Exponents and Logarithms

Solutions
1. We have logx (2 · 4 · 8) = 1 which implies that logx (64) = 1. Therefore, x = 64.

2. Since 12 = 22 · 3 it follows that

122x+1 = 23x+7 · 33x−4

22(2x+1) · 32x+1 = 23x+7 · 33x−4
22(2x+1)−3x−7 = 33x−4−2x−1
2x−5 = 3x−5

The graphs of y = 2x−5 and y = 3x−5 intersect only at x = 5 and y = 1. (Since 2z = 3z only
for z = 0.) Therefore, x = 5 is the only solution.
   
3 4 5 200 200
3. This expression equals log10 · · ...... = log10 = log10 100 = 2.
2 3 4 199 2
4. For the first equation we know that x, y 6= 0. The second equation states xy −2 = 3−3 which
gives us x3 y −6 = 3−9 after cubing both sides. Since neither side of this new equation is 0 we
can divide the first equation by this new equation to eliminate x.

x3 y 5 211 313
=
x3 y −6 3−9
y 11 = 211 · 322
y 11 = 211 · (32 )11
y 11 = (2 · 9)11
y = 18.

y2
Therefore, x = = 12.
27
1 1 1 1
5. log8 (18) = log8 2 + log8 9 = + log8 32 = + 2 log8 3 = + 2k = 2k +
3 3 3 3
6. Solution 1
We express the logarithms in exponential form to arrive at: 2x = 2y and x = 4y . Thus,

2y = 2(4y )
2y = 2(22y )
2y = 22y+1
y = 2y + 1
y = −1
 
1 1
Thus, x = 4−1 = . Therefore, the point of intersection is , −1 .
4 4
 Solution 2
Substituting one equation into the other we obtain

log2 2x = log4 x
log 2x log x
=
log 2 log 4
log 2x log x
=
log 2 log 22
log 2x log x
=
log 2 2 log 2
2 log 2x = log x
2 log 2 + 2 log x = log x
log x = −2 log 2
log x = log 2−2
1
log x = log
4
1
Therefore, x = . Substituting this value back into either of the original equations and we
4  
1
obtain that y = −1. Therefore, the point of intersection is , −1 .
4
y
7. 
We note first that  x = a for all points on the curve. The midpoint of AB is given by
x1 + x2 y 1 + y 2 y1 + y2
, . Since we draw a horizontal line through the midpoint, y3 = , we
2 2 2
have that

(x3 )2 = (ay3 )2
y1 +y2
= (a 2 )2
= (ay1 )(ay2 )
= x1 x2 .

8. Given that the graph of the function passes through (2, 1), we know that a 6= 0. We have
1 = a(2r ) and 4 = a(32r ). Since neither side of the first equation is 0, we can divide the second
32r 2r · 16r 1
equation by the first to obtain 4 = r = r
= 16r . Therefore, r = .
2 2 2
9. Factoring both sides of the equation 2x+3 + 2x = 3y+2 − 3y gives

(23 + 1)2x = (32 − 1)3y

9 · 2x = 8 · 3y
32 · 2x = 23 · 3y
2x−3 = 3y−2

Since x and y are integers, and the only integer power of 2 that is also an integer power of 3 is
the number 1 = 20 = 30 , we have x = 3 and y = 2.
10. If f (x) = 24x−2 , then f (x) · f (1 − x) = 24x−2 · 24(1−x)−2 = 24x−2+4−4x−2 = 20 = 1.

11. Observe that the argument of both logarithms must be positive and so x > 6. Now

log5 (x − 2) + log5 (x − 6) = 2
log5 ((x − 2)(x − 6)) = 2
(x − 2)(x − 6) = 25
x2 − 8x − 13 = 0
√
x=4± 29
√
However, since x > 6, we have that x = 4 + 29.
   
b c b c
12. If a, b, c is a geometric sequence, then = . It follows that logx = logx which
a b a b
implies logx b − logx a = logx c − logx b. Therefore, the logarithms form an arithmetic sequence.
If logx a, logx b, logx c form an arithmetic sequence, then

logx b − logx a = logx c − logx b

 
b c
logx = logx
a b
b c
= since the log function takes on each value only once
a b
Thus, a, b, c form a geometric sequence.

13. Using exponent rules and arithmetic, we manipulate the given equation:

3x+2 + 2x+2 + 2x = 2x+5 + 3x

3x 32 + 2x 22 + 2x = 2x 25 + 3x
9(3x ) + 4(2x ) + 2x = 32(2x ) + 3x
8(3x ) = 27(2x )
3x 27
=
2x 8
 x  3
3 3
=
2 2

Since the two expressions are equal and the bases are equal, then the exponents must be equal,
and so x = 3.

14. Let a = log10 x. Then (log10 x)log10 (log10 x) = 10 000 becomes alog10 a = 104 .
Taking the base 10 logarithm of both sides and using the fact that log10 (ab ) = b log10 a, we obtain
(log10 a)(log10 a) = 4 or (log10 a)2 = 4. Therefore, log10 a = ±2 and so log10 (log10 x) = ±2.
If log10 (log10 x) = 2, then log10 x = 102 = 100 and so x = 10100 .
If log10 (log10 x) = −2, then log10 x = 10−2 = 1
100
and so x = 101/100 . Therefore, x = 10100 or
x = 101/100 .
15. Note that x 6= 1 since 1 cannot be the base of a logarithm. This tells us that log x 6= 0.
log b
Using the fact that loga b = and then using other logarithm laws, we obtain the following
log a
equivalent equations:
7
log4 x − logx 16 = 6
− logx 8
log x log 16 7 log 8
− = − (note that x 6= 1, so log x 6= 0)
log 4 log x 6 log x
log x 7 log 16 − log 8
= +
log 4 6 log x
16
log x 7 log( 8 )
= +
log(22 ) 6 log x
log x 7 log 2
= +
2 log 2 6 log x
 
1 log x 7 log 2
= +
2 log 2 6 log x

log x
Letting t = = log2 x and noting that t 6= 0 since x 6= 1, we obtain the following equations
log 2
equivalent to the previous ones:
t 7 1
= +
2 6 t
2
3t = 7t + 6 (multiplying both sides by 6t)
2
3t − 7t − 6 = 0
(3t + 2)(t − 3) = 0

Therefore, the original equation is equivalent to t = − 23 or t = 3.

Converting back to the variable x, we obtain log2 x = − 32 or log2 x = 3, which gives x = 2−2/3
or x = 23 = 8.
 Trigonometry
Solutions
1
1. (a) Here sin 2θ = − . The smallest positive value of 2θ which satisfies this equation is 210◦ .
2
Therefore, θ = 105◦ .
(b) We know that cos2 θ = 1 − sin2 θ. Substituting this fact into our given equation we obtain

2(2 sin2 θ − 1) = 8 sin θ − 5

4 sin2 θ − 8 sin θ + 3 = 0
(2 sin θ − 1)(2 sin θ − 3) = 0
1 3
sin θ = ,
2 2
1
But we know that | sin θ| ≤ 1 and so sin θ = . Therefore, for −π ≤ θ ≤ π, we have
2
π 5π
θ= , .
6 6
2. Solution 1
Let θ = ∠AM C. Therefore, ∠BM A = 180◦ − θ. Using the cosine law in 4ABM gives

49 = 9 + 25 − 30 cos(180◦ − θ)
15 = −30 cos(180◦ − θ)
1
cos(180◦ − θ) = −
2
1
The fact that cos(180◦ − θ) = − cos θ gives us cos θ = .
2
Therefore, using the cosine law in 4AM C gives
 
2 1
AC = 9 + 36 − 36 cos θ = 45 − 36 = 27
2
√
Therefore, AC = 3 3.
Solution 2
13
Using the cosine law in 4ABM gives 9 = 49 + 25 − 70 cos(∠ABM ). Thus, cos(∠ABM ) = .
14
Using the cosine law in 4ABC gives AC 2 = 49 + 121 − 154 cos(∠ABC). But ∠ABC = ∠ABM
13 √
and so cos(∠ABC) = . Therefore, AC 2 = 27 and so AC = 3 3.
14
3. We determine that ∠BN A = 180◦ −  47◦ − 108 ◦
= 25◦ . Using the sine law in 4BN A gives
◦
BN 100 sin 108 MN
◦
= ◦
and so BN = 100 ◦
. But from 4BN M we get = tan 32◦
sin 108 sin 25 sin 25 BN
sin 108◦
 
◦
and so M N = BN tan 32 . Therefore, M N = 100 ◦
tan 32◦ ≈ 141 m.
sin 25
 5π 5π π
4. Since the area of the rectangle is , its height is ÷ = 5. Since the cosine graph is
3 3 3   √
π π π 3
symmetrical about the y-axis, P O = OQ = ÷ 2 = . But cos = . Since we are
√ 3 √ 6 6 2
3 10 3
graphing y = k cos x, we see that k = 5 ÷ = .
2 3
5. Since the minimum point has a y-coordinate of −2, the amplitude is a = 2. Also, since the
3π 3π 3π 3π
minimum occurs at x = (rather than where it is found for sin x), k = ÷ = 2.
4 2 2 4
1 π
Setting y = 1, we obtain that 1 = 2 sin 2x and so sin 2x = . Therefore, 2x = (since we
2 6
π
are looking for the intersection that occurs before the first maximum) and so x = . Thus,
π  12
D= ,1 .
12
6. Solution 1
Since one side of each of these triangles is parallel with one of the sides of the square and
another side of each of the triangles is parallel to another side of the square, these four triangles
are right-angled. We let the length of the side of the triangles opposite θ be a and the length
a
of the side adjacent to θ be b. Then tan θ = .
b
Also, a − b is equal to the length of the sides of the small square which is 3. The area of the
large square is equalto the
 sum of the areas of the triangles along with the area of the small
1 40 40
square. Therefore, 4 ab + 9 = 89 and so b = . Thus, a − = 3 and so a2 − 3a − 40 = 0,
2 a a
8
which gives a = 8 or −5. Now since a is positive, a = 8 and b = 5. Therefore, tan θ = .
5
Solution 2
As in the previous solution,
√ we note that all of the four triangles are right-angled. Each has a
hypotenuse of length 89, since the area of the large square is 89. If we let the longer of the
two legs of the triangle be a, then the length of the other leg is a − 3, since the small square
has sides of length 3.
Using the Pythagorean Theorem on this triangle gives a2 + (a − 3)2 = 89 which simplifies to
a2 − 3a − 40 = 0. Therefore, a = 8 or a = −5. But a is positive and so a = 8. Therefore, the
8
lengths of the legs of the triangle are 8 and 5 and thus, tan θ = .
5
√
7. Using the Pythagorean theorem, we find that F A = 2, AC = 2 and F C = 2. The cosine law
in 4F AC gives

F C 2 = F A2 + AC 2 − 2 · F A · AC · cos(∠F AC)
√
4 = 4 + 2 − 2 · 2 · 2 cos(∠F AC)
1
cos(∠F AC) = √ .
2 2
8. Consider the diagram below which makes use of four of the small equilateral triangles. Using
√
the cosine law we get AT 2 = 12 + 42 − 2(1)(4) cos 60◦ . Therefore, AT 2 = 13 and so AT = 13.

We can form similar diagrams using √four of the small equilateral triangles and again apply
√ to get W A = W T = 13. Therefore, 4W AT is an equilateral triangle with
the cosine law
side length 13. Using the formula for the area of an equilateral
√ triangle (see the Toolkit for
13 3
Euclidean Geometry) we have the area of 4W AT = .
4
9. Using the cosine law we get

a2 = 64 + b2 − 16b(cos 60◦ )
= b2 − 8b + 64
= (b − 4)2 + 48
a2 − (b − 4)2 = 48
(a + b − 4)(a − b + 4) = 48

Since 48 is positive, the factors must be both positive or both negative. The sum of the
two factors is 2a and since a is a positive integer, the sum of the factors must be even and
positive. Therefore, they must be both even or both odd and they must both be positive.
Since 48 is even, both of the factors must be even. The even-even factorizations of 48 are
2 · 24, 4 · 12, 6 · 8, 8 · 6, 12 · 4, 24 · 2.
We have already discovered that a is the sum of the factors divided by 2. The difference of the
d
factors is a + b − 4 − (a − b + 4) = 2b − 8 and so b = + 4, where d is the difference of factors.
2
factors sum difference, d a b
2, 24 26 −22 13 −7
4, 12 16 −8 8 0
6, 8 14 −2 7 3
8, 6 14 2 7 5
12, 4 16 8 8 8
24, 2 26 22 13 15

Since b must be positive, the possible values for a and b are (a, b) = (7, 3), (7, 5), (8, 8), (13, 15).
 Functions, Equations and Polynomials
Solutions
1. Subtract the first equation from the second, rearrange the resulting expression and then factor
to obtain

−8x + y + xy − 8 = 0
xy − 8x + y − 8 = 0
x(y − 8) + y − 8 = 0
(x + 1)(y − 8) = 0

Therefore, x = −1 or y = 8. If x = −1, then substituting into the first equation and solving we
√ y = −9. If y = 8, then substituting into the first √
obtain that equation and solving we obtain
x = 4 ± 2 2. Therefore, the solutions are (−1, −9) and (4 ± 2 2, 8).

2. Solution 1
We are asked for the x value of the midpoint of zeros, which is the x value of the vertex of the
parabola. The equation is written in vertex form already and so a = 1.

Solution 2
Find the x-intercepts:

(x − 1)2 − 4 = 0
(x − 1)2 = 4
x=1±2

−1 + 3
Thus, x = 3 or x = −1. Thus, a = = 1.
2
3. (a) Consider a = 0 and a = 1 and find the intersection point of the resulting equations,
y = x2 and y = x2 + 2x + 1. Subtracting the
 equations
 we obtain 0 = 2x + 1. Therefore,
1 1 1 1
x = − and so the intersection point is − , . Now substitute x = − into the
2 2 4 2
general equation. Therefore,

y = x2 + 2ax + a
 
1 −1
= + 2a · +a
4 2
1
=
4
 
1 1
Since − , satisfies the general equation, it is a point on all of the parabolas.
2 4
 (b) Now y = x2 +2ax+a = (x+a)2 +a−a2 and so the vertex is at (−a, a−a2 ). If we represent
the coordinates of the vertex by (p,q) we have p = −a and q = a − a2 or q = −p2 − p, the
required parabola. Completing the square we obtain
   2
2 1 1 1 1
q =− p +p+ + =− p+ +
4 4 2 4
 
1 1
and so we see that the vertex of this parabola is − , , the common point found in
2 4
part (a)
4. Factoring both equations we arrive at:
p(1 + r + r2 ) = 26 (1)
p2 r(1 + r + r2 ) = 156 (2)

From equation (1) we can see neither of the factors of its left-hand side are 0. Dividing (2) by
(1) gives pr = 6. Substituting this relation back into (1) we get
6
+ 6 + 6r = 26
r
6 − 20r + 6r2 = 0
3r2 − 10r + 3 = 0
(3r − 1)(r − 3) = 0
 
1 1
Therefore, r = or r = 3. Hence (p, r) = (2, 3) or 18, .
3 3
b c
5. We assume, on the contrary, that the coefficients are in geometric sequence. Then = which
a b
implies that b2 = ac. But now the discriminant b2 − 4ac = −3b2 < 0, so that the roots are not
real. Thus, we have a contradiction to the condition set out in the statement of the problem
and our assumption is false.
6. Let r and s be the integer roots. The equation can be written as
a(x − r)(x − s) = a(x2 − (r + s)x + rs)
= ax2 − a(r + s)x + ars
= ax2 + bx + c
with b = −a(r + s) and c = ars. Since a, b, c are in arithmetic sequence, we have
c−b=b−a
a + c − 2b = 0
a + ars + 2a(r + s) = 0
1 + rs + 2(r + s) = 0 (we can divide by a since a 6= 0)
rs + 2r + 2s + 4 = 3
(r + 2)(s + 2) = 3
 Ignoring the order of the factors, we can factor 3 as a product of two integers in two ways:
3 = 1(3) or 3 = (−1)(−3). Therefore, the two possibilities for the roots of quadratic are: (i)
−1 and 1 or (ii) −3 and −5.
7. Solution 1
Multiplying out and collecting terms results in x4 − 6x3 + 8x2 + 2x − 1 = 0. We look for a
factoring with integer coefficients, using the fact that the first and last coefficients are 1 and
−1, respectively. So
x4 − 6x3 + 8x2 + 2x − 1 = (x2 + ax + 1)(x2 + bx − 1)
where a and b are undetermined coefficients. However, expanding and comparing coefficients
gives a + b = −6 and −a + b = 2 and ab = 8. Since all three equations are satisfied by a = −4
and b = −2, we have factored the original expression as
x4 − 6x3 + 8x2 + 2x − 1 = (x2 − 4x + 1)(x2 − 2x − 1)
√ √
Factoring these two quadratics gives the roots x = 2 ± 3 and x = 1 ± 2.
Solution 2
We observe that the original equation is of the form f (f (x)) = x, where f (x) = x2 − 3x + 1.
Now if we can find x such that f (x) = x, then f (f (x)) = x. So we solve f (x) = x2 − 3x + 1 = x
which gives the first factor x2 − 4x + 1 above. With polynomial division, we can then determine
that
x4 − 6x3 + 8x2 + 2x − 1 = (x2 − 4x + 1)(x2 − 2x − 1)
and continue as in Solution 1.
8. The vertex has x = 2 and y = −16 and so A = (2, −16). When y = 0 we get 0 = x2 − 4x − 12
which factors to give us intercepts at −2 and 6. The larger value is 6, and so B = (6, 0).
Therefore, we want the line through (2, −16) and (6, 0). Finding the slope of the line and using
the second point, the equation of the line is
 
0 + 16
y= (x − 6)
6−2
which simplifies to y = 4x − 24.
9. Solution 1
Multiplying gives
x2 − (b + c)x + bc = a2 − (b + c)a + bc
x2 − (b + c)x + a(−a + b + c) = 0
The roots are
p
b+c± (b + c)2 − 4a(−a + b + c)
x=
p 2
b + c ± (b + c) + 4a2 − 4a(b + c)
2
=
p 2
b + c ± (b + c − 2a)2
=
2
 Thus, x = −a + b + c or x = a.
Solution 2
Observe that x = a is one solution. Rearranging as in the first solution we get
x2 − (b + c)x + a(−a + b + c) = 0
Using the sum (or the product) of the roots, we determine that other root is x = −a + b + c.
10. Since x = −2 is a solution of x3 − 7x − 6 = 0, we know that x + 2 is a factor of x3 − 7x − 6.
Factoring (or using long division) we obtain
x3 − 7x − 6 = (x + 2)(x2 − 2x − 3)
= (x + 2)(x + 1)(x − 3)
Thus, the roots are −2, −1 and 3.
11. Let the roots be r and s. Using the sum of the roots and the product of the roots we obtain
−4(a − 2)
r+s=
4
=2−a
and
−8a2 + 14a + 31
rs =
4
7 31
= −2a2 + a +
2 4
Then
r2 + s2 = (r + s)2 − 2rs
 
2 2 7 31
= (2 − a) − 2 −2a + a +
2 4
31
= 4 − 4a + a2 + 4a2 − 7a −
2
2 23
= 5a − 11a − .
2
23
It appears that the minimum value should be at the vertex of the parabola f (a) = 5a2 −11a− ,
2
11
that is, at a = (found by completing the square). But we have ignored the condition that
10
the roots are real. The discriminant of the original equation is
B 2 − 4AC = [4(a − 2)]2 − 4(4)(−8a2 + 14a + 31)
= 16(a2 − 4a + 4) + 128a2 − 224a − 496
= 144a2 − 288a − 432
= 144(a2 − 2a − 3)
= 144(a − 3)(a + 1).
 11
Thus, we have real roots only when a ≥ 3 or a ≤ −1. Therefore, a = cannot be our final
10
23
answer, since the roots are not real for this value. However f (a) = 5a2 − 11a − is a parabola
2
11
opening up and is symmetrical about its axis of symmetry a = . So we move to the nearest
10
value of a to the axis of symmetry that gives real roots, which is a = 3.
12. Let g(2) = k. Since f and g are inverse functions, we know that f (k) = 2. We need to solve
3k − 7
=2
k+1
3k − 7 = 2(k + 1)
k=9
Thus, g(2) = 9.
13. Complete the square to obtain
y = −2x2 − 4ax + k
= −2(x2 + 2ax + a2 ) + k + 2a2
= −2(x + a)2 + k + 2a2
The vertex is at (−a, k + 2a2 ) which we know is (−2, 7). Therefore, solving we obtain a = 2
and k = −1.
14. Using the sum and the product of the roots we have the four equations:
a + b = −c
ab = d
c + d = −a
cd = b

Therefore,
−(c + d) + cd = −c
cd − d = 0
d(c − 1) = 0
But none of a, b, c or d are zero, so c = 1. Then we get d = b. Substituting d = b into ab = d
we get a = 1. Then d = b = −2. Thus, a + b + c + d = −2.
15. The most common way to do this problem uses calculus. However, we make the substitution
z = x − 4. To get y in terms of z, try
y = x2 − 2x − 3
= (x − 4)2 + 6x − 19
= (x − 4)2 + 6(x − 4) + 5
= z 2 + 6z + 5
 y−4 z 2 + 6z + 1 6 1
Therefore, the value we want to minimize is 2
= 2
= 1 + + 2 . If we now let
(x − 4) z z z
1
u = , we have the parabola 1 + 6u + u2 which opens up and has its minimum at u = −3 with
z
minimum value of −8. Note that since x can assume any real value except 4, we know that z
and u will assume all real values except zero. Thus, the minimum value of this expression is
−8.

16. Solution 1
Since the function g is linear and has positive slope, it is one-to-one and so it is invertible.
This means that g −1 (g(a)) = a for every real number a and g(g −1 (b)) = b for every real number
b.
Therefore, g(f (g −1 (g(a)))) = g(f (a)) for every real number a.
This means that

g(f (a)) = g(f (g −1 (g(a))))

= 2(g(a))2 + 16g(a) + 26
= 2(2a − 4)2 + 16(2a − 4) + 26
= 2(4a2 − 16a + 16) + 32a − 64 + 26
= 8a2 − 6

Furthermore, if b = f (a), then g −1 (g(f (a))) = g −1 (g(b)) = b = f (a). Therefore,

f (a) = g −1 (g(f (a))) = g −1 (8a2 − 6)

1
Since g(x) = 2x − 4, we have y = 2g −1 (y) − 4 and so g −1 (y) = y + 2. Therefore,
2
f (a) = 12 (8a2 − 6) + 2 = 4a2 − 1

and so f (π) = 4π 2 − 1.
Solution 2
Since the function g is linear and has positive slope, it is one-to-one and so it is invertible.
To find a formula for g −1 (y), we start with the equation g(x) = 2x−4, convert to y = 2g −1 (y)−4
y+4
and then solve for g −1 (y) to obtain 2g −1 (y) = y + 4 and so g −1 (y) = .
2
We are given that g(f (g −1 (x))) = 2x2 + 16x + 26.
We can apply the function g −1 to both sides successively to obtain

f (g −1 (x)) = g −1 (2x2 + 16x + 26)

−1 (2x2 + 16x + 26) + 4
f (g (x)) = (knowing a formula for g −1 )
2
f (g −1 (x)) = x2 + 8x + 15
 
x+4
f = x2 + 8x + 15 (knowing a formula for g −1 )
2
 
x+4
f = x2 + 8x + 16 − 1
2
 
x+4
f = (x + 4)2 − 1
2

We want to determine the value of f (π).

x+4
Thus, we can replace with π, which is equivalent to replacing x + 4 with 2π.
2
Thus, f (π) = (2π)2 − 1 = 4π 2 − 1.
 Analytic Geometry
Solutions
1. Using the line segment from O(0, 0) to C(9, 0) as the base and noting that the height is 4,
the area of triangle OCD is 18. We let the vertical line be x  = k. The
 line from O(0, 0) to
1 1
D(8, 4) is y = x and this line intersects the vertical line at K k, . Let L = (k, 0) be the
2 2k
18 1
x-intercept of the vertical line. The area of triangle OKL must be = 9. Therefore, k 2 = 9
2 4
and so the vertical line required is x = 6. (The value of k = −6 is not admissible since the line
x = −6 does not intersect the triangle.)
2. There are several ways to solve this question. We will give two solutions using analytic geometry.
Solution 1
If the line is tangent to the circle, then the distance from the
√ centre (0,0) to the line y = x + c
(or x − y + c = 0) equals the radius of the circle which is 2 2. Using the formula for distance
from a point to a line,
√ |c|
2 2= √
2
Therefore, we have c = ±4.
Solution 2
We substitute y = x + c into the equation of the circle to obtain x2 + (x + c)2 = 8. Expanding
we obtain 2x2 + 2xc + c2 − 8 = 0. If the line is tangent to the circle, then we need to find values
of c such that this quadratic has exactly one root. Therefore, it must be the case that

4c2 − 4(2)(c2 − 8) = 0
4c2 − 8c2 − 64 = 0
4c2 = 64
c2 = 16
c = ±4

−5 −4 −3 −2 −1 0 1 2 3 4 5
−1
x

−2
y
−3

−4

−5

3. There are two circles, the first with its centre at (0,0) and with radius |k|, and the second with
its centre at (5,−12) and with radius 7. The distance between the centres can be calculated to
 p
be (−5)2 + (12)2 = 13. Since the two circles intersect only once, they can be either externally
or internally tangent. If they are externally tangent, |k| + 7 = 13 and so k = 6 or k = −6. If
they are internally tangent, |k| − 7 = 13 and so k = 20 or −20. So we have 4 possible values of
k: −6, 6, −20 and 20.

20
5
16
x
-5.0 -2.5 0.0 2.5 5.0 7.5 10.0
12
0 y
8
6
4
-5
0

-20 -10 0 6 10 20
7 -4
y -10 x

-8
7
-12
-15
-16
7

-20

4. Solution 1
All lines that cut a circle in half pass through the centre. Now the perpendicular bisector of
any chord passes through the centre. If we consider the vertical chord from (0, 0) to (0, 10),
the perpendicular bisector is the horizontal line y = 5. Similarly, if we consider the horizontal
chord from (0, 0) to (8, 0), the perpendicular bisector is the vertical line x = 4. These two
lines intersect at (4, 5) and therefore, the centre is (4, 5). We require the y-intercept of the line
through (4, 5) and P (2, −3). This line is y = 4x − 11 and the y-intercept is −11.
Solution 2
Observe that 4AOB is right-angled at O, thus AB is the diameter of the circle, and its midpoint
(4, 5) is the centre of the circle. As in solution 1, we require the y-intercept of the line that
goes through (4, 5) and P (2, −3). This line is y = 4x − 11 and the y-intercept is −11.
Solution 3
The general equation of a circle with centre (h, k) and radius r is (x − h)2 − (y − k)2 = r2 .
Substituting in the point (0, 0) we obtain the equation h2 + k 2 = r2 . Substituting in the point
(0, 10) we obtain the equation h2 + (10 − k)2 = r2 . Therefore, (10 − k)2 = k 2 which gives k = 5.
Substituting the point (8, 0) into the equation of the circle gives (8 − h)2 + k 2 = r2 . Therefore,
(8 − h)2 = h2 which gives h = 4. As in solution 1, we require the y-intercept of the line that
goes through (4, 5) and P (2, −3). This line is y = 4x − 11 and the y-intercept is −11.

10
A

B
0

0.0 5 10
P
-5

-10
5. Since the slopes of AB and AC are 1 and −1 respectively, the required line is vertical and its
equation is x = 0.

6. Consider a triangle with the following vertices: the origin, the centre of the circle and a point
of tangency of one of the two tangents we are considering. Since there are two tangents, we
have two such triangles. A tangent is perpendicular to the radius at the point of tangency and
so these two triangles are right-angled. The two known sides of one of these right triangles are
the radius
√ 2 and the segment from (0,0) to (3,4) which has length 5. Thus, the other side has
length 21.
6

5
2
4

2
y 3

5
2

0 2 4 6
x
-1

Since each of the tangents pass through the origin, their equations are of the form y = mx.
We are interested in values of m for which the line y = mx intersects the circle only once.
Substituting into the equation of the circle we get
(x − 3)2 + (mx − 4)2 = 4
x2 − 6x + 9 + m2 x2 − 8mx + 16 = 4
(1 + m2 )x2 − (6 + 8m)x + 21 = 0
Now this quadratic will have one solution when its discriminant is zero. Thus, we are looking
for values of m that give a discriminant of 0. So
(6 + 8m)2 − 4 · 21 · (1 + m2 ) = 0
36 + 96m + 64m2 − 84 − 84m2 = 0
−20m2 + 96m − 48 = 0
√
12 ± 2 21
m=
5
 √
√ 12 ± 2 21
Therefore, both tangents we are considering have length 21 and their slopes are .
5
7. The required set of points is the line that is the
 perpendicular
 bisector of the line segment CD.
1 3
Since CD has slope − and midpoint M = 3, , the required line passes through M and
2 2
has slope 2. The equation of the resulting line is 4x − 2y − 9 = 0.
8. We present the solution that uses analytic geometry most directly. Let the coordinates of the
points
 be  K(0,0),W (x,y),A(a,b) and D(d,0). Therefore, the coordinates of M and N are
x y a+d b
M , and N , . Now we are given that 2M N = AW + DK. Therefore,
2 2 2 2
s 2  2
a+d−x b−y p
2 + = (a − x)2 + (b − y)2 + d
2 2
Squaring both sides and simplifying (using the fact that d 6= 0) gives
p
(a + d − x)2 + (b − y)2 = (a − x)2 + (b − y)2 + 2d (a − x)2 + (b − y)2 + d2
p
2d(a − x) = 2d (a − x)2 + (b − y)2
p
(a − x) = (a − x)2 + (b − y)2
Squaring both sides again and simplifying gives
(a − x)2 = (a − x)2 + (b − y)2
(b − y)2 = 0

This result gives b = y and implies that the slope of AW is 0. Therefore, AW is parallel to
KD.
9. Let the coordinates of A and B be (a, c) and (b, d), respectively. Point A satisfies the equation
of the first line and point B satisfies the equation of the second line and so 4a + 3c − 48 = 0 and
a+b c+d
b+3d+10 = 0 . Moreover, since (4,2) is the midpoint, we know = 4 and = 2. Thus,
2 2
b = 8 − a and d = 4 − c. Substituting these into the second equation above and simplifying we
obtain −a − 3c + 30 = 0. Adding this equation to the first equation gives 3a − 18 = 0 and so
a = 6. From the first equation we obtain that c = 8. Therefore, b = 2 and d = −4. So the
coordinates of A and B are (6, 8) and B(2, −4), respectively.
 Sequences and Series
Solutions
1. It is known that
t11 + t13 187500
=
t5 + t7 1500
ar + ar12
10
= 125
ar4 + ar6
ar10 (1 + r2 )
= 125
ar4 (1 + r2 )
r6 = 125
√ 4 6
Thus, r = ± 5. Therefore, √ ar + ar = 25a√+ 125a = 150a = 1500 and so a = 10. Therefore,
the sequence begins 10, 10 5, 50 or 10, −10 5, 50.
2. Let d be the common difference in the arithmetic sequence. Since the sequence has distinct
terms, we know that d 6= 0. Then b − c = −d, c − a = 2d and a − b = −d. Thus,
−dx2 + 2dx − d = 0
−d(x − 1)2 = 0
and since d 6= 0, we have x = 1.
3. From the arithmetic sequence we have that 4 = x + d and y = 4 + d, where d is the common
difference. Therefore, x + y = 4 − d + 4 + d = 8. From the geometric sequence we have xr = 3
and 3r = y, where r is the common ratio (which is not 0 since the second term is 3). Therefore,
3 1 1 x+y 8
xy = (3r) = 9. Thus, + = = .
r x y xy 9
4. Solution 1
Since the product of the three numbers is non-zero, so is r, the common ratio of the geometric
a
sequence. We let the numbers be , a, and ar. Thus, a3 = 125 and so a = 5. Therefore,
r
5
the numbers are , 5, and 5r. Let d be the common difference of the arithmetic sequence. We
r
5
know that is the first term of the arithmetic sequence and 5 is the third term. Therefore,
r
5
5 − = 2d. We also know that 5r is the sixth term of the arithmetic sequence and therefore,
r
5r − 5 = 3d. Therefore,
5
5−
r = 2d
 5r −  5 3d
5
3 5− = 2(5r − 5)
r
3
3 − = 2r − 2
r
0 = 2r2 − 5r + 3
0 = (2r − 3)(r − 1)
 The solution r = 1 gives the sequence 5, 5, 5, but we were told that the three numbers are
3 10 15
distinct and so we discard this solution. The solution r = gives the sequence , 5, .
2 3 2
Solution 2
Since the product of the three numbers is non-zero, so is r, the common ratio of the geometric
a
sequence. We let the numbers be , a, and ar. Thus, a3 = 125 and so a = 5. We know that
r
the middle term, 5, is the third term of an arithmetic sequence.
Let d be the common difference of this arithmetic sequence. The first term is the first term
of the arithmetic sequence and therefore, it is 5 − 2d. The third term is the sixth term of the
arithmetic sequence and therefore, it is 5 + 3d. The product of these three terms is 125 and
so (5 − 2d)5(5 + 3d) = 125. Dividing both sides by 5 and simplifying gives −6d2 + 5d = 0.
5
Therefore, d = 0 or d = . If d = 0, then the sequence is 5, 5, 5, but we were told that the
6
5 10 15
three numbers are distinct and so we discard this solution. If d = , the sequence is , 5, .
6 3 2
5. Our sum is
N PN PN
X k2 + k k=1 k2 + k=1 k
=
k=1
2 2
N (N + 1)(2N + 1) N (N + 1)
+
= 6 2
 2 
N (N + 1) 2N + 1
= +1
4 3
 
N (N + 1) 2N + 4
=
4 3
N (N + 1)(N + 2)
=
6
Taking N = 200 we get
200
X k2 + k 200 · 201 · 202
=
k=1
2 6
= 1353400.

6. Represent the angles as a − 2d, a − d, a, a + d and a + 2d. The sum of these values is 540◦ .
Therefore, 5a = 540◦ and so a = 108◦ . So either a − d = 90◦ or a − 2d = 90◦ . So the largest
angle is either 126◦ or 144◦ .

7. We let the four positive integers be represented by k, kr, kr2 and kr3 . Then

kr + kr2 = 30 (1)
k + kr3 = 35 (2)
 Dividing (2) by (1) gives
k + kr3 35
2
=
kr + kr 30
1 + r3 7
= (k 6= 0 since kr + kr2 = 30)
r + r2 6
6r3 − 7r2 − 7r + 6 = 0
By inspection, we find that r = −1 is a solution. Using the factor theorem and long division,
we arrive at
(r + 1)(2r − 3)(3r − 2) = 0
2 3
So r = −1, or
3 2
Using r = −1, equation (2) gives 0k = 35, which is impossible. (It would also violate the
condition a < b < c < d. )
2
Using r = in (1), we find k = 27.
3
3
Using r = in (1) we find k = 8.
2
Both of these value give the same list of numbers, and when arranged in increasing order they
are (a, b, c, d) = (8, 12, 18, 27).
8. The sequence is arithmetic if and only if t1 + t3 = 2t2 . There are 27 equally likely ways to pick
three numbers, of which only five lead to such a sequence:
1, 4, 7
1, 5, 9
2, 5, 8
3, 5, 7
3, 6, 9
5
So the probability is .
27
9. Solution 1
Since there are an odd number of integers, the average of the integers is the middle integer.
500
Therefore, the middle integer is = 20. Thus, the smallest integer is 8.
25
Solution 2
The common difference is 1 and the number of terms is 25. Therefore, using the sum of an
25
arithmetic sequence we get 500 = (a + (a + 24)), which simplifies to 40 = 2a + 24. Therefore,
2
a = 8,
10. The common difference is d = 2, the first term is a = −1994 and so
−1994 + 2(n − 1) = −1994
Solving for n gives n = 1995.
11. (a) S1 = t1 = 31 − 1 = 2.
S2 = t1 + t2 = 32 − 1 = 8 and so t2 = 8 − 2 = 6.
S3 = t1 + t2 + t3 = 33 − 1 = 26 and so t3 = 26 − 8 = 18.
(b)
tn+1 Sn+1 − Sn
=
tn Sn − Sn−1
(3n+1 − 1) − (3n − 1)
= n
(3 − 1) − (3n−1 − 1)
3n · (3 − 1)
= n−1
3 · (3 − 1)
= 3.

12. We can see that the nth term of the sequence is 7n. The smallest multiple of 7 that is greater
than 40 is 42 and the largest multiple of 7 that is less than 28001 is 28000 (We see that
28001 28000 − (42)
≈ 4000.1 and 7 · 4000 = 28000.) So n − 1 = and n = 3995.
7 7
1
13. We know f (n + 1) = f (n) + and so the the function evaluated at positive integers gives
3
1
a sequence that is arithmetic. Its first term is 2 and its common difference is . Therefore,
  3
1
f (100) = 2 + 99 = 35.
3
14. Substituting for x and y, −p + 2q = r so q − p = r − q and we are done!
15. If the common difference is 0, then the sequence is also a geometric sequence with a common
ratio of 1. In this case, any three terms form a three-term geometric sequence.
Let’s consider what happens when d 6= 0. For any three-term geometric sequence, x1 , x2 , x3 we
have x1 x3 = (x2 )2 . So
(a + 4d)(a + 15d) = (a + 8d)2
a2 + 19ad + 60d2 = a2 + 16ad + 64d2
3ad = 4d2
3
d = a (Since d 6= 0)
4
Thus, the general term is
3
tk = a + (k − 1) a
4
a
= (3k + 1)
4
Therefore,
a
t9 (3 · 9 + 1) 7
r= 4
= a =
t5 (3 · 5 + 1) 4
4
 We need to find an infinite number of triples (i, j, k) such that
tj tk 7
= =
ti tj 4

which is to say that

3j + 1 3k + 1 7
= =
3i + 1 3j + 1 4
Therefore, 4(3j + 1) = 7(3i + 1), which implies that 3j + 1 is a multiple of 7 and 3i + 1 is a
multiple of 4.
Also, 4(3k + 1) = 7(3j + 1), which implies that 3k + 1 is a multiple of 7 and 3j + 1 is a multiple
of 4. So 3j + 1 must be a multiple of 28. Let 3j + 1 = 28n for some integer n.
We also have that (3j + 1)2 = (3i + 1)(3k + 1) and so (3i + 1)(3k + 1) must be a multiple of 282 .
So if we make 3i + 1 = 16n and 3k + 1 = 49n, then we will have satisfied all the conditions.
However, we need to guarantee that i, j and k are positive integers. We note that

3i + 1 = 16n = 3(5n) + n
3j + 1 = 28n = 3(9n) + n
3k + 1 = 49n = 3(16n) + n

So if we choose n such that it is 1 more than a multiple of 3, then i, j and k will be integers.
Therefore, let n = 3m + 1 for some non-negative integer m and we obtain

16(3m + 1) − 1
i= = 16m + 5
3
28(3m + 1) − 1
j= = 28m + 9
3
49(3m + 1) − 1
k= = 49m + 16
3
7
For each value of m we will obtain a three-term geometric sequence with common ratio .
4
16. The sequence goes 5, 3, −2, −5, −3, 2, 5, 3, . . .. The sequence repeats in groups of 6 whose sum
is 0. So the sum of 32 terms is 5 + 3 = 8.

17.
1995
t1998 = × t1996
1997
1995 1997
= × × t1994
1997 1995
1995 1997 1995 3 1
= × × × · · · × × × t2
1997 1995 1993 5 3
1
=−
1997
18. The first term is t1 = 555 − 7 = 548 and the common difference is −7. Therefore, the sum is
n
Sn = [2(548) + (n − 1)(−7)]. Thus, the sum is negative when 1096 + (n − 1)(−7) < 0. Solving
2
the equality 1096 − 7n + 7 = 0 we obtain n ≈ 157.6. We note that S157 = 314 and S158 = −237
Therefore, the smallest value of n for which Sn is negative is n = 158.
 Euclidean Geometry
Solutions
1. The area of quadrilateral ABCD is the sum of the areas of 4ABC and 4ACD.
1 1
Since 4ABC is right-angled at B, its area equals (AB)(BC) = (3)(4) = 6.
2 2
Since 4ABC is right-angled at B, by the Pythagorean Theorem,
√ √ √
AC = AB 2 + BC 2 = 32 + 42 = 25 = 5

because AC > 0. (We could have also observed that 4ABC must be a “3-4-5” triangle.)
Since 4ACD is right-angled at A, by the Pythagorean Theorem,
√ √ √
AD = CD2 − AC 2 = 132 − 52 = 144 = 12

because AD > 0. (We could have also observed that 4ACD must be a “5-12-13” triangle.)
1 1
Thus, the area of 4ACD equals (AC)(AD) = (5)(12) = 30. Finally, the area of quadrilat-
2 2
eral ABCD is thus 6 + 30 = 36.

2. Solution 1
Suppose that the rectangular prism has dimensions a cm by b cm by c cm.
Suppose further that one of the faces that is a cm by b cm is the face with area 27 cm2 and
that one of the faces that is a cm by c cm is the face with area 32 cm2 . (Since every pair of
non-congruent faces shares exactly one side length, there is no loss of generality in picking these
particular variables for these faces.) Therefore, ab = 27 and ac = 32.
Further, we are told that the volume of the prism is 144 cm3 , and so abc = 144.
a2 b 2 c 2 (abc)2 1442
Thus, bc = 2 = = = 24. (We could also note that abc = 144 means
a bc (ab)(ac) (27)(32)
1442
a2 b2 c2 = 1442 or (ab)(ac)(bc) = 1442 and so bc = .)
(27)(32)
In other words, the third type of face of the prism has area 24 cm2 .
Thus, since the prism has two faces of each type, the surface area of the prism is equal to
2(27 cm2 + 32 cm2 + 24 cm2 ) or 166 cm2 .
Solution 2
Suppose that the rectangular prism has dimensions a cm by b cm by c cm.
Suppose further that one of the faces that is a cm by b cm is the face with area 27 cm2 and
that one of the faces that is a cm by c cm is the face with area 32 cm2 . (Since every pair of
non-congruent faces shares exactly one side length, there is no loss of generality in picking these
particular variables for these faces.) Therefore, ab = 27 and ac = 32.
Further, we are told that the volume of the prism is 144 cm3 , and so abc = 144.
144 16
Since abc = 144 and ab = 27, we have c = = .
27 3
 144 9
Since abc = 144 and ac = 32, we have b = = .
32 2
16 9
This means that bc = · = 24.
3 2
In cm2 , the surface area of the prism equals 2ab + 2ac + 2bc = 2(27) + 2(32) + 2(24) = 166.
Thus, the surface area of the prism is 166 cm2 .
3. Let the radius of the smaller circle be r cm and let the radius of the larger circle be R cm.
Thus, the circumference of the smaller circle is 2πr cm, the circumference of the larger circle is
2πR cm, the area of the smaller circle is πr2 cm2 , and the area of the larger circle is πR2 cm2 .
Since the sum of the radii of the two circles is 10 cm, we have r + R = 10.
Since the circumference of the larger circle is 3 cm larger than the circumference of the smaller
circle, it follows that 2πR − 2πr = 3, or 2π(R − r) = 3.
Then the difference, in cm2 , between the area of the larger circle and the area of the smaller
circle is
1 1
πR2 − πr2 = π(R − r)(R + r) = [2π(R − r)](R + r) = (3)(10) = 15
2 2
Therefore, the difference between the areas is 15 cm2 .
4. Since ABC is a quarter of a circular pizza with centre A and radius 20 cm, we have
AC = AB = 20 cm. We are also told that ∠CAB = 90◦ (one-quarter of 360◦ ).
Since ∠CAB = 90◦ and A, B and C are all on the circumference of the circle, it follows that
CB is a diameter of the pan. (This is a property of circles: if X, Y and Z are three points on
a circle with ∠ZXY = 90◦ , then Y Z must be a diameter of the circle.)
√ √
Since 4CAB is a right-angled isosceles triangle, we have CB = 2AC = 20 2 cm. Therefore,
1 √
the radius of the circular plate is CB or 10 2 cm. Thus, the area of the circular pan is
√ 2
π(10 2 cm)2 = 200π cm2 .
The area of the slice of pizza is one-quarter of the area of a circle with radius 20 cm, or
1
π(20 cm)2 = 100π cm2 .
4
Finally, the fraction of the pan that is covered is the area of the slice of pizza divided by the
100π cm2 1
area of the pan, or 2
= .
200π cm 2
5. Solution 1
Since 4AF D is right-angled at F , by the Pythagorean Theorem,
√ √ √ √
AD = AF 2 + F D2 = 42 + 22 = 20 = 2 5

since AD > 0.
Let ∠F AD = β. Since ABCD is a rectangle, we have ∠BAF = 90◦ − β.
Since 4AF D is right-angled at F , we have ∠ADF = 90◦ − β.
Since ABCD is a rectangle, we have ∠BDC = 90◦ − (90◦ − β) = β.
Therefore, 4BF A, 4AF D, and 4DF E are all similar as each is right-angled and has either
an angle of β or an angle of 90◦ − β (and√hence both of these angles).
AB DA 4(2 5) √
Therefore, = and so AB = = 4 5.
AF DF 2
FE FD 2(2)
Also, = and so F E = = 1.
FD FA 4
√ √
Since ABCD is a rectangle, we have BC = AD = 2 5, and DC = AB = 4 5.
Finally, the area of quadrilateral BCEF equals the area of 4DCB minus the area 4DF E.
Thus, the required area is
1 1 1 √ √ 1
(DC)(CB) − (DF )(F E) = (4 5)(2 5) − (2)(1) = 20 − 1 = 19
2 2 2 2

Solution 2
Since 4AF D is right-angled at F , by the Pythagorean Theorem,
√ √ √ √
AD = AF 2 + F D2 = 42 + 22 = 20 = 2 5

since AD > 0.
Let ∠F AD = β. Since ABCD is a rectangle, we have ∠BAF = 90◦ − β. Since 4BAF is
right-angled at F , we have ∠ABF = β.
Since 4AF D is right-angled at F , we have ∠ADF = 90◦ − β.
Since ABCD is a rectangle, we have ∠BDC = 90◦ − (90◦ − β) = β.

FD 2 1 AF 4 2
Looking at 4AF D, we see that sin β = = √ = √ , cos β = = √ = √ , and
AD 2 5 5 AD 2 5 5
FD 2 1
tan β = = = .
AF 4 2
AF 4 √
Since AF = 4 and ∠ABF = β, we have AB = = 1 = 4 5.
sin β √
5
 1
Since F D = 2 and ∠F DE = β, we have F E = F D tan β = 2 · = 1.
2
√ √
Since ABCD is a rectangle, we have BC = AD = 2 5, and DC = AB = 4 5.
Finally, the area of quadrilateral EF BC equals the area of 4DCB minus the area 4DF E.
Thus, the required area is
1 1 1 √ √ 1
(DC)(CB) − (DF )(F E) = (4 5)(2 5) − (2)(1) = 20 − 1 = 19
2 2 2 2

6. Join BE.

Since 4F BD is congruent to 4AEC, we have F B = AE.

Since 4F AB and 4AF E are each right-angled, share a common side AF and have equal
hypotenuses (F B = AE), it follows that these triangles are congruent, and so AB = F E.
Now BAF E has two right angles at A and F (so AB and F E are parallel) and has equal sides
AB = F E so must be a rectangle. This means that BCDE is also a rectangle.
Now the diagonals of a rectangle partition it into four triangles of equal area. (Diagonal AE
of the rectangle splits the rectangle into two congruent triangles, which have equal area. The
diagonals bisect each other, so the four smaller triangles all have equal area.)
Since one quarter of rectangle ABEF is shaded and one quarter of rectangle BCDE is shaded,
it follows that one quarter of the total area is shaded. (If the area of ABEF is x and the area
1 1
of BCDE is y, then the total shaded area is x + y, which is one quarter of the total area
4 4
x + y.)
Since AC = 200 and CD = 50, the area of rectangle ACDF is 200(50) = 10 000, so the total
1
shaded area is (10 000) = 2500.
4
7. Suppose that M is the midpoint of Y Z. Suppose that the centre of the smaller circle is O and
the centre of the larger circle is P . Suppose that the smaller circle touches XY at C and XZ
at D, and that the larger circle touches XY at E and XZ at F . Join OC, OD and P E.
 Since OC and P E are radii that join the centres of circles to points of
tangency, it follows that OC and P E are perpendicular to XY .
Construct XM . Since 4XY Z is isosceles, XM (which is a median by
construction) is an altitude (that is, XM is perpendicular to Y Z) and
an angle bisector (that is, ∠M XY = ∠M XZ).
Now XM passes through O and P . (Since XC and XD are tangents
from X to the same circle, we have XC = XD. This means that
4XCO is congruent to 4XDO by side-side-side. This means that
∠OXC = ∠OXD and so O lies on the angle bisector of ∠CXD, and
so O lies on XM . Using a similar argument, P lies on XM .)
Draw a perpendicular from O to T on P E. Note that OT is parallel to
XY (since each is perpendicular to P E) and that OCET is a rectangle
(since it has three right angles).
Consider 4XM Y and 4OT P . Each triangle is right-angled (at M and at T , respectively).
Also, ∠Y XM = ∠P OT . (This is because OT is parallel to XY , since both are perpendicular
XY OP
to P E.) Therefore, 4XM Y is similar to 4OT P . Thus, = .
YM PT
1
Now XY = a and Y M = b. Also, OP is the line segment joining the centres of two tangent
2
circles, so OP = r + R.
Lastly, P T = P E − ET = R − r, since P E = R, ET = OC = r, and OCET is a rectangle.
Therefore,
a R+r
=
b/2 R−r
2a R+r
=
b R−r
2a(R − r) = b(R + r)
2aR − bR = 2ar + br
R(2a − b) = r(2a + b)
R 2a + b
= (since 2a > b we have 2a − b 6= 0, and r > 0)
r 2a − b
R 2a + b
Therefore, = .
r 2a − b
8. Let ∠P EQ = θ. Join P to B.
We use the fact that the angle between a tangent to a circle and a chord in that circle that
passes through the point of tangency equals the angle inscribed by that chord. We prove this
fact below.
More concretely, ∠DEP = ∠P BE (using the chord EP and the tangent through E) and
∠ABP = ∠P EQ = θ (using the chord BP and the tangent through B).
Now ∠DEP is exterior to 4F EP and so ∠DEP = ∠F P E + ∠EF P = 25◦ + 30◦ , and
so ∠P BE = ∠DEP = 55◦ . Furthermore, ∠AQB is an exterior angle of 4P QE. Thus,
∠AQB = ∠QP E + ∠P EQ = 25◦ + θ.
 In 4ABQ, we have ∠BAQ = 35◦ , ∠ABQ = θ + 55◦ , and ∠AQB = 25◦ + θ.
Thus, 35◦ + (θ + 55◦ ) + (25◦ + θ) = 180◦ or 115◦ + 2θ = 180◦ , and so 2θ = 65◦ . Therefore
1
∠P EQ = θ = (65◦ ) = 32.5◦ .
2
As an addendum, we prove that the angle between a tangent to a circle and a chord in that
circle that passes through the point of tangency equals the angle inscribed by that chord.
Consider a circle with centre O and a chord XY , with tangent ZX
meeting the circle at X. We prove that if ZX is tangent to the circle,
then ∠ZXY equals ∠XW Y whenever W is a point on the circle on
the opposite side of XY as XZ (that is, the angle subtended by XY
on the opposite side of the circle).
We prove this in the case that ∠ZXY is acute. The cases where
∠ZXY is a right angle or an obtuse angle are similar.
Draw diameter XOV and join V Y .
Since ∠ZXY is acute, points V and W are on the same arc of chord XY . This means that
∠XV Y = ∠XW Y , since they are angles subtended by the same chord.
Since OX is a radius and XZ is a tangent, it follows that ∠OXZ = 90◦ . Therefore, we have
∠OXY + ∠ZXY = 90◦ .
Since XV is a diameter, we have ∠XY V = 90◦ .
From 4XY V , we see that ∠XV Y + ∠V XY = 90◦ .
But ∠OXY + ∠ZXY = 90◦ and ∠XV Y + ∠V XY = 90◦ and ∠OXY = ∠V XY tells us that
∠ZXY = ∠XV Y . This gives us that ∠ZXY = ∠XW Y , as required.

9. Solution 1
Draw a line segment through M in the plane of 4P M N parallel to P N and extend this line
until it reaches the plane through P , A and D at Q on one side and the plane through N , B
and C at R on the other side.
Join Q to P and A. Join R to N and B.
So the volume of solid ABCDP M N equals the volume of solid ABCDP QRN minus the
volumes of solids P M QA and N M RB.
Solid ABCDP QRN is a trapezoidal prism. This is because N R and BC are parallel (since
they lie in parallel planes), which makes N RBC a trapezoid. Similarly, P QAD is a trapezoid.
Also, P N , QR, DC, and AB are all perpendicular to the planes of these trapezoids and equal
in length, since they equal the side lengths of the squares.
Solids P M QA and N M RB are triangular-based pyramids. We can think of their bases as
being 4P M Q and 4N M R. Their heights are each equal to 2, the height of the original solid.
(The volume of a triangular-based pyramid equals 31 times the area of its base times its height.)
The volume of ABCDP QRN equals the area of trapezoid N RBC times the width of the prism,
which is 2.
1 1
That is, this volume equals (N R + BC)(N C)(N P ) = (N R + 2)(2)(2) = 2 · N R + 4.
2 2
So we need to find the length of N R.
Consider quadrilateral P N RQ. This quadrilateral is a rectangle since P N and QR are perpen-
dicular to the two side planes of the original solid. Thus, N R equals the height of 4P M N .
Join M to the midpoint T of P N . Since 4P M N is isosceles, M T is perpendicular to P N .

1
Since N T = P N = 1 and ∠P M N = 90◦ and ∠T N M = 45◦ , it follows that 4M T N is also
2
right-angled and isosceles with M T = T N = 1.
Therefore, N R = M T = 1 and so the volume of ABCDP QRN is 2 · 1 + 4 = 6.
The volumes of solids P M QA and N M RB are equal. Each has height 2 and their bases 4P M Q
and 4N M R are congruent, because each is right-angled (at Q and at R) with P Q = N R = 1
and QM = M R = 1.
 
1 1 1
Thus, using the formula above, the volume of each is (1)(1) 2 = .
3 2 3
1 16
Finally, the volume of the original solid equals 6 − 2 · = .
3 3
Solution 2
We determine the volume of ABCDP M N by splitting it into two solids: ABCDP N and
ABN P M by slicing along the plane of ABN P .
Solid ABCDP N is a triangular prism, since 4BCN and 4ADP are each right-angled (at C
and D), BC = CN = AD = DP = 2, and segments P N , DC and AB are perpendicular to
each of the triangular faces and equal in length.
Thus, the volume of ABCDP N equals the area of 4BCN times the length of DC. Therefore,
1 1
(BC)(CN )(DC) = (2)(2)(2) = 4. (This solid can also be viewed as “half” of a cube.)
2 2
Solid ABN P M is a pyramid with rectangular base ABN P . (Note that P N and AB are
perpendicular to the planes of both
√ of the side
√ triangular faces of the original solid, that
P N = AB = 2 and BN = AP = 22 + 22 = 2 2, by the Pythagorean Theorem.)
√
1 4 2
Therefore, the volume of ABN P M equals (AB)(BN )h = h, where h is the height of the
3 3
pyramid (that is, the distance that M is above plane ABN P ).
So we need to calculate h.
Join M to the midpoint, T , of P N and to the midpoint, S,
of AB. Join S and T . By symmetry, M lies directly above ST .
Since ABN P is a rectangle and S and √ T are the midpoints of
opposite sides, we have ST = AP = 2 2.
Since 4P M N is right-angled and isosceles, M T is perpen-
1
dicular to P N . Since N T = P N = 1 and ∠T N M = 45◦ ,
2
it follows that 4M T N is also right-angled and isosceles with
M T = T N = 1.

Also, M S is the hypotenuse of the triangle formed by dropping

a perpendicular from M to U in the plane of ABCD (a distance
of 2) and joining U to S. Since M is 1 unit horizontally from
P N , we have √
U S = 1. √
Thus, M S = 22 + 12 = 5 by the Pythagorean Theorem.

We can now consider 4SM T . h is the height of this triangle, from M to base ST .

Now h = M T sin(∠M T S) = sin(∠M T S).

By the cosine law in 4SM T , we have

M S 2 = ST 2 + M T 2 − 2(ST )(M T ) cos(∠M T S)

 √ √
Therefore, 5 = 8 + 1 − 4 2 cos(∠M T S) or 4 2 cos(∠M T S) = 4.
1 1
Thus, cos(∠M T S) = √ and so ∠M T S = 45◦ which gives h = sin(∠M T S) = √ .
2 2
(Alternatively, we note that the plane of ABCD is parallel to the plane of P M N , and so since
the angle between plane ABCD and plane P N BA is 45◦ , it follows that the angle between
plane P N BA and plane P M N is also 45◦ , and so ∠M T S = 45◦ .)
√
4 2 1 4
Finally, this means that the volume of ABN P M is · √ = , and so the volume of solid
3 2 3
4 16
ABCDP M N is 4 + = .
3 3
10. Let ∠EAF = θ. Since ABCD is a parallelogram, AB and DC are parallel with AB = DC,
and DA and CB are parallel with DA = CB.
Since AE is perpendicular to DC and AB and DC are parallel, it follows that AE is perpen-
dicular to AB. In other words, ∠EAB = 90◦ , and so ∠F AB = 90◦ − θ.
Since 4AF B is right-angled at F and ∠F AB = 90◦ − θ, we have ∠ABF = θ.
Using similar arguments, we obtain that ∠DAE = 90◦ − θ and ∠ADE = θ.

1
Since cos(∠EAF ) = cos θ = and cos2 θ + sin2 θ = 1, we have
3
√
√
r r
1 8 2 2
sin θ = 1 − cos2 θ = 1 − = =
9 9 3
(Note that sin θ > 0 since θ is an angle in a triangle.)
AF FB
In 4AF B, sin θ = and cos θ = .
AB AB
√
2 2 AF 32 48 √
Since AF = 32 and sin θ = , we have AB = = √ = √ = 24 2.
3 sin θ 2 2/3 2
√ 1 √ 1 √
Since AB = 24 2 and cos θ = , we have F B = AB cos θ = 24 2( ) = 8 2.
3 3
AE DE
In 4AED, sin θ = and cos θ = .
AD AD
√
2 2 AE 20 30 √
Since AE = 20 and sin θ = , we have AD = = √ = √ = 15 2.
3 sin θ 2 2/3 2
√ √ 1 √
 
1
Since AD = 15 2 and cos θ = , we have DE = AD cos θ = 15 2 = 5 2. (To calculate
3 3
AD and DE, we could also have used the fact that 4ADE is similar to 4ABF .)
 Finally, the area of quadrilateral AECF equals the area of parallelogram ABCD minus the
combined areas of 4AF B and 4ADE.
√ √
The area of parallelogram ABCD equals AB · AE = 24 2 · 20 = 480 2.
1 1 √ √
The area of 4AF B equals (AF )(F B) = (32)(8 2) = 128 2.
2 2
1 1 √ √
The area of 4AED equals (AE)(DE) = (20)(5 2) = 50 2.
2 2
√ √ √ √
Thus, the area of quadrilateral AECF is 480 2 − 128 2 − 50 2 = 302 2.

11. Solution 1
Consider 4BCE and 4ACD.

Since 4ABC is equilateral, we have BC = AC. Since 4ECD is equilateral, we have that
CE = CD.
Since BCD is a straight line and ∠ECD = 60◦ , we have ∠BCE = 180◦ − ∠ECD = 120◦ .
Since BCD is a straight line and ∠BCA = 60◦ , we have ∠ACD = 180◦ − ∠BCA = 120◦ .
Therefore, 4BCE is congruent to 4ACD (“side-angle-side”).
Since 4BCE and 4ACD are congruent and CM and CN are line segments drawn from
the corresponding vertex (C in both triangles) to the midpoint of the opposite side, we have
CM = CN .
Since ∠ECD = 60◦ , 4ACD can be obtained by rotating 4BCE through an angle of 60◦
clockwise about C. This means that after this 60◦ rotation, CM coincides with CN . In other
words, ∠M CN = 60◦ .
But since CM = CN and ∠M CN = 60◦ , we have
1
∠CM N = ∠CN M = (180◦ − ∠M CN ) = 60◦
2
Therefore, 4M N C is equilateral, as required.
Solution 2
We prove that 4M N C is equilateral by introducing a coordinate system.
Suppose that C is at the origin (0, 0) with BCD along the x-axis, with B having coordinates
(−4b, 0) and D having coordinates (4d, 0) for some real numbers b, d > 0.
Drop a perpendicular from E to P on CD.
Since 4ECD is equilateral, P is the midpoint of CD.
Since C has coordinates (0, 0) and D has coordinates (4d, 0), it follows that the coordinates of
P are (2d, 0).
Since 4ECD√ is equilateral,
√ we have ∠ECD = 60◦ and so 4EP C is a 30◦ -60◦ -90◦ triangle and
so EP = 3CP = 2 3d.
√
Therefore, the coordinates of E are (2d, 2 3d).
√
In a similar way, we can show that the coordinates of A are (−2b, 2 3b).
√
Now M is the midpoint of B(−4b, 0) and E(2d, 2 3d), and therefore, the coordinates of M are
√ √
 
1 1
(−4b + 2d), (0 + 2 3d) or (−2b + d, 3d).
2 2
√
Also, N is the midpoint of A(−2b, 2 3b) and D(4d, 0), and therefore, the coordinates of N are
1 √ √
 
1
(−2b + 4d), (2 3b + 0) or (−b + 2d, 3b).
2 2
To show that 4M N C is equilateral, we show that CM = CN = M N or equivalently that
CM 2 = CN 2 = M N 2 :
√
CM 2 = (−2b + d − 0)2 + ( 3d − 0)2
√
= (−2b + d)2 + ( 3d)2
= 4b2 − 4bd + d2 + 3d2
= 4b2 − 4bd + 4d2
√
CN 2 = (−b + 2d − 0)2 + ( 3b − 0)2
√
= (−b + 2d)2 + ( 3b)2
= b2 − 4bd + 4d2 + 3b2
= 4b2 − 4bd + 4d2
√ √
M N 2 = ((−2b + d) − (−b + 2d))2 + ( 3d − 3b)2
= (−b − d)2 + 3(d − b)2
= b2 + 2bd + d2 + 3d2 − 6bd + 3b2
= 4b2 − 4bd + 4d2

Therefore, CM 2 = CN 2 = M N 2 and so 4M N C is equilateral, as required.

12. Since P Q is parallel to AB, it is parallel to DC and is perpendicular to BC.
Drop perpendiculars from A to E on P Q and from P to F on DC.

Then ABQE and P QCF are rectangles. Thus, EQ = x, which means that P E = r − x and
F C = r, which means that DF = y − r.
Let BQ = b and QC = c. Thus, AE = b and P F = c.
1
The area of trapezoid ABQP is (x + r)b.
2
1
The area of trapezoid P QCD is (r + y)c.
2
1 1 x+r c
Since these areas are equal, we have (x + r)b = (r + y)c, which gives = .
2 2 r+y b
Since AE is parallel to P F , we have ∠P AE = ∠DP F and 4AEP is similar to 4P F D.
AE PF b c c y−r
Thus, = which gives = or = .
PE DF r−x y−r b r−x
x+r c c y−r x+r y−r
Combining = and = gives = or (x + r)(r − x) = (r + y)(y − r).
r+y b b r−x r+y r−x
From this, we get r2 − x2 = y 2 − r2 or 2r2 = x2 + y 2 , as required.

13. Suppose that the parallel line segments EF and W X are a distance of x apart.
This means that the height of trapezoid EF XW is x.
Since the side length of square EF GH is 10 and the side length of square W XY Z is 6, it
follows that the distance between parallel line segments ZY and HG is 10 − 6 − x or 4 − x.
Recall that the area of a trapezoid equals one-half times its height times the sum of the lengths
of the parallel sides.
1 1
Thus, the area of trapezoid EF XW is x(EF + W X) = x(10 + 6) = 8x.
2 2
1 1
Also, the area of trapezoid GHZY is (4 − x)(HG + ZY ) = (4 − x)(10 + 6) = 32 − 8x.
2 2
Therefore, the sum of the areas of trapezoids EF XW and GHZY is 8x + (32 − 8x) = 32.
This sum is a constant and does not depend on the position of the inner square within the
outer square, as required.
 Counting and Probability
Solutions
1. When an integer is divided by 3, there are only three possibilities for the remainder: 0, 1, or
2. Therefore, each of these integers is of the form 3k, 3k + 1 or 3k + 2 for some non-negative
integer k. If we don’t have three integers each of a different form, then the five integers have
two possible forms and so there must at least three integers of the same form.
If we have three integers of the same form, then the sum of these three integers is

3k1 + 3k2 + 3k3 = 3(k1 + k2 + k3 )

or
3k1 + 1 + 3k2 + 1 + 3k3 + 1 = 3(k1 + k2 + k3 + 1)
or
3k1 + 2 + 3k2 + 2 + 3k3 + 2 = 3(k1 + k2 + k3 + 2)

If we have three integers all of different form, then the sum of these three integers is

3k1 + 3k2 + 1 + 3k3 + 2 = 3(k1 + k2 + k3 + 1)

In all cases, these sums are divisible by 3 and so we can always choose three of the five integers
such that their sum is divisible by 3.

2. There are 7! = 7 · 6 · 5 · 4 · 3 · 2 · 1 permutations of 1, 2, 3, 4, 5, 6, 7, because there are 7 choices

for a1 , then 6 choices for a2 , and so on.
We determine the average value of a1 − a2 + a3 − a4 + a5 − a6 + a7 over all of these permutations
by determining the sum of all 7! values of this expression and dividing by 7!.
To determine the sum of all 7! values, we determine the sum of the values of a1 in each of these
expressions and call this total s1 , the sum of the values of a2 in each of these expressions and
call this total s2 , and so on.
The sum of the 7! values of the original expression must equal s1 − s2 + s3 − s4 + s5 − s6 + s7 .
This uses the fact that, when adding, the order in which we add the same set of numbers does
not matter.
By symmetry, the sums of the values of a1 , a2 , a3 , a4 , a5 , a6 , a7 will all be equal. That is,
s1 = s2 = s3 = s4 = s5 = s6 = s7 .
This means that the desired average value equals

s1 − s2 + s3 − s4 + s5 − s6 + s7 (s1 + s3 + s5 + s7 ) − (s2 + s4 + s6 ) 4s1 − 3s1 s1

= = =
7! 7! 7! 7!
So we need to determine the value of s1 .
Now a1 can equal each of 1, 2, 3, 4, 5, 6, 7.
If a1 = 1, there are 6! combinations of values for a2 , a3 , a4 , a5 , a6 , a7 , since there are still 6
choices for a2 , 5 for a3 , and so on.
 Similarly, there are 6! combinations with a1 equal to each of 2, 3, 4, 5, 6, 7.
Thus, s1 = 1 · 6! + 2 · 6! + 3 · 6! + 4 · 6! + 5 · 6! + 6 · 6! + 7 · 6! = 6!(1 + 2 + 3 + 4 + 5 + 6 + 7) = 28(6!).
28(6!) 28(6!) 28
Therefore, the average value of the expression is = = = 4.
7! 7(6!) 7
3. We consider choosing the three numbers all at once.
We list the possible sets of three numbers that can be chosen:

{1, 2, 3} {1, 2, 4} {1, 2, 5} {1, 3, 4} {1, 3, 5} {1, 4, 5} {2, 3, 4} {2, 3, 5} {2, 4, 5} {3, 4, 5}

We have listed each in increasing order because once the numbers are chosen, we arrange them
in increasing order.
There are 10 sets of three numbers that can be chosen.
Of these 10 sets, the 4 sequences 1, 2, 3 and 1, 3, 5 and 2, 3, 4 and 3, 4, 5 are arithmetic sequences.
4
Therefore, the probability that the resulting sequence is an arithmetic sequence is 10
or 25 .

4. There are five odd digits: 1, 3, 5, 7, 9.

We consider the positive integers less than 1000 in three sets: those with one digit, those with
two digits, and those with three digits.
There are 5 positive one-digit integers with one odd digit (namely 1, 3, 5, 7, 9).
Consider the two-digit positive integers with only odd digits. Such an integer has the form XY
where X and Y are digits. There are five possibilities for each of X and Y (since each must be
odd). Therefore, there are 5 × 5 = 25 two-digit positive integers with only odd digits.
Consider the three-digit positive integers with only odd digits. Such an integer has the form
XY Z where X, Y and Z are digits. There are five possibilities for each of X, Y and Z (since
each must be odd). Therefore, there are 5 × 5 × 5 = 125 three-digit positive integers with only
odd digits.
In total, there are 5 + 25 + 125 = 155 positive integers less than 1000 with only odd digits.

5. We make a table of the 36 possible combinations of rolls and the resulting sums:

2 3 5 7 11 13
2 4 5 7 9 13 15
3 5 6 8 10 14 16
5 7 8 10 12 16 18
7 9 10 12 14 18 20
11 13 14 16 18 22 24
13 15 16 18 20 24 26

Of the 36 entries in the table, 6 are prime numbers (two entries each of 5, 7 and 13). Therefore,
6
the probability that the sum is a prime number is 36 or 16 .
(Note that each sum is at least 4 and so must be odd to be prime. Since odd plus odd equals
even, then the only possibilities that really need to be checked are even plus odd and odd plus
even, that is, the first row and first column of the table.)
6. Solution 1
There are two possibilities: either each player wins three games or one player wins more games
than the other.
5
Since the probability that each player wins three games is 16 , the probability that any one
5 11
player wins more games than the other is 1 − = .
16 16
Since each of Blaise and Pierre is equally likely to win any given game, each must be equally
likely to win more games than the other.
1 11 11
Therefore, the probability that Blaise wins more games than Pierre is × = .
2 16 32
Solution 2
We consider the results of the 6 games as a sequence of 6 Bs or Ps, with each letter a B if Blaise
wins the corresponding game or P if Pierre wins.
Since the two players are equally skilled, the probability that each wins a given game is 12 . This
means that the probability of each letter being a B is 12 and the probability of each letter being
a P is also 21 .
Since
 6 each sequence consists of 6 letters, the probability of a particular sequence occurring is
1 1
= , because each of the letters is specified.
2 64
Since they play 6 games in total, the probability that Blaise wins more games than Pierre is the
sum of the probabilities that Blaise wins 4 games, that Blaise wins 5 games, and that Blaise
wins 6 games.
1
If Blaise wins 6 games, then the sequence consists of 6 Bs. The probability of this is 64
, since
there is only one way to arrange 6 Bs.
If Blaise wins 5 games, then the sequence consists of 5 Bs and 1 P. The probability of this is
1 6
6× = , since there are 6 possible positions in the list for the 1 P (eg. PBBBBB, BPBBBB,
64 64
BBPBBB, BBBPBB, BBBBPB, BBBBBP).
   
6 1 15 6
The probability that Blaise wins 4 games is × = , since there are = 15 ways
2 64 64 2
for 4 Bs and 2 Ps to be arranged.
1 6 15 22 11
Therefore, the probability that Blaise wins more games than Pierre is + + = = .
64 64 64 64 32
7. Solution 1
Suppose that the bag contains g gold balls. We assume that Feridun reaches into the bag and
removes the two balls one after the other.
There are 40 possible balls that he could remove first and then 39 balls that he could remove
second. In total, there are 40(39) pairs of balls that he could choose in this way.
If he removes 2 gold balls, then there are g possible balls that he could remove first and then
g − 1 balls that he could remove second. In total, there are g(g − 1) pairs of gold balls that he
could remove.
 5
We are told that the probability of removing 2 gold balls is . Since there are 40(39) total
12
pairs of balls that can be chosen and g(g − 1) pairs of gold balls that can be chosen in this way,
g(g − 1) 5 5
we have = which is equivalent to g(g − 1) = (40)(39) = 650.
40(39) 12 12
Therefore, g 2 − g − 650 = 0 or (g − 26)(g + 25) = 0, and so g = 26 or g = −25. Since g > 0,
we have g = 26, so there are 26 gold balls in the bag.
Solution 2
Suppose that the bag contains g gold balls. We assume that Feridun reaches into the bag and
removes the two balls together.
 
40
Since there are 40 balls in the bag, there are pairs of balls that he could choose in this
2
way.
 
g
Since there are g gold balls in the bag, there are pairs of gold balls that he could choose
2
in this way.
 
5 40
We are told that the probability of removing 2 gold balls is . Since there are pairs in
  12 2
g
total that can be chosen and pairs of gold balls that can be chosen in this way, we have
  2
g
     
2 5 g 5 40 n n(n − 1)
  = which is equivalent to = . Since = , this equation is
40 12 2 12 2 2 2
2
g(g − 1) 5 40(39)
equivalent to = · = 325.
2 12 2
Therefore, g(g − 1) = 650 or g 2 − g − 650 = 0 or (g − 26)(g + 25) = 0, and so g = 26 or g = −25.
Since g > 0, we have g = 26, so there are 26 gold balls in the bag.

8. The number of integers between 100 and 999 inclusive is 999 − 100 + 1 = 900.
An integer n in this range has three digits, say a, b and c, with the hundreds digit equal to a.
Note that 0 ≤ b ≤ 9 and 0 ≤ c ≤ 9 and 1 ≤ a ≤ 9.
To have a + b + c = 24, then the possible triples for a, b, c in some order are 9,9,6; 9,8,7; 8,8,8.
(There cannot be three 9’s. If there are two 9’s, the the other digit equals 6. If there is one 9,
the second and third digits add to 15 but are both less than 9, so must equal 8 and 7. If there
are zero 9’s, the maximum for each digit is 8, and so each digit must be 8 in order for the sum
of all three to equal 24.)
If the digits are 9, 9 and 6, there are 3 arrangements: 996, 969, 699.
If the digits are 9, 8 and 7, there are 6 arrangements: 987, 978, 897, 879, 798, 789.
If the digits are 8, 8 and 8, there is only 1 arrangement: 888.
Therefore, there are 3 + 6 + 1 = 10 integers n in the range 100 to 999 with the sum of the digits
of n equal to 24.
 The required probability equals the number of possible values of n with the sum of digits equal
10 1
to 24 divided by the total number of integers in the range, or = .
900 90
9. Suppose that Billy removes the ball numbered x from his bag and that Crystal removes the
ball numbered y from her bag.
Then b = 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 − x = 45 − x.
Also, c = 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 − y = 45 − y.
Hence, b − c = (45 − x) − (45 − y) = y − x.
Since 1 ≤ x ≤ 9 and 1 ≤ y ≤ 9, we have −8 ≤ y − x ≤ 8. (This is because y − x is maximized
when y is largest (that is, y = 9) and x is smallest (that is, x = 1), so y − x ≤ 9 − 1 = 8.
Similarly, y − x ≥ −8.)
Since b − c = y − x is between −8 and 8, for it to be a multiple of 4, b − c = y − x must be −8,
−4, 0, 4, or 8.
Since each of Billy and Crystal chooses 1 ball from 9 balls and each ball is equally likely to be
chosen, the probability of any specific ball being chosen from one of their bags is 91 . Thus, the
1 1 1
probability of any specific pair of balls being chosen (one from each bag) is × = .
9 9 81
Therefore, to compute the desired probability, we must count the number of pairs (x, y) where
1
y − x is −8, −4, 0, 4, 8, and multiply this result by 81 .
Method 1
y − x = −8, then (x, y) must be (9, 1).
If
y − x = 8, then (x, y) must be (1, 9).
If
y − x = −4, then (x, y) can be (5, 1), (6, 2), (7, 3), (8, 4), (9, 5).
If
y − x = 4, then (x, y) can be (1, 5), (2, 6), (3, 7), (4, 8), (5, 9).
If
y − x = 0, then (x, y) can be (1, 1), (2, 2), (3, 3), (4, 4), (5, 5), (6, 6), (7, 7), (8, 8), (9, 9).
If
21 7
There are thus 21 pairs (x, y) that work, so the desired probability is = .
81 27
Method 2
If x = 9, then for y−x to be a multiple of 4, y must be 9, 5 or 1.
If x = 8, then for y−x to be a multiple of 4, y must be 8 or 4.
If x = 7, then for y−x to be a multiple of 4, y must be 7 or 3.
If x = 6, then for y−x to be a multiple of 4, y must be 6 or 2.
If x = 5, then for y−x to be a multiple of 4, y must be 9, 5 or 1.
If x = 4, then for y−x to be a multiple of 4, y must be 8 or 4.
If x = 3, then for y−x to be a multiple of 4, y must be 7 or 3.
If x = 2, then for y−x to be a multiple of 4, y must be 6 or 2.
If x = 1, then for y−x to be a multiple of 4, y must be 9, 5 or 1.
21 7
There are thus 21 pairs (x, y) that work, so the desired probability is = .
81 27
10. If she tosses 3 heads on the first toss, then she has no coins to toss on the second toss, so she
cannot toss exactly 1 head.
 If she tosses 2, 1 or 0 heads on the first toss, then she has at least one coin to toss on the second
toss, so it is possible for her to toss exactly 1 head on the second toss.
The following possibilities exist:
• She tosses 2 heads out of 3 coins on the first toss and she tosses 1 head out of 1 coin on
the second toss
• She tosses 1 head out of 3 coins on the first toss and she tosses 1 head out of 2 coins on
the second toss
• She tosses 0 heads out of 3 coins on the first toss and she tosses 1 head out of 3 coins on
the second toss
We calculate the various probabilities.
If 3 coins are tossed, there are 8 equally likely possibilities: 
HHH,3 HHT, HTH, THH, TTH,
1 1
THT, HTT, TTT. Each of these possibilities has probability = . Therefore,
2 8
1
• the probability of tossing 0 heads out of 3 coins is 8
3
• the probability of tossing 1 head out of 3 coins is 8
3
• the probability of tossing 2 heads out of 3 coins is 8
1
• the probability of tossing 3 heads out of 3 coins is 8

If 2 coins are tossed, there are

 42equally likely possibilities: HH, HT, TH, TT. Each of these
1 1
possibilities has probability = . Therefore, the probability of tossing 1 head out of 2
2 4
2 1
coins is = .
4 2
If 1 coin is tossed, the probability of tossing 1 head is 12 .
To summarize, the possibilities are
• She tosses 2 heads out of 3 coins on the first toss (with probability 38 ) and she tosses 1
head out of 1 coin on the second toss (with probability 12 )
• She tosses 1 head out of 3 coins on the first toss (with probability 38 ) and she tosses 1 head
out of 2 coins on the second toss (with probability 12 )
• She tosses 0 heads out of 3 coins on the first toss (with probability 18 ) and she tosses 1
head out of 3 coins on the second toss (with probability 38 )
3 1 3 1 1 3 27
Therefore, the overall probability is · + · + · = .
8 2 8 2 8 8 64
11. Let P (10) be the probability the product of the numbers on the two balls chosen is divisible by
10.
Since there are 10 balls in each bag, there are 10 · 10 = 100 pairs of balls that can be chosen.
Let a be the number on the first ball chosen and b be the number on the second ball chosen.
To determine P (10), we count the number of pairs (a, b) for which ab is divisible by 10. If the
m
number of pairs is m, then P (10) = .
100
 For ab to be divisible by 10, at least one of a and b must be a multiple of 5 and at least one of
a and b must be even.
If a = 10 or b = 10, then the pair (a, b) gives a product ab divisible by 10. In this case, we
obtain the 19 pairs

(a, b) = (1, 10), (2, 10), . . . , (9, 10), (10, 10), (10, 9), . . . , (10, 2), (10, 1)

If neither a nor b equals 10, then either a = 5 or b = 5 in order for a or b to be divisible by 5.

In this case, the other of a and b must be even and not equal to 10. (We have already counted
the pairs where a = 10 or b = 10.) In this case, we obtain the 8 pairs

(a, b) = (5, 2), (5, 4), (5, 6), (5, 8), (2, 5), (4, 5), (6, 5), (8, 5)

From our work above, there are no additional pairs for which ab is divisible by 10.
Thus, there are 19 + 8 = 27 pairs (a, b) for which ab is divisible by 10, which means that
27
P (10) = . (We note that we could have made a 10 by 10 table that listed all possible
100
combinations of a and b and their product, from which we could obtain P (10).)

12. There are 210 = 1024 strings of ten letters, each of which is A or B, because there are 2 choices
for each of the 10 positions in the string.
We determine the number of these strings that do not include the “substring” ABBA (that
is, that do not include consecutive letters ABBA) by counting the number of strings that do
include the substring ABBA and subtracting this total from 1024.
If a string includes the substring ABBA, there are 7 possible positions in which this substring
could start (ABBAxxxxxx, xABBAxxxxx, . . ., xxxxxxABBA). There are 2 choices for each
of the remaining 6 letters in such a string, so there are 7 · 26 = 448 occurrences of the substring
ABBA among the 1024 strings.
This does not mean that there are 448 strings that contain the substring ABBA. Since ABBA
can appear multiple times in a single string, this total will count some strings more than once.
(For example, the string ABBAAAABBA is included in both the first and seventh of these
categories, so this string is counted twice.)
So we must “correct” this total of 448 by accounting for the strings in which ABBA occurs
more than once.
Since two substrings of ABBA can overlap in 0 letters (for example, ABBAABBAxx) or in 1
letter (for example, ABBABBAxxx), the maximum number of times that the substring ABBA
can appear is 3, and there is only one such string: ABBABBABBA.
If a string contains two copies of ABBA that overlap, then it must be of one of the following
forms:

ABBABBAxxx xABBABBAxx xxABBABBAx xxxABBABBA

Since there are 4 choices for the starting position of ABBABBA and 2 choices for each of the
three unknown letters, then there are 4 · 23 = 32 occurrences of ABBABBA among all of these
strings.
 But the string ABBABBABBA is counted in each of the first and last categories, so we subtract
2 occurrences from this total to obtain 30, the total number of strings of ten letters that included
exactly two overlapping copies of ABBA. (We’ll count the string ABBABBABBA later.)
If a string contains exactly two substrings of ABBA and these do not overlap, then it must be
of one of the following forms:

ABBAABBAxx ABBAxABBAx ABBAxxABBA

xABBAABBAx xABBAxABBA xxABBAABBA
Since there are 6 such forms and 2 choices for each of the 2 unknown letters in each case, there
appear to be 6 · 22 = 24 such strings.
This total includes the string ABBABBABBA in the third category, so we subtract 1 from
this total to obtain 23, the total number of strings of ten letters that include exactly two copies
of ABBA which do not overlap.
So there are 30 strings that contain exactly two overlapping substrings ABBA, 23 strings that
contain exactly two non-overlapping substrings ABBA, and 1 string that contains exactly three
substrings ABBA.
To get the total number of strings with one or more substrings ABBA we take the total number
of occurrences of ABBA (448), subtract the number of strings with exactly two substrings
ABBA (since these were included twice in the original count), and subtract two times the
number of strings with exactly three substrings ABBA (since these were included three times
in the original count).
Therefore, there are 448−23−30−2·1 = 393 strings that include at least one substring ABBA,
and so there are 1024−393 = 631 strings of ten letters that do not include the substring ABBA.

13. The Eden sequences from {1, 2, 3, 4, 5} are

1 3 5 1, 2 1, 4 3, 4 1, 2, 3 1, 2, 5 1, 4, 5 3, 4, 5 1, 2, 3, 4 1, 2, 3, 4, 5

There are 12 such sequences.

We present a brief justification of why these are all of the sequences.

• An Eden sequence of length 1 consists of a single odd integer. The possible choices are 1
and 3 and 5.
• An Eden sequence of length 2 consists of an odd integer followed by a larger even integer.
Since the only possible even integers here are 2 and 4, the possible sequences are 1, 2 and
1, 4 and 3, 4.
• An Eden sequence of length 3 starts with an Eden sequence of length 2 and appends (that
is, adds to the end) a larger odd integer. Starting with 1,2, we form 1,2,3 and 1,2,5.
Starting with 1,4, we form 1,4,5. Starting with 3,4, we form 3,4,5.
• An Eden sequence of length 4 starts with an Eden sequence of length 3 and appends a
larger even integer. Since 2 and 4 are the only possible even integers, the only possible
sequence here is 1,2,3,4.
• An Eden sequence of length 5 from {1, 2, 3, 4, 5} must include all 5 elements, so is 1,2,3,4,5.
14. Throughout this problem, we represent the states of the n plates as a string of 0’s and 1’s (called
a binary string) of length n of the form p1 p2 · · · pn , with the rth digit from the left (namely pr )
equal to 1 if plate r contains a gift and equal to 0 if plate r does not. We call a binary string
of length n allowable if it satisfies the requirements – that is, if no two adjacent digits both
equal 1. Note that digit pn is also “adjacent” to digit p1 , so we cannot have p1 = pn = 1.

(a) Suppose that p1 = 1. Then p2 = p7 = 0, so the string is of the form 10p3 p4 p5 p6 0.

Since k = 3, we know that 2 of p3 , p4 , p5 , p6 equal 1, but in such a way that no two
adjacent digits are both 1. The possible strings in this case are 1010100, 1010010 and
1001010.
Suppose that p1 = 0. Then p2 can equal 1 or 0.
If p2 = 1, then p3 = 0 as well. This means that the string is of the form 010p4 p5 p6 p7 ,
which is the same as the general string in the first case, but shifted by 1 position around
the circle, so there are again 3 possibilities.
If p2 = 0, then the string is of the form 00p3 p4 p5 p6 p7 and 3 of the digits p3 , p4 , p5 , p6 , p7
equal 1 in such a way that no 2 adjacent digits equal 1. There is only 1 way in which this
can happen: 0010101.
Overall, this gives 7 possible configurations, so f (7, 3) = 7.
(b) Solution 1
An allowable string p1 p2 · · · pn−1 pn has (p1 , pn ) = (1, 0), (0, 1), or (0, 0).
Define g(n, k, 1, 0) to be the number of allowable strings of length n, containing k 1’s, and
with (p1 , pn ) = (1, 0). We define g(n, k, 0, 1) and g(n, k, 0, 0) in a similar manner.
Note that f (n, k) = g(n, k, 1, 0) + g(n, k, 0, 1) + g(n, k, 0, 0).
Consider the strings counted by g(n, k, 0, 1). Since pn = 1, we know that pn−1 = 0. Since
p1 = 0, the digit p2 can equal 0 or 1.
We remove the first and last digits of these strings. We obtain strings p2 p3 · · · pn−2 pn−1
that is strings of length n − 2 containing k − 1 1’s.
Since pn−1 = 0, the first and last digits of these strings are not both 1. Also, since the
original strings did not contain two consecutive 1’s, these new strings does not either.
Therefore, p2 p3 · · · pn−2 pn−1 are allowable strings of length n − 2 containing k − 1 1’s, with
pn−1 = 0 and p2 = 1 or p2 = 0.
The number of such strings with p2 = 1 and pn−1 = 0 is g(n−2, k −1, 1, 0) and the number
of such strings with p2 = 0 and pn−1 = 0 is g(n − 2, k − 1, 0, 0). Thus, g(n, k, 0, 1) =
g(n − 2, k − 1, 1, 0) + g(n − 2, k − 1, 0, 0).
Consider the strings counted by g(n, k, 0, 0). Since p1 = 0 and pn = 0, we can remove pn
to obtain strings p1 p2 · · · pn−1 of length n − 1 containing k 1’s. These strings are allowable
since p1 = 0 and the original strings were allowable. Note that we have p1 = 0 and pn−1
is either 0 or 1.
So the strings p1 p2 · · · pn−1 are allowable strings of length n − 1 containing k 1’s, starting
with 0, and ending with 0 or 1.
The number of such strings with p1 = 0 and pn−1 = 0 is g(n − 1, k, 0, 0) and the number
of such strings with p1 = 0 and pn−1 = 1 is g(n − 1, k, 0, 1). Thus, g(n, k, 0, 0) = g(n −
1, k, 0, 0) + g(n − 1, k, 0, 1).
Consider the strings counted by g(n, k, 1, 0). Here, p1 = 1 and pn = 0. Thus, pn−1 can
equal 0 or 1. We consider these two sets separately.
If pn−1 = 0, then the string p1 p2 · · · pn−1 is an allowable string of length n − 1, containing
k 1’s, beginning with 1 and ending with 0. Therefore, the number of strings counted by
g(n, k, 1, 0) with pn−1 = 0 is equal to g(n − 1, k, 1, 0).
If pn−1 = 1, then the string p2 p3 · · · pn−1 is of length n − 2, begins with 0 and ends with 1.
Also, it contains k −1 1’s (having removed the original leading 1) and is allowable since the
original string was. Therefore, the number of strings counted by g(n, k, 1, 0) with pn−1 = 1
is equal to g(n − 2, k − 1, 0, 1).
Therefore,
f (n, k) = g(n, k, 1, 0) + g(n, k, 0, 1) + g(n, k, 0, 0)
= (g(n − 1, k, 1, 0) + g(n − 2, k − 1, 0, 1))
+(g(n − 2, k − 1, 1, 0) + g(n − 2, k − 1, 0, 0))
+(g(n − 1, k, 0, 0) + g(n − 1, k, 0, 1))
= (g(n − 1, k, 1, 0) + g(n − 1, k, 0, 1) + g(n − 1, k, 0, 0))
+(g(n − 2, k − 1, 0, 1) + g(n − 2, k − 1, 1, 0) + g(n − 2, k − 1, 0, 0))
= f (n − 1, k) + f (n − 2, k − 1)
as required.
Solution 2
We develop an explicit formula for f (n, k) by building these strings.
Consider the allowable strings of length n that include k 1’s. Either pn = 0 or pn = 1.
Consider first the case when pn = 0. Here, p1 can equal 0 or 1. These strings are all of the
form p1 p2 p3 · · · pn−1 0. In this case, since a 1 is always followed by a 0 and the strings end
with 0, we can build these strings using blocks of the form 10 and 0. Any combination
of these blocks will be an allowable string, as each 1 will always be both preceded and
followed by a 0. Thus, these strings can all be built using k 10 blocks and n − 2k 0 blocks.
This gives k 1’s and k + (n − 2k) = n − k 0’s. Note that any string built with these blocks
will be allowable and will end with a 0, and any such allowable string can be built in this
way. The number of ways  of arranging k blocks  kind and n − 2k blocks of another
 of one
k + (n − 2k) n−k
kind is , which simplifies to .
k k
Consider next the case when pn = 1. Here, we must have pn−1 = p1 = 0, since these are
the two digits adjacent to pn . Thus, these strings are all of the form 0p2 p3 · · · 01.
Consider the strings formed by removing the first and last digits. These strings are al-
lowable, are of length n − 2, include k − 1 1’s, end with 0, and can begin with 0 or 1.
Again, since a 1 is always followed by a 0 and the strings end with 0, we can build these
strings using blocks of the form 10 and 0. Any combination of these blocks will be an
allowable string, as each 1 will always be both preceded  and followed by a 0.Translating 
(n − 2) − (k − 1) n−k−1
our method of counting from the first case, there are or
k−1 k−1
such strings.    
n−k n−k−1
Thus, f (n, k) = + such strings.
k k−1
      
m m−1 m−1
To prove the desired fact, we will use the fact that = + , which
r r r−1
we prove at the end. Now

f (n − 1, k) + f (n − 2, k − 1)
       
(n − 1) − k (n − 1) − k − 1 (n − 2) − (k − 1) (n − 2) − (k − 1) − 1
= + + +
k k−1 k−1 (k − 1) − 1
       
n−k−1 n−k−2 n−k−1 n−k−2
= + + +
k k−1 k−1 k−2
       
n−k−1 n−k−1 n−k−2 n−k−2
= + + +
k k−1 k−1 k−2
   
n−k n−k−1
= + (using the identity above)
k k−1
= f (n, k)

as required.
To prove the identity, we expand the terms on the right side:
   
m−1 m−1 (m − 1)! (m − 1)!
+ = +
r r−1 r!(m − r − 1)! (r − 1)!(m − r)!
(m − 1)!(m − r) r(m − 1)!
= +
r!(m − r − 1)!(m − r) r(r − 1)!(m − r)!
(m − 1)!(m − r) r(m − 1)!
= +
r!(m − r)! r!(m − r)!
(m − 1)!(m − r + r)
=
r!(m − r)!
(m − 1)!m
=
r!(m − r)!
m!
=
r!(m − r)!
 
m
=
r

as required.
 Properties of Numbers
Solutions
1. Suppose that a palindrome p is the sum of the three consecutive integers a − 1, a, a + 1.
In this case, p = (a − 1) + a + (a + 1) = 3a, so p is a multiple of 3.
The largest palindromes less than 200 are 191, 181, 171.
Note that 191 and 181 are not divisible by 3, but 171 is divisible by 3. We can easily check this
using the divisibility by 3 test. For each of these integers, the sum of their digits is 11, 10 and
9, respectively. Only 9 is divisible by 3 and so 171 is the only one that is divisible by 3.
The integer 171 can be written as 56 + 57 + 58, so 171 is the largest palindrome less than 200
that is the sum of three consecutive integers.
A+B
2. Suppose that n has digits AB. Then n = 10A + B. The average of the digits of n is .
2
Putting a decimal point between the digits of n is equivalent to dividing n by 10, so the resulting
10A + B
number is .
10
So we want to determine A and B so that
10A + B A+B
=
10 2
10A + B = 5(A + B)
5A = 4B

Since A and B are digits such that 5A = 4B, we have A = 4 and B = 5 is the only possibility.
Therefore, n = 45. (We can quickly check that the average of the digits of n is 4.5, the number
obtained by putting a decimal point between the digits of n.)

3. Solution 1
The integer equal to 1020 consists of the digit 1 followed by 20 0s. The integer equal to 1020 − 1
thus consists of 20 9s.
Now, n = 1020 − 20 is 19 less than 1020 − 1 which is the integer that consists of 20 9s.
So n = 1020 − 20 = 99 · · · 980, where this integer has 18 9s.
Therefore, the sum of the digits of n is 18(9) + 8 + 0 = 162 + 8 = 170.
Solution 2
Since 1020 − 20 = 10(1019 − 2) and 1019 − 2 = 99 · · · 98 (where this integer has 18 9s), we have
1020 − 20 = 99 · · · 980, where this integer has 18 9s.
Therefore, the sum of the digits of n is 18(9) + 8 + 0 = 162 + 8 = 170.

4. The parity of an integer is whether it is even or odd.

Since the Fibonacci sequence begins 1, 1, 2, 3, 5, 8, 13, 21, . . ., it follows that the parities of the
first eight terms are Odd, Odd, Even, Odd, Odd, Even, Odd, Odd.
In the sequence, if x and y are consecutive terms, then the next term is x + y.
 In general, suppose that x and y are integers. If x is even and y is even, then x + y is even. If
x is even and y is odd, then x + y is odd. If x is odd and y is even, then x + y is odd. If x is
odd and y is odd, then x + y is even. Therefore, the parities of two consecutive terms x and y
in the Fibonacci sequence determine the parity of the following term x + y.
Also, once there are two consecutive terms whose parities match the parities of two earlier
consecutive terms in the sequence, then the parities will repeat in a cycle. In particular, the
parities of the fourth and fifth terms (Odd, Odd) are the same as the parities of the first and
second terms (Odd, Odd). Therefore, the parities in the sequence repeat the cycle Odd, Odd,
Even. This cycle has length 3. Therefore, the 99th term in the Fibonacci sequence ends one of
these cycles, since 99 is a multiple of 3. In particular, the 99th term ends the 33rd cycle.
Each cycle contains two odd terms. Therefore, the first 99 terms in the sequence include
2 × 33 = 66 odd terms.
Finally, the 100th term in the sequence begins a new cycle, so it is odd. Therefore, the first
100 terms include 66 + 1 = 67 odd terms.

5. Since 900 = 302 and 30 = 2 × 3 × 5, we have 900 = 22 32 52 .

The positive divisors of 900 are those integers of the form d = 2a 3b 5c , where each of a, b, c is 0,
1 or 2.
For d to be a perfect square, the exponent on each prime factor in the prime factorization of d
must be even. Thus, for d to be a perfect square, each of a, b, c must be 0 or 2.
There are two possibilities for each of a, b, c so 2 × 2 × 2 = 8 possibilities for d.
These are 20 30 50 = 1, 22 30 50 = 4, 20 32 50 = 9, 20 30 52 = 25, 22 32 50 = 36, 22 30 52 = 100,
20 32 52 = 225, and 22 32 52 = 900.
Thus, 8 of the positive divisors of 900 are perfect squares.

6. Since each list contains 6 consecutive positive integers and the smallest integers in the lists are
a and b, it follows that the positive integers in the first list are a, a+1, a+2, a+3, a+4, a+5 and
the positive integers in the second list are b, b + 1, b + 2, b + 3, b + 4, b + 5. Note that 1 ≤ a < b.
We first determine the pairs (a, b) for which 49 will appear in the third list, then determine
which of these pairs give a third list that contains no multiple of 64, and then finally keep only
those pairs for which there is a number in the third list larger than 75.
The first bullet tells us that 49 is the product of an integer in the first list and an integer in
the second list.
Since 49 = 72 and 7 is prime, it follows that these integers are either 1 and 49 or 7 and 7.
If 1 is in one of the lists, then either a = 1 or b = 1. Since 1 ≤ a < b, it must be that a = 1.
If 49 is in the second list, then one of b, b+1, b+2, b+3, b+4, b+5 equals 49, and so 44 ≤ b ≤ 49.
Therefore, for 1 and 49 to appear in the two lists, then (a, b) must be one of

(1, 49), (1, 48), (1, 47), (1, 46), (1, 45), (1, 44) .

If 7 appears in the first list, then one of a, a + 1, a + 2, a + 3, a + 4, a + 5 equals 7, so 2 ≤ a ≤ 7.

Similarly, if 7 appears in the second list, then 2 ≤ b ≤ 7.
 Therefore, for 7 to appear in both lists, then, knowing that a < b, then (a, b) must be one of
(2, 3), (2, 4), (2, 5), (2, 6), (2, 7), (3, 4), (3, 5), (3, 6), (3, 7), (4, 5), (4, 6), (4, 7), (5, 6), (5, 7), (6, 7) .
The second bullet tells us that no pair of numbers in the first and second lists have a product
that is a multiple of 64.
Given that the possible values of a and b are 1, 2, 3, 4, 5, 6, 7, 44, 45, 46, 47, 48, 49, then the pos-
sible integers in the two lists are those integers from 1 to 12, inclusive, and from 44 to 54,
inclusive. (For example, if the first number in one list is 7, then the remaining numbers in this
list are 8, 9, 10, 11, 12.)
There is no multiple of 32 or 64 in these lists.
Thus, for a pair of integers from these lists to have a product that is a multiple of 64, one is a
multiple of 4 and the other is a multiple of 16, or both are multiples of 8.
If (a, b) = (1, 48), (1, 47), (1, 46), (1, 45), (1, 44), then 4 appears in the first list and 48 appears
in the second list; these have a product of 192, which is 3 · 64.
If (a, b) = (1, 49), there is a multiple of 4 but not of 8 in the first list, and a multiple of 4 but
not of 8 in the second list, so there is no multiple of 64 in the third list.
If (a, b) = (3, 4), (3, 5), (3, 6), (3, 7), (4, 5), (4, 6), (4, 7), (5, 6), (5, 7), (6, 7), then 8 appears in both
lists, so 64 appears in the third list.
If (a, b) = (2, 3), (2, 4), (2, 5), (2, 6), (2, 7), then there is no multiple of 8 or 16 in the first list
and no multiple of 16 in the second list, so there is no multiple of 64 in the third list.
Therefore, after considering the first two bullets, the possible pairs (a, b) are (1, 49), (2, 3),
(2, 4), (2, 5), (2, 6), (2, 7).
The third bullet tells us that there is at least one number in the third list that is larger than
75.
Given the possible pairs (a, b) are (1, 49), (2, 3), (2, 4), (2, 5), (2, 6), (2, 7), the corresponding pairs
of largest integers in the lists are (6, 54), (7, 8), (7, 9), (7, 10), (7, 11), (7, 12).
The corresponding largest integers in the third list are the products of the largest integers in
the two lists; these products are 324, 56, 63, 70, 77, 84, respectively.
Therefore, the remaining pairs (a, b) are (1, 49), (2, 6), (2, 7).
Having considered the three conditions, the possible pairs (a, b) are (1, 49), (2, 6), (2, 7).
7. First, √
we determine the perfect squares between 1300 and 1400 and between 1400 and 1500.
Since 1300 ≈ 36.06, the first perfect square larger than 1300 is 372 = 1369. The next perfect
squares are 382 = 1444 and 392 = 1521.
Since Charles was born between 1300 and 1400 in a year that was a perfect square, Charles
must have been born in 1369.
Since Louis was born between 1400 and 1500 in a year that was a perfect square, Louis must
have been born in 1444.
Suppose that on April 7 in some year, Charles was m2 years old and Louis was n2 years old
for some positive integers m and n. Thus, Charles was m2 years old in the year 1369 + m2 and
Louis was n2 years old in the year 1444 + n2 .
 Since these expressions represent the same years, we have that 1369 + m2 = 1444 + n2 , or
m2 − n2 = 1444 − 1369 = 75. In other words, we want to find two perfect squares less than 110
(since their ages are less than 110) whose difference is 75.
The perfect squares less than 110 are 1, 4, 9, 16, 25, 36, 49, 64, 81, 100. The two that differ by 75
are 100 and 25. Thus, m2 = 100 and n2 = 25.
This means that the year in which the age of each of Charles and Louis was a perfect square
was the year 1369 + 100 = 1469.
1 x
8. Since 10y 6= 0, the equation = y is equivalent to 10y = 32x.
32 10
So the given question is equivalent to asking for the smallest positive integer x for which 32x
equals a positive integer power of 10.
Now 32 = 25 and so 32x = 25 x. For 32x to equal a power of 10, each factor of 2 must be
matched with a factor of 5. Therefore, x must be divisible by 55 (that is, x must include at
least 5 powers of 5), and so x ≥ 55 = 3125.
But 32(55 ) = 25 55 = 105 , and so if x = 55 = 3125, then 32x is indeed a power of 10, namely
1 x
105 . This tells us that the smallest positive integer x for which = y for some positive
32 10
integer y is x = 55 = 3125.

9. Since the average of three consecutive multiples of 3 is a, a is the middle of these three integers,
and so the integers are a − 3, a, a + 3.
Since the average of four consecutive multiples of 4 is a + 27, we have that a + 27 is halfway in
between the second and third of these multiples (which differ by 4), so the second and third of
the multiples are (a + 27) − 2 = a + 25 and (a + 27) + 2 = a + 29. Therefore, the four integers
are a + 21, a + 25, a + 29, a + 33.
The smallest of these seven integers is a − 3 and the largest is a + 33.
1 1
The average of these two integers is (a − 3 + a + 33) = (2a + 30) = a + 15.
2 2
Since a + 15 = 42, we have a = 27.

10. First, we factor the left side of the given equation to obtain a(a2 + 2b) = 2013.
Next, we factor the integer 2013 as 2013 = 3 × 671 = 3 × 11 × 61. Note that each of 3, 11 and
61 is prime, so we can factor 2013 no further. (We can find the factors of 3 and 11 using tests
for divisibility by 3 and 11, or by systematic trial and error.)
Since 2013 = 3 × 11 × 61, the positive divisors of 2013 are

1, 3, 11, 33, 61, 183, 671, 2013

Since a and b are positive integers, a and a2 + 2b are both positive integers.
Since a and b are positive integers, we have a2 ≥ a and 2b > 0, so a2 + 2b > a.
Since a(a2 + 2b) = 2013, a and a2 + 2b must be a divisor pair of 2013 (that is, a pair of positive
integers whose product is 2013) with a < a2 + 2b.
 We make a table of the possibilities:

a a2 + 2b 2b b
1 2013 2012 1006
3 671 662 331
11 183 62 31
33 61 −1028 N/A

Note that the last case is not possible, since b must be positive. Therefore, the three pairs of
positive integers that satisfy the equation are (1, 1006), (3, 331), (11, 31). (We can verify by
substitution that each is a solution of the original equation.)

11. Suppose that the auditorium with these properties has r rows and c columns of chairs. Then
there are rc chairs in total.
Each chair is empty, is occupied by a teacher, or is occupied by a student.
Since there are 14 teachers in each row, there are 14r chairs occupied by teachers. Since
there are 10 students in each column, there are 10c chairs occupied by students. Since there
are exactly 3 empty chairs, the total number of chairs can also be written as 14r + 10c + 3.
Therefore, rc = 14r + 10c + 3.
We proceed to find all pairs of positive integers r and c that satisfy this equation. We note that
since there are 14 teachers in each row, there must be at least 14 columns (that is, c ≥ 14) and
since there are 10 students in each column, there must be at least 10 rows (that is, r ≥ 10).
Manipulating the equation,

rc = 14r + 10c + 3
rc − 14r = 10c + 3
r(c − 14) = 10c + 3
10c + 3
r =
c − 14
10c − 140 + 143
r =
c − 14
10c − 140 143
r = +
c − 14 c − 14
143
r = 10 +
c − 14
143 143
Since r is an integer, 10 + is an integer, and so must be an integer.
c − 14 c − 14
Therefore, c − 14 is a divisor of 143. Since c ≥ 14, we have c − 14 ≥ 0, and so c − 14 is a positive
divisor of 143.
Since 143 = 11 × 13, its positive divisors are 1, 11, 13, 143.
We make a table of the possible values of c−14 along with the resulting values of c, r (calculated
 
143
using r = 10 + and rc:
c − 14

c − 14 c r rc
1 15 153 2295
11 25 23 575
13 27 21 567
143 157 11 1727

Therefore, the four possible values for rc are 567, 575, 1727, 2295. That is, the smallest possible
number of chairs in the auditorium is 567. (Can you create a grid with 27 columns and 21 rows
that has the required properties?)