GROUPS

Definition:
A non-empty set G with a binary operation  on G is called a group if the
following axioms hold.
(i)  is associative thus, a(bc) = (ab)c.

(ii) There exists e G s.t. ea = a  a G.

(iii) For every a G  a-1 G s/t a-1a = e = (a-1)-1 a-1

(iv) If the group also satisfies.

For all a,b G, a.b = b.a, then the group is called an abelian
(or commutative).

Eamples: I
Let G be the set of integers with the operations of addition. The sum of two
integers is a unique integer and the associative law also holds. The identity
element is 0 (zero), and the inverse of an integer a is an integer – a. (Z). This is
called the additive group of the integers represented by (Z, + ).
(ii)(a) The set of all rational numbers Q with the operation of addition is an
additive group of the rational numbers, (Q, + ).
(b) The set of all rational numbers with the operation of
multiplication is a group (Q, ·). Similarly, the set of all real numbers with the
operation of multiplication is a group. (R, ).
(iii) Integers modulo n under addition is a group where n > 1.

1
Example: The group of integers modulo 6 with an operation of addition is a
group:
 0 1 2 3 4 5
0 0 1 2 3 4 5 

1 1 2 3 4 5 0
 
2 2 3 4 5 0 1
3 3 4 5 0 1 2
 
4 4 5 0 1 2 3
5 4 
 5 0 1 2 3

ie. 3 + 3 = 0, 3 + 5 = 2, 2 + 2 = 4 etc
.
(iv) For each positive integer n, let GL(n, k) be the set of all nonsingular
(invertible) n x n matrices over K, where K is a field of all complex
numbers or the field of all real numbers. Then for each pair
A, B, Є GL(n, k, .) is a group.

Proof:
A Є GL(n, k) iff det A  0. If A, B Є GL(n,k), then det A.B= det A det B0 and
so
A.B Є GL(n, k). Thus multiplication is a binary operation on GL(n, k).
(a) When A,B,C Є GL(n, k), then A.(B.C) = (A.B).C.
(b) Let I be the identity n x n matrix, then IA = A  A  GL(n, k).
(c) For each AЄ GL(n, k), A-1 Є GL(n,k) and A-1 A = A A-1 = I.
But A.B  B.A, hence GL(n,k) is not abelian group.

2
Definition:
A group (G, ) is said to be abelian if for all a, b  G, a.b = b.a.

Example I:
If R is the Set of all real numbers, then (R, +) is an abelian group.

Proof:
For every pair a, b ЄR, a+b Є R.
(i) x + (y +z) = (x + y) + z. where x, y, z ЄR.
(ii) 0 Є R and 0 + x = x  x R.
(iii) For each a R, - a R and - a + a = 0.
(v) x + y = y + x for every pair x, y ЄR.

Example II:
If R* = R – {0} then (Rx, .) is a group, then (Rx, .) is an abelian group.

Proof:
If x, y  Rx, then x. y  Rx therefore . is a binary operation on Rx.
(i) If x, y, z  R then x.(y.z) = (x.y).z Rx.
(ii) 1  Rx and 1.x = x  X Є Rx.
(iii) For each x Rx,  x-1  Rx s.t. x-1 x=1
(iv) For each pair x.y ЄRx, x.y = y.x ЄRx.

3
Example III: (a)
If G is a group such that (ab)n = anbn for three consecutive integers, then ab=ba.

Proof: We have (i) (ab)n = anbn, (iii) (ab)n+1 = an+1 bn+1 and (ab)n+2 = an+2bn+2.
From (i) and (ii) we get (anbn)(ab) = (ab)n+1 = an+1 bn+1. By the appropriate,
cancellations. This implies bn a = abn. Similarly, from (ii) and (iii) we get bn+1a =
abn+1
But bn+1 a = b(bn)a = b(abn). So abn+1 = babn. This gives ab = ba.
(b) Let G be a group such that (ab)2 = (ba)2,  a, b ЄG. Suppose G also has
the property that x2 = e, x ЄG  x = e. then G is abelian.

Proof:
Let a, b ЄG then, by hypothesis, (ab-1 b)2 = (bab-1)2, that is a2 = ba2b-1
or a2b = ba2. So in other words, the square of every element in G commutes
with every element in G.
Set C = aba-1b-1 and so
C2 = ab(a-1 b-1a) (ba-1 b-1) = ab(aa-2 b-1 a) (ba-1 b-1)
= ab(ab-1 a-2 a) ( ba-1 b-1) =ab(ab-1 a-1)(ba-1 b-1)
= ab(abb-2 a-1)(ba-1 b-1) = (ab)(aba-1 b-2)(ba-1b-1)
= (ab)(aba-1)(b-1 a-1 b-1) = (ab)2(a-1 b-1)2
= (ba)2(a-1b-1)2
= e
 C = e and hence ab = ba.

4
Properties of Groups
Theorem I:
If G is a group and a, b, c are elements of G, then (i) ab = ac  b = c and
(ii) ba = ca  b = c

Proof:
(i) Suppose ab = ac
Then a-1(ab) = a-1(ac)
(a-1 a)b = (a-1 a)c by the associative law.
eb = ec
 b = c
(ii) Suppose ba = ca
Then (ba)a-1 = (ca)a-1
b(aa-1) = c(aa-1) by associativity.
be = ce
 b =c
(iii) A left identity is also a right identity
Thus ae = a

Proof:
a-1(ae) = (a-1a) e
= e.e
= e
= a-1 a

5
hence by left cancellation, ae = a. (For this reason a left identity is called an
identity).
(iv) Uniqueness of e
If e, I,  G and ea = a  a  G and
Ia = a  a G then I = e.

Proof:
I = e I
= I e by (iii)
= e
(iv) A left inverse of an element is also a right inverse of it.
For each a G, a.a-1 = e.

Proof:
a-1 (a.a-1) = (a-1, a) a-1
= e.a-1
= a-1
= a-1.e  a.a-1 = e
Theorem 2:
If G is a group and a, b are elements of G, then ab = e  a = b-1 and b = a-1

Proof:
If ab = e, then ab = aa-1, so by cancellation law b = a-1
Similarly, if ab= e, then ab = bb-1  a = b-1.

6
Theorem: 3
If G is a group and a, b are elements of G, then:
(i) (ab)-1 = b-1 a-1 and
(ii) (a-1)-1 = a

Proof:
(i) (ab)(b-1 a-1) = a (bb-1) a-1
= a(ea-1)
= aa-1
= e
Since the product of ab and (b-1 a-1) is the identity element e, it follows by
theorem 2. that one is the inverse of the other. Thus (ab)-1 = b-1 a-1

Proof : (ii)
aa-1 = e so by theorem 2, a is the inverse of a-1, that is a = (a-1)-1.

Uniqueness of the Identity:

Let  e G. If  g  G such that  g = g then.
 = e
Proof:
Choose gG s.t. I.g. = g
Then  = e =  (gg-1).
= (.g) g-1
= g.g-1

7
 = e
Questions:
Let G be a group s. t a2 = e  aG show that G is abelian.

Solution:
For all a, b G,
ab = aeb = a(ab)2b = aababb = ebae = ba
Hence G is abelian.
(2) Prove that the matrices :

1 0 1 0  1 0  1 0
      and   form a.
0 1 0  1 0 1 0  1

multiplicative group. (This group is known as the Klein four-group).

Solution:
1 0 1 0  1 0  1 0
Let e =   a    b    c   
0 1 0  1 0 1  0  1

1 0  1 0
a2 =     = b2 = c2 = e
0  1 0  1

We have also that ab = c = ba, bc = a = cb, and ca = b = ac.

Hence, G = {e, a, b, c} is closed under multiplication, e is the identity element,
and each element is its own inverse, because matric multiplication is associative,
G is a group.

8
Example:
i 0 0 1
If A =    GL (2, C) and B =   . Show that C = A4 = I
0 1  1 0 

(ii) B2 = 
(iii) BA = A3B. Deduce that 8 = {A, A2, A3 A4, B, AB, A2B, A3B} is
a group under multiplication. 8 is called the dihedral group of order 8.

Functions:
If A and B are Sets, then the function from A to B is a rule which is to every
element X in A assigns a unique element Y in B. We usually write f: A  B as
f(x) = y and we call y the image of x under the function f.

Example:
Given Sets S, X, Y and mappings g: S  X and f: X Y define.
Fog: S Y by fog(a) = f{g(a)}.
Then fog is a mapping, and fog is called a composite.
Let f(x) = 2x g(x) = x + 1 gof = g(f(x)
fog = f(g(x) = 2(x + 1) = 2x + 1

Definition: (I)
Let X and Y be sets and f: X  Y , a mapping.
(ii) f is said to be injective (one – one) if for a, b X ,f(a) = f(b)  a = b.

9
 In order words, a function f: X  Y is called injective if each element of Y
is the image of not more than one element of X.
X f Y
a w
eg b x
c y
z

Definition: (ii)
A function f: X  Y is called surjective (onto) if each element of Y is the image
of at least one element of x. If x  X and y  Y then f(x) = y.

X f
eg a x
b
c y

Definition: (iii)
A function f: x  y is called bijective if it is both injective and surjective.
e.g f
a x
b y
c z

10
Subgroups

Definition:
Let (G, ) be a group, and let H be a subset of G, H is called a subgroup of G,
written H<G, if H is a group relative to the binary operation in G.

Example:
(a) The set of all the even integers is a subgroup of the additive group Z of the
integers. Eg. The sum of any two even integers is an even integer, and the
negative of any even integer is an even integer.
(b) Q* (the group of nonzero rational number under multiplication) is a subgroup
of R* (the group of non-zero real numbers under multiplication.

Note: If S is a subgroup of G, the operation of S is the same as the operation of

G.

Theorem:
Let G be a group. A nonempty subset H of G is a subgroup of G if and only
if either of the following holds.
(i) For all a,b  H, ab  H and a-1 єH.
(ii) For all a,b  H, ab-1  H

11
Proof:
If H is a subgroup, (i) and (ii) are obviously true. Conversely, suppose H
satisfies (i). Then for any a H, a-1  H. Hence, e = aa-1  H. Therefore, H is a
subgroup. Next, suppose that H satisfies (ii).
Let a, b  H, Then e = bb-1  H. Hence b-1 = eb-1  H. Therefore a(b-1)-1  H.
Hence H is a subgroup of G.
Example 1
a. Let S  {0,2,4}  Z6 under addition mod 6 is a subgroup. Exercise.

If H is a subgroup of G then,
1. H must be closed under the operation * of G.
2. If
3. For any group (G,) the set {e} is a subgroup. It is the trivial subgroup.

Example 2.

b. The set is a subgroup of the multiplicative group . Exercise.

COSETS
Definition:
Let H be a subgroup of G. Given a єG, the set aH = {ah/h  H} is called the left
coset of H determined by a.
Similarly the set Ha = {ha/h  H} is the right coset of H. In the two cases, the set
of all products in either ah or ha, a remains fixed and h ranges over H.

Theorem:
If a  Hb, then Ha = Hb.

12
Proof:
Since a  Hb  a = h, b for some h,  H. We are to prove that Ha = Hb.
Let x  Ha which means that x = h2a for some h2  H.
But a = h b  x = h2a = (h2h)b  Hb which shows that every x  Ha is in Hb.
Similarly, Let y  Hb which means that y = h2b
Since b  ha  b = h1a for some h1  H. b = h1a  y = h2b = (h1h2)a  Ha
 Ha = Hb

Theorem:
The family of all cosets Ha, as a ranges over G is a partition of G

Proof:
Ha
G He Hb
Hc

First we are to show that any two cosets Ha and Hb are either disjoint or equal.
If they are disjoint, we are done.
Suppose Ha = Hb.
Let X  Hb  Ha. Because X ha, X = h1 a for some h1  H.
Because X Hb, X = h2b for some h2 H.
Thus h a = h2b and solving for a, we have a = (h1-1 h2)b,  a  Hb but we know
that If a  Hb, then Ha = Hb.
Next we are to show that every element c  G is in one of the cosets of H.
This is obvious; because c = ec and e  H.

13
 c = ec Hc.
Thus, the family of all cosets of H is a partition of G.

Definition
If there exists a nonzero integer n such that an = e, then the order of the elements a
is defined to be the least positive integer n such that an = e.
If there does not exist any nonzero integer n such that an = e. we say that a has
order infinity

Example:
In S3 all the permutations of the set (1, 2, 3) are
1 2 3 1 2 3 1 2 3 ( )(1)   (1)
 =          
1 2 3 1 3 2 3 1 2    (3)  2

( )(2)   (2)

1 2 3 1 2 3 1 2 3
 =       K      (1)  1
2 1 3 2 3 1 3 2 1 ( )(3)   (3)
  (2)  3

The table of the group S3 is as follow.

o      k
 O     k
     K 
  k    
   k   
    k  
k K     

The order of  = 2 because 2 = . The order of  = 3 because 3 = .

14
Find the orders of the other elements.
In Z6, the number 2 has order 3 because 2 + 2 + 2 = 0 which is a multiple 2.
Fine the order of other numbers in Z6. If the order of a is n, there are exactly n
different powers of a namely; a0, a1, a2, a3, … an-1
If the order of a is n , then where n is the smallest positive integer which
satisfies this equation.
Lemma:
Every power of a is equal to one of the powers ,
Proof
Let be any power of a. Using division algorithm we have,

, Then .
Thus where r is one of the integers 0,1,2,………, . Next we prove that
are different . Assume they are not and where r and s are
distinct integers between 0 and n-1. We have either
and consequently
But and

This is not possible because r-s is a positive integer less than n, whereas n (the
order of a) is the smallest positive integer such that . Hence

Hence are distinct.

Poblems:
1 What is the order of 10 in

15
 2 What is the order of
3 What is the order of 9 in
4 What is the order of 8 in ?

Theorem:
If G is a group and H is a subgroup of G, then,

(i) G = (ii) for every pair a, b  G, either Ha  Hb =  or Ha = Hb.

(iii) For x, y  G, Hx = Hy iff, xy-1  H.
Proof:
For each g G, Hg  G and so U Hg  G. Let e be the identity in G, then e єH
g G

and so for each X  G, x = ex  Hx, That implies G CU Hg. Hence G = UHg .

gG gG

(ii) Let a, b  G. If Ha  Hb   choose c  Ha  Hb.

Next choose h1, h2  H s.t. c = h1a and c1 = h2b.
Then h1 a = h2b.  G = (h1-1 h2)b and b = (h2-1 h1)a
And so for every h  H, ha = (hh1-1h2)b  Hb and Hb = (hh2-1h1)a  Ha.
Thus Ha  Hb and Hb  Ha  Ha = Hb.
If Ha  Hb  . Hence either Ha  Hb =  or Ha = Hb.
(iii) Let x, y  G. If Hx = Hy then x = xe  Hx = Hy and so x = Hy
where h  H  xy-1 = h. Conversely if xy-1  H, let xy –1 = h.
Then x = hy  Hy and x = ex  Hx.
Therefore x  Hx n Hy  Hx = Hy by (ii).

16
Definition: Cyclic groups:

A multiplicative group G is said to be cyclic if there is an element a  G such that

for any b єG, there is some integer (j) with b = ai. Such an element a is called a
generator of the cyclic group, and we write G =  a  . From the definition, every
cyclic group is commutative.
If G =  a  is the cyclic group generated by a, and a has order n, then G is a
cyclic group of order n. If the generator of G has order infinity, then G is a cyclic
group of order infinity. A multiplicative cyclic group is simply the set of all powers

of one of its elements. Eg G= If (G,+) is an additive cyclic

group, then . The element a is called the generator of G.

Example:
(i) The cyclic group of Z consisting of all the multiples of 1 (in additive
notation, we speak of multiples instead of powers). Z is a cyclic group of
order infinity generated by 1.
(ii) z6 is also a cyclic group consisting of all the multiples of 1. z6 is a cyclic
group of order 6.1 is the generator of z6. But z6 has another generator. Find
it out Exx.
(iii) Z8 has four generators

3x1=3
3x2=6 mod 8

17
3x3=9=1mod8
3x4=12=4mod8
3x5=15=7mod8
3x6=18=2mod8
3x7=21=5mod8
3x8=24=0mod8
So the set of multiples of 3 taken mod8, is the set {3,6,1,4,7,2,5,0}=
Theorem
Every cyclic group is abelian.
Proof
Let G be a cyclic group and let be a generator of G so that
If g1 and g2 are any two elements of G, there exists
integers r and s such that

. So G is abelian.
Example:
The multiplicative group of units in the ring of
They are units because they have inverse.
The cyclic subgroup generated by 7.
.Therefore the element 7 has order 4
in . Thus the cyclic subgroup

18
Theorem: Every cyclic group is isomorphic to Z or to z/(n) for some nN.

Proof:
If G =  a  is an infinite cyclic group, consider the mapping :Z  G Given
by (i) = ai. It is clear that  is a subjective homomorphism.
Moreover, i  j  ai, = aj for otherwise ‘a’ would be of finite order.
Hence,  is injective. Therefore,  is an isomorphism.
Next, suppose G =  a  is a cyclic group of finite order n.
Then G = {e, a, - -, an-1 } and o(a) = a. Consider the mapping   : z/(n) 
G given by (i) = ai.  is well defined and also injective; for let i, i  Z/(n).
Then  = J  n/I - j  aI – j = e  aI = aj. Clearly,  is surjective.
Further:  (i  j )   (i  j) = ai+j = aiaj. Hence,  is an isomrophism

Theorem: Every subgroup of a cyclic group is cyclic.

Proof:
Let G be a cycle group of generator a and let H be a subgroup of G
G is the cycle group generated by a, and H is a subgroup of G Implies every
element of H is therefore in G. which means that every element of H is some
power of a. Hence the generator of H has some powers of a. let m be the smallest
positive integer such that am H. Let at be any element of H. Then by division
algorithm (t/m)
t = mq + r o  r < m.
Then at = amq+r = amq ar. making ar the subgroup  ar = (amq)-1 at = (am)-q at
But am H and at H Thus (am)-q at H.

19
 ar H. But r < m and m is the smallest positive integer s.t am H  r = 0 and 
t = mq.
Hence every element at H is of the form at = (am)q that is a power of am. Hence H
is a cycle group generated by am

Example I
Let H =  a  and K=  b  be cyclic groups of order m and n, respectively.
Such that (m,n) = 1. Show that HxK is a cyclic group of order mn.

Solution:
Let 0(a, b) = d. Now (a,b)mn = (amn, bmn) = (e, e)  d1mn. Also (e,e) = (a,b)d = (ad,
bd)  ad = e = bd. So m/d and n/d. Therefore, mn/d. Consequently, mn = d.
Since ||H x K|| = mn, it follows that (a,b) generates the group H x K.

Definition: Finite groups

A group G is called finite if it contains finitely many elements. The number of
elements in a finite group is called its order.eg. If G contains exactly n elements,
, then the order of G is n.
Lemma
Let H be a subgroup of a group G, and let then
1. In particular,
2. If
3.

20
Lagrange:
Theorem: Let G be a finite group, and H any subgroup of G. The order of G is a
multiple of the order of H. Thus the order of H divides the order of G

Proof:
Let be the family of all the distinct cosets of G. The
cosets of H partition G because they are equivalent classes. Hence G is the disjoint
union. . It follows that
for all I by the above lemma.
Hence

Example:
If G has 15 elements, 15 proper subgroups may have either 3 or 5 elements. If G
has 7 elements,. It has no proper subgroups, thus 7 has no factors except 1 and 7.
Let G be a group with prime number P of elements m. If a  G where ae, then the
order of a is some integer m1. But the cyclic group a has m elements. By
Lagrange’s theorem, must be a factor of P. But P is a prime number, and therefore
m = P. It follows that  a  has P elements, and is therefore all of G.
Corollary: let G be a finite group of order n. Then for every a  G, o(a)/n and
hence, an = e.
Consequently, every finite group of prime order is cyclic and hence abelian.

Proof:
Let a єG. By Lagrange’s theorem the order of the cyclic group  a  divide n. so
o(a)/n. If n is prime and ae, the order of  a  must be n. Hence a = G, and
therefore G is cyclic.

21
Homomorphism
Definition: Let G, H be groups. A mapping : G  H is called a
homomorphism if for all x, y  G. : (xy) = (x) (y).
Furthermore, H  is bijective, then  is called an isomorphism of G onto H, and we
write G  H. If  is just injective, that is 1 – 1, then we say that  is a
homomorphism of G into H. If  is surjective, that is, onto, then  is called an
epimorphism. A homomorphism of G into itself is called an endomorphism of G.
An endomorphism of G that is both 1–1 and onto is called an automorphism of G.

Example 1:
Let G be a group. For a given a єG consider the mapping a : G  G given by
a(x) = a x a-1  x G, then the mapping is an isomorphism of G.

Proof:
Since a(xy) = axya-1 = (axa-1)(aya-1) = a (x) a(y). a is a homomorphism.
By cancellation laws, axa-1 = aya-1  x=y. Hence, a is injective. For any x in G,
x = a(a-1 x a)a-1. Hence, a is surjective. Consequently, Ia is an automorphism of
G.
Lemma:
Let
i. f(1)=1
ii.
iii for all

22
Theorem
Let be homomorphism of groups.
i. kerf is a subgroup of G and imf is a subgroup of H
ii. If
iii. f is injective if and only if
Proof
I. The above lemma shows that 1 . We know that kerf
and imf are both nonempty. Let a,b, , hence f(a)=1=f(b), then
f(ab)=f(a)f(b)=1x1=1, and so .Finally, if

therefore kerf is a subgroup of G. We now show that imf is a subgroup

of H. Let Then

Hence, imf is a subgroup of H.

II. If

III. If f is injective, then

Conversely, assume that ker f
=
and
f is injective.

23
(2) Given a group (G, 0) defined a binary operation  on G by x  y = y o x
for every pair x, y,  G. Then
(i) (G, ) is also a group.
(ii) The mapping h: G  G defined by h(x) = x-1 is an isomorphism of (G, 0) unto
(G, ).

Proof:
(i) If a, b, C G, the a  (b C) = a  (c o b) a  (C0b)
= c (b o a)
= C o (a  b)
= (a  b) C
Let e be the identity in (G, o). Then for every a G e  a = aoe = a. Hence e is
a left identity in (G, ). Finally if aG then a-1  a = a0a-1 = e. and so a-1 is also
the inverse of a(G, ). Thus (G, ) is also a group.
(ii) For every pair x, y G, h(xoy) = (xoy)-1 = y-1 0 x-1
y -1 0 x-1 =
= x-1  y-1
= h(x)  h(y)
 h is a homomorphism.
If a, b  G and h(a) = h(b) then a-1 = b-1 taking inverses  a = b. Thus h is
injective. If g G, then g-1 G and h(g-1) = (g-1)-1 = g.
 h is surjective.

Theorem:
Let G and H be groups with identities e and e/ respectively, and let : G  H be a
homomorphism. Then (i) (e) = e (ii) (x-1) = (x)-1, x G

24
Proof (I):
ø (e) ø (e) = ø ee) = ø (e) = é ø (e). Hence by cancellation law, ø (e) = é
(i) ø (x)(x-1) = ø (xx-1) = ø (e) = é. Hence, ø (x-1) = (x)-1

Definition: Let G and H be groups, and let Ø : G  H be a homomorphism. The

kernel of Ø is defined to be the Set..
Ker Ø = {x  G/ Ø (x) = é} where é is the identity in H. because Ø (e) = é, it
follows that Ker Ø is not empty.

Theorem:
A homomorphism Ø : G  H is injective  Ker Ø = { e }.

Proof:
Suppose Ø is injective, and let x  Ker Ø . then Ø (x) = é = Ø (e). Hence x = e.
Therefore Ker Ø = {e}. Conversely, suppose Ker Ø = {e}.
Then Ø(x) = Ø (y)  Ø (xy-1) = Ø(x) (y)-1 = é
 xy-1  Ker Ø  xy-1 = e  X = y .
Hence Ø is injective.

Theorem:
Let G be a group with the identity e and L a group. Then these two statements
on G homomorphism h : G  L are equivalent.
(i) h is a homomorphism
(ii) Ker h = e

25
Proof:
Suppose (I) is true. If u  Ker h then h(u) =  the identity in L but  =h(e)  u =
e.
Thus Kerh c{e}.
 Ker h = {e } since we know that e Ker h.
This proves 1 2 .
Next suppose 2 is true.
If a, b G and h(a) = h(b), then
h(ab-1) = h(a) h(b-1)
= h(a) {h(b)}-1
= h(a){h(a)}-1
=  the identity in G.
 ab-1  Ker h = e  ab-1 = e  a = b.
Thus h is injective and so h is a monomorphism
Thus proves 2  1

Theorem:
If G and L are groups. h: G  L a homomorphism and S is a subgroup of G,
then h(s) is a subgroup of L.

Proof:
Let e be the identity in G and  the identity in L. Then  = h(e) and so   h(s).
Since e  S if y  h(s), choose x S s.t. h(x) = y . Then y–1 = h(x-1)  h(s) since
X-1 S.
Finally, if u, v h(s), choose a, bS s.t. h(a) = u and h(b) = v. then ab  S and
UV = h(a) h(b) = h(ab)  h(S). Hence h(S) is a subgroup of L.

26
Theorem:
Let G and B be finite groups and h: G  B an isomorphism. If y єG then the
order of y is the same as the order of h(y).

Proof:
Let m be the order of y and n be the order of h(y). Then {h(ym) } = h( ym) =
h(e) where e is the identity in G. h{e) =  the identity in B.
 n divides m.
Also Xn = {h-1 (h(x))}n = h-1 {(h(x)n} = h-1( () = e.
 m also divides n  m=n. [ m = na, n = mb then m= mba = m(ba -1) =
0.  ba =1 hence b = 1 = a hence m = n].

Definition: Permutation groups

Let X be a nonempty set. The group of all permutation of X under composition
of mappings is called the symmetric group in X and is denoted by Sx . A
subgroup of Sx is called a permutation group on X. It is easily seen that a bijection
X  Y induces in a natural way an isomorphism Sx = Sy. If |X| = n, Sx is denoted
by Sn and called the symmetric group of degree n.
A permutation  sn can be exhibited in the form
 1 2 3.......... ..........n 
  consisting of two rows of integers.
  (1)  (2)  (3)................ (n) 

The Set {1, 2, 3, 4, 5 } can be re-arranged to form the set {3, 2, 1, 5, 4} s.t we
have 1 2 3 4 5
     
3 2 1 5 4

27
Example: In the symmetric group of S3, we have the following permutations of
the Set {1, 2, 3}.

1 2 3 1 2 3 1 2 3 1 2 3
 =              
1 2 3 1 3 2 3 1 2 2 1 3

1 2 3 1 2 3
       
2 3 1 3 2 1

1 2 3
For example    is the function s.t
3 2 1 

1 2 3
(1) = 3, (2) = 1, (3) = 2 or B =  
3 1 2 

Considering the above permutation of S3, we see that S2 is a composition. To

find o, we have
(o)(1) = (1) = (3) = 2
(o)(2) = (2) = (1) = 1
(o)(b) = (3) = (2) = 3.

1 2 3
Thus o =   = 
2 1 3

1 2 3
Thus o =  
2 1 3 

1 2 3
Thus o =   =
2 1 3

28
Complete the following table.
      







In the S3 permutation we can examine that: ε = (1)  = (1, 2, 3)  = (1 3 2)

 = (2 3)  = (3 1)  = (1 2). Which are all cycles.
1 2 3
We see that   =   
3 2 1 

Let  = (1 2 3) and  = (2 3). We can easily verity that

2 = , 3 = ε, 2 = ε,  = , 2 =  = 
Hence S3 = {ε, , 2 =  , , , 2} and the product of any two elements
in S3 can be computed by using the relations:
3 = ε = 2,  = 2  and the associative property of multiplication.

Theorem: (cayley)
Every group is isomorphic to a permutation group.

Proof: Let G be a group. For any given aG, the mapping fa : G  G, given by
fa (x) = ax  x  G, is a bijection, because ax = ax1
 x = x1 and y = fa(a-1 y)  x, x1 y G Consider the mapping.

29
Ø : G  Sa given by Ø(a)= f(a)  aG.
where Sa is the symmetric group on the set G. For all a, b, x G, fab(x) = abx = fa
(bx) = fa(fb (x) = (fafb)(x).
Hence Ø(a,b) = Ø(a) Ø(b)
Therefore, Ø is a homomorphism, and Im Ø is a subgroup of Sa.
Moreover, Ø(a) = Ø(b)  ax = bx  xG  a = b.
Hence, Ø is an injective homomorphism.
Therefore, G is isomorphic to a subgroup of sa.

Groups of symmetries:
Let X be a Set of points in space, so that the distance d(x, y) between points x,y,
is given for all x, y  X. A permutation of X is called a symmetry of X if
d( 6 (x), 6 (y)) = d(x,y)  x, y  X. A symmetry therefore preserves distance
between every two points.
Let Tx denote the set of all symmetries of X.
Then for all 6,   Tx and xyX, show that Tx is a subgroup of Sx..
Let 6,   Tx and  x, y  X,
d( 6-1(x), 6-1(y)) = d(6-1 (x), 6-1(y)
= d(6-1 (x), 66-1 (y))
= d(x, y).
1
Hence,  6  Tx. Hence Tx is a subgroup of Sx and therefore is itself a group
under composition of mapping. It is called the group 7 symmetric of X.

Definition:
The group of symmetries of a regular polygon Pn of n sides is called the dihedral
group of degree n and denoted by Dn.

30
The eight symmetries of a square form a group under the operation  of
composition, called the group of symmetries of the square.

Eg. 1 2
.p Rotation through 90, 180, 270 clockwise.
4 3
1 2 3 4 1 2 3 4
90 = R =   180, R2 =   2700 = R3 =
2 3 4 1  3 4 1 2

1 2 3 4
 
4 1 2 3

1 2 3 4
A C B R4 =  
3 4 1 2 

1 2
1 2 3 4
D R5 =  
3 2 1 4 

4 3
1 2 3 4
R6 =  
2 1 4 3 

1 2 3 4
R7 =  
4 3 2 1 

1 2 3 4
Ro =  
1 2 3 4 

1 2 3 4
R10 R4 =   = R6
2 1 4 3 

For example, the group of the square is D4, the group of the Pentagon is D5 etc.

31
Example:
The symmetries of a rectangle are easily seen to:

4 3
1 2 3 4
be, e =  
1 2 3 4 

1 2
1 2 3 4 1 2 3 4 1 2 3 4
a   , b   , c   
3 4 1 2 2 1 4 3 4 3 2 1

a is a rotation through  and b and c are reflections in the axes of

symmetry parallel to the sides. We see that:
a2 = b2 = c2 = e, ab = c, bc = a ca = b. Thus the group of symmetries of a
rectangle is the Klein four-group.

Example: Find all subgroups of S3 and S4.

Proper subgroups of S3 are the cyclic group generated by
C = (,2,3);  = (1 2);  = (1 3);  = (2 3)

Theorem
(cayley) Every group is isomorphic to a permutation group

Proof:

32
 Let G be a group. For any a G, the mapping fa: G  G. Given by fa(x) = ax for
all x G is a injection, because ax = ax1  x = x1 and y = fa (a-1 y) for all x, x-1, y
G consider the mapping.
: G  Sa given by (a) = fa  a G, where Sa is a symmetric group on the set G.
For all a, b, x G
fab (x) = ab x = fa(bx) = fa(fb (x)) = (fafb) (x).
Hence (a, b) = (a) (b).
 is a homomorphism and Im  is a subgroup of SG. Moreover (a) = (b)  ax
= bx for all x G  a = b.
Hence  is injective homomorphism. Therefore G is isumphic to a subgroup of SG

Normal subgroups
Definition:
Let G be a group. A subgroup N of G is called a normal subgroup of G,
written N  G, if for every nN and every X G , XnX-1  N.

Example:
If G and B are groups and h: G  B is a homomorphism, then Kerh is a
normal subgroup of G.

Proof:
Let  be the identity in G and  the identity in B. then h(e) =   e  Kerh. If
x  Kerh, then h(X-1) = {h(X)} –1 =  -1 = .
 x-1  Ker h, then h(ab) = h(a) h(b) = .  = 
 a, b  Ker h.

33
Finally if VG and X Ker h, then h(V –1 X V) = h(V-1) h(X) h(V).
= h(V –1) h(V).
= 
 V-1 XV  Ker h if V G and X Ker h.
Thus Ker h is a normal subgroup of G.
(2) If G is an abelian group and N is a subgroup of G, then is a normal
subgroup of G.

Proof:
For every gG and every nN, g-1 ng = g-1 gn = nN.

Example 3:
Given a group G. Let Z(a) = {x G/Xa = ax for every aG}, then Z(a) is called
the centre of G. Za has the following important properties;
(a) z(a) is a normal subgroup of G.
(b) If N is a subgroup of G s.t. Nc Z(a), then N is a normal subgroup of G.

Proof:
(a) Let e be the identity in G. Then e  Z(a) since  gG . eg. = g = ge.
If y Z(G), then for every gG, gy = yg  y –1g = gy –1  y –1  Z(a).
If for every x, y  Z(a), then for every gG, (Xy)g = x (yg) = X(gy) = (Xg)y =
g(xy)
 Xy  Z(a)..
The above show that Z(a) is a subgroup of G.
Finally, for every gG and every yG, g –1 yg = yg-1 g = y Z(G).
Hence Z(G) is a normal subgroup of G

34
(b) For each nN and for each gG, g-1n g = g-1gn = nN.
 N is a normal subgroup of G.

Theorem:
Let N be a subgroup of G. Then the following are equivalent.
(i) N is a normal subgroup of G. (ii) XnX-1 = n for every X G
(iii) Xn = nX for every XG. (iv) (Xn)yn) = Xyn for all X, y G.

Proof: (I)  (ii)

Suppose N is a normal subgroup of G.
Let X G, then by definition XnX-1 CN. Also, X-1 G.
Hence, X-1n XXn.
Therefore n = X((X-1nX)X-1 CXnX-1. Hence, XnX-1 = n.
(iii)  (iii) nX = (XnX-1) X = Xn X-1X = Xne = Xn
(iii) (iv). (Xn)(yn) = X(ny)n = X(yn) n = Xy(nn).
Now nn CN because N is closed under multiplication .
On the otherhand, n = en C nn
Hence nn = n. , (Xn)(yn) = (Xy)n.
(iv)  (I). XnX-1 = XnX-1 e  XnX-1n = XX-1n = en = n
Hence N is a normal subgroup of G.
Let N be a normal subgroup of G. The above theorem has shown that any left
coset of N in G is a right coset and vice versa. Hence we need not distinguish
between them and so all cosets of N shall be written as left cosets and denote the
set of all cosets of N in G by G|N. (gN = Ng).

35
Theorem: Let N be a normal subgroup of the group G. Then G|N is a group
under multiplication. The mapping Ø: G  G|N, given by X  Xn, is a
subjective homomorphism, and Ker Ø = N.

Proof: By the above theorem, (xn)(yn) = (xy)n  X, y G. Then G|N is a

group under multiplication, because multiplication is also associative in G.
multiplication is also associative in G|N. The coset en = n is the identity for
multiplication in G|N, and for any XG, (Xn)X-1n = (XX-1)n = en. Hence G|N is a
group.
The mapping Ø is obviously subjective. For all X, yG, Ø(Xy) = (Xy)n =
(Xn)(yn) = Ø(X)(Ø(y).
Hence, Ø is a homomorphism. Further, Xn = en
 XN. Hence, Ker Ø = {XG/Ø(X) = en} = n

Example:
(a) Prove that the centre Z(G) of a group G is a normal subgroup.

Solution:
For all a Z (G) and XG. X a X-1 = aXX-1 = a Z(G).
(b) If N and M are normal subgroups of G, prove that NM is also a normal
subgroup of G.

Solution:
For all nN. mM and XG.
X(nm) X-1 = (XnX-1) (XmX-1)  NM .
Hence the prove.

36
(c) Let H be a subgroup of a group G such that X2  H for every XG. Prove
that H is a normal subgroup of G and G|H is abelian.

Solution:
For all hH and XG. Xhx-1 = (Xh)2 h-1 (x-1)2 H. Hence HG. Moreover, in
every coset Xhє GIH, (Xh)(Xh) = hx2 = h. Hence G|H is abelian.

Theorem: If H is a normal subgroup of G, then aH = Ha for every a G.

Let H be a normal subgroup of G. If Ha = Hc and Hb = Hd, then H(ab) = H(c d).

Proof:
If ha = Hc then a  Hc , Hence a = hc for some h1  H.
If Hb = Hd, then b = h2 d for some h2 єH, thus ab = h1 ch2 d = h1 (ch2)d. But Ch2
 cH = Hc by the above theorem. Thus, ch2 =h3 c for some h3 H.
But ab = h1 (c h2)d = h1(h3 c)d = (h1h3) (c d).  H c d .
 ab  H(c d). But we know that if aє Hb then Ha = Hb. Hence H(a b) = H(c d).

Direct Products of Groups

If G and H are any two groups, then direct product denoted by G X H is a new
group.
Let G x H consists of all the ordered pairs (x, y) where X є G and y  H , thus
G x H = {(x, y)/x  G and y  H}.
The operation of G x H consists of multiplying corresponding components.
Eg. (x, y)  (x1, y1) = (xx1, yy1)
If G and H are denoted additively, it is customary to also denote G X H

37
additively.
Eg. (x, y) + (x1, y1) = (x + x1), (y + y1)
If G and H are groups, prove that g X H is a group.

Solution:
Let x1, x2, x3  G (x1, y1) [(x2, y2)] (x3, y3)] = [(x1y1) (x2, y2)](x3, y3)
Let ea be the identity in G and eH the identity in H.
Then the identity in G x H = (ea , eH) = e
(3) For each (a, b)  G x H, the inverse of (a, b) is
Then (a, b)-1 = (a-1, b-1) s.t.
(a,b) (a-1, b-1) = (aa-1, bb-1) = e.

Finitely Generated Abelian Groups

In this section we show that any finitely generated abelian group can be
decomposed as a finite direct sum of cyclic groups. This decomposition, when
applied to finite abelian groups enable us to find the number of nonismorphic
abelian groups of a given order.

Theorem (I):
(Fundamental theorem of finitely generated abelian groups).
Let A be a finitely generated abelian group. Then A can be decomposed as a
direct sum of a finite number of cyclic groups G. Precisely, A = C1 () …….
(+)Ck, such that either C1, ..Ck are all infinite, or for some j  k1 C1 …C j are of
order M1, ……..mj respectively, with m1/m2 …… / mj and C1j + 11 …… C1c are
infinite.

38
Theorem:
Let A be a finite abelian group. Then there exists a unique list of integers m1,
…mk |A| = m1 ………. Mk m1/m2………. mk and A = C, (+) C2 (+) …….. (+) Ck.
Where c1, …. Ck are cyclic subgroups of a of order m1, m2, …… mk, respectively
– consequently, A  Zm1, (+) …. (+) Zmk.

Proof:
By theorem 1.1 above; a = C1 (+) …. Ck, where C1, ….. Ck are cyclic subgroups
of order m1, …… mk, respectively, such that m1/m2/ …../mk.
Further, since |S||T| = |SIIT| for any finite sets S and T, It follows that:
|A| = |C1| ……….. + |Ck| = m1, ……. Mk.
Moreover, any cyclic group of order in S isomorphic to Zm. Hence
A = Zm……. (+) Zmk.

Uniqueness of m1 ……….. mk.

Suppose A = C1 (+) ……. Ck. = D1 (+) ….. (+) DI where Ci, Di are all cyclic
subgroups of A, with
|Ci| = mi, m1/m2/….. /mk
|Dj| = nj, n1/n2/ ………… / ni.
Now Di has an element of order ni. But every element in A is of order  m1c mi.
Hence ni  mk
By the same argument mk  mi. Hence mk= ni
Now consider mk-1 A = {mk-1, G/a  A}. From the two decomposition of A
assumed above, we get

39
mk-1 A = (mk-1 c1)  …. (mk-1 Ck-1)  (mk-1 Ck=1)
= (mk-1 D1)  …… (mk-1 Dk-1 )  mk-1 Di)
Because mi /mk-1 . For i = 1, ----- k-1. if follows that mk-1 ci is trivial for i = 1,
……… k-1. |mk-1 A| = |mk-1 ck| = |mk-1 Di|. Therefore, |m1k-1 = Di| = 1, for j = 1
……… i – 1
Hence, ni-1 |mk-1. By symmetric argument mk-1| ni-1. Therefore mk-1 = ni-1.
Proceeding in this manner, we show that mk-r = ni-r, v = 0,1,2, - …….
But m1 …. mk = |A| = n1 …….. ni. Hence k = i and m1 = ni, i = 1,------ k.

Definition:
Let A be a finite abelian group.
If A  Zm  Zm2  ………  Zmk, where 1  m1|m2|…….mk, then A is said to
be of type (m1, …. Mk) and the integers m1, …mk are called the invanants of A.
The fundamental theorem of finite abelian groups enables me to determine the
number of isomorphism choses of abelian groups of order n.

Lemma:
There ;is a 1-1 correspondence between the family F of nonisomorphic abelian
groups of order Pe, P Prime, and the Set P(e) of partition of e.

Proof:
Let AєF. Then A determine a unique type (Pe … Pek) were ei  e2  …  ek. Cd
e1 + e2 + …… + e1c) = e. Define a mapping Ø: F  P(e) by Ø(A) = (e1,……ek
). Ø is clearly injective . ( By theorem 1) To show that Ø is subjective, let (e1,
….. es)  P(e). Then the group Zpe,  …… Zpes, is obviously the pre-image
of (e1 ……. es)

40
Theorem:
Let n = kj=1 Pj f , Pj distinct primes. Then number of nonisomorphic abelian
groups of order n is kj =1 |P(fj)|.
Example:
Find the nonismorphic abelian groups of order 360 = 23. 32.51.

Solution: n = 360 = 23.51 /P(3)/ = 3 /P(2)/ =2

|P(1)| = 1 Hence these are six nonismorphic abelian groups of order 360, and these
are:
Z8  Z9  Z5
Z8  Z3  Z3  Z5
Z2  Z4  Z3  Z 3  Z5
Z2  Z2  Z2  Z9  Z5
Z2  Z2  Z2  Z3  Z3  Z5

Sylow theorem:
Definition:
Let P be a prime number. A P-group is any group whose order is a power of P.

Definition
Let G be a finite group, and let P be a prime. Let Pm |G|, Pm+1 |G|, m > 0. Then
any subgroup of order Pm is called Sylow P-subgroup of G.

Definition:

41
Let P be prime. A group G is called a P-group if the order of every element in G
is some power of P. Likewise, a subgroup H of any group G is called a subgroup
of G if the order of every element in H is some power of P.

Lemma:
(Cauchy’s theorem for abelian groups).
Let A be a finite abelian group, and let P be a prime. If P divides /A/, then A has
an element of order P.

Theorem:
(First Sylow Theorem).
Let G be a finite group, and let P be a prime. If Pm divides the order of G, then G
has a subgroup of order Pm.

Proof:
The prove is by induction m n = IGI. If n = 1, the result is obvious. assure the
result holds for all groups of orders less than n.
If the order of the centre of G is divisible by P, then it follows from Cauchy’s
Theorem for abelian groups that the centre of G contains an element a of order P.
The cyclic group C generated by a is clearly nomal in G, and the quotient group
G/C has order n/p, which is divisible by Pm-1. Therefore, by the induction
hypothesis, G/C contains a subgroup H/C of order Pm-1. Consider the case when
the order of the centre is not divisible by P.
The class equation of G is,

42
n = |G| = |Z(G)| +  (G : N(a)), where summation runs over one element from
each conjugate class having more than one element. Now P/n, but p  |Z(G)|.
Hence, p  [G : N(a)] for some a  G, a  Z(G). Thus implies that Pm /[N(a)]
and |N(a)|  |G|. By the induction hypothesis N(a) has a subgroup of order Pm,
which is the required subgroup of G.
Corollary (Cauchy’s theorem). If the order of a finite group G is divisible by a
prime number P, then G has an element of order P.

Proof:
By Sylow’s Theorem there exists a subgroup of order P. Therefore, there are at
least P-1 element of order P. (Namely all but the identity element of the cyclic
subgroup of order P).
Corollary: A finite group G is a p-group if its order is a power of P.

Proof:
If G is of prime power order, then each element has, invially, order a power of P.
So let us assure that each element in G has order of G is a power of P. Let q  p
be a prime number dividing the order of G. then by Cauchy’s theorem  an
element of order q, a contradiction. Hence, G is of order a power of P.

Corollary:
Every group of order P2 (P prime) is abelian.
Suppose G is a non-abelian group of order P2. By a theorem, G has a non-trivial
centre Z that must be of order P. Let a єG, a  Z. Then Z is a proper subset of

43
N(a). Hence N(a) must be of order P2. This implies N(a) = G Hence, a є Z, a
contradiction.

Example:
Every group of order q is either a cyclic group or abelian group, {e, a, a2, b, b2,
ab, a2b, ab2, a2b2}. Where e is the identity and a, b are elements of order 3 and
ab = ba.

Groups of Order P2
We know that a group of order P2 is abelian. Hence these are two and only two
groups of order P2.
(i) abelian group of type (P2) and (ii) abelian group of type (P, P)
Corollay: a Sylow P-subgroup of a finite group G is unique if and only if it is
normal.

Example:
(a) If a group of order Pn contains exactly one subgroup each of orders
P, P2, ---- Pn-1, then it is cyclic.

Solution:
Let G be a group of order Pn, and let H be a sybgroup of order Pn-1. By sylow’s
first theorem, H contains subgroups of orders P, P2, ----- Pn-2. Because these
subgroups are also subgroups of G, it follows that all the proper subgroups of G are
contained in H. Now if a  G, a  H, then the order of G must be Pn, otherwise
a will generate a subgroup K of order  Pn. This would mean a  H, because aє
K c H. Thus, the order of a is Pn, and G is therefore a cyclic group generated by
a. he only proper subgroups of Sy are the cyclic groups generated by:

44
{1,2,3,4}, {1 2}, {1 3}, {1 4}, {1 2 3} (1 2 4} {2 3 4} {34} {2 3} {2, 4}
{12} {13} {14} {23} {24} {34} {123} {124} {234} {1, 3, 4} {1 2 3 4}
(ab)2 = a2b2
(ab)2 a.a1 = a2. B2 a.a-1
= ab
= b2 a2 a.a-1
ab = b2 a2

45