1

Sections 1.4 Continuity

Intuitive Definition or Pencil Test If you can trace the graph of a func-
tion without lifting your pencil, the function is continuous.

For motivation we will consider graphically 4 fundamental discontinuities:

Case 1: f(a) is not defined – a hole in the graph case.

x2 −1
Consider the graph of the function f (x) = x−1

Note that limx→1 f (x) exists but f (1) is not defined.

Case 2 f (a) is defined but still a hole in the graph.

( 2
x −1
x−1 if x 6= 1
Consider the graph of f (x) =
−1 if x = 1

Note that limx→1 f (x) exists and f (1) is defined but limx→1 f (x) 6= f (1)
2

Case 3 Jump Discontinuity

½ 2
x if x ≥ 1
Consider the graph of f (x) =
−x2 if x < 1

f (1) exists but limx→1 f (x) does not exist.

Case 4 Vertical Asymptotes

1
Consider f (x) = (x−1)2

Here f (x) grows unbounded as x → 1+ or x → 1− . Hence f (1) does not

exist and limx→1 f (x) does not exist.

So based on our observations above the below definition should not come as
a surprise.
 3

Definition A function f is continuous at x = a if

1) limx→a f (x) exists

2) f (a) is defined

3) limx→a f (x) = f (a)

Remark: If f(x) is not continuous at x = a, it is discontinuous at x = a.

There are a few important things to notice about continuous functions:

1) evaluating limits of continuous functions is easy, because its the same

as evaluating the function;

2) a function can only be continuous on its domain, since the definition

involves evaluating f at a (in particular if f(a) is not defined, then f(x)
cant be continuous at a);

3) a continuous function has a limit at a (in particular, if limx→a f (x)

does not exist, f cant be continuous).

Types of discontinuity A function can fail to be continuous in a few dif-

ferent ways. The two big ways well see for a function to fail to be continuous
at a point are jump discontinuities and removable discontinuities.
• A jump discontinuity is a point a so that limx→a− f (x) and limx→a+ f (x)
both exist, but are not equal.

• A removable discontinuity is a point a so that limx→a f (x) exists, but

limx→a f (x) 6= f (a).
Jump discontinuities are irredeemable, in the sense that unless we give a
major make-over to the graph of f (x) we cannot get rid of them. However,
Removable discontinuities needs only fixing at only one point, namely at ”a”
. We can remove this type of discontinuity and get a continuous function.
Here is how such an operation is done on an example:
4

Example Recall we have showed using the Squeeze Theorem

lim x2 sin( x1 ) = 0. Clearly f (0) is not defined. So this function is a discon-
x→0
tinuous function. But since the limit exists at x = 0,
½ we2 can ”fill in the hole”
x sin( x1 ) if x 6= 0
by defining a new function h(x) as follows h(x) =
0 if x = 0
h(x) now is continuous at x = 0 lim h(x) = h(0) so we have removed the
x→0
discontinuity.
Definition A function is continuous from the right at x = a if
lim f (x) = f (a)
x→a+
√
Example f (x) = x − 2. Applying the Pencil Test to the graph of f(x)
we see that f is continuous at every point in (2, ∞)

1.5

1.0
Out[3]=

0.5

-1 1 2 3 4 5

Also we have previously showed that lim f (x) = 0. We say that f is con-
x→2+
tinuous on its domain [2, ∞). Closed bracket at 2 implies that the function
is continuous from right.

Definition A function f is continuous on an interval [a, b] if it is continuous

at every point in the interval. At the end points of the interval ”a” and
”b” by continuity we understand one sided-continuity. At ”a” we require
continuity from right and at ”b” we require continuity from left.
Remark If a function is continuous on all of < it is referred to simply
as continuous.

Example Let p(x) be any polynomial. In Section 1.3 we have showed

that lim p(x) = p(a) at all a in <. So by definition above p(x) is continuous.
x→a

In fact, almost all the functions that we have talked about in our library
are continuous: polynomials, rational functions (on their domain –so in par-
ticular where the denominator is not 0), trig functions (on their domain),
 5

exponential functions, log functions (on their domain), and the absolute
value function are all continuous.
We can also put together two continuous functions and get continuous func-
tions: the sum, difference, product, scaling, quotient, and composition of
two continuous functions are all continuous on the domain of the resultant
function!

Example Find lim sin(x)?

x→e7 /π

Since sin(x) is continuous by definition of continuity, lim sin(x) = sin(e7 /π)

x→e7 /π

Since continuity is defined as a limit, the limit laws have a direct correlation
to continuous functions.

Theorem If f and g are continuous at x=a and c is a constant, then the

following are also continuous at x=a

1) f ± g

2) cf

3) (f · g)

f
4) If g(a) 6= 0 g

ln x+tan−1 (x)
Example Where is f (x) = x2 −1
continuous?

Divide and conquer: ln(x) is continuous on its domain (0, ∞) and tan−1 (x)
is continuous on all of <. Like wise x2 − 1. By the property (1) above
the numerator ln(x) + tan− (x) continuous on (0, ∞). By Property (4) f(x)
is continuous on (0, ∞) where x2 − 1 6= 0. Hence f (x) is continuous on
(0, 1) ∪ (1, ∞)
Theorem If f is continuous at b and limx→a g(x) = b, then limx→a f (g(x)) =
g(b). In other words;

lim f (g(x)) = f ( lim g(x))

x→a x→a
6
√
Example Find lim arcsin( 1− x
1−x )
x→1
√
Since arcsin( 1− x
1−x ) is not defined at x=1 we can’t simply push x = 1 into
√
1− x
the function. But, if lim 1−x = b and arcsin(x) is continuous at b, we can
x→1
use the last theorem.
√ √ √
1− x 1− x 1+ x
lim = lim ·( √ )
x→1 1 − x x→1 1 − x 1+ x
√ √
1+ x− x−x
= lim √
x→1 (1 − x)(1 + x)
1−x
= lim √
x→1 (1 − x)(1 + x)
1 1
= lim √ =
x→1 1 + x 2

Recall that the domain of arcsin(x) is [−1, 1]. Since 1/2 is in this domain
arcsin(x) is continuous at x = 1. By last theorem,
√ µ √ ¶
1− x 1− x 1 π
lim arcsin( ) = arcsin lim = arcsin( ) =
x→1 1−x x→1 1 − x 2 6

Theorem If g is continuous at a and f is continuous at g(a), then the com-

posite function f ◦ g(x) = f (g(x)) is continuous at a.

Example Where is F (x) = ln(1 + cos(x)) continuous?

Note that F (x) = f (g(x)) where f (x) = ln(x) and g(x) = 1 + cos(x). F (x)
is only defined when 1 + cos(x) > 0. So F (x) is undefined when cos(x) = −1
or x = ±π, ±3π, ±5π, ...... So F is continuous on the intervals between these
values.
 7

The intermediate Value Theorem(IVT)

Theorem Suppose that f is continuous on the closed interval [a, b] and ei-
ther;
• f (a) < N and f (b) > N or
• f (a) > N and f (b) < N
Then there is a number c in the interval (a, b) for which f (c) = N .

Remarks: 1)In geometric terms IVT says that if any horizontal line y = N
is given between y = f (a) and y = f (b) then graph of f cannot jump over
the line. It must intersect y = N somewhere

2) The IVT states that there is such a number c it neither says what this c
is nor claims c to be unique. It is shown in the figures below that the value
N can be taken on once or more that once.
fHbL
fHbL
N
N
Out[75]=
a Out[17]=
c b
fHaL
fHxL fHaL
a c c b
fHxL
3) It is important that the function f in the theorem be continuous, IVT is
not true in general for discontinuous functions as can be seen in the example
below. ½
1 if x > 0
f (x) =
−1 if x ≤ 0
f HxL
3

1
Out[30]=

-2 -1 1 2

-1

-2

Note that there are no c-values on [−1, 1] where −1 < f (c) = 1/2 < 1 or
any other N where −1 < N < 1 for that matter.
8

IVT has many useful applications. We will see two of them.

Example Show that the equation x2008 + 2008x − 1 has a root.
A root of the equation f (x) = x2008 + 2008x − 1 is a solution to the equation
f (x) = 0,so we have to prove that f(x) = 0 has a solution.
First, we notice that f(x) is a continuous function, since f(x) is a polynomial,
and polynomials are continuous. Further we can evaluate

f (0) = 02008 + 2008 · 0 − 1 = −1 < 0 = N

f (1) = 12008 + 2008 · 1 − 1 = 1 + 2008 − 1 = 2008 > 0 = N
Since f(x) is a continuous function on [0, 1] with f (0) < 0 and f (1) > 0, we
may apply the IVT and conclude that there exists a number c in the interval
(0, 1) with f (c) = 0 or c2008 + 2008 · c − 1 = 0, which is exactly what we
wanted to show.

Side NoteIndeed, the conclusion of the intermediate value theorem is some-

what mysterious in that it states the existence of a number c which has a
nice property f(c) = N but does not actually tell you what that number is.
This is about the only rule we have to show that a certain number exists
without actually producing it, so if someone asks you to show a certain num-
ber exists but does not make you actually find it, you should start thinking
intermediate value theorem.

Heres a trickier example that is a good example of this principle

Example Show that there exists a number x in the interval [0, π2 ] with
x = cos(x).

This problem asks us to show there exists a number x which has a nice prop-
erty (this time, nice property means x = cos(x)) but does not ask us to say
exactly what it is, so we are thinking that we should use the intermediate
value theorem.

The first thing we need to do is translate this problem into a problem that
we can use the intermediate value theorem to solve. In particular, given a
continuous function g(x) and some hypotheses, the intermediate value the-
orem lets us conclude the existence of a number c with g(c) = N , where N
is some real number. So lets translate our problem into the solution of an
equation.
 9

Specifically, we notice that finding a number x with x = cos(x) is the same

as finding a solution to the equation f (x) = 0, where f (x) = cos(x) − x.
Why is this true? A solution to f (x) = 0 is a number so that cos(x) − x = 0,
or equivalently, so that cos(x) = x. This is exactly the condition we want!

Great, so lets try to show that for the function f (x) = cos(x)−x there exists
a solution to f (x) = 0. Since we know we have to use the intermediate value
theorem, we start by observing that f(x) is the difference of two continuous
functions, and is therefore itself continuous. Furthermore we can see that

f (0) = cos(0) − 0 = 1 − 0 = 1 > 0

π π π π π
f ( ) = cos( ) − = 0 − = − < 0
2 2 2 2 2
Therefore, since f(x) is a continuous function on [0, π2 ] with f (0) > 0 and
f ( π2 ) < 0, we may apply the IVT and conclude that there exists a number
c in the interval (0, π2 ) with f (c) = 0. This magical value of c therefore
satisfies cos(c) − c = 0, and so cos(c) = c. Thats just what we want.