers m
and n with m + n < p for which p is a divisor of 5m 7n − 1.
36
Solutions
1. First suppose that n is odd. Then we must have V ≥ 1, as the difference
must be odd. We show that V = 1 is always possible. Colour the vertical
domino tiles in the odd numbered columns red and the vertical domino
tiles in the even numbered columns blue. As in every row, every horizontal
domino tile covers a square in an even numbered column and one in an odd
numbered column, every row contains one more square covered by a red
domino tile than squares covered by a blue domino tile. Now colour the
horizontal domino tile in each row alternatingly blue and red (starting with
blue). If the number of horizontal domino tiles is even, then at the end,
the number of red squares will be one more than that of blue squares; if
the number of horizontal domino tiles is odd, then at the end, the number
of blue squares will be one more than that of red squares. The difference
will therefore always be equal to 1.
Now suppose that n ≡ 2 mod 4. Then we have V ≥ 2 if Jetze places
every domino tile horizontally; then every row contains an odd number of
horizontal domino tiles. We show that V = 2 is always possible. Use the
same strategy as in the odd case. After colouring the vertical domino tiles,
the numbers of red and blue squares are equal. If we alternatingly colour
the horizontal domino tiles in each row blue and red again, we see that in
the end, in every row the difference between the number of red and blue
squares is 0 or 2.
Finally, suppose that n ≡ 0 mod 4. We show that V = 0 is always possible.
Number the rows from top to bottom from 1 up to m, and let bi be the
number of vertical domino tiles of which the top square is in row i. By
induction on i, we easily show that bi is even, using the fact that a horizontal
domino tile always covers an even number of squares in a row. We now
colour the vertical domino tiles in rows i and i + 1 as follows: if bi ≡ 0
mod 4, we colour half of them red, and the other half blue, and if bi ≡ 2
mod 4 we colour two more domino tiles red than we colour blue if i is even,
and we colour two more domino tiles blue than we colour red if i is odd.
We show that we can now colour the horizontal domino tiles in each row k
in such a way that every row has the same number of red and blue squares.
If bk−1 ≡ bk ≡ 0 mod 4, then vertical domino tiles in row k cover the same
number of red squares as blue squares. Moreover, the number of horizontal
domino tiles in row k is even, so we can simply colour half of them red and
half of them blue. If bk−1 ≡ bk ≡ 2 mod 4, then vertical domino tiles in
row k again cover the same number of red squares as blue squares, since of
k − 1 and k, one is odd and one is even. Again, the number of horizontal
domino tiles in row k is even, so we can again simply colour half of them
red and half of them blue. If bk−1 6≡ bk mod 4, then the difference in the
37
number of squares covered by red vertical domino tiles and blue vertical
ones is equal to 2. The number of horizontal domino tiles is odd, so we can
colour those in such a way that in the end, the number of red and blue
squares are equal.
Hence the minimal values for V are: V = 1 if n is odd, V = 2 if n ≡ 2
mod 4, and V = 0 if n ≡ 0 mod 4.
E
Z F
Q
A B
D
2. Let M , N , R, S be the midpoints of line segments BC, CA, BD, AD. Let
Z be the centroid of 4ABC. Quadrilateral QF M C is cyclic as ∠QF C =
90◦ = ∠QM C. Note that therefore CQ is a diameter of the circumcircle of
QF M C. Analogously, we see that P N EC is a cyclic quadrilateral, with
diameter CP .
We show that Z also lies on the circumcircles of these cyclic quadrilaterals.
The similarity transforming triangle BCA into triangle BDC transforms
triangle CZM into DF R, as C is mapped to D, the centroid Z is mapped
to the centroid F , and the midpoint M of BC is mapped to the midpoint of
BD, which is R. So 4CZM ∼ 4DF R, in particular ∠CZM = ∠DF R =
∠CF M (opposite angles). Therefore Z lies on the circumcircle of the cyclic
quadrilateral QF M C. Analogously, we have ∠CZN = ∠DES = ∠CEN ,
from which follows that Z lies on the circumcircle of the cyclic quadrilateral
P N EC.
We can now show that Z lies on P Q. As CQ is a diameter of the circle
through Q, F , M , C, Z, we have ∠QZC = 90◦ . As CP is a diameter of
the circle through P , N , E, Z, C, we also have ∠CZP = 90◦ . Hence P , Z,
and Q are collinear.
38
3. Note that the function f (x) = 0 for all x does not satisfy the condition.
Hence there is some a with f (a) 6= 0. Substituting x = a and y = 0, we
get f (a) =f (a)f (0), so f (0) = 1. Substituting x = 1 and y = −1, we get
f 1 − f (0) = 1 + f (1)f (−1). As f (0) = 1, this reads 1 = 1 + f (1)f (−1),
so f (1) = 0 or f (−1) = 0. We consider these two cases separately.
First suppose that f (1) = 0. Substituting x = t and y = 1 − t, and x = 1 − t
and y = t then gives
f (t) = (1 − t)2 + f (t)f (1 − t),

f (1 − t) = t2 + f (t)f (1 − t).
Subtracting these two gives f (t) − f (1 − t) = (1 − t)2 − t2 = 1 − 2t, so

f (1 − t) = f (t) + 2t − 1. Substituting this expression into the first of the
above two equations gives
f (t) = (1 − t)2 + f (t)2 + (2t − 1)f (t),
or equivalently
f (t)2 + (2t − 2)f (t) + (1 − t)2 = 0,
2
or equivalently, f (t) − (1 − t) = 0. Therefore f (t) = 1 − t for all t.
We verify this function in the original equation: the left hand side is
1 − (x + y(1 − x − y)) = 1 − (x + y − xy − y 2 ) = 1 − x − y + xy + y 2 , and
the right hand side is y 2 + (1 − x)(1 − y) = y 2 + 1 − x − y + xy, which is
equal to the left hand side. Therefore the function f (x) = 1 − x for all x is
a solution to the original equation.
Now suppose that f (−1) = 0. Substituting x = t and y = −1 − t, and
x = −1 − t and y = t, then gives
f (t) = (−1 − t)2 + f (t)f (−1 − t),

f (−1 − t) = t2 + f (t)f (−1 − t).
Subtracting these two gives f (t) − f (−1 − t) = (−1 − t)2 − t2 = 1 + 2t, so

f (−1 − t) = f (t) − 2t − 1. Substituting this expression into the first of the
above two equations gives
f (t) = (−1 − t)2 + f (t)2 + (−2t − 1)f (t),
or equivalently
f (t)2 − (2t + 2)f (t) + (t + 1)2 = 0,
2
or equivalently, f (t) − (t + 1) = 0. We deduce that f (t) = t + 1 for all t.
39
We verify this function satisfies the original equation. The left hand side
is x + y(x + y + 1) + 1 = x + xy + y 2 + y + 1 and the right hand side is
y 2 + (x + 1)(y + 1) = y 2 + xy + x + y + 1, which is equal to the left hand side.
Hence the function f (x) = x + 1 for all x satisfies the original equation.
We conclude that there are exactly two solutions of the original equation:
f (x) = 1 − x for all x, and f (x) = x + 1 for all x.
4. By Fermat’s Little Theorem, we have ap−1 ≡ 1 mod p for all a such that
p - a. As p > 10, p is odd, so p − 1 is even. We have
p−1 p−1
a 2 − 1 a 2 + 1 = ap−1 − 1 ≡ 0 mod p.
p−1 p−1
So p | (a 2 − 1)(a 2 + 1), so p is a divisor of at least one of the factors.
p−1
Hence a 2 is congruent to 1 or −1 modulo p.
We apply this to a = 5 and a = 7. Note that p > 10, so p 6= 5, 7. If
p−1 p−1
5 2 ≡ 1 mod p and 7 2 ≡ 1 mod p, we choose m = n = p−1 2 , which
p−1 p−1
satisfy the condition. The same holds if 5 2 ≡ −1 mod p and 7 2 ≡ −1
mod p.
The remaining case is that one is congruent to 1 and the other is congruent
p−1 p−1
to −1. Assume that 5 2 ≡ 1 mod p and 7 2 ≡ −1 mod p. The case in
which it is the other way around is analogous.
If there exists an n such that 0 < n < p−1 n
2 and 7 ≡ 1 mod p, then we
p−1
choose this n and m = 2 , which satisfy the condition. If not, then no
n such that p−1 n
2 < n < p − 1 and 7 ≡ 1 mod p can exist either, since
otherwise 7 p−1−n
≡ 7 (7 ) ≡ 1 mod p, while 0 < p − 1 − n < p−1
p−1 n −1
2 ,
which is a contradiction. Moreover, no i, j such that 1 ≤ i < j ≤ p − 1
and 7i ≡ 7j mod p can exist, since otherwise 7j−i ≡ 1 mod p with 1 ≤
j − i < p − 1. Hence 7i assumes distinct values for 1 ≤ i ≤ p − 1 modulo
p, which are all non-zero. So 7i assumes all non-zero values modulo p for
1 ≤ i ≤ p − 1.
In particular, there exists an n such that 7n ≡ 5−1 mod p. Then n ≤ p − 2,
since 7p−1 ≡ 1 6≡ 5−1 mod p. Choosing this n and m = 1 gives 7n · 5m ≡
5−1 · 5 ≡ 1 mod p.
Therefore it is always possible to find m and n satisfying the conditions.
Sectorgelden
Outreach Bèta
and the team selection for IMO 2021 Russia
MY

CY
CMY

Dutch MO &amp; TST 2008-20 With Solutions

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dutch MO &amp; TST 2008-20 With Solutions

Uploaded by

Copyright:

Available Formats

The 46th

and the team selection

for IMO 2008 Madrid

First Round, January 2007 1

Second Round, September 2007 5

Team Selection Test, June 2008 8

First Round, January 2008 13

Board of the Training committee

prof.dr. Frits Beukers, dr. Quintijn Puite,

• Raymond van Bommel (16 y.o., participated in IMO 2007 as well)

• Remy van Dobben de Bruyn (17 y.o.)

• Floris van Doorn (17 y.o., observer C in IMO 2007)

• Alexander van Hoorn (18 y.o.)

• Milan Lopuhaä (18 y.o., participated to in IMO 2007 as well)

• Maarten Roelofsma (17 y.o.)

The Dutch delegation for IMO 2008 Madrid further consists of

• dr. Quintijn Puite, team leader

• Birgit van Dalen MSc, deputy team leader

• drs. Wim Berkelmans, project manager IMO 2011, observer A

• prof.dr. Gerhard Wöginger, observer A

• Teunis van Wijngaarden, observer C

A) 2007 B) 18036 C) 18063 D) 18084 E) 4028049

A5. We use five different paths to connect five grid points.

A) 667 B) 669 C) 1003 D) 2006 E) No solutions exist

A3. A) 76 There are 9×8×7

A4. B) 7 First we rewrite the equation as ab+1b = 13 × ( ba+1

A5. B) First, we see that path A is longer than path B, since

A6. C) 4 We see that the sequence of final digits is equal to

A7. C) 1003 Since 9n + 9n + 9n = 3 × 9n = 32n+1 , n = 1003 is a solution of our equation.

B3. 120 Recursively, we compute

1. Consider the equilateral triangle ABC with |BC| = |CA| = |AB| = 1.

4. Determine the number of integers a satisfying 1 6 a 6 100 such that aa is a perfect

5. A triangle ABC and a point P inside this triangle are given.

of the position of P inside the triangle. D E

Let M be the circumcentre of 4ABC and D be the midpoint of BC.

We consider two cases; namely a being even, or a being odd.

- Now suppose that a is odd, and write a = 2c + 1. Then

We denote the area of 4ABC as [ABC]. Similarly,

Let us first consider 4ABP . The segments AE and BD F

[DRESF T ] = 13 ([ABP ] + [BCP ] + [CAP ]) = 31 [ABC],

which is independent of the location of P .

Valkenswaard, June 7, 2008

1. Find all funtions f : Z>0 → Z>0 that satisfy

f (f (f (n))) + f (f (n)) + f (n) = 3n

for all n ∈ Z>0 .

3. Let m, n be positive integers. Consider a sequence of positive integers a1 , a2 , . . . , an

bi = #{j ∈ {1, 2, . . . , n} : aj ≥ i},

so bi is the number of terms aj of the given sequence for which aj ≥ i. Similarly, we

Problem 1. If f (m) = f (n), then 3m = 3n, thus m = n. So f is injective. Now we use

(a) Solution 1. Note that, for 1 ≤ i ≤ m, 1 ≤ j ≤ n:

Now let 1 ≤ j ≤ n, and i = aj . Then aj ≥ i, from which follows that cj ≥ i = aj .

Solution 2. Since the sequences are non-increasing, bi = max{l : al ≥ i} and cj =

cj = max{i : bi ≥ j} = max{i : max{l : al ≥ i} ≥ j}.

For fixed i, we have

(b) Solution 1. For 1 ≤ k ≤ m,

which is what we needed to prove.

Problem 4. Let a be such that 1 + 12n2 = a2 . Rewrite this as

12n2 = a2 − 1 = (a + 1)(a − 1).

a + 1 = 2b2 and a − 1 = 6c2

for certain integers b, c such that bc = n.

which was what we needed to prove.

α = ∠BCP = ∠CP B = ∠QP A = ∠AQP.

Now note that

∠CP Q = ∠CP B + ∠BP Q = ∠QP A + ∠BP Q = ∠BP A = 90◦ .

Dutch MO & TST 2008-20 With Solutions

Dutch MO & TST 2008-20 With Solutions