A Corner of Mathematical

Olympiad and Competition

Book IV
(Selection Problems from Canada)

1. Canadian Open Mathematics Challenge 1996-2014

2. Canadian Mathematical Olympiad 1969-2014

Problems with Solutions

Phnom Penh, October 24, 2014

Prepare by: Keo Sodara
Contact: [email protected]
[email protected]
www.facebook.com/aprissigi
(+855) 0967308278
 Part I

Canadian Open
Mathematics
Challenge From
1996-2014
 The Canadian Mathematical Society
in collaboration with
The Center for Education
in Mathematics and Computing

The Second
Canadian Open
Mathematics Challenge
Wednesday, November 26, 1997
Examination Paper
c Canadian Mathematical Society 1997

Part A
Note: All questions in part A will be graded out of 5 points.

1. In triangle ABC , \A equals 120 de- A

grees. A point D is inside the trian-
gle such that \DBC = 2
\ABD D
and \DCB = 2
\ACD. Deter-
mine the measure, in degrees, of B C
\BDC .
2. Solve the following system of equations:
xy2 = 108; xy = 1010 :
3

3. Determine all points on the straight line which joins (,4; 11) to (16; ,1) and
whose coordinates are positive integers.
4. Given three distinct digits a; b and c, it is possible, by choosing two digits at
a time, to form six two-digit numbers. Determine all possible sets fa; b; cg for
which the sum of the six two-digits numbers is 484.
5. Two cubes have their faces painted either red or blue. The rst cube has ve
red faces and one blue face. When the two cubes are rolled simultaneously, the
probability that the two top faces show the same colour is 21 . How many red
faces are there on the second cube?
 6. The triangle ABC has sides AB = A
137; AC = 241, and BC =
200. There is a point D, on BC ,
such that both incircles of trian-
gles ABD and ACD touch AD at
the same point E . Determine the E
length of CD.
B D C
7. Determine the minimum value of f (x) where
f (x) = (3 sin x , 4 cos x , 10)(3 sin x + 4 cos x , 10):
8. An hourglass is formed from two identical cones. Initially, the upper cone is
lled with sand and the lower one is empty. The sand ows at a constant rate
from the upper to the lower cone. It takes exactly one hour to empty the upper
cone. How long does it take for the depth of sand in the lower cone to be half
the depth of sand in the upper cone? (Assume that the sand stays level in both
cones at all times.)

Part B
Note: All questions in part B will be graded out of 10 points.

1. The straight line l1 with equation x , 2y +10 = 0 meets the circle with equation
x2 + y2 = 100 at B in the rst quadrant. A line through B , perpendicular to l1
cuts the y-axis at P (0; t). Determine the value of t.
2. Consider the ten numbers ar; ar2; ar3;


; ar10. If their sum is 18 and the sum
of their reciprocals is 6, determine their product.
3. In an isosceles right-angled triangle AOB , B
points P; Q and S are chosen on sides
OB; OA and AB respectively such that a
square PQRS is formed as shown. If the P S
lengths of OP and OQ are a and b respec-
tively, and the area of PQRS is 52 that of a
triangle AOB , determine a : b. R
Ob Q A
4. Find all real values of x; y and z such that
x,p pyz = 42
y , pxz = 6
z , xy = ,30:
 The Canadian Mathematical Society
in collaboration with
The Center for Education
in Mathematics and Computing

The Second
Canadian Open
Mathematics Challenge
Wednesday, November 26, 1997
Solutions
c Canadian Mathematical Society 1997

Part A
Note: All questions in part A were graded out of 5 points.

1. In triangle ABC , \A equals 120 A

degrees. A point D is inside the tri-
angle such that \DBC = 2
\ABD D
and \DCB = 2
\ACD. Deter-
mine the measure, in degrees, of B C
\BDC .

By letting \DBC = 2x and \DCB = 2y, one obtains an equation involving

3x + 3y + 120, leading to \BDC = 140.
The average score was 4.0.
2. Solve the following system of equations:
3
xy2 = 108; xy = 1010 :

There are many approaches. Probably the most straightforward is to determine

x in terms of y in the rst equation, and then to substitute this in the second
equation. The answer is x = 104; y = 102.
The average score was 3.8
3. Determine all points on the straight line which joins (,4; 11) to (16; ,1) and
whose coordinates are positive integers.
By using the given points, the slope of the line segment is , 35 . Using this
slope, the points are easily determined to be (11; 2); (6; 5), and (1; 8).
The average score was 3.7.
4. Given three distinct digits a; b and c, it is possible, by choosing two digits at
a time, to form six two-digit numbers. Determine all possible sets fa; b; cg for
which the sum of the six two-digits numbers is 484.
The six possible numbers are 10a + b; 10a + c; 10b + a; 10b + c; 10c + a; 10c + b.
Their sum is 22(a + b + c). From this, the acceptable sets are f6; 7; 9g and
f5; 8; 9g, since the digits are distinct.
The average score was 2.8.
5. Two cubes have their faces painted either red or blue. The rst cube has ve
red faces and one blue face. When the two cubes are rolled simultaneously, the
probability that the two top faces show the same colour is 21 . How many red
faces are there on the second cube?
The colour on the top of the rst cube is irrelevant. Once it is rolled, we
must have three red faces and three blue faces on the second cube if the prob-
ability of like faces is 21 .
The average score was 2.7.
6. The triangle ABC has sides A
AB = 137; AC = 241, and
BC = 200. There is a point D, on
BC , such that both incircles of tri-
angles ABD and ACD touch AD
at the same point E . Determine E
the length of CD.
B D C
The solution to this problem is obtained from using the property that tangents
to a circle from an external point are equal. Using this fact and applying al-
gebraic variables as needed, one obtains, from the resulting equation, CD = 152.
The average score was 0.5.
 7. Determine the minimum value of f (x) where
f (x) = (3 sin x , 4 cos x , 10)(3 sin x + 4 cos x , 10):
While calculus can be used, it is not necessary. Multiply the given expressions
together and substitute for cos2 x. The result is a quadratic in sin x. Complet-
ing the square and noting that j sin xj  1 yields a minimum value of 49.
The average score was 1.2.
8. An hourglass is formed from two identical cones. Initially, the upper cone is
lled with sand and the lower one is empty. The sand ows at a constant rate
from the upper to the lower cone. It takes exactly one hour to empty the upper
cone. How long does it take for the depth of sand in the lower cone to be half
the depth of sand in the upper cone? (Assume that the sand stays level in both
cones at all times.)
At the required time the depth of sand in the upper cone is two-thirds its
original depth. Since volume varies
, 2
3
as the cube of any dimension in regular
gures, the time required is 1 , 3 = 19 27 of an hour.
The average score was 0.7.

Part B
Note: All questions in part B were graded out of 10 points.

1. The straight line l1 with equation x , 2y +10 = 0 meets the circle with equation
x2 + y2 = 100 at B in the rst quadrant. A line through B , perpendicular to l1
cuts the y-axis at P (0; t). Determine the value of t.
This is a straightforward problem. Solve for point B by substitution from the
rst equation to the second equation to obtain B as (6; 8). The line through B
perpendicular to l1 meets the y-axis at (0; 20).
The average score was 5.9
2. Consider the ten numbers ar; ar2; ar3;


; ar10. If their sum is 18 and the sum
of their reciprocals is 6, determine their product.
Consider the given equations, and divide the rst by the second. This yields
a2 r11 = 3. The required result is then a10 r55 = 35.
The average score was 2.1.
3. In an isosceles right-angled triangle AOB , B
points P; Q and S are chosen on sides
OB; OA and AB respectively such that a
square PQRS is formed as shown. If the P S
lengths of OP and OQ are a and b respec-
tively, and the area of PQRS is 52 that of a
triangle AOB , determine a : b. R
Ob Q A
One method is to draw ST perpendicular to OB . Congruent triangles are ob-
tained, yielding OB = 2a + b. Another method is to use the sine and cosine
laws on 4BPS , and to use analytic geometry. The result is a : b = 2 : 1.
The average score was 0.7.
4. Find all real values of x; y and z such that
x,p pyz = 42
y , pxz = 6
z , xy = ,30:
Let x = a2 ; y = b2 ; z = c2; thereby eliminating the radicals. Combining the
equations in pairs leads to the fact that b = a+2 c . This allows a reduction from
three to two variables, and hence the result x = 54; y = 24; z = 6.
The average score was 0.4.
 The Canadian Mathematical Society
in collaboration with
The Center for Education
in Mathematics and Computing

The Third
Canadian Open
Mathematics Challenge
Wednesday, November 25, 1998
Examination Paper
c Canadian Mathematical Society 1998

Part A
Note: All questions in part A will be graded out of 5 points.

1. Solve for x, given that 3x+2 = 3x + 216.

2. A rectangular closed box
(shown) with dimensions a,
2a and 1 has a surface area
of 54, where a is an integer.
Determine the volume of the
box.
1
a

2a

3. In the figure, each region T B

represents an equilateral tri-
angle and each region S a S
semicircle. The complete fig- A C
ure is a semicircle of radius 6 S S
with its centre O . The three T
smaller semicircles touch the
T T
large semicircle at points A, B
O 6
and C . What is the radius of a
semicircle S ?
 4. In an arithmetic sequence t1 ; t2 ; t3 ; :::; t47 , the sum of the odd numbered terms is
1272. What is the sum of all 47 terms in the sequence?
5. Compute the sum of the first 99 terms of the series
loga 2 + log a3
loga a
4 + log a5 6 +:::
a a loga a a loga a

6. The lengths of the sides of tri- B

angle ABC are 60; 80 and 100
with \A = 90Æ . The line AD
D
divides triangle ABC into two
triangles of equal perimeter.
Calculate the length of AD .

A C
7. There are ten prizes, five A’s, three B ’s and two C ’s, placed in identical sealed en-
velopes for the top ten contestants in a mathematics contest. The prizes are awarded
by allowing winners to select an envelope at random from those remaining. When
the eight contestant goes to select a prize, what is the probability that the remaining
three prizes are one A, one B and one C ?
8. Nine spheres are placed in a closed cubical box of side lenth 32 cm. Four small
spheres of radius r are first placed in the bottom corners of the box so that they
touch adjacent sides of the box but not each other. A large sphere of radius 15 cm
is then placed in the box so that it touches each of the four smaller spheres but not
the bottom. Four spheres of radius r are then added in the upper corners and the box
closed so that the lid just touches the four smaller spheres. Calculate r .

Part B
Note: Answer all questions. The problems in this section are worth 10 marks each. Marks
will be based on presentation. A correct solution poorly presented will not earn full marks.

1. Triangle ABC has its sides determinded in the following way: side AB by line
3x 2y + 3 = 0; side BC by line x + y 14 = 0; and side AC by line y = 3. If
the point P is chosen so that P A = P B = P C , determine the equation of the line
containing A and P .
2. ABC D is a rectangle and lines DX , A X B
DY and X Y are drawn where X is

on AB and Y is on BC . The area of

triangle AX D is 5, the area of trian-
gle BX Y is 4 and the area of triangle Y
C Y D is 3. Determine the area of tri-

angle DX Y .
D C
 3. Alphonse and Beryl play a game by
alternately moving a disk on a circu-
lar board. The game starts with the
disk already on the board as shown.
A player may move either clockwise
one position or one position toward
the centre but cannot move to a po-
sition that has been previously occu-
pied. The last person who is able to
move wins the game.

(a) If Alphonse moves first, is there a strategy which guarantees that he will always win?

(b) Is there a winning strategy for either of the players if the board is changed to five
concentric circles with nine regions in each ring and Alphonse moves first? (The
rules for playing this new game remain the same)

4. A line segment BC has length 6. Point A is chosen such that \BAC is a right angle.
For any position of A a point D is chosen in BC so that AD is perpendicular to BC .
A circle with AD as diameter has tangents drawn from C and B to touch the circle at
M and N , respectively, with these tangents intersecting at Z . Prove that Z B + Z C

is constant.
 The Canadian Mathematical Society
in collaboration with
The Center for Education
in Mathematics and Computing

The Third
Canadian Open
Mathematics Challenge
Wednesday, November 25, 1998
Solutions
c Canadian Mathematical Society 1998

Part A
Note: All questions in part A were graded out of 5 points.

1. This question is most easily solved by bringing the 3 x to the left side,
factoring and then arriving at, 3x (8) = 216. This leads to the solution
x = 3.

The average was 4.0.

2. If we recognize that the area of opposite faces in the box are equal we
arrive at the equations 2(2a2)+2(2a)+2(2a) = 54 or 2a2 +3a 27 = 0.
This leads to a = 3 with a volume of 18.

The average was 4.3.

3. If we cut out part of B
the diagram and label ap-
propriately we find that r
DL
p = r and DO = r
r 3. This
p gives OB =
K
D
L
r(1 + 3) = 6 or r =
6
p .
r3
1+ 3

O
The average was 1.7.
4. The easiest way to do this question is to recognize that the average of
the 24 odd terms is 1272
24 = 53. This is also the average of all 47 terms.
The sum of 47 terms is then 47  53 = 2491. This could also have
been done with the use of formulae and a more standard approach.

The average was 1.8.

5. Using the fact that loga an = nlogaa = n we see that the numerical
value of the series is 1 2 + 3 4 + : : : 98 + 99 = 50.

The average was 3.2.

6. We first note that DC = 60 and BD = 40. If we usepthe cosine law in
4ACD and the fact that cos C = 35 , we find AD = 2880. We could
also have done the problem by constructing a perpendicular D to AC
and then solving similar triangles.

The average was 1.7.

 thisproblem each of which leads

7. There are numerous approaches to
10
quickly to the answer. There are groups of three letters to be
3
chosen. There are 5 ways to choose in A, 3 ways to choose a B and 2
ways to choose a C or 5  3  2 ways of selecting an A, B or C . The
required probability is  3 2 = 41 .
5
10
3
 The average was 0.6.
8. Using basic properties of symmetry we observe that the corner of the
box, the centre of the small sphere and the centre of the large sphere
p
all lie along the same straight line drawn from the origin. The distance
from the corner of the box to the centre of the small sphere is r 3. If
p from the corner ofp the box to the centre
the distance p of the large
16
p sphere
3 15 .
is 16 3 we can now write, r 3 + r + 15 = 16 3 or r = p3+1

The average was 0.6.

Part B
Note: All questions in part B were graded out of 10 points.

1. This problem can be easily solved by first finding the coordinates of

the vertices of 4ABC . The coordinates of P are then found by us-
ing the properties of right bisectors of the sides. Doing this, we find
P (6; 4) as the required point. The equation of the required line is then
x 5y + 14 = 0. Note: Students should first draw a diagram in doing
this problem.

The average was 5.0.

2. If we let DC be y and AD be x then we can represent Y C as y6
xy 6
and AX 10
= x . From this, BY = y and BX = xy x 10 giving

y =p 8 since j4BXY j = 4. This leads to the solution
xy 10 xy 6
x
j4DXY j = 2 21.
The average was 1.8.
3.
(a) In essence, Alphonse adopts a strategy that will make Beryl enter the
last ring first. This guarantees a win because in the final ring the moves
are necessarily successive and there are an even number of regions
thus guaranteeing Alphonse the 2nd, 4th, 6th and 8th position, i.e. the
winning position
(b) Beryl adopts the strategy that will allow him to be the first to enter ring
three and five. This guaranteese Beryl that he will always win because
Alphonse will always be in an even position in these rings (if we label
the regions 1,2,3,. . . ,9) when it is Beryl’s turn to move.

The average was 2.7.

4. The best approach is to complete the Z
diagram as shown and label appro-
a
priately. We now observe that  + A
 +  = 180Æ and using the dou-
a
M
ble tangent formula we conclude that N
tan( +  ) = tan  and then by the   r
 
2
substitution 4r = (6 x)(x) we can 6-x  I x

arrive at a = 2. It is easy to see that r
ZB + ZC = 2 + (6 x) + 2 + x = 10
as required. This problem can also be B 6-x D x C
done using Heron’s formula.
The average was 0.8.
 The Canadian Mathematical Society
in collaboration with
The Center for Education
in Mathematics and Computing

The Fourth
Canadian Open
Mathematics Challenge
Wednesday, November 24, 1999
Examination Paper
c Canadian Mathematical Society 1999

Part A
Note: All questions in part A will be graded out of 5 points.

1. Determine the sum of all odd positive two-digit integers that are divisible by 5.
2. A rough sketch
y
of the graph of
y = x +
(2
4)(x2 3x) is
shown. For x
what values of
x is y  0?
3. Solve

for

xx :
4 x 8 1 =
2 9. 27
3
4. Solve the system of equations for x.
x + 2y z=5
3x + 2y + z = 11
(x + 2y )
2 z 2 = 15

5. Determine all x which satisfy:

3 x + 6 sin2 x
2 sin sin x 3=0 0; < x < 2
 6. A trapezoid, DEF G, is circum- D E
scribed about a circle that has centre
C and radius 2, as is shown. The
shorter of the two parallel sides, DE ,
has length 3 and angles DEF and
.
C

EF G are right angles. Determine the

area of the trapezoid. G F
7. The sector OAB of a
circle, with centre O ,
has a perimeter of 12.
Determine the radius O
of the circle which
maximizes the area of
the sector.
A B
8. Find the smallest positive integer k so that the expression 14kk+17
9 becomes a fraction in
the form qd where p; q and d are positive integers, p and q have no common divisors,
pd

and neither q nor d equals 1.

Part B
Note: All questions in part B will be graded out of 10 points.

1.

(a) Two identical triangles each have an area of 24. Their vertices are determined by the
intersection of the lines with equations y = 4,x = 0 and y = 43 x + b. Determine
the two possible values for b.
(b) For either of the two given triangles, a circle can be drawn to pass through its three
vertices. What is the radius of this circle?
2. If (bd + cd) is an odd integer, show that the cubic polynomial x 3 + bx2 + cx + d
cannot be expressed in the form (x + r )(x2 + px + q ) where b; c; d; r; p and q are all
integers.
3. Triangle ABC is right angled A
with its right angle at A. The
points P and Q are on the hy-
potenuse BC such that BP =
P Q = QC , AP = 3 and 3 4
AQ = 4. Determine the
length of each side of 4ABC .

B P Q C
4. Triangle ABC is any one of the set of
p triangles having base BC equal to a and height
from A to BC equal to h, with h < 23 a. P is a point inside the triangle such that the
value of \P AB = \P BA = \P CB = . Show that the measure of  is the same
for every triangle in the set.
 The Canadian Mathematical Society
in collaboration with
The Center for Education
in Mathematics and Computing

The Fourth
Canadian Open
Mathematics Challenge
Wednesday, November 24, 1999
Solutions
c Canadian Mathematical Society 1999

Part A
Note: All questions in part A were graded out of 5 points.

1. Answer
495
The average on this question was 3.7.

Comments
This question is most easily solved by adding the series from ’front to back’ or by
using the formula. Students could also have just added the terms mechanically to get
the same answer.
2. Answer
2  x  0 or x  3; x 2 R
The average on this question was 3.9.

Comments
The intercepts of the function are -2, 0 or 3 which gives the answer 2  x  0
or x  3. Generally speaking, students should have solved for the intercepts first
and then used the diagram to read off the correct intervals. Students should take care
in dealing with inequality signs. Many students had the signs going in the wrong
direction.
3. Solution If we convert to base 23 , this gives,
 x  8 1 x
4 2
=
9 27 3
  2 2x " 2 3 #1 x
2
=
3 3 3
 2x  3 3x  1
2 2 2
=
3 3 3
2 x+3  2 1
Therefore =
3 3
Since bases are equal, x + 3 = 1, x = 2:
The average on this question was 3.7.

Comments
This question was generally well done. It should be noted that converting to base 23
is by far the easiest way to approach the problem. The biggest mistake here was in
the inappropriate use of the power rules for exponents.

4. Solution
Factoring equation 3, (x + 2y z )(x + 2y + z ) = 15.
Substituting x + 2y z = 5, 5(x + 2y + z ) = 15 or, x + 2y + z = 3.
Subtracting this from (2): 2x = 8
Therefore, x = 4.
The average on this question was 3.4.

Comments
A variety of good solutions were given by students who used the first two equations
to arrive at y = 4 x and z = 3 x. This allowed substitutions into the third
equation. The best way to proceed, howeve, was to factor the third equation as a
difference of squares and then make the direct substitution on x + 2y z = 5 as
shown above.

5. Solution
Factoring gives, 2 sin2 x(sin x + 3) (sin x + 3) = 0 , (2 sin2 x 1)(sin x + 3) = 0.
Either sin x = 3 which is inadmissible since j sin xj  1, or

2 sin
2x =1

x = p
1
sin
2
 3 5 7
Therefore, x = ; ; ; .
4 4 4 4
The average on this question was 1.8.
Comments
This problem could have been solved by either factoring directly or using the factor
theorem. Students should make the comment in their solutions that sin x = 3 is
inadmissible. From there, the recognition that sin 2 x = 12 has four solutions should
be an easy matter.
6. Solution
From C draw D 1 2 E
a line perpen- 1
dicular to both Z 2
DE and EF C
and label the
k
. A
diagram as
shown. From 2
D draw a line G
perpendicular J Y 2 F
to GF to meet
the line at J . k

Since DE = JF = 3, JY =3 2 = 1. Thus, GJ = k 1. Since 4DJG is right

angled, (k + 1)2 k= ( 1)2 + 42 or, k 2 + 2k + 1 = 16 + k 2 2k + 1, 4k = 16,
(3 + 6)
k = 4. This makes GF and the area of trapezoid DEF G is
=6 (4) = 18.
2
The average on this question was 2.1.

Comments
This question could be done in a variety of ways. The easiest way is to use properties
of tangents to a circle and the Pythagorean theorem. When solving problems of this
type, it is almost always a matter of dropping perpendiculars and using simple prop-
erties of circles and triangles to get the appropriate equations. There were a number
of very unusual and insightful solutions to this problem

7. Solution
We are given that a + 2r = 12.
Therefore a = 12 2r . The
formula for the area (A) of a
1 O
sector is, A = ar where a
2
is arc lenght and r is radius. r  r
Using the formula for area,

A B
a
1
A= (12 2 ) rr
2

A = r2 + 6r
To maximize the area we complete the square or use calculus to find r = 3. Thus the
radius that maximizes the area is r = 3.
The average for this question was 1.6.
Comments
This question was nicely done by a large number of competitors. The formula for
1
the area of a sector can be easily derived to be A = ar. From there if we use the
2
 relationship, a + 2r = 12 it is not difficult to get the required expression for the area
of the sector of the circle. It was graifying to see the number of students who solved
this problem correctly.

8. Solution
k
14 + 17
If we rewrite in the following way,
k 9

k + 17 [14(k k
; k 6= 9
14 9) + 126] + 17 14( 9) 143 143
= = + = 14 +
k 9 k 9 k 9 k 9 k 9

Since 143 = 1:11:13, it is not difficult to see that k 9 = qd where d is a number

contained in 1:11:13 and neither q nor d equals 1. The smallest possible value for d
and q is 11 and 2 respectively. This makes k = 31.
The average on this question was 0.9.

Comments
It was delighted to see how many competitors solved this problem and the variety of
solutions. Some students made the observation that djk 9 and dj14k + 17 and so
dj14k + 17 14(k 9) or dj143. This also leads quickly to the solution.

Part B
Note: All questions in part B were graded out of 10 points.

1.

(a) Solution
The line y = 43 x + b y
meets the y -axis at C
and the line y = 4 at
B . Since the slope of
this line is 43 , we let C
AC be 3a and AB be x
4a. Then the area of
the triangle is given by B
A

1
a a
(3 )(4 ) = 24
2

6a2 = 24
a2 = 4
a = 2
 Then 3a = 6. Now the coordinates of C are (0; b), so AC = 6+4. Then b +4 = 6
b = 2 or 10
Comments
The best way of solving this problem was to recognize that a slope of 43 means that
the ratio of the height to the base for the triangle is 3a : 4a. If the triangle has an area
of 24, then the actual lenghts of the sides are 6 and 8. This gives b = 2 or b = 10.
In part (b), the triangle is right angled with a hypotenuse of 10. This gives the radius
5. Students should always draw diagrams for problems of this type. Many students
lost marks here because they didn’t explain how they arrived at their answers and did
not visualize a solution.
(b) Solution
Since either of the triangles have side lenghts 6 and 8 then the hypotenuse has a
lenght of 10. The semi-circle must have a radius of 5.
The average on this question was 3.4.
2. Solution
Assume that we can expand and compare coefficients. Expanding,
(x + r)(x2 + px + q ) = x3 + (p + r)x2 + (pr + q )x + qr

Comparing coefficeints,
p+r =b (1)
pr + q = c (2)
qr = d (3)
If bd + cd is odd, so is d(b + c). From this, d and b + c are both odd. From (3), if d
is odd then q and r are both odd.(4)
Adding (1) and (2), b + c = p + r + pr + q = (q + r ) + p(1 + r ). Since b + c is odd then
(q + r ) + p(1 + r ) is also odd. From (4), if q and r are both odd then q + r is even. This
implies that p(1 + r ) must be odd but this is not possible because r is odd and r + 1
is then even making p(1 + r ) both odd and even at the same time. This contradiction
implies that our original assumption was incorrect and thus x 3 + bx2 + cx + d cannot
be expressed in the form (x + r )(x2 + px + q ).
The average on this question was 2.1.

Comments
There were a variety of gorgeous solutions to this problem. We provide one solution
in its entirety. The method of proof here is that of contradiction. In essence we as-
sume that is possible to compare coefficients by expanding the left side. From this,
we show that this leads to a contradiction. Since there is a contridictory conclusion,
our original assumption must in fact have been false and thus it is not possible to
compare coefficients, as is required. A large number of students developed a large
varity of proofs, some of which were quite unique and very interesting and in fact
correct.
3. Solution
Label 4ABP as A
shown. From
4ABP , 32 = b
2
p +c 2 2pc cos B . 3 4
Because \BAC
is a right angle,
c B p P p Q p C
cos B=
3p c
so 9 = p2 + c2 2pc
3 p
or, 9 = p2 + c2 (1)
1
3
Following the same procedure in 4ACQ we have,
p2 + b2 (2)
1
16 =
3
Adding (1) and (2) gives, 25 = 2p2 + b2 + c2 ).
1
(
3
Since b2 + c2 = 9p2 , 25 = 2p2 + (9p2 ) = 5p2 .
1

p 3 p p
p = 5(p > 0) and BC = 3 5 or 45.
Therefore, p p p
Since p = 5, substituting in equation (1) and (2) gives AB = 12 and AC = 33.
The average on this question was 1.0.

Comments
This problem could be approached in a wide variety of ways. We provide one of the
many possible proofs above. Many students attempted this proof by drawing lines
parallel to AB throught P and Q and then using the side splitting theorem. This leads
to a simple application of Pythagoras and a nice system of equations to solved. The
number of students who presented unique solutions was quite gratifying.

4. Solution
Let P be a point inside 4ABC such that \P AB = \P BA = \P CB = .
Let \ABC be  . \BP C = 180 [( ) + ] = 180  .
A

N
h
P
 180-

B a C
From P draw a line perpendicular to AB meeting AB at N . Applying the Sine Law
in 4P BC ,
PB a a
= =
sin  sin(180 ) sin 
a sin 
Therefore, P B = (1)
sin 
 BN AB AB
From 4BP N , cos  = = or P B = (2)
P B 2P B 2 cos 
Equating our two expressions, (1) and (2), we have
a sin  AB
=
sin  2 cos 
or
AB sin 
2 sin  cos  =
a
AB sin 
sin 2 =
a
Since AB sinp3  = h then sin
p3 2 = a
h

Since h < 2 a, sin 2 < 2 (by substitution)

Then 2 < 3 and  < 6 or 2 > 23 and  > 3 .
But  > 3 is impossible because the sum of the angles in the triangle is  . Hence
there is one value of  for any given a and h.
The average on this question was 0.1.

Comments
This was a very hard problem. It can be done in three or four ways. A number of
competitors got this question correct. We provide here just one solution. Some stu-
p but only one or two students made an attempt at dealing
dents solved the problem
with restrictions, h 
3
a. This problem could have been attempted using coordi-
2
nates but was awkward and hard mechanically.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

The
Canadian Open
Mathematics Challenge

Wednesday, November 29, 2000

1
Time: 2 hours © 2000 Canadian Mathematical Society
2

Calculators are NOT permitted.

Do not open this booklet until instructed to do so.

There are two parts to the paper.

PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value for each
question by entering the correct answer in the space provided. Any work you do in obtaining an answer
will be considered for part marks if you do not have the correct answer, provided that it is done in the
space allocated to that question in your answer booklet.

PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must be written
in the appropriate location in the answer booklet. Rough work should be done separately. If you require
extra pages for your finished solutions, foolscap will be provided by your supervising teacher. Any extra
papers should be placed inside your answer booklet.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution poorly
presented will not earn full marks.

NOTE: At the completion of the contest, insert the information sheet inside the answer booklet.
 Canadian Open Mathematics Challenge

NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations and answers will be expressed as exact numbers such
as 4 p, 2 + 7 , etc.
4. Calculators are not allowed.

PART A

a
1. An operation “ D ” is defined by a D b = 1 – , b π 0.
b
What is the value of (1 D 2)D(3 D 4) ?

2. The sequence 9, 18, 27, 36, 45, 54, … consists of successive multiples of 9. This sequence is then
altered by multiplying every other term by –1, starting with the first term, to produce the new
sequence – 9, 18, – 27, 36, – 45, 54,... . If the sum of the first n terms of this new sequence is 180,
determine n.

3. The symbol n! is used to represent the product n(n – 1)(n – 2) L(3)(2)(1) .

( )( )( )( )
For example, 4!= 4(3)(2)(1) . Determine n such that n!= 215 36 53 72 (11)(13).

4. The symbol Î x û means the greatest integer less than or equal to x. For example,

Î5.7û = 5 , Îp û = 3 and Î4 û = 4 .
Calculate the value of the sum

Î 1û + Î 2 û + Î 3 û + Î 4 û + L + Î 48 û + Î 49 û + Î 50 û .

5. How many five-digit positive integers have the property that the product of their digits is 2000?

6. Solve the equation 4 ÊË16sin x ˆ¯ = 2 6 sin x , for 0 £ x £ 2 p .

2

7. The sequence of numbers …, a–3 , a– 2 , a–1, a0 , a1, a2 , a3 , … is defined by an – (n + 1)a2 – n = (n + 3)2 ,

for all integers n. Calculate a0 .

8. In the diagram, D ABC is equilateral and the radius of its

E C D
inscribed circle is 1. A larger circle is drawn through the vertices
of the rectangle ABDE . What is the diameter of the larger circle?

A B
 PART B

1. Triangle ABC has vertices A(0, 0) , B(9, 0) and C(0, 6) . The points P and Q lie on side AB such
that AP = PQ = QB . Similarly, the points R and S lie on side AC so that AR = RS = SC .
The vertex C is joined to each of the points P and Q. In the same way, B is joined to R and S.
(a) Determine the equation of the line through the points R and B.
(b) Determine the equation of the line through the points P and C.
(c) The line segments PC and RB intersect at X, and the line segments QC and SB intersect at Y.
Prove that the points A, X and Y lie on the same straight line.

2. In D ABC , the points D, E and F are on sides BC , CA and AB , A

respectively, such that – AFE = – BFD, – BDF = –CDE, and
E
–CED = – AEF . F
(a) Prove that – BDF = – BAC .
(b) If AB = 5, BC = 8 and CA = 7, determine the length of BD.

B D C

3. (a) Alphonse and Beryl are playing a game, starting with the
geometric shape shown in Figure 1. Alphonse begins the
game by cutting the original shape into two pieces along
one of the lines. He then passes the piece containing the Figure 1
black triangle to Beryl, and discards the other piece.
Beryl repeats these steps with the piece she receives; that is to say, she cuts along the length of
a line, passes the piece containing the black triangle back to Alphonse, and discards the other
piece. This process continues, with the winner being the player who, at the beginning of his or
her turn, receives only the black triangle. Show, with justification, that there is always a
winning strategy for Beryl.

(b) Alphonse and Beryl now play a game with the same rules
as in (a), except this time they use the shape in Figure 2
and Beryl goes first. As in (a), cuts may only be made
along the whole length of a line in the figure. Is there a
strategy that Beryl can use to be guaranteed that she will
win? (Provide justification for your answer.)

Figure 2

4. A sequence t1, t2 , t3 , ..., tn of n terms is defined as follows:

t1 = 1, t2 = 4 , and tk = tk –1 + tk – 2 for k = 3, 4, ..., n .
Let T be the set of all terms in this sequence; that is, T = {t1, t2 , t3 , ..., tn } .
(a) How many positive integers can be expressed as the sum of exactly two distinct elements of
the set T ?
(b) How many positive integers can be expressed as the sum of exactly three distinct elements of
the set T ?
 Canadian Open Mathematics Challenge

Comments on the Paper

Part A
a
1. An operation “ ∆ ” is defined by a ∆b = 1– , b ≠ 0.
b
What is the value of (1∆ 2)∆(3 ∆ 4) ?

Solution
By the definition of “ ∆ ”
1∆ 2 = 1 – 1 = 1
2 2
3 1
3∆ 4 = 1– =
4 4
1
and so (1∆ 2)∆(3∆ 4) =   ∆  = 1– = 1– 2 = –1
1 1 2
2 4 1
4
ANSWER: –1

2. The sequence 9, 18, 27, 36, 45, 54, … consists of successive multiples of 9. This sequence is then altered by multiplying
every other term by –1, starting with the first term, to produce the new sequence –9, 18, – 27, 36, – 45, 54,... . If the sum
of the first n terms of this new sequence is 180, determine n.

Solution
The terms in the sequence are paired, by combining each odd-numbered term with the next term
(that is, we combine terms 1 and 2, 3 and 4, 5 and 6, etc).

The sum of each of these pairs is 9.

So we need 20 of these pairs to reach a sum of 180.

Thus we need 2 × 20 or 40 terms.

ANSWER: 40
3. The symbol n! is used to represent the product n(n – 1)(n – 2) … (3)(2)(1) .
For example,4!= 4(3)(2 )(1) . Determine n such that n!= 2( 15 )(36 )(5 3)(72 )(11)(13) .
Solution
Since n! has a prime factor of 13, n must be at least 13.
Since n! has no prime factor of 17, n must be less than 17.
These two facts are true because if m ≤ n , then m divides n!.
Since n! has 53 as a factor, then n ≥ 15 , since we need n! to have 3 factors which are multiples of 5.
We must thus determine if n = 15 or n = 16 .
So we look at the number of factors of 2 in 16!.
16! gets 1 factor of 2 each from 2, 6, 10, 14
2 factors of 2 each from 4, 12
3 factors of 2 from 8
4 factors of 2 from 16
We have a total of 15 twos which then corresponds to n = 16 .
ANSWER: 16

4. The symbol x  means the greatest integer less than or equal to x. For example,

5.7 = 5 , π = 3 and 4 = 4 .

Calculate the value of the sum

 1 +  2 +  3 +  4 +      
… + 48 + 49 + 50 .

Solution
We note that for k a positive integer and k ≤ n < (k + 1) , then k ≤ n < k +1 and so  n = k .
2 2

Thus for 1 ≤ n ≤ 3,  n = 1
4 ≤ n ≤ 8,  n = 2

9 ≤ n ≤ 15,  n = 3

etc.
So the sum equals
(1+1+ 1) + (2 + 2 + 2 + 2 + 2) + (3 + … + 3) + … + (6 + … + 6) + (7 + 7)
= 3(1) + 5(2) + 7(3) + 9(4 ) +11(5) +13(6) + 2 (7)
= 3 +10 + 21+ 36 + 55 + 78+ 14
= 217
ANSWER: 217

5. How many five-digit positive integers have the property that the product of their digits is 2000?

Solution
Let a five-digit number have the form a b c d e where 0 ≤ a, b, c, d, e ≤ 9, a ≠ 0 .
Since the product of the digits is 2000, we must have the product abcde = 2000 = 24 53 .
 Since the product of the digits is 2000, then 3 of the digits have to be 5. The remaining 2 digits must have a product of
4
16 or 2 .
Thus the two remaining digits must be 4 and 4, or 2 and 8.

Possibility 1
5!
Case 1 Using the numbers 5, 5, 5, 4, 4 there are = 10 possible numbers.
3!2!
5!
Case 2 Using the numbers 5, 5, 5, 2, 8 there are = 20 possible numbers.
3!
There are 30 possible such numbers.

OR

Possibility 2
 5
We choose 3 of the 5 positions for the “5s” in   ways; there are 3 possibilities for the remaining two digits (including
 3
order): 2, 8; 4, 4; 8, 2.
5 
So there are 3×   = 3 ×10 = 30 possible such 5 digit numbers.
3 
ANSWER: 30

6. Solve the equation 4 16sin ( 2

x
)= 2 6 sin x
, for 0 ≤ x ≤ 2π .

Solution
We write all factors as powers of 2. Thus
(
4 16 sin
2
x
)= 2 6 sin x

2 (2 )= 2
2 4 sin2 x 6 sin x

2
x +2
2 4sin = 26 sin x (*)
Equating exponents (which we can do by taking base 2 logarithms),
4 sin 2 x + 2 = 6 sin x
2 sin 2 x – 3 sin x + 1 = 0
(2 sin x – 1)(sin x – 1) = 0
1
Therefore, sin x = or sin x = 1 .
2
π 5π π
Since 0 ≤ x ≤ 2 π , x = , or .
6 6 2
π 5π π
ANSWER: , ,
6 6 2
 The sequence of numbers …, a –3 , a– 2 , a–1 , a0 , a1 , a2 , a3 , … is defined by an – (n + 1)a2– n = (n + 3) , for all integers n.
2
7.
Calculate a0 .

Solution
Using the given general equation, we note that there are only two choices of n which yield equations containing a2 ,
n = 0 or 2.
i.e. a0 – a2 = 9 from n = 0
a2 – 3a0 = 25 from n = 2
Adding the first equation to the second, we obtain –2a0 = 34 , so a0 = –17 .
ANSWER: –17

8. In the diagram, ∆ ABC is equilateral and the radius of its E C D

inscribed circle is 1. A larger circle is drawn through the vertices
of the rectangle ABDE .
What is the diameter of the larger circle?
A B

Solution
First, we calculate the side length of the equilateral triangle ABC . C
Let O be the centre of the smaller circle and P the point of tangency of
the circle to the side AB .
O 30°
Join OP and OB . Then ∠OPB = 90° by tangency and ∠OBP = 30°
by symmetry since ∠CBA = 60° . A P B
Since OP = 1 and ∆ BOP is 30°-60°-90°, then OB = 2 and BP = 3 . Thus AB = 2 3 .
Also by symmetry, CO = OB = 2 , so CP = 3 .

Since ABDE is a rectangle and CP ⊥ AB , then AE = 3 .

We now look at the rectangle ABDE and its circumcircle.
Since ABDE is a rectangle, ∠ EAB = 90° . E D
So BE is a diameter.
By Pythagoras, 3
BE 2 = EA2 + AB 2

( )
2
=3 + 2 3
2 A B
2 3
= 21
The diameter is 21 .
ANSWER: 21
 Part B

1. Triangle ABC has vertices A(0, 0) , B(9, 0 ) and C (0, 6) . The points P and Q lie on side AB such that AP = PQ = QB .
Similarly, the points R and S lie on side AC so that AR = RS = SC . The vertex C is joined to each of the points P and Q.
In the same way, B is joined to R and S.
(a) Determine the equation of the line through the points R and B.
(b) Determine the equation of the line through the points P and C.
(c) The line segments PC and RB intersect at X, and the line segments QC and SB intersect at Y. Prove that the
points A, X and Y lie on the same straight line.

Solution
Since A(0, 0) , B(9, 0 ) and AP = PQ = QB , then P has coordinates (3, 0 ) and Q has coordinates (6, 0) .
Similarly, R is the point (0, 2) and S is the point (0, 4).
2–0 2
(a) Since R is (0, 2) and B is (9, 0), then the slope of RB is m = = – and so the equation of the line is
0–9 9
2
y – 2 = – (x – 0)
9
2
y = – x+ 2
9
0–6
(b) Since P is (3, 0) and C is (0, 6), then the slope of PC is m = = – 2 and so the equation of the line is
3–0
y – 0 = –2(x – 3)
y = –2x + 6
C(0, 6)
(c) First, we determine the coordinates of X.
Equating the lines from (a) and (b), we have S(0, 4)
2 Y
– x + 2 = – 2x + 6
9 R(0, 2)
16 X
x=4
9
9 A(0, 1) P(3, 0)
x= Q(6, 0) B(9, 0)
4

and substituting into y = – 2 x + 6 = –2  + 6 = , so X is the point  ,  .

9 3 9 3
 4 2  4 2
We calculate the equations of the lines QC and SB as in (a) and (b).
0–6
For QC , slope = = –1 and so y – 0 = –1(x – 6) or y = – x + 6.
6–0
0– 4 4 4 4
For SB , slope = = – and so y – 0 = – (x – 9) or y = – x + 4 .
9–0 9 9 9
So the point Y, which lies at the intersection QC and SB , we obtain by equating these lines
4
–x + 6 = – x + 4
9
5
2= x
9
18
x=
5
and thus Y is the point  ,  .
18 12 18 12
and so y = – x + 6 = – +6 =
5 5  5 5
3
–0
Now the line through A(0, 0) and X  ,  has slope m = 29
9 3 2 2
= and so is y = x .
 4 2 –0 3 3
4
 12 2  18 
The point Y lies on this line, as = . [This could be done with L.S./R.S. format using equation of line.]
5 3 5 
Therefore A, X, Y lie on the same line.

2. In ∆ ABC , the points D, E and F are on sides BC , CA and AB , A

respectively, such that ∠ AFE = ∠ BFD , ∠ BDF = ∠CDE , and E
F
∠CED = ∠ AEF .
(a) Prove that ∠ BDF = ∠ BAC .
(b) If AB = 5 , BC = 8 and CA = 7 , determine the length of
B D C
BD .

Solution
(a) Let ∠ AFE = ∠ BFD = x A
∠BDF = ∠CDE = y z E
∠CED = ∠AEF = z F x z
x
Thus ∠ FAE = 180° – x – z
∠FBD = 180° – x – y y y
∠ECD = 180° – y – z B D C
and these 3 angles add to 180° , so
540° – 2(x + y + z) = 180°
x + y + z = 180°
Since ∠ FAE + ∠ AFE + ∠ AEF = 180° (from ∆ AEF )
∠FAE + x + z = x + y + z
∠FAE = y
Therefore ∠ BDF = ∠ BAC .

(b) Similarly to part (a), ∠ ECD = ∠ BFD = x , ∠ FBD = ∠CED = z .

BD BA 5 CD CA 7 AE AB 5
By equal angles, ∆ ABC ~ ∆ DBF ~ ∆ DEC ~ ∆ AEF and so = = , = = , = = .
BF BC 8 CE CB 8 AF AC 7
Therefore, let BD = 5k , BF = 8k , CD = 7l , CE = 8l , AE = 5m , AF = 7m for some k, l, m.
Then 5k + 7l = 8 (1) A
7m y 5m E
7m + 8k = 5 (2)
x z
5m + 8l = 7 (3) F z
x 8l
Determining 7 × (3) – 5 × (1) to eliminate m, we get 8k
56l – 40k = 49 – 25 = 24 z y y x
7l – 5k = 3 (4) B 5k D 7l C
5
Calculating (1) – (4), we get 10 k = 5 or BD = 5k = .
2
3. (a) Alphonse and Beryl are playing a game, starting with the
geometric shape shown in Figure 1. Alphonse begins the
game by cutting the original shape into two pieces along one
of the lines. He then passes the piece containing the black
Figure 1
triangle to Beryl, and discards the other piece.
Beryl repeats these steps with the piece she receives; that is to say she cuts along the length of a line, passes the
piece containing the black triangle back to Alphonse, and discards the other piece. This process continues, with
the winner being the player who, at the beginning of his or her turn receives only the black triangle. Show, with
justification, that there is always a winning strategy for Beryl.

Solution
We first consider Alphonse’s possible moves to begin the game. We can assume, without loss of generality, that he cuts
on the left side of the black triangle.

Case 1
Alphonse removes two white triangles, leaving .
In this case, Beryl removes only one white triangle, and passes the shape back to Alphonse, forcing him to
remove the last white triangle and lose.

Case 2

Alphonse removes one white triangle only, leaving .

Beryl removes both of the white triangles on the right, leaving Alphonse in the same position as in Case 1 for his second
turn.
Therefore Beryl can always win, regardless of Alphonse’s strategy.
(b) Alphonse and Beryl now play a game with the same rules
as in (a), except this time they use the shape in Figure 2
and Beryl goes first. As in (a), cuts may only be made
along the whole length of a line in the figure. Is there a
strategy that Beryl can use to be guaranteed that she will
win? (Provide justification for your answer.)

Figure 2

Solution
We show that, again, Beryl always has a winning strategy.
The strategy is to reduce the shape in Figure 2 to the shape in Figure 1, and to have Alphonse make the first cut at this
stage. Beryl also knows that if she is forced into a position of being the first to cut when Figure 2 is reduced to Figure 1,
then Alphonse can force her to lose.
We number the lines on the diagram for convenience. (9)
(1)
(8)
(2)
(7)
(3)

(4) (5) (6)

We can assume without loss of generality (because of symmetry) that Beryl cuts along (1), (2) or (3) to begin.
If she cuts (2) or (3), then Alphonse cuts the other of these two and leaves Beryl with Figure 1, where she will lose.
Therefore Beryl cuts (1) to begin.

If Alphonse now cuts (2) or (3), Beryl cuts the other of these two and passes the shape in Figure 1 back to Alphonse, and
so he loses.
If Alphonse cuts (8) or (9), Beryl cuts the opposite and passes the shape in Figure 1 to Alphonse, and so he loses.
(Similarly, if he cuts (5) or (6)).
So assume that Alphonse cuts (4) or (7), say (4) by symmetry.

If Beryl now cuts any of (2), (3), (5), (6), (8), or (9), then Alphonse can force Beryl to lose, in the same way as she could
have forced him to lose, as above. So Beryl cuts (7).

Now Alphonse is forced to cut one of (2), (3), (5), (6), (8), or (9), and so Beryl makes the appropriate cut, passing the
shape in Figure 1 back to Alphonse, and so he must lose.
Therefore Beryl always can have a winning strategy.
4. A sequence t1, t2 , t3 , ..., tn of n terms is defined as follows:
t1 = 1 , t2 = 4 , and tk = tk –1 + tk –2 for k = 3, 4, ..., n .
Let T be the set of all terms in this sequence; that is, T = {t1 , t2 , t3 , ..., tn }.
(a) How many positive integers can be expressed as the sum of exactly two distinct elements of the set T ?

Summary
Part (a) 4✓
Part (b) 6✓

Solution
tk > 0 for all k, 1 ≤ k ≤ n .
Also tk < tk+1 for all k ≤ n – 1 since tk +1 = tk + tk –1 .
Hence the sequence is monotone increasing.
 n
The sum of any pair of terms is an integer and there are   pairs.
 2
Can any two pairs produce the same integer?
Consider ta + tb and tc + td . Clearly if tb = td then ta = tc and vice versa, implying the same pair.
Hence none of the four terms is equal, so we can assume one term, say td to be the largest.
Then td = td –1 + td –2 ≥ ta + tb , since the maximum values of ta and tb are td –1 and td –2 and they cannot be alike.
 n
But since tc > 0 , tc + td > ta + td and there are no two pairs that add to the same integer, so there are exactly   integers
 2
possible.
(b) How many positive integers can be expressed as the sum of exactly three distinct elements of the set T ?

Solution
Consider ta + tb + tc and td + te + t f . If any of the first three equals any of the second three we are left with pair
sums of the remaining ones being equal, which is impossible from part (a). Hence there are six unlike terms, and
again we can assume one, say t f , to be the greatest.
It is clearly possible for equality by setting ta and tb equal to t f –1 and t f –2 and then td and te equal to tc –1 and
tc –2 .
In how many ways can this be done for given f. Clearly,6 ≤ f ≤ n , and since 2 < c < f – 2 , for any given f there
n
 n – 4
are f – 5 choices for c and the number of ways possible is ∑
( f – 5) = 1 + 2 + 3 + + (n – 5) = 
 2 
.
f =6
 n  n – 4
There are a maximum of   –   possible integers.
 3  2 
Of these, are any two of like sum?
In ta + tb + tc , the maximal values are t f +1 , t f –3 , and t f –4 , since if one is t f –1 and one t f –2 we revert to the
discussed state. Hence
ta + tb + tc ≤ t f –1 + t f –3 + t f –4
= t f –1 + t f –2
= tf .
But td + te + t f > t f .
Hence there are no other triples for which equality of sums exist, and the number of possible integers is
 n  n – 4
  – .
 3  2 
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

The
Canadian Open
Mathematics Challenge
Wednesday, November 28, 2001

1 © 2001 Canadian Mathematical Society

Time: 2 hours
2

Calculators are NOT permitted.

Do not open this booklet until instructed to do so.

There are two parts to the paper.

PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value for each
question by entering the correct answer in the space provided. Any work you do in obtaining an answer
will be considered for part marks if you do not have the correct answer, provided that it is done in the
space allocated to that question in your answer booklet.

PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must be written
in the appropriate location in the answer booklet. Rough work should be done separately. If you require
extra pages for your finished solutions, paper will be provided by your supervising teacher. Any extra
papers should be placed inside your answer booklet.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution poorly
presented will not earn full marks.

NOTE: At the completion of the contest, insert the information sheet inside the answer booklet.
 Canadian Open Mathematics Challenge

NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations and answers will be expressed as exact numbers such
as 4 p, 2 + 7 , etc.
4. Calculators are not allowed.

1. An operation “ —” is defined by a — b = a + 3 .
2 b

What is the value of (2 — 0)—(0 —1) ?

2. In the given diagram, what is the value of x? D

5x∞
A C E G
3x∞ 2x∞
6x∞
F
4x∞

3. A regular hexagon is a six-sided figure which has all of its angles equal and all of its side lengths equal.
If P and Q are points on a regular hexagon which has a side length of 1, what is the maximum possible
length of the line segment PQ?

4. Solve for x:
( )
2 2 2 x = 4 x + 64 .

5. Triangle PQR is right-angled at Q and has side lengths PQ = 14 P

and QR = 48. If M is the midpoint of PR, determine the cosine of
M
– MQP .

Q R

tn – 1
6. The sequence of numbers t1, t2 , t3 , ... is defined by t1 = 2 and tn +1 = , for every positive integer n.
tn + 1
Determine the numerical value of t999 .

7. If a can be any positive integer and

2x + a = y
a+y= x
x+y=z
determine the maximum possible value for x + y + z.
8. The graph of the function y = g( x ) is shown. y

Determine the number of solutions of the equation

g( x ) – 1 = .
1
2
4

x
–4 –2 0 2 4

–2

–4

PART B
1. The triangular region T has its vertices determined by the intersections of the three lines x + 2y = 12,
x = 2 and y = 1.
(a) Determine the coordinates of the vertices of T, and show this region on the grid provided.
(b) The line x + y = 8 divides the triangular region T into a quadrilateral Q and a triangle R.
Determine the coordinates of the vertices of the quadrilateral Q.
(c) Determine the area of the quadrilateral Q.

2. (a) Alphonse and Beryl are playing a game, starting with a pack of 7 cards. Alphonse begins by
discarding at least one but not more than half of the cards in the pack. He then passes the remaining
cards in the pack to Beryl. Beryl continues the game by discarding at least one but not more than half
of the remaining cards in the pack. The game continues in this way with the pack being passed back
and forth between the two players. The loser is the player who, at the beginning of his or her turn,
receives only one card. Show, with justification, that there is always a winning strategy for Beryl.
(b) Alphonse and Beryl now play a game with the same rules as in (a), except this time they start with
a pack of 52 cards, and Alphonse goes first again. As in (a), a player on his or her turn must discard
at least one and not more than half of the remaining cards from the pack. Is there a strategy that
Alphonse can use to be guaranteed that he will win? (Provide justification for your answer.)

3. (a) If f ( x ) = x + 6 x + c , where c is an integer, prove that f (0) + f ( –1) is odd.

2

(b) Let g( x ) = x 3 + px 2 + qx + r , where p, q and r are integers. Prove that if g(0) and g( –1) are both odd,
then the equation g( x ) = 0 cannot have three integer roots.

4. Triangle ABC is isosceles with AB = AC = 5 and BC = 6. Point D A

lies on AC and P is the point on BD so that – APC = 90∞ . If
– ABP = – BCP , determine the ratio AD:DC.
D

B C
 Comments on the Paper Commentaires sur l'épreuve

PART A PARTIE A
1. Solution 1. Solution
By the definition, 2∇0 = 2 2 + 30 = 4 + 1 = 5 D’après la définition, on a : 2∇0 = 2 2 + 30 = 4 + 1 = 5
0∇1 = 0 2 + 31 = 0 + 3 = 3 0∇1 = 0 2 + 31 = 0 + 3 = 3
and so (2∇0)∇(0∇1) = 5∇3 Donc : (2∇0)∇(0∇1) = 5∇3

= 52 + 33 = 52 + 33
= 25 + 27 = 25 + 27
= 52 = 52
Commentaires
Comments
Les élèves ont bien réussi cette question. La plupart des
This question was quite well done. Most students correctly
élèves ont correctement interprété les opérations données
interpreted the given operation to do the required calcula-
nécessaires aux calculs requis.
tions. Moyenne: 3,6
Average: 3.6
2. Solution D
2. Solution D D’après le diagramme, 180° – 7x°
5x°
180° – 8x°

From the diagram, 180° – 7x°

5x°
180° – 8x° ∠ACB = 180o − 7 x o et A C E G
3x° 2x°

∠ACB = 180o − 7 x o and

A
3x°
C E
2x°
G
∠FEG = 180o − 8 x o . 6x°

∠FEG = 180o − 8 x o .
6x°
Donc ∠DCE = 180o − 7 x o F
F 4x°

Therefore, ∠DCE = 180o − 7 x o 4x° et ∠DEC = 180o − 8 x o . B

and ∠DEC = 180o − 8 x o , so B Dans le triangle CDE, on a donc :
from ∆CDE , 5 x o + 180o − 7 x o + 180o − 8 x o = 180o (*)
5 x o + 180o − 7 x o + 180o − 8 x o = 180o (*)
360o − 10 x o = 180o
o o o
360 − 10 x = 180
10 x o = 180o
10 x o = 180o x = 18
x = 18 Commentaires
Comments Chapeau! La quasi-totalité des concurrents ont bien maîtrisé
Extremely well done! Almost all of the competitors had cette question qui nécessitait l’application des connaissances
a good handle on dealing with angles in triangles. des angles aux triangles.
Average: 4.1 Moyenne: 4,1
3. Solution 1 3. Solution 1
A B A B
Let ABCDEF be a regular hexagon with Soit ABCDEF un hexagone régulier dont
o les côtés ont une longueur de 1. Chacun
a side length of 1. Each angle is 120 .
Thus, if we join FC, EB, DA, each of the
F C de ses angles mesure 120o . Les segments F C

interior angles is bisected, and so each FC, EB et DA sont des bissectrices des
E D angles. L’hexagone est donc décomposé E D
part is 60o . Thus the hexagon is decom-
en 6 triangles équilatéraux.
posed into 6 equilateral triangles, as shown. La distance maximale possible entre deux points sur
The maximum distance between any two points on the l’hexagone est celle entre deux sommets opposés. Puisque
hexagon is the distance between two opposite vertices. chaque triangle équilatéral a des côtés de longueur 1, la
Since each of the triangles is equilateral with a side length longueur maximale possible du segment est égale à 2.
of 1, the diagonal distance is 2, ie. the maximum possible
length of PQ is 2. Version abrégée de la solution 1
Comme l’indique le diagramme, l’hexagone régulier peut
Brief version of Solution 1 être divisé en 6 triangles équilatéraux ayant des côtés de
A regular hexagon with side length 1 can be decomposed longueur 1. La distance maximale possible entre deux
into 6 equilateral triangles with a side length of 1, as shown. points sur l’hexagone est celle entre deux sommets opposés.
Elle est égale à 2.
The maximum distance between any two points is between
opposite vertices, and this distance is 2. 6
 Comments on the Paper Commentaires sur l'épreuve

Solution 2 Solution 2
The maximum distance is between A B Par symétrie, la distance maximale A B

two opposite vertices, say F and C entre deux points sur l’hexagone est
celle entre deux sommets opposés,
by symmetry. 60° 60° 60° 60°
F R S C disons F et C. Aux points A et B, on F R S C
Drop perpendiculars from A and B abaisse des perpendiculaires AR et AS au segment FC.
to meet FC at R and S respectively. Puisque AB = 1 et que AB est parallèle à RS, alors RS = 1.
Since AB = 1 and AB is parallel to RS, then RS = 1.
Par symétrie, FR = CS . Or FR = AF cos60o = 1( 12 ) = 12 .
By symmetry, FR = CS . But
Donc CF = 2 et la longueur maximale possible est donc
FR = AF cos60o = 1( 12 ) = 12 . Therefore, CF = 2, and so égale à 2.
the maximum possible distance is 2.
Commentaires
Comments Une grande partie des difficultés de cette question résidait
The key problems here were to interpret the question and dans son interprétation. Les élèves devaient par la suite
to then figure out that the longest distance between any two déduire que la distance entre deux sommets opposés
points is the distance between opposite vertices. The représentait la distance maximale entre deux points. La façon
easiest way to calculate this length was to break the la plus rapide de la calculer consistait à diviser l’hexagone en
hexagon up into 6 equilateral triangles each of side length six triangles équilatéraux dont chacune des arêtes équivalait
one. If you didn’t notice this, have a look at this idea. à un. Nous invitons les élèves qui n’ont pas envisagé cette
Average: 3.6 approche de résolution de l’examiner de plus près.
Moyenne: 3,6
4. Solution
Solving the equation, 4. Solution

( )
2 2 2 x = 4 x + 64 or ( )
2 2 2 x = 4 x + 64
On a :
( )
2 2 2 x = 4 x + 64 ou ( )
2 2 2 x = 4 x + 64
2( 4 ) = 4
x x
+ 64 2( 2 ) = 2
2x 2x
+ 64
2( 4 ) = 4
x x
+ 64 2( 2 ) = 2
2x 2x
+ 64
4 = 64
x 2x
2 = 64
x=3 4 x = 64 2 2 x = 64
2x = 6
x=3 2x = 6
x=3
x=3
Comments
This question was reasonably well done. Students who are Commentaires
comfortable dealing with exponents had a great deal of Les élèves ont généralement bien réussi cette question, tout
success on this question. Many students continue to have particulièrement ceux qui maîtrisaient bien le concept
difficulty with exponents. d’exposant (qui d’ailleurs pose encore des difficultés pour
Average: 3.8 bon nombre d’élèves).
Moyenne: 3,8
5. Solution 1
5. Solution 1
Join M to Q. P P
On joint M et Q.
M M
Through M, draw a line parallel F Au point M, on trace un segment F
to QR meeting PQ at F. parallèle au côté QR . Ce segment
Q R Q R
Therefore, PF = FQ = 7 and MF = 24 . coupe le côté PQ en F.
By Pythagoras, MQ = 25, and so Donc PF = FQ = 7 et MF = 24 .
D’après le théorème de Pythagore dans le triangle MFQ,
cos( ∠MQP ) = cos( ∠MQF ) MQ = 25, d’où :
FQ cos( ∠MQP ) = cos( ∠MQF )
=
MQ FQ
=
7 MQ
=
25 7
=
7 25
 Comments on the Paper Commentaires sur l'épreuve

Solution 2 Solution 2
P On joint M et Q.
Join M to Q. By Pythagoras, P
M D’après le théorème de M
14
PR = 14 2 + 482 = 50 . Pythagore dans le triangle
14

Since M is the midpoint of the Q R Q R

PQR, PR = 14 2 + 482 = 50 .
hypotenuse, then MQ = MP = MR . (This is because the
Puisque M est le milieu de l’hypoténuse,
circle circumscribed around ∆PQR has PR as diameter
o
MQ = MP = MR . (Puisque ∠PQR = 90o , PQ est le
(since ∠PQR = 90 ) and so M is the centre and thus MP, diamètre du cercle circonscrit au triangle et M en est le
MQ and MR are radii.) Therefore, ∠MQP = ∠MPQ and centre. Donc MP, MQ et MR sont des rayons.)
so cos( ∠MQP ) = cos( ∠MPQ ) = 14
50 =
7
25 . Donc ∠MQP = ∠MPQ , d’où
cos( ∠MQP ) = cos( ∠MPQ ) = 14 50 = 25
7 .
Comments
This is a nice question because it can be done with a Commentaires
Euclidean approach or a more analytic approach. Some Il s’agissait d’une question intéressante en raison de ses
students had difficulty when they encountered a triangle that deux méthodes de résolution : l’approche euclidienne ou
was not right-angled in which they had to calculate a cosine. une approche analytique plus générale. Quelques élèves
Average: 2.7 n’ont pu contourner la difficulté que posait la présence d’un
triangle sans angle droit, ce qui nécessitait le calcul du
6. Solution cosinus.
We calculate the first few terms in the series Moyenne: 2,7
t1 = 2
t1 − 1 2 − 1 1 6. Solution
t2 = = = Les premiers termes de la suite sont :
t1 + 1 2 + 1 3
t1 = 2
t −1 1
−1 1
t3 = 2 = 3
=− t1 − 1 2 − 1 1
t2 + 1 1
3 +1 2 t2 = = =
t1 + 1 2 + 1 3
t3 − 1 − 12 − 1
t4 = = = −3 t2 − 1 1
−1 1
t3 + 1 − 12 + 1 t3 = = 3
=−
t2 + 1 1
3 +1 2
t4 − 1 −3 − 1
t5 = = =2 t3 − 1 − 12 − 1
t4 + 1 −3 + 1 t4 = = = −3
t3 + 1 − 12 + 1
Since a term in the sequence depends only on the previous
one, then the sequence will cycle with a period of 4. Thus, t4 − 1 −3 − 1
t5 = = =2
t1 = t5 = L = t997 = t1001 = 2 . Therefore, t998 = 13 and t999 = − 12 . t4 + 1 −3 + 1
Puisque chaque terme de la suite dépend seulement du
Comments terme précédent, les valeurs se reproduisent à tous les 4
This question was very well done! Most students quickly termes. Les valeurs ont donc un cycle de longueur 4.
determined after a few calculations that the sequence was Donc t1 = t5 = L = t997 = t1001 = 2 . Donc t998 = 13 et
cyclic, and on this basis determined the value of the required t999 = − 12 .
term. The nicest solution by a student was to repeatedly
Commentaires
apply the definition:
t n −1 −1 Cette question a été très bien réussie! La plupart des élèves
t n − 1 tn −1 +1 − 1 (t n −1 − 1) − (t n −1 + 1) 1
t n +1 = = t −1 = =− ont conclu, moyennant quelques calculs, qu’ils se trouvaient
tn + 1 n −1
+ 1 (t n −1 − 1) + (t n −1 + 1)
t n −1 +1
t n −1
en présence d’une séquence cyclique : ils ont par la suite été
In a similar way, en mesure de trouver la valeur du terme requis. La solution
1 la plus astucieuse consistait à appliquer la définition de
t n −1 = −
tn− 3 façon répétitive :
and so t n −1 −1
1
t n +1 = − 1 = t n − 3 t −1
t n +1 = n =
t n −1 +1 − 1
=
(tn−1 − 1) − (tn−1 + 1) = − 1
−t
n −3
tn + 1 t n −1 −1
t n −1 +1 + 1
(tn−1 − 1) + (tn−1 + 1) tn−1
8
 Comments on the Paper Commentaires sur l'épreuve

Thus, the sequence repeats every four terms. What a nice De la même manière,
solution! 1
t n −1 = −
Average: 2.7 tn− 3
On peut conclure que
1
7. Solution t n +1 = − 1 = t n − 3
−t
We treat a as a constant and solve for x, y, z in terms of n −3

a. Par conséquent, on peut affirmer que la séquence se répétera

From the second equation, y = x − a (*) à tous les quatre termes. Il suffisait d’y penser!
Therefore, from the first equation, Moyenne: 2,7
2x + a = x − a
7. Solution
x = −2 a On exprime d’abord x, y et z en fonction de a.
Substituting into (*), y = −3a . D’après la deuxième équation, y = x − a (*) .
Substituting into the 3rd equation, z = x + y = −5a . On reporte y = x – a dans la première équation :
So x + y + z = −10 a . 2x + a = x − a
Since a is a positive integer, the maximum value for x = −2 a
x + y + z is –10 (which occurs when a = 1). On reporte x = –2a dans l’équation (*) pour obtenir y = −3a .
On reporte y = −3a dans la troisième équation pour obtenir
Comments z = x + y = −5a .
This question was done very well. We anticipated that this Donc x + y + z = −10 a .
would be quite a difficult question, and so we were Puisque a peut prendre la valeur de n’importe quel entier
extremely happy with the results. Many students handled strictement positif, la valeur maximale possible de
this system of equations with relative ease by figuring out l’expression x + y + z est –10. On l’obtient lorsque a = 1.
that they had to solve for x, y and z in terms of a.
Commentaires
Average: 2.8
Nous avons été agréablement surpris du succès des élèves à
8. Solution 1 (Graphical) résoudre cette question qui se voulait extrêmement difficile;
y y de nombreux élèves ont facilement résolu ce système
d’équations où il fallait trouver la valeur de x, y et de z en
fonction de a.
4 4
Moyenne: 2,8
2 2

x 0 x
8. Solution 1 (Graphique)
–4 –2 0 2 4 –4 –2 2 4 y y

–2 –2

y = g( x ) y = g( x ) – 1 4 4

y 2 2
so the number of solutions
x 0 x
of g( x ) − 1 = 1
2 is 8, –4 –2 0 2 4 –4 –2 2 4

4 –2 –2
from the third graph.
2
1 y = g( x ) y = g( x ) – 1
y=2
x
–4 –2 0 2 4
y
–2
Selon le troisième graphique,
l’équation g( x ) − 1 = 1
2
y = g( x ) –1 4
admet 8 solutions.
2
1
y=2
x
–4 –2 0 2 4

–2

9 y = g( x ) –1
 Comments on the Paper Commentaires sur l'épreuve

Solution 2 (Algebraic) Solution 2 (Algébrique)

From the original equation g( x ) − 1 = 1
2, using the D’après l’équation g( x ) − 1 = 12 , on a :
definition of absolute value we obtain, g( x ) − 1 = 1
or
ou g( x ) − 1 = − 12
2
g( x ) − 1 = 1
2 or g( x ) − 1 = − 12 g( x ) = 3
2 or
ou g( x ) = 1
2
g( x ) = 3
2 or g( x ) = 1
2 g( x ) = ± 23 or
ou g( x ) = ± 12
g( x ) = ± 23 or g( x ) = ± 12 D’après le graphique donné :
a) g( x ) = 23 admet 1 solution,
From the original graph,
(a) g( x ) = 23 has 1 solution, b) g( x ) = − 23 admet 1 solution,

(b) g( x ) = − 23 has 1 solution, c) g( x ) = 1

2 admet 3 solutions,

(c) g( x ) = 1
has 3 solutions, d) g( x ) = − 12 admet 3 solutions.
2
(d) g( x ) = − 12 has 3 solutions. Donc g( x ) − 1 = 1
2 admet 8 solutions.

Therefore, g( x ) − 1 = 1
2 has 8 solutions. Commentaires
Cette question a permis aux élèves de mettre à l’épreuve
Comments
leurs connaissances de la valeur absolue dans le contexte
This question was a good test of the concept of absolute
d’une représentation graphique. Certains élèves ont choisi
value from a graphical perspective. Some students used a
une approche graphique pour déterminer le graphique
graphical approach to convert the original graph to the
requis à partir de la transposition du premier graphique. En
desired one. Quite a few students used the graphical ap-
fait, plusieurs élèves ont eu recours à cette méthode de
proach to determine the potential values for either g( x ) or
résolution pour trouver les valeurs possibles de g( x ) ou
g( x ) and then read the appropriate number of solutions off
g( x ) ; ils ont ensuite déterminer le nombre de solutions
the graph. Students who tried to determine the actual
appropriées à partir des informations du graphique. Les
equation of the curve tended to get bogged down in their
élèves qui ont tenté de déterminer l’équation de la courbe
calculations.
ont en général éprouvé des difficultés au niveau des calculs.
Average: 2.0 Moyenne: 2,0
Part B
1. Solution Partie B
y
(a) The lines x = 2 and 1. Solution
y = 1 intersect at (2,1) .
x=2
a) Les droites définies y
The lines x = 2 and (2, 5) par x = 2 et y = 1 se x=2

x + 2 y = 12 intersect x + 2y = 12
coupent au point (2,1) .
(2, 5)
at (2, 5) , since (2, 1)
(10, 1)
y=1
x
Les droites définies
x = 2 ⇒ 2 + 2 y = 12 ⇒ y = 5. par x = 2 et x + 2 y = 12 x + 2y = 12

(10, 1)
se coupent au point y=1
The lines y = 1 and x + 2 y = 12 intersect at (10,1) , (2, 1)
x
(2, 5) , car en reportant
since y = 1 ⇒ x + 2 = 12 ⇒ x = 10 .
x = 2 dans x + 2 y = 12 , on obtient 2 + 2y = 12, d’où y
(b) x + y = 8 intersects y
= 5. Les droites définies par y = 1 et x + 2 y = 12 se
x = 2 at (2, 6) , which x=2

is above the point of coupent au point (10,1) , car en reportant y = 1 dans

intersection of x = 2
(2, 5)
(4, 4)
x + 2 y = 12 , on obtient x + 2 = 12, d’où x = 10.
x + 2y = 12
b) Les droites d’équations y
and x + 2 y = 12 . Q
R (10, 1)
y=1 x + y = 8 et se coupent x = 2
x + y = 8 intersects (2, 1) (7, 1)
x
au point (2, 6) . Ce
y = 1 at (7,1). (2, 5)
point est au-dessus du (4, 4)
To find the intersection point of x + y = 8 and point d’intersection des Q
x + 2y = 12

R
x + 2 y = 12 , subtract the first equation from the sec- droites d’équations (2, 1) (7, 1)
(10, 1)
y=1
x
ond to obtain y = 4 , so x = 4. x = 2 et x + 2 y = 12 .
10
 Comments on the Paper Commentaires sur l'épreuve

Therefore, the vertices of Q are (2,1) , (2, 5) , ( 4, 4) , Les droites d’équations x + y = 8 ett y = 1 se coupent au
(7,1). point (7,1).
(c) Area of Q = Area of T − Area of R Pour déterminer le point d’intersection des droites
définies par x + y = 8 et x + 2 y = 12 , on soustrait la
= 1
2
( 8)( 4 ) − 12 ( 3)( 3) première équation de la deuxième, membre par membre,
= 16 − 29 pour obtenir y = 4 , d’où x = 4.
Les coordonnées du quadrilatère Q sont (2,1) , (2, 5) ,
= 23
2
(4, 4) et (7,1).
Comments c) Aire de Q = Aire de T – Aire de R
This question was exceptionally well done. Students = 1
2
( 8)( 4 ) − 12 ( 3)( 3)
either approached this strictly graphically or with a com-
bination of graphical and analytic approaches. In either = 16 − 29
case, they tended to do very well. Part (c) was quite well = 23
2
done. Students managed to determine one of the many
Commentaires
ways to calculate the area of Q, either by subtracting the
Les élèves ont exceptionnellement bien réussi cette question
area of R from the area of T, by breaking Q up into two
peu importe la méthode de résolution utilisée. Des élèves ont
triangles, or by breaking Q up into one rectangle and two ainsi choisi la méthode de résolution graphique tandis que
triangles. d’autres ont opté pour un agencement des méthodes graphique
Average: 8.4 et analytique. Ils ont tout particulièrement bien réussi la partie
(c). Ils ont trouvé l’une des nombreuses façons de déterminer
2. (a) Solution 1 l’aire de Q, soit en soustrayant l’aire de R de celle de T, ou en
We define a “losing position” to be a number of divisant Q en deux triangles ou encore en un rectangle et deux
cards, such that if a player receives this number of triangles.
Moyenne: 8,4
cards at the beginning of his or her turn, he or she is
guaranteed to lose assuming best play by both play- 2. a) Solution 1
ers. A “winning position” is defined similarly. On définit une « position perdante » comme étant un
nombre de cartes qu’un joueur ou une joueuse reçoit au
Clearly, by the rules of the game, 1 is a losing début de son tour et qui lui assure une défaite si les deux
position. adversaires jouent à leur meilleur. On définit une « position
For a player to receive 1 card at the beginning of a gagnante » de façon semblable.
turn, the previous player must start with 2 cards. Selon les règles du jeu, 1 est une position perdante.
(This is true since a player can never remove more Pour qu’un joueur reçoive 1 carte au début de son tour, le
than half of the deck, so the number of cards at the joueur précédent doit commencer avec 2 cartes. (Un joueur
beginning of the previous turn can never be more ne peut retirer plus de la moitié des cartes, donc le nombre
than double that of the current turn.) Therefore, 2 is de cartes qu’il reçoit ne peut être supérieur au double du
nombre de cartes remises au joueur suivant.)
a winning position, since a player starting with 2
Donc 2 est une position gagnante, puisqu’un joueur qui
cards can only remove 1, and so passes 1 card to the reçoit 2 cartes peut seulement en enlever une et il remet 1
other player, who loses. carte à l’adversaire qui perd.
Is 3 a winning position or a losing position? Est-ce que 3 est une position gagnante ou perdante?
Given a pack of 3 cards, the rules of the game say that Si on reçoit 3 cartes, on peut seulement retirer une carte du
a player can only remove 1 card, and so pass a pack jeu et on remet 2 cartes à l’adversaire qui reçoit une position
gagnante. Donc 3 est une position perdante.
of 2 cards (a winning position) to the other player.
Therefore, 3 is a losing position. On peut constater que 4, 5 et 6 sont des positions gagnantes,
car on peut retirer respectivement 1, 2 ou 3 cartes et
We can then see that 4, 5 and 6 are all winning remettre 3 cartes à l’adversaire qui reçoit alors une position
positions, as given 4, 5 or 6 cards, a player can perdante. On constate que 7 est une position perdante, car
remove 1, 2 or 3 cards respectively to pass the other on doit retirer 1, 2 ou 3 cartes et on remet respectivement
player 3 cards, a losing position. 6, 5 ou 4 cartes à l’adversaire qui reçoit à chaque fois une
11
position gagnante.
Comments on the Paper Commentaires sur l'épreuve

Therefore, 7 is a losing position, since a player remov- Si Alain reçoit 7 cartes, Brigitte peut donc toujours
gagner.
ing 7 cards must remove 1, 2 or 3 cards, thus giving the
other player 6, 5 or 4 cards respectively, any of which Résumé de la stratégie de Brigitte
is a winning position. So if Alphonse starts with 7 • Elle recevra 4, 5 ou 6 cartes d’Alain.
cards, Beryl can always win. • Elle retirera 1, 2 ou 3 cartes de manière à remettre 3
cartes à Alain.
Summary of Beryl’s Strategy • Alain est forcé à retirer 1 carte et à remettre 2 cartes à
• She will receive 4, 5 or 6 cards from Alphonse. Brigitte.
• Remove 1, 2 or 3 cards in order to pass 3 cards back • Brigitte retire 1 carte et remet 1 carte à Alain qui perd.
to Alphonse.
• Alphonse is forced to remove 1 only, and pass back Solution 2 (suffisante pour recevoir le maximum de points)
2 to Beryl. Alain reçoit 7 cartes et il peut retirer 1, 2 ou 3 cartes pour
remettre 6, 5 ou 4 cartes à Brigitte.
• Beryl removes 1 and passes 1 back, so Alphonse
Brigitte doit retirer 3, 2 ou 1 carte de manière à remettre 3
loses. cartes à Alain.
Solution 2 (Sufficient for full marks) Alain est forcé à retirer 1 carte et à remettre 2 cartes à
Alphonse starts with 7 cards, and so can remove 1, 2 or Brigitte.
Brigitte retire 1 carte, (elle n’a pas d’autre choix) et elle
3 cards, passing 6, 5 or 4 cards to Beryl.
remet 1 carte à Alain qui perd.
Beryl should remove 3, 2 or 1 cards respectively, Brigitte a donc une stratégie gagnante.
leaving 3 cards only, and pass these 3 cards back to
Alphonse. b) Solution 1
Alphonse now is forced to remove 1 card only, and On doit déterminer si 52 est une position gagnante ou
pass 2 back to Beryl. perdante.
Beryl removes 1 card (her only option) and passes 1 Comme dans la partie précédente, on peut démontrer que 8,
9, 10, …, 14 sont des positions gagnantes, puisqu’on peut
back to Alphonse, who thus loses.
retirer suffisamment de cartes pour remettre 7 cartes à
Therefore, Beryl is guaranteed to win. l’adversaire qui reçoit alors une position perdante.
(b) Solution 1 Donc 15 est une position perdante, puisqu’en recevant 15
We must determine if 52 is a winning position or a cartes, on doit remettre de 8 à 14 cartes à l’adversaire qui
reçoit alors des positions gagnantes.
losing position. By a similar argument to above, since
7 is a losing position, 8 through 14 are all winning De la même manière, les nombres de 16 à 30 sont des
positions, since they can all be reduced to 7 in one turn. positions gagnantes, 31 est une position perdante et les
nombres de 32 à 62 sont des positions gagnantes.
Therefore, 15 is a losing position, since given 15 cards,
a player is forced to reduce to some number between Donc 52 est une position gagnante. Alain peut donc utiliser
8 and 14, since no more than 7 cards can be removed. une stratégie gagnante.

Similarly, 16 through 30 are winning positions, 31 is Résumé de la stratégie d’Alain

a losing position, and 32 through 62 are winning • Alain retire 21 cartes du jeu de 52 cartes et remet 31
positions. cartes à Brigitte.
Therefore, 52 is a winning position, so Alphonse has • Si Brigitte retire b1 cartes, 1 ≤ b1 ≤ 15 , Alain retire alors
a winning strategy. 16 − b1 cartes et remet 15 cartes à Brigitte. [On remarque
que c’est toujours permis, car
Summary of Alphonse’s strategy 2(16 − b1 ) = 32 − 2b1 ≤ 31 − b1 et 16 − b1 n’est jamais
• Alphonse removes 21 cards from original 52, and supérieur à la moitié du nombre de cartes reçues.]
pass 31 cards to Beryl. • Si Brigitte retire b2 cartes, 1 ≤ b2 ≤ 7 , retire alors 8 − b2
• If Beryl removes b1 cards with 1 ≤ b1 ≤ 15 , cartes et remet 7 cartes à Brigitte. [Ce nombre est
Alphonse removes 16 − b1 cards to reduce the toujours permis selon un argument semblable.]
pack to 15 cards. [Notice that this is always a • Alain adopte maintenant la même stratégie que Brigitte
dans la partie a).
legal move, since 2(16 − b1 ) = 32 − 2b1 ≤ 31 − b1,
so 16 − b1 is never more than half of the pack.]
12
 Comments on the Paper Commentaires sur l'épreuve

• If Beryl removes b2 cards with 1 ≤ b2 ≤ 7 ,

Solution 2
Alphonse removes 8 − b2 to reduce the pack to 7 Alain retire 21 cartes du jeu de 52 cartes et remet 31 cartes à
cards. [This move is always legal by a similar Brigitte.
argument.]
Si Brigitte retire b1 cartes, 1 ≤ b1 ≤ 15 , Alain retire alors
• Beryl now has 7 cards, so Alphonse should adopt
Beryl’s strategy from (a). 16 − b1 cartes et remet 15 cartes à Brigitte. [On remarque que
c’est toujours permis, car 2(16 − b1 ) = 32 − 2b1 ≤ 31 − b1 et
Solution 2
16 − b1 n’est jamais supérieur à la moitié du nombre de cartes
Alphonse removes 21 cards from original 52, and
reçues.]
passes 31 cards to Beryl.
Si Brigitte retire b2 cartes, 1 ≤ b2 ≤ 7 , retire alors 8 − b2
If Beryl removes b1 cards with 1 ≤ b1 ≤ 15 , Alphonse
cartes et remet 7 cartes à Brigitte. [Ce nombre est toujours
removes 16 − b1 cards to reduce the pack to 15 cards. permis selon un argument semblable.]
[This is always a legal move, since
Alain adopte maintenant la même stratégie que Brigitte dans
2(16 − b1 ) = 32 − 2b1 ≤ 31 − b1, so 16 − b1 is never la partie a). Il a donc une stratégie gagnante.
more than half of the pack.]
Commentaires
If Beryl removes b2 cards with 1 ≤ b2 ≤ 7 , Alphonse Les questions pertinentes à « Alphonse et Beryl » demeurent
removes 8 − b2 to reduce the pack to 7 cards. [This sans contredit un des points marquants du concours du Défi
move is always legal by a similar argument.] ouvert canadien de mathématiques (DOCM). Les élèves ont
Beryl now has 7 cards, so Alphonse should adopt bien réussi la partie (a). La plupart ont rapidement compris
que la position des trois cartes constituait l’aspect essentiel
Beryl’s strategy from (a), so Alphonse has a winning
du problème. Les élèves ont cependant éprouvé des difficultés
strategy. à la partie (b) où bon nombre ont supposé que la méthode de
Comments résolution était liée au nombre égal de cartes paires et
impaires ou à la correspondance du nombre de cartes (si
The "Alphonse and Beryl" questions continue to be a
Beryl prend cinq cartes Alphonse devrait faire de même). Les
highlight of the COMC. Part (a) met with a good degree concurrents devraient lire la solution puis inviter un camarade
of success. Most students quickly realized that the posi- à se laisser prendre au jeu!
tion of 3 cards was the important one on which to focus. Moyenne: 4,0
Part (b) did not meet with as much success – many students
thought that the strategy had something to do with parity 3. a) Solution
(ie. even or odd numbers of cards) or with a "matching" f (0) + f ( −1) = c + (1 − 6 + c)
strategy (if Beryl takes 5 cards, Alphonse should take 5 = 2c − 5
cards). Competitors should have a look at the solution and Puisque c est un entier, 2c est toujours pair et 2c − 5 est
then try playing the game with an unsuspecting friend! donc toujours impair.
Average: 4.0
b) Solution 1
3. (a) Solution Supposons que l’équation g( x ) = 0 admet trois racines
Calculating, entières, a, b et c.
f (0) + f ( −1) = c + (1 − 6 + c) Donc g( x ) = ( x − a)( x − b)( x − c) = x 3 + px 2 + qx + r .
= 2c − 5 Donc g(0) = − abc et puisque g(0) est impair, alors
Now 2c is always even for c an integer, so 2c − 5 chacune des racines doit être impaire pour que leur
produit soit impair.
is always odd.
On a g( −1) = ( −1 − a)( −1 − b)( −1 − c).
(b) Solution 1 Puisque a, b et c sont impairs, alors −1 − a, –1 – b et
Assume that g( x ) = 0 has three integer roots a, b, c, –1 – c sont pairs et g( −1) est donc pair, ce qui est une
ie. contradiction.
g( x ) = ( x − a)( x − b)( x − c) = x 3 + px 2 + qx + r . Donc l’équation g( x ) = 0 ne peut admettre trois racines
Now g(0) = − abc from the above and is odd, so entières.
each of a, b and c must be odd for their product to be
odd. 13
Comments on the Paper Commentaires sur l'épreuve

Therefore, g( −1) = ( −1 − a)( −1 − b)( −1 − c). Solution 2

Since a is odd, −1 − a is even, and so g( −1) is even, a Supposons que l’équation g( x ) = 0 admet trois
racines entières, a, b et c.
contradiction. (In fact, g( −1) is divisible by 8.) Donc
Thus, g( x ) = 0 cannot have three integer roots.
g( x ) = ( x − a)( x − b)( x − c) = x 3 + px 2 + qx + r .
Solution 2 On développe pour obtenir :
Assume that g( x ) = 0 has three integer roots a, b, c, x 3 − ( a + b + c) x 2 + ( ab + ac + bc) x − abc
ie. g( x ) = ( x − a)( x − b)( x − c) = x 3 + px 2 + qx + r = x 3 + px 2 + qx + r
or, expanding,
Puisque g(0) et g(–1) sont impairs, et que g(0) = r
x 3 − ( a + b + c) x 2 + ( ab + ac + bc) x − abc et g( −1) = −1 + p − q + r alors r et –1 + p – q + r
= x 3 + px 2 + qx + r sont impairs.
Donc p − q est impair.
Now g(0) = r so r is odd, and
Donc p ou q doit être pair (ils ne peuvent être
g( −1) = −1 + p − q + r , which is odd. impairs tous les deux, car leur différence serait
Combining these, since r is odd, then p − q is odd too. paire).
Therefore, one of p and q is even (they cannot both be
Puisque r est impair et que r = − abc , alors a, b et c
odd, since odd – odd = even). sont tous impairs. Donc les nombres
Since r is odd and r = − abc , then each of a, b and c p = −( a + b + c )
is odd. This implies that q = ab + ac + bc
p = −( a + b + c ) sont impairs tous les deux, ce qui est une
q = ab + ac + bc contradiction car on vient de conclure que p ou q
doit être pair.
are both odd, a contradiction (since we have shown
Donc l’équation g( x ) = 0 ne peut admettre trois
above that one of p and q must be even.)
racines entières.
Therefore, g( x ) = 0 cannot have three integer roots.
Comments Commentaires
The purpose of part (a) was to have a relatively straightfor- Les élèves ont généralement très bien réussi la partie (a)
ward "proof" question, which required students to write a où il fallait résoudre une question au moyen d’un
raisonnement logique, mais ont éprouvé davantage de
logical argument. In general, students did exceptionally
difficultés avec la partie (b). De nombreux élèves ont
well on (a). Part (b) was a fair bit more difficult. Many compris que r était impair et que de p et q l’un devait
students recognized that r had to be odd and then that one être pair et l’autre impair sans toutefois aller plus loin
of p and q is even and the other odd, but were then stuck. A dans la résolution du problème. Quelques élèves
few students ingeniously pointed out that not only can this ingénieux ont indiqué que l’équation cubique ne pouvait
cubic equation not have 3 integer roots, it cannot even have avoir trois racines entières ni même une comme le
1 integer root, as they showed with the following proof: démontre le raisonnement suivant :
Suppose that q and r are both odd and p even.
Supposons que q et r sont tous deux pairs et que p est
Let a be an integer root of x 3 + px 2 + qx + r = 0 , ie. impair.
a 3 + pa 2 + qa + r = 0 . Soit a est une racine entière de l’équation

If a is even, then a 3 , pa 2 , and qa are all even, and r is odd, x 3 + px 2 + qx + r = 0 , c’est-à-dire

so a 3 + pa 2 + qa + r is odd and so cannot be 0. a 3 + pa 2 + qa + r = 0 .

If a is odd, then a 3 , qa, and r are all odd, and is even (since Si a est impair, alors a 3 , pa 2 et qa sont tous pairs, et

p is even), so a 3 + pa 2 + qa + r is odd and so cannot be 0. r est impair, a 3 + pa 2 + qa + r a pour solution un

Therefore, there cannot be any integer roots. nombre impair autre que 0.
(A similar argument works for p odd and q even.) Si a est impair, alors a 3 , qa et r sont tous impairs et pa 2
Congratulations to those students who had this brilliant est impair (puisque p est impair, alors a 3 + pa 2 + qa + r
insight! a pour solution un nombre pair autre que 0).
Average: 4.0 14
Comments on the Paper Commentaires sur l'épreuve

4. Solution 1 (Trigonometry) Par conséquent, il n’existe pas de racine entière.

A
Let ∠BCP = ∠ABP = α 90 – θ (Le raisonnement semblable voulant que p soit impair et q
and ∠ACP = θ . pair est également valable.)
x α D Félicitations aux élèves qui sont arrivés à cette brillante
Then ∠PBC = θ since 90 + θ + α 2θ
+
déduction!
∆ABC is isosceles. Also, Moyenne: 4,0
α P θ
∠PAC = 90o − θ (from θ α
B C
∆APC), ∠ADP = 2θ + α 4. Solution 1 (par trigonométrie et géométrie analytique)
A
(exterior angle), and ∠APB = 90 + θ + α o Soit ∠BCP = ∠ABP = α et ∠ACP = θ . 90 – θ

(exterior angle). Donc ∠PBC = θ , puisque le triangle

x D
Let AP = x . Then from ∆ APC , ABC est isocèle. 90 + θ + α 2θ
+α

x x Dans le triangle APC,

sin θ = = ⇒ x = 5 sin θ (*) ∠PAC = 90o − θ . ∠ ADP = 2θ + α α P θ
AC 5 θ α
(angle extérieur du triangle BCD) B C
By the sine law in ∆ABP , o
x 5 ∠APB = 90 + θ + α (angle extérieur du triangle ADP)
= o
(
sin α sin 90 + θ + α ) Soit AP = x . Dans le triangle APC, on a sin θ =
x
AC 5
x
= ,
5
= d’où x = 5 sin θ (*).
cos(θ + α )
D’après la loi des sinus dans le triangle ABP :
5
= x 5
cos( ∠ABC ) =
5
o
(
sin α sin 90 + θ + α )
= 3 5
5 =
cos(θ + α )
x 25
= (**) 5
sin α 3 =
[Note that cos(∠ABC ) = 3
since drawing a perpen- cos( ∠ABC )
5
dicular from A bisects BC. Also, sin(∠ABC ) = 45 .] 5
= 3
Therefore combining (*) and (**) 5
x 25
= (**)
5 sin θ 25 sin α 3
=
sin α 3
3 sin θ = 5 sin α [On remarque que cos(∠ABC ) = 35 , car la hauteur au point A

3 sin θ = 5 sin(∠ABC − θ ) est aussi la médiatrice de BC. On a aussi sin(∠ABC ) = 45 .]

On utilise (*) et (**) pour obtenir :
3 sin θ = 5 sin(∠ABC ) cosθ − 5 cos(∠ABC ) sin θ
5 sin θ 25
3 sin θ = 4 cosθ − 3 sin θ =
sin α 3
6 sin θ = 4 cosθ 3 sin θ = 5 sin α
tan θ = 2
3 3 sin θ = 5 sin(∠ABC − θ )
3 sin θ = 5 sin(∠ABC ) cosθ − 5 cos(∠ABC ) sin θ
To determine the ratio of AD to DC, we use coordinates.
3 sin θ = 4 cosθ − 3 sin θ
Let B have coordinates (0, 0) and C have coordinates
6 sin θ = 4 cosθ
(6, 0). Thus, A has coordinates (3, 4) , since the altitude
from A to BC has length 4. tan θ = 2
3

Since tan θ = 23 , the line from B to D has equation Pour déterminer le rapport de AD à DC, on utilise un repère
y = 23 x . Also, the line from A to C has equation cartésien.
Soit (0, 0) les coordonnées du point B et (6, 0) les coordonnées
y = − 43 ( x − 6) . To find D, we find the intersection of
du point C. Les coordonnées du point A sont (3, 4), puisque
these two lines:
la hauteur de A à BC a une longueur de 4. Puisque tan θ = 23 ,
15
Comments on the Paper Commentaires sur l'épreuve

la droite qui passe par B et D a pour équation y = 23 x . De

2
3 x = − 43 x + 8
plus, la droite qui passe par A et C a pour équation
2x = 8 y = − 43 ( x − 6) . Le point D est le point d’intersection de ces
x=4 deux droites. Donc :
Therefore, D has coordinates ( 4, 83 ) . Since the x-coordi- 2
3 x = − 43 x + 8
nate of D is 13 of the way between those of A and C, then 2x = 8
AD : DC = 1 : 2 . x=4
Solution 2 (Similar triangles) Les coordonnées de D sont donc ( 4, 83 ) .
Draw a perpendicular from A to A On compare les déplacements horizontaux de A à D et de
DàC: 4–3 1
meet BC at M. Then since α D =
5 6–4 2
AB = AC , BM = MC = 3 and so
Donc AD : DC = 1 : 2 .
AM = 4 . α P
θ θ
θ α
Let ∠BCP = ∠ABP = α and B 3 M 3 C Solution 2 (par triangles semblables)
A
∠ACP = θ . Then ∠PBC = θ On abaisse une perpendiculaire AM de
since ∆ABC is isosceles. A à BC. Puisque AB = AC , alors α D
5
Draw circle with AC has diameter. This circle passes BM = MC = 3, d’où AM = 4 .
through both P and M, since ∠APC = ∠AMC = 90o . Soit ∠BCP = ∠ABP = α et α P
θ
θ
α
θ

Join P to M. Then ∠PAM = α since ∠PAM = ∠PCM ∠ACP = θ . Donc ∠PBC = θ , B 3 M 3 C

puisque le triangle ABC est isocèle.
(subtended by the same chord). Also, ∠AMP = θ for
On trace un cercle ayant AC pour diamètre. Ce cercle
similar reasons. Therefore, ∆MPA is similar to ∆BPC.
passe aux points P et M, puisque ∠APC = ∠AMC = 90o .
Thus,
On joint P et M. Donc ∠PAM = α , puisque
PA MA 4 PA 2
= = ⇒ tan θ = = ∠PAM = ∠PCM (ces angles interceptent le même arc).
PC BC 6 PC 3 De même, ∠AMP = θ . Les triangles MPA et BPC sont
So now we must compute the length of DC. Consider donc semblables.
∆BDC . By the sine law, PA MA 4 PA 2
Donc = = , d’où tan θ = = .
PC BC 6 PC 3
On cherche maintenant à déterminer DC. Selon la loi des
DC BC sinus dans le triangle BDC :
=
sin θ sin(∠BDC ) DC BC
=
6 sin θ sin θ sin(∠BDC )
DC = o
(
sin 180 − θ − ∠DCB ) DC =
6 sin θ

=
6 sin θ ( o
sin 180 − θ − ∠DCB )
sin(θ + ∠DCB) 6 sin θ
=
6 sin θ sin(θ + ∠DCB)
=
sin θ cos(∠DCB) + cosθ sin(∠DCB) 6 sin θ
=
=
6 sin θ cos(∠DCB) + cosθ sin(∠DCB)
cos(∠DCB) + cot θ sin(∠DCB) 6
=
=
6 cos(∠DCB) + cot θ sin(∠DCB)
3
5 + 3 4
2 5() 6
=
=
10 3
5 + 3 4
2 5()
3 10
=
3
which yields also that AD = 5 − 103 = 5
and so
3 Donc AD = 5 − 103 = 5
3 , d’où AD : DC = 1 : 2 .
AD : DC = 1 : 2 .
16
Comments on the Paper Commentaires sur l'épreuve

Solution 3 (Geometry) Solution 3 (par géométrie)

Draw a line from A perpen- A K On abaisse une perpendiculaire AM de A A K
dicular to BC at M. D à BC. Soit O le milieu de AC. On trace un
D
Let O be the midpoint of AC.
X O cercle de centre O et de rayon OC. Ce X O
cercle passe au point A, puisque
Draw circle with centre O P P
AO = OC , et aux points P et M,
and radius OC. B M C o
puisque ∠APC = ∠AMC = 90 . B M C
Then this circle passes through A (since AO = OC ), P
On joint M et O et on prolonge le segment obtenu jusqu’au
and M (since ∠APC = ∠AMC = 90o ). cercle au point K.
Join M to O and extend this line segment to meet the Puisque CO = 12 CA et CM = 12 CB , alors MK est parallèle à
circle at K. BA. On prolonge BD jusqu’à MK au point K ′ .
Since CO = 12 CA and CM = 12 CB , then MK is parallel Donc ∠MK ′B = ∠ABK ′ à cause des segments parallèles.
to BA. Or ∠ ABK ′ = ∠ ABP = ∠ BCP , d’où ∠MK ′P = ∠MCP .
Extend BD to meet MK at K ′ . Donc K ′ est situé sur le cercle, ce qui indique que K ′ et K
Then ∠MK ′B = ∠ABK ′ because of parallel lines. coïncident.
On joint A et K, ainsi que K et C.
But ∠ABK ′ = ∠ABP = ∠BCP ⇒ ∠MK ′P = ∠MCP.
Puisque KM et AC sont des diamètres du cercle, le quadrilatère
Therefore, K ′ lies on the circle; that is, K ′ coincides with AKCM est un rectangle.
K. Donc AK = MC = BM .
Next, join A to K and K to C. Donc AKMB est un parallélogramme. Donc AM et BK se
Then AKCM is a rectangle, since KM and AC are diameters coupent en leur milieu au point X.
On considère maintenant le triangle AKM. KX et AO sont des
of the circle, so the quadrilateral has four right angles.
médianes de ce triangle. Elles se coupent donc dans le rapport
Therefore, AK = MC = BM .
2:1. Donc AD : DO = 2 : 1. Puisque AO : OC = 1 : 1, alors
Then AKMB is a parallelogram, which implies that AM and
AD : DC = 1 : 2 .
BK bisect each other (meeting at X).
Commentaires
Consider now ∆AKM . Then KX and AO are medians, and
Nous nous réjouissons des tentatives de bon nombre d’élèves
so intersect in the ratio 2 : 1 , ie. AD : DO = 2 : 1. Since à résoudre cette question d’un haut niveau de difficulté. Dans
AO : OC = 1 : 1, then AD : DC = 1 : 2 . un premier temps, il fallait déterminer la valeur de quelques
angles du diagramme, une étape somme toute simple. Dans
Comments un deuxième temps, il fallait discerner la présence d’un
This question was extremely difficult, but it was gratifying cercle, puis trouver des triangles semblables par une approche
déductive complexe. Parmi les trois solutions données par les
to see many students at least making an effort to start the
élèves, la seconde comportait sans doute les éléments les plus
question. The first step was to determine some of the intéressants tandis que la première nécessitait un raisonnement
angles in the diagram, which was relatively straightfor- moins complexe mais davantage de calculs difficiles.
ward. The second step of recognizing the presence of a Moyenne: 0,3
circle and then finding similar triangles required a great
deal of insight. Of the three solutions presented, the
second is probably the nicest. The first solution requires
less insight, but is more difficult computationally.
Average: 0.3

17
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

The
Canadian Open
Mathematics Challenge
Wednesday, November 27, 2002

1 © 2002 Canadian Mathematical Society

Time: 2 hours
2

Calculators are NOT permitted.

Do not open this booklet until instructed to do so.

There are two parts to the paper.

PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value for each
question by entering the correct answer in the space provided. Any work you do in obtaining an answer
will be considered for part marks if you do not have the correct answer, provided that it is done in the
space allocated to that question in your answer booklet.

PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must be written
in the appropriate location in the answer booklet. Rough work should be done separately. If you require
extra pages for your finished solutions, paper will be provided by your supervising teacher. Any extra
papers should be placed inside your answer booklet.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution poorly
presented will not earn full marks.

NOTE: At the completion of the contest, insert the information sheet inside the answer booklet.
 Canadian Open Mathematics Challenge

NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations and answers will be expressed as exact numbers such
as 4 π, 2 + 7 , etc.
4. Calculators are not allowed.

PART A
1. In triangle PQR, F is the point on QR so that PF is perpendicular P
to QR. If PR = 13, RF = 5 , and FQ = 9 , what is the perimeter
of ∆PQR ?
13

Q 9 F 5 R

2. If x + y = 4 and xy = −12 , what is the value of x 2 + 5 xy + y 2 ?

3. A regular pentagon is a five-sided figure which has all of its A

angles equal and all of its side lengths equal. In the diagram, R
TREND is a regular pentagon, PEA is an equilateral triangle, and
P
OPEN is a square. Determine the size of ∠ EAR . E
T

O N
D

4. In a sequence of numbers, the sum of the first n terms is equal to 5 n 2 + 6 n . What is the sum of the
3rd, 4th and 5th terms in the original sequence?

5. If m and n are non-negative integers with m < n , we define m∇n to be the sum of the integers from
m to n, including m and n. For example, 5∇8 = 5 + 6 + 7 + 8 = 26 .

For every positive integer a, the numerical value of

[(2a − 1)∇(2a + 1)] is the same. Determine this
[(a − 1)∇(a + 1)]
value.

6. Two mirrors meet at an angle of 30o at the point V. A beam of

A
light, from a source S, travels parallel to one mirror and strikes S
the other mirror at point A, as shown. After a number of
reflections, the beam comes back to S. If SA and AV are both 30°
1 metre in length, determine the total distance travelled by the V
beam.

7. N is a five-digit positive integer. A six-digit integer P is constructed by placing a 1 at the right-hand

end of N. A second six-digit integer Q is constructed by placing a 1 at the left-hand end of N. If P is
three times Q, determine the value of N.
8. Suppose that M is an integer with the property that if x is randomly chosen from the set
{1, 2, 3,K, 999,1000} , the probability that x is a divisor of M is 100
1
. If M ≤ 1000 , determine the
maximum possible value of M.

PART B

1. Square ABCD has vertices A(0, 0) , B(0, 8) , C (8, 8) , and D(8, 0) . The points P (0, 5) and Q(0, 3) are on
side AB , and the point F (8,1) is on side CD.
(a) What is the equation of the line through Q parallel to the line through P and F?
(b) If the line from part (a) intersects AD at the point G, what is the equation of the line through F
and G?
(c) The centre of the square is the point H ( 4, 4 ) . Determine the equation of the line through H
perpendicular to FG.
(d) A circle is drawn with centre H that is tangent to the four sides of the square. Does this circle
intersect the line through F and G? Justify your answer. (A sketch is not sufficient justification.)

2. (a) Let A and B be digits (that is, A and B are integers between 0 and 9 inclusive). If the product of
the three-digit integers 2 A5 and 13B is divisible by 36, determine with justification the four
possible ordered pairs ( A, B) .
(b) An integer n is said to be a multiple of 7 if n = 7 k for some integer k.
(i) If a and b are integers and 10 a + b = 7 m for some integer m, prove that a − 2b is a multiple
of 7.
(ii) If c and d are integers and 5c + 4 d is a multiple of 7, prove that 4c − d is also a multiple
of 7.

3. There are some marbles in a bowl. Alphonse, Beryl and Colleen each take turns removing one or two
marbles from the bowl, with Alphonse going first, then Beryl, then Colleen, then Alphonse again, and
so on. The player who takes the last marble from the bowl is the loser, and the other two players are
the winners.
(a) If the game starts with 5 marbles in the bowl, can Beryl and Colleen work together and force
Alphonse to lose?
(b) The game is played again, this time starting with N marbles in the bowl. For what values of N
can Beryl and Colleen work together and force Alphonse to lose?

4. Triangle DEF is acute. Circle C1 is drawn with DF as its D

diameter and circle C2 is drawn with DE as its diameter.
Points Y and Z are on DF and DE respectively so that EY and C1
FZ are altitudes of ∆DEF . EY intersects C1 at P, and FZ C2
intersects C2 at Q. EY extended intersects C1 at R, and FZ
extended intersects C2 at S. Prove that P, Q, R, and S are
E F
concyclic points.
The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

The
Canadian Open
Mathematics Challenge
Wednesday, November 27, 2002

Solutions

© 2002 Canadian Mathematical Society

2002 COMC Solutions 2

Part A

1. By Pythagoras in ∆PFR , PF 2 = 132 − 5 2 = 144 , or P

PF = 12 .
By Pythagoras in ∆PFQ , PQ2 = 9 2 + 12 2 = 225 , or
13
PQ = 15 . 12
Therefore, the side lengths of ∆PQR are 13, 14 and 15,
i.e. the perimeter is 42.
Q 9 F 5 R

2. Solution 1
x 2 + 5 xy + y 2 = x 2 + 2 xy + y 2 + 3 xy
= ( x + y ) + 3 xy
2

= 4 2 + 3(−12)
= −20

Solution 2
Examining the two given equations, we see that x = 6 and y = −2 is a solution.
Therefore, x 2 + 5 xy + y 2 = 6 2 + 5(6)(−2) + (−2) = 36 − 60 + 4 = −20 .
2

Solution 3
We solve the first equation for x and substitute into the second equation.
From the first equation, x = 4 − y .
Substituting into the second equation, ( 4 − y ) y = −12 or 0 = y 2 − 4 y − 12 .
Factoring, 0 = ( y − 6)( y + 2) , i.e. y = 6 or y = −2 . The corresponding values of x are
x = −2 and x = 6 , which give the same answer as in Solution 2, i.e. x 2 + 5 xy + y 2 = −20 .

3. To determine ∠EAR , we look at the angles around the A

point E. R
We know that ∠AER + ∠REN + ∠NEP + ∠PEA = 360o . P 60° 102°
90° 108°
Since ∠PEA is an angle in an equilateral triangle, E T
∠PEA = 60o .
Since ∠NEP is an angle in a square, ∠NEP = 90o . O N
D
Since ∠REN is an angle in a regular pentagon, ∠REN = 1
5 (540 ) = 108 .
o o
2002 COMC Solutions 3

Therefore,
∠AER = 360o − ∠REN − ∠NEP − ∠PEA
= 360o − 108o − 90o − 60o
= 102o
Now since PEA is an equilateral triangle, OPEN is a square, and TREND is a regular
pentagon, then their side lengths must all be the same, since OPEN and TREND share a
side, and since OPEN and PEA share a side. In particular, AE = ER .
Therefore, ∆ARE is an isosceles triangle, and so
( ) ( )
∠ARE = 12 180o − ∠AER = 12 180o − 102o = 39o .

4. Solution 1
The sum of the 3rd, 4th and 5th terms of the sequence is equal to the sum of the first five
terms of the sequence minus the sum of the first two terms of the sequence.
[ ][ ]
Thus, the sum is 5(5) + 6(5) − 5(2) + 6(2) = 155 − 32 = 123.
2 2

Solution 2
We determine the first 5 terms in the sequence and then add up the 3rd, 4th and 5th terms.
From the formula given, the sum of the first 1 terms is 11.
This tells us that the first term is 11.
From the formula given, the sum of the first 2 terms is 32. Since the first term is 11, then
the second term is 21.
Next, the sum of the first 3 terms is 63, and so the third term is 31, since the first two terms
are 11 and 21. (We could use the fact that the sum of the first two terms is 32, instead.)
Next, the sum of the first 4 terms is 104, and so the fourth term is 41.
Lastly, the sum of the first 5 terms is 155, and so the fifth term is 51.
Therefore, the sum of the 3rd, 4th and 5th terms is 31 + 41 + 51 = 123 .

Solution 3
Since the sum of the first n terms has a quadratic formula, then the terms in the sequence
have a common difference, i.e. The sequence is an arithmetic sequence.
Therefore, the sum of the 3rd, 4th and 5th terms is equal to three times the 4th term.
The 4th term is the sum of the first four terms minus the sum of the first three terms, i.e.
104 − 63 = 41.
Thus, the sum of the 3rd, 4th and 5th terms is 3( 41) = 123.
2002 COMC Solutions 4

5. Solution 1
Since the value of this expression is the same for every positive integer a, then we can find
the value by substituting in a = 1.
Thus,
[(2a − 1)∇(2a + 1)] = [1∇3] = 1 + 2 + 3 = 6 = 2
[(a − 1)∇(a + 1)] [0∇2] 0 + 1 + 2 3
Therefore, the value required is 2.

Solution 2
If a is a positive integer, the only integer between 2 a − 1 and 2 a + 1 is 2a . Similarly, the
only integer between a − 1 and a + 1 is a.
Thus,
[(2a − 1)∇(2a + 1)] = (2a − 1) + 2a + (2a + 1) = 6a = 2
[(a − 1)∇(a + 1)] (a − 1) + a + (a + 1) 3a
Therefore, the value required is 2.

6. Label the two ends of the mirrors U and W, as shown.

Since the initial beam is parallel to the mirror WV, then ∠UAS = 30o . Since the angle of
incidence equals the angle of reflection, then the reflected beam of light also makes an
angle of 30o with the mirror UV.

Let B be the point on the mirror WV that the light next strikes.
Since ∠VAB = ∠AVB = 30o , then the angle of incidence, ∠ABW , is equal to 60o , because
it is an external angle of ∆ABV . ( ∠ABW could also have been calculated using the facts
that ∠SAB = 120o and SA is parallel to WV.)
Therefore, the angle of reflection is also 60o .
Let C be the point on the mirror UV where the light next U
o o
strikes. Since ∠CVB = 30 and ∠VBC = 60 , then S 30° A 1
∠BCV = 90o . This tells us that the light is reflected 1 2
30° C 1
straight back along its path from C back to S. 2
Therefore, the required distance is 2( SA + AB + BC ) . W
60° 60° 30°
B V
Since we are given that SA = AV = 1, then since ∆ABV is isosceles with BC an altitude,
then AC = CV = 12 , and so BC = 1 ( AC ) = 1 and AB = 2 ( AC ) = 1 .
3 2 3 3 3
Therefore, the required distance is
(
2( SA + AB + BC ) = 2 1 + 1
3
+ 1
2 3 )=2+ 3
3
=2+ 3

( )
Thus, the total distance travelled by the beam is 2 + 3 m, or about 3.73 m.
2002 COMC Solutions 5

7. Solution 1
Since P is formed by adding a 1 at the end of N, then P = 10 N + 1.
Since Q is formed by adding a 1 in front of the 5 digits of N, then Q = 100000 + N .
Since P = 3Q ,
10 N + 1 = 3(100000 + N )
10 N + 1 = 300000 + 3N
7 N = 299999
N = 42857
Therefore, N is 42857.

Solution 2
Suppose N has digits abcde. Then since P = 3Q , we have abcde1 = 3(1abcde) .
Since the units digit on the left side is 1, then the units digit on the right is also 1, which
means that e = 7 .
Thus, abcd 71 = 3(1abcd 7) . Since the tens digit on the left side is 7 and we get a “carry” of
2 from multiplying the last digit on the right side by 3, then 3 × d has a units digit of 5, i.e.
d = 5.
Thus, abc 571 = 3(1abc 57) . Since the hundreds digit on the left side is 5 and we get a carry
of 1 from multiplying the last two digits on the right side by 3, then the units digit of 3 × c
must be a 4, i.e. c = 8 .
Thus, ab8571 = 3(1ab857) . In a similar fashion, we see that b = 2 and a = 4 .
Therefore, N = 42857 .

8. We are not told that M must be a positive integer, but it makes sense to look for a positive
integer M that satisfies these conditions, since we want the maximum possible value of M.
Since there are 1000 numbers in the set {1, 2, 3,K, 999,1000} and the probability that an x
1
chosen randomly from this set is a divisor of M is 100 , then M must have 10 divisors
between 1 and 1000.
Since we are told that M ≤ 1000 , then M must have exactly 10 positive divisors.
Therefore, M must be of the form p 9 where p is a prime number, or p 4 q where p and q are
both primes.
(Recall that to find the number of positive divisors of M, we find the prime factorization of
M and then take each of the exponents, add 1, and find the product of these numbers. For
example, if M = 48 = 2 4 3, then the number of positive divisors is ( 4 + 1)(1 + 1) = 10 .)
Now, we want to determine the maximum M in each of these two forms.
Case 1 M = p 9
Since M ≤ 1000 , then we must have p = 2 , i.e. M = 512 .
(If p = 3 , then p 9 = 19683 is too large.)
2002 COMC Solutions 6

Case 2 M = p 4 q
Since M ≤ 1000 and 5 4 = 625 , then we must have p = 2 or p = 3 .
If p = 2 , then the largest q can be so that q is prime and M ≤ 1000 is 61, i.e.
M = (16)(61) = 976 .
If p = 3 , then the largest q can be so that q is prime and M ≤ 1000 is 11, i.e.
M = (81)(11) = 891.
Therefore, the maximum possible value of M is 976.

Part B

1. (a) The slope of the line through P and F is

5 −1 1
=−
0−8 2
and so the slope of the desired line is also − 12 .
Since the point Q lies on the y-axis, then the y-intercept of the line is 3.
Therefore, the line is y = − 12 x + 3 .

(b) Since AD lies along the x-axis, then G is the y

point where the line from (a) crosses the x-axis. C(8, 8)
To find the coordinates of G, we set y = 0 in B(0, 8)

the line from (a) to get 0 = − 12 x + 3 or x = 6 .

P(0, 5)
Therefore, the desired line passes through the
points G(6, 0) and F (8,1) . Thus its slope is Q(0, 3)
1− 0 1
= , and so its equation is F(8, 1)
8−6 2 x
y − 0 = 12 ( x − 6) or y = 12 x − 3. A(0, 0) G(6, 0) D(8, 0)

(c) Since FG has slope 1

2
, then a line perpendicular to FG has slope −2 , the negative
1
reciprocal of 2
.
Since the desired line passes through H ( 4, 4 ) , it has equation y − 4 = −2( x − 4 ) or
y = −2 x + 12 .
 2002 COMC Solutions 7

(d) The circle has centre H ( 4, 4 ) , and it is tangent to y

all four sides of the square, and so its radius must C(8, 8)
B(0, 8)
be 4, since the distance from the centre to each of
the four sides is 4.
P(0, 5)
Does this circle intersect the line y = 12 x − 3, i.e. H(4, 4)
the line through F and G? Q(0, 3)

F(8, 1)
x
A(0, 0) G(6, 0) D(8, 0)

We must find the shortest distance between the centre of the circle and the line, i.e. the
perpendicular distance. We already have the equation of a line through H that is
perpendicular to the line through F and G, the line y = −2 x + 12 . Where do these lines
intersect? Setting y-coordinates equal,
1
2
x − 3 = −2 x + 12
5
2
x = 15
x=6
i.e. the lines intersect at the point G(6, 0) ! Therefore, the shortest distance from H
to the line through F and G is the distance from H to G, which is
(6 − 4) 2 + (4 − 0) 2 = 20
i.e. is greater than than 4 = 16 , the radius of the circle.
Therefore, the circle does not intersect the line.

2. (a) For the product (2 A5)(13B) to be divisible by 36, we need it to be divisible by both 4
and 9. Since 2A5 is odd, it does not contain a factor of 2.
Therefore, 13B must be divisible by 4.
For a positive integer to be divisible by 4, the number formed by its last two digits must
be divisible by 4, i.e. 3B is divisible by 4, i.e. B = 2 or B = 6 .

Case 1 B = 2
In this case, 132 is divisible by 3, but not by 9. Therefore, for the original product to be
divisible by 9, we need 2A5 to be divisible by 3.
For a positive integer to be divisible by 3, the sum of its digits is divisible by 3, i.e.
2 + A + 5 = A + 7 is divisible by 3.
Therefore, A = 2 or 5 or 8 .

Case 2 B = 6
In this case, 136 contains no factors of 3, so for the original product to be divisible by 9,
we need 2A5 to be divisible by 9.
2002 COMC Solutions 8

For a positive integer to be divisible by 9, the sum of its digits is divisible by 9, i.e.
2 + A + 5 = A + 7 is divisible by 9. Therefore, A = 2 .

Therefore, the four possible ordered pairs are ( A, B) = (2, 2), (8, 2), (5, 2), (2, 6) .

(b) (i) If 10 a + b = 7 m , then b = 7 m − 10 a . Thus,

a − 2b = a − 2( 7 m − 10 a) = 21a − 14 m = 7( 3a − 2 m)
Since 3a − 2 m is an integer, then by definition, a − 2b is divisible by 7.

(ii) Solution 1
If 5c + 4 d is divisible by 7, then 5c + 4 d = 7 k for some integer k.
Therefore, d = 41 ( 7 k − 5c ) .
7( 3c − k )
So 4 c − d = 4 c − 41 ( 7 k − 5c ) = 41 (21c − 7 k ) = .
4
Since 4c − d is an integer, then 7( 3c − k ) must be divisible by 4. But 4 has no
3c − k
common factors with 7, so 4 must divide into 3c − k , i.e. is an integer.
4
 3c − k 
Therefore, 4 c − d = 7  , i.e. 4c − d is divisible by 7.
 4 

Solution 2
We note that 4 c − d = (14 c + 7 d ) − 2(5c + 4 d ) .
Since both terms on the right side are divisible by 7, then 4c − d is divisible by 7.

Solution 3
Multiplying the expression 4c − d by 5 does not affect its divisibility by 7.
Thus, we can consider whether or not 20c − 5 d is divisible by 7, and this will be
equivalent to considering 4c − d .
Since we are told that 5c + 4 d = 7 t for some integer t, then we know that
4 (5c + 4 d ) = 20c + 16 d = 28 t or 20c = 28 t − 16 d .
If we now consider 20c − 5 d , we see
20c − 5 d = (28 t − 16 d ) − 5 d
= 28 t − 21d
= 7( 4 t − 3d )
Since 20c − 5 d is divisible by 7 by definition, then 4c − d is divisible by 7.
2002 COMC Solutions 9

3. (a) We consider the possible cases. On his first turn, Alphonse can take either 1 marble or
2 marbles.
If Alphonse takes 1 marble, Beryl can take 2 marbles and then Colleen 1 marble, to
leave Alphonse with 1 marble left in the bowl. Therefore, Alphonse loses. (Note that
Beryl and Colleen can agree on their strategy before the game starts.)
If Alphonse takes 2 marbles, Beryl can take 1 marble and then Colleen 1 marble, to
leave Alphonse again with 1 marble left in the bowl. Therefore, Alphonse loses.
In either case, Beryl and Colleen can work together and force Alphonse to lose.

(b) Solution 1
On their two consecutive turns, Beryl and Colleen remove in total 2, 3 or 4 marbles.
On his turn, Alphonse removes either 1 marble or 2 marbles. Therefore, by working
together, Beryl and Colleen can ensure that the total number of marbles removed on
any three consecutive turns beginning with Alphonse’s turn is 4 or 5. (Totals of 3 and 6
cannot be guaranteed because of Alphonse’s choice.)

Therefore, if N is a number of marbles in which Alphonse can be forced to lose, then so

are N + 4 and N + 5 , because Beryl and Colleen can force Alphonse to choose from N
marbles on his second turn.

From (a), we know that 5 is a losing position for Alphonse. Also, 1 is a losing position
for Alphonse. (Since 1 is a losing position, then 5 and 6 are both losing positions,
based on our earlier comment.)

Since 5 and 6 are losing positions, then we can determine that 9, 10 and 11 are also
losing positions, as are 13, 14, 15, and 16. If we add 4 to each of these repeatedly, we
see that N is a losing position for every N ≥ 13 .

What about the remaining possibilities, i.e. 2, 3, 4, 7, 8, and 12?

For N = 2 or N = 3, if Alphonse chooses 1 marble, then either Beryl or Colleen is
forced to take the last marble, so these are not losing positions for Alphonse, i.e. they
are winning positions.
For N = 4 , if Alphonse chooses 2 marbles, then either Beryl or Colleen is forced to
take the last marble, so this is also not a losing position for Alphonse.

Next, we notice that if Alphonse chooses 1 marble, then the total number of marbles
chosen by the three players will be 3, 4 or 5, and if Alphonse chooses 2 marbles, then
the total number chosen will be 4, 5 or 6.
So if N = 7 , then Alphonse can choose 1 marble and ensure that he receives 2, 3 or 4
marbles on his next turn. So 7 is a winning position for Alphonse.
2002 COMC Solutions 10

If N = 8 , then Alphonse can choose 2 marbles and ensure that he receives 2, 3 or 4

marbles on his next turn. So 8 is also a winning position for Alphonse.

Lastly, we consider N = 12 .
If Alphonse chooses 1 marble, Beryl and Colleen can choose 1 each and return 9
marbles to Alphonse. As we have shown, this is a losing position for Alphonse.
If Alphonse chooses 2 marbles, Beryl and Colleen can choose 2 each and return 6
marbles to Alphonse. This is a losing position for Alphonse.

Therefore, the values of N for which Beryl and Colleen can force Alphonse to lose are
1, 5, 6, and all N for which N ≥ 9 .

Solution 2
First, we notice that if Alphonse chooses 1 marble, then the total number of marbles
chosen by the three players will be 3, 4 or 5, and if Alphonse chooses 2 marbles, then
the total number chosen will be 4, 5 or 6.

We define a “losing position” to be a number of marbles in the bowl so that if Alphonse

starts with this number, he can be forced to lose.

From (a), we know that 5 is a losing position for Alphonse. Also, 1 is a losing position
for Alphonse.
For N = 2 or N = 3, if Alphonse chooses 1 marble, then either Beryl or Colleen is
forced to take the last marble, so these are not losing positions (ie. they are winning
positions) for Alphonse.
For N = 4 , if Alphonse chooses 2 marbles, then either Beryl or Colleen is forced to
take the last marble, so this is a winning position for Alphonse.

How can we ensure that a starting position N ≥ 6 is not a losing position?

N will not be a losing position if either none of N − 3, N − 4 or N − 5 are losing
positions, or none of N − 4 , N − 5 or N − 6 are losing positions. (If either group
consists of three non-losing positions, then Alphonse can ensure that he gets a position
from the appropriate set at the beginning of his next turn by choosing either 1 or 2
marbles respectively.)

Also, N will be a losing position as long as at least one of N − 3, N − 4 , N − 5 and at

least one of N − 4 , N − 5 , N − 6 are losing positions. (If there is a losing position in
each group of 3, then no matter whether Alphonse chooses 1 or 2 marbles, then Beryl
and Colleen will be able to force Alphonse into one of these previously known losing
positions.)

Using these two criteria for checking whether a position is a losing position or not a
losing position, we can see
2002 COMC Solutions 11

i) N = 6 is a losing position, since N − 5 = 1 is a losing position in both groups of

three
ii) N = 7 is not a losing position, since N − 3, N − 4 , N − 5 (namely, 4, 3, 2) are not
losing positions.
iii) N = 8 is not a losing position, since N − 4 , N − 5 , N − 6 (namely, 4, 3, 2) are not
losing positions.
iv) N = 9 is a losing position, since N − 4 = 5 is a losing position in both groups of
three
v) N = 10 is a losing position, since N − 5 = 5 is a losing position in both groups of
three
vi) N = 11 is a losing position, since N − 5 = 6 is a losing position in both groups of
three
vii) N = 12 is a losing position, since N − 3 = 9 and N − 6 = 6 are both losing
positions

And so we have obtained 4 consecutive losing positions, which guarantees us that any
N ≥ 13 will also be a losing position, since N − 4 will be a losing position in both
groups of 3.

Therefore, the values of N for which Beryl and Colleen can force Alphonse to lose are
1, 5, 6, and all N for which N ≥ 9 .

But among the first eight possibilities, there are now no more sets of three consecutive
non-losing positions. This tells us that every position for N ≥ 9 is a losing position,
since we cannot find three consecutive non-losing positions as described above.

Therefore, the values of N for which Beryl and Colleen can force Alphonse to lose are
1, 5, 6, and all N for which N ≥ 9 .

4. Solution 1
Join E to P, Y and R, and join F to Q, Z and S. D
Let O be the point of intersection of EY and FZ. R
Since EY and FZ are altitudes in ∆DEF , then the S
third altitude, DX say, passes through O. Y
Z O
If we look at altitude DX, we see that ∠DXE = 90o .
Since circle C2 has DE as its diameter, then point X P Q
must lie on circle C2 , since a right angle is subtended E X F
by the diameter at point X.

Similarly, point X lies on circle C1 .

Therefore, DX is a chord of both circle C1 and circle C2 .
We can now use the “Chord-Chord Theorem” in each of circle C1 and C2 , to say
2002 COMC Solutions 12

SO ⋅ OQ = DO ⋅ OX (from circle C2 )
RO ⋅ OP = DO ⋅ OX (from circle C1 )
From this we can conclude that SO ⋅ OQ = RO ⋅ OP .

Why does this allow us to conclude that P, Q, R, and S lie on the same circle?
SO RO
From the equation, we obtain = , which tells us that ∆SOP is similar to ∆ROQ ,
OP OQ
and so ∠PSQ = ∠PSO = ∠ORQ = ∠PRQ .
Since the chord PQ subtends the equal angles ∠PSQ and ∠PRQ (in an undrawn circle),
then the points P, Q, R, and S are concyclic.

Solution 2
In order to show that the four points lie on a circle, we will show that the points are
equidistant from a fifth point, which will thus be the centre of the circle on which the four
points lie.
Consider first the points Q and S. Any point equidistant from Q and S lies on the
perpendicular bisector of the line joining these points. Since Q and S both lie on circle C2 ,
DE is a diameter of C2 , and QS is perpendicular to DE (since they lie on an altitude of the
triangle), then DE is the perpendicular bisector of QS.
Similarly, DF is the perpendicular bisector of PR.
Therefore, any point that is equidistant from all four of the given points must lie on both
DE and DF. Thus, the only possible candidate is point D. (And we already know that
DS = DQ and DP = DR from our discussion of perpendicular bisectors.)

Thus, if we can show that DS = DR , then we will have shown what we need to show.

Method 1
Let SZ = c , DZ = a and EZ = b . D
Then DS 2 = DZ 2 + SZ 2 = c 2 + a 2 (Pythagoras). R
Now if we extract ∆DSE , we see that ∠DSE = 90o , S a
Z Y
since DE is a diameter of circle C2 . Therefore, c O
DZ SZ b
∆DSZ is similar to ∆SEZ , or = or c 2 = ab .
SZ EZ P Q
Thus, DS 2 = a 2 + ab = a( a + b) = DZ ⋅ DE . E F
Similarly, DR 2 = DY ⋅ DF , looking at ∆DRF .

Now consider the points E, Z, Y, and F. Since ∠EZF = ∠EYF = 90o , then EF must be the
diameter of the circle containing points Y and Z (and points E and F).
Therefore, DE and DF are secants of the circle which intersect the circle at Z and Y,
respectively. By the “Secant-Secant Theorem”, DZ ⋅ DE = DY ⋅ DF .
2002 COMC Solutions 13

From above, we can conclude that DS 2 = DR 2 , or DS = DR , and thus

DP = DQ = DR = DS .

Method 2
As above, we can obtain that DS 2 = a 2 + ab = a( a + b) = DZ ⋅ DE .
Since ∠DZF = 90o , then DZ = DF cos(∠ZDF ) = DF cos(∠EDF ) , and so
DS 2 = DZ ⋅ DE = DE ⋅ DF cos(∠EDF ) .
Repeating the process on the other side of the triangle gives us that
DR 2 = DY ⋅ DF = DF ⋅ DE cos(∠EDF ) , or DR 2 = DS 2 .
Therefore, DP = DQ = DR = DS .
Therefore, we can conclude that the points P, Q, R, and S are concyclic.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

The
Canadian Open
Mathematics Challenge
Wednesday, November 26, 2003

1 © 2003 Canadian Mathematical Society

Time: 2 hours
2

Calculators are NOT permitted.

Do not open this booklet until instructed to do so.

There are two parts to the paper.

PART A
This part of the paper consists of 8 questions, each worth 5. You can earn full value for each question
by entering the correct answer in the space provided. Any work you do in obtaining an answer will be
considered for part marks if you do not have the correct answer, provided that it is done in the space
allocated to that question in your answer booklet.

PART B
This part of the paper consists of 4 questions, each worth 10. Finished solutions must be written in the
appropriate location in the answer booklet. Rough work should be done separately. If you require extra
pages for your finished solutions, paper will be provided by your supervising teacher. Any extra papers
should be placed inside your answer booklet.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution poorly
presented will not earn full marks.

NOTE: At the completion of the contest, insert the information sheet inside the answer booklet.
 Canadian Open Mathematics Challenge

NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations and answers will be expressed as exact numbers such
as 4 π, 2 + 7 , etc.
4. Calculators are not allowed.

PART A
1. Jeff, Gareth and Ina all share the same birthday. Gareth is one year older than Jeff, and Ina is two years
older than Gareth. This year the sum of their ages is 118. How old is Gareth?

2. The point ( 4, − 2 ) is reflected in the x-axis. The resulting

point is then reflected in the line with equation y = x . What y y=x
are the coordinates of the final point?

x
(4, – 2)

3. A circle of radius 1 is centred at the origin. Two particles start y

moving at the same time from the point (1 , 0) and move around
the circle in opposite directions. One of the particles moves
counterclockwise with constant speed v and the other moves
clockwise with constant speed 3v. After leaving (1 , 0) , the two x
(1, 0)
particles meet first at point P, and continue until they meet next
at point Q. Determine the coordinates of the point Q.

4. Two different numbers are chosen at random from the set {0, 1, 2, 3, 4} . What is the probability that
their sum is greater than their product?

5. In the diagram, square ABCD has a side length of 6. Circular A B

arcs of radius 6 are drawn with centres B and D. What is the area
of the shaded region?

D C

6. The symbol a  means the greatest integer less than or equal to a.

For example, 5.7  = 5, 4  = 4 and −4.2  = −5 .
 3 4 
Determine all values of x for which   +   = 5 .
x x
7. Each of the points P ( 4 , 1) , Q( 7 ,−8) and R(10, 1) is the midpoint of a radius of the circle C. Determine
the length of the radius of circle C.

8. Determine the number of triples ( k, l, m) of positive integers such that

k + l + m = 97
4 k 5l 6m
+ + = 82
5 6 7

PART B
1. In the diagram shown, whole numbers are to be k
placed in the ten circles so that the sum of the
numbers in the circles along any of the ten straight
lines is 15. For example, a + g + k = 15 and i
e + i = 15 .
(a) If k = 2 and e = 5 , fill in the whole numbers g h e
that go in all of the circles in the diagram. f
(b) Suppose that k = 2 and the value of e is d
unknown.
(i) Find a formula for each of b and c in terms a b c
of e. A clearly labelled diagram is sufficient
explanation.
(ii) Show that e must be equal to 5.
(c) Suppose now that k = x , where x is unknown. Prove that e must still be equal to 5.

2. A barn has a foundation in the shape of a trapezoid, with three sides D 6 A

of length 6 m, and one side of length 12 m, as shown.
(a) Determine each of the interior angles in the trapezoid. 6 6
(b) Chuck the Llama is attached by a chain to a point on the outside
wall of the barn. Chuck is smarter than the average llama, and so C 12 B
realizes that he can always reach the area between the barn and
where the chain is fully extended.
(i) If Chuck is attached at the point A with a chain of length 8 m, what is the area outside the barn
that Chuck can reach?
(ii) If Chuck is attached at some point P along the wall between A and B with a chain of length
15 m, determine the location of P which restricts Chuck to the minimum area.

3. (a) In the diagram, the two circles C1 and C2 have a P

A
common chord AB. Point P is chosen on C1 so that it is
C1
outside C2 . Lines PA and PB are extended to cut C2 at
X
X and Y, respectively. If AB = 6 , PA = 5 , PB = 7 and
AX = 16, determine the length of XY. B
C2

Y
(b) Two circles C3 and C4 have a common chord GH. Point Q is chosen on C3 so that it is outside
C4 . Lines QG and QH are extended to cut C4 at V and W, respectively. Show that, no matter
where Q is chosen, the length of VW is constant.

over ...
 Canadian Open
Mathematics
Challenge
(English)

2003
4. The polynomial equation x 3 − 6 x 2 + 5 x − 1 = 0 has three real roots a, b and c.
(a) Determine the value of a 5 + b 5 + c 5 .
(b) If a < b < c , show that c 2004 is closer to its nearest integer than c 2003 is to its nearest integer.
The Canadian Mathematical Society
in collaboration with

The CENTRE for Education

in MATHEMATICS and COMPUTING

The
Canadian Open
Mathematics Challenge

Wednesday, November 26, 2003

Solutions

© 2003 Canadian Mathematical Society

2004 Canadian Open Mathematics Challenge Solutions 2

Part A

1. Let Gareth’s present age, in years, be G.

Then Jeff’s age is G − 1, and Ina’s age is G + 2.
Since the sum of their three ages is 118, then
(G − 1) + G + (G + 2) = 118
G = 39
Therefore, Gareth’s age is 39.

2. When the point (4,−2 ) is reflected in the x-axis, its image is (4,2) .
When the point (4,2) is reflected in the line y = x , its image is (2,4) .
Therefore, the coordinates of the final point are (2,4) .

3. The particle which moves clockwise is moving three times as fast as the particle moving
counterclockwise. Therefore, the particle moving clockwise moves three times as far as the
particle moving counterclockwise in the same amount of time.

This tells us that in the time that the clockwise particle y

travels 34 of the way around the circle, the counterclockwise
P(0, 1)
particle will travel 14 of the way around the circle, and so the
two particles will meet at P(0,1) .

x
(1, 0)

Using the same reasoning, the particles will meet at Q(−1, 0) y

when they meet the second time.
P
Q(–1, 0)
x
2004 Canadian Open Mathematics Challenge Solutions 3

4. Solution 1
In choosing a pair of numbers from the five given numbers, there are 10 different ways of choosing
these numbers. These pairs are {(0,1), (0,2),(0,3),(0,4), (1,2), (1,3),(1,4),(2,3), (2,4 ), (3,4 )} . The only
pairs in which the sum is greater than the product are those containing a 0 or a 1. Since there are 7
of these, the required probability is 107 .

Solution 2
The most straightforward way to approach this problem is to make a chart:
Numbers chosen Sum Product
0,1 1 0
0,2 2 0
0,3 3 0
0,4 4 0
1,2 3 2
1,3 4 3
1,4 5 4
2,3 5 6
2,4 6 8
3,4 7 12

So there are 10 possible ways that two different numbers can be chosen, and for 7 of these
possibilities, the sum of the two numbers is greater than the product.
7
Therefore, the probability is 10 .

5. Join A to C.
This line divides the shaded region into two identical pieces. A B

D C
Consider the shaded region above AC.
This piece of the region is formed by taking the sector DAC of A
the circle, centre D and radius 6, and then removing ∆ADC .
6

D 6 C
2004 Canadian Open Mathematics Challenge Solutions 4

o
Since ∠ADC = 90 , then the sector is one quarter of the whole circle, and has area
2
( )
2
4 π r = 4 π 6 = 9π .
1 1

Also, ∆ADC is right-angled with base DC of length 6 and height DA of length 6, and so has area
2 bh = 2 (6)(6) = 18.
1 1

Therefore, the area of the region above the line is 9π − 18 , and so the area of the entire shaded
region is 2(9π − 18) = 18π − 36 square units.

< 0 , so ⎢⎢ ⎥⎥ ≤ < 0. Similarly, ⎢⎢ ⎥⎥ < 0 , so we cannot possibly have

3 3 3 4
6. If x < 0 , then
x ⎣x⎦ x ⎣x⎦
⎢ 3 ⎥ + ⎢ 4 ⎥ = 5 . Therefore, x > 0 .
⎢⎣ x ⎥⎦ ⎢⎣ x ⎥⎦

< , so ⎢⎢ ⎥⎥ ≤ ⎢⎢ ⎥⎥ .
3 4 3 4
When x > 0 , we have
x x ⎣x⎦ ⎣ x⎦
⎢ ⎥
3 ⎢ ⎥
4
Since each of ⎢ ⎥ and ⎢ ⎥ is an integer, then there are three possibilities:
⎣x⎦ ⎣x⎦
i) ⎢⎢ ⎥⎥ = 0 and ⎢⎢ ⎥⎥ = 5
3 4
⎣x⎦ ⎣x⎦
ii) ⎢⎢ ⎥⎥ = 1 and ⎢⎢ ⎥⎥ = 4
3 4
⎣x⎦ ⎣x⎦
iii) ⎢⎢ ⎥⎥ = 2 and ⎢⎢ ⎥⎥ = 3
3 4
⎣x⎦ ⎣x⎦

If ⎢⎢ ⎥⎥ = 0 , then 0 ≤ < 1 or x > 3. If ⎢⎢ ⎥⎥ = 5, then 5 ≤ < 6 or < x ≤ . These intervals do

3 3 4 4 2 4
⎣x⎦ x ⎣x⎦ x 3 5
not overlap, so there are no solutions in this case.
If ⎢⎢ ⎥⎥ = 1, then 1 ≤ < 2 or < x ≤ 3 . If ⎢⎢ ⎥⎥ = 4 , then < x ≤ 1 . These intervals do not
3 3 3 4 4
⎣x⎦ x 2 ⎣x⎦ 5
overlap, so there are no solutions in this case.
If ⎢⎢ ⎥⎥ = 2 , then 2 ≤ < 3 or 1 < x ≤ . If ⎢⎢ ⎥⎥ = 3, then 1 < x ≤ . In this case, the intervals do
3 3 3 4 4
⎣x⎦ x 2 ⎣x⎦ 3
overlap. When we combine these intervals, we see that if 1 < x ≤ , then ⎢⎢ ⎥⎥ + ⎢⎢ ⎥⎥ = 5 .
4 3 4
3 ⎣x⎦ ⎣x⎦
4
Therefore, the range of values is 1 < x ≤ .
3
2004 Canadian Open Mathematics Challenge Solutions 5

7. Solution 1
Let the radius of circle C be r.
1 1
r
Since P, Q and R are given as midpoints of the radius of the 2 2
r

large circle, they themselves lie on a circle with the same centre P(4, 1) R(10, 1)
1 1
as the given circle, but with half its radius. 2
r
2
r
1
r
2
1
r
Q(7, – 8)
2

Method 1 – Perpendicular bisectors

To find the centre of the circle passing through P, Q and R, we must find the intersection of the
perpendicular bisectors of the sides of the triangle formed by the three points.

Consider first side PR. Since PR is a line segment parallel to (7, 1)

P(4, 1) R(10, 1)
the x-axis, its perpendicular bisector has equation x = 7 .

⎛ 11, 7 ⎞
⎝ 2 2⎠
Q(7, – 8)

Consider next side PQ. Since P has coordinates (4,1) and Q has coordinates (7,−8 ) , then PQ
has slope –3 and has midpoint ( 112 ,− 72) . Therefore, the perpendicular bisector of PQ has slope
3 and has equation y + 2 = 3 (x − 2 ) .
1 7 1 11

Therefore, at the intersection of these two perpendicular bisectors, y + 72 = 1

3 (7 − 112) or y = −3 .
Thus the centre of the circle is the point (7,−3) , and since (7,−8 ) lies on the circle, the radius of
the small circle is 5. Therefore, the radius of circle C is 10.

Method 2 – Geometric
As in Method 1, we proceed by trying to find the centre of P(4, 1) (7, 1) R(10, 1)
the circle. Also, we again know that the centre is the 1
r (7, b)
2
intersection of the perpendicular bisectors of the sides of the 1
r
∆PQR . One perpendicular bisector is very easy to find – 2

that of PR, which has equation x = 7 as we found above.

Q(7, – 8)
This tells us that the centre lies on the line x = 7 . Thus, the
centre of the circle can be represented by O( 7,b) .
2004 Canadian Open Mathematics Challenge Solutions 6

Since radii of a circle are equal,

2 2
OP = OQ
(7 − 4) 2 + (b − 1)2 = (7 − 7)2 + (b + 8)2
9 + b 2 − 2b + 1= b2 + 16b + 64
b = −3
32 + (−4 ) = 5 and the radius of the larger circle is 10.
2
The radius of the circle P, Q and R is

8. The first thing that we must notice in this problem is because we are looking for positive integers k,
l and m such that
4k 5l 6m
+ + = 82
5 6 7
then k must be divisible by 5, l must be divisible by 6, and m must be divisible by 7.
So we make the substitution k = 5K , l = 6L and m = 7M , where K, L and M are positive integers.
Therefore, we obtain, by substitution
5K + 6L + 7M = 97
4K + 5L + 6M = 82
Subtracting the second equation from the first, we get
K + L + M = 15
4K + 5L + 6M = 82
Subtracting six times the first equation from the second equation, we get
K + L + M = 15
−2K − L = −8
or
K + L + M = 15
2K + L = 8
Since K, L and M are all positive integers, we obtain from the second equation the following
possibilities for K and L which give us M from the first equation, and thus k, l and m:

K L M k l m
1 6 8 5 36 56
2 4 9 10 24 63
3 2 10 15 12 70

Therefore, there are three triples (k,l,m) of positive integers which are solutions to the system of
equations.
2004 Canadian Open Mathematics Challenge Solutions 7

Part B

1. (a) We will proceed systematically to fill in the circles based on the initial knowledge that
k = 2 and e = 5 :
i = 10 (5, i)
[Throughout the solution to this problem, we have used notation such as (5, 7, c) to indicate
that there is a straight line joining the circles containing 5, 7 and c. Thus, 5 + 7 + c = 15 or
c = 3.]
b=8 (2, 5, b) k 2
d=7 (8, d)
c=3 (5, 7, c)
a= 4 (a, 8, 3) i 10
g= 9 (4, g, 2)
1
f =6 (4, f, 5)
9 g h e 5
h =1 (9, h, 5) f 6
d 7

4 a b c 3
8
We can verify that the numbers along each of the ten straight lines add to 15.

(b) (i) Solution 1

We start with knowing that k = 2 and e is unknown.
Then k 2
i = 15 − e (e, i)
c = 15 − 2 − (15 − e) = e − 2 (2, 15 − e , c)
b = 15 − 2 − e = 13− e (2, e, b) i 15 – e
Therefore, b = 13 − e and c = e − 2 .
g h e
f
d

a b c
13 – e e–2
2004 Canadian Open Mathematics Challenge Solutions 8

Solution 2
We start with knowing that k = 2 and e is k 2
unknown.
Then
i = 15 − e (e, i) i 15 – e
c = 15 − 2 − (15 − e) = e − 2 (2, 15 − e , c)
d = 15 − e − (e − 2) = 17 − 2e (e, d, e − 2 ) g h e
b = 15 − (17 − 2e) = 2e − 2 (b, 17 − 2e ) f 17 – 2e
Therefore, b = 2e − 2 and c = e − 2 . d

a b c
2e – 2 e–2

(ii) Solution
We know from (i) that k = 2 , b = 13 − e , k 2
and c = e − 2 .
Therefore, d = 15 − b = 15 − (13− e ) = 2 + e .
But we also know that i 15 – e
e + d + c = 15
e + (2 + e ) + (e − 2) = 15 g h e
3e = 15 f 2+e
e=5 d
Therefore, e = 5 .
a b c
13 – e e–2

(c) Solution
We can model our approach from (b). k x
Starting with e being unknown and k = x , we
express some of the circles in terms of x and e:
i = 15 − e (e, i) i 15 – e
c = 15 − x − (15 − e) = e − x (x, 15 − e , c)
b = 15 − x − e (x, e, b) g h e
d = 15 − (15 − x − e) = x + e (15 − x − e , d) f x+e
But we know that d
e + d + c = 15
e + ( x + e ) + (e − x ) = 15 a b c
15 – x – e e–x
3e = 15
e=5
Therefore, e must still be equal to 5.
2004 Canadian Open Mathematics Challenge Solutions 9

2. (a) Solution 1
Drop perpendiculars from D and A to BC, meeting BC at E and F respectively.
Since DA is parallel to CB, then DE and AF are also perpendicular to DA.
Since DAFE is a rectangle, then EF = 6 .
Since DC = AB , DE = AF and ∆DEC and ∆AFB are right- D 6 A
angled, then they are congruent triangles, and so CE = BF ,
6 6
and so both of these lengths must be equal to 3.

C 3E 6 F 3 B

2 2 2 2
By the Pythagorean Theorem, DE = DC − CE = 6 − 3 = 27 = 3 3 .
Therefore, the sides of ∆DEC are in the ratio 1 to 3 to 2, so DEC is a 30-60-90 triangle, with
o o
∠DCE = 60 and ∠CDE = 30 .
o
Therefore, since we have congruent triangles, ∠DCB = ∠ABC = ∠DCE = 60 and
o o
∠CDA = ∠DAB = 90 + ∠CDE = 120 .

Solution 2
Join D to the midpoint M of CB.
Then CM = MB = 6 .
Since DM and MB are parallel and of equal length, then AB D 6 A
and DM will also be parallel and equal length.
6 6
Thus, DM = 6 , and so ∆DCM is equilateral.

C 6 M 6 B
o o
Therefore, ∠DCB = ∠DCM = 60 . By symmetry, ∠ABC = ∠DCB = 60 .
o
Since DA and CB are parallel, ∠CDA = ∠DAB = 120 .

(b) (i) If Chuck was attached to a point P and there were no obstructions, he would be able to
reach a circle of radius 8 m. (If Chuck stays at the end of his chain, he could trace out a
circle of radius 8 m, but Chuck can move everywhere inside this circle, since his chain does
not have to be tight.) However, here we have an obstruction – the trapezoidal barn.

o
Since the interior angle of the barn at point A is 120 , then the exterior angle of the barn is
o o
240 . So Chuck can certainly reach the area which is a 240 sector of radius 8 m, centred
at A. (If Chuck extends the chain as far as possible in a straight line in the direction of D
from A, can then walk in a clockwise direction, keeping the chain at its full length until the
o
chain lies along AB. He will have moved through 240 , and the region is the sector of a
circle.)
2004 Canadian Open Mathematics Challenge Solutions 10

However, when the chain is fully extended in the

direction of D, Chuck will be 2 m past point D.
He will thus be free to move towards side DC of
the barn. If he does this and keeps the chain tight,
he will trace out part of a circle of radius 2 m
centred at D. (Point D now serves as a “pivot” 240°
point for the chain.) Since the exterior angle of 2 D 6 A
o 60° 120°
the barn at point D is 240 , then the angle 2
o 6
between AD extended and DC is 60 . Therefore,
o 60° B
Chuck can reach a 60 sector of a circle of radius
C 120° 2
2 m, centred at D. 2

When the chain is fully extended in the direction of B, Chuck will be 2 m past point B. He
will thus be free to move towards side BC of the barn. If he does this and keeps the chain
tight, he will trace out part of a circle of radius 2 m centred at B. (Point B now serves as a
o
“pivot” point for the chain.) Since the exterior angle of the barn at point B is 300 (the
o o
interior angle at B is 60 ), then the angle between AB extended and BC is 120 . Therefore,
o
Chuck can reach a 120 sector of a circle of radius 2 m, centred at B.

o θ
The area of a sector of angle θ of a circle of radius r is πr 2 .
360

Therefore, the total area that Chuck can reach is

240 60 120 2 1 1 128π 134π
(π )82 + (π )2 2 + (π )2 2 = (64π ) + (4π ) + (4π ) = + 2π =
360 360 360 3 6 3 3 3
square metres.

(ii) Let x be the distance along AB from A to P.

Since the total perimeter of the barn is 30 m and Chuck is attached with a 15 m chain, then
Chuck can reach the same point on the barn whether he wraps the chain around the barn in
a clockwise direction or a counterclockwise direction. This point will move, however, as P
moves. For example, if Chuck was attached at point A (ie. if x = 0 ), then he could reach a
point 3 m along CB from C towards B wrapping in either the clockwise or counterclockwise
direction. If Chuck was attached at point B (ie. if x = 6 ), he could reach the midpoint of
CD in either direction. As point P moves from A towards B, this furthest point on the barn
that Chuck can reach will slide along BC towards C and then up CD towards D. If P is at
the midpoint of AB (ie. if x = 3), the furthest point along the barn that he can reach will be
point C. So in our analysis, we must be careful as to whether 0 ≤ x ≤ 3 or 3 ≤ x ≤ 6 .

o
Regardless of the value of x, Chuck can certainly reach a 180 sector of a circle of radius 15
centred at P.
2004 Canadian Open Mathematics Challenge Solutions 11

We start in the counterclockwise

direction.
Also, regardless of the value of x, Chuck
o 15 – x
can reach a 60 sector of a circle of radius
15 − x centred at A (using A as the new
pivot point for the chain). 9–x D 6 60° A
Still regardless of the value of x, Chuck 60° x
o 6 6–x
can reach a 60 sector of a circle of radius
9 − x = (15 − x ) − 6 centred at D (using D C 9+x B
3–x 120° 120°
as the new pivot point for the chain).
9+x
If 3 ≤ x ≤ 6 , then 9 − x ≤ 6 , so Chuck
cannot reach past point C.

o
If 0 ≤ x ≤ 3, then 9 − x ≥ 6 , so Chuck can reach past point C, and so can reach a 120 sector
of a circle of radius 3− x = (9 − x ) − 6 centred at C (using C as the new pivot point for the
chain).
Next, we consider the clockwise direction.
o
Regardless of the value of x, Chuck can reach a 120 sector of a circle of radius
9 + x = 15 − (6 − x ) (the distance from B to P is 6 − x ) centred at B (using B as the new
pivot point for the chain).
If 0 ≤ x ≤ 3, then 9 + x ≤ 12 , so Chuck cannot reach past point C.
o
If 3 ≤ x ≤ 6 , then 9 + x ≥ 12 , so Chuck can reach past point C, and so can reach a 120
sector of a circle of radius x − 3 = (9 + x ) − 12 centred at C (using C as the new pivot point
for the chain).

We now calculate the total area that Chuck can reach.

If 0 ≤ x ≤ 3, then the area that Chuck can reach is
180 60 60 120 120
π15 2 + π (15 − x ) 2 + π (9 − x )2 + π (3 − x )2 + π (9 + x )2
360 360 360 360 360
1
2
1
6
( ) (
1
6
) 1
(
= π (225) + π 225 − 30x + x 2 + π 81− 18x + x 2 + π 9 − 6x + x 2 + π 81+ 18x + x 2
3
1
3
) ( )
1
(
= π 675 + 225 − 30x + x 2 + 81− 18x + x 2 + 18 − 12x + 2x 2 + 162 + 36x + 2x 2
6 )
1
(
= π 1171− 24x + 6x 2
6
)
387
= πx 2 − 4πx + π
2
2004 Canadian Open Mathematics Challenge Solutions 12

If 3 ≤ x ≤ 6 , then the area that Chuck can reach is

180 60 60 120 120

π15 2 + π (15 − x ) 2 + π (9 − x )2 + π (9 + x ) 2 + π ( x − 3)2
360 360 360 360 360
387
= πx 2 − 4πx + π
2

(Notice that the one term between these two initial expressions that seems to be different is
actually the same!)

Therefore, no matter what the value of x is, the area that Chuck can reach is
387
πx 2 − 4πx + π . This is a parabola opening upwards, so the vertex of the parabola gives
2
−4π
us the minimum of the parabola. This vertex is at x = − = 2 . Since x = 2 is between
2(π )
the endpoints of the allowable interval (0 and 6), then this will give the minimum.

Therefore, the location of P which minimizes the area that Chuck can reach is 2 m along the
wall from A towards B.

3. (a) Solution 1
Let ∠PAB = θ . P
A 5
o
Then ∠XAB = 180 − θ , and so ∠XYB = θ since XYBA is 16 θ C1
0° –θ 7
a cyclic quadrilateral, and so opposite angles sum to X 18 6
o
180 .
B
Therefore, ∆PAB is similar to ∆PYX (common angle at C2
θ
P, equal angle θ ).
XY PX BA ⋅ PX 6(5 + 16) Y
So, = or XY = = = 18 .
BA PB PB 7

Solution 2
By the cosine law in ∆APB ,
AB2 = PA 2 + PB 2 − 2(PA)( PB) cos(∠APB)
36 = 25 + 49 − 2(5)(7) cos(∠APB)
cos(∠APB) = 38
70
= 19
35
2004 Canadian Open Mathematics Challenge Solutions 13

Now, PX and PY are both secants of circle C2 , so by the P

A 5
Secant-Secant Theorem, 16
PA ⋅ PX = PB ⋅ PY 6
7
X
5(5 + 16) = 7(7 + BY )
8 B
105 = 7(7 + BY )
15 = 7 + BY Y
BY = 8

Now in ∆PXY we know the lengths of sides PX (length 21), PY (length 15) and the cosine of
∠XPY = ∠APB , so we can use the cosine law to calculate the length of XY.
XY 2 = PX 2 + PY 2 − 2(PX )(PY ) cos(∠XPY )
XY 2 = 441+ 225 − 2(21)(15) ( 1935 )
XY 2 = 441+ 225 − 2( 3)( 3)(19)
2
XY = 441+ 225 − 342
XY 2 = 324
XY = 18
Therefore, the length of XY is 18.

(b) Solution 1
Since the circle C4 is fixed, the length VW will be fixed if the angle it subtends on the circle is
fixed, ie. if the angle ∠VHW does not depend on the position of Q.
Now Q
o
G
∠VHW = 180 − ∠VHQ C 3
o
( o
= 180 − 180 − ∠GVH − ∠GQH ) V
= ∠GVH + ∠GQH H
C4

W
But since chord GH is a chord of fixed length in both circles (ie. it doesn’t change as Q moves),
then the angles that it subtends in both circles are constant. In particular, ∠GVH and ∠GQH
are both fixed (that is, they do not depend on the position of Q).
Since both of these angles are fixed, then ∠VHW = ∠GVH + ∠GQH is also fixed.
Therefore, the length of VW is fixed.
2004 Canadian Open Mathematics Challenge Solutions 14

Solution 2
We start by noting that chord GH has a constant length, Q
G
ie. does not depend on the the position of Q. Thus, GH is C3
a constant chord in both C3 and C4 .
V
In C3 , let ∠GQH = α .
H
In C4 , let ∠GVH = ∠GWH = β . C4
These angles are constant since GH is of constant length.
W
Therefore, ∠VHQ = ∠QGW = 180 − (α + β ) , and so ∠VGW = ∠WHV = α + β .
o

Since these last two angles are constant for all position of Q, then VW is a chord of constant
length.

Solution 3
We proceed by considering two different positions for the point Q, which we call Q1 and Q2 .
These points will create two different positions for the line segment VW, which we call V1 W1
and V2 W2 .
To show that the length of VW is constant, we must show that V1 W1 and V2 W2 have the same
length. We do know that the points G and H are fixed, so the length of GH does not vary.

Q1 V2
G
G Q2

H
H
V1
W1
W2

As in part (a) Solution 1, ∆Q1 HG and ∆Q1V1W1 are similar triangles, as are ∆Q2 HG and
∆Q2V2W2 . (The two positions for Q play the same role as P, points G and H play the same role
as A and B, and the points V and W play the same role as X and Y.)
VW QV QV
Therefore, by similar triangles, 1 1 = 1 1 or V1 W1 = HG ⋅ 1 1 .
HG Q1 H Q1 H
VW QV QV
Also, by similar triangles, 2 2 = 2 2 or V2 W2 = HG ⋅ 2 2 .
HG Q2 H Q2 H
Since the length of HG (or GH) does not change, to show that V1 W1 and V2 W2 have the same
QV Q V
length, we must show that 1 1 = 2 2 .
Q1 H Q2 H

Join H to each of V1 and V2 .

Now GH is a chord of fixed length in both circles, so the angle that it subtends at any point on
the circumference of each circle is the same.
2004 Canadian Open Mathematics Challenge Solutions 15

Therefore, ∠GQ1 H = ∠GQ2 H and ∠GV1 H = ∠GV2 H .

But this tells us that ∆Q1 HV1 and ∆Q2 HV2 are similar.
QV QH QV Q V
Thus, 1 1 = 1 or 1 1 = 2 2 .
Q2 V2 Q2 H Q1 H Q2 H
QV QV
Therefore, V1 W1 = HG ⋅ 1 1 = HG ⋅ 2 2 = V2 W2 , ie. the length of VW is constant.
Q1 H Q2 H

4. (a) Solution 1
3 2
Since a, b and c are the roots of the equation x − 6x + 5x − 1 = 0 , then using the properties of
the coefficients of a cubic equation,
a +b+c =6
ab + ac + bc = 5
abc = 1
Since we know that each of a, b and c is a roots of the equation, then
a 3 − 6a2 + 5a − 1 = 0
b3 − 6b2 + 5b − 1 = 0
c3 − 6c 2 + 5c − 1 = 0
or after rearranging
a 3 = 6a2 − 5a + 1
b 3 = 6b2 − 5b + 1 (*)
3 2
c = 6c − 5c + 1
Adding these three equations, we obtain
a 3 + b3 + c3 = 6a2 + 6b 2 + 6c 2 − 5a − 5b − 5c + 3
( )
= 6 a 2 + b 2 + c 2 − 5( a + b + c ) + 3
2 2 2
We already know that a + b + c = 6, so if we could determine the value of a + b + c , then
3 3 3
we would know the value of a + b + c .
But
(a + b + c) 2 = a2 + b2 + c 2 + 2ab + 2bc + 2ac
6 2 = a2 + b2 + c 2 + 2( ab + bc + ac)
a 2 + b 2 + c 2 = 36 − 2(5)
a 2 + b 2 + c 2 = 26
and so
( )
a 3 + b3 + c3 = 6 a 2 + b 2 + c 2 − 5( a + b + c ) + 3 = 6(26) − 5(6) + 3 = 129

If we know take the equations in (*) and multiply both sides in the first, second and third
equations by a, b and c, respectively, we get
2004 Canadian Open Mathematics Challenge Solutions 16

a 4 = 6a 3 − 5a2 + a
b 4 = 6b 3 − 5b 2 + b (**)
4 3 2
c = 6c − 5c + c
which we then can add to obtain
4 3
(3 3 2 2
) (
a + b + c = 6 a + b + c − 5 a + b + c + (a + b + c )
4 4 2
)
= 6(129) − 5(26) + 6
= 650
Repeating the process one more time by multiplying the first, second and third equations in
(**) by a, b and c, respectively, and adding, we obtain
5 5 5 4
( 4 4
) ( 3
a +b +c = 6 a +b +c −5 a +b +c + a +b +c
3 3 2
) (
2 2
)
= 6(650) − 5(129) + 26
= 3281
5 5 5
Therefore, the value of a + b + c is 3281.

Solution 2
3 2
Since a, b and c are the roots of the equation x − 6x + 5x − 1 = 0 , then using the properties of
the coefficients of a cubic equation,
s= a +b+ c =6
t = ab + ac + bc = 5
p = abc = 1
5 5 5
We will attempt to express a + b + c in terms of s, t and p, which will thus allow us to
5 5 5
calculate the value of a + b + c .

First,
(a + b + c) 2 = a2 + b2 + c 2 + 2ab + 2bc + 2ac
a 2 + b 2 + c 2 = s2 − 2t
Next,
( )
a 2 + b2 + c2 (a + b + c) = a3 + b3 + c 3 + a2 b + a 2 c + b2 a + b2 c + c 2 a + c 2b
3 3 3 2
( 2
) [ 2 2 2 2 2
a + b + c = s s − 2t − a b + a c + b a + b c + c a + c b ]
= s( s2 − 2t ) − [(ab + ac + bc )( a + b + c ) − 3abc]

= s( s2 − 2t ) − [ts − 3p ]

= s3 − 3st + 3p
5 5 5
We can now attempt to express a + b + c as
2004 Canadian Open Mathematics Challenge Solutions 17

a 5 + b5 + c 5
( )( ) [
= a 2 + b2 + c 2 a3 + b 3 + c3 − a2 b3 + a 2 c3 + b 2 a3 + b2 c 3 + c 2 a3 + c 2 b3 ]
= (a 2 + b2 + c 2 )(a3 + b 3 + c3 ) − [(a 2 b2 + a2 c 2 + b 2c 2 )(a + b + c ) − (a 2 b2 c + a 2 bc 2 + ab2 c 2 )]

= (s2 − 2t )( s3 − 3st + 3p) − [[(ab + ac + bc )2 − 2( a2 bc + ab 2 c + abc 2 )](a + b + c ) − abc (ab + ac + bc )]

= (s2 − 2t )( s3 − 3st + 3p) − [[t 2 − 2abc( a + b + c )]s − pt ]

= (s2 − 2t )( s3 − 3st + 3p) − [[t 2 − 2ps]s − pt ]

= (62 − 2( 5))(63 − 3(6 )(5) + 3(1)) − [[ 52 − 2(1)(6)](6 ) − 1(5)]

= (26)(129) − [[13](6 ) − 5]
= 3354 − [73]
= 3281

5 5 5
Therefore, the value of a + b + c is 3281.

(b) We will proceed by dividing our proof into several steps.

Step 1: Estimate the values of a, b and c
n n n
Step 2: Show that a + b + c is an integer for every positive integer n
Step 3: Final conclusion

Step 1: Estimate the values of a, b and c

Define f ( x ) = x 3 − 6x 2 + 5x − 1.
If x < 0 , then x < 0 , −6 x < 0 , 5x < 0 , and −1 < 0 , so f ( x ) = x 3 − 6x 2 + 5x − 1< 0.
3 2

This tells us that f ( x ) = 0 cannot have any negative roots. It is also clear that 0 is not a
root of f ( x ) = 0 , so each of a, b and c is positive.
If we calculate a few values of f ( x ) = x 3 − 6x 2 + 5x − 1, we obtain f (0) = −1 , f (1) = −1 ,
f (2) = −7 , f ( 3) = −13 , f ( 4) = −13 , f (5) = −1 , and f (6) = 29 .
Therefore, one of the roots is between 5 and 6.
However, we know from part (a) that a + b + c = 6 , so since all three roots are positive, then
we must have 5 < c < 6 and 0 < a,b < 1 . (Since one root is bigger than 5, each root is
positive, and the sum of the three roots is 6, then neither a nor b is bigger than 1.)
Since 5 < c < 6 and a + b + c = 6 , then 0 < a + b < 1.
We also know from part (a) that abc = 1. Since 5 < c < 6 , then 16 < ab < 15 and since each
1
of a and b is less than 1, then each of a and b must be bigger than 6.
Since a and b are each bigger than 1
6 and 0 < a + b < 1, then 1
6 < a,b < 56 .
(We could have proceeded less formally by doing some quick calculations to see that
f (0.1) = −0.559 , f (0.2) = −0.232 , f (0.3) = −0.013 , f (0.4 ) = 0.104 , f (0.5) = 0.125 ,
2004 Canadian Open Mathematics Challenge Solutions 18

f (0.6) = 0.056 , f (0.7) = −0.097 , and so a must be between 0.3 and 0.4, and b must be
between 0.6 and 0.7.)

n n n
Step 2: Show that a + b + c is an integer for every positive integer n
n n n
In part (a), we saw that a + b + c is definitely an integer for n equal to 1, 2 and 3.
If we return to the set of equations (*) in Solution 1 of part (a) and multiply the three
n −3 n− 3 n− 3
equations by a , b and c , respectively, we obtain
n −3
a = 6a
n n− 1
− 5a n− 2
+a
bn = 6bn −1 − 5b n− 2 + b n− 3 (***)
n n −1
c = 6c − 5c n− 2 + c n− 3
and adding, we get
n n n
a +b +c =6 a (
n− 1
+b
n− 1
+c
n− 1
−5 a ) (
n− 2
+b
n− 2
+c
n− 2
) (
+ a
n −3
+b
n −3
+c
n −3
) (****)
for every n greater than or equal to 4.
k k k
If we set n equal to 4, then since a + b + c is an integer for k equal to 1, 2 and 3, then by
4 4 4
(****), a + b + c is also an integer.
k k k
If we set n equal to 5, then since a + b + c is an integer for k equal to 2, 3 and 4, then by
5 5 5
(****), a + b + c is also an integer.
k k k
It is now clear than we can continue this process inductively, since if a + b + c is an
n n n
integer for k equal to n − 3, n − 2 , and n − 1, then a + b + c will also be an integer, by
(****).
2003
In particular, we can conclude that a + b2003 + c 2003 and a2004 + b 2004 + c 2004 are both
2003
integers, say a + b2003 + c 2003 = M and a2004 + b 2004 + c 2004 = N .

Step 3: Final conclusion

2003
Since each of a and b is between 0 and 1, then a > a2004 and b 2003 > b2004 , so
a2003 + b2003 > a2004 + b 2004 .
Since a is less than 56 , then a is less than 0.9, so a < 0.81, so a4 < (0.81) < 0.7, so
2 2

a8 < (0.7) < 0.5 , so a16 < (0.5) < 0.25 .

2 2

16
Similarly, since b is less than 56 , then b < 0.25 .
16 16
Therefore, a + b < 0.5 , and since each of a and b is less than 1, then
a2004 + b 2004 < a2003 + b2003 < a16 + b16 < 0.5 .
Therefore, since c
2003
( )
= M − a2003 + b2003 and a2003 + b2003 < 0.5 , then the closest
2003 2003 2003
integer to c is M and the distance between them is a +b . Similarly, the closest
2004 2004 2004
integer to c is N and the distance between them is a +b .
2004 2004 2003 2003 2004 2003
But a +b <a +b , so c is closer to N than c is to M, as required.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING
presents the

Canadian Open
Mathematics Challenge
Wednesday, November 24, 2004

Time: 2 12 hours 2004

c Canadian Mathematical Society

Calculators are NOT permitted.

Do not open this booklet until instructed to do so.

There are two parts to this paper.

PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value for each
question by entering the correct answer in the space provided. Any work you do in obtaining an answer
will be considered for part marks if you do not have the correct answer, provided that it is done in the
space allocated to that question in your answer booklet.

PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must be written
in the appropriate location in the answer booklet. Rough work should be done separately. If you require
extra pages for your finished solutions, paper will be provided by your supervising teacher. Any extra
papers should be placed inside your answer booklet.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution poorly
presented will not earn full marks.

NOTE: At the completion of the contest, insert the information sheet inside the answer
booklet.
 Canadian Open Mathematics Challenge

NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations
√ and answers will be expressed as
exact numbers such as 4π, 2 + 7, etc.
4. Calculators are not allowed.

PART A
1. If x + 2y = 84 = 2x + y, what is the value of x + y?

2. Let S be the set of all three-digit positive integers whose digits are 3, 5 and 7, with no digit
repeated in the same integer. Calculate the remainder when the sum of all of the integers in
S is divided by 9.

3. In the diagram, point E has coordinates

√ (0, 2), and B lies on y
the positive x-axis so that BE = 7. Also, point C lies on
the positive x-axis so that BC = OB. If point D lies in the E
first quadrant such that ∠CBD = 30◦ and ∠BCD = 90◦ ,
what is the length of ED?
D

x
O B C
4. A function f (x) has the following properties:

i) f (1) = 1
ii) f (2x) = 4f (x) + 6
iii) f (x + 2) = f (x) + 12x + 12

Calculate f (6).

5. The Rice Tent Company sells tents in two different sizes, large and small. Last year, the
Company sold 200 tents, of which one quarter were large. The sale of the large tents pro-
duced one third of the company’s income. What was the ratio of the price of a large tent to
the price of a small tent?

6. In the diagram, a square of side length 2 has semicircles

drawn on each side. An “elastic band” is stretched tightly
around the figure. What is the length of the elastic band in
this position?
7. Let a and b be real numbers, with a > 1 and b > 0.
a
If ab = ab and = a3b , determine the value of a.
b
8. A rectangular sheet of paper, ABCD, has AD = 1 and AB = r, where 1 < r < 2. The
paper is folded along a line through A so that the edge AD falls onto the edge AB. Without
unfolding, the paper is folded again along a line through B so that the edge CB also lies
on AB. The result is a triangular piece of paper. A region of this triangle is four sheets
thick. In terms of r, what is the area of this region?

PART B
1. The points A(−8, 6) and B(−6, −8) lie on the circle x2 + y 2 = 100.

(a) Determine the equation of the line through A and B.

(b) Determine the equation of the perpendicular bisector of AB.
(c) The perpendicular bisector of AB cuts the circle at two points, P in the first quadrant
and Q in the third quadrant. Determine the coordinates of P and Q.
(d) What is the length of P Q? Justify your answer.

2. (a) Determine the two values of x such that x2 − 4x − 12 = 0.

√
(b) Determine the one value of x such that x − 4x + 12 = 0. Justify your answer.
(c) Determine all real values of c such that
p
x2 − 4x − c − 8x2 − 32x − 8c = 0

has precisely two distinct real solutions for x.

3. A map shows all Beryl’s Llamaburgers restaurant locations in North America. On this map,
a line segment is drawn from each restaurant to the restaurant that is closest to it. Every
restaurant has a unique closest neighbour. (Note that if A and B are two of the restaurants,
then A may be the closest to B without B being closest to A.)

(a) Prove that no three line segments on the map can form a triangle.
(b) Prove that no restaurant can be connected to more than five other restaurants.

4. In a sumac sequence, t1 , t2 , t3 , . . ., tm , each term is an integer greater than or equal to 0.

Also, each term, starting with the third, is the difference of the preceding two terms (that
is, tn+2 = tn − tn+1 for n ≥ 1). The sequence terminates at tm if tm−1 − tm < 0. For
example, 120, 71, 49, 22, 27 is a sumac sequence of length 5.

(a) Find the positive integer B so that the sumac sequence 150, B, . . . has the maximum
possible number of terms.
(b) Let m be a positive integer with m ≥ 5. Determine the number of sumac sequences
of length m with tm ≤ 2000 and with no term divisible by 5.
The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING
presents the

Canadian Open
Mathematics Challenge
Wednesday, November 24, 2004

Solutions

2004
c Canadian Mathematical Society
2004 COMC Solutions Page 2 of 19

Part A

1. If x + 2y = 84 = 2x + y, what is the value of x + y?

Solution 1
Since x + 2y = 84 and 2x + y = 84, then adding these two equations together, we obtain
3x + 3y = 168 or x + y = 56.

Solution 2
Since x + 2y = 84, then x = 84 − 2y.
Substituting into the second equation, we get

2(84 − 2y) + y = 84
168 − 3y = 84
84 = 3y
y = 28

Therefore, x = 84 − 2(28) = 28 and so x + y = 56.

Solution 3
Since 2x + y = 84, then y = 84 − 2x.
Substituting into the first equation, we get

x + 2(84 − 2x) = 84
168 − 3x = 84
84 = 3x
x = 28

Therefore, y = 84 − 2(28) = 28 and so x + y = 56.

Solution 4
SInce these two expressions are identical when x is replaced by y and y is replaced by x, then
x = y.
Therefore, 3x = 84 or x = 28 and so y = 28.
Thus, x + y = 56.

Answer: 56
2004 COMC Solutions Page 3 of 19

2. Let S be the set of all three-digit positive integers whose digits are 3, 5 and 7, with no digit
repeated in the same integer. Calculate the remainder when the sum of all of the integers in S
is divided by 9.

Solution 1
We can write down the elements of S: 357, 375, 537, 573, 735, 753.
The sum of these elements is 357 + 375 + 537 + 573 + 735 + 753 = 3330.
Since 3330 is divisible by 9 (because the sum of its digits is divisible by 9), the remainder when
we divide by 9 is 0.

Solution 2
There are six numbers formed with the three given numbers.
Two of these numbers have a 3 in the 100s position, two have a 5 in the 100s position, and two
have a 7 in the 100s position.
The same can be said about the distribution of numbers in the 10s and units positions.
Therefore, the sum of the six numbers is

2(3 + 5 + 7)(100) + 2(3 + 5 + 7)(10) + 2(3 + 5 + 7)(1) = 3330

The remainder is 0 when 3330 is divided by 9.

Answer: 0

3. In the diagram, point E has coordinates (0, 2), and B lies on y

√
the positive x-axis so that BE = 7. Also, point C lies on the
positive x-axis so that BC = OB. If point D lies in the first E

quadrant such that ∠CBD = 30◦ and ∠BCD = 90◦ , what is the
length of ED? D

x
O B C
Solution
In order to find the length of ED, we will try to find the coordinates of D. Let the coordinates
of B be (b, 0).
√
Since BE = 7 and the coordinates of E are (0, 2), then
p √
(b − 0)2 + (0 − 2)2 = 7
b2 + 4 = 7
b2 = 3
√ √
Since the point B lies on the positive x-axis, then b = 3 (not − 3), so B has coordinates
√
( 3, 0).
2004 COMC Solutions Page 4 of 19

(Note that it would have also been possible to find the coordinates of B by using Pythagoras
√ √
– OE 2 + OB 2 = EB 2 so OB 2 = 7 − 4 = 3 so OB = 3 and so B has coordinates ( 3, 0).)
√
Since BC = OB, then C has coordinates (2 3, 0).
√
Since ∠BCD = 90◦ and D lies in the first quadrant, then D has coordinates (2 3, d), with
d > 0.
Since 4DBC has ∠BCD = 90◦ and ∠CBD = 30◦ , then it is a 30◦ − 60◦ − 90◦ triangle. Since
√
CB = 3 (and CB is opposite the 60◦ angle), then DC (which is opposite the 30◦ angle) has
length 1.
√
Therefore, D has coordinates (2 3, 1).

E (0,2 )

(
D 2 3, 1 )
x
O
(
C 2 3, 0 )
B ( 3, 0 )
q √ √ √
Thus, ED = (2 3 − 0)2 + (1 − 2)2 = 12 + 1 = 13.
√
Answer: 13

4. A function f (x) has the following properties:

i) f (1) = 1
ii) f (2x) = 4f (x) + 6
iii) f (x + 2) = f (x) + 12x + 12

Calculate f (6).

Solution 1
Using property ii) with x = 1,

f (2) = 4f (1) + 6 = 4(1) + 6 = 10

since f (1) = 1 by property i).

Using property ii) with x = 2,

f (4) = 4f (2) + 6 = 4(10) + 6 = 46

2004 COMC Solutions Page 5 of 19

Using property iii) with x = 4,

f (6) = f (4) + 12(4) + 12 = 46 + 48 + 12 = 106

Therefore, the value of f (6) is 106.

Solution 2
Using property iii) with x = 1,

f (3) = f (1) + 12(1) + 12 = 1 + 12 + 12 = 25

since f (1) = 1 by property i).

Using property ii) with x = 3,

f (6) = 4f (3) + 6 = 4(25) + 6 = 106

Therefore, the value of f (6) is 106.

Solution 3
Working backwards,

f (6) = 4f (3) + 6 (by property ii) with x = 3)

= 4(f (1) + 12(1) + 12) + 6 (by property iii) with x = 1)
= 4f (1) + 4(24) + 6
= 4(1) + 102 (by property i))

Therefore, the value of f (6) is 106.

Answer: f (6) = 106

5. The Rice Tent Company sells tents in two different sizes, large and small. Last year, the Com-
pany sold 200 tents, of which one quarter were large. The sale of the large tents produced one
third of the company’s income. What was the ratio of the price of a large tent to the price of
a small tent?

Solution
Since the Rice Tent Company sold 200 tents, of which one quarter were large, then they sold
50 large tents and 150 small tents last year.
Let L be the price of a large tent and S the price of a small tent.
Then their income from large tents was 50L and from small tents was 150S.
Their total income last year was 50L + 150S.
2004 COMC Solutions Page 6 of 19

From the given information,

1
50L = (50L + 150S)
3
150L = 50L + 150S
100L = 150S
L 150 3
= =
S 100 2
Therefore, the ratio of the price of a large tent to the price of a small tent was 3 : 2.

Answer: 3 : 2

6. In the diagram, a square of side length 2 has semicircles drawn on

each side. An “elastic band” is stretched tightly around the figure.
What is the length of the elastic band in this position?

Solution
Label the four vertices of the square as W , X, Y , Z, in clockwise order.
Label the four midpoints of the sides of the square (that is, the centres of the four semicircles)
as M , N , O, P , in clockwise order, starting with M being the midpoint of W X.
In each semicircle, join the centre to the two points on that semicircle where the band just
starts (or stops) to contact the circle. Label these eight points as A, B, C, D, E, F , G, and H
in clockwise order, starting with A and B on the semicircle with centre M .

A B

H W M X C

P N

G Z O Y D

F E

By symmetry, the four straight parts of the band will be equal in length (that is, BC =
DE = F G = HA) and the four arc segments of the band will be equal in length (that is,
AB = CD = EF = GH).
Therefore, the total length of the band is 4(Length of arc AB) + 4(Length of BC).

Now, BC will actually be tangent to the two semicircles (with centres M and N ) where it
initially just touches them.
Thus, M B and N C are both perpendicular to BC.
2004 COMC Solutions Page 7 of 19

Since M B = N C = 1 (because they are radii of the semicircles and each semicircle has diame-
ter 2), then M BCN must actually be rectangle, so BC is equal and parallel to M N .
Since M and N are the midpoints of the sides of the square of side length 2, then M Y = Y N =
√ √
1, so M N = 2, so BC = 2.

Next, we determine the length of AB. Previously, we saw that M BCN is a rectangle, so
BC was parallel to M N . Similarly, HA is parallel to P M .
But P M is perpendicular to M N , so HA is perpendicular to BC.
Therefore, ∠AM B = 90◦ , ie. AB is one-quarter of the circumference of a circle with radius 1
or 41 (2π(1)) = 12 π.

√ √
Therefore, the total length of the band is 4( 21 π) + 4( 2) = 2π + 4 2.
√
Answer: 2π + 4 2

7. Let a and b be real numbers, with a > 1 and b > 0.

a
If ab = ab and = a3b , determine the value of a.
b

Solution 1
a
Since ab = ab and = a3b , then multiplying these two equations together, we get
b
a2 = ab · a3b = a4b .

Dividing both sides by a2 (since a 6= 0), we get a4b−2 = 1.

Since a > 1, then 4b − 2 = 0 or b = 12 .
√ √
Substituting back into the first equation, we get 21 a = a1/2 = a or a = 2 a.
Squaring both sides gives a2 = 4a or a2 − 4a = 0 or a(a − 4) = 0.
Since a > 1, then a = 4.

Solution 2
Since ab = ab , then, dividing both sides by a which is not equal to 0, we get b = ab−1 .
a
Since = a3b , then a = ba3b = ab−1 a3b = a4b−1 .
b
Comparing exponents, we get 1 = 4b − 1 or b = 21 .
Substituting b = 12 into ab = ab , we have 21 a = a1/2 or a1/2 = 2 or a = 4. So a = 4.

Solution 3
Since a > 1 and b > 0, we can take logarithms of both sides of both equations.


In the first equation, using log rules on log(ab) = log ab gives log(a) + log(b) = b log(a).
In the first equation, using log rules on log ab = log a3b gives log(a) − log(b) = 3b log(a).

Adding these two new equations gives 2 log(a) = 4b log(a) or (4b − 2) log(a) = 0.
Since a > 1, then log(a) > 0, so we must have 4b − 2 = 0 or b = 12 .
2004 COMC Solutions Page 8 of 19

Substituting this back into the first log equation gives log(a) + log 12 = 21 log(a) or


1
log(a) = − log 12 = log(2) or log(a) = 2 log(2) = log(4), so a = 4.


2

Answer: 4

8. A rectangular sheet of paper, ABCD, has AD = 1 and AB = r, where 1 < r < 2. The paper is
folded along a line through A so that the edge AD falls onto the edge AB. Without unfolding,
the paper is folded again along a line through B so that the edge CB also lies on AB. The
result is a triangular piece of paper. A region of this triangle is four sheets thick. In terms of
r, what is the area of this region?

Solution
Start with the rectangular sheet of paper, ABCD, with A in the top left and B in the bottom
left.
Fold AD across so that AD lies along AB. Let D0 be the point were D touches AB and let E
be the point on DC where the fold hits DC.
Since AD0 is the old AD, then AD0 = 1.
Since D0 E is perpendicular to D0 A (since DC was perpendicular to AD) then D0 E is parallel
to BC, so D0 E = 1 as well.

A D

1
D′ E
r −1
B C

We can also conclude that D0 B = EC = r − 1, since AB has length r.

Next, we fold the paper so that BC lies along AB. Unfold this paper and lay it flat so
that we can see the crease.
Since BC is folded onto AB, then the crease bisects ∠ABC, that is the crease makes an angle
of 45◦ with both AB and BC.
Suppose that the crease crosses D0 E at X and AE at Y .
2004 COMC Solutions Page 9 of 19

D′ E
X

B C

Now when the paper had been folded the second time (before we unfolded it!), the only way to
obtain a region four sheets thick was to fold a region two sheets thick on top of a region which
is also two sheets thick.
Since 4XY E is the only part of the paper “below” the second crease which is two sheets thick,
and, when the second fold is made, it lies entirely over another region which is two sheets thick,
then the desired area is the area of 4XY E.

Since ∠ABX = 45◦ , then 4BD0 X is isosceles and right-angled, so D0 X = D0 B = r − 1.

Thus, EX = 1 − D0 X = 1 − (r − 1) = 2 − r.
Since ∠D0 XB = 45◦ , then ∠Y XE = 45◦ . Also, since 4AD0 E is right isosceles, then ∠Y EX =
45◦ , so 4XY E is isosceles and right-angled.

s s

X 2−r E

√
Suppose XY = s. Then 2s = XE = 2 − r or 2s2 = (2 − r)2 .
The area of 4XY E is 21 s2 or 14 (2 − r)2 .
Therefore, the area of the desired region is 41 (2 − r)2 .

Answer: 14 (2 − r)2
2004 COMC Solutions Page 10 of 19

Part B

1. The points A(−8, 6) and B(−6, −8) lie on the circle x2 + y 2 = 100.

(a) Determine the equation of the line through A and B.

Solution
6 − (−8)
First, we determine the slope of the line segment AB. The slope is = −7.
−8 − (−6)
We could now proceed to find the equation of the line in several different ways.
Using the point-slope form, we obtain y − 6 = −7(x − (−8)) or y = −7x − 50.

(b) Determine the equation of the perpendicular bisector of AB.

Solution
Since the slope of AB is −7, then the slope of the perpendicular bisector of AB is 71 .
Also, the perpendicular bisector passes through the midpoint of AB, which is
 
1 1
((−8) + (−6)), (6 + (−8)) = (−7, −1).
2 2

Therefore, the equation of the perpendicular bisector is y − (−1) = 17 (x − (−7)) or y = 17 x.

(c) The perpendicular bisector of AB cuts the circle at two points, P in the first quadrant
and Q in the third quadrant. Determine the coordinates of P and Q.

Solution 1
y

A
P
x
Q

We want to find the points of intersection of the circle x2 + y 2 = 100 and the line y = 17 x.
From the equation of the line, x = 7y. Substituting this into the equation of the circle we
2004 COMC Solutions Page 11 of 19

obtain

(7y)2 + y 2 = 100
49y 2 + y 2 = 100
50y 2 = 100
y2 = 2
√
y = ± 2
√ √ √ √
Since x = 7y, then if y = 2, then x = 7 2, and if y = − 2, then x = −7 2.
√ √
Since P is in the first quadrant, then P has coordinates (7 2, 2).
√ √
Since Q is in the third quadrant, then Q has coordinates (−7 2, − 2).

Solution 2
We want to find the points of intersection of the circle x2 + y 2 = 100 and the line y = 17 x.
Substituting y = 71 x into the equation of the circle we obtain
 2
2 1
x + x = 100
7
1
x2 + x2 = 100
49
50 2
x = 100
49
x2 = 98
√ √
x = ± 98 = ±7 2
√ √ √ √
Since y = 17 x, then if x = 7 2, then y = 2, and if x = −7 2, then y = − 2.
√ √
Since P is in the first quadrant, then P has coordinates (7 2, 2).
√ √
Since Q is in the third quadrant, then Q has coordinates (−7 2, − 2).

(d) What is the length of P Q? Justify your answer.

Solution 1
The points P and Q are joined by the line y = 71 x, which passes through the origin.
Since the origin is the centre of the circle, then P Q must be a diameter of the circle.
Since the circle has equation x2 + y 2 = 100 = 102 , then its radius is 10, so its diameter is
20.
Therefore, P Q = 20.

Solution 2
√ √ √ √
Since we know that P (7 2, 2) and Q(−7 2, − 2), then we can determine the distance
2004 COMC Solutions Page 12 of 19

P Q by direct calculation:
r
√  √ 2 √  √ 2
PQ = 7 2 − −7 2 + 2− − 2
r
√ 2  √ 2
= 14 2 + 2 2
r
√ 2
= 2 [142 + 22 ]
p
= 2[196 + 4]
√
= 400
= 20

Therefore, the length of P Q is 20.

2. (a) Determine the two values of x such that x2 − 4x − 12 = 0.

Solution
Factoring the given equation x2 − 4x − 12 = 0, we obtain (x − 6)(x + 2) = 0.
Therefore, the two solutions are x = 6 and x = −2.

√
(b) Determine the one value of x such that x − 4x + 12 = 0. Justify your answer.

Solution
We first eliminate the square root by isolating it on one side and squaring:
√
x− 4x + 12 = 0
√
x = 4x + 12
x2 = 4x + 12
x2 − 4x − 12 = 0
(x − 6)(x + 2) = 0

Therefore, the two possible solutions are x = 6 and x = −2. (Since we have squared both
sides, it is possible that we have introduced an extraneous root, so we should verify both
of these.)
p √
If x = 6, then 6 − 4(6) + 12 = 6 − 36 = 0.
p √
If x = −2, then (−2) − 4(−2) + 12 = −2 − 4 = −4 6= 0.
Therefore, the one value of x that solves the equation is x = 6.
2004 COMC Solutions Page 13 of 19

(c) Determine all real values of c such that

√
x2 − 4x − c − 8x2 − 32x − 8c = 0

has precisely two distinct real solutions for x.

Solution
We start by attempting to solve this equation and then seeing what conditions on c arise.
Since 8x2 − 32x − 8c = 8(x2 − 4x − c), we let T = x2 − 4x − c.
Then the equation is
√
T − 8T = 0 (∗)
√
T = 8T
T 2 = 8T
T (T − 8) = 0

Therefore, T = 0 or T = 8. We can check that neither root is extraneous, so x2 − x − c = 0

or x2 − 4x − c = 8.
Let’s look at these last two equations.
First, we look at x2 − 4x − c = 0. The discriminant of this quadratic equation is (−4)2 −
4(−c) = 16 + 4c. Therefore, this equation has
• zero solutions if 16 + 4c < 0, so c < −4,
• exactly one solution if 16 + 4c = 0, so c = −4, and
• two distinct solutions if 16 + 4c > 0, so c > −4.
Next, we look at x2 − 4x − c = 8 or x2 − 4x − (c + 8) = 0. The discriminant of this
quadratic equation is (−4)2 − 4(−(c + 8)) = 48 + 4c. Therefore, this equation has
• zero solutions if 48 + 4c < 0, so c < −12,
• exactly one solution if 48 + 4c = 0, so c = −12, and
• two distinct solutions if 48 + 4c > 0, so c > −12.
We see also that any value of x that satisfies one of these two equations cannot satisfy the
other, since we cannot have both x2 − 4x − c = 0 and x2 − 4x − c = 8. (In other words,
no roots overlap between these two equations.)
We make a table to combine our observations:
c < −12 c = −12 −12 < c < −4 c = −4 c > −4
2
x − 4x − c = 0 0 solutions 0 solutions 0 solutions 1 solution 2 solutions
x2 − 4x − c = 8 0 solutions 1 solution 2 solutions 2 solutions 2 solutions
Total solutions 0 solutions 1 solution 2 solutions 3 solutions 4 solutions
Therefore, for there to be exactly two distinct solutions, we must have −12 < c < −4.
2004 COMC Solutions Page 14 of 19

3. A map shows all Beryl’s Llamaburgers restaurant locations in North America. On this map,
a line segment is drawn from each restaurant to the restaurant that is closest to it. Every
restaurant has a unique closest neighbour. (Note that if A and B are two of the restaurants,
then A may be the closest to B without B being closest to A.)

(a) Prove that no three line segments on the map can form a triangle.

Solution 1
We start by assuming that three line segments on the map do form a triangle, and show
that this is in fact impossible.
Notice that if restaurants X and Y are joined by a line segment, then either X is the
closest restaurant to Y or Y is the closest restaurant to X (or both).

Assume that A, B and C are the three points on the map connect by segments.
B

A C

To begin, we focus on the segment joining A to B. Let’s assume that A is the closest
restaurant to B. (It doesn’t matter which direction we assume here.) This means that C
is not the closest restaurant to B, so BA < BC.
But B and C are connected and C is not the closest restaurant to B. Therefore, B is the
closest restaurant to C, which means CB < CA.
But C and A are also connected and A is not the closest restaurant to C. Therefore, C is
the closest restaurant to A, which means AC < AB.

But this means that BA < BC, BC < AC and AC < BA. This cannot be the case.
Therefore, it is impossible for three line segments to form a triangle.

Solution 2
We prove this by showing that constructing a triangle is impossible.
We start by considering two locations A and B and the line segment AB.
Since A and B are connected, we can assume without loss of generality that A is closest
to B. (The case B closest to A involves interchanging A and B, and the case of A and B
closest to each other is included in the case of A closest to B.)

If A is closest to B and we add a new location C which is connected to B, then B

must be closest to C (since C can’t also be closest to B along with A).
2004 COMC Solutions Page 15 of 19

If we join C to A, then either C is closest to A or A is closest to C.

But A can’t be closest to C since B is closest to C.
Therefore, we must have C closest to A.
But then AC is shorter than AB, along with AB being shorter than BC (since A is closest
to B), which means that AC is shorter than BC or A is closer to C than B is, which isn’t
true. This is a contradiction.

Therefore, we can’t construct a triangle.

(b) Prove that no restaurant can be connected to more than five other restaurants.

Solution
We start by assuming that one restaurant can be connected to six others and show that
this is impossible. From this we can conclude that no restaurant can be connected to
more than five other restaurants (for if it could be joined to 8 others, say, then we could
consider six of them only and reach a contradiction).

Assume that restaurant A can be connected to restaurants B, C, D, E, F , and H, where

these restaurants are listed in clockwise order of their line segments joining to A.

H B

A C
F

D
E

Consider restaurants A, B and C.

We know that B and C are both connected to A and both cannot be the closest neighbour
to A. Thus, A must be the closest neighbour to one of these, say B. (It doesn’t matter
which we choose).
Since A is the closest restaurant to B, then BA < BC.
Now consider the line joining C to A.
If C is the closest neighbour to A, then AC < AB, so AC < AB < BC.
If A is the closest neighbour to C, then CA < CB so CA < CB and BA < CB.
In either case, BC is the (strictly) longest side in 4ABC, and so must be opposite the
(strictly) largest angle.
Since the angles in a triangle add to 180◦ , then if there is a largest angle, then this angle
must be larger than 60◦ . Therefore, from the above reasoning, ∠BAC > 60◦ .
2004 COMC Solutions Page 16 of 19

But we can reapply this reasoning to conclude that ∠CAD, ∠DAE, ∠EAF , ∠F AH,
and ∠HAB are each greater than 60◦ . But the sum of these six angles is 360◦ , since they
will form a full circle around A, and six angles, each greater than 60◦ , cannot add to 360◦ .
So we have a contradiction.

Therefore, it is impossible for a restaurant to be connected to more than five other restau-
rants.

4. In a sumac sequence, t1 , t2 , t3 , . . ., tm , each term is an integer greater than or equal to 0.

Also, each term, starting with the third, is the difference of the preceding two terms (that is,
tn+2 = tn − tn+1 for n ≥ 1). The sequence terminates at tm if tm−1 − tm < 0. For example, 120,
71, 49, 22, 27 is a sumac sequence of length 5.

(a) Find the positive integer B so that the sumac sequence 150, B, . . . has the maximum
possible number of terms.

Solution
Suppose that we have a sumac sequence with t1 = 150 and t2 = B. Let’s write out the
next several terms (assuming that they exist) in terms of B:

t3 = 150 − B t4 = 2B − 150 t5 = 300 − 3B t6 = 5B − 450

t7 = 750 − 8B t8 = 13B − 1200 t9 = 1950 − 21B t10 = 34B − 3150
In order to maximize the length of this sumac sequence, we would like to choose B so that
as many terms as possible starting from t1 are non-negative. (When we reach the first
negative “term”, we know that the sequence terminated at the previous term.)
For t2 ≥ 0, B ≥ 0.
For t3 ≥ 0, 150 − B ≥ 0 or B ≤ 150.
For t4 ≥ 0, 2B − 150 ≥ 0 or B ≥ 75.
For t5 ≥ 0, 300 − 3B ≥ 0 or B ≤ 100.
For t6 ≥ 0, 5B − 450 ≥ 0 or B ≥ 90.
For t7 ≥ 0, 750 − 8B ≥ 0 or B ≤ 750
8
= 93 68 .
For t8 ≥ 0, 13B − 1200 ≥ 0 or B ≥ 1200
13
= 92 134
.
1950 18
For t9 ≥ 0, 1950 − 21B ≥ 0 or B ≤ 21 = 92 21 .
Therefore, since B must be a positive integer, if we choose B = 93, then we can ensure
that each of t1 through t8 are non-negative. This will maximize the number of terms
4
starting from the beginning, since B must satisfy 92 13 ≤ B ≤ 93 86 in order for at least the
first eight terms to be non-negative. (Note that t9 will in fact be negative when B = 93.)

When we set B = 93, we obtain the sumac sequence 150, 93, 57, 36, 21, 15, 6, 9.
2004 COMC Solutions Page 17 of 19

(b) Let m be a positive integer with m ≥ 5. Determine the number of sumac sequences of
length m with tm ≤ 2000 and with no term divisible by 5.

Solution
We begin our solution by making some observations about sumac sequences.
• A sumac sequence is completely determined by its first two terms. This is true since
the first two terms give us the third, the second and third give us the fourth, and so
on. The sequence will terminate when the “next term” would be negative.
• In a sumac sequence, since for every (valid) n we have tn+2 = tn − tn+1 , then tn =
tn+1 + tn+2 . This means that we can “reverse engineer” a sumac sequence – if we know
terms (n + 1) and (n + 2), then we can determine term n. Thus, if we know the final
two terms in a sumac sequence, then we can determine all of the previous terms.
• From the first observation, the first two terms of a sumac sequence completely de-
termines the sequence. Is the same true of the last two terms? No. When we start
looking at a sumac sequence from the back, every new term as we proceed towards
the front will always be non-negative (since we are adding non-negative terms). Thus,
there is no “stopping condition” as there is when we work forwards. (For example, 3,
1, 2 is a sumac sequence ending with 1, 2, as is 4, 3, 1, 2.)
• However, if we know the final two terms and the length of the sequence, this completely
determines the sumac sequence (and we will always be able to find such a sequence).
Now we proceed. Let m be a fixed positive integer with m ≥ 5.
Suppose that t1 , t2 , . . . , tm−1 , tm is a sumac sequence of length m.
Because we are given a condition on the final term of the sequence, we will examine the
sequence from the back.
Let x = tm and y = tm−1 . Note that x, y and m determine the sequence.
Since x and y are the last two terms in the sumac sequence, then tm−1 − tm = y − x < 0
or x > y.
Since we have m fixed, we would like to determine how many sumac sequences we can
form with tm = x ≤ 2000, tm−1 = y < x and no term divisible by 5.

Let’s write out the last five terms of the sequence (in reverse order): x, y, x + y, x + 2y,
2x + 3y. (Since m ≥ 5, we know that there are at least five terms in the sequence.)
Since we want no term divisible by 5, let us consider x and y modulo 5 to see what hap-
pens. (There are 25 possible pairs for (x, y) modulo 5.)
Since no term is divisible by 5, then we don’t want x ≡ 0 (mod 5) or y ≡ 0 (mod 5). This
cuts us down to 16 possibilities for (x, y).
We make a table of these possibilities to determine which pairs can be eliminated simply
by looking at the last five terms. (All entries in the table are modulo 5. In any given row,
we stop if we reach a 0, since this possibility can then be discarded.)
2004 COMC Solutions Page 18 of 19

x y x+y x + 2y 2x + 3y
1 1 2 3 0
1 2 3 0
1 3 4 2 1
1 4 0
2 1 3 4 2
2 2 4 1 0
2 3 0
2 4 1 0
3 1 4 0
3 2 0
3 3 1 4 0
3 4 2 1 3
4 1 0
4 2 1 3 4
4 3 2 0
4 4 3 2 0
So the only possible pairs for (x, y) modulo 5 are (1, 3), (2, 1), (3, 4) and (4, 2).
If we start with (x, y) = (1, 3) modulo 5, then the terms in the sequence modulo 5 are 1,
3, 4, 2, 1, 3, 4, 2, 1, . . ., ie. the terms cycle modulo 5 with no terms divisible by 5.
This similar cycling will happen with each of the other 3 pairs, so each of these 4 pairs
give no terms divisible by 5.

So for each of these pairs, we need to determine the number of pairs of non-negative
integers (x, y) with x ≤ 2000, y < x and congruent to the appropriate things modulo
5. Each such pair will give a sumac sequence of length m ≥ 5 with no term divisible by
5. (Since the divisibility of the terms is independent of length, this also means that the
number of such sequences will be independent of m!)

Case 1: (x, y) congruent to (1, 3) modulo 5

Here x can take the values 1996, 1991, . . ., 6, 1.
If x = 1996, y can be 1993, 1988, . . ., 8, 3. (399 possibilities)
If x = 1991, y can be 1988, 1983, . . ., 8, 3. (398 possibilities)
This pattern continues, with one fewer possibility each time x decreases by 5, until we
reach x = 6, where y = 3 is the only possibility.
Thus, there are 399 + 398 + · · · + 2 + 1 possibilities for this case.
2004 COMC Solutions Page 19 of 19

Case 2: (x, y) congruent to (2, 1) modulo 5

Here x can take the values 1997, 1992, . . ., 7, 2.
If x = 1997, y can be 1996, 1991, . . ., 6, 1. (400 possibilities)
If x = 1992, y can be 1991, 1986, . . ., 6, 1. (399 possibilities)
This pattern continues, with one fewer possibility each time x decreases by 5, until we
reach x = 2, where y = 1 is the only possibility.
Thus, there are 400 + 399 + · · · + 2 + 1 possibilities for this case.

Case 3: (x, y) congruent to (3, 4) modulo 5

Here x can take the values 1998, 1993, . . ., 8, 3.
If x = 1998, y can be 1994, 1989, . . ., 9, 4. (399 possibilities)
If x = 1993, y can be 1989, 1984, . . ., 9, 4. (398 possibilities)
This pattern continues, with one fewer possibility each time x decreases by 5, until we
reach x = 8, where y = 4 is the only possibility.
Thus, there are 399 + 398 + · · · + 2 + 1 possibilities for this case.

Case 4: (x, y) congruent to (4, 2) modulo 5

Here x can take the values 1999, 1994, . . ., 9, 4.
If x = 1999, y can be 1997, 1992, . . ., 7, 2. (400 possibilities)
If x = 1994, y can be 1992, 1987, . . ., 7, 2. (399 possibilities)
This pattern continues, with one fewer possibility each time x decreases by 5, until we
reach x = 4, where y = 2 is the only possibility.
Thus, there are 400 + 399 + · · · + 2 + 1 possibilities for this case.

Therefore, overall there are

2(399+398+· · ·+2+1)+2(400+399+· · ·+2+1) = 399(400)+400(401) = 400(800) = 320 000

possibilities. Therefore, there are exactly 320 000 sumac sequences of length m with no
term divisible by 5 and with tm ≤ 2000.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

presents the

Canadian Open
Mathematics Challenge
Wednesday, November 23, 2005

Time: 2 12 hours 2005

c Canadian Mathematical Society

Calculators are NOT permitted.

Do not open this booklet until instructed to do so.

There are two parts to this paper.

PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value
for each question by entering the correct answer in the space provided. If you do not have the
correct answer, any work you do in obtaining an answer will be considered for part marks,
provided that it is done in the space allocated to that question in your answer booklet.
PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must
be written in the appropriate location in the answer booklet. Rough work should be done
separately. If you require extra pages for your finished solutions, paper will be provided by
your supervising teacher. Any extra papers should be placed inside your answer booklet. Be
sure to write your name and school name on any inserted pages.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution
poorly presented will not earn full marks.
NOTE: At the completion of the contest, insert the information sheet inside the
answer booklet.
 Canadian Open Mathematics Challenge
NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations
√ and answers will be expressed as
exact numbers such as 4π, 2 + 7, etc.
4. Calculators are not allowed.

PART A
1. Determine the value of 102 − 92 + 82 − 72 + 62 − 52 + 42 − 32 + 22 − 12 .

2. A bug in the xy-plane starts at the point (1, 9). It moves first to the point (2, 10) and
then to the point (3, 11), and so on. It continues to move in this way until it reaches
a point whose y-coordinate is twice its x-coordinate. What are the coordinates of
this point?

3. If ax3 + bx2 + cx + d = (x2 + x − 2)(x − 4) − (x + 2)(x2 − 5x + 4) for all values of x,

what is the value of a + b + c + d?
p
4. A fraction is in lowest terms if p and q have no common factor larger than 1.
q
1 2 70 71
How many of the 71 fractions , ,..., , are in lowest terms?
72 72 72 72
5. An office building has 50 storeys, 25 of which are painted black and the other 25 of
which are painted gold. If the number of gold storeys in the top half of the building
is added to the number of black storeys in the bottom half of the building, the sum
is 28. How many gold storeys are there in the top half of the building?

6. In the grid shown, each row has a value assigned

to it and each column has a value assigned to it. 3 0 5 6 −2
The number in each cell is the sum of its row and −2 −5 0 1 y
column values. For example, the “8” is the sum 5 2 x 8 0
of the value assigned to the 3rd row and the value 0 −3 2 3 −5
assigned to the 4th column. Determine the values −4 −7 −2 −1 −9
of x and y.
7. In the diagram, the semi-circle has centre O and diameter AB. A ray of light leaves
point P in a direction perpendicular to AB. It bounces off the semi-circle at point D
in such a way that ∠P DO = ∠EDO. (In other words, the angle of incidence equals
the angle of reflection at D.) The ray DE then bounces off the circle in a similar
way at E before finally hitting the semicircle again at B. Determine ∠DOP .

A B
P O
8. The number 18 is not the sum of any 2 consecutive positive integers, but is the sum
of consecutive positive integers in at least 2 different ways, since 5 + 6 + 7 = 18 and
3 + 4 + 5 + 6 = 18. Determine a positive integer less than 400 that is not the sum of
any 11 consecutive positive integers, but is the sum of consecutive positive integers
in at least 11 different ways.

PART B
1. A line with slope −3 intersects the positive x-axis at A and the positive y-axis at B.
A second line intersects the x-axis at C(7, 0) and the y-axis at D. The lines intersect
at E(3, 4).
(a) Find the slope of the line through C and E.
y

(b) Find the equation of the line through C B

and E, and the coordinates of the point D.

(c) Find the equation of the line through A D

and B, and the coordinates of the point B.
E (3, 4)
(d) Determine the area of the shaded region.

x
O A C (7, 0)
2. (a) Determine all possible ordered pairs (a, b) such that
a−b = 1
2 2
2a + ab − 3b = 22

(b) Determine all possible ordered triples (x, y, z) such that

x2 − yz + xy + zx = 82
2
y − zx + xy + yz = −18
z 2 − xy + zx + yz = 18

3. Four tiles identical to the one shown, with a > b > 0,

b
are arranged without overlap to form a square with a
square hole in the middle. b
a
a

(a) If the outer square has area (a + b)2 , show that the area of the inner square
is (a − b)2 .
(b) Determine the smallest integer value of N for which there are prime numbers
a and b such that the ratio of the area of the inner square to the area of the
outer square is 1 : N .
(c) Determine, with justification, all positive integers N for which there are odd
integers a > b > 0 such that the ratio of the area of the inner square to the
area of the outer square is 1 : N .
 Mathematics
Challenge

Canadian
(English)

Open

2005
4. Triangle ABC has its base on line segment P N and vertex A on line P M . Circles
with centres O and Q, having radii r1 and r2 , respectively, are tangent to the triangle
ABC externally and to each of P M and P N .

E
A
D Q
O L
K
P F B C G N

(a) Prove that the line through K and L cuts the perimeter of triangle ABC into
two equal pieces.
(b) Let T be the point of contact of BC with the circle inscribed in triangle ABC.
Prove that (T C)(r1 ) + (T B)(r2 ) is equal to the area of triangle ABC.
The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

presents the

Canadian Open
Mathematics Challenge
Wednesday, November 23, 2005

Solutions

2005
c Canadian Mathematical Society
2005 COMC Solutions Page 2 of 26

Part A

1. Determine the value of 102 − 92 + 82 − 72 + 62 − 52 + 42 − 32 + 22 − 12 .

Solution 1
Using differences of squares,

102 − 92 + 82 − 72 + 62 − 52 + 42 − 32 + 22 − 12
= (10 − 9)(10 + 9) + (8 − 7)(8 + 7) + (6 − 5)(6 + 5) + (4 − 3)(4 + 3) + (2 − 1)(2 + 1)
= 1(10 + 9) + 1(8 + 7) + 1(6 + 5) + 1(4 + 3) + 1(2 + 1)
= 10 + 9 + 8 + 7 + 6 + 5 + 4 + 3 + 2 + 1
= 55

(We can get the answer 55 either by computing the sum directly, or by using the fact that
1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 + 10 = 21 (10)(11) = 55.)

Solution 2
Computing directly,

102 − 92 + 82 − 72 + 62 − 52 + 42 − 32 + 22 − 12
= 100 − 81 + 64 − 49 + 36 − 25 + 16 − 9 + 4 − 1
= 19 + 15 + 11 + 7 + 3 (computing difference of each pair)
= 55

Answer: 55

2. A bug in the xy-plane starts at the point (1, 9). It moves first to the point (2, 10) and then to
the point (3, 11), and so on. It continues to move in this way until it reaches a point whose
y-coordinate is twice its x-coordinate. What are the coordinates of this point?

Solution 1
The bug starts at (1, 9) and each time moves 1 unit to the right and 1 unit up.
Thus, after k moves, the bug will be at the point (1 + k, 9 + k).
When its y-coordinate is twice its x-coordinate, we have 9 + k = 2(1 + k) or 9 + k = 2 + 2k
or k = 7.
When k = 7, the bug is at point (1 + 7, 9 + 7) = (8, 16), and the bug stops here.

Solution 2
The bug starts at (1, 9) and each time moves 1 unit to the right and 1 unit up.
Thus, at any point to which the bug moves, the y-coordinate will be 8 more than
2005 COMC Solutions Page 3 of 26

the x-coordinate, so every such point is of the form (n, n + 8).

For the y-coordinate to be twice the x-coordinate, n + 8 = 2n or n = 8.
When n = 8, the bug is at the point (8, 16), and the bug stops here.

Solution 3
We write out the sequence of points to which the bug moves and stop when we get to a point
where the y-coordinate is twice the x-coordinate:

(1, 9), (2, 10), (3, 11), (4, 12), (5, 13), (6, 14), (7, 15), (8, 16)

Thus, the bug stops at (8, 16).

Answer: (8, 16)

3. If ax3 + bx2 + cx + d = (x2 + x − 2)(x − 4) − (x + 2)(x2 − 5x + 4) for all values of x, what is

the value of a + b + c + d?

Solution 1
We use the fact that a + b + c + d = a(13 ) + b(12 ) + c(1) + d, so a + b + c + d must be equal to
the right side of the given equation with x set equal to 1.
Thus,

a + b + c + d = (12 + 1 − 2)(1 − 4) − (1 + 2)(12 − 5 + 4) = 0(−3) − 3(0) = 0

Solution 2
We simplify the right side of the given equation by factoring the two quadratic polynomials:

(x2 + x − 2)(x − 4) − (x + 2)(x2 − 5x + 4) = (x − 1)(x + 2)(x − 4) − (x + 2)(x − 1)(x − 4)

= 0

Therefore, ax3 + bx2 + cx + d = 0 for all values of x. (In other words, ax3 + bx2 + cx + d = 0 is
the zero polynomial, so all of its coefficients are equal to 0.)
Therefore, a = b = c = d = 0, so a + b + c + d = 0.

Solution 3
We expand and simplify the right side:

ax3 + bx2 + cx + d = (x2 + x − 2)(x − 4) − (x + 2)(x2 − 5x + 4)

= x3 + x2 − 2x − 4x2 − 4x + 8 − (x3 + 2x2 − 5x2 − 10x + 4x + 8)
= x3 − 3x2 − 6x + 8 − (x3 − 3x2 − 6x + 8)
= 0

Therefore, each of the coefficients of ax3 + bx2 + cx + d are 0, so a = b = c = d = 0 and

2005 COMC Solutions Page 4 of 26

thus a + b + c + d = 0.

Answer: 0

p
4. A fractionis in lowest terms if p and q have no common factor larger than 1.
q
1 2 70 71
How many of the 71 fractions , ,..., , are in lowest terms?
72 72 72 72

Solution 1
First, we note that 72 = 23 × 32 .
a
For one of the fractions to be in lowest terms, then a and 72 have no common factors.
72
In other words, a cannot be divisible by 2 or 3 (since 2 and 3 are the only prime numbers which
are divisors of 72).
How many of the positive integers from 1 to 71 are not divisible by 2 or 3?
Of these integers, 35 of the them are divisible by 2 (namely, 2, 4, 6, . . ., 70).
Also, 23 of them (namely, 3, 6, . . ., 69) are divisible by 3.
But some numbers are counted twice in these lists: all of the multiples of both 2 and 3 (ie. the
multiples of 6). These are 6, 12, . . ., 66 – that is, 11 numbers in total.
So the number of positive integers from 1 to 71 which are divisible by 2 or 3 is 35 + 23 − 11 = 47
(11 is subtracted to remove the double-counted numbers).
So the number of positive integers from 1 to 71 which are not divisible by 2 or 3 is 71 − 47 = 24.
1 2 70 71
Therefore, 24 of the 71 fractions , ,··· , , are irreducible.
72 72 72 72

Solution 2
First, we note that 72 = 23 × 32 .
a
For one of the fractions to be in lowest terms, then a and 72 have no common factors.
72
Since the only primes which are divisors of 72 are 2 and 3, then a and 72 have no common
factors when a is not divisible by 2 or 3.
Look at the first few fractions in the list:
1 2 3 4 5 6 7 8 9 10 11 12
, , , , , , , , , , ,
72 72 72 72 72 72 72 72 72 72 72 72
From this list, the ones which are in lowest terms are
1 5 7 11
, , ,
72 72 72 72
So the 1st and 5th of each of the two sets of 6 fractions above are in lowest terms.
72
This pattern will continue, so if we include the fraction (which we know is not in lowest
72
terms) at the end of the list, we obtain 12 sets of 6 fractions, and 2 fractions out of each set
will be in lowest terms, giving 12 × 2 = 24 fractions in lowest terms.
2005 COMC Solutions Page 5 of 26

(Why does this pattern continue? Each of the fractions in the list can be written in one
of the following forms:
6k + 1 6k + 2 6k + 3 6k + 4 6k + 5 6k + 6
, , , , ,
72 72 72 72 72 72
Since the numerators 6k + 2, 6k + 4 and 6k + 6 are divisible by 2 and the numerator 6k + 3 is
divisible by 3, then none of the fractions with these as numerators is in lowest terms.
Also, 6k + 1 and 6k + 5 are never divisible by 2 or 3, so these corresponding fractions are always
in lowest terms.
Thus, 2 out of each set of 6 fractions is in lowest terms.)

Answer: 24

5. An office building has 50 storeys, 25 of which are painted black and the other 25 of which are
painted gold. If the number of gold storeys in the top half of the building is added to the
number of black storeys in the bottom half of the building, the sum is 28. How many gold
storeys are there in the top half of the building?

Solution 1
Let G be the number of gold storeys in the top half of the building.
Then there are 25 − G black storeys in the top half of the building.
Since there are 25 black storeys in total, then the number of black storeys in the bottom half
of the building is 25 − (25 − G) = G.
Since the sum of the number of gold storeys in the top half of the building and the number of
black storeys in the bottom half of the building is 28, then G + G = 28, or G = 14.
Thus, there are 14 gold storeys in the top half of the building.

Solution 2
Let G and g be the number of gold storeys in the top and bottom halfs of the building, and
B and b the number of black storeys in the top and bottom halfs of the building.
Then G + B = 25 and g + b = 25, looking at the top and bottom halfs of the building.
Also, G + g = 25 and B + b = 25, since 25 of the storeys are painted in each colour.
Also, G + b = 28 from the given information, or b = 28 − G.
Since B + b = 25, then B + 28 − G = 25, so B = G − 3.
Since G + B = 25, then G + G − 3 = 25 or 2G = 28 or G = 14.
Thus, there are 14 gold storeys in the top half of the building.

Answer: 14
2005 COMC Solutions Page 6 of 26

6. In the grid shown, each row has a value as-

signed to it and each column has a value as- 3 0 5 6 −2
signed to it. The number in each cell is the −2 −5 0 1 y
sum of its row and column values. For exam- 5 2 x 8 0
ple, the “8” is the sum of the value assigned to 0 −3 2 3 −5
the 3rd row and the value assigned to the 4th −4 −7 −2 −1 −9
column. Determine the values of x and y.
Solution 1
First, we label the values assigned to the five columns A, B, C, D, E and the values assigned to
the five rows a, b, c, d, e.
0 1
Look at the sub-grid .
x 8
Since the 0 is in row 2 and column 3, then 0 = b + C.
Similarly, 1 = b + D, 8 = c + D and x = c + C.
But then 0 + 8 = (b + C) + (c + D) = (c + C) + (b + D) = x + 1 or x = 7.
1 y
In a similar way, we can show by looking at the sub-grid that we must have
8 0
1 + 0 = y + 8 or y = −7.
Thus, x = 7 and y = −7.
p q
(In fact, in any sub-grid of the form , we must have p + s = q + r.)
r s

Solution 2
First, we label the values assigned to the five columns A, B, C, D, E and the values assigned to
the five rows a, b, c, d, e.
Suppose that we try A = 0.
Looking at the “3” in the first row and first column, A + a = 3, so a = 3.
Since a = 3 and the entry in the first row and second column is 0, then a + B = 0, or B = −3.
Similarly, C = 2, D = 3 and E = −5.
Since A = 0 and the entry in the second row and first column is −2, then b + A = 0, then
b = −2.
Since y = b + E, then y = −2 + (−5) = −7.
Since A = 0 and the entry in the third row and first column is 5, then c + A = 5, so c = 5.
Since x = c + C, then x = 5 + 2 = 7.
Thus, x = 7 and y = −7.

Solution 3
First, we label the values assigned to the five columns A, B, C, D, E and the values assigned to
the five rows a, b, c, d, e.
If we choose five entries from the table which include one from each row and one from each
2005 COMC Solutions Page 7 of 26

column, then the sum of these entries is constant no matter how we choose the entries, as it is
always equal to A + B + C + D + E + a + b + c + d + e.
Here are three ways in which this can be done (looking at the bolded numbers):

3 0 5 6 −2 3 0 5 6 −2 3 0 5 6 −2
−2 −5 0 1 y −2 −5 0 1 y −2 −5 0 1 y
5 2 x 8 0 5 2 x 8 0 5 2 x 8 0
0 −3 2 3 −5 0 −3 2 3 −5 0 −3 2 3 −5
−4 −7 −2 −1 −9 −4 −7 −2 −1 −9 −4 −7 −2 −1 −9

Therefore, 3 + (−5) + 2 + 8 + (−9) = (−4) + (−3) + x + 1 + (−2) = 3 + y + 2 + (−2) + 3 or

−1 = x − 8 = y + 6.
Thus, x = 7 and y = −7.

Solution 4
First, we label the values assigned to the five columns A, B, C, D, E and the values assigned to
the five rows a, b, c, d, e.
Consider the first two entries in row 1.
We have 3 = A + a and 0 = B + a.
Subtracting these, we obtain 3 = 3 − 0 = (A + a) − (B + a) = A − B.
Notice that whenever we take entries in columns 1 and 2 from the same row, their difference
will always equal A − B, which is equal to 3.
Similarly, since the difference between the 0 and the 5 in the first row is 5, then every entry in
column 3 will be 5 greater than the entry in column 2 from the same row.
Thus, x = 2 + 5 = 7.
Also, since the difference between the 6 and the −2 in the first row is 8, then every entry in
column 5 will be 8 less than the entry in column 4 from the same row.
Thus, y = 1 − 8 = −7.
Therefore, x = 7 and y = −7.

Answer: x = 7 and y = −7
2005 COMC Solutions Page 8 of 26

7. In the diagram, the semi-circle has centre O and diameter AB. A ray of light leaves point P
in a direction perpendicular to AB. It bounces off the semi-circle at point D in such a way
that ∠P DO = ∠EDO. (In other words, the angle of incidence equals the angle of reflection
at D.) The ray DE then bounces off the circle in a similar way at E before finally hitting the
semicircle again at B. Determine ∠DOP .

A B
P O

Solution 1
Join D and E to O, and let ∠DOP = x.
Since DP ⊥ AB, then ∠P DO = 90◦ − x.
Since the angle of incidence equals the angle of reflection at D, then ∠EDO = ∠P DO = 90◦ −x.

E
90 x
90 x

D 90 x

90 x

x 90 x
A B
P O

Since DO and EO are both radii, then DO = EO, so 4EDO is isosceles, and so
∠DEO = ∠EDO = 90◦ − x.
Since the angle of incidence equals the angle of reflection at E, then ∠DEO = ∠BEO = 90◦ −x.
Since EO and BO are both radii, then EO = BO, so 4BEO is isosceles, and so
∠EBO = ∠BEO = 90◦ − x.
Consider quadrilateral P DEB.
We have ∠DP B = 90◦ , ∠P DE = (90◦ − x) + (90◦ − x) = 180◦ − 2x,
∠DEB = (90◦ − x) + (90◦ − x) = 180◦ − 2x, and ∠EBP = 90◦ − x.
Since the sum of the angles in the quadrilateral is 360◦ , then
90◦ + 180◦ − 2x + 180◦ − 2x + 90◦ − x = 360◦ or 540◦ − 5x = 360◦ or 5x = 180◦ or x = 36◦ .
Therefore, ∠DOP = x = 36◦ .

Solution 2
Join D and E to O, and let ∠DOP = x.
Since DP ⊥ AB, then ∠P DO = 90◦ − x.
Since the angle of incidence equals the angle of reflection at D, then ∠EDO = ∠P DO = 90◦ −x.
2005 COMC Solutions Page 9 of 26

E
90 x
90 x

D 90 x

90 x
2x
x 2x 90 x
A B
P O

Since DO and EO are both radii, then DO = EO, so 4EDO is isosceles, and so
∠DEO = ∠EDO = 90◦ − x. Also, ∠DOE = 180◦ − 2(90◦ − x) = 2x.
Since the angle of incidence equals the angle of reflection at E, then ∠DEO = ∠BEO = 90◦ −x.
Since EO and BO are both radii, then EO = BO, so 4BEO is isosceles, and so
∠EBO = ∠BEO = 90◦ − x. Also, ∠EOB = 180◦ − 2(90◦ − x) = 2x.
Since P OB is a straight line, then ∠P OD + ∠DOE + ∠EOB = 180◦ or x + 2x + 2x = 180◦ or
5x = 180◦ or x = 36◦ .
Therefore, ∠DOP = x = 36◦ .

Solution 3
Reflect the diagram across AB to complete the circle and form the pentagon DEBE 0 D0 . (Note
that DP D0 is a straight line since ∠DP O = ∠D0 P O = 90◦ .)

A O
B

Since DO, EO, BO, E 0 O and D0 O are all radii, then DO = EO = BO = E 0 O = D0 O.

Let ∠DOP = x. Since DP ⊥ AB, then ∠P DO = 90◦ − x.
By reflection, ∠D0 OP = ∠DOP = x, so ∠DOD0 = 2x.
Since the angle of incidence equals the angle of reflection at D, then ∠EDO = ∠P DO = 90◦ −x.
Since DO and EO are both radii, then DO = EO, so 4EDO is isosceles, and so
∠DEO = ∠EDO = 90◦ − x. Also, ∠DOE = 180◦ − 2(90◦ − x) = 2x.
Since the angle of incidence equals the angle of reflection at E, then ∠DEO = ∠BEO = 90◦ −x.
Since EO and BO are both radii, then EO = BO, so 4BEO is isosceles, and so
∠EBO = ∠BEO = 90◦ − x. Also, ∠EOB = 180◦ − 2(90◦ − x) = 2x.
2005 COMC Solutions Page 10 of 26

Therefore, the triangles DOE, EOB, BOE 0 , E 0 OD0 and D0 OD are all congruent by side-angle-
side. Therefore, pentagon DEBE 0 D0 is a regular pentagon.
Thus, ∠DOD0 = 51 (360◦ ) = 72◦ since the central angles of each of the five sides of the pentagon
are equal.
Since 4DOD0 is isosceles and OP is perpendicular to DD0 , then ∠P OD = 12 ∠DOD0 = 36◦ .
Thus, ∠P OD = 36◦ .

Answer: ∠DOP = 36◦

8. The number 18 is not the sum of any 2 consecutive positive integers, but is the sum of consec-
utive positive integers in at least 2 different ways, since 5 + 6 + 7 = 18 and 3 + 4 + 5 + 6 = 18.
Determine a positive integer less than 400 that is not the sum of any 11 consecutive positive
integers, but is the sum of consecutive positive integers in at least 11 different ways.

Solution
Suppose that the positive integer N is the sum of an odd number of consecutive integers, say
2k + 1 consecutive integers. Then for some integer a,

N = (a − k) + (a − (k − 1)) + · · · + (a − 1) + a + (a + 1) + · · · + (a + k) = (2k + 1)a

Thus, 2k + 1 is a divisor of N (ie. the number of integers in the representation is a divisor of N ).

Next, suppose that N is the sum of an even number of consecutive integers, say 2k consecutive
integers. Then for some integer b,
1
N = (b−k)+(b−(k−1))+· · ·+(b−1)+b+(b+1)+· · · (b+(k−1)) = 2kb−k = k(2b−1) = (2k)(2b−1)
2
Thus, k is a divisor of N and 2k is not a divisor of N (since 2b−1 is odd and so has no factor of 2).

We would like to find a positive integer N which is not the sum of 11 consecutive positive
integers (and so is not a multiple of 11) but is the sum of consecutive positive integers in 11
different ways.

Let’s consider the number of integers in each of the ways in which we write N as the sum
of consecutive integers. Note that if N is the sum of m consecutive positive integers, then N
is at least 1 + 2 + · · · + m. We make a table of what properties N must have for N to be the
sum of m consecutive integers for m = 2 to m = 10:
2005 COMC Solutions Page 11 of 26

m N at least Property of N
2 3 Divisible by 1, not by 2
3 6 Divisible by 3
4 10 Divisible by 2, not by 4
5 15 Divisible by 5
6 21 Divisible by 3, not by 6
(ie. divisible by 3, not by 2)
7 28 Divisible by 7
8 36 Divisible by 4, not by 8
9 45 Divisible by 9
10 55 Divisible by 5, not by 10
(ie. divisible by 5, not by 2)

How can we combine as many of these as possible? If we make N at least 55 and divisible
by 5, 7 and 9 and not divisible by 2, then N will be the sum of 2, 3, 5, 6, 7, 9 and 10 con-
secutive positive integers (7 representations in total). In this case, N must be divisible by
5 × 7 × 9 = 315. So following this line of thought, if N is less than 400, then we must have
N = 315. Now, 315 is also

• the sum of 15 consecutive positive integers since 315 is divisible by 15 and is at least 120
(ie. 1 + 2 + 3 + · · · + 15),
• the sum of 14 consecutive positive integers since 315 is divisible by 7, not by 14, and is at
least 105 (ie. 1 + 2 + 3 + · · · + 14)
• the sum of 18 consecutive positive integers since 315 is divisible by 9, not by 18, and is at
least 171 (ie. 1 + 2 + 3 + · · · + 18)
• the sum of 21 consecutive positive integers since 315 is divisible by 21 and is at least 231
(ie. 1 + 2 + 3 + · · · + 21)

So 315 is the sum of consecutive positive integers in at least 11 ways, and is not the sum
of 11 consecutive positive integers. (In fact, 315 is the unique answer, but we are not asked to
justify this.)

(Note: A good way to write a solution to this problem would be to first figure out in rough
that 315 was the answer, and then begin the solution by claiming that 315 is the answer. We
could then demonstrate that 315 works by showing that it can be represented in the correct
number of ways. While this approach is perfectly correct, it would not give much of a clue as
to how the answer was obtained.)

Answer: 315
2005 COMC Solutions Page 12 of 26

Part B

1. A line with slope −3 intersects the positive x-axis at A and the positive y-axis at B. A second
line intersects the x-axis at C(7, 0) and the y-axis at D. The lines intersect at E(3, 4).

D
E (3, 4)

x
O A C (7, 0)

(a) Find the slope of the line through C and E.

Solution
Since C has coordinates (7, 0) and E has coordinates (3, 4), then the slope of the line
through C and E is
0−4 −4
= = −1
7−3 4

(b) Find the equation of the line through C and E, and the coordinates of the point D.

Solution 1
Since the line through C and E has slope −1 and passes through the point (7, 0), then
the line has equation y − 0 = (−1)(x − 7) or y = −x + 7.
From the equation of the line, y = 7 is the y-intercept of the line.
Since D is the point where this line crosses the y-axis, then D has coordinates (0, 7).

Solution 2
Since the line through C and E has slope −1 and passes through the point (3, 4), then
the line has equation y − 4 = (−1)(x − 3) or y = −x + 7.
From the equation of the line, y = 7 is the y-intercept of the line.
Since D is the point where this line crosses the y-axis, then D has coordinates (0, 7).

(c) Find the equation of the line through A and B, and the coordinates of the point B.

Solution
Since the line through A and B has slope −3 and passes through the point E(3, 4), then
2005 COMC Solutions Page 13 of 26

the line has equation y − 4 = (−3)(x − 3) or y = −3x + 13.

From the equation of the line, y = 13 is the y-intercept of the line.
Since B is the point where this line crosses the y-axis, then B has coordinates (0, 13).

(d) Determine the area of the shaded region.

Solution 1
The area of the shaded region is the sum of the areas of 4DOC and 4BDE.
4DOC is right-angled at O, so the area of 4DOC is 21 (DO)(OC) = 12 (7)(7) = 49
2
.
We can consider 4BDE as having base BD of length 13 − 7 = 6 and height equal to the
distance of E from the y-axis (a distance of 3).
Therefore, the area of 4BDE is 21 (6)(3) = 9.
Thus, the area of the shaded region is 49
2
+ 9 = 67
2
.

Solution 2
The area of the shaded region is the sum of the areas of 4BOA and 4AEC.
4BOA is right-angled at O, so the area of 4BOA is 21 (BO)(OA).
Point A is the point where the line y = −3x + 13 crosses the x-axis, so it has x-coordinate
which satisfies −3x + 13 = 0, ie. x = 13
3
.
Therefore, the area of 4BOA is 12 (13) 13

169
3
= 6 .
We can consider 4AEC as having base AC of length 7 − 13 3
= 83 and height equal to the
distance of E from the x-axis (a distance of 4).
Therefore, the area of 4AEC is 21 (4) 83 = 16


3
.
Thus, the area of the shaded region is 6 + 3 = 201
169 16
6
= 67
2
.

Solution 3
Drop perpendiculars from E to point X on the x-axis and to point Y on the y-axis.
y

D
E (3, 4)
Y

x
O X A C (7, 0)

Then Y has coordinates (0, 4), X has coordinates (3, 0), and OXEY is a rectangle.
The area of the shaded region is thus the sum of the areas of 4BY E, rectangle OXEY
and 4EXC.
2005 COMC Solutions Page 14 of 26

Since 4BY E is right-angled at Y , its area is 12 (BY )(Y E) = 21 (13 − 4)(3 − 0) = 27

2
.
The area of rectangle OY EX is 3 × 4 = 12.
Since 4EXC is right-angled at X, its area is 21 (EX)(XC) = 12 (4 − 0)(7 − 3) = 8.
Therefore, the area of the shaded region is 27
2
+ 12 + 8 = 672
.

2. (a) Determine all possible ordered pairs (a, b) such that

a−b = 1
2a2 + ab − 3b2 = 22

Solution 1
Factoring the left side of the second equation, we get 2a2 + ab − 3b2 = (a − b)(2a + 3b).
Since a − b = 1, we get (1)(2a + 3b) = 22 or 2a + 3b = 22.
So we now have a − b = 1 and 2a + 3b = 22.
Adding 3 times the first equation to the second equation, we get 5a = 25 or a = 5.
Substituting back into the first equation, we get b = 4.
Thus, the only solution is (a, b) = (5, 4).

Solution 2
From the first equation, a = b + 1.
Substituting into the second equation, we obtain

2(b + 1)2 + (b + 1)(b) − 3b2 = 22

(2b2 + 4b + 2) + (b2 + b) − 3b2 = 22
5b = 20
b = 4

Substituting back into the first equation, we get a = 5, so the only solution is (a, b) = (5, 4).

Solution 3
From the first equation, b = a − 1.
Substituting into the second equation, we obtain

2a2 + a(a − 1) − 3(a − 1)2 = 22

2a2 + (a2 − a) − (3a2 − 6a + 3) = 22
5a = 25
a = 5

Substituting back into the first equation, we get b = 4, so the only solution is (a, b) = (5, 4).
2005 COMC Solutions Page 15 of 26

(b) Determine all possible ordered triples (x, y, z) such that

x2 − yz + xy + zx = 82
y 2 − zx + xy + yz = −18
z 2 − xy + zx + yz = 18

Solution 1
If we add the second equation to the third equation, we obtain

y 2 − zx + xy + yz + z 2 − xy + zx + yz = −18 + 18
y 2 + 2yz + z 2 = 0
(y + z)2 = 0
y+z = 0
z = −y

Substituting back into the three equations, we obtain

x2 + y 2 = 82
2xy = −18
−2xy = 18

Thus, x2 + y 2 = 82 and xy = −9.

Therefore, (x + y)2 = x2 + 2xy + y 2 = 82 + (−18) = 64, so x + y = ±8.

If x + y = 8, then y = 8 − x and so since xy = −9, then x(8 − x) = −9 or x2 − 8x − 9 = 0

or (x − 9)(x + 1) = 0 so x = 9 or x = −1.
Since x + y = 8, then if x = 9, we have y = −1 and z = −y = 1.
Since x + y = 8, then if x = −1, we have y = 9 and z = −y = −9.

If x+y = −8, then y = −8−x and so since xy = −9, then x(−8−x) = −9 or x2 +8x−9 = 0
or (x + 9)(x − 1) = 0 so x = −9 or x = 1.
Since x + y = −8, then if x = −9, we have y = 1 and z = −y = −1.
Since x + y = −8, then if x = 1, we have y = −9 and z = −y = 9.

Therefore, the four solutions are (x, y, z) = (9, −1, 1), (−1, 9, −9), (−9, 1, −1), (1, −9, 9).
2005 COMC Solutions Page 16 of 26

Solution 2
If we add the first equation to the second equation, we obtain

x2 − yz + xy + zx + y 2 − zx + xy + yz = 82 − 18
x2 + 2xy + y 2 = 64
(x + y)2 = 64
x + y = ±8

Similarly, adding the first equation to the third equation, we obtain x2 + 2xz + z 2 = 100
or x + z = ±10.
Also, adding the second equation to the third equation, we obtain y 2 + 2yz + z 2 = 0
or y + z = 0, and so z = −y.
Using x + z = ±10 and z = −y, we obtain x − y = ±10.
Thus, we have x + y = ±8 and x − y = ±10.

If x + y = 8 and x − y = 10, then adding these equations, we get 2x = 18 or x = 9

and so y = −1 and z = −y = 1.
If x + y = 8 and x − y = −10, then adding these equations, we get 2x = −2 or x = −1
and so y = 9 and z = −y = −9.
If x + y = −8 and x − y = 10, then adding these equations, we get 2x = 2 or x = 1 and
so y = −9 and z = −y = 9.
If x + y = −8 and x − y = −10, then adding these equations, we get 2x = −18 or x = −9
and so y = 1 and z = −y = −1.

Therefore, the four solutions are (x, y, z) = (9, −1, 1), (−1, 9, −9), (−9, 1, −1), (1, −9, 9).

3. Four tiles identical to the one shown, with a > b > 0, b

are arranged without overlap to form a square with
b
a square hole in the middle.
a
a

(a) If the outer square has area (a + b)2 , show that the area of the inner square
is (a − b)2 .

Solution 1
Each tile can be split into two right-angled triangles along a diagonal, each with legs of
lengths a and b.
The area of each of these triangles is 21 ab, so the area of each tile is ab.
2005 COMC Solutions Page 17 of 26

If the outer square has area (a + b)2 and this is partially covered with four tiles each of
area ab, then the area of the leftover portion (ie. the square hole) is

(a + b)2 − 4ab = a2 + 2ab + b2 − 4ab = a2 − 2ab + b2 = (a − b)2

Solution 2
If the outer square has area (a + b)2 , then the side length of the outer square is a + b.
In order to get a side length of a + b, we need to line up the “a” side of a tile with the “b”
side of a second tile, as shown.
b a

b b
a
a b

(Note that the tiles do fit together in this way, since each is a quadrilateral with two right
angles, so the remaining two angles add to 180◦ , that is, a straight line.)
We can complete the square as follows:
b a

b b
a

b
a
b
b
a b

Now the inner hole is clearly a rectangle (as it has four right angles) and is in fact a square
as its four sides are all of length a − b (as each of its sides are the remaining portion of a
line segment of length a when a segment of length b is cut off from one end).
Since the inner square has side length a − b, then its area is (a − b)2 .

(b) Determine the smallest integer value of N for which there are prime numbers a and b such
that the ratio of the area of the inner square to the area of the outer square is 1 : N .

Solution
(a − b)2
From (a), the ratio of the area of the inner square to the area of the outer square is .
(a + b)2
We would like to find integers N for which there are prime numbers a and b such
(a − b)2 1
that 2
= (and in fact find the minimum such N ).
(a + b) N
a−b 1
Taking the positive square root of both sides, we obtain =√ .
a+b N
2005 COMC Solutions Page 18 of 26

√
Since the left side is a rational number (since a and b are integers), then N must be
rational, so N must be a perfect square.
Suppose N = k 2 , for some positive integer k.
a−b 1
Thus, we have = or a + b = k(a − b) or (k − 1)a = (k + 1)b.
a+b k

Since we would like to find the smallest value of N which works, then we try to find
the smallest value of k which works.
Does k = 1 work? Are there prime numbers a and b so that 0 = 2b? No, since this means
b = 0.
Does k = 2 work? Are there prime numbers a and b so that a = 3b? No, since here a is a
multiple of 3, so the only possible prime value of a is 3, which would make b = 1, which
is not a prime.
Does k = 3 work? Are there prime numbers a and b so that 2a = 4b (ie. a = 2b)? No,
since here a is a multiple of 2, so the only possible prime value of a is 2, which would make
b = 1, which is not a prime.
Does k = 4 work? Are there prime numbers a and b so that 3a = 5b?
Yes: a = 5 and b = 3.
Therefore, the smallest value of k which works is k = 4, so the smallest value of N which
works is N = 16.

(c) Determine, with justification, all positive integers N for which there are odd integers
a > b > 0 such that the ratio of the area of the inner square to the area of the outer
square is 1 : N .

Solution
Suppose that N is a positive integer for which there are odd integers a > b > 0 such that
(a − b)2 1
2
= .
(a + b) N
Then, as in (b), N must be a perfect square, say N = k 2 , for some positive integer k.
Since a and b are odd, then set a = 2A + 1 and b = 2B + 1, for some integers A and B.
(2A − 2B)2 1 A−B 1
Thus we have 2
= 2 or = or k(A − B) = A + B + 1.
(2A + 2B + 2) k A+B+1 k
If A and B have the same parity (ie. both even or both odd), then A − B is even so the
left side is even and A + B + 1 is odd, so the right side is odd. Since we cannot have an
odd number equal to an even number, then this cannot happen.
Thus, A and B must have opposite parity (ie. one even and the other odd). In this case,
A − B is odd and A + B + 1 is even. Since k(A − B) = A + B + 1, then k is even.
Therefore, N must be an even perfect square.

We must now check if every even perfect square is a possible value for N .
2005 COMC Solutions Page 19 of 26

Suppose N = (2m)2 .
Using our substitutions from above, can we find integers A and B so that
2m(A − B) = A + B + 1?
If A = m and B = m − 1, then A − B = 1 and A + B + 1 = 2m, so 2m(A − B) = A + B + 1.
(a − b)2 1 1
So if a = 2A + 1 = 2m + 1 and b = 2B + 1 = 2m − 1, then 2
= 2
= .
(a + b) (2m) N
Therefore, the positive integers N which have the required property are all even perfect
squares.

4. Triangle ABC has its base on line segment P N and vertex A on line P M . Circles with centres
O and Q, having radii r1 and r2 , respectively, are tangent to the triangle ABC externally and
to each of P M and P N .
M

E
A
D Q
O L
K
P F B C G N

(a) Prove that the line through K and L bisects the perimeter of triangle ABC.

Solution
We must show that KB + BC + CL = KA + AL.
Since BK and BF are tangents to the left circle from the same point B, then BK = BF .
Since CL and CG are tangents to the right circle from the same point C, then CL = CG.
Since AK and AD are tangents to the left circle from the same point A, then AK = AD.
Since AL and AE are tangents to the right circle from the same point A, then AL = AE.
Therefore, KB + BC + CL = F B + BC + CG = F G and KA + AL = DA + AE = DE.
Now F G = P G − P F and DE = P E − P D.
Since P E and P G are tangents to the right circle from the same point P , then P E = P G.
Since P D and P F are tangents to the left circle from the same point P , then P D = P F .
Therefore, F G = P G − P F = P E − P D = DE, so KB + BC + CL = KA + AL, ie. the
line through K and L bisects the perimeter of triangle ABC.

(b) Let T be the point of contact of BC with the circle inscribed in triangle ABC.
Prove that (T C)(r1 ) + (T B)(r2 ) is equal to the area of triangle ABC.

Solution 1
2005 COMC Solutions Page 20 of 26

Let I be the centre of the circle inscribed in 4ABC, T be the point of contact of this
circle with BC, and r the radius of this circle.
Join O to K and B, and I to B and T .
M

D Q

O L
K I
P N
F B T C G

(Note that the circle with centre I is not necessarily tangent to AB at K or AC at L.)
Note that OK is perpendicular to KB and IT is perpendicular to BC.
Now OB bisects ∠F BK and IB bisects ∠KBC, since the circles with centres O and I
are tangent to F B and BK, and BA and BC, respectively.
Now ∠KOB = 90◦ − ∠KBO = 90◦ − 21 ∠F BK = 12 (180◦ − ∠F BK) = 21 ∠KBC = ∠IBT ,
so 4OKB is similar to 4BT I.
BK IT BK r (T B)(BK)
Therefore, = or = or r1 = .
KO TB r1 TB r
(T C)(LC)
Similarly, r2 = .
r
Therefore,

(T C)(T B)(BK) (T B)(T C)(LC) (T B)(T C)

(T C)(r1 ) + (T B)(r2 ) = + = (BK + LC)
r r r
Let BC = a, AB = c, AC = b, and let s be the semi-perimeter of 4ABC (that is, s is
half of the perimeter of 4ABC).
Now, from (a), since KB + BC + LC = s, then BK + LC = s − BC = s − a.
(T B)(T C)
Therefore, (T C)(r1 ) + (T B)(r2 ) = (s − a).
r

We can now focus entirely on 4ABC.

Let X and Y be the points where the circle with centre I is tangent to sides AB and AC,
respectively.
2005 COMC Solutions Page 21 of 26

Y
X I

B C
T

Using tangent arguments as in (a), we see that AX = AY , BX = BT and CY = CT .

Since AX + AY + BX + BT + CY + CT = 2s, then BT + AY + Y C = s,
so T B = s − (AY + Y C) = s − AC = s − b.
Similarly, T C = s − c.
(s − b)(s − c)(s − a) s(s − a)(s − b)(s − c)
Therefore, (T C)(r1 ) + (T B)(r2 ) = = .
r sr
Let |4ABC| denote the area of 4ABC.
Then s(s − a)(s − b)(s − c) = |4ABC|2 by Heron’s formula.
Also,
1 1 1 1
sr = r(AB + BC + AC) = (IX)(AB) + (IT )(BC) + (IY )(AC) (∗)
2 2 2 2
since IX, IT and IY are all radii of the circle with centre I.
Since IX, IT and IY are perpendicular to AB, BC and AC, respectively, then the three
terms on the right side of (∗) are the areas of 4IAB, 4IBC and 4ICA, respectively,
and so their sum is |4ABC|, ie. sr = |4ABC|.
|4ABC|2
Thus, (T C)(r1 ) + (T B)(r2 ) = = |4ABC|, as required.
|4ABC|

Solution 2
Join O to F and B and Q to C and G.
Since the circle with centre O is tangent to P B and AB at F and K, then OF is perpen-
dicular to P B and OB bisects ∠F BK.
Similarly, QG is perpendicular to CN and QC bisects ∠GCL.
Extend AB and AC through B and C, respectively, and construct the circle which is
tangent to AB extended, BC, and AC extended, and lies outside 4ABC. This circle is
called an excircle of 4ABC.
The centre of this excircle, which we label U , is on the angle bisector of the angle formed
by AB extended and BC, as the circle is tangent to these two lines, so U lies on OB
extended. Similarly, U lies on QC extended.
2005 COMC Solutions Page 22 of 26

D Q

O L
K
P V N
F B C G
Y
X
U

Let AB = c, AC = b, BC = a, let s be the semi-perimeter (that is, half of the perimeter)

of 4ABC, and let |4ABC| denote the area of 4ABC.
|4ABC|
Then the radius of the excircle, which we will denote rA is equal to . (See the
s−a
end of this solution for a proof of this fact.)
Let V be the point where the excircle is tangent to BC.
Then U V is perpendicular to BC.
Thus, 4OF B is similar to 4U V B and 4QGC is similar to 4U V C (since they have
opposite angles which are equal and right angles).
OF UV r1 rA QG UV r2 rA
This tells us that = or = and = or = .
FB VB FB VB GC VC GC VC
Therefore, since F B = BK and CG = CL and KB + BC + CL = s by (a), then

(V B)(r1 )+(V C)(r2 ) = rA (F B+CG) = rA (KB+LC) = rA (s−BC) = rA (s−a) = |4ABC|

Now suppose that the excircle is tangent to AB extended and AC extended at X and Y ,
respectively.
Then AX = AY , and AX = AB + BX = AB + BV and AY = AC + CY = AC + CV (by
equal tangents from B and C), so AX +AY = AB+AC+BV +V C = AB+AC+BC = 2s,
ie. AX = AY = s.
Thus, V B = BX = AX − AB = s − c and similarly V C = s − b.
But T B = s − b = V C and T C = s − c = V B (see Solution 1), so

|4ABC| = (T C)(r1 ) + (T B)(r2 )

as required.

|4ABC|
(Why is rA = ?
s−a
2005 COMC Solutions Page 23 of 26

Join X, Y and V to U . Note that ∠AXU = ∠AY U = ∠BV U = ∠CV U = 90◦ .

Then AXU Y is a shape of the same type as in Problem 3, so its area is equal to
AX · U X = srA .
Similarly, the areas of BV U X and CV U Y are rA (s − c) and rA (s − b).
Thus,

|4ABC| = Area of AXU Y − Area of BV U X − Area of CV U Y = rA (s − (s − c) − (s − b))

But, s − (s − c) − (s − b) = b + c − s = a + b + c − a − s = 2s − a − s = s − a, so
|4ABC| = rA (s − a), which is what we wanted to show.)

Solution 3
Let AB = c, AC = b, BC = a, and let s denote the semi-perimeter of 4ABC (that is,
half of its perimeter).
Then by (a), AK + AL = KB + BC + LC = s.
Since P M and P N are tangent to both circles, then the line through O and Q passes
through P .
Join O to D, F and K, and Q to L, E and G.
M

D Q

O L
K
P N
F B C G

In each case, the centre of a circle is being joined to a point where the circle is tangent to
a line, so creates a right angle.
PF PG
Therefore, 4P OF is similar to 4P QG, so = or r1 (P G) = r2 (P F ).
OF QG
Each of the shapes P DOF , ADOK, BF OK, AEQL, CGQL and P EQG has two right
angles and two pairs of equal sides (ie. each is a shape as in Problem 3).
The area of each of these shapes is the product of the lengths of two of the sides which
meet at a right angle.
We use |P EQG| to denote the area of the shape P EQG, and so on.
2005 COMC Solutions Page 24 of 26

Therefore,

|P EQG| = |4ABC| + |P DOF | + |ADOK| + |BF OK| + |AEQL| + |CGQL|

(P G)(QG) = |4ABC| + (P F )(OF ) + (AK)(OK) + (KB)(OK) + (AL)(QL) + (CL)(LQ)
|4ABC| = r2 (P G) − r2 (AL) − r2 (CL) − r1 (P F ) − r1 (AK) − r1 (KB)
= r2 (P G − AL − CL) + r1 (−P F − AK − KB)
= r2 (P G − CG − AL) + r1 (−P F − AB) (equal tangents)
= r2 (P C − AL) + r1 (−P F − AB)
= r2 (P F + F B + BC − AL) + r1 (−P F − AB)
= r2 (P F + BK + a − (s − AK)) + r1 (−P F − AB)
(since AK + AL = s, and BK = F B by equal tangents)
= r2 (P F + AK + BK + a − s) + r1 (−P F − AB)
= r2 (P F + AB + a − s) + r1 (−P F − AB)
= r2 (P F ) + r2 (c + a − s) + r1 (−P F − AB)
= r1 (P G) + r2 (a + b + c − b − s) + r1 (−P F − AB) (since r1 (P G) = r2 (P F ))
= r2 (2s − b − s) + r1 (P G − P F − AB)
= r2 (s − b) + r1 (GF − AB)
= r2 (s − b) + r1 (F B + BC + CG − c)
= r2 (s − b) + r1 (KB + BC + CL − c) (equal tangents)
= r2 (s − b) + r1 (s − c)

As in Solution 1, T B = s − c and T C = s − b.
Therefore, |4ABC| = (T C)(r1 ) + (T B)(r2 ), as required.

Solution 4
Let ∠ABC = 2β, ∠ACB = 2γ and ∠M P N = 2θ.
Then ∠P AB = 2β − 2θ and ∠M AC = 2γ + 2θ, using external angles in 4P AB and
4P AC. Also, ∠ABP = 180◦ − 2β.
Since the circle with centre O is tangent to AP and AK, then O lies on the bisector of
∠P AK, so ∠KAO = β − θ. Similarly, ∠LAQ = γ + θ and ∠KBO = 90◦ − β.
Since 4OKB is right-angled at K (since AB is tangent to the circle with centre O at K),
then ∠KOB = β.
KO KB
Thus, tan(∠KAO) = tan(β − θ) = and tan(∠KOB) = tan(β) = .
AK KO
2005 COMC Solutions Page 25 of 26

Therefore,

AB = AK + KB
KO
AB = + KO tan(β)
tan(β − θ)
 
1 + tan(β) tan(θ)
AB = r1 + tan(β) (since KO = r1 )
tan(β) − tan(θ)
1 + tan(β) tan(θ) tan2 (β) − tan(β) tan(θ)
 
AB = r1 +
tan(β) − tan(θ) tan(β) − tan(θ)
2
 
1 + tan (β)
AB = r1
tan(β) − tan(θ)
AB(tan(β) − tan(θ))
r1 =
1 + tan2 (β)
AB(tan(β) − tan(θ))
r1 =
sec2 (β)
r1 = AB sin(β) cos(β) − AB cos2 (β) tan(θ)
1
r1 = 2
AB sin(2β) − AB cos2 (β) tan(θ)

But AB sin(2β) is the length of the height, h, of 4ABC from A to BC.

Thus r1 = 21 h − AB cos2 (β) tan(θ).
Similarly, r2 = 21 h + AC cos2 (γ) tan(θ).

Since the circle with centre I is tangent to AB and BC, then I lies on the angle bi-
IT
sector of ∠ABC, so ∠IBT = β, so tan(β) = .
TB
IT r
Thus, T B = = .
tan(β) tan(β)
r
Similarly, T C = .
tan(γ)
Therefore,

r1 (T C) + r2 (T B) = T C 21 h − AB cos2 (β) tan(θ) + T B 21 h + AC cos2 (γ) tan(θ)

   

1
= 2
h(T C + T B) + tan(θ) [−T C · AB cos2 (β) + T B · AC cos2 (γ)] (∗)

The first term on the right side of (∗) equals 12 h(BC) which equals the area of 4ABC.
Considering the second factor of the second term, we obtain

T B · AC cos2 (γ) − T C · AB cos2 (β)

r r
= AC cos2 (γ) − AB cos2 (β)
tan(β) tan(γ)
r
2AC tan(γ) cos2 (γ) − 2AB tan(β) cos2 (β)


=
2 tan(β) tan(γ)
r
= (2AC sin(γ) cos(γ) − 2AB sin(β) cos(β))
2 tan(β) tan(γ)
r
= (AC sin(2γ) − AB sin(2β))
2 tan(β) tan(γ)
2005 COMC Solutions Page 26 of 26

But AC sin(2γ) = AB sin(2β) = h, so this second factor equals 0, so the second term of
the right side of (∗) equals 0.
Therefore, r1 (T C) + r2 (T B) equals the area of 4ABC, as required.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING
presents the

Canadian Open
Mathematics Challenge
Wednesday, November 22, 2006

Supported by:

Time: 2 12 hours 2006

c Canadian Mathematical Society

Calculators are NOT permitted.

Do not open this booklet until instructed to do so.

There are two parts to this paper.

PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value
for each question by entering the correct answer in the space provided. If you do not have
the correct answer, any work you do in obtaining an answer will be considered for part marks,
provided that it is done in the space allocated to that question in your answer booklet.
PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must
be written in the appropriate location in the answer booklet. Rough work should be done
separately. If you require extra pages for your finished solutions, paper will be provided by
your supervising teacher. Any extra papers should be placed inside your answer booklet. Be
sure to write your name and school name on any inserted pages.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution poorly
presented will not earn full marks.
NOTES:
At the completion of the contest, insert the information sheet inside the answer
booklet.
The names of top scoring competitors will be published on the Web sites of the
CMS and CEMC.
 Canadian Open Mathematics Challenge
NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations
√ and answers will be expressed as
exact numbers such as 4π, 2 + 7, etc.
4. Calculators are not allowed.

PART A
1 1 1 1





1. What is the value of 1 + 2 1+ 3 1+ 4 1+ 5 ?

2. If f (2x + 1) = (x − 12)(x + 13), what is the value of f (31)?

3. In 4ABC, M is the midpoint of BC, as shown. If ∠ABM = 15◦

and ∠AM C = 30◦ , what is the size of ∠BCA? A

B M C

4. Determine all solutions (x, y) to the system of equations

4 5
+ = 12
x y2
3 7
+ = 22
x y2

x+8
5. In 4ABC, BC = 4, AB = x, AC = x + 2, and cos(∠BAC) = .
2x + 4
Determine all possible values of x.

6. Determine the number of integers n that satisfy all three of the conditions below:
• each digit of n is either 1 or 0,
• n is divisible by 6, and
• 0 < n < 107 .

7. Suppose n and D are integers with n positive and 0 ≤ D ≤ 9.

n
Determine n if = 0.9D5 = 0.9D59D59D5 . . . .
810
8. What is the probability that 2 or more successive heads will occur at least once in
10 tosses of a fair coin?
 PART B
1. Piotr places numbers on a 3 by 3 grid using the following rule, called “Piotr’s
Principle”:

For any three adjacent numbers in a horizontal, vertical or diagonal

line, the middle number is always the average (mean) of its two
neighbours.

(a) Using Piotr’s principle, determine the missing numbers in the 3 19

grid to the right. (You should fill in the missing numbers in
8
the grid in your answer booklet.)

(b) Determine, with justification, the total of the nine numbers x

when the grid to the right is completed using Piotr’s
5 23
Principle.

(c) Determine, with justification, the values of x and y when the x 7

grid to the right is completed using Piotr’s Principle.
9 y
20

2. In the diagram, the circle x2 + y 2 = 25 intersects

the x-axis at points A and B. The line x = 11 y x = 11
intersects the x-axis at point C. Point P moves
along the line x = 11 above the x-axis and AP P
intersects the circle at Q. Q
(a) Determine the coordinates of P when
4AQB has maximum area. Justify your x
A B C
answer.

(b) Determine the coordinates of P when Q is

the midpoint of AP . Justify your answer.

(c) Determine the coordinates of P when the

area of 4AQB is 41 of the area of 4AP C.
Justify your answer.
3. (a) In the diagram, trapezoid ABCD has parallel sides
AB and DC of lengths 10 and 20, respectively. Also, A B
the length of AD is 6 and the length of BC is 8.
Determine the area of trapezoid ABCD.
D C
(b) In the diagram, P QRS is a rectangle and
T is the midpoint of RS. The inscribed R T S
circles of 4P T S and 4RT Q each have
radius 3. The inscribed circle of 4QP T
has radius 4. Determine the dimensions of
rectangle P QRS.

Q P
 Mathematics
Challenge

Canadian
(English)

Open

2006
p
4. (a) Determine, with justification, the fraction , where p and q are positive integers
q
and q < 100, that is closest to, but not equal to, 73 .
a c a+c
(b) The baseball sum of two rational numbers and is defined to be .
b d b+d
(A rational number is a fraction whose numerator and denominator are both
integers and whose denominator is not equal to 0.) Starting with the rational
numbers 01 and 11 as Stage 0, the baseball sum of each consecutive pair of
rational numbers in a stage is inserted between the pair to arrive at the next
stage. The first few stages of this process are shown below:
0 1
STAGE 0: 1 1

0 1 1
STAGE 1: 1 2 1

0 1 1 2 1
STAGE 2: 1 3 2 3 1

0 1 1 2 1 3 2 3 1
STAGE 3: 1 4 3 5 2 5 3 4 1
Prove that
(i) no rational number will be inserted more than once,
(ii) no inserted fraction is reducible, and
(iii) every rational number between 0 and 1 will be inserted in the pattern at
some stage.
The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

presents the

Canadian Open
Mathematics Challenge
Wednesday, November 22, 2006

Supported by:

Solutions

2006
c Canadian Mathematical Society
2006 COMC Solutions Page 2

Part A
1 1 1 1





1. What is the value of 1 + 2
1+ 3
1+ 4
1+ 5
?

Solution 1

1 1 1 1 3 4 5 6









1+ 2
1+ 3
1+ 4
1+ 5
= 2 3 4 5
    
=
3
4
5 6
2
3
4
5
(simplifying numerators and denominators)
6
= 2

= 3

Solution 2

1 1 1 1 3 4 5 6









1+ 2
1+ 3
1+ 4
1+ 5
= 2 3 4 5
360
= 120

= 3

2. If f (2x + 1) = (x − 12)(x + 13), what is the value of f (31)?

Solution 1
Since f (2x + 1) = (x − 12)(x + 13), then

f (31) = f (2(15) + 1) = (15 − 12)(15 + 13) = 3(28) = 84

Solution 2
w−1
If w = 2x + 1, then x = .
2
Since f (2x + 1) = (x − 12)(x + 13), then
     
w−1 w−1 w − 25 w + 25
f (w) = − 12 + 13 =
2 2 2 2

Therefore,   
31 − 25 31 + 25
f (31) = = 3(28) = 84
2 2
2006 COMC Solutions Page 3

3. In 4ABC, M is the midpoint of BC, as shown. If

A
∠ABM = 15◦ and ∠AM C = 30◦ , what is the size of
∠BCA?
B M C

Solution
Since ∠AM C = 30◦ , then ∠AM B = 180◦ − ∠AM C = 150◦ .
Since ∠ABM = 15◦ and ∠AM B = 150◦ , then ∠BAM = 180◦ − ∠ABM − ∠AM B = 15◦ .
Since ∠ABM = ∠BAM , then BM = M A.
Since BM = M A and BM = M C, then M A = M C, so ∠M AC = ∠M CA.
Thus, ∠M CA = 21 (180◦ − ∠AM C) = 75◦ .
Therefore, ∠BCA = ∠M CA = 75◦ .

4. Determine all solutions (x, y) to the system of equations

4 5
+ 2 = 12
x y
3 7
+ 2 = 22
x y
Solution 1
Subtracting 5 times the second equation from 7 times the first equation, we obtain
   
4 5 3 7
7 + −5 + = 7(12) − 5(22)
x y2 x y2
13
= −26
x
x = − 21

4 5 5
Substituting x = − 12 into the first equation, we obtain 1 + 2 = 12 or −8 + 2 = 12 or
−2 y y
5
= 20 or y 2 = 14 .
y2
Therefore, y = ± 12 .
Thus, the solutions are − 12 , 12 and − 12 , − 12 .

(We can check by substitution that both of these solutions work.)

2006 COMC Solutions Page 4

Solution 2
Subtracting 4 times the second equation from 3 times the first equation, we obtain
   
4 5 3 7
3 + −4 + = 3(12) − 4(22)
x y2 x y2
13
− 2 = −52
y
y 2 = 14
y = ± 12

4 5 4
Substituting y = ± 12 into the first equation, we obtain + 2 = 12 or + 20 = 12 or
x ± 12 x


4
= −8 or x = − 12 .
x
Thus, the solutions are − 21 , 12 and − 12 , − 12 .

(We can check by substitution that both of these solutions work.)

x+8
5. In 4ABC, BC = 4, AB = x, AC = x + 2, and cos(∠BAC) = .
2x + 4
Determine all possible values of x.

Solution
Using the cosine law in 4ABC,

BC 2 = AB 2 + AC 2 − 2(AB)(AC) cos(∠BAC)
x+8
42 = x2 + (x + 2)2 − 2x(x + 2)
2x + 4
2 2
16 = x + x + 4x + 4 − x(x + 8)
0 = x2 − 4x − 12
0 = (x − 6)(x + 2)

Therefore, x = 6 or x = −2.
Since AB = x, then x must be positive, so x = 6.

6. Determine the number of integers n that satisfy all three of the conditions below:

• each digit of n is either 1 or 0,

• n is divisible by 6, and
• 0 < n < 107 .

Solution 1
Since 0 < n < 107 , then n is a positive integer with fewer than 8 digits.
2006 COMC Solutions Page 5

Since n is divisible by 6, then n is even. Since each digit of n is either 1 or 0, then n must end
with a 0.
Since n is divisible by 6, then n is divisible by 3, so n has the sum of its digits divisible by 3.
Since each digit of n is 0 or 1 and n has at most 6 non-zero digits, then the sum of the digits
of n must be 3 or 6 (that is, n contains either 3 or 6 digits equal to 1).

Since n has at most 7 digits, we can write n in terms of its digits as abcdef 0, where each
of a, b, c, d, e, f can be 0 or 1. (We allow n to begin with a 0 in this representation.)
If n contains 6 digits equal to 1, then there is no choice in where the 1’s are placed so
n = 1111110.
If n contains 3 digits equal to 1, then 3 of!the 6 digits a through f are 1 (and the other 3 are
6
0). The number of such possibilities is = 20.
3
Therefore, there are 20 + 1 = 21 such integers n.

Solution 2
Since 0 < n < 107 , then n is a positive integer with fewer than 8 digits.
Since n is divisible by 6, then n is even. Since each digit of n is either 1 or 0, then n ends with
a 0.
Since n is divisible by 6, then n is divisible by 3, so has the sum of its digits divisible by 3.
Since each digit of n is 0 or 1 and n has at most 6 non-zero digits, then the sum of the digits
of n must be 3 or 6 (that is, n contains either 3 or 6 digits equal to 1).

If n contains 6 digits equal to 1, then n = 1111110, since n has at most 7 digits.

If n contains 3 digits equal to 1, then n has between 4 and 7 digits, and must begin with 1.
If n has 4 digits, then n must be 1110. !
3
If n has 5 digits, then n has the form 1abc0 with 2 of a, b, c equal to 1. There are =3
2
such possibilities. !
4
If n has 6 digits, then n has the form 1abcd0 with 2 of a, b, c, d equal to 1. There are =6
2
such possibilities.
If n!has 7 digits, then n has the form 1abcde0 with 2 of a, b, c, d, e equal to 1. There are
5
= 10 such possibilities.
2
Therefore, there are 1 + 1 + 3 + 6 + 10 = 21 possibilities for n.
2006 COMC Solutions Page 6

7. Suppose n and D are integers with n positive and 0 ≤ D ≤ 9.

n
Determine n if = 0.9D5 = 0.9D59D59D5 . . . .
810

Solution
9D5
First, we note that 0.9D5 = since
999
1000(0.9D5) = 9D5.9D5
1000(0.9D5) − 0.9D5 = 9D5.9D5 − 0.9D5
999(0.9D5) = 9D5
9D5
0.9D5 =
999
Alternatively, we could derive this result by noticing that

0.9D5 = 0.9D59D59D5 . . .
9D5 9D5 9D5
= + + + ···
103 106 109
9D5
= 103 (summing the infinite geometric series)
1
1− 3
10
9D5
=
1000 − 1
9D5
=
999
Therefore,
n 9D5
=
810 999
999n = 810(9D5)
111n = 90(9D5)
37n = 30(9D5)

Thus, 30(9D5) is divisible by 37. Since 30 is not divisible by 37 and 37 is prime, then 9D5
must be divisible by 37.
The multiples of 37 between 900 and 1000 are 925, 962 and 999.
Thus, 9D5 must be 925, so D = 2.
So 37n = 30(925) or n = 30(25) = 750.
2006 COMC Solutions Page 7

8. What is the probability that 2 or more successive heads will occur at least once in 10 tosses of
a fair coin?

Solution 1
For a given toss, we use T to represent a result of tails and H for heads.
There are 210 = 1024 possible sequences of outcomes when a fair coin is tossed 10 times.
Let tn be the number of sequences of n tosses of a fair coin which do not contain 2 or more
successive heads.
(So the number of sequences of length 10 that contain 2 or more successive heads is 1024 − t10
1024 − t10
which means that the desired probability is .)
1024

Note that t1 = 2, as both T and H do not contain 2 successive H’s.

Also, t2 = 3. (These are the sequences T T , T H, HT .)
Consider a sequence of n tosses of a fair coin which do not contain 2 or more successive heads,
where n ≥ 3.
Such a sequence must begin with H or T .
If the sequence begins with H, the second toss must be T , and the last n − 2 can be any
sequence of n − 2 tosses that does not contain 2 or more successive heads. There are tn−2 such
sequences of n − 2 tosses, so there are tn−2 sequences of n tosses beginning with H and not
containing 2 or more successive heads.
If the sequence begins with T , the last n − 1 tosses can be any sequence of n − 1 tosses that
does not contain 2 or more successive heads. There are tn−1 such sequences of n − 1 tosses, so
there are tn−1 sequences of n tosses beginning with T and not containing 2 or more successive
heads.
Therefore, tn = tn−1 + tn−2 , since each sequence begins with H or T .

Starting with t1 = 2 and t2 = 3, we can then generate the sequence tn for n = 1 to n = 10 by

adding the two previous terms to obtain the next term:

2, 3, 5, 8, 13, 21, 34, 55, 89, 144

1024 − 144 880 55
So t10 = 144, so the desired probability is = = .
1024 1024 64

Solution 2
For a given toss, we use T to represent a result of tails and H for heads.
There are 210 = 1024 possible sequences of outcomes when a fair coin is tossed 10 times.
Let us count the number of such sequences which do not contain 2 or more successive H’s, by
grouping them by the number of H’s that they contain. (Note that not containing 2 or more
successive H’s is equivalent to not containing the pair HH.)
2006 COMC Solutions Page 8

If a sequence of length 10 consists of 0 H and 10 T ’s, there is only !

1 possibility.
10
If a sequence of length 10 consists of 1 H and 9 T ’s, there are = 10 possibilities.
1

If a sequence of length 10 consists of 2 H’s and 8 T ’s, then we start with

T T T T T T T T

Each of the two H’s must be placed in separate spaces. We can then eliminate any unused
spaces to obtain a sequence of 8 T ’s and!2 H’s containing no consecutive H’s (and we get all
9
such sequences this way). There are = 36 ways of positioning the H’s, and so 36 such
2
sequences.
!
8
In a similar way, with 3 H’s and 7 T ’s, there are = 56 such sequences.
3
!
7
With 4 H’s and 6 T ’s, there are = 35 such sequences.
4
!
6
With 5 H’s and 5 T ’s, there are = 6 such sequences.
5

Therefore, there is a total of 1 + 10 + 36 + 56 + 35 + 6 = 144 sequences of 10 tosses which do

not contain 2 or more successive H’s.
Thus, there are 1024 − 144 = 880 sequences of 10 tosses which do contain 2 or more successive
H’s. Therefore, the probability that a given sequence contains 2 or more successive H’s is
880
1024
= 55
64
.
2006 COMC Solutions Page 9

Part B

1. Piotr places numbers on a 3 by 3 grid using the following rule, called “Piotr’s Principle”:

For any three adjacent numbers in a horizontal, vertical or diagonal line, the
middle number is always the average (mean) of its two neighbours.

(a) Using Piotr’s principle, determine the missing numbers

3 19
in the grid to the right. (You should fill in the missing
8
numbers in the grid in your answer booklet.)

Solution
Since the average of 3 and 19 is 12 (3 + 19) = 11, then 11 goes between the 3 and 19.
The number which goes below 8 is the number whose average with 3 is 8, so 13 goes
below 8.
The average of 13 and 19, or 16, goes in the middle square.
The number which goes to the right of the 16 is the number whose average with 8 is 16,
or 24.
The number which goes below 24 is the number whose average with 19 is 24, or 29.
The number which goes between 13 and 29 is their average, which is 21.
3 11 19
Therefore, the completed grid is 8 16 24 .
13 21 29
(We can check that each line obeys Piotr’s Principle.)

Note
There are other orders in which the squares can be filled.

(b) Determine, with justification, the total of the nine num-

x
bers when the grid to the right is completed using Piotr’s
5 23
Principle.

Note
When we have the three numbers a, X, b on a line, then X is the average of a and b, so
X = 12 (a + b).
When we have the three numbers a, b, X on a line, then b is the average of a and X, so
b = 12 (a + X) or 2b = a + X or X = 2b − a.
These facts will be useful as we solve (b) and (c).
2006 COMC Solutions Page 10

Solution 1
The average of 5 and 23 is 12 (5 + 23) = 14, which goes in the square between the 5 and 23.
Since the average of the numbers above and below the 5 equals 5, then their sum is
2(5) = 10. (Note that we do not need to know the actual numbers, only their sum.)
Similarly, the sum of the numbers above and below the 14 is 2(14) = 28 and the sum of
the numbers above and below the 23 is 2(23) = 46.
Therefore, the sum of the numbers in the whole grid is 5 + 10 + 14 + 28 + 23 + 46 = 126.

Solution 2
The average of 5 and 23 is 12 (5 + 23) = 14, which goes in the square between the 5 and 23.
Since the average of the x and the number below the 5 is 5, then the number below the 5
is 10 − x.
Since the average of the x and the bottom right number is 14, then the bottom right
number is 28 − x.
The average of 10 − x and 28 − x is 21 (10 − x + 28 − x) = 19 − x, which goes in the middle
square on the bottom row.
Since the average of 19 − x and the number above the 14 is 14, then the number above
the 14 is 2(14) − (19 − x) = 9 + x.
Since the average of 28 − x and the number above the 23 is 23, then the number above
the 23 is 2(23) − (28 − x) = 18 + x.
x 9 + x 18 + x
Thus, the completed grid is 5 14 23 and so the sum of the entries is
10 − x 19 − x 28 − x
x + 9 + x + 18 + x + 5 + 14 + 23 + 10 − x + 19 − x + 28 − x = 126.

(c) Determine, with justification, the values of x and y when

x 7
the grid to the right is completed using Piotr’s Principle.
9 y
20

Solution
The centre square is the average of 9 and y and is also the average of x and 20.
Comparing these facts, 21 (9 + y) = 21 (x + 20) or 9 + y = x + 20 or x − y = −11.
The number in the top right corner gives an average of 7 when combined with x (so equals
2(7) − x = 14 − x) and gives an average of y when combined with 20 (so equals 2y − 20).
Therefore, 14 − x = 2y − 20 or x + 2y = 34.
Subtracting the first equation from the second, we obtain 3y = 45 or y = 15.
Substituting back into the first equation, we obtain x = 4.
2006 COMC Solutions Page 11

4 7
We check by completing the grid. Starting with 9 15 gives, after some work,
20
4 7 10
9 12 15 , which does obey Piotr’s Principle.
14 17 20
Therefore, x = 4 and y = 15.

2. In the diagram, the circle x2 +y 2 = 25 intersects

y x = 11
the x-axis at points A and B. The line x = 11
intersects the x-axis at point C. Point P moves P
along the line x = 11 above the x-axis and AP Q
intersects the circle at Q.
x
A B C

(a) Determine the coordinates of P when 4AQB has maximum area. Justify your answer.

Solution
Since the circle has equation x2 + y 2 = 25, then to find the coordinates of A and B, the
x-intercepts of the circle, we set y = 0 to obtain x2 = 25 or x = ±5. Therefore, A and B
have coordinates (−5, 0) and (5, 0), respectively.
Since 4AQB has a base AB of constant length and a variable height, then the area of
4AQB is maximized when the height of 4AQB is maximized (that is, when Q is furthest
from AB).
To maximize the height of 4AQB, we would like Q to have as large a y-coordinate as
possible. Thus, we would like Q to be at the “top” of the circle – that is, at the place
where the circle intersects the y-axis.
Since the circle has equation x2 + y 2 = 25, then setting x = 0, we obtain y 2 = 25 or
y = ±5, so Q has coordinates (0, 5) as Q lies above the x-axis.
Therefore, P lies on the line through A(−5, 0) and Q(0, 5). This line has slope 1 and
y-intercept 5, so has equation y = x + 5.
Since P has x-coordinate 11 and lies on the line with equation y = x + 5, then P has
coordinates (11, 16).
2006 COMC Solutions Page 12

(b) Determine the coordinates of P when Q is the midpoint of AP . Justify your answer.

Solution
Suppose the coordinates of P are (11, p).
We will determine p so that the midpoint of P A lies on the circle. (This is equivalent to
finding P so that the point on the circle is the midpoint of P and A.)
Since A has coordinates (−5, 0), then for Q to be the midpoint of AP , Q must have coor-
dinates 21 (−5 + 11), 12 (0 + p) = (3, 21 p).

For (3, 12 p) to lie on the circle,

2
32 + 1
2
p = 25
1 2
4
p = 16
2
p = 64
p = ±8

Since P must lie above the x-axis, then p = 8.

Therefore, P has coordinates (11, 8).

1
(c) Determine the coordinates of P when the area of 4AQB is 4
of the area of 4AP C. Jus-
tify your answer.

Solution 1
Join Q to B.
y x = 11

P
Q

x
A B C

Since AB is a diameter of the circle, then ∠AQB = 90◦ .

Thus, since 4AQB and 4ACP are both right-angled and share a common angle at A,
then 4AQB and 4ACP are similar.
Since the area of 4ACP is 4 times the area of 4AQB, then the sides of 4ACP are
√
4 = 2 times as long as the corresponding sides of 4AQB.
Since AB = 10, then AP = 2AB = 20.
2006 COMC Solutions Page 13

We also know that AC = 16 (since C has coordinates (11, 0) and A has coordinates
(−5, 0)).
√ √
Therefore, by the Pythagorean Theorem, P C = AP 2 − AC 2 = 202 − 162 = 12.
Thus, P has coordinates (11, 12).

Solution 2
Let the coordinates of P be (11, p), and the coordinates of Q be (a, b). Thus, the height
of 4AQB is b.
The area of 4AQB is 21 (AB)(b) = 5b since AB = 10.
The area of 4AP C is 21 (AC)(p) = 8p since AC = 16.
Since the area of 4AQB is 14 that of 4AP C, then 5b = 2p or b = 25 p.
This tells us that Q must be 52 of the way along from A to P .
Since A has x-coordinate −5 and P has x-coordinate 11, then Q has x-coordinate
−5 + 25 (11 − (−5)) = 75 . Therefore, Q has coordinates 57 , 25 p .

Since the circle has equation x2 + y 2 = 25, then

7 2

2

2
5
+ 5
p = 25
49 4 2 625
25
+ 25
p = 25
2
4p = 576
p2 = 144

and so p = 12 since p > 0. Therefore, P has coordinates (11, 12).

3. (a) In the diagram, trapezoid ABCD has parallel

A B
sides AB and DC of lengths 10 and 20 and sides
AD and BC of lengths 6 and 8. Determine the
area of trapezoid ABCD. D C
Solution 1
Extend DA and CB to meet at P .
P

A B

D C

Since AB is parallel to DC, then ∠P AB = ∠P DC and ∠P BA = ∠P CD.

Therefore, 4P AB is similar to 4P DC.
2006 COMC Solutions Page 14

Since AB = 12 DC, then the sides of 4P AB are 21 the length of the corresponding sides of
4P DC.
Therefore, P A = AD = 6 and P B = BC = 8.
Thus, the sides of 4P DC have lengths 12, 16 and 20. Since 122 + 162 = 202 , then 4P DC
is right-angled at P by the Pythagorean Theorem.
Thus, the area of 4P DC is 21 (12)(16) = 96.
Since 4P AB is right-angled at P , its area is 12 (6)(8) = 24.
Therefore, the area of trapezoid ABCD is 96 − 24 = 72.

Solution 2
Drop perpendiculars from A and B to P and Q on DC.
A B

D P Q C

Let AP = BQ = h and let DP = x.

Since AB = 10 and ABQP is a rectangle, then P Q = 10.
Since DC has length 20, then QC = 20 − x − 10 = 10 − x.
Using the Pythagorean Theorem in 4DP A, we obtain x2 + h2 = 62 .
Using the Pythagorean Theorem in 4CQB, we obtain (10 − x)2 + h2 = 82 .
Subtracting the first of these equations from the second, we obtain 100 − 20x = 28 or
20x = 72 or x = 185
.

2 576
Substituting back into the first equation, h2 = 36 − 18
5
= 25 so h = 24
5
.
1 24


Therefore, the area of trapezoid ABCD is 2 5 (10 + 20) = 72.

Solution 3
Drop perpendiculars from A and B to P and Q on DC.
A B

D P Q C

Cut out rectangle ABQP and join the two remaining pieces along the cut line.
The remaining shape is a triangle DCX with side lengths DX = 6, XC = 8 and DC =
20 − 10 = 10. Since 62 + 82 = 102 , then 4DCX is right-angled by the Pythagorean
Theorem.
XC 8 4
Since sin(∠XDC) = = = , then the length of the altitude from X to DC is
DC 10 5
XD sin(∠XDC) = 6 45 = 24


5
2006 COMC Solutions Page 15

which is also the height of the original trapezoid.

(We could also determine the length of this altitude by calculating the area of 4XDC in
two ways: once as 21 (DX)(XC) = 24 and once as 21 h(DC) = 5h.)
Therefore, the area of the original trapezoid is 21 24


5
(10 + 20) = 72.
(Alternatively, the area of the original trapezoid is the sum of the areas of rectangle ABQP
( 24
5
× 10 = 48) and 4XDC (24), for a total of 72.)

Solution 4
Draw BX from B to X on DC so that BX is parallel to AD.
A B

D X C

Then ABXD is a parallelogram so BX = AD = 6 and DX = AB = 10.

Therefore, XC = DC − DX = 10.
Thus, 4BXC has sides of length 6, 8 and 10. Since 62 + 82 = 102 , then 4BXC is right-
angled at B by the Pythagorean Theorem. The area of 4BXC is thus 12 (6)(8) = 24.
Join B to D. BD divides the area of ABXD in half.
Also, BX divides the area of 4BDC in half, since it is a median.
Therefore, the areas of 4ABD, 4BDX and 4XBC are all equal.
So the area of trapezoid ABCD is 3(24) = 72.

Solution 5
Let X be the midpoint of DC. Join X to A and B.
Then AB = DX = XC = 10.
Since AB = DX and AB is parallel to DX, then AD and BX are parallel and equal, so
BX = 6.
Since AB = XC and AB is parallel to XC, then AX and BC are parallel and equal, so
AX = 6.
Therefore, the trapezoid is divided into three triangles, each of which has side lengths 6,
8 and 10.
A triangle with side lengths 6, 8 and 10 is right-angled (since 62 + 82 = 102 ), so has area
1
2
(6)(8) = 24.
Therefore, the area of the trapezoid is 3 × 24 = 72.
2006 COMC Solutions Page 16

(b) In the diagram, P QRS is a rectangle and

R T S
T is the midpoint of RS. The inscribed
circles of 4P T S and 4RT Q each have
radius 3. The inscribed circle of 4QP T
has radius 4. Determine the dimensions of
rectangle P QRS. Q P
Solution 1
Let RT = a (so RS = 2a) and RQ = b.
Drop a perpendicular from T to Z on QP . By symmetry, Z is also the point of tangency
of the middle circle to QP .
Let O be the centre of the circle inscribed in 4QT P and C be the centre of the circle
inscribed in 4RT Q.
Let A, B and D be the points of tangency of the circle with centre C to QR, RT and QT ,
respectively.
Let Y be the point of tangency of the circle with centre O to QT .
Join C to A, B and D, and O to Y . These radii are perpendicular to QR, RT , QT , and
QT , respectively.

R B T S
A C
Y
D
O

Q Z P

We know that OY = OZ = 4.
Since T Z = b, then T O = b − 4.
Since QZ = a, then QY = a (equal tangents).
Since CA = CB = 3 and RACB is a rectangle (as it has three right angles), then RACB
is a square and RA = RB = 3.
Therefore, AQ = b − 3 and BT = a − 3.
By equal tangents, T D = BT = a − 3 and QD = QA = b − 3.
Now, T Y = QT − QY = QD + T D − QY = (b − 3) + (a − 3) − a = b − 6.
Therefore, 4T OY is right-angled at Y with sides of length T O = b − 4, T Y = b − 6 and
OY = 4.
By the Pythagorean Theorem, 42 + (b − 6)2 = (b − 4)2 or 4b = 36 or b = 9.
OY 4
Therefore, T Y = 9 − 6 = 3 and tan(∠OT Y ) = = .
TY 3
4 a
Also, 3 = tan(∠QT Z) = .
b
Since b = 9, a = 12.
2006 COMC Solutions Page 17

Therefore, the rectangle is 24 by 9.

Solution 2
Let RT = a (so RS = 2a) and RQ = b.
Drop a perpendicular from T to Z on QP . By symmetry, Z is also the point of tangency
of the middle circle to QP , and QZ = a.
Since the incircle of 4QRT has radius 3, then so does the incircle of 4QZT .
Let O be the centre of the circle inscribed in 4QT P and C 0 the centre of the circle in-
scribed in 4QZT .
Since C 0 and O both lie on the angle bisector of ∠T QP , then tan(∠C 0 QP ) = tan(∠OQP ).
3 4
Since C 0 is 3 units from the line T Z, then tan(∠C 0 QP ) = and tan(∠OQP ) = so
a−3 a
3 4
= or 3a = 4a − 12 or a = 12.
a−3 a
We can calculate b = 9 as in Solution 1, to obtain that the rectangle is 24 by 9.

Solution 3
Let RT = a (so RS = 2a) and RQ = b.
We calculate the areas of each of 4QRT and 4QT P in two ways: once using the stan-
dard 12 bh formula and once using the less well-known Area = sr formula, where s is the
semi-perimeter of the triangle (that is, half of the perimeter) and r is the inradius (that
is, the radius of the inscribed circle).
√
In 4QRT , RT = a, RQ = b, QT = a2 + b2 and the inradius is 3, so

1
√
2
ab = 12 (a + b + a2 + b2 )(3)
√
In 4QT P , QT = T P = a2 + b2 , QP = 2a, the height is b, and the inradius is 4, so

1
√
2
(2a)b = 12 (2 a2 + b2 + 2a)(4)

Simplifying these equations, we get the system of equations

√
ab = 3(a + b) + 3 a2 + b2
√
ab = 4a + 4 a2 + b2
√
Eliminating the ab terms, we obtain 3b − a = a2 + b2 .
Squaring both sides, we obtain 9b2 − 6ab + a2 = a2 + b2 or 8b2 − 6ab = 0.
a 4
Since b 6= 0, = .
b 3
2006 COMC Solutions Page 18

Substituting a = 43 b into the first equation yields

q
4 2
3
b = 4b + 3b + 3 16
9
b2 + b2
√
4 2
3
b = 7b + 25b2
4 2
3
b = 7b + 5b (since b > 0)
2
4b = 36b
b = 9

since b 6= 0. Therefore, a = 12 and the rectangle is 24 by 9.

Solution 3
We use the notation and diagram from Solution 1.
Since RS and QP are parallel, then ∠BT D = ∠Y QZ.
Since C and O are the centres of inscribed circles, then C lies on the angle bisector of
∠BT D and O lies on the angle bisector of ∠Y QZ.
Therefore, ∠BT C = 12 ∠BT D = 12 ∠Y QZ = ∠OQZ.
Therefore, 4BT C and 4ZQO are similar, as each is right-angled.
Thus, BT : QZ = BC : OZ = 3 : 4.
Suppose that BT = 3x and QZ = 4x.
Then RT = RB + BT = AC + 3x = 3 + 3x, since RBCA is a square.
But RT = QZ so 4x = 3 + 3x or x = 3.
Let QA = y.
Then QT = QD + DT = QA + BT by equal tangents, so QT = y + 3x = y + 9.
Since 4QRT is right-angled, then

QR2 + RT 2 = QT 2
(y + 3)2 + 122 = (y + 9)2
y 2 + 6y + 9 + 144 = y 2 + 18y + 81
12y = 72
y = 6

Therefore, since RT = 3 + 3x, then RS = 24 and RQ = 3 + y = 9, so the rectangle is 24

by 9.
2006 COMC Solutions Page 19

p
4. (a) Determine, with justification, the fraction , where p and q are positive integers and
q
q < 100, that is closest to, but not equal to, 37 .

Solution
We would like to find positive integers p and q with q < 100 which minimizes
   
 p 3   7p − 3q  |7p − 3q|
 − =
 q 7   7q  =

7q

To minimize such a fraction, we would like to make the numerator small while making the
denominator large.
Since the two fractions pq and 73 are not equal, the numerator of their difference cannot be
0. Since the numerator is a positive integer, its minimum possible value is 1.
We consider the largest possible values of q (starting with 99) and determine if 7p − 3q
can possibly be equal to 1 or −1.

If q = 99, 7p − 3q = 7p − 297, which cannot equal ±1 since the nearest multiple of 7

to 297 is 294.
If q = 98, 7p − 3q = 7p − 294, which cannot equal ±1 since 7p − 294 is always divisible
by 7.
If q = 97, 7p − 3q = 7p − 291, which cannot equal ±1 since the nearest multiple of 7 to
291 is 294.
If q = 96, 7p − 3q = 7p − 288, which equals −1 if p = 41.

3 p 1
If p = 41 and q = 96, the difference between the fractions and is .
7 q 7(96)
|7p − 3q|
If q > 96, the numerator of is always at least 2, so the difference is at least
7q
2 1
> .
7(99) 7(96)
p 3 1 1
If q < 96, the difference between and is at least > .
q 7 7(95) 7(96)
41
So 96
minimizes the difference, so it is the closest fraction to 37 under the given conditions.
2006 COMC Solutions Page 20

a c a+c
(b) The baseball sum of two rational numbers and is defined to be . (A rational
b d b+d
number is a fraction whose numerator and denominator are both integers and whose de-
nominator is not equal to 0.) Starting with the rational numbers 01 and 11 as Stage 0, the
baseball sum of each consecutive pair of rational numbers in a stage is inserted between
the pair to arrive at the next stage. The first few stages of this process are shown below:
0 1
STAGE 0: 1 1

0 1 1
STAGE 1: 1 2 1

0 1 1 2 1
STAGE 2: 1 3 2 3 1

0 1 1 2 1 3 2 3 1
STAGE 3: 1 4 3 5 2 5 3 4 1

Prove that
(i) no rational number will be inserted more than once,

Solution
a c
Consider two rational numbers and which occur next to each other at a given
b d
a c
stage with < .
b d
Note that this means that ad < bc or bc − ad > 0.
a+c
The rational number that will be inserted between them at the next stage is .
b+d
Now
a a+c
< ⇔ a(b + d) < b(a + c) ⇔ 0 < bc − ad
b b+d
which we know to be true, and
a+c c
< ⇔ d(a + c) < c(b + d) ⇔ 0 < bc − ad
b+d d
which is again true.
a a+c c
Therefore, < < .
b b+d d
This tells us that every rational number which is inserted at any given stage is strictly
between the two rational numbers on either side.
Therefore, once a given rational number is inserted, every other rational number which
is inserted must be either strictly larger or strictly smaller, as the list at each stage
must be strictly increasing.
Therefore, no rational number will be inserted more than once.
2006 COMC Solutions Page 21

(ii) no inserted fraction is reducible,

Solution
First, we prove a lemma.

Lemma
a c
If < are consecutive rational numbers in a given stage, then bc − ad = 1.
b d

Proof
At Stage 0, the two fractions obey this property.
Assume that the property holds for all fractions in Stage k.
a c
Consider two consecutive fractions < at Stage k.
b d
a c a+c
The fraction that will be inserted between and at Stage k + 1 is ,
b d b+d
a a+c c
giving < < .
b b+d d
Note that b(a + c) − a(b + d) = bc − ad = 1 and c(b + d) − d(a + c) = bc − ad = 1.
This tells us that each pair of consecutive fraction at Stage k + 1 obeys this property.
Therefore, by induction, the required property holds.

kp a c
Suppose then that a fraction (k, p, q ∈ Z+ ) is inserted between and .
kq b d
By the Lemma, we must have b(kp) − a(kq) = 1 or k(bp − aq) = 1.
Therefore, k divides 1, so k = 1.
Thus, any inserted fraction can only have a common factor of 1 between its numerator
and denominator, so is irreducible. Thus, no inserted fraction is reducible.

(iii) every rational number between 0 and 1 will be inserted in the pattern at some stage.

Solution 1
1 n−1
We note first that every rational number of the forms and for n ≥ 2 do enter
n n
the pattern as the first and last new entry in Stage n − 1. (These rational numbers
0 1 n−2 1
enter between and , and and , respectively.)
1 n−1 n−1 1

Assume that there are rational numbers between 0 and 1 which are not inserted
in the pattern at the some stage.

p
Suppose that with p, q ∈ Z+ and gcd(p, q) = 1 is such a rational number with
q
minimal denominator. (Note that all irreducible fractions with denominators 1, 2, 3
are inserted already.)
2006 COMC Solutions Page 22

Since gcd(p, q) = 1, the Diophantine equation py − qx = 1 has solutions.

In fact, this Diophantine equation has a unique solution with 0 ≤ y < q. We can say
further that 0 < y < q since q > 1. (If y = 0, we have −qx = 1 so q would have to
be 1.)
Since 0 < y < q and qx = py − 1, then 0 < qx < pq so 0 < x < p as well.

Suppose that (x, y) = (a, b) is this unique solution.

a p
Note that pb − qa = 1 so pb > qa so < .
b q
p−a
Consider the fraction .
q−b
Its numerator and denominator are each a positive integer since 0 < a < p and
0 < b < q. Also, note that b(p − a) − a(q − b) = bp − aq = 1.
a p−a
Among other things, this tells us that < .
b q−b
a p−a p
If we can prove that and are consecutive at some stage, then will be
b q−b q
inserted between them at the next stage.

a p−a
Since each of and has a denominator less than q, it appears in the pat-
b q−b
p
tern as has the smallest denominator among those fractions which do not appear.
q
a 0
Note next that cannot be .
b 1
(If it was, a = 0 so bp − aq = 1 gives bp = 1 so p = 1.
We know that every fraction with p = 1 enters the pattern, so p 6= 1.)
p−a 1
Also, cannot be .
q−b 1
(If it was, then p − a = q − b so b(p − a) − a(q − b) = 1 gives (b − a)(q − b) = 1 so
b − a = 1.
Since p − a = q − b then q − p = b − a = 1.
n−1
But every fraction of the form enters the pattern, so q − p 6= 1.)
n
a p−a
This tells us that each of and actually entered the pattern at some stage.
b q−b

There are now three cases: b < q − b, b > q − b and b = q − b.

• Assume that b < q − b.
p−a
Consider the point when the fraction was inserted into the pattern.
q−b
p−a m M
Suppose that was inserted immediately between < .
q−b n N
2006 COMC Solutions Page 23

(That is, m + M = p − a and n + N = q − b).

Note that 0 < n < q − b. (n and N cannot be 0 since denominators cannot be 0.)

m p−a
Since < are consecutive fractions at this stage, we must also have
n q−b
(p − a)n − (q − b)m = 1.

But since gcd(p − a, q − b) = 1, the Diophantine equation (p − a)Y − (q − b)X = 1

must have a unique solution with 0 < Y < q − b.
But (X, Y ) = (a, b) is such a solution since 0 < b < q −b and b(p−a)−a(q −b) = 1
(from above).
a p−a
Therefore, must be the fraction immediately to the left of at the stage
b q−b
p−a p
where enters, which means that will be inserted into the pattern, contra-
q−b q
dicting our assumption.

• Assume that q − b < b.

a
Consider the point when the fraction was inserted into the pattern.
b
a m M
Suppose that was inserted immediately between < (that is, m + M = a
b n N
and n + N = b).
Note that 0 < n < b. (n and N cannot be 0 since denominators cannot be 0.)

m a
Since < are consecutive fractions at this stage, we must also have
n b
an − bm = 1.

But since gcd(a, b) = 1, the Diophantine equation aY − bX = 1 must have a

unique solution with 0 < Y < b.
But (X, Y ) = (p − a, q − b) is such a solution since 0 < q − b < b and
b(p − a) − a(q − b) = 1 (from above).
p−a a
Therefore, must be the fraction immediately to the left of at the stage
q−b b
a p
where enters, which means that will be inserted into the pattern, contradict-
b q
ing our assumption.

• Assume that q − b = b.
In this case, q = 2b.
But bp − aq = 1 so b(p − 2a) = 1, and so b = 1 giving q = 2.
But we know that every irreducible fraction with denominator 2 does enter.
(Namely, the fraction 12 .)
So this case cannot occur.
2006 COMC Solutions Page 24

Thus, every rational number between 0 and 1 will be inserted in the pattern at some
stage.

Solution 2
a c
Suppose and are consecutive (irreducible) fractions at some stage.
b a c d
Define S , = a + b + c + d, the sum of the numerators and denominators of the
b d
consecutive fractions.
We consider the minimum value of S at a given stage.
a+c a a+c c
When is inserted with < < , the two new sums are
b+d b b+d d
 
a a+c
S , = a + b + a + c + b + d = 2a + 2b + c + d
b b+d

and  
a+c c
S , = a + c + b + d + c + d = a + b + 2c + 2d
b+d d
a c 
each of which is larger than S , .
b d
So the minimum value of these sums must increase from one stage to the next.

a
Suppose that the fraction between 0 and 1 with a, b ∈ Z+ and gcd(a, b) = 1 is
b
never inserted into the pattern.
a
At any given Stage, since the fraction does not occur, it must be strictly between
b
m1 a m2 a
two consecutive fractions, say < < . (Since never occurs, we must be able
n1 b n2 b
to find such a pair of fractions at every Stage.)
(We know that m2 n1 − n2 m1 = 1 from (b).)
Thus, m1 b < n1 a and n2 a < m2 b. Since each of these quantities is a positive integer,
n1 a − m1 b ≥ 1 and m2 b − n2 a ≥ 1.
Now

m2 + n2 + m1 + n1
= (m2 + n2 )(1) + (m1 + n1 )(1)
≤ (m2 + n2 )(n1 a − m1 b) + (m1 + n1 )(m2 b − n2 a)
= m2 n1 a + n1 n2 a − m1 m2 b − m1 n2 b + m1 m2 b + m2 n1 b − m1 n2 a − n1 n2 a
= a(m2 n1 − m1 n2 ) + b(m2 n1 − m1 n2 )
= a+b
a
But for a fixed fraction , a + b is fixed and the minimum possible value of
b
m2 + n2 + m1 + n1
2006 COMC Solutions Page 25

increases from one stage to the next, so eventually a + b < m2 + n2 + m1 + n1 , a

contradiction, since once this happens we cannot find two fractions in that Stage
a
between which to put .
b
That is, there will be a stage beyond which we cannot find two consecutive fractions
a a
with between them. This means that must actually occur in the pattern.
b b
Therefore, every rational number between 0 and 1 will be inserted in the pattern at
some stage.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING
presents the

Sun Life Financial Canadian

Open Mathematics Challenge

Wednesday, November 21, 2007

Time: 2 12 hours 2007

c Canadian Mathematical Society
Calculators are NOT permitted.
Do not open this booklet until instructed to do so.
There are two parts to this paper.
PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value
for each question by entering the correct answer(s) in the space provided. If your answer is
incorrect, any work that you do will be considered for part marks, provided that it is done
in the space allocated to that question in your answer booklet.
PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must
be written in the appropriate location in the answer booklet. Rough work should be done
separately. If you require extra pages for your finished solutions, paper will be provided by
your supervising teacher. Any extra papers should be placed inside your answer booklet. Be
sure to write your name and school name on any inserted pages.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution poorly
presented will not earn full marks.
NOTES:
At the completion of the contest, insert the information sheet inside the answer
booklet.
The names of top scoring competitors will be published on the Web sites of the
CMS and CEMC.
Sun Life Financial Canadian Open Mathematics Challenge
NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations
√ and answers will be expressed as
exact numbers such as 4π, 2 + 7, etc., rather than as 12.566 . . .
or 4.646 . . ..
4. Calculators are not allowed.

PART A
1. If a = 15 and b = −9, what is the value of a2 + 2ab + b2 ?

2. A circular wind power generator turns at a rate of 30 complete revolutions per

minute. Through how many degrees does it turn in one second? y
3. In the diagram, ABCD is a rectangle with A on the
line y = x + 10, B on the line y = −2x + 10, and
C and D on the x-axis. If AD = 4, what is the area
of rectangle ABCD?

A B

x
D C

4. In June, the ratio of boys to girls in a school was 3 : 2. In September, there were
80 fewer boys and 20 fewer girls in the school and the ratio of boys to girls was 7 : 5.
What was the total number of students at the school in June?

5. The numbers 1, 2, 3, . . . , 9 are placed in a square array. The sum of the three rows,
the sum of the three columns, and the sum of the two diagonals are added together
to form a “grand sum”, S.
For example, if the numbers are placed as shown, the grand sum is

1 2 3 S = row sums + column sums + diagonal sums

4 5 6 = 45 + 45 + 30
7 8 9 = 120 .

What is the maximum possible value of the grand sum S?

6. In the diagram, O is the centre of the circle, AN is tangent to

the circle at A, P lies on the circle, and P N is perpendicular
to AN . If AN = 15 and P N = 9, determine the radius of O
the circle.
P

A N
7. Determine all ordered triples of real numbers, (x, y, z), that satisfy the system of
equations

xy = z 2
x+y+z = 7
x + y 2 + z 2 = 133 .
2

8. In the diagram, there are 28 line segments of length 1 arranged as shown to form
9 squares. There are various routes from A to B travelling along the segments so
that no segment is travelled more than once. Of these possible routes, determine

• the length of route that occurs the most often, and

• the number of different routes of this length.

A
B

PART B
1. An arithmetic sequence a, a + d, a + 2d, . . . is a sequence in which successive terms
have a common difference d. For example, 2, 5, 8, . . . is an arithmetic sequence with
common difference d = 3 because 5 − 2 = 8 − 5 = 3.

(a) If x − 1, 2x + 2 and 7x + 1 are the first three terms of an arithmetic sequence,

determine the value of x.
(b) For the value of x from (a), what is the middle term of the arithmetic sequence
x − 1, 2x + 2, 7x + 1, . . . , 72?

A geometric sequence a, ar, ar2 , . . . is a sequence in which successive terms have a

common ratio r. For example, the sequence 2, 10, 50, . . . is a geometric sequence with
10 50
common ratio r = 5 because = = 5.
2 10
(c) If y − 1, 2y + 2 and 7y + 1 are the first three terms of a geometric sequence,
determine all possible values of y.
(d) For each of the values of y from (c), determine the 6th term of the geometric
sequence y − 1, 2y + 2, 7y + 1, . . . .

2. In the diagram, ∠ABC = ∠BCD = 90◦ . Also, AB = 9, BC = 24 and CD = 18.

The diagonals AC and BD of quadrilateral ABCD meet at E.

(a) Determine the area of the quadrilateral ABCD.

(b) Show that the ratio DE : EB = 2 : 1. 9 A

B
(c) Determine the area of triangle DEC.
E
(d) Determine the area of triangle DAE.
24

C 18 D
 Canadian Open
Mathematics
Challenge

Financial
(English)

Sun Life
2007
3. Alphonse and Beryl are back! They are playing a two person game with the following
rules:

• Initially there is a pile of N stones, with N ≥ 2.

• The players alternate turns, with Alphonse going first. On his first
turn, Alphonse must remove at least 1 and at most N − 1 stones from
the pile.
• If a player removes k stones on their turn, then the other player must
remove at least 1 and at most 2k − 1 stones on their next turn.
• The player who removes the last stone wins the game.

(a) Determine who should win the game when N = 7, and explain the winning
strategy.
(b) Determine who should win the game when N = 8, and explain the winning
strategy.
(c) Determine all values of N for which Beryl has a winning strategy. Explain this
strategy.

4. A cat is located at C, 60 metres directly west of a mouse located at M . The mouse

is trying to escape by running at 7 m/s in a direction 30◦ east of north. The cat, an
expert in geometry, runs at 13 m/s in a suitable straight line path that will intercept
the mouse as quickly as possible.
(a) If t is the length of time, in seconds, that it
takes the cat to catch the mouse, determine N
the value of t.

(b) Suppose that the mouse instead chooses a

different direction to try to escape. Show that
no matter which direction it runs, all points of
13 m/s 30
interception lie on a circle.
7 m/s

(c) Suppose that the mouse is intercepted after W 60 m M E

C
running a distance of d1 metres in a particular
direction. If the mouse would have been S
intercepted after it had run a distance of d2
metres in the
√ opposite direction, show that
d1 + d2 ≥ 14 30.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

presents the

Sun Life Financial

Canadian Open Mathematics Challenge
Wednesday, November 21, 2007

Solutions

2007
c Canadian Mathematical Society
2007 COMC Solutions Page 2

Part A

1. Solution 1
If a = 15 and b = −9, then

a2 + 2ab + b2 = (a + b)2 = (15 + (−9))2 = 62 = 36

Solution 2
If a = 15 and b = −9, then

a2 + 2ab + b2 = 152 + 2(15)(−9) + (−9)2 = 225 − 270 + 81 = 36

Answer: 36

30 1
2. Since there are 60 seconds in a minute, the wind power generator turns = of a revolution
60 2
each second.
Since a full revolution is 360◦ , then the generator turns 21 (360◦ ) = 180◦ each second.
(Alternatively, the generator turns through 30×360◦ in one minute, so through 30×360◦ ÷60 =
180◦ in one second.)

Answer: 180

3. Since AD = 4 and AD is perpendicular to the x-axis, then A has y-coordinate 4.

Suppose that the coordinates of A are (a, 4). (This tells us also that D has coordinates (a, 0).)
Since A lies on the line y = x + 10, then 4 = a + 10, or a = −6.
Therefore, A has coordinates (−6, 4) and D has coordinates (−6, 0).
Since ABCD is a rectangle, then AB is parallel to the x-axis, so B has y-coordinate 4.
Suppose that the coordinates of B are (b, 4). (This tells us also that C has coordinates (b, 0)
since BC is perpendicular to the x-axis.)
Since B lies on the line y = −2x + 10, then 4 = −2b + 10 so 2b = 6 or b = 3.
Therefore, B has coordinates (3, 4) and C has coordinates (3, 0).
Now the height of rectangle ABCD equals the length of AD, so is 4.
The width of rectangle ABCD equals the length of CD, which is 3 − (−6) = 9.
Therefore, the area of rectangle is 9 × 4 = 36.

Answer: 36

4. Solution 1
Suppose that there were 3k boys and 2k girls in the school in June, for some positive integer k.
2007 COMC Solutions Page 3

In September, there were thus 3k − 80 boys and 2k − 20 girls in the school. Since the new ratio
is 7 : 5, then
3k − 80 7
=
2k − 20 5
5(3k − 80) = 7(2k − 20)
15k − 400 = 14k − 140
k = 260

Therefore, the total number of the students in the school in June was 3k + 2k = 5k = 5(260),
or 1300 students.

Solution 2
Suppose that there were b boys and g girls in the school in June.
In September, there were thus b − 80 boys and g − 20 girls in the school.
b 3 b − 80 7
From the given information, we know that = and = .
g 2 g − 20 5
Eliminating fractions gives the equations 2b = 3g and 5(b−80) = 7(g−20) or 5b−400 = 7g−140
or 5b − 7g = 260.
Multiplying the second equation by 2 gives 10b − 14g = 520, and substituting 10b = 15g gives
g = 520.
Therefore, b = 32 (520) = 780, so there were b + g = 780 + 520 = 1300 students in the school in
June.

Answer: 1300

5. Solution 1
When the nine numbers are placed in the array in any arrangement, the sum of the row sums
is always 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 = 45, because each of the nine numbers appears in
exactly one row.
Similarly, the sum of the column sums is also always 45, as again each of the nine numbers
appears in exactly one column.
Therefore, the grand sum S equals 90 plus the sum of the diagonal sums, and so depends only
on the diagonal entries, labelled in the array below:

a c
e
g k

So S = 90 + (a + e + k) + (c + e + g) = 90 + 2e + a + c + g + k. To make S as large as possible,

we must make 2e + a + c + g + k as large as possible.
2007 COMC Solutions Page 4

Since a, c, e, g, k can be any of the numbers from 1 to 9, then S is largest when e = 9 and
a, c, g, k are 5, 6, 7, 8 in some order, for example in the configuration below:

5 1 6
2 9 3
7 4 8

Therefore, the maximum possible value of S is 90 + 2(9) + 8 + 7 + 6 + 5 = 90 + 44 = 134.

Solution 2
Suppose that a, b, c, d, e, f, g, h, k represent the numbers 1 to 9 in some order, and are entered
in the array as shown:

a b c
d e f
g h k

The grand sum is thus

S = (a + b + c) + (d + e + f ) + (g + h + k) + (a + d + g) + (b + e + h) + (c + f + k) +
(a + e + k) + (c + e + g)
= 4e + 3a + 3c + 3g + 3k + 2b + 2d + 2f + 2h

To make S as large as possible, we should assign the values of a, b, c, d, e, f, g, h, k so that the

largest values go to the variables with the largest coefficients in the expression for S.
In other words, S will be maximized when e = 9, a, c, g, k equal 8, 7, 6, 5 in some order, and
b, d, f, h equal 4, 3, 2, 1 in some order.
Therefore, the maximum possible value of S is

S = 4(9) + 3(8 + 7 + 6 + 5) + 2(4 + 3 + 2 + 1) = 36 + 3(36) + 2(10) = 134

Answer: 134

6. Suppose that r is the radius of the circle.

Join O to P and drop a perpendicular from P to Q on OA.

O
Q P

A N
2007 COMC Solutions Page 5

Since OA and P N are perpendicular to AN and P Q is perpendicular to OA, then QP N A is

a rectangle. Therefore, QP = AN = 15 and QA = P N = 9.
Since O is the centre of the circle and A and P are on the circumference, then OA = OP = r.
Since OA = r and QA = 9, then OQ = r − 9.
Since 4OQP is right-angled at Q, then, by the Pythagorean Theorem,

OP 2 = OQ2 + QP 2
r2 = (r − 9)2 + 152
r2 = r2 − 18r + 81 + 225
18r = 306
r = 17

Therefore, the radius is 17.

Answer: 17

7. From the second equation, x + y = 7 − z, so after squaring both sides, we obtain

x2 + 2xy + y 2 = 49 − 14z + z 2 (∗)

From the third equation, x2 + y 2 = 133 − z 2 , so using this and the first equation to substitute
into (∗), we get

(133 − z 2 ) + 2(z 2 ) = 49 − 14z + z 2

14z = −84
z = −6

Substituting this value for z back into the first two equations, we get xy = (−6)2 = 36 and
x + y = 7 − (−6) = 13.
Therefore, y = 13 − x and so x(13 − x) = 36 or 0 = x2 − 13x + 36.
This tells us that 0 = (x − 4)(x − 9) so x = 4 or x = 9.
If x = 4, then y = 13 − x tells us that y = 9.
If x = 9, then y = 13 − x tells us that y = 4.
Therefore, the solutions are (x, y, z) = (4, 9, −6), (9, 4, −6).

Answer: (4, 9, −6), (9, 4, −6)

8. In order to travel from A to B along the segments without travelling along any segment more
than once, we must always move up, down or to the right. (In other words, we can never
travel to the left without retracing our steps.) To see this, we note that if we do travel along
2007 COMC Solutions Page 6

a segment to the left, then we must have travelled along the other horizontal segment in this
square to the right at an earlier stage. We would then need to travel back along one of these
segments to get to B, thus retracing our steps.
Any route from A to B involves exactly 9 moves to the right and some number of moves up
and down.
Any route from A to B involves exactly one more move down than moves up, as we start at the
top of the grid and end up at the bottom. Therefore, the total number of up and down moves
must be odd, as it equals (x + 1) + x = 2x + 1, where x is the total number of up moves.
There are 10 vertical segments. Any choice of an odd number of these vertical segments uniquely
determines a route from A to B, as we must start at A, travel to the top of the leftmost of
these segments, travel down the segment, travel to the right to the bottom of the next segment,
travel up it, and so on.
Therefore, the routes from A to B are in exact correspondence with choices of an odd number
of the 10 vertical segments.
We compute the number of routes using n of these segments, for n = 1, 3, 5, 7, 9. In each case,
the length of the route will be 9 + n. ! !
10 10
For n = 1 and n = 9, the number of routes is = = 10.
1 9
! !
10 10 10(9)(8)
For n = 3 and n = 7, the number of routes is = = = 120.
3 7 3(2)(1)
!
10 10(9)(8)(7)(6) 10(9)(8)(7)
For n = 5, the number of routes is = = = 2(9)(2)(7) = 252.
5 5(4)(3)(2)(1) 5(4)
Therefore, the route length with the maximum number of routes is when n = 5. In this case,
the route length is 14 and the number of routes is 252.
(Instead of going !through all of the above calculations, we could have remarked that among
10
the numbers , the largest occurs when n is exactly half of 10.)
n

Answer: Length= 14, Number of Routes= 252

2007 COMC Solutions Page 7

Part B

1. (a) Since x − 1, 2x + 2, and 7x + 1 form an arithmetic sequence, then

(2x + 2) − (x − 1) = (7x + 1) − (2x + 2)

x + 3 = 5x − 1
4 = 4x
x = 1

so x = 1.
(b) Solution 1
Since x = 1, the first term of the sequence is 0.
Since the last term is 72, the sequence is arithmetic, and we are told that there is a middle
0 + 72
term, then this middle term is equal to = 36.
2
(Note that if there was an even number of terms, there would not necessarily be a middle
term. Since we are asked to find the middle term, we can safely assume that there is one!)

Solution 2
Since x = 1, the first three terms of the sequence are 0, 4, 8.
Since the common difference is 4 and the first term is 0, the number of times that the
72 − 0
difference needs to be added to get to the final term of 72 is = 18.
4
Therefore, 72 is the 19th term.
The middle term is thus the 10th term, or 0 + 4(10 − 1) = 36.
(c) Since y − 1, 2y + 2, and 7y + 1 form a geometric sequence, then
2y + 2 7y + 1
=
y−1 2y + 2
2
(2y + 2) = (y − 1)(7y + 1)
4y 2 + 8y + 4 = 7y 2 − 6y − 1
0 = 3y 2 − 14y − 5
0 = (3y + 1)(y − 5)

Therefore, y = − 13 or y = 5.
(d) If y = − 13 , the first three terms of the sequence are − 34 , 43 , − 34 .
4
3
In this case, the common ratio between successive terms is = −1.
− 34
Therefore, the 6th term in this sequence is − 34 (−1)5 = 43 .

If y = 5, the first three terms of the sequence are 4, 12, 36.

2007 COMC Solutions Page 8

12
In this case, the common ratio between successive terms is = 3.
4
Therefore, the 6th term in this sequence is 4(35 ) = 4(243) = 972.

2. (a) Solution 1
Since ∠ABC = ∠BCD = 90◦ , then BA and CD are parallel, so ABCD is a trapezoid.
Thus, the area of ABCD is 12 (24)(9 + 18) = 12(27) = 324.

Solution 2
Since ∠ABC = 90◦ , then the area of 4ABC is 12 (9)(24) = 9(12) = 108.
Also, since ∠BCD = 90◦ , then 4ACD has height 24.
Therefore, the area of 4ACD is 12 (18)(24) = 9(24) = 216.
Thus, the area of quadrilateral ABCD is 108 + 216 = 324.
(b) Solution 1
Since BA is parallel to CD, then ∠ABD = ∠BDC.
Since ∠BEA = ∠DEC as well, then 4ABE is similar to 4CDE.
DE CD 18
Therefore, = = = 2, so DE : EB = 2 : 1, as required.
BE AB 9

Solution 2
As suggested by the diagram, we coordinatize the diagram.
Put C at the origin, D on the positive x-axis (with coordinates (18, 0)) and B on the
positive y-axis (with coordinates (0, 24)).
Since ∠ABC = 90◦ , then A has coordinates (9, 24).
Therefore, the line through C and A has slope 24 9
= 83 so has equation y = 83 x.
Also, the line through B and D has slope −24 18
= − 43 , so has equation y = − 34 x + 24.
Point E lies at the point of intersection of these lines, so we combine the equations to find
the coordinates of E, getting 38 x = − 43 x + 24 or 4x = 24 or x = 6.
Therefore, E has y-coordinate 38 (6) = 16, so E has coordinates (6, 16).
To show that DE : EB = 2 : 1, we can note that E lies one-third of the way along from B
to D since the x-coordinate of E is one-third that of D (and the x-coordinate of B is 0),
or since the y-coordinate of E is two-thirds that of B (and the y-coordinate of D is 0).
Alternatively, we could calculate the length BE (which is 10) and the length of ED (which
is 20).
Using any of these methods, DE : EB = 2 : 1.
(c) Solution 1
From (b), 4ABE is similar to 4CDE and their sides are in the ratio 1 : 2.
This also tells us that the height of 4CDE is twice that of 4ABE.
Since the sum of the heights of the two triangles is 24, then the height of 4CDE is
2
3
(24) = 16.
2007 COMC Solutions Page 9

Therefore, the area of 4DEC is 12 (18)(16) = 144.

Solution 2
From (b), the coordinates of E are (6, 16).
Therefore, the height of 4DEC is 16.
Therefore, the area of 4DEC is 12 (18)(16) = 144.
(d) Solution 1
From (c), the area of 4DEC is 144.
From Solution 2 of (a), the area of 4ACD is 216.
The area of 4DAE is the difference in these areas, or 216 − 144 = 72.

Solution 2
Using the coordinatization from (b), the coordinates of A are (9, 24), the coordinates of
E are (6, 16), and the coordinates of D are (18, 0).
Using the “up products and down products” method, the area of the triangle is
 
 9 24 
 
 
1  6 16  1
= |9(16) + 6(0) + 18(24) − 24(6) − 16(18) − 0(9)|
2  18 0 
  2
1
 
 9 24 
= |144 + 0 + 432 − 144 − 288 − 0|
2
= 72

3. (a) Alphonse should win.

If Alphonse starts by taking 1 stone, then by Rule #3, Beryl must remove at least 1 stone
and at most 2(1) − 1 = 1 stone. In other words, Beryl must remove 1 stone.
This in turn forces Alphonse to remove 1 stone, and so on.
Continuing in this way, Alphonse removes 1 stone from an odd-sized pile at each turn and
Beryl removes 1 stone from an even-sized pile at each turn. Thus, Alphonse removes the
last stone.
Therefore, Alphonse wins by removing 1 stone initially since 7 is odd.
(In fact, this argument shows that Alphonse should win whenever N is odd.)
(b) Beryl should win.
We make a chart in which the rows enumerate a possible combination of moves. Each
move indicates the number of stones removed.
2007 COMC Solutions Page 10

A1 B1 A2 B2 A3 B3 A4 B4 Winner
1 1 1 1 1 1 1 1 B
2 2 1 1 1 1 B
2 2 2 2 B
2 2 3 1 B
3 5 B
4 4 B
5 3 B
6 2 B
7 1 B
Thus, no matter what number of stones Alphonse removes initially, there is a move that
Beryl can make which allows her to win. (There are possible combinations of moves where
Alphonse wins that are not listed in this chart.) Therefore, Beryl should win when N = 8.
Her strategy is:
• If Alphonse removes 3 or more stones, then he can remove the remaining stones in
the pile and win.
• If Alphonse removes 1 or 2 stones, then he can win by using the table above, choosing
a row for which she wins. In effect, Beryl repeats Alphonse’s move on her first turn.
This ensures that Alphonse receives a pile with an even number of stones and that he
can remove no more than 3 stones on his next turn. Thus, she can win, as the table
shows.
(c) Solution 1
We show that Beryl has a winning strategy if and only if N = 2m , with m a positive
integer.

First, if N is odd, we know that Alphonse has a winning strategy as in (a) (Alphonse
removes 1 stone, forcing Beryl to remove 1 stone, and so on).
Second, if N = 2, then Beryl wins as Alphonse must remove 1 stone to begin, so Beryl
removes the remaining stone.

Next, we show that if N = 2k, then the player who has the winning strategy for N = k
also has a winning strategy for N = 2k. This will tell us that Beryl has a winning strategy
for N = 2, 4, 8, 16, . . . (in general, for N = 2m ) and that Alphonse has a winning strategy
if N = 2m q where q is an odd integer (since Alphonse wins for N = q, 2q, 4q, . . .). Since
every even integer can be written in one of these two forms, this will complete our proof.

So consider N = 2k.
• If either player removes an odd number of stones from an even-sized pile (leaving an
2007 COMC Solutions Page 11

odd-sized pile), then they can be forced to lose, as the other player can then remove
1 stone from an odd-sized pile and force a win as in (a). So if Alphonse removes
an even number of stones to start, then Beryl should next remove an even number
of stones (so that Alphonse can’t immediately force her to lose), so the pile size will
always remain even and each player’s move will always be to remove an even number
of stones.
• Suppose that Alphonse has a winning strategy for N = k of the form a1 , b1 , a2 , b2 , . . . , aj .
Here, we mean that Alphonse removes a1 stones on his first turn and responds to
Beryl’s first move b1 by removing a2 and so on. (Of course, a2 will depend on b1
which could take a number of values, and so on.) Since these are valid moves, then
1 ≤ a1 < k, and b1 < 2a1 , and a2 < 2b1 , and so on.
Then 2a1 , 2b1 , 2a2 , 2b2 , . . . , 2aj will be a winning strategy for Alphonse for N = 2k
since 1 < 2a1 < 2k, and 2b1 < 2(2a1 ), and 2a2 < 2(2b1 ), and so on, so this is a valid
sequence of moves and they exhaust the pile with Alphonse taking the last stone.
In other words, to win when N = 2k, Alphonse consults his winning strategy for
N = k. He removes twice his initial winning move for N = k. If Beryl removes 2b
stones next, Alphonse then removes 2a stones, where a is his winning response to
Beryl removing b stones in the N = k game. This guarantees that he will win.
• Suppose that Beryl has a winning strategy for N = k.
By an analogous argument, Beryl has a winning strategy for N = 2k, for if Alphonse
removes 2a stones, then she removes 2b stones, where b is her winning responding
move to Alphonse removing a stones in the N = k game.
Therefore, Beryl wins if and only if N = 2m , with m a positive integer.

Solution 2
We show that Beryl has a winning strategy if and only if N = 2m , with m a positive
integer.

Suppose first that N is not a power of 2.

We can write N as a sum of distinct powers of 2, in the form N = 2k1 + 2k2 + · · · + 2kj ,
where k1 > k2 > · · · > kj ≥ 0. (In essence, we are writing N in binary.) Since N is not
itself a power of 2, then this representation uses more than one power of 2 (that is, j ≥ 2).
We will show that Alphonse has a strategy where he can always reduce the number of
powers of 2 being used, while Beryl can never reduce the number of powers of 2 being
used. This will show that Alphonse can always remove the final stone, as only he can
reduce the number of powers of 2 being used to 0.
Alphonse’s strategy is to initially remove the smallest power of 2 from the representation
of N (that is, he removes 2kj stones).
2007 COMC Solutions Page 12

On her first turn, Beryl thus receives a pile with 2k1 + 2k2 + · · · + 2kj−1 stones. By rule #3,
she must remove fewer than 2(2kj ) = 2kj +1 stones. Since kj−1 > kj , then kj−1 ≥ kj + 1, so
Beryl must remove fewer than 2kj−1 stones.
When she removes these stones, the 2kj−1 will be removed from the representation of the
number of remaining stones, but will be replaced by at least one (if not more) smaller
powers of 2.
Thus, Beryl cannot reduce the number of powers of 2 in the representation.
Suppose that Alphonse thus receives a pile with 2k1 + 2k2 + · · · + 2kj−2 + 2d1 + 2d2 + · · · + 2dh
stones, with k1 > k2 > · · · > kj−2 > d1 > d2 > · · · > dh and h ≥ 1.
This means that Beryl removed B = 2kj−1 − (2d1 + 2d2 + · · · + 2dh ) stones.
But B > 0 and B is divisible by 2dh (since B = 2dh 2kj−1 −dh − (2d1 −dh + 2d2 −dh + · · · + 20

and each of the exponents initially were larger than dh ), so B ≥ 2dh .

Therefore, Alphonse can remove 2dh stones on his turn (that is, the smallest power of 2 in
the representation of the number of remaining stones) since 2dh stones satisfies Rule #3,
and so his strategy can continue.
Therefore, Alphonse has a winning strategy if N is not a power of 2.

If N is a power of 2, then Alphonse on his first turn cannot decrease the number of
powers of 2 in the representation of N . (This is a similar argument to the one above for
Beryl’s first turn.) On Beryl’s first turn, though, she can reduce the number of powers of
2 (as in Alphonse’s second turn above).
Therefore, the roles are reversed, and Beryl can always reduce the number of powers of 2,
while Alphonse cannot. Therefore, Beryl has a winning strategy when N is a power of 2.

Therefore, Beryl has a winning strategy if and only if N is a power of 2.

4. (a) Solution 1
In t seconds, the mouse runs 7t metres and the cat runs 13t metres.
Using this, we get a triangle with the cat and mouse meeting at point P .
P

13t
7t
120
C 60 M
2007 COMC Solutions Page 13

By the cosine law,

CP 2 = CM 2 + M P 2 − 2(CM )(M P ) cos(∠CM P )

(13t)2 = 602 + (7t)2 − 2(60)(7t) cos(120◦ )
169t2 = 3600 + 49t2 − 120(7t)(− 21 )
169t2 = 3600 + 49t2 + 60(7t)
120t2 − 420t − 3600 = 0
2t2 − 7t − 60 = 0
(2t − 15)(t + 4) = 0

Therefore, t = 15
2
or t = −4.
15
Since t represents a time, then t > 0, so t = 2
.

Solution 2
In t seconds, the mouse runs 7t metres and the cat runs 13t metres.
Using this, we get a triangle with the cat and mouse meeting at point P . Drop a perpen-
dicular from P to N on CM extended.
P

13t
7t

60
C 60 M N

Since ∠P M N = 60◦ , then 4P M N is a 30◦ -60◦ -90◦ triangle.

√ √
Therefore, M N = 21 P M = 72 t and P N = 3M N = 7 2 3 t.
√
This gives us right-angled 4CP N with CP = 13t, P N = 7 2 3 t, and CN = 60 + 72 t.
By the Pythagorean Theorem,

CP 2 = CN 2 + N P 2
 √ 2
2 7 2
(13t) = 60 + 2 t + 7 2 3 t

169t2 = 3600 + 420t + 49 2

4
t + 147 2
4
t
169t2 = 3600 + 420 + 49t 2

120t2 − 420t − 3600 = 0

2t2 − 7t − 60 = 0
(2t − 15)(t + 4) = 0

Therefore, t = 15
2
or t = −4.
15
Since t represents a time, then t > 0, so t = 2
.
2007 COMC Solutions Page 14

(b) Solution 1
We coordinatize the situation, as suggested in the diagram with C having coordinates
(−60, 0) and M having coordinates (0, 0).
Suppose that the cat intercepts the mouse at point P (x, y).
CP 13
Since the cat runs at 13 m/s and the mouse at 7 m/s, then = . Thus,
MP 7
p
(x + 60)2 + y 2 13
p =
x2 + y 2 7
(x + 60)2 + y 2 169
2 2
=
x +y 49
49((x + 60) + y ) = 169(x2 + y 2 )
2 2

0 = 120x2 − 2(49)(60)x − 49(602 ) + 120y 2

0 = x2 − 49x − 49(30) + y 2

Since this equation is of the form 0 = x2 + ax + y 2 + by + c and there is at least one

point whose coordinates satisfy the equation (for example, setting y = 0 gives a quadratic
equation with positive discriminant), then it is the equation of a circle, so all points of
interception lie on a circle. (We could also complete the square to obtain the equation
(x − 49
2
)2 + y 2 = ( 91
2
)2 , which is the equation of the circle with centre ( 49
2
, 0) and radius 91
2
.)

Solution 2
We coordinatize the situation, as suggested in the diagram, with C having coordinates
(−60, 0) and M having coordinates (0, 0).
Suppose that the cat intercepts the mouse at point P (x, y).
Suppose that the cat intercepts the mouse after t seconds and that the mouse runs in the
direction θ East of North. (θ here could be negative. We can assume that −90◦ ≤ θ ≤ 90◦
to keep the situation in the upper half of the plane. If θ did not lie in this range, then
P would be in the lower half plane and we could reflect it in the x-axis and use this
argument.)
P(x, y)

13t
7t

C 60 m M
2007 COMC Solutions Page 15

As in (a),

CP 2 = CM 2 + M P 2 − 2(CM )(M P ) cos(∠CM P )

(13t)2 = 602 + (7t)2 − 2(60)(7t) cos(90◦ + θ)
120t2 = 3600 + 120(7t) sin θ
t2 − 7t sin θ = 30

But M P 2 = 49t2 = x2 + y 2 and x = 7t cos(90◦ − θ) = 7t sin θ, so

x2 + y 2
− x = 30
49
x2 − 49x − 49(30) + y 2 = 0

so all possible points P lie on a circle, as in Solution 1.

Solution 3
We coordinatize the situation, as suggested in the diagram with C having coordinates
(−60, 0) and M having coordinates (0, 0).
Suppose that the cat intercepts the mouse at point P (x, y).
Suppose that the cat intercepts the mouse after t seconds and that the mouse runs in the
direction θ East of North. (θ here could be negative. We can assume that −90◦ ≤ θ ≤ 90◦
to keep the situation in the upper half of the plane. If θ did not lie in this range, then
P would be in the lower half plane and we could reflect it in the x-axis and use this
argument.)
P(x, y)

13t
7t

C 60 m M

If the mouse decides to run due East, then it will be caught when −60+13t = 7t or t = 10,
so will be caught at B(70, 0).
If the mouse decides to run due West, then it will be caught when −60 + 13t = −7t or
t = 3, so will be caught at A(−21, 0).
The positions above the x-axis where the mouse will be caught should be exactly symmetric
with the positions below the x-axis where the mouse will be caught. Therefore, if these
positions lie on a circle, then a diameter of this circle should lie on the x-axis.
Since the only positions on the x-axis where the mouse will be caught are A(−21, 0) and
2007 COMC Solutions Page 16

B(70, 0), then these must be endpoints of the diameter.

Therefore, the circle will have centre E with coordinates 21 (−21 + 70), 0 = 49



2
,0 and
radius 12 (70 − (−21)) = 91
2
.
From the diagram above, the coordinates of the point P of intersection will be

(7t cos(90◦ − θ), 7t sin(90◦ − θ)) = (7t sin θ, 7t cos θ)

If we can show that P E = 91 2

for every value of θ, then we will have shown that every
point of intersection lies on the circle with centre E and radius 91
2
.
Now

2
P E 2 = 7t sin θ − 49 2
+ (7t cos θ)2
492
= 49t2 sin2 θ − 7(49)t sin θ + 4
+ 49t2 cos2 θ
492
= 49t2 (sin2 θ + cos2 θ) − 7(49)t sin θ + 4
492
= 49t2 − 7(49)t sin θ + 4

From Solution 2, t2 − 7t sin θ = 30, so

492 72 132 91 2
P E 2 = 49(30) + 49 169




4
= 49 30 + 4
= 49 4
= 4
= 2

so P E = 91
2
, as required.
Therefore, all points of intersection lie on a circle.
(c) From (b), we know that the points of intersection lie on the circle with diameter AB,
where A has coordinates (−21, 0) and B has coordinates (70, 0).
Suppose that the mouse is intercepted at point P1 after running d1 metres and at point
P2 after running d2 metres.
P1

d1
21 70
A B
M
d2

P2

By the Intersecting Chords Theorem, d1 d2 = 21(70).

By the Arithmetic Mean-Geometric Mean Inequality,
d1 + d2 p p √
≥ d1 d2 = 21(70) = 7 30
2
√ √
Therefore, d1 + d2 ≥ 2(7 30) = 14 30.
2007 COMC Solutions Page 17

(The Arithmetic Mean-Geometric Mean Inequality (known as the AM-GM Inequality)

comes from the fact that if d1 and d2 are non-negative, then (d1 − d2 )2 ≥ 0.
 2
2 2 d1 + d2 d1 + d2 p
Thus, d1 + 2d1 d2 + d ≥ 4d1 d2 , so ≥ d1 d2 , so ≥ d1 d2 .)
2 2
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING
presents the

Sun Life Financial Canadian

Open Mathematics Challenge

Wednesday, November 19, 2008

Time: 2 12 hours 2008

c Canadian Mathematical Society
Calculators are NOT permitted.
Do not open this booklet until instructed to do so.
There are two parts to this paper.
PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value
for each question by entering the correct answer(s) in the space provided. If your answer is
incorrect, any work that you do will be considered for part marks, provided that it is done
in the space allocated to that question in your answer booklet.
PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must
be written in the appropriate location in the answer booklet. Rough work should be done
separately. If you require extra pages for your finished solutions, paper will be provided by
your supervising teacher. Any extra papers should be placed inside your answer booklet. Be
sure to write your name and school name on any inserted pages.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution,
poorly presented, will not earn full marks.
NOTES:
At the completion of the contest, insert the information sheet inside the answer
booklet.
The names of top scoring competitors will be published on the Web sites of the
CMS and CEMC.
Sun Life Financial Canadian Open Mathematics Challenge
NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations
√ and answers will be expressed as
exact numbers such as 4π, 2 + 7, etc., rather than as 12.566 . . .
or 4.646 . . ..
4. Calculators are not allowed.
5. Diagrams are not drawn to scale. They are intended as aids only.

PART A
1. If 2x + y = 13 and x + 2y = 11, what is the value of x + y?
2. Determine the units digit of the integer equal to 9 + 92 + 93 + 94 .
(The units digit of an integer is its rightmost digit. For example, the units digit of
the integer 1234 is 4.)
3. If the average of four different positive integers is 8, what is the largest possible value
of any one of these integers?
4. In the diagram, D is on AB and E is on AC with DE parallel A
to BC. Also, DE = 1, BC = 6, AE = x, and EC = x2 + 4. x
Determine all possible values of x.
D E
1

x2 + 4

B C
6
5. Four consecutive integers p, q, r, s with p < q < r < s satisfy 12 p + 13 q + 14 r = s.
What is the value of s?
6. Rectangle ABCD intersects a circle at points A H G D
E, F, G, and H, as shown. If AH = 4, HG = 5 B C
E F
and BE = 3, determine the length of EF .

7. A star is placed in the bottom right corner square of a 3 × 3

grid, as shown. A fair coin is flipped repeatedly. Each time that
the coin shows heads, the star is moved one space upwards;
each time that the coin shows tails, the star is moved one space
to the left. (The star may move off the grid.) Determine the
probability that the star reaches the top left corner square of
the grid.

8. Determine the sum of all integer values of the parameter r for which the equation
x3 − rx + r + 11 = 0 has at least one positive integer solution for x.
 PART B
1. In the diagram, 4P SR is right-angled at S and
4P RQ is right-angled at R. Also, P S = 12, P
SR = 9, and P Q = 25.
12 25
(a) Determine the length of RQ.
(b) Determine the area of figure P QRS. S
9 Q
(c) Show that ∠QP R = ∠P RS. R
(d) Determine the length of SQ.

2. (a) Determine all real numbers x such that (x + 3)(x − 6) = −14.

(b) Determine all real numbers x such that 22x − 3(2x ) − 4 = 0.
(c) Determine all real numbers x such that (x2 − 3x)2 = 4 − 3(3x − x2 ).

3. (a) An infinite sequence a0 , a1 , a2 , a3 , . . . satisfies

am−n + am+n = 12 a2m + 12 a2n

for all non-negative integers m and n with m ≥ n ≥ 0.

(i) Show that a0 = 0.
(ii) If a1 = 1, determine the value of a2 and the value of a3 .
(b) An infinite sequence b0 , b1 , b2 , b3 , . . . satisfies

bm−n + bm+n = b2m + b2n

for all non-negative integers m and n with m ≥ n ≥ 0. Prove that all terms in
the sequence have the same value.

4. A triangle is called automedian if its three medians can be used to form a triangle
that is similar to the original triangle.

(a) Show that the triangle with sides of length 7, 13 and 17 is automedian.
(b) 4ABC has side lengths AB = c, AC = b and BC = a, with a < b < c. If
4ABC is automedian, prove that a2 + c2 = 2b2 .
(c) Determine, with proof, an infinite family of automedian triangles with integer
side lengths, such that no two of the triangles in the family are similar.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

presents the

Sun Life Financial

Canadian Open Mathematics Challenge
Wednesday, November 19, 2008

Solutions

2008
c Canadian Mathematical Society
2008 COMC Solutions Page 2

Part A

1. Solution 1
Adding the two equations gives (2x + y) + (x + 2y) = 13 + 11 or 3x + 3y = 24.
Thus, x + y = 31 (24) = 8.

Solution 2
We solve for x first, then y.
Since 2x + y = 13, then 4x + 2y = 26.
Subtracting the second equation from this, we obtain (4x + 2y) − (x + 2y) = 26 − 11 or 3x = 15
or x = 5.
Substituting into the original first equation, we obtain 2(5) + y = 13 or y = 13 − 10 = 3.
Thus, x + y = 5 + 3 = 8.

Solution 3
We solve for y first, then x.
Since x + 2y = 11, then 2x + 4y = 22.
Subtracting the first equation from this, we obtain (2x + 4y) − (2x + y) = 22 − 13 or 3y = 9 or
y = 3.
Substituting into the original second equation, we obtain x + 2(3) = 11 or x = 11 − 6 = 5.
Thus, x + y = 5 + 3 = 8.

Answer: x + y = 8

2. Solution 1
We note that 9 + 92 + 93 + 94 = 9(1 + 91 ) + 93 (1 + 91 ) = (9 + 93 )(1 + 9) = 10(9 + 93 ).
Therefore, 9 + 92 + 93 + 94 is an integer that is divisible by 10, so its units digit is 0.

Solution 2
Calculating each term, 92 = 81, 93 = 92 91 = 81(9) = 729 and 94 = 93 91 = 729(9) = 6561.
Thus, 9 + 92 + 93 + 94 = 9 + 81 + 729 + 6561 = 90 + 7290 = 7380.
Therefore, the units digit of the integer equal to 9 + 92 + 93 + 94 is 0.

Solution 3
Since 92 = 81, its units digit is 1.
Since 93 = 92 91 , then we can calculate the units digit of 93 by multiplying the units digits of 92
and 91 and finding the units digit of this product. (This is because the units digit of a product
depends only on the units digits of the numbers that we are multiplying together.) Thus, the
units digit of 93 is 1(9) = 9.
2008 COMC Solutions Page 3

Since 94 = 92 92 and the units digit of 92 is 1, then the units digit of 94 is 1(1) = 1, in a similar
way.
We can now calculate the units digit of 9 + 92 + 93 + 94 by adding the units digits of the four
terms, and finding the units digit of this sum of units digits.
The sum of the units digits is 9 + 1 + 9 + 1 = 20, so the units digit of the integer equal to
9 + 92 + 93 + 94 is 0.

Answer: 0

3. Suppose the four positive integers are a, b, c, d.

a+b+c+d
Since the average of the four positive integers is 8, then = 8 or a + b + c + d = 32.
4
We try to find the maximum possible value of d.
We know that d = 32 − a − b − c = 32 − (a + b + c).
To make d as large as possible, we make a + b + c as small as possible.
Since a, b and c are different positive integers, then the smallest possible value of a + b + c is
1 + 2 + 3 or 6.
Thus, the largest possible value of d is 32 − 6 = 26.

Answer: 26

4. Consider 4AED and 4ACB.

These triangles have a common angle at A.
Also, since DE is parallel to BC, then ∠AED = ∠ACB.
Therefore, 4AED is similar to 4ACB.
Thus,
AE AC
=
DE BC
x x + x2 + 4
=
1 6
2
6x = x + x + 4
0 = x2 − 5x + 4
0 = (x − 1)(x − 4)

so x = 1 or x = 4.
(We can check that each of these values of x actually gives a triangle.)

Answer: x = 1 or x = 4

5. Since p, q, r, s are four consecutive integers with p < q < r < s, then r = s − 1, q = s − 2, and
p = s − 3.
2008 COMC Solutions Page 4

Thus,
1
2
p + 31 q + 14 r = s
1
2
(s − 3) + 13 (s − 2) + 14 (s − 1) = s
6(s − 3) + 4(s − 2) + 3(s − 1) = 12s (multiplying through by 12)
13s − 18 − 8 − 3 = 12s
s = 29 .

(Checking, 21 (26) + 13 (27) + 14 (28) = 13 + 9 + 7 = 29.)

Answer: s = 29

6. Solution 1
Let O be the centre of the circle and join O to the midpoints M and N of EF and HG,
respectively.
Since O is the centre, then O, M and N are collinear and ON is perpendicular to both EF
and HG.

A H G D
N

B E M
F C
O

Since N is the midpoint of HG, then HN = 12 HG = 25 , so AN = AH + HN = 4 + 52 = 13

2
.
Since ON is perpendicular to AN and BM , and AN and BM are parallel, then AN M B is a
rectangle.
Therefore, BM = AN = 13 2
.
Thus,
EF = 2(EM ) = 2(BM − BE) = 2( 13 2
− 3) = 2( 72 ) = 7 .
Solution 2
Let O be the centre of the circle.
Drop perpendiculars from G and H to K and L on EF , respectively.

A H G D
F
B E K L C
O

Since ABCD is a rectangle, then AHKB, HGLK and GDCL are rectangles.
We want the length of EF . We note that EF = EK + KL + LF .
2008 COMC Solutions Page 5

Since HGLK is a rectangle, then KL = HG = 5.

By symmetry, EK = LF .
Now EK = BK − BE = AH − BE = 4 − 3 = 1.
Thus, EF = 1 + 5 + 1 = 7.

Answer: EF = 7

7. We make a list of the possible paths that the star could take until it either reaches the top left
corner or it moves off the grid. We use “L” to represent a move to the left and “U” to represent
a move upwards. The possible paths are:

LLL, LLUL, LLUU, LULL, LULU, LUUL, LUUU,

UUU, UULU, UULL, ULUU, ULUL, ULLU, ULLL

Of these paths, 6 of them reach the top left corner square, namely LLUU, LULU, LUUL, UULL,
ULUL, and ULLU.
Each time the star moves, the probability that it moves to the left is 21 and the probability that
it moves upwards is 12 . Thus, the probability that the star follows a particular path of length 3

3
4
is 12 = 18 and the probability that the star follows a particular path of length 4 is 12 = 16 1
.
(We can note that we have 2 paths of length 3 and 12 paths of length 4, which is consistent
with the probabilities above, because 2 18 + 12 16 1


1 3
= 4 + 4 = 1.)
Since there are 6 paths of length 4 that reach the top left corner, the probability that the star
1

3
reaches this square is 6 16 = 8.
3
Answer: 8

8. Suppose that x is an integer that satisfies this equation.

Then

x3 − rx + r + 11 = 0
x3 + 11 = rx − r
r(x − 1) = x3 + 11
x3 + 11 x3 − 1 12
r = = +
x−1 x−1 x−1
12
r = x2 + x + 1 +
x−1
We note that x 6= 1 (which we can confirm from the original equation, as if x = 1 we obtain
12 = 0).
2008 COMC Solutions Page 6

12
Thus, for r to be an integer, we need to be an integer, because x2 + x + 1 is an integer
x−1
already.
Hence, x − 1 must be a divisor of 12 and x must be positive (so x − 1 is non-negative). We
make a table to enumerate the possible values of x − 1, and thus of x and r:

12
x−1 x r = x2 + x + 1 +
x−1
12 13 184
6 7 59
4 5 34
3 4 25
2 3 19
1 2 19

Therefore, the sum of the possible values of r is

184 + 59 + 34 + 25 + 19 = 321

(Note that we only include 19 once in this sum, not two times.)
Answer: 321
2008 COMC Solutions Page 7

Part B

1. (a) By the Pythagorean Theorem in 4P SR, we have

P R2 = SR2 + SP 2 = 92 + 122 = 81 + 144 = 225

√
so P R = 225 = 15, since P R > 0.
By the Pythagorean Theorem in 4P RQ, we have

RQ2 = P Q2 − P R2 = 252 − 152 = 625 − 225 = 400

√
so RQ = 400 = 20, since RQ > 0.
(b) The area of figure P QRS is the sum of the areas of 4P SR and 4P RQ.
Since these triangles are right-angled at S and R, respectively, the area is

1
2
(12)(9) + 12 (15)(20) = 54 + 150 = 204 .

(c) Solution 1
We know all of the side-lengths of each of the two right-angled triangles 4RSP and
4P RQ.
Therefore, we can calculate trigonometric ratios.
RQ 20 4 PS 12 4
In particular, sin(∠QP R) = = = and sin(∠P RS) = = = .
PQ 25 5 PR 15 5
Since ∠QP R and ∠P RS are each acute and have equal sines, then ∠QP R = ∠P RS, as
required.
(We could instead have calculated the cosines or tangents of these two angles.)

Solution 2
Since each of 4RSP and 4P RQ is right-angled (at S and at R, respectively) and
RS PR 3
= = , then 4RSP is similar to 4P QR.
SP RQ 4
Thus, ∠QP R = ∠SRP = ∠P RS.
(d) Since ∠QP R = ∠P RS, then P Q is parallel to SR.
Since P S is perpendicular to SR and P Q is parallel to SR, then P S is also perpendicular
to P Q, so ∠SP Q = 90◦ .
Therefore, by the Pythagorean Theorem, SQ2 = SP 2 +P Q2 = 122 +252 = 144+625 = 769,
√
so SQ = 769 since SQ > 0.

2. (a) Since (x + 3)(x − 6) = −14, then x2 − 3x − 18 = −14 or x2 − 3x − 4 = 0.

Factoring, we obtain (x − 4)(x + 1) = 0, so x = 4 or x = −1.
(b) Let u = 2x . Thus, u2 = (2x )2 = 22x , so the equation becomes u2 − 3u − 4 = 0.
By (a), u = 4 or u = −1 so 2x = 4 or 2x = −1.
2008 COMC Solutions Page 8

The first gives x = 2 and the second does not have a solution since 2x > 0 for every real
number x.
Thus, x = 2.
(c) Let w = x2 −3x. Thus, the equation can be rewritten as w2 = 4−3(−w) or w2 −3w−4 = 0.
By (a), w = 4 or w = −1.
If w = 4, then x2 − 3x = 4 or x2 − 3x − 4 = 0 so x = 4 or x = −1 (by (a) again).
If w = −1, then x2 − 3x = −1 or x2 − 3x + 1 = 0. By the quadratic formula,
p √
3 ± 32 − 4(1)(1) 3± 5
x= =
2(1) 2
√
3± 5
Therefore, the solutions are x = 4, x = −1, and x = .
2
3. (a) (i) Setting m = n = 0, the given equation becomes

a0 + a0 = 12 a0 + 12 a0

or 2a0 = a0 or a0 = 0.
(ii) Setting m = 1 and n = 0, the given equation becomes

a1 + a1 = 12 a2 + 12 a0

Since a0 = 0 and a1 = 1, we obtain 1 + 1 = 21 a2 + 12 (0) or 21 a2 = 2 or a2 = 4.

One way to obtain a3 in an equation by using values of m and n that are as small as
possible is by setting m = 2 and n = 1. In this case, the equation becomes

a1 + a3 = 12 a4 + 12 a2 (∗)

Since we already know the values of a1 and a2 , then (∗) requires us to determine a4
in order to determine a3 .
We can determine a4 by setting m = 2 and n = 0, which gives

a2 + a2 = 12 a4 + 12 a0

from which we get 4 + 4 = 21 a4 + 21 (0) or 12 a4 = 8 or a4 = 16.

Substituting back into (∗), we get 1 + a3 = 12 (16) + 12 (4) or 1 + a3 = 8 + 2 or a3 = 9.
Thus, if a1 = 1, then a2 = 4 and a3 = 9.
(If we set m = k and n = 0, the given equation would become ak + ak = 12 a2k + 12 a0 .
Since a0 = 0, then a2k = 4ak , which is a more general statement than a2 = 4a1 and
a4 = 4a2 that were discovered above.)
2008 COMC Solutions Page 9

(b) Solution 1
Let M be a positive integer.
Setting m = n = M , the given equation becomes

b0 + b2M = b2M + b2M

or b0 + b2M = 2b2M or b0 = b2M .

This tells us that each even-numbered term in the sequence has the same value.
Next, set m = M and n = 0. In this case, the given equation becomes

bM + bM = b2M + b0

Since b2M = b0 , this equation becomes 2bM = 2b0 or bM = b0 .

Therefore, bM = b0 for every positive integer M .
Hence, all terms in the sequence have the same value, as required.

Solution 2
Let M be a positive integer.
Setting m = n = M , the given equation becomes

b0 + b2M = b2M + b2M

or b0 + b2M = 2b2M or b0 = b2M .

This tells us that each even-numbered term in the sequence has the same value.
Next, set m = 1 and n = 0. In this case, the given equation becomes

b1 + b1 = b2 + b0

Since b2 = b0 , this equation becomes 2b1 = 2b0 or b1 = b0 .

Next, setting m = M and n = M − 1, the equation becomes

b1 + b2M −1 = b2M + b2M −2

Since b1 = b0 (by the above) and b2M = b2M −2 = b0 (because these are even-numbered
terms), this equation becomes b0 + b2M −1 = b0 + b0 or b2M −1 = b0 .
Therefore, for any positive integer M , we have b2M = b0 and b2M −1 = b0 . In other words,
every even-numbered term and every odd-numbered term is equal to b0 .
Hence, all terms in the sequence have the same value, as required.
2008 COMC Solutions Page 10

4. Before solving the individual parts, we develop a formula that will allow us to calculate the
lengths of the medians of any triangle, given its side lengths.

Consider 4ABC with AB = c, AC = b and BC = a. Let P , Q, R be the midpoints of

BC, AC and AB, respectively, and ma , mb and mc the lengths of the medians AP , BQ and
CR, respectively.
A

B C
P

We calculate the length of ma first.

Consider 4ABC and 4ABP .
By the cosine law in 4ABC,

AC 2 = AB 2 + BC 2 − 2(AB)(BC) cos(∠ABC)
b2 = c2 + a2 − 2ca cos(∠ABC)
2ac cos(∠ABC) = c2 + a2 − b2
c 2 + a2 − b 2
cos(∠ABC) =
2ac
By the cosine law in 4ABP , using BP = 12 BC = 12 a,

AP 2 = AB 2 + BP 2 − 2(AB)(BP ) cos(∠ABP )
(ma )2 = c2 + ( 12 a)2 − 2c( 12 a) cos(∠ABC)
 2
c + a2 − b 2

2 2 1 2
(ma ) = c + 4 a − ac
2ac
2 2 1 2
(ma ) = c + 4 a − 2 c − 2 a + 12 b2
1 2 1 2

(ma )2 = 1 2
2
b + 12 c2 − 41 a2

(A second approach to calculating the length of ma would be to consider 4AP B and

4AP C, apply the cosine law in each focusing on vertex P , and use the fact that
cos(∠AP B) = − cos(∠AP C) since ∠AP B + ∠AP C = 180◦ .)
So (ma )2 = 21 b2 + 21 c2 − 14 a2 .
Using a similar approach (or by relabelling the diagram), we can determine that

(mb )2 = 21 a2 + 12 c2 − 14 b2

and
(mc )2 = 12 a2 + 21 b2 − 14 c2
2008 COMC Solutions Page 11

We are now ready to tackle parts (a) and (b).

(a) Suppose that a = 7, b = 13 and c = 17. From above,

(ma )2 = 12 (132 ) + 21 (172 ) − 14 (72 ) = 169

2
+ 289
2
− 49
4
= 229 − 49
4
= 867
4

and
(mb )2 = 21 (72 ) + 12 (172 ) − 14 (132 ) = 49
2
+ 289
2
− 169
4
= 169 − 169
4
= 507
4

and
(mc )2 = 21 (72 ) + 21 (132 ) − 14 (172 ) = 49
2
+ 169
2
− 289
4
= 109 − 289
4
= 147
4

Thus,
q q √
867 3(289) 3
ma = 4
= 4
= 17
2
q q √
507 3(169) 3
mb = 4
= 4
= 13 2
q q √
147 3(49) 3
mc = 4
= 4
= 7 2

Since a : b : c = 7 : 13 : 17 and mc : mb : ma = 7 : 13 : 17, then the lengths of the three

medians mc , mb , ma are in the same ratio as the lengths of the sides a, b, c, so the triangle
formed with side lengths mc , mb , ma is similar to the original triangle.
Therefore, the triangle with sides of length 7, 13 and 17 is automedian.
(Note that the fact that the lengths mc , mb , ma are in the correct ratio (and in the same
ratio as the side lengths of a triangle that already exists) also implies that a triangle can
be formed with sides of lengths mc , mb , ma .)
(b) Suppose that 4ABC is automedian with a < b < c.
The first thing that we need to do is to determine the relative lengths of ma , mb and mc .
In part (a), we saw that mc < mb < ma . We verify that this is always the case whenever
a < b < c.
First, we note that

(mc )2 = 21 a2 + 21 b2 − 14 c2 < 12 a2 + 12 c2 − 14 b2 = (mb )2

since b < c, so mc < mb since mc > 0 and mb > 0.

Similarly,
(mb )2 = 12 a2 + 21 c2 − 14 b2 < 12 b2 + 12 c2 − 14 a2 = (ma )2

since a < b, so mb < ma , which gives mc < mb < ma .

Therefore, since the triangles with side lengths a < b < c and mc < mb < ma are similar,
then mc = ka, mb = kb and ma = kc for some real number k.
2008 COMC Solutions Page 12

Thus, we have

(mc )2 = k 2 a2 = 1 2
2
a + 12 b2 − 14 c2
(mb )2 = k 2 b2 = 1 2
2
a + 12 c2 − 14 b2
(ma )2 = k 2 c2 = 1 2
2
b + 12 c2 − 14 a2

Adding these three equations, we obtain

k 2 (a2 + b2 + c2 ) = 34 a2 + 43 b2 + 34 c2
√
and so k 2 = 43 or k = 23 since k > 0.
We can then substitute this value of k 2 into any of the three equations for the lengths of
the medians. We substitute into the equation for mc , obtaining 34 a2 = 12 a2 + 12 b2 − 41 c2 or
1 2
4
a + 14 c2 = 21 b2 or a2 + c2 = 2b2 , as required.
(It is interesting to note that, even if a triangle is not automedian, its three medians will
always form a triangle, and a triangle whose area is 43 of the area of the original triangle.
This is consistent with the ratio of similarity found here.)
(c) We approach this part by trying to find a few triples of positive integers (a, b, c) with
a < b < c that satisfy a2 + c2 = 2b2 then by finding a pattern to write down a general form
for an infinite family that appears to satisfy a2 + c2 = 2b2 . We will then need to prove
that this infinite family works. We do not yet know that if a2 + c2 = 2b2 , then the triangle
is automedian, but we will prove this too. (We proved the converse in (b).)
We are looking for triples (a, b, c) that satisfy a2 +c2 = 2b2 , or equivalently, c2 −b2 = b2 −a2 .
(In other words, a2 , b2 , c2 form an arithmetic sequence.)
After some trial and error, we can find the triples (1, 5, 7), (7, 13, 17), (17, 25, 31) that
satisfy this equation. Note that the largest number in each triple becomes the smallest
in the next triple and the differences between consecutive numbers in each triple are two
more than in the previous triple. (Note that the triple (1, 5, 7) is a triple that satisfies
a2 + c2 = 2b2 , but a triangle cannot be formed with sides of these lengths. We keep this
triple in our list in any event as it helps us find a pattern.)
We try to write down an infinite family (an , bn , cn ) that follows from these three triples.
If we label the first triple with n = 1, the second with n = 2 and the third with n = 3, we
might see that 7 = 2(22 ) − 1, 17 = 2(32 ) − 1 and 31 = 2(42 ) − 1.
Thus, we try cn = 2(n + 1)2 − 1 = 2n2 + 4n + 1.
The difference between b and c in the first is 2, in the second is 4 and in the third is 6, so
we try bn = cn − 2n = 2n2 + 2n + 1.
The difference between a and b in the first is 4, in the second is 6 and in the third is 8, so
we try an = bn − 2(n + 1) = 2n2 − 1.
Thus, we try the triples (an , bn , cn ) = (2n2 − 1, 2n2 + 2n + 1, 2n2 + 4n + 1) for n a positive
2008 COMC Solutions Page 13

integer. (The three triples found above do fit this formula.)

Since n is a positive integer, these are triples of positive integers. We need to check that
• these triples actually form triangles,
• the triangles formed are automedian, and
• no two of the triangles formed are similar.
Step 1: If n ≥ 2, then (an , bn , cn ) are the side lengths of a triangle
For this to be the case, the three lengths must obey the triangle inequality. That is, the
sum of any two of the lengths must be greater than the third length. Since an < bn < cn ,
this means that we only need to check if an + bn > cn (since an + cn > bn and bn + cn > an
are automatic).
Here, (2n2 − 1) + (2n2 + 2n + 1) > (2n2 + 4n + 1) is equivalent to 4n2 + 2n > 2n2 + 4n + 1
which is equivalent to 2n2 − 2n > 1 which is equivalent to 2n(n − 1) > 1, which is true
when n ≥ 2, as the left side will be at least 2(2)(1) = 4.

Step 2: If n ≥ 2, then (an , bn , cn ) are the side lengths of an automedian triangle

We do this in two steps – we show that (an )2 + (cn )2 = 2(bn )2 and then show that if a
triangle has a2 + c2 = 2b2 , then it is automedian.
We show that (an )2 + (cn )2 = 2(bn )2 by showing that (cn )2 − (bn )2 = (bn )2 − (an )2 .
We have

(cn )2 −(bn )2 = (cn +bn )(cn −bn ) = (4n2 +6n+2)(2n) = 4n(2n2 +3n+1) = 4n(n+1)(2n+1)

and

(bn )2 − (an )2 = (bn + an )(bn − an ) = (4n2 + 2n)(2n + 2) = 4n(2n + 1)(n + 1)

as required.
Next, suppose that a triangle with side lengths (a, b, c) satisfies a2 + c2 = 2b2 .
Then, from our formulae above,

(mc )2 = 1 2
2
a + 21 b2 − 14 c2 = 12 a2 + 21 b2 − 14 (2b2 − a2 ) = 34 a2

(mb )2 = 1 2
2
a + 21 c2 − 14 b2 = 12 a2 + 21 (2b2 − a2 ) − 14 b2 = 34 b2

(ma )2 = 1 2
2
b + 21 c2 − 14 a2 = 12 b2 + 21 c2 − 41 (2b2 − c2 ) = 34 c2
√ √ √
so mc = 23 a, mb = 23 b, and ma = 23 c.
Thus, if a2 + c2 = 2b2 , then the triangle is automedian.
Therefore, if n ≥ 2, then (an , bn , cn ) are the side lengths of an automedian triangle.
2008 COMC Solutions Page 14

Step 3: No two (an , bn , cn ) with n ≥ 2 form similar triangles

Suppose that (an , bn , cn ) and (am , bm , cm ) are two similar triangles, with am = dan , bm =
dbn and cm = dcn for some real number d.

2m2 − 1 = d(2n2 − 1)
2m2 + 2m + 1 = d(2n2 + 2n + 1)
2m2 + 4m + 1 = d(2n2 + 4n + 1)

Subtracting the second of these equations from the third, we obtain 2m = 2dn or m = dn.
Subtracting the first of these equations from the second, we obtain 2m + 2 = d(2n + 2) or
m + 1 = d(n + 1).
Since m = dn, we obtain dn + 1 = dn + d or d = 1.
Since d = 1, then n = m. Therefore, if n 6= m, the triangles with side lengths (am , bm , cm )
and (an , bn , cn ) are not similar.

Therefore, the infinite family (an , bn , cn ) = (2n2 − 1, 2n2 + 2n + 1, 2n2 + 4n + 1) with

n a positive integer and n ≥ 2 is an infinite family of automedian triangles with integer
side lengths, no two of which are similar.
(There are other possible infinite families that work too.)
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING
presents the

Sun Life Financial Canadian

Open Mathematics Challenge

Wednesday, November 25, 2009

Time: 2 12 hours 2009

c Canadian Mathematical Society
Calculators are NOT permitted.
Do not open this booklet until instructed to do so.
There are two parts to this paper.
PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value
for each question by entering the correct answer(s) in the space provided. If your answer is
incorrect, any work that you do will be considered for part marks, provided that it is done
in the space allocated to that question in your answer booklet.
PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must
be written in the appropriate location in the answer booklet. Rough work should be done
separately. If you require extra pages for your finished solutions, paper will be provided by
your supervising teacher. Any extra papers should be placed inside your answer booklet. Be
sure to write your name and school name on any inserted pages.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution,
poorly presented, will not earn full marks.
NOTES:
At the completion of the contest, insert the information sheet inside the answer
booklet.
The names of top scoring competitors will be published on the Web sites of the
CMS and CEMC.
Sun Life Financial Canadian Open Mathematics Challenge
NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations
√ and answers will be expressed as
exact numbers such as 4π, 2 + 7, etc., rather than as 12.566 . . .
or 4.646 . . ..
4. Calculators are not allowed.
5. Diagrams are not drawn to scale. They are intended as aids only.

PART A
1. Determine the value of
−1 + 2 − 3 + 4 − 5 + 6 − 7 + 8 − 9 + 10 − 11 + 12 − 13 + 14 − 15 + 16 − 17 + 18 .

2. If 3 × 10a + 5 × 10b + 7 × 10c = 5073, and a, b and c are non-negative integers, what
is the value of a + b + c?
3. Soroosh has 10 coins, each of which is either a quarter (worth 25 cents) or a dime
(worth 10 cents). The total value of the dimes is greater than the total value of the
quarters. What is the smallest possible number of dimes that he could have?
4. The positive integers 15, 12 and n have the property that the product of any two of
them is divisible by the third. Determine the smallest possible value of n.
5. In the diagram, there are three islands labelled
A, B and C. Islands A and B are joined by a
bridge, as are islands B and C. Maya begins her A
B
journey on island A and travels between the islands
by bridge only. She records the sequence of islands
that she visits. She does not necessarily visit all
three islands. If Maya makes 20 bridge crossings C
in total, how many possible sequences of islands
A, B and C could she travel along?

6. A polygon is called regular if all of its sides are

equal in length and all of its interior angles are B C
equal in size. In the diagram, a portion of a regular D
polygon is shown. If ∠ACD = 120◦ , how many A
sides does the polygon have?

7. Determine all angles θ with 0◦ ≤ θ ≤ 360◦ such that log2 (−3 sin θ) = 2 log2 (cos θ)+1.
8. Determine all triples (a, b, c) of positive integers such that a! = 4(b!) + 10(c!).
Note: If n is a positive integer, the symbol n! (read as “n factorial”) is used to
represent the product of the positive integers from 1 to n; that is,
n! = n(n − 1)(n − 2) · · · (3)(2)(1)
For example, 5! = 5(4)(3)(2)(1).
 PART B
C
1. (a) In the diagram, ∠CAB = 90◦ , AB = 9 and
BC = 15. Determine the area of 4ABC.

15

A 9 B

(b) From part (a), 4ABC has side BA extended C

to D. If the area of 4CDB is 84, determine
the length of CD.
15

D A 9 B

(c) In 4P QR, P Q = 25 and QR = 25. If the area P

of 4P QR is 300, determine the length of P R.

Q T R

2. Triangle P QR has vertices P (7, 13), Q(19, 1) and R(1, 1). Point M (4, 7) is the
midpoint of P R; the midpoint of P Q is N .

(a) Determine the equation of the median of the triangle that passes through points
Q and M .
(b) Determine the coordinates of G, the point of intersection of RN and QM .
(c) Point F is on P R so that QF is perpendicular to P R. Point T is on P Q so
that RT is perpendicular to P Q. Determine the coordinates of H, the point of
intersection between altitudes QF and RT .
(d) Determine which of G and H is closer to the origin.
 Canadian Open
Mathematics
Challenge

Financial
(English)

Sun Life
2009
3. Suppose that f and g are functions.
We say that the real number c is a real fixed point of f if f (c) = c.
We say that f and g commute if f (g(x)) = g(f (x)) for all real numbers x.

(a) If f (x) = x2 − 2, determine all real fixed points of f .

(b) If f (x) = x2 − 2, determine all cubic polynomials g that commute with f .
(c) Suppose that p and q are real-valued functions that commute.
If 2 [q(p(x))]4 + 2 = [p(x)]4 + [p(x)]3 for all real numbers x, prove that q has no
real fixed points.

4. For each positive integer n, define f (n) to be the smallest positive integer s for which
1+2+3+· · ·+(s−1)+s is divisible by n. For example, f (5) = 4 because 1+2+3+4
is divisible by 5 and none of 1, 1 + 2, or 1 + 2 + 3 is divisible by 5.

(a) Determine all positive integers a for which f (a) = 8.

(b) Prove that there are infinitely many odd positive integers b for which

f (b + 1) − f (b) > 2009 .

(c) Determine, with proof, the smallest positive integer k for which the equation
f (c) = f (c + k) has an odd positive integer solution for c.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

presents the

Sun Life Financial

Canadian Open Mathematics Challenge
Wednesday, November 25, 2009

Solutions

©2009 Canadian Mathematical Society

2009 COMC Solutions Page 2

Part A

1.
−1 + 2 − 3 + 4 − 5 + 6 − 7 + 8 − 9 + 10 − 11 + 12 − 13 + 14 − 15 + 16 − 17 + 18
= (2 − 1) + (4 − 3) + (6 − 5) + (8 − 7) + (10 − 9) + (12 − 11) +
(14 − 13) + (16 − 15) + (18 − 17)
= 1+1+1+1+1+1+1+1+1
= 9

Answer: 9

2. We write 5073 in place value notation as 5 × 1000 + 7 × 10 + 3 or 5 × 103 + 7 × 101 + 3 × 100 .

Thus, if a = 0, b = 3 and c = 1, then the left side (3 × 10a + 5 × 10b + 7 × 10c ) equals 5073.
Any other combination of values for a, b and c will not give 5073.
Therefore, a + b + c = 0 + 3 + 1 = 4.
(We can show that the only possibility is a = 0, b = 3 and c = 1. We start by noting that
the remainder when the right side is divided by 10 is 3, so the remainder when the left side is
divided by 10 must also be 3. If a = 0 and each of b and c is larger than 0, then the remainder
on the left side will be 3. If more than one of a, b and c equals 0, then we can see by trying
the possibilities that the remainder on the left side cannot be 3.
Therefore, a = 0 and b and c are positive.
We can then subtract 3 from both sides and divide by 10 to obtain the new equation

5 × 10b−1 + 7 × 10c−1 = 507

and repeat the argument to show that c = 1 and then b = 3.)

Answer: 4

3. Solution 1
Suppose that Soroosh has d dimes.
Since he has 10 coins, then he has 10 − d quarters.
The value of the dimes is 10d cents and the value of the quarters is 25(10 − d) cents.
Since we want the value of the dimes to be larger than the value of the quarters, then

10d > 25(10 − d)

10d > 250 − 25d
35d > 250
7d > 50
50
d > 7
= 7 17
2009 COMC Solutions Page 3

Since d is an integer, then d ≥ 8, so the smallest possible number of dimes is 8.

Solution 2
We proceed by systematic trial and error.
If Soroosh has 4 quarters and 6 dimes, then the quarters are worth 4 × 25 = 100 cents and
the dimes are worth 6 × 10 = 60 cents. (Any smaller number of dimes than 6 makes the value
of the dimes smaller and the value of quarters larger, so the number of dimes must be greater
than 6.)
If Soroosh has 3 quarters and 7 dimes, then the quarters are worth 3 × 25 = 75 cents and the
dimes are worth 7 × 10 = 70 cents.
If Soroosh has 2 quarters and 8 dimes, then the quarters are worth 2 × 25 = 50 cents and the
dimes are worth 8 × 10 = 80 cents.
Therefore, the smallest number of dimes for which the value of the dimes is greater than the
value of the quarters is 8.

Answer: 8

4. Solution 1
From the given conditions, we want 15(12) = 180 to be divisible by n, and 15n to be divisible
by 12, and 12n to be divisible by 15.
For 15n to be divisible by 12, then 15n is a multiple of 12, or 15n = 12m for some positive
integer m. Simplifying, we see that 5n = 4m.
Since the right side is divisible by 4, then the left side must be divisible by 4, so n must be
divisible by 4.
For 12n to be divisible by 15, we must have 12n = 15k for some positive integer k. Simplifying,
we see that 4n = 5k.
Since the right side is a multiple of 5, then the left side must be a multiple of 5, so n must be
a multiple of 5.
Therefore, n must be a multiple of 4 and a multiple of 5.
This tells us that n must be a multiple of 20.
Since we want n to be as small as possible, then we try n = 20, since this is the smallest positive
multiple of 20.
If n = 20, then it is true that 15(12) = 180 is divisible by 20, and 15(20) = 300 is divisible by
12, and 12(20) = 240 is divisible by 15.
Thus, the smallest possible value of n is 20.
(We could instead have started with the condition 180 is divisible by n, listed the positive
divisors of 180, and then tried these divisors starting from the smallest until we found a divisor
that satisfied the other two conditions.)
2009 COMC Solutions Page 4

Solution 2
First, we note that the prime factorizations of 12 and 15 are 12 = 22 · 3 and 15 = 3 · 5.
Since 12 | 15n, then n must contain a factor of 22 since 12 does and 15 is not divisible by 2.
Since 15 | 12n, then n must contain a factor of 5 since 15 does and 12 does not.
Since n | 12(15) = 22 · 32 · 5, then n cannot contain more than 2 factors of 2 and 1 factor of 5,
since 12(15) contains only 2 factors of 2 and 1 of 5.
Therefore, to make n as small as possible, n must be exactly 22 · 5 = 20.
(Can you find the other values of n that work?)

Answer: 20

5. Solution 1
We represent the sequence of islands that Maya visits as a sequence of letters starting with A.
Since she makes 20 bridge crossings, then she visits 21 islands in total, so the sequence contains
21 letters.
If Maya is on island A or on island C, then the next island that she visits must be island B,
since it is the only island connected to A and the only island connected to C.
If Maya is on island B, then Maya has two choices: cross to island A or cross to island C.
Since she starts at island A, then the second letter in the sequence must be B, since she must
cross to island B.
The third letter can be either A or C, as she has a choice from island B.
Once on island A or C, she must cross back to island B, so the fourth letter is B.
She is thus in the same situation as she was after her first crossing, and so the pattern continues.
In other words, the letters in odd positions in the sequence, starting at the third, can be either
A or C, and the letters in the even positions must be B.
We can represent the sequence then as follows:

A A A A A A A A A A
AB B B B B B B B B B
C C C C C C C C C C

Thus, there are 10 positions in the sequence where Maya has 2 choices and the rest of the
positions are fixed.
Thus, there are 210 = 1024 possible sequences.

Solution 2
Define Sn to be the number of sequences starting at island A with n crossings. We want to
determine S20 .
Note that S2 = 2 (A to B to A, and A to B to C are the possible routes).
First, we note that islands A and C are interchangeable, since we could switch the labels and
there would be no structural difference to the diagram.
2009 COMC Solutions Page 5

Thus, the number of sequences of a given length starting at A is the same as the number of
sequences of the same length starting at C.
Suppose that Maya is going to make a journey with t crossings, where t is an even integer with
t ≥ 4. There are St such sequences.
After two crossings, Maya would be at either island A or island C and would have t−2 crossings
remaining.
But starting at either island A or C, there are St−2 sequences that she could follow.
Therefore, St = St−2 + St−2 = 2St−2 .
Now,
S20 = 2S18 = 2(2S16 ) = 22 S16 = 23 S14 = · · · = 29 S2 = 29 (2) = 210 = 1024
Therefore, there are 1024 possible sequences.

Answer: 210 = 1024

6. Solution 1
Suppose that the polygon has n sides.
Extend CB outside ofthe polygon.
◦ Since the sum of the exterior angles in a polygon is always
360
360◦ , then ∠ABE = , since there will be n equal exterior angles.
n

B C
E D
A
 ◦
◦ 360
Thus, ∠ABC = 180 − and this will also be the measure of ∠BCD, since the polygon
n
is regular.
Since the polygon is regular, then AB = BC, so 4ABC is isosceles, which means that we have
∠BAC = ∠BCA.
Therefore,
   ◦   ◦
1 ◦ 1 ◦ ◦ 360 180
∠BCA = (180 − ∠ABC) = 180 − 180 − =
2 2 n n
But ∠BCD = ∠BCA + ∠ACD, so
 ◦  ◦
◦ 360 180
180 − = + 120◦
n n
540
60 =
n
n = 9

Therefore, the polygon has 9 sides.

2009 COMC Solutions Page 6

Solution 2
Suppose that the polygon has n sides.
Let O be the centre of the polygon. Join O to each of A, B, C, and D.
Since the polygon is regular, then the angle subtended at O by each of the n sides will be equal,
and these angles all add to 360◦ .

B C
D
A

O
 ◦
360
Since there are n equal central angles, then ∠AOB = ∠BOC = ∠COD = .
 ◦  ◦ n
360 720
This also tells us that ∠AOC = ∠AOB + ∠BOC = 2 = .
n n
Since the polygon is regular, then OA = OC = OD, which tells us that 4AOC and 4COD
are both isosceles.
Thus,   ◦   ◦
1 ◦ 1 ◦ 720 ◦ 360
∠ACO = (180 − ∠AOC) = 180 − = 90 −
2 2 n n
and   ◦   ◦
1 1 360 180
∠DCO = (180◦ − ∠COD) = ◦
180 − ◦
= 90 −
2 2 n n
Now, ∠ACD = ∠ACO + ∠DCO, so
 ◦  ◦
◦ ◦ 360 ◦ 180
120 = 90 − + 90 −
n n
540
= 60
n
n = 9

Therefore, the polygon has 9 sides.

Answer: 9
2009 COMC Solutions Page 7

7. Using the rules for manipulating logarithms and trigonometric functions,

log2 (−3 sin θ) = 2 log2 (cos θ) + 1

log2 (−3 sin θ) = log2 (cos2 θ) + log2 2
log2 (−3 sin θ) = log2 (2 cos2 θ)
2
2log2 (−3 sin θ) = 2log2 (2 cos θ)

−3 sin θ = 2 cos2 θ
−3 sin θ = 2(1 − sin2 θ) (since cos2 θ + sin2 θ = 1)
2 sin2 θ − 3 sin θ − 2 = 0
(2 sin θ + 1)(sin θ − 2) = 0

Therefore, sin θ = − 21 or sin θ = 2.

The second possibility is inadmissible, so sin θ = − 12 .
Since 0◦ ≤ θ ≤ 360◦ and sin θ = − 21 , then θ = 210◦ or θ = 330◦ .
But, we also need cos θ > 0 to satisfy the domains of the logarithms in the original equation.
Therefore, θ = 210◦ is inadmissible (since it is in the third quadrant and cos θ < 0), but
θ = 330◦ is the admissible (since it is in the fourth quadrant and cos θ > 0).
√
Checking, sin(330◦ ) = − 21 and cos(330◦ ) = 23 , so the left side of the original equation equals
√
log2 23 and the right side equals 2 log2 23 + 1 = log2 43 + log2 (2) = log2 32 , as required.

Therefore, θ = 330◦ .

Answer: 330◦

8. We examine three cases: b = c, b > c and b < c.

Note that, in any of these cases, we have a! > 4(b!) and a! > 10(c!) so a > b and a > c.

Case 1: b = c
a!
Here, the equation becomes a! = 14(b!) or = 14 or a(a − 1) · · · (b + 2)(b + 1) = 14.
b!
The expression on the left side is a single integer (if a = b + 1) or the product of 2 or more
consecutive integers.
Since 14 = 2(7), then 14 cannot be written as the product of two or more consecutive integers.
Therefore, the expression on the left must be a single integer.
Therefore, a = b + 1 = 14, so b = c = 13.
Thus, the only solution in this case is (a, b, c) = (14, 13, 13).

Case 2: b > c
a! 10(c!)
Dividing both sides by b!, the equation becomes =4+ or
b! b!
10
a(a − 1) · · · (b + 2)(b + 1) = 4 +
b(b − 1) · · · (c + 2)(c + 1)
2009 COMC Solutions Page 8

Since the left side is an integer, then the right side must be an integer. Thus,
10
b(b − 1) · · · (c + 2)(c + 1)
is an integer, which means that 10 is divisible by b(b − 1) · · · (c + 2)(c + 1), which is again a
single integer (if b = c + 1) or the product of 2 or more consecutive integers, each of which is
at least 2 (since c ≥ 1).
As in Case 1, the only possibility is that the denominator is one of the single integers 10, 5
and 2.
The possibilities are thus b = c + 1 = 10 (whence b = 10 and c = 9), b = c + 1 = 5 (whence
b = 5 and c = 4), or b = c + 1 = 2 (whence b = 2 and c = 1).
If b = 10 and c = 9, the right side of the initial equation becomes 4(10!)+10(9!) or 4(10!)+(10!),
which equals 5(10!). This number is not a factorial because it is bigger than 10! and less than
11!. There is thus no possible value for a.
If b = 5 and c = 4, the right side of the initial equation becomes 4(5!) + 10(4!), which equals
480 + 240 = 720 = 6!, and so a = 6.
If b = 2 and c = 1, the right side of the initial equation becomes 4(2!) + 10(1!) = 8 + 10 = 18.
This number is not a factorial because it is bigger than 3! and less than 4!. There is thus no
possible value for a.
Therefore, the only solution in this case is (a, b, c) = (6, 5, 4).

Case 3: b < c
a! 4(b!)
Dividing both sides by c!, the equation becomes = + 10 or
c! c!
4
a(a − 1) · · · (c + 2)(c + 1) = + 10
c(c − 1) · · · (b + 2)(b + 1)
Since the left side is an integer, then the right side must be an integer. Thus,
4
c(c − 1) · · · (b + 2)(b + 1)
is an integer, which means that 4 is divisible by c(c − 1) · · · (b + 2)(b + 1), which is again a single
integer (if c = b + 1) or the product of 2 or more consecutive integers, each of which is at least
2 (since b ≥ 1).
As in Case 2, the only possibility is that the denominator is one of the single integers 4 and 2.
The possibilities are thus c = b + 1 = 4 (whence c = 4 and b = 3) or c = b + 1 = 2 (whence
c = 2 and b = 1).
If c = 4 and b = 3, the right side of the initial equation becomes 4(3!) + 10(4!) which equals
24 + 240 = 264. This number is not a factorial because it is bigger than 5! and less than 6!.
There is thus no possible value for a.
If c = 2 and b = 1, the right side of the initial equation becomes 4(1!) + 10(2!) which equals
2009 COMC Solutions Page 9

4 + 20 = 24 = 4!, and so a = 4.
Therefore, the only solution in this case is (a, b, c) = (4, 1, 2).

Therefore, the three solutions are (a, b, c) = (14, 13, 13), (6, 5, 4), and (4, 1, 2).

Answer: (a, b, c) = (14, 13, 13), (6, 5, 4), and (4, 1, 2)

2009 COMC Solutions Page 10

Part B

1. (a) Solution 1
By the Pythagorean Theorem, since AC > 0,
√ √ √ √
AC = CB 2 − AB 2 = 152 − 92 = 225 − 81 = 144 = 12

Therefore, the area of 4ABC is 21 (AB)(AC) = 21 (9)(12) = 54.

Solution 2
Since 4ABC is right-angled at A and AB : CB = 9 : 15 = 3 : 5, then 4ABC is similar
to a 3 : 4 : 5 triangle.
Therefore, AC = 43 AB = 43 (9) = 12.
Therefore, the area of 4ABC is 21 (AB)(AC) = 21 (9)(12) = 54.
(b) From (a), AC = 12.
Since the area of 4CDB is 84, then 12 (DB)(AC) = 84 or 21 (DB)(12) = 84 or 6(DB) = 84.
Therefore, DB = 14 and so DA = DB − AB = 14 − 9 = 5.
Lastly, by the Pythagorean Theorem, since CD > 0, we have
√ √ √ √
CD = DA2 + AC 2 = 52 + 122 = 25 + 144 = 169 = 13

(c) Since the area of 4P QR is 300, then 21 (QR)(P T ) = 300 or 1

2
(25)(P T ) = 300
or 25(P T ) = 600 or P T = 24.
By the Pythagorean Theorem, since QT > 0,
p √ √ √
QT = P Q2 − P T 2 = 252 − 242 = 625 − 576 = 49 = 7

Thus, T R = QR − QT = 25 − 7 = 18.
In 4P T R, we now have P T = 24, ∠P T R = 90◦ , and T R = 18.
Lastly, by the Pythagorean Theorem, since P R > 0, we have
√ √ √ √
PR = P T 2 + T R2 = 242 + 182 = 576 + 324 = 900 = 30

(The given diagram implies that T , the foot of the altitude from P to QR, lies between
Q and R, although the problem does not explicitly state this. If this implied restriction is
removed, there is a second case with P R = 40, ∠P QR obtuse, and T to the left of Q.)

7−1 6 2
2. (a) The line through points Q and M has slope = = − and so has equation
4 − 19 −15 5
y − 7 = − 25 (x − 4) or y = − 25 x + 43
5
.
(b) Solution 1
1
+ 19), 12 (13 + 1) = (13, 7).


The midpoint, N , of P Q has coordinates 2
(7
2009 COMC Solutions Page 11

7−1 6 1
The line through points R and N has slope = = and so has equation
13 − 1 12 2
y − 1 = 12 (x − 1) or y = 21 x + 12 .
At the point of intersection of y = − 25 x + 43
5
and y = 21 x + 21 , the values of y are equal so:

1 1
2
x + 2
= − 25 x + 43
5
5 4 86 5
10
x + 10
x = 10
− 10
9 81
10
x = 10

x = 9

Since the x-coordinate of G is 9, then the y coordinate is 12 (9) + 1

2
= 10
2
= 5, so the
coordinates of G are (9, 5).

Solution 2
Point G is the intersection of two of the medians of 4P QR, and so is the centroid of
4P QR. (In fact, all three medians will pass through G.)
The coordinates of the centroid are the averages of the coordinates of the three vertices.
Thus, the coordinates of G are 13 (7 + 1 + 19), 31 (13 + 1 + 1) = (9, 5).

(c) Solution 1
13 − 1 12
The slope of P R is = = 2.
7−1 6
Since QF is perpendicular to P R, then its slope is the negative reciprocal of 2, or − 12 .
Thus, the line through Q and F has equation y − 1 = − 21 (x − 19) or y = − 21 x + 212
.
13 − 1 12
The slope of P Q is = = −1.
7 − 19 −12
Since RT is perpendicular to P Q, then its slope is the negative reciprocal of −1, or 1.
Thus, the line through R and T has equation y − 1 = 1(x − 1) or y = x.
At the point of intersection of these lines, the values of y are equal so:

x = − 12 x + 21
2
3 21
2
x = 2

x = 7

Since the x-coordinate of H is 7, then the y coordinate is also 7, since H lies on the line
y = x.
Thus, the coordinates of H are (7, 7).

Solution 2
The three altitudes of 4P QR all pass through H.
Since side QR of 4P QR is horizontal, then the altitude from P must be vertical.
Since the x-coordinate of Q is 7, then the equation of the altitude through P is x = 7.
We can then determine the equation of a second altitude, say the altitude through R and
2009 COMC Solutions Page 12

T as in Solution 1, to be y = x.
Therefore, point H lies at the intersection of y = x and x = 7, which is the point (7, 7).
p √ √
(d) The distance between O(0, 0) and G is (9 − 0)2 + (5 − 0)2 = 81 + 25 = 106.
p √ √
The distance between O and H is (7 − 0)2 + (7 − 0)2 = 49 + 49 = 98.
√ √
Since 98 < 106, then H is closer to the origin than G.

3. (a) To find all real fixed points, we need to solve the equation f (c) = c.
Since f (x) = x2 − 2, we solve c2 − 2 = c or c2 − c − 2 = 0.
Thus, (c − 2)(c + 1) = 0, so the real fixed points are c = 2 and c = −1.
(b) Solution 1
Suppose that g(x) = ax3 + bx2 + dx + e for some real coefficients a, b, d, e with a 6= 0 (since
g(x) is cubic). Suppose also that f and g commute (that is, f (g(x)) = g(f (x)) for all real
numbers x). Now,

f (g(x)) = f (ax3 + bx2 + dx + e)

= (ax3 + bx2 + dx + e)2 − 2
= a2 x6 + b2 x4 + d2 x2 + e2 + 2abx5 + 2adx4 + 2aex3 + 2bdx3 + 2bex2 + 2dex − 2
= a2 x6 + 2abx5 + (b2 + 2ad)x4 + (2ae + 2bd)x3 + (d2 + 2be)x2 + 2dex + (e2 − 2)

and

g(f (x)) = g(x2 − 2)

= a(x2 − 2)3 + b(x2 − 2)2 + d(x2 − 2) + e
= a(x6 − 6x4 + 12x2 − 8) + b(x4 − 4x2 + 4) + d(x2 − 2) + e
= ax6 + (−6a + b)x4 + (12a − 4b + d)x2 + (−8a + 4b − 2d + e)

Since f (g(x)) = g(f (x)) for all real numbers x, then the coefficients on the left side must
equal the coefficients on the right side.
Therefore,

a2 = a (1)
2ab = 0 (2)
b2 + 2ad = −6a + b (3)
2ae + 2bd = 0 (4)
d2 + 2be = 12a − 4b + d (5)
2de = 0 (6)
e2 − 2 = −8a + 4b − 2d + e (7)
2009 COMC Solutions Page 13

From (1), a2 − a = 0 or a(a − 1) = 0 and so a = 1 or a = 0. Since a 6= 0, then a = 1.

Substituting a = 1 into (2), we obtain 2b = 0 or b = 0.
Substituting a = 1 and b = 0 into (3), we obtain 0 + 2(1)d = −6(1) + 0 or 2d = −6, so
d = −3.
Substituting d = 3 into (6), we obtain −6e = 0 so e = 0.
We can check that a = 1, b = 0, d = −3, and e = 0 satisfy equations (4), (5) and (7).
Therefore, g(x) = 1x3 + 0x2 + (−3)x + 0 = x3 − 3x is the only cubic polynomial that
commutes with f (x).
(We can check by expanding that (x3 − 3x)2 − 2 = (x2 − 2)3 − 3(x2 − 2).)

Solution 2
Suppose that g(x) = ax3 + bx2 + dx + e for some real coefficients a, b, d, e with a 6= 0 (since
g(x) is cubic). Suppose also that f and g commute (that is, f (g(x)) = g(f (x)) for all real
numbers x). Now,

f (g(x)) = f (ax3 + bx2 + dx + e)

= (ax3 + bx2 + dx + e)2 − 2

and

g(f (x)) = g(x2 − 2)

= a(x2 − 2)3 + b(x2 − 2)2 + d(x2 − 2) + e

Since f (g(x)) = g(f (x)) for all real numbers x, then the coefficients on the left side must
equal the coefficients on the right side when expanded.
On the left side, the only term involving x6 will come from squaring the term ax3 , so the
coefficient of x6 is a2 .
On the right side, the only term involving x6 comes from a(x2 − 2)3 ; since the coefficient
of x6 in (x2 − 2)3 is 1, then the coefficient of x6 on the right side is a.
Therefore, a2 = a or a2 − a = 0 or a(a − 1) = 0 and so a = 1 or a = 0. Since a 6= 0, then
a = 1.
When the expansion on the right side is done, there will be only even powers of x.
Thus, the left side cannot contain any odd powers of x.
When the left side is expanded, we will obtain a term 2abx5 . Thus, 2ab = 0. Since a = 1,
then b = 0.
Therefore, we have

(x3 + dx + e)2 − 2 = (x2 − 2)3 + d(x2 − 2) + e

On the left side, the only way to obtain an x4 term is by multiplying x3 and dx, so the x4
term on the left side is 2dx4 . On the right side, the only x4 term is from the expansion of
2009 COMC Solutions Page 14

(x2 − 2)3 and so is 3(−2)(x2 )2 = −6x4 .

Comparing coefficients, 2d = −6 or d = −3.
Since the right side contains no term involving x1 , then the coefficient of x on the left side
is 0.
When (x3 − 3x + e)2 − 2 is expanded, the term involving x1 will be 2(−3x)e = −6ex and
so −6e = 0 or e = 0.
Therefore, g(x) = 1x3 + 0x2 + (−3)x + 0 = x3 − 3x is the only cubic polynomial that
commutes with f .
(We can check by expanding that (x3 − 3x)2 − 2 = (x2 − 2)3 − 3(x2 − 2).)
(c) We prove the desired result by contradiction.
Suppose that q(x) has a real fixed point c; that is, suppose that q(c) = c.
Since p and q commute, then p(q(x)) = q(p(x)) for all real numbers x.
In particular, p(q(c)) = q(p(c)).
Since q(c) = c, this equation becomes p(c) = q(p(c)).
Since the equation that we are given is true for all real numbers x, then it is true for x = c,
so

2 [q(p(c))]4 + 2 = [p(c)]4 + [p(c)]3

2 [p(c)]4 + 2 = [p(c)]4 + [p(c)]3 (from above)
[p(c)]4 − [p(c)]3 = −2

Define u = p(c). Note that u is a real number since c is a real number and p(x) has real
coefficients.
To arrive at our contradiction, we show that there are no real numbers u for which
u4 − u3 = −2.
We do this by looking at three cases: u ≥ 1, u ≤ 0, and 0 < u < 1.
If u ≥ 1, then u4 = u(u3 ) ≥ 1(u3 ) = u3 so u4 − u3 ≥ 0, which means that u4 − u3 6= −2.
If u ≤ 0, then u3 ≤ 0, which means that u4 − u3 ≥ u4 ≥ 0, so u4 − u3 6= −2.
If 0 < u < 1, then u4 > 0 and u3 < 1, and so −u3 > −1. Thus, u4 − u3 > 0 + (−1) = −1,
so u4 − u3 6= −2.
In all cases, u4 − u3 6= −2.
This is a contradiction, since we have determined that [p(c)]4 − [p(c)]3 = −2.
Therefore, our original assumption must be incorrect, so q(x) cannot have a real fixed
point.

4. (a) We want to find all positive integers a for which the smallest positive s with the property
that a divides into 1 + 2 + 3 + · · · + s is s = 8.
Note that 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 = 36. (We use the term triangular number to
2009 COMC Solutions Page 15

mean a positive integer of the form 1 + 2 + 3 + · · · + s.) The previous triangular numbers
are 1, 3, 6, 10, 15, 21, 28. Therefore, we want to find all positive integers a that are divisors
of 36 but not of any of 1, 3, 6, 10, 15, 21, 28.
The divisors of 36 are 1, 2, 3, 4, 6, 9, 12, 18, 36.
The divisors 1, 2, 3, 4, 6 each divide into at least one of 6 and 28, so f (1) 6= 8, f (2) 6= 8,
f (3) 6= 8, f (4) 6= 8, and f (6) 6= 8.
The divisors 9, 12, 18, 36 do not divide into any triangular number smaller than 36.
Therefore, the complete solution to f (a) = 8 is a = 9, 12, 18, 36.
(b) For m a positive integer, we define T (m) = 1 + 2 + · · · + m = 21 m(m + 1). (T (m) is the
mth triangular number.)
First, we show that if T (z) is a multiple of w, then f (w) ≤ z:
We know that f (w) is the smallest integer m for which 1 + 2 + 3 + · · · + m = T (m)
is a multiple of w.
Suppose that T (z) is a multiple of w.
If z is the smallest positive integer with this property, then f (w) = z; otherwise,
z is not the smallest such positive integer, so f (w) < z.
In either case, f (w) ≤ z.
Next, we show that if y is an odd positive integer with y > 1, then f (y) ≤ y − 1:
Suppose that y = 2Y + 1 for some positive integer Y .
Then T (y − 1) = 12 (y − 1)y = 21 (2Y )(2Y + 1) = Y (2Y + 1) = Y y.
Thus, T (y − 1) is a multiple of y, and so by the first fact above, f (y) ≤ y − 1.
Next, we show that f (2a ) = 2a+1 − 1 for every positive integer a:
Suppose that f (2a ) = m.
Then 21 m(m + 1) is a multiple of 2a , so 21 m(m + 1) = q2a for some positive integer
q or 2a+1 q = m(m + 1).
Since one of m and m + 1 is even and the other is odd, then the even one of these
must contain at least a + 1 factors of 2 and so must be at least 2a+1 .
The smallest m for which this is possible is m = 2a+1 −1 which makes m+1 = 2a+1 .
This tells us that f (2a ) ≥ 2a+1 − 1.
But T (2a+1 − 1) = 21 (2a+1 − 1)2a+1 = 2a (2a+1 − 1), which is divisible by 2a , so
f (2a ) ≤ 2a+1 − 1.
Therefore, f (2a ) = 2a+1 − 1.
We can now look at f (b + 1) − f (b) when b = 2a − 1 for some positive integer a.
Note that b is odd.
In this case,

f (b + 1) − f (b) = f (2a ) − f (2a − 1) = 2a+1 − 1 − f (2a − 1) ≥ 2a+1 − 1 − (2a − 2) = 2a + 1

2009 COMC Solutions Page 16

If a ≥ 11, then 2a + 1 ≥ 2049 > 2009.

Therefore, if b = 2a − 1 and a is a positive integer with a ≥ 11, then f (b + 1) − f (b) > 2009,
so there are infinitely many odd positive integers b for which f (b + 1) − f (b) > 2009.
(c) From (a), we know that f (c) = f (c + 3) has a solution, namely c = 9, since f (9) = 8 and
f (12) = 8.
We show that k = 3 is the smallest possible value of k by showing that the equations

f (c) = f (c + 1) and f (c) = f (c + 2)

are not satisfied by any odd positive integer c.

If a and b are positive integers, we use the notation “a | b” to mean that b is divisible by
a (in other words, b is a multiple of a or equivalently a divides b).

Case 1: f (c) = f (c + 1)
Suppose that f (c) = f (c + 1) = m for some odd positive integer c.
Then c | T (m) and T (m) = 12 m(m + 1); say, 12 m(m + 1) = qc for some positive integer q.
Since c is odd, then m ≤ c − 1 by (b), so qc = 21 m(m + 1) ≤ 12 (c − 1)(c) which tells us that
q ≤ 21 (c − 1).
But if f (c + 1) = m as well, then c + 1 | T (m) and T (m) = qc, so c + 1 | qc.
Since c and c + 1 are consecutive integers, then gcd(c, c + 1) = 1.
(This is true since if d is a positive common divisor of c and c + 1, then d divides into their
difference (which equals 1), so d itself must equal 1.)
Since c + 1 | qc, and q and c are positive integers, and gcd(c, c + 1) = 1, then c + 1 | q, so
q ≥ c + 1.
But q ≤ 21 (c − 1), which is a contradiction, since we cannot have q ≥ c + 1 > 12 (c − 1) ≥ q.
Thus, f (c) = f (c + 1) has no odd solutions.

Case 2: f (c) = f (c + 2)
Suppose that f (c) = f (c + 2) = m for some odd positive integer c.
Then c | T (m); say, 12 m(m + 1) = qc for some positive integer q.
Since m ≤ c − 1 by (c), then qc ≤ 21 (c − 1)(c) which tells us that q ≤ 12 (c − 1).
But if f (c + 2) = m as well, then c + 2 | T (m) and T (m) = qc, so c + 2 | qc.
Since c and c + 2 are odd integers, then gcd(c, c + 2) = 1.
(This is true since if d is a positive common divisor of c and c + 2, then c divides into their
difference (which equals 2), so d itself must equal 1 or 2. But both c and c + 2 are odd, so
d is odd, so d = 1.)
Since c + 2 | qc, and q and c are positive integers, and gcd(c, c + 2) = 1, then c + 2 | q, so
q ≥ c + 2.
But q ≤ 12 (c − 1), so we have a contradiction as above.
2009 COMC Solutions Page 17

Thus, f (c) = f (c + 2) has no odd solutions.

Therefore, k = 3 is the smallest positive integer k for which f (c) = f (c + k) has solutions
with c odd, since if k = 1 or k = 2, there are no solutions, and there is at least 1 solution
for k = 3.
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING
presents the

Sun Life Financial Canadian

Open Mathematics Challenge

Wednesday, November 24, 2010

Time: 2 12 hours 2010

c Canadian Mathematical Society
Calculators are NOT permitted.
Do not open this booklet until instructed to do so.
There are two parts to this paper.
PART A
This part of the paper consists of 8 questions, each worth 5 marks. You can earn full value
for each question by entering the correct answer(s) in the space provided. If your answer is
incorrect, any work that you do will be considered for part marks, provided that it is done
in the space allocated to that question in your answer booklet.
PART B
This part of the paper consists of 4 questions, each worth 10 marks. Finished solutions must
be written in the appropriate location in the answer booklet. Rough work should be done
separately. If you require extra pages for your finished solutions, paper will be provided by
your supervising teacher. Any extra papers should be placed inside your answer booklet. Be
sure to write your name and school name on any inserted pages.
Marks are awarded for completeness, clarity, and style of presentation. A correct solution,
poorly presented, will not earn full marks.
NOTES:
At the completion of the contest, insert the information sheet inside the answer
booklet.
The names of top scoring competitors will be published on the Web sites of the
CMS and CEMC.
Sun Life Financial Canadian Open Mathematics Challenge
NOTE: 1. Please read the instructions on the front cover of this booklet.
2. Write solutions in the answer booklet provided.
3. It is expected that all calculations
√ and answers will be expressed as
exact numbers such as 4π, 2 + 7, etc., rather than as 12.566 . . .
or 4.646 . . ..
4. Calculators are not allowed.
5. Diagrams are not drawn to scale. They are intended as aids only.

PART A
(9 + 5)2 − (9 − 5)2
1. Determine the integer equal to .
(9)(5)
2. Determine all values of x for which x − (8 − x) = 8 − (x − 8).

3. In the diagram, each of the three circles has centre O. C

Diameter CD of the largest circle passes through points B
B, A and O. The lengths of the radii of the circles are A
OA = 2, OB = 4, and OC = 6. What is the area of the
shaded region? O

D
(3.1 × 107 )(8 × 108 )
4. Determine the number of digits of the integer equal to .
2 × 103
5. What point on the line with equation y = x is closest to the point P (−3, 9)?

6. On a calculus exam, the average of those who studied was 90% and the average of
those who did not study was 40%. If the average of the entire class was 85%, what
percentage of the class did not study?

7. In the diagram, rectangle ABCD has AB = 20 and W

BC = 10. Points W and K are outside of the
rectangle with W A = KC = 12 and W B = KD = 16.
Determine the length of W K and express your answer A B
√
in the form W K = m n, where m and n are positive
integers with m > 1.

D C

8. Determine all values of x for which (x2 + 3x + 2)(x2 − 2x − 1)(x2 − 7x + 12) + 24 = 0.

 PART B
1. In each part of this problem, each of the variables in the
grid is to be replaced with an integer. The sum of the
X 13 30
integers in a row is given to the right of the row. The
Y 11 23
sum of the integers in a column is given at the bottom of
the column. For example, from the grid to the right we 29 24
can conclude that X +13 = 30, Y +11 = 23, X +Y = 29,
and 13 + 11 = 24.
(a) Determine the value of C.
A A 50
B C 44
37 57
(b) Determine the value of n, the sum of the integers in the second column.

D D D 30
F F E 55
F E E 50
50 n 40

(c) Determine the value of P + Q.

P Q T R 20
Q P T R 20
R R R T 33
T T T R 19
20 20 19 33

2. The parabola with equation y = x2 − 4x + 12 intersects the line with equation

y = −2x + 20 at points A and B.
(a) Determine the coordinates of the points A and B.
(b) Determine the coordinates of the midpoint, M , of the segment AB.
(c) A line parallel to the line with equation y = −2x + 20 intersects the parabola at
distinct points P (p, p2 − 4p + 12) and Q(q, q 2 − 4q + 12). Prove that p + q = 2.
(d) Point N is the midpoint of P Q. Explain why line segment M N is vertical.
3. In the diagram, the circle has centre O, diameter AC,
and radius 1. A chord is drawn from A to an arbitrary P
point B (different from A) on the circle and extended B
to the point P with BP = 1. Thus P can take many
positions. Let S be the set of points P .

(a) Let U be a point in S for which U O is

A C
perpendicular to AC. Determine the length of U O. O
(b) Let V be a point in S for which V C is
perpendicular to AC. Determine the length of V C.

(c) Determine whether or not there is a circle on which

all points of S lie.
 Canadian Open
Mathematics
Challenge

Financial
(English)

Sun Life
2010
4. Let bxc denote the greatest integer less than or equal to x.
For example, b3.1c = 3 and
 b−1.4c =−2. 
1 1
For x > 0, define f (x) = x + − x+ .
x x
For example, f 49 = 49 + 49 − 49 + 94 = 97 25


 
36 − 2 = 36 .

(a) Determine all x > 0 so that f (x) = x.

a
(b) Suppose that x = for some positive integer a > 1.
a+1
Prove that x 6= f (x), but that f (x) = f (f (x)).
(c) Prove that there are infinitely many rational numbers u so that
• 0 < u < 1,
• u, f (u) and f (f (u)) are all distinct, and
• f (f (u)) = f (f (f (u))).
 The Canadian Mathematical Society
in collaboration with

The CENTRE for EDUCATION

in MATHEMATICS and COMPUTING

presents the

Sun Life Financial

Canadian Open Mathematics Challenge
Wednesday, November 24, 2010

Solutions

©2010 Canadian Mathematical Society

2010 COMC Solutions Page 2

Part A

1. Solution 1
(9 + 5)2 − (9 − 5)2 142 − 42 196 − 16 180
Calculating, = = = = 4.
(9)(5) 45 45 45

Solution 2
For a general x and y with x and y not zero,
(x + y)2 − (x − y)2 (x2 + 2xy + y 2 ) − (x2 − 2xy + y 2 ) 4xy
= = =4
xy xy xy
(9 + 5)2 − (9 − 5)2
Since this expression equals 4 for any values of x and y, then = 4.
(9)(5)

Solution 3
For a general x and y with x and y not zero, we can factor as a difference of squares:
(x + y)2 − (x − y)2 [(x + y) + (x − y)][(x + y) − (x − y)] (2x)(2y)
= = =4
xy xy xy
(9 + 5)2 − (9 − 5)2
Since this expression equals 4 for any values of x and y, then = 4.
(9)(5)

Answer: 4

2. Simplifying both sides,

x − (8 − x) = 8 − (x − 8)
x−8+x = 8−x+8
3x = 24
x = 8

Therefore, x = 8.

Answer: x = 8

3. Solution 1
We call the ring between the middle and inner circles the “inner ring”.
We reflect the shaded portion of the inner ring across line segment CD. The area of the shaded
region does not change when we do this.
The shaded region is now the entire semi-circle to the right of CD.
Thus, the area of the shaded region is half of the area of the outer circle.
Since OC = 6, then the outer circle has radius 6 and so has area π62 = 36π.
Therefore, the area of the shaded region is 21 (36π) = 18π.
2010 COMC Solutions Page 3

Solution 2
We call the ring between the outer and middle circles the “outer ring”, and the ring between
the middle and inner circles the “inner ring”.
Since OC = 6, then the outer circle has radius 6 and so has area π62 = 36π.
Since OB = 4, then the middle circle has radius 4 and so has area π42 = 16π.
Since OA = 2, then the inner circle has radius 2 and so has area π22 = 4π.
Since the outer circle has area 36π and the middle circle has area 16π, then the area of the
outer ring is 36π − 16π = 20π.
Since the diameter CD divides each ring into two parts of equal area, then the shaded region
of the outer ring has area 21 (20π) = 10π.
Since the middle circle has area 16π and the inner circle has area 4π, then the area of the inner
ring is 16π − 4π = 12π.
Since the diameter CD divides each ring into two parts of equal area, then the shaded region
of the inner ring has area 12 (12π) = 6π.
Since the inner circle has area 4π and line segment CD passes through the centre of this circle,
then the shaded region of the inner circle has area 12 (4π) = 2π.
Therefore, the total shaded area is 10π + 6π + 2π = 18π.

Answer: 18π

4. First, we simplify the given expression:

(3.1 × 107 )(8 × 108 ) 3.1 × 8 107 × 108

3
= × 3
= 3.1 × 4 × 107+8−3 = 12.4 × 1012 = 124 × 1011
2 × 10 2 10
Therefore, this integer consists of the digits 124 followed by 11 zeroes, so has 14 digits.

Answer: 14

5. Solution 1
Let Q be the point on the line y = x that is closest to P (−3, 9).
Then P Q is perpendicular to the line y = x.
Since the line with equation y = x has slope 1 and P Q is perpendicular to this line, then P Q
has slope −1.
Note that a general point Q on the line with equation y = x has coordinates (t, t) for some real
number t.
t−9
For the slope of P Q to equal −1, we must have = −1 or t − 9 = −(t + 3) or 2t = 6
t − (−3)
or t = 3.
Therefore, the point on the line with equation y = x that is closest to P is the point (3, 3).
2010 COMC Solutions Page 4

Solution 2
Let Q be the point on the line y = x that is closest to P (−3, 9).
Then P Q is perpendicular to the line y = x.
Let R be the point on the line y = x so that P R is horizontal, as shown.

P R

Q
x

Since P R is horizontal and P has y-coordinate 9, then R has y-coordinate 9.

Since R lies on the line with equation y = x, then R has coordinates (9, 9).
Since P R is horizontal and QR has slope 1 (because it lies along the line with equation y = x),
then ∠P RQ = 45◦ .
Since ∠P QR = 90◦ , then 4P QR is isosceles and right-angled.
Let M be the midpoint of P R. Since P has coordinates (−3, 9) and R has coordinates (9, 9),
then M has coordinates (3, 9).
Since 4P QR is isosceles and M is the midpoint of P R, then QM is perpendicular to P R.
Thus, QM is vertical, so Q has x-coordinate 3.
Since Q lies on the line with equation y = x, then Q has coordinates (3, 3).

Solution 3
Note that a general point Q on the line with equation y = x has coordinates (t, t) for some real
number t.
p
Then P Q = (t − (−3))2 + (t − 9)2 or P Q2 = (t + 3)2 + (t − 9)2 .
Since we want to find the point on the line with equation y = x that is closest to P , then we
want to minimize the value of P Q, or equivalently to minimize the value of P Q2 .
In other words, we want to find the value of t that minimizes the value of

P Q2 = t2 + 6t + 9 + t2 − 18t + 81 = 2t2 − 12t + 90

Since this equation represents a parabola opening upwards, then its minimum occurs at its
−12
vertex, which occurs at t = − 2(2) = 3. Thus, t = 3 minimizes the length of P Q.
Therefore, the point on the line with equation y = x that is closest to P is the point (3, 3).

Answer: (3, 3)
2010 COMC Solutions Page 5

6. Let x be the number of people who studied for the exam and let y be the number of people
who did not study.
We assume without loss of generality that the exam was out of 100 marks.
Since the average of those who studied was 90%, then those who studied obtained a total of
90x marks.
Since the average of those who did not study was 40%, then those who did not study obtained
a total of 40y marks.
90x + 40y
Since the overall average was 85%, then = 85.
x+y
Therefore, 90x + 40y = 85x + 85y or 5x = 45y or x = 9y.
Therefore, x : y = 9 : 1 = 90 : 10. This means that 10% of the class did not study for the exam.

Answer: 10%

7. Solution 1
Since ABCD is a rectangle, then AD = BC = 10 and DC = AB = 20.
Since W A = 12, W B = 16, AB = 20, and 122 + 162 = 144 + 256 = 400 = 202 , then
W A2 + W B 2 = AB 2 . Thus, 4AW B is right-angled at W .
Note that 4CKD is congruent to 4AW B, so 4CKD is right-angled at K.
Extend W A and KD to meet at Y and W B and KC to meet at Z.

W
16
12
A B
20 Z
10
Y
D C

Suppose that ∠W AB = ∠KCD = θ. Then ∠W BA = ∠KDC = 90◦ − θ.

Now ∠Y AD = 180◦ − ∠W AB − ∠BAD = 180◦ − θ − 90◦ = 90◦ − θ.
Also, ∠Y DA = 180◦ − ∠KDC − ∠ADC = 180◦ − (90◦ − θ) − 90◦ = θ.
Therefore, 4Y DA is similar to 4W AB. This means that ∠DY A = 90◦ .
Also, since DA = 21 AB, then the sides of 4Y DA are half as long as the corresponding sides of
4W AB. Thus, Y D = 12 W A = 6 and Y A = 12 W B = 8.
Similarly, ∠BZC = 90◦ . Therefore, W Y KZ is a rectangle.
We have W Y = W A + AY = 12 + 8 = 20 and Y K = Y D + DK = 6 + 16 = 22.
2010 COMC Solutions Page 6

By the Pythagorean Theorem, since W K > 0, then

√ √ √ √ √
W K = W Y 2 + Y K 2 = 202 + 222 = 400 + 484 = 884 = 2 221

as required.

Solution 2
Since ABCD is a rectangle, then AD = BC = 10 and DC = AB = 20.
We coordinatize the diagram, putting D at the origin, A at (0, 10), C at (20, 0), and B at
(20, 10).
Since W A = 12, W B = 16, AB = 20, and 122 + 162 = 144 + 256 = 400 = 202 , then
W A2 + W B 2 = AB 2 . Thus, 4AW B is right-angled at W .
Note that 4CKD is congruent to 4AW B, so 4CKD is right-angled at K.
Since 4AW B is right-angled and we know its side lengths, then we can compute the trigono-
metric ratios of its angles.
WB 16 4 WA 12 3
In particular, sin(∠W AB) = = = and cos(∠W AB) = = = .
AB 20 5 AB 20 5
We drop perpendiculars from W to X on AB and from K to Y on DC.

W
12 16

A (0, 10) B (20, 10)

X 20
10
Y
D
x
C (20, 0)

Then AX = W A cos(∠W AB) = 12( 35 ) = 36 5

and W X = W A sin(∠W AB) = 12( 45 ) = 48 5
.
Since A has coordinates (0, 10), then W has coordinates ( 36 5
, 10 + 485
) = ( 36 , 98 ).
5 5
Since 4CKD is congruent to 4AW B, then in a similar way we can find that the coordinates
of K are (20 − 36
5
, − 48
5
) = ( 64
5
, − 48
5
).
Since we have the coordinates of W and K, then the distance between W and K is
q q
282 2
WK = ( 64
5
− 36 2
5
) + (− 48
5
− 98 2
5
) = 52
+ 146
52
√ √
= 25 142 + 732 = 52 196 + 5329
√ q
= 5 5525 = 2 5525
2
25
√
= 2 221
2010 COMC Solutions Page 7

√
Therefore, W K = 2 221.
√
Answer: W K = 2 221

8. Solution 1
First, we factor the first and third quadratic factors to obtain

(x + 1)(x + 2)(x2 − 2x − 1)(x − 3)(x − 4) + 24 = 0

Next, we rearrange the factors to obtain

(x + 1)(x − 3)(x2 − 2x − 1)(x + 2)(x − 4) + 24 = 0

and expand to obtain

(x2 − 2x − 3)(x2 − 2x − 1)(x2 − 2x − 8) + 24 = 0

Next, we make the substitution w = x2 − 2x to obtain

(w − 3)(w − 1)(w − 8) + 24 = 0

This is a cubic equation in w so we expand, simplify and factor:

(w2 − 4w + 3)(w − 8) + 24 = 0
w3 − 12w2 + 35w = 0
w(w2 − 12w + 35) = 0
w(w − 5)(w − 7) = 0

Therefore, the solutions in terms of w are w = 0 or w = 5 or w = 7.

If w = x2 − 2x = 0, then x(x − 2) = 0 which gives x = 0 or x = 2.
If w = x2 − 2x = 5, then x2 − 2x − 5 = 0. √
2 ± 24 √
The quadratic formula gives the roots x = = 1 ± 6.
2
If w = x2 − 2x = 7, then x2 − 2x − 7 = 0. √
2 ± 32 √
The quadratic formula gives the roots x = = 1 ± 8.
√ 2 √
Therefore, x = 0 or x = 2 or x = 1 ± 6 or x = 1 ± 8. (This last pair can be rewritten as
√
x = 1 ± 2 2.)

Solution 2
First, we factor the first and third quadratic factors and complete the square in the second
quadratic factor to obtain

(x + 1)(x + 2)((x − 1)2 − 2)(x − 3)(x − 4) + 24 = 0

2010 COMC Solutions Page 8

Next, we make the substitution y = x − 1 (which makes x + 1 = y + 2 and x + 2 = y + 3 and

x − 3 = y − 2 and x − 4 = y − 3) to obtain

(y + 2)(y + 3)(y 2 − 2)(y − 2)(y − 3) + 24 = 0

(We made this substitution because it made the algebra more “symmetric”; that is, after making
this substitution, factors of the form y − a are paired with factors of the form y + a.)
Next, we rearrange the factors to obtain

(y + 2)(y − 2)(y + 3)(y − 3)(y 2 − 2) + 24 = 0

Next, we multiply out pairs of factors to obtain

(y 2 − 4)(y 2 − 9)(y 2 − 2) + 24 = 0

Next, we make the substitution z = y 2 to obtain

(z − 4)(z − 9)(z − 2) + 24 = 0

This is a cubic equation in z so we expand and simplify:

(z 2 − 13z + 36)(z − 2) + 24 = 0
z 3 − 15z 2 + 62z − 48 = 0

By inspection, z = 1 is a solution so z − 1 is a factor of the cubic equation.

We factor out this linear factor to obtain (z − 1)(z 2 − 14z + 48) = 0.
The quadratic factor can be factored as (z − 6)(z − 8).
Therefore, we have (z − 1)(z − 6)(z − 8) = 0.
Therefore, the solutions in terms of z are z = 1 or z = 6 or z = 8.
√ √
Since z = y 2 , then the solutions in terms of y are y = ±1 or y = ± 6 or y = ± 8.
Since y = x − 1, then x = y + 1, and so the solutions in terms of x are x = 0 or x = 2 or
√ √ √
x = 1 ± 6 or x = 1 ± 8. (This last pair can be rewritten as x = 1 ± 2 2.)
√ √
Answer: x = 0, 2, 1 ± 6, 1 ± 2 2
2010 COMC Solutions Page 9

Part B

1. (a) Solution 1
From the first row, A + A = 50 or A = 25.
From the second column, A + C = 57. Since A = 25, then C = 57 − 25 = 32.

Solution 2
From the first row, A + A = 50 or A = 25.
From the first column, A + B = 37. Since A = 25, then B = 37 − 25 = 12.
From the second row, B + C = 44. Since B = 12, then C = 44 − 12 = 32.

(b) Solution 1
The sum of the nine entries in the table equals the sum of the column sums, or 50+n+40 =
90 + n. (This is because each entry in the table is part of exactly one column sum.)
Similarly, the sum of the nine entries in the table also equals the sum of the row sums, or
30 + 55 + 50 = 135.
Therefore, 90 + n = 135 or n = 45.

Solution 2
The sum of the nine entries in the table equals the sum of the row sums, or 30 + 55 + 50 =
135. (This is because each entry in the table is part of exactly one row sum.)
Since the entries in the table include three entries equal to each of D, E and F , then the
sum of the entries in the table is also 3D + 3E + 3F = 3(D + E + F ).
Therefore, 3(D + E + F ) = 135 or D + E + F = 45.
From the second column, D + E + F = n. Thus, n = 45.

Solution 3
From the first row, D + D + D = 30 or D = 10.
From the first column, D + 2F = 50. Since D = 10, then 2F = 50 − 10 and so F = 20.
From the third column, D + 2E = 40. Since D = 10, then 2E = 40 − 10 and so E = 15.
Therefore, n = D + E + F = 10 + 15 + 20 = 45.

(c) Solution 1
From the third row, 3R + T = 33.
From the fourth row, R + 3T = 19.
Adding these equations, we obtain 4R + 4T = 52 or R + T = 13.
From the first row, P + Q + R + T = 20.
Since R + T = 13, then P + Q = 20 − 13 = 7.
2010 COMC Solutions Page 10

Solution 2
The sum of the sixteen entries in the table equals the sum of the row sums, or 20 + 20 +
33 + 19 = 92. (This is because each entry in the table is part of exactly one row sum.)
The table includes two entries equal to each of P and Q and six entries equal to each of
R and T .
Therefore, 2P + 2Q + 6R + 6T = 92.
The last two rows of the table include four entries equal to each of R and T , so 4R + 4T =
33 + 19 = 52, or R + T = 13.
Therefore, 2P + 2Q = 92 − 6(R + T ) = 92 − 6(13) = 14, and so P + Q = 7.

Solution 3
From the third row, 3R + T = 33.
From the fourth row, R + 3T = 19.
Multiplying the first equation by 3 and subtracting the second equation gives
(9R + 3T ) − (R + 3T ) = 99 − 19 or 8R = 80 or R = 10.
Since 3R + T = 33, then T = 33 − 3(10) = 3.
From the first row, P + Q + R + T = 20.
Since R = 10 and T = 3, then P + Q = 20 − 10 − 3 = 7.

2. (a) To determine the coordinates of A and B, we equate values of y using the equations
y = x2 − 4x + 12 and y = −2x + 20 to obtain

x2 − 4x + 12 = −2x + 20
x2 − 2x − 8 = 0
(x − 4)(x + 2) = 0

Therefore, x = 4 or x = −2.
To determine the y-coordinates of points A and B, we can use the equation of the line.
If x = 4, then y = −2(4) + 20 = 12.
If x = −2, then y = −2(−2) + 20 = 24.
Therefore, the coordinates of A and B are (4, 12) and (−2, 24).

(b) Using the coordinates of A and B from (a), the coordinates of the midpoint M of AB are
( 21 (4 + (−2)), 12 (24 + 12)) or (1, 18).

(c) Solution 1
The line with equation y = −2x + 20 has slope −2.
Therefore, we have a line with slope −2 that intersects the parabola at points
2010 COMC Solutions Page 11

P (p, p2 − 4p + 12) and Q(q, q 2 − 4q + 12).

In other words, line segment P Q has slope −2.
Therefore,

(p2 − 4p + 12) − (q 2 − 4q + 12)

= −2
p−q
p2 − q 2 − 4p + 4q
= −2
p−q
(p − q)(p + q) − 4(p − q)
= −2
p−q
(p + q) − 4 = −2 (since p 6= q)
p+q = 2

Therefore, p + q = 2, as required.

Solution 2
The line with equation y = −2x + 20 has slope −2.
Therefore, we have a line with slope −2 (say with equation y = −2x + b) that intersects
the parabola at points P and Q.
Since y = −2x + b and y = x2 − 4x + 12 intersect when x = p, then p2 − 4p + 12 = −2p + b,
which gives p2 − 2p + 12 − b = 0.
Since y = −2x + b and y = x2 − 4x + 12 intersect when x = q, then q 2 − 4q + 12 = −2q + b,
which gives q 2 − 2q + 12 − b = 0.
Since we have two expressions equal to 0, then

p2 − 2p + 12 − b = q 2 − 2q + 12 − b
p2 − 2p = q 2 − 2q
p2 − q 2 − 2p + 2q = 0
(p − q)(p + q) − 2(p − q) = 0
(p − q)(p + q − 2) = 0

Therefore, p − q = 0 or p + q − 2 = 0.
Since p 6= q, then p + q = 2.

(d) Since P has coordinates (p, p2 − 4p + 12) and Q has coordinates (q, q 2 − 4q + 12), then the
x-coordinate of the midpoint N of P Q is 21 (p + q).
Since p + q = 2 by (c), then the x-coordinate of N is 1.
Since the x-coordinate of M is 1 and the x-coordinate of N is 1, then line segment M N
is vertical.
2010 COMC Solutions Page 12

3. (a) Solution 1
Let U be the point in S vertically above O and let B be the point where AU intersects
the circle. (There will be one other point U in S with U O perpendicular to AC; this point
will be vertically below O. By symmetry, the length of U O is the same in either case.)
Join U O and BO.
U

A C
O

Let ∠BU O = θ.
Note that AO = BO = 1 since they are radii and BU = 1 by definition.
Therefore, 4U BO is isosceles and so ∠BOU = ∠BU O = θ.
Now ∠ABO is an exterior angle in this triangle, so ∠ABO = ∠BU O + ∠BOU = 2θ.
Since OB = OA, then 4ABO is isosceles and so ∠BAO = ∠ABO = 2θ.
But 4U AO is right-angled at O, and so ∠U AO + ∠AU O = 90◦ or 2θ + θ = 90◦ .
Therefore, 3θ = 90◦ or θ = 30◦ .
√ √
This tells us that 4U AO is a 30◦ -60◦ -90◦ triangle, and so U O = 3AO = 3.

Solution 2
Let U be the point in S vertically above O and let B be the point where AU intersects
the circle. (There will be one other point U in S with U O perpendicular to AC; this point
will be vertically below O. By symmetry, the length of U O is the same in either case.)
Join U O and BC.
U

A C
O
2010 COMC Solutions Page 13

Since AC is a diameter, then ∠ABC = 90◦ .

Therefore, 4ABC is similar to 4AOU since each is right-angled and each includes the
angle at A.
AB AO
Thus, = .
AC AU
Since U is in S, then BU = 1, so AU = AB + BU = AB + 1.
Also, AO = 1 and AC = 2 since the radius of the circle is 1.
AB 1
Therefore, = or AB 2 + AB − 2 = 0.
2 AB + 1
Factoring, we obtain (AB − 1)(AB + 2) = 0. Since AB > 0, then AB = 1 and so AU = 2.
√ √ √
By the Pythagorean Theorem, since U O > 0, then U O = AU 2 − AO2 = 22 − 12 = 3.
(b) As in (a), we can choose the point V in S that is vertically above C.
B is the point where AV intersects the circle. Note that BV = 1 by definition.
Join V C and BC.
Since AC is a diameter, then ∠ABC = 90◦ .
V

A C
O

Let V C = x.
Since 4V CA is right-angled at C and AV > 0, then by the Pythagorean Theorem,
√ √
AV = AC 2 + CV 2 = 4 + x2 .
Now 4V BC is similar to 4V CA since both are right-angled and the triangles share a
common angle at V .
Since these triangles are similar, then
VB VC
=
VC VA
1 x
= √
x x2 + 4
√
x2 + 4 = x2
x2 + 4 = x4
0 = x4 − x2 − 4
0 = (x2 )2 − x2 − 4
2010 COMC Solutions Page 14

This is a quadratic equation in x2 . By the quadratic formula,

p √
2 1 ± (−1)2 − 4(1)(−4) 1 ± 17
x = =
2 2
√
2 2 1 + 17
Since x is positive, then x = .
2 r √
1 + 17
Since V C = x is positive, then V C = x = .
2
(c) We prove that such a circle does not exist by contradiction.
Suppose that there is a circle Z on which all of the points in S lie.
We coordinatize the original diagram, putting O at (0, 0), C at (1, 0), and A at (−1, 0).
For every point X in S that is above AC, there will be a corresponding point Y in S that
is below AC which is the reflection of X in AC.
Therefore, S is symmetric across the x-axis. Thus, Z is also symmetric across the x-axis
and so its centre lies on the x-axis.
Suppose Z has centre (p, 0) and radius r.
Then the equation of Z is (x − p)2 + y 2 = r2 .
√
From (a), the point (0, 3) lies on Z. Thus, p2 + 3 = r2 .
Also, the point W (2, 0) lies on Z. This point comes from choosing B to coincide with C
and extending AB horizontally by 1 unit. Thus, (2 − p)2 + 02 = r2 or p2 − 4p + 4 = r2 .
From the equations p2 + 3 = r2 and p2 − 4p + 4 = r2 , we equate values of r2 to obtain
p2 + 3 = p2 − 4p + 4 or 4p = 1 or p = 41 .
1
Thus, r2 = p2 + 3 = 16 + 3 = 49
16
and so r = 74 since r > 0.
Therefore, the equation
 q of Z must be (x − 14 )2 + y 2 = ( 47 )2 .
√
From (b), the point 1, 1+2 17 lies on the circle.
Therefore,
q √
2
1 2
(1 − 4
) + 1+ 17
2
= ( 74 )2
√
( 43 )2 + 1+2 17 = 49
16
√
9 8 17 49
16
+ 16
+ 2
= 16
√
17
2
= 2
√
17 = 4

This statement is false, so we have reached a contradiction.

Therefore, our assumption is false and there is no circle on which all of the points in S lie.
1 1 1
4. (a) First we note that if x > 0, then x + > 1, since if x ≥ 1, then x + ≥ 1 + > 1 and if
x x x
1 1
0 < x < 1, then > 1, so x + > 1.
x x
2010 COMC Solutions Page 15

We note that f (x) = x is equivalent to

   
1 1
x+ − x+ = x
x x
 
1 1
= x+
x x

In this last equation, the right side is a positive integer, so the left side is also a positive
integer.
1 1
Suppose that = n for some positive integer n. Then x = .
x   n  
1 1 1
Therefore, the equation = x + is equivalent to the equation n = +n .
x x n
 
1 1 1
Note that if n ≥ 2, then < 1, so n < n + < n + 1, which says that + n = n so
n n n
this result is true for all positive integers n ≥ 2.
 
1 1
Note also that if n = 1, then n + = 2, so + n 6= n.
n n
 
1
Therefore, if n is a positive integer, then n = + n if and only if n ≥ 2.
n
1
Therefore, the solution set of the equation f (x) = x is x = , where n is a positive integer
n
with n ≥ 2.
a
(b) Suppose that x = for some positive integer a > 1.
a+1

First, we calculate f (x).

Note that
1 a a+1 a2 + (a + 1)2 2a2 + 2a + 1 2(a2 + a) + 1 1
x+ = + = = 2
= 2
=2+ 2
x a+1 a a(a + 1) a +a a +a a +a
1 1 1 1
Since a > 1, then 2 < 2 = and so 2 < 2 + 2 < 3.
 a
 +a 1 +1 2 a +a
1 1
Therefore, x + = 2+ 2 = 2.
x a +a
     
a 1 1 1 1
Thus, if x = , then f (x) = x + − x+ = 2+ 2 −2= 2 .
a+1 x x a +a a +a

Second, we show that x 6= f (x).

a a2 1
Note that x = = and f (x) = , so x and f (x) would be equal if
a+1 a(a + 1) a(a + 1)
a2 1
and only if = which is true if and only if a2 = 1.
a(a + 1) a(a + 1)
Since a > 1, this is not true, so x 6= f (x).
1
(Alternatively, we could note that, from (a), x = f (x) if and only if x is of the form for
n
2010 COMC Solutions Page 16

a
some positive integer n > 1. Here, x = which is not of this form when a > 1, so
a+1
x 6= f (x).)

Third, we show that f (x) = f (f (x)).

1
We set y = f (x) = 2 .
a +a
1
Since a is a positive integer with a > 1, then y is of the form for some positive integer
n
n with n > 2.
Thus, y is of the form discovered in (a), so f (y) = y; in other words, f (f (x)) = f (x).

a
Therefore, if x = for some positive integer a > 1, then x 6= f (x), but f (x) = f (f (x)).
a+1
(c) Solution 1
We want to find an infinite family of rational numbers u with the properties that
• 0 < u < 1,
• u, f (u), and f (f (u)) are all distinct, and
• f (f (u)) = f (f (f (u))).
We will do this by finding an infinite family of rational numbers u with 0 < u < 1 with
a
the property that f (u) = for some positive integer a > 1.
a+1
1 1
In this case, (b) shows that f (f (u)) = 2 and that f (f (f (u))) = 2 .
a +a a +a
Thus, we will have f (f (u)) = f (f (f (u))) and f (u) 6= f (f (u)).
a 1
As long as we have u 6= and u 6= 2 , then we will have found a family of rational
a+1 a +a
numbers u with the required properties.
1
Note that in fact we cannot have u = 2 because in this case we would have f (u) = u
a +a
a
and so we would not have f (u) = .
a+1

We now show the existence of an infinite family of rational numbers u with 0 < u < 1
a
with f (u) = for some positive integer a > 1.
a+1
b
Let us consider candidate rational numbers u = with b and c positive integers and
b+c
c > 1.
Since b + c > b, then each is a rational number with 0 < u < 1.
1 b b+c b2 + (b + c)2 2b2 + 2bc + c2 c2
In this case, u + = + = = = 2 + .
u b+c b b(b + c) b2 + bc b2 + bc
c2 1 c2
If we suppose further that c2 < b2 + bc, then 2 < 1 and so u + = 2 + 2 < 3,
b + bc u b + bc
which gives
c2 c2
   
1 1
f (u) = u + − u+ =2+ 2 −2= 2
u u b + bc b + bc
2010 COMC Solutions Page 17

a c2
We want f (u) to be of the form . In other words, we want 2 to be of the form
a+1 b + bc
a
, which would be true if b2 + bc − c2 = 1.
a+1
Note that if b2 + bc − c2 = 1, then c2 = b2 + bc − 1 < b2 + bc, so the additional assumption
above is included in this equation. Also, if b2 + bc − c2 = 1, then b and c can have no
b
common divisor larger than 1 so u = is irreducible. Combining this with the fact
b+c
b a
that c 6= 1, we see that cannot be of the form .
b+c a+1
To summarize so far, if b2 + bc − c2 = 1 has an infinite family of positive integer solutions
b
(b, c), then the infinite family of rational numbers u = has the required properties.
b+c

Consider the equation b2 + bc − c2 = 1.

This is equivalent to the equations 4b2 + 4bc − 4c2 = 4 and 4b2 + 4bc + c2 − 5c2 = 4 and
(2b + c)2 − 5c2 = 4.
If we let d = 2b + c, we obtain the equation d2 − 5c2 = 4.
This is a version of Pell’s equation. It is known that if such an equation has one positive
integer solution, then it has infinitely many positive integer solutions.
Since d2 − 5c2 = 4 has one positive integer solution (d, c) = (7, 3), then it has infinitely
many positive integer solutions (d, c).
If d2 = 5c2 + 4 and c is odd, then c2 is odd, so d2 = 5c2 + 4 is odd, which means that d is
odd.
If d2 = 5c2 + 4 and c is even, then c2 is even, so d2 = 5c2 + 4 is even, which means that d
is even.
Therefore, if (d, c) satisfies d2 − 5c2 = 4, then d and c have the same parity so b = 21 (d − c)
is an integer.
In addition, since d2 = 5c2 + 4 > c2 then d > c which means that b = 12 (d − c) is a positive
integer.
Therefore, each positive integer solution (d, c) of the equation d2 − 5c2 = 4 gives a solution
(b, c) of the equation b2 + bc − c2 = 1 which is also a positive integer solution.

b
Therefore, there exists an infinite family of rational numbers u = with the required
b+c
properties.

Solution 2
As in Solution 1, we want to show the existence of an infinite family of rational numbers
a
u with 0 < u < 1 with f (u) = for some positive integer a > 1.
a+1
Consider the Fibonacci sequence which has F1 = 1, F2 = 1, and Fn = Fn−1 + Fn−2 for
n ≥ 3.
2010 COMC Solutions Page 18

F2n−1
Define un = for each positive integer n ≥ 2.
F2n+1
Note that 0 < F2n−1 < F2n+1 so 0 < u < 1.
F3 2
For example, u2 = = .
F5 5
In this case, f (u2 ) = ( 5 + 52 ) − b 25 + 52 c = 10
2 29
− b 29
10
9
c = 10 , which has the desired properties.
a 1
We must show that un is not of the form or of the form 2 :
a+1 a +a
F2n−1 a
If = , then aF2n−1 + F2n−1 = aF2n+1 or F2n−1 = a(F2n+1 − F2n−1 ) or
F2n+1 a+1
F2n−1 = aF2n . Since a is a positive integer and F2n > F2n−1 , this cannot be the
case.
F2n−1 1
If = 2 , then F2n+1 is divisible by F2n−1 . But Fj+1 and Fj−1 never
F2n+1 a +a
have a common divisor larger than 1, so this cannot be the case. (If Fj+1 and
Fj−1 have a common divisor larger than 1, then Fj = Fj+1 − Fj−1 also has this
divisor. We can continue this process using the equation Fj−2 = Fj − Fj−1 to
show that Fj−2 also has this divisor, and so on, until we obtain that F2 and F1
both have this divisor. Since F2 = F1 = 1, we have a contradiction.)
In general, note that
1 F2n−1 F2n+1
un + = +
un F2n+1 F2n−1
(F2n−1 )2 + (F2n+1 )2
=
F2n−1 F2n+1
(F2n−1 )2 + (F2n + F2n−1 )2
=
F2n−1 (F2n + F2n−1 )
2(F2n−1 )2 + 2F2n F2n−1 + (F2n )2
=
(F2n−1 )2 + F2n F2n−1
(F2n )2
= 2+
(F2n−1 )2 + F2n F2n−1
(F2n )2
= 2+
F2n−1 F2n+1

It is known that (F2n )2 − F2n−1 F2n+1 = −1 for all positive integers n. (See the end of the
solution for a proof of this.)
Set an = (F2n )2 , which is a positive integer.
1 an
Then, un + =2+ .
un an + 1
2010 COMC Solutions Page 19

Therefore,
   
1 1
f (un ) = un + − un +
un un
   
an an
= 2+ − 2+
an + 1 an + 1
 
an
= 2+ −2
an + 1
an
=
an + 1
Therefore, the infinite family of rational numbers un has the desired properties.

As a postscript, we prove that (Fm )2 − Fm−1 Fm+1 = (−1)m+1 for all positive integers
m ≥ 2.
We prove this result by induction on m.
When m = 2, we obtain (F2 )2 − F1 F3 = 12 − 1(2) = −1 = (−1)2+1 , as required.
Suppose that the result is true for m = k, for some positive integer k ≥ 2.
That is, suppose that (Fk )2 − Fk−1 Fk+1 = (−1)k+1 .
Consider m = k + 1. Then

(Fk+1 )2 − Fk Fk+2 = (Fk + Fk−1 )2 − Fk (Fk + Fk+1 )

= (Fk )2 + 2Fk Fk−1 + (Fk−1 )2 − (Fk )2 − Fk Fk+1
= 2Fk Fk−1 + (Fk−1 )2 − Fk Fk+1
= 2Fk Fk−1 + (Fk−1 )2 − Fk (Fk + Fk−1 )
= 2Fk Fk−1 + (Fk−1 )2 − (Fk )2 − Fk Fk−1
= Fk Fk−1 + (Fk−1 )2 − (Fk )2
= Fk−1 (Fk + Fk−1 ) − (Fk )2
= Fk−1 Fk+1 − (Fk )2
= (−1)((Fk )2 − Fk−1 Fk+1 )
= (−1)(−1)k+1 (by our inductive assumption)
= (−1)(k+1)+1

as required.
Therefore, (Fm )2 − Fm−1 Fm+1 = (−1)m+1 for all positive integers m ≥ 2 by induction,
which shows that (F2k )2 − F2k−1 F2k+1 = (−1)2k+1 = −1.
COMC 2011 1

2011 COMC contest

Short Answer Problems

A1. If r is a number such that r2 − 6r + 5 = 0, what is the value of (r − 3)2 ?

A2. Carmen selects four different numbers from the set {1, 2, 3, 4, 5, 6, 7} whose sum is 11. If ℓ is
the largest of these four numbers, what is the value of ℓ?

A3. The faces of a cube contain the numbers 1, 2, 3, 4, 5, 6 such that the sum of the numbers on
each pair of opposite faces is 7. For each of the cube’s eight corners, we multiply the three
numbers on the faces incident to that corner, and write down its value. (In the diagram, the
value of the indicated corner is 1 × 2 × 3 = 6.) What is the sum of the eight values assigned
to the cube’s corners?

3
1

A4. In the figure, AQP B and ASRC are squares, and AQS is an equilateral triangle. If QS = 4
and BC = x, what is the value of x?
Q 4 S

P R
A

B x C

1
COMC 2011 2

B1. Arthur is driving to David’s house intending to arrive at a certain time. If he drives at 60
km/h, he will arrive 5 minutes late. If he drives at 90 km/h, he will arrive 5 minutes early.
If he drives at n km/h, he will arrive exactly on time. What is the value of n?

B2. Integers a, b, c, d, and e satisfy the following three properties:

(i) 2 ≤ a < b < c < d < e < 100

(ii) gcd(a, e) = 1
(iii) a, b, c, d, e form a geometric sequence.

What is the value of c?

√
B3. In the figure, BC is a diameter of the circle, where BC = 901, BD = 1, and DA = 16. If
EC = x, what is the value of x?
A
E

B C

B4. A group of n friends wrote a math contest consisting of eight short-answer problems S1 ,
S2 , S3 , S4 , S5 , S6 , S7 , S8 , and four full-solution problems F1 , F2 , F3 , F4 . Each person in
the group correctly solved exactly 11 of the 12 problems. We create an 8 × 4 table. Inside
the square located in the ith row and j th column, we write down the number of people who
correctly solved both problem Si and problem Fj . If the 32 entries in the table sum to 256,
what is the value of n?
F1 F2 F3 F4
S1
S2
S3
S4
S5
S6
S7
S8

2
COMC 2011 3

Full Solution Problems

C1. ABC is a triangle with coordinates A = (2, 6), B = (0, 0), and C = (14, 0).
A(2, 6)

B(0, 0) C(14, 0)

(a) Let P be the midpoint of AB. Determine the equation of the line perpendicular to AB
passing through P .
(b) Let Q be the point on line BC for which P Q is perpendicular to AB. Determine the
length of AQ.
(c) There is a (unique) circle passing through the points A, B, and C. Determine the radius
of this circle.

C2. Charlotte writes a test consisting of 100 questions, where the answer to each question is either
TRUE or FALSE. Charlotte’s teacher announces that for every five consecutive questions on
the test, the answers to exactly three of them are TRUE. Just before the test starts, the teacher
whispers to Charlotte that the answers to the first and last questions are both FALSE.

(a) Determine the number of questions for which the correct answer is TRUE.
(b) What is the correct answer to the sixth question on the test?
(c) Explain how Charlotte can correctly answer all 100 questions on the test.

C3. Let n be a positive integer. A row of n + 1 squares is written from left to right, numbered
0, 1, 2, . . . , n, as shown.

0 1 2 ··· n

Two frogs, named Alphonse and Beryl, begin a race starting at square 0. For each second
that passes, Alphonse and Beryl make a jump to the right according to the following rules: if
there are at least eight squares to the right of Alphonse, then Alphonse jumps eight squares
to the right. Otherwise, Alphonse jumps one square to the right. If there are at least seven
squares to the right of Beryl, then Beryl jumps seven squares to the right. Otherwise, Beryl
jumps one square to the right. Let A(n) and B(n) respectively denote the number of seconds
for Alphonse and Beryl to reach square n. For example, A(40) = 5 and B(40) = 10.

(a) Determine an integer n > 200 for which B(n) < A(n).
(b) Determine the largest integer n for which B(n) ≤ A(n).

3
COMC 2011 4

C4. Let f (x) = x2 − ax + b, where a and b are positive integers.

(a) Suppose a = 2 and b = 2. Determine the set of real roots of f (x) − x, and the set of real
roots of f (f (x)) − x.
(b) Determine the number of pairs of positive integers (a, b) with 1 ≤ a, b ≤ 2011 for which
every root of f (f (x)) − x is an integer.

4
COMC 2011 5

COMC 2011 Solutions

A1. If r is a number for which r2 − 6r + 5 = 0, what is the value of (r − 3)2 ?

Solution: The answer is 4.

Solution 1: Note that (r −3)2 = r2 −6r +9. Since r2 −6r +5 = 0, r2 −6r +9 = 4. Therefore,
the answer is 4.

Solution 2: The quadratic equation r2 − 6r + 5 factors as

(r − 1)(r − 5).

Therefore, r = 1 or r = 5. If r = 1, then (r−3)2 = (−2)2 = 4. If r = 5, then (r−3)2 = 22 = 4.

In either case, (r − 3)2 = 4.

Solution 3: By completing the square on r2 − 6r + 5, we have r2 − 6r + 5 = (r − 3)2 − 4.

Since r2 − 6r + 5 = 0, (r − 3)2 − 4 = 0. Hence, (r − 3)2 = 4.

5
COMC 2011 6

A2. Carmen selects four different numbers from the set {1, 2, 3, 4, 5, 6, 7} whose sum is 11. If ℓ is
the largest of these four numbers, what is the value of ℓ?

Solution: The answer is 5.

Solution 1: Note that the sum of the smallest four integers in the list is 1 + 2 + 3 + 4 = 10.
Hence, 1 + 2 + 3 + 5 = 11. The largest positive integer in this sum is 5. Therefore, ℓ = 5.

Solution 2: Since ℓ is the largest of four numbers from {1, 2, 3, 4, 5, 6, 7}, ℓ ≥ 4. Therefore,
l is equal to one of 4, 5, 6 and 7. If ℓ = 7, then the smallest possible sum of the four numbers
is 1 + 2 + 3 + 7 = 13 > 11. Therefore, ℓ 6= 7. Similarly, if ℓ = 6, then the smallest possible
sum of the four numbers is 1 + 2 + 3 + 6 = 12 > 11. Similarly, ℓ 6= 4. Therefore, ℓ = 5.

6
COMC 2011 7

A3. The faces of a cube contain the numbers 1, 2, 3, 4, 5, 6 such that the sum of the numbers on
each pair of opposite faces is 7. For each of the cube’s eight corners, we multiply the three
numbers on the faces incident to that corner, and write down its value. (In the diagram, the
value of the indicated corner is 1 × 2 × 3 = 6.) What is the sum of the eight values assigned
to the cube’s corners?

3
1

Solution: The answer is 343.

2 2
4 1 3 3 6 4
5 5

Solution 1: The left picture shows the corners and the faces touching the side labeled 1 and
the right picture shows the opposite side of the die, whose label is 6, which is incident to the
other four corners.

We compute the eight numbers individually and sum the eight numbers. The eight triplets
of integers at the eight corners are
(1, 2, 3), (1, 2, 4), (1, 3, 5), (1, 4, 5), (6, 2, 3), (6, 2, 4), (6, 3, 5), (6, 4, 5).
These eight values are
1×2×3 = 6
1×2×4 = 8
1×3×5 = 15
1×4×5 = 20
6×2×3 = 36
6×2×4 = 48
6×3×5 = 90
6×4×5 = 120 .
The sum of these eight positive integers is 343.

Solution 2: Since no corner contains two numbers that sum to 7, the sum in solution 1 can
be computed as
(1 + 6)(2 + 5)(3 + 4) = 73 = 343.

7
COMC 2011 8

A4. In the figure, AQP B and ASRC are squares, and AQS is an equilateral triangle. If QS = 4
and BC = x, what is the value of x?
Q 4 S

P R
A

B x C
√
Solution: The answer is 4 3.

Solution 1: Since ∆AQS is equilateral, AQ = QS = AS. Since QS = 4, AQ = AS = 4.

Since AQP B and ASRC are squares, AB = AQ = 4 and AC = AS = 4. Since ∆AQS is
equilateral, ∠QAS = 60◦ . Therefore, ∠BAC = 360◦ − 90◦ − 90◦ − ∠QAS = 120◦ .

B M C
Drop the perpendicular from A to side BC and let this perpendicular intersect BC at M .
Then by symmetry, M is the midpoint of BC and ∠BAM = ∠CAM = ∠BAC/2 = 120/2 =
60◦ . Therefore, ∆ABM is a 30 − 60 − 90 triangle. Consequently,
√
BM 3
= .
BA 2
√ √ √
Hence,
√ we have BM = 4 3/2 = 2 3. Similarly, CM = 2 3. Therefore, BC = BM + CM =
4 3.

Solution 2: By Solution 1, AB = AC = 4 and ∠BAC = 120◦ . By the Cosine Law, we have

√
BC = √AB 2 + AC 2 − 2 · AB · AC · cos ∠BAC
= p42 + 42 − 2 · 4 · 4 · cos 120◦ √ √
√
= 32 − 32 · (−1/2) = 32 + 16 = 48 = 4 3.
√
Therefore, x = 4 3.

8
COMC 2011 9

Solution 3:
Q N S

P R
A

B M C
Let M, N be the midpoints of BC and QS, respectively. By symmetry, M, A, N are collinear
and the line M N is perpendicular to lines QS and BC. By Solution 1, ∠QAS = 60◦ and
∠BAC = 120◦ . Therefore, by symmetry, ∠QAN = 30◦ and ∠BAM = 60◦ . Since ∆AQS
is equilateral, ∠AQN = 60◦ and ∠ABM = 180◦ − ∠BAM − ∠AM B = 180◦ − 60◦ − 90◦ =
30◦ . Since AB = AQ, ∆AN Q is congruent to ∆BM A. Therefore, BM = AN . By the
Pythagorean Theorem,
p p √ √
BM = AN = AQ2 − QN 2 = 42 − 22 = 12 = 2 3.
√
Hence, x = BC = 2 · BM = 2 3.

9
COMC 2011 10

B1. Arthur is driving to David’s house intending to arrive at a certain time. If he drives at 60
km/h, he will arrive 5 minutes late. If he drives at 90 km/h, he will arrive 5 minutes early.
If he drives at n km/h, he will arrive exactly on time. What is the value of n?

The answer is 72.

Solution 1: Let d be the distance from Arthur to David’s house in km and t the time, in
hours, for Arthur to drive to David’s place driving at n km/h. If he drives at 60 km/h,
Arthur will drive for t hours + 5 minutes = t + 5/60 hours. If he drives at 90km/h, Arthur
will drive for t hours - 5 minutes = t − 5/60 hours. Therefore, using the distance = speed ×
time formula, we have
d = nt = 60(t + 5/60) = 90(t − 5/60). (1)
This simplifies to
15
d = nt = 60t + 5 = 90t − , (1)
2
We first determine t. Using the right-most equation of (1), we have 30t = 5 + 15 25
2 = 2 .
Therefore, t = 25/60. Hence, d = 60t + 5 = 60(25/60) + 5 = 30. Consequently, n = d/t =
30/(25/60) = 30 × 60/25 = 72 km/h.

Solution 2: Let d be the distance from Arthur to David’s house. Note that the time it takes
for Arthur to drive to David’s place at n km/h is the average of the times it take for Arthur
to drive to David’s place at 60 and 90 km/h, respectively. Hence,
d d
d 60 + 90
= .
n 2
Dividing both sides by d and cross multiplying yields
2 1 1 5
= + = .
n 60 90 180
Hence, 5n = 360. Therefore, n = 72.

10
COMC 2011 11

B2. Integers a, b, c, d, and e satisfy the following three properties:

(i) 2 ≤ a < b < c < d < e < 100

(ii) gcd(a, e) = 1
(iii) a, b, c, d, e form a geometric sequence.

What is the value of c?

Solution: The answer is 36.

Let r be the common ratio of the geometric sequence a, b, c, d, e. Since a < b < c < d < e,
r > 1. Then a = a, b = ar, c = ar2 , d = ar3 , e = ar4 . Since a, e have no common factors
and a > 1, r is not an integer. Let x/y be this common ratio, where x, y are positive
integers and gcd(x, y) = 1. Since r > 1 and is not an integer, x > y > 1. Therefore,
b = ax/y, c = ax2 /y 2 , d = ax3 /y 3 and e = ax4 /y 4 . Since e is an integer and gcd(x, y) = 1,
a is divisible by y 4 . Then a = ky 4 for some positive integer k. Then a = ky 4 , b = kxy 3 , c =
kx2 y 2 , d = kx3 y, e = kx4 . Since gcd(a, e) = 1, k = 1. Hence, a = y 4 and e = x4 . Since
2 ≤ a < e < 100 and 34 < 100 < 44 , 2 ≤ y < x ≤ 3, which implies that x = 3 and y = 2.
Then c = kx2 y 2 = 1 · 32 · 22 = 62 = 36.

11
COMC 2011 12
√
B3. In the figure, BC is a diameter of the circle, where BC = 901, BD = 1, and DA = 16. If
EC = x, what is the value of x?
A
E

B C

Solution: The answer is 26.

Solution 1: Since BC is the diameter of the circle, ∠BDC = ∠BEC = 90◦ . By the
Pythagorean Theorem, we have
p p √
CD = BC 2 − BD2 = 901 − 12 = 900 = 30.

Since ∠BDC = 90◦ , ∠ADC = 90◦ . Then by the Pythagorean Theorem, we have
p p
AC = AD2 + DC 2 = 162 + 302 = 34.

Since x = CE,
√ AE = 34 − x.
√ We need to determine x. By the Pythagorean Theorem, we
have BE = BA2 − AE 2 = BC 2 − CE 2 . Hence,

BA2 − AE 2 = BC 2 − CE 2 .

Note that BA = BD + DA = 16 + 1 = 17. Therefore,

172 − (34 − x)2 = 901 − x2

⇒ x2 + 289 = (x − 34)2 + 901
⇒ x2 + 289 = x2 − 68x + 1156 + 901
⇒ 68x = 1768.

Therefore, x = 1768/68 = (17 × 104)/(17 × 4) = 104/4 = 26. Hence, EC = 26.

Solution 2: As in Solution 1, ∠BDC = ∠BEC = 90◦ , CD = 30 and AC = 34. By

computing the area of ∆ABC in two different ways, we have
1 1
the area of ∆ABC = × AB × DC = × AC × BE.
2 2

12
COMC 2011 13

Therefore, AB ·DC = AC ·BE. Hence, 17·30 = 34·BE. Therefore, 30 = 2·BE. Equivalently,

BE = 15. Therefore, by the Pythagorean Theorem,
p p √
EC = BC 2 − BE 2 = 901 − 152 = 676 = 26.

Solution 3: As in Solution 1, ∠BDC = ∠BEC = 90◦ , CD = 30 and AC = 34. Compare

∆ADC and ∆AEB. Note that ∠ADC = ∠AEB = 90◦ and ∠DAC = ∠EAB. Therefore,
∆ADC is similar to ∆AEB. Consequently,
AD AC
= .
AE AB
Therefore, AD · AB = AE · AC. Note that AB = AD + BD = 16 + 1 = 17. Hence,
16 × 17 = AE · 34. Therefore, AE = 8. We then conclude that EC = AC − AE = 34 − 8 = 26.

13
COMC 2011 14

B4. A group of n friends wrote a math contest consisting of eight short-answer problems S1 ,
S2 , S3 , S4 , S5 , S6 , S7 , S8 , and four full-solution problems F1 , F2 , F3 , F4 . Each person in
the group correctly solved exactly 11 of the 12 problems. We create an 8 × 4 table. Inside
the square located in the ith row and j th column, we write down the number of people who
correctly solved both problem Si and problem Fj . If the 32 entries in the table sum to 256,
what is the value of n?
F1 F2 F3 F4
S1
S2
S3
S4
S5
S6
S7
S8

Solution: The answer is 10.

Solution 1: The sum of all of the numbers written is the sum of all of the number of pairs
of problems of the form (Si , Fj ) each student solved. The contribution of each student to this
sum is the product of the number of problems this student solved in the short-answer part
and the number of problems this student solved in the full-solution part. Since each student
solved 11 problems, each student solved either 8 short-answer problems and 3 full-solution
problems, or 7 short-answer problems and 4 full-solution problems. Let x be the number of
students who solved 8 short-answer problems and 3 full-solution problems and y the number
of students who solved 7 short-answer problems and 4 full-solution problems. Then the sum
of the numbers written down is 8 × 3 × x + 7 × 4 × y = 256. Hence, 24x + 28y = 256. Dividing
both sides by 4 yields 6x + 7y = 64. Note that 0 ≤ x ≤ 10. Substituting each such value of
x, we get the following values of y:
x y
0 64/7
1 58/7
2 52/7
3 46/7
4 40/7
5 34/7
6 4
7 22/7
8 16/7
9 10/7
10 4/7

14
COMC 2011 15

We note that only (x, y) = (6, 4) yields a non-negative integer solution for x and y. Hence,
the number of students is x + y = 6 + 4 = 10.

Solution 2: Since each person solved 11 of the 12 problems, there was one problem that
each person did not correctly solve. Let si be the number of people who missed problem Si
(for i = 1, . . . 8) and let fj be the number of people who missed problem Fj (for j = 1, . . . 4).

As in Solution 1, let x be the number of students who solved 8 short-answer problems and 3
full-solution problems, and let y be the number of students who solved 7 short-answer prob-
lems and 4 full-solution problems. By definition, y = s1 + s2 + ... + s8 and x = f1 + f2 + f3 + f4
and n = x + y.

Consider the entry in the ith row and j th column of our 8 × 4 table. This number must be
n − si − fj . Adding all 32 entries, we find that 256 = 32n − 4(s1 + ... + s8 ) − 8(f1 + ... + f4 ) =
32n − 4y − 8x = 32(x + y) − 4y − 8x = 24x + 28y. Therefore, 24x + 28y = 256. We then
complete the problem as in Solution 1.

Solution 3: Let si , fj be as in Solution 2. Then

n = (s1 + s2 + . . . + s8 ) + (f1 + f2 + f3 + f4 ).

Also as in Solution 2, we have

256 = 32n − 4(s1 + s2 + . . . + s8 ) − 8(f1 + f2 + f3 + f4 )

= 32n − 4n − 4(f1 + f2 + f3 + f4 )
= 28n − 4(f1 + f2 + f3 + f4 )

Therefore, 64 = 7n − (f1 + f2 + f3 + f4 ). Hence, n ≥ 10. But note that if n ≥ 11, then

(f1 + f2 + f3 + f4 ) = 7n − 64 = n + (6n − 64) > n.

Since f1 +f2 +f3 +f4 is the number of people that missed a full-solution problem, f1 +f2 +f3 +f4
is at most the number of people in the group, which is n. This contradicts f1 +f2 +f3 +f4 > n.
Hence, n 6≥ 11. This result in conjunction with n ≥ 10 yields n = 10.

15
COMC 2011 16

1 Full Solution Problems

C1. ABC is a triangle with coordinates A = (2, 6), B = (0, 0), and C = (14, 0).
A(2, 6)

B(0, 0) C(14, 0)

(a) Let P be the midpoint of AB. Determine the equation of the line perpendicular to AB
passing through P .
(b) Let Q be the point on line BC for which P Q is perpendicular to AB. Determine the
length of AQ.
(c) There is a (unique) circle passing through the points A, B, and C. Determine the radius
of this circle.

Solution:
(a) The answer is y = −1/3 · x + 10/3 or x + 3y = 10.

The midpoint of AB is  
0+2 0+6
P = , = (1, 3).
2 2
The slope of AB is 6/2 = 3. Therefore, the slope of the line perpendicular to AB is −1/3.
Hence, the equation of the line perpendicular to AB passing through P is
−1
y−3= (x − 1).
3
This is equivalent to
−1 10
y= x+ .
3 3
Rewriting this yields
x + 3y = 10.

(b) The answer is 10.

Solution 1: The line BC is the line y = 0. Since Q lies on BC, the y-coordinate of Q is 0.
Since Q also lies on the line passing through P perpendicular to AB and the equation of this
line is x + 3y = 10, we substitute y = 0 into x + 3y = 10 to yield x = 10. Hence, Q = (10, 0).
Since A = (2, 6), by the Pythagorean Theorem,
p p
AQ = (10 − 2)2 + (0 − 6)2 = 82 + 62 = 10.

16
COMC 2011 17

Solution 2: As in Solution 1, Q = (10, 0). Since Q lies on the perpendicular bisector of AB,
QA = QB. Since Q = (10, 0) and B = (0, 0), QA = QB = 10.

√ √
(c) The answer is 5 2 or 50.

Solution 1: Let O = (x, y) be the centre of the circle. Since O lies on the perpendicular
bisector of BC, x = (0 + 14)/2 = 7. Since O lies on the line perpendicular to AB passing
through P and the equation of the line passing through P perpendicular to AB is x+3y = 10,
we substitute x = 7 into x + 3y = 10 to yield y = 1. Hence, the centre of the circle is at (7, 1).
The radius of the circle is the distance from O to any of A, B, C. For simplicity’s sake, we
compute the length
√ of OB, √ since B √
= (0, 0). By the Pythagorean Theorem, the radius of the
circle is OB = 72 + 12 = 50 = 5 2.

Solution 2: We will use the following property of a triangle; let a, b, c be the side lengths
of a triangle, R the circumradius of the triangle and K the area of the triangle. Then the
quantities a, b, c, R, K have the following relationship;
abc
.
K=
4R
√ √ √ p
In
√ this triangle,
√ AB = 22 + 62 = 40 = 2 10, BC = 14 and CA = (14 − 2)2 + 62 =
180 = 3 20. Note that
1 1
K= × BC × {the height to side BC} = × 14 × 6 = 42.
2 2
Therefore,
√ √ √
AB · BC · CA 2 10 × 14 × 3 20 60 2 × 14 √
R= = = = 5 2.
4K 4 × 42 4 × 42

17
COMC 2011 18

C2. Charlotte writes a test consisting of 100 questions, where the answer to each question is either
TRUE or FALSE. Charlotte’s teacher announces that for every five consecutive questions on
the test, the answers to exactly three of them are TRUE. Just before the test starts, the teacher
whispers to Charlotte that the answers to the first and last questions are both FALSE.

(a) Determine the number of questions for which the correct answer is TRUE.
(b) What is the correct answer to the sixth question on the test?
(c) Explain how Charlotte can correctly answer all 100 questions on the test.

Solution

(a) The answer is 60.

Split the 100 problems into groups of 5, namely 1−5, 6−10, 11−15, . . . , 91−95, 96−100.
Since there are 100 problems and five problems per group and every set of five consecutive
problems contain exactly three problems whose answer is TRUE, each group contains
three problems whose answers are TRUE. Since there are 20 groups, there are 20×3 = 60
problems whose answers are TRUE on the test.

(b) Consider the problems 1, 2, 3, 4, 5, 6. Among problems 1−5, there are exactly three prob-
lems whose answer is TRUE. Since the answer to the first problem is FALSE, among
problems 2 − 5, exactly three of these problems have answer TRUE. Now consider prob-
lem 6. Since problems 2 − 6 contains exactly three problems whose answers are TRUE
and problems 2 − 5 already contain 3 such problems, the answer to problem 6 is FALSE.

(c) Solution 1: We claim that the answer to problem n has the same answer as problem
n + 5. Consider the problems n, n + 1, n + 2, n + 3, n + 4, n + 5. Note that problems
n, n+1, n+2, n+3, n+4 contain three problems whose answers are TRUE and problems
n + 1, n + 2, n + 3, n + 4, n + 5 contain three problems whose answers are TRUE. Note
that problems n + 1, n + 2, n + 3, n + 4 contain either 2 or 3 problems whose answers
are TRUE. In the former case, the answers to both problem n and problem n + 5 are
TRUE. In the latter case, the answers to both problem n and problem n + 5 are FALSE.
In either case, problems n and n + 5 have the same answer.

By this claim, problems {1, 6, 11, 16, . . . , 91, 96} have the same answers. So do {2, 7, 12,
17, . . . , 92, 97}, {3, 8, 13, 18, . . . , 93, 98}, {4, 9, 14, 19, . . . , 94, 99} and {5, 10, 15, 20, . . . , 95,
100}. For each of these five groups of problems, if we can determine the answer to one
problem in the group, we can determine the answers to every problem in the group. Since
the answer to problem 1 is FALSE, the answers to problems {1, 6, 11, 16, . . . , 91, 96}
are all FALSE. Since problem 100 is FALSE, then the answers to problems {5, 10, 15,
20, . . . , 95, 100} are also FALSE. Since problems 1 and 5 have answers FALSE, and

18
COMC 2011 19

exactly three of problems 1, 2, 3, 4, 5 have answer TRUE, problems 2, 3, 4 have answer

TRUE. Therefore, the answers to the remaining problems {2, 7, 12, 17, . . . , 92, 97}, {3, 8,
13, 18, . . . , 93, 98}, {4, 9, 14, 19, . . . , 94, 99} are all TRUE. Having determined the correct
answer to each question, Charlotte achieves a perfect score by answering FALSE, TRUE,
TRUE, TRUE, FALSE to the first five questions, and repeating this pattern for each
block of five consecutive questions.

Solution 2: As in Solution 1, problems {1, 6, 11, . . . , 96} and {5, 10, 15, . . . , 100} have
answers FALSE. There are 40 such problems. By part (a), 60 of the 100 problems have
answer TRUE. Therefore, the remaining 60 problems, mainly, {2, 7, . . . , 97}, {3, 8, . . . , 98}
and {4, 9, . . . , 99}, have answer FALSE.

Comment: An analogue solution to this solution is to define a variable xi for Problem i,

with xi = 1 if the answer to Problem i is TRUE and xi = 0 if the answer to Problem i is
FALSE. Then based on the information given we have the following system of equations:

xj + xj+1 + xj+2 + xj+3 + xj+4 = 3, ∀1 ≤ j ≤ 96

and x1 = 0 and x100 = 0. Charlotte needs to determine all xi where 1 ≤ i ≤ 100. Since
xj ∈ {0, 1}, solving this system of the equations yield

x1 = x6 = x11 = x16 = . . . = x91 = x96 = 0

x2 = x7 = x12 = x17 = . . . = x92 = x97 = 1
x3 = x8 = x13 = x18 = . . . = x93 = x98 = 1
x4 = x9 = x14 = x19 = . . . = x94 = x99 = 1
x5 = x10 = x15 = x20 = . . . = x95 = x100 = 0

19
COMC 2011 20

C3. Let n be a positive integer. A row of n + 1 squares is written from left to right, numbered
0, 1, 2, . . . , n, as shown.

0 1 2 ··· n

Two frogs, named Alphonse and Beryl, begin a race starting at square 0. For each second
that passes, Alphonse and Beryl make a jump to the right according to the following rules: if
there are at least eight squares to the right of Alphonse, then Alphonse jumps eight squares
to the right. Otherwise, Alphonse jumps one square to the right. If there are at least seven
squares to the right of Beryl, then Beryl jumps seven squares to the right. Otherwise, Beryl
jumps one square to the right. Let A(n) and B(n) respectively denote the number of seconds
for Alphonse and Beryl to reach square n. For example, A(40) = 5 and B(40) = 10.

(a) Determine an integer n > 200 for which B(n) < A(n).
(b) Determine the largest integer n for which B(n) ≤ A(n).

Solution 1: Note that if we write n = 8q1 + r1 where q1 , r1 are non-negative integers and
0 ≤ r1 < 8, then Alphonse performs q1 8-square jumps and r1 1-square jump. Then the
number of jumps Alphonse performs is A(n) = q1 + r1 . Similarly, if we write n = 7q2 + r2
where q2 , r2 are non-negative integers and 0 ≤ r2 < 7, then B(n) = q2 + r2 .

(a) Since Alphonse’s 8-square jump is farther than Beryl’s 7-square jump, in order for Beryl
to finish faster than Alphonse, n must be an integer such that Beryl performs very few
1-square jumps and Alphonse performs many 1-square jumps, i.e. n should be an integer
that is divisible by 7 and has a high remainder upon division by 8, i.e. 7. Note that 7
is such an integer. Note that adding 7 × 8 = 56 repeatedly to 7 preserves this property,
i.e. 63, 119, 175, 231. Since 231 = 33 × 7, B(231) = 33. Since 231 = 28 × 8 + 7, Alphonse
performs 28 + 7 = 35 jumps, i.e. A(231) = 35. Therefore, B(231) < A(231). Hence,
n = 231 is such a positive integer.

(b) Since B(n) ≤ A(n), we have q2 +r2 ≤ q1 +r1 . Since 8q1 +r1 = 7q2 +r2 and r2 ≤ q1 +r1 −q2 ,

8q1 + r1 ≤ 7q2 + q1 + r1 − q2 .

Equivalently, 7q1 ≤ 6q2 . Therefore, q2 ≥ 7q1 /6. Substituting this into 8q1 + r1 = 7q2 + r2
yields
49
8q1 + r1 ≥ q1 + r2 .
6
Therefore,
q1
≤ r1 − r2 .
6

20
COMC 2011 21

Since r1 ≤ 7 and r2 ≥ 0, r1 − r2 ≤ 7, which implies that q1 ≤ 42. Since r1 ≤ 7,

n = 8q1 + r1 ≤ 8 × 42 + 7 = 343.

To prove that 343 is indeed the maximum, note that 343 = 42 × 8 + 7, which implies that
A(343) = 42 + 7 = 49. Also, note that 343 = 49 × 7, which implies that B(343) = 49.
Therefore, A(343) = B(343). Hence, n = 343 is the maximum positive integer such that
B(n) ≤ A(n).

Solution 2:
Using the notation in Solution 1, we have A(n) = q1 + r1 and B(n) = q2 + r2 . Let ⌊x⌋ be the
greatest integer less than or equal to x. For example, ⌊ 23
8 ⌋ = 2. Note that q = ⌊n/8⌋. Then
r = n − 8q = n − 8⌊n/8⌋. Hence, A(n)q + r = n − 7⌊n/8⌋. Similarly, B(n) = n − 6⌊n/7⌋.

(a) We seek an integer n > 200 for which B(n) = n − 6⌊ n7 ⌋ < n − 7⌊ n8 ⌋ = A(n), i.e.,
7⌊ n8 ⌋ < 6⌊ n7 ⌋. If we were to remove the floor notation, the inequality would reduce to
7n 6n n n
8 < 7 , which is not true. Thus, in order to achieve the inequality 7⌊ 8 ⌋ < 6⌊ 7 ⌋, we
n n n n
want to make 8 − ⌊ 8 ⌋ as large as possible and 7 − ⌊ 7 ⌋ as small as possible. One way
to achieve this is to make n8 just less than an integer, so that ⌊ n8 ⌋ will be approximately
n n n n
8 − 1, while ensuring that 7 is equal to an integer, so that ⌊ 7 ⌋ = 7 .

Let n = 56k + 7, for some integer k > 0. Then ⌊ n8 ⌋ = ⌊ 56k+7 7

8 ⌋ = ⌊7k + 8 ⌋ = 7k, and
⌊ n7 ⌋ = ⌊ 56k+7
7 ⌋ = 8k + 1. Then our inequality becomes 7 · 7k < 6 · (8k + 1), which is
equivalent to k < 6. For example, if k = 4, then n = 56 · 4 + 7 = 231 is an integer
satisfying 7⌊ n8 ⌋ < 6⌊ n7 ⌋, which implies that B(231) < A(231). Checking, we see that
A(231) = 231 − 7⌊ 231 231
8 ⌋ = 35 and B(231) = 231 − 6⌊ 7 ⌋ = 33. Thus, n = 231 is indeed
a solution to the problem. Another solution is n = 56 · 5 + 7 = 287, found by letting
k = 5. Other solutions include n = 238 and n = 239.

(b) For each positive integer n, there exist unique integers p, q, r for which n = 56p + 8q + r,
where 0 ≤ q ≤ 6 and 0 ≤ r ≤ 7. The inequality B(n) ≤ A(n) is equivalent to
7⌊ n8 ⌋ ≤ 6⌊ n7 ⌋.

We have ⌊ n8 ⌋ = ⌊ 56p+8q+r
8 ⌋ = 7p + q + ⌊ 8r ⌋ = 7p + q, since 0 ≤ r ≤ 7. And also we have
56p+8q+r
⌊ n7 ⌋ = ⌊ 7 ⌋ = 8p + q + ⌊ q+r
7 ⌋.

Thus, the inequality 7⌊ n8 ⌋ ≤ 6⌊ n7 ⌋ is equivalent to 7(7p + q) ≤ 6(8p + q) + 6⌊ q+r

7 ⌋, which
simplifies to p + q ≤ 6⌊ q+r
7 ⌋. Since q + r ≤ 6 + 7 = 13, we must have p + q ≤ 6⌊ 13
7 ⌋ = 6.

We wish to determine the largest integer n = 56p + 8q + r for which the above inequality
is satisfied. To do this, we want to maximize p. Since p + q ≤ 6, let us first try p = 6.
Then this forces q = 0. This case satisfies the inequality p + q ≤ 6⌊ q+r
7 ⌋ if and only if

21
COMC 2011 22

r = 7. We remark that the triplet (p, q, r) = (6, 0, 7) yields n = 56 × 6 + 7 = 343 which

is indeed a solution because A(343) = 343−7⌊ 343 343
8 ⌋ = 49 and B(343) = 343−6⌊ 7 ⌋ = 49.

To show that n = 343 is indeed the largest value of n satisfying B(n) ≤ A(n), we note
that (p, q, r) = (6, 0, 7) is the only triplet satisfying the inequality for p = 6, from the
analysis above. And so any other solution must have p ≤ 5. But then such a solution
would have n = 56p + 8q + r ≤ 56 × 5 + 8 × 6 + 7 = 335 < 343.

22
COMC 2011 23

C4. Let f (x) = x2 − ax + b, where a and b are positive integers.

(a) Suppose a = 2 and b = 2. Determine the set of real roots of f (x) − x, and the set of real
roots of f (f (x)) − x.
(b) Determine the number of pairs of positive integers (a, b) with 1 ≤ a, b ≤ 2011 for which
every root of f (f (x)) − x is an integer.

Solution:
(a) If a = 2 and b = 2, then f (x) = x2 − 2x + 2. Hence, f (x) − x = x2 − 3x + 2 = (x − 2)(x − 1).
Therefore, the roots of f (x) − x are 1 and 2.

We now determine f (f (x)) − x. Note that f (f (x)) = (x2 − 2x + 2)2 − 2(x2 − 2x + 2) + 2 =

x4 − 4x3 + 6x2 − 4x + 2. Therefore,

f (f (x)) − x = x4 − 4x3 + 6x2 − 5x + 2.

Note that 1 is a root of f (f (x)) − x. Then

x4 − 4x3 + 6x2 − 5x + 2 = (x − 1)(x3 − 3x2 + 3x − 2) = (x − 1)(x − 2)(x2 − x + 1).

Note that x2 −x+1 has no real roots since its discriminant is 12 −4·1·1 = −3 < 0. Therefore,
the real roots of f (f (x)) − x are 1 and 2.

(b) The answer is 43.

First, we claim that if r is a root of f (x) − x, then r is a root of f (f (x)) − x. Since r is a root
of f (x) − x, f (r) − r = 0, i.e. f (r) = r. Therefore,

f (f (r)) − r = f (r) − r = 0.

Hence, any root of f (x) − x is a root of f (f (x)) − x. Consequently, f (x) − x is a factor of

f (f (x)) − x.

Note that f (f (x)) − x = f (x2 − ax + b) − x = (x2 − ax + b)2 − a(x2 − ax + b) + b − x,

= x4 − 2ax3 + (a2 + 2b − a)x2 − (2ab − a2 + 1)x + (b2 − ab + b).

Since f (x) − x = x2 − (a + 1)x + b, f (f (x)) − x factors as

f (f (x)) − x = (x2 − (a + 1)x + b)(x2 − (a − 1)x + (b − a + 1)).

Since both factors are monic, every root of f (f (x)) − x is an integer if and only if the
discriminants of both of these quadratic factors are perfect squares. These two discriminants
are
(a + 1)2 − 4b = a2 + 2a + 1 − 4b

23
COMC 2011 24

and
(a − 1)2 − 4(b − a + 1) = a2 + 2a + 1 − 4b − 4.
The first discriminant is four larger than the second discriminant. The only two perfect
squares that differ by 4 are 4 and 0. This statement is true since if r, s are non-negative integers
such that r2 − s2 = 4, then (r − s)(r + s) = 4. Since r, s are non-negative, (r − s, r + s) = (2, 2)
or (1, 4). In the latter case, r − s = 1 and r + s = 4. Therefore, r = 5/2 and s = 3/2, which
are not integers. Therefore, (r − s, r + s) = (2, 2), i.e. r = 2, s = 0. Hence, the larger perfect
square is 22 = 4 and the smaller perfect square is 0.

Therefore, a2 + 2a + 1 − 4b = 4. Rearranging this and factoring yields

(a + 1)2 = 4(b + 1).

Since (a+1)2 and 4 are perfect squares, b+1 is a perfect square. Therefore, there exists a pos-
itive integer m such that b + 1 = m2 . Then b = m2 − 1. Consequently, (a + 1)2 = 4m2 . Since
a is a positive integer, a + 1 = 2m. Hence, a = 2m − 1. Therefore, (a, b) = (2m − 1, m2 − 1).

We now verify that all such (a, b) have the property that the roots of x2 − (a + 1)x + b and
x2 − (a − 1)x + (b − a + 1) are all integers, implying that every root of f (f (x)) − x is an integer.
Substituting (a, b) = (2m − 1, m2 − 1) into these two polynomials yield x2 − 2mx + m2 − 1 =
(x−(m−1))(x−(m+1)) and x2 −(2m−2)x+(m2 −2m+1) = (x−(m−1))(x−(m−1)). Since
m is a positive integer, all four roots of f (f (x)) − x is an integer. (Alternately, note that since
the leading coefficient of each of the quadratic factors is 1, the roots of the quadratic factors
are all integers if and only if the discriminant of the quadratic factors are both perfect square.)

Since 1 ≤ a, b ≤ 2011, it remains to find the number of positive integers m such that
1 ≤ 2m − 1, 2 2 2
√m − 1 ≤ 2011. Since 1 ≤ m − 1 ≤ 2011, 2 ≤ m ≤ 2012. Hence,
2 ≤ m ≤ ⌊ 2012⌋ = 44, where ⌊t⌋ denotes the largest integer less than or equal to t.
There are 43 solutions for m, namely m = 2, 3, . . . 44. These values of m clearly satisfy
1 ≤ 2m − 1 ≤ 2011.

Therefore, the number of ordered positive integer pairs (a, b) that results in f (f (x))−x having
all integer roots is 43.

24
COMC 2012 Official Solutions 1

COMC 2012 Official Solutions

by Adrian Tang ([email protected])

A1 Determine the positive integer n such that 84 = 4n .

Solution: The answer is n = 6.

Solution 1: Note that 84 = (23 )4 = 212 = 46 . Therefore, n = 6. 

Solution 2: We write 84 and 4n as an exponent with base 2.

84 = 4n
(23 )4 = (22 )n
212 = 22n

Therefore, 2n = 12. Hence, n = 6. 

Solution 3: Note that 84 = (82 )2 = 642 = 4096. Hence, 4n = 4096. We check each positive
integer n starting from 1.

n 4n
1 4
2 16
3 64
4 256
5 1024
6 4096

All positive integers n > 6 yield a value of 4n larger than 4096. Therefore, n = 6. 

1
COMC 2012 Official Solutions 2

A2 Let x be the average of the following six numbers: {12, 412, 812, 1212, 1612, 2012}. Determine
the value of x.

Solution: The answer is x = 1012.

Solution 1: The sum of the first and sixth terms is 2024. The sum of the second and fifth
terms is 2024 and the sum of the third and fourth terms is 2024. Hence, the sum of the six
terms is 2024 × 3. Hence, the average of the six terms is
2024 × 3 2024
= = 1012. 
6 2

Solution 2: The average of the six numbers is

12 + 412 + 812 + 1212 + 1612 + 2012
6
0 + 400 + 800 + 1200 + 1600 + 2000 12 + 12 + 12 + 12 + 12 + 12
= +
6 6
100(4 + 8 + 12 + 16 + 20) 100 × 60
= + 12 = + 12 = 1000 + 12 = 1012. 
6 6

Solution 3: Note that the sequence is arithmetic.1 Therefore, the average of the six num-
bers is the average of the middle two numbers, which is the halfway point between 812, 1212.
Hence, the answer is 1012. 

1
A sequence is said to be arithmetic if successive terms in the sequence have a common difference.

2
COMC 2012 Official Solutions 3

A3 Let ABCDEF be a hexagon all of whose sides are equal in length and all of whose angles
are equal. The area of hexagon ABCDEF is exactly r times the area of triangle ACD.
Determine the value of r.
A B

F C

E D
Solution 1: The answer is r = 3.

Divide the hexagon into six regions as shown, with the centre point denoted by P .

A B

P
F C

E D
This is possible since the hexagon is regular. By symmetry, note that P A = P B = P C =
P D = P E = P F and the six interior angles about P are equal. Then since the sum of the
six interior angles about P sum to 360◦ ,

∠AP B = ∠BP C = ∠CP D = ∠DP E = ∠EP F = ∠F P A = 60◦ .

Therefore, the six triangles ∆P AB, ∆P BC, ∆P CD, ∆P DE, ∆P EF, ∆P F A are all equilat-
eral and have the same area. Let K be the area of any one of these triangles. Therefore, the
hexagon has area 6K.

Note that the area of ∆ACD is equal to the area of ∆P CD plus the area of ∆P AC. Since
∆P AB, ∆P BC are both equilateral, P A = AB and P C = CB. Therefore, triangles ∆BAC
and ∆P AC are congruent and hence have the same area. Note that the area of ∆P AC plus
the area of ∆BAC is the sum of the areas of the equilateral triangles ∆P AB and ∆P BC,
which is 2K. Therefore, ∆P AC has area K. We already noted that the area of ∆ACD is

3
COMC 2012 Official Solutions 4

equal to the area of ∆P CD plus that of ∆P AC. This quantity is equal to K + K = 2K.
Hence, the area of ABCDEF is 6K/2K = 3 times the area of ∆ACD. The answer is 3. 

Solution 2: Divide the hexagon into six regions and define K as in Solution 1. Note
that ∆AP C and ∆DP C have a common height, namely the height from C to AD. Since
P A = P D, ∆AP C and ∆DP C have the same area, namely K. Therefore, the area of ∆ACD
is the sum of the areas of ∆AP C and that of ∆DP C, which is K + K = 2K. Hence, the
ratio of the area of ABCDEF to the area of ∆ACD is 6K/2K = 3. 

Solution 3: The sum of the angles of a hexagon is 180◦ × (6 − 2) = 720◦ . Therefore,

∠ABC = 120◦ . Since BA = BC, ∠BAC = ∠BCA. Then since the sum of the angles
of ∆ABC is 180◦ and ∠ABC = 120◦ , ∠BAC = ∠BCA = 30◦ . Since ∠BCA = 30◦ and
∠BCD = 120◦ , ∠ACD = 90◦ .
A B

F C

E D
Suppose that each side of the hexagon has length 1. We now determine the length AC to
determine the area of ∆ACD. By the cosine law,
AC 2 = BA2 + BC 2 − 2 · BA · BC · cos ∠ABC = 12 + 12 − 2 · 1 · 1 · cos 120◦ = 2 − 2 · (−1/2) = 3.
√ √ √
Therefore, AC = 3. Hence, the area of ∆ACD is 1/2 · CD · CA = 1/2 · 1 · 3 = 3/2.

We now find the area of the hexagon. As in Solution 1, the hexagon √ consists of 6 equilateral
triangles each with side
√ 1. The√area of each equilateral triangle is 3/4. Therefore, the area
of the hexagon is 6 · 3/4 = 3 3/2. Therefore, the ratio of the area of the hexagon to the
area of ∆ACD is √
3 3/2
√ = 3.
3/2
Therefore, the answer is 3. 

As in Solution 3, suppose each side of the hexagon has length 1. Then the area
Solution 4: √
of ∆ACD is 3/2. Note that the area of ∆ABC is
√ √
1 1 3 3
· BA · BC · sin 120 = · 1 · 1 ·
◦
= .
2 2 2 4

4
COMC 2012 Official Solutions 5

Therefore, ∆ACD is twice the area of ∆ABC and then ∆ACD is 2/3 the area of the quadri-
lateral ABCD. But the line AD splits the hexagon ABCDEF in half. Therefore, ∆ACD
is 1/3 the area of the entire hexagon. Therefore, the ratio of the area of the hexagon to the
area of ∆ACD is 3. 

Solution 5: We will follow Solution 4, but provide a different way to show that the area
of ∆ACD is twice the area of ∆ABC. Note that these two triangles have a common height
with base AD and BC, respectively. Since AD is twice the length of BC, ∆ACD is twice
the area of ∆ABC. Then as in Solution 4, we can conclude that the ratio of the area of the
hexagon to the area of ∆ACD is 3. 

Solution 6: Join the segment DF . The hexagon is cut into four triangles. By symme-
try, ∆ACD and ∆AF B are congruent, as are ∆ABC and ∆DEF . Note that AD||BC and
F C||BF . Let AD, F C meet at P . Then ∆AP C and ∆ABC are congruent (parallelogram
cut by diagonal). ∆AP C has half the height of ∆ABC on base ∆AC (by symmetry), so
[ACD] = 2[ABC], where [· · · ] denotes the area of a figure. Similarly, [AF D] = 2[ABC] Thus
the hexagon’s area is [ACD] + [ADF ] + [ABC] + [DEF ] = 6[ABC] = 3[ACD]. Therefore,
the answer is 3. 

5
COMC 2012 Official Solutions 6

A4 Twelve different lines are drawn on the coordinate plane so that each line is parallel to exactly
two other lines. Furthermore, no three lines intersect at a point. Determine the total number
of intersection points among the twelve lines.
. . . .
. ..... ..... .. ....
. . . .. .. .
.. .. .. .. ..

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

Solution: The answer is 54.

Solution 1: Since no point lies on three or more lines, the number of intersection points
is equal to the number of pairs of lines that intersect. The total number of pairs of lines is
12 × 11/2 = 6 × 11 = 66. Each line is parallel to two other lines. Hence, each line is part of
two pairs of lines that do not intersect. Since there are twelve lines, there are 12 × 2/2 = 12
pairs of lines that do not intersect. Therefore, there are 66 − 12 = 54 pairs of lines that
intersect. Hence, the answer is 54. 

Solution 2: Since each line is parallel to exactly two other lines, each line is not parallel to
nine other lines. Hence, each line intersects nine other lines.

Since no point lies on three or more lines, each point of intersection lies on exactly two lines.
Combining these two observations yields that the total number of intersection points is
12 × 9
= 54. 
2

Solution 3: Recall that two lines are parallel if and only if they have the same slope. Con-
sider the set of slopes of the twelve lines; since each line is parallel to exactly two other lines,
each slope is the slope of three lines among the twelve lines. Hence, there are four different
slopes represented among the twelve lines.

6
COMC 2012 Official Solutions 7

Since each slope contains three lines, each pair of slopes contains 3 × 3 = 9 points of in-
tersection. There are four different slopes. Hence, the number of pairs of different slopes
is 4 × 3/2 = 6. Since no three lines intersect at a common point, the number of points of
intersection is 9 × 6 = 54. 

7
COMC 2012 Official Solutions 8

B1 Alice and Bob run in the clockwise direction around a circular track, each running at a
constant speed. Alice can complete a lap in t seconds, and Bob can complete a lap in 60
seconds. They start at diametrically-opposite points.

Alice Bob

When they meet for the first time, Alice has completed exactly 30 laps. Determine all possible
values of t.

Solution: The answer is t = 59 or t = 61.

Since Alice ran exactly 30 laps, Bob meets Alice at where Alice started. Since Bob started
diametrically across from Alice, Bob ran n + 21 laps for some positive integer n. Since Alice
and Bob meet only the first time they meet, the number of laps that Alice ran and the number
of laps Bob ran cannot differ by more than 1. Therefore, Bob ran either 29.5 laps or 30.5 laps.

Note that Alice and Bob ran for the same amount of time and the number of seconds each
person ran is the number of laps he/she ran times the number of seconds it takes he/she to
complete a lap.

If Bob ran 29.5 laps, then 30t = 29.5 × 60. Hence, t = 29.5 × 2 = 59.

If Bob ran 30.5 laps, then similarly, 30t = 30.5 × 60. Hence, t = 30.5 × 2 = 61.

Therefore, t = 59 or t = 61. 

8
COMC 2012 Official Solutions 9

B2 For each positive integer n, define ϕ(n) to be the number of positive divisors of n. For exam-
ple, ϕ(10) = 4, since 10 has 4 positive divisors, namely {1, 2, 5, 10}.

Suppose n is a positive integer such that ϕ(2n) = 6. Determine the minimum possible value
of ϕ(6n).

Solution: The answer is 8.

Solution 1: Recall that if a positive integer m has prime factorization pe11 pe22 . . . pet t , where
p1 , . . . , pt are distinct primes, then the number of positive divisors of m is ϕ(m) = (e1 +
1)(e2 + 1) . . . (et + 1) (*). Note that each term in this product is at least 2.

Since 2n is an even positive integer with 6 positive divisors, 2n = 25 , 22 · p or 2 · p2 , where p is

some odd prime number. Therefore, n = 24 = 16, 2p or p2 . Therefore, 6n = 6 × 16 = 96, 12p
or 6p2 .

Note that ϕ(96) = ϕ(25 × 31 ) = 6 × 2 = 12.

If p = 3, then ϕ(12p) = ϕ(36) = ϕ(22 × 32 ) = 3 × 3 = 9 and ϕ(6p2 ) = ϕ(54) = ϕ(21 × 33 ) =

2 × 4 = 8.

It remains to show the case when p > 3. So far the minimum value obtained for ϕ(6n) = 8.
If p > 3, then 6n contains at least 3 different prime divisors. Then by (*), the number of
positive divisors of 6n is at least 2 × 2 × 2 = 8. Therefore, ϕ(6n) ≥ 8 for all positive integers
n. As we have shown, n = 9 yields ϕ(6n) = 8. Therefore, the answer is 8. 

Solution 2: Note that four positive divisors of 2n are 1, 2, n and 2n. Note also that n = 2
does not satisfy ϕ(2n) = 6. Therefore, n ≥ 2 and consequently, 1, 2, n and 2n are all distinct.

Since 2n has 6 positive divisors, there are two other positive divisors a, b of 2, n, with a, b > 2.
Then the set of positive divisors of 2n is {1, 2, a, b, n, 2n}.

Now consider the positive divisors of 6n. Note that the set of positive divisors of 6n contains
those of 2n. Further note that 3n and 6n are positive divisors of 6n, which are not posi-
tive divisors of 2n. Hence, the set of positive divisors of 6n contains {1, 2, a, b, n, 2n, 3n, 6n}.
Therefore, ϕ(6n) ≥ 8.

We will show that this minimum can be obtained. Since {1, 2, a, b, n, 2n, 3n, 6n} are the posi-
tive divisors of 6n and appear in increasing order, a · 2n = 6n and bn = 6n. Multiplying both

9
COMC 2012 Official Solutions 10

equations and dividing both sides by 2n2 yield ab = 18. But since {1, 2, a, b, n, 2n} are the
positive divisors of 2n, ab = 2n. Therefore, 2n = 18, from which we can conclude that n = 9
is a candidate which yields ϕ(2n) = 6 and ϕ(6n) = 8.

This can be easily verified, since the positive divisors of 18 are {1, 2, 3, 6, 9, 18}. Since the
positive divisors of 54 are {1, 2, 3, 6, 9, 18, 27, 54}, ϕ(54) = 8. 

10
COMC 2012 Official Solutions 11

B3 Given the following 4 by 4 square grid of points, determine the number of ways we can label
ten different points A, B, C, D, E, F, G, H, I, J such that the lengths of the nine segments
AB, BC, CD, DE, EF, F G, GH, HI, IJ
are in strictly increasing order.

Solution: The answer is 24.

First, we count the number of possible

√ lengths of
√ the segments.√By the Pythagorean
√ The-
orem, 2 2 2 2 2 2 2 2 =
√ √ the different√ lengths
√ are √0 +√1 = 1, 0 √+ 2√ = 2, 0 + √3 √ = 3, 1 + 1√
2 2 2 2 2 2 2
2, 1 + 2 = 5, 1 + 3 = 10, 2 + 2 = 8, 2 + 3 = 13, 3 + 3 = 18.2 2 2

These nine lengths are all different. Therefore, all nine lengths are represented among
AB, BC, CD, DE, EF, F G, GH, HI, IJ. Furthermore, these nine lengths in increasing or-
der are: p p p p p
02 + 12 < 12 + 12 < 02 + 22 < 12 + 22 < 22 + 22
p p p p
< 02 + 32 < 12 + 32 < 22 + 32 < 32 + 32 .
Hence, the longest length must be a segment that goes from one corner to the diagonally-
opposite corner.

We will construct the ten points in the order J, I, H, G, F, E, D, C, B, A.

For simplicity, we place the points on the coordinate plane, with the bottom left corner at
(0, 0) and the top right corner at (3, 3).

Note that J must be a corner of the grid, and there are four such corners. Furthermore, I
must be the diagonally opposite corner from J. Without loss of generality, suppose J = (0, 0).
Then I = (3, 3).

11
COMC 2012 Official Solutions 12
√
The point H has the property that HI = 32 + 22 , i.e. H is a point which is distance three
horizontally from I and distance two vertically from I, or vice versa. By symmetry along the
diagonal JI, there two choices for H, namely (0, 1) or (1, 0). Without loss of generality, sup-
pose H = (1, 0).

J H

√
The segment GH has length 32 + 12 . Hence, √ G is either (0, 3) or (2, 3). But if G = (0, 3)
then F is a point such that F G = 3 = 02 + 32 . Then F = (0, 0) or (3, 3), which are
already occupied by J, I, respectively. Therefore, G cannot be (0, 3), and thus must be (2, 3).
Consequently, F = (2, 0).

G I

J H F

√ √ √ √
EF has length 8= 22 + 22 . Hence, E = (0, 2). DE has length 5= 22 + 12 . Hence,
D = (2, 1).

G I
E
D
J H F

Then C = (0, 1) and B = (1, 2).

12
COMC 2012 Official Solutions 13
G I
E B
C D
J H F

From B, there are three remaining points A such that AB = 1, namely (1, 1), (1, 3), (2, 2).

By our construction, the points J, H and A were the only points where there was more than
one choice. Every other point was determined from our construction. There were 4 choices
for J, 2 choices for H and 3 choices for A. Hence, the number of ways to select 10 points
that satisfy the condition given in the problem is 4 × 3 × 2 = 24. The answer is 24. 

13
COMC 2012 Official Solutions 14

B4 In the following diagram, two lines that meet at a point A are tangent to a circle at points
B and C. The line parallel to AC passing through B meets the circle again at D. Join the
segments CD and AD. Suppose AB = 49 and CD = 28. Determine the length of AD.

B D

A C
Solution 1: The answer is AD = 63.

Join the segment BC. Since the two lines are both tangent to the circle, AB = AC. There-
fore, ∠ABC = ∠ACB.

B D

A C
Furthermore, since BD is parallel to AC, ∠ACB = ∠DBC. Since AC is tangent to the
circle at C, by the tangent-chord theorem, ∠BDC = ∠ACB. Hence, we have the following
sequence of equal angles:

∠ABC = ∠ACB = ∠CBD = ∠CDB.

Furthermore, AB = AC and CB = CD. Therefore, ∆ABC is similar to ∆CBD. Hence.

AB CB
= .
BC BD
Since AB = 49 and BC = CD = 28, BD = BC 2 /AB = 282 /49 = 42 = 16.

Let M be the foot of the perpendicular from D on AC and N the foot of the perpendicular
on BD from C.

14
COMC 2012 Official Solutions 15

B N D

A C M
Since CB = CD, N is the midpoint of BD. Since BD is parallel to CM , N DM C is a
rectangle. Therefore, CM = N D = 21 · BD = 8. We now determine the length of DM .
p p √ √
DM = N C = DC 2 − DN 2 = 282 − 82 = 784 − 64 = 720.

Therefore,
p p p √
AD = AM 2 + M D2 = (AC + CM )2 + M D2 = (49 + 8)2 + 720 = 3969 = 63.

Hence, the answer is 63. 

Solution 2: As in Solution 1, join segment BC and conclude that BC = 28 and BD = 16.

Also as in Solution 1, ∠ABC = ∠ACB = ∠CBD = ∠CDB. Let θ be this angle.

B N D
θθ

A C
Let N be the foot of the perpendicular on BD from C. As in Solution 1, N is the midpoint
of BD. Therefore, BN = 8. We can now determine cos θ from ∆CBN , which is
BN 8 2
cos θ = = = .
BC 28 7
(Alternatively, we can use ∆ABC to determine cos θ.) Note that ∠ABD = 2θ. We then
apply the cosine law on ∆ABD to determine AD. By the cosine law, we have

AD2 = BA2 + BD2 − BA · BD · cos(2θ).

We first determine cos(2θ), which by the double-angle formula is

cos(2θ) = 2 cos2 θ − 1 = 2 · (2/7)2 − 1 = −41/49.

15
COMC 2012 Official Solutions 16

Therefore,

AD2 = 492 + 162 − 2(49)(16)(−41/49) = 2401 + 256 + 2 · 16 · 41 = 3969.

√
Hence, AD = 3969 = 63. 

Solution 3: Let θ be defined as in Solution 2. Then as shown in Solution 2, cos θ = 2/7.

Then note that
∠BCD = 180 − ∠CBD − ∠CDB = 180 − 2θ.
Therefore, ∠ACD = ∠ACB + ∠BCD = θ + (180 − 2θ) = 180 − θ. We now apply the cosine
law on ∆ACD.

AD2 = CA2 + CD2 − 2 · CA · CD · cos ∠ACD = 492 + 282 − 2 · 49 · 28 · cos(180 − θ)

2
= 2401 + 784 + 2 · 49 · 28 · cos θ = 3185 + 2 · 49 · 28 · = 3185 + 4 · 7 · 28 = 3969.
7
√
Therefore, AD = 3969 = 63. 

16
COMC 2012 Official Solutions 17

Part C

C1 Let f (x) = x2 and g(x) = 3x − 8.

(a) (2 marks) Determine the values of f (2) and g(f (2)).

(b) (4 marks) Determine all values of x such that f (g(x)) = g(f (x)).
(c) (4 marks) Let h(x) = 3x − r. Determine all values of r such that f (h(2)) = h(f (2)).

Solution:

(a) The answers are f (2) = 4 and g(f (2)) = 4.

Substituting x = 2 into f (x) yields f (2) = 22 = 4.

Substituting x = 2 into g(f (x)) and noting that f (2) = 4 yields g(f (2)) = g(4) =
3 · 4 − 8 = 4. 

(b) The answers are x = 2 and x = 6.

Note that
f (g(x)) = f (3x − 8) = (3x − 8)2 = 9x2 − 48x + 64
and
g(f (x)) = g(x2 ) = 3x2 − 8.
Therefore, we are solving
9x2 − 48x + 64 = 3x2 − 8.
Rearranging this into a quadratic equation yields

6x2 − 48x + 72 = 0 ⇒ 6(x2 − 8x + 12) = 0.

This factors into 6(x − 6)(x − 2) = 0. Hence, x = 2 or x = 6. We now verify these are
indeed solutions.

If x = 2, then f (g(2)) = f (3(2) − 8) = f (−2) = (−2)2 = 4 and g(f (2)) = 4 by part(a).

Hence, f (g(2)) = g(f (2)). Therefore, x = 2 is a solution.

If x = 6, then f (g(6)) = f (3 · 6 − 8) = f (10) = 102 = 100 and g(f (6)) = g(62 ) = g(36) =
3·36−8 = 108−8 = 100. Hence, f (g(6)) = g(f (6)). Therefore, x = 6 is also a solution. 

17
COMC 2012 Official Solutions 18

(c) The answers are r = 3 and r = 8.

We first calculate f (h(2)) and h(f (2)) in terms of r.

f (h(2)) = f (3 · 2 − r) = f (6 − r) = (6 − r)2

and
h(f (2)) = h(22 ) = h(4) = 3 · 4 − r = 12 − r.
Therefore, (6 − r)2 = 12 − r ⇒ r2 − 12r + 36 = 12 − r. Re-arranging this yields

r2 − 11r + 24 = 0,

which factors as
(r − 8)(r − 3) = 0.
Hence, r = 3 or r = 8. We will now verify that both of these are indeed solutions.

If r = 3, then h(x) = 3x − 3. Then f (h(2)) = f (3 · 2 − 3) = f (3) = 9 and h(f (2)) =

h(22 ) = h(4) = 3 · 4 − 3 = 9. Therefore, f (h(2)) = h(f (2)). Consequently, r = 3 is a
solution. From the result of part (b), we also verified that r = 8 is a solution. 

18
COMC 2012 Official Solutions 19

C2 We fill a 3 × 3 grid with 0s and 1s. We score one point for each row, column, and diagonal
whose sum is odd.

1 1 0 1 1 1
1 0 1 1 0 1
0 1 1 0 1 1

For example, the grid on the left has 0 points and the grid on the right has 3 points.

(a) (2 marks) Fill in the following grid so that the grid has exactly 1 point. No additional
work is required. Many answers are possible. You only need to provide one.

Solution: Any of the following is a solution:

0 0 0 1 1 0 0 1 1 0 0 0
1 1 0 1 1 0 0 1 1 0 1 1
1 1 0 0 0 0 0 0 0 0 1 1

0 1 1 1 0 1 1 0 1 1 1 0
1 1 0 1 1 0 0 1 1 0 1 1
1 0 1 0 1 1 1 1 0 1 0 1

(b) (4 marks) Determine all grids with exactly 8 points.

Solution: Note that there are three rows, three columns and two diagonals. Hence,
every row, column and diagonal has an odd sum.

We will consider two cases; the first case is when the middle number is 0 and second
case is when the middle number is 1.

Case 1: If the middle number is 0, then let A, B, C, D be the values provided in the
following squares.

19
COMC 2012 Official Solutions 20

A B C
0 D

Then since each row, column and diagonal has an odd sum, each term diametrically
opposite from A, B, C, D has a different value from A, B, C, D, respectively. Denote
0 = 1 and 1 = 0. Then we have the following values in the grid:

A B C
D 0 D
C B A

Note that X + X = 1 for any value X. Then note the sum of A, B, C, A, B, C is

1 + 1 + 1 = 3. Hence, one of A + B + C and A + B + C is even. Therefore, either the
top row or bottom row sum to an even number. Hence, there are no grids with 8 points
in this case.

Case 2: If the middle number is 1, then again, let A, B, C, D be the values provided in
the following squares.

A B C
1 D

Then since each row, column and diagonal has an odd sum, each term diagonally oppo-
site from A, B, C, D has the same value as A, B, C, D, respectively. Then we have the
following values in the grid:

A B C
D 1 D
C B A

Since A + B + C and A + D + C are both odd, B = D.

A B C
B 1 B
C B A

20
COMC 2012 Official Solutions 21

Hence, the only remaining restriction is that A + B + C is odd. Since A, B, C = 0 or

1, A + B + C = 1 or 3. The only triples (A, B, C) that give this result are (A, B, C) =
(1, 0, 0), (0, 1, 0), (0, 0, 1) or (1, 1, 1). The following are the grids corresponding to these
results, which completes the problem. 

1 0 0 0 1 0 0 0 1 1 1 1
0 1 0 1 1 1 0 1 0 1 1 1
0 0 1 0 1 0 1 0 0 1 1 1

(c) (4 marks) Let E be the number of grids with an even number of points, and O be the
number of grids with an odd number of points. Prove that E = O.

Solution 1: Consider the set of all grids. Pair the grids so that each grid G is paired
with the grid G∗ formed by switching the top-left number of G. (By switching, we mean
if the top left number of G is 0, we switch it to a 1. If the top left number of G is 1, we
switch it to a 0.) The following is an example of the action provided by G∗ .

0 0 1 1 0 1
G= 1 1 1 G∗ = 1 1 1
1 0 1 1 0 1

Note that the sum of the elements in the top row, the left most column and the diagonal
going from the top-left to the bottom-right switches parity, i.e. switches either from odd
to even, or even to odd and the sum of the elements of the other rows / columns /
diagonals remain unchanged. Hence, the total number of rows/columns/diagonals which
have odd sum in G and G∗ differ by an odd number. Hence, exactly one of G, G∗ has an
even number of points and the other has an odd number of points. Since each grid lies
in exactly one pair, there is the same number of grids with an even number of points as
grids with an odd number of points, i.e. E = O. 

Comment: The solution also applies if we switch any one of the four corners of the grid.

Solution 2: Note that the grid consisting of all zeros has an even number of points,
namely zero. Note that for any grid, switching the centre square keeps the parity of the
number of points the same. Switching any of the four side squares keeps the parity of
the number of points the same. As in Solution 1, switching the centre changes the parity
of the number of points the same.

Therefore, if a grid has 0, 2 or 4 of its corners as 1, then the number of points of the grid
is even. If a grid has 1 or 3 of its corner as 1, then number of points in the grid is odd.

21
COMC 2012 Official Solutions 22

We will count the number of grids of based on the number of corner squares containing 1.

There are five non-corner squares. Therefore, there are 25 grids with zero corners con-
taining 1.

There are 41 = 4 ways to choose one corner to be 1. Therefore, there are 4 × 25 grids

with one corner containing 1. Similarly, there are 42 ×25 = 6×25 grids with two corners

containing 1, 43 × 25 = 4 × 25 grids with three corners containing 1 and 25 grids with

four corners containing 1.

Hence, there are 25 (1 + 6 + 1) = 8 × 25 grids with an even number of points and

25 (4 + 4) = 8 × 25 grids with an odd number of points. Therefore, E = O, as desired. 

22
COMC 2012 Official Solutions 23

C3 Let ABCD be a parallelogram. We draw in the diagonal AC. A circle is drawn inside ∆ABC
tangent to all three sides and touches side AC at a point P .
A B

D C
(a) (2 marks) Prove that DA + AP = DC + CP .

Solution: Let the circle inside ∆ABC touch AB, BC at X, Y , respectively.

A X B

P Y

D C
Then by equal tangents, we have

DA + AP = DA + AX = DA + AB − BX

and
DC + CP = DC + CY = DC + CB − BY.
By equal tangents, we have BX = BY . Since opposite sides of a parallelogram have
equal lengths, AB = DC and DA = CB. Therefore, DA + AB − BX = DC + CB − BY .
Consequently, DA + AP = DC + CP , as desired. 

(b) (4 marks) Draw in the line DP . A circle of radius r1 is drawn inside ∆DAP tangent to
all three sides. A circle of radius r2 is drawn inside ∆DCP tangent to all three sides.
Prove that
r1 AP
= .
r2 PC

23
COMC 2012 Official Solutions 24

Solution 1: Consider the triangles ∆AP D and ∆CP D and note that the heights of
these triangles to side AP, P C are the same. Therefore,

AP [AP D]
= ,
PC [CP D]

where [· · · ] denotes area.

Q R

S
Given any triangle QRS with a circle on the inside touching all three sides, let O be the
centre of the circle and r the radius of the circle. Then the distance from O to each of
the sides QR, RS, SQ is the same, and is the radius of the circle. Join OQ, OR, OS.
Q
.... .. R
. .... .
. . . ...
.... ..r ...
. . . . ... . . . . .
. ....O .
r.... ........r..
.. .. ... .
... ..
..
..
..
S
Then
r · QR r · RS r · SQ
[QRS] = [OQR] + [ORS] + [OSQ] = + +
2 2 2
r r
= · (QR + RS + SQ) = · (Perimeter of ∆QRS) .
2 2
Then
r1
[AP D] = · (Perimeter of ∆AP D)
2
and
r2
[CP D] = · (Perimeter of ∆CP D)
2
Then
AP [AP D] r1 Perimeter of ∆AP D
= = · .
PC [CP D] r2 Perimeter of ∆CP D
Hence, to prove that AP/P C = r1 /r2 , it suffices to show that ∆AP D, ∆CP D have the
same perimeter.

24
COMC 2012 Official Solutions 25

By part (a), we have DA+AP = DC+CP . The perimeter of ∆AP D is DA+AP +P D =

DC + CP + P D, which is the perimeter of ∆CP D. This solves the problem. 

Solution 2: Let X, Y, Z be the centres of the circles inside ∆ABC, ∆AP D and ∆CP D,
respectively, M the point where the circle inside ∆ADP touch AC and N the point where
the circle inside ∆CDP touch AC. Note that XP , Y M and ZN are each perpendicular
to AC.
A B
.. . .
.. . . ...
.. ...
...
.. M ......
.
.. ... ... X
.. .
.. ... . .
.. . . ..
.
..
Y P. .
.. N ..
.. ..
.. ..
... .
Z . . . . . ..
. . ..
.
D C
Note also that AY bisects ∠DAC, CZ bisects ∠DCA, AX bisects ∠BAC and CX
bisects ∠BCA. Since AD is parallel to BC, ∠DAC = ∠BCA. Therefore, ∠CAY =
∠ACX, which implies that ∠M AY = ∠P CX. Since ∆AY M and ∆CXP are both
right-angled triangles, ∆AY M ∼ ∆CXP . Similarly, ∆CZN ∼ ∆AXP . Therefore,
AM CP CN AP
= , and = .
MY PX NZ PX
Note that M Y = r1 and N Z = r2 . This yields
AM CP CN AP
= , and = .
r1 PX r2 PX
Dividing the second equation by the first equation yields
AP AM r1
= · .
PC CN r2
Therefore, to solve the problem, it suffices to show that AM = CN .
A B
.. . .
.. . . . . . .
.. ...
..
.. .M ......
.. ... ... X
R ..
.. ... . . . ..
. . . . ..
Y P
. .. ..
. N ..
.
. .. ..
... .
S .
Z . . . . . . ..
. . ..
D C
25
COMC 2012 Official Solutions 26

Let the circle inside ∆DAP touch AD, DP at R, S, respectively. Then note that
AR = AM, DR = DS and P M = P S. Therefore, DA + AP = DR + RA + AM + M P =
DS + AM + AM + SP = 2AM + DP . Similarly, DC + CP = 2CN + DP . By part
(a), DA + AP = DC + CP . Therefore, 2AM + DP = 2CN + DP , from which we can
conclude that AM = CN . This solves the problem. 

(c) (4 marks) Suppose DA + DC = 3AC and DA = DP . Let r1 , r2 be the two radii defined
in (b). Determine the ratio r1 /r2 .

Solution: The answer is r1 /r2 = 4/3.

Solution 1: By part (b). r1 /r2 = AP/P C. Let x = AP and y = P C. The answer is

the ratio x/y.

By part (a), DA + AP = DC + CP . Let s = DA + AP = DC + CP . Then DA = s − x

and DC = s − y. Since DA + DC = 3AC, (s − x) + (s − y) = 3(x + y). Hence,
2s = 4(x + y). Therefore, s = 2(x + y). Therefore, DA = x + 2y and DC = 2x + y.
Since DP = DA, DP = x + 2y.
A

D C

Drop the perpendicular from D to AC and let the perpendicular intersect AC at M .

Since DA = DP , M is the midpoint of AP . Therefore, M P = x/2. By the Pythagorean
Theorem, we have M D2 + M C 2 = DC 2 and M D2 + M P 2 = DP 2 . Therefore, DC 2 −
M C 2 = DP 2 − M P 2 . Therefore,

(2x + y)2 − (x/2 + y)2 = (x + 2y)2 − (x/2)2 .

Simplifying this yields

x2 x2
4x2 + 4xy + y 2 − − xy − y 2 = x2 + 4xy + 4y 2 − .
4 4

26
COMC 2012 Official Solutions 27

Hence, 3x2 − xy − 4y 2 = 0. Factoring this yields (3x − 4y)(x + y) = 0. Since x, y are

lengths, x + y 6= 0. Therefore, 3x − 4y = 0. Therefore, x/y = 4/3. 

Solution 2: We define x, y as in Solution 1. Then we have DA = x+2y and DC = 2x+y

and DP = x + 2y. Consider triangles ∆ADP and ∆CDP . Then

P A2 + P D2 − AD2 x2 + (x + 2y)2 − (x + 2y)2 x2 x

cos ∠AP D = = = =
2 · PA · PD 2 · x · (x + 2y) 2x(x + 2y) 2(x + 2y)

and
P C 2 + P D2 − CD2 y 2 + (x + 2y)2 − (2x + y)2 −3x2 + 4y 2
cos ∠CP D = = = .
2 · PC · PD 2 · y · (x + 2y) 2y(x + 2y)

Since ∠AP D and ∠CP D sum to 180◦ , their cosine values are negatives of each other.
Hence,
−x −3x2 + 4y 2 −3x2 + 4y 2
= ⇒ −x = .
2(x + 2y) 2y(x + 2y) y
This simplifies to 3x2 − xy − 4y 2 = 0. Factoring this yields (3x − 4y)(x + y) = 0. As in
Solution 1, we get x/y = 4/3. 

Solution 3: We define x, y as in Solution 1. Then we have DA = x+2y and DC = 2x+y

and DP = x + 2y. We now determine cos ∠DAP using cosine law in both ∆DAP and
∆DAC.
AD2 + AP 2 − DP 2
cos ∠DAP =
2 · AD · AP
(x + 2y)2 + x2 − (x + 2y)2 x2 x
= = =
2 · (x + 2y) · x 2x(x + 2y) 2(x + 2y)
and
AD2 + AC 2 − DC 2
cos ∠DAC =
2 · AD · AC
(x + 2y)2 + (x + y)2 − (2x + y)2 −2x2 + 2xy + 4y 2 −(x − 2y)(x + y) −x + 2y
= = = = .
2 · (x + 2y)(x + y) 2(x + 2y)(x + y) (x + 2y)(x + y) x + 2y
Therefore,
x −x + 2y
= .
2(x + 2y) x + 2y
Hence, x = 2(−x + 2y). This simplifies to x/y = 4/3. 

27
COMC 2012 Official Solutions 28

C4 For any positive integer n, an n-tuple of positive integers (x1 , x2 , · · · , xn ) is said to be super-
squared if it satisfies both of the following properties:

(1) x1 > x2 > x3 > · · · > xn .

(2) The sum x21 + x22 + · · · + x2k is a perfect square for each 1 ≤ k ≤ n.

For example, (12, 9, 8) is super-squared, since 12 > 9 > 8, and each of 122 , 122 + 92 , and
122 + 92 + 82 are perfect squares.

(a) (2 marks) Determine all values of t such that (32, t, 9) is super-squared.

Solution: The only answer is t = 24.

Note that 322 + t2 = 1024 + t2 and 322 + t2 + 92 = 1105 + t2 are perfect squares. Then
there exist positive integers a, b such that

1024 + t2 = a2
1105 + t2 = b2 .

Subtracting the first equation from the second equation gives

b2 − a2 = 81 ⇒ (b − a)(b + a) = 81.

The only ways 81 can be written as the product of two distinct positive integers is
81 = 1 × 81 and 81 = 3 × 27.

If (b − a, b + a) = (1, 81), then b − a = 1 and b + a = 81. Summing these two equations

yield 2b = 82. Therefore, b = 41. Hence, a = 40. Therefore, t2 = a2 − 322 = 402 − 322 =
82 (52 − 42 ) = 82 · 32 . Hence, t = 24.

We now verify that (32, 24, 9) is indeed super-squared. Clearly, the tuple is strictly de-
creasing, i.e. satisfies condition (1). Finally, 322 + 242 = 82 (42 + 32 ) = 82 · 52 = 402 and
322 + 242 + 92 = 402 + 92 = 1681 = 412 . Therefore, the tuple also satisfies condition (2).

If (b−a, b+a) = (3, 27), then b−a = 3 and b+a = 27. Summing these two equations gives
2b = 30. Therefore, b = 15. Hence, a = 12. Therefore, t2 = a2 − 322 = 122 − 322 < 0.
Hence, there are no solutions for t in this case.

Therefore, t = 24 is the only solution.

(b) (2 marks) Determine a super-squared 4-tuple (x1 , x2 , x3 , x4 ) with x1 < 200.

28
COMC 2012 Official Solutions 29

Solution: Note that if (x1 , · · · , xn ) is super-squared, then (ax1 , · · · , axn ) is also super-
squared for any positive integer a. We will show that this tuple satisfies both (1)
and (2) to show that it is indeed super-squared. Clearly, since x1 > x2 > · · · > xn ,
ax1 > ax2 > · · · axn . Since x21 +x22 +· · ·+x2k is a perfect square, x21 +x22 +· · ·+x2k = m2 for
some positive integer m. Therefore, (ax1 )2 +· · ·+(axk )2 = (am)2 . Hence, (ax1 , · · · , axn )
is super-squared.

From the example in the problem statement, (12, 9, 8) is super-squared. Therefore,

12(12, 9, 8) = (144, 108, 96) is also super-squared. Note that 122 + 92 + 82 = 172 . Hence,
1442 + 1082 + 962 = 2042 = 122 · 172 .

Note that 132 · 172 = (122 + 52 ) · 172 = 122 · 172 + 52 · 172 = 122 · 172 + 852 . Therefore,
2212 = 132 × 172 = 1442 + 1082 + 962 + 852 . And so we conclude that (144, 108, 96, 85)
is super-squared.

Comment: The list of all super-squared 4-tuples (x1 , x2 , x3 , x4 ) with x1 < 200 is

(132, 99, 88, 84), (144, 108, 75, 28), (144, 108, 96, 85), (156, 117, 104, 60), (180, 96, 85, 60),

(180, 135, 120, 32), and (192, 144, 100, 69).

(c) (6 marks) Determine whether there exists a super-squared 2012-tuple.

Solution: There does indeed exist a super-squared 2012-tuple.

We will show that there exists a super-squared n-tuple for any positive integer n ≥ 3.
We will prove this by induction on n. In the problem statement and in part (b), we
showed that this statement holds for n = 3, 4.

Suppose there exists a super-squared k-tuple (x1 , x2 , · · · , xk ) for some positive integer
k ≥ 3. We will show from this k-tuple that there exists a super-squared (k + 1)-tuple.

Let a, b, c be a tuple of positive integers such that a2 + b2 = c2 . We will provide the

additional conditions on (a, b, c) shortly.

Let r be the positive integer such that x21 + x22 + · · · + x2k = r2 . As in part (b), we
note that if (x1 , · · · , xk ) is super-squared, then (ax1 , · · · , axk ) is also super-squared and
(ax1 )2 +· · ·+(axk )2 = (ar)2 . Then we claim that (ax1 , · · · , axk , br) satisfies property (2)
of super-squared. Clearly, (ax1 )2 + · · · + (axt )2 is a perfect square, since (ax1 , · · · , axk )
is super-squared, for all 1 ≤ t ≤ k. To prove the claim, it remains to show that
(ax1 )2 + · · · + (axk )2 + (br)2 is a perfect square. This is clear since this quantity is equal

29
COMC 2012 Official Solutions 30

to (ar)2 + (br)2 = r2 (a2 + b2 ) = (cr)2 . This proves the claim.

To make the tuple (ax1 , · · · , axk , br) super-squared, we require that axk > br, or equiv-
alently, a/b > r/xk . Note that r, xk are determined from the tuple (x1 , · · · , xk ). Hence,
it suffices to show that there exists a Pythagorean triple (a, b, c), with a2 + b2 = c2 such
that a/b > r/xk . In general, we need to show that a/b can be arbitrarily large.

Note that (a, b, c) = (m2 − 1, 2m, m2 + 1) is a Pythagorean triple for any positive integer
m. This is clear since (m2 −1)2 +(2m)2 = m4 −2m2 +1+4m2 = m4 +2m2 +1 = (m2 +1)2 .
In such a case,
a m2 − 1 m 1 m
= = − > − 1,
b 2m 2 2m 2
which can be made arbitrarily large. This completes the induction proof. 

30
COMC 2013 http://comc.math.ca/ Official Solutions

A1 Determine the positive integer n that satisfies the following equation:

1 1 1 n
+ + = 10 .
210 29 28 2

Solution
Adding the left hand side of the given equation with with a common denominator of 210 , we
have,
1 1 1 1 2 22 1+2+4 7
10
+ 9
+ 8
= 10
+ 10
+ 10
= 10
= 10 .
2 2 2 2 2 2 2 2
Therefore, n = 7.

c 2013, 2014 Canadian Mathematical Society Page 1
Official Solutions http://comc.math.ca/ COMC 2013

A2 Determine the positive integer k for which the parabola y = x2 − 6 passes through the point
(k, k).
Solution
If the curve passes through the point (k, k), then we can substitute x = k, y = k into the
given equation to get k 2 − k − 6 = 0. We can factor this as (k − 3)(k + 2) = 0, so k = 3 or
k = −2. Since we want the positive value of k, we get k = 3.

Page 2
c 2013, 2014 Canadian Mathematical Society
COMC 2013 http://comc.math.ca/ Official Solutions

A3 In the figure below, the circles have radii 1, 2, 3, 4, and 5. The total area that is contained
inside an odd number of these circles is mπ for a positive number m. What is the value of m?

Solution
A point is inside an odd number of circles if it is in the outermost ring, the third ring, or
the middle circle. The area of the middle circle is π. The third ring is the area contained
in the circle of radius 3 but not contained in the circle of radius 2. The area of the third
ring is 32 π − 22 π = 5π. The outer ring is the area contained in the circle of radius 5 but not
contained in the circle of radius 4. The area of the fifth ring is 52 π − 42 π = 9π. Thus, the
total area is π + 5π + 9π = 15π, so m = 15.

c 2013, 2014 Canadian Mathematical Society Page 3
Official Solutions http://comc.math.ca/ COMC 2013

A4 A positive integer is said to be bi-digital if it uses two different digits, with each digit used
exactly twice. For example, 1331 is bi-digital, whereas 1113, 1111, 1333, and 303 are not.
Determine the exact value of the integer b, the number of bi-digital positive integers.
Solution 1
There are 9 choices for what the left-most digit of the number is (it cannot be 0) and there
are 3 choices for where the second copy of this digit is. There are 9 possibilities for the digit
that fills the remaining positions. Thus, b = 9 × 3 × 9 = 243.
Solution 2
We consider two
cases. Either 0 is one of the digits, or it is not. If 0 is not one of the digits,
then we have 92 = 36 ways to choose 2 digits which are not 0. There are (2!) 4!
2 = 6 ways

to arrange these digits, for a total of 216 numbers. If 0 is one of the digits, it cannot be
the first digit of the

number, since then the number would have fewer than 4 digits. In this
9
case, there are 1 = 9 ways to choose the other digit. The first digit must be the non-zero
digit and there are 3 places for the other non-zero digit, so there are 27 such numbers. Thus,
b = 216 + 27 = 243.

Page 4
c 2013, 2014 Canadian Mathematical Society
COMC 2013 http://comc.math.ca/ Official Solutions

B1 Given a triangle ABC, X, Y are points on side AB, with X closer to A than Y , and Z is
a point on side AC such that XZ is parallel to Y C and Y Z is parallel to BC. Suppose
AX = 16 and XY = 12. Determine the length of Y B.
A
@
 @
16 @
 @
 @
X @ Z
@
   @
12  
 
@
  
   @
 @

Y  C
@

 
 

 
  
 
  

 

B


Solution
Triangles AXZ and AY C are similar, so AZ : AX = ZC : XY and so AZ/ZC = 4/3. Also,
triangles AY Z and ABC are similar, so AZ : ZC = 28 : Y B. Combining the two results gives
4/3 = Y28B so Y B = 21.

c 2013, 2014 Canadian Mathematical Society Page 5
Official Solutions http://comc.math.ca/ COMC 2013

B2 There is a unique triplet of positive integers (a, b, c) such that a ≤ b ≤ c and

25 1 1 1
= + + .
84 a ab abc
Determine a + b + c.
Solution:
1 25
Note that 4 < 84 < 13 . Therefore, a ≥ 4. But if a ≥ 5, then b, c ≥ 5. Consequently,

25 1 1 1 1 1 1 52 + 5 + 1 31 1 25
= + + ≤ + 2+ 3 = = < < ,
84 a ab abc 5 5 5 53 125 4 84
which is a contradiction. Therefore, a 6≥ 5. Hence, a = 4.

Substituting this into the equation given in the problem yields

25 1 1 1
= + + .
84 4 4b 4bc
Multiplying both sides by 4 and rearranging yields
4 1 1
= + . (1)
21 b bc
1 4
Note that 6 < 21 < 15 . Therefore, b ≥ 6. If b ≥ 7, then c ≥ 7. Hence,

4 1 1 1 1 7+1 8 1 4
= + ≤ + 2 = 2
= < < ,
21 b bc 7 7 7 49 6 21
which is a contradiction. Therefore, b 6≥ 7. Consequently, b = 6. Substituting this into (1)
yields
4 1 1
= + .
21 6 6c
Multiplying both sides by 6 and rearranging yields
1 1
= .
7 c
Therefore, c = 7.

Hence, (a, b, c) = (4, 6, 7), which yields a + b + c = 17.

Page 6
c 2013, 2014 Canadian Mathematical Society
COMC 2013 http://comc.math.ca/ Official Solutions

B3 Teams A and B are playing soccer until someone scores 29 goals. Throughout the game
the score is shown on a board displaying two numbers – the number of goals scored by A
and the number of goals scored by B. A mathematical soccer fan noticed that several times
throughout the game, the sum of all the digits displayed on the board was 10. (For example,
a score of 12 : 7 is one such possible occasion). What is the maximum number of times
throughout the game that this could happen?
Solution 1
When the sum of all the digits on the scoreboard is 10, the sum of the scores must be 1 more
than a multiple of 9. The highest possible sum of the scores is 29 + 28 = 57. The numbers
less than 57 that are 1 more than a multiple of 9 are 1, 10, 19, 28, 37, 46, and 55. If the sum
of the scores is 1, then the sum of the digits is 1, not 10. If the sum of the scores is 55, then
the scores are 26 and 29 or 27 and 28, both of which have a digit sum of 19. Thus, we cannot
have this happen more than 5 times.
We see that the scores (5, 5), (5, 14), (14, 14), (23, 14), (23, 23) each have a digit sum of 10, and
can all be acheived in the same game. Thus, the maximum number of times is 5.
Solution 2
Denote by (a1 a2 , b1 b2 ) the score displayed on the board where a1 , a2 , b1 , b2 are digits (we allow
a1 and b1 to be 0), and a1 a2 , b1 b2 are the numbers of goals scored by the two teams. We will
call a score good if a1 + a2 + b1 + b2 = 10.
Lemma: Suppose scores (x, y) and (z, t) occurred throughout the game. Then at most one of
x > z and y < t can hold.
Proof: Suppose that x > z. Then the first team scored x goals after it scored z goals, so the
score (x, y) occurred later in the game than the score (z, t). Therefore y ≥ t, and the result
follows.
We now show that we cannot have two good scores occurring throughout the game of the
form (a1 a2 , a1 b2 ) and (a1 a02 , a1 b02 ). Suppose the scores did occur; then a2 + b2 = a02 + b02 .
WLOG a2 > a02 . Then b2 < b02 ; hence a1 a2 > a1 a02 ; a1 b2 < a1 b02 , which is impossible by the
Lemma.
We next claim that if a1 > b1 , then at most one of the good scores (a1 a2 , b1 b2 ), (b1 a02 , a1 b02 )
could occur throughout the game. This follows immediately from the Lemma since a1 a2 >
b1 a02 ; a1 b02 > b1 b2 .
Since the game ends when someone scores 29 goals, the tens digit for each team is 0, 1, or
2. By the first claim have at most nine possibilities for the good scores: (0a, 0b), (0a, 1b),
(0a, 2b), (1a, 0b), (1a, 1b), (1a, 2b), (2a, 0b), (2a, 1b), (2a, 2b) for some digits a, b (possibly
different for each case). By the second claim, at most one of (0a, 1b) and (1a, 0b); (0a, 2b)
and (2a, 0b); (1a, 2b) and (2a, 1b) can occur, eliminating three possibilities. Furthermore, if
(0a, 2b) or (2a, 0b) occurred then (1a, 1b) could not occur and vice versa (since if WLOG
(0a, 2b) occurred, then the second team had at least 20 points by the time the first team got
to 10 points). This eliminates one more possibility.

c 2013, 2014 Canadian Mathematical Society Page 7
Official Solutions http://comc.math.ca/ COMC 2013

Hence at most 9 - 3 - 1 = 5 good scores occurred. It remains to give an example when this
occurrence is indeed possible. One such example is (3, 7), (8, 11), (14, 14), (16, 21), (23, 23).

Page 8
c 2013, 2014 Canadian Mathematical Society
COMC 2013 http://comc.math.ca/ Official Solutions

B4 Let a be the largest real value of x for which x3 − 8x2 − 2x + 3 = 0. Determine the integer
closest to a2 .
Solution 1
Since the equation has degree 3, there are at most 3 values of x for which it will hold.
Let f (x) = x3 − 8x2 − x + 3, and b, c the other two roots of f (x).
Note that
f (−1) = (−1)3 − 8(−1)2 − 2(−1) + 3 = −4 < 0
and    3  2
−1 −1 −1 −1 −1
f = −8 −2· +3= − 2 + 1 + 3 > 0.
2 2 2 2 8
Hence, there is a root between −1 and −1/2.
Similarly,
   3  2
1 1 1 1 1 1
f = −8 −2· +3= −2−1+3= >0
2 2 2 2 8 8
and
f (1) = 1 − 8 − 2 + 3 = −6 < 0.
Hence, there is a root between 1/2 and 1.
Hence, suppose −1 < b < −1/2 and 1/2 < c < 1.
Consider the quantity a2 +b2 +c2 . By the factor theorem, x3 −8x2 −2x+3 = (x−a)(x−b)(x−
c) = x3 − (a + b + c)x2 + (ab + bc + ca)x − abc. Therefore, a + b + c = 8 and ab + bc + ca = −2.
Then a2 + b2 + c2 = (a + b + c)2 − 2(ab + bc + ca) = 82 − 2 · (−2) = 68.
Now, we consider the quantity b2 + c2 . Since b < −1/2 and c > 1/2, b2 + c2 > 1/2. Now we
need an upper bound on b2 + c2 . Note that
1 3 1 2
       
1 1 1 2 −3
f √ = √ −8 √ −2 √ + 3 = √ − 4 − √ + 3 = √ − 1 < 0.
2 2 2 2 2 2 2 2 2
√
Since f (1/2) > 0, 1/2 < c < 1/ 2.
Therefore, b2 + c2 < 1 + 1/2 = 3/2. Since 1/2 < b2 + c2 < 3/2 and a2 + b2 + c2 = 68,
66.5 < a2 < 67.5. Therefore, the integer closest to a2 is 67.
Solution 2
As in solution 1, we can verify that there are two values of x between −1 and 1 for which the
equation holds. Note that since the equation is cubic there are at most 3 distinct solutions.
13
We can rewrite the equation as x2 (x − 8) = 2x − 3, which simplifies to x2 = 2 + x−8 . Letting
2 13
x = 8.2 we get the left hand side is 8.2 = 67.24 and the right side is 2 + .2 = 67. As we
decrease x, from 8.2 to 8.1, the left hand side decreases from 67.24 to 65.61 and the right hand
side increases from 67 to 132. Since both functions are continuous, there is a point between
where they will have the same value, and that value will be between 67 and 67.24. Thus, the
integer closest to x2 is 67.

c 2013, 2014 Canadian Mathematical Society Page 9
Official Solutions http://comc.math.ca/ COMC 2013

C1 In the diagram, 4AOB is a triangle with coordinates O = (0, 0), A = (0, 30), and B = (40, 0).
Let C be the point on AB for which OC is perpendicular to AB.

A(0, 30)
Z
Z
Z
Z
Z C
Z
ZZ
 Z
 Z
 Z
Z
 Z
 Z
 Z
Z
 Z
O(0, 0) B(40, 0)
(a) Determine the length of OC.
(b) Determine the coordinates of point C.
(c) Let M be the centre of the circle passing through O, A, and B. Determine the length
of CM .

Solution 1
√
(a) By the Pythagorean Theorem, the length of AB is 302 + 402 = 50. By calculating the
area of the triangle as AB × CO/2 and AO × OB/2 we get that 50 × OC = 1200, and
OC = 24.
(b) Since OC is perpendicular to AB, angle ACO is a right angle. Thus, triangle ACO is
similar to triangle AOB, so AC : AO = AO : AB and AC = 18. So point C
is 1850 of the
18
× 40, 32 72 96


way along the line from A to B. Thus, the coordinates are 50 50 × 30 = 5 , 5
(c) Since the angle AOB is a right angle, AB is a diameter of the circle through O, A, and
B. Thus, M must be the midpoint of the line AB. We already calculated that AC = 18,
and we know that AM = AB/2 = 25, so CM = AM − AC = 25 − 18 = 7.

Solution 2
√
(a) By the Pythagorean Theorem, the length of AB is 302 + 402 = 50. Since OC is per-
AO AB
pendicular to AB, angle ACO is a right angle and thus OC = OB , so OC = AO×OB
AB =
1200
50 = 24.
x
(b) The equation of the line through A and B has the form y−30 = 40−0
0−30 , which we can
3
rewrite as y = − 4 x + 30. The equation of the line through O and C is perpendicular to
y = − 43 x + 30, 4 4

so it has slope 3 and the equation is y = 3 x. These lines intersect at the
72 96
point 5 , 5 , which are the coordinates of C.

Page 10
c 2013, 2014 Canadian Mathematical Society
COMC 2013 http://comc.math.ca/ Official Solutions

(c) Let (x, y) be the coordinates of M. Since M is the centre of a circle containing the
points A, B, O we have M A = M M O = M B. This gives x2 + (y − 30)2 = x2 + y 2 =
(x−40)2 +y 2 . The first equality gives y = 15 and the second equality
q gives x = 20, so M =

2
2
(20, 15). By the Pythagorean theorem, the length of M C is 20 − 725 + 15 − 965 =
√ 2 2
28 +(−21)
5 = 35
5 = 7.

c 2013, 2014 Canadian Mathematical Society Page 11
Official Solutions http://comc.math.ca/ COMC 2013

C2 (a) Determine all real solutions to a2 + 10 = a + 102 .

(b) Determine two positive real numbers a, b > 0 such that a 6= b and a2 + b = b2 + a.
(c) Find all triples of real numbers (a, b, c) such that a2 + b2 + c = b2 + c2 + a = c2 + a2 + b.
Solution
We can rearrange the equation as follows:

a2 − b2 = a − b
(a − b)(a + b) = (a − b)
(a − b)(a + b − 1) = 0

This tells us that our two solutions are a = b and a = 1 − b.

(a) By the above result, the solutions are a = 10, a = −9.

(b) By the above result, the pair a = 14 and b = 34 is such a pair of positive real numbers.
Any pair of positive real numbers a, b with a + b = 1 will suffice.
(c) Applying the above result to the first two parts of the equality gives a = c or a = 1 − c.
Applying it to the first and third gives b = c or b = 1 − c. Applying to the second and
third gives a = b or b = 1 − a.
Fix any real number a. Then b = a or b = 1 − a and c = a or c = 1 − a. Note any
pair (b, c) formed satisfies b = c or b = 1 − c. Hence, all four solutions (a, a, a), (a, a, 1 −
a), (a, 1 − a, a), (a, 1 − a, 1 − a) are solutions to the given equation.

Page 12
c 2013, 2014 Canadian Mathematical Society
COMC 2013 http://comc.math.ca/ Official Solutions

C3 Alphonse and Beryl play the following game. Two positive integers m and n are written on
the board. On each turn, a player selects one of the numbers on the board, erases it, and
writes in its place any positive divisor of this number as long as it is different from any of
the numbers previously written on the board. For example, if 10 and 17 are written on the
board, a player can erase 10 and write 2 in its place (as long as 2 has not appeared on the
board before). The player who cannot make a move loses. Alphonse goes first.

(a) Suppose m = 240 and n = 351 . Determine which player is always able to win the game
and explain the winning strategy.
(b) Suppose m = 240 and n = 251 . Determine which player is always able to win the game
and explain the winning strategy.

Solution

(a) Notice that for (a) m and n have greatest common divisor equal to 1, therefore on each
turn a player can always make a move of replacing the number k with its divisor l strictly
less than k, as long as l > 1, or as long as l = 1 and 1 has not yet appeared on the board.
Instead of dealing with the actual numbers we will deal with the number of prime factors
they have. Then, the game becomes equivalent to the following. Two numbers m and
n are written on the board. On each turn a player can select a number k greater than
0 and replace it with any positive integer less than k, or replace it with 0, as long as 0
is not already written on the board. A player who cannot make a move loses.
It immediately follows that m = 0, n = 1 is a losing position. Therefore, m = 0, n ≥ 2 is a
winning position (since a player replaces n with 1 and wins). Furthermore, m = 1, n ≥ 1
is a winning position (since a player replaces n with 0 and wins). Hence m = 2, n = 2 is
a losing position; m = 2, n ≥ 3 is a winning position; m = 3, n = 3 is a losing position,
m = 3, n ≥ 4 is a winning position. By induction it follows that for k ≥ 2, m = k, n = k
is a losing position, while m = k, n ≥ k + 1 is a winning position.
We are in the case of m = 40, n = 51 ≥ 41 in the “transformed” game, thus this is a
winning position and Alphonse wins.
(b) This case is different, since now m and n have more than one divisor in common. We
will deal with the original game and not make any transformations. Note that m and n
are both powers of 2, so throughout the whole game only powers of 2 can appear on the
board.
We first note that the player who first writes down a number less than or equal to 2 loses.
This is because if they write down 1, then 2 has not yet been written; the opponent on
the next turn replaces the other number with 2 wins. (Note that this move is legal since
at the start m > 2, n > 2 so at the time that 1 is written, the other number on the
board must be greater than 2). If they write down 2, then 1 has not yet been written;
the opponent on the next turn replaces the other number with 1 and wins.
Similarly, the player who first writes down a number less than or equal to 8 loses. This is
because if they write down 4, the other player writes 8 – thus forcing the original player

c 2013, 2014 Canadian Mathematical Society Page 13
Official Solutions http://comc.math.ca/ COMC 2013

to write down a number less than or equal to 2 (note they cannot replace 8 with 4 since
4 has already appeared on the board). Similarly, if they write down 8, the other player
writes down 4 and wins.
By induction it follows that if m, n > 22k−1 then the player who first writes down a
number less than or equal to 22k−1 loses for every positive integer k. Thus for the case
m = 240 , n = 253 , the player to first write down a number less than or equal to 239 loses.
Therefore on his first turn, Alphonse replaces 253 with 241 and wins – because on her
turn, Beryl is faced with 240 and 241 on the board and has to write down a number less
than or equal to 239 .

Page 14
c 2013, 2014 Canadian Mathematical Society
COMC 2013 http://comc.math.ca/ Official Solutions

C4 For each real number x, let [x] be the largest integer less than or equal to x. For example,
[5] = 5, [7.9] = 7 and [−2.4] = −3. An arithmetic progression of length k is a sequence
a1 , a2 , . . . , ak with the property that there exists a real number b such that ai+1 − ai = b for
each 1 ≤ i ≤ k − 1.
Let α > 2 be a given irrational number. Then S = {[n · α] : n ∈ Z}, is the set of all integers
that are equal to [n · α] for some integer n.

(a) Prove that for any integer m ≥ 3, there exist m distinct numbers contained in S which
form an arithmetic progression of length m.
(b) Prove that there exist no infinite arithmetic progressions contained in S.

Solution

(a) We first prove the following statement: For each positive integer m there exist positive
1
integers n ≤ m and xm such that |nα − xm | < m .
We consider the fractional parts of the numbers nα for n = 0, . . . , m, i.e., consider
{nα} := nα − [nα]. By the definition of the integer part of a real number we conclude
that each {nα} ∈ [0, 1).
Using the pigeonhole principle we conclude that there must exist two distinct integers 0 ≤
n1 < n2 ≤ m such that both the corresponding fractional parts {n1 α} and {n2 α} belong
to the same interval [(k − 1)/m, k/m), for some k = 1, . . . , m. Hence |{n2 α} − {n1 α}| <
1
m.
1
Thus |n2 α − [n2 α] − n1 α + [n1 α]| < m , and therefore letting n := n2 − n1 and xm :=
1
[n2 α] − [n1 α], we conclude that |nα − xm | < m .
Furthermore, since 0 ≤ n1 < n2 ≤ m, we get that n ≤ m is a positive integer. Also,
using that α > 2 while n2 > n1 we conclude that xm = [n2 α] − [n1 α] ≥ [α] ≥ 2 is also a
positive integer.
As proved above, for each integer m ≥ 3, there exist positive integers n ≤ m and xm
1
such that |nα − xm | < m . At the expense of replacing n by −n and replacing xm by
1
−xm , we may assume that 0 < {nα} < m , and thus 0 < nα − xm < 1/m.
Then xm = [nα] and so, nα = xm + {nα}. We deduce that for each k ∈ {1, 2, · · · , m}
we have kxm < nkα = kxm + k{nα} < kxm + 1. So, [nkα] = kxm , which proves that
indeed the numbers [nα], [2nα], · · · , [mnα] form an arithmetic progression.
(b) Assume there exists an infinite arithmetic progression in S: [n1 α], [n2 α], · · · , [ni α], · · · .
For each i ∈ N, using the fact that [ni α] + [ni+2 α] = 2[ni+1 α], we conclude that (ni+2 −
2ni+1 + ni ) · α = {ni+2 α} − 2{ni+1 α} + {ni α} ∈ (−2, 2), where in the last inequality we
used the fact that the fractional part of any real number is in the interval [0, 1).
However, since each ni ∈ Z and moreover, α > 2 we conclude that for each i ∈ N we
have ni+2 − 2ni+1 + ni = 0. So, n1 α, n2 α, · · · , ni α, · · · is itself an arithmetic progression.
Therefore, the difference of the two arithmetic progressions is another infinite arithmetic
progression: {n1 α}, {n2 α}, · · · , {ni α}, · · · .

c 2013, 2014 Canadian Mathematical Society Page 15
Official Solutions http://comc.math.ca/ COMC 2013

However, the arithmetic progressions cannot be bounded, unless their ratio is 0. Hence
{n2 α} = {n1 α}, which yields that n2 α − n1 α = [n2 α] − [n1 α] ∈ Z and therefore α ∈ Q,
which is a contradiction with our assumption (also note that n2 6= n1 since they belong
to an infinite arithmetic progression).

Page 16
c 2013, 2014 Canadian Mathematical Society
 The Sun Life Financial
Canadian Open Mathematics Challenge
November 6/7, 2014

STUDENT INSTRUCTION SHEET

General Instructions

1) Do not open the exam booklet until instructed to do so by your supervising

teacher.
2) The supervisor will give you five minutes before the exam starts to fill in the
identification section on the exam cover sheet. You don’t need to rush. Be sure
to fill in all information fields and print legibly. Mobile phones and
calculators are NOT
3) Once you have completed the exam and given it to your supervising teacher
permitted.
you may leave the exam room.
4) The contents of the COMC 2014 exam and your answers and solutions must not be publicly discussed
(including online) for at least 24 hours.

Exam Format

You have 2 hours and 30 minutes to complete the COMC. There are three sections to the exam:

PART A: Four introductory questions worth 4 marks each. Partial marks may be awarded for work shown.

PART B: Four more challenging questions worth 6 marks each. Partial marks may be awarded for work
shown.

PART C: Four long-form proof problems worth 10 marks each. Complete work must be shown. Partial marks
may be awarded.

Diagrams are not drawn to scale; they are intended as aids only.

Work and Answers

All solution work and answers are to be presented in this booklet in the boxes provided – do not include
additional sheets. Marks are awarded for completeness and clarity. For sections A and B, it is not necessary to
show your work in order to receive full marks. However, if your answer or solution is incorrect, any work that
you do and present in this booklet will be considered for partial marks. For section C, you must show your
work and provide the correct answer or solution to receive full marks.

It is expected that all calculations and answers will be expressed as exact numbers such as 4π, 2 + √7, etc.,
rather than as 12.566, 4.646, etc. The names of all award winners will be published on the Canadian
Mathematical Society web site https://cms.math.ca/comc.
 The 2014 Sun Life Financial
Canadian Open Mathematics Challenge

Please print clearly and complete all information below. Failure to print legibly or provide
complete information may result in your exam being disqualified. This exam is not considered
valid unless it is accompanied by your test supervisor’s signed form.
First Name: Grade:
8 9 10
11 12 Cégep
Last Name: Other:

T-Shirt Size:
(Optional. For prize draw)
Are you currently registered in full-time attendance at an elementary, secondary or XS S M
Cégep school, or home schooled and have been since September 15th of this year? L XL XXL
Yes No Date of Birth:

Are you a Canadian Citizen or a Permanent Resident of Canada (regardless of current y y y y m m d d

Gender: (Optional)
address)?
Yes No
Male Female

E-mail Address:

Signature: ________________________________

For official use only:

A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 C3 C4 TOTAL

Marker initials Data entry initials Verification initials

 Page 2 of 16 SUN LIFE FINANCIAL CANADIAN OPEN MATHEMATICS CHALLENGE 2014

Section A
Part A: Question 1 (4 marks)
1. In triangle ABC, there is a point D on side BC such that BA = AD = DC. Suppose
∠BAD = 80◦ . Determine the size of ∠ACB.
A

 
  
  
  
  
  
  
  
  
B D C
Your Solution:
2. The equations x2 − a = 0 and 3x4 − 48 = 0 have the same real solutions. What is the value

of a?
3. A positive integer m has the property that when multiplied by 12, the result is a four-digit
Section
number n ofAthe form 20A2 for some digit A. What is the 4 digit number, n?
1.4. In
Alana, Beatrix,
triangle ABC,Celine,
there and
is a Deanna
point D played
on side6 BC
games of tennis
such that BAtogether.
= AD In = each
DC. game, the
Suppose
four of =
∠BAD them
80◦ .split into two
Determine theteams of ∠ACB.
size of two and one of the teams won the game. If Alana was
on the winning team for 5 games, A Beatrix for 2 games, and Celine for 1 game, for how many
games was Deanna on the winning  team?
 
  
   Your final answer:
  
Section B 
 

 

  
 passes through thepoints (1, 1), (1, 7),
1. The area of the circle that and (9, 1) can be expressed
  
Part A:What
as kπ. Question
is the2 (4
Bmarks)
value of k? D C
2.2. The equations
Determine x2 −a=
all integer values 3x4 − 48
0 andof n for = 0 nhave
which the+same
2 + 6n 24 is real solutions.
a perfect square.What is the value
of a?
3. 5 Xs and 4 Os are arranged in the below grid such that each number is covered by either an
3. Your Solution:
AX positive
or an O.integer m has
There are the of
a total property that when
126 different multiplied
ways that the Xsby and 12,Osthe
canresult is a four-digit
be placed. Of these
number n of the form 20A2 for some digit A. What is the 4
126 ways, how many of them contain a line of 3 Os and no line of 3 Xs?digit number, n?
4. Alana, Beatrix,
A line of Celine,
3 in a row can beand Deanna played
a horizontal line, a6vertical
games line,
of tennis
or onetogether. In each
of the diagonal game,
lines the
1−5−9
four
or 7 of
− 5them
− 3. split into two teams of two and one of the teams won the game. If Alana was
on the winning team for 5 games, Beatrix for 2 games, and Celine for 1 game, for how many
games1 was 2Deanna 3 on the winning team?

4 5 6
Section B
7 8 9
1. The area of the circle that passes through the points (1, 1), (1, 7), and (9, 1) can be expressed
as kπ. What is the value of k?
2. Determine all integer values of n for which 1n2 + 6n + 24 is a perfect square.
3. 5 Xs and 4 Os are arranged in the below grid such that each number is covered by either an
X or an O. There are a total of 126 different ways that the Xs and Os can be placed. Of these
126 ways, how many of them contain a line of 3 Os and no line of 3 Xs?
Your final answer:
A line of 3 in a row can be a horizontal line, a vertical line, or one of the diagonal lines 1−5−9
or 7 − 5 − 3.
cms.math.ca © 2014 CANADIAN MATHEMATICAL SOCIETY
1 2 3
   
  
  
  
  
  
 
SUN LIFE FINANCIAL CANADIAN
B OPEN MATHEMATICS CHALLENGE
D 2014 C Page 3 of 16
2. The equations x2
− a = 0 and 3x4 − 48 = 0 have the same real solutions. What is the value
of a?
Part A: Question 3 (4 marks)
3. A positive integer m has the property that when multiplied by 12, the result is a four-digit
number n of the form 20A2 for some digit A. What is the 4 digit number, n?
4. Your Solution:
Alana, Beatrix, Celine, and Deanna played 6 games of tennis together. In each game, the
four of them split into two teams of two and one of the teams won the game. If Alana was
on the winning team for 5 games, Beatrix for 2 games, and Celine for 1 game, for how many
games was Deanna on the winning team?

Section B
A
1. The
1. In triangle
area of ABC, there
the circle thatispasses
a point D on the
through sidepoints
BC such
(1, 1),that BAand
(1, 7), = (9,
AD 1) =
canDC. Suppose
be expressed
∠BAD
as
◦
= 80 is. Determine
kπ. What k? size of ∠ACB.
the value ofthe
A
2. Determine all integer values of n forwhich n2 + 6n + 24 is a perfect square.
 
  
3. 5 Xs and 4 Os are arranged in the  below grid such that each number is covered by either an
X or an O. There are a total of126 different
 ways that
 the Xs and Os can be placed. Of these
   no line of 3 Xs?
126 ways, how many of them  contain a lineof 3 Os and 
  of the diagonal lines 1−5−9
A line of 3 in a row can bea horizontal line, a vertical
 line, or one
  
or 7 − 5 − 3.
B D C
Your final answer:
2. The1 equations
2 x2 4
3 − a = 0 and 3x − 48 = 0 have the same real solutions. What is the value
of a?
4
3. A positive5 integer
6 m has the property that when multiplied by 12, the result is a four-digit
Part A: Question
number 4 (4 marks)
n of the form 20A2 for some digit A. What is the 4 digit number, n?
7
4. Alana, 8
Beatrix, 9
Celine, and Deanna played 6 games of tennis together. In each game, the
four of them split into two teams of two and one of the teams won the game. If Alana was
on the winning team for 5 games, Beatrix for 2 games, and Celine for 1 game, for how many
games was Deanna on the winning team? 1
Your Solution:

Section B
1. The area of the circle that passes through the points (1, 1), (1, 7), and (9, 1) can be expressed
as kπ. What is the value of k?
2. Determine all integer values of n for which n2 + 6n + 24 is a perfect square.
3. 5 Xs and 4 Os are arranged in the below grid such that each number is covered by either an
X or an O. There are a total of 126 different ways that the Xs and Os can be placed. Of these
126 ways, how many of them contain a line of 3 Os and no line of 3 Xs?
A line of 3 in a row can be a horizontal line, a vertical line, or one of the diagonal lines 1−5−9
or 7 − 5 − 3.

1 2 3
Your final answer:
4 5 6

© 2014 CANADIAN MATHEMATICAL SOCIETY cms.math.ca

7 8 9
 four of them split into two teams of two and one of the teams won the game. If Alana was
on the winning team for 5 games, Beatrix for 2 games, and Celine for 1 game, for how many
games was Deanna on the winning team?
Page 4 of 16 SUN LIFE FINANCIAL CANADIAN OPEN MATHEMATICS CHALLENGE 2014

Section B
Part B: Question 1 (6 marks)
1. The area of the circle that passes through the points (1, 1), (1, 7), and (9, 1) can be expressed
as kπ. What is the value of k?
Your Solution:
2. Determine all integer values of n for which n2 + 6n + 24 is a perfect square.
3. 5 Xs and 4 Os are arranged in the below grid such that each number is covered by either an
X or an O. There are a total of 126 different ways that the Xs and Os can be placed. Of these
126 ways, how many of them contain a line of 3 Os and no line of 3 Xs?
A line of 3 in a row can be a horizontal line, a vertical line, or one of the diagonal lines 1−5−9
or 7 − 5 − 3.

1 2 3

4 5 6

7 8 9

Your final answer:

cms.math.ca © 2014 CANADIAN MATHEMATICAL SOCIETY

 games was Deanna on the winning team?

Section BCANADIAN OPEN MATHEMATICS CHALLENGE 2014

SUN LIFE FINANCIAL Page 5 of 16

1. The area of the circle that passes through the points (1, 1), (1, 7), and (9, 1) can be expressed
Part
as kπ.B:What
Question
is the2 value
(6 marks)
of k?
2. Determine all integer values of n for which n2 + 6n + 24 is a perfect square.
3. 5Your
Xs Solution:
and 4 Os are arranged in the below grid such that each number is covered by either an
X or an O. There are a total of 126 different ways that the Xs and Os can be placed. Of these
126 ways, how many of them contain a line of 3 Os and no line of 3 Xs?
A line of 3 in a row can be a horizontal line, a vertical line, or one of the diagonal lines 1−5−9
or 7 − 5 − 3.

1 2 3

4 5 6

7 8 9

Your final answer:

© 2014 CANADIAN MATHEMATICAL SOCIETY cms.math.ca

Section B
1. The
Page 6 ofarea of the circle that passes through the points
16 (1, 1),
SUN LIFE (1, 7), and
FINANCIAL (9, 1) OPEN
CANADIAN can be expressed CHALLENGE 2014
MATHEMATICS
as kπ. What is the value of k?
2. Part
Determine all integer
B: Question values of n for which n2 + 6n + 24 is a perfect square.
3 (6 marks)
3. 5 Xs and 4 Os are arranged in the below grid such that each number is covered by either an
X or an O. There are a total of 126 different ways that the Xs and Os can be placed. Of these
126 ways, how many of them contain a line of 3 Os and no line of 3 Xs?
A line of 3 in a row can be a horizontal line, a vertical line, or one of the diagonal lines 1−5−9
or 7 − 5 − 3.

1 2 3

4 5 6

7 8 9

Your Solution: 1

Your final answer:

cms.math.ca © 2014 CANADIAN MATHEMATICAL SOCIETY

 SUN LIFE FINANCIAL CANADIAN OPEN MATHEMATICS CHALLENGE 2014 Page 7 of 16

Part B: Question 4 (6 marks)

1
4. Let f (x) = . Determine the smallest positive integer n such that
x3 + 3x2 + 2x
503
f (1) + f (2) + f (3) + · · · + f (n) > .
2014
Your solution:
Section C
1. A sequence of the form {t1 , t2 , ..., tn } is called geometric if t1 = a, t2 = ar, t3 = ar2 , . . . , tn =
arn−1 . For example, {1, 2, 4, 8, 16} and {1, −3, 9, −27} are both geometric sequences. In all
three questions below, suppose {t1 , t2 , t3 , t4 , t5 } is a geometric sequence.

(a) If t1 = 3 and t2 = 6, determine the value of t5 .

(b) If t2 = 2 and t4 = 8, determine all possible values of t5 .
(c) If t1 = 32 and t5 = 2, determine all possible values of t4 .

2. A local high school math club has 12 students in it. Each week, 6 of the students go on a
field trip.

(a) Jeffrey, a student in the math club, has been on a trip with each other student in the
math club. Determine the minimum number of trips that Jeffrey could have gone on.
(b) If each pair of students have been on at least one field trip together, determine the
minimum number of field trips that could have happened.

3. The line L given by 5y + (2m − 4)x − 10m = 0 in the xy-plane intersects the rectangle with
vertices O(0, 0), A(0, 6), B(10, 6), C(10, 0) at D on the line segment OA and E on the line
segment BC.

(a) Show that 1 ≤ m ≤ 3.

1
(b) Show that the area of quadrilateral ADEB is 3 the area of rectangle OABC.
(c) Determine, in terms of m, the equation of the line parallel to L that intersects OA at F
and BC at G so that the quadrilaterals ADEB, DEGF , F GCO all have the same area.

4. A polynomial f (x) with real coefficients is said to be a sum of squares if there are polynomials
p1 (x), p2 (x), . . . , pn (x) with real coefficients for which

f (x) = p21 (x) + p22 (x) + · · · + p2n (x)

For example, 2x4 + 6x2 − 4x + 5 is a sum of squares because

√
2x4 + 6x2 − 4x + 5 = (x2 )2 + (x2 + 1)2 + (2x − 1)2 + ( 3)2

(a) Determine all values of a for which f (x) = x2 + 4x + a is a sum of squares.

2
Your final answer:

© 2014 CANADIAN MATHEMATICAL SOCIETY cms.math.ca

 4. Let f (x) = . Determine the smallest positive integer n such that
x3 + 3x2 + 2x
503
f (1) + f (2) + f (3) + · · · + f (n) > .
2014
Page 8 of 16 SUN LIFE FINANCIAL CANADIAN OPEN MATHEMATICS CHALLENGE 2014

Section C
Section C 1 (10 marks)
Part C: Question
1. A sequence of the form {t1 , t2 , ..., tn } is called geometric if t1 = a, t2 = ar, t3 = ar2 , . . . , tn =
1. ar
A n−1
sequence
. For of
example, {t2,
the form{1, 1 , t4,
2 ,8, tn }and
...,16} is called
{1, −3, geometric
9, −27} ifare
t1 = a, tgeometric
both = ar2 , . . . ,Intnall
2 = ar, t3 sequences. =
n−1
ar
three .questions below,{1,
For example, 2, 4, 8, 16}
suppose {t1 , tand {1, −3, 9, −27} are both geometric sequences. In all
2 , t3 , t4 , t5 } is a geometric sequence.
three questions below, suppose {t1 , t2 , t3 , t4 , t5 } is a geometric sequence of real numbers.
(a) If t1 = 3 and t2 = 6, determine the value of t5 .
(a) If t = 3 and t = 6, determine the value of t5 .
(b) If t21 = 2 and t42 = 8, determine all possible values of t5 .
Solution: t1 = 3 = a (1 mark) and t2 = ar = 6 , so r = 6/3 = 2. 1 mark.
(c) If t1 = 32 and t5 = 2,4 determine all possible values of t4 .
This gives t5 = 3 × 2 = 48. 1 mark
Your
2. A(b) If solution:
local thigh
2 = 2school
and t4math = 8, determine
club has 12 allstudents
possible in values
it. Each of t5 .week, 6 of the students go on a
Solution: t2 = 2 = ar and t4 = 8 = ar , Dividing the two equations gives r2 = 4, so
field trip. 3

r = ±2. 1 mark
(a) Jeffrey,
When r a=student2 we have in the
a = math
1, so tclub, has
4 been on a trip with each other student in the
5 = 2 = 16. 1 mark.
math club. Determine the minimum number of4 trips that Jeffrey could have gone on.
When r = −2 we have a = −1, so t5 = −1 × 2 = −16. 1 mark.
(b) If each pair of students have been on at least one field trip together, determine the
(c) If t1 = 32 and t5 = 2, determine all possible values of t4 .
minimum number of field trips that could have happened. 1
Solution: We have t1 = 32 = a and t5 = 2 = ar4 . This gives a = 32, and r4 = 16 .1
mark
3. The line L given by 5y + (2m − 4)x − 10m = 0 in the xy-plane intersects the rectangle with
vertices
WhenO(0,r40),= A(0,1
16 we6),getB(10, 1
r2 =6), r2 =0)−1
C(10,
4 or 4 at
. D on the line segment OA and E on the line
segment BC. 2 −1
When r = 4 , r is not a real number, so this is not a valid sequence. 1 mark
When r2 = 1 we get r = ± 12 . 1 mark.
(a) Show that 14≤ m ≤ 3.1
This gives t4 = 32 × 8 = 4 and t4 = 32 × −1 8 = −4. 1 mark
(b) Show that the area of quadrilateral ADEB is 13 the area of rectangle OABC.
2. The
(c) line L given in
Determine, byterms
5y + of
(2mm,−the − 10m =of0 the
4)xequation in the
linexy-plane
parallel tointersects the rectangle
L that intersects with
OA at F
vertices O(0, 0), A(0, 6), B(10, 6), C(10, 0) at D on the line
and BC at G so that the quadrilaterals ADEB, DEGF , F GCO all have the same area.segment OA and E on the line
segment BC.
4. A polynomial f (x) with real coefficients is said to be a sum of squares if there are polynomials
p(a)
1 (x),Show . . . ,1pn≤(x)
p2 (x),that ≤ 3. real coefficients for which
m with
Solution: Since D is on OA, the2x-coordinate 2
of D is 0. 2
The y-coordinate D is then the
f (x) = p 1 (x) + p 2 (x) + · · · + p n (x)
solution to the equation 5y − 10m = 0, i.e. y = 2m. Hence L intersects OA at D(0, 2m).
For D to be on OA, 0 ≤ 2m ≤ 6, or equivalently 0 ≤ m ≤ 3. 1 mark
For example, 2x4 + 6x2 − 4x + 5 is a sum of squares because
Similarly, the x-coordinate of E is 10, so the y-coordinate is the solution √ to 5y + (2m −
4 0, whose
4)(10) − 10m 2x= + 6x2 − 4xsolutions
+ 5 = (xis2 )y2 =
+ (x8− 2 2m. 2Hence 0 ≤ 82 − 2m ≤2 6 or equivalently
+ 1) + (2x − 1) + ( 3)
1 ≤ m ≤ 4. 1 mark
(a) Determine 2
Thus 0 ≤ mall≤values 3 and of1≤ amfor≤which
4 so 1f (x)
≤ m=≤x 3.+14xmark + a is a sum of squares.

cms.math.ca © 2014 CANADIAN MATHEMATICAL SOCIETY

2. A (c) If high
local t1 = 32 and math
school t5 = 2,club
determine
has 12 all possible
students in values
it. Each of tweek,
4. 6 of the students go on a
1
Solution: We have t1 = 32 = a and t5 = 2 = ar . This gives a = 32, and r4 = 16
field trip. 4 .1
mark
(a) Jeffrey,
When ra4 student
= 161
weinget
the r2 math
= 14 orclub,
r2 =has
−1 been on a trip with each other student in the
4 .
math
Page 10 of club.
16
When 2 Determine
−1 the minimum number
r = 4 , r is not a real number, so this of istrips
SUN LIFE that Jeffrey
notFINANCIAL
a valid could 1
CANADIAN
sequence. have
OPEN gone on.
MATHEMATICS
mark CHALLENGE 2014
(b) IfWhen
each rpair
2 = of
1 students have 1 been on at least one field trip together, determine the
4 we get r = ± 2 . 1 mark.
minimum number 32of
×field
1 trips that could have
−1 happened.
This
Part C: gives t42=(10
Question 8 = 4 and t4 = 32 × 8 = −4. 1 mark
marks)
3.2. The
Theline
lineLLgiven
givenby
by5y (2m−−4)x
5y++(2m 4)x−−10m
10m==00ininthe
thexy-plane
xy-planeintersects
intersectsthe
therectangle
rectanglewith
with
vertices
vertices O(0, 0), A(0, 6), B(10, 6), C(10, 0) at D on the line segment OA and Eon
O(0, 0), A(0, 6), B(10, 6), C(10, 0) at D on the line segment OA and E onthe
theline
line
segment BC.
segment BC.
(a) Show that 1 ≤ m ≤ 3.
(a) Show that 1 ≤ m ≤ 3.
(b) Show that the
Solution: areaDofis quadrilateral
Since ADEB is 13 ofthe
on OA, the x-coordinate D area
is 0. of
Therectangle OABC.
y-coordinate D is then the
solution to in
(c) Determine, theterms
equation
of m,5ythe− 10m = 0, of
equation i.e.the
y= 2m.
line HencetoL Lintersects
parallel OA at OA
that intersects D(0,at
2m).
F
ForBC
and D to atbe G on OA, the
so that 0 ≤quadrilaterals
2m ≤ 6, or equivalently
ADEB, DEGF 0 ≤ m, F ≤GCO
3. 1 mark
all have the same area.
Similarly, the x-coordinate of E is 10, so the y-coordinate is the solution to 5y + (2m −
4. A Your solution:
polynomial
4)(10) −f10m (x) with
= 0, real
whose coefficients
solutionsisissaid
y =to 8−be2m.
a sum of squares
Hence 0 ≤ 8 −if 2m
there
≤ are
6 orpolynomials
equivalently
p1 (x), 1p2≤
(x), . . . , p
m ≤ 4. n1 mark(x) with real coefficients for which
Thus 0 ≤ m ≤ 3 and 1 f≤(x) m=≤ p42 so
(x)1+≤pm2 ≤ 3. 1 mark
(x) + · · · + p2 (x)
1 2 n

For example, 2x4 + 6x2 − 4x + 5 is a sum of squares because

√
2x4 + 6x2 − 4x + 5 = (x2 )2 + (x2 + 1)2 + (2x − 1)2 + ( 3)2

(a) Determine all values of a for which f (x) = x2 + 4x + a is a sum of squares.

2
5

cms.math.ca © 2014 CANADIAN MATHEMATICAL SOCIETY

 arn−1 .this
Forline.
example,
1 mark {1, 2, 4, 8, 16} and {1, −3, 9, −27} are both geometric sequences. In all
three questions below, suppose {t1 , t2 , t3 , t4 , t5 } is a geometric 4sequence.
− 2m
The slope of this line is the same as L, so it is given by . 1 mark
5
(a) IfThus
t1 = the
3 and
linet2is= 6, determine thevalue of t5 .
4 − 2m
Page 12 If
(b) of t16
2 = 2 and t4 = 8, determiney all SUN
x +LIFE
= possible values FINANCIAL
of −
(2 t52m).
. CANADIAN OPEN MATHEMATICS CHALLENGE 2014
5
(c) If t1 = 32 and t5 = 2, determine all possible values of t4 .
1 mark
Part C: Question 3 (10 marks)
2. A local high school math club has 12 students in it. Each week, 6 of the students go on a
3. field
A local
trip.high school math club has 12 students in it. Each week, 6 of the students go on a
field trip.
(a) Jeffrey, a student in the math club, has been on a trip with each other student in the
(a) math
Jeffrey, a student
club. in the
Determine themath club, number
minimum has been of on a trip
trips thatwith each
Jeffrey other
could student
have gone in
on.the
math club. Determine the minimum number of trips that Jeffrey could have gone on.
(b) If each pair of students have been on at least one field trip together, determine the
Solution: There are 11 students in the club other than Jeffrey and each field trip that
minimum number of field trips that could have happened.
Jeffrey is on has 5 other students. In order for Jeffrey to go on a field trip with each
Yourother
solution:  11
3. The line student,
L given he +
by 5y must
(2mgo− on
4)xat
− least
10m = 50in 2.2
= the = 3 field
xy-plane trips 2 marks.
intersects the rectangle with
vertices O(0, 0), A(0, 6), B(10, 6), C(10, 0) at D on the line segment OA and E on the line
segment BC.
6
(a) Show that 1 ≤ m ≤ 3.
1
(b) Show that the area of quadrilateral ADEB is 3 the area of rectangle OABC.
(c) Determine, in terms of m, the equation of the line parallel to L that intersects OA at F
and BC at G so that the quadrilaterals ADEB, DEGF , F GCO all have the same area.

4. A polynomial f (x) with real coefficients is said to be a sum of squares if there are polynomials
p1 (x), p2 (x), . . . , pn (x) with real coefficients for which

f (x) = p21 (x) + p22 (x) + · · · + p2n (x)

For example, 2x4 + 6x2 − 4x + 5 is a sum of squares because

√
2x4 + 6x2 − 4x + 5 = (x2 )2 + (x2 + 1)2 + (2x − 1)2 + ( 3)2

(a) Determine all values of a for which f (x) = x2 + 4x + a is a sum of squares.

cms.math.ca © 2014 CANADIAN MATHEMATICAL SOCIETY

 vertices O(0, 0), A(0, 6), B(10, 6), C(10, 0) at D on the line segment OA and E on the line
segment BC.

(a) Show that 1 ≤ m ≤ 3.

Page (b)
14 ofShow
16 is 13LIFE
that the area of quadrilateral ADEBSUN theFINANCIAL
area of rectangle
CANADIANOABC.
OPEN MATHEMATICS CHALLENGE 2014
(c) Determine, in terms of m, the equation of the line parallel to L that intersects OA at F
and BC at G so that the quadrilaterals ADEB, DEGF , F GCO all have the same area.
Part C: Question 4 (10 marks)
4. A polynomial f (x) with real coefficients is said to be a sum of squares if there are polynomials
p1 (x), p2 (x), . . . , pn (x) with real coefficients for which

f (x) = p21 (x) + p22 (x) + · · · + p2n (x)

For example, 2x4 + 6x2 − 4x + 5 is a sum of squares because

√
2x4 + 6x2 − 4x + 5 = (x2 )2 + (x2 + 1)2 + (2x − 1)2 + ( 3)2

(a) Determine all values of a for which f (x) = x2 + 4x + a is a sum of squares.

(b) Determine all values of a for which f (x) = x4 + 2x3 + (a − 7)x2 + (4 − 2a)x + a is a sum
2
of squares, and for such values of a, write f (x) as a sum of squares.
(c) Suppose f (x) is a sum of squares. Prove there are polynomials u(x), v(x) with real
coefficients such that f (x) = u2 (x) + v 2 (x).

Your solution:

cms.math.ca © 2014 CANADIAN MATHEMATICAL SOCIETY

Sun Life Financial
Canadian Open
Mathematics Challenge
2014
Unofficial Solutions v2.0

COMC exams from other years, with or without the solutions included, are free to download online.
Please visit http://comc.math.ca/practice.html

Section A – 4 marks each

1. In triangle ABC, there is a point D on side BC such that BA = AD = DC. Suppose
∠BAD = 80◦ . Determine the size of ∠ACB.
A
HH
S
 S HH
 S HH
 S HH
 S HH
 S HH
 S HH
 S HH
 S H
B D C

Solution 1: Let x = ∠ADB. Then since AB = AD, ∠ABD = x.

Summing the angles of ∆ABD, 2x + ∠BAD = 180◦ . Hence, 2x + 80◦ = 180◦ ⇒ 2x = 100◦ ⇒
x = 50◦ . Therefore, ∠ADB = 50◦ .
Hence, ∠ADC = 180◦ − ∠ADB = 180◦ − 50◦ = 130◦ .
Since AD = DC, ∠ACD = ∠DAC. Let y be this common angle. Then summing the an-
gles of ∆ACD yield 2y + 130◦ = 180◦ . Therefore, 2y = 50◦ . Hence, y = 25◦ . Therefore,
∠ACB = ∠ACD = 25◦ .

Solution 2: Since AB = AD, ∠ABD = ∠ADB.

Since DA = DC, ∠DAC = ∠DCA. Let x = ∠ABD = ∠ADB and y = ∠DAC = ∠DCA.
The solution that we are looking for is y. Then by external angle theorem on ∆ADC, x = 2y.
Summing the angles of ∆ABD then yields 4y + 80◦ = 180◦ .
Therefore, 4y = 100◦ , which yields the solution y = 25◦ .

c 2014 Canadian Mathematical Society Page 1
Unofficial Solutions v2.0 http://comc.math.ca/ COMC 2014

2. The equations x2 − a = 0 and 3x4 − 48 = 0 have the same real solutions. What is
the value of a?

Solution: The equation 3x4 − 48 can be factored as 3(x4 − 16) = 3(x2 − 4)(x2 + 4)

a The real solutions of the equation are x = ±2.

Thus, x2 = 4, so a = 4.
or b Since the equations must have the same real solutions, we must have x2 − a = x2 − 4
Thus a = 4.

3. A positive integer m has the property that when multiplied by 12, the result is a
four-digit number n of the form 20A2 for some digit A. What is the 4 digit number,
n?

Solution 1: For a number to be divisible by 3, the sum of the digits must be a multiple of
3, so 3|(A + 4), which means A ∈ {2, 5, 8}.
For a number to be divisible by 4, the number formed by the last two digits must be divisible
by 4, so 4|(10A + 2). This gives A ∈ {1, 3, 5, 7, 9}.
The only number in common is A = 5, so n = 2052.
Solution 2: The number must be of the form n − 12 is divisible by 5, so n is of the form
60k + 12 for some k.
60 × 34 + 12 = 2052, so n = 2052.

4. Alana, Beatrix, Celine, and Deanna played 6 games of tennis together. In each game,
the four of them split into two teams of two and one of the teams won the game. If
Alana was on the winning team for 5 games, Beatrix for 2 games, and Celine for 1
game, for how many games was Deanna on the winning team?

Solution: Each game has two winners, to there will be a total of 6 × 2 = 12 winners.
Let A, B, C, D be the number of wins for each player. Then the total number of wins can
also be expressed as A + B + C + D.
These two quantities are equal, so A + B + C + D = 12.
With A = 5, B = 2, C = 1 we have D = 12 − 5 − 2 − 1 = 4. .

Page 2
c 2014 Canadian Mathematical Society
COMC 2014 http://comc.math.ca/ Unofficial Solutions v2.0

Section B – 6 marks each

1. The area of the circle that passes through the points (1, 1), (1, 7), and (9, 1) can be
expressed as kπ. What is the value of k?

Solution: The triangle has two sides parallel to the axes, so it is right angled.
The sides parallel to the axes have lengths 7 − 1 = 6 and 9 − 1 = 8.
√
By the Pythagorean Theorem, the length of the hypotenuse is 62 + 82 = 10.
Since the triangle is right angled, the hypotenuse is the diameter of the circle.
The area of the circle is 52 π = 25π so k = 25.

2. Determine all integer values of n for which n2 + 6n + 24 is a perfect square.

Solution: We can write n2 + 6n + 24 = a2 as (n + 3)2 + 15 = a2 .

Let x = n + 3. We can rewrite the given equation as a2 − x2 = 15, or (a − x)(a + x) = 15.
We may assume that a is positive, since we get the same values of x no matter whether we use
the positive or negative a value. This gives us that (a+x, a−x) ∈ {(1, 15), (3, 5), (5, 3), (15, 1)}.
These give solutions for (a, x) of (8, 7), (4, 1), (4, −1), (8, −7).
The corresponding n values for these are 4, −2, −4, −10.

3. 5 Xs and 4 Os are arranged in the below grid such that each number is covered by
either an X or an O. There are a total of 126 different ways that the Xs and Os can
be placed. Of these 126 ways, how many of them contain a line of 3 Os and no line
of 3 Xs?
A line of 3 in a row can be a horizontal line, a vertical line, or one of the diagonal
lines 1 − 5 − 9 or 7 − 5 − 3.

Solution: If we have a horizontal (or vertical) line of all Os, then since there are 5Xs for the
other two lines, there must be a horizontal (or vertical) line of all Xs. Thus, our line of 3 Os
must be a diagonal.
When one of the diagonal lines is all Os, then no other line can be all Xs, since each diagonal
line intersects all other lines. Thus, each configuration with one diagonal line of Os is a
desired solution.

c 2014 Canadian Mathematical Society Page 3
Unofficial Solutions v2.0 http://comc.math.ca/ COMC 2014

When we have the diagonal line 1 − 5 − 9, there are 6 places that the last O could be:
2, 3, 4, 6, 7, 8. Each of these will give a valid solution. Similarly, we have 6 solutions when we
have the diagonal line 7 − 5 − 3. .
It is not possible for both diagonal lines to have only Os, since there are only 4 Os, thus we
have not counted the same configuration twice.
Thus 12 of the 126 ways contain a line of 3 Os and no line of 3 Xs.

1
4. Let f (x) = . Determine the smallest positive integer n such that
x3 + 3x2 + 2x
503
f (1) + f (2) + f (3) + · · · + f (n) > .
2014

Solution 1: Note that x3 + 3x2 + 2x = x(x + 1)(x + 2). We first write 1/(x3 + 3x2 + 2x) as
a b c
+ +
x x+1 x+2
where a, b, c are real numbers.
This expression simplifies to

a(x + 1)(x + 2) + bx(x + 2) + cx(x + 1) (a + b + c)x2 + (3a + 2b + c)x + 2a

= .
x(x + 1)(x + 2) x(x + 1)(x + 2)
Hence, a + b + c = 0, 3a + 2b + c = 0 and 2a = 1.
From the last equation, we get a = 1/2. The first two equations simplifies to b + c = −1/2
and 2b + c = −3/2. Subtracting the former from the latter yields b = −1. Hence, c = 1/2.
Therefore,  
1 1 2 1
f (x) = − + .
2 x x+1 x+2

We now focus on the sum f (1) + f (2) + f (3) + . . . + f (n). This is equal to
     
1 1 1 1 1 1 1 1 1 1 1 1
+ + + ... + − + + ... + + + + ... + .
2 1 2 3 n 2 3 n+1 2 3 4 n+2

All terms except 1, 1/2, 1/(n + 1), 1/(n + 2) cancel out completely. Hence, the result simplifies
to
1 1 1 1 1 1 1 1
− − · + · = − .
2 4 2 (n + 1) 2 n + 2 4 2(n + 1)(n + 2)
We can now solve for the desired answer, i.e. find the smallest positive integer n such that
1 1 503
− > .
4 2(n + 1)(n + 2) 2014
This simplifies to
1 1 503 2 1
< − = = ,
2(n + 1)(n + 2) 4 2014 4 · 2014 2 · 2014
which is equivalent to (n + 1)(n + 2) > 2014.

Page 4
c 2014 Canadian Mathematical Society
COMC 2014 http://comc.math.ca/ Unofficial Solutions v2.0

By trial and error and estimation, we see that 44 · 45 = 1980 and 45 · 46 = 2070. Hence,
n = 44 is the desired answer.

Solution 2: We obtain (n + 1)(n + 2) > 2014 as in Solution 1. This simplifies to n2 + 3n −

2012 > 0. By the quadratic formula, we obtain
√ √
−3 + 32 + 4 · 2012 −3 + 8057
n> = .
2 2
√
The largest positive integer less than 8057 is 89. Hence, n > (−3 + 89)/2 = 43. Therefore,
n = 44 is the smallest positive integer satisfying the given eqation.

c 2014 Canadian Mathematical Society Page 5
Unofficial Solutions v2.0 http://comc.math.ca/ COMC 2014

Section C – 10 marks each

1. A sequence of the form {t1 , t2 , ..., tn } is called geometric if t1 = a, t2 = ar, t3 =
ar2 , . . . , tn = arn−1 . For example, {1, 2, 4, 8, 16} and {1, −3, 9, −27} are both geo-
metric sequences. In all three questions below, suppose {t1 , t2 , t3 , t4 , t5 } is a geometric
sequence of real numbers.

(a) If t1 = 3 and t2 = 6, determine the value of t5 .

Solution: t1 = 3 = a and t2 = ar = 6 , so r = 6/3 = 2. .

This gives t5 = 3 × 24 = 48.

(b) If t2 = 2 and t4 = 8, determine all possible values of t5 .

Solution: t2 = 2 = ar and t4 = 8 = ar3 , Dividing the two equations gives r2 = 4, so

r = ±2.
When r = 2 we have a = 1, so t5 = 24 = 16. .
When r = −2 we have a = −1, so t5 = −1 × 24 = −16.

(c) If t1 = 32 and t5 = 2, determine all possible values of t4 .

1
Solution: We have t1 = 32 = a and t5 = 2 = ar4 . This gives a = 32, and r4 = 16 .
When r4 = 16 1
we get r2 = 14 or r2 = −1
4 .
2 −1
When r = 4 , r is not a real number, so this is not a valid sequence.
When r2 = 14 we get r = ± 12 .
This gives t4 = 32 × 81 = 4 and t4 = 32 × −1
8 = −4.

2. The line L given by 5y + (2m − 4)x − 10m = 0 in the xy-plane intersects the rectangle
with vertices O(0, 0), A(0, 6), B(10, 6), C(10, 0) at D on the line segment OA and
E on the line segment BC.

(a) Show that 1 ≤ m ≤ 3.

Solution: Since D is on OA, the x-coordinate of D is 0. The y-coordinate D is then the

solution to the equation 5y − 10m = 0, i.e. y = 2m. Hence L intersects OA at D(0, 2m).
For D to be on OA, 0 ≤ 2m ≤ 6, or equivalently 0 ≤ m ≤ 3.
Similarly, the x-coordinate of E is 10, so the y-coordinate is the solution to 5y + (2m −
4)(10) − 10m = 0, whose solutions is y = 8 − 2m. Hence 0 ≤ 8 − 2m ≤ 6 or equivalently
1 ≤ m ≤ 4.
Thus 0 ≤ m ≤ 3 and 1 ≤ m ≤ 4 so 1 ≤ m ≤ 3.

Page 6
c 2014 Canadian Mathematical Society
COMC 2014 http://comc.math.ca/ Unofficial Solutions v2.0

1
(b) Show that the area of quadrilateral ADEB is 3 the area of rectangle OABC.

Solution: Observe ADEB is a trapezoid with base AB and parallel sides are AD and
BE, so its area is

AD + BE (6 − 2m) + (6 − (8 − 2m)) 4
AB · = 10 · = 10 · = 20,
2 2 2
and since the area of OABC is 6 · 10, the result follows.

(c) Determine, in terms of m, the equation of the line parallel to L that intersects
OA at F and BC at G so that the quadrilaterals ADEB, DEGF , F GCO all
have the same area.

Solution: In order for the quadrilaterals to have equal area, it is sufficient to demand
F GCO has area 20 (i.e. 31 the area of OABC).
Let M (5, b) be the midpoint of F and G. Then the average of the y-coordinates of F
and G is b, so the area of F GCO is b · 10 = 10b, so b = 2. Hence the point M (5, 2) is on
this line.
4 − 2m
The slope of this line is the same as L, so it is given by .
5
Thus the line is  
4 − 2m
y= x + (2m − 2).
5

3. A local high school math club has 12 students in it. Each week, 6 of the students go
on a field trip.

(a) Jeffrey, a student in the math club, has been on a trip with each other student
in the math club. Determine the minimum number of trips that Jeffrey could
have gone on.

Solution: There are 11 students in the club other than Jeffrey and each field trip that
Jeffrey is on has 5 other students. In order for Jeffrey to go on a field trip with each
other student, he must go on at least d 11
5 e = d2.2e = 3 field trips.

c 2014 Canadian Mathematical Society Page 7
Unofficial Solutions v2.0 http://comc.math.ca/ COMC 2014

To see that 3 trips is sufficient, we let the other students be {s1 , s2 , . . . , s11 }. On the first
trip, Jeffrey can go with {s1 , . . . , s5 }, on the second trip with {s6 , . . . , s10 } and on the
third trip with s11 and any other 4 students.

(b) If each pair of students have been on at least one field trip together, determine
the minimum number of field trips that could have happened.

Solution: From part (a), we know that each student must go on at least 3 trips.
Since there are 12 students in total, if we count the number of trips that each student
went on, we would get a minimum of 12 × 3 = 36.
Since 6 students attend each field trip, that means there must be at least 36
6 = 6 trips.
We now show that we can do this with 6 trips.
Divide the students into 4 groups of 3 students each (groups A, B, C, D). There are 6
different pairs of groups (AB, AC, AD, BC, BD, CD). Let these pairs of groups be our
6 field trips.
We see that since each group goes on a trip with each other group, that each pair of
students goes on a trip together. Hence, this can be done with 6 trips.

4. A polynomial f (x) with real coefficients is said to be a sum of squares if there are
polynomials p1 (x), p2 (x), . . . , pn (x) with real coefficients for which

f (x) = p21 (x) + p22 (x) + · · · + p2n (x).

For example, 2x4 + 6x2 − 4x + 5 is a sum of squares because

√
2x4 + 6x2 − 4x + 5 = (x2 )2 + (x2 + 1)2 + (2x − 1)2 + ( 3)2 .

(a) Determine all values of a for which f (x) = x2 + 4x + a is a sum of squares.

Solution: If f (x) is a sum of squares of polynomials, then f (x) must be nonnegative

for all values of x. Completing the square gives us f (x) = (x + 2)2 + (a − 4).
So f (x) is nonnegative for all x provided that a − 4 ≥ 0, i.e. a ≥ 4. This is in fact
sufficient for f (x) to be a sum of squares, since if a − 4 ≥ 0 then
√
2
f (x) = (x + 2)2 + a−4 .

Thus f (x) is a sum of squares if and only if a ≥ 4.

(b) Determine all values of a for which f (x) = x4 + 2x3 + (a − 7)x2 + (4 − 2a)x + a
is a sum of squares, and for such values of a, write f (x) as a sum of squares.

Solution: The sum of the coefficients of f (x) is 0, so x − 1 is a factor. Factoring this

out we have
f (x) = (x − 1)(x3 + 3x2 + (a − 4)x − a).

Page 8
c 2014 Canadian Mathematical Society
COMC 2014 http://comc.math.ca/ Unofficial Solutions v2.0

Since the sum of the coefficients of x3 + 3x2 + (a − 4)x − a is also 0, x − 1 is a factor of

it. Factoring this out we have

f (x) = (x − 1)2 (x2 + 4x + a).

If f (x) is a sum of squares, then it must be nonnegative. Since (x − 1)2 is always

nonnegative, we require x2 + 4x + a to be nonnegative, which as in part (a), requires
a ≥ 4.
For such a, we have

f (x) = (x − 1)2 (x2 + 4x + a)

 √
2 
= (x − 1)2 (x + 2)2 + a−4
√
= [(x − 1)(x + 2)]2 + [ a − 4(x − 1)]2 .

Hence f (x) is a sum of squares if and only if a ≥ 4, and we can express f (x) as a sum
of squares as shown above.

(c) Suppose f (x) is a sum of squares. Prove there are polynomials u(x), v(x) with
real coefficients such that f (x) = u2 (x) + v 2 (x).

Solution: Since f (x) is nonnegative for all x, its leading coefficient must be positive,
and we can therefore assume it is monic by factoring out the square root of its leading
coefficient. Now the non-real roots of f (x) come in pairs (complex conjugates), so f (x)
factors into a product of linear and irreducible quadratic polynomials over the reals,
raised to certain powers, say
m
Y n
Y
f (x) = pi (x)ki qi (x)ji .
i=1 j=1

where pi ’s are the distinct linear polynomials, qi ’s are the distinct irreducible quadratic
polynomials, and ki , ji > 1 for each i.
For each i, let qi (x) = x2 + ai x + bi . Since qi (x) is irreducible over the reals, a2i − 4bi < 0,
so qi (x) can be written as a sum of squares, namely
r !2
 ai 2 a2
qi (x) = x + + bi − i .
2 4

Now for each i, let pi (x) = x − ci .

We claim ki is even for all i. Suppose otherwise, and say the exponents k1 , k2 , . . . , km
which are odd are ki1 , ki2 , . . . , kil where without loss of generality ci1 < ci2 < · · · < cil .
Then
f (x) = (x − ci1 )(x − ci2 ) · · · (x − cil )g(x).
where g(x) is nonnegative for all x (since g(x) is the product of irreducible quadratics
which we proved are sums
 c +cof squares
 and hence nonnegative, and even powers of linear
il il−1
polynomials). Then f 2 < 0, which is impossible if f (x) is a sum of squares.

c 2014 Canadian Mathematical Society Page 9
Unofficial Solutions v2.0 http://comc.math.ca/ COMC 2014

We then deduce that

n
Y
f (x) = h(x)2 rj2 (x) + s2j (x) .


(1)
j=1
Qm ki
where h(x) and rj (x), sj (x) q
are polynomials, for all j, namely ( h(x) = i=1 pi (x)
2 ,
a a2j
rj (x) = x + 2j , and sj (x) = bj − 4 ).
Now observe that for any i, j,

(ri2 + s2i )(rj2 + s2j ) = (ri rj + si sj )2 + (ri sj − rj si )2 .

Repeatedly applying this identity to(1) gives us polynomials P (x), Q(x) for which
Qn  2
P (x) + Q (x) = j=1 rj (x) + s2j (x) , and hence
2 2

f (x) = (h(x)P (x))2 + (h(x)Q(x))2 .

Page 10
c 2014 Canadian Mathematical Society
 Part II

Canadian
Mathematical
Olympiad From
1969-2014
 Canadian Mathematical Olympiad
1994
PROBLEM 1
Evaluate the sum
X(,1) n + n + 1 :
1994
n
2

n=1
n!
PROBLEM 2 p
Show that every positive integralppower of 2 , 1pis of pthe form
pm , pm , 1 for
p
some positive integer m. (e.g. ( 2 , 1)2 = 3 , 2 2 = 9 , 8).
PROBLEM 3
Twenty-ve men sit around a circular table. Every hour there is a vote, and each
must respond yes or no. Each man behaves as follows: on the nth vote, if his
response is the same as the response of at least one of the two people he sits
between, then he will respond the same way on the (n + 1)th vote as on the nth
vote; but if his response is dierent from that of both his neighbours on the n-th
vote, then his response on the (n + 1)-th vote will be dierent from his response
on the nth vote. Prove that, however everybody responded on the rst vote, there
will be a time after which nobody's response will ever change.
PROBLEM 4
Let AB be a diameter of a circle
and P be any point not on the line through A
and B . Suppose the line through P and A cuts
again in U , and the line through
P and B cuts
again in V . (Note that in case of tangency U may coincide with
A or V may coincide with B . Also, if P is on
then P = U = V .) Suppose
that jPU j = sjPAj and jPV j = tjPB j for some nonnegative real numbers s and t.
Determine the cosine of the angle APB in terms of s and t.
PROBLEM 5
Let ABC be an acute angled triangle. Let AD be the altitude on BC , and let H
be any interior point on AD. Lines BH and CH , when extended, intersect AC
and AB at E and F , respectively. Prove that EDH = FDH .
6 6

{1{
 Canadian Mathematical Olympiad
1995
PROBLEM 1
Let ( ) = 9x9+3
f x
x
. Evaluate the sum
1 ) + ( 2 ) + ( 3 ) +


+ ( 1995 )
( 1996
f
1996f
1996 f
1996 f

PROBLEM 2
Let , , and be positive real numbers. Prove that
a b c

a b c
a b c  ( ) a+3b+c
abc :

PROBLEM 3
Dene a boomerang as a quadrilateral whose op-
posite sides do not intersect and one of whose in-
ternal angles is greater than 180 degrees. (See
Figure displayed.) Let be a convex polygon
C

having 5 sides. Suppose that the interior region of

C is the union of quadrilaterals, none of whose
q

interiors intersect one another. Also suppose that

b of these quadrilaterals are boomerangs. Show
that  + s,2 2 .
q b

PROBLEM 4
Let be a xed positive integer. Show that for only nonnegative integers , the
n k

diophantine equation
3 + 3 +


+ 3 = 3k+2
1 2 x x n x y

has innitely many solutions in positive integers i and . x y

PROBLEM 5
Suppose that is a real parameter with 0
u 1. Dene
< u <

(
0 if 0  
( ) = 1 , p + p(1 , )(1 , )2 if   1
x u
f x
ux u x u x

and dene the sequence f ng recursively as follows:

u

u1 = f (1); and un = f (un,1) for all n > 1:

Show that there exists a positive ineger for whichk k
u = 0.
{1{
 Canadian Mathematical Olympiad
1996
PROBLEM 1
If ; ; are the roots of x3 , x , 1 = 0, compute
1 +  + 1 +  + 1 + :
1, 1, 1,

PROBLEM 2
Find all real solutions to the following system of equations. Carefully justify your
answer. 8 4x2
>>
>< 1 +4y42x2 = y
>> 1 + 42y2 = z
>: 4z = x
1 + 4z 2
PROBLEM 3
We denote an arbitrary permutation of the integers 1; : : : ; n by a1 ; : : : ; an . Let f (n)
be the number of these permutations such that
(i) a1 = 1;
(ii) jai , ai+1 j  2; i = 1; : : : ; n , 1.
Determine whether f (1996) is divisible by 3.
PROBLEM 4
Let 4ABC be an isosceles triangle with AB = AC . Suppose that the angle bisector
of B meets AC at D and that BC = BD + AD. Determine A.
6 6

PROBLEM 5
Letm r1 ; r2 ; : : : ; rm be a given set of m positive rational
P Pm [r numbers such that
r
k=1 k = 1. Dene the function f by f ( n ) = n , k=1 k n ] for each positive
integer n. Determine the minimum and maximum values of f (n). Here [x] denotes
the greatest integer less than or equal to x

{1{
 CMO 1996
SOLUTIONS

QUESTION 1

Solution .

If f (x) = x3 − x − 1 = (x − α)(x − β)(x − γ) has roots α, β, γ standard results about roots of

polynomials give α + β + γ = 0, αβ + αγ + βγ = −1, and αβγ = 1.

Then
1+α 1+β 1+γ N
S= + + =
1−α 1−β 1−γ (1 − α)(1 − β)(1 − γ)

where the numerator simplifies to

N = 3 − (α + β + γ) − (αβ + αγ + βγ) + 3αβγ

= 3 − (0) − (−1) + 3(1)
= 7.

The denominator is f (1) = −1 so the required sum is −7.

 QUESTION 2

Solution 1.
4t2
For any t, 0 ≤ 4t2 < 1 + 4t2 , so 0 ≤ < 1. Thus x, y and z must be non-negative and less
1 + 4t2
than 1.

Observe that if one of x y or z is 0, then x = y = z = 0.

If two of the variables are equal, say x = y, then the first equation becomes

4x2
= x.
1 + 4x2

1 1
This has the solution x = 0, which gives x = y = z = 0 and x = which gives x = y = z = .
2 2
Finally, assume that x, y and z are non-zero and distinct. Without loss of generality we may
assume that either 0 < x < y < z < 1 or 0 < x < z < y < 1. The two proofs are similar, so we do
only the first case.
4t2
We will need the fact that f (t) = is increasing on the interval (0, 1).
1 + 4t2
To prove this, if 0 < s < t < 1 then

4t2 4s2
f (t) − f (s) = −
1 + 4t2 1 + 4s2
4t2 − 4s2
=
(1 + 4s2 )(1 + 4t2 )
> 0.

So 0 < x < y < z ⇒ f (x) = y < f (y) = z < f (z) = x, a contradiction.

1
Hence x = y = z = 0 and x = y = z = are the only real solutions.
2
Solution 2.

Notice that x, y and z are non-negative. Adding the three equations gives

4z 2 4x2 4y 2
x+y+z = 2
+ 2
+ .
1 + 4z 1 + 4x 1 + 4y 2

This can be rearranged to give

x(2x − 1)2 y(2y − 1)2 z(2z − 1)2

+ + = 0.
1 + 4x2 1 + 4y 2 1 + 4z 2

1
Since each term is non-negative, each term must be 0, and hence each variable is either 0 or . The
2
1
original equations then show that x = y = z = 0 and x = y = z = are the only two solutions.
2

2
Solution 3.

Notice that x, y, and z are non-negative. Multiply both sides of the inequality
y
≥0
1 + 4y 2

by (2y − 1)2 , and rearrange to obtain

4y 2
y− ≥ 0,
1 + 4y 2
and hence that y ≥ z. Similarly, z ≥ x, and x ≥ y. Hence, x = y = z and, as in Solution 1, the
two solutions follow.

Solution 4.

As for solution 1, note that x = y = z = 0 is a solution and any other solution will have each of
x, y and z positive.
1 + 4x2 √
The arithmetic-geometric mean inequality (or direct computation) shows that ≥ 1 · 4x2 = 2x
2
4x2 1
and hence x ≥ 2
= y, with equality if and only if 1 = 4x2 – that is, x = . Similarly, y ≥ z
1 + 4x 2
1 1
with equality if and only if y = and z ≥ x with equality if and only if z = . Adding x ≥ y, y ≥ z
2 2
1
and z ≥ x gives x+y +x ≥ x+y +z. Thus equality must occur in each inequality, so x = y = z = .
2

3
 QUESTION 3

Solution.

Let a1 , a2 , . . . , an be a permutation of 1, 2, . . . , n with properties (i) and (ii).

A crucial observation, needed in Case II (b) is the following: If ak and ak+1 are consecutive integers
(i.e. ak+1 = ak ± 1), then the terms to the right of ak+1 (also to the left of ak ) are either all less
than both ak and ak+1 or all greater than both ak and ak+1 .

Since a1 = 1, by (ii) a2 is either 2 or 3.

CASE I: Suppose a2 = 2. Then a3 , a4 , . . . , an is a permutation of 3, 4, . . . , n. Thus a2 , a3 , . . . , an

is a permutation of 2, 3, . . . , n with a2 = 2 and property (ii). Clearly there are f (n − 1) such
permutations.

CASE II: Suppose a2 = 3.

(a) Suppose a3 = 2. Then a4 , a5 , . . . , an is a permutation of 4, 5, . . . , n with a4 = 4 and property

(ii). There are f (n − 3) such permutations.

(b) Suppose a3 ≥ 4. If ak+1 is the first even number in the permutation then, because of (ii),
a1 , a2 , . . . , ak must be 1, 3, 5, . . . , 2k − 1 (in that order). Then ak+1 is either 2k or 2k − 2, so
that ak and ak+1 are consecutive integers. Applying the crucial observation made above, we
deduce that ak+2 , . . . , an are all either greater than or smaller than ak and ak+1 . But 2 must
be to the right of ak+1 . Hence ak+2 , . . . , an are the even integers less than ak+1 . The only
possibility then, is

1, 3, 5, . . . , ak−1 , ak , . . . , 6, 4, 2.

Cases I and II show that

f (n) = f (n − 1) + f (n − 3) + 1, n ≥ 4. (∗)

Calculating the first few values of f (n) directly gives

f (1) = 1, f (2) = 1, f (3) = 2, f (4) = 4, f (5) = 6.

Calculating a few more f (n)’s using (*) and mod 3 arithmetic, f (1) = 1, f (2) = 1, f (3) =
2, f (4) = 1, f (5) = 0, f (6) = 0, f (7) = 2, f (8) = 0, f (9) = 1, f (10) = 1, f (11) = 2. Since
f (1) = f (9), f (2) = f (10) and f (3) = f (11) mod 3, (*) shows that f (a) = f (a mod 8), mod 3, a ≥
1.

Hence f (1996) ≡ f (4) ≡ 1 (mod 3) so 3 does not divide f (1996).

4
 QUESTION 4

Solution 1.

Let BE = BD with E on BC, so that AD = EC:

D
4x
2x

x
x 4x 2x
B C
E

AB AD
By a standard theorem, = ; so in
CB DC

4 CED and 4 CAB we have a common angle and

CE AD AB CA
= = = .
CD CD CB CB

Thus 4 CED ∼ 4 CAB , so that ∠ CDE = ∠ DCE = ∠ ABC = 2x .

Hence ∠BDE = ∠ BED = 4x , whence 9x = 180◦ so x = 20◦ .
Thus ∠A = 180◦ − 4x = 100◦ .

Solution 2.

Apply the law of sines to 4ABD and 4BDC to get

AD sin x AD BC sin 3x
= and 1+ = = .
BD sin 4x BD BD sin 2x

Now massage the resulting trigonometric equation with standard identities to get

sin 2x (sin 4x + sin x) = sin 2x (sin 5x + sin x) .

Since 0 < 2x < 90◦ , we get

5x − 90◦ = 90◦ − 4x ,

so that ∠A = 100◦ .

5
 QUESTION 5

Solution.

Let m
X
f (n) = n − [rk n]
k=1
m
X m
X
=n rk − [rk n]
k=1 k=1
m
X
= {rk n − [rk n]}.
k=1

Now 0 ≤ x − [x] < 1, and if c is an integer, (c + x) − [c + x] = x − [x].

m
X
Hence 0 ≤ f (n) < 1 = m. Because f (n) is an integer, 0 ≤ f (n) ≤ m − 1.
k=1
ak
To show that f (n) can achieve these bounds for n > 0, we assume that rk = where ak , bk are
bk
integers; ak < bk .

Then, if n = b1 b2 . . . bm , (rk n) − [rk n] = 0, k = 1, 2, . . . , m and thus f (n) = 0.

Letting n = b1 b2 . . . bn − 1, then

rk n = rk (b1 b2 . . . bm − 1)
= rk {(b1 b2 . . . bm − bk ) + bk − 1)}
= integer + rk (bk − 1).

This gives
rk n − [rk n] = rk (bk − 1) − [rk (bk − 1)]
· ¸
ak ak
= (bk − 1) − (bk − 1)
bk bk
µ ¶ · ¸
ak ak
= ak − − ak −
bk bk
µ ¶
ak
= ak − − (ak − 1)
bk
ak
=1− = 1 − rk .
bk

Pm
Hence f (n) = k=1 (1 − rk ) = m − 1.

6
 Canadian Mathematical Olympiad
1997
PROBLEM 1
How many pairs of positive integers x; y are there, with x  y, and such that
gcd(x; y) = 5! and lcd(x; y) = 50!.
NOTE. gcd(x; y) denotes the greatest common divisor of x and y, lcd(x; y) denotes
the least common multiple of x and y, and n! = n  (n , 1) 


 2  1.
PROBLEM 2
The closed interval A = [0; 50] is the union of a nite number of closed intervals,
each of length 1. Prove that some of the intervals can be removed so that those
remaining are mutually disjoint and have total length  25.
NOTE. For a  b, the closed interval [a; b] := fx 2 R : a  x  bg has length b , a;
disjoint intervals have empty intersection.
PROBLEM 3
Prove that
1 1 3 5 1997 1
1999 < 2
4
6


1998 < 44 :
PROBLEM 4
The point O is situated inside the parallelogram ABCD so that
6 AOB + 6 COD = 180:
Prove that 6 OBC = 6 ODC .
PROBLEM 5
Write the sum
,

n
X (,1)k nk
k=0
k3 + 9k2 + 26k + 24
in the form p(n)
q(n) , where p and q are polynomials with integer coecients.

{1{
 1997
SOLUTIONS

Problem 1 – Deepee Khosla, Lisgar Collegiate Institute, Ottawa, ON

Let p1 , . . . , p12 denote, in increasing order, the primes from 7 to 47. Then

5! = 23 · 31 · 51 · p01 · p02 . . . p012

and

50! = 2a1 · 3a2 · 5a3 · pb11 · pb22 . . . pb1212 .

Note that 24 , 32 , 52 , p1 , . . . , p12 all divide 50!, so all its prime powers differ from those of 5!

Since x, y|50!, they are of the form

x = 2n1 · 3n2 · . . . pn1215

y = 2m1 · 3m2 · . . . pm 15
12 .

Then max(ni , mi ) is the ith prime power in 50!

and min(ni , mi ) is the ith prime power in 5!

Since, by the above note, the prime powers for p12 and under differ in 5! and 50!, there are 215
choices for x, only half of which will be less than y. (Since for each choice of x, y is forced and
either x < y or y < x.) So the number of pairs is 215 /2 = 214 .
Problem 2 – Byung Kuy Chun, Harry Ainlay Composite High School, Edmonton, AB

Look at the first point of each given unit interval. This point uniquely defines the given unit
interval.

Lemma. In any interval [x, x + 1) there must be at least one of these first points (0 ≤ x ≤ 49).

Proof. Suppose the opposite. The last first point before x must be x − ε for some ε > 0. The
corresponding unit interval ends at x − ε + 1 < x + 1. However, the next given unit interval cannot
begin until at least x + 1.

This implies that points (x − ε + 1, x + 1) are not in set A, a contradiction.

∴ There must be a first point in [x, x + 1). 2

Note that for two first points in intervals [x, x + 1) and [x + 2, x + 3) respectively, the corresponding
unit intervals are disjoint since the intervals are in the range [x, x + 2) and [x + 2, x + 4) respectively.

∴ We can choose a given unit interval that begins in each of

[0, 1)[2, 3) . . . [2k, 2k + 1) . . . [48, 49).

Since there are 25 of these intervals, we can find 25 points which correspond to 25 disjoint unit
intervals.

Problem 2 – Colin Percival, Burnaby Central Secondary School, Burnaby, BC

[
I prove the more general result, that if [0, 2n] = Ai , |Ai | = 1, Ai are intervals then ∃a1 . . . an , such
T i
that Aai Aaj = ∅.
2
Let 0 < ε ≤ and let bi = (i − 1)(2 + ε), i = 1 . . . n. Then
n−1
µ ¶ µ ¶
2 2n
min{bi } = 0, max{bi } = (n − 1)(2 + ε) ≤ (n − 1) 2 + = (n − 1) = 2n.
n−1 n−1

So all the bi are in [0, 2n].

[
Let ai be such that bi ∈ Aai . Since Ai = [0, 2n], this is possible.

Then since (bi − bj ) = (i − j)(2 +

\
ε) ≥ 2 + ε > 2, and the Ai are intervals of length 1, min Aai −
max Aaj > 2 − 1 − 1 = 0, so Aai Aaj = ∅.

Substituting n = 25, we get the required result. Q.E.D.

2
Problem 3 – Mihaela Enachescu, Dawson College, Montréal, PQ
1 3 1997 1 1 3 3
Let P = · · ... · . Then > because 2 < 3, > because 4 < 5, . . .,
2 4 1998 2 3 4 5
1997 1997
... > because 1998 < 1999 .
1998 1999
So
1 3 1997 1
P > · · ... · = . (1)
3 5 1999 1999
1 2 3 4
Also < because 1 · 3 < 2 · 2, < because 3 · 5 < 4 · 4, . . .
2 3 4 5

1997 1998
< because 1997 · 1999 = 19982 − 1 < 19982 .
1998 1999
µ ¶
2 4 1998 2 4 6 1998 1
So P < · · . . . · = · · · ... · .
3 5 1999 1 3 5 1997 1999
| {z }
1
P

1 1 1 1
Hence P 2 < < = 2 and P < . (2)
1999 1936 44 44
1 1
Then (1) and (2) give <P < (q.e.d.)
1999 44

Problem 4 – Joel Kamnitzer, Earl Haig Secondary School, North York, ON

O’

A B

D C

Consider a translation which maps D to A. It will map 0 → 00 with OO0 = DA, and C will be
mapped to B because CB = DA.

This translation keeps angles invariant, so ∠AO0 B = ∠DOC = 180◦ − ∠AOB.

∴ AOBO 0 is a cyclic quadrilateral.

∴ ∠ODC = ∠O 0 AB = ∠O 0 OB

but, since O0 O is parallel to BC,

∠O 0 OB = ∠OBC
∴ ∠ODC = ∠OBC.

3
Problem 4 – Adrian Chan, Upper Canada College, Toronto, ON
A B

θ
180−α α
Ο
180−θ

D C

Let ∠AOB = θ and ∠BOC = α. Then ∠COD = 180◦ − θ and ∠AOD = 180◦ − α.

Since AB = CD (parallelogram) and sin θ = sin(180◦ − θ), the sine law on 4OCD and 4OAB
gives
sin ∠CDO sin(180◦ − θ) sin θ sin ∠ABO
= = =
OC CD AB OA
so
OA sin ∠ABO
= . (1)
OC sin ∠CDO

Similarily, the sine law on 4OBC and 4OAD gives

sin ∠CBO sin α sin(180◦ − α) sin ∠ADO
= = =
OC BC AD OA
so
OA sin ∠ADO
= . (2)
OC sin ∠CBO

Equations (1) and (2) show that sin ∠ABO · sin ∠CBO = sin ∠ADO · sin ∠CDO hence
1 1
[cos(∠ABO + ∠CBO)−cos(∠ABO − ∠CBO)] = [cos(∠ADO + ∠CDO)−cos(∠ADO − ∠CDO)].
2 2
Since ∠ADC = ∠ABC(parallelogram) and ∠ADO + ∠CDO = ∠ADC and ∠ABO + ∠CBO =
∠ABC it follows that cos(∠ABO − ∠CBO) = cos(∠ADO − ∠CDO).

There are two cases to consider.

Case (i): ∠ABO − ∠CBO = ∠ADO − ∠CDO.

Since ∠ABO + ∠CBO = ∠ADO + ∠CDO, subtracting gives 2 ∠CBO = 2 ∠CDO so ∠CBO =
∠CDO, and we are done.

Case (ii): ∠ABO − ∠CBO = ∠CDO − ∠ADO.

Since we know that ∠ABO + ∠CBO = ∠CDO + ∠ADO, adding gives 2 ∠ABO = 2 ∠CDO so
∠ABO = ∠CDO and ∠CBO = ∠ADO.

Substituting this into (1), it follows that OA = OC.

Also, ∠ADO + ∠ABO = ∠CBO + ∠ABO = ∠ABC.

Now, ∠ABC = 180◦ − ∠BAD since ABCD is a parallelogram.

Hence ∠BAD + ∠ADO + ∠ABO = 180◦ so ∠DOB = 180◦ and D, O, B are collinear.

We now have the diagram

4
 A B

θ
O
180−θ
D C

Then ∠COD + ∠BOC = 180◦ , so ∠BOC = θ = ∠AOB.

4AOB is congruent to 4COB (SAS, OB is common, ∠AOB = ∠COB and AO = CO), so

∠ABO = ∠CBO. Since also ∠ABO = ∠CDO we conclude that ∠CBO = ∠CDO.

Since it is true in both cases, then ∠CBO = ∠CDO.

Q.E.D.

5
Problem 5 – Sabin Cautis, Earl Haig Secondary School, North York, ON

We first note that

k 3 + 9k 2 + 26k + 24 = (k + 2)(k + 3)(k + 4).

à !
n
n (−1)k
X k
Let S(n) = .
k=0
k 2 + 9k 2 + 26k + 24

Then

n
X (−1)k n!
S(n) =
k=0
k!(n − k)!(k + 2)(k + 3)(k + 4)
n
à ! µ ¶
X (−1)k (n + 4)! k+1
= × .
k=0
(k + 4)!(n − k)! (n + 1)(n + 2)(n + 3)(n + 4)

Let

n
à à ! !
X n+4
k
T (n) = (n + 1)(n + 2)(n + 3)(n + 4)S(n) = (−1) (k + 1) .
k+4
k=0

Now, for n ≥ 1,

n
à !
X n
i
(−1) =0 (∗)
i=0
i

since
à ! à ! à ! à !
n n n n n n
(1 − 1) = − + + . . . + (−1) = 0.
0 1 2 n

Also

n
à ! n
X n X i · n! 0 · n!
(−1)i i = (−1)i + (−1)0 ·
i=0
i i=1
i! · (n − i)! 0! · n!
n
X n!
= (−1)i
i=1
(i − 1)!(n − i)!
n
à !
X n−1
i
= (−1) n
i=1
i−1
n
à !
X n−1
= n (−1)i
i=1
i−1
n
à !
X n−1
i−1
= −n (−1) .
i=1
i−1

6
Substituting j = i − 1, (*) shows that

n
à ! n−1
à !
X n X n−1
i j
(−1) i = −n (−1) = 0. (∗∗)
i=0
i j=0
j

Hence

n
à !
X n+4
k
T (n) = (−1) (k + 1)
k+4
k=0
n
à !
X n+4
k+4
= (−1) (k + 1)
k+4
k=0
n
à ! à à !!
X n+4 n+4
k+4
= (−1) (k + 1) − −3 + 2(n + 4) − .
k+4 2
k=−4
Substituting j = k + 4,
n+4
à ! µ ¶
X n+4 (n + 4)(n + 3)
j
= (−1) (j − 3) − 2n + 8 − 3 −
j=0
j 2
n+4
à ! n+4
à !
X n+4 X n+4 1
j j
= (−1) j−3 (−1) − (4n + 10 − n2 − 7n − 12)
j=0
j j=0
j 2

The first two terms are zero because of results (*) and (**) so

n2 + 3n + 2
T (n) = .
2

Then
T (n)
S(n) =
(n + 1)(n + 2)(n + 3)(n + 4)
n2 + 3n + 2
=
2(n + 1)(n + 2)(n + 3)(n + 4)
(n + 1)(n + 2)
=
2(n + 1)(n + 2)(n + 3)(n + 4)
1
= .
2(n + 3)(n + 4)

à !
n
n (−1)k
X k 1
∴ =
k=0
k3 + 9k 2 + 26k + 24 2(n + 3)(n + 4)

7
 THE 1998 CANADIAN MATHEMATICAL
OLYMPIAD

1. Determine the number of real solutions a to the equation

· ¸ · ¸ · ¸
1 1 1
a + a + a =a.
2 3 5

Here, if x is a real number, then [ x ] denotes the greatest integer that is less than or
equal to x.

2. Find all real numbers x such that

µ ¶1/2 µ ¶1/2
1 1
x= x− + 1− .
x x

3. Let n be a natural number such that n ≥ 2. Show that

µ ¶ µ ¶
1 1 1 1 1 1 1
1+ +···+ > + +···+ .
n+1 3 2n − 1 n 2 4 2n

4. Let ABC be a triangle with ∠BAC = 40◦ and ∠ABC = 60◦ . Let D and E be
the points lying on the sides AC and AB, respectively, such that ∠CBD = 40◦ and
∠BCE = 70◦ . Let F be the point of intersection of the lines BD and CE. Show that
the line AF is perpendicular to the line BC.

5. Let m be a positive integer. Define the sequence a0 , a1 , a2 , . . . by a0 = 0, a1 = m,

and an+1 = m2 an − an−1 for n = 1, 2, 3, . . . . Prove that an ordered pair (a, b) of
non-negative integers, with a ≤ b, gives a solution to the equation

a2 + b2
= m2
ab + 1
if and only if (a, b) is of the form (an , an+1 ) for some n ≥ 0.
 1998
SOLUTIONS

The solutions to the problems of the 1998 CMO presented below are taken from students papers.
Some minor editing has been done - unnecesary steps have been eliminated and some wording has
been changed to make the proofs clearer. But for the most part, the proofs are as submitted.

Solution to Problem 1 – David Arthur, Upper Canada College, Toronto, ON

Let a = 30k + r, where k is an integer and r is a real number between 0 and 29 inclusive.
· ¸ · ¸ · ¸ · ¸ · ¸ · ¸ · ¸
1 1 r 1 r 1 r
Then a = (30k + r) = 15k + . Similarly a = 10k + and a = 6k + .
2 2 2 3 3 5 5
· ¸ · ¸ · ¸ µ · ¸¶ µ · ¸¶ µ · ¸¶
1 1 1 r r r
Now, a + a + a = a, so 15k + + 10k + + 6k + = 30k + r and
2 3 5 2 3 5
· ¸ · ¸ · ¸
r r r
hence k = r − − − .
2 3 5
· ¸ · ¸ · ¸
r r r
Clearly, r has to be an integer, or r − − − will not be an integer, and therefore, cannot
2 3 5
equal k.
· ¸ · ¸ · ¸
r r r
On the other hand, if r is an integer, then r − − − will also be an integer, giving
2 3 5
exactly one solution for k.

For each r(0 ≤ r ≤ 29), a = 30k + r will have a different remainder mod 30, so no two different
values of r give the same result for a.

Since there are 30 possible values for r(0, 1, 2, . . . , 29), there are then 30 solutions for a.

1
Solution to Problem 2 – Jimmy Chui, Earl Haig S.S., North York, ON
µ ¶1/2 µ ¶1/2 µ ¶1/2 µ ¶1/2
1 1 1 1
Since x− ≥ 0 and 1− ≥ 0, then 0 ≤ x − + 1− =x.
x x x x
1
Note that x 6= 0. Else, would not be defined so x > 0.
x
Squaring both sides gives,
µ ¶ µ ¶ sµ ¶µ ¶
2 1 1 1 1
x = x− + 1− +2 x− 1−
x x x x
r
2 2 1 1
x = x + 1 − + 2 x − 1 − + 2.
x x x

Multiplying both sides by x and rearranging, we get

p
x3 − x2 − x + 2 = 2 x3 − x2 − x + 1
p
(x3 − x2 − x + 1) − 2 x3 − x2 − x + 1 + 1 = 0
p
( x3 − x2 − x + 1 − 1)2 = 0
p
x3 − x2 − x + 1 = 1

x3 − x2 − x + 1 = 1

x(x2 − x − 1) = 0

x2 − x − 1 = 0 since x 6= 0.

√
1± 5
Thus x = . We must check to see if these are indeed solutions.
2
√ √
1+ 5 1− 5
Let α = , β= . Note that α + β = 1, αβ = −1 and α > 0 > β.
2 2
Since β < 0, β is not a solution.

Now, if x = α, then

µ ¶1/2 µ ¶1/2
1 1
α− + 1− = (α + β)1/2 + (1 + β)1/2 (since αβ − −1)
α α
= 11/2 + (β 2 )1/2 (since α + β = 1 and β 2 = β + 1)

= 1−β (since β < 0)

= α (since α + β = 1).

So x = α is the unique solution to the equation.

2
Solution 1 to Problem 3 – Chen He, Columbia International Collegiate, Hamilton, ON

1 1 1 1 1 1 1
1+ + ... + = + + + + ... (1)
3 2n − 1 2 2 3 5 2n − 1

Since
1 1 1 1 1 1
> , > , ... , > ,
3 4 5 6 2n − 1 2n

(1) gives
µ ¶
1 1 1 1 1 1 1 1 1 1 1 1
1 + + ... + > + + + + ... + = + + + + ... + . (2)
3 2n − 1 2 2 4 6 2n 2 2 4 6 2n

Since
1 1 1 1 1 1 1 1
> , > , > , ... , >
2 4 2 6 2 8 2 2n

then
n 1 1 1 1 1 1 1 1
= + + + ... + > + + + ... +
2 |2 2 2{z 2} 2 4 6 2n
n

so µ ¶
1 1 1 1 1 1
> + + + ... + . (3)
2 n 2 4 6 2n

Then (1), (2) and (3) show

µ ¶ µ ¶
1 1 1 1 1 1 1 1 1 1 1
1 + + ... + > + + + ... + + + + + ... +
3 2n − 1 n 2 4 6 2n 2 4 6 2n
µ ¶µ ¶
1 1 1 1
= 1+ + + ... +
n 2 4 2n
µ ¶
n+1 1 1 1
= + + ... + .
n 2 4 2n

µ ¶ µ ¶
1 1 1 1 1 1 1
Therefore 1 + + ... + > + + ... + for all n ∈ N and n ≥ 2.
n+1 3 2n − 1 n 2 4 2n

3
Solution 2 to Problem 3 – Yin Lei, Vincent Massey S.S., Windsor, ON

Since n ≥ 2, n(n + 1) ≥ 0. Therefore the given inequality is equivalent to

µ ¶ µ ¶
1 1 1 1 1
n 1+ + ... + ≥ (n + 1) + + ... + .
3 2n − 1 2 4 2n

We shall use mathematical induction to prove this.

µ ¶ µ ¶
1 1 4 1 1 1 3
For n = 2, obviously 1+ = > + = .
3 3 9 2 2 4 8
Suppose that the inequality stands for n = k, i.e.

µ ¶ µ ¶
1 1 1 1 1
k 1+ + ... + > (k + 1) + + ... + . (1)
3 2k − 1 2 4 2k

Now we have to prove it for n = k + 1.

We know

µ ¶ µ ¶
1 1 1 1 1
1+ + ... + − + + ... +
3 2k − 1 2 4 2k
µ ¶ µ ¶ µ ¶ µ ¶
1 1 1 1 1 1 1
= 1− + − + − + ... + −
2 3 4 5 6 2k − 1 2k
1 1 1 1
= + + + ... + .
1×2 3×4 5×6 (2k − 1)(2k)

Since

1 × 2 < 3 × 4 < 5 × 6 < . . . < (2k − 1)(2k) < (2k + 1)(2k + 2)

then
1 1 1 k
+ + ... + >
1×2 3×4 (2k − 1)(2k) (2k + 1)(2k + 2)

hence
1 1 1 1 1 k
1+ + ... + > + + ... + + . (2)
3 2k − 1 2 4 2k (2k + 1)(2k + 2)

Also
k+1 k+2 2k 2 + 2k + 2k + 2 − 2k 2 − 4k − k − 2 k
− = =−
2k + 1 2k + 2 (2k + 1)(2k + 2) (2k + 1)(2k + 2)

therefore
k+1 k+2 k
= − . (3)
2k + 1 2k + 2 (2k + 1)(2k + 2)

4
Adding 1, 2 and 3:
µ ¶ µ ¶
1 1 1 1 k+1
k 1+ + ... + + 1+ + ... + +
3 2k − 1 3 2k − 1 2k + 1
µ ¶ µ ¶
1 1 1 1 1 1 k k+2 k
> (k + 1) + + ... + + + + ... + + + −
2 4 2k 2 4 2k (2k + 1)(2k + 2) 2k + 2 (2k + 1)(2k + 2)
Rearrange both sides to get

µ ¶ µ ¶
1 1 1 1 1
(k + 1) 1 + + . . . + > (k + 2) + + ... + .
3 2k + 1 2 4 2k + 2

Proving the induction.

5
Solution 1 to Problem 4 – Keon Choi, A.Y. Jackson S.S., North York, ON

Suppose H is the foot of the perpendicular line from A to BC; construct equilateral 4ABG, with
C on BG. I will prove that if F is the point where AH meets BD, then 6 F CB = 70◦ . (Because
that means AH, and the given lines BD and CE meet at one point and that proves the question.)
Suppose BD extended meets AG at I.
A

E
I
F D

B H C G

Now BF = GF and 6 F BG = 6 F GB = 40◦ so that 6 IGF = 20◦ . Also 6 IF G = 6 F BG + 6 F GB =

80◦ , so that

6 F IG = 180◦ − 6 IF G − 6 IGF
= 180◦ − 80◦ − 20◦
= 80◦ .

Therefore 4GIF is an isoceles triangle, so

GI = GF = BF. (1)

But 4BGI and 4ABC are congruent, since BG = AB, 6 GBI = 6 BAC, 6 BGI = 6 ABC.

Therefore

GI = BC. (2)

From (1) and (2) we get

BC = BF.

So in 4BCF ,

180◦ − 6 F BC 180◦ − 40◦

6 BCF = = = 70◦ .
2 2

Thus 6 F CB = 70◦ and that proves that the given lines CE and BD and the perpendicular line
AH meet at one point.

6
Solution 2 to Problem 4 – Adrian Birka, Lakeshore Catholic H.S., Port Colborne, ON

First we prove the following lemma:

In 4ABC, AA0 , BB 0 , CC 0 intersect if-f

sin α1 sin β1 sin γ1

· · = 1,
sin α2 sin β2 sin γ2

where α1 , α2 , β1 , β2 , γ1 , γ2 are as shown in the diagram just below.

(Editor: This is a known variant of Ceva’s Theorem.)

B

a2
c1 β1 β2

A’
C’
D
c2 a1
α2
γ1
α1 γ2
A b1 B’ b2 C

Proof: Let 6 BB 0 C = x, then 6 BB 0 A = 180◦ − x. Using the sine law in 4BB 0 C yields

b2 a
= . (1)
sin β2 sin x

Similarly using the sine law in 4BB 0 A yields

b1 c c
= ◦
= . (2)
sin β1 sin(180 − x) sin x

Hence,

c sin β1
b1 : b2 = (3)
a sin β2

(from (1),(2)). (Editor: Do you recognize this when β1 = β2 ?)

Similarly,

b sin α1 a sin γ1
a1 : a2 = , c1 : c 2 = . (4)
c sin α2 b sin γ2

By Ceva’s theorem, the necessary and sufficient condition for AA0 , BB 0 , CC 0 to intersect is:
(a1 : a2 ) · (b1 : b2 ) · (c1 : c2 ) = 1. Using (3), (4) on this yields:

b sin α1 a sin γ1 c sin β1

· · · · · =1
c sin α2 b sin γ2 a sin β2

7
so

sin α1 sin β1 sin γ1

· · = 1. (5)
sin α2 sin β2 sin γ2

This is just what we needed to show, therefore the lemma is proved.

Now, in our original question, give 6 BAC = 40◦ , 6 ABC = 60◦ . It follows that 6 ACB = 80◦ .

Since 6 CBD = 40◦ , 6 ABD = 6 ABC − 6 DBC = 20◦ . Similarly, 6 ECA = 20◦ .
B

E K
F

A D C

Now let us show that 6 F AD = 10◦ . Suppose otherwise. Let F 0 be such that F, F 0 are in the same
side of AC and 6 DAF 0 = 10◦ . Then 6 BAF 0 = 6 BAC − 6 DAF 0 = 30◦ .

Thus

sin 6 ABD sin 6 BCE sin 6 CAF 0 sin 20◦ sin 70◦ sin 10◦
· · = · ·
sin 6 DBC sin 6 ECA sin 6 F 0 AB sin 40◦ sin 10◦ sin 30◦
sin 20◦ cos 20◦
= ·
2 sin 20◦ cos 20◦ sin 30◦
1
= = 1.
2 sin 30◦

By the lemma above, AF 0 passes through CE ∩ BD = F . Therefore AF 0 = AF , and 6 F AD = 10◦ ,

contrary to assumption. Thus 6 F AD must be 10◦ . Now let AF ∩ BC = K. Since 6 KAC =
10◦ , 6 KCA = 80◦ , it follows that 6 AKC = 90◦ . Therefore AK⊥BC ⇒ AF ⊥BC as needed.

8
Solution to Problem 5 – Adrian Chan, Upper Canada College, Toronto, ON
a2n + a2n+1
Let us first prove by induction that = m2 for all n ≥ 0.
an · an+1 + 1
Proof:
a20 + a21 0 + m2
Base Case (n = 0) : = = m2 .
a0 · a1 + 1 0+1
Now, let us assume that it is true for n = k, k ≥ 0. Then,

a2k + a2k+1
= m2
ak · ak+1 + 1
a2k + a2k+1 = m2 · ak · ak+1 + m2
a2k+1 + m4 a2k+1 − 2m2 · ak · ak+1 + a2k = m2 + m4 a2k+1 − m2 · ak · ak+1
a2k+1 + (m2 ak+1 − ak )2 = m2 + m2 ak+1 (m2 ak+1 − ak )
a2k+1 + a2k+2 = m2 + m2 · ak+1 · ak+2 .

a2k+1 + a2k+2
So = m2 ,
ak+1 · ak+2 + 1
a2 + b2
proving the induction. Hence (an , an+1 ) is a solution to = m2 for all n ≥ 0.
ab + 1

a2 + b2
Now, consider the equation = m2 and suppose (a, b) = (x, y) is a solution with 0 ≤ x ≤ y.
ab + 1
Then

x2 + y 2
= m2 . (1)
xy + 1

If x = 0 then it is easily seen that y = m, so (x, y) = (a0 , a1 ). Since we are given x ≥ 0, suppose
now that x > 0.

Let us show that y ≤ m2 x.

Proof by contradiction: Assume that y > m2 x. Then y = m2 x + k where k ≥ 1.

Substituting into (1) we get

x2 + (m2 x + k)2
= m2
(x)(m2 x + k) + 1
x2 + m4 x2 + 2m2 xk + k 2 = m4 x2 + m2 kx + m2
(x2 + k 2 ) + m2 (kx − 1) = 0.

Now, m2 (kx − 1) ≥ 0 since kx ≥ 1 and x2 + k 2 ≥ x2 + 1 ≥ 1 so (x2 + k 2 ) + m2 (kx − 1) 6= 0.

Thus we have a contradiction, so y ≤ m2 x if x > 0.

9
Now substitute y = m2 x − x1 , where 0 ≤ x1 < m2 x, into (1).

We have
x2 + (m2 x − x1 )2
= m2
x(m2 x − x1 ) + 1
x2 + m4 x2 − 2m2 x · x1 + x21 = m4 x2 − m2 x · x1 + m2
x2 + x21 = m2 (x · x1 + 1)
x2 + x21
= m2 . (2)
x · x1 + 1

If x1 = 0, then x2 = m2 . Hence x = m and (x1 , x) = (0, m) = (a0 , a1 ). But y = m2 x − x1 = a2 , so

(x, y) = (a1 , a2 ). Thus suppose x1 > 0.

Let us now show that x1 < x.

Proof by contradiction: Assume x1 ≥ x.

à !
x2 + y 2
Then m2 x − y ≥ x since y = m2 x − x1 , and x − y ≥ x since
xy + 1
a2 + b2
(x, y) is a solution to = m2 .
ab + 1
So x3 + xy 2 ≥ x2 y + xy 2 + x + y, hence x3 ≥ x2 y + x + y which is a contradiction since y ≥ x > 0.

With the same proof that y ≤ m2 x, we have x ≤ m2 x1 . So the substitution x = m2 x1 − x2 with

x2 ≥ 0 is valid.
x21 + x22
Substituting x = m2 x1 − x2 into (2) gives = m2 .
x1 · x2 + 1
x2j + x2j+1
If x2 6= 0, then we continue with the substitution xi = m2xi+1 −xi+2 (*) until we get = m2
xj · xj+1 + 1
and xj+1 = 0. (The sequence xi is decreasing, nonnegative and integer.)

So, if xj+1 = 0, then x2j = m2 so xj = m and (xj+1 , xj ) = (0, m) = (a0 , a1 ).

Then (xj , xj−1 ) = (a1 , a2 ) since xj−1 = m2 xj − xj+1 (from (*)).

Continuing, we have (x1 , x) = (an−1 , an ) for some n. Then (x, y) = (an , an+1 ).
a2 + b2
Hence = m2 has solutions (a, b) if and only if (a, b) = (an , an+1 ) for some n.
ab + 1

10
 GRADERS’ REPORT

Each question was worth a maximum of 7 marks. Every solution on every paper was graded by two
different markers. If the two marks differed by more than one point, the solution was reconsidered
until the difference resolved. If the two marks differed by one point, the average was used in
computing the total score.

The various grades assigned each solution are displayed below, as a percentage.

MARKS #1 #2 #3 #4 #5

0 7.6 9.8 40.2 31.0 73.9

1 14.1 27.7 7.1 27.7 9.2
2 10.9 16.8 16.8 21.7 12.0
3 6.5 16.3 3.8 1.6 1.1
4 3.3 2.2 1.6 2.2 0.5
5 6.0 14.1 4.3 3.8 0.0
6 16.3 6.0 7.1 2.2 1.1
7 35.3 7.1 19.0 9.8 2.2

PROBLEM 1

This question was well done. 47 students received 6 or 7 and only 6 students received no marks.
Many students came up with a proof similar to David Arthur’s proof. Another common approach
was to find bounds for a (either 0 ≤ a < 60 or 0 ≤ a < 90) and to then check which of these a
satisfy the equation.

PROBLEM 2

Although most students attempted this problem, there were only 6 perfect solutions. A further 6
solutions earned a mark of 6/7 and 13 solutions earned a mark of 5/7.

The most common approach was to square both sides of the equation, rearrange the terms to isolate
the radical, and to then square both sides again. This resulted in the polynomial x6 − 2x5 − x4 +
2x3 + x2 = 0. Many students were unable to factor this polynomial, and so earned only 2 or 3
points.
√ √
1+ 5 1− 5
The polynomial has three distinct roots: 0, , and . Most students recognized that
2 2 √
1− 5
0 is extraneous. One point was deducted for not finding that is extraneous, and a further
√ 2
1+ 5
point was deducted for not checking that is a solution. (It’s not obvious that the equation
2
has any solutions.) Failing to check for extraneous roots is considered to be a major error. The
graders should, perhaps, have deducted more points for this mistake.

The solution included here avoids the 6th degree polynomial, thus avoiding the difficult factoring.

11
However, the solutions must still be checked.

PROBLEM 3

There were 17 perfect solutions and eleven more contestants earned either 5 or 6 points.
1 1 1
The most elegant solution uses two simple observations: that 1 = + and that is greater than
2 2 2
1 1 1
the average of , , . . . , . A telescope argument also works, adding the first and last terms
2 4 2n
from each side, and so on. The key to a successful proof by induction is to be careful with algebra
and to avoid the temptation to use inequalities. For example, many students used the induction
hypothesis to deduce that

µ ¶ µ ¶
1 1 1 n+1 1 1 1
1 + + ... + > 1 + + ... + +
n+2 3 2n + 1 (n + 2)n 2 2n (n + 2)(2n + 1)

n+1 1
then used > , which is too sloppy for a successful induction proof.
(n + 2)n n+1

PROBLEM 4

Many contestants attempted this question, though few got beyond labeling the most apparent
angles. Nine students successfully completed the problem, while another six made a significant
attempt.

Most of these efforts employed trigonometry or coordinates to set up a trigonometric equation for
an unknown angle. This yields to an assault by identities. Adrian Birka produced a very clean
solution of this nature.

Only Keon Choi managed to complete a (very pretty) synthetic solution. One other contestant
made significant progress with the same idea.

PROBLEM 5

Many students were successful in finding the expression for the terms of the sequence {an } by a
variety of methods: producing an explicit formula, by means of a generating function and as a
sum of binomial coefficients involving parameter m. Unfortunately this does not help solving the
problem. Nevertheless seventeen contestants were able to prove by induction that the terms of the
sequence satisfy the required relation.

To prove the ”only if” part one should employ the method of descent which technically is the same
calculation as in the direct part of the problem. Three students succeeded in this, but only two
obtained a complete solution by showing that the sequence constructed by descent is decreasing
and must have m and 0 as the last two terms.

12
 THE 1999 CANADIAN MATHEMATICAL
OLYMPIAD

1. Find all real solutions to the equation 4x2 − 40[x] + 51 = 0.

Here, if x is a real number, then [x] denotes the greatest integer that is less than
or equal to x.

2. Let ABC be an equilateral triangle of altitude 1. A circle with radius 1 and

center on the same side of AB as C rolls along the segment AB. Prove that
the arc of the circle that is inside the triangle always has the same length.

3. Determine all positive integers n with the property that n = (d(n))2 . Here d(n)
denotes the number of positive divisors of n.

4. Suppose a1 , a2 , . . . , a8 are eight distinct integers from {1, 2, . . . , 16, 17}. Show
that there is an integer k > 0 such that the equation ai − aj = k has at least
three different solutions. Also, find a specific set of 7 distinct integers from
{1, 2, . . . , 16, 17} such that the equation ai − aj = k does not have three distinct
solutions for any k > 0.

5. Let x, y, and z be non-negative real numbers satisfying x + y + z = 1. Show

that
4
x2 y + y 2 z + z 2 x ≤ ,
27
and find when equality occurs.
 1999
SOLUTIONS

Most of the solutions to the problems of the 1999 CMO presented below are taken from students’
papers. Some minor editing has been done - unnecessary steps have been eliminated and some
wording has been changed to make the proofs clearer. But for the most part, the proofs are as
submitted.

Solution to Problem 1 – Adrian Chan, Upper Canada College, Toronto, ON

Rearranging the equation we get 4x2 + 51 = 40[x]. It is known that x ≥ [x] > x − 1, so

4x2 + 51 = 40[x] > 40(x − 1)

4x2 − 40x + 91 > 0
(2x − 13)(2x − 7) > 0

Hence x > 13/2 or x < 7/2. Also,

4x2 + 51 = 40[x] ≤ 40x

4x2 − 40x + 51 ≤ 0
(2x − 17)(2x − 3) ≤ 0

Hence 3/2 ≤ x ≤ 17/2. Combining these inequalities gives 3/2 ≤ x < 7/2 or 13/2 < x ≤ 17/2 .
CASE 1: 3/2 ≤ x < 7/2.
For this case, the possible values for [x] are 1, 2 and 3.
If [x] = 1 then 4x2 + 51 = 40 · 1 so 4x2 = −11, which has no real solutions.
√ √ √ √
If [x] = 2 then 4x2 + 51 = 40 · 2 so 4x2 = 29 and x = 229 . Notice that 216 < 229 < 236 so
2 < x < 3 and [x] = 2.
√ √ √
If [x] = 3 then 4x2 + 51 = 40 · 3 and x = 69/2. But 269 > 264 = 4. So, this solution is rejected.
CASE 2: 13/2 < x ≤ 17/2.
For this case, the possible values for [x] are 6, 7 and 8.
√ √ √ √
189 144 189 196
If [x] = 6 then 4x2 + 51 = 40 · 6 so x = 2 . Notice that 2 < 2 < 2 so 6 < x < 7 and
[x] = 6.
√ √ √ √
229 196 229 256
If [x] = 7 then 4x2 + 51 = 40 · 7 so x = 2 . Notice that 2 < 2 < 2 so 7 < x < 8 and
[x] = 7.
√ √ √ √
If [x] = 8 then 4x2 + 51 = 40 · 8 so x = 269
2 . Notice that
256
2 < 269
2 < 324
2 so 8 < x < 9 and
[x] = 8.
√ √ √ √
29 189 229 269
The solutions are x = , , , .
2 2 2 2
(Editor: Adrian then checks these four solutions.)

1
Solution 1 to Problem 2 – Keon Choi, A.Y. Jackson S.S., North York, ON

E
Let D and E be the intersections of BC and extended
AC respectively with the circle.
Since CO||AB (because both the altitude and the ra-
dius are 1) 6 BCO = 60◦ and therefore 6 ECO =
C O
180◦ − 6 ACB − 6 BC0 = 60◦ .
Since a circle is always symmetric in its diameter and
line CE is reflection of line CB in CO, line segment
CE is reflection of line segment CB. D
Therefore CE = CD. A B
Therefore 4CED is an isosceles.
Therefore 6 CED = 6 CDE and 6 CED + 6 CDE = 6 ACB = 60◦ .
6 CED = 30◦ regardless of the position of centre 0. Since 6 CED is also the angle subtended from
the arc inside the triangle, if CED is constant, the arc length is also constant.
Editor’s Note: This proof has had no editing.

2
Solution 2 to Problem 2 – Jimmy Chui, Earl Haig S.S., North York, ON

³ ´
Place C at the origin, point A at √1 , 1 and y
³ ´ 3
point B at − √13 , 1 . Then 4ABC is equilateral B A
with altitude of length 1. A’
B’
Let O be the center of the circle. Because the
circle has radius 1, and since it touches line AB,
the locus of O is on the line through C parallel to
AB (since C is length 1 away from AB), i.e., the x
C O(a,0)
locus of O is on the x-axis.
Let point O be at (a, 0). Then − √13 ≤ a ≤ √1
3
since we have the restriction that the circle rolls
along AB.
Now, let A0 and 0
√ B be the intersection of the circle with
√ CA and CB respectively. The equation
of CA is y = 3 x, 0 ≤ x ≤ √13 , of CB is y = − 3 x, − √13 ≤ x ≤ 0, and of the circle is
(x − a)2 + y 2 = 1.
√
0
√ 2 2 a ± 4 − 3a2
We solve for A by substituting y = 3 x into (x − a) + y = 1 to get x = .
4
Visually, we can see that solutions represent the intersection of AC extended and the circle, but
we are only concerned with the greater x-value – this is the solution that is on AC, not on AC
extended. Therefore
√ Ã √ !
a+ 4 − 3a2 √ a+ 4 − 3a2
x= , y= 3 .
4 4

Likewise we solve for B 0 , but we take the lesser x-value to get

√ Ã √ !
a− 4 − 3a2 √ a+ 4 − 3a2
x= , y=− 3 .
4 4

Let us find the length of A0 B 0 :

à √ √ !2 Ãà √ ! à √ !!2
0 0 2 a+ 4 − 3a2 a − 4 − 3a2 √ a+ 4 − 3a2 √ a− 4 − 3a2
|A B | = − + 3 − − 3
4 4 4 4
4 − 3a2 a2
= +3
4 4
= 1

which is independent of a.
Consider the points 0, A0 and B 0 . 40A0 B 0 is an equilateral triangle (because A0 B 0 = 0A0 = 0B 0 = 1).
π
Therefore 6 A0 0B 0 = 3 and arc A0 B 0 = π3 , a constant.

3
Solution to Problem 3 – Masoud Kamgarpour, Carson S.S., North Vancouver, BC

Note that n = 1 is a solution. For n > 1 write n in the form n = P1α1 P2α2 ...Pm
αm where the P ’s,
i
1 ≤ i ≤ m, are distinct prime numbers and αi > 0. Since d(n) is an integer, n is a perfect square,
so αi = 2βi for integers βi > 0.
Using the formula for the number of divisors of n,

d(n) = (2β1 + 1)(2β2 + 1) . . . (2βm + 1)

which is an odd number. Now because d(n) is odd, (d(n))2 is odd, therefore n is odd as well, so
Pi ≥ 3, 1 ≤ i ≤ m. We get

P1α1 · P2α2 . . . Pm
αm
= [(α1 + 1)(α2 + 1) . . . (αm + 1)]2

or using αi = 2βi
P1β1 P2β2 . . . Pm
βm
= (2β1 + 1)(2β2 + 1) . . . (2βm + 1).

Now we prove a lemma:

Lemma: P t ≥ 2t + 1 for positive integers t and P ≥ 3, with equality only when P=3 and t=1.
Proof: We use mathematical induction on t. The statement is true for t = 1 because P ≥ 3. Now
suppose P k ≥ 2k + 1, k ≥ 1; then we have

P k+1 = P k · P ≥ P k (1 + 2) > P k + 2 ≥ (2k + 1) + 2 = 2(k + 1) + 1

Thus P t ≥ 2t + 1 and equality occurs only when P = 3 and t = 1.

Let’s say n has a prime factor Pk > 3; then (by the lemma) Pkβk > 2βk +1 and we have P1β1 . . . Pm
βm >

(2β1 + 1) . . . (2βm + 1), a contradiction.

Therefore, the only prime factor of n is P = 3 and we have 3α = 2α + 1. By the lemma α = 1.
The only positive integer solutions are 1 and 9.

Solution 1 to Problem 4 – David Nicholson, Fenelon Falls S.S., Fenelon Falls, ON

Without loss of generality let a1 < a2 < a3 . . . < a8 .

Assume that there is no such integer k. Let’s just look at the seven differences di = ai+1 − ai . Then
amongst the di there can be at most two 1s, two 2s, and two 3s, which totals to 12.
Now 16 = 17 − 1 ≥ a8 − a1 = d1 + d2 + . . . + d7 so the seven differences must be 1,1,2,2,3,3,4.
Now let’s think of arranging the differences 1,1,2,2,3,3,4. Note that the sum of consecutive differ-
ences is another difference. (Eg d1 + d2 = a3 − a1 , d1 + d2 + d3 = a4 − a1 )
We can’t place the two 1s side by side because that will give us another difference of 2. The 1s
can’t be beside a 2 because then we have three 3s. They can’t both be beside a 3 because then we
have three 4s! So we must have either 1, 4, −, −, −, 3, 1 or 1, 4, 1, 3, −, −, − (or their reflections).
In either case we have a 3,1 giving a difference of 4 so we can’t put the 2s beside each other. Also we
can’t have 2,3,2 because then (with the 1,4) we will have three 5s. So all cases give a contradiction.
Therefore there will always be three differences equal.
A set of seven numbers satisfying the criteria are {1, 2, 4, 7, 11, 16, 17}. (Editor: There are many
such sets)

4
Solution 2 to Problem 4 – The CMO committee

Consider all the consecutive differences (ie, di above) as well as the differences bi = ai+2 − ai , i =
1 . . . 6. Then the sum of these thirteen differences is 2·(a8 −a1 )+(a7 −a2 ) ≤ 2(17−1)+(16−2) = 46.
Now if no difference occurs more than twice, the smallest the sum of the thirteen differences can
be is 2 · (1 + 2 + 3 + 4 + 5 + 6) + 7 = 49, giving a contradiction.

Solution 1 to Problem 5 – The CMO committee

Let f (x, y, z) = x2 y + y 2 z + z 2 x. We wish to determine where f is maximal. Since f is cyclic,

without loss of generality we may assume that x ≥ y, z. Since

f (x, y, z) − f (x, z, y) = x2 y + y 2 z + z 2 x − x2 z − z 2 y − y 2 x
= (y − z)(x − y)(x − z),

we may also assume y ≥ z. Then

f (x + z, y, 0) − f (x, y, z) = (x + z)2 y − x2 y − y 2 z − z 2 x
= z 2 y + yz(x − y) + xz(y − z) ≥ 0,

so we may now assume z = 0. The rest follows from the arithmetic-geometric mean inequality:
µ ¶3
2x2 y 1 x + x + 2y 4
f (x, y, 0) = ≤ =
2 2 3 27

Equality occurs when x = 2y, hence at (x, y, z) = ( 23 , 13 , 0). (As well as (0, 23 , 13 ) and ( 13 , 0, 23 ).

Solution 2 to Problem 5 – The CMO committee

With f as above, and x ≥ y, z

µ ¶
z z xz z2y z3
f x + , y + , 0 − f (x, y, z) = yz(x − y) + (x − z) + +
2 2 2 4 8
so we may assume that z = 0. The rest follows as for solution 1.

5
 GRADERS’ REPORT
Each question was worth a maximum of 7 marks. Every solution on every paper was graded by two
different markers. If the two marks differed by more than one point, the solution was reconsidered
until the difference resolved. If the two marks differed by one point, the average was used in
computing the total score.
The various grades assigned to each solution are displayed below, as a percentage.

MARKS #1 #2 #3 #4 #5

0 7.6 45.6 18.4 38.6 51.3

1 13.9 15.8 43.0 0.0 32.3
2 12.0 12.7 15.2 41.8 13.9
3 5.7 2.5 5.1 7.6 2.5
4 4.4 1.3 8.9 4.4 0.0
5 8.9 2.5 3.2 5.7 1.3
6 8.9 0.0 1.3 1.9 0.0
7 39.9 20.9 6.3 1.3 0.0

PROBLEM 1 The aim of the question was to give the competitors an encouraging start (it was
not a give away!). Over half of the students had good scores of 5, 6 or 7.
The general approach was to find bounds for x and then to find the exact value for x by substituting
in the resulting possible values of [x]. Depending on how the bounds were determined, this meant
checking 6 - 10 different cases.
Points were lost for not adequately verifying the bounds on x. For example, 2 points were deducted
for assuming, without proof, that 4x2 + 51 > 40[x] for x ≥ 9.

PROBLEM 2 Many competitors saw that the key here is to prove that the angle subtended by
the arc at its centre is constant, namely π/3. In all, 16 students managed a complete proof. Most
attempted an analytic solution – indeed, the problem is nearly routine if one chooses coordinates
wisely and later on notes that two such x-coordinates are roots of the same quadratic. A few
students used trigonometry, namely the law of sines on a couple of useful triangles. Two students
found essentially the same synthetic solution, which is very elegant.

PROBLEM 3 Most competitors determined by direct calculation that n = 1 and n = 9 are

solutions. The difficulty was to show that these are the only solutions, which boils down to proving
that pk ≥ 2k + 1 for all primes p > 2 and all k > 0 with equality only for k = 1 and p = 3. This
can be done by induction or by calculus. Only 5 students obtained perfect marks.

PROBLEM 4
Many students found a specific set of seven integers such that the equation did not have three
different solutions. This earned two points. (One student found such a set with maximum value
14. A maximum value of 13 is not possible.)

6
Only eight competitors received high marks on the question (5, 6, or 7), and only one student scored
a perfect 7. All of the successful solvers considered differences of consecutive integers, showing that
they must be 1, 1, 2, 2, 3, 3, and 4, and then showed that every ordering of these differences led
to at least three repetitions of the same value. Most competitors recognized that the 1s could not
be together, nor could they be beside a 2. They then proceed by considering all such possible
arrangements, which often resulted in close to a dozen cases (depending on how the the cases were
handled.) David Nicholson was the most efficient at pruning the cases. (See Solution 1 to Problem
4.) Most students failed to consider one or two (easily dismissed) cases, hence lost 1 or 2 points.
A number of the contestants attempted to solve the problem by examining the odd-even character
of the set of eight integers, counting how many of the differences were odd or even, and using the
pigeon-hole principle. Although this approach looked promising, no one was able to handle the
case that 3 of the integers were of one parity, and 5 were of the other parity.

PROBLEM 5 No students received full marks for this problem. One student received 5 marks
for a proof that had minor errors. This proof was by Calculus. The committee was aware that
the problem could be solved using Calculus but (erroneously) thought it unlikely high school
students
³ would
´ ³ attempt´ such ³a solution.´ Many students received 1 point for “guessing” that
2 1 2 1 1 2
3 , 3 , 0 , 0, 3 , 3 and 3 , 0 , 3 are where equality occurred. Some students received
a further point for verifying the inequality on the boundary of the region.

7
 THE 2000 CANADIAN MATHEMATICAL
OLYMPIAD

1. At 12:00 noon, Anne, Beth and Carmen begin running laps around a circular
track of length three hundred meters, all starting from the same point on the
track. Each jogger maintains a constant speed in one of the two possible di-
rections for an indefinite period of time. Show that if Anne’s speed is different
from the other two speeds, then at some later time Anne will be at least one
hundred meters from each of the other runners. (Here, distance is measured
along the shorter of the two arcs separating two runners.)

2. A permutation of the integers 1901, 1902, . . . , 2000 is a sequence a1 , a2 , . . . , a100

in which each of those integers appears exactly once. Given such a permutation,
we form the sequence of partial sums

s1 = a1 , s2 = a1 + a2 , s3 = a1 + a2 + a3 , . . . , s100 = a1 + a2 + · · · + a100 .

How many of these permutations will have no terms of the sequence s1 , . . . , s100
divisible by three?

3. Let A = (a1 , a2 , . . . , a2000 ) be a sequence of integers each lying in the inter-

val [−1000, 1000]. Suppose that the entries in A sum to 1. Show that some
nonempty subsequence of A sums to zero.

4. Let ABCD be a convex quadrilateral with

6 CBD = 26 ADB,
6 ABD = 26 CDB
and AB = CB.

Prove that AD = CD.

5. Suppose that the real numbers a1 , a2 , . . . , a100 satisfy

a1 ≥ a2 ≥ · · · ≥ a100 ≥ 0,
a1 + a2 ≤ 100
and a3 + a4 + · · · + a100 ≤ 100.

Determine the maximum possible value of a21 + a22 + · · · + a2100 , and find all
possible sequences a1 , a2 , . . . , a100 which achieve this maximum.
 2000 Canadian Mathematics Olympiad Solutions
Chair: Luis Goddyn, Simon Fraser University, [email protected]

The Year 2000 Canadian Mathematics Olympiad was written on Wednesday April 2, by 98 high
school students across Canada. A correct and well presented solution to any of the ve questions
was awarded seven points. This year’s exam was a somewhat harder than usual, with the mean
score being 8.37 out of 35. The top few scores were: 30, 28, 27, 22, 20, 20, 20. The rst, second and
third prizes are awarded to: Daniel Brox (Sentinel Secondary BC), David Arthur (Upper Canada
College ON), and David Pritchard (Woburn Collegiate Institute ON).

1. At 12:00 noon, Anne, Beth and Carmen begin running laps around a circular track of length
three hundred meters, all starting from the same point on the track. Each jogger maintains
a constant speed in one of the two possible directions for an indenite period of time. Show
that if Anne’s speed is dierent from the other two speeds, then at some later time Anne will
be at least one hundred meters from each of the other runners. (Here, distance is measured
along the shorter of the two arcs separating two runners.)
Comment: We were surprised by the difficulty of this question, having awarded an average
grade of 1.43 out of 7. We present two solutions; only the rst appeared among the graded
papers.
Solution 1: By rotating the frame of reference we may assume that Anne has speed zero, that
Beth runs at least as fast as Carmen, and that Carmen’s speed is positive. If Beth is no more
than twice as fast as Carmen, then both are at least 100 meters from Anne when Carmen has
run 100 meters. If Beth runs more that twice as fast as Carmen, then Beth runs a stretch of
more than 200 meters during the time Carmen runs between 100 and 200 meters. Some part
of this stretch lies more than 100 meters from Anne, at which time both Beth and Carmen
are at least (in fact, more than) 100 meters away from Anne.
Solution 2: By rotating the frame of reference we may assume Anne’s speed to equal zero,
and that the other two runners have non-zero speed. We may assume that Beth is running at
least as fast as Carmen. Suppose that it takes t seconds for Beth to run 200 meters. Then at
all times in the innite set T = {t, 2t, 4t, 8t, . . .}, Beth is exactly 100 meters from Anne. At
time t, Carmen has traveled exactly d meters where 0 < d  200. Let k be the least integer
such that 2k d  100. Then k  0 and 100  2k d  200, so at time 2k t ∈ T both Beth and
Carmen are at least 100 meters from Anne.

2. A permutation of the integers 1901, 1902, . . ., 2000 is a sequence a1, a2, . . . , a100 in which each
of those integers appears exactly once. Given such a permutation, we form the sequence of
partial sums

s1 = a1, s2 = a1 + a2 , s3 = a1 + a2 + a3 , . . . , s100 = a1 + a2 +


+ a100.

How many of these permutations will have no terms of the sequence s1 , . . ., s100 divisible by
three?
Comment: This question was the easiest and most straight forward, with an average grade of
3.07.
Solution: Let {1901, 1902, . . ., 2000} = R0 ∪ R1 ∪ R2 where each integer in Ri is congruent
to i modulo 3. We note that |R0| = |R1| = 33 and |R2| = 34. Each permutation S =
(a1, a2, . . . , a100) can be uniquely specied by describing a sequence S 0 = (a01, a02, . . . , a0100) of
 residues modulo 3 (containing exactly 33 zeros, 33 ones and 34 twos), and three permutations
(one each of R0 , R1 , and R2). Note that the number of permutations of Ri is exactly |Ri|! =
1
2


|Ri|.
The condition on the partial sums of S depends only on the sequence of residues S 0 . In
order to avoid a partial sum divisible by three, the subsequence formed by the 67 ones and
twos in S 0 must equal either 1, 1, 2, 1, 2, . . . , 1, 2 or 2, 2, 1, 2, 1, . . ., 2, 1. Since |R2| = |R1| + 1,
only the second pattern is possible. The 33 zero entries

in S 0 may appear anywhere among
a01, a02, . . . , a0100 provided that a01 6= 0. There are 99 99!
33 = 33! 66! ways to choose which entries in
0 99
0
S equal zero. Thus there are exactly 33 sequences S whose partial sums are not divisible
by three. Therefore the total number of permutations S satisfying this requirement is exactly
!
99 99!
33!
34!

33!
33!
34! = .
33 66!

Incidently, this number equals approximately 4.4
10138.

3. Let A = (a1, a2, . . . , a2000) be a sequence of integers each lying in the interval [−1000, 1000].
Suppose that the entries in A sum to 1. Show that some nonempty subsequence of A sums
to zero.
Comment: This students found this question to be the most difficult, with an average grade
of 0.51, and only one perfect solution among 100 papers.
Solution: We may assume no entry of A is zero, for otherwise we are done. We sort A into
a new list B = (b1, . . . , b2000) by selecting elements from A one at a time in such a way that
b1 > 0, b2 < 0 and, for each i = 2, 3, . . ., 2000, the sign of bi is opposite to that of the partial
sum
si−1 = b1 + b2 +


+ bi−1.
(We can assume that each si−i 6= 0 for otherwise we are done.) At each step of the selection
process a candidate for bi is guaranteed to exist, since the condition a1 + a2 +


+ a2000 = 1
implies that the sum of unselected entries in A is either zero or has sign opposite to si−1 .
From the way they were dened, each of s1 , s2, . . . , s2000 is one of the 1999 nonzero integers
in the interval [−999, 1000]. By the Pigeon Hole Principle, sj = sk for some j, k satisfying
1  j < k  2000. Thus bj+1 + bj+2 +


+ bk = 0 and we are done.

4. Let ABCD be a convex quadrilateral with

\CBD = 2\ADB,
\ABD = 2\CDB
and AB = CB.

Prove that AD = CD.

Comment: There are several dierent solutions to this, including some using purely trigono-
metric arguments (involving the law of sines and standard angle sum formulas). We present
here two prettier geometric arguments (with diagrams). The rst solution is perhaps the more
attractive of the two. Average grade: 1.84 out of 7.
Solution 1 (from contestant Keon Choi): Extend DB to a point P on the circle through A
and C centered at B. Then \CP D = 12 \CBD = \ADB and \AP D = 12 \ABD = \CDB,

2
 so AP CD is a parallelogram. Now P D bisects AC so BD is an angle bisector of isosceles
triangle ABC. We have
1 1
\ADB = \CBD = \ABD = \CDB
2 2
so DB is the angle bisector of \ADC. As DB bisects the base of triangle ADC, this triangle
must be isosceles and AD = CD.

Solution 2: Let the bisector of \ABD meet AD at E. Let the bisector of \CBD meet CD
at F . Then \F BD = \BDE and \EBD = \BDF , which imply BE k F D and BF k ED.
Thus BEDF is a parallelogram whence

BD intersects EF at its midpoint M . (1)

AB AE
On the other hand since BE is an angle bisector, we have BD = ED . Similarly we have
CB CF AE CF
BD = F D . By assumption AB = CB so ED = F D which implies EF k AC. Thus 4DEF
and 4DAC are similar, which implies by (1) that BD intersects AC at its midpoint N . Since
4ABC is isosceles, this implies AC ⊥ BD. Thus 4N AD and 4N CD are right triangles
with equal legs and hence are congruent. Thus AD = CD.
P

B B

A C A C
N

E F E M F

D D
Diagram for Solution 1 Diagram for Solution 2

5. Suppose that the real numbers a1, a2, . . . , a100 satisfy

a1  a2 


 a100  0

a1 + a2  100
a3 + a4 +


+ a100  100.
Determine the maximum possible value of a21 + a22 +


+ a2100 , and nd all possible sequences
a1, a2, . . . , a100 which achieve this maximum.
Comment: All of the correct solutions involved a sequence of adjustments to the variables,
each of which increase a21 + a22 +


+ a2100 while satisfying the constraints, eventually arriving
at the two optimal sequences: 100, 0, 0, . . ., 0 and 50, 50, 50, 50, 0, 0, . . . , 0. We present here a
sharper proof, which might be arrived at after guessing that the optimal value is 1002. Average
grade: 1.52 out of 7.

3
Solution: We have a1 + a2 +


+ a100  200, so

a21 + a22 +


+ a2100  (100 − a2)2 + a22 + a23 +


+ a2100

= 1002 − 200a2 + 2a22 + a23 +


+ a2100
 1002 − (a1 + a2 +


+ a100)a2 + 2a22 + a23 +


+ a2100
= 1002 + (a22 − a1a2 ) + (a23 − a3 a2 ) + (a24 − a4 a2 ) +


+ (a2100 − a100a2)
= 1002 + (a2 − a1)a2 + (a3 − a2 )a3 + (a4 − a2 )a4 +


+ (a100 − a2)a100

Since a1  a2 


 a100  0, none of the terms (ai − aj )ai is positive. Thus a21 + a22 +


+
a2100  10, 000 with equality holding if and only if

a1 = 100 − a2 and a1 + a2 +


+ a100 = 200

and each of the products

(a2 − a1 )a2 , (a3 − a2 )a3, (a4 − a2)a4 ,


, (a100 − a2 )a100

equals zero. Since a1  a2  a3 


 a100  0, the last condition holds if and only if for
some i  1 we have a1 = a2 =


= ai and ai+1 =


= a100 = 0. If i = 1, then we get the
solution 100, 0, 0, . . ., 0. If i  2, then from a1 + a2 = 100, we get that i = 4 and the second
optimal solution 50, 50, 50, 50, 0, 0, . . ., 0.

4
THE 2001 CANADIAN MATHEMATICAL OLYMPIAD

1. Randy: “Hi Rachel, that’s an interesting quadratic equation you have written down. What
are its roots?”
Rachel: “The roots are two positive integers. One of the roots is my age, and the other root
is the age of my younger brother, Jimmy.”
Randy: “That is very neat! Let me see if I can figure out how old you and Jimmy are. That
shouldn’t be too difficult since all of your coefficients are integers. By the way, I notice that
the sum of the three coefficients is a prime number.”
Rachel: “Interesting. Now figure out how old I am.”
Randy: “Instead, I will guess your age and substitute it for x in your quadratic equation
. . . darn, that gives me −55, and not 0.”
Rachel: “Oh, leave me alone!”
(a) Prove that Jimmy is two years old.
(b) Determine Rachel’s age.
2. There is a board numbered −10 to 10 as shown. Each square is coloured either red or white,
and the sum of the numbers on the red squares is n. Maureen starts with a token on the square
labeled 0. She then tosses a fair coin ten times. Every time she flips heads, she moves the token
one square to the right. Every time she flips tails, she moves the token one square to the left.
At the end of the ten flips, the probability that the token finishes on a red square is a rational
number of the form ab . Given that a + b = 2001, determine the largest possible value for n.
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10

3. Let ABC be a triangle with AC > AB. Let P be the intersection point of the perpen-
dicular bisector of BC and the internal angle bisector of 6 A. Construct points X on AB
(extended) and Y on AC such that P X is perpendicular to AB and P Y is perpendicular
to AC. Let Z be the intersection point of XY and BC. Determine the value of BZ/ZC.
C

M
P

A X
B

4. Let n be a positive integer. Nancy is given a rectangular table in which each entry is a positive
integer. She is permitted to make either of the following two moves:
 (a) select a row and multiply each entry in this row by n.
(b) select a column and subtract n from each entry in this column.

Find all possible values of n for which the following statement is true:

Given any rectangular table, it is possible for Nancy to perform a finite sequence of
moves to create a table in which each entry is 0.

5. Let P0 , P1 , P2 be three points on the circumference of a circle with radius 1, where P1 P2 =

t < 2. For each i ≥ 3, define Pi to be the centre of the circumcircle of 4Pi−1 Pi−2 Pi−3 .

(a) Prove that the points P1 , P5 , P9 , P13 , . . . are collinear.

(b) Let x be the distance from P1 to P1001
q , and let y be the distance from P1001 to P2001 .
Determine all values of t for which 500
x/y is an integer.
 2001
SOLUTIONS

Several solutions are edited versions of solutions submitted by the contestants whose names appear
in italics..

1. (Daniel Brox)
Let R be Rachel’s age, and let J be Jimmy’s age. Rachel’s quadratic is

a(x − R)(x − J) = ax2 − a(R + J)x + aRJ.

for some number a. We are given that the coefficient a is an integer. The sum of the
coefficients is
a − a(R + J) + aRJ = a(R − 1)(J − 1).
Since this is a prime number, two of the three integers a, R − 1, J − 1 multiply to 1. We are
given that R > J > 0, so we must have that a = 1, J = 2, R − 1 is prime, and the quadratic
is
(x − R)(x − 2).
We are told that this quadratic takes the value −55 = −5 · 11 for some positive integer x.
Since R > 2, the first factor, (x − R), must be the negative one. We have four cases:
x − R = −55 and x − 2 = 1, which implies x = 3, R = 58.
x − R = −11 and x − 2 = 5, which implies x = 7, R = 18.
x − R = −5 and x − 2 = 11, which implies x = 13, R = 18.
x − R = −1 and x − 2 = 53, which implies x = 57, R = 58.
Since R − 1 is prime, the first and last cases are rejected, so R = 18 and J = 2.

2. (Lino Demasi )
After ten coin flips, the token finishes on the square numbered 2k − 10, where k is the number
¡ ¢
of heads obtained. Of the 210 = 1024 possible results of ten coin flips, there are exactly 10 k
ways to¡ obtain exactly k heads, so the probability of finishing on the square labeled 2k − 10
10¢
equals k /1024.
The probability of landing on a red square equals c/1024 where c is the sum of a selection of
the numbers from the list
à ! à ! à ! à !
10 10 10 10
, , ,..., = 1, 10, 45, 120, 210, 252, 210, 120, 45, 10, 1. (1)
0 1 2 10

We are given that for some integers a, b satisfying a + b = 2001,

a/b = c/1024.

If we assume (as most contestants did!) that a and b are relatively prime, then the solution
proceeds as follows. Since 0 ≤ a/b ≤ 1 and a + b = 2001, we have 1001 ≤ b ≤ 2001. Also b
divides 1024, so we have b = 1024. Thus a = c = 2001 − 1024 = 977. There is only one way
to select terms from (1) so that the sum equals 977.

977 = 10 + 10 + 45 + 120 + 120 + 210 + 210 + 252. (2)

(This is easy to check, since the remaining terms in (1) must add to 1024 − 977 = 47, and
47 = 45 + 1 + 1 is the only possibility for this.)

1
In order to maximize n, we must colour the strip as¡ follows. Odd numbered squares are
10¢
red if positive, and white if negative. Since 252 = 5 is in the sum, the square labeled
¡ ¢
2 · 5 − 10 = 0 is red. For k = 0, 1, 2, 3, 4, if 10
k appears twice in the sum (2), then both 2k − 10
¡ ¢
and 10 − 2k are coloured red. If 10 does not appear in the sum, then both 2k − 10 and
¡10¢ k
10 − 2k are coloured white. If k appears once in the sum, then 10 − 2k is red and 2k − 10 is
white. Thus the maximum value of n is obtained when the red squares are those numbered
{1, 3, 5, 7, 9, −8, 8, −4, 4, −2, 2, 0, 6} giving n = 31.

(Alternatively) If we do not assume a and b are relatively prime, then there are several more
possibilities to consider. The greatest common divisor of a and b divides a + b = 2001, so
gcd(a, b) is one of
1, 3, 23, 29, 3 · 23, 3 · 29, 23 · 29, 3 · 23 · 29.
Since a/b = c/1024, dividing b by gcd(a, b) results in a power of 2. Thus the prime factoriza-
tion of b is one of the following, for some integer k.

2k , 3 · 2k , 23 · 2k , 29 · 2k , 69 · 2k , 87 · 2k , 667 · 2k , 2001.
Again we have 1001 ≤ b ≤ 2001, so b must be one of the following numbers.
1024, 3 · 512 = 1536, 23 · 64 = 1472, 29 · 64 = 1856, 69 · 16 = 1104, 87 · 16 = 1392, 667 · 2 = 1334, 2001.

Thus a/b = (2001 − b)/b is one of the following fractions.

977 465 529 145 897 609 667 0
, , , , , , , .
1024 1536 1472 1856 1104 1392 1334 2001
Thus c = 1024a/b is one of the following integers.

977, 310, 368, 80, 832, 448, 512, 0.

After some (rather tedious) checking, one finds that only the following sums with terms from
(1) can add to a possible value of c.

977 = 10 + 10 + 45 + 120 + 120 + 210 + 210 + 252

310 = 10 + 45 + 45 + 210
512 = 10 + 10 + 120 + 120 + 252
512 = 1 + 1 + 45 + 45 + 210 + 210
0 = 0.

Again only those terms appearing exactly once in a sum can affect maximum value of n. We
make the following table.

c Terms appearing once in sum Corresponding red squares

977 {45, 252} {6, 0}
310 {10, 210} {8, 2}
512 {252} {0}
512 ∅ ∅
0 ∅ ∅
¡ ¢ ¡ ¢ ¡ ¢ ¡ ¢
Evidently, the maximum possible value of n is obtained when c = 310 = 10 10 10
2 + 6 + 8 + 9 ,
10

the red squares are {1, 3, 5, 7, 9, −6, 6, 2, 8}, the probability of landing on a red square is
a/b = 465/1536 = 310/1024 = 155/512, and n = 35.

2
3. Solution 1: (Daniel Brox)
Set O be the centre of the circumcircle of 4ABC. Let the angle bisector of 6 BAC meet this
circumcircle at R. We have

6 BOR = 26 BAR = 26 CAR = 6 COR

Thus BR = CR and R lies on the perpendicular bisector of BC. Thus R = P and ABCP
are concyclic. The points X, Y , M are the bases of the three perpendiculars dropped from P
onto the sides of 4ABC. Thus by Simson’s rule, X, Y, M are collinear. Thus we have M = Z
and BZ/ZC = BM/M C = 1.
Note: XY Z is called a Simson line, Wallace line or pedal line for 4ABC. To prove Simson’s
rule, we note that BM P X are concyclic, as are AY P X, thus

6 BXM = 6 BP M = 90 − 6 P BC = 90 − 6 P AC = 6 AP Y = 6 AXY

Solution 2: (Kenneth Ho)

Since 6 P AX = 6 P AY and 6 P XA = 6 P Y A = 90, triangles 4P AX and 4P AY are con-
gruent, so AX = AY and P X = P Y . As P is on the perpendicular bisector of BC, we have
P C = P B. Thus 4P Y C and 4P XB are congruent right triangles, which implies CY = BX.
Since X, Y and Z are collinear, we have by Menelaus’ Theorem
AY CZ BX
= −1.
Y C ZB XA
Applying AX = AY and CY = BX, this is equivalent to BZ/ZC = 1.

4. We shall # the only solution is n = 2. First we show that if n 6= 2, then the table
" see that
1
T0 = can not be changed into a table containing two zeros. For n = 1, this is
n−1
" #
a
very easy to see. Suppose n ≥ 3. For any table T = , let d(T ) be the quantity b − a
b
(mod n − 1). We shall show that neither of the two permitted moves can change the value of
d(T ). If we subtract n from both elements in T , then b − a does not change. If we multiply
the first row by n, then the element a changes to na, for a difference of (n − 1)a, which is
congruent to 0 (mod n − 1). Similarly, multiplying the second row by n does not change
d(T ). Since d(T0 ) = (n − 1) − 1 ≡ −1 (mod n − 1), we can never obtain the table with two
zeros by starting with T0 , because 0 − 0 is not congruent to −1 modulo n − 1.
For n = 2, and any table of positive integers, the following procedure will always result in a
table of zeros. We shall begin by converting the first column into a column of zeros as follows.
We repeatedly subtract 2 from all entries in the first column until at least one of the entries
equals either 1 or 2. Now we repeat the following sequence of three steps:

(a) multiply by 2 all rows with 1 in the first column

(b) now multiply by 2 every row having a 2 in the first column (there is at least one such
row)
(c) subtract 2 from all entries in the first column.

Each iteration of the three steps decreases the sum of those entries in the first column which
are greater than 2. Thus the first column eventually consists entirely of ones and twos, at
which time we apply (a) and (c) once again to obtain a column of zeros. We now repeat the

3
 above procedure for each successive column of the table. The procedure does not affect any
column which has already been set to zero, so we eventually obtain a table with all entries
zero.

5. (Daniel Brox)
Let 6 P1 P3 P2 = 2α. As 4P1 P2 P3 is isosceles, we have that

t = P1 P2 = 2 sin α.

The line P3 P4 is the perpendicular bisector of P1 P2 . Since 4P2 P3 P4 is isosceles, we calculate

its length,
P2 P3 /2 1
P3 P4 = = .
cos α 2 cos α
As P5 is the circumcentre of 4P2 P3 P4 , we have 6 P3 P5 P4 = 26 P3 P2 P4 = 26 P2 P3 P4 = 2α.
The isosceles triangle 4P3 P4 P5 is therefore similar to 4P1 P2 P3 . As P3 P4 ⊥ P1 P2 , we have
6 P1 P3 P5 = 90. Furthermore, the ratio P3 P5 : P1 P3 equals r where

P3 P4 1 1
r= = = .
P1 P2 (2 sin α)(2 cos α) 2 sin(2α)

P3 P0

P2
P1

P5

P4

By the same argument, we see that each 6 Pi Pi+2 Pi+4 is a right angle with Pi+2 Pi+4 : Pi Pi+2 =
r. Thus the points P1 , P3 , P5 , . . . lie on a logarithmic spiral of ratio r and period four as shown
below. It follows that P1 , P5 , P9 , . . . are collinear, proving part (a).

4
 P3

1
r
P1 P11
P9
P13

r3 P5
r2
P7

By the self-similarity of the spiral, we have that P1 P1001 = r500 P1001 P2001 , so
q
500
x/y = 1/r = 2 sin(2α).

This is an integer when sin(2α) ∈ p {0, ±1/2, ±1}. Since 0 < α < 90, this is equiva-
lent to α ∈ {15, 45, 75}. Thus 500 x/y is an integer exactly when t belongs to the set
{2 sin 15, 2 sin 45, 2 sin 75}. This answers part (b).

5
 GRADERS’ REPORT

Eighty four of the eighty five eligible students submitted an examination paper. Each paper con-
tained proposed solutions the some or all of the five examination questions. Each correct and well
presented solution was awarded seven marks for a maximum total score of 35. The mean score was
10.8/35. The top three scores were 28, 27, and 22, thus special scrutiny was required to separate
the top two papers.
Each solution was independently marked by two graders. If the two marks differed, then the
solution was reconsidered until the difference was resolved. The top twenty papers were then
carefully regraded by the chair to ensure that nothing was amiss.
The grade distribution and average mark for each question appears in the following table. For
example, 13.1% of students were awarded 3 marks for question #1.

Marks #1 #2 #3 #4 #5

0 10.7 11.9 45.2 60.7 90.5

1 8.3 8.3 26.2 10.7 3.6
2 8.3 6.0 4.8 8.3 1.2
3 13.1 9.5 3.6 9.5 1.2
4 10.7 17.9 0.0 4.8 1.2
5 9.5 32.1 3.6 2.4 2.4
6 20.2 13.1 0.0 1.2 0.0
7 19.0 1.2 16.7 2.4 0.0
Ave.
4.05 3.64 1.79 1.09 .26
Mark

PROBLEM 1 Ninety five percent of students found the correct solution, although a surprising
number arrived at a solution through trial and error or by guessing and verifying a solution. Many
assumed without proof that the leading coefficient equals one, which resulted in a two-point penalty.
Another common error was not to consider all four possibilities for the pair (x − R), (x − 2).

PROBLEM 2 There was a flaw in question 2. The proposers intended that the integers a, b be
relatively prime. This was made explicit in an early draft, but somehow was lost with the ambiguous
phrase “of the form a/b.” Without this assumption, the problem is much more tedious to solve.
Remarkably, one student (Lino Demasi) considered more (but not all) possible values for gcd(a, b)
and obtained the correct solution n = 35. All other students assumed implicitly (and in two cases,
explicitly) that gcd(a, b) = 1. Solutions to both problems are presented in this publication.

PROBLEM 3 Most students either completely solved or were baffled by this basic geometry
problem. There were at least four types of solutions: one trigonometric, one using basic geometry,
and two which refer to standard theorems relating to the triangle. The first two tended to be
lengthy or cumbersome, and the last two are presented here. There were complaints from some
participants regarding the inaccurate angles appearing in the diagram supplied with the question.
The inaccuracy was intensional, since the key observation M = Z would have otherwise been given

6
away. Unfortunately, this caused some students to doubt their own proofs that BZ : ZC = 1; as
the ratio appears to be closer to 2 in the misleading diagram!

PROBLEM 4 This problem was left unanswered by about 60% of students. Several solutions
consisted only of a proof that n = 1 is not possible. About 25% described a procedure which works
when n = 2. Indeed the procedure for n = 2 seems to be unique. About 10% proved that for
no other value of n was possible, and all of the proofs explicitly or implicitly involved considering
residues modulo n − 1.

PROBLEM 5 This problem proved to be very difficult. Only two students completely answered
part (a), and no students correctly answered part (b). Of the students receiving more than 0 marks,
only two were were not among the top 15 This suggests that the question effectively resolved the
ranking of the strongest participants, which is arguably the purpose of Problem 5.

7
THE 2002 CANADIAN MATHEMATICAL OLYMPIAD

1. Let S be a subset of {1, 2, . . . , 9}, such that the sums formed by adding each unordered pair of
distinct numbers from S are all different. For example, the subset {1, 2, 3, 5} has this property,
but {1, 2, 3, 4, 5} does not, since the pairs {1, 4} and {2, 3} have the same sum, namely 5.
What is the maximum number of elements that S can contain?

2. Call a positive integer n practical if every positive integer less than or equal to n can be
written as the sum of distinct divisors of n.
For example, the divisors of 6 are 1, 2, 3, and 6. Since
1=1, 2=2, 3=3, 4=1+3, 5=2+ 3, 6=6,
we see that 6 is practical.
Prove that the product of two practical numbers is also practical.

3. Prove that for all positive real numbers a, b, and c,

a3 b3 c3
+ + ≥ a + b + c,
bc ca ab
and determine when equality occurs.
√
4. Let Γ be a circle with radius r. Let A and B be distinct points on Γ such that AB < 3r.
Let the circle with centre B and radius AB meet Γ again at C. Let P be the point inside Γ
such that triangle ABP is equilateral. Finally, let the line CP meet Γ again at Q.
Prove that P Q = r.

5. Let N = {0, 1, 2, . . .}. Determine all functions f : N → N such that

xf (y) + yf (x) = (x + y)f (x2 + y 2 )

for all x and y in N .

 2002 Canadian Mathematical Olympiad
Solutions
1. Let S be a subset of {1, 2, . . . , 9}, such that the sums formed by adding each unordered pair of
distinct numbers from S are all different. For example, the subset {1, 2, 3, 5} has this property,
but {1, 2, 3, 4, 5} does not, since the pairs {1, 4} and {2, 3} have the same sum, namely 5.
What is the maximum number of elements that S can contain?

Solution 1
It can be checked that all the sums of pairs for the set {1, 2, 3, 5, 8} are different.
Suppose, for a contradiction, that S is a subset of {1, . . . , 9} containing 6 elements such that
all the sums of pairs are different. Now the smallest possible sum for two numbers from S is
1 + 2 = 3 and the
µ largest
¶ possible sum is 8 + 9 = 17. That gives 15 possible sums: 3, . . . , 17.
6
Also there are = 15 pairs from S. Thus, each of 3, . . . , 17 is the sum of exactly one
2
pair. The only pair from {1, . . . , 9} that adds to 3 is {1, 2} and to 17 is {8, 9}. Thus 1, 2, 8, 9
are in S. But then 1 + 9 = 2 + 8, giving a contradiction. It follows that the maximum number
of elements that S can contain is 5.

Solution 2.
It can be checked that all the sums of pairs for the set {1, 2, 3, 5, 8} are different.
Suppose, for a contradiction, that S is a subset of {1, . . . 9} such that all the sums of pairs
are different and that a1 < a2 < . . . < a6 are the members of S.
Since a1 + a6 6= a2 + a5 , it follows that a6 − a5 6= a2 − a1 . Similarly a6 − a5 6= a4 − a3 and
a4 − a3 6= a2 − a1 . These three differences must be distinct positive integers, so,

(a6 − a5 ) + (a4 − a3 ) + (a2 − a1 ) ≥ 1 + 2 + 3 = 6 .

Similarly a3 − a2 6= a5 − a4 , so

(a3 − a2 ) + (a5 − a4 ) ≥ 1 + 2 = 3 .

Adding the above 2 inequalities yields

a6 − a5 + a5 − a4 + a4 − a3 + a3 − a2 + a2 − a1 ≥ 6 + 3 = 9 ,

and hence a6 − a1 ≥ 9. This is impossible since the numbers in S are between 1 and 9.
2. Call a positive integer n practical if every positive integer less than or equal to n can be
written as the sum of distinct divisors of n.
For example, the divisors of 6 are 1, 2, 3, and 6 . Since
1=1, 2=2, 3=3, 4=1+3, 5=2+ 3, 6=6,
we see that 6 is practical.
Prove that the product of two practical numbers is also practical.

Solution
Let p and q be practical. For any k ≤ pq, we can write

k = aq + b with 0 ≤ a ≤ p, 0 ≤ b < q.

Since p and q are practical, we can write

a = c 1 + . . . + cm , b = d 1 + . . . + d n

where the ci ’s are distinct divisors of p and the dj ’s are distinct divisors of q. Now

k = (c1 + . . . + cm )q + (d1 + . . . + dn )
= c1 q + . . . + cm q + d1 + . . . + dn .

Each of ci q and dj divides pq. Since dj < q ≤ ci q for any i, j, the ci q’s and dj ’s are all distinct,
and we conclude that pq is practical.
3. Prove that for all positive real numbers a, b, and c,
a3 b3 c3
+ + ≥ a + b + c,
bc ca ab
and determine when equality occurs.

Each of the inequalities used in the solutions below has the property that equality holds if
and only if a = b = c. Thus equality holds for the given inequality if and only if a = b = c.
Solution 1.
(a4 + b4 ) (b4 + c4 ) (c4 + a4 )
Note that a4 + b4 + c4 = + + . Applying the arithmetic-geometric
2 2 2
mean inequality to each term, we see that the right side is greater than or equal to

a2 b2 + b2 c2 + c2 a2 .

We can rewrite this as

a2 (b2 + c2 ) b2 (c2 + a2 ) c2 (a2 + b2 )
+ + .
2 2 2
Applying the arithmetic mean-geometric mean inequality again we obtain a4 + b4 + c4 ≥
a2 bc + b2 ca + c2 ab. Dividing both sides by abc (which is positive) the result follows.

Solution 2.
Notice the inequality is homogeneous. That is, if a, b, c are replaced by ka, kb, kc, k > 0 we
get the original inequality. Thus we can assume, without loss of generality, that abc = 1.
Then
µ 3 ¶
a3 b3 c3 a b3 c3
+ + = abc + +
bc ca ab bc ca ab
= a4 + b4 + c4 .
So we need prove that a4 + b4 + c4 ≥ a + b + c.
By the Power Mean Inequality,
µ ¶4
a4 + b4 + c4 a+b+c
≥ ,
3 3

(a + b + c)3
so a4 + b4 + c4 ≥ (a + b + c) · .
27
a+b+c √ 3
By the arithmetic mean-geometric mean inequality, ≥ abc = 1, so a + b + c ≥ 3.
3
(a + b + c)3 33
Hence, a4 + b4 + c4 ≥ (a + b + c) · ≥ (a + b + c) = a + b + c.
27 27

Solution 3.
Rather than using the Power-Mean inequality to prove a4 + b4 + c4 ≥ a + b + c in Proof 2,
the Cauchy-Schwartz-Bunjakovsky inequality can be used twice:
(a4 + b4 + c4 )(12 + 12 + 12 ) ≥ (a2 + b2 + c2 )2
(a2 + b2 + c2 )(12 + 12 + 12 ) ≥ (a + b + c)2
a4 + b4 + c4 (a2 + b2 + c2 )2 (a + b + c)4
So ≥ ≥ . Continue as in Proof 2.
3 9 81
 √
4. Let Γ be a circle with radius r. Let A and B be distinct points on Γ such that AB < 3r.
Let the circle with centre B and radius AB meet Γ again at C. Let P be the point inside
Γ such that triangle ABP is equilateral. Finally, let CP meet Γ again at Q. Prove that
P Q = r.

Γ
O
Q P C

A B

Solution 1.
Let the center of Γ be O, the radius r. Since BP = BC, let θ = ]BP C = ]BCP .
Quadrilateral QABC is cyclic, so ]BAQ = 180◦ − θ and hence ]P AQ = 120◦ − θ.
Also ]AP Q = 180◦ − ]AP B − ]BP C = 120◦ − θ, so P Q = AQ and ]AQP = 2θ − 60◦ .
Again because quadrilateral QABC is cyclic, ]ABC = 180◦ − ]AQC = 240◦ − 2θ .
Triangles OAB and OCB are congruent, since OA = OB = OC = r and AB = BC.
1
Thus ]ABO = ]CBO = ]ABC = 120◦ − θ.
2
We have now shown that in triangles AQP and AOB, ]P AQ = ]BAO = ]AP Q = ]ABO.
Also AP = AB, so 4AQP ∼ = 4AOB. Hence QP = OB = r.

Solution 2.
Let the center of Γ be O, the radius r. Since A, P and C lie on a circle centered at B,
60◦ = ]ABP = 2]ACP , so ]ACP = ]ACQ = 30◦ .
Since Q, A, and C lie on Γ, ]QOA = 2]QCA = 60◦ .
So QA = r since if a chord of a circle subtends an angle of 60◦ at the center, its length is the
radius of the circle.
Now BP = BC, so ]BP C = ]BCP = ]ACB + 30◦ .
Thus ]AP Q = 180◦ − ]AP B − ]BP C = 90◦ − ]ACB.
Since Q, A, B and C lie on Γ and AB = BC, ]AQP = ]AQC = ]AQB + ]BQC = 2]ACB.
Finally, ]QAP = 180 − ]AQP − ]AP Q = 90 − ]ACB.
So ]P AQ = ]AP Q hence P Q = AQ = r.
5. Let N = {0, 1, 2, . . .}. Determine all functions f : N → N such that

xf (y) + yf (x) = (x + y)f (x2 + y 2 )

for all x and y in N.

Solution 1.
We claim that f is a constant function. Suppose, for a contradiction, that there exist x and
y with f (x) < f (y); choose x, y such that f (y) − f (x) > 0 is minimal. Then
xf (x) + yf (x) xf (y) + yf (x) xf (y) + yf (y)
f (x) = < < = f (y)
x+y x+y x+y
so f (x) < f (x2 + y 2 ) < f (y) and 0 < f (x2 + y 2 ) − f (x) < f (y) − f (x), contradicting the
choice of x and y. Thus, f is a constant function. Since f (0) is in N, the constant must be
from N.
Also, for any c in N, xc + yc = (x + y)c for all x and y, so f (x) = c, c ∈ N are the solutions
to the equation.

Solution 2.
We claim f is a constant function. Define g(x) = f (x) − f (0). Then g(0) = 0, g(x) ≥ −f (0)
and
xg(y) + yg(x) = (x + y)g(x2 + y 2 )
for all x, y in N.
Letting y = 0 shows g(x2 ) = 0 (in particular, g(1) = g(4) = 0), and letting x = y = 1 shows
g(2) = 0. Also, if x, y and z in N satisfy x2 + y 2 = z 2 , then
y
g(y) = − g(x). (∗)
x
Letting x = 4 and y = 3, (∗) shows that g(3) = 0.
For any even number x = 2n > 4, let y = n2 − 1. Then y > x and x2 + y 2 = (n2 + 1)2 . For
any odd number x = 2n + 1 > 3, let y = 2(n + 1)n. Then y > x and x2 + y 2 = ((n + 1)2 + n2 )2 .
Thus for every x > 4 there is y > x such that (∗) is satisfied.
Suppose for a contradiction, that there is x > 4 with g(x) > 0. Then we can construct a
xi+1
sequence x = x0 < x1 < x2 < . . . where g(xi+1 ) = − g(xi ). It follows that |g(xi+1 )| >
xi
|g(xi )| and the signs of g(xi ) alternate. Since g(x) is always an integer, |g(xi+1 )| ≥ |g(xi )| + 1.
Thus for some sufficiently large value of i, g(xi ) < −f (0), a contradiction.
As for Proof 1, we now conclude that the functions that satisfy the given functional equation
are f (x) = c, c ∈ N.

Solution 3. Suppose that W is the set of nonnegative integers and that f : W → W satisfies:
xf (y) + yf (x) = (x + y)f (x2 + y 2 ). (∗)

We will show that f is a constant function.

Let f (0) = k, and set S = {x | f (x) = k}.
Letting y = 0 in (∗) shows that f (x2 ) = k ∀ x > 0, and so

x2 ∈ S ∀ x > 0 (1)
In particular, 1 ∈ S.
Suppose x2 + y 2 = z 2 . Then yf (x) + xf (y) = (x + y)f (z 2 ) = (x + y)k. Thus,

x ∈ S iff y ∈ S. (2)

whenever x2 + y 2 is a perfect square.

For a contradiction, let n be the smallest non-negative integer such that f (2n ) 6= k. By (l) n
n−1 n−1 n−1 n−1
must be odd, so is an integer. Now < n so f (2 2 ) = k. Letting x = y = 2 2
2 2
in (∗) shows f (2n ) = k, a contradiction. Thus every power of 2 is an element of S.

For each integer n ≥ 2 define p(n) to be the largest prime such that p(n) | n.
Claim: For any integer n > 1 that is not a power of 2, there exists a sequence of integers
x1 , x2 , . . . , xr such that the following conditions hold:

a) x1 = n.
b) x2i + x2i+1 is a perfect square for each i = 1, 2, 3, . . . , r − 1.
c) p(x1 ) ≥ p(x2 ) ≥ . . . ≥ p(xr ) = 2.

Proof: Since n is not a power of 2, p(n) = p(x1 ) ≥ 3. Let p(x1 ) = 2m + 1, so n = x1 =

b(2m + 1)a , for some a and b, where p(b) < 2m + 1.

Case 1: a = 1. Since (2m+1, 2m2 +2m, 2m2 +2m+1) is a Pythagorean Triple, if x2 = b(2m2 +
2m), then x21 + x22 = b2 (2m2 + 2m + 1)2 is a perfect square. Furthermore, x2 = 2bm(m + 1),
and so p(x2 ) < 2m + 1 = p(x1 ).

Case 2: a > 1. If n = x1 = (2m + 1)a · b, let x2 = (2m + 1)a−1 · b · (2m2 + 2m), x3 =

(2m + 1)a−2 · b · (2m2 + 2m)2 , . . ., xa+1 = (2m + 1)0 · b · (2m2 + 2m)a = b · 2a ma (m + 1)a . Note
that for 1 ≤ i ≤ a, x2i + x2i+1 is a perfect square and also note that p(xa+1 ) < 2m + 1 = p(x1 ).

If xa+1 is not a power of 2, we extend the sequence xi using the same procedure described
above. We keep doing this until p(xr ) = 2, for some integer r.

By (2), xi ∈ S iff xi+1 ∈ S for i = 1, 2, 3, . . . , r − 1. Thus, n = x1 ∈ S iff xr ∈ S. But xr is

a power of 2 because p(xr ) = 2, and we earlier proved that powers of 2 are in S. Therefore,
n ∈ S , proving the claim.

We have proven that every integer n ≥ 1 is an element of S, and so we have proven that
f (n) = k = f (0), for each n ≥ 1. Therefore, f is constant, Q.E.D.
 The Canadian Mathematical Olympiad - 2003

1. Consider a standard twelve-hour clock whose hour and minute hands move continu-
ously. Let m be an integer, with 1 ≤ m ≤ 720. At precisely m minutes after 12:00, the
angle made by the hour hand and minute hand is exactly 1◦ . Determine all possible
values of m.
2001
2. Find the last three digits of the number 20032002 .

3. Find all real positive solutions (if any) to

x3 + y 3 + z 3 = x + y + z, and

x2 + y 2 + z 2 = xyz.

4. Prove that when three circles share the same chord AB, every line through A different
from AB determines the same ratio XY : Y Z, where X is an arbitrary point different
from B on the first circle while Y and Z are the points where AX intersects the other
two circles (labelled so that Y is between X and Z).

l
A
X
Y
Z

5. Let S be a set of n points in the plane such that any two points of S are at least 1
is a subset T of S with at least n/7 points such that any two
unit apart. Prove there √
points of T are at least 3 units apart.
 Solutions to the 2003 CMO
written March 26, 2003

1. Consider a standard twelve-hour clock whose hour and minute hands move continu-
ously. Let m be an integer, with 1 ≤ m ≤ 720. At precisely m minutes after 12:00, the
angle made by the hour hand and minute hand is exactly 1◦ . Determine all possible
values of m.

Solution
The minute hand makes a full revolution of 360◦ every 60 minutes, so after m minutes
it has swept through 360
60
m = 6m degrees. The hour hand makes a full revolution every
12 hours (720 minutes), so after m minutes it has swept through 360
720
m = m/2 degrees.
Since both hands started in the same position at 12:00, the angle between the two
hands will be 1◦ if 6m − m/2 = ±1 + 360k for some integer k. Solving this equation
we get
720k ± 2 5k ± 2
m= = 65k + .
11 11
Since 1 ≤ m ≤ 720, we have 1 ≤ k ≤ 11. Since m is an integer, 5k ± 2 must be divisible
by 11, say 5k ± 2 = 11q. Then
q±2
5k = 11q ± 2 ⇒ k = 2q + .
5
If is now clear that only q = 2 and q = 3 satisfy all the conditions. Thus k = 4 or
k = 7 and substituting these values into the expression for m we find that the only
possible values of m are 262 and 458.
 2001
2. Find the last three digits of the number 20032002 .

Solution
2001
We must find the remainder when 20032002 is divided by 1000, which will be the
20022001
same as the remainder when 3 is divided by 1000, since 2003 ≡ 3 (mod 1000).
To do this we will first find a positive integer n such that 3n ≡ 1 (mod 1000) and then
try to express 20022001 in the form nk + r, so that
2001
20032002 ≡ 3nk+r ≡ (3n )k · 3r ≡ 1k · 3r ≡ 3r (mod 1000).

Since 32 = 10 − 1, we can evaluate 32m using the binomial theorem:

m(m − 1)
32m = (10 − 1)m = (−1)m + 10m(−1)m−1 + 100 (−1)m−2 + · · · + 10m .
2
After the first 3 terms of this expansion, all remaining terms are divisible by 1000, so
letting m = 2q, we have that

34q ≡ 1 − 20q + 100q(2q − 1) (mod 1000). (1)

Using this, we can check that 3100 ≡ 1 (mod 1000) and now we wish to find the
remainder when 20022001 is divided by 100.
Now 20022001 ≡ 22001 (mod 100) ≡ 4 · 21999 (mod 4 · 25), so we’ll investigate powers of
2 modulo 25. Noting that 210 = 1024 ≡ −1 (mod 25), we have

21999 = (210 )199 · 29 ≡ (−1)199 · 512 ≡ −12 ≡ 13 (mod 25).

Thus 22001 ≡ 4 · 13 = 52 (mod 100). Therefore 20022001 can be written in the form
100k + 52 for some integer k, so
2001
20032002 ≡ 352 (mod 1000) ≡ 1 − 20 · 13 + 1300 · 25 ≡ 241 (mod 1000)
2001
using equation (1). So the last 3 digits of 20032002 are 241.
3. Find all real positive solutions (if any) to

x3 + y 3 + z 3 = x + y + z, and

x2 + y 2 + z 2 = xyz.

Solution 1
Let f (x, y, z) = (x3 − x) + (y 3 − y) + (z 3 − z). The first equation above is equivalent
to f (x, y, z) = 0. If x, y, z ≥ 1, then f (x, y, z) ≥ 0 with equality only if x = y = z = 1.
But if x = y = z = 1, then the second equation is not satisfied. So in any solution to
the system of equations, at least one of the variables is less than 1. Without loss of
generality, suppose that x < 1. Then

x2 + y 2 + z 2 > y 2 + z 2 ≥ 2yz > yz > xyz.

Therefore the system has no real positive solutions.

Solution 2
We will show that the system has no real positive solution. Assume otherwise.
The second equation can be written x2 − (yz)x + (y 2 + z 2 ). Since this quadratic in x
has a real solution by hypothesis, its discrimant is nonnegative. Hence

y 2 z 2 − 4y 2 − 4z 2 ≥ 0.

Dividing through by 4y 2 z 2 yields

1 1 1 1
≥ 2 + 2 ≥ 2.
4 y z y

Hence y 2 ≥ 4 and so y ≥ 2, y being positive. A similar argument yields x, y, z ≥ 2.

But the first equation can be written as

x(x2 − 1) + y(y 2 − 1) + z(z 2 − 1) = 0,

contradicting x, y, z ≥ 2. Hence, a real positive solution cannot exist.

Solution 3
Applying the arithmetic-geometric mean inequality and the Power Mean Inequalities
to x, y, z we have
r r
√ x + y + z x 2 + y2 + z2 3 3 3
3 x + y + z
3
xyz ≤ ≤ ≤ .
3 3 3
Letting S = x + y + z = x3 + y 3 + z 3 and P = xyz = x2 + y 2 + z 2 , this inequality can
be written r r
√3 S P 3 S
P ≤ ≤ ≤ .
3 3 3
√ q q
Now P ≤ 3 implies P ≤ P /27, so P ≥ 27. Also 3 ≤ 3 S3 implies S 3 /27 ≤ S/3,
3 P 2 3 S

√ q √ q
3 S
3
so S ≤ 3. But then P ≥ 3 and 3
≤ 1 which is inconsistent with P ≤ 3 S3 .
3

Therefore the system cannot have a real positive solution.

4. Prove that when three circles share the same chord AB, every line through A different
from AB determines the same ratio XY : Y Z, where X is an arbitrary point different
from B on the first circle while Y and Z are the points where AX intersects the other
two circles (labelled so that Y is between X and Z).

l
A
X
α Y
β
γ Z

Solution 1
Let l be a line through A different from AB and join B to A, X, Y and Z as in the
above diagram. No matter how l is chosen, the angles AXB, AY B and AZB always
subtend the chord AB. For this reason the angles in the triangles BXY and BXZ are
the same for all such l. Thus the ratio XY : Y Z remains constant by similar triangles.
Note that this is true no matter how X, Y and Z lie in relation to A. Suppose X, Y and
Z all lie on the same side of A (as in the diagram) and that ]AXB = α, ]AY B = β
and ]AZB = γ. Then ]BXY = 180◦ − α, ]BY X = β, ]BY Z = 180◦ − β and
]BZY = γ. Now suppose l is chosen so that X is now on the opposite side of A from
Y and Z. Now since X is on the other side of the chord AB, ]AXB = 180◦ − α, but
it is still the case that ]BXY = 180◦ − α and all other angles in the two pertinent
triangles remain unchanged. If l is chosen so that X is identical with A, then l is
tangent to the first circle and it is still the case that ]BXY = 180◦ − α. All other
cases can be checked in a similar manner.
 l
A
X
P Q R Y
Z
O1 O2 O3

Solution 2
Let m be the perpendicular bisector of AB and let O1 , O2 , O3 be the centres of the
three circles. Since AB is a chord common to all three circles, O1 , O2 , O3 all lie on m.
Let l be a line through A different from AB and suppose that X, Y , Z all lie on the
same side of AB, as in the above diagram. Let perpendiculars from O1 , O2 , O3 meet l
at P , Q, R, respectively. Since a line through the centre of a circle bisects any chord,

AX = 2AP, AY = 2AQ and AZ = 2AR.

Now
XY = AY − AX = 2(AQ − AP ) = 2P Q and, similarly, Y Z = 2QR.

Therefore XY : Y Z = P Q : QR. But O1 P || O2 Q || O3R, so P Q : QR = O1 O2 : O2 O3 .

Since the centres of the circles are fixed, the ratio XY : Y Z = O1 O2 : O2 O3 does not
depend on the choice of l.
If X, Y , Z do not all lie on the same side of AB, we can obtain the same result with
a similar proof. For instance, if X and Y are opposite sides of AB, then we will have
XY = AY + AX, but since in this case P Q = AQ + AP , it is still the case that
XY = 2P Q and result still follows, etc.
5. Let S be a set of n points in the plane such that any two points of S are at least 1
unit apart. Prove there √
is a subset T of S with at least n/7 points such that any two
points of T are at least 3 units apart.

Solution
We will construct the set T in the following way: Assume the points of S are in the
xy-plane and let P be a point in S with maximum y-coordinate. This point P will be a
member of√the set T and now, from S, we will remove P and all points in S which are
less than 3 units from P . From the remaining points we choose one with maximum
y-coordinate
√ to be a member of T and remove from S all points at distance less than
3 units from this new point. We continue in this way,
√ until all the points of S are
exhausted. Clearly any two points in T are at least 3 units apart. To show that T
has at least n/7 points, we must prove that at each stage no more than 6 other points
are removed along with P .
At a typical stage in this process, we’ve√ selected a point P with maximum y-coordinate,
so any points
√ at distance less than 3 from P must lie inside the semicircular region
of radius 3 centred at P shown in the first diagram below. Since points of S are at
least 1 unit apart, these points must lie outside (or on) the semicircle of radius 1. (So
they lie in the shaded region of the first diagram.) Now divide this shaded region into
6 congruent regions R1 , R2 , . . . , R6 as shown in this diagram.
We will show that each of these regions contains at most one point of S. Since all 6
regions are congruent, consider one of them as depicted in the second diagram below.
The distance between any two points in this shaded region √ must be less than the length
of the line segment AB. The lengths of P A and P B are 3 and 1, respectively, and
angle AP B = 30◦ . If√we construct a perpendicular from B to P A at C, then the length
of P C is cos 30◦ = 3/2. Thus BC is a perpendicular bisector of P A and therefore
AB = P B = 1. So the distance between any two points in this region is less than 1.
Therefore each of R1 , . . . , R6 can contain at most one point of S, which completes the
proof.

√
1 P 3
P C A
R1 R6 30◦

R2 R5
B
R3 R4
 36th Canadian Mathematical Olympiad
Wednesday, March 31, 2004

1. Find all ordered triples (x, y, z) of real numbers which satisfy the following system of
equations: 
 xy = z − x − y
xz = y − x − z

yz = x − y − z

2. How many ways can 8 mutually non-attacking rooks be placed

on the 9 × 9 chessboard (shown here) so that all 8 rooks are on
squares of the same colour?
[Two rooks are said to be attacking each other if they are placed
in the same row or column of the board.]

3. Let A, B, C, D be four points on a circle (occurring in clockwise order), with AB < AD

and BC > CD. Let the bisector of angle BAD meet the circle at X and the bisector
of angle BCD meet the circle at Y . Consider the hexagon formed by these six points
on the circle. If four of the six sides of the hexagon have equal length, prove that BD
must be a diameter of the circle.

4. Let p be an odd prime. Prove that

p−1
X p(p + 1)
k 2p−1 ≡ (mod p2 ).
k=1
2

[Note that a ≡ b (mod m) means that a − b is divisible by m.]

5. Let T be the set of all positive integer divisors of 2004100 . What is the largest possible
number of elements that a subset S of T can have if no element of S is an integer
multiple of any other element of S?
 Solutions to the 2004 CMO
written March 31, 2004

1. Find all ordered triples (x, y, z) of real numbers which satisfy the following system of
equations: 
 xy = z − x − y
xz = y − x − z

yz = x − y − z

Solution 1
Subtracting the second equation from the first gives xy − xz = 2z − 2y. Factoring y − z
from each side and rearranging gives
(x + 2)(y − z) = 0,
so either x = −2 or z = y.
If x = −2, the first equation becomes −2y = z + 2 − y, or y + z = −2. Substituting
x = −2, y + z = −2 into the third equation gives yz = −2 − (−2) = 0. Hence either y
or z is 0, so if x = −2, the only solutions are (−2, 0, −2) and (−2, −2, 0).
If z = y the first equation becomes xy = −x, or x(y + 1) = 0. If x = 0 and z = y,
the third equation becomes y 2 = −2y which gives y = 0 or y = −2. If y = −1 and
z = y = −1, the third equation gives x = −1. So if y = z, the only solutions are
(0, 0, 0), (0, −2, −2) and (−1, −1, −1).
In summary, there are 5 solutions: (−2, 0, −2), (−2, −2, 0), (0, 0, 0), (0, −2, −2) and
(−1, −1, −1).

Solution 2
Adding x to both sides of the first equation gives

x(y + 1) = z − y = (z + 1) − (y + 1) ⇒ (x + 1)(y + 1) = z + 1.

Similarly manipulating the other two equations and letting a = x + 1, b = y + 1,

c = z + 1, we can write the system in the following way.

 ab = c
ac = b

bc = a

If any one of a, b, c is 0, then it’s clear that all three are 0. So (a, b, c) = (0, 0, 0) is
one solution and now suppose that a, b, c are all nonzero. Substituting c = ab into the
second and third equations gives a2 b = b and b2a = a, respectively. Hence a2 = 1,
b2 = 1 (since a, b nonzero). This gives 4 more solutions: (a, b, c) = (1, 1, 1), (1, −1, −1),
(−1, 1, −1) or (−1, −1, 1). Reexpressing in terms of x, y, z, we obtain the 5 ordered
triples listed in Solution 1.
2. How many ways can 8 mutually non-attacking rooks be placed
on the 9 × 9 chessboard (shown here) so that all 8 rooks are on
squares of the same colour?
[Two rooks are said to be attacking each other if they are placed
in the same row or column of the board.]

Solution 1
We will first count the number of ways of placing 8 mutually non-attacking rooks on
black squares and then count the number of ways of placing them on white squares.
Suppose that the rows of the board have been numbered 1 to 9 from top to bottom.
First notice that a rook placed on a black square in an odd O O O O O
numbered row cannot attack a rook on a black square in an even E E E E
O O O O O
numbered row. This effectively partitions the black squares into E E E E
a 5 × 5 board and a 4 × 4 board (squares labelled O and E O O O O O
respectively, in the diagram at right) and rooks can be placed O E O E O E O E O
independently on these two boards. There are 5! ways to place E E E E
5 non-attacking rooks on the squares labelled O and 4! ways to O O O O O
place 4 non-attacking rooks on the squares labelled E.
This gives 5!4! ways to place 9 mutually non-attacking rooks on black squares and
removing any one of these 9 rooks gives one of the desired configurations. Thus there
are 9 · 5!4! ways to place 8 mutually non-attacking rooks on black squares.
Using very similar reasoning we can partition the white squares O O O O
as shown in the diagram at right. The white squares are par- E E E E E
O O O O
titioned into two 5 × 4 boards such that no rook on a square E E E E E
marked O can attack a rook on a square mark E. At most 4 O O O O
E E E E E
non-attacking rooks can be placed on a 5 × 4 board and they O O O O
can be placed in 5 · 4 · 3 · 2 = 5! ways. Thus there are (5!)2 ways E E E E E
O O O O
to place 8 mutually non-attacking rooks on white squares.
In total there are 9 · 5!4! + (5!)2 = (9 + 5)5!4! = 14 · 5!4! = 40320 ways to place 8
mutually non-attacking rooks on squares of the same colour.
Solution 2
Consider rooks on black squares first. We have 8 rooks and 9 rows, so exactly one row
will be without rooks. There are two cases: either the empty row has 5 black squares
or it has 4 black squares. By permutation these rows can be made either last or second
last. In each case we’ll count the possible number of ways of placing the rooks on the
board as we proceed row by row.
In the first case we have 5 choices for the empty row, then we can place a rook on any
of the black squares in row 1 (5 possibilities) and any of the black squares in row 2 (4
possibilities). When we attempt to place a rook in row 3, we must avoid the column
containing the rook that was placed in row 1, so we have 4 possibilities. Using similar
reasoning, we can place the rook on any of 3 possible black squares in row 4, etc. The
total number of possibilities for the first case is 5 · 5 · 4 · 4 · 3 · 3 · 2 · 2 · 1 = (5!)2 . In the
second case, we have 4 choices for the empty row (but assume it’s the second last row).
We now place rooks as before and using similar logic, we get that the total number of
possibilities for the second case is 4 · 5 · 4 · 4 · 3 · 3 · 2 · 1 · 1 = 4(5!4!).
Now, do the same for the white squares. If a row with 4 white squares is empty (5
ways to choose it), then the total number of possibilities is (5!)2. It’s impossible to
have a row with 5 white squares empty, so the total number of ways to place rooks is

(5!)2 + 4(5!4!) + (5!)2 = (5 + 4 + 5)5!4! = 14(5!4!).

3. Let A, B, C, D be four points on a circle (occurring in clockwise order), with AB < AD
and BC > CD. Let the bisector of angle BAD meet the circle at X and the bisector
of angle BCD meet the circle at Y . Consider the hexagon formed by these six points
on the circle. If four of the six sides of the hexagon have equal length, prove thatBD
must be a diameter of the circle.

B X

α
A α γ C
γ

Y D

Solution 1
We’re given that AB < AD. Since CY bisects ]BCD, BY = Y D, so Y lies between
D and A on the circle, as in the diagram above, and DY > Y A, DY > AB. Similar
reasoning confirms that X lies between B and C and BX > XC, BX > CD. So if
ABXCDY has 4 equal sides, then it must be that Y A = AB = XC = CD.
Let ]BAX = ]DAX = α and let ]BCY = ]DCY = γ. Since ABCD is cyclic,
]A+]C = 180◦ , which implies that α+γ = 90◦ . The fact that Y A = AB = XC = CD
means that the arc from Y to B (which is subtended by ]Y CB) is equal to the arc from
X to D (which is subtended by ]XAD). Hence ]Y CB = ]XAD, so α = γ = 45◦ .
Finally, BD is subtended by ]BAD = 2α = 90◦ . Therefore BD is a diameter of the
circle.

Solution 2
We’re given that AB < AD. Since CY bisects ]BCD, BY = Y D, so Y lies between
D and A on the circle, as in the diagram above, and DY > Y A, DY > AB. Similar
reasoning confirms that X lies between B and C and BX > XC, BX > CD. So
if ABXCDY has 4 equal sides, then it must be that Y A = AB = XC = CD.
This implies that the arc from Y to B is equal to the arc from X to D and hence
that Y B = XD. Since ]BAX = ]XAD, BX = XD and since ]DCY = ]Y CB,
DY = Y B. Therefore BXDY is a square and its diagonal, BD, must be a diameter
of the circle.
4. Let p be an odd prime. Prove that
p−1
X p(p + 1)
k 2p−1 ≡ (mod p2 ).
2
k=1

[Note that a ≡ b (mod m) means that a − b is divisible by m.]

Solution
Since p − 1 is even, we can pair up the terms in the summation in the following way
(first term with last, 2nd term with 2nd last, etc.):
p−1
p−1
X 2 
X 
k 2p−1 = k 2p−1 + (p − k)2p−1 .
k=1 k=1

Expanding (p − k)2p−1 with the binomial theorem, we get

   
2p − 1 2p − 1
(p − k)2p−1 = p2p−1 − · · · − p2 k 2p−3 + pk 2p−2 − k 2p−1 ,
2 1

where every term on the right-hand side is divisible by p2 except the last two. Therefore
 
2p−1 2p−1 2p−1 2p − 1
k + (p − k) ≡k + pk 2p−2 − k 2p−1 ≡ (2p − 1)pk 2p−2 (mod p2 ).
1

For 1 ≤ k < p, k is not divisible by p, so k p−1 ≡ 1 (mod p), by Fermat’s Little Theorem.
So (2p − 1)k 2p−2 ≡ (2p − 1)(12 ) ≡ −1 (mod p), say (2p − 1)k 2p−2 = mp − 1 for some
integer m. Then
(2p − 1)pk 2p−2 = mp2 − p ≡ −p (mod p2 ).
Finally,
p−1
X
p−1
X
2  
2p−1 p−1
k ≡ (−p) ≡ (−p) (mod p2 )
2
k=1 k=1
p − p2 p(p + 1)
≡ + p2 ≡ (mod p2 ).
2 2
5. Let T be the set of all positive integer divisors of 2004100 . What is the largest possible
number of elements that a subset S of T can have if no element of S is an integer
multiple of any other element of S?

Solution
Assume throughout that a, b, c are nonnegative integers. Since the prime factorization
of 2004 is 2004 = 22 · 3 · 167,
n  o

T = 2a 3b 167c  0 ≤ a ≤ 200, 0 ≤ b, c ≤ 100 .

Let n  o

S = 2200−b−c 3b 167c  0 ≤ b, c ≤ 100 .

For any 0 ≤ b, c ≤ 100, we have 0 ≤ 200− b− c ≤ 200, so S is a subset of T . Since there

are 101 possible values for b and 101 possible values for c, S contains 1012 elements.
We will show that no element of S is a multiple of another and that no larger subset
of T satisfies this condition.
Suppose 2200−b−c 3b 167c is an integer multiple of 2200−j−k 3j 167k . Then

200 − b − c ≥ 200 − j − k, b ≥ j, c ≥ k.

But this first inequality implies b + c ≤ j + k, which together with b ≥ j, c ≥ k gives

b = j and c = k. Hence no element of S is an integer multiple of another element of S.
Let U be a subset of T with more than 1012 elements. Since there are only 1012 distinct
pairs (b, c) with 0 ≤ b, c ≤ 100, then (by the pigeonhole principle) U must contain two
elements u1 = 2a1 3b1 167c1 and u2 = 2a2 3b2 167c2 , with b1 = b2 and c1 = c2, but a1 6= a2.
If a1 > a2, then u1 is a multiple of u2 and if a1 < a2 , then u2 is a multiple of u1 . Hence
U does not satisfy the desired condition.
Therefore the largest possible number of elements that such a subset of T can have is
1012 = 10201.
 37th Canadian Mathematical Olympiad
Wednesday, March 30, 2005

1. Consider an equilateral triangle of side length n, which is divided into unit triangles, as shown. Let
f (n) be the number of paths from the triangle in the top row to the middle triangle in the bottom
row, such that adjacent triangles in our path share a common edge and the path never travels up
(from a lower row to a higher row) or revisits a triangle. An example of one such path is illustrated
below for n = 5. Determine the value of f (2005).

2. Let (a, b, c) be a Pythagorean triple, i.e., a triplet of positive integers with a2 + b2 = c2 .

a) Prove that (c/a + c/b)2 > 8.

b) Prove that there does not exist any integer n for which we can find a Pythagorean triple (a, b, c)
satisfying (c/a + c/b)2 = n.

3. Let S be a set of n ≥ 3 points in the interior of a circle.

a) Show that there are three distinct points a, b, c ∈ S and three distinct points A, B, C on the
circle such that a is (strictly) closer to A than any other point in S, b is closer to B than any
other point in S and c is closer to C than any other point in S.
b) Show that for no value of n can four such points in S (and corresponding points on the circle)
be guaranteed.

4. Let ABC be a triangle with circumradius R, perimeter P and area K. Determine the maximum
value of KP/R3 .

5. Let’s say that an ordered triple of positive integers (a, b, c) is n-powerful if a ≤ b ≤ c, gcd(a, b, c) = 1,
and an + bn + cn is divisible by a + b + c. For example, (1, 2, 2) is 5-powerful.

a) Determine all ordered triples (if any) which are n-powerful for all n ≥ 1.
b) Determine all ordered triples (if any) which are 2004-powerful and 2005-powerful, but not 2007-
powerful.

[Note that gcd(a, b, c) is the greatest common divisor of a, b and c.]

 Solutions to the 2005 CMO
written March 30, 2005

1. Consider an equilateral triangle of side length n, which is divided into unit triangles, as
shown. Let f (n) be the number of paths from the triangle in the top row to the middle
triangle in the bottom row, such that adjacent triangles in our path share a common
edge and the path never travels up (from a lower row to a higher row) or revisits a
triangle. An example of one such path is illustrated below for n = 5. Determine the
value of f (2005).

Solution
We shall show that f (n) = (n − 1)!.
Label the horizontal line segments in the triangle l1 , l2 , . . . as in the diagram below.
Since the path goes from the top triangle to a triangle in the bottom row and never
travels up, the path must cross each of l1 , l2 , . . . , ln−1 exactly once. The diagonal lines
in the triangle divide lk into k unit line segments and the path must cross exactly one
of these k segments for each k. (In the diagram below, these line segments have been
highlighted.) The path is completely determined by the set of n − 1 line segments
which are crossed. So as the path moves from the kth row to the (k + 1)st row,
there are k possible line segments where the path could cross lk . Since there are
1 · 2 · 3 · · · (n − 1) = (n − 1)! ways that the path could cross the n − 1 horizontal lines,
and each one corresponds to a unique path, we get f (n) = (n − 1)!.
Therefore f (2005) = (2004)!.

l1
l2
l3
l4
2. Let (a, b, c) be a Pythagorean triple, i.e., a triplet of positive integers with a2 + b2 = c2 .
a) Prove that (c/a + c/b)2 > 8.
b) Prove that there does not exist any integer n for which we can find a Pythagorean
triple (a, b, c) satisfying (c/a + c/b)2 = n.

a) Solution 1
Let (a, b, c) be a Pythagorean triple. View a, b as lengths of the legs of a right
angled triangle with hypotenuse of length c; let θ be the angle determined by the
sides with lengths a and c. Then
2

c c 2 sin2 θ + cos2 θ + 2 sin θ cos θ

1 1
+ = + =
a b cos θ sin θ (sin θ cos θ)2
 
1 + sin 2θ 4 4
= 4 2 = 2 +
sin 2θ sin 2θ sin 2θ
Note that because 0 < θ < 90◦ , we have √ 0 < sin 2θ ≤ 1, with equality only if
θ = 45◦ . But then a = b and we obtain 2 = c/a, contradicting a, c both being
integers. Thus, 0 < sin 2θ < 1 which gives (c/a + c/b)2 > 8.
Solution 2
Defining θ as in Solution 1, we have c/a √ + c/b = sec θ + csc θ. By the AM-GM
inequality, we have (sec θ + csc θ)/2 ≥ sec θ csc θ. So
√
2 2 2 √
c/a + c/b ≥ √ =√ ≥ 2 2.
sin θ cos θ sin 2θ
√
Since a, b, c are integers, we have c/a + c/b > 2 2 which gives (c/a + c/b)2 > 8.
Solution 3
By simplifying and using the AM-GM inequality,
2 √ √

c c 2 a+b (a2 + b2 )(a + b)2 2 a2 b2 (2 ab)2

2
+ =c = ≥ = 8,
a b ab a2 b 2 a2 b 2
with equality only if a = b. By using the same argument as in Solution 1, a cannot
equal b and the inequality is strict.
Solution 4

c c 2 c2 c2 2c2 b 2 a2 2(a2 + b2 )
+ = 2+ 2+ =1+ 2 + 2 +1+
a b a b ab a b ab
2
a b

2

= 2+ − +2+ (a − b)2 + 2ab
b a ab
2
a b 2(a − b)2

= 4+ − + + 4 ≥ 8,
b a ab
with equality only if a = b, which (as argued previously) cannot occur.
b) Solution 1
Since c/a + c/b is rational, (c/a + c/b)2 can only be an integer if c/a + c/b is an
integer. Suppose c/a + c/b = m. We may assume that gcd(a, b) = 1. (If not,
divide the common factor from (a, b, c), leaving m unchanged.)
Since c(a+b) = mab and gcd(a, a+b) = 1, a must divide c, say c = ak. This gives
a2 + b2 = a2 k 2 which implies b2 = (k 2 − 1)a2 . But then a divides b contradicting
the fact that gcd(a, b) = 1. Therefore (c/a + c/b)2 is not equal to any integer n.
Solution 2
We begin as in Solution 1, supposing that c/a + c/b = m with gcd(a, b) = 1.
Hence a and b are not both even. It is also the case that a and b are not both
odd, for then c2 = a2 + b2 ≡ 2 (mod 4), and perfect squares are congruent to
either 0 or 1 modulo 4. So one of a, b is odd and the other is even. Therefore
c must be odd.
Now c/a + c/b = m implies c(a + b) = mab, which cannot be true because c(a + b)
is odd and mab is even.
 3. Let S be a set of n ≥ 3 points in the interior of a circle.
a) Show that there are three distinct points a, b, c ∈ S and three distinct points
A, B, C on the circle such that a is (strictly) closer to A than any other point in
S, b is closer to B than any other point in S and c is closer to C than any other
point in S.
b) Show that for no value of n can four such points in S (and corresponding points
on the circle) be guaranteed.

Solution 1
a) Let H be the smallest convex set of points in the plane which contains S.† Take
3 points a, b, c ∈ S which lie on the boundary of H. (There must always be at
least 3 (but not necessarily 4) such points.)
Since a lies on the boundary of the convex region H, we can construct a chord L
such that no two points of H lie on opposite sides of L. Of the two points where
the perpendicular to L at a meets the circle, choose one which is on a side of L
not containing any points of H and call this point A. Certainly A is closer to a
than to any other point on L or on the other side of L. Hence A is closer to a
than to any other point of S. We can find the required points B and C in an
analogous way and the proof is complete.
[Note that this argument still holds if all the points of S lie on a line.]
P
A √
3
a 2 r
r
a b
c
L
H r
b
Q c R

(a) (b)

b) Let P QR be an equilateral triangle inscribed in the circle and let a, b, c be mid-

points of the three sides of P √ QR. If r is the radius of the circle, then every
point on the circle is within ( 3/2)r of one of a, b or c. (See figure (b) above.)
√
Now 3/2 < 9/10, so if S consists of a, b, c and a cluster of points within r/10 of
the centre of the circle, then we cannot select 4 points from S (and corresponding
points on the circle) having the desired property.

†
By the way, H is called the convex hull of S. If the points of S lie on a line, then H will be the shortest
line segment containing the points of S. Otherwise, H is a polygon whose vertices are all elements of S and
such that all other points in S lie inside or on this polygon.
Solution 2
a) If all the points of S lie on a line L, then choose any 3 of them to be a, b, c. Let
A be a point on the circle which meets the perpendicular to L at a. Clearly A is
closer to a than to any other point on L, and hence closer than other other point
in S. We find B and C in an analogous way.
Otherwise, choose a, b, c from S so that the triangle formed by these points has
maximal area. Construct the altitude from the side bc to the point a and extend
this line until it meets the circle at A. We claim that A is closer to a than to any
other point in S.
Suppose not. Let x be a point in S for which the distance from A to x is less than
the distance from A to a. Then the perpendicular distance from x to the line bc
must be greater than the perpendicular distance from a to the line bc. But then
the triangle formed by the points x, b, c has greater area than the triangle formed
by a, b, c, contradicting the original choice of these 3 points. Therefore A is closer
to a than to any other point in S.
The points B and C are found by constructing similar altitudes through b and c,
respectively.
b) See Solution 1.
4. Let ABC be a triangle with circumradius R, perimeter P and area K. Determine the
maximum value of KP/R3 .

Solution 1
Since similar triangles give the same value of KP/R3 , we can fix R = 1 and maximize
KP over all triangles inscribed in the unit circle. Fix points A and B on the unit circle.
The locus of points C with a given perimeter P is an ellipse that meets the circle in at
most four points. The area K is maximized (for a fixed P ) when C is chosen on the
perpendicular bisector of AB, so we get a maximum value for KP if C is where the
perpendicular bisector of AB meets the circle. Thus the maximum value of KP for
a given AB occurs when ABC is an isosceles triangle. Repeating this argument with
BC fixed, we have that the maximum occurs when ABC is an equilateral triangle.
Consider
√ an equilateral √triangle with side length a. It has P = 3a. It has height equal
2
to a 3/2
√ giving K = a 3/4. ¿From the extended law of sines, 2R = a/ sin(60) giving
R = a/ 3. Therefore the maximum value we seek is
 √   √ 3
a2 3 3 27
KP/R3 = (3a) = .
4 a 4

Solution 2
From the extended law of sines, the lengths of the sides of the triangle are 2R sin A,
2R sin B and 2R sin C. So
1
P = 2R(sin A + sin B + sin C) and K = (2R sin A)(2R sin B)(sin C),
2
giving
KP
= 4 sin A sin B sin C(sin A + sin B + sin C).
R3
We wish to find the maximum value of this expression over all A + B + C = 180◦ .
Using well-known identities for sums and products of sine functions, we can write

KP cos(B − C) cos(B + C) B+C B−C

     
= 4 sin A − sin A + 2 sin cos .
R3 2 2 2 2

If we first consider A to be fixed, then B + C is fixed

 also and this expression takes
its maximum value when cos(B − C) and cos B−C 2
equal 1; i.e. when B = C. In a
similar way, one can show that for any fixed value of B, KP/R3 is maximized when
A = C. Therefore the maximum value of KP/R3 occurs when A = B = C = 60◦ ,
and it is now an easy task to substitute this into the above expression to obtain the
maximum value of 27/4.
Solution 3
As in Solution 2, we obtain
KP
= 4 sin A sin B sin C(sin A + sin B + sin C).
R3
From the AM-GM inequality, we have
3
sin A + sin B + sin C

sin A sin B sin C ≤ ,
3

giving
KP 4
≤ (sin A + sin B + sin C)4 ,
R 3 27
with equality when sin A = sin B = sin C. Since the sine function is concave on the
interval from 0 to π, Jensen’s inequality gives
√
sin A + sin B + sin C A+B+C π
 
3
≤ sin = sin = .
3 3 3 2

Since equality occurs here when sin A = sin B = sin C also, we can conclude that the

√ 4
4 3 3
maximum value of KP/R3 is 27 2
= 27/4.
5. Let’s say that an ordered triple of positive integers (a, b, c) is n-powerful if a ≤ b ≤ c,
gcd(a, b, c) = 1, and an + bn + cn is divisible by a + b + c. For example, (1, 2, 2) is
5-powerful.
a) Determine all ordered triples (if any) which are n-powerful for all n ≥ 1.
b) Determine all ordered triples (if any) which are 2004-powerful and 2005-powerful,
but not 2007-powerful.
[Note that gcd(a, b, c) is the greatest common divisor of a, b and c.]

Solution 1
Let Tn = an + bn + cn and consider the polynomial
P (x) = (x − a)(x − b)(x − c) = x3 − (a + b + c)x2 + (ab + ac + bc)x − abc.
Since P (a) = 0, we get a3 = (a + b + c)a2 − (ab + ac + bc)a + abc and multiplying both
sides by an−3 we obtain an = (a + b + c)an−1 − (ab + ac + bc)an−2 + (abc)an−3 . Applying
the same reasoning, we can obtain similar expressions for bn and cn and adding the
three identities we get that Tn satisfies the following 3-term recurrence:
Tn = (a + b + c)Tn−1 − (ab + ac + bc)Tn−2 + (abc)Tn−3 , for all n ≥ 3.
¿From this we see that if Tn−2 and Tn−3 are divisible by a + b + c, then so is Tn . This
immediately resolves part (b)—there are no ordered triples which are 2004-powerful
and 2005-powerful, but not 2007-powerful—and reduces the number of cases to be
considered in part (a): since all triples are 1-powerful, the recurrence implies that any
ordered triple which is both 2-powerful and 3-powerful is n-powerful for all n ≥ 1.
Putting n = 3 in the recurrence, we have
a3 + b3 + c3 = (a + b + c)(a2 + b2 + c2 ) − (ab + ac + bc)(a + b + c) + 3abc
which implies that (a, b, c) is 3-powerful if and only if 3abc is divisible by a + b + c.
Since
a2 + b2 + c2 = (a + b + c)2 − 2(ab + ac + bc),
(a, b, c) is 2-powerful if and only if 2(ab + ac + bc) is divisible by a + b + c.
Suppose a prime p ≥ 5 divides a + b + c. Then p divides abc. Since gcd(a, b, c) = 1, p
divides exactly one of a, b or c; but then p doesn’t divide 2(ab + ac + bc).
Suppose 32 divides a + b + c. Then 3 divides abc, implying 3 divides exactly one of a,
b or c. But then 3 doesn’t divide 2(ab + ac + bc).
Suppose 22 divides a + b + c. Then 4 divides abc. Since gcd(a, b, c) = 1, at most one
of a, b or c is even, implying one of a, b, c is divisible by 4 and the others are odd. But
then ab + ac + bc is odd and 4 doesn’t divide 2(ab + ac + bc).
So if (a, b, c) is 2- and 3-powerful, then a + b + c is not divisible by 4 or 9 or any prime
greater than 3. Since a + b + c is at least 3, a + b + c is either 3 or 6. It is now a
simple matter to check the possibilities and conclude that the only triples which are
n-powerful for all n ≥ 1 are (1, 1, 1) and (1, 1, 4).
Solution 2
Let p be a prime. By Fermat’s Little Theorem,

1 (mod p), if p doesn’t divide a;

p−1
a ≡
0 (mod p), if p divides a.

Since gcd(a, b, c) = 1, we have that ap−1 + bp−1 + cp−1 ≡ 1, 2 or 3 (mod p). Therefore if
p is a prime divisor of ap−1 +bp−1 +cp−1 , then p equals 2 or 3. So if (a, b, c) is n-powerful
for all n ≥ 1, then the only primes which can divide a + b + c are 2 or 3.
We can proceed in a similar fashion to show that a + b + c is not divisible by 4 or 9.
Since
0 (mod 4), if p is even;

2
a ≡
1 (mod 4), if p is odd
and a, b, c aren’t all even, we have that a2 + b2 + c2 ≡ 1, 2 or 3 (mod 4).
By expanding (3k)3 , (3k + 1)3 and (3k + 2)3 , we find that a3 is congruent to 0, 1 or
−1 modulo 9. Hence

0 (mod 9), if 3 divides a;

6
a ≡
1 (mod 9), if 3 doesn’t divide a.

Since a, b, c aren’t all divisible by 3, we have that a6 + b6 + c6 ≡ 1, 2 or 3 (mod 9).

So a2 + b2 + c2 is not divisible by 4 and a6 + b6 + c6 is not divisible by 9. Thus if (a, b, c)
is n-powerful for all n ≥ 1, then a + b + c is not divisible by 4 or 9. Therefore a + b + c
is either 3 or 6 and checking all possibilities, we conclude that the only triples which
are n-powerful for all n ≥ 1 are (1, 1, 1) and (1, 1, 4).
See Solution 1 for the (b) part.
 38th Canadian Mathematical Olympiad
Wednesday, March 29,2006

1. Let f (n; k ) be the number of ways of distributing k candies to n children so that each child receives at
most 2 candies. For example, if n = 3, then f (3; 7) = 0, f (3; 6) = 1 and f (3; 4) = 6.

Determine the value of

f (2006; 1) + f (2006; 4) + f (2006; 7) +


+ f (2006; 1000) + f (2006; 1003) :

2. Let ABC be an acute-angled triangle. Inscribe a rectangle DE F G in this triangle so that D is on AB ,

E is on AC and both F and G are on BC . Describe the locus of (i.e., the curve occupied by) the
intersections of the diagonals of all possible rectangles DE F G.

3. In a rectangular array of nonnegative real numbers with m rows and n columns, each row and each
column contains at least one positive element. Moreover, if a row and a column intersect in a positive
element, then the sums of their elements are the same. Prove that m = n.

4. Consider a round-robin tournament with 2n + 1 teams, where each team plays each other team exactly
once. We say that three teams X, Y and Z, form a cycle triplet if X beats Y , Y beats Z, and Z

beats X. There are no ties.

(a) Determine the minimum number of cycle triplets possible.

(b) Determine the maximum number of cycle triplets possible.

5. The vertices of a right triangle ABC inscribed in a circle divide the circumference into three arcs.
The right angle is at A, so that the opposite arc BC is a semicircle while arc AB and arc AC are
supplementary. To each of the three arcs, we draw a tangent such that its point of tangency is the
midpoint of that portion of the tangent intercepted by the extended lines AB and AC . More precisely,
the point D on arc BC is the midpoint of the segment joining the points D
0
and D
00
where the tangent
at D intersects the extended lines AB and AC . Similarly for E on arc AC and F on arc AB .

Prove that triangle DE F is equilateral.

 38th Canadian Mathematical Olympiad
Wednesday, March 29, 2006

Solutions to the 2006 CMO paper

1. Let f (n, k) be the number of ways of distributing k candies to n children so that each child receives at most 2 candies.
For example, if n = 3, then f (3, 7) = 0, f (3, 6) = 1 and f (3, 4) = 6.
Determine the value of
f (2006, 1) + f (2006, 4) + f (2006, 7) + · · · + f (2006, 1000) + f (2006, 1003) .

Comment. Unfortunately, there was an error in the statement of this problem. It was intended that the sum should
continue to f (2006, 4012).
Solution 1. The number of ways of distributing k candies to 2006 children is equal to the number of ways of distributing
0 to a particular child and k to the rest, plus the number of ways of distributing 1 to the particular child and k − 1
to the rest, plus the number of ways of distributing 2 to the particular child and k − 2 to the rest. Thus f (2006, k) =
f (2005, k) + f (2005, k − 1) + f (2005, k − 2), so that the required sum is
1003
X
1+ f (2005, k) .
k=1
¡n¢
In evaluating f (n, k), suppose that there are r children who receive 2 candies; these r children can be chosen in r ways.
Then there are k − 2r candies from which at most one is given to each of n − r children. Hence
bk/2c µ ¶µ ¶ ∞ µ ¶µ ¶
X n n−r X n n−r
f (n, k) = = ,
r=0
r k − 2r r=0
r k − 2r
¡x¢
with y = 0 when x < y and when y < 0. The answer is

1003
XX ∞ µ ¶µ ¶ ∞ µ
X ¶ 1003 µ ¶
2005 2005 − r 2005 X 2005 − r
= .
r k − 2r r k − 2r
k=0 r=0 r=0 k=0

Solution 2. The desired number is the sum of the coefficients of the terms of degree not exceeding 1003 in the expansion
of (1 + x + x2 )2005 , which is equal to the coefficient of x1003 in the expansion of

(1 + x + x2 )2005 (1 + x + · · · + x1003 ) = [(1 − x3 )2005 (1 − x)−2005 ](1 − x1004 )(1 − x)−1

= (1 − x3 )2005 (1 − x)−2006 − (1 − x3 )2005 (1 − x)−2006 x1004 .

Since the degree of every term in the expansion of the second member on the right exceeds 1003, we are looking for the
coefficient of x1003 in the expansion of the first member:
2005
X µ ¶ ∞ µ ¶
2005 3i X −2006 j
(1 − x3 )2005 (1 − x)−2006 = (−1)i x (−1)j x
i=0
i j=0
j

1
 2005
XX ∞ µ ¶µ ¶
2005 2005 + j 3i+j
= (−1)i x
i=0 j=0
i j
∞ µ 2005
X X µ ¶µ ¶¶
2005 2005 + k − 3i
= (−1)i xk .
i=1
i 2005
k=0

The desired number is µ ¶µ ¶ X

334
X 334
i 2005 3008 − 3i (3008 − 3i)!
(−1) = (−1)i .
i=1
i 2005 i=1
i!(2005 − i)!(1003 − 3i)!
¡3008−3i¢
(Note that 2005 = 0 when i ≥ 335.)

2. Let ABC be an acute-angled triangle. Inscribe a rectangle DEF G in this triangle so that D is on AB, E is on AC
and both F and G are on BC. Describe the locus of (i.e., the curve occupied by) the intersections of the diagonals of all
possible rectangles DEF G.
Solution. The locus is the line segment joining the midpoint M of BC to the midpoint K of the altitude AH. Note that
a segment DE with D on AB and E on AC determines an inscribed rectangle; the midpoint F of DE lies on the median
AM , while the midpoint of the perpendicular from F to BC is the centre of the rectangle. This lies on the median M K of
the triangle AM H.
Conversely, any point P on M K is the centre of a rectangle with base along BC whose height is double the distance from
K to BC.

3. In a rectangular array of nonnegative real numbers with m rows and n columns, each row and each column contains at
least one positive element. Moreover, if a row and a column intersect in a positive element, then the sums of their elements
are the same. Prove that m = n.
Solution 1. Consider first the case where all the rows have the same positive sum s; this covers the particular situation
in which m = 1. Then each column, sharing a positive element with some row, must also have the sum s. Then the sum of
all the entries in the matrix is ms = ns, whence m = n.
We prove the general case by induction on m. The case m = 1 is already covered. Suppose that we have an m × n array
not all of whose rows have the same sum. Let r < m of the rows have the sum s, and each of the of the other rows have a
different sum. Then every column sharing a positive entry with one of these rows must also have sum s, and these are the
only columns with the sum s. Suppose there are c columns with sum s. The situation is essentially unchanged if we permute
the rows and then the column so that the first r rows have the sum s and the first c columns have the sum s. Since all
the entries of the first r rows not in the first c columns and in the first c columns not in the first r rows must be 0, we can
partition the array into a r × c array in which all rows and columns have sum s and which satisfies the hypothesis of the
problem, two rectangular arrays of zeros in the upper right and lower left and a rectangular (m − r) × (n − c) array in the
lower right that satisfies the conditions of the problem. By the induction hypothesis, we see that r = c and so m = n.
Solution 2. [Y. Zhao] Let the term in the ith row and the jth column of the array be denoted by aij , and let S = {(i, j) :
aij > 0}. Suppose that ri is the sum of the ith row and cj the sum of the jth column. Then ri = cj whenever (i, j) ∈ S.
Then we have that X aij X aij
{ : (i, j) ∈ S} = { : (i, j) ∈ S} .
ri cj

We evaluate the sums on either side independently.

X aij X aij Xm n m µ ¶ m
1 X X 1 X
{ : (i, j) ∈ S} = { : 1 ≤ i ≤ m, 1 ≤ j ≤ n} = aij = ri = 1=m.
ri ri r
i=1 i j=1 i=1
ri i=1

X aij X aij Xn m n µ ¶ n
1 X X 1 X
{ : (i, j) ∈ S} = { : 1 ≤ i ≤ m, 1 ≤ j ≤ n} = aij = cj = 1=n.
cj cj c
j=1 j i=1 j=1
cj j=1

Hence m = n.

2
 Comment. The second solution can be made cleaner and more elegant by defining uij = aij /ri for all (i, j). When aij = 0,
then uij = 0. When aij > 0, then, by hypothesis, uij = aij /cj , a relation that in fact holds for all (i, j). We find that
n
X n
X
uij = 1 and uij = 1
j=1 i=1

for 1 ≤ i ≤ m and 1 ≤ j ≤ n, so that (uij ) is an m × n array whose row sums and column sums are all equal to 1. Hence
m µX
X n ¶ X n µX
X m ¶
m= uij = {uij : 1 ≤ i ≤ m, 1 ≤ j ≤ n} = uij =n
i=1 j=1 j=1 i=1

(being the sum of all the entries in the array).

4. Consider a round-robin tournament with 2n + 1 teams, where each team plays each other team exactly once. We say
that three teams X, Y and Z, form a cycle triplet if X beats Y , Y beats Z, and Z beats X. There are no ties.
(a) Determine the minimum number of cycle triplets possible.
(b) Determine the maximum number of cycle triplets possible.
Solution 1. (a) The minimum is 0, which is achieved by a tournament in which team Ti beats Tj if and only if i > j.
(b) Any set of three teams constitutes either a cycle triplet or a¡ “dominated
¢ triplet” in which one team beats the other
two; let there be c of the former and d of the latter. Then c + d = 2n+1 . Suppose that team Ti beats xi other teams; then
¡ ¢ 3P
2n+1 ¡ ¢
it is the winning team in exactly x2i dominated triples. Observe that i=1 xi = 2n+1 2 , the total number of games. Hence

Xµ
2n+1
xi
¶ 2n+1
1 X 2 1 2n + 1
µ ¶
d= = xi − .
i=1
2 2 i=1 2 2
P2n+1 P2n+1
By the Cauchy-Schwarz Inequality, (2n + 1) i=1 x2i ≥ ( i=1 xi )2 = n2 (2n + 1)2 , whence
µ ¶ 2n+1
X µxi ¶ µ2n + 1¶ n2 (2n + 1) 1 µ2n + 1¶ n(n + 1)(2n + 1)
2n + 1
c= − ≤ − + = .
3 i=1
2 3 2 2 2 6

To realize the upper bound, let the teams be T1 = T2n+2 , T2 = T2n+3 . · · ·, Ti = T2n+1+i , · · ·, T2n+1 = T4n+2 . For
each i, let team Ti beat Ti+1 , Ti+2 , · · · , Ti+n and lose to Ti+n+1 , · · · , Ti+2n . We need to check that this is a consistent
assignment of wins and losses, since the result for each pair of teams is defined twice. This can be seen by noting that
(2n + 1 + i) − (i + j) = 2n + 1 − j ≥ n + 1 for 1 ≤ j ≤ n . The cycle triplets are (Ti , Ti+j , Ti+j+k ) where 1 ≤ j ≤ n and
(2n + 1 + i) − (i + j + k) ≤ n, i.e., when 1 ≤ j ≤ n and n + 1 − j ≤ k ≤ n. For each i, this counts 1 + 2 + · · · + n = 12 n(n + 1)
cycle triplets. When we range over all i, each cycle triplet gets counted three times, so the number of cycle triplets is
µ ¶
2n + 1 n(n + 1) n(n + 1)(2n + 1)
= .
3 2 6

Solution 2. [S. Eastwood] (b) Let t be the number of cycle triplets and u be the number of ordered triplets of teams
(X, Y, Z) where X beats Y and Y beats Z. Each cycle triplet generates three ordered triplets while other triplets generate
exactly one. The total number of triplets is µ ¶
2n + 1 n(4n2 − 1)
= .
3 3
The number of triples that are not cycle is
n(4n2 − 1)
−t .
3
Hence µ ¶
n(4n2 − 1)
u = 3t + −t =⇒
3

3
 3u − n(4n2 − 1) u − (2n + 1)n2 n(n + 1)(2n + 1)
t= = + .
6 2 6
If team Y beats a teams and loses to b teams, then the number of ordered triples with Y as the central element is ab.
Since a + b = 2n, by the Arithmetic-Geometric Means Inequality, we have that ab ≤ n2. Hence u ≤ (2n + 1)n2, so that

n(n + 1)(2n + 1)
t≤ .
6
The maximum is attainable when u = (2n + 1)n2, which can occur when we arrange all the teams in a circle with each team
beating exactly the n teams in the clockwise direction.
Pn
Comment. Interestingly enough, the maximum is i=1 i2 ; is there a nice argument that gives the answer in this form?

5. The vertices of a right triangle ABC inscribed in a circle divide the circumference into three arcs. The right angle is
at A, so that the opposite arc BC is a semicircle while arc AB and arc AC are supplementary. To each of the three arcs, we
draw a tangent such that its point of tangency is the midpoint of that portion of the tangent intercepted by the extended
lines AB and AC. More precisely, the point D on arc BC is the midpoint of the segment joining the points D0 and D00 where
the tangent at D intersects the extended lines AB and AC. Similarly for E on arc AC and F on arc AB.
Prove that triangle DEF is equilateral.

Solution 1. A prime indicates where a tangent meets AB and a double prime where it meets AC. It is given that
DD0 = DD00 , EE 0 = EE 00 and F F 0 = F F 00 . It is required to show that arc EF is a third of the circumference as is arc
DBF .
AF is the median to the hypotenuse of right triangle AF 0 F 00 , so that F F 0 = F A and therefore

arc AF = 2∠F 00 F A = 2(∠F F 0 A + ∠F AF 0 ) = 4∠F AF 0 = 4∠F AB = 2 arc BF ,

whence arc F A = (2/3) arc BF A. Similarly, arc AE = (2/3) arc AEC. Therefore, arc F E is 2/3 of the semicircle, or 1/3
of the circumference as desired.
As for arc DBF , arc BD = 2∠BAD = ∠BAD + ∠BD0 D = ∠ADD00 = (1/2) arc ACD. But, arc BF = (1/2) arc AF ,
so arc DBF = (1/2) arc F AED. That is, arc DBF is 1/3 the circumference and the proof is complete.
Solution 2. Since AE 0 E 00 is a right triangle, AE = EE 0 = EE 00 so that ∠CAE = ∠CE 00 E. Also AD = D0 D = DD00 , so
that ∠CDD00 = ∠CAD = ∠CD00 D. As EADC is a concyclic quadrilateral,
180◦ = ∠EAD + ∠ECD
= ∠DAC + ∠CAE + ∠ECA + ∠ACD
= ∠DAC + ∠CAE + ∠CEE 00 + ∠CE 00 E + ∠CDD00 + ∠CD00 D
= ∠DAC + ∠CAE + ∠CAE + ∠CAE + ∠CAD + ∠CAD
= 3(∠DAC + ∠DAE) = 3(∠DAE)
Hence ∠DF E = ∠DAE = 60◦ . Similarly, ∠DEF = 60◦ . It follows that triangle DEF is equilateral.

4
 39th Canadian Mathematical Olympiad
Wednesday, March 28,2007

1. What is the maximum number of non-overlapping 2 × 1 dominoes that can be placed on a 8 × 9 checkerboard if six of
them are placed as shown? Each domino must be placed horizontally or vertically so as to cover two adjacent squares of
the board.
































2. You are given a pair of triangles for which

(a) two sides of one triangle are equal in length to two sides of the second triangle, and
(b) the triangles are similar, but not necessarily congruent.
√ √
Prove that the ratio of the sides that correspond under the similarity is a number between 12 ( 5 − 1) and 12 ( 5 − 1).
3. Suppose that f is a real-valued function for which

f (xy) + f (y − x) ≥ f (y + x)

for all real numbers x and y.

(a) Give a nonconstant polynomial that satisfies the condition.
(b) Prove that f (x) ≥ 0 for all real x.
4. For two real numbers a, b, with ab 6= 1, define the ∗ operation by
a + b − 2ab
a∗b= .
1 − ab
Start with a list of n ≥ 2 real numbers whose entries x all satisfy 0 < x < 1. Select any two numbers a and b in the list;
remove them and put the number a ∗ b at the end of the list, thereby reducing its length by one. Repeat this procedure
until a single number remains.
(a) Prove that this single number is the same regardless of the choice of pair at each stage.
(b) Suppose that the condition on the numbers x in S is weakened to 0 < x ≤ 1. What happens if S contains exactly one
1?
5. Let the incircle of triangle ABC touch sides BC, CA and AB at D, E and F , respectively. Let Γ,Γ1 ,Γ2 and Γ3 denote the
circumcircles of triangle ABC, AEF , BDF and CDE respectively. Let Γ and Γ1 intersect at A and P , Γ and Γ2 intersect
at B and Q, and Γ and Γ3 intersect at C and R.
(a) Prove that the circles Γ1 , Γ2 and Γ3 intersect in a common point.
(b) Show that P D, QE and RF are concurrent.
 39th Canadian Mathematical Olympiad
Wednesday, March 28, 2007

Solutions to the 2007 CMO paper

Solution to 1. Identify ve subsets A; B; C; D; E of the board, where C consists of the squares occupied by the six dominos
already placed, B is the upper right corner, D is the lower left corner, A consists of the squares above and to the left of
those in B [ C [ D and E consists of the squares below and to the right of those in B [ C [ D. The board can be coloured
checkerboard fashion so that A has 13 black and 16 white squares, B a single white square, E 16 black and 13 white squares
and D a single black square. Each domino beyond the original six must lie either entirely in A [ B [ D or C [ B [ D, either of
which contains at most 14 dominos. Thus, altogether, we cannot have more that 2  14 + 6 = 34 dominos. This is achievable,
by placing 14 dominos in A [ D and 14 in E [ B .
Solution to 2. If the triangles are isosceles, then they must be congruent and the desired ratio is 1. For, if they share
equal side lengths, at least one of these side lengths on one triangle corresponds to the same length on the other. And if they
share unequal side lengths, then either equal sides correspond or unequal sides correspond in both directions and the ratio is
1. This falls within the bounds.
Let the triangles be scalene. It is not possible for the same length to be an extreme length (largest or smallest) of both
triangles. Therefore, we must have a situation in which the corresponding side lengths of the two triangles are (x; y; z ) and
(y; z; u) with x < y < z and y < z < u. We are given that y=x = z=y = u=z = r > 1. Thus, y = rx andpz = ry = r2 x. From
the triangle inequality z < x + y, we have that r2 < 1 + r. Since r2 r 1 < p 0 and r > 1, 1 < r < 12 ( 5 + 1). The ratio of
the dimensions from the smaller to the larger triangle is 1=r which satises 2 ( 5 1) < 1=r < 1. The result follows.
1

Solution to 3. (a) Let f (x) = x2 + 4. Then

f (xy ) + f (y x) f (y + x) = (x2 y2 + 4) + (y x)2 + 4 (y + x)2 4
= (xy)2 4xy + 4 = (xy 2)2  0 : (1)
Thus, f (x) = x2 + 4 satises the condition.
(b) Consider (x; y) for which xy = x + y. Rewriting this as (x 1)(y 1) = 1, we nd that this has the general solution
(x; y) = (1 + t 1 ; 1 + t), for t 6= 0. Plugging this into the inequality, we get that f (t t 1 )  0 for all t 6= 0. For arbitrary real

1
u,the equation t t 1 = u leads to the quadratic t2 ut 1 = 0 which has a positive discriminant and so a real solution.
Hence f (u)  0 for each real u.
Comment. The substitution v = y x, u = y + x whose inverse is x = 12 (u v), y = 12 (u + v) renders p the condition as
f ( (u
1
4
2
v )) + f (v )  f (u). The same strategy as in the foregoing solution leads to the choice u = 2 + v 2 + 4 and f (v )  0
2

for all v.
Solution to 4 (b). It is straightforward to verify that a  1 = 1 for a 6= 1, so that once 1 is included in the list, it can never
by removed and so the list terminates with the single value 1.
Solution to 4 (a). There are several ways of approaching (a). It is important to verify that the set fx : 0 < x < 1g is
closed under the operation  so that it is always dened.
If 0 < a; b < 1, then
a + b 2ab
0< <1:
1 ab
The left inequality follows from
a + b 2ab = a(1 b) + b(1 a) > 0
and the right from
a + b 2ab (1 a)(1 b)
1 = >0:
1 ab 1 ab
Hence, it will never happen that a set of numbers will contain a pair of reciprocals, and the operation can always be performed.
Solution 1. It can be shown by induction that any two numbers in any of the sets arise from disjoint subsets of S .

Use an induction argument on the number of entries that one starts with. At each stage the number of entries is reduced
by one. If we start with n numbers, the nal result is
1 22 + 33


+ ( 1)n 1 nn
;
1 2 + 23 34 +


+ ( 1)n 1 (n 1)n
n

where i is the symmetric sum of all i i fold products of the n elements xi in the list.
Solution 2. Dene
a+b 2ab
ab= :
1 ab
This operation is commutative and also associative:
a+b+c 2(ab + bc + ca) + 3abc
a  (b  c) = (a  b)  c = :
1 (ab + bc + ca) + 2abc
Since the nal result amounts to a  product of elements of S with some arrangement of brackets, the result follows.
Solution 3. Let (x) = x=(1 x) for 0 < x < 1. This is a one-one function from the open interval (0; 1) to the half line
(0; 1). For any numbers a; b 2 S , we have that
 
a+b 2ab a+b 2ab a+b 2ab
 = ab) (a+b 2ab) = 1 a b+ab
1 ab (1

= a + b = (a) + (b) : (2)

1 a 1 b

Let T = f(s) : s 2 S g. Then replacing a; b in S as indicated corresponds to P

replacing (a) and (b) in T by (a) + (b) to
get a new pair of sets related by . The nal result is the inverse under  of f(s) : s 2 S g.
Solution 4. Let f (x) = (1 x) be dened for positive x unequal to 1. Then f (x) > 1 if and only if 0 < x < 1. Observe
1

that
1 xy 1 1
f (x  y ) = = + 1:
1 x y + xy 1 x 1 y
If f (x) > 1 and f (y) > 1, then also f (x  y) > 1. It follows that if x and y lie in the open interval (0; 1), so does x  y. We
also note that f (x) is a one-one function.

2
 To each list L, we associate the function g(L) dened by
X
g ( L) = ff (x) : x 2 Lg :
Let Ln be the given list, and let the subsequent lists be Ln 1 ; Ln 2 ;


; L1 , where Li has i elements. Since f (x  y) =
f (x) + f (y ) 1, g(Li ) = g(Ln ) (n i) regardless of the choice that creates each list from its predecessors. Hence
g (L1 ) = g (Ln ) (n 1) is xed. However, g(L1 ) = f (a) for some number a with 0 < a < 1. Hence a = f 1 (g(Ln ) (n 1))
is xed.
Solution to 5 (a). Let I be the incentre of triangle ABC . Since the quadrilateral AEIF has right angles at E and F , it
is concyclic, so that passes through I . Similarly, 2 and 3 pass through I , and (a) follows.
1

Solution to 5 (b). Let ! and I denote the incircle and incentre of triangle ABC , respectively. Observe that, since AI
bisects the angle F AE and AF = AE , then AI right bisects the segment F E . Similarly, BI right bisects DF and CI right
bisects DE .
We invert the diagram through !. Under this inversion, let the image of A be A0 , etc. Note that the centre I of inversion
is collinear with any point and its image under the inversion. Under this inversion, the image of 1 is EF , which makes A0
the midpoint of EF . Similarly, B 0 is the midpoint of DF and C 0 is the midpoint of DE . Hence, 0 , the image of under
this inversion, is the circumcircle of triangle A0 B 0 C 0 , which implies that 0 is the nine-point circle of triangle DEF .
Since P is the intersection of and 1 other than A, P 0 is the intersection of 0 and EF other than A0 , which means
that P 0 is the foot of the altitude from D to EF . Similarly, Q0 is the foot of the altitude from E to DF and R0 is the foot
of the altitude from F to DE .
Now, let X , Y and Z be the midpoints of arcs BC , AC and AB on respectively. We claim that X lies on P D.
Let X 0 be the image of X under the inversion, so I , X and X 0 are collinear. But X is the midpoint of arc BC , so A,
A0 , I , X 0 and X are collinear. The image of line P D is the circumcircle of triangle P 0 ID, so to prove that X lies on P D, it
suces to prove that points P 0 , I , X 0 and D are concyclic.
We know that B 0 is the midpoint of DF , C 0 is the midpoint of DE and P 0 is the foot of the altitude from D to EF .
Hence, D is the re ection of P 0 in B 0 C 0 .
Since IA0 ? EF , IB 0 ? DF and IC 0 ? DE , I is the orthocentre of triangle A0 B 0 C 0 . So, X 0 is the intersection of the
altitude from A0 to B 0 C 0 with the circumcircle of triangle A0 B 0 C 0 . From a wellknown fact, X 0 is the re ection of I in B 0 C 0 .
This means that B 0 C 0 is the perpendicular bisector of both P 0 D and IX 0 , so that the points P 0 , I , X 0 and D are concyclic.
Hence, X lies on P D. Similarly, Y lies on QE and Z lies on RF . Thus, to prove that P D, QE and RF are concurrent,
it suces to prove that DX , EY and F Z are concurrent.
To show this, consider tangents to at X , Y and Z . These are parallel to BC , AC and AB , respectively. Hence, the
triangle
that these tangents dene is homothetic to the triangle ABC . Let S be the centre of homothety. Then the
homothety taking triangle ABC to
takes ! to , and so takes D to X , E to Y and F to Z . Hence DX , EY and F Z
concur at S .
Comment. The solution uses the following result: Suppose ABC is a triangle with orthocentre H and that AH intersects
BC at P and the circumcircle of ABC at D. Then HP = P D. The proof is straightforward: Let BH meet AC at Q. Note
that AD ? BC and BQ ? AC . Since \ACB = \ADB ,
\HBC = \QBC = 90 \QCB = 90 \ACB = 90 \ADB = \DBP ;
from which follows the congruence of triangle HBP and DBP and equality of HP and P D.
Solution 2. (a) Let 2 and 3 intersect at J . Then BDJF and CDJE are concyclic. We have that

\F JE = 360 (\DJF + \DJE )

= 360 (180 \ABC + 180 \ACB )

= \ABC + \ACB = 180 \F AE : (3)
Hence AF JE is concyclic and so the circumcircles of AEF , BDF and CED pass through J .
(b) [Y. Li] Join RE, RD, RA and RB. In 3 , \ERD = \ECD = \ACB and \REC = \RDC . In , \ARB = \ACB .
Hence, \ERD = \ARB =) \ARE = \BRD. Also,
\AER = 180 \REC = 180 \RDC = \BDR :
Therefore, triangle ARE and BRD are similar, and AR : BR = AE : BD = AF : BF . If follows that RF bisects angle
ARB , so that RF passes through the midpoint of minor arc AB on . Similarly, P D and QE are respective bisectors of
angles BP C and CQA and pass through the midpoints of the minor arc BC and CA on ..

3
 Let O be the centre of circle , and U , V , W be the respective midpoints of the minor arc BC , CA, AB on this circle,
so that P U contains D, QV contains E and RW contains F . It is required to prove that DU , EV and F W are concurrent.
Since ID and OU are perpendicular to BC , IDkOU . Similarly, IE kOV and IF kOW . Since jIDj = jIE j = jIF j = r (the
inradius) and jOU j = jOV j = jOW j = R (the circumradius), a translation IO ! followed by a dilatation of factor R=r takes
triangle DEF to triangle U V W , so that these triangles are similar with corresponding sides parallel.
Suppose that EV and F W intersect at K and that DU and F W intersect at L. Taking account of the similarity of the
triangles KEF and KV W , LDF and LU W , DEF and U V W , we have that
KF : F W = EF : V W = DF : U W = LF : LW ;
so that K = L and the lines DU , EV and F W intersect in a common point K , as desired.

4
 40th Canadian Mathematical Olympiad
Wednesday, March 26, 2008

1. ABCD is a convex quadrilateral for which AB is the longest side. Points M and N are located on
sides AB and BC respectively, so that each of the segments AN and CM divides the quadrilateral
into two parts of equal area. Prove that the segment M N bisects the diagonal BD.

2. Determine all functions f defined on the set of rational numbers that take rational values for which

f (2f (x) + f (y)) = 2x + y ,

for each x and y.

3. Let a, b, c be positive real numbers for which a + b + c = 1. Prove that

a − bc b − ca c − ab 3
+ + ≤ .
a + bc b + ca c + ab 2

4. Determine all functions f defined on the natural numbers that take values among the natural numbers
for which
(f (n))p ≡ n mod f (p)
for all n ∈ N and all prime numbers p.

5. A self-avoiding rook walk on a chessboard (a rectangular grid of unit squares) is a path traced by
a sequence of moves parallel to an edge of the board from one unit square to another, such that
each begins where the previous move ended and such that no move ever crosses a square that has
previously been crossed, i.e., the rook’s path is non-self-intersecting.

Let R(m, n) be the number of self-avoiding rook walks on an m × n (m rows, n columns) chessboard
which begin at the lower-left corner and end at the upper-left corner. For example, R(m, 1) = 1 for
all natural numbers m; R(2, 2) = 2; R(3, 2) = 4; R(3, 3) = 11. Find a formula for R(3, n) for each
natural number n.
 Report - Fortieth Canadian Mathematical Olympiad 2008

40th Canadian Mathematical Olympiad

Wednesday, March 26, 2008

1. ABCD is a convex quadrilateral forSolutions

which AB- isCMO 2008 side. Points M and N are located on
the longest
sides AB and BC respectively, so that each of the segments AN and CM divides the quadrilateral
1. ABCD is parts
into two a convex quadrilateral
of equal in which
area. Prove is the longest
ABsegment
that the side. Points
MN bisects M and BD.
the diagonal N are located on sides
AB and BC respectively, so that each of the segments AN and CM divides the quadrilateral into two
2.parts of equal
Determine allarea. Provef that
functions the on
defined segment bisects numbers
the setMofNrational the diagonal
thatBD.
take rational values for which

Solution. Since [M ADC] = 12 [ABCD]f (2f

= [N(x)ADC], it =
+ f (y)) follows
2x + that
y , [AN C] = [AM C], so that M N �AC.
Let m be a line through D parallel to AC and M N and let BA produced meet m at P and BC produced
meet m
forat Q. xThen
each and y.

3. Let a, b, c be [M
positive
P C] =real
[M numbers for which
AC] + [CAP ] = [MaAC]
+ b+
+ [CAD]
c = 1. Prove that = [BM C]
= [M ADC]
a − bc b − ca c − ab 3
whence BM = M P . Similarly BN = N Q, so that
+ MN + ≤ of. triangle BP Q and must bisect BD.
is a midline
a + bc b + ca c + ab 2

2.4.Determine
Determineall
allfunctions
functions ff defined on the
defined on the natural
set of rationals
numbersthat
that take
take rational valuesthe
values among fornatural
which numbers
for which
f (2f
(f (x)
(n)) p + f (y)) = 2x + y
≡n mod f (p)
for all n ∈ N and all prime numbers p.
for each x and y.
5. A self-avoiding rook walk on a chessboard (a rectangular grid of unit squares) is a path traced by
Solution 1. The
a sequence only solutions
of moves parallel toarean
f (x) = xoffor
edge theallboard
rational x and
from (x) =
one funit −x for
square toall rationalsuch
another, x. Both
that of
these each
readily check out.
begins where the previous move ended and such that no move ever crosses a square that has
Setting y = been
previously x yields f (3f (x))
crossed, i.e., = 3xrook’s
the for allpath
rational x. Now replacing x by 3f (x), we find that
is non-self-intersecting.

Let R(m, n) be the numberfof(9x)

self-avoiding
= f (3f (3frook
(x)) walks
= 3[3fon an=m9f×(x)
(x)] n (m
, rows, n columns) chessboard
which begin at the lower-left corner and end at the upper-left corner. For example, R(m, 1) = 1 for
all natural numbers m; R(2, 2) = 2; R(3, 2) = 4; R(3, 3) = 11. Find a formula for R(3, n) for each
for all rational x. Setting x = 0 yields f (0) = 9f (0), whence f (0) = 0.
natural number n.
Setting x = 0 in the given functional equation yields f (f (y)) = y for all rational y. Thus f is one-one
onto. Applying f to the functional equation yields that

2f (x) + f (y) = f (2x + y)

for every rational pair (x, y).

Setting y = 0 in the functional equation yields f (2f (x)) = 2x, whence 2f (x) = f (2x). Therefore
f (2x) + f (y) = f (2x + y) for each rational pair (x, y), so that

f (u + v) = f (u) + f (v)

for each rational pair (u, v). 1

Since 0 = f (0) = f (−1) + f (1), f (−1) = −f (1). By induction, it can be established that for each intger
n and rational x, f (nx) = nf (x). If k = f (1), we can establish from this that f (n) = nk, f (1/n) = k/n and
f (m/n) = mk/n for each integer pair (m, n). Thus f (x) = kx for all rational x. Since f (f (x)) = x, we must
have k2 = 1. Hence f (x) = x or f (x) = −x. These check  out.
Solution 2. In the functional equation, let
 Setting y = 0 in the functional equation yields f (2f (x)) = 2x, whence 2f (x) = f (2x). Therefore
f (2x) + f (y) = f (2x + y) for each rational pair (x, y), so that

f (u + v) = f (u) + f (v)
Report - Fortieth Canadian Mathematical Olympiad 2008
for each rational pair (u, v).
Since 0 = f (0) = f (−1) + f (1), f (−1) = −f (1). By induction, it can be established that for each intger
n and rational x, f (nx) = nf (x). If k = f (1), we can establish from this that f (n) = nk, f (1/n) = k/n and
f (m/n) = mk/n for each integer pair (m, n). Thus f (x) = kx for all rational x. Since f (f (x)) = x, we must
have k2 = 1. Hence f (x) = x or f (x) = −x. These check out.

Solution 2. In the functional equation, let

x = y = 2f (z) + f (w)

to obtain f (x) = f (y) = 2z + w and

f (6z + 3w) = 6f (z) + 3f (w)

1 f (0) = 0, w = 0 to obtain f (6z) = 6f (z) and z = 0

for all rational pairs (z, w). Set (z, w) = (0, 0) to obtain
to obtain f (3w) = 3f (w) for all rationals z and w. Hence f (6z + 3w) = f (6z) + f (3w). Replacing (6z, 3w)
by (u, v) yields
f (u + v) = f (u) + f (v)
for all rational pairs (u, v). Hence f (x) = kx where k = f (1) for all rational x. Substitution of this into the
functional equation with (x, y) = (1, 1) leads to 3 = f (3f (1)) = f (3k) = 3k2, so that k = ±1. It can be
checked that both f (x) ≡ 1 and f (x) ≡ −1 satisfy the equation.
Acknowledgment. The first solution is due to Man-Duen Choi and the second to Ed Doolittle.

3. Let a, b, c be positive real numbers for which a + b + c = 1. Prove that

a − bc b − ca c − ab 3
+ + ≤ .
a + bc b + ca c + ab 2

Solution 1. Note that

a − bc 2bc 2bc
1− = = .
a + bc 1 − b − c + bc (1 − b)(1 − c)
The inequality is equivalent to
2bc 2ca 2ab 3
+ + ≥ .
(1 − b)(1 − c) (1 − c)(1 − a) (1 − a)(1 − b) 2
Manipulation yields the equivalent

4(bc + ca + ab − 3abc) ≥ 3(bc + ca + ab + 1 − a − b − c − abc) .

This simplifies to ab + bc + ca ≥ 9abc or

1 1 1
+ + ≥9.
a b c
This is a consequence of the harmonic-arithmetic means inequality.

Solution 2. Observe that

a + bc = a(a + b + c) + bc = (a + b)(a + c)

and that a + b = 1 − c, with analogous relations for other permutations of the variables. Then

(b + c)(c + a)(a + b) = (1 − a)(1 − b)(1 − c) = (ab + bc + ca) − abc .

Putting the left side of the desired inequality over a common denominator, we find that it is equal to
(a − bc)(1 − a) + (b − ac)(1 − b) + (c − ab)(1 − c) (a + b + c) − (a2 + b2 + c2) − (bc + ca + ab) + 3abc
=
(b + c)(c + a)(a + b) (b + c)(c + a)(a + b)
1 − (a + b + c)2 + (bc + ca + ab) + 3abc
=
 (ab + bc + ca) − abc
(bc + ca + ab) + 3abc
=
(bc + bc + ab) − abc
4abc
=1+ .
 a + bc = a(a + b + c) + bc = (a + b)(a + c)

and that a + b = 1 − c, with analogous relations for other permutations of the variables. Then

(b + c)(c + a)(a + b) = (1 − a)(1 − b)(1 − c) = (ab + bc + ca) − abc .

Report - Fortieth Canadian Mathematical Olympiad 2008
Putting the left side of the desired inequality over a common denominator, we find that it is equal to
(a − bc)(1 − a) + (b − ac)(1 − b) + (c − ab)(1 − c) (a + b + c) − (a2 + b2 + c2) − (bc + ca + ab) + 3abc
=
(b + c)(c + a)(a + b) (b + c)(c + a)(a + b)
1 − (a + b + c)2 + (bc + ca + ab) + 3abc
=
(ab + bc + ca) − abc
(bc + ca + ab) + 3abc
=
(bc + bc + ab) − abc
4abc
=1+ .
(a + b)(b + c)(c + a)

Using the arithmetic-geometric means inequality, we obtain that

(a + b)(b + c)(c + a) = (a2 b + b2 c + c2 a) + (ab2 + bc2 + ca2) + 2abc

≥ 3abc + 3abc + 2abc = 8abc ,

whence 4abc/[(a + b)(b + c)(c + a)] ≤ 21 . The desired2result follows. Equality occurs exactly when a = b =
c = 31 .

4. Find all functions f defined on the natural numbers that take values among the natural numbers for
which
(f (n))p ≡ n mod f (p)

for all n ∈ N and all prime numbers p.

Solution. The substitution n = p, a prime, yields p ≡ (f (p))p ≡ 0 (mod f (p)), so that p is divisible by
f (p). Hence, for each prime p, f (p) = 1 or f (p) = p.
Let S = {p : p is prime and f (p) = p}. If S is infinite, then f (n)p ≡ n (mod p) for infinitely many
primes p. By the little Fermat theorem, n ≡ f (n)p ≡ f (n), so that f (n) − n is a multiple of p for infinitely
many primes p. This can happen only if f (n) = n for all values of n, and it can be verified that this is a
solution.
If S is empty, then f (p) = 1 for all primes p, and any function satisfying this condition is a solution.
Now suppose that S is finite and non-empty. Let q be the largest prime in S. Suppose, if possible, that
q ≥ 3. Therefore, for any prime p exceeding q, p ≡ 1 (mod q). However, this is not true. Let Q be the
product of all the odd primes up to q. Then Q + 2 must have a prime factor exceeding q and at least one
of them must be incongruent to 1 (mod q). (An alternative argument notes that Bertrand’s postulate can
turn up a prime p between q and 2q which fails to satisfy p ≡ 1 mod q.)
The only remaining case is that S = {2}. Then f (2) = 2 and f (p) = 1 for every odd prime p. Since
f (n)2 ≡ n (mod 2), f (n) and n must have the same parity. Conversely, any function f for which f (n) ≡ n
(mod 2) for all n, f (2) = 2 and f (p) = 1 for all odd primes p satisfies the condition.
Therefore the only solutions are
• f (n) = n for all n ∈ N;
• any function f with f (p) = 1 for all primes p;
• any function for which f (2) = 2, f (p) = 1 for primes p exceeding 2 and f (n) and n have the same
parity.

5. A self-avoiding rook walk on a chessboard (a rectangular grid of squares) is a path traced by a sequence
of rook moves parallel to an edge of the board from one unit square to another, such that each begins
where the previous move ended and such that no move ever crosses a square that has previously been
crossed, i.e., the rook’s path is non-self-intersecting.
Let R(m, n) be the number of self-avoiding rook walks on an m × n (m rows, n columns) chessboard
which begin at the lower-left corner and end at the upper-left corner. For example, R(m, 1) = 1 for all
natural numbers m; R(2, 2) = 2; R(3, 2) = 4; R(3, 3) = 11. Find a formula for R(3, n) for each natural
number n.

Solution 1. Let rn = R(3, n). It can be checked  directly that r1 = 1 and r2 = 4. Let 1 ≤ i ≤ 3 and
1 ≤ j; let (i, j) denote the cell in the ith row from the bottom and the jth column from the left, so that the
paths in question go from (1, 1) to (3, 1).
 5. A self-avoiding rook walk on a chessboard (a rectangular grid of squares) is a path traced by a sequence
of rook moves parallel to an edge of the board from one unit square to another, such that each begins
where the previous move ended and such that no move ever crosses a square that has previously been
crossed, i.e., the rook’s path is non-self-intersecting.
Report -ofFortieth
Let R(m, n) be the number Canadian
self-avoiding rook Mathematical
walks on an mOlympiad 2008n columns) chessboard
× n (m rows,
which begin at the lower-left corner and end at the upper-left corner. For example, R(m, 1) = 1 for all
natural numbers m; R(2, 2) = 2; R(3, 2) = 4; R(3, 3) = 11. Find a formula for R(3, n) for each natural
number n.

Solution 1. Let rn = R(3, n). It can be checked directly that r1 = 1 and r2 = 4. Let 1 ≤ i ≤ 3 and
1 ≤ j; let (i, j) denote the cell in the ith row from the bottom and the jth column from the left, so that the
paths in question go from (1, 1) to (3, 1).
Suppose that n ≥ 3. The rook walks fall into exactly one of the following six categories:
(1) One walk given by (1, 1) → (2, 1) → (3, 1).
(2) Walks that avoid the cell (2, 1): Any such walk must start with (1, 1) → (1, 2) and finish with (3, 2) →
(3, 1); there are rn−1 such walks.
(3) Walks that begin with (1, 1) → (2, 1) → (2, 2) and never return to the first row: Such walks enter the
third row from (2, k) for some k with 2 ≤ k ≤ n and then go along the third row leftwards to (3, 1); there
are n − 1 such walks.
(4) Walks that begin with (1, 1) → (2, 1) → · · · → (2, k) → (1, k) → (1, k + 1) and end with (3, k + 1) →
(3, k) → (3, k − 1) → · · · → (3, 2) → (3, 1) for some k3with 2 ≤ k ≤ n − 1; there are rn−2 + rn−3 + · · · + r1
such walks.
(5) Walks that are the horizontal reflected images of walks in (3) that begin with (1, 1) → (2, 1) and never
enter the third row until the final cell; there are n − 1 such walks.
(6) Walks that are horizontal reflected images of walks in (5); there are rn−2 + rn−3 + · · · + r1 such walks.
Thus, r3 = 1 + r2 + 2(2 + r1 ) = 11 and, for n ≥ 3,

rn = 1 + rn−1 + 2[(n − 1) + rn−2 + rn−3 + · · · + r1 ]

= 2n − 1 + rn−1 + 2(rn−2 + · · · + r1 ) ,

and
rn+1 = 2n + 1 + rn + 2(rn−1 + rn−2 + · · · + r1 ) .
Therefore
rn+1 − rn = 2 + rn + rn−1 =⇒ rn+1 = 2 + 2rn + rn−1 .
Thus
rn+1 + 1 = 2(rn + 1) + (rn−1 + 1) ,
whence
1 √ 1 √
rn + 1 = √ (1 + 2)n+1 − √ (1 − 2)n+1 ,
2 2 2 2
and
1 √ 1 √
rn = √ (1 + 2)n+1 − √ (1 − 2)n+1 − 1 .
2 2 2 2

Solution 2. Employ the same notation as in Solution 1. We have that r1 = 1, r2 = 4 and r3 = 11. Let
n ≥ 3. Consider the situation that there are rn+1 columns. There are basically three types of rook walks.
Type 1. There are four rook walks that enter only the first two columns.
Type 2. There are 3rn−1 rooks walks that do not pass between the second and third columns in the
middle row (in either direction), viz. rn−1 of each of the types:

(1, 1) −→ (1, 2) −→ (1, 3) −→ · · · −→ (3, 3) −→ (3, 2) −→ (3, 1) ;

(1, 1) −→ (2, 1) −→ (2, 2) −→ (1, 2) −→ (1, 3) −→ · · · −→ (3, 3) −→ (3, 2) −→ (3, 1) ;

(1, 1) −→ (1, 2) −→ (1, 3) −→ · · · −→ (3, 3) −→ (3, 2) −→ (2, 2) −→ (2, 1) −→ (3, 1) .

Type 3. Consider the rook walks that pass between the second and third column along the middle row.
They are of Type 3a:

(1, 1) −→ ∗ −→ (2, 2) −→ (2, 3) −→ · · · −→ (3, 3) −→ (3, 2) −→ (3, 1) ,

or Type 3b: 
(1, 1) −→ (1, 2) −→ (1, 3) −→ · · · −→ (2, 3) −→ (2, 2) −→ ∗ −→ (3, 1) ,
where in each case the asterisk stands for one of two possible options.
 (1, 1) −→ (1, 2) −→ (1, 3) −→ · · · −→ (3, 3) −→ (3, 2) −→ (3, 1) ;

(1, 1) −→ (2, 1) −→ (2, 2) −→ (1, 2) −→ (1, 3) −→ · · · −→ (3, 3) −→ (3, 2) −→ (3, 1) ;

(1, 1) −→ (1, 2) −→ (1, 3) −→ · · · −→ (3, 3) −→ (3, 2) −→ (2, 2) −→ (2, 1) −→ (3, 1) .
Report - Fortieth Canadian Mathematical Olympiad 2008
Type 3. Consider the rook walks that pass between the second and third column along the middle row.
They are of Type 3a:

(1, 1) −→ ∗ −→ (2, 2) −→ (2, 3) −→ · · · −→ (3, 3) −→ (3, 2) −→ (3, 1) ,

or Type 3b:
(1, 1) −→ (1, 2) −→ (1, 3) −→ · · · −→ (2, 3) −→ (2, 2) −→ ∗ −→ (3, 1) ,
where in each case the asterisk stands for one of two possible options.
We can associate in a two-one way the walks of Type 3a to a rook walk on the last n columns, namely

(1, 2) −→ (2, 2) −→ (2, 3) −→ · · · −→ (3, 3) −→ (3, 2)

and the walks of Type 3b to a rook walk on the last n columns, namely

(1, 2) −→ (1, 3) −→ · · · −→ (2, 3) −→ (2, 2) −→ (3, 2) .

4
The number of rook walks of the latter two types together is rn − 1 − rn−1 . From the number of rook walks
on the last n columns, we subtract one for (1, 2) → (2, 2) → (3, 2) and rn−1 for those of the type

(1, 2) −→ (1, 3) −→ · · · −→ (3, 3) −→ (2, 3) .

Therefore, the number of rook walks of Type 3 is 2(rn − 1 − rn−1 ) and we find that

rn+1 = 4 + 3rn−1 + 2(rn − 1 − rn−1 ) = 2 + 2rn + rn−1 .

We can now complete the solution as in Solution 1.

Solution 3. Let S(3, n) be the set of self-avoiding rook walks in which the rook occupies column n but
does not occupy column n + 1. Then R(3, n) = |S(3, 1)| + |S(3, 2)| + · · · + |S(3, n)|. Furthermore, topological
considerations allow us to break S(3, n) into three disjoint subsets S1 (3, n), the set of paths in which corner
(1, n) is not occupied, but there is a path segment (2, n) −→ (3, n); S2 (3, n), the set of paths in which corners
(1, n) and (3, n) are both occupied by a path (1, n) −→ (2, n) −→ (3, n); and S3 (3, n), the set of paths in
which corner (3, n) is not occupied but there is a path segment (1, n) −→ (2, n). Let si (n) = |Si (3, n)| for
i = 1, 2, 3. Note that s1 (1) = 0, s2 (1) = 1 and s3 (1) = 0. By symmetry, s1 (n) = s3 (n) for every positive n.
Furthermore, we can construct paths in S(3, n + 1) by “bulging” paths in S(3, n), from which we obtain

s1 (n + 1) = s1 (n) + s2 (n) ;
s2 (n + 1) = s1 (n) + s2 (n) + s3 (n) ;
s3 (n + 1) = s2 (n) + s3 (n) ;

or, upon simplification,

s1 (n + 1) = s1 (n) + s2 (n) ;
s2 (n + 1) = 2s1 (n) + s2 (n) .
Hence, for n ≥ 2,
s1 (n + 1) = s1 (n) + 2s1 (n − 1) + s2 (n − 1)
= s1 (n) + 2s1 (n − 1) + s1 (n) − s1 (n − 1)
= 2s1 (n) + s1 (n − 1) .
and
s2 (n + 1) = 2s1 (n) + s2 (n) = 2s1 (n − 1) + 2s2 (n − 1) + s2 (n)
= s2 (n) − s2 (n − 1) + 2s2 (n − 1) + s2 (n)
= 2s2 (n) + s2 (n − 1) .

We find that
1 √ 1 √
s1 (n) = √ (1 + 2)n−1 − √ (1 − 2)n−1 ;
2 2 2 2
1 √ n−1 1 √
s2 (n) = (1 + 2) + (1 − 2)n−1 .
2 2
Summing a geometric series yields that 10

R(3, n) = (s2 (1) + · · · + s2 (n)) + 2(s1 (1) + · · · + s1 (n))

� �� √ � � �� √ �
1 1 (1 + 2)n − 1 1 1 (1 − 2)n − 1
 = 2s1 (n) + s1 (n − 1) .
and
s2 (n + 1) = 2s1 (n) + s2 (n) = 2s1 (n − 1) + 2s2 (n − 1) + s2 (n)
= s2 (n) − s2 (n − 1) + 2s2 (n − 1) + s2 (n)
Report -=Fortieth
2s2 (n) +Canadian
s2 (n − 1) Mathematical
. Olympiad 2008

We find that
1 √ 1 √
s1 (n) = √ (1 + 2)n−1 − √ (1 − 2)n−1 ;
2 2 2 2
1 √ n−1 1 √
s2 (n) = (1 + 2) + (1 − 2)n−1 .
2 2
Summing a geometric series yields that

R(3, n) = (s2 (1) + · · · + s2 (n)) + 2(s1 (1) + · · · + s1 (n))

� �� √ � � �� √ �
1 1 (1 + 2)n − 1 1 1 (1 − 2)n − 1
= +√ √ + −√ √
2 2 2 2 2 − 2
� �
1 √ √
= √ [(1 + 2)n+1 − (1 − 2)n+1 ] − 1 .
2 2
The formula agrees with R(3, 1) = 1, R(3, 2) = 4 and R(3, 3) = 11.

Acknowledgment. The first two solutions are due to Man-Duen Choi, and the third to Ed Doolittle.

11
 41st Canadian Mathematical Olympiad
Wednesday, March 25, 2009

Problem 1. Given an m × n grid with squares coloured either black or white, we say
that a black square in the grid is stranded if there is some square to its left in the same
row that is white and there is some square above it in the same column that is white (see
Figure 1.).

Figure 1. A 4 × 5 grid with no stranded black squares

Find a closed formula for the number of 2 × n grids with no stranded black squares.

Problem 2. Two circles of different radii are cut out of cardboard. Each circle is subdi-
vided into 200 equal sectors. On each circle 100 sectors are painted white and the other
100 are painted black. The smaller circle is then placed on top of the larger circle, so that
their centers coincide. Show that one can rotate the small circle so that the sectors on
the two circles line up and at least 100 sectors on the small circle lie over sectors of the
same color on the big circle.

Problem 3. Define
(xy + yz + zx)(x + y + z)
f (x, y, z) = .
(x + y)(x + z)(y + z)
Determine the set of real numbers r for which there exists a triplet (x, y, z) of positive
real numbers satisfying f (x, y, z) = r.

Problem 4. Find all ordered pairs (a, b) such that a and b are integers and 3a + 7b is a
perfect square.

Problem 5. A set of points is marked on the plane, with the property that any three
marked points can be covered with a disk of radius 1. Prove that the set of all marked
points can be covered with a disk of radius 1.
 Report - Forty First Canadian Mathematical Olympiad 2009

CANADIAN MATHEMATICAL OLYMPIAD 2009

SOLUTIONS

Problem 1. Given an m × n grid with squares coloured either black or white, we say
that a black square in the grid is stranded if there is some square to its left in the same
row that is white and there is some square above it in the same column that is white (see
Figure).

Figure 1. A 4 × 5 grid with no stranded black squares

Find a closed formula for the number of 2 × n grids with no stranded black squares.

Solution. There is no condition for squares in the first row. A square in the second row
can be black only if the square above it is black or all squares to the left of it are black.
Suppose the first k squares in the second row are black and the (k + 1)-st square is white
or k = n. When k < n then for each of the first k + 1 squares in the first row we have 2
choices, and for each of the remaining n − k − 1 columns we have 3 choices. When k = n,
there are 2n choices for the first row. The total number of choices is thus:
n−1

2k+1 3n−k−1 + 2n .
k=0
This expression simplifies to
2 · 3n − 2n .



 Report - Forty First Canadian Mathematical Olympiad 2009

2 CANADIAN MATHEMATICAL OLYMPIAD 2009 SOLUTIONS

Problem 2. Two circles of different radii are cut out of cardboard. Each circle is subdi-
vided into 200 equal sectors. On each circle 100 sectors are painted white and the other
100 are painted black. The smaller circle is then placed on top of the larger circle, so that
their centers coincide. Show that one can rotate the small circle so that the sectors on
the two circles line up and at least 100 sectors on the small circle lie over sectors of the
same color on the big circle.

Solution. Let x0 , . . . , x199 be variables. Assign the value of +1 or −1 to xi depending on

whether the (i + 1)st segment of the larger circle (counting counterclockwise) is black or
white, respectively. Similarly, assign the value of +1 or −1 to the variable yi depending
on whether the (i + 1)th segment of the smaller circle is black or white. We can now
restate the problem in the following equivalent way: show that
200

Sj = xi yi+j ≥ 0 ,
i=1
for some j = 0, . . . , 199. Here the subscript i + j is understood modulo 200.
Now observe that y0 + · · · + y199 = 0 and thus
199

S0 + · · · + S199 = xi (y0 + · · · + y199 ) = 0 .
I=0
Thus Sj ≥ 0 for some j = 0, . . . , 199, as claimed. 


 Report - Forty First Canadian Mathematical Olympiad 2009

2 CANADIAN
CANADIAN MATHEMATICAL
MATHEMATICAL OLYMPIAD
OLYMPIAD 2009 2009 SOLUTIONS
SOLUTIONS 3

Problem
Problem 2. 3. Two
Define circles of different radii are cut out of cardboard. Each circle is subdi-
vided into 200 equal sectors. On each(xy + yz
circle 100+ zx)(x
sectors+arey +painted
z) white and the other
f (x, y, z) = .
100 are painted black. The smaller circle(xis + then
y)(xplaced
+ z)(yon+topz) of the larger circle, so that
their
Determine the set of real numbers r for which there exists circle
centers coincide. Show that one can rotate the small so that
a triplet (x, y,the
z) sectors on
of positive
the
realtwo circlessatisfying
numbers line up and f (x,aty,least
z) = 100
r. sectors on the small circle lie over sectors of the
same color on the big circle.
Solution. We prove that 1 < f (x, y, z) ≤ 98 , and that f (x, y, z) can take on any value
Solution.
within the Let x0 ,(1,
range . . .9,].x199 be variables. Assign the value of +1 or −1 to xi depending on
8
whether the (i +for
The expression 1)st segment
f (x, y, z) can of the larger circle
be simplified to (counting counterclockwise) is black or
white, respectively. Similarly, assign the value of +1 xyzor −1 to the variable yi depending
f (x,
on whether the (i + 1)th segment of the (x y, z) = 1 + smaller .
+ y)(xcircle+ z)(y is +
black
z) or white. We can now
restate the problem in the following equivalent way: show that
Since x, y, z are positive, we get 1 < f (x, 200
y, z).
The inequality f (x, y, z) ≤ 98 can be simplified  to
Sj = xi yi+j ≥ 0 ,
x2 y + x2 z + y 2 x + i=1 y 2 z + z 2 x + z 2 y − 6xyz ≥ 0.
for some j =
Rearrange 0, left
the . . . , hand
199. Here side as thefollows:
subscript i + j is understood modulo 200.
Now observe that2 y0 + ·2· · + y2199 = 02 and 2thus 2
x y + x z + y x + y z + z x + z y − 6xyz =
199

x(y 2 +Sz02+ ) −· ·2xyz
· + S199 = + x
+ y(x 2
z 2i (y
)− 2xyz + z(x2 + y 2 ) − 2xyz =
0 + · · · + y199 ) = 0 .
x(y − z)2 + y(x − z)2 + z(x I=0 − y) .
2

Thus
This S j ≥ 0 for some
expression is clearly j = non-negative
0, . . . , 199, as when
claimed. x, y, z are non-negative. 
9
To prove that f (x, y, z) takes any values in the interval (1, 8 ], define
t
g(t) = f (t, 1, 1) = 1 + .
2(1 + t)2
Then g(1) = 98 and g(t) approaches 1 as t approaches 0. It follows from the continuity
of g(t) for 0 < t ≤ 1 that it takes all values in the interval (1, 98 ]. (Alternatively, one
can check that the quadratic equation g(t) = r has a solution t for any number r in the
interval (1, 98 ].) 


 Report - Forty First Canadian Mathematical Olympiad 2009

4 CANADIAN MATHEMATICAL OLYMPIAD 2009 SOLUTIONS

Problem 4. Find all ordered pairs (a, b) such that a and b are integers and 3a + 7b is a
perfect square.

Solution. It is obvious that a and b must be non-negative.

Suppose that 3a + 7b = n2 . We can assume that n is positive. We first work modulo 4.
Since 3a + 7b = n2 , it follows that
n2 ≡ (−1)a + (−1)b (mod 4).
Since no square can be congruent to 2 modulo 4, it follows that we have either (i) a is
odd and b is even or (ii) a is even and b is odd.
Case (i): Let b = 2c. Then
3a = (n − 7c )(n + 7c ).
It cannot be the case that 3 divides both n − 7c and n + 7c . But each of these is a power
of 3. It follows that n − 7c = 1, and therefore
3a = 2 · 7c + 1.
If c = 0, then a = 1, and we obtain the solution a = 1, b = 0. So suppose that c ≥ 1.
Then 3a ≡ 1 (mod 7). This is impossible, since the smallest positive value of a such that
3a ≡ 1 (mod 7) is given by a = 6, and therefore all a such that 3a ≡ 1 (mod 7) are even,
contradicting the fact that a is odd.
Case (ii): Let a = 2c. Then
7b = (n − 3c )(n + 3c ).
Thus each of n − 3c and n + 3c is a power of 7. Since 7 cannot divide both of these, it
follows that n − 3c = 1, and therefore
7b = 2 · 3c + 1.
Look first at the case c = 1. Then b = 1, and we obtain the solution a = 2, b = 1. So
from now on we may assume that c > 1. Then 7b ≡ 1 (mod 9). The smallest positive
integer b such that 7b ≡ 1 (mod 9) is given by b = 3. It follows that b must be a multiple
of 3. Let b = 3d. Note that d is odd, so in particular d ≥ 1.
Let y = 7d . Then y 3 − 1 = 2 · 3c , and therefore
2 · 3c = (y − 1)(y 2 + y + 1).
It follows that y − 1 = 2 · 3u for some positive u, and that y 2 + y + 1 = 3v for some v ≥ 2.
But since
3y = (y 2 + y + 1) − (y − 1)2 ,
it follows that 3 | y, which is impossible since 3 | (y − 1). 

10
 Report - Forty First Canadian Mathematical Olympiad 2009

2 CANADIAN MATHEMATICAL OLYMPIAD 2009 SOLUTIONS

CANADIAN MATHEMATICAL OLYMPIAD 2009 SOLUTIONS 5

Problem 2. Two circles of different radii are cut out of cardboard. Each circle is subdi-
Problem 5. A set of points is marked on the plane, with the property that any three
vided into 200 equal sectors. On each circle 100 sectors are painted white and the other
marked points can be covered with a disk of radius 1. Prove that the set of all marked
100 are painted black. The smaller circle is then placed on top of the larger circle, so that
points can be covered with a disk of radius 1.
their centers coincide. Show that one can rotate the small circle so that the sectors on
the two circles line up and at least 100 sectors on the small circle lie over sectors of the
Solution. (For a finite set of points only.) Let D be a disk of smallest radius that covers
same color on the big circle.
all marked points. Consider the marked points on the boundary C of this disk. Note that
if all marked points on C lie on an arc smaller than the half circle (ASTTHC for short),
Solution. Let x0 , . . . , x199 be variables. Assign the value of +1 or −1 to xi depending on
then the disk can be moved a little towards these points on the boundary and its radius
whether the (i + 1)st segment of the larger circle (counting counterclockwise) is black or
can be decreased. Since we assumed that our disk has minimal radius, the marked points
white, respectively. Similarly, assign the value of +1 or −1 to the variable yi depending
on its boundary do not lie on an ASTTHC.
on whether the (i + 1)th segment of the smaller circle is black or white. We can now
If the two endpoints of a diagonal of D are marked, then D is the smallest disk containing
restate the problem in the following equivalent way: show that
these two points, hence must have radius at most 1.
200
If there are 3 marked points on C that  do not lie on an ASTTHC, then D is the smallest
disk covering these 3 points and hence must S j = xi yi+j ≥ 0radius
have , at most 1. (In this case the
i=1
triangle formed by the three points is acute and C is its circumcircle.)
for some are
If there j =more
0, . . .than
, 199.3 Here
marked the points
subscript i + boundary
on the j is understood
that domodulo
not lie200.
on an ASTTHC,
Now observe that
then we can remove y + · · · +
0 one of themy 199 = 0 and thus
so that the remaining points again do not lie on an
ASTTHC. By induction this leads us to 199the case of 3 points. Indeed, given 4 or more

points on C, choose S + · · · +that

3 0 points S199 lie
= on axi (y · · · + y199
0 +circle.
half ) = the
Then 0 . middle point can be
removed. I=0
Thus Sj ≥ 0 for some j = 0, . . . , 199, as claimed. 


11
 42nd Canadian Mathematical Olympiad
Wednesday, March 24, 2010

(1) For a positive integer n, an n-staircase is a figure consisting of unit squares, with one
square in the first row, two squares in the second row, and so on, up to n squares in
the nth row, such that all the left-most squares in each row are aligned vertically. For
example, the 5-staircase is shown below.

Let f (n) denote the minimum number of square tiles required to tile the n-staircase,
where the side lengths of the square tiles can be any positive integer. For example,
f (2) = 3 and f (4) = 7.

(a) Find all n such that f (n) = n.

(b) Find all n such that f (n) = n + 1.
(2) Let A, B, P be three points on a circle. Prove that if a and b are the distances from P
to the tangents at A and B and c is the distance from P to the chord AB, then c2 = ab.
(3) Three speed skaters have a friendly “race” on a skating oval. They all start from the
same point and skate in the same direction, but with different speeds that they maintain
throughout the race. The slowest skater does 1 lap a minute, the fastest one does 3.14
laps a minute, and the middle one does L laps a minute for some 1 < L < 3.14. The race
ends at the moment when all three skaters again come together to the same point on
the oval (which may differ from the starting point.) Find how many different choices for
L are there such that exactly 117 passings occur before the end of the race. (A passing
is defined when one skater passes another one. The beginning and the end of the race
when all three skaters are together are not counted as passings.)
(4) Each vertex of a finite graph can be coloured either black or white. Initially all vertices
are black. We are allowed to pick a vertex P and change the colour of P and all of its

1
 neighbours. Is it possible to change the colour of every vertex from black to white by a
sequence of operations of this type?
(A finite graph consists of a finite set of vertices and a finite set of edges between
vertices. If there is an edge between vertex A and vertex B, then B is called a neighbour
of A.)
(5) Let P (x) and Q(x) be polynomials with integer coefficients. Let an = n! + n. Show that
if P (an )/Q(an ) is an integer for every n, then P (n)/Q(n) is an integer for every integer
n such that Q(n) 6= 0.

2
 CANADIAN MATHEMATICAL OLYMPIAD 2010
PROBLEMS AND SOLUTIONS

(1) For a positive integer n, an n-staircase is a figure consisting of unit squares, with
one square in the first row, two squares in the second row, and so on, up to n
squares in the nth row, such that all the left-most squares in each row are aligned
vertically. For example, the 5-staircase is shown below.

Let f (n) denote the minimum number of square tiles required to tile the n-
staircase, where the side lengths of the square tiles can be any positive integer.
For example, f (2) = 3 and f (4) = 7.

(a) Find all n such that f (n) = n.

(b) Find all n such that f (n) = n + 1.

Solution. (a) A diagonal square in an n-staircase is a unit square that lies on

the diagonal going from the top-left to the bottom-right. A minimal tiling of an
n-staircase is a tiling consisting of f (n) square tiles.
Observe that f (n) ≥ n for all n. There are n diagonal squares in an n-staircase,
and a square tile can cover at most one diagonal square, so any tiling requires at
least n square tiles. In other words, f (n) ≥ n. Hence, if f (n) = n, then each
square tile covers exactly one diagonal square.
Let n be a positive integer such that f (n) = n, and consider a minimal tiling of
an n-staircase. The only square tile that can cover the unit square in the first row
is the unit square itself.
Now consider the left-most unit square in the second row. The only square tile
that can cover this unit square and a diagonal square is a 2 × 2 square tile.

1
 Next, consider the left-most unit square in the fourth row. The only square tile
that can cover this unit square and a diagonal square is a 4 × 4 square tile.

Continuing this construction, we see that the side lengths of the square tiles
we encounter will be 1, 2, 4, and so on, up to 2k for some nonnegative integer k.
Therefore, n, the height of the n-staircase, is equal to 1+2+4+· · ·+2k = 2k+1 −1.
Alternatively, n = 2k − 1 for some positive integer k. Let p(k) = 2k − 1.
Conversely, we can tile a p(k)-staircase with p(k) square tiles recursively as
follows: We have that p(1) = 1, and we can tile a 1-staircase with 1 square tile.
Assume that we can tile a p(k)-staircase with p(k) square tiles for some positive
integer k.
Consider a p(k + 1)-staircase. Place a 2k × 2k square tile in the bottom left
corner. Note that this square tile covers a digaonal square. Then p(k + 1) − 2k =
2k+1 − 1 − 2k = 2k − 1 = p(k), so we are left with two p(k)-staircases.

p(k)

2k

2k p(k)

Furthermore, these two p(k)-staircases can be tiled with 2p(k) square tiles, which
means we use 2p(k) + 1 = p(k + 1) square tiles.
Therefore, f (n) = n if and only if n = 2k − 1 = p(k) for some positive integer
k. In other words, the binary representation of n consists of all 1s, with no 0s.
(b) Let n be a positive integer such that f (n) = n + 1, and consider a minimal
tiling of an n-staircase. Since there are n diagonal squares, every square tile
except one covers a diagonal square. We claim that the square tile that covers
the bottom-left unit square must be the square tile that does not cover a diagonal
square.
If n is even, then this fact is obvious, because the square tile that covers the
bottom-left unit square cannot cover any diagonal square, so assume that n is odd.
Let n = 2m + 1. We may assume that n > 1, so m ≥ 1. Suppose that the square
tile covering the bottom-left unit square also covers a diagonal square. Then the
side length of this square tile must be m + 1. After this (m + 1) × (m + 1) square
tile has been placed, we are left with two m-staircases.
2
 m

m+1

m+1 m
Hence, f (n) = 2f (m) + 1. But 2f (m) + 1 is odd, and n + 1 = 2m + 2 is even,
so f (n) cannot be equal to n + 1, contradiction. Therefore, the square tile that
covers the bottom-left unit square is the square tile that does not cover a diagonal
square.
Let t be the side length of the square tile covering the bottom-left unit square.
Then every other square tile must cover a diagonal square, so by the same con-
struction as in part (a), n = 1 + 2 + 4 + · · · + 2k−1 + t = 2k + t − 1 for some positive
integer k. Furthermore, the top p(k) = 2k − 1 rows of the n-staircase must be tiled
the same way as the minimal tiling of a p(k)-staircase. Therefore, the horizontal
line between rows p(k) and p(k) + 1 does not pass through any square tiles. Let
us call such a line a fault line. Similarly, the vertical line between columns t and
t + 1 is also a fault line. These two fault lines partition two p(k)-staircases.

p(k)

t p(k)

If these two p(k)-staircases do not overlap, then t = p(k), so n = 2p(k). For

example, the minimal tiling for n = 2p(2) = 6 is shown below.

Hence, assume that the two p(k)-staircases do overlap. The intersection of the
two p(k)-staircases is a [p(k) − t]-staircase. Since this [p(k) − t]-staircase is tiled
the same way as the top p(k) − t rows of a minimal tiling of a p(k)-staircase,
p(k) − t = p(l) for some positive integer l < k, so t = p(k) − p(l). Then

n = t + p(k) = 2p(k) − p(l).

Since p(0) = 0, we can summarize by saying that n must be of the form

n = 2p(k) − p(l) = 2k+1 − 2l − 1,

3
 where k is a positive integer and l is a nonnegative integer. Also, our argument
shows how if n is of this form, then an n-staircase can be tiled with n + 1 square
tiles.
Finally, we observe that n is of this form if and only if the binary representation
of n contains exactly one 0:
2k+1 − 2l − 1 = 11 . . . 1} 0 |11 {z
| {z . . . 1} .
k − l 1s l 1s

(2) Let A, B, P be three points on a circle. Prove that if a and b are the distances
from P to the tangents at A and B and c is the distance from P to the chord AB,
then c2 = ab.

Solution. Let r be the radius of the circle, and let a’ and b’ be the respective
lengths of P A and P B. Since b0 = 2r sin ∠P AB = 2rc/a0 , c = a0 b0 /(2r). Let AC
be the diameter of the circle and H the foot of the perpendicular from P to AC.
The similarity of the triangles ACP and AP H imply that AH : AP = AP : AC
or (a0 )2 = 2ra. Similarly, (b0 )2 = 2rb. Hence
(a0 )2 (b0 )2
c2 = = ab
2r 2r
as desired. 

Alternate Solution. Let E, F, G be the feet of the perpendiculars to the

tangents at A and B and the chord AB, respectively. We need to show that
P E : P G = P G : GF , where G is the foot of the perpendicular from P to AB.
This suggest that we try to prove that the triangles EP G and GP F are similar.
Since P G is parallel to the bisector of the angle between the two tangents,
∠EP G = ∠F P G. Since AEP G and BF P G are concyclic quadrilaterals (having
opposite angles right), ∠P GE = ∠P AE and ∠P F G = ∠P BG. But ∠P AE =
∠P BA = ∠P BG, whence ∠P GE = ∠P F G. Therefore triangles EP G and GP F
are similar.
The argument above with concyclic quadrilaterals only works when P lies on
the shorter arc between A and B. The other case can be proved similarly. 

(3) Three speed skaters have a friendly race on a skating oval. They all start from
the same point and skate in the same direction, but with different speeds that
they maintain throughout the race. The slowest skater does 1 lap a minute, the
fastest one does 3.14 laps a minute, and the middle one does L laps a minute for
some 1 < L < 3.14. The race ends at the moment when all three skaters again
come together to the same point on the oval (which may differ from the starting
4
point.) Find how many different choices for L are there such that 117 passings
occur before the end of the race. (A passing is defined when one skater passes
another one. The beginning and the end of the race when all three skaters are at
together are not counted as a passing.)

Solution. Assume that the length of the oval is one unit. Let x(t) be the
difference of distances that the slowest and the fastest skaters have skated by time
t. Similarly, let y(t) be the difference between the middle skater and the slowest
skater. The path (x(t), y(t)) is a straight ray R in R2 , starting from the origin,
with slope depending on L. By assumption, 0 < y(t) < x(t).
One skater passes another one when either x(t) ∈ Z, y(t) ∈ Z or x(t) − y(t) ∈ Z.
The race ends when both x(t), y(t) ∈ Z.
Let (a, b) ∈ Z2 be the endpoint of the ray R. We need to find the number of
such points satisfying:
(a) 0 < b < a
(b) The ray R intersects Z2 at endpoints only.
(c) The ray R crosses 357 times the lines x ∈ Z, y ∈ Z, y − x ∈ Z.
The second condition says that a and b are relatively prime. The ray R crosses
a − 1 of the lines x ∈ Z, b − 1 of the lines y ∈ Z and a − b − 1 of the lines x − y ∈ Z.
Thus, we need (a − 1) + (b − 1) + (a − b − 1) = 117, or equivalently, 2a − 3 = 117.
That is a = 60.
Now b must be a positive integer less than and relatively prime to 60. The
number of such b can be found using the Euler’s φ function:
φ(60) = φ(22 · 3 · 5) = (2 − 1) · 2 · (3 − 1) · (5 − 1) = 16.
Thus the answer is 16. 

Alternate Solution. First, let us name our skaters. From fastest to slowest,
call them: A, B and C. (Abel, Bernoulli and Cayley?)
Now, it is helpful to consider the race from the viewpoint of C. Relative to C,
both A and B complete a whole number of laps, since they both start and finish
at C.
Let n be the number of laps completed by A relative to C, and let m be the
number of laps completed by B relative to C. Note that: n > m ∈ Z+
Consider the number of minutes required to complete the race. Relative to C,
A is moving with a speed of 3.14 − 1 = 2.14 laps per minute and completes the
n
race in 2.14 minutes. Also relative to C, B is moving with a speed of (L − 1) laps
m
per minute and completes the race in L−1 minutes. Since A and B finish the race
together (when they both meet C):
n m m
= ⇒ L = 2.14 + 1.
2.14 L−1 n
Hence, there is a one-to-one relation between values of L and values of the postive
proper fraction mn
. The fraction should be reduced, that is the pair (m, n) should
5
 be relatively prime, or else, with k = gcd(m, n), the race ends after n/k laps for
A and m/k laps for B when they first meet C together.
It is also helpful to consider the race from the viewpoint of B. In this frame
of reference, A completes only n − m laps. Hence A passes B only (n − m) − 1
times, since the racers do not ”pass” at the end of the race (nor at the beginning).
Similarily A passes C only n − 1 times and B passes C only m − 1 times. The
total number of passings is:
117 = (n − 1) + (m − 1) + (n − m − 1) = 2n − 3 ⇒ n = 60
m
Hence the number of values of L equals the number of m for which the fraction 60 is
positive, proper and reduced. That is the number of positive integer values smaller
than and relatively prime to 60. One could simply count: {1,7,11,13,17,...}, but
Euler’s φ function gives this number:
φ(60) = φ(22 · 3 · 5) = (2 − 1) · 2 · (3 − 1) · (5 − 1) = 16.
Therefore, there are 16 values for L which give the desired number of passings.
Note that the actual values for the speeds of A and C do not affect the result.
They could be any values, rational or irrational, just so long as they are different,
and there will be 16 possible values for the speed of B between them. 

(4) Each vertex of a finite graph can be colored either black or white. Initially all
vertices are black. We are allowed to pick a vertex P and change the color of P
and all of its neighbours. Is it possible to change the colour of every vertex from
black to white by a sequence of operations of this type?

Solution. The answer is yes. Proof by induction on the number n of vertices.

If n = 1, this is obvious. For the induction assumption, suppose we can do this for
any graph with n − 1 vertices for some n ≥ 2 and let X be a graph with n vertices
which we will denote by P1 , . . . , Pn+1 .
Let us denote the “basic” operation of changing the color of Pi and all of its
neighbours by fi . Removing a vertex Pi from X (along with all edges connecting
to Pi ) and applying the induction assumption to the resulting smaller graph, we
see that there exists a sequence of operations gi (obtained by composing some fj ,
with j 6= i) which changes the colour of every vertex in X, except for possibly Pi .
If gi it also changes the color of Pi then we are done. So, we may assume that
gi does not change the colour of P for every i = 1, . . . , n. Now consider two cases.
Case 1: n is even. Then composing g1 , . . . , gn we will change the color of every
vertex from white to black.
Case 2: n is odd. I claim that in this case X has a vertex with an even number
of neighbours.
Indeed, denote the number of neighbours of Pi (or equivalently, the number of
edges connected to P ) by ki . Then P1 + · · · + Pn+1 = 2e, where e is the number
of edges of X. Thus one of the numbers ki has to be even as claimed.
6
 After renumbering the vertices, we may assume that P1 has 2k neighbours, say
P2 , . . . , P2k+1 . The composition of f1 with g1 , g2 , . . . , g2k+1 will then change the
colour of every vertex, as desired.


(5) Let P (x) and Q(x) be polynomials with integer coefficients. Let an = n! + n.
Show that if P (an )/Q(an ) is an integer for every n, then P (n)/Q(n) is an integer
for every integer n such that Q(n) 6= 0.

Solution. Imagine dividing P (x) by Q(x). We find that

P (x) R(x)
= A(x) + ,
Q(x) Q(x)
where A(x) and R(x) are polynomials with rational coefficients, and R(x) is either
identically 0 or has degree less than the degree of Q(x).
By bringing the coefficients of A(x) to their least common multiple, we can find
a polynomial B(x) with integer coefficients, and a positive integer b, such that
A(x) = B(x)/b. Suppose first that R(x) is not identically 0. Note that for any
integer k, either A(k) = 0, or |A(k)| ≥ 1/b. But whenever |k| is large enough,
0 < |R(k)/Q(k)| < 1/b, and therefore if n is large enough, P (an )/Q(an ) cannot
be an integer.
So R(x) is identically 0, and P (x)/Q(x) = B(x)/b (at least whenever Q(x) 6= 0.)
Now let n be an integer. Then there are infinitely many integers k such that
n ≡ ak (mod b). But B(ak )/b is an integer, or equivalently b divides B(ak ). It
follows that b divides B(n), and therefore P (n)/Q(n) is an integer. 

7
 43rd Canadian Mathematical Olympiad
Wednesday, March 23, 2011

(1) Consider 70-digit numbers n, with the property that each of the digits 1, 2, 3, . . . , 7
appears in the decimal expansion of n ten times (and 8, 9, and 0 do not appear).
Show that no number of this form can divide another number of this form.

(2) Let ABCD be a cyclic quadrilateral whose opposite sides are not parallel, X the
intersection of AB and CD, and Y the intersection of AD and BC. Let the
angle bisector of ∠AXD intersect AD, BC at E, F respectively and let the angle
bisector of ∠AY B intersect AB, CD at G, H respectively. Prove that EGF H is
a parallelogram.

(3) Amy has divided a square up into finitely many white and red rectangles, each
with sides parallel to the sides of the square. Within each white rectangle, she
writes down its width divided by its height. Within each red rectangle, she writes
down its height divided by its width. Finally, she calculates x, the sum of these
numbers. If the total area of the white rectangles equals the total area of the red
rectangles, what is the smallest possible value of x?

(4) Show that there exists a positive integer N such that for all integers a > N , there
exists a contiguous substring of the decimal expansion of a that is divisible by 2011.
(For instance, if a = 153204, then 15, 532, and 0 are all contiguous substrings of
a. Note that 0 is divisible by 2011.)

(5) Let d be a positive integer. Show that for every integer S, there exists an integer
n > 0 and a sequence ²1 , ²2 , . . . , ²n , where for any k, ²k = 1 or ²k = −1, such that
S = ²1 (1 + d)2 + ²2 (1 + 2d)2 + ²3 (1 + 3d)2 + · · · + ²n (1 + nd)2 .

1
 43rd Canadian Mathematical Olympiad
Wednesday, March 23, 2011

Problems and Solutions

(1) Consider 70-digit numbers n, with the property that each of the digits 1, 2, 3, . . . , 7
appears in the decimal expansion of n ten times (and 8, 9, and 0 do not appear). Show
that no number of this form can divide another number of this form.

Solution. Assume the contrary: there exist a and b of the prescribed form, such that
b ≥ a and a divides b. Then a divides b − a.
Claim: a is not divisible by 3 but b − a is divisible by 9. Indeed, the sum of the digits
is 10(1 + · · · + 7) = 280, for both a and b. [Here one needs to know or prove that an
integer n is equivalent of the sum of its digits modulo 3 and modulo 9.]
We conclude that b − a is divisible by 9a. But this is impossible, since 9a has 71 digits
and b has only 70 digits, so 9a > b > b − a. ¤

(2) Let ABCD be a cyclic quadrilateral whose opposite sides are not parallel, X the inter-
section of AB and CD, and Y the intersection of AD and BC. Let the angle bisector
of ∠AXD intersect AD, BC at E, F respectively and let the angle bisector of ∠AY B
intersect AB, CD at G, H respectively. Prove that EGF H is a parallelogram.

Solution. Since ABCD is cyclic, ∆XAC ∼ ∆XDB and ∆Y AC ∼ ∆Y BD. There-

fore,
XA XC AC YA YC
= = = = .
XD XB DB YB YD
Let s be this ratio. Therefore, by the angle bisector theorem,
AE XA XC CF
= = = = s,
ED XD XB FB
and
AG YA YC CH
= = = = s.
GB YB YD HD
AG
Hence, GB = FCF AE
B and ED =
DH
HC . Therefore, EH||AC||GF and EG||DB||HF . Hence,
EGF H is a parallelogram. ¤

1
(3) Amy has divided a square up into finitely many white and red rectangles, each with sides
parallel to the sides of the square. Within each white rectangle, she writes down its width
divided by its height. Within each red rectangle, she writes down its height divided by
its width. Finally, she calculates x, the sum of these numbers. If the total area of the
white rectangles equals the total area of the red rectangles, what is the smallest possible
value of x?

Solution. Let ai and bi denote the width and height of each white rectangle, and let
ci and di denote the width and height of each red rectangle. Also, let L denote the side
length of the original square.
P P
Lemma: Either ai ≥ L or di ≥ L.

Proof of lemma: Suppose there exists a horizontal line across the square that is
covered entirely with white rectangles. Then, the total width of these rectangles is at
least L, and the claim is proven. Otherwise, there is a red rectangle intersecting every
horizontal line, and hence the total height of these rectangles is at least L. ¤
P
Now, let us assume without loss of generality that ai ≥ L. By the Cauchy-Schwarz
inequality,
µX ¶ ³X ´ ³X ´2
ai
· ai bi ≥ ai
bi
≥ L2 .

P L2 P ai
But we know ai bi = 2 , so it follows that bi ≥ 2. Furthermore, each ci ≤ L, so

X di 1 X 1
≥ 2
· ci di = .
ci L 2

Therefore, x is at least 2.5. Conversely, x = 2.5 can be achieved by making the top half
of the square one colour, and the bottom half the other colour. ¤

(4) Show that there exists a positive integer N such that for all integers a > N , there exists
a contiguous substring of the decimal expansion of a that is divisible by 2011. (For
instance, if a = 153204, then 15, 532, and 0 are all contiguous substrings of a. Note that
0 is divisible by 2011.)

Solution. We claim that if the decimal expansion of a has at least 2012 digits, then
a contains the required substring. Let the decimal expansion of a be ak ak−1 . . . a0 . For
i = 0, . . . , 2011, Let bi be the number with decimal expansion ai ai−1 . . . a0 . Then by
pidgenhole principle, bi ≡ bj mod 2011 for some i < j ≤ 2011. It follows that 2011
divides bj − bi = c · 10i . Here c is the substring aj . . . ai+1 . Since 2011 and 10 are
relatively prime, it follows that 2011 divides c. ¤

2
(5) Let d be a positive integer. Show that for every integer S, there exists an integer n > 0
and a sequence ²1 , ²2 , . . . , ²n , where for any k, ²k = 1 or ²k = −1, such that
S = ²1 (1 + d)2 + ²2 (1 + 2d)2 + ²3 (1 + 3d)2 + · · · + ²n (1 + nd)2 .

Solution. Let Uk = (1 + kd)2 . We calculate Uk+3 − Uk+2 − Uk+1 + Uk . This turns

out to be 4d2 , a constant. Changing signs, we obtain the sum −4d2 .
Thus if we have found an expression for a certain number S0 as a sum of the desired
type, we can obtain an expression of the desired type for S0 + (4d2 )q, for any integer q.
It remains to show that for any S, there exists an integer S 0 such that S 0 ≡ S
(mod 4d2 ) and S 0 can be expressed in the desired form. Look at the sum
(1 + d)2 + (1 + 2d)2 + · · · + (1 + N d)2 ,
where N is “large.” We can at will choose N so that the sum is odd, or so that the sum
is even.
By changing the sign in front of (1 + kd)2 to a minus sign, we decrease the sum by
2(1 + kd)2 . In particular, if k ≡ 0 (mod 2d), we decrease the sum by 2 (modulo 4d2 ). So
If N is large enough, there are many k < N such that k is a multiple of 2d. By
switching the sign in front of r of these, we change (“downward”) the congruence class
modulo 4d2 by 2r. By choosing N so that the original sum is odd, and choosing suitable
r < 2d2 , we can obtain numbers congruent to all odd numbers modulo 4d2 . By choosing
N so that the original sum is even, we can obtain numbers congruent to all even numbers
modulo 4d2 . This completes the proof. ¤

3
 44th CANADIAN MATHEMATICAL OLYMPIAD 
44ème OLYMPIADE MATHÉMATIQUE DU CANADA 

1. Let , and be positive real numbers. Show that 4 4.

Soit , et trois nombres réels positifs. Démontrez que 4 4.

2. For any positive integers and , let , be the least common multiple of the consecutive integers
, 1, …., 1. Show that for any integer , there exist integers and such that
, 1, .

Soit , le plus petit commun multiple de la suite des k entiers consécutifs , 1, … . , 1,

où et sont deux entiers positifs quelconques. Montrez que pour tout entier , il existe des nombres entiers
et tels que , 1, .

3. Let be a convex quadrilateral and let be the point of intersection of and . Suppose that
. Prove that the internal angle bisectors of , , and meet at a common point.

Soit un quadrilatère convexe et le point d’intersection des droites et . Supposons que

. Démontrez que les bissectrices des angles internes de , , et se coupent
en un point.

4. A number of robots are placed on the squares of a finite, rectangular grid of squares. A square can hold any number of
robots. Every edge of the grid is classified as either passable or impassable. All edges on the boundary of the grid are
impassable.
You can give any of the commands up, down, left, or right. All of the robots then simultaneously try to move in the
specified direction. If the edge adjacent to a robot in that direction is passable, the robot moves across the edge and
into the next square. Otherwise, the robot remains on its current square. You can then give another command of up,
down, left, or right, then another, for as long as you want.
Suppose that for any individual robot, and any square on the grid, there is a finite sequence of commands that will
move that robot to that square. Prove that you can also give a finite sequence of commands such that all of the robots
end up on the same square at the same time.

Un certain nombre de robots sont placés sur les carrés composant une grille rectangulaire de dimension finie. Chaque
carré peut contenir un nombre quelconque de robots. Les bords des carrés de la grille sont classés comme
franchissable ou infranchissable. Les côtés qui forment le pourtour de la grille sont infranchissables.
Vous pouvez donner n’importe laquelle des commandes suivantes : en haut, en bas, à gauche ou à droite. Tous les
robots tentent alors de se déplacer simultanément dans la direction précisée. Si le bord adjacent au carré vers lequel se
déplace un robot est franchissable, le robot le franchit et se place dans le carré suivant. Sinon, le robot reste dans le
carré où il se trouve. Vous pouvez ensuite lancer une autre commande de déplacement vers le haut, le bas, la gauche
ou la droite, et encore une autre, aussi longtemps que vous le désirez.
Supposons que pour chaque robot, et ce, pour n’importe quel carré, il existe une suite finie de commandes qui
amèneront ce robot au carré donné. Démontrez que vous pouvez aussi lancer une suite finie de commandes de sorte
que tous les robots finiront par se retrouver simultanément dans le même carré.
 44th CANADIAN MATHEMATICAL OLYMPIAD 
44ème OLYMPIADE MATHÉMATIQUE DU CANADA 

5. A bookshelf contains n volumes, labelled 1 to in some order. The librarian wishes to put them in the correct order as
follows. The librarian selects a volume that is too far to the right, say the volume with label , takes it out, and inserts
it so that it is in the -th place. For example, if the bookshelf contains the volumes 1, 3, 2, 4 in that order, the
librarian could take out volume 2 and place it in the second position. The books will then be in the correct order
1, 2, 3, 4.

(a) Show that if this process is repeated, then, however the librarian makes the selections, all the volumes will
eventually be in the correct order.

(b) What is the largest number of steps that this process can take?

Une étagère contient n volumes étiquetés de 1 à , rangés dans un certain ordre. Le bibliothécaire souhaite les mettre
dans le bon ordre de la façon suivante : il choisit un volume qui se trouve trop à droite, par exemple le volume étiqueté
, le retire de son emplacement et l’insère à la –ième place. Par exemple, si les volumes sont rangés dans l’ordre
1, 3, 2, 4, le bibliothécaire peut prendre le volume 2 et le mettre à la deuxième place. Les livres sont alors rangés dans
le bon ordre, soit 1, 2, 3, 4.

a) Démontrez que si l’on répète ce processus, tous les volumes finiront par être dans le bon ordre, et ce, qu’elle que
soit la manière dont le bibliothécaire les range.

b) Quel est le plus grand nombre d’étapes que peut exiger un tel processus?
1. Let x, y and z be positive real numbers. Show that x2 + xy 2 + xyz 2 ≥ 4xyz − 4.

Solution. Note that

x2 ≥ 4x − 4, y 2 ≥ 4y − 4, and z 2 ≥ 4z − 4,

and therefore

x2 + xy 2 + xyz 2 ≥ (4x − 4) + x(4y − 4) + xy(4z − 4) = 4xyz − 4.

2. For any positive integers n and k, let L(n, k) be the least common multiple of the
k consecutive integers n, n + 1, . . . , n + k − 1. Show that for any integer b, there
exist integers n and k such that L(n, k) > b L(n + 1, k).

Solution. I. Let p > b be prime, let n = p3 and k = p2 . If p3 < i < p3 + p2 , then no

power of p greater than 1 divides i, while p divides p3 + p. It follows that L(p3 , p2 ) =
p2 L(p3 + 1, p2 − 1). A similar calculation shows that L(p3 + 1, p2 ) = pL(p3 + 1, p2 − 1).
Thus L(p3 , p2 ) = pL(p3 + 1, p2 ) > bL(p3 + 1, p2 ).
II. Let m > 1. Then L(m!−1, m+1) is the least common multiple of the integers from
m!−1 to m!+m−1. But m!−1 is relatively prime to all of m!, m!+1, . . . , m!+m−1. It
follows that L(m!−1, m+1) = (m!−1)M , where M = lcm(m!, m!+1, . . . , m!+m−1).
Now consider L(m!, m+1). This is lcm(M, m!+m). But m!+m = m((m−1)!+1),
and m divides M . Thus lcm(M, m! + m) ≤ M ((m − 1)! + 1), and

L(m! − 1, m + 1) m! − 1
≥ .
L(m!, m + 1) (m − 1)! + 1

Since m can be arbitrarily large, so can L(m! − 1, m + 1)/L(m!, m + 1). Therefore

taking n = m! − 1 for sufficiently large m, and k = m + 1, works.

3. Let ABCD be a convex quadrilateral and let P be the point of intersection of

AC and BD. Suppose that AC + AD = BC + BD. Prove that the internal angle
bisectors of ∠ACB, ∠ADB, and ∠AP B meet at a common point.

Solution. I. Construct A0 on CA so that AA0 = AD and B 0 on CB such that

BB 0 = BD. Then we have three angle bisectors that correspond to the perpendicular
bisectors of A0 B 0 , A0 D, and B 0 D. These perpendicular bisectors are concurrent, so
the angle bisectors are also concurrent. This tells us that the external angle bisectors
at A and B meet at the excentre of P DB. A symmetric argument for C finishes the
problem.
II. Note that the angle bisectors ∠ACB and ∠AP B intersect at the excentres of
4P BC opposite C and the angle bisectors of ∠ADB and ∠AP B intersect at the
excentres of 4P AD opposite D. Hence, it suffices to prove that these two excentres
coincide.
Let the excircle of 4P BC opposite C touch side P B at a point X, line CP at a
point Y and line CB at a point Z. Hence, CY = CZ, P X = P Y and BX = BZ.
Therefore, CP + P X = CB + BX. Since CP + P X + CB + BX is the perimeter
of 4CBP , CP + P X = CB + BX = s, where s is the semi-perimeter of 4CBP .
Therefore,
s CB + BP + P C CB + BP − P C
P X = CB + BX − CP = − CP = − CP = .
2 2 2
Similarly, if we let the excircle of 4P AD opposite D touch side P A at a point
0
X , then
DA + AP − P D
P X0 = .
2
Since both excircles are tangent to AC and BD, if we show that P X = P X 0 ,
then we would show that the two excircles are tangent to AC and BD at the same
points, i.e. the two excircles are identical. Hence, the two excentres coincide.
We will use the fact that AC + AD = BC + BD to prove that P X = P X 0 . Since
AC +AD = BC +BD, AP +P C +AD = BC +BP +P D. Hence, AP +AD −P D =
BC + BP − P C. Therefore, P X = P X 0 , as desired.

4. A number of robots are placed on the squares of a finite, rectangular grid of

squares. A square can hold any number of robots. Every edge of each square of the
grid is classified as either passable or impassable. All edges on the boundary of the
grid are impassable.
You can give any of the commands up, down, left, or right. All of the robots
then simultaneously try to move in the specified direction. If the edge adjacent to
a robot in that direction is passable, the robot moves across the edge and into the
next square. Otherwise, the robot remains on its current square. You can then give
another command of up, down, left, or right, then another, for as long as you want.
Suppose that for any individual robot, and any square on the grid, there is a
finite sequence of commands that will move that robot to that square. Prove that
you can also give a finite sequence of commands such that all of the robots end up
on the same square at the same time.

Solution. We will prove any two robots can be moved to the same square. From
that point on, they will always be on the same square. We can then similarly move
a third robot onto the same square as these two, and then a fourth, and so on, until
all robots are on the same square.
Towards that end, consider two robots A and B. Let d(A, B) denote the mini-
mum number of commands that need to be given in order to move A to the square
on which B is currently standing. We will give a procedure that is guaranteed to de-
crease d(A, B). Since d(A, B) is a non-negative integer, this procedure will eventually
decrease n to 0, which finishes the proof.
Let n = d(A, B), and let S = {s1 , s2 , . . . , sn } be a minimum sequence of moves
that takes A to the square where B is currently standing. Certainly A will not run
into an impassable edge during this sequence, or we could get a shorter sequence by
removing that command. Now suppose B runs into an impassable edge after some
command si . From that point, we can get A to the square on which B started with
the commands si+1 , si+2 , . . . , sn and then to the square where B is currently with
the commands s1 , s2 , . . . , si−1 . But this was only n − 1 commands in total, and so
we have decreased d(A, B) as required.
Otherwise, we have given a sequence of n commands to A and B, and neither
ran into an impassable edge during the execution of these commands. In particular,
the vector v connecting A to B on the grid must have never changed. We moved
A to the position B = A + v, and therefore we must have also moved B to B + v.
Repeating this process k times, we will move A to A + kv and B to B + kv. But if
v 6= (0, 0), this will eventually force B off the edge of the grid, giving a contradiction.
5. A bookshelf contains n volumes, labelled 1 to n, in some order. The librarian
wishes to put them in the correct order as follows. The librarian selects a volume
that is too far to the right, say the volume with label k, takes it out, and inserts it
in the k-th position. For example, if the bookshelf contains the volumes 1, 3, 2, 4 in
that order, the librarian could take out volume 2 and place it in the second position.
The books will then be in the correct order 1, 2, 3, 4.
(a) Show that if this process is repeated, then, however the librarian makes the
selections, all the volumes will eventually be in the correct order.
(b) What is the largest number of steps that this process can take?
Solution. (a) If tk is the number of times that volume k is selected, then we have
tk ≤ 1 + (t1 + t2 + · · · + tk−1 ). This is because volume k must move to the right
between selections, which means some volume was placed to its left. The only way
that can happen is if a lower-numbered volume was selected. This leads to the bound
tk ≤ 2k−1 . Furthermore, tn = 0 since the nth volume will never be too far to the
right. Therefore if N is the total number of moves then
N = t1 + t2 + · · · + tn−1 ≤ 1 + 2 + · · · + 2n−2 = 2n−1 − 1,
and in particular the process terminates.
(b) Conversely, 2n−1 − 1 moves are required for the configuration (n, 1, 2, 3, . . . , n − 1)
if the librarian picks the rightmost eligible volume each time.
This can be proved by induction: if at a certain stage we are at (x, n − k, n −
k + 1, . . . , n − 1), then after 2k − 1 moves, we will have moved to (n − k, n − k +
1, . . . , n − 1, x) without touching any of the volumes further to the left. Indeed,
after 2k−1 − 1 moves, we get to (x, n − k + 1, n − k + 2, . . . , n − 1, n − k), which
becomes (n − k, x, n − k + 1, n − k + 2, . . . , n − 1) after 1 more move, and then
(n − k, n − k + 1, . . . , n − 1, x) after another 2k−1 − 1 moves. The result follows by
taking k = n − 1.
 45th Canadian Mathematical Olympiad
Wednesday, March 27, 2013

1. Determine all polynomials P (x) with real coefficients such that

(x + 1)P (x − 1) − (x − 1)P (x)
is a constant polynomial.
2. The sequence a1 , a2 , . . . , an consists of the numbers 1, 2, . . . , n in some order. For
which positive integers n is it possible that the n+1 numbers 0, a1 , a1 +a2 , a1 +a2 +a3 ,
. . ., a1 + a2 + · · · + an all have different remainders when divided by n + 1?
3. Let G be the centroid of a right-angled triangle ABC with ∠BCA = 90◦ . Let P
be the point on ray AG such that ∠CP A = ∠CAB, and let Q be the point on ray
BG such that ∠CQB = ∠ABC. Prove that the circumcircles of triangles AQG and
BP G meet at a point on side AB.
4. Let n be a positive integer. For any positive integer j and positive real number
r, define fj (r) and gj (r) by
µ ¶ µ» ¼ ¶
j j
fj (r) = min (jr, n) + min , n , and gj (r) = min (djre, n) + min ,n ,
r r
where dxe denotes the smallest integer greater than or equal to x. Prove that
n
X n
X
fj (r) ≤ n2 + n ≤ gj (r)
j=1 j=1

for all positive real numbers r.

5. Let O denote the circumcentre of an acute-angled triangle ABC. Let point P on
side AB be such that ∠BOP = ∠ABC, and let point Q on side AC be such that
∠COQ = ∠ACB. Prove that the reflection of BC in the line P Q is tangent to the
circumcircle of triangle AP Q.
 45th Canadian Mathematical Olympiad
Wednesday, March 27, 2013

Problems and Solutions

1. Determine all polynomials P (x) with real coefficients such that
(x + 1)P (x − 1) − (x − 1)P (x)
is a constant polynomial.

Solution 1: The answer is P (x) being any constant polynomial and P (x) ≡
kx2 + kx + c for any (nonzero) constant k and constant c.

Let Λ be the expression (x + 1)P (x − 1) − (x − 1)P (x), i.e. the expression in the
problem statement.

Substituting x = −1 into Λ yields 2P (−1) and substituting x = 1 into Λ yield

2P (0). Since (x+1)P (x−1)−(x−1)P (x) is a constant polynomial, 2P (−1) = 2P (0).
Hence, P (−1) = P (0).

Let c = P (−1) = P (0) and Q(x) = P (x) − c. Then Q(−1) = Q(0) = 0. Hence,
0, −1 are roots of Q(x). Consequently, Q(x) = x(x + 1)R(x) for some polynomial R.
Then P (x) − c = x(x + 1)R(x), or equivalently, P (x) = x(x + 1)R(x) + c.

Substituting this into Λ yield

(x + 1)((x − 1)xR(x − 1) + c) − (x − 1)(x(x + 1)R(x) + c)
This is a constant polynomial and simplifies to
x(x − 1)(x + 1)(R(x − 1) − R(x)) + 2c.
Since this expression is a constant, so is x(x − 1)(x + 1)(R(x − 1) − R(x)). Therefore,
R(x − 1) − R(x) = 0 as a polynomial. Therefore, R(x) = R(x − 1) for all x ∈ R.
Then R(x) is a polynomial that takes on certain values for infinitely values of x.
Let k be such a value. Then R(x) − k has infinitely many roots, which can occur
if and only if R(x) − k = 0. Therefore, R(x) is identical to a constant k. Hence,
Q(x) = kx(x+1) for some constant k. Therefore, P (x) = kx(x+1)+c = kx2 +kx+c.

Finally, we verify that all such P (x) = kx(x + 1) + c work. Substituting this into
Λ yield

(x + 1)(kx(x − 1) + c) − (x − 1)(kx(x + 1) + c)
= kx(x + 1)(x − 1) + c(x + 1) − kx(x + 1)(x − 1) − c(x − 1) = 2c.

Hence, P (x) = kx(x + 1) + c = kx2 + kx + c is a solution to the given equation

for any constant k. Note that this solution also holds for k = 0. Hence, constant
polynomials are also solutions to this equation. ¤

Solution 2: As in Solution 1, any constant polynomial P satisfies the given

property. Hence, we will assume that P is not a constant polynomial.

Let n be the degree of P . Since P is not constant, n ≥ 1. Let

n
X
P (x) = a i xi ,
i=0

with an 6= 0. Then
n
X n
X
i
(x + 1) ai (x − 1) − (x − 1) ai xi = C,
i=0 i=0

for some constant C. We will compare the coefficient of xn of the left-hand side of
this equation with the right-hand side. Since C is a constant and n ≥ 1, the coeffi-
cient of xn of the right-hand side is equal to zero. We now determine the coefficient
of xn of the left-hand side of this expression.

The left-hand side of the equation simplifies to

n
X n
X n
X n
X
x ai (x − 1)i + ai (x − 1)i − x a i xi + ai xi .
i=0 i=0 i=0 i=0
We will determine the coefficient xn of each of these four terms.

By the Binomial Theorem, the coefficient¡ of x¢n of the first term is equal to that
n
of x (an−1 (x − 1)n−1 + an (x − 1)n ) = an−1 − n−1 an = an−1 − nan .

The coefficient of xn of the second term is equal to that of an (x − 1)n , which is an .

The coefficient of xn of the third term is equal to an−1 and that of the fourth
term is equal to an .

Summing these four coefficients yield an−1 − nan + an − an−1 + an = (2 − n)an .

This expression is equal to 0. Since an 6= 0, n = 2. Hence, P is a quadratic

polynomial.

Let P (x) = ax2 + bx + c, where a, b, c are real numbers with a 6= 0. Then

(x + 1)(a(x − 1)2 + b(x − 1) + c) − (x − 1)(ax2 + bx + c) = C.

Simplifying the left-hand side yields

(b − a)x + 2c = 2C.

Therefore, b − a = 0 and 2c = 2C. Hence, P (x) = ax2 + ax + c. As in Solution 1,

this is a valid solution for all a ∈ R\{0}. ¤
2. The sequence a1 , a2 , . . . , an consists of the numbers 1, 2, . . . , n in some order. For
which positive integers n is it possible that 0, a1 , a1 + a2 , . . . , a1 + a2 + . . . + an all
have different remainders when divided by n + 1?

Solution: It is possible if and only if n is odd.

n
If n is even, then a1 + a2 + . . . + an = 1 + 2 + . . . + n = 2
· (n + 1), which is
congruent to 0 mod n + 1. Therefore, the task is impossible.

Now suppose n is odd. We will show that we can construct a1 , a2 , . . . , an that sat-
isfy the conditions given in the problem. Then let n = 2k + 1 for some non-negative
integer k. Consider the sequence: 1, 2k, 3, 2k − 2, 5, 2k − 3, . . . , 2, 2k + 1, i.e. for each
1 ≤ i ≤ 2k + 1, ai = i if i is odd and ai = 2k + 2 − i if i is even.

We first show that each term 1, 2, . . . , 2k + 1 appears exactly once. Clearly, there
are 2k + 1 terms. For each odd number m in {1, 2, . . . , 2k + 1}, am = m. For each
even number m in this set, a2k+2−m = 2k + 2 − (2k + 2 − m) = m. Hence, every
number appears in a1 , . . . , a2k+1 . Hence, a1 , . . . , a2k+1 does consist of the numbers
1, 2, . . . , 2k + 1 in some order.

We now determine a1 + a2 + . . . + am (mod 2k + 2). We will consider the cases

when m is odd and when m is even separately. Let bm = a1 + a2 . . . + am .

If m is odd, note that a1 ≡ 1 (mod 2k + 2), a2 + a3 = a4 + a5 = . . . = a2k +

a2k+1 = 2k + 3 ≡ 1 (mod 2k + 2). Therefore, {b1 , b3 , . . . , b2k+1 } = {1, 2, 3, . . . , k + 1}
(mod 2k + 2).

If m is even, note that a1 + a2 = a3 + a4 = . . . = a2k−1 + a2k = 2k + 1 ≡ −1

(mod 2k + 2). Therefore, {b2 , b4 , . . . , b2k } = {−1, −2, . . . , −k} (mod 2k + 2) ≡
{2k + 1, 2k, . . . , k + 2} (mod 2k + 2).

Therefore, b1 , b2 , . . . , b2k+1 do indeed have different remainders when divided by

2k + 2. This completes the problem. ¤
3. Let G be the centroid of a right-angled triangle ABC with ∠BCA = 90◦ . Let P
be the point on ray AG such that ∠CP A = ∠CAB, and let Q be the point on ray
BG such that ∠CQB = ∠ABC. Prove that the circumcircles of triangles AQG and
BP G meet at a point on side AB.

Solution 1. Since ∠C = 90◦ , the point C lies on the semicircle with diameter AB
which implies that, if M is te midpoint of side AB, then M A = M C = M B. This
implies that triangle AM C is isosceles and hence that ∠ACM = ∠A. By definition,
G lies on segment M and it follows that ∠ACG = ∠ACM = ∠A = ∠CP A. This
implies that triangles AP C and ACG are similar and hence that AC 2 = AG · AP .
Now if D denotes the foot of the perpendicular from C to AB, it follows that triangles
ACD and ABC are similar which implies that AC 2 = AD·AB. Therefore AG·AP =
AC 2 = AD·AB and, by power of a point, quadrilateral DGP B is cyclic. This implies
that D lies on the circumcircle of triangle BP G and, by a symmetric argument, it
follows that D also lies on the circumcircle of triangle AGQ. Therefore these two
circumcircles meet at the point D on side AB.
Solution 2. Define D and M as in Solution 1. Let R be the point on side AB
such that AC = CR and triangle ACR is isosceles. Since ∠CRA = ∠A = ∠CP A,
it follows that CP RA is cyclic and hence that ∠GP R = ∠AP R = ∠ACR = 180◦ −
2∠A. As in Solution 1, M C = M B and hence ∠GM R = ∠CM B = 2∠A = 180◦ −
∠GP R. Therefore GP RM is cyclic and, by power of a point, AM · AR = AG · AP .
Since ACR is isosceles, D is the midpoint of AR and thus, since M is the midpoint
of AB, it follows that AM · AR = AD · AB = AG · AP . Therefore DGP B is cyclic,
implying the result as in Solution 1.
4. Let n be a positive integer. For any positive integer j and positive real number
r, define
µ ¶ µ» ¼ ¶
j j
fj (r) = min (jr, n) + min , n , and gj (r) = min (djre, n) + min ,n ,
r r

where dxe denotes the smallest integer greater than or equal to x. Prove that
n
X n
X
2
fj (r) ≤ n + n ≤ gj (r).
j=1 j=1

Solution 1: We first prove the left hand side inequality. We begin by drawing
an n × n board, with corners at (0, 0), (n, 0), (0, n) and (n, n) on the Cartesian plane.

Consider the line ` with slope r passing through (0, 0). For each j ∈ {1, . . . , n},
consider the point (j, min(jr, n)). Note that each such point either lies on ` or the
top edge of the board. In the j th column from the left, draw the rectangle of height
min(jr, n). Note that the sum of the n rectangles is equal to the area of the board
under the line ` plus n triangles (possibly with area 0) each with width at most 1
and whose sum of the heights is at most P n. Therefore, the sum of the areas of these
n triangles is at most n/2. Therefore, nj=1 min(jr, n) is at most the area of the
square under ` plus n/2.

Consider the line with slope 1/r. By symmetry about the line y = x, the area of
the square under the line with slope 1/r is equal to the area
Pn of the square above the
line `. Therefore, using the same reasoning as before, j=1 min(j/r, n) is at most
the area of the square above ` plus n/2.
P P
Therefore, nj=1 fj (r) = nj=1 (min(jr, n) + min( rj , n)) is at most the area of the
board plus n, which is n2 + n. This proves the left hand side inequality.

To prove the right hand side inequality, we will use the following lemma:

Lemma: Consider the line ` with slope s passing through (0, 0). P Then the num-
ber of squares on the board that contain an interior point below ` is nj=1 min (djse, n).

Proof of Lemma: For each j ∈ {1, . . . , n}, we count the number of squares in the
j th column (from the left) that contain an interior point lying below the line `. The
line x = j intersect the line ` at (j, js). Hence, since each column contains n squares
total, the number of such squares is min(djse, n). Summing over all j ∈ {1, 2, . . . , n}
proves the lemma. End Proof of Lemma

By the lemma, the rightmost expression of the inequality is equal to the number
of squares containing an interior point below the line with slope r plus the number
of squares containing an interior point below the line with slope 1/r. By symmetry
about the line y = x, the latter number is equal to the number of squares containing
an interior point above the line with slope r. Therefore, the rightmost expression
of the inequality is equal to the number of squares of the board plus the number of
squares of which ` passes through the interior. The former is equal to n2 . Hence, to
prove the inequality, it suffices to show that every line passes through the interior of
at least n squares. Since ` has positive slope, each ` passes through either n rows
and/or n columns. In either case, ` passes through the interior of at least n squares.
Hence, the right inequality holds. ¤

PnSolution 2: We first prove the left inequality. Define the function f (r) =
j=1 fj (r). Note that f (r) = f (1/r) for all r > 0. Therefore, we may assume
that r ≥ 1.

Let m = bn/rc, where bxc denotes the largest integer less than or equal to x. Then
min(jr, n) = jr for all j ∈ {1, . . . , m} and min(jr, n) = n for all j ∈ {m + 1, . . . , n}.
Note that since r ≥ 1, min(j/r, n) ≤ n for all j ∈ {1, . . . , n}. Therefore,
n
X 1
f (r) = fj (r) = (1 + 2 + . . . m)r + (n − m)n + (1 + 2 + . . . + n) ·
j=1
r

m(m + 1) n(n + 1) 1
= ·r+ · + n(n − m) (1)
2 2 r
2
Then by (??), note that f (r) ≤ n + n if and only if
m(m + 1)r n(n + 1)
+ ≤ n(m + 1)
2 2r
if and only if
m(m + 1)r2 + n(n + 1) ≤ 2rn(m + 1) (2)
Since m = bn/rc, there exist a real number b satisfying 0 ≤ b < r such that
n = mr + b. Substituting this into (??) yields

m(m + 1)r2 + (mr + b)(mr + b + 1) ≤ 2r(mr + b)(m + 1),

if and only if

2m2 r2 + mr2 + (2mb + m)r + b2 + b ≤ 2m2 r2 + 2mr2 + 2mbr + 2br,

which simplifies to mr + b2 + b ≤ mr2 + 2br ⇔ b(b + 1 − 2r) ≤ mr(r − 1) ⇔

b((b − r) + (1 − r)) ≤ mr(r − 1). This is true since

b((b − r) + (1 − r)) ≤ 0 ≤ mr(r − 1),

which holds since r ≥ 1 and b < r. Therefore, the left inequality holds.
P
We now prove the right inequality. Define the function g(r) = nj=1 = gj (r).
Note that g(r) = g(1/r) for all r > 0. Therefore, we may assume that r ≥ 1. We
will consider two cases; r ≥ n and 1 ≤ r < n.

If r ≥ n, then min(djre, n) = n and min(dj/re, n) = 1 for all j ∈ {1, . . . , n}.

Hence, gj (r) = n + 1 for all j ∈ {1, . . . , n}. Therefore, g(r) = n(n + 1) = n2 + n,
implying that the inequality is true.

Now we consider the case 1 ≤ r < n. Let m = bn/rc. Hence, jr ≤ n for all
j ∈ {1, . . . , m}, i.e. min(djr, e, n) = djre and jr ≥ n for all j ∈ {m + 1, . . . , n}, i.e.
min(djre, n) = n. Therefore,
n
X m
X
min(djre, n) = djre + (n − m)n. (3)
j=1 j=1

Pn
We will now consider the second sum j=1 min{dj/re, n}.

Since r ≥ 1, min(dj/re, n) ≤ min(dn/re, n) ≤ n. Therefore, min(dj/re, n) =

dj/re. Since m = bn/rc, dn/re ≤ m + 1. Since r > 1, m < n, which implies that
m + 1 ≤ n. Therefore, min{dj/re, n} = dj/re ≤ dn/re ≤ m + 1 for all j ∈ {1, . . . , n}.

For each positive integer k ∈ {1, . . . , m + 1}, we now determine the number of
positive integers j ∈ {1, . . . , n} such that dj/re = k. We denote this number by sk .

Note that dj/re = k if and only if k − 1 < j/r ≤ k if and only if (k − 1)r < j ≤
min(kr, n), since j ≤ n. We will handle the cases k ∈ {1, . . . , m} and k = m + 1
separately. If k ∈ {1, . . . , m}, then min(kr, n) = kr, since r ≤ m and m = bn/rc.
The set of positive integers j satisfying (k − 1)r < j ≤ kr is {b(k − 1)rc + 1, b(k −
1)rc + 2, . . . , bkrc}. Hence,

sk = brkc − (br(k − 1)c + 1) + 1 = brkc − br(k − 1)c

for all k ∈ {1, . . . , m}. If k = m + 1, then (k − 1)r < j ≤ min(kr, n) = n. The set
of positive integers j satisfying (k − 1)r < j ≤ kr is {b(k − 1)rc + 1, . . . , n}. Then
sm+1 = n − br(k − 1)c = n − bmrc. Note that this number is non-negative by the
definition of m. Therefore, by the definition of sk , we have
n
X µ» ¼ ¶ m+1
X
j
min ,n = ksk
j=1
r k=1
m
X m
X
= (k (bkrc − b(k − 1)rc)) + (m + 1)(n − brmc) = (m + 1)n − bkrc.
k=1 k=1
(4)

Summing (??) and (??) yields that

m
X
2
g(r) = n + n + (djre − bjrc) ≥ n2 + n,
j=1

which proves the right inequality. ¤

5. Let O denote the circumcentre of an acute-angled triangle ABC. A circle Γ
passing through vertex A intersects segments AB and AC at points P and Q such
that ∠BOP = ∠ABC and ∠COQ = ∠ACB. Prove that the reflection of BC in the
line P Q is tangent to Γ.

Solution. Let the circumcircle of triangle OBP intersect side BC at the points R
and B and let ∠A, ∠B and ∠C denote the angles at vertices A, B and C, respectively.
Now note that since ∠BOP = ∠B and ∠COQ = ∠C, it follows that

∠P OQ = 360◦ −∠BOP −∠COQ−∠BOC = 360◦ −(180−∠A)−2∠A = 180◦ −∠A.

This implies that AP OQ is a cyclic quadrilateral. Since BP OR is cyclic,

∠QOR = 360◦ − ∠P OQ − ∠P OR = 360◦ − (180◦ − ∠A) − (180◦ − ∠B) = 180◦ − ∠C.

This implies that CQOR is a cyclic quadrilateral. Since AP OQ and BP OR are

cyclic,

∠QP R = ∠QP O + ∠OP R = ∠OAQ + ∠OBR = (90◦ − ∠B) + (90◦ − ∠A) = ∠C.

Since CQOR is cyclic, ∠QRC = ∠COQ = ∠C = ∠QP R which implies that the
circumcircle of triangle P QR is tangent to BC. Further, since ∠P RB = ∠BOP =
∠B,

∠P RQ = 180◦ − ∠P RB − ∠QRC = 180◦ − ∠B − ∠C = ∠A = ∠P AQ.

This implies that the circumcircle of P QR is the reflection of Γ in line P Q. By

symmetry in line P Q, this implies that the reflection of BC in line P Q is tangent to
Γ.
 46th Canadian Mathematical Olympiad
Wednesday, April 2, 2014

1. Let a1 , a2 , . . . , an be positive real numbers whose product is 1. Show that the sum
a1 a2 a3 an
+ + +· · ·+
1 + a1 (1 + a1 )(1 + a2 ) (1 + a1 )(1 + a2 )(1 + a3 ) (1 + a1 )(1 + a2 ) · · · (1 + an )
n
2 −1
is greater than or equal to .
2n
2. Let m and n be odd positive integers. Each square of an m by n board is coloured
red or blue. A row is said to be red-dominated if there are more red squares than
blue squares in the row. A column is said to be blue-dominated if there are more
blue squares than red squares in the column. Determine the maximum possible value
of the number of red-dominated rows plus the number of blue-dominated columns.
Express your answer in terms of m and n.
3. Let p be a fixed odd prime. A p-tuple (a1 , a2 , a3 , . . . , ap ) of integers is said to be
good if
(i) 0 ≤ ai ≤ p − 1 for all i, and
(ii) a1 + a2 + a3 + · · · + ap is not divisible by p, and
(iii) a1 a2 + a2 a3 + a3 a4 + · · · + ap a1 is divisible by p.
Determine the number of good p-tuples.
4. The quadrilateral ABCD is inscribed in a circle. The point P lies in the interior
of ABCD, and ∠P AB = ∠P BC = ∠P CD = ∠P DA. The lines AD and BC meet
at Q, and the lines AB and CD meet at R. Prove that the lines P Q and P R form
the same angle as the diagonals of ABCD.
5. Fix positive integers n and k ≥ 2. A list of n integers is written in a row on a
blackboard. You can choose a contiguous block of integers, and I will either add 1 to
all of them or subtract 1 from all of them. You can repeat this step as often as you
like, possibly adapting your selections based on what I do. Prove that after a finite
number of steps, you can reach a state where at least n − k + 2 of the numbers on
the blackboard are all simultaneously divisible by k.
 46th Canadian Mathematical Olympiad
Wednesday, April 2, 2014

Problems and Solutions

1. Let a1 , a2 , . . . , an be positive real numbers whose product is 1. Show that the sum
a1 a2 a3 an
+ + +· · ·+
1 + a1 (1 + a1 )(1 + a2 ) (1 + a1 )(1 + a2 )(1 + a3 ) (1 + a1 )(1 + a2 ) · · · (1 + an )
2n − 1
is greater than or equal to .
2n
Solution. Note for that every positive integer m,
am 1 + am 1
= −
(1 + a1 )(1 + a2 ) · · · (1 + am ) (1 + a1 )(1 + a2 ) · · · (1 + am ) (1 + a1 )(1 + a2 ) · · · (1 + am )
1 1
= − .
(1 + a1 ) · · · (1 + am−1 ) (1 + a1 ) · · · (1 + am )
Therefore, if we let bj = (1 + a1 )(1 + a2 ) · · · (1 + aj ), with b0 = 0, then by telescoping
sums,
Xn X n µ ¶
aj 1 1 1
= − =1− .
j=1
(1 + a1 ) · · · (1 + aj ) j=1
bj−1 bj bn
√ √ √
Note that bn = (1 + a1 )(1 + a2 ) · · · (1 + an ) ≥ (2 a1 )(2 a2 ) · · · (2 an ) = 2n , with
equality if and only if all ai ’s equal to 1. Therefore,
1 1 2n − 1
1− ≥1− n = .
bn 2 2n
To check that this minimum can be obtained, substituting all ai = 1 to yield
1 1 1 1 2n−1 + 2n−2 + . . . + 1 2n − 1
+ 2 + 3 + ... + n = = ,
2 2 2 2 2n 2n
as desired.
2. Let m and n be odd positive integers. Each square of an m by n board is coloured
red or blue. A row is said to be red-dominated if there are more red squares than
blue squares in the row. A column is said to be blue-dominated if there are more
blue squares than red squares in the column. Determine the maximum possible value
of the number of red-dominated rows plus the number of blue-dominated columns.
Express your answer in terms of m and n.
Solution. The answer is m + n − 2 if m, n ≥ 3 and max{m, n} if one of m, n is equal
to 1.

Note that it is not possible that all rows are red-dominated and all columns are
blue-dominated. This is true, since the number of rows and columns are both odd,
the number of squares is odd. Hence, there are more squares of one color than the
other. Without loss of generality, suppose there are more red squares than blue
squares. Then it is not possible that for every column, there are more blue squares
than red squares. Hence, every column cannot be blue-dominated.

If one of m, n is equal to 1, say m without loss of generality, then by the claim, the
answer is less than n + 1. The example where there are n blue-dominated columns
is by painting every square blue. There are 0 red-dominated rows. The sum of the
two is n = max{m, n}.

Now we handle the case m, n ≥ 3.

There are m rows and n columns on the board. Hence, the answer is at most
m + n. We have already shown that the answer cannot be m + n.
Since m, n are odd, let m = 2a − 1 and n = 2b − 1 for some positive integers
a, b. Since m, n ≥ 3, a, b ≥ 2. We first show that the answer is not m + n − 1. By
symmetry, it suffices to show that we cannot have all rows red-dominated and all-but-
one column blue-dominated. If all rows are red dominated, then each row has at least
b red squares. Hence, there are at least bm = (2a − 1)b red squares. Since all-but-one
column is blue-dominated, there are at least 2b − 2 blue-dominated columns. Each
such column then has at least a blue squares. Therefore, there are at least a(2b − 2)
blue squares. Therefore, the board has at least (2a − 1)b + a(2b − 2) = 4ab − b − 2a
squares. But the total number of squares on the board is
(2a − 1)(2b − 1) = 4ab − 2a − 2b + 1 = 4ab − 2a − b − b + 1 < 4ab − 2a − b,
which is true since b ≥ 2. This is a contradiction. Therefore, the answer is less than
m + n − 1.
 We now claim that there is a colouring of the board such that the number of blue-
dominated columns plus the number of red-dominated rows is m + n − 2; Colour the
first column entirely red, and the first row, minus the top-left corner, entirely blue.
The remaining uncoloured square is an even-by-even board. Colour the remaining
board in an alternating pattern (i.e. checkerboard pattern). Hence, on this even-
by-even board, each row has the same number of red squares as blue squares and
each column has the same number of red squares as blue squares. Then on the
whole board, since the top row, minus the top-left square is blue, all columns, but
the leftmost column, are blue-dominated. Hence, there are n − 1 blue-dominated
columns. Since the left column is red, all rows but the top row are red dominated.
Hence, there are m − 1 red-dominated rows. The sum of these two quantities is
m + n − 2, as desired.

3. Let p be a fixed odd prime. A p-tuple (a1 , a2 , a3 , . . . , ap ) of integers is said to be

good if

(i) 0 ≤ ai ≤ p − 1 for all i, and

(ii) a1 + a2 + a3 + · · · + ap is not divisible by p, and

(iii) a1 a2 + a2 a3 + a3 a4 + · · · + ap a1 is divisible by p.

Determine the number of good p-tuples.

Solution. Let S be the set of all sequences (b1 , b2 , . . . , bp ) of numbers from the
set {0, 1, 2, . . . , p − 1} such that b1 + b2 + · · · + bp is not divisible by p. We show
that |S| = pp − pp−1 . For let b1 , b2 , . . . , bp−1 be an arbitrary sequence of numbers
chosen from {0, 1, 2, . . . , p − 1}. There are exactly p − 1 choices for bp such that
b1 + b2 + · · · + bp−1 + bp 6≡ 0 (mod p), and therefore |S| = pp−1 (p − 1) = pp − pp−1 .
Now it will be shown that the number of good sequences in S is p1 |S|. For a
sequence B = (b1 , b2 , . . . , bp ) in S, define the sequence Bk = (a1 , a2 , . . . , ap ) by

ai = bi − b1 + k mod p

for 1 ≤ i ≤ p. Now note that B in S implies that

a1 + a2 + · · · + ap ≡ (b1 + b2 + · · · + bp ) − pb1 + pk ≡ (b1 + b2 + · · · + bp ) 6≡ 0 (mod p)

and therefore Bk is in S for all non-negative k. Now note that Bk has first element
k for all 0 ≤ k ≤ p − 1 and therefore the sequences B0 , B1 , . . . , Bp−1 are distinct.
 Now define the cycle of B as the set {B0 , B1 , . . . , Bp−1 }. Note that B is in its
own cycle since B = Bk where k = b1 . Now note that since every sequence in S is in
exactly one cycle, S is the disjoint union of cycles.
Now it will be shown that exactly one sequence per cycle is good. Consider
an arbitrary cycle B0 , B1 , . . . , Bp−1 , and let B0 = (b1 , b2 , . . . , bp ) where b0 = 0, and
note that Bk = (b1 + k, b2 + k, . . . , bp + k) mod p. Let u = b1 + b2 + · · · + bp , and
v = b1 b2 + b2 b3 + · · · + bp b1 and note that (b1 + k)(b2 + k) + (b2 + k)(b3 + k)) + · · · +
(bp + k)(b1 + k) = u + 2kv mod p for all 0 ≤ k ≤ p − 1. Since 2v is not divisible by
p, there is exactly one value of k with 0 ≤ k ≤ p − 1 such that p divides u + 2kv
and it is exactly for this value of k that Bk is good. This shows that exactly one
sequence per cycle is good and therefore that the number of good sequences in S is
1
p
|S|, which is pp−1 − pp−2 .

4. The quadrilateral ABCD is inscribed in a circle. The point P lies in the interior
of ABCD, and ∠P AB = ∠P BC = ∠P CD = ∠P DA. The lines AD and BC meet
at Q, and the lines AB and CD meet at R. Prove that the lines P Q and P R form
the same angle as the diagonals of ABCD.

Solution. . Let Γ be the circumcircle of quadrilateral ABCD. Let α = ∠P AB =

∠P BC∠P CD = ∠P DA and let T1 , T2 , T3 and T4 denote the circumcircles of
triangles AP D, BP C, AP B and CP D, respectively. Let M be the intersection of
T1 with line RP and let N be the intersection of T3 with line SP . Also let X denote
the intersection of diagonals AC and BD.
By power of a point for circles T1 and Γ, it follows that RM · RP = RA · RD =
RB · RC which implies that the quadrilateral BM P C is cyclic and M lies on T2 .
Therefore ∠P M B = ∠P CB = α = ∠P AB = ∠DM P where all angles are directed.
This implies that M lies on the diagonal BD and also that ∠XM P = ∠DM P = α.
By a symmetric argument applied to S, T3 and T4 , it follows that N lies on T4 and
that N lies on diagonal AC with ∠XN P = α. Therefore ∠XM P = ∠XN P and
X, M , P and N are concyclic. This implies that the angle formed by lines M P and
N P is equal to one of the angles formed by lines M X and N X. The fact that M
lies on BD and RP and N lies on AC and SP now implies the desired result.

5. Fix positive integers n and k ≥ 2. A list of n integers is written in a row on a

blackboard. You can choose a contiguous block of integers, and I will either add 1 to
all of them or subtract 1 from all of them. You can repeat this step as often as you
like, possibly adapting your selections based on what I do. Prove that after a finite
number of steps, you can reach a state where at least n − k + 2 of the numbers on
the blackboard are all simultaneously divisible by k.
Solution. We will think of all numbers as being residues mod k. Consider the
following strategy:

• If there are less than k − 1 non-zero numbers, then stop.

• If the first number is 0, then recursively solve on the remaining numbers.

• If the first number is j with 0 < j < k, then choose the interval stretching from
the first number to the jth-last non-zero number.

First note that this strategy is indeed well defined. The first number must have
value between 0 and k − 1, and if we do not stop immediately, then there are at least
k − 1 non-zero numbers, and hence the third step can be performed.
For each j with 1 ≤ j ≤ k − 2, we claim the first number can take on the value of
j at most a finite number of times without taking on the value of j − 1 in between. If
this were to fail, then every time the first number became j, I would have to add 1 to
the selected numbers to avoid making it j − 1. This will always increase the j-th last
non-zero number, and that number will never be changed by other steps. Therefore,
that number would eventually become 0, and the next last non-zero number would
eventually become zero, and so on, until the first number itself becomes the j-th last
non-zero number, at which point we are done since j ≤ k − 2.
Rephrasing slightly, if 1 ≤ j ≤ k − 2, the first number can take on the value of j
at most a finite number of times between each time it takes on the value of j − 1. It
then immediately follows that if the first number can take on the value of j − 1 at
most a finite number of times, then it can also only take on the value of j a finite
number of times. However, if it ever takes on the value of 0, we have already reduced
the problem to n − 1, so we can assume that never happens. It then follows that the
first number can take on all the values 0, 1, 2, . . . , k − 2 at most a finite number of
times.
Finally, every time the first number takes on the value of k − 1, it must subse-
quently take on the value of k − 2 or 0, and so that can also happen only finitely
many times.