G12ALN

Groups

Edjvet, M

University of Nottingham

2019

1 / 146
Section 1

Groups

2 / 146
3 / 146
4 / 146
 1 Groups
1.1 Permutations

Recall that a permutation of a non-empty set X is a bijection from

X to X. The identity permutation is the mapping idX : X → X
defined by idX (x) = x (∀x ∈ X).
Example If X = {1, 2, . . . , n} then X has n! permutations; and
the set of permutations of X is denoted by Sn .
We adopt cycle notation for permutation. For example a k-cycle is
a permutation π ∈ Sn which has the form π = (a1 a2 . . . ak ) where
{a1 , a2 , . . . , ak } ⊆ {1, 2, . . . , n} and is defined by π(aj ) = aj+1
(1 ≤ j < k), while π(ak ) = a1 and π does not move any other
points.
Example π = (1743) ∈ S8 is a 4-cycle defined by π(1) = 7;
π(2) = 2; π(3) = 1; π(4) = 3; π(5) = 5; π(7) = 4; and π(8) = 8.
Observe that π = (1743) = (7431) = (4317) = (3174).
5 / 146
The k1 -cycle (a1 . . . ak1 ) and k2 -cycle (b1 . . . bk2 ) are said to be
disjoint if {a1 , . . . , ak1 } ∩ {b1 , . . . , bk2 } = ∅.
Theorem 1.1.1 Every permutation of a finite set can be written
as a product of disjoint cycles.
Convention We usually omit 1-cycles, so for example, for
(13)(275)(4)(6) ∈ S7 we write (13)(275).
Example

S2 = {(1)(2), (12)}
S3 = {(1)(2)(3), (12), (13), (23), (123), (132)}
S4 = {(1)(2)(3)(4), (12), (13), (14), (23), (24), (34),
(12)(34), (13)(24), (14)(23), (123), (132), (124),
(142), (134), (143), (234), (243), (1234), (1243),
(1324), (1342), (1423), (1432)}.

6 / 146
7 / 146
8 / 146
Thus the possible cycle structures for S4 are (∗)(∗)(∗)(∗); (∗∗);
(∗∗)(∗∗); (∗ ∗ ∗); and (∗ ∗ ∗∗) of which there are 1, 6, 3, 8 and 6
respectively.
If we compose two permutations of a set X then we always end up
with another permutation of X. Therefore if τ : X → X and
σ : X → X are permutations then so are τ σ = τ ◦ σ : X → X and
στ = σ ◦ τ : X → X defined by (τ σ)(x) = (τ ◦ σ)(x) = τ (σ(x))
and (στ )(x) = (σ ◦ τ )(x) = σ(τ (x)) (∀x ∈ X).
Example If τ = (12)(34) and σ = (123) then
τ σ = (12)(34)(123) = (243) and στ = (123)(12)(34) = (134).
Notice that τ s 6= στ . However if π and ρ are disjoint permutations
of X then πρ = ρπ.
Definition Sn will now denote the set of permutations of
X = {1, . . . , n} together with the binary operation of composition
of permutations.

9 / 146
Example The composition or multiplication table for S3 .
(1)(2)(3) (12) (13) (23) (123) (132)
(1)(2)(3) (1)(2)(3) (12) (13) (23) (123) (132)
(12) (12) (1)(2)(3) (132) (123) (23) (13)
(13) (13) (123) (1)(2)(3) (132) (12) (23)
(23) (23) (132) (123) (1)(2)(3) (13) (12)
(123) (123) (13) (23) (12) (132) (1)(2)(3)
(132) (132) (23) (12) (13) (1)(2)(3) (123)

Recall that a mapping is a bijection if and only if it is invertible . It

follows that every permutation π of X has a unique inverse π −1
for which π ◦ π −1 = π −1 ◦ π = idX .

10 / 146
11 / 146
12 / 146
Example We see from the table for S3 that
(1)(2)(3)−1 = (1)(2)(3); (12)−1 = (12); (13)−1 = (13);
(23)−1 = (23); (123)−1 = (132); and (132)−1 = (123). Indeed to
obtain the inverse of π given by disjoint cycles, simply reverse the
order of each cycle. So (149)(2765)−1 = (941)(5672)
(= (194)(2567) for example).
Notation π m = π ◦ π ◦ . . . ◦ π (m times).
Example If π = (14738) then π 2 = (17843); π 3 = (13487);
π 4 = (18374); and π 5 = (1)(2) . . . (8) = idX .
Theorem 1.1.2 (i) If π is a k-cycle then π k = idX but
0
π k 6= idX for 1 ≤ k 0 < k. (ii) If π is the product of r ≥ 2 pairwise
disjoint ki -cycles πi (1 ≤ i ≤ r) then π l = idX when
0
l = l.c.m.{ki : 1 ≤ i ≤ r} but π l 6= idX for 1 ≤ l0 < l.

13 / 146
Example If π = (1234)(56)(789) then l.c.m.{4, 2, 3} = 12 so
π 12 = idX whereas π m 6= idX for 1 ≤ m ≤ 11. But we need
disjoint. For example if π = (234568)(139)(3579)(124) then
l.c.m.{6, 3, 4, 3} = 12 but π = (1368257) and so π 7 = idX .
Definition A cycle of length 2 or 2-cycle is called a transposition.
For example S4 contains the 6 transpositions (12), (13), (14),
(23), (24) and (34). More generally, Sn contains n(n − 1)/2
distinct transpositions.
Observe that
(a1 a2 . . . ak ) = (a1 ak ) . . . (a1 a3 )(a1 a2 ) (k − 1),
= (a1 a2 )(a2 a3 ) . . . (ak−1 ak ) (k − 1).

Since every permutation is a product of disjoint cycles we have the

following result.

14 / 146
15 / 146
16 / 146
Theorem 1.1.3 Every permutation can be decomposed into a
product of transpositions.
Example (153)(264) = (13)(15)(24)(26) = (15)(53)(26)(64)
Remark Unfortunately the decomposition into transpositions is
in general neither disjoint nor unique. For example
(19643) = (13)(14)(16)(19) = (16)(24)(36)(19)(46)(26). However
we have
Theorem 1.1.4 If a permutation can be written as an even
number of transpositions then it cannot be written as a product of
an odd number of transpositions and vice-versa.
Definition An element of Sn is called even or odd according to
whether it can be written as a product of an even or odd number
of transpositions. The set of even permutations of Sn is denoted
by An and An contains exactly n! 2 elements.

17 / 146
Note Observe that idX = (12)(12) for example and so is even.
Example

A2 = {(1)(2)}
A3 = {(1)(2)(3), (123), (132)}
A4 = {(1)(2)(3)(4), (12)(34), (13)(24), (14)(13), (123),
(132), (124), (142), (134), (143), (234), (243)}

Concluding Remark Observe that Sn is closed in the sense that

the composition of any two elements in Sn yields another element
in Sn ; that Sn has an identity permutation idX such that
idX π = π = π idX (∀π ∈ Sn ); that each permutation π ∈ Sn has
an inverse π −1 ∈ Sn ; and that associativity holds since it holds for
composition
of mappings:



(f ◦ g) ◦ h
(x) = (f ◦ g) h(x)

= f g h(x)

and
f ◦ (g ◦ h) (x) = f ◦ g h(x) = f g h(x) .

18 / 146
19 / 146
20 / 146
 1 Groups
1.2 Basic concepts

A group G = hG, ∗i is a non-empty set G together with a binary

operation ∗ on G satisfying the following axioms.

(G1) (closure) x ∗ y ∈ G (∀x, y ∈ G)

(G2) (associativity) x ∗ (y ∗ z) = (x ∗ y) ∗ z (∀x, y, z ∈ G)
(G3) (identity) ∃e ∈ G such that x ∗ e = x = e ∗ x (∀x ∈ G), e is
called an identity element
(G4) (inverses) corresponding to each x ∈ G there is an element
y ∈ G such that x ∗ y = e = y ∗ x, y is called an inverse of x.

21 / 146
Remarks

1. We will often write xy for x ∗ y.

2. A group G is Abelian or commutative if

x∗y =y∗x (∀x, y ∈ G).

3. The order of G, denoted |G|, is the number of elements in G.

A group is said to be infinite if it has an infinite number of
elements – otherwise finite.
4. Each group G has exactly one identity element, often denoted
by e or eG , and each element g has exactly one inverse,
denoted g −1 .
5. If |G| = n < ∞ then we can form the n × n Cayley table for
G. The (i, j)th entry is gi ∗ gj where
G = {g1 = e, g2 , . . . , gn }. Thus G is Abelian iff (i, j)th entry
= (j, i)th entry (1 ≤ i, j ≤ n).

22 / 146
23 / 146
24 / 146
Examples

1. hZ, +i hQ, +i hR, +i hC, +i are infinite Abelian groups. The

identity is 0 and x−1 = −x.
2. hQ \ {0}, ×i hR \ {0}, ×i hC \ {0}, ×i are infinite Abelian
groups. The identity is 1 and x−1 = x1 .

Examples

1. Sn the symmetric group of degree n ≥ 2. Recall that

|Sn | = n! and Sn is Abelian if and only if n = 2. The identity
element is (1)(2) . . . (n).
2. An the alternating group of degree n ≥ 2. Recall that
|An | = 12 (n!) and that An consists of the even permutations.
Clearly the identity permutation is even; the product of two
even permutations is even; and the inverse of an even
permutation is even.

25 / 146
Examples

1. hMn (R), +i The set of n × n matrices over R is an infinite

Abelian group. The identity is the n × n zero matrix; and
A−1 = −A.
2. hGLn (R), ×i The set of invertible n × n matrices over R is
an infinite non-Abelian group. The identity is In .

Examples

1. hZ/nZ, +i; {0, 1, . . . , n − 1}; addition modulo n.

2. hZ/pZ \ {0}, ×i; {1, 2, . . . , p − 1}; p a prime; multiplication
mod p.
3. Cm = hxi = {xi : 0 ≤ i ≤ m − 1} xi xj = xk where
k ≡ (i + j) mod m (cyclic group of order m).

Example Dn the nth dihedral group where n ≥ 3. Dn is the

group of symmetries (reflections and rotations) of the regular
n-gon and |Dn | = 2n.
26 / 146
27 / 146
28 / 146
Example D4

1 2

4 3

rotations: (1)(2)(3)(4), (1234), (13)(24), (1432)

reflections: (12)(34), (14)(23), (13), (24)

29 / 146
Cayley table for D4 .

(1)(2)(3)(4) (1234) (13)(24) (1432)

(1)(2)(3)(4) (1)(2)(3)(4) (1234) (13)(24) (1432)
(1234) (1234) (13)(24) (1432) (1)(2)(3)(4)
(13)(24) (13)(24) (1432) (1)(2)(3)(4) (1234)
(1432) (1432) (1)(2)(3)(4) (1234) (13)(24)
(12)(34) (12)(34) (24) (14)(23) (13)
(14)(23) (14)(23) (13) (12)(34) (24)
(13) (13) (12)(34) (24) (14)(23)
(24) (24) (14)(23) (13) (12)(34)

30 / 146
31 / 146
32 / 146
 (12)(34) (14)(23) (13) (24)
(1)(2)(3)(4) (12)(34) (14)(23) (13) (24)
(1234) (13) (24) (14)(23) (12)(34)
(13)(24) (14)(23) (12)(34) (24) (13)
(1432) (24) (13) (12)(34) (14)(23)
(12)(34) (1)(2)(3)(4) (13)(24) (1432) (1234)
(14)(23) (13)(24) (1)(2)(3)(4) (1234) (1432)
(13) (1234) (1432) (1)(2)(3)(4) (13)(24)
(24) (1432) (1234) (13)(24) (1)(2)(3)(4)

33 / 146
A subset H of a group G = hG, ∗i is a subgroup of G, written
H ≤ G, if and only if the following conditions are satisfied.
(S1) eG ∈ H
(S2) x, y ∈ H ⇒ x ∗ y ∈ H
(S3) x ∈ H ⇒ x−1 ∈ H
Examples
1. hZ, +i ≤ hQ, +i ≤ hR, +i.
2. hkZ, +i ≤ hZ, +i kZ =
{. . . , −3k, −2k, −k, 0, k, 2k, 3k, . . .}.
Example An ≤ Sn
Example SLn (Z) ≤ GLn (Z) Special linear group of degree n
over Z consisting of matrices with determinant = 1.
Example Any subgroup of a cyclic group is cyclic.
Example The intersection of subgroups is itself a subgroup.
Example (transitivity) If A, B, C are subgroups of G such
that A ≤ B and B ≤ C then A ≤ C.
34 / 146
35 / 146
36 / 146
Let g ∈ G. The order of g, denoted |g|, is the least positive integer
n such that g n = eG . If no such n exists then g is said to have
infinite order.
Examples

1. |g| = 1 if and only if g = eG .

2. If π is the product of r pairwise disjoint cycles πi of length ki
(1 ≤ i ≤ r) in Sn then |π| = l.c.m.{ki : 1 ≤ i ≤ r}.
3. Recall that if g ∈ G then hgi ≤ G is the cyclic subgroup
generated by g and consists of all powers of g. Thus
|hgi| = |g|. For example if |g| = 6 then
hgi = {e(= g 0 = g 6 ), g (= g −5 ), g 2 (= g −4 ), g 3 (=
g −3 ), g 4 (= g −2 ), g 5 (= g −1 )}. If |g| = ∞ then
hgi = {. . . g −3 , g −2 , g −1 , e, g, g 2 , g 3 , . . .}.

37 / 146
Let G = hG, ∗i and H = hH, #i be groups. The direct product of
G and H is the set G × H = {(g, h) : g ∈ G, h ∈ H} together with
the binary operation (g1 , h1 )(g2 , h2 ) = (g1 ∗ g2 , h1 #h2 ).
Remarks

1. |G × H| = |G||H|.
2. If (g, h) ∈ G × H and g, h have finite order then
|(g, h)| = l.c.m.{|g|, |h|}.
3. Can form G1 × . . . × Gn for n ≥ 2.
4. G × H is Abelian if and only if G and H are Abelian.
(Exercise Sheet.)

Example C2 = hxi, C3 = hyi, |(x, y 2 )| = 6 in C2 × C3 .

38 / 146
39 / 146
40 / 146
Let S be a non-empty subset of the group G. Denote by hSi the
set of finite products of elements in S and the inverses of elements
in S. Thus x ∈ hSi if and only if ∃n ∈ N and elements t1 , . . . , tn
such that ti ∈ S or t−1
i ∈ S, and such that

x = t1 t2 . . . tn .

Example S = {a, b, c} x = bca−3 bac−11 acb−1 ∈ hSi.

Lemma 1.2.1 (a) hSi ≤ G.
(b) hSi = ∩{H : H ≤ G and S ⊆ H} = K.
Proof (a) Exercise Sheet.
(b) Let x ∈ S. Then certainly x ∈ K and, since the intersection of
subgroups is a subgroup, it follows that any product of elements of
S or their inverses is in K. So hSi ⊆ K. But by (a), hSi is a
subgroup of G that contains S and so K ⊆ hSi since K is the
intersection of such subgroups. 
41 / 146
We call hSi the subgroup of G generated by S. If hSi = G then
we say G is generated by (the elements of) S. It follows from
1.2.1(b) that hSi is the smallest subgroup of G to contain S.
If S = {s1 , . . . , sk } we sometimes write hs1 , s2 , . . . , sk i for hSi.
Example If S = {s} then hSi = hsi is the cyclic subgroup
generated by s.
Example Sn = h{(ij) : 1 ≤ i < j ≤ n}i – the symmetric group is
generated by transpositions.
Example A4 = h(123), (124)i (Exercise Sheet)
We can also work additively. In this case x ∈ hSi means
n
P
x= λi si where λi ∈ Z and si ∈ S.
i=1

Example Z = hZ, +i = h7, 11i since if x ∈ Z then

x = λ1 7 + λ2 11 where λ1 = −3x and λ2 = 2x.

42 / 146
43 / 146
44 / 146
A mapping θ : hG, ∗i → hH, #i is called a homomorphism if

θ(x ∗ y) = θ(x) # θ(y) (∀x, y ∈ G).

Examples

1. θ : hGLn (R), ×i → hR \ {0}, ×i defined by θ(A) = det A.

2. θ(eG ) = eH ; more generally θ(g k ) = [θ(g)]k (∀g ∈ G,
∀k ∈ Z).
3. If the homomorphism θ : G → H is bijective then θ is an
isomorphism and we write G ∼
= H. For example D3 ∼ = S3 .
4. If θ : G → G is an isomorphism then θ is called an
automorphism of G.
5. (Exercise Sheet.) A × B ∼= B × A. More generally, if π ∈ Sn
then G1 × . . . × Gn ∼
= Gπ(1) × . . . × Gπ(n) .

45 / 146
Let θ : G → H be a homomorphism. Then the image of θ is
Im θ = {θ(g) : g ∈ G} and the kernel of θ is
ker θ = {g ∈ G : θ(g) = eH }.
Theorem 1.2.2

(i) Im θ ≤ H and ker θ ≤ G;

(ii) θ is surjective if and only if Im θ = H;
(iii) θ is injective if and only if ker θ = {eG }.

[Note: For f : X → Y
f injective: f (x) = f (y) ⇒ x = y
f surjective: ∀y ∈ Y ∃x ∈ X s.t. f (x) = y.]

46 / 146
47 / 146
48 / 146
Proof

(i) (S1) θ(eG ) = eH ⇒ eH ∈ Im θ and eG ∈ ker θ.

(S2) h1 , h2 ∈ Im θ ⇒ ∃g1 , g2 such that θ(g1 ) = h1 and
θ(g2 ) = h2 ⇒ θ(g1 g2 ) = θ(g1 )θ(g2 ) = h1 h2 ⇒ h1 h2 ∈ Im θ.
(S2) g1 , g2 ∈ ker θ ⇒ θ(g1 g2 ) = θ(g1 )θ(g2 ) = eH eH = eH ⇒
g1 g2 ∈ ker θ.
(S3) h ∈ Im θ ⇒ ∃g ∈ G such that
θ(g) = h ⇒ θ(g −1 ) = [θ(g)]−1 = h−1 ⇒ h−1 ∈ Im θ.
(S3)
g ∈ ker θ ⇒ θ(g −1 ) = [θ(g)]−1 = e−1H = eH ⇒ g
−1 ∈ ker θ.

(ii) Immediate from definition of surjective.

(iii) Suppose θ is injective and let g ∈ ker θ. Then
θ(g) = eH = θ(eG ) ⇒ g = eG ⇒ ker θ = {eG }.
Suppose ker θ = {eG }. Then θ(g1 ) = θ(g2 ) ⇒
θ(g1 )[θ(g2 )]−1 = θ(g2 )[θ(g2 )]−1 ⇒ θ(g1 )θ(g2−1 ) = eH ⇒
θ(g1 g2−1 ) = eH ⇒ g1 g2−1 = eG ⇒ g1 = g2 . 

49 / 146
Example

C∞ = hxi = {. . . x−3 , x−2 , x−1 , x0 = 1, x, x2 , x3 , . . .}

C9 = hyi = {1, y, y 2 , . . . , y 8 }

Let θ : C∞ → C9 be a homomorphism such that θ(x) = y 4 .

Then, for example, θ(x3 ) = [θ(x)]3 = (y 4 )3 = y 12 = y 3 .

Im θ = {. . . θ(x−3 ), θ(x−2 ), θ(x−1 ), θ(1), θ(x), θ(x2 ), θ(x3 ), . . .}

= {. . . θ(x)−3 , θ(x)−2 , θ(x)−1 , 1, θ(x), θ(x)2 , θ(x)3 , . . .}
= hθ(x)i = hy 4 i
= {y 4 , y 8 , y 3 , y 7 , y 2 , y 6 , y, y 5 , 1} = C9 (surjective)
ker θ = {. . . x−18 , x−9 , x0 = 1, x9 , x18 , . . .} = hx9 i (not injective).

50 / 146
51 / 146
52 / 146
 1 Groups
1.3 Finite Abelian groups
We now consider the structure of finite Abelian groups. We recall
the structure of finite cyclic groups first.
Theorem 1.3.1 Let G be a finite cyclic group of order n ≥ 2,
say,
G = hxi = {e, x, x2 , · · · , xn−1 } .
Then to each divisor d of n, there exists exactly one subgroup of
order d, namely (with n = dm):

hxm i = {e, xm , x2m , · · · , x(d−1)m } .

These are all the subgroups of G, and they are all cyclic.
0
For two such subgroups hxm i and hxm i (where n = dm = d0 m0 ),
0
we have hxm i ≤ hxm i if and only if m0 divides m.
Proof We omit the proof. 
53 / 146
With this theorem, it is now quite easy to draw the lattice of
subgroups of a finite cyclic group G. Such a lattice is a graph in
which all the subgroups are arranged from bottom to top by
inclusion (the group {e} being the bottom vertex, the group G
itself being the top vertex). A subgroup H in this lattice is
contained in another subgroup H 0 if there is a strictly upward
moving path from H to H 0 .
Example Consider C36 = hxi. By the above theorem, the
subgroups are therefore hxm i with m ∈ {1, 2, 3, 4, 6, 9, 12, 18, 36},
and hxm i is of order 36/m = d.
For example, we have hx4 i ≤ hx2 i because 2 divides 4, but
hx3 i 6≤ hx2 i because 2 does not divide 3. Thus, by checking
mutual divisibility, we obtain the following lattice of subgroups of
C36 (the respective orders of the subgroups are added as indices):

54 / 146
55 / 146
56 / 146
hxm id

hxi36 = C36
 Q
 Q
 QQ
hx2 i18 hx3 i12
 Q  Q
 Q  Q
 QQ  QQ
4 6
hx i9 hx i6 hx9 i4
Q  Q 
Q  Q 
QQ  QQ 
hx12 i3 hx18 i2
Q 
Q 
QQ 
36
hx i1 = {e}

57 / 146
Recall the following.
Theorem 1.3.2 Cmn ∼
= Cm × Cn iff gcd(m, n) = 1.
Proof. (⇒) If prime p divides m and n then by Theorem 1.3.1 we
have Cp × Cp ≤ Cm × Cn ∼ = Cmn which contradicts the fact that a
subgroup of a cyclic group is cyclic since Cp × Cp is not a cyclic
group. (⇐) Observe that |(x, y)| = l.c.m.{|x|, |y|} = |x||y| and so
Cm × Cn is a cyclic group with generator (x, y). 

58 / 146
59 / 146
60 / 146
WARNING gcd(m, n) 6= 1 ⇒ Cm × Cn 6 Cmn , for example
C2 × C4  C8 .
Example Consider C4 × C10 . Now 4 does not divide 10 however
we have

C4 × C10 ∼
= C4 × C2 × C5 (since C10 ∼
= C2 × C5 )
∼
= C2 × C4 × C5 (since C4 × C2 ∼= C2 × C4 )
∼
= C2 × C20 (since C20 ∼
= C4 × C5 )

and 2 does divide 20. It turns out that we can always do this.

61 / 146
Theorem 1.3.3 (Classification Theorem for Finite Abelian
Groups) Let A be a non-trivial finite Abelian group. Then

A∼
= A1 × · · · × Ak

where (i) each Ai is cyclic of order mi > 1, say; and (ii)

m1 | m2 . . . | mk .
Moreover if also A ∼
= B1 × · · · × Bl where each Bj is cyclic of order
nj > 1 and n1 | n2 | . . . | nl then l = k and ni = mi (1 ≤ i ≤ k).
Proof We omit the proof. 
Examples

|A| = p (prime) Cp
|A| = 4 C2 × C2 ; C4
|A| = 6 C6
|A| = 8 C2 × C2 × C2 ; C2 × C4 ; C8

62 / 146
63 / 146
64 / 146
Example |A| = 8575 = 52 .73
Partitions of 2: 1 + 1; 2.
Partitions of 3: 1 + 1 + 1; 1 + 2; 3.
Six possibilities:

0 + 1 + 1, 1 + 1 + 1 C7 × C35 × C35 C50 71 × C51 71 × C51 71

0 + 1 + 1, 0 + 1 + 2 C35 × C245
0 + 1 + 1, 0 + 0 + 3 C5 × C1715
0 + 0 + 2; 1 + 1 + 1 C7 × C7 × C175
0 + 0 + 2; 0 + 1 + 2 C7 × C1225
0 + 0 + 2; 0 + 0 + 3 C8575
5 7

65 / 146
Notation If x ∈ Z then P (x) denotes the number of partitions of
x. For example P (1) = 1, P (2) = 2, P (3) = 3, P (4) = 5 and
P (5) = 7.
Corollary 1.3.4 If n = pn1 1 . . . pnk k where ni ≥ 1 and the pi are
different primes then the number of Abelian groups of order n is
P (n1 ) · P (n2 ) . . . · P (nk ).
Example The number of Abelian groups of order 526500?
Now 526500 = 22 .34 .53 .13 so number is
P (2).P (4).P (3).P (1) = 2.5.3.1 = 30.

66 / 146
67 / 146
68 / 146
 1 Groups
1.4 Cosets and Lagrange’s Theorem

Let G be a group and let H be a subgroup of G. A left coset of H

in G is a subset of G of the form

xH = {xh : h ∈ H}

for some x in G.
A right coset of H in G is a subset of the form

Hx = {hx : h ∈ H}

for some x in G.

69 / 146
Remarks

1. If G is an Abelian group then

xH = {xh : h ∈ H} = {hx : h ∈ H} = Hx. For Abelian
groups we need not distinguish between left and right cosets.
2. In general xH 6= Hx (see example below).
3. eH = H = He.
4. xH contains x since x = xe and e ∈ H ≤ G.
5. xH = H if and only if x ∈ H if and only if Hx = H (Exercise
Sheet).
6. |xH| = |H| = |Hx| (Exercise Sheet).

70 / 146
71 / 146
72 / 146
Example D3 = {(1)(2)(3), (123), (132), (12), (23), (13)}
Let H = h(12)i = {(1)(2)(3), (12)}. Then

(1)(2)(3)H = (12)H = H H(1)(2)(3) = H(12) = H

(123)H = (13)H = {(123), (13)} H(123) = H(23) = {(123), (23)}
(132)H = (23)H = {(132), (23)} H(132) = H(13) = {(132), (13)}

If H is a subgroup of an Abelian group A then we can often use

additive notation. A coset of H in A is of the form

x + H = {x + h : h ∈ H}

for some x in A.

73 / 146
Example A = Z H = 3Z

0 + 3Z = {. . . , −9, −6, −3, 0, 3, 6, 9, . . .} = 3Z = H

1 + 3Z = {. . . , −8, −5, −2, 1, 4, 7, 10, . . .}
2 + 3Z = {. . . , −7, −4, −1, 2, 5, 8, 11, . . .}

and Z = 3Z ∪ (1 + 3Z) ∪ (2 + 3Z) (disjoint).

Given H ≤ G we can define two relations on G as follows:
a ∼L b if and only if a−1 b ∈ H
a ∼R b if and only if ab−1 ∈ H

74 / 146
75 / 146
76 / 146
Theorem 1.4.1
(i) a ∼L b is an equivalence relation on G and if c ∈ G then the
equivalence class [c] containing c is the left coset cH. In
particular, the left cosets of H in G form a partition of G.
(ii) a ∼R b is an equivalence relation on G and if c ∈ G then the
equivalence class [c] containing c is the right coset Hc. In
particular, the right cosets of H in G form a partition of G.
Proof
(i) If a ∈ G then a−1 a = e ∈ H and so a ∼L a. If a ∼L b then
a−1 b ∈ H and so b−1 a = (a−1 b)−1 ∈ H which implies
b ∼L a. If a ∼L b and b ∼L c then a−1 b and b−1 c ∈ H so
a−1 c = (a−1 b)(b−1 c) ∈ H which implies a ∼L c.
Now let x ∈ [c]. Then c ∼L x and so c−1 x ∈ H ⇒ c−1 x = h
for some h ∈ H ⇒ x = ch ⇒ x ∈ cH and thus [c] ⊆ cH. On
the other hand if y ∈ cH then y = ch for some
h ∈ H ⇒ c−1 y = h ∈ H ⇒ c ∼L y ⇒ y ∈ [c] ⇒ cH ⊆ [c],
therefore equality. (ii) Similar to (i). 
77 / 146
The theory of equivalence relations yields
Corollary The cosets of H in G are pairwise equal or disjoint and
they form a partition of G.
Lemma 1.4.2 Let G be a group, H ≤ G and x, y ∈ G. The
following are equivalent:

(i) (a) x−1 y ∈ H (ii) (a) xy −1 ∈ H

(b) y ∈ xH (b) x ∈ Hy
(c) xH ∩ yH 6= ∅ (c) Hx ∩ Hy 6= ∅
(d) xH = yH (d) Hx = Hy
(e) x ∈ yH (e) y ∈ Hx
(f ) y −1 x ∈ H (f ) yx−1 ∈ H

Note. Therefore

xH = yH iff x−1 y ∈ H iff y −1 x ∈ H.

78 / 146
79 / 146
80 / 146
Lemma 1.4.3 The number of left cosets of H in G equals the
number of right cosets of H in G.
Proof. Exercise. 
Put

|G : H| = number of left cosets of H in G

= number of right cosets of H in G.

Then |G : H| is called the index of H in G.

Let G be a group and H ≤ G. We have seen that, for example,
the right cosets of H in G form a partition of G.

81 / 146
In particular, there exist distinct elements {xi : i ∈ I} ⊆ G (where
I is a suitably chosen index set) such that
[˙
G= Hxi (disjoint).
i∈I

The set {xi : i ∈ I} is called a system of right coset representatives

of H in G. Clearly we have |I| = |G : H|.
But by Remark 6 above we have |H| = |Hxi | (∀i ∈ I) and so we
have.
Theorem 1.4.4 If H ≤ G then |G| = |G : H||H|.
Consequently we get one of the fundamental results of group
theory.
Corollary 1.4.5 (Lagrange’s Theorem) If G is a finite group and
H ≤ G then |H| divides |G|.
If g ∈ G then |g| = |hgi| and hgi ≤ G so another immediate
consequence is.
Corollary 1.4.6 If g ∈ G then |g| divides |G|.
82 / 146
83 / 146
84 / 146
Example G = D3 H = {(1)(2)(3), (12)}.
D3 = H ∪ {(123), (13)} ∪ {(132), (23)} =
(1)(2)(3)H ∪ (13)H ∪ (132)H.
So

|D3 | = 6 |H| = 2 |D3 : H| = 3 and {(1)(2)(3), (13), (132)}

is a system of left coset representatives. Note that

{(1)(2)(3), (13), (132)} is not a system of right coset
representatives (see earlier Example).
Example G = Z H = 3Z

Z = (0 + 3Z) ∪ (1 + 3Z) ∪ (2 + 3Z) (disjoint)

and so |Z : 3Z| = 3 and |Z| = |3Z| = ∞.

85 / 146
Example An ≤ Sn and consists of all the even permutations.
Now consider the coset (12)An . Clearly if π ∈ (12)An then π is an
odd permutation. On the other hand if σ ∈ Sn is an odd
permutation then τ = (12)σ is even and so τ ∈ An . Thus
σ = (12)(12)σ = (12)τ ∈ (12)An . Therefore there are exactly two
left cosets of An in Sn , namely An (even) and (12)An (odd). Any
subset of Sn with 2 elements, one an even permutation, the other
an odd permutation, will be a system of left coset representatives
of An in Sn . We have |Sn : An | = 2, |Sn | = n! and |An | = 12 |Sn |.
Thus

˙
Sn = An ∪(12)An.
even odd

86 / 146
87 / 146
88 / 146
Example Consider the subgroup
SLn (R) = {A ∈ GLn (R) | det(A) = 1} of GLn (R).
For x ∈ R∗ = R\{0}, consider the n × n diagonal matrix
 
x 0 ··· 0
.. 
 0 1 ...

. 
Ax =  .. . . ..


 . . . 0 
0 ··· 0 1

We have det(Ax ) = x 6= 0, so Ax ∈ GLn (R). In fact Ax Ay = Axy

so Ax Ax−1 = Axx−1 = A1 = In ⇒ A−1 x = Ax−1 .

89 / 146
Now B ∈ Ax SLn (R) ⇒ B = Ax C where det C = 1

⇒ det B = det(Ax C) = det Ax det C = x.

Clearly then if x 6= y then Ax SLn (R) ∩ Ay SLn (R) = ∅.

Let B ∈ GLn (R) and let det B = x. Then
det Ax−1 B = det Ax−1 det B = x−1 x = 1
⇒ Ax−1 B = C ∈ SLn (R) ⇒ B = (Ax−1 )−1 C = Ax C ∈
Ax SLn (R).
It follows that {Ax : x ∈ R\{0}} is a set of left coset
representatives for SLn (R) in GLn (R) and
|GLn (R) : SLn (R)| = ∞. Thus

˙ x SLn (R) : x ∈ R\{0}}.

GLn (R) = ∪{A

90 / 146
91 / 146
92 / 146
Example Let p be a prime integer and let G be a group of order
p. If H is a subgroup of G then by Lagrange |H| divides p. Thus
either |H| = 1 and H = {eG } or |H| = p and H = G. Moreover,
let g ∈ G\{eG }. Then |g| divides p by Corollary 1.4.6. Since
|g| =
6 1 it follows that |g| = p ⇒ hgi = G and so G is cyclic.

Conclusion |G| = p prime ⇒ G ∼

= Cp .
Thus: for each prime p there is (up to isomorphism) exactly one
group of order p, namely Cp .

93 / 146
Given two subsets A and B of a group G we define the product
AB of A and B by

AB = {ab : a ∈ A, b ∈ B} ⊆ G.

Theorem 1.4.7 Let A and B be subgroups of G. Then AB is a

subgroup of G if and only if AB = BA.
Proof. Suppose that AB ≤ G. Let ab ∈ AB. Then
c = (ab)−1 ∈ AB and so we can write c = a1 b1 for some a1 ∈ A,
b1 ∈ B. Thus ab = c−1 = (a1 b1 )−1 = b−1 −1
1 a1 ∈ BA. This shows
AB ⊆ BA. To show the reverse inclusion let ba ∈ BA. Then
ba = (a−1 b−1 )−1 ∈ AB since AB is a subgroup. It follows that
BA ⊆ AB.

94 / 146
95 / 146
96 / 146
Conversely, suppose that AB = BA. We check the subgroup
axioms for AB. Since both A and B contain eG it follows that
eG = eG eG ∈ AB and (S1) holds. Let a1 b1 and a2 b2 ∈ AB. Then
b1 a2 ∈ BA = AB and so b1 a2 = a3 b3 for some a3 ∈ A, b3 ∈ B.
Therefore
(a1 b1 )(a2 b2 ) = a1 (b1 a2 )b2 = a1 (a3 b3 )b2 = (a1 a3 )(b3 b2 ) ∈ AB and
(S2) holds. Finally if ab ∈ AB then (ab)−1 = b−1 a−1 ∈ BA = AB
and (S3) holds. 
Example If A ≤ G then Ak = A for all k ≥ 1.

97 / 146
WARNING AB = BA does not mean ab = ba. It means
ab = b1 a1 for some a1 ∈ A, b1 ∈ B.
Theorem 1.4.8 Let A and B be finite subgroups of a group G.
Then
|A||B|
|AB| = .
|A ∩ B|

Proof. We omit the proof. 

98 / 146
99 / 146
100 / 146
 1 Groups
1.5 Normal Subgroups and Quotient Groups
Let x be an element of a group G. Any element of G of the form
gxg −1 where g ∈ G is called a conjugate of x and is denoted by
xg . Note: |x| = |gxg −1 |.
For any subgroup H of G and any g ∈ G the subset
H g = {ghg −1 : h ∈ H} is called a conjugate of H.
Lemma 1.5.1 H g ≤ G
Proof

(S1) eG = geG g −1 ∈ H g .
(S2) If x, y ∈ H g then x = gh1 g −1 and y = gh2 g −1 for some
h1 , h2 ∈ H and so xy = gh1 g −1 gh2 g −1 = g(h1 h2 )g −1 ∈ H g .
(S3) If z ∈ H g then z = ghg −1 for some h ∈ H and
z −1 = (ghg −1 )−1 = gh−1 g −1 ∈ H g . 
101 / 146
A subgroup N of G is normal and we write N E G if and only if

gng −1 ∈ N (∀g ∈ G, ∀n ∈ N ) (closed under conjugation).

Proposition 1.5.2 The following conditions on a subgroup N of

G are equivalent:

(a) N E G;
(b) gN = N g (∀g ∈ G).

Proof (a) ⇒ (b) Suppose N E G. Let x ∈ gN . Then x = gn for

some n ∈ N . Thus xg −1 = gng −1 = n̂1 for some n̂1 ∈ N and so
x = n̂1 g ∈ N g. Therefore gN ⊆ N g. Now let y ∈ N g. Then
y = ng for some n ∈ N . Thus
g −1 y = g −1 ng = g −1 n(g −1 )−1 = n̂2 for some n̂2 ∈ N and so
y = gn̂2 ∈ gN . Therefore N g ⊆ gN whence equality.
(b) ⇒ (a) Now let gN = N g (∀g ∈ G). If then g ∈ G and n ∈ N
then gng −1 = (gn)g −1 = n̂gg −1 = n̂ ∈ N so N E G. 
102 / 146
103 / 146
104 / 146
Remarks

1. Thus if N E G then we need only talk about the cosets of N

in G.
2. gN = N g does not mean that, given g ∈ G and n ∈ N ,
gn = ng. It means that gn ∈ N g, that is, there is an n̂ ∈ N
such that gn = n̂g.
3. If N ≤ G then N E G iff gN = N g (∀g ∈ G) iff gN g −1 = N
(∀g ∈ G) iff N = N g (∀g ∈ G).

Example For any group G, {eG } E G and G E G. If these are

the only normal subgroups of G and G 6= {eG } then G is called a
simple group.
Example If A is an Abelian group and B ≤ A then B E A. This
is because if a ∈ A, b ∈ B then aba−1 = aa−1 b = b ∈ B.
[Warning: Abelian subgroups of a non-Abelian group need not be
normal.]

105 / 146
Example If N ≤ G and |G : N | = 2 then N E G. [Exercise
Sheet.]
As a consequence we have An E Sn . Also, the subgroup of Dn
consisting of the rotations is a normal subgroup.
Example SLn (R) E GLn (R). For if A ∈ GLn (R) and
B ∈ SLn (R) then det(ABA−1 ) = det(A) det(B) det(A−1 ) =
det(A) det(A−1 ) det(B) = det(B) = 1 ⇒ ABA−1 ∈ SLn (R).
Example H = {(1)(2)(3), (12)} is an Abelian subgroup of S3
but H is not a normal subgroup of S3 . For example
(13)(12)(13)−1 = (13)(12)(13) = (23) ∈ / H.
Example (Kernels are normal subgroups.) If θ : G → H is a
homomorphism then ker θ E G. We already know that ker θ ≤ G
so let g ∈ G and x ∈ ker θ. Then
θ(gxg −1 ) = θ(g)θ(x)θ(g −1 ) = θ(g)eH θ(g −1 ) = θ(g)θ(g −1 ) =
θ(g)[θ(g)]−1 = eH ⇒ gxg −1 ∈ ker θ.

106 / 146
107 / 146
108 / 146
Lemma 1.5.3 Let N E G and H ≤ G.

(i) H ∩ N E H.
(ii) HN ≤ G.
(iii) N E HN .

Proof.

(i) Clearly H ∩ N ≤ H since H ∩ N ≤ G and H ∩ N ⊆ H. Let

x ∈ H ∩ N and h ∈ H. Since x ∈ H it follows that
hxh−1 ∈ H. But N E G and x ∈ N implies hxh−1 ∈ N .
Thus hxh−1 ∈ H ∩ N as required.
(ii) Exercise sheet.
(iii) N E G and N ⊆ HN ⇒ N E HN . 

109 / 146
We turn now to a fundamental construction – the quotient group.
Let G be a group and let N E G. Let

G/N denote the set of cosets gN of N in G

and define a binary operation on G/N by

(xN )(yN ) := (xy)N = xyN.

Clearly for this to make any sense we must show that it is

independent of the choice of coset representative. If so we say that
the binary operation is well-defined. Thus, suppose that
xN = x0 N and yN = y 0 N . We must check that
xN yN = x0 N y 0 N , that is, xyN = x0 y 0 N . But
xN = x0 N ⇒ x0 = x0 e ∈ x0 N = xN ⇒ x0 = xn1 for some
n1 ∈ N ; and yN = y 0 N ⇒ y 0 = yn2 for some n2 ∈ N . Hence
x0 y 0 = x(n1 y)n2 = x(yn3 )n2 for some n3 ∈ N , since N E G, and
so x0 y 0 N = xyn3 n2 N = xyN , as required.

110 / 146
111 / 146
112 / 146
Theorem 1.5.4 Let G be a group and N E G. Then the set
G/N of cosets of N in G together with the binary operation
xN yN := xyN is a group – called the quotient group (or factor
group) of G by N ; and |G/N | = |G : N |.
Proof. We need only check the group axioms. Clearly the binary
operation is closed since xyN ∈ G/N . For associativity observe
that (xN yN )zN = xyN zN = (xy)zN = x(yz)N = xN yzN =
xN (yN zN ). The identity eG/N of G/N is given by
eG/N = eG N = N eG = N ; and the inverse of xN is x−1 N , that
is, (xN )−1 = x−1 N . 
Example nZ E Z and we can form the quotient to obtain
hZ/nZ, +i whose elements are {[0], [1], . . . , [n − 1]} which, for
ease of notation, we usually write as {0, 1, 2, . . . , n − 1}. Note
then that [j] = j + nZ and so

[j] + [k] = (j + Z) + (k + Z) = (j + k) + Z = [j + k].

113 / 146
 √
Example hR/Z, +i, | 34 + Z| = 4 and | 2 + Z| = ∞.

Note |G| < ∞ ⇒ |G/N | = |G : N | = |G|/|N |.

Example G = S3 , N = h(123)i = {(1)(2)(3), (123), (132)}.
Then |G : N | = 2 ⇒ N E G. The elements of G/N are N and
(12)N = {(12), (23), (13)}. The Cayley table for G/N is

N (12)N
N N (12)N
(12)N (12)N N

and G/N ∼
= C2 .

114 / 146
115 / 146
116 / 146
Example Let G = C12 × C4 where C12 = hxi and C4 = hyi.

N = hx3 i × hy 2 i.

Then N E G since G is Abelian. Moreover

hx3 i = {1, x3 , x6 , x9 } and hy 2 i = {1, y 2 }.

Thus N =
{(1, 1), (1, y 2 ), (x3 , 1), (x3 , y 2 ), (x6 , 1), (x6 , y 2 ), (x9 , 1), (x9 , y 2 )}.

|N | = 4.2 = 8 ⇒ |G/N | = 48/8 = 6.

117 / 146
The elements of G/N are

N = (1, 1)N = {(1, 1), (1, y 2 ), (x3 , 1), (x3 , y 2 ), (x6 , 1), (x6 , y 2 ), (x9 , 1), (x9 , y 2 )}
(1, y)N = {(1, y), (1, y 3 ), (x3 , y), (x3 , y 3 ), (x6 , y), (x6 , y 3 ), (x9 , y), (x9 , y 3 )}
(x, 1)N = {(x, 1), (x, y 2 ), (x4 , 1), (x4 , y 2 ), (x7 , 1), (x7 , y 2 ), (x10 , 1), (x10 , y 2 )}
(x, y)N = {(x, y), (x, y 3 ), (x4 , y), (x4 , y 3 ), (x7 , y), (x7 , y 3 ), (x10 , y), (x10 , y 3 )}
(x2 , 1)N = {(x2 , 1), (x2 , y 2 ), (x5 , 1), (x5 , y 2 ), (x8 , 1), (x8 , y 2 ), (x11 , 1), (x11 , y 2 )}
(x2 , y)N = {(x2 , y), (x2 , y 3 ), (x5 , y), (x5 , y 3 ), (x8 , y), (x8 , y 3 ), (x11 , y), (x11 , y 3 )}

A system of coset representatives for N in G is

{(1, 1), (1, y), (x, 1), (x, y), (x2 , 1), (x2 , y)}.

118 / 146
119 / 146
120 / 146
The Cayley table for G/N is

(1, 1)N (1, y)N (x, 1)N (x, y)N (x2 , 1)N (x2 , y)N
(1, 1)N (1, 1)N (1, y)N (x, 1)N (x, y)N (x2 , 1)N (x2 , y)N
(1, y)N (1, y)N (1, 1)N (x, y)N (x, 1)N (x2 , y)N (x2 , 1)N
(x, 1)N (x, 1)N (x, y)N (x2 , 1)N (x2 , y)N (1, 1)N (1, y)N
(x, y)N (x, y)N (x, 1)N (x2 , y)N (x2 , 1)N (1, y)N (1, 1)N
(x2 , 1)N (x2 , 1)N (x2 , y)N (1, 1)N (1, y)N (x, 1)N (x, y)N
(x2 , y)N (x2 , y)N (x2 , 1)N (1, y)N (1, 1)N (x, y)N (x, 1)N

Note that (1, 1)N has order 1; (1, y)N has order 2; (x, 1)N and
(x2 , 1)N have order 3; (x, y)N and (x2 , y)N have order 6. Thus
G/N ∼ = C6 .

121 / 146
An example of multiplication in G/N is the following.

(x2 , y)N (x2 , y)N = (x4 , y 2 )N

= (x4 , y 2 ){(1, 1), (1, y 2 ), (x3 , 1), (x3 , y 2 ),
(x6 , 1), (x6 , y 2 ), (x9 , 1), (x9 , y 2 )}
= {(x4 , y 2 ), (x4 , 1), (x7 , y 2 ), (x7 , 1),
(x10 , y 2 ), (x10 , 1), (x, y 2 ), (x, 1)}
= (x, 1)N.

Alternatively (better):

(x2 , y)(x2 , y)N = (x4 , y 2 )N

= (x, 1)(x3 , y 2 )N = (x, 1)N

since (x3 , y 2 ) ∈ N .

122 / 146
123 / 146
124 / 146
We have already seen that the kernel of a homomorphism is a
normal subgroup. The next result gives the converse.
Theorem 1.5.5 Let G be a group and N E G. Then the natural
map φ : G → G/N defined by φ(g) = gN (∀g ∈ G) is a surjective
homomorphism with ker φ = N .
Proof. Clearly φ is surjective and g ∈ ker φ if and only if
φ(g) = eG/N = N iff gN = N iff g ∈ N . Moreover
φ(g1 g2 ) = g1 g2 N = g1 N g2 N = φ(g1 )φ(g2 ). 

125 / 146
We conclude this section with an important bijection concerning
subgroups of quotient groups.
Theorem 1.5.6 (The Correspondence Theorem) Let N be a
normal subgroup of a group G. Then every subgroup of the
quotient group G/N is of the form H/N where N E H ≤ G.
Conversely if H ≤ G and H contains N then H/N ≤ G/N . The
correspondence between subgroups of G/N and subgroups of G
containing N is a bijection. The bijection maps normal subgroups
of G/N onto normal subgroups of G containing N .
Proof We omit the proof. 

126 / 146
127 / 146
128 / 146
Example (Exercise Sheet) G = D4 and N = h(13)(24)i. Then
N E G and |G/N | = 8/2 = 4. In fact G/N ∼ = C2 × C2 . (Recall
that |K| = 4 ⇒ K Abelian.) Now C2 × C2 has a total of five
subgroups so by the Correspondence Theorem D4 has a total of
five subgroups containing N , namely D4 , N and three others.

129 / 146
 1 Groups
1.6 Isomorphism Theorems

We begin this section with another fundamental result.

Theorem 1.6.1 (First Isomorphism Theorem) Let
φ : G → H be a homomorphism. Then we have G/ ker φ ∼
= Im φ.
Proof. We have already seen that K = ker φ E G and Im φ ≤ H.
Define θ : G/K → Im φ by θ(xK) = φ(x) (∀x ∈ G). First we
must show that θ is well-defined. Let xK = yK, so that x = yk
for some k ∈ K. Then

θ(xK) = φ(x) = φ(yk) = φ(y)φ(k) = φ(y)eH = φ(y) = θ(yK)

as required.

130 / 146
131 / 146
132 / 146
Next, for all x, y in G

θ(xK)θ(yK) = φ(x)φ(y) = φ(xy) = θ(xyK) = θ((xK)(yK))

so θ is a homomorphism.
Let z ∈ Im φ. Then φ(x) = z for some x ∈ G which implies
θ(xK) = z and so θ is surjective. If θ(xK) = θ(yK) then
φ(x) = φ(y) ⇒ [φ(y)]−1 φ(x) = eH ⇒ y −1 x ∈ K ⇒ xK = yK by
Lemma 1.4.2. Thus θ is injective and so is an isomorphism. 
General principle: if a homomorphism θ is defined on cosets then
you must check that θ is well-defined.

133 / 146
Example Consider θ : GLn (R) → R× = hR\{0}, ×i defined by
θ(A) = det A. Then Im θ = R× and ker θ = SLn (R) and so
GLn (R)/SLn (R) ∼
= R× .
Example Consider S 1 = {z ∈ C : |z| = 1} = {e2πix : x ∈ R} the
unit circle in the complex plane.
Then S 1 is a subgroup of C× = hC\{0}, ×i. Consider the map

exp : hR, +i → S 1 ,
x → e2πix .

Now exp(x + y) = e2πi(x+y) = e2πix e2πiy = exp(x) exp(y) which

shows that exp is a homomorphism. Clearly exp is surjective and
exp(x) = 1 if and only if e2πix = 1 if and only if x ∈ Z, hence
ker(exp) = Z and so R/Z ∼ = S1.

134 / 146
135 / 146
136 / 146
Theorem 1.6.2 (Second Isomorphism Theorem) Let
H ≤ G and N E G. Then N E HN ≤ G, N ∩ H E H and
H ∼ HN
= .
N ∩H N
Proof. The fact that N E HN ≤ G and N ∩ H E H follow from
Lemma 1.5.3. Now define φ : H → HN/N by φ(h) = hN . Then

φ(xy) = xyN = (xN )(yN ) = φ(x)φ(y)

shows that φ is a homomorphism.

The kernel of φ is given by ker φ = {h ∈ H : φ(h) = eHN/N }
= {h ∈ H : hN = N } = {h ∈ H : h ∈ N } = H ∩ N .
Finally φ is surjective since the element hnN = hN of HN/N is
φ(h). The result now follows from Theorem 1.6.1. 

137 / 146
Theorem 1.6.3 (Third Isomorphism Theorem) Let H and
N be normal subgroups of the group G with N contained in H.
Then

(G/N )/(H/N ) ∼
= G/H.

Proof. Define a map φ : G/N → G/H by φ(gN ) = gH. We must

check that φ is well-defined. Suppose that xN = yN so that
y −1 x ∈ N . Then since N is contained in H we see that y −1 x ∈ H
and so xH = yH, that is, φ(xN ) = φ(yN ). Next

φ(xN )φ(yN ) = (xH)(yH) = xyH = φ(xyN ) = φ((xN )(yN ))

shows that φ is a homomorphism.

138 / 146
139 / 146
140 / 146
Clearly φ is surjective, that is, Im φ = G/H, and

ker φ = {gN ∈ G/N : φ(gN ) = eG/H }

= {gN ∈ G/N : gH = H}
= {gN ∈ G/N : g ∈ H} = H/N

and the result now follows from Theorem 1.6.1. 

141 / 146
Example Consider On (R) = {A ∈ GLn (R) : AA> = In } where
A> denotes the transpose of the n × n matrix A. Then
On (R) ≤ GLn (R) and is called the orthogonal group of degree n
with coefficients in R.

(S1) In In> = In ⇒ In ∈ On (R)

(S2) A, B ∈ On (R) ⇒ (AB)(AB)> = ABB > A>

= AIn A> = AA> = In
⇒ AB ∈ On (R)

(S3) Let A ∈ On (R). Observe that

AA> = In ⇒ A> = A−1 ⇒ A> A = In

and so
(A−1 )(A−1 )> = A> (A> )> = A> A = In ⇒ A−1 ∈ On (R).

142 / 146
143 / 146
144 / 146
Consider the group homomorphism

det : On (R) → R\{0} A → det(A)

whose kernel is denoted by SOn (R) the special orthogonal group

of degree n. In fact SOn (R) = SLn (R) ∩ On (R). Now
 
−1 0
 1 
det(In ) = 1 det   = −1
 
..
 . 
0 1

145 / 146
shows that {1, −1} ⊆ Im(det). But

A ∈ On (R) ⇒ In = AA>
⇒ 1 = det(AA> ) = det(A) det(A> ) = [det(A)]2
⇒ det(A) ∈ {1, −1}
⇒ Im(det) ⊆ {1, −1}.

Thus the image is h{1, −1}, ×i ∼

= C2 so by the First Isomorphism
Theorem
On (R) On (R) ∼
= = Im(det) ∼
= C2
SOn (R) ker(det)

and by the Second Isomorphism Theorem

SLn (R)On (R) ∼ On (R) ∼ On (R) ∼

= = = C2 .
SLn (R) (SLn (R) ∩ On (R)) SOn (R)

146 / 146