FUNCTIONS & CONTINUITY

Abstract:
Calculus is the study of the rate of change and the total accumulation of processes
described by functions. In this chapter we review some familiar notions of function and
explore functions that are defined by sequences of functions.

2.1 The Notion of a Function

The idea of a function is the most important concept in mathematics. There are many
sources of functions, and they carry information of a special kind. Some are based on
observations, like the maximum daily temperature T in your town every day last year:

temperature = T(day).

Some express a causal relation between two quantities, such as the force f exerted by a
spring as a function of the displacement:

force = f (displacement).

Research in science is motivated by finding functions to express such causal relations. A

function might also express a purely arbitrary relationship like

F=
relating the Fahrenheit temperature scale to the Celsius scale. Or it could express a
mathematical theorem such as the quadratic formula below

,
where x is the root of the quadratic equation

Functions can be represented in different ways. Some of these ways include graphs,
tables, and equations. Other methods, such as representing a function through a sequence
of functions, or as a solution to a differential equation, are made possible by calculus.

Example 1.
The vertical distance h (measured in kilometers) traveled by a rocket depends on the time
t (measured in seconds) that has elapsed since the rocket was launched.

1
Time Seconds)

Example 2
The distance d traveled by a body falling freely from rest near the surface of the Earth,
measured in meters, and t the time of fall measured in seconds.

Example 3
The national debt D in billions of dollars in year y.
Y 2004 2005 2006 2007 2008 2009 2010
D 7,354 7,905 8,451 8,951 9,654 10,413 13,954

Example 4
The Internal Revenue Service’s 2010 tax rates for single-filing status are given in the
table below. The tax rates can be described as a function. Let x be adjusted gross income
in dollars, and f (x) the rate at which each dollar within that income level is taxed. The
function f can be described by the following rule

To compute the tax on 10,000 dollars, for example, the first 8,375 is taxed at 10%, and
the next 10,000−8,375=1,625 is taxed at 15%. So the tax is (0.15)(1,625) + (0.10)
(8,375)= 1,121.25.

Definition 2.
A function f is a rule that assigns to every number x in a collection D, a number f (x). The
set D is called the domain of the function, and f (x) is called the value of the function at x.

2
The set of all values of a function is called its range. The set of ordered pairs (x, f (x)) is
called the graph of f .

When we describe a function by a rule, we assume, unless told otherwise, that the set of
inputs is the largest set of numbers for which the rule makes sense.

Examples.
i)
The domain of f is all numbers
ii)
The domain of g is x ≥ 1
iii)
The domain of h is any number other than 1 or −1.

Bounded Functions
Definition
We say that a function f is bounded if there is a positive number m such that for all values
of f , −m ≤ f (x) ≤ m. We say that a function g is bounded away from 0 if there is a
positive number p such that no value of g
falls in the interval from −p to p.

A function that is not bounded, or is not bounded away from 0, may have one or both of
those properties on a subset of its domain.

Example
Let . Then h is not bounded. It has arbitrarily large values (both positive and
negative) as x tends to 1 or −1. Furthermore, h is not bounded away from 0, because h(x)
tends to 0 as x becomes arbitrarily large (positive or negative). However, if we restrict the

3
domain of h to, say, the interval [−0.8,0.8], then h is both bounded and bounded away
from 0 on [−0.8,0.8].

from the diagram, the function . is neither bounded nor bounded away
from 0

Arithmetic of Functions
With the knowledge of functions, its worth noting that you can use them to make new
functions. The sum of functions f and g is denoted by f +g, and the difference by f −g:
Thus i) ( f +g)(x) = f (x)+g(x),
ii) ( f −g)(x) = f (x)−g(x).

The product and quotient of functions f and g are denoted by f g(x) and
respectively.
i) f g(x) = f (x)g(x),
ii) , for

N/B In applications, it makes sense to add or subtract two functions only if their values
are measured in the same units.

POLYNOMIALS.

Starting with the simplest functions, the constant functions c(x) = c and the identity
function i(x) = x,
We can build more complicated functions by forming sums and products. All functions
that are obtained from constant and identity functions through repeated additions and
multiplications are of the form

4
where the a’s are constants and n is a positive integer. Such a function is called a
polynomial. A quotient of two polynomials, , is called a rational function.

Linear Functions.
A simple but highly important class of functions is that of linear functions. Every linear
function is of the form

or

where m and b are some given numbers. Linear functions have the property that

which means that when the input x in increased by h, the output changes by an amount
that does not depend on x.
i.e

Example
In changing temperature from Celsius to Fahrenheit, we use the formula

A change in temperature in degrees Celsius produces a change in temperature in degrees

Fahrenheit that is always as large, independent of the temperature

Example.
Show the interval onto which the linear function maps the interval [0,2]

For x=0, y=3(0) -1 =-1

For x=2, y= 3(2)-1 = 5

The interval [0,2] is mapped onto [-1,5]

Exercise
1. For each of these functions, state whether or not
i) f bounded?
ii) f bounded away from zero?

a)
b)

5
c)
d)

2. Let , and
a) Show that when x=0, -3
(b) Find the domains of f , g, and h.
(c) Sketch the graphs of f , g, and h.

3. Let with domain [−0.8, 0.8]. Find bounds p and q on the range of h:

4. The gravitational force between masses M and m with centers separated by distance r
is, according to Newton’s law,

The value of G depends on the units in which we measure mass, distance, and force. Take
the domain to be r > 0. Is f rational? bounded? bounded away from 0?

N/B A function f is said to be continuous at ‘C’ when: for any tolerance , there
exists a precision , such that
whenever
Thus, the precision depends on the tolerance

Continuity

In this section, we scrutinize the definition of function given in the previous section.
According to that definition, a function f assigns a value f (x) to each number x in the
domain of f . Clearly, in order to find the value of f (x), we have to know x.

Continuity at a Point Using Limits

The concept of the limit of a function gives another way to define continuity at a point.

Definition

6
The limit of a function f (x) as x tends to c is L,

when for any tolerance > 0, there is a precision > 0 such that f (x) differs from L by
less than whenever x differs from c by less than , x = c.

By comparing the definitions of limit as x tends to c and continuity at c, we find

a new way to define continuity of f at c.

Alternative definition.
A function f is said to be continuous at c when:

N/B If f is not continuous at c, we say that f is discontinuous at c.

Theorem 1.
If , and
Then,
a)
b)

c)

7
Theorem 2. ( Squeeze theorem)
If f (x) ≤ g(x) ≤ h(x) for all x in an open interval containing c, except possibly at x = c,
and if
, then
.
Combining Theorem 1 and the limit definition of continuity, one can prove theorem 3
below

Theorem 3.
Suppose f , g, and h are continuous at c, and h(c) = 0. Then f +g, f g, and are
continuous at c.

We have noted before that any constant function, and the identity function, are
continuous at each point c. According to Theorem 3, products and sums built from these
functions are continuous at each c.

Every polynomial

can be constructed by taking sums and products of functions that are continuous at c. This
shows that polynomials are continuous at each c. It also follows from the theorem that a
rational function
is continuous at each number c for which q(c) = 0.

Example
So from Theorem 3, the rational function

=
is continuous at every point except 0.

Sometimes a function is undefined at a point c, but the limit of f (x) as x tends to c exists.

Example
Consider the graph of

8
.
The function f is not defined at 3. Notice, however, that for x 3,

The graph of f looks like a straight line with a small hole at the point x = 3.
The functions and x+3 are quite different at x = 3, but they are equal when x = 3.
This means that their limits are the same as x tends to 3:

Example

Let d(x) be defined as follows:

Then d is not continuous at x = 1, because d(1) equals 1, yet for x greater than 1 and no
matter how close to 1, d(x) is negative. A negative number is not close to 1. See Fig.
below

It is useful to have a way to describe the behavior of f (x) as x approaches c from one side
or the other. If f (x) tends to L as x approaches c from the right, c < x, we say that the
right-hand limit of f at c is L, and write

The function d(x) in figure above is not continuous at x = 1

9
If f (x) tends to L as x approaches c from the left, x < c, we say that the left-hand limit of f
at c is L, and write

If f (x) becomes arbitrarily large and positive as x tends to c, we write

and say that f (x) tends to infinity as x tends to c. If f (x) becomes arbitrarily large and
negative as x tends to c, we write

and say that f (x) tends to minus infinity as x tends to c. Neither of these limits exists, but
we use the notation to describe the behavior of the function near c. We also use the one-
sided versions of these notations, as in example below.

Example
Let , for then

and

It is also useful to have a way to describe one-sided continuity. If we

say that f is left continuous at c. If we say that f is right continuous at c.

Example
The function d

in is left continuous at 1, and not right continuous at 1:

and

Left and right continuity give us a way to describe continuity on an interval that includes
endpoints. For example, we say that f is continuous on [a,b] if f is continuous at each c in
(a,b) as well as right continuous at a, and left continuous at b.

10
Continuity on an Interval
We have seen that functions can be continuous at some points and not at others. The most
interesting functions are the ones that are continuous at every point on an interval where
they are defined.

Example
Let us analyze the continuity of the function f (x) = on the interval [2,4]. Let c be any
point of this interval; how close must x be to c in order for f (x) to differ from f (c) by
less than ? Recall the identity

On the left, we have the difference f (x)− f (c) of two values of f . Since both x and c are
between 2 and 4, we have (x+c) ≤ 8. It follows that | f (x)− f (c)| = |x+c||x−c| ≤ 8|x−c|.
If we want to be within of , it suffices to take x within of c. That is, take
or less. This proves the continuity of f on [2,4].

Extreme and Intermediate Value Theorems

Next, we state two key theorems about continuous functions on a closed interval.

Theorem 1
The intermediate value theorem.
If f is a continuous function on a closed interval [a,b], then f takes on all values between
f (a) and f (b).

The theorem says in a careful way that the graph of f does not skip values.

The intermediate value theorem shows that there exists at least one number c between a
and b at which f (c) = m

Theorem 2.
The extreme value theorem.

11
If f is a continuous function on a closed interval [a,b], then f takes both a maximum value
and a minimum value at some points in [a,b].

One consequence of the extreme value theorem is that every function that is continuous
on a closed interval is bounded. Although the extreme value theorem does not tell us how
or where to find the bounds, it is still very useful.

Corollary .
If f is continuous on an open interval (a,b) and f (x) tends to infinity as x tends to each of
the endpoints, then f has a minimum value at some point in (a,b). Similarly, if f (x) tends
to minus infinity as x tends to each of the endpoints, then f has a maximum value at some
point in (a,b).

Exercise
1. Evaluate the following limits.
i)

ii)

iii)

2 Evaluate the following limits.

i)

ii)

iii)

3. Let when x = 0, and f (0) = 1.

(a) Sketch the graph of f .
(b) Is f continuous on [0,1]?
(c) Is f continuous on [−1,0]?
(d) Is f continuous on [−1,1]?

4. Suppose that functions f , g, and h are each defined on an interval containing c, that
they are continuous at c, and that h(c) = 0. Show that f +g, f g, and are continuous
at c
5. Let on the interval [-20,120]. Is f bounded? Explain.

6. Show that the equation has a solution on [-2,2]

12
13
Composition and Inverses of Functions

Composition
We start with a simple example:
A rocket is launched vertically from point L. The distance (in kilometers) of the rocket
from the launch point at time t is h(t). An observation post O is located 1 km from the
launch site (see figure below). To determine the distance d of the rocket from the
observation post as a function of time, we can use the Pythagorean theorem to express d
as a function of h,

Tracking the rocket from the observation post

Therefore, the distance from R to O at time t is

The process that builds a new function in this way is called composition; the resulting
function is called the composition of the two functions.

Definition
Let f and g be two functions, and suppose that the range of g is included in the domain of
f . Then the composition of f with g, denoted by f ◦ g, is defined by ( f ◦ g)(x) = f (g(x)).
We also say that we have composed the functions.

The construction is well described by Fig. below;

Composition of functions, using the box picture

Example

14
Let g and f be the linear functions y = g(x) = 2x + 3, and z = f (y) = 3y+1. The
composition z = f (g(x)) = 3(2x+3)+1 = 6x+10 is illustrated in Fig. below

A composition of two linear functions

Earlier we saw that that the linear function mx+b stretches every interval by a factor of |
m|. In Fig. above, we see that when the linear functions are composed, these stretching
factors are multiplied.

Example
The effect of composing a function f with g(x) = x+1 depends on the order of
composition. For example f (g(x)) = f (x+1) shifts the graph of f one unit to the left, since
the output of f at x is the same as the output of f ◦ g at x−1.

On the other hand, g( f (x)) = f (x)+1 shifts the graph of f up one unit. See figure below.

Composition with the translation x+1, in Example above. It makes a difference which function is applied first

Example

15
Let h(x) = 3x. The graph of f (h(x)) looks as though the domain of f has been compressed
by a factor of 3. This is because the output of f at x is the same as the output of f ◦ h at .

If we compose f and h in the opposite order, the graph of h( f (x)) = 3 f (x) is the graph of f
stretched by a factor of three in the vertical direction. See Fig. below

Composition with multiplication 3x results in stretching or compressing the graph

.

Example
Let h(x) = −x. The graph of h( f (x)) = −f (x) is the reflection of the graph of f across the
x-axis, while the graph of f (h(x)) = f (−x) is the reflection of the graph of f across the y-
axis.

Example
If and , then

i) ( f ◦ g)(x) = and

ii)

N/B Notice that f ◦ g and g ◦ f are quite different functions. Thus composition is not a
commutative operation. This is not surprising: using the output of g as input for f is quite
different from using the output of f as input for g.

Theorem.
The composition of two continuous functions is continuous.

Proof.
We give an intuitive proof of this result. We want to compare the values of f (g(x)) with
those of f (g(z)) as the numbers x and z vary. Since f is continuous, these values will differ
by very little when the numbers g(x) and g(z) are close. But since g is also continuous,
those values g(x) and g(z) will be close whenever x and z are sufficiently close.

16
Theorem .
Suppose f ◦ g is defined on an interval containing c, that and that f is
continuous at L. Then , that is,

Inverse Functions

We look at some examples of compositions of functions that undo each other.

Example
For f (x) = and g(x) = , we see that

i) =

ii)

Example
Let , when and , when . Then if , we have

N/B You may also check that when x = −1, we have g( f (x)) = x.

In both of the examples above, we see that f applied to the output of g returns the input of
g, and similarly, g applied to the output of f returns the input of f. We may ask the
following question about a function: if we know the output, can we determine the input?

Example 1
Find the inverse of f (x) =

Solution

17
In finding the inverse of the function, we take the function and make the variable x the
subject of the formula and interchange f(x) with x
i.e Given f (x) =
we make the variable x the subject of the formula

Interchange x with f(x)

Or which is the inverse function

Example 2
Find the inverse of

Solution

We make the variable x the subject of the formula

i.e
Expanding we have,

Interchange x with f(x)

which is the inverse function.

Definition.
If a function g has the property that different inputs always lead to different outputs, i.e.,
if implies , then we can determine its input from the output. Such a
function g is called invertible; its inverse f is defined in words: the domain of f is the
range of g, and f (y) is defined as the number x for which g(x) = y. We denote the inverse
of g by .

By the way in which it is defined, we see that undoes, or reverses, g: it works

backward from the output of g to the input. If g is invertible, then g−1 is also invertible,
and its inverse is g. Furthermore, the composition of a function and its inverse, in either
order, is the identity function:
i.e., and

18
Example
Let , and restrict the domain of g to be x ≥ 0. Since the squares of two different
nonnegative numbers are different, g is invertible. Its inverse is (x) = √ x. Note that if
we had defined g(x) = and taken its domain to be all numbers, not just the nonnegative
ones, then g would not have been invertible, since (−x) = x . Thus, invertibility depends
crucially on what we take to be the domain of the function (Fig. below).

Left: x is plotted with the domain all numbers. Right: the domain is the positive numbers. Only one of these functions is
invertible

Monotonicity. The graph of a function can be very helpful in determining whether the
function is invertible. If lines parallel to the x-axis intersect the graph in at most one
point, then different domain values are assigned different range values, and the function
is invertible. Two kinds of functions that pass this “horizontal line test” are the increasing
functions and the decreasing functions.

Definition.
An increasing function is one for which f (a) < f (b) whenever a < b. A decreasing
function is one for which f (a) > f (b) whenever a < b. A nondecreasing function is one
for which f (a) ≤ f (b) whenever a < b. A nonincreasing function is one for which f (a) ≥ f
(b) whenever a < b.

Example
Suppose f is increasing and f (x ) > f (x 2 ). Which of the following is true?
(a) x 1 = x
(b) x > x
(c) x < x

Explanation
Item (a) is certainly not true, because then we would have f (x ) = f (x ). Item (b) is
consistent with f increasing, but this does not resolve the question. If item (c) were true,
then f (x ) < f (x ), which is not possible. So it is (b) after all.

19
Two graphs of monotonic functions. Left: increasing, Right: decreasing

The figure above shows the graphs of an increasing function and a decreasing function.
Both pass the horizontal line test and both are invertible.

Definition .
Functions that are either increasing or decreasing are called strictly monotonic. Function
that are either nonincreasing or nondecreasing are called monotonic.

If f is strictly monotonic, then the graph of its inverse is simply the reflection of the graph
of f across the line y = x ( see figure below).

Left: graphs of an increasing function f and its inverse g. Right: if you write f (x) = y and x = g(y), then the graph of
f (x) = y is also the graph of g(y) = x

The Inversion Theorem.

The graphs suggest the following theorem:

Inversion theorem.
Suppose that f is a continuous and strictly monotonic function defined on an interval
[a,b]. Then its inverse g is a continuous strictly monotonic function defined on the closed
interval between f (a) and f (b).

Proof.
A strictly monotonic function is invertible, because different inputs always result in
different outputs. The inverse is strictly monotonic.

20
What remains to be shown is that the domain of the inverse function is precisely the
closed interval between f (a) and f (b), no more no less, and that is continuous.

According to the intermediate value theorem, for every m between f (a) and f (b), there is
a number c such that m = f (c). Thus every number between f (a) and f (b) is in the
domain of the inverse function. On the other hand, the value f (c) of a strictly monotonic
function at the point c between a and b must lie between
f (a) and f (b). This shows that the domain of f−1 is precisely the closed interval between
f (a) and f (b).
Next, we show that f−1 is continuous. Let be any tolerance. Divide the interval [a,b]
into n subintervals of length less than with end points , , -----, =b

The values f ( ) divide the range of f into an equal number of subintervals. Denote By
the length of the smallest of these. (See Fig. below) Let and be numbers in the
range that are within of each other. Then and are in either the same or adjacent
subintervals of the range. Correspondingly and lie in the same or
adjacent subintervals of [a,b]. Since the lengths of the subintervals of [a,b]
were made less than , we have < .
Thus we have shown that given any tolerance , there is a such that if and differ
by less than , then and differ by less than . This shows that f is
uniformly continuous on [a,b], hence continuous.

The inverse of a continuous strictly monotonic function is continuous

As an application of the inversion theorem, take any positive integer. Then f

is continuous and increasing on every interval [0,b], so it has an inverse g. The value of g
at a is the nth root of a and is written with a fractional exponent:

By the inversion theorem, the nth-root function is continuous and strictly monotonic.
Then powers of such functions, such as x2/3 = (x1/3)2, are continuous and strictly
monotonic on [0,b]. The figure below shows some of these functions and their inverses.

21
The power functions

We shall see later that many important functions can be defined as the inverse of a strictly
monotonic continuous function and that we can make important deductions about a
function f from properties of its inverse f−1.

Exercise
1. Find the inverse function of . Sketch the graphs of f and .

2. The volume of water V in a bottle is a function of the height H of the water, say
V = f (H). (See Figure below). Similarly, the height of the water is a function of the
volume of water in the bottle, say H =g(V). Show that f and g are inverse functions.

A bottle of water for question 2 above

3. Let , , ,and . Find formulas for the

22
compositions
(a) (h ◦ g)(x)
(b) (g ◦ h)(x)
(c) ( f ◦ g)(x)
(d) (k ◦ h)(x)
(e) (h ◦ k)(x)
(f) (k ◦ g ◦ h)(x)

4. Is there a function f (x) = that is its own inverse function? Is there more than one
such function?
5. Show that the function , on domain x > 0, is increasing by explaining
each of the following items.
(a) The sum of two increasing functions is increasing.
(b) The functions x and are increasing.

6. Tell how to compose some of the functions defined in Problem 3 to produce the
functions
(a)
(b)
(c)

7. The graph of a function f on [0,a] is given in Fig. below.

The graph of the function f in Problem 7

Use the graph of f to sketch the graphs of the following functions.

(a)
(b)
(c)
(d)
(e)

8. Use the intermediate value theorem to show that the equation

has a solution in [−1,0].

9. (a) Show that the inverse of an increasing function is increasing.

23
(b) Then state the analogous result for decreasing functions.

10. (a) Suppose f is increasing. Is f ◦ f increasing? Give a proof or a counterexample.

(b) Suppose f is decreasing. Is f ◦ f decreasing? Give a proof or a counterexample.

11. Assume that functions f and g are increasing. Is f g increasing? If so give a proof, and
if not, explain why not.

24