M.

SC DISSERTATION

ON

ORDINAL SPACE TOPOLOGY

Under the supervision of :

Dr. Atasi Deb Ray
Department of Pure Mathematics
University of Calcutta
Kolkata

Submitted by :
Manali Sajjan
Sayantan Banerjee

Sayantan Banerjee
Roll No:91/PMT/171007
Registration No:A01-1112-0622-14
 DECLARATION

I do hereby declare that Ms.Manali Sajjan and Mr.Sayantan Banerjee have done their Dissertation under
my guidance and this project report entitled ”ORDINAL SPACE TOPOLOGY” has been submitted
for partial fulfillment of the dissertation in M.Sc 4th Semester for the year 2019 at the Department of Pure
Mathematics, University of Calcutta.

Dr. Atasi Deb Ray

Department of Pure Mathematics
University of Calcutta
Kolkata

1
 ACKNOWLEDGEMENTS

We take immense pleasure in thanking our supervisor Dr.Atasi Deb Ray, Associate Professor,Department
of Pure Mathematics, University of Calcutta, for giving us a wonderful opportunity to carry out this project
work.

We wish to express our deep sense of gratitude to all the faculty members, Department of Pure Math-
ematics, University of Calcutta.

Finally, We would like to express our heartfelt thanks to our beloved parents for their blessings and to
our friends for their help and wishes for the successful completion of this project.

Manali Sajjan
Sayantan Banerjee
students
Department of Pure Mathematics
University of Calcutta
Kolkata

2
 ORDINAL SPACE TOPOLOGY

June 18, 2019

1 Introduction

When we try to find counterexamples in Topology we take a large class of topological spaces satisfying
certain topological properties and then look for the spaces which satisfy some of those properties and do
not satisfy the rest of them. For example, we take the class of all metrizable spaces. A metrizable space
is both 1st countable and completely normal. We find that there are also non metrizable spaces which are
both 1st countable and completely normal. Sorgenfrey line(Rl ) is one such topological space which is non
metrizable but satisfies both 1st countability and T5 ness.
Clearly, if we try to find topolgical spaces which are completely normal but do not satisfy 1st countability
those spaces must be non metrizable. In this dissertation, our aim is to construct such a non metrizable
topological space which is T5 but does not satisfy 1st countability.

2 Linear and Partial Ordering

A binary relation < on a set P is a partial ordering of P if

1. p < p is not true for any p ∈ P

2. if p < q and q < r then p < r

(P, <) is called a partially ordered set.

A partial ordering < of P is a linear ordering if moreover p < q or p = q or q < p for all p, q ∈ P
If < is a partial(linear) ordering, then the relation ≤ (where p ≤ q if either p < q or p = q) is also called a
partial(linear) ordering and < is sometimes called a strict ordering.

Definition If (P, <) is a partially ordered set , X is a non empty subset of P and a ∈ P then

• a is a maximal element of X if a ∈ X and for all x ∈ X a < x is not true.

• a is a minimal element of X if a ∈ X and for all x ∈ X x < a is not true.

3
 • a is the greatest element of X if a ∈ X and for all x ∈ X x ≤ a

• a is the least element of X if a ∈ X and for all x ∈ X a ≤ x

• a is an upper bound of X if for all x ∈ X x ≤ a

• a is an lower bound of X if for all x ∈ X a ≤ x

• a is the supremum of X if a is the least upper bound of X

• a is the infimum of X if a is the greatest lower bound of X

If X is linearly ordered by < then a maximal element of X is its greatest element.

If (P, <) and (Q, <) are partially ordered sets and f : P → Q then f is order preserving if

x < y ⇒ f (x) < f (y)

A linear ordering < of a set P is a well ordering if every non empty subset of P has a least element.

3 The Order Topology

Let X be a linearly ordered set having the order relation ≤. Given elements a and b of X such that a < b
there are four subsets of X that are called the intervals determined by a and b. They are the following

(a, b) = {x | a < x < b}

(a, b] = {x | a < x ≤ b}
[a, b) = {x | a ≤ x < b}
[a, b] = {x | a ≤ x ≤ b}
Here a set of the first type is called an open interval in X a set of the last type is called a closed interval
in X and sets of the second and third types are called half open intervals.

Let X be a set with a linear order relation; assume that X has more than one element. Let B be the
collection of all sets of the following types:

1. All open intervals (a, b) in X

2. All intervals of the form [a0 , b) where a0 is the least element(if any) of X

3. All intervals of the form (a, b0 ] where b0 is the greatest element (if any) of X

The collection B is a basis for a topology on X. Indeed, every element of X lies in at least one element of
B. The least element (if any) lies in all sets of type (2), the greatest element (if any) lies in all sets of type
(3) and every other element lies in a set of type (1). Also the intersection of any two sets of the preceding
types is again a set of one of those types, or empty.
The generated topology on X by B is called the order topology on X

4
 If X is an ordered set and a is an element of X, there are four subsets of X that are called the rays
determined by a.They are the following:

(a, +∞) = {x | x > a}

(−∞, a) = {x | x < a}
[a, +∞) = {x | x ≥ a}
(−∞, a] = {x | x ≤ a}
Sets of the first two types are called open rays and sets of the last two types are called closed rays. In
fact, open rays in X are open sets in the order topology. Same applies for the closed rays also. Consider
the ray (a, +∞). If X has a greatest element b0 then (a, +∞) equals the basis element (a, b0 ].If X has no
greatest element then (a, +∞) equals the union of all basis elements of the form (a, x) for x > a. In either
case (a, +∞) is open.
The open rays, in fact, form a subbasis for the order topology on X. Because the open rays are open in the
order topology, the topology they generate is contained in the order topology. On the other hand, every
basis element for the order topology equals a finite intersection of open rays; the interval (a, b) equals the
intersection of (−∞, b) and (a, +∞) while (a, b0 ] and [a0 , b) ,if they exist, are themselves open rays. Hence
the topology generated by the open rays contains the order topology.

Theorem 3.1 A linear ordered topological space is Hausdorff.

Proof Suppose (X, ≤) is a linear ordered topological space; and let p, q be distinct element of X. Without
loss of generality, we assume that p < q. If there exists an element a such that p < a < q then (−∞, a)
and (a, +∞) separate the two points , and we are done. If there is no such a then (−∞, q) = (−∞, p] and
(p, +∞) = [q, +∞) are disjoint open sets containing p,q respectively.

Theorem 3.2 A linearly ordered topological space is completely normal.

Proof Suppose (X, τ, ≤) is a linear ordered topological space and let H and K be separated subsets of X.
(H ∩ cl(K) = cl(H) ∩ K = φ). For each x ∈ H, there is a convex set Vx ∈ τ such that x ∈ Vx ⊂ X − K.
S each y ∈ K thereSis a convex set Vy ∈ τ such that y ∈ Vy ⊂ X − H
Similarly, for
Let VH = x∈H Vx and VK = y∈K Vy . Clearly H ⊂ VH and K ⊂ VK . Also, VH ∩ VK ⊂ X − (H ∪ K).
Indeed for each x ∈ H, Vx ⊂ X − K so that VH ⊂ X − K. Similarly VK ⊂ X − H and hence
VH ∩ VK ⊂ (X − H) ∩ (X − K) = X − (H ∪ K).
Let V = VH ∩ VK . If V = φ then we are done. So suppose that V 6= φ.
Define a relation ∼ on V by p ∼ q iff [min{p, q}, max{p, q}] ⊂ V . Clearly p ∼ p and if p ∼ q then q ∼ p.
Let p ∼ q and q ∼ r. Without loss of generality let p < q and q < r. Then [p, q] ⊂ V and [q, r] ⊂ V .
Since V is the intersection of union of convex sets it follows that [p, r] ⊂ V . Therefore ∼ is an equivalence
relation.
Let T ⊂ V contains exactly one point of each sim class. Suppose that x ∈ H and p, q ∈ Vx ∩ T with
p < q.
Claim : p < x < q
Suppose that x < p. Since p ∈ T ⊂ V there is a y ∈ K such that p ∈ Vy ; [x, q] ⊂ Vx ⊂ X − K so that
y∈/ [x, q]. If y < x then [p, q] ⊂ Vx ∩ Vy ∩ V so p ∼ q.
But that contradicts the choice of p and q showing that p < x. A similar argument shows that x < q.
Similarly if y ∈ K and p, q ∈ Vy ∩ T with p < q then p < y < q. Hence it follows immediately that the
cardinality of Vx ∩ T ≤ 2 for all x ∈ H ∩ K.
Now fix p ∈ T . Let Hp = {x ∈ H| p ∈ Vx } and Kp = {y ∈ K| p ∈ Vy }. Clearly Hp 6= Kp 6= φ , since
p ∈ V = VH ∩ VK . Suppose that x < p for some x ∈ Hp . If y ∈ Kp and y < p then either x < y and y ∈ Vx

5
or y < x and x ∈ Vy , since the sets Vx and Vy are convex but neither is possible so p < y. Since y ∈ Kp is
arbitrary hence it follows that p < Kp . A similar argument shows that Hp < p and hence Hp < p < Kp .
If instead p < x for some x ∈ Hp it follows similarly that Kp < p < Hp .
For each x ∈ H ∪ K define Wx ∈ τ as follows
Wx = Vx , if Vx ∩ T = φ
= Vx ∩ (p, ∞), if Vx ∩ T = {p} and p < x
= Vx ∩ (−∞, p), if Vx ∩ T = {p} and x < p
= Vx ∩ (p, q), if Vx ∩ T = {p, q} and p < x < q
S S
Let WH = x∈H Wx and WK = x∈K Wx
By construction WH and WK are open. Also H ⊂ WH , K ⊂ WK .
Claim : WH ∩ WK = φ
Suppose not, then there are x ∈ H and y ∈ K such that Wx ∩ Wy 6= φ.
Without loss of generality suppose that x < y. Fix q ∈ Wx ∩ Wy . Then x < q < y. If not, then suppose
that q < x. Since q ∈ Wy and q < x < y and also Wy is convex it follows that x ∈ Wy which is a
contradiction since x ∈ Wy ⊂ Vy ⊂ X − H. Similarly we arrive at a contradiction assuming that y < q.
Therefore x < q < y. Moreover q ∈ Wx ∩ Wy ⊂ Vx ∩ Vy ⊂ V . So q ∼ p for a unique p ∈ T . Let I be the
closed interval with endpoints p and q. Then I ⊂ V ⊂ X − (H ∪ K). So x, y ∈ / I and therefore x < p < y.
If p 6 q then p ∈ Wx ∪ T ⊂ Vx ∪ T . By construction Wx ⊂ (−∞, p) and so p ∈
/ Wx which is a contradiction.
If on the other hand q 6 p then p ∈ Wy ∩ T ⊂ Vy ∩ T so that Wy ⊂ (p, ∞) and p ∈ / Wy which is again a
contradiction.
It follows that WH ∩ WK = φ and hence the proof is complete.

4 Compactness and Connectedness in the Order Topology

We will define an inductive subset of an ordered set and show that an ordered set X is Dedekind Complete
iff the only inductive subset of X is X itself. The result will be used in characterization of connectedness
and compactness in order topology.

Theorem 4.1 Principle of Real Induction Let S ⊂ [a, b] and suppose

(i) a ∈ S
(ii) for all x ∈ S if x 6= b there exists y > x such that [x, y] ⊂ S
(iii) for all x ∈ R IF [a, x) ⊂ S then x ∈ S
Then S = [a, b].

Definition X is said to be complete if every subset of X has a supremum and an infimum. If X is

complete then the least element of X is denoted as 0 and the greatest element of X is denoted as 1. X is
said to be Dedekind complete if every non empty bounded above subset has a supremum, equivalently if
every non empty bounded below subset has an infimum. We call this the least upper bound property. X
is complete iff it is Dedekind complete and has 0 and 1.
We will say a subset S ⊂ X is inductive if it satisfies all of the following
(I) There exists a ∈ X such that (−∞, a) ⊂ S
(II) For all x ∈ S either x = 1 or there exists y > x such that [x, y] ⊂ S.
(III) For all x ∈ X if (−∞, x) ⊂ S then x ∈ S

Theorem 4.2 For a linearly ordered set X T F AE

(i) X is Dedekind complete.
(ii) The only inductive subset of X is X itself.

6
Proof (i)⇒(ii) Let S ⊂ X be inductive. Seeking a contradiction we suppose that S 0 = X − S is non
empty. Fix a ∈ X satisfying (I). Then a is a lower bound for S 0 . So by hypothesis S 0 has an infimum say
y. Any element less than y is strictly less than every element of S 0 so (−∞, y) ⊂ S. By (III) y ∈ S. If
y = 1, then S 0 = 1 or S 0 = φ. But according to our assumption S 0 is non empty. Also (−∞, 1) ⊂ S and
S is inductive so 1 ∈ S which is again a contradiction. So y < 1. Then by(II) there exists z > y such
that [y, z] ⊂ S and thus (−∞, z] ⊂ S. Thus z is a lower bound for S 0 which is strictly larger than y, a
contradiction.
(ii)⇒(i) Let T ⊂ X be non empty and bounded below by a. Let S be the set of lower bounds of T . Then
(−∞, a] ⊂ S. So S satisfies (I).
Case 1: Suppose S does not satisfy (II). there is x ∈ S with no y ∈ X such that [x, y] ⊂ S. Since S is
downward closed, x = inf (T )
Case 2: Suppose S does not satisfy (III). There is x ∈ X such that (−∞, x) ⊂ S but x ∈ / S i.e, there exists
t ∈ T such that t < x. Then also t ∈ S so t is the least element of T i.e, t = inf (T ).
Case 3: If S satisfies (I) and (II) then S = X. So T = 1 and inf (T ) = 1

Definition A linearly ordered set X is dense if for all a < b ∈ X there exists e with a < e < b.

Theorem 4.3 For a linearly ordered set X T F AE :

(i) X is dense and Dedekind complete.
(ii) X is connected in the order topology

Proof (i)⇒(ii): We suppose 0 ∈ X. Let S ⊂ X contains 0 and is both open and closed in the order
topology. Hence S satisfies (I). Now let x ∈ S and x 6= 1. Since S is open, there exists y ∈ X such that
[x, y) ⊂ S. Now since X is dense, there exists z such that x < z < y so that [x, z] ⊂ S. S satisfies (II).
Also since S is closed for x ∈ S if (−∞, x) ⊂ S then x ∈ S. Then by previous theorem S = X i.e; X is
connected.
(ii)⇒(i): Suppose we have S ⊂ X non S empty bounded below by a and with no infimum. Let L be the set
of lower bounds for S and put U = b∈L (−∞, b). So U is open and U < S. We have a 6= inf (S) so a ∈ U
and thus U 6= φ. If x ∈/ U then x ≥ L and indeed since L has no maximal element x > L so there exists
s ∈ S such that s < x. Since the order is dense there is y with s < y < x and then the entire open set
(y, ∞) lies in the complement of U . Thus U is also closed. Since X is connected, U = X contradicting
U < S. Also if the order is not dense there are a < b in X with [a, b] = {a, b}, so A = (−∞, a] B = [b, ∞)
is a separation of X.

Theorem 4.4 For a non empty linearly ordered set X T F AE

(i) X is complete
(ii) X is compact in the order topology.

Proof (i)⇒(ii): Let U = {Ui }i∈I be an open covering of X. Let S be the set of x ∈ X such that the covering
U ∩ [0, x] of [0, x] admits a finite subcovering . 0 ∈ S so S satisfies (I). Suppose U1 ∩ [0, x]........Un ∩ [0, x]
covers [0, x]. If there exists y ∈ X such that [x, y] = {x, y} then adding to the covering any element Uy
containing y gives a finite covering of [0, y]. Otherwise some Ui contains x and hence also [x, y] for some
y > x. So S satisfies (II). Now suppose that (−∞, x) ⊂ S. Let ix ∈ I beSsuch that xS∈ Uix and let y < x
be such that (y, x] ⊂ Uix . Since y ∈ S there is a finite J ⊂ I with [0, y] ⊂ i∈J Ui . So i∈J Ui ∪ Uix ⊃ [0, x].
Thus x ∈ S and S satisfies (III). Thus S is an inductive subset of the dedekind complete ordered set X so
S = X.
(ii)⇒(i): If possible let X be not complete. Then there exists a set A ⊂ X such that A has no least upper
bound, then the sets {x | x < α} and {x | x > β} for α ∈ A and β is an upper bound of A cover X
but they contain no finite subcover which is a contradiction to our assumption.

7
Theorem 4.5 Let X be a linearly ordered set having the least upper bound property. In the order topology,
each closed interval in X is compact.

Proof Step 1: Given a < b, let A be a covering of [a, b] by sets open in [a, b] in the subspace topology
(which is same as the order topology). We wish to prove the existence of a finite subcollection of A covering
[a, b]. First we prove the following : if x is a point of [a, b] different from b then there is a point y > x of
[a, b] such that the interval [x, y] can be covered by at most two elements of A.
If x has an immediate successor in X let y be this immediate successor. Then [x, y] consists of the two
points x and y so that it can be covered by at most two elements of A. If x has no immediate successor in
X choose an element A0 of A containing x. Because x 6= b and A0 is open, A0 contains an interval of the
form [x, c) for some c in [a, b]. Choose a point y in (x, c); then the interval [x, y] is covered by the single
element A0 of A.
Step 2: Let C be the set of all points y > a of [a, b] such that the interval [a, y] can be covered by finitely
many elements of A. Applying Step 1 to the case x = a, we see that there exists at least one such y, so C
is not empty. Let c be the least upper bound of the set C; then a < c 6 b.
Step 3: We show that c belongs to C; that is, we show that the interval [a, c] can be covered by finitely
many elements of A. Choose an element A0 of A containing c; since A0 is open, it contains an interval
of the form (d, c] for some d in [a, b]. If c is not in C, there must be a point z of C lying in the interval
(d, c) because otherwise d would be a smaller upper bound on C than c. Since z is in C, the interval [a, z]
can be covered by finitely many , say n, elements of A. Now [z, c] lies in the single element A0 of A hence
[a, c] = [a, z] ∪ [z, c] can be covered by n + 1 elements of A. Thus c is in C, contrary to assumption.
Step 4: Finally, we show that c = b, and our theorem is proved. Suppose that c < b. Applying Step 1
to the case x = c we conclude that there exists a point y > c of [a, b] such that the interval [c, y] can be
covered by finitely many elements of A. We proved in Step 3 that c is in C so [a, c] can be covered by
finitely many elements of A. Therefore, the interval

[a, y] = [a, c] ∪ [c, y]

can also be covered by finitely many elements of A. This means that y is in C, contradicting the fact that
c is an upper bound of C.

5 Construction of the Ordinal Spaces SΩ and SΩ

We recall that a well ordered set is a simply ordered set (A, <) for which every non empty subset contains
a least element. Every finite ordered set is well ordered. The set of positive integers Z+ and the set
{1 − 1/n | n ∈ Z+ } ∪ {2 − 1/n | n ∈ Z+ } where both are endowed with Archimedian ordering are
well ordered. On the other hand neither Z nor (0, ∞) are well ordered with the archimedian ordering.
Despite such examples, Zermelo’s theorem( which is known to be equivalent of axiom of choice) asserts
that every set admits an ordering that is a well ordering.
If A is a well ordered set then A itself contains a smallest element which we will denote by a0 . Moreover,
for each x ∈ A, if it is non empty the set {y ∈ A | y > x} must have a smallest element and this
smallest element is called an immediate successor of x. Thus every non maximal element in a well ordered
set has an immediate successor. For each element x in a well ordered set A the section at x is defined to
be the subset

Sx = (−∞, x) = [a0 , x) = {y ∈ A | y < x}

Theorem 5.1 Well Ordering theorem If A is a set there exists an order relation on A that is well ordering.

8
Corollary 5.2 There exists an uncountable well ordered set.

lemma 5.3 There exists an uncountable well ordered set A having a largest element Ω such that the section
SΩ of A by Ω is uncountable but every other section of A is countable.

Proof We begin with an uncountable well ordered set B (Existence of such a set is guaranteed by Zer-
melo’s theorem). Let C be the well ordered set 1, 2 × B in the dictionary order then some section of C
is uncountable. Indeed the section of C by any element of the form 2 × b is uncountable. Let Ω be the
smallest element of C for which the section of C by Ω is uncountable. Then we take A to consist of this
section along with the element Ω.

Note that SΩ is an uncountable well ordered set every section of which is countable.

Definition We define SΩ to be the uncountable ordinal space. It is a subset of the closed uncountable
ordinal space SΩ which is defined by SΩ = SΩ ∪ {Ω} with the well ordering given by (a) if x, y ∈ SΩ then
x < y in SΩ iff x < y in SΩ and (b) if x ∈ SΩ then x < Ω. Ω is a maximal element in SΩ but SΩ does not
have a maximal element. Also the notation SΩ is justified by the observation that SΩ is the section of Ω
in SΩ . Also with respect to the order topology the closure of SΩ in SΩ is SΩ , which justifies the use of the
bar notation

We recall that the collection of intervals of the forms (i) [a0 , y), (ii) (x, y) (iii) (x, Ω] where x, y ∈ SΩ form
a basis of the order topology on SΩ and that the intervals of the forms (i) and (ii) comprise a basis for the
order topology on SΩ .

6 Topological properties of the open ordinal space SΩ

Our objective in this section is to discuss that the open ordinal space SΩ satisfies which topological
properties and does not satisfy which of them. First, to begin with, we prove a couple of lemmas.

lemma 6.1 Every countable subset of SΩ has an upper bound in SΩ

Proof Let B be a countable subset of SΩ . Then the set B 0 = b∈B Sb is a countable set, since it is a
S
countable union of countable sets. As SΩ is uncountable, there must be an element x ∈ SΩ − B 0 . If x < b
for some b ∈ B then x ∈ Sb ⊂ B 0 which is a contradiction. Therefore x is an upper bound for B.

Definition Let (xn ) be a sequence in a linearly ordered set. An index n ≥ 1 is a peak point of (xn ) if
xn ≥ xk for all k ≥ n.

lemma 6.2 Peak point lemma Let (xn ) be a sequence in a linearly ordered set. Then (xn ) has a monotone
subsequence.

Proof If (xn ) has infinitely many peak points n1 , n2 , .... then (xnk ) is a monotone non increasing subse-
quence. Otherwise suppose that (xn ) has finitely many peak points n1 < ....... < nk . Let m1 > nk . Then
m1 is not a peak point so there is m2 > m1 such that xm1 6 xm2 . Similarly m2 is not a peak point so
there is m3 > m2 such that xm2 6 xm3 . Continuing this process we obtain a monotone non decreasing
subsequence (xmk ).

Theorem 6.3 SΩ is sequentially compact.

9
Proof Let (αn ) be a sequence in SΩ . We will show that (αn ) has a convergent subsequence.
By the peak point lemma (αn ) has a monotone subsequence (αnk ).
Case 1: (αnk ) is non decreasing.
The set {αnk | k ∈ Z+ } is countable therefore by lemma 5.4 it has an upper bound in SΩ . Therefore
the set of upper bounds of the set {αnk | k ∈ Z+ } is non empty. Due to well orderedness of SΩ it has a
least element, let us call it α. Then (αnk ) converges to α, since for any open interval around α, contains
some element of the set {αnk | k ∈ Z+ }, suppose αnk0 , therefore (αnk ) belongs to that open interval for
all nk > nk0
Case 2: (αnk ) is non increasing
Consider the set {αn1 , αn2 , .....}. Due to well orderedness of SΩ , it has a least element α. Then (αnk )
converges to α.
Hence the theorem is proved.

We now look into some other topological properties of SΩ

Theorem 6.4 SΩ is first countable

Proof Let x be an element of SΩ with immediate successor x1 .

Claim : The collection of intervals {(t, x1 ) | t ∈ Sx } is a countable neighborhood base at x.
Indeed the collection {(t, x1 ) | t ∈ Sx } is countable since Sx is countable. Also for any interval containing
x suppose (a, b) where a < x < b we get (a, x1 ) ⊂ (a, b) so the claim is justified.

Theorem 6.5 SΩ is not lindelof.

Proof Let U = {Sy | y ∈ SΩ }. Each element of the collection U is basic open set and for each x ∈ SΩ ,
x ∈ Sx1 ∈ U where x1 is the immediate successor of x. Therefore U is an open cover of SΩ . Suppose that
V is countable subcover of U . Then V = {Sy | y ∈ B} for some countable set B ⊂ SΩ . By lemma 5.4,
B has an upper bound z ∈ SΩ . But then z ∈/ Sy for each y ∈ B and this implies that z ∈
/ ∪V contradicting
our assumption that V is an open cover of SΩ . This shows that U has no countable subcover and so SΩ is
not lindelof.

Corollary 6.6 SΩ is not compact

Proof Since for compact topological spaces every open cover admits of a finite subcover and in the proof
of previous theorem we constructed an open cover which does not admit of even a countable subcover,
hence the result follows.

SΩ is an example of a sequentially compact topological space which is not compact.

Corollary 6.7 SΩ is not metrizable

Proof By immediate consequences of the fact that SΩ is sequentially compact but not compact, we
conclude that SΩ is not metrizable, since for a metrizable space, these two conditions are equivalent.

Corollary 6.8 SΩ is not second countable

Proof Since second countable spaces are always lindelof, hence the result follows.

Theorem 6.9 SΩ is not separable

10
Proof Let B be a countable subset of SΩ . By lemma 5.4, B has an upper bound z. So, B ⊂ [a0 , z]
and [a0 , z] is a closed set. Thus B ⊂ [a0 , z] is a proper subset of SΩ , since [a0 , z] is countable. Indeed,
[a0 , z] = Sz ∪ Z and Sz is countable, but SΩ is uncountable. It follows that B is not dense in SΩ and since
B is arbitrary, we therefore conclude that SΩ has no countable dense subset.

Theorem 6.10 SΩ is locally compact hausdorff space

Proof Since SΩ is linearly ordered, it follows that SΩ is hausdorff. Now to show local compactness, for
each x ∈ SΩ , a basic neighborhood of x is an interval, and this interval is contained in a closed interval.
Now by theorem 4.5, if X is a linearly ordered set having the least upper bound property, then in the order
topology, each closed interval in X is compact. Since SΩ is well ordered, it satisfies the least upper bound
property. Therefore every basic neighborhood of x is contained in a compact set in the order topology,and
hence SΩ is locally compact.

Theorem 6.11 SΩ is not connected.

Proof Let x be an element of SΩ and let x1 be its immediate successor.

Claim : [a0 , x1 ) ∩ (x, +∞) = φ
Indeed, if possible let y ∈ [a0 , x1 ) ∩ (x, +∞). y < x1 but x < y and that contradicts the fact that x1 is the
least element of the set {z ∈ SΩ | x < z}.
Both [a0 , x1 ) and (x, +∞) are open and non empty, since x ∈ [a0 , x1 ) and x1 ∈ (x, +∞). Therefore they
form a separation of SΩ showing that SΩ is not connected.

Corollary 6.12 SΩ is not path connected

Proof This follows from the previous theorem, since path connected spaces are connected.

Theorem 6.13 SΩ is not locally connected

Proof Let x be an element of SΩ which does not have an immediate predecessor. For example, x could be
chosen to be the set {y ∈ SΩ | Sy is inf inite}. Let U be a neighborhood of x. Then U must contain
an open interval I with x ∈ I and I must contain an element y < x since otherwise the left endpoint of
I would be an immediate predecessor of x. Let y1 be the immediate successor of y. Then [a0 , y1 ) ∩ U
and (y, +∞) ∩ U are open sets in the subspace topology on U which form a separation of U . Hence the
neighborhood U of x cannot be connected. We conclude that x has no connected neighborhoods and SΩ
is not locally connected.

Definition A space X is locally metrizable if each point x of X has a neighborhood that is metrizable in
the subspace topology.

Theorem 6.14 SΩ is locally metrizable.

Proof Let x be an element of SΩ with immediate successor x1 . The open subset [a0 , x1 ) = [a0 , x] = Sx1 is
countable. Therefore [a0 , x] is second countable, since intervals of the form [a0 , x) or (x, y) form a countable
family. Also [a0 , x] is a compact hausdorff space and is therefore regular. Now Urysohn metrization theorem
states that every regular space X with a countable basis is metrizable. Therefore we conclude that [a0 , x]
is metrizable. Therefore, every element of SΩ has a metrizable neighborhood and hence SΩ is locally
metrizable.

Theorem 6.15 SΩ is completely normal

Proof As SΩ is a linearly ordered topological space, by theorem 3.2, the result follows.

11
7 Topological properties of the closed ordinal space SΩ
Our main objective for this dissertation was to construct a topological space which is completely normal
but does not satisfy 1st countability. In this section we will see that the closed ordinal space SΩ is an
example of one such topological space

Theorem 7.1 SΩ is compact

Proof Since SΩ is the closed interval [a0 , Ω] and SΩ satisfies the least upper bound property, therefore by
theorem 4.5 SΩ is compact.

Corollary 7.2 SΩ is sequentially compact.

Corollary 7.3 SΩ is limit point compact.

Corollary 7.4 SΩ is lindelof.

Proof Since compact spaces are lindelof, the result follows from theorem 7.1

Theorem 7.5 SΩ is not connected.

Proof Let x be an element of SΩ and let x1 be its immediate successor.

Claim : [a0 , x1 ) ∩ (x, Ω] = φ
Indeed, if possible let y ∈ [a0 , x1 ) ∩ (x, Ω]. y < x1 but x < y and that contradicts the fact that x1 is the
least element of the set {z ∈ SΩ | x < z}.
Both [a0 , x1 ) and (x, Ω] are open and non empty, since x ∈ [a0 , x1 ) and x1 ∈ (x, Ω]. Therefore they form a
separation of SΩ showing that SΩ is not connected.

Corollary 7.6 SΩ is not path connected.

Theorem 7.7 SΩ is not locally connected.

Proof Let x be an element of SΩ which does not have an immediate predecessor. For example, x could be
chosen to be the set {y ∈ SΩ | Sy is inf inite}. Let U be a neighborhood of x. Then U must contain
an open interval I with x ∈ I and I must contain an element y < x since otherwise the left endpoint of
I would be an immediate predecessor of x. Let y1 be the immediate successor of y. Then [a0 , y1 ) ∩ U
and (y, Ω] ∩ U are open sets in the subspace topology on U which form a separation of U . Hence the
neighborhood U of x cannot be connected. We conclude that x has no connected neighborhoods and SΩ
is not locally connected.

Corollary 7.8 SΩ is not locally path connected.

Theorem 7.9 SΩ is locally compact hausdorff space

Proof Since SΩ is compact and linearly ordered topological space hence by previous results the theorem
follows.

Theorem 7.10 SΩ is completely normal but not first countable

12
Proof Since SΩ is linearly ordered topological space hence it follows that SΩ is completely normal.
Now to prove that SΩ is not first countable, we observe that every neighborhood of Ω in SΩ contains
an interval (x, Ω] for some x ∈ SΩ and that interval contains an element of SΩ .( in fact, it contains the
immediate successor of x). This implies that Ω is a limit point of SΩ and that Ω is in the closure of SΩ .
However if (xn ) is a sequence in SΩ then (xn ) is contained in a closed interval [a0 , z] for some z ∈ SΩ .
As a result there is no sequence in SΩ which converges to Ω. Therefore we conclude that SΩ is not first
countable.

Corollary 7.11 SΩ is not second countable.

Proof Since second countable topological spaces are always first countable, hence the result follows.

Corollary 7.12 SΩ is not metrizable.

Proof The result follows from the fact that metrizable spaces are first countable.

Corollary 7.13 SΩ is not locally metrizable.

Proof This follows from the fact that every locally metrizable space is first countable. Indeed if X is
a locally metrizable space then for each x ∈ X has a neighborhood that is metrizable in the subspace
topology. Let us call it U . We can define a metric d on U and take the collection {Bd (x, 1/n) : n ∈ N.
Then it will be a countable base at x.

Theorem 7.14 SΩ is not separable.

Proof If B is a countable dense subset of SΩ then B ∩ SΩ is a countable dense subset of SΩ but this is
impossible since by theorem 6.9, SΩ is not separable.

We observe that SΩ is first countable but not lindelof and not separable. On the other hand, SΩ is not
first countable and separable but satisfy lindelofness. Therefore we can assert that first countability and
lindelofness do not depend on each other.

8 Ordinal numbers
In set theory, an ordinal number, or ordinal, is one generalization of the concept of a natural number that
is used to describe a way to arrange a collection of objects in order, one after another. Any finite collection
of objects can be put in order just by the process of counting: labeling the objects with distinct natural
numbers. Ordinal numbers are thus the ”labels” needed to arrange collections of objects in order.
An ordinal number is used to describe the order type of a well-ordered set (though this does not work
for a well-ordered proper class).
Two well-ordered sets have the same order type if and only if there is a bijection from one set to the other
that converts the relation in the first set to the relation in the second set.
Whereas ordinals are useful for ordering the objects in a collection, they are distinct from cardinal
numbers, which are useful for saying how many objects are in a collection. Although the distinction
between ordinals and cardinals is not always apparent in finite sets (one can go from one to the other just
by counting labels), different infinite ordinals can describe the same cardinal. Like other kinds of numbers,
ordinals can be added, multiplied, and exponentiated, although the addition and multiplication are not
commutative.
Ordinals were introduced by Georg Cantor in 1883 to accommodate infinite sequences and to classify
derived sets, which he had previously introduced in 1872 while studying the uniqueness of trigonometric
series.

13
8.1 Ordinals extend the natural numbers
A natural number (which, in this context, includes the number 0) can be used for two purposes: to describe
the size of a set, or to describe the position of an element in a sequence. When restricted to finite sets these
two concepts coincide, there is only one way to put a finite set into a linear sequence, up to isomorphism.
When dealing with infinite sets one has to distinguish between the notion of size, which leads to cardinal
numbers, and the notion of position, which is generalized by the ordinal numbers described here. This is
because while any set has only one size (its cardinality), there are many nonisomorphic well-orderings of
any infinite set.
Whereas the notion of cardinal number is associated with a set with no particular structure on it, the
ordinals are intimately linked with the special kind of sets that are well-ordered. A well-ordered set is
a totally ordered set in which there is no infinite decreasing sequence (however, there may be infinite
increasing sequences); equivalently, every non-empty subset of the set has a least element. Ordinals may
be used to label the elements of any given well-ordered set (the smallest element being labelled 0, the one
after that 1, the next one 2, ”and so on”) and to measure the ”length” of the whole set by the least ordinal
that is not a label for an element of the set. This ”length” is called the order type of the set.
Any ordinal is defined by the set of ordinals that precede it: in fact, the most common definition of
ordinals identifies each ordinal as the set of ordinals that precede it. For example, the ordinal 42 is the
order type of the ordinals less than it, i.e., the ordinals from 0 (the smallest of all ordinals) to 41 (the
immediate predecessor of 42), and it is generally identified as the set {0, 1, 2, ....., 41}. Conversely, any set
S of ordinals that is downward-closed meaning that for any ordinal α in S and any ordinal β < α, β is
also in S is an ordinal.
There are infinite ordinals as well: the smallest infinite ordinal is ω, which is the order type of the natural
numbers (finite ordinals) and that can even be identified with the set of natural numbers (indeed, the set of
natural numbers is well-ordered as is any set of ordinals and since it is downward closed it can be identified
with the ordinal associated with it, which is exactly how ω is defined). Perhaps a clearer intuition of
ordinals can be formed by examining a first few of them: as mentioned above, they start with the natural
numbers, 0, 1, 2, 3, 4, 5, After all natural numbers comes the first infinite ordinal, ω, and after that come
ω + 1, ω + 2, ω + 3, and so on. (As per arithmatic of ordinal numbers) After all of these come ω.2 (which
is ω + ω), ω.2 + 1, and so on, then ω.3, and then later on ω.4. Now the set of ordinals formed in this way
(the ω.m + n, where m and n are natural numbers) must itself have an ordinal associated with it: and
that is ω2. Further on, there will be ω3, then ω4, and so on, and ωω, then ωωω, and so on. This can be
continued indefinitely far. The smallest uncountable ordinal is the set of all countable ordinals, expressed
as Ω.

8.2 Von Neumann definition of ordinals

The standard definition, suggested by John von Neumann and now called definition of von Neumann
ordinals, is: ”each ordinal is the well-ordered set of all smaller ordinals.” In symbols, λ = [0, λ).
A set S is an ordinal if and only if S is strictly well-ordered with respect to set membership and every
element of S is also a subset of S. The natural numbers are thus ordinals by this definition. For instance,
2 is an element of 4 = {0, 1, 2, 3}, and 2 is equal to {0, 1} and so it is a subset of {0, 1, 2, 3}.
It can be shown that every well-ordered set is order-isomorphic to exactly one of these ordinals, that
is, there is an order preserving bijective function between them.
Furthermore, the elements of every ordinal are ordinals themselves. Given two ordinals S and T , S is
an element of T if and only if S is a proper subset of T . Moreover, either S is an element of T , or T is an
element of S, or they are equal. So every set of ordinals is totally ordered. Further, every set of ordinals
is well-ordered. This generalizes the fact that every set of natural numbers is well-ordered.
Consequently, every ordinal S is a set having as elements precisely the ordinals smaller than S

14
8.3 Burali-Forti paradox
In set theory, a field of mathematics, the Burali-Forti paradox demonstrates that constructing ”the set of
all ordinal numbers” leads to a contradiction and therefore shows an antinomy in a system that allows
its construction. It is named after Cesare Burali-Forti, who in 1897 published a paper proving a theorem
which, unknown to him, contradicted a previously proved result by Cantor. Bertrand Russell subsequently
noticed the contradiction, and when he published it in his 1903 book Principles of Mathematics, he stated
that it had been suggested to him by Burali-Forti’s paper, with the result that it came to be known by
Burali-Forti’s name.
We will prove this by reductio ad absurdum.
1. Let Ω be a set that contains all ordinal numbers.
2. Ω is transitive because for every element x of Ω (which is an ordinal number and can be any ordinal
number) and every element y of x (i.e. under the definition of Von Neumann ordinals, for every ordinal
number y < x), we have that y is an element of Ω because any ordinal number contains only ordinal
numbers, by the definition of this ordinal construction.
3. Ω is well ordered by the membership relation because all its elements are also well ordered by this
relation.
4. So, by steps 2 and 3, we have that Ω is an ordinal class and also, by step 1, an ordinal number, because
all ordinal classes that are sets are also ordinal numbers.
5. This implies that Ω is an element of Ω.
6. Under the definition of Von Neumann ordinals, Ω < Ω is the same as Ω being an element of Ω. This
latter statement is proven by step 5.
7. But we have that no ordinal class is less than itself, including Ω because of step 4 (Ω is an ordinal class).
Therefore we arrive at a contradiction assuming the sethood of Ω. Under the Von Newmann definition of
ordinals therefore, the class of all ordinals is not a set.

8.4 limit and successor ordinal

Any nonzero ordinal has the minimum element. It may or may not have a maximum element. For example,
42 has maximum 41 and ω + 6 has maximum ω + 5. On the other hand, ω does not have a maximum since
there is no largest natural number. If an ordinal has a maximum α, then it is the next ordinal after α, and
it is called a successor ordinal, namely the successor of α, written α + 1. In the von Neumann definition
of ordinals, the successor of is α ∪ {α} since its elements are those of α and α itself.
A nonzero ordinal that is not a successor is called a limit ordinal. One justification for this term is that a
limit ordinal is the limit in a topological sense of all smaller ordinals (under the order topology).
When hαι |ι < γi is an ordinal-indexed sequence, indexed by a limit γ and the sequence is increasing, i.e.
αι < αρ whenever ι < ρ, its limit is defined as the least upper bound of the set {αι |ι < γ}, that is, the
smallest ordinal (it always exists) greater than any term of the sequence. In this sense, a limit ordinal is
the limit of all smaller ordinals (indexed by itself). Put more directly, it is the supremum of the set of
smaller ordinals.
Another way of defining a limit ordinal is to say that α is a limit ordinal if and only if:
There is an ordinal less than α and whenever β is an ordinal less than α, then there exists an ordinal
 such that β <  < α. So in the following sequence:
0, 1, 2, ...., ω, ω + 1 ω is a limit ordinal because for any smaller ordinal (in this example, a natural
number) there is another ordinal (natural number) larger than it, but still less than ω.
Thus, every ordinal is either zero, or a successor (of a well-defined predecessor), or a limit.

15
8.5 Ordinals and Cardinals
Each ordinal associates with one cardinal, its cardinality. If there is a bijection between two ordinals, then
they associate with the same cardinal. Any well-ordered set having an ordinal as its order-type has the
same cardinality as that ordinal. The least ordinal associated with a given cardinal is called the initial
ordinal of that cardinal. Every finite ordinal (natural number) is initial, and no other ordinal associates
with its cardinal. But most infinite ordinals are not initial, as many infinite ordinals associate with the
same cardinal. The axiom of choice is equivalent to the statement that every set can be well-ordered, i.e.
that every cardinal has an initial ordinal.
The α-th infinite initial ordinal is written ωα , it is always a limit ordinal. Its cardinality is written ℵα .
For example, the cardinality of ω is ℵ0 , which is also the cardinality of ω2. So ω can be identified with ℵ0 ,
except that the notation ℵ0 is used when writing cardinals, and ω when writing ordinals (this is important
since, for example, ℵ20 = ℵ0 where as ω 2 > ω .

16
References
[1] JAMES R.MUNKRES TOPOLOGY PEARSON

[2] THOMAS JECH Set Theory SPRINGER

[3] PETE.L.CLARK, DEPARTMENT OF MATHEMATICS,UNIVERSITY OF GEORGIA Induction and

completeness in ordered sets Expository notes
http://math.uga.edu/ pete/induction-completeness-brief.pdf

[4] Arnold W.Miller,UNIVERSITY OF WISCONSIN Topology-II,Spring,2009

17