MAT 1348/1748

SOLUTIONS TO SUPPLEMENTAL EXERCISES

by J. Khoury, L. Dumitrescu, and M. Šajna

1 Propositional Logic
p q P1 P2 P3 P4 P5 P6
T T T T T F F F
1. T F T F F T F F
F T T F T F T F
F F F T T T F T

(a) From the table, the corresponding DNFs are

P1 ≡ (p ∧ q) ∨ (p ∧ ¬q) ∨ (¬p ∧ q)
P2 ≡ (p ∧ q) ∨ (¬p ∧ ¬q)
P3 ≡ (p ∧ q) ∨ (¬p ∧ q) ∨ (¬p ∧ ¬q)
P4 ≡ (p ∧ ¬q) ∨ (¬p ∧ ¬q)
P5 ≡ ¬p ∧ q
P6 ≡ ¬p ∧ ¬q

(b) Analyzing each compound proposition and then using equivalence laws we have

P1 ≡ p ∨ q ≡ ¬(¬p ∧ ¬q)
P2 ≡ p ↔ q ≡ (p → q) ∧ (q → p) ≡ (¬p ∨ q) ∧ (¬q ∨ p) ≡ ¬(p ∧ ¬q) ∧ ¬(q ∧ ¬p)
P3 ≡ p → q ≡ ¬p ∨ q ≡ ¬(p ∧ ¬q)
P4 ≡ ¬q
P5 ≡ ¬p ∧ q
P6 ≡ ¬p ∧ ¬q.

(c) Using the form of each compound proposition and then equivalence laws,

P1 ≡ p ∨ q ≡ ¬¬p ∨ q ≡ ¬p → q
P2 ≡ p ↔ q ≡ (p → q) ∧ (q → p) ≡ ¬¬(p → q) ∧ (q → p)
≡ ¬(¬(p → q) ∨ ¬(q → p)) ≡ ¬((p → q) → ¬(q → p))
P3 ≡ p → q
P4 ≡ ¬q
P5 ≡ ¬p ∧ q ≡ ¬(p ∨ ¬q) ≡ ¬(q → p)
P6 ≡ ¬(¬p → q).

1
 p q p⊕q
T T F
2. (a) T F T
F T T
F F F

p q r p⊕q (p ⊕ q) ⊕ r q⊕r p ⊕ (q ⊕ r)
T T T F T F T
T T F F F T F
T F T T F T F
(b) The truth table T F F T T F T
F T T T F F F
F T F T T T T
F F T F T T T
F F F F F F F
Since the fifth and the seventh columns are the same, we conclude that the cor-
responding propositions, (p ⊕ q) ⊕ r and p ⊕ (q ⊕ r), are equivalent.
p q p⊕q p↔q ¬(p ↔ q)
T T F T F
(c) T F T F T
F T T F T
F F F T F
Since the third and the fifth columns are the same, we conclude that p ⊕ q ≡
¬(p ↔ q).
3. Using equivalence laws,
¬(a → b) → c ≡ ¬(¬(a → b)) ∨ c
≡ (a → b) ∨ c
≡ (¬a ∨ b) ∨ c.

4. (a) Truth tables.

(i) Denote P1 = (x ∨ y) ∧ (¬x ∨ z) and note that the compound proposition
((x ∨ y) ∧ (¬x ∨ z) ∧ (y → z)) → z can be written as (P1 ∧ (y → z)) → z.
x y z x∨y ¬x ∨ z P1 y→z P1 ∧ (y → z) (P1 ∧ (y → z)) → z
T T T T T T T T T
T T F T F F F F T
T F T T T T T T T
T F F T F F F F T
F T T T T T T T T
F T F T T T F F T
F F T F T F T F T
F F F F T F T F T
2
 Hence, ((x ∨ y) ∧ (¬x ∨ z) ∧ (y → z)) → z is a tautology.
(ii) Denote P2 = (x → y) ∧ (x → z) and note that the proposition (x → (y ∨ z)) →
((x → y) ∧ (x → z)) can be written as (x → (y ∨ z)) → P2 .
x y z y∨z x → (y ∨ z) x→y x→z P2 (x → (y ∨ z)) → P2
T T T T T T T T T
T T F T T T F F F
T F T T T F T F F
T F F F F F F F T
F T T T T T T T T
F T F T T T T T T
F F T T T T T T T
F F F F T T T T T
Hence, (x → (y ∨ z)) → ((x → y) ∧ (x → z)) is not a tautology. If either (x
is T, y is T and z is F) or (x is T, y is F and z is T), then the proposition
(x → (y ∨ z)) → ((x → y) ∧ (x → z)) is F.

(b) Truth trees.

(i) Consider the negation of the given compound proposition,

¬(((x ∨ y) ∧ (¬x ∨ z) ∧ (y → z)) → z) ≡ ¬(¬((x ∨ y) ∧ (¬x ∨ z) ∧ (y → z)) ∨ z),

and apply the truth trees method.

¬(¬((x ∨ y) ∧ (¬x ∨ z) ∧ (y → z)) ∨ z)X

(x ∨ y) ∧ (¬x ∨ z) ∧ (y → z)X
¬z

x ∨ yX
¬x ∨ zX
y → zX

x y

¬x z ¬x z
× × ×
¬y z
× ×
Since all the paths are inactive, we conclude that the negation proposition considered is
a contradiction. This implies that the initial given compound proposition is a tautology.

(ii) Consider the negation of the given compound proposition is

¬((x → (y ∨ z)) → ((x → y) ∧ (x → z))),

and apply the method of truth trees.

3
 ¬((x → (y ∨ z)) → ((x → y) ∧ (x → z)))X

x → (y ∨ z)X
¬((x → y) ∧ (x → z))X

¬x y ∨ zX

¬(x → y)X ¬(x → z)X ¬(x → y)X ¬(x → z)X

x x x x
¬y ¬z ¬y ¬z
× × y z y z
× ×
Since there are complete active paths, we conclude that the negation proposition is
not a contradiction and therefore, the initial proposition is not a tautology. The active
paths provide the two counterexamples, i.e. truth values of x, y and z for which the
x y z
proposition (x → (y ∨ z)) → ((x → y) ∧ (x → z)) is false: T T F .
T F T

5. The argument can be written as

Hypothesis 1: (P → J) → (¬C → M )
Hypothesis 2: ¬J → ¬P
Hypothesis 3: ¬J ∧ E → ¬C
Hypothesis 4: ¬M → P

Conclusion: ¬(J ∧ ¬P ) → C

Using truth tables or truth trees the above argument can be shown to be invalid. The
P M J E C
following is a counterexample. If , then each hypothesis is true
F T F T F
whereas the conclusion is false.
6. The argument can be written as
Hypothesis 1: B → (D → S)
Hypothesis 2: ¬D → P
Hypothesis 3: (D ∨ S) → B
Hypothesis 4: P → (¬D ∧ ¬S)

Conclusion: B ∧ D
4
 Using truth tables or truth trees the above argument can be shown to be invalid.
B D P S
The following is a counterexample. If , then each hypothesis is true
F F T F
whereas the conclusion is false.

7. For each compound proposition, we construct its truth table and also find its DNF by
algebraic manipulations. For the truth trees method - see other file.

A B A→B B → A (A → B) → (B → A)
T T F T T
(i) T F F T T
F T T F F
F F T T T
Hence, a DNF for (A → B) → (B → A) is (A ∧ B) ∨ (A ∧ ¬B) ∨ (¬A ∧ ¬B).
Algebraically,

(A → B) → (B → A) ≡ ¬(A → B) ∨ (B → A)
≡ ¬(¬A ∨ B) ∨ (¬B ∨ A)
≡ (A ∧ ¬B) ∨ ¬B ∨ A.

Apply the truth trees method.

(A → B) → (B → A)X

¬(A → B)X B → AX

A ¬B A
¬B
Therefore, a DNF of the compound proposition (A → B) → (B → A) is (A ∧
¬B) ∨ ¬B ∨ A.
P Q R P →Q R∨Q P → (R ∨ Q) (P → Q) ↔ (P → (R ∨ Q))
T T T T T T T
T T F T T T T
T F T F T T F
(ii) T F F F F F T
F T T T T T T
F T F T T T T
F F T T T T T
F F F T F T T
Then, a DNF for (P → Q) ↔ (P → R ∨ Q) is (P ∧ Q ∧ R) ∨ (P ∧ Q ∧ ¬R) ∨ (P ∧
¬Q ∧ ¬R) ∨ (¬P ∧ Q ∧ R) ∨ (¬P ∧ Q ∧ ¬R) ∨ (¬P ∧ ¬Q ∧ R) ∨ (¬P ∧ ¬Q ∧ ¬R).

5
 Algebraically,

(P → Q) ↔ (P → R ∨ Q)
≡ ((P → Q) ∧ (P → R ∨ Q)) ∨ (¬(P → Q) ∧ ¬(P → R ∨ Q))
≡ ((¬P ∨ Q) ∧ (¬P ∨ (R ∨ Q))) ∨ (¬(¬P ∨ Q) ∧ ¬(¬P ∨ (R ∨ Q)))
≡ ((¬P ∨ Q) ∧ (¬P ∨ R ∨ Q)) ∨ ((P ∧ ¬Q) ∧ (P ∧ ¬R ∧ ¬Q))
≡ ((¬P ∨ Q) ∧ ((¬P ∨ Q) ∨ R)) ∨ ((P ∧ ¬Q) ∧ ((P ∧ ¬Q) ∧ ¬R))
≡ ((¬P ∨ Q) ∧ ((¬P ∨ Q) ∨ R)) ∨ ((P ∧ ¬Q) ∧ ((P ∧ ¬Q) ∧ ¬R))
≡ (¬P ∨ Q) ∨ ((P ∧ ¬Q) ∧ ¬R)
≡ ¬P ∨ Q ∨ (P ∧ ¬Q ∧ ¬R).

Apply the truth tree method.

(P → Q) ↔ (P → R ∨ Q)X

P → QX ¬(P → Q)X
P → R ∨ QX ¬(P → R ∨ Q)X

P
¬P Q ¬Q
¬P R ∨ QX ¬P R ∨ QX P
¬(R ∨ Q)X
R Q R Q
¬R
¬Q
It follows that a DNF of the compound proposition (P → Q) ↔ (P → R ∨ Q) is
¬P ∨ (¬P ∧ R) ∨ (¬P ∧ Q) ∨ (Q ∧ R) ∨ Q ∨ (P ∧ ¬R ∧ ¬Q).
(iii) Let P denote the proposition ¬A → (B → (A → (B ∧ C))).
A B C B∧C A → (B ∧ C) B → (A → (B ∧ C)) P
T T T T T T T
T T F F F F T
T F T F F T T
T F F F F T T
F T T T T T T
F T F F T T T
F F T F T T T
F F F F T T T
Since P is a tautology, a DNF of it is T.

6
 Algebraically,
¬A → (B → (A → (B ∧ C))) ≡ ¬¬A ∨ (B → (A → (B ∧ C)))
≡ A ∨ (¬B ∨ (A → (B ∧ C)))
≡ A ∨ (¬B ∨ (¬A ∨ (B ∧ C)))
≡ A ∨ ¬B ∨ ¬A ∨ (B ∧ C)
≡ T.

Apply the truth tree method.

¬A → (B → (A → (B ∧ C)))X

A B → (A → (B ∧ C))X

¬B A → (B ∧ C)X

¬A B ∧ CX

B
C
It follows that a DNF of ¬A → (B → (A → (B∧C))) is A∨¬B∨¬A∨(B∧C) ≡ T.
8. From the truth table, the compound proposition F is T only in the following cases.
x y z F
T T T T
T F F T
F F T T
Hence, a DNF for F is (x ∧ y ∧ z) ∨ (x ∧ ¬y ∧ ¬z) ∨ (¬x ∧ ¬y ∧ z).

2 Knights and Knaves

Unless otherwise stated, the following propositional variables are used:
a: “Person A is a knight.”
b: “Person B is a knight.”
c: “Person C is a knight.”
1. A says ¬a ∧ ¬b ∧ ¬c, and B says (¬a ∧ ¬b ∧ c) ∨ (a ∧ ¬b ∧ c) ∨ (a ∧ b ∧ ¬c). For a
and A’s statement, as well as b and B’s statement to be logically equivalent, the only
possibility is that A is a knave and C is a knight, while B can be either.
2. See the handout A Method of Solving Knights-And-Knaves Questions.
3. A says ¬b, and B says a ↔ c. For a and A’s statement, as well as b and B’s statement
to be logically equivalent, the only possibility is that C is a knave, and A and B are of
opposite types.
7
 4. A says b ↔ c, and C says either a ↔ b or ¬(a ↔ b). A truth table shows that a and
A’s statement, as well as c and C statement, can be simultaneously logically equivalent
only in the first case, that is, if C says a ↔ b. In other words, C answers yes.

5. See the handout A Method of Solving Knights-And-Knaves Questions.

6. Let h be the proposition “A eats his hat.” Then A says a → h. For a and A’s statement
to have the same truth value, we must have a and h both true, that is, A eats his hat.

7. A says a → T. For a and A’s statement to have the same truth value, a must be true.

8. A says a → F. For a and A’s statement can not have the same truth value; A can not
be an Islander.

9. See the handout A Method of Solving Knights-And-Knaves Questions.

10. Let x be the proposition “X is innocent,” and y be the proposition “Y is innocent.”

Then A says ¬x → ¬y and B says x ⊕ ¬y. Assuming A and B are of distinct types,
proposition a and A’s statement, as well as proposition b and B’s statement, will be
simultaneously equivalent precisely when a and x are true, and b and y are false. Hence
A and B need not be of the same type.

11. A says b and B says a → c. This is possible only when A, B, C are all knights.

12. Let b be the proposition “I love Betty,” and j be the proposition “I love Jane.” It is
given that b ∨ j and b → j are both true propositions. From a truth table we see that
j must be true, while b can be either.

13. The speaker makes the statement (b → j) → b. Assuming this is true, b must be true,
while j can be either.

14. Let l be the proposition “I love Linda,” and k be the proposition “I love Kathy.” The
speaker (say A) says l as well as l → k. For a and his two statements to all be logically
equivalent, the propositions a, l, k must all be true. That is, the speaker is a knight.

15. Let g be the proposition “There is gold on the island.” Then A says a ↔ g. This is
equivalent to a if and only if g is true and a is either. Hence there is gold on the island.

16. Now A says ¬(a ↔ g). This is equivalent to a if and only if g is false and a is either.
Hence there is no gold on the island.

17. Let m be the proposition “This is the island of Maya.” Then A said b ∧ m and B said
¬a ∧ m. Since a must be logically equivalent to A’s statement, and at the same time,
b logically equivalent to B’s statement, we find that a, b, m must all be false. Thus it
is not Maya.

18. Now A and B both say ¬a ∧ ¬b ∧ m. As above, we find that a, b, m must all be false.
Thus it is not Maya.

8
19. Now A and B both say ¬(a ∧ b) ∧ m. As above, we find that a, b, m must all be false.
Thus it is not Maya.

20. Omitted (question unclear).

3 Proofs
1. The statement is equivalent with the following conditional: ”If r is a rational number
and x is an irrational number, then r + x is an irrational number.”
Let
p : ”r is a rational number and x is an irrational number”
q : ”r + x is an irrational number”.
We need to prove p → q. To construct a proof by contradiction we assume that the
negation, ¬(p → q) is true. Note that

¬(p → q) ≡ ¬(¬p ∨ q) ≡ p ∧ ¬q.

Hence, we assume p is true and ¬q is true. Since p is true, r and r + x are rational
numbers. However, the difference between two rational numbers is always rational.
Hence, (r + x) − r is a rational number (or x is a rational number). This statement
gives a contradiction because p is true (x is irrational).
Therefore, ¬(p → q) is false and hence p → q is true.
√
2. Define the propositional variable p : ” 3 3 is irrational”. To show p is true we construct
a proof by contradiction.
Assume ¬p is true. Then, there exist integers m and n (that have only 1 as a common
√ m
divisor), with n ̸= 0 such that 3 = . This implies that m3 = 3n3 from where it
3

n
follows that 3 is a divisor of m3 . Hence, 3 is a divisor of m: there exists an integer k
such that m = 3k.
We have 27k 3 = 3n3 , or 9k 3 = n3 . Then, 3 is a divisor of n3 and so 3 is a divisor of n.
It follows that 3 is a divisor of m and n which gives a contradiction with the fact that
m and n have only 1 as a common divisor.
Therefore, p is true.

3. Note that

• max(x, y) = x, when x ≥ y and max(x, y) = y, when y > x

• min(x, y) = x, when x ≤ y and max(x, y) = y, when y < x.

Hence, we need to consider two cases x < y and x ≥ y. Define the propositional
variables

9
 p1 : ”x < y”
p2 : ”x ≥ y”
q : max(x, y) + min(x, y) = x + y.
We need to prove that q is true in each of the two cases (note that the two cases cover
all possible situations). Hence, we prove: p1 → q and p2 → q.
Case (i). Assume p1 is true. Then, max(x, y) = y and min(x, y) = x. This implies
that q is true.
Case (ii). Assume p2 is true. Then, max(x, y) = x and min(x, y) = y. This implies
that q is true.
4. We construct a proof by cases.
Define the propositional variables
p1 : ”aa is rational”
p2 : ”aa is irrational”.
Case (i). Assume p1 is true, i.e. aa is rational. Then the statement: ”at least one of
the numbers aa and (aa )a is rational” is true.
√ √2 √ √ 2
Case (ii). Assume that p2 is true, i.e. aa is irrational. Note that ( 2 ) 2 = 2 = 2
is rational. Hence, the statement: ”at least one of the numbers aa and (aa )a is rational”
is true.
5. We construct a proof by contradiction. Assume that the statement is not true, i.e.
each number a1 , a2 , ..., an is smaller than the average. Denote this average by
a1 + . . . an
m= . Hence, we assume that a1 < m and a2 < m, . . . , and an < m.
n
a1 + . . . an
It follows that a1 + . . . an < m + . . . + m = n · m. This implies that < m,
n
which means m < m. This is a contradiction.
Therefore, at least one number must be grater or equal than the average.
6. Let m and n (with m < n) be two distinct rational numbers. To construct a proof by
contradiction, we consider the negation of the given proposition and assume it is true.
This means we assume that between m and n there is a finite number (p) of distinct
rational numbers q1 , . . . , qp .
m + q1 m + q1
Consider the number . Clearly, this is a rational number and m < < q1 .
2 2
Hence, we found another distinct rational that is between m and n (and the procedure
can be repeated). This contradicts the fact that we assumed there were only p distinct
rationals between m and n.
7. Define the propositional variables
p : ”0 < a < 1”
q : ”a > a2 ”.
We need to prove that p → q. Remember that the implication is equivalent to ¬q → ¬p.
To construct an indirect proof, we assume that ¬q is true and prove that ¬p is true.
10
 If ¬q is true then a ≤ a2 . This implies that a2 − a ≥ 0, or a(a − 1) ≥ 0. The solution
of the inequality is (−∞, 0] ∪ [1, ∞). Hence, ¬p is true.

8. Define the propositional variables

p : ”n5 + 7 is even”
q : ”n is odd”.
To prove p → q using an indirect proof, we assume ¬q is true and show that ¬p is true.
If ¬q is true, then n is even: there exists an integer k such that n = 2k. Hence,
n5 + 7 = (2k)5 + 7 = 32k 5 + 7 = 2(16k 5 + 3) + 1. If we denote by m = 16k 5 + 3, we can
write n5 + 7 = 2m + 1, with m an integer. Therefore, n5 + 7 is odd, and consequently,
¬p is true.

9. Note that |x| = x, when x ≥ 0 and |x| = −x, when x < 0.

We have −|x| ≤ x ≤ |x| and similarly, −|y| ≤ y ≤ |y|. These imply

−|x| − |y| ≤ x + y ≤ |x| + |y|. (1)

We consider two cases corresponding to the sign of the sum x + y. Note that these
cases cover all possible situations. Define the propositional variables
p1 : ”x + y ≥ 0”
p2 : ”x + y < 0”
q : ”|x + y| ≤ |x| + |y|”.
To prove that q is true, we use a proof by cases and show that p1 → q and p2 → q.
Case (i). Assume p1 is true, x + y ≥ 0. Then |x + y| = x + y which together with the
right hand side of (1) implies that q is true.
Case (ii). Assume p2 is true, x + y < 0. Then |x + y| = −(x + y) which together with
the left hand side of (1) implies that q is true.

10. Define the propositional variables

p : 3 divides n2
q : 3 divides n.
We need to prove that p → q. We use the method of contradiction: we assume that
the negation ¬(p → q) is true. Since ¬(p → q) ≡ p ∧ ¬q we assume p is true and ¬q is
true. This implies that 3 is not a divisor of n. Hence, there exists an integer m such
that n = 3m + 1 or n = 3m + 2. We continue using the method of proof by cases.
Case (i). Assume n = 3m + 1 is true. Then n2 = 9m2 + 6m + 1 = 3(3m2 + 2m) + 1 =
3k + 1, with k = 3m2 + 2m - integer. This gives a contradiction with the fact that p
is true (3 divides n2 ).
Case (ii). Assume n = 3m+2 is true. Then n2 = 9m2 +12m+4 = 3(3m2 +4m+1)+1 =
3l + 1, with l = 3m2 + 4m + 1 - integer. This gives a contradiction with the fact that
p is true.
Since each case leads to a contradiction we conclude that ¬(p → q) is false (or p → q
is true).
11
11. The statement of the theorem is of type p → q, where
p : ”x2 does not divide a2 + b2 ”
q : ”x does not divide a or x does not divide b”.
To construct an indirect proof, we assume that ¬q is true: x divides a and x divides
b. We need to prove that ¬p is true: ”x2 divides a2 + b2 .
Since x divides a and x divides b, there exist integers k and l such that a = kx and
b = lx. It follows that a2 + b2 = k 2 x2 + l2 x2 = (k 2 + l2 )x2 , where (k 2 + l2 ) is an integer.
This proves that x2 divides a2 + b2 .

12. Define p : ”x3 + 3x + 5 = 0 has no rational roots”.

The proof can be done by contradiction and then use a proof by cases.
Assume the ¬p is true, i.e. the equation x3 + 3x + 5 = 0 has at least one rational root.
m
Denote it by r = , where m and n are integers, n ̸= 0 and m and n have no common
n
divisors other than 1. Then, the initial cubic equation can be written as

m3 + 3mn2 + 5n3 = 0. (2)

For the rest of the proof we consider 4 cases, corresponding to the parity of m and n.
Define the propositional variables
p1 : ”m and n are both even”
p2 : ”m is even and n is odd”
p3 : ”m is odd and n is even”
p4 : ”m and n are both odd”.
Note that the 4 cases cover all possible situations. To prove that ¬p leads to a contra-
diction, we prove that each case gives a contradiction.
Case (i). Assume p1 is true. This implies that 2 is a divisor of m and also 2 is a divisor
of n. This is a contradiction with the fact that m and n have no other common divisor
but 1.
Case (ii). Assume p2 is true. Since n is odd, 5n3 is odd. Since m is even, it follows
that m3 and 3mn2 are even and therefore, their sum is even. Hence, the left hand side
of (2) is odd. This is a contradiction since 0 is even.
Case (iii). Assume p3 is true. Since m is odd, m3 is odd. Since n is even, it follows
that 3mn2 , 5n3 are even and therefore their sum is even. These imply that the left
hand side of (2) is odd. This is a contradiction since 0 is even.
Case (iv). Assume p4 is true. Then, each term of the sum appearing at the left hand
side of (2) is odd. It follows that the sum is odd which contradicts the fact that 0 is
even.
Therefore, ¬p is false (and p is true).

12
4 Sets
1. (a) A ∩ B = {1, 3, 4}
(b) A ∪ B = {1, 2, 3, 4, 5, 6, 9}
(c) A − B = {2, 5}
(d) B − A = {6, 9}
(e) A ⊕ B = (A − B) ∪ (B − A) = {2, 5, 6, 9}
(f) (A ⊕ B) ∩ A = {2, 5}
(g) (A ⊕ B) ∪ B = {1, 2, 3, 4, 5, 6, 9}

2. (a) ∅ ⊆ A is true
(b) ∅ ∈ A is true
(c) {∅} ∈ A is true
(d) {∅} ⊆ A is true
(e) {∅, {∅}} ∈ A is true
(f) {{∅, {∅}}} ∈ A is false
(g) {{∅}} ∈ A is false
(h) {{∅}} ⊆ A is true
(i) {{∅}, {∅, {∅}}} ⊆ A is true
(j) {∅, {∅}, {{{∅}}}} ⊆ A is false
(k) {∅, {{∅}}} ∈ P(A) is false
(l) {{{{∅}}}} ⊆ P(A) is false

3. (a) (i) |A| = 3,

P(A) = {∅, {a}, {b}, {{a, b}}, {a, b}, {a, {a, b}}, {b, {a, b}}, {a, b, {a, b}}},
|P(A)| = 23 = 8
(b) (ii) |B| = 2, P(B) = {∅, {∅}, {{∅}}, {∅, {∅}}} and |P(B)| = 22 = 4

4. To prove that B = C, we need to show B ⊆ C and C ⊆ B.

Proof B ⊆ C. Let x ∈ B.
Remember that A ⊕ B = (A − B) ∪ (B − A) = (A ∪ B) − (A ∩ B). There are two cases.
Case 1. Assume x ∈ A. Since x ∈ B, it follows that x ∈ A ∩ B and hence, x ∈
/ A ⊕ B.
Since A ⊕ B = A ⊕ C, we obtain x ∈
/ A ⊕ C and hence, it follows that x ∈
/ A − C and
x∈/ C − A.
From x ∈
/ C − A and x ∈ A, we obtain x ∈ A ∩ C. This concludes the proof of x ∈ C.
Case 2. Assume x ∈ / A. Since x ∈ B, this implies x ∈ B − A and hence x ∈ A ⊕ B.
From the last relation it follows that x ∈ A ⊕ C and hence x ∈ (A − C) ∪ (C − A).
13
 Therefore, we have x ∈ (A − C) or x ∈ (C − A) and since x ∈
/ A we conclude that
x ∈ C − A and so x ∈ C.
Proof C ⊆ B follows similarly if we replace B by C in the previous proof.
5. (a)

(A − B) ∩ B = (A ∩ B) ∩ B
= (A ∪ B) ∩ B
= (A ∩ B) ∪ (B ∩ B)
= (A ∩ B) ∪ ∅
= (A ∩ B)
= (A − B).

(b)

(A − B) ∩ C = (A ∩ B) ∩ C
= (A ∪ B) ∩ C
= (A ∩ C) ∪ (B ∩ C)

(c)

(A − C) − (B − C) = (A ∩ C) ∩ (B ∩ C)
= (A ∩ C) ∩ (B ∪ C)
= (A ∩ B ∩ C) ∪ (A ∩ C ∩ C)
= ((A − B) ∩ C) ∩ ∅
= (A − B) − C

(d)
A ∩ (B − A) = A ∩ (B ∩ A)
= A∩B∩A
= ∅

(e)
(B − A) ∪ (C − A) = (B ∩ A) ∪ (C ∩ A)
= (B ∪ C) ∩ A
= (B ∪ C) − A

6. To prove that the two sets are equal, we first prove that A×(B∩C) ⊆ (A×B)∩(A×C).
Let (x, y) ∈ A × (B ∩ C). Then x ∈ A and y ∈ B ∩ C. Since y ∈ B ∩ C, it follows
that y ∈ B and y ∈ C. Therefore, (x, y) ∈ A × B and (x, y) ∈ A × C. These imply
(x, y) ∈ (A × B) ∩ (A × C).
A similar argument can be used to prove (A × B) ∩ (A × C) ⊆ A × (B ∩ C).
14
 7. (a) [−3, 6] ∩ (−2, 7] = (−2, 6]
(b) (−5, 7] ∩ Z = {−4, −3, −2, −1, 0, 1, 2, 3, 4, 5, 6, 7}
8. We construct proof by contradiction. Assume that there exist sets A ⊆ N and B ⊆ N
such that A × B = {(0, 0), (1, 1)}. It follows that (0, 0) ∈ A × B and (1, 1) ∈ A × B.
Hence, {0, 1} ⊆ A and {0, 1} ⊆ B.
Take x = 0 and y = 1. We know that x ∈ A and y ∈ B. However, (x, y) = (0, 1) ∈
/
A × B which is a contradiction.
9. The left hand side of the set identity is (A − B) − C = (A ∩ B) ∩ C = A ∩ (B ∩ C).
The right hand side is A − (B − C) = A ∩ (B − C) = A ∩ (B − C) = A ∩ (B ∩ C) =
A ∩ (B ∪ C).
In general, the identity is not true. The following is a counterexample. Let U =
{1, 2, 3, 4}, A = {4}, B = {1, 2, 3, 4} and C = {3, 4}.
Then, A − B = ∅ and so (A − B) − C = ∅. However, B − C = {1, 2} and so
A − (B − C) = {4}.
10. We have |A ∪ B| = |A| + |B| − |A ∩ B| = 6 + 8 − 3 = 11. Hence, |P(A ∪ B)| = 211 .
11. (a) The statement is false since A∩B = ∅ implies that they have no common elements.
E.g. Take A = {1} and B = {2}. Here, A ∩ B = ∅, however, A ̸= B.
(b) The statement is false since A−B = ∅ implies that A∩B = ∅. E.g. Take A = {1}
and B = {1, 2}. Here, A − B = ∅, however, A ̸= B.
(c) The statement is true.
To prove this let x ∈ A. We need to prove that x ∈ B.
We construct a proof by contradiction. Assume that x ∈ A but x ∈ / B. Then
x ∈ A and x ∈ B and hence x ∈ A − B. This is a contradiction with the fact that
A − B = ∅.
Hence, A ⊆ B.
(d) The statement is true. If A ∪ B = ∅, then each set A and B is empty. Hence,
A = B = ∅.
(e) The statement is true. Note that since A⊕B = (A−B)∪(B −A), then A⊕B = ∅
implies (see part (d)) that
i. A − B = ∅. This gives A ⊆ B (see part (c)).
ii. B − A = ∅. This gives B ⊆ A (see part (c)).
Hence, A = B.
(f) The statement is true. Since A × B = ∅, then A = B = ∅.
(g) The statement is false. If A − B = ∅, then (see part (c)) A ⊆ B and therefore
B ⊆ A. The following is a counterexample.
Let U = {1, 2, 3}, A = {1, 2} and B = {2}. Then, A = {3} and therefore,
A − B = A ∩ B = ∅. However, A ̸= B.
15
 (h) The statement is false. Note that A − B = A implies A ∩ B = A and hence,
A ⊆ B. The following is a counterexample.
Let A = {1}, B = ∅ and U = {1, 2}. Then A − B = A ∩ B = A ∩ U = A.
However, A is not a subset of B.
(i) The statement is false. A ∪ B = A implies that B ⊂ A. Let A = {1, 2}, B = {1}
and U = {1, 2, 3}. Note that A ∪ B = {1, 2} = A but B ̸= ∅.
(j) The statement is true. Let x ∈ A. We prove that x ∈ B.
Assume, by contradiction that x ∈ A but x ∈ / B. Then x ∈ B and since B ⊆ A
then x ∈ A. This is a contradiction with the fact that x ∈ A.
Hence, A ⊆ B.
12. Note that −x2 + x + 2 = −(x2 − x − 2) = −(x − 2)(x + 1). The solutions of the equation
−x2 + x + 2 = 0 are x = −1 and x = 2. The quadratic function −x2 + x + 2 has a
positive sign between its roots. Therefore, S = [−1, 2]. Then, S ∩ Z = {−1, 0, 1, 2}
and so |S ∩ Z| = 4.
13. (a) A100 = {−100, . . . , −1, 0, 1, 2, 3, . . .} and A96 = {−96, . . . , −1, 0, 1, 2, 3, . . .}
and so A100 − A96 = {−100, −99, −98, −97}.
(b) ∪∞n=1 An = Z. To see this, note that since each set An ⊆ Z, then ∪n=1 An ⊆ Z.
∞

It is left to prove that Z ⊆ ∪∞

n=1 An . Let n ∈ Z. We split the rest of the proof into
three cases, according to the sign of n.
Case 1. Assume that n > 0 then n ∈ An and so n ∈ ∪∞ n=1 An .
Case 2. Assume that n < 0 so n ∈ A−n . Hence, n ∈ ∪n=1 An .
∞

Case 3. If n = 0, then n ∈ A1 and therefore n ∈ ∪∞ n=1 An .

This concludes the proof of Z ⊆ ∪n=1 An , and so Z = ∪∞
∞
n=1 An .
(c) We claim that ∩∞ n=1 An = A1 , with A1 = {−1, 0, 1, . . .}. To prove it we should
prove both inclusions.
Clearly, ∩∞n=1 An ⊆ A1 (if x ∈ ∩n=1 An , then x ∈ An , for every n ≥ 1 and so
∞

x ∈ A1 ).
We need to prove that A1 ⊆ ∩∞ n=1 An . Let x ∈ A1 . Then, x ∈ An , for each n ≥ 2
(in addition, note that A1 ⊆ A2 ⊆ . . . An ⊆ . . ..). Hence, x ∈ ∩∞
n=1 An .

5 Functions
1. (a) f (n) = n + 5 is not onto. To see this, take m = 1 ∈ N. There is no n ∈ N
such that f (n) = 1. However, the function is one-to-one. If f (n1 ) = f (n2 ), i.e.
n1 + 5 = n2 + 5 then n1 = n2 .
[n]
(b) g(n) = is not one-to-one. Take n1 = 0 and n2 = 1. Then g(n1 ) = [0] = 0 and
[2 ]
1
g(n2 ) = = 0. But g is onto. Let m ∈ N. Then there exists n ∈ N such that
2
g(n) = m. (You may take n = 2m, for example).
16
 (c) Let f : N −→ N be given by
{
n + 1, if n is even
f (n) =
n − 1, if n is odd

To show that f is one-to-one we consider n1 ∈ N and n2 ∈ N such that f (n1 ) =

f (n2 ). Then, we consider the following cases.
Case 1. Assume that n1 + 1 = n2 + 1. Then n1 = n2 .
Case 2. Assume that n1 − 1 = n2 − 1. Then n1 = n2 .

To show that f is onto, take any m ∈ N. There are two cases.

Case 1. If m is even, then m + 1 is odd and f (m + 1) = (m + 1) − 1 = m. Hence,
there exists n ∈ N, (n = m + 1) such that f (n) = m.
Case 2. If m is odd, then m − 1 is even and f (m − 1) = (m − 1) + 1 = m. Hence,
there exists n ∈ N, (n = m − 1) such that f (n) = m.
Hence, f is bijective.
(d) Let f : N −→ N be given by f (n) = 1. Clearly, f is not bijective (is neither a
one-to-one function nor onto.)

2. Let g : A −→ B and f : B −→ C be two functions.

(a) Let x1 ∈ A and x2 ∈ A such that g(x1 ) = g(x2 ). Then f (g(x1 )) = f (g(x2 )), or
(f ◦ g)(x1 ) = (f ◦ g)(x2 ). Since f ◦ g is one-to-one, it follows that x1 = x2 . Hence,
g is one-to-one.
(b) The statement is not true, in general. The following provides a counterexample.
Let A = {a}, B = {b1 , b2 } and C = {c}. We define the functions as follows.
Let g(a) = b1 , f (b1 ) = f (b2 ) = c. Then (f ◦ g)(a) = c and f are clearly onto.
However, g is not onto since there is no element in the domain that is mapped
into b2 (or b2 ∈
/ Image(g)).

3. Let f : A −→ B be a function, and S and T two subsets of A. Show that:

(a) The equality of the two sets can be proved using the definition.
Let y ∈ f (S ∪ T ). Then there exists x ∈ S ∪ T such that f (x) = y. It follows that
x ∈ S or x ∈ T , and f (x) = y. This gives y ∈ f (S) or y ∈ f (T ) which proves
f (S ∪ T ) ⊆ f (S) ∪ f (T ).
The proof of f (S) ∪ f (T ) ⊆ f (S ∪ T ) can be done similarly.
(b) Let y ∈ f (S ∩ T ). Then, there exists x ∈ S ∩ T such that f (x) = y. This implies
that x ∈ S and x ∈ T , and f (x) = y. Hence, y ∈ f (S) ∩ f (T ).

4. We show that f is not one-to-one. Take m1 = 1, n1 = 1, m2 = 2 and n2 = 2. Clearly,

(m1 , n1 ) ̸= (m2 , n2 ) but f (m1 , n1 ) = 1 and f (m2 , n2 ) = 1.

17
 We show that f is onto. Let q ∈ Q. From the definition of a rational number, there
m
exist integers m and n, with n ̸= 0 such that q = . This proves that for any q ∈ Q,
n
m
there exist (m, n) ∈ Z × (Z − {0}) such that f (m, n) = .
n
5. For each of the following assignments, determine whether it is a function or not. If it
is a function, is it one-to-one? Is it onto?

(a) f1 : R −→ R, f1 (x) = −x3 +1 is a bijective function (one-to-one since −x31 +1 =

−x32 + 1 implies x1 = x2 and onto since for any y ∈ R, there exist x ∈ R:
√
x = 3 1 − y such that f (x) = y).
(b) f2 : N −→ Z, f2 (n) = n2 + 3 is one-to-one (n21 + 3 = n22 + 3 implies n1 = n2 since
they are both natural numbers) but not onto (If z = 0 ∈ Z, there is no n ∈ N
such that f (n) = 0).
(c) f3 : R −→ [0, +∞), f3 (x) = 2x is one-to-one (2x1 = 2x2 implies x1 = x2 ) but
not onto (for y = 0, there does not exist x ∈ R such that 2x = 0).
√ √
(d) f4 : N −→ N, f4 (n) = n + 1 is not a function since n + 1 ∈ / N.
(e) f5 : R × R −→ R, f5 (x, y) = x + y is not one-to-one (if (x1 , y1 ) = (0, 0) and
(x2 , y2 ) = (1, −1), f (x1 , y1 ) = f (x2 , y2 ) = 0) but f is onto (if y ∈ R, there
exists a pair, e.g. (0, y) ∈ R × R, with f (0, y) = y).
(f) f5 : R × R −→ R × N, f6 (x, y) = (2x, 0) is not on-to-one ((1, 2) and (1, 3) are
both mapped into (2, 0)) and not onto (for (1, 5) ∈ R × N there does not exist
a pair (x, y) ∈ R × R such that f (x, y) = (1, 5)).

6. Let (a1 , b1 ) ∈ A × B and (a2 , b2 ) ∈ A × B with λ(a1 , b1 ) = λ(a2 , b2 ). Then,

2φ(a1 ) 3ψ(b1 ) = 2φ(a2 ) 3ψ(b2 ) and so 2φ(a1 )−φ(a2 ) = 3ψ(b1 )−ψ(b2 ) . It follows that ϕ(a1 )−ϕ(a2 ) =
ψ(b1 ) − ψ(b2 ) = 0. Since ϕ and ψ are one-to-one functions, we obtain a1 = a2 and
b1 = b2 and so (a1 , b1 ) = (a2 , b2 ).

7. (a) Let x1 ∈ (0, +∞) and x2 ∈ (0, +∞) be such that f (x1 ) = f (x2 ). Then, 2x21 + 3 =
2x22 +3. This implies x21 = x22 and since they are positive numbers, we can conclude
that x1 = x2 . Hence, f is one-to-one.
(b) For any x ∈ (0, +∞), we have x2 > 0 and so 2x2 + 3 > 3. Hence, Image(f ) =
(3, ∞). Since Image(f ) ̸= (0, ∞), we conclude that f is not onto.

8. The
√ function f ◦ g : R −→ R is given by (f ◦ g)(x) = f (g(x)) = f (x3 − 2) =
3
(x3 − 2) + 2 = x, for any x ∈ R.
√
The
√ function g ◦ f : R −→ R is given by (g ◦ f )(x) = g(f (x)) = g( 3 x + 2) =
( 3 x + 2)3 − 2 = x, for any x ∈ R.
Since f ◦ g = 1R and g ◦ f = 1R we conclude that f −1 = g. This implies that f is a
bijection and henceg is also a bijection.

9. (a) Since |A| = 4 < |B| = 6 the statement is true.

18
 (b) Since |P(A)| = 24 < 26 = |P(B)| the statement is false.
(c) Since |P(A)| = 24 > 6 = |B| the statement is true.
(d) Since |A × B| = |A| · |B| = 4 · 6 = 24 < 26 |P(B)| the statement is true.
4
(e) Since |P(P(A))| = 22 = 216 < 224 = |P(A × B)| the statement is false.

10. The functions are not one-to-one.

If (x1 = 1.2 and x2 = 1.5) then (f (x1 ) = ⌊1.2⌋ = 1 and f (x2 ) = ⌊1.5⌋ = 1) and
(g(x1 ) = ⌈1.2⌉ = 2 and g(x2 ) = ⌈1.5⌉ = 2).

However, the functions are onto.

Let m ∈ Z be arbitrary.
Then there exists x ∈ R such that the equation f (x) = m has at least one solution
in R (x = m is one solution). Similarly, there exists x ∈ R such that the equation
g(x) = m has at least one solution in R (x = m is one solution).

11. (a) The conclusion follows once we prove that g ◦ f and (g ◦ f )−1 are invertible. We
show this by actually identifying their inverses.
Since (g ◦ f ) ◦ (f −1 ◦ g −1 ) = g ◦ f ◦ f −1 ◦ g −1 = g ◦ 1B ◦ g −1 = g ◦ g −1 = 1C ,
and similarly, (f −1 ◦ g −1 ) ◦ (g ◦ f ) = 1A , it follows that (g ◦ f )−1 = (f −1 ◦ g −1 )
and hence, g ◦ f and f −1 ◦ g −1 are invertible (f −1 and g −1 exist since f and g are
assumed to be bijective).
√
(b) Let f : [0, ∞) −→ R, f (x) = x and g : R −→ [0, ∞), g(x) = x2 .
The functions f and g are not bijective since f is not onto (for y = −1 ∈ R
the equation f (x) = −1 has no solution in [0, ∞)) and g is not one-to-one
(g(−1) = g(1) = 1).
√ √
However, (g ◦ f )(x) = g(f (x)) = g( x) = ( x)2 = x is bijective. (Note that in
this case, (f ◦ g)(x) = |x|, for x ∈ R.)

6 Relations
1. R1 is reflexive since (x, x) ∈ R1 , for any x ∈ A
R1 is symmetric since (x, y) ∈ R1 implies (y, x) ∈ R1 , for any x ∈ A and y ∈ A (or
equivalently, there is no pair (x, y) ∈ A × A such that (x, y) ∈ R1 and (y, x) ∈
/ R1 )
(details are omitted)
R1 is transitive since (x, y) ∈ R1 and (y, z) ∈ R1 implies (x, z) ∈ R1 (details are
omitted)
R1 is not antisymmetric since (1, 2) ∈ R1 and (2, 2) ∈ R1 but 1 ̸= 2
R2 is reflexive since (x, x) ∈ R2 , for any x ∈ A

19
 R2 is symmetric since (x, y) ∈ R2 implies (y, x) ∈ R2 , for any x ∈ A and y ∈ A
(details are omitted)
R2 is transitive since (x, y) ∈ R2 and (y, z) ∈ R2 implies (x, z) ∈ R2
R2 is antisymmetric since (x, y) ∈ R2 and (y, x) ∈ R2 implies x = y
R3 is not reflexive since (1, 1) ∈
/ R3 , and 1 ∈ A
R3 is not symmetric since (1, 4) ∈ R3 but (4, 1) ∈ R3
R3 is not transitive since (1, 3) ∈ R3 and (3, 1) ∈ R3 but (1, 1) ∈
/ R3
R3 is not antisymmetric since (1, 3) ∈ R3 and (3, 1) ∈ R3 but 1 ̸= 3

2. R1 is not reflexive since (a, a) ∈

/ R1
R1 is symmetric: if (x, y) ∈ R1 then (y, x) ∈ R1 is true vacuously
R1 is transitive: if (x, y) ∈ R1 and (y, z) ∈ R1 then (x, z) ∈ R1 is true vacuously
R1 is antisymmetric: if (x, y) ∈ R1 and (y, x) ∈ R1 then x = y is true vacuously
R2 is not reflexive, not symmetric, transitive and antisymmetric
R3 is not reflexive, symmetric, not transitive and not antisymmetric
R4 is reflexive, not symmetric, transitive and antisymmetric

3. R1 is not reflexive since e.g. (1, 1) ∈

/ R1
R1 is not symmetric since e.g. (1, 2) ∈ R1 but (2, 1) ∈
/ R1
R1 is transitive since (a, b) ∈ R1 and (b, c) ∈ R1 implies (a, c) ∈ R1 (that is if a < b
and b < c then a < c)
R1 is antisymmetric since (a, b) ∈ R1 and (b, a) ∈ R1 implies a = b vacuously
R2 is not reflexive since e.g. (4, 4) ∈
/ R2
R2 is not symmetric since e.g. (4, 2) ∈ R2 but (2, 4) ∈
/ R2
R2 is not transitive since e.g. (−3, −2) ∈ R2 and (−2, 0) ∈ R2 but (−3, 0) ∈
/ R2
R2 is not antisymmetric since e.g. (−1, −2) ∈ R2 and (−2, −1) ∈ R2 but −1 ̸= −2

4. (a) The statement is true since |P(A × B)| = 215 = 32768

(b) The statement is false. The number of binary relations from B to A equals to the
number of binary relations from A to B
(c) The statement is false. A binary relation from P(A) to P(B) is a subset of
P(P(A) × P(B)) which is not equal to P(A × B).
(d) The statement is true since the number of binary relation from A to B that contain
the subset {(x, 0); x ∈ A} is equal to the number of binary relations that can be
defined from A to the set {1, 2, 3, 4}, which is 23·5 = 4096.

20
 (e) The statement is false. The number of binary relation from A to B that contain
the subset {(a, y); y ∈ B} is equal to the number of binary relations that can be
defined from the set {b, c} to B, which is 22·5 ̸= 4096.

5. R1 is not a function since (1, a) ∈ R1 and (1, b) ∈ R1 (the element 1 would be mapped
into two different values).
R1 is a function. It is not one-to-one since (0, c) ∈ R2 and (1, c) ∈ R2 (two different
inputs have the same output). It is not onto since there is no x ∈ A such that
(x, b) ∈ R2 (or (x, e) ∈ R2 )
R3 is a function. It is not one-to-one since e.g. (0, c) ∈ R3 and (1, c) ∈ R3 . It is not
onto since there is no x ∈ A such that (x, b) ∈ R3 (or (x, e) ∈ R3 )
R4 is a function. It is not one-to-one since (1, a) ∈ R4 and (3, a) ∈ R4 . It is not onto
since there is no x ∈ A such that (x, e) ∈ R4 .
R5 is not a function since it does not contain an element of the form (1, y) (the element
1 is not mapped into an element on B).
R6 is a function. It is one-to-one and onto (and therefore bijective).

6. (a) R1 is not reflexive since e.g. (1, 1) ∈

/ R1 .
R1 is symmetric
R1 is not transitive e.g. (1, 2) ∈ R1 and (2, 3) ∈ R1 but (1, 3) ∈
/ R1
(b) R1 is not reflexive since e.g. (2, 2) ∈
/ R2 .
R2 is symmetric
R2 is transitive (x, y) ∈ R2 and (y, z) ∈ R2 implies (x, z) ∈ R2
(c) R3 is reflexive since for any x ∈ Z we have xR3 x = x + x2 = x(x + 1) which is an
even integer
R3 is not symmetric. Let (2, 1) ∈ Z × Z. Since 2 + 2 · 1 = 4 is even, 2R3 1.
However, (1, 2) ∈ / R since 1 + 1 · 2 = 3 is not even
R3 is transitive. Let (x, y) ∈ R3 and (y, z) ∈ R3 . Then x + xy is even and y + yz
is even. The proof continues by considering two cases, according to the parity of
y.
Case 1. Assume that y is even. Then xy is even and since x + xy is even, it follows
that x is even. Hence, x + xz is even.
Case 2. Assume that y is odd. Then, since y + yz is even it follows that z is odd
and so x + xz is even.

7. (a) R1 is reflexive (A ⊆ A, for any A ∈ P(U )), not symmetric (A ⊆ B does not
imply B ⊆ A), transitive (A ⊆ B and B ⊆ C implies A ⊆ C) and antisymmetric
(A ⊆ B and B ⊆ A implies A = B)
(b) R2 is not reflexive (A ∩ A = A ̸= ∅ not true for any A ∈ P(U )), symmetric
(A ∩ B = ∅ implies B ∩ A = ∅), not transitive (e.g. if A = {1, 2}, B = {4} and

21
 C = {2, 3}, then A ∩ B = ∅, B ∩ C = ∅ but A ∩ C = {2}) and not antisymmetric
(e.g. A ={1}, B = {2} and A ∩ B = B ∩ A = ∅)
(c) First show that A − B = ∅ if and only if A ⊆ B.
To prove the equivalence, we first assume A − B = ∅ is true and show that A ⊆ B
holds (see also Exercise 11, part (c) in Section 4).
Let x ∈ A. Then, since A − B = A ∩ B = ∅, it follows that x ∈ / B and hence
x ∈ B. This proves A ⊆ B.
Now we assume that A ⊆ B holds and show A − B = ∅.
Case 1. If A = ∅, then A − B = ∅ is true.
Case 2. Assume A ̸= ∅. Let x ∈ A. Then, since A ⊆ B, we have x ∈ B and so
x∈/ B. This implies A − B = A ∩ B = ∅.
Therefore, A R3 B ⇔ A ⊆ B.
Therefore, R3 is equivalent to R1 and so it has the same properties.
(d) Using Exercise 11, part (e) in Section 4, we conclude that A ⊕ B = ∅ if and only
if A = B.
Therefore, R4 becomes A R4 B ⇔ A = B.
R4 is reflexive (A = A for any A ∈ P(U )), symmetric, transitive (A = B and
B = C implies A = C) and antisymmetric (if A = B and B = A, then A = B)
8. (a) We first show that R is symmetric.
Let (a, b) ∈ R. Since R is reflexive, (b, b) ∈ R. Now, since R is cyclic, we apply
the definition (with c = b) and obtain that (b, a) ∈ R which proves that R is
symmetric.
We now prove that R is transitive.
Let (a, b) ∈ R and (b, c) ∈ R. Since R is cyclic we obtain (c, a) ∈ R which
together with the fact that R is symmetric (proved before) proves that (a, c) ∈ R.
(b) Let (a, b) ∈ R and (b, c) ∈ R. Then, since R is transitive, (a, c) ∈ R. This
together with the fact that R is symmetric implies that (c, a) ∈ R which proves
that R is cyclic.
9. (a) Let R = {(2, 2), (2, 3), (3, 2), (3, 3)}. R is symmetric and transitive but not
reflexive since (1, 1) ∈
/ R.
(b) The proof in not correct since the following assumption is used at the beginning
of the argument: for any a ∈ A, there exists b ∈ A such that (a, b) ∈ R. This
statement is not true in general (see example given at the previous part: for
a = 1 ∈ A there is no element b ∈ A such that (1, b) ∈ R).

7 Equivalence Relations
x
1. (a) The relation R is reflexive (for any x ∈ A, xRx, since = 30 ), symmetric (for
x
x y
any x, y ∈ A, xRy implies yRx, since = 3k implies = 3−k for some k ∈ Z)
y x
22
 x
and transitive (for any x, y, z ∈ A, xRy and yRz imply xRz, since = 3k and
y
y l x k+l
= 3 imply = 3 for some k ∈ Z and l ∈ Z).
z z
{ } { } { } { }
1 1 1 1 9 2
(b) The partition is {1}, , , 3 , , , , {2}, , {5}.
3 27 4 36 4 9
2. (a) The relation R is reflexive (for any f ∈ A, f Rf , since f (x)−f (x) = 0), symmetric
(for any f, g ∈ A, f Rg implies gRf , since f (x)−g(x) = c implies g(x)−f (x) = −c
for some constant c ∈ Z) and transitive (for any f, g, h ∈ A, f Rg and gRh imply
f Rh, since f (x) − g(x) = c and g(x) − h(x) = d imply f (x) − h(x) = c + d for
some c ∈ Z and d ∈ Z).
(b) [f (x)]R = {g : Z → R, g is a function such that f (x)−g(x) = c, for some constant c ∈
Z} = {g : Z → R, g is a function such that g(x) = 2x + c, for some constant c ∈
Z}. Hence, f1 ,f5 ∈ [f (x)]R

3. (a) The fact that R is an equivalence relation on W follows from properties of real
numbers. Note that two pairs are included in R if and only if their components
have same sign, respectively, i.e. (x, y)R(a, b) if and only if (x and a have same
sign and y and b have same sign)
(b) There are four equivalence classes on W (corresponding to the four quadrants in
the plane):
[(1, 1)]R , [(1, −1)]R , [(−1, 1)]R , [(−1, −1)]R .

4. (a) Follows easily from properties of real numbers.

(b) The following two sets form the partition on A defined by R.
{( ) ( )}
1 0 2 1
, ,
0 1 3 2
{( ) ( ) ( ) ( )}
1 0 2 4 0 1 0 0
, , , .
0 0 3 6 0 0 0 0

5. The given binary relation on R2 can be equivalently defined as

(x, y)R (x′ , y ′ ) ⇔ x − y = x′ − y ′ .

(a) Follows easily from properties of real numbers.

(b) [(1, 1)]R = {(x, y) ∈ R2 , x − y = 0} = {(x, x) ∈ R2 }. The equivalence class of
the vector (1, 1) represents the points located on the first diagonal, i.e the points
located on the line y = x.

6. (a) R is an equivalence relation since it is:

- reflexive (a = a · 1)

23
 1
- symmetric (if b = a · c then a = b, for some c > 0)
c
- transitive (if b = a · c and d = b · e, for some c > 0 and e > 0, then d = a · (ce),
with ce > 0)
(b) The partition consists of [1]R = (0, ∞), [−1]R = (−∞, 0) and [0]R = {0}.

7. (a) Note that κ(a) ∈ N∗ , for any a ∈ N∗ . R is an equivalence relation since it is:
- reflexive (κ(a) = κ(a))
- symmetric (if κ(a) = κ(b) then κ(b) = κ(a))
- transitive (if κ(a) = κ(b) and κ(b) = κ(c), then κ(a) = κ(c))
(b) [2]R = {n ∈ N∗ , κ(2) = κ(n)}. Since κ(2) = 2, [2]R = {n ∈ N∗ , n is even}.

8. (a) omitted
(b) The partition is the following sequence of sets.
{011, 111} , {010, 110} , {001, 101} , {000, 100}

9. (a) omitted
(b) [(0, 2)]R = {(x, y) ∈ R2 , x2 + y 2 = 4}. The sets consists of the points located on
the circle of radius 2 and origin (0, 0).
Other elements in the equivalence class of (0, 2) are e.g. (2, 0), (−2, 0), (0, −2).

10. (a) R is an equivalence relation since it is

- reflexive: ARA since A = 1 · A
1
A, for some λ ∈ R∗
- symmetric: ARB implies BRA, since A = λB implies B =
λ
- transitive: ARB and BRC imply ARC, since A = λB and B = γC imply
A = δC, where δ = λ · γ ∈ R∗
(b) The partition is {A1 , A5 , A8 }, {A2 , A6 }, {A3 , A7 } and {A4 }.
a b
11. Let a, b ∈ R∗ such that aRb. Then |a|b = a|b|, or = . Hence aRb if a and b have
|a| |b|
the same sign, so the given relation can be equivalently written as

aR b ⇔ a and b have the same sign.

(a) omitted
(b) [1]R = (0, ∞).

8 Basic Counting Techniques

1. Note that in this question we assume that a license plate has six characters.
Let A be the set of license plates that can be made with three digits followed by three
letters and B be the set of license plates that can be made with three letters followed
by three digits. We need to compute the cardinality of A ∪ B.
24
 Since A ∩ B = ∅, using the principle of exclusion-inclusion, we have |A ∪ B| = |A| + |B|.
Using the product rule, |A| = 10 · 10 · 10 · 26 · 26 · 26 = 2603 , |B| = 2603 and hence
|A ∪ B| = 2 · 2603 = 35152000.
2. Assume that n is even. Then, there exists k ∈ N∗ such that n = 2k. Hence, we
only need to count the number of binary strings of length k that can be formed. This
n
number is equal to 2k , using the product rule. Therefore, there are 2 2 palindromes of
length n, if n is even.
Assume that n is odd. Then, there exists k ∈ Nk such that n = 2k + 1. The number
of binary strings of length k that can be formed is 2k , but since n is odd, we need
to multiply by 2 (at position k + 1 we can have either 0 or 1). Therefore, there are
n−1 n+1
2 2 +1 = 2 2 palindromes of length n, if n is odd.
3. There are 5 odd digits and 26 letters in total. Using the product rule and excluding
repetitions, the number of postal codes of the given form is 5 · 26 · 4 · 25 · 3 · 24 = 936000.
4. Assume that A = {x1 , . . . , x7 } and B = {y1 , . . . , y4 }.
(a) There are 4 ways in which f (x1 ) can be defined, 4 ways in which f (x2 ) can be
defined, etc. In total, using the product rule, there are 47 = 16384 ways in which
a function from A to B can be defined.
(b) There are 7 ways in which f (y1 ) can be defined, 7 ways in which f (y2 ) can be
defined, etc. In total, using the product rule, there are 74 = 2401 ways in which
a function from B to A can be defined.
(c) Since |A| > |B|, we cannot define an one-to-one function from A to B.
(d) There are 7 ways in which f (x1 ) can be defined. Since the function must be
one-to-one, there are 6 ways in which f (x2 ) can be defined (after choosing the
value of f (x1 )). After specifying the values of f (x1 ) and f (x2 ), there are 5 ways of
defining f (x3 ) and finally, using the same procedure, there are 4 ways of defining
f (x4 ). In total, using the product rule, there are 7 · 6 · 5 · 4 = 840 ways in which
a one-to-one function from B to A can be defined.
(e) We first compute the number of functions that can be defined from A to B that
are not onto.
For every 1 ≤ i ≤ 4, we define the set Fi = {f : A → B, yi ∈/ Image(f )}. The
number of functions that can be defined from A to B that are not onto is
|F1 ∪ F2 ∪ F3 ∪ F4 | = |F1 | + |F2 | + |F3 | + |F4 | − |F1 ∩ F2 | − |F1 ∩ F3 |
− |F1 ∩ F4 | − |F2 ∩ F3 | − |F2 ∩ F4 | − |F3 ∩ F4 |
+ |F1 ∩ F2 ∩ F3 | + |F1 ∩ F2 ∩ F4 | + |F1 ∩ F3 ∩ F4 | + |F2 ∩ F3 ∩ F4 |
− |F1 ∩ F2 ∩ F3 ∩ F4 |
= 3 7 + 3 7 + 37 + 3 7 − 27 − 27 − 27 − 2 7 − 27 − 27
+ 1 7 + 1 7 + 17 + 1 7 − 0
= 7984.
25
 Now, the number of onto functions that can be defined from A to B is equal to
16384 − 7984 = 8400.
(f) Since |B| < |A|, we cannot define an onto function from B to A.

5. Note that there are 9 digits that can be placed on the first position of the number
(corresponding to the cardinality of {1, . . . 9}), 9 digits that can be placed on the
second position, after the first position was occupied (that is 10 − 1), etc. The number
of integers with the given property is 9 · 9 · 8 · 7 · 6 = 27216.

6. Let A = {n ∈ N : 7 ≤ n ≤ 2125, n is divisible by 3} and B = {n ∈ N : 7 ≤ n ≤

2125, n is divisible by 11}. Then A∩B = {n ∈ N : 7 ≤ n ≤ 2125, n is divisible by 33}.

(a) The number of integers between 7 and 2125 (inclusive) that are divisible by 3
or 11 is |A ∪ B|. To compute it we need |A| = ⌊ 2125 3
⌋ − ⌊ 36 ⌋ = 708 − 2 = 706,
|B| = ⌊ 2125
11
6
⌋ − ⌊ 11 ⌋ = 193 − 0 = 193, and |A ∩ B| = ⌊ 2125
33
6
⌋ − ⌊ 33 ⌋ = 64 − 0 = 64.
Using the Principle of Inclusion-Exclusion, |A ∪ B| = |A| + |B| − |A ∩ B| =
706 + 193 − 64 = 835.
(b) Let U = {n ∈ N : 7 ≤ n ≤ 2125} be our universal set. This question is asking for
the number of integers between 7 and 2125 (inclusive) that are not divisible by 11
(since 11 is a prime, an integer n is co-prime with 11 — that is, has no common
divisors with 11 except 1 — precisely when n is not divisible by 11). This number
is |U − B| = |U | − |B| = (2125 − 6) − 193 = 1926.
(c) We need to compute |A − B|. We write A = (A − B) ∪ (A ∩ B), and since
(A − B) ∩ (A ∩ B) = ∅, using the Sum Rule we obtain |A| = |A − B| + |A ∩ B|.
From here, we conclude that |A − B| = |A| − |A ∩ B| = 706 − 64 = 642.

7. Let A-be the set of length 13 binary strings that begin with 0110 and B-be the set of
length 13 binary strings that end with 1000. We need |A ∪ B|. Using the principle of
inclusion-exclusion, |A ∪ B| = |A| + |B| + |A ∩ B|.
We have |A| = 29 , |B| = 29 and |A ∩ B| = 25 and so |A ∪ B| = 29 + 29 − 25 = 992.

8. Let A = {n ∈ N : 1 ≤ n ≤ 250, n is divisible by 4} and B = {n ∈ N : 1 ≤ n ≤

250, n is divisible by 6}. Then A ∩ B = {n ∈ N : 1 ≤ n ≤ 250, n is divisible by 12}.
We need |A ∪ B|.
We have |A| = ⌊ 250
4
⌋ = 62, |B| = ⌊ 250
6
⌋ = 41 and |A ∩ B| = ⌊ 250
12
⌋ = 20. It follows that
|A ∪ B| = 62 + 41 − 20 = 83.

9. For every 0 ≤ i ≤ n we define Ai = {the set of binary strings of length i}. Note that
the binary string of length 0 is the empty string so A0 = {empty string} and |A0 | = 1.
Since the sets Ai are mutually disjoint, the number of binary strings of length at most

26
 n is
n
∑
2 3 n
|A0 ∪ . . . An | = |A0 | + |A1 | + . . . + |An | = 1 + 2 + 2 + 2 . . . + 2 = 2i
i=0
n+1
2 −1
= = 2n+1 − 1.
2−1

10. In this question it is assumed that the plates can have either 4, 5 or 6 characters.
Let
A be the set consisting of all 4 characters license plates that are formed using 2 letters
followed by 2 digits
B be the set consisting of all 5 characters license plates that are formed using 2 letters
followed by 3 digits
C be the set consisting of all 5 characters license plates that are formed using 3 letters
followed by 2 digits
D be the set consisting of all 6 characters license plates that are formed using 3 letters
followed by 3 digits.
We need to find A ∪ B ∪ C ∪ D. Using the product rule, |A| = 262 · 102 = 67600,
|B| = 262 · 103 = 676000, |C| = 263 · 102 = 1757600 and |D| = 263 · 103 = 17576000.
Since the sets are mutually disjoint, it follows that A∪B∪C ∪D = |A|+|B|+|C|+|D| =
20077200.

11. For 6 ≤ i ≤ 9, we define

Pi = {i characters passwords that contain at least two distinct characters}.

We need to find |P6 ∪ P7 ∪ P8 ∪ P9 |.

For 6 ≤ i ≤ 9, since the length i password can contain characters that are either a
lower letter, an upper case or a digit, it follows that there are 2 · 26 + 10 options for a
certain position. Because a password cannot have identical characters it follows that
|Pi | = 62i − 62.
Therefore, since Pi are mutually disjoint, for 6 ≤ i ≤ 9, we have |P6 ∪ P7 ∪ P8 ∪ P9 | =
|P6 | + |P7 | + |P8 | + |P9 | = (626 − 62) + (627 − 62) + (628 − 62) + (629 − 62) =.

12. (a) Since the bride should stand next to the groom (i.e. either to the left or to the
right), the number of ways the group of 6 people can be arranged equals to twice
the number of ways in which we can arrange a group of 5 people. Hence, there
are 2 · 5! = 240 ways in which a group with the given property can be arranged.
(b) There are 6! ways in which a group of 6 can be arranged and so, using part (a),
there are 6! − 2 · 5! = 480 ways to arrange the group so that the bride does not
stand next to the groom.
(c) The arranging procedure can be divided into the following steps.

27
 T1 : arrange the 4 people in the group that are neither the bride neither the groom
nor the bride
T2 : arrange the groom in one of the 5 gaps between the guests and the bride to
the left of groom
There are 4! ways to perform T1 .
According to the position of the groom, there are different ways in which the
bride can be placed. Using the summation rule, the number of ways T2 can be
performed is 1 + 2 + 3 + 4 + 5. The first term corresponds to the situation when
the groom is placed on the fifth position and the bride is placed on the sixth, the
second term corresponds to the situation when the groom is placed on the fourth
position, and hence the bride can be placed either on the fifth or on the sixth
position, etc.
Using the product rule, there are 15 · 24 = 360 ways to arrange people in a picture
such that the bride is situated to left of the groom.

9 The Pigeonhole Principle

1. Draw the three diagonals to divide the regular hexagon into 5 equilateral triangles,
each with side of length 1. In each triangle, any two points in the interior or on the
perimeter are at distance at most one. Since there are 6 triangles and 7 points, by PP,
two points will have to lie in the same triangle, and hence be at distance at most 1.
2. For each computer, the number of computers it is directly linked to (call them neigh-
bours) is in the set {1, 2, 3, 4, 5}. There are 6 computers and 5 possible numbers of
neighbours; hence by PP, at least two computers must have the same number of neigh-
bours.

3. There are 7 · 12 = 84 possible pairs (day of the week, month). Hence by PP, we need
at least 85 people to guarantee that at least two among them were born on the same
day of the week and in the same month.

4. Assuming that every package contains 20 distinct cards, we need at least 28 packages,
since 27 · 20 < 551 ≤ 28 · 20.

5. There are 14 possible remainders (0, 1, 2, . . . , 13) when dividing by 14. Hence any set
of 15 integers will contain two that give the same remainder when divided by 14. The
difference of these numbers will be then divisible by 14.

6. There are 9 possible remainders (0, 1, 2, . . . , 8) when dividing by 9. The “worst-case

scenario” is to have 9 · 5 = 45 integers such that no 6 have the same remainder.
However, by PP, any set of 46 integers will contain 6 that have the same remainder.
7. The number of such birth certificate codes is n = 104 · 263 . By PP, to guarantee that
at least 26 certificates carry the same code, we need at least 25n + 1 certificates. Hence
the number of people (certificates) is at least 25 · 104 · 263 + 1 = 4, 394, 000, 001.
28
 8. The number of subsets of A of cardinality at most 3 is 90 + 91 + 92 + 93 = 1 +
() () () ()

9 + 36 + 84 = 130. The sum of the elements that such a set can have is in the set
{0, 1, 2, . . . , 24}. Since 130 > 24 · 5, any there exist 6 subsets of cardinality at most 3
that have the same sum of elements.

9. Draw a grid that divides the rectangle into squares of side length 10cm. There will be
200 such squares. Any three points within (or on the perimeter) of one of these squares
define a triangle of area at most half the area of the square, that is, at most 50cm2 .
Since there are 500 points and 200 squares, by PP, some square indeed contains at
least 3 points (otherwise there would be at most 400 points).

10. Consider the integers 1, 11, 111, ..., up to the integer with 7778 repeated ones, and
their remainders when divided by 7777. Since there are only 7777 possible remainders
when divided by 7777, two of these 7778 integers (say x with a ones and y with b ones,
for x < y) have the same remainder. Hence their difference y − x is divisible by 7777.
But y − x has b − a ones followed by a zeros. Let z be the number with b − a repeated
ones. Then y − x = z · 10a . So 7777 divides z · 10a , but since 7777and 10a have no
common divisors, 7777 must in fact divide z. Hence z is the integer with repeated ones
that is divisible by 7777.

11. The possible numbers of mistakes are 0, 1, 2, . . . , 12 (boxes). Placing the 30 objects
(students) into 12 boxes, by PP, we’ll end up with at least one box with at least 3
objects.
x +x y +y z +z
12. For each pair i, j ∈ {1, 2, . . . , 9}, i < j, let Mij = ( i 2 j , i 2 j , i 2 j ) denote the
midpoint of the line segment joining points Pi = (xi , yi , zi ) and Pj = (xj , yj , zj ). Mij
will have integer coordinates if and only if xi and xj , yi and j , and zi and zj have equal
parity. There are 8 possible triples (q1 , q2 , q3 ) where each qi is either “odd” or “even”;
these are our boxes. Placing the 9 points into these 8 boxes we’ll end up with a box
with at least two points, say Pi and Pj . Then their midpoint Mij will have integer
coordinates as explained above.

10 Permutations and Combinations

1. 2 · (n!)2

2. 40
( )
17
100
( )
3. The coefficient of xk is 0 if k is odd, and 50− k2
if k is even.

4. Essentially, we have 2 symbols,(9)five 011s (which jointly contain ten 1s) and four more
1s. These can be arranged in 5 ways.
(200) 99 101
5. (a) −
(24)1012 2 3
(b) ( 2) 2
(c) 84 24 and 0, respectively.
29
 6. (a) (94 ) ( ) ( ) ( ) ( )
()

(b) 90 +((91) + (92) + (93) + (94))

(c) 29 − 90 + 91 + 92 + 93

7. (a) 2(44)
(b) ( 44
14
(c) 63 44
)( )
9
(12)(13)
8. (a) 3 2 (since S contains 12 even and 13 odd elements)
(b) 9i=1 10i 9i
∑ ( )( )

9. (a) (13
( )
4)
(b) 12
∑4 ( )( 8 )
(c) 3i=0 5i 5−i
(d) 10
( )
4
(38) ∑2 (10)( 28 )
10. (a) 6 − i=0 i 6−i
(b) ( 83 )(30
( )( )
3 )
(c) (10 20
4 )( 2)( )
(d) 28 5 15
(28)(4)(14)
2 2 2
− 2 1 1

11. P (26, 3)P (10, 3) = 26 · 25 · 24 · 10 · 9 · 8

12. omitted

13. 52
( )(47)(42)(37)(32)(27)
5 5 5 5 5 5

11 Mathematical Induction
1. For all n ∈ Z+ , define the proposition P (n) as
n(n + 1)(2n + 1)
12 + 22 + . . . + n2 = . (∗)
6
1(1+1)(2·1+1)
BI: to prove P (1). The LHS of (*) is 1 and the RHS is 6
= 1 Hence P (1)
holds.
IS: We must show P (k) → P (k + 1) for all k ∈ Z+ . Take any k ∈ Z+ and assume P (k)
holds; that is, 12 + 22 + . . . + k 2 = k(k+1)(2k+1)
6
. Now consider P (k + 1). The LHS is
(using IH)
k(k + 1)(2k + 1)
12 + 22 + . . . + k 2 + (k + 1)2 = + (k + 1)2
6
1
= (k + 1)(2k 2 + k + 6k + 6)
6
1
= (k + 1)(k + 2)(2(k + 1) + 1),
6
30
 which is the RHS of P (k + 1). Thus P (k) → P (k + 1) holds.
By Mathematical induction, since P (1) holds and P (k) → P (k+1) holds for all k ∈ Z+ ,
we conclude that P (n) holds for all n ∈ Z+ .

2. omitted

3. Let n = 2m − 1. We thus have to show that

P (n) : (2m − 1)2 − 1 is divisible by 8

for all m ∈ Z+ .
BI: to prove P (1). Now (2 · 1 − 1)2 − 1 = 0, which is divisible by 8. Hence P (1) holds.
IS: We must show P (k) → P (k + 1) for all k ∈ Z+ . Take any k ∈ Z+ and assume P (k)
holds; that is, (2k − 1)2 − 1 is divisible by 8. Consider P (k + 1).

(2(k + 1) − 1)2 − 1 = ((2k − 1) + 2)2 − 1 = (2k − 1)2 + 4(2k − 1) + 4 − 1

= ((2k − 1)2 − 1) + 8k

Since (2k−1)2 −1 is divisible by 8 by IH, and 8k is clearly divisible by 8, (2(k+1)−1)2 −1

is divisible by 8. Thus P (k) → P (k + 1) holds.
By Mathematical induction, since P (1) holds and P (k) → P (k+1) holds for all k ∈ Z+ ,
we conclude that P (m) holds for all m ∈ Z+ .

4. We thus have to show that

P (n) : 4n+1 + 52n−1 is divisible by 21

for all m ∈ Z+ .
BI: to prove P (1). Now 41+1 + 52·1−1 = 42 + 51 = 21, which is divisible by 21. Hence
P (1) holds.
IS: We must show P (k) → P (k + 1) for all k ∈ Z+ . Take any k ∈ Z+ and assume P (k)
holds; that is, 4k+1 + 52k−1 is divisible by 21. Consider P (k + 1).

4(k+1)+1 + 52(k+1)−1 = 4k+2 + 52k+1 = 4 · 4k+1 + 25 · 52k−1

= 4(4k+1 + 52k−1 ) + 21 · 52k−1

Since 4k+1 + 52k−1 is divisible by 21 by IH, and 21 · 52k−1 is clearly divisible by 21,
4(k+1)+1 + 52(k+1)−1 is divisible by 21. Thus P (k) → P (k + 1) holds.
By Mathematical induction, since P (1) holds and P (k) → P (k+1) holds for all k ∈ Z+ ,
we conclude that P (n) holds for all n ∈ Z+ .

5. It can be checked that 2n < n3 for n ∈ {1, 2, . . . , 9}. We shall prove

P (n) : 2n > n3
31
 for all n ∈ Z, n ≥ 10.
BI: to prove P (10). Now 210 = 1024 > 1000 = 103 . Hence P (10) holds.
IS: We must show P (k) → P (k + 1) for all k ∈ Z, k ≥ 10. Take any k ∈ Z, k ≥ 10 and
assume P (k) holds; that is, 2k > k 3 . Consider P (k + 1).

k3
( 3
k3 k3
)
k+1 k 3 3 3 3 3 k
2 = 2·2 >2·k =k +k =k +3· =k + + +
3 3 3 3
( 3 2
) ( 2 )
k k k k k
> k3 + + + > k 3 + 9 + 9 + 1 = (k + 1)3
3 3 3 3 3

Thus P (k) → P (k + 1) holds.

By Mathematical Induction, since P (1) holds and P (k) → P (k + 1) holds for all k ∈ Z,
k ≥ 10, we conclude that P (n) holds for all n ∈ Z, n ≥ 10.

6. We must prove ( √ )n−2

1+ 5
P (n) : fn >
2
for all n ∈ Z, n ≥ 3. We shall use Strong Induction.
( √ )3−2 √
1+ 5 1+ 5
BI: to prove P (3). Note f3 = f2 + f1 = f1 + f0 + f1 = 3 and 2
= 2
<
√
1+ 9
2
= 2. Hence P (3) holds.
IS: We must show (P (3) ∧ P (4) ∧ . . . ∧ P (k)) → P (k + 1) for all k ∈ Z, k ≥ 3. Take
( √ )i−2
any k ∈ Z, k ≥ 3 and assume P (3), P (4), . . . , P (k) all hold; that is, fi > 1+2 5 for
all i ∈ {3, 4, . . . , k}. Examine fk+1 . If k = 3, then fk+1 = f4 = f3 + f2 = 3 + 2 = 5 and
( √ )4−2 ( √ )2 √ √
1+ 5 1+ 5 3+ 5 3+ 9
2
= 2
= 2
< 2
= 3. Hence P (4) holds.
Otherwise, k ≥ 4 and by IH we obtain
( √ )k−2 ( √ )k−3
1+ 5 1+ 5
fk+1 = fk + fk−1 > +
2 2
( √ )k−3 ( √ ) ( √ )k−3 √ ( √ )k−1
1+ 5 1+ 5 1+ 5 3+ 5 1+ 5
= +1 = =
2 2 2 2 2

Thus (P (3) ∧ P (4) ∧ . . . ∧ P (k)) → P (k + 1) holds.

By Strong Induction, since P (3) holds and (P (3) ∧ P (4) ∧ . . . ∧ P (k)) → P (k + 1) holds
for all k ∈ Z, k ≥ 3, we conclude that P (n) holds for all n ∈ Z, n ≥ 3.

7. We shall prove
P (n) : n2 ≥ 2n + 3
for all n ∈ Z, n ≥ 3.
32
 BI: to prove P (3). Now 32 = 9 ≥ 2 · 3 + 3. Hence P (3) holds.
IS: We must show P (k) → P (k + 1) for all k ∈ Z, k ≥ 3. Take any k ∈ Z, k ≥ 3 and
assume P (k) holds; that is, k 2 ≥ 2k + 3. Consider P (k + 1). Using the IH we obtain

(k + 1)2 = k 2 + 2k + 1
≥ (2k + 3) + (2k + 1) = 4k + 4 = 2(k + 1) + 3 + (2k − 1) > 2(k + 1) + 3

Thus P (k) → P (k + 1) holds.

By Mathematical Induction, since P (1) holds and P (k) → P (k + 1) holds for all k ∈ Z,
k ≥ 3, we conclude that P (n) holds for all n ∈ Z, n ≥ 3.

8. We shall prove
P (n) : 7n − 2n is divisible by 5
for all n ∈ N.
BI: to prove P (0). Now 70 − 20 = 0, which is divisible by 5. Hence P (0) holds.
IS: We must show P (k) → P (k + 1) for all k ∈ N. Take any k ∈ N and assume P (k)
holds; that is, 7k − 2k is divisible by 5. Consider P (k + 1). Using the IH we obtain

7k+1 − 2k+1 = 7 · 7k − 2 · 2k = 7(7k − 2k ) + 5 · 2k

Since 7k − 2k is divisible by 5 by IH, and 5 · 2k is clearly divisible by 5, we conclude

that 7k+1 − 2k+1 is divisible by 5. Thus P (k) → P (k + 1) holds.
By Mathematical Induction, since P (0) holds and P (k) → P (k + 1) holds for all k ∈ N,
we conclude that P (n) holds for all n ∈ N.

9. omitted (this is very similar to Question 1).

10. We shall prove

P (n) : 1 · 1! + 2 · 2! + . . . + n · n! = (n + 1)! − 1
for all n ∈ Z+ .
BI: to prove P (1). Now 1 · 1! = 1 = (1 + 1)! − 1. Hence P (1) holds.
IS: We must show P (k) → P (k + 1) for all k ∈ Z+ . Take any k ∈ Z+ and assume P (k)
holds; that is, 1 · 1! + 2 · 2! + . . . + k · k! = (k + 1)! − 1. Consider P (k + 1). Using the
IH we obtain

1 · 1! + 2 · 2! + . . . + k · k! + (k + 1) · (k + 1)! = (k + 1)! − 1 + (k + 1) · (k + 1)!

= (1 + k + 1) · (k + 1)! − 1
= (k + 2)! − 1

Thus P (k) → P (k + 1) holds.

By Mathematical Induction, since P (1) holds and P (k) → P (k+1) holds for all k ∈ Z+ ,
we conclude that P (n) holds for all n ∈ Z+ .
33
11. We must prove
P (n) : an = 2n + (−1)n
for all n ∈ N. We shall use Strong Induction.
BI: to prove P (0). Note a0 = 2 = 20 + (−1)0 . Hence P (0) holds.
IS: We must show (P (0) ∧ P (1) ∧ . . . ∧ P (k)) → P (k + 1) for all k ∈ N. Take any k ∈ N
and assume P (0), P (1), . . . , P (k) all hold; that is, ai = 2i +(−1)i for all i ∈ {0, 1, . . . , k}.
Examine ak+1 . If k = 0, then ak+1 = a1 = 1 by definition, and 21 + (−1)1 = 1, so P (1)
holds.
Otherwise, k ≥ 1 and by IH we obtain

ak+1 = ak + 2ak−1 = 2k + (−1)k + 2(2k−1 + (−1)k−1 )

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

Thus (P (0) ∧ P (1) ∧ . . . ∧ P (k)) → P (k + 1) holds.

By Strong Induction, since P (0) holds and (P (0) ∧ P (1) ∧ . . . ∧ P (k)) → P (k + 1) holds
for all k ∈ N, we conclude that P (n) holds for all n ∈ N.

12. (1) a1 = a3 = a5 = a + 7 = 2, a2 = 4, a4 = 16, a6 = 4, a8 = 256

(2) We must prove
P (n) : an ≤ 2n
for all n ∈ Z+ . We shall use Strong Induction.
BI: to prove P (1). Clearly a1 = 2 ≤ 21 . Hence P (1) holds.
IS: We must show (P (1)∧P (2)∧. . .∧P (k)) → P (k+1) for all k ∈ Z+ . Take any k ∈ Z+
and assume P (1), P (2), . . . , P (k) all hold; that is, ai ≤ 2i for all i ∈ {1, 2, . . . , k}.
Examine ak+1 . If k is even, then ak+1 = 2 ≤ 2k+1 and so P (k + 1) holds.
Otherwise, k is odd and k ≥ 1, and by IH we obtain
( k+1 )2
ak+1 = a2k+1 ≤ 2 2 = 2k+1
2

Thus (P (1) ∧ P (2) ∧ . . . ∧ P (k)) → P (k + 1) holds.

By Strong Induction, since P (1) holds and (P (1) ∧ P (2) ∧ . . . ∧ P (k)) → P (k + 1) holds
for all k ∈ Z+ , we conclude that P (n) holds for all n ∈ Z+ .

13. omitted

14. omitted

34
12 Graphs
1. (a) Does not exist, as every graph has an even number of vertices of odd degree.
(b) Does not exist. Such a graph would have a vertex of degree 5, but there are only
4 other vertices it can be adjacent to.
(c) Exists: this is a star on 5 vertices.
(d) Does not exist: if u and v are 2 vertices of degree 3, then each must be adjacent
to the remaining two vertices. Hence there can not be a vertex of degree 1.
(e) Exists.
(f) Does not exist: if u and v are 2 vertices of degree 6, then each must be adjacent to
the remaining five vertices. Hence there can not be a vertex of degree 1.
(g) Exists: C3 .
(h) Does not exist: the maximum degree a simple graph with 6 vertices can have is 5.

2. (i) n2 = n(n−1)
( )
2
. The complete graph Kn achieves this upper bound.
(ii) 12 (2 + 2 + 3 + 3 + 4) = 7.
(iii) |E| = 21 v∈V deg(v) ≥ 12 v∈V 3 = 32 n.
∑ ∑

3. Let G be a simple graph with n ≥ 2 vertices. The possible degrees for a vertex in G
are 0, 1, 2, . . . , n − 1. However, if there is a vertex of degree n − 1, there can not be a
vertex of degree 0. Hence there are at most n − 1 possible degrees. By the Pigeonhole
Principle, at least two of the n vertices must hence have the same degree.

4. (1) |V (G)| = 94 = 126.

()

(2) Let A be a vertex in G. To construct a subset B of S such that |A ∩ B| = 1, we

choose(1)vertex from A for the intersection, and then 3 out of S − A for B − A. There
are 4 · 53 = 40 such subsets. Hence A as a vertex in G has degree 40, and G is regular
of degree 40. By the Handshaking Theorem, |E| = 12 v∈V deg(v) = 21 126 · 40 = 2520.
∑

5. For all n ∈ Z+ , define a proposition P (n) : “Kn has n(n−1)

2
edges.”
1(1−1)
BI: n = 1. Clearly K1 has 0 edges, and 2 = 0. Hence BI holds.
IS: We must show P (k) → P (k + 1) for all k ∈ Z+ . Take any k ∈ Z+ and assume
P (k) holds; that is, Kk has k(k−1)
2
edges. Now consider Kk+1 . Take any vertex in
Kk+1 . Removing this vertex and all edges incident with it (observe there are k of
them) we obtain a graph H isomorphic to Kk . By IH, H has k(k−1) 2
edges. Hence G
k(k−1) (k+1)k
has 2 + k = 2 edges, and P (k + 1) holds. Thus P (k) → P (k + 1) holds.
By Mathematical induction, since P (1) holds and P (k) → P (k+1) holds for all k ∈ Z+ ,
we conclude that P (n) holds for all n ∈ Z+ .

6. (1) Let G = (V, E) be a graph with n vertices that is regular of odd degree k. We have
|E| = 12 v∈V deg(v) = 12 v∈V k = 21 nk. Since k is odd, and |E| is an integer, n must
∑ ∑

be even.
(2) Since n must be even from (1), n = 2m for an integer m. Then |E| = mk as seen
above, and E is a multiple of k.

35
 7. omitted

8. omitted

9. Isomorphic with the indicated isomorphism.

0
0

4 5 5
1 4 9 1
6 6
9
8
3 8 7
7
2 3 2

10. The first and third graph are isomorphic (with the indicated isomorphism), and they
are bipartite graphs. The second graph is not bipartite (it contains C3 as a subgraph)
and hence not isomorphic to the other two.
1
1 3 5
6
2

5 3
4 2 4 6

11. The first and third graph are isomorphic, and they are bipartite graphs. The second
graph is not bipartite (it contains C5 as a subgraph) and hence not isomorphic to the
other two.
1 2 1 7
5
6 6 5
8 2
8
7
4 3 4 3

12. Bipartite. Not bipartite (it contains C5 as a subgraph). Bipartite. Not bipartite (it
contains C3 as a subgraph).

13. (a) No as it has vertices of odd degree.

(b) Yes since it has exactly two vertices of odd degree, namely, u and v. An open
Euler trail has to start at u and end at v or vice-versa. Example of an open Euler
tour: uacdabcvdubv.
36
 (c) In G + uv, all vertices have even degree. Hence, G + uv has an Euler tour, but
no open Euler trail. Example of an Euler tour: uacdabcvdubvu.

14. Consider the graph G whose vertices are the five quarters and he edges are the fourteen
bridges.

(a) Here we are looking for an Euler tour in G. Since G has vertices of odd degree
(A and B), it has no Euler tour.
(b) Yes, an open Euler tour exists since G has exactly two vertices of odd degree, A
and B. An open Euler tour must have these two vertices and endpoints. One
possible open Euler tour: b2 b1 b5 b6 b7 b8 b12 b4 b3 b14 b13 b10 b9 b11 .

15. omitted

16. omitted

17. (a) Let n and e be the number of vertices and edges, respectively, in a full 5-ary tree T
with 101 leaves. Then n = 5i + 1 = i + 101, where i is the number of internal vertices.
Solving for i we obtain i = 25. Now e = n − 1 = (i + 101) − 1 = 125.
(b) Since in a full m-ary tree of height h the number of leaves is at most mh , we have
51 ≤ m3 . That gives m ≥ 4. On the other hand, n = ℓ + i = mi + 1 (where n is
the number of vertices, ℓ the number of leaves, an di the number of internal vertices).
From here we obtain ℓ − 1 = (m − 1)i. In our case, 50 = (m − 1)i or 2 · 52 = (m − 1)i.
Since m − 1 is an integer, m − 1 ∈ {1, 2, 5, 10, 25, 50}. Since m ≥ 4 from above, we
have m ∈ {6, 11, 26, 51}. It can be verified that m = 51 and m = 26 can not give a
tree of height 3, while trees with m = 6 and m = 11 can be easily constructed.

37