An Exploration of the Cantor Set

Christopher Shaver, Rockhurst University 09

James and Elizabeth Monahan Summer Research Fellowship Summer 2008
MT4960: Mathematics Seminar Spring 2009
Abstract
This paper is a summary of some interesting properties of the Cantor ternary set and a few investigations of
other general Cantor sets. The ternary set is discussed in detail, followed by an explanation of three ways
of creating general Cantor sets developed by the author. The focus is on the dimension of these sets, with
a detailed explaination of Hausdor dimension included, and how they act as interesting examples of fractal
sets.
Keywords: Cantor set, fractal, Hausdor dimension, self-similarity
1
Introduction
Georg Cantor (1845-1918) rst introduced the set that became known as the Cantor ternary set in the
footnote to a statement saying that perfect sets do not need to be everywhere dense. This footnote gave
an example of an innte, perfect set that is not everywhere dense in any interval, a set he dened as real
numbers of the form
r =
c
1
3
+ ... +
c
u
3
u
+ ... (1.1)
where c
u
is 0 or 2 for each integer .[1]
Denition (1.2): A set S is said to be perfect if S = S, where S is the set of all the limit points of S. In
other words, S is perfect if S is a closed set in which every point is a limit point of S.
Denition (1.3): A set S is said to be nowhere dense if the interior of the closure of S is empty.
The paper in which Cantors statement and footnote were made was written in October of 1882, but
in a paper published in 1875, H. J. S. Smith made the discovery of nowhere dense sets with positive outer
content, meaning that they still take up space. This paper, in which Smith gave an example of a set that
would today be classied as a type of Cantor set, went largely unnoticed until well after Cantors discoveries
were made. Now, in this paper, I attempt to study the Cantor ternary set from the perspective of fractals
and Hausdor dimension and make my own investigations of other general Cantor sets I constructed and
their dimensions.
The term "Cantor set" is most often used to refer to what is known as the Cantor ternary set, which is
constructed as follows:
Let 1 be the interval [0, 1]. Divide 1 into thirds. Remove the open set that makes up the middle third,
that is, (
1
3
,
2
3
), and let
1
be the remaining set. Then

1
= [0,
1
3
] ' [
2
3
, 1]. (1.4)
2
Continue by removing the open middle third segment from each of the two closed sets in
1
and call the
remaining set
2
. So,

2
= [0,
1
9
] ' [
2
9
,
1
3
] ' [
2
3
,
7
9
] ' [
8
9
, 1]. (1.5)
Continue in this fashion at each step / for / N, removing the open middle third segment from each
of the closed sets in
|
and calling the remaining set
|+1
. For each / N,
|
is the union of 2
|
closed
intervals each of length 3
|
.
Denition (1.6): The Cantor ternary set, which we denote C
3
, is the "limiting set" of this process, i.e.
C
3
=
1

|=1

|
. [2]
The Cantor ternary set is interesting in mathematics because of the apparent paradoxes of it. By the
way it is constructed, an innite number of intervals whose total length is 1 are removed from an interval
of length 1, so the set cannot contain any interval of non-zero length. Yet the set does contain an innite
number of points, and, in fact, it has the cardinality of the full interval [0, 1]. So the Cantor set contains as
many points as the set it is carved out of, but contains no intervals and is nowhere dense. We know that the
set contains an innite number of points because the endpoints of each closed interval will always remain in
the set, but the Cantor set actually contains more than just the endpoints of the closed intervals
|
. In fact,
3
1
4
C but is not an endpoint of any of the intervals in any of the sets
|
. Notice rst that each element of
C can be written in a ternary (base 3) expansion of only 0s and 2s. At every level of removal, every number
with a ternary expansion involving a 1 is removed. At the rst stage of removal, for instance, any number
remaining would have the digit c
1
= 0 or 2 where r = 0.c
1
c
2
c
3
..., since if r [0,
1
3
], c
1
= 0 and if r [
2
3
, 1],
c
1
= 2. Repeating this argument for each level of removal, it can be shown that if r remains after removal
:, c
n
is 0 or 2. Now, we can write
1
4
as 0.020202... in ternary expansion. At the /
||
stage of removal, any
new endpoint has a form of either a 2 in the 3
|
th
ternary place which repeats innitely or terminates at
the 3
(|1)
ternary place. Since
1
4
= 0.020202, it does not follow the pattern of the endpoints. Therefore,
1
4
is not an endpoint of the Cantor set, and there are innitely many points like that.
Properties of the Cantor set
Certain easily proven properties of the Cantor ternary set, when they are pieced together, help to show the
special nature of Cantor sets. The Cantor ternary set, and all general Cantor sets, have uncountably many
elements, contain no intervals, and are compact, perfect, and nowhere dense.
C
3
has uncountably many elements.
We will show this by contradiction. Let C
3
= r [0, 1) : r has a ternary expansion involving only zeros
and twos. Suppose C
3
is countable. Then there exists ) : N
11
on|o
C
3
by the denition of countablility.
Let A
n
= )(:) for all : N. So C
3
= r
1
, r
2
, r
3
, ..., r
n
, ... where:
r
1
= 0.c
11
c
12
c
13
...
r
2
= 0.c
21
c
22
c
23
...
.
.
. (2.1)
r
n
= 0.c
n1
c
n2
c
n3
...
4
.
.
.
where c
nm
is either 0 or 2 for all :, :. Dene c = 0.c
1
c
2
c
3
... by
c
1
=
_
2 if c
1
1
= 0
0 if c
1
1
= 2
_
, c
2
=
_
2 if c
2
2
= 0
0 if c
2
2
= 2
_
, ..., c
n
=
_
2 if c
nn
= 0
0 if c
nn
= 2
_
, ... (2.2)
Clearly, c C
3
. But, c ,= A
n
for any :, since c ,= r
n
in its 3
n
th
place. This is a contradiction.
Therefore, C
3
is uncountable.
C
3
contains no intervals.
We will show that the length of the complement of the Cantor set C
3
is equal to 1, hence C
3
contains no
intervals. At the /
||
stage, we are removing 2
|1
intervals from the previous set of intervals, and each one
has a length of
1
3
k
. The length of the complement within [0, 1] after an innite number of removals is:
1

|=1
2
|1
(
1
3
|
) =
1
3
1

|=1
(
2
3
)
|1
=
1
3
1

|=0
(
2
3
)
|
=
1
3
_
1
1
2
3
_
= 1. (2.3)
Thus, we are removing a length of 1 from the unit interval [0, 1] which has a length of 1. Therefore, the
Cantor set must have a length of 0, which means it has no intervals.
C
3
is compact.
Using the Heine-Borel Theorem, which states that a subset of R is compact i it is closed and bounded,
it can be shown rather easily that C
3
is compact. C
3
is the intersection of a collection of closed sets, so C
3
itself is closed. Since each
|
is within the interval [0, 1], C
3
, as the intersection of the sets
|
, is bounded.
Hence, since C
3
is closed and bounded, C
3
is compact.
So far we have that the Cantor set is a subset of the interval [0, 1] that has uncountably many elements
yet contains no intervals. It has the cardinality of the real numbers, yet it has zero length.
C
3
is perfect.
5
By (1.2) a set is perfect if the set is closed and all the points of the set are limit points of the set. Since
C
3
is compact, it is necessarily closed. For each endpoint in the set C
3
there will always exist another point
in the set within a deleted neighborhood of some radius - 0 on one side of that point since the remaining
intervals at each step are being divided into innitely small subintervals and since the real numbers are
innitely dense. Likewise, for each nonendpoint in the set there will always exist another point in the set
within a deleted neighborhood of some radius - 0 on both sides of that point. Hence, there must exist
a deleted neighborhood of some radius - 0 around each point of the set C
3
for which the intersection of
that deleted neighborhood and the set is nonempty. Therefore, each point in the set is a limit point of the
set, and since the set is closed, the set C
3
is perfect.
C
3
is nowhere dense.
By (1.3) a set is nowhere dense if the interior of the closure of the set is empty. The closure of a set
is the union of the set with the set of its limit points, so since every point in the set C
3
is a limit point of
the set the closure of C
3
is simply the set itself. Now, the interior of the set must be empty since no two
points in the set are adjacent to each other. At the innite level of removal, if there did exist a series of
adjacent points, that is an interval of points, the middle third section of that interval will be removed and
the removal would continue on an innitely small scale, ultimately removing anything between two points.
Hence the set C
3
is nowhere dense.
Dimension of the Cantor Set
The Cantor set seems to be merely a collection of non-adjacent points, so should intuitively have a dimension
of zero, as any random collection of non-adjacent points would have. In this sense, the topological dimension
of the Cantor set is 0. But utilizing a dierent denition of dimension, such as Hausdor dimension, allows
us to see the fractional dimension of the Cantor set while still maintaining the integer dimensions of points,
lines, and planes.
6
Hausdor Dimension
Denition: Let 1 _ R
n
and let : and c be positive real numbers. Let C
o
(1) be the collection of all
countable c-covers of 1, where a c-cover of 1 is a sequence l

of subsets of R
n
whose union contains 1
and for which 0 < [l

[ < c for all , = 1, 2, .... Then,

H
s
o
(1) := inf
1

=1
[l

[
s
: l

C
o
(1) (3.1)
[3].
Denition (3.2): For any subset 1 of R
n
, H
s
(1) := sup
o!0
H
s
o
(1).
Denition (3.3): The Hausdor s-dimensional measure is the restriction of H
s
(+) to the o-eld of H
s
(+)-
measureable sets. The Hausdor dimension of 1 is dened to be:
dim(1) = inf: : H
s
(1) = 0 (3.4)
[3].
In simple terms, Hausdor dimension involves rst considering the c-covers of the set 1. If you take
the sum of the diameters of all c-covers of 1 raised to the power of :, the inmum of that sum is H
s
o
(1).
The Hausdor s-dimensional measure of 1 is then the sup
o,0
H
s
o
(1). Finally, the Hausdor dimension of 1,
denoted by dim(1), is the inmum of all the values of the real number : such that the Hausdor s-dimensional
measure of 1 is 0.
Hausdor dimension generalizes the concept of dimension of a vector space in such a way that points
have Hausdor dimension 0, lines have Hausdor dimension 1, etc., but in general the Hausdor dimension
of a set is not necessarily integer. Fractals are dened as sets whose Hausdor dimension is greater than its
topological dimension, with the Hausdor dimension of fractals specically non-integer.
Theorem 1: The dimension of the Cantor ternary set (C
|
) is:
7
d =
log
1
2
log(
1
2

1
2|
)
(3.5)
Proof: Let us create a series of c-covers of C
|
:
l
01
= [0, 1], a 1 cover (3.6)
l
11
= [0,
1
2

1
2/
], l
12
= [
1
2
+
1
2/
, 1], a (
1
2

1
2/
) cover
l
21
= [0,
1
4

1
2/
+
1
4/
2
], l
22
= [
1
4

1
4/
2
,
1
2

1
2/
], l
23
= [
1
2
+
1
2/
,
3
4
+
1
4/
2
],
l
24
= [
3
4
+
1
2/

1
4/
2
, 1], a (
1
4

1
2/
+
1
4/
2
) cover (1)
In general, a (
1
2

1
2|
)

-cover is
l
j

2
g
=1
(3.7)
For : 0,
H
s
o
(C
|
) : = inf
1

I=1
[l
I
[
s
: l
I
C
o
(C
|
) (3.8)
_
2
g

I=1
[l
I
[
s
, for (
1
2

1
2/
)

_ c
= 2

((
1
2

1
2/
)

)
s
=
2

((
1
2

1
2|
)

)
s
=
_
2
(
1
2

1
2|
)
s
_

So, if (
1
2

1
2|
)
s
= 2, then H
s
o
(C
|
) _ 1 for all c 0. Now, (
1
2

1
2|
)
s
= 2 i log(
1
2

1
2|
)
s
= log 2 i
: log(
1
2

1
2|
) = log 2. That is, : =
log 2
log(
1
2

1
2k
)
. If : =
log 2
log(
1
2

1
2k
)
, then H
s
(C
|
) _ 1 and
dim(C
|
) _
log 2
log(
1
2

1
2|
)
(3.9)
8
Now we show H
s
(C
|
) _ 1, where : =
log 2
log(
1
2

1
2k
)
. For c 0, let \
I
be a c-cover of C
|
. We may assume
\
I
is open for each i. Since C
|
is compact, there is a nite subcover \
I

n
I=1
; \
I
= (a
I
, /
I
), where a
I
< a
I+1
and /
I
< /
I+1
for 1 _ i _ : 1. Now, there exists a closed subinterval [c
I
, ,
I
] of (a
I
, /
I
) such that
C
|
_
n
_
I=1
[c
I
, ,
I
], c
I
< ,
I
< c
I+1
_ ,
I+1
for 1 _ i _ : 1; (3.10)
where c
I
and ,
I
have the form :(
1
2

1
2|
)

for all i.
Let
1 = max/ : c
I
= :(
1
2

1
2/
)

or ,
I
= :(
1
2

1
2/
)

for some i, 1 _ i _ :; (
1
2

1
2/
) doesnt divide :
(3.11)
Then
2
g
_
=1
l
j
_
n
_
I=1
[c
I
, ,
I
], where l
j

2
g
=1
is as dened in the rst part of the proof.
Let C

=
2
g
_
=1
l
j
. Now consider [c
I
, ,
I
] C

for some i, 1 _ i _ :. [c
I
, ,
I
] C

contains at least one open

interval, 1, of length at least
1
(
1
2

1
2k
)
(,
I
c
I
). Now [c
I
, ,
I
] 1 is the disjoint union of two closed intervals, 1
1
and 1
2
. By the concavity of )(t) = t
s
and since (
1
2

1
2|
)
s
= 2 and (,
I
c
I
) _
(
1
2

1
2k
)
2
([1
1
[ +[1
2
[) we obtain:
(,
I
c
I
)
s
_ [
(
1
2

1
2|
)
2
([1
1
[ +[1
2
[)]
s
(3.12)
= 2[
1
2
[1
1
[ +
1
2
[1
2
[]
s
_ 2(
1
2
[1
1
[
s
+
1
2
[1
2
[
s
)
= [1
1
[
s
+[1
2
[
s
Reduce 1
1
and 1
2
similarly and obtain that (,
I
c
I
)
s
_

:Igi[oi,o
i
]
[l
I
[
s
Consequently,
9
n

I=1
[\
I
[
s
_
n

I=1
(,
I
c
I
)
s
(3.13)
_
n

I=1
_
_

:Ig
j
[oi,o
i
]
[l
I
[
s
_
_
=
2
g

=1
[l

[
s
= 2

_
(
1
2

1
2/
)

_
s
=
_
2
(
1
2

1
2|
)
s
_

= 1
It follows that H
s
o
(C
|
) _ 1 and hence H
s
(C
|
) _ 1. Therefore, dim(C
|
) =
log 2
log(
1
2

1
2k
)
.

[3]
There is a shortcut for computing the Hausdor dimension of sets that are self-similar in nature. To be
self-similar, the mappings of the set must be a nite collection of similitudes, that is, the mappings preserve
the geometry of the set, such that the set is invariant with respect to the set of mappings, and there must
exist a positive real number : such that H
s
of the set is positive but H
s
of the intersection of two dierent
mappings of the set is zero.
Denition (3.14): Let c : R
n
R
n
. If there exists a c, 0 < c < 1, such that [c(r) c(j)[ _ c [r j[
for any x and y in R
n
, then c is called a contraction.
Denition (3.15): Let c be a contraction. Then infc : [c(r) c(j)[ _ c [r j[ \r, jcR
n
is called the
ratio of the contraction c.
Denition (3.16): Let c : R
n
R
n
be a contraction and 1 be a subset of R
n
. If c preserves the
geometry of 1 (c is a combination of a translation, a rotation, a reection, and/or a dilation) then c is
called a similitude.
Denition (3.17): Let 1 _ R
n
and c
I

|
I=1
be a nite collection of contractions. Then 1 is said to be
invariant with respect to c
I

|
I=1
if 1 =
|
_
I=1
c
I
(1).
Denition (3.18): Let 1 _ R
n
and c
I

|
I=1
be a nite collection of similitudes such that 1 is invariant
with respect to c
I

|
I=1
. If there exists : 0 such that H
s
(1) 0 but H
s
(c
I
(1) c

(1)) = 0 for i ,= ,,
then 1 is called self-similar.
10
Denition (3.19): A nite collection of contractions c
I

|
I=1
is said to have the open set condition if there
exists a bounded open set \ such that
|
_
I=1
c
I
(\) _ \, and this union is disjoint.
The Cantor set is self-similar by (3.18) since its mappings, all variations on
1
3
r, preserve the geometry
of the set and there is a positive Hausdor s-dimensional measure of the set but not of the intersection of
two dierent mappings since no two mappings of the set have a non-empty intersection. The Hausdor
dimension of a self-similar set can be found by using the following theorem:
Theorem 2: Let
I

|
I=1
be a collection of similitudes such that 1 _ R
n
is invariant with respect to

|
I=1
. If
I

|
I=1
satises the open set condition and r
I
is the ratio of the i-th similitude
I
, then the
Hausdor dimension of 1 is equal to the unique positive numbers for which
|

I=1
(r
I
)
s
= 1. [3]
The computation of the Hausdor dimension of the Cantor ternary set, C
3
, follows very easily from
Theorem 2.
Proposition: The dimension of the Cantor ternary set C
3
is d =
log 2
log 3
.
Proof: Let c
1
(r) and c
2
(r) be dened as:
c
1
(r) =
1
3
r
c
2
(r) =
1
3
r +
2
3
(3.20)
Notice that C
3
=
2
_
I=1

I
(C
3
). Also,
I

2
I=1
satises the open set condition for \ = (0, 1). Applying
the theorem with r
1
=
1
3
and r
2
=
1
3
, we need to nd : such that
2

I=1
(r
I
)
s
= 1.
2(
1
3
)
s
= 1 i : =
log 2
log 3
. (3.21)
dim(C
3
) =
log 2
log 3
.

[3].
11
General Cantor Sets
Up to this point, our discussion of the Cantor set has been limited to what is known as the Cantor ternary
set, dened in the Introduction. We will now discuss some generalizations. My further investigations were
motivated by a curiosity as to what would happen to the dimension of the set if the removal process was
dened dierently. I consider three dierent general methods of removal from the interval [0, 1] that depend
upon a natural number /. As will be shown, all three methods of removal are equivalent when / = 3,
yielding the Cantor ternary set.
1.) Method C: Let C
|
be the collection of sets dened in terms of /, for / _ 2, in which each set
in the sequence is formed by the repetitive removal of an open interval of length
1
|
from the center of each
closed interval, starting with the interval [0, 1]. In this way, the size of the closed intervals remaining on
either side of the open interval removed will be (
1
2

1
2|
).
Since each of the sets C
|
are self-similar sets, we can use Theorem 2 to nd the Hausdor dimension of
each of the sets, which we will do in general for any / _ 2.
Let c
1
(r) and c
2
(r) be dened as:
c
1
(r) = (
1
2

1
2/
)r
c
2
(r) = (
1
2

1
2/
)r +
1
2
+
1
2/
(4.1)
Notice that C
|
=
2
_
I=1
c
I
(C
|
). Applying the theorem for self-similar sets with r
1
= (
1
2

1
2|
) and
r
2
= (
1
2

1
2|
), we need to nd : such that
2

I=1
(r
I
)
s
= 1. (4.2)
So,
2(
1
2

1
2/
)
s
= 1 i : =
log
1
2
log(
1
2

1
2|
)
. (4.3)
12
Thus,
dim(C
|
) =
log
1
2
log(
1
2

1
2|
)
.

(4.4)
2.) Method 1: Let 1
|
be the collection of sets dened in terms of /, for / _ 2, in which each set
in the sequence is formed by the repetitive removal of an open interval of length (1
2
|
) from the center
of each closed interval starting with [0, 1], with intervals of length
1
|
remaining on each side. Notice that
with this method of removal, we are varying the length of the side intervals in terms of / then removing
the interval inbetween, whereas in the rst method of removal, method C, you are varying the length of the
center interval in terms of /.
Again, since each of the sets 1
|
are self-similar sets, we can use Theorem 2 to nd the Hausdor dimension
of each of the sets, which we will do in general for any / _ 2.
Let c
1
(r) and c
2
(r) be dened as:
c
1
(r) =
1
/
r
c
2
(r) =
1
/
r + 1
1
/
(4.5)
Notice that 1
|
=
2
_
I=1
c
I
(1
|
). Applying the theorem for self-similar sets with r
1
=
1
|
and r
2
=
1
|
, we
need to nd : such that
2

I=1
(r
I
)
s
= 1. (4.2)
So,
2(
1
/
)
s
= 1 i : =
log 2
log /
. (4.6)
13
Thus,
dim(1
|
) =
log 2
log /
.

(4.7)
3.) Method 1: Let 1
|
be the collection of sets dened in terms of /, for / _ 2, in which each set in
the sequence is formed by the repetitive removal of alternating open intervals of length
1
|
from each closed
interval, starting with [0, 1], when each closed interval is divided into / subintervals. This method of removal
leads to two similar but distinctly dierent cases. When / is odd,
|1
2
alternating sections are removed
from each closed interval, leaving each interval of length
1
|
. When / is even, (
|
2
1) alternating sections
are removed from each closed interval, leaving one interval of length
1
|
on the left end and two full intervals
each of length
1
|
adjacent to each other on the right end.
Since dierent mappings will be required in order to generate the sets, each of the two cases yields sets
with dierent Hausdor dimensions. In the case where / is odd, using Theorem 2 to nd the Hausdor
dimension of these self-similar sets, we let c
1
(r), c
2
(r), ... , ck+1
2
(r) be dened as:
c
1
(r) =
1
/
r
c
2
(r) =
1
/
r +
2
/
(4.8)
, ...,
ck+1
2
(r) =
1
/
r +
/ 1
/
Note that the number of c
I
mappings needed is determined by the value of the natural number /, with
|+1
2
mappings needed when / is odd. Applying the theorem for self-similar sets with r
1
=
1
|
, r
2
=
1
|
, ... ,
rk+1
2
=
1
|
, we need to nd : such that
k+1
2

I=1
(r
I
)
s
= 1. (4.2)
So,
14
(
/ + 1
2
)(
1
/
)
s
= 1 i : =
log
|+1
2
log /
. (4.9)
Thus, when / is odd,
dim(1
|
) =
log
|+1
2
log /
.

(4.10)
In the case where / is even, we let c
1
(r), c
2
(r), ... , ck+1
2
(r) be dened as:
c
1
(r) =
1
/
r
c
o
(r) =
1
/
r +
2
/
(4.11)
, ...,
ck
2
(r) =
2
/
r +
/ 2
/
for c = 1, 2, ..., (
|
2
1). Note that the number of c
I
mappings is determined by the natural number /,
with
|
2
mappings needed when / is even. Applying the theorem for self-similar sets with r
1
=
1
|
, r
o
=
1
|
,
rk
2
=
2
|
, we need to nd : such that
k
2

I=1
(r
I
)
s
= 1. (4.2)
So,
(
/
2
1)(
1
/
)
s
+ (
2
/
)
s
= 1. (4.12)
Simplifying this equation to (
|
2
1) = /
s
2
s
, it is not noticeably solvable directly for a general /, but
the dimension of each set is equal to the value of : that satises the equation for its given /.
15
The dierences between the sets formed under each method of removal are apparent when / = 4, and
the dierences between dierent values of the natural number / for a given method of removal are apparent
by comparing / = 4 with / = 5.
Set C
4
:
Set C
5
:
Set 1
4
:
Set 1
5
:
Set 1
4
:
16
Set 1
5
:
Note that when / = 2, the amount being removed under methods 1 and 1 is equal to 0, and the
dimension is consequently equal to 1 in all cases, since we are left with the full interval [0, 1].
Note: C
3
= 1
3
= 1
3
.
Proof: The set C
3
is formed by the repetitive removal of an open interval of length
1
3
from the center of
each closed interval, starting with the interval [0, 1] which leaves closed intervals of size
2
3
on each side. Since
the set 1
3
is formed by the repetitive removal of an open interval of length (1
2
3
) =
1
3
from the center of
each closed interval starting with [0, 1], with intervals of length
1
3
remaining on each side, it is equivalent to
the set C
3
. Also, since the set 1
3
is formed by dividing the interval [0, 1] into 3 subintervals and removing
alternating sections, which is only the center section, with intervals of length
1
3
on each end, it is equivalent
to both C
3
and 1
3
. Hence, C
3
= 1
3
= 1
3
.

Also note that the calculations of dimension for each set yield the same dimension, which must be the
case since the three sets are the same. So,
dim(C
3
) =
log
1
2
log(
1
2

1
6
)
=
log 2
log 3
. dim(1
3
) =
log 2
log 3
. dim(1
3
) =
log
3+1
2
log 3
=
log 2
log 3
(4.13)
Thus, dim(C
3
) = dim(1
3
) = dim(1
3
).
In each method of removal, what happens to the dimension as / approaches innity?
In method C, dim(C
|
) =
log
1
2
log(
1
2

1
2k
)
. So,
lim
|!1
dim(C
|
) = lim
|!1
_
log
1
2
log(
1
2

1
2|
)
_
=
log
1
2
log
1
2
= 1. (4.14)
Hence, when [0, 1] is divided into 3 subintervals, the smaller the length of the interval
1
|
removed, the
closer the Hausdor dimension gets to 1.
17
In method 1, dim(1
|
) =
log 2
log |
. So,
lim
|!1
dim(1
|
) = lim
|!1
_
log 2
log /
_
= 0. (4.15)
Hence, when [0, 1] is divided into 3 subintervals, the smaller the length of the side intervals
1
|
, that is,
the larger the length of the interval (1
2
|
) removed, the closer the Hausdor dimension gets to 0.
In method 1, dim(1
|
) =
log
k+1
2
log |
when / is odd. So,
lim
|!1
dim(1
|
) = lim
|!1
_
log
|+1
2
log /
_
= 1. (4.16)
Hence, when [0, 1] is divided into an odd number of equal length intervals, the larger the number of such
intervals, the closer the Hausdor dimension gets to 1.
Conclusion
The Cantor ternary set and the general Cantor sets are all examples of fractal sets. Their self-similarity
allows their Hausdor dimention to be calculated easily, and each is shown to be non-integer. Many questions
for further investigation remain, including expoloring other methods of removal. The ultimate question is
whether or not it is possible to nd a method of removal that will yield a specic given Hausdor dimension.
Considerable work and exploration would need to be done in order to determine this.
Acknowledgements
This paper is the product of a research project undertaken in Summer 2008 through the Department of
Mathematics and Physics at Rockhurst University in Kansas City, Missouri with Dr. Zdeka Guadarrama,
Assistant Professor, and Jeanette Powers, undergraduate student. The project was extended with further
research for the MT4960 Mathematics Seminar course in Spring 2009 at Rockhurst. The author and the
project were generously supported through the James and Elizabeth Monahan Summer Research Fellowship.
The author wishes to thank Mr. and Mrs. Monahan and especially Dr. Guadarrama and Ms. Powers.
18
References
1. J. Fleron, "A Note on the History of the Cantor Set and Cantor Functions", Mathematics Magazine.
67 (1994) 136-140.
2. S. Lay, Analysis, With an Introduction to Proof, Fourth Edition, Prentice Hall, Upper Saddle River,
NJ, 2005.
3. K.J. Falconer, The Geometry of Fractal Sets, Cambridge University Press, Cambridge, 1985.
19