INTERVALS
Rajendra Pant
CONTENTS
1 Intervals
Nested Intervals
Intervals: The Order Relation on R determines a natural collection of
subsets called “intervals.” If a, b ∈ R satisfy a < b, then the open interval
determined by a and b is the set
(a, b) := {x ∈ R : a < x < b}.
The points a and b are called the endpoints of the interval. However, the
endpoints are not included in an open interval.
If both endpoints are adjoined to this open interval, then we obtain the
closed interval determined by a and b namely, the set
[a, b] := {x ∈ R : a ≤ x ≤ b}.
The two half-open (or half-closed) intervals determined by a and b are [a, b),
which includes the endpoint a, and (a, b], which includes the endpoint b.
Each of these four intervals is bounded and has length defined by b – a. If
a = b, the corresponding open interval is the empty set (a, a) = ∅, whereas
the corresponding closed interval is the singleton set [a, a] = {a}.
The infinite open intervals are the sets of the form
(a, ∞) := {x ∈ R : x > a} and (–∞, b) := {x ∈ R : x < b}.
The infinite closed intervals are the sets of the form
[a, ∞) := {x ∈ R : x ≥ a} and (–∞, b] := {x ∈ R : x ≥ b}.
Theorem 2.5.1 (Characterization Theorem)
If S ⊂ R such that it contains at least two points and has the property
if x, y ∈ S and x < y ⇒ [x, y] ⊆ S
then S is an interval.
Nested Intervals: A sequence of intervals In (n ∈ N) is nested if
I1 ⊇ I2 ⊇ · · · In ⊇ In+1 ⊇ · · · .
Example
• In = 0, n1 ;
h i
• In = 0, n1 .
Nested-Intervals
Theorem 2.5.2 (Nested Intervals Property)
If In = [an , bn ] (n ∈ N) is a nested sequence of closed bounded intervals, then
there exists a number ζ ∈ R such that ζ ∈ In for all n ∈ N.
Proof. Firstly, we have In ⊆ I1 for all n ∈ N (since the intervals are
nested).Hence an ≤ b1 for all n ∈ N. Thus the set {an : n ∈ N} is bounded
above, and so from the Completeness Property of R it follows that there
exists a supremum, ζ for the set, that is,
an ≤ ζ for all n ∈ N.
We want to show that ζ ≤ bn for all n ∈ N, from which it will follow that
an ≤ ζ ≤ bn for all n ∈ N and hence ζ ∈ In for all n ∈ N.
If k < n, then In ⊆ Ik
Let n ∈ N. We consider two cases:
1. If n ≤ k, then In ⊇ Ik (since the intervals are nested), and so
ak ≤ bk ≤ bn .
2. If k < n, then Ik ⊇ In , and so ak ≤ an ≤ bn .
In both cases then, ak ≤ bn for all k ∈ N. Hence it follows that bn is an
upper bound for the set {ak : k ∈ N}, and so ζ ≤ bn . Since n was chosen
arbitrarily, it follows that ζ ≤ bn for all n ∈ N, and so the result follows.
Theorem 2.5.3
If In = [an , bn ] (n ∈ N) is a nested sequence of closed bounded intervals such
that the interval lengths satisfy
inf{bn – an : n ∈ N} = 0
then the number ζ contained in In for all n ∈ N (from the Nested Intervals
Properly) is unique.
Proof. Let η = inf{bn : n ∈ N}. From the two cases considered in Theorem
2.5.2 (with k fixed) we have that ak ≤ bn for all n ∈ N, and so ak is a lower
bound for the set {bn : n ∈ N}. Therefore
ak ≤ η for all k ∈ N.
So, ζ ≤ η where ζ = sup{ak : k ∈ N} (see Theorem 2.5.2). It follows that for
a fixed n ∈ N, η ≤ bn and an ≤ ζ which implies –ζ ≤ –an . So,
0 ≤ η – ζ ≤ bn – an ( as ζ ≤ η).
Since inf{bn – an : n ∈ N} = 0, it follows that for every ε > 0 there is an
m ∈ N such that bm – am < ε, and so from the preceding remarks we have
0 ≤ η – ζ ≤ bm – am < ε. Since ε > 0 was chosen arbitrarily, so η – ζ = 0 (by
Theorem 2.1.9). Therefore, we conclude that η = ζ is the only point that
belongs to In for every n ∈ N.
Theorem 2.5.4
R is uncountable.
Proof. We will show that the unit interval I = [0, 1] is uncountable, which
will imply that R is uncountable, since I ⊂ R. Assume, to the contrary, that I
is countable, and so it can be enumerated as follows
I = {x1 , x2 , · · · xn , · · · }.
Select a closed subinterval I1 of I such that x1 ∉ I1 . Then select a closed
subinterval I2 of I1 such that x2 ∉ I2 . Hence
I1 ⊇ I2 ⊇ · · · In ⊇ In+1 ⊇ · · · .
is a nested sequence of closed bounded intervals such that xn ∉ In for all
n ∈ N. The Nested Intervals Property implies that there is a point ζ ∈ In for
all n ∈ N, and so we have an element of I not listed in {x1 , x2 , · · · xn , · · · }. It
follows that the enumeration is not a complete listing of the elements of I,
and so we have a contradiction.
