Calculus I Workbook

Julie C. La Corte, PhD

Georgia State University

T
Dunwoody Campus

AF For use with OpenStax Calculus, Volume 1

DR
Created: September 25, 2020
Last revised: September 30, 2021

This document is covered by a Creative Commons Attribution-NonCommercial 4.0 International public license.
 Disclaimer about sources

This document presents lecture notes developed by the author while teaching Calculus at Texas
State University, UWM, Berry College, and GSU in the years 2006 through 2020. The handwritten
originals drew material from a variety of editions of a number of textbooks, including the Five
College Consortium’s Calculus in Context, George Thomas’ Calculus, James Stewart’s Calculus,
Claudia Neuhauser’s Calculus for Biology and Medicine, and Ross Middlemiss’s Di↵erential and
Integral Calculus.
A serious e↵ort has been made to remove direct quotations of expository material and exercises
the author suspects to have been taken from these copyrighted textbooks, and to provide citations
where the sources are known. As time allows, and to the best of her ability, she will continue to
purge and replace material whose original attribution is unknown to her.
However, the author no longer has access to every edition of every textbook from which she
taught Calculus over the years, and her memory regarding the origin of every exercise included in
this document is far from infallible. Doubtless some of the material appearing herein can also be
found in the sources from which she drew while compiling her handwritten notes.
The author also notes the exigency imposed by the COVID-19 pandemic, which was ongoing
when this document was created and hastily deployed.
To the best of the author’s knowledge, in the production of auxiliary teaching materials used for
pedagogical reasons and not for personal profit, the practice of recycling mathematical exercises
that may be found in other authors’ printed material is is protected by the legal doctrine of Fair
Use, as well as being consistent with long-established practices in tertiary mathematics education.
 Lesson 0: Sets and functions
Sets

A set is a well-defined collection of objects called members or elements.

If S is a set, the symbols
a2S
mean that a is a member of S, and the symbols
a2
/S
mean that a is not a member of S.

A set can be specified in one of several ways, which we will now discuss.

Defining a set by a rule:

+ R is the set of all numbers on the number line. The numbers on the number line are
called real numbers.

+ The symbols (-1, 1) mean the same thing as R: that is, the set of all numbers on
the number line.
Ex. 1. True/False/Makes no sense:
(a) ⇡ 2 R.
(b) 32/ R.
(c) R 2 2.
(d) 1 is a number on the number line.

Defining a set by listing its members (“roster notation”):

+ Curly braces { } mean “the set consisting of.”
Example:
The statement
S = {2, 4, 6, 8, . . . }
means
“S is the set consisting of 2, 4, 6, 8, and so on.”
This way of writing a set (that is, by listing its members) is called roster notation.

Ex. 2. True/False/Makes no sense:

(a) 10 2 {2, 4, 6, 8}.
(b) 10 2 {2, 4, 6, 8, . . . }.
(c) 10 2 2, 4, 6, 8.
Set-builder notation:
+ The vertical bar | means “such that.”
Example:
The statement
S = {x 2 R | x < 0}
means
“S is the set consisting of x in the set of real numbers such that x < 0.”
This way of writing a set (that is, by specifying a test for membership after the vertical bar)
is called set-builder notation.

Ex. 3. Write in set-builder notation.

(a) The set of numbers x such that 0 6 x < 1.
(b) The set of numbers x such that x > 0.

Some special sets:

The set of integers is
Z = {. . . , -3, -2, -1, 0, 1, 2, 3, . . . }.

The set of natural numbers is

N = {n 2 Z | n > 1}.

The Cartesian plane R2 is the set of points (or ordered pairs) (x, y), where x, y 2 R.
We can write R2 in set-builder notation as follows:
R2 = (x, y) | x 2 R and y 2 R .

Ex. 4. True/False/Makes no sense:

(a) Every member of Z is a member of R.
(b) Every member of N is a member of R.
(c) Every member of R2 is a number.

Ex. 5. Write in set-builder notation:

The set of fractions n/d such that n and d are real numbers, and d is not equal to 0.

The set in Ex. 5 is called the set of rational numbers. It is denoted by Q.

+ You’re not required to memorize the meaning of the symbols Z, N, R2 , and Q. (But
you should memorize the meaning of the symbol R.)
Intersections and unions:
+ Let A and B be two sets. Their intersection A \ B is the set of members both of
A and of B. Their union A [ B is the set of members either of A or of B. In set-builder
notation,

A \ B = {x x 2 A and x 2 B},
A [ B = {x x 2 A or x 2 B}.

Ex. 6. Write the set [2, 4] \ (3, 1) in the simplest possible notation. (Hint: Draw the
number line and, using two di↵erent colors, shade the regions corresponding to the two sets
[2, 4] and (3, 1).)

Ex. 7.
(a) How many numbers are members of the set {1, 2}?
(b) How many numbers are members of the set (1, 2)?
(c) How many numbers are members of the set [1, 2)?
 Functions

Definition: A variable y is said to be a function of a second variable x if a relation exists between

them such that to each of a certain set of values of x, there corresponds exactly one value of y.
• The amount y of postage you pay to ship a package is a function of the weight x of the
package.
• The volume V of a cubical box (that is, the amount of space inside the box) is a function of
the length s of its side.
• The position D of a car’s gasoline gauge (see image below) is a function of the amount v of
gas in the tank.

Definition: If y is a function of x, we call y the dependent variable (or output), we call x the
independent variable (or input), and we say that y depends on x.

Ex. 8. The equation

4x + 2y = 12
is a relation between x and y such that to every value of x, there corresponds exactly one value of
y.
• If x = 0, then y = .
• If x = 1, then y = .
It appears that y is a function of x. Is x a function of y?

Ex. 9. Consider the following relation between the two variables x and y.

y = x2 .

Is y a function of x? Is x a function of y?
Some functions are not described by algebraic formulas.
Instead, a procedure may be given for determining the output for a given input, as in the following
example.

Ex. 10. Define two functions cos and sin as follows.

Recall that the unit circle is the circle of radius 1 in the Cartesian plane centered at (0, 0).
Given a real number t, locate a point P on the unit circle by tracing an arc of length |t| along the
circle starting from the point (1, 0).
• If t is positive, trace the arc counter-clockwise.
• Trace the arc clockwise if t is negative.
Then the outputs cos(t) and sin(t) are defined to be the coordinates of the point P, as shown:

1.0 1.0

P=HcosHtL, sinHtLL

»t»
0.5 0.5

t
-1.0 -0.5 0.5 1.0 -1.0 -0.5 0.5 1.0
t
»t»
-0.5 -0.5

P=HcosHtL, sinHtLL
-1.0 -1.0

Ex. 11. Explain why we can’t solve the equation cos(t) = 1/t by writing t = 1/ cos.
 Function notation

A function is a relationship between two variables—an input variable and an output variable whose
value depends on the value of the input.
In Ex. 10, we saw that a function can be given a name (like cos).
When we give a function a name, we often use function notation:
If “f” is the name of a function, we write

f(t)

(read as “f of t”) for the value of the output of f when the input value is t.

This does not mean “f times t.” (See Ex. 11.)

Ex. 12. V(s) = volume of a cube with side length s. Evaluate the expression V(2).

Ex. 13. Define a function y = f(x) by the relation y = x2 . Evaluate the expression

f(3 + h) - f(3).

Hint: Start by writing

⇥ ⇤ ⇥ ⇤
f(3 + h) - f(3) = -

and then fill in the first blank with the value of y = f(x) = x2 when x = 3 + h, and the
second blank with the value of f(x) = x2 when x = h. Then simplify.

Ex. 14. Define a function S(n) by the relation

-n if n is even,
S(n) =
0 if n is odd.

(a) What’s the value of S(0)?

(b) What’s the value of S(-3)?
(c) (Sneaky question:) What’s the value of S(⇡)?
 The domain of a function

The output of a function f need not be defined for every input value. (See Ex. 14(c).)

Definition. The set of all input values for which f is defined is called the domain of f.

Ex. 15.
(a) What is the domain of the function f(x) = x2 - 1?
(b) What is the domain of the function A(r) = area of a circle of radius r? (Is 0 a member of the
domain of A?)
p p
(c) What is the domain of the function g(x) = - x? (Recall: the symbol means the
nonnegative square root.)

Convention: In this class, our variables will always represent real numbers, unless explicitly stated
otherwise.

Graphs and the Vertical Line Test

Definition. The graph of an equation involving x and y is the set of ordered pairs (x, y) 2 R2
satisfying the equation.
y

Hx-2L2 +Hy-2L2 =4
5

x
-1 1 2 3 4 5
-1

Definition. The graph of a function f is the graph of the equation y = f(x).

y= 4 - Hx - 2L2 + 2
5

x
-1 1 2 3 4 5
-1

+ In symbols, the graph of a function f is the set (x, y) 2 R2 y = f(x) .

Recall: To determine whether a graph is the graph of a function, use the Vertical Line Test:
• If no vertical line meets the graph in more than one point, then the graph is the graph of a
function.

Ex. 16.
(a) Graph y = -x + 2. Is y a function of x?
(b) Graph y = x4 . Is y a function of x?
(c) Graph x = y4 . Is y a function of x?

Symmetry
Definition.
• A function f is even if f(x) = f(-x) for all x 2 Domain(f).
• A function f is odd if -f(x) = f(-x) for all x 2 Domain(f).

To determine if a function f is even, odd, both, or neither using algebra only, simplify the expressions
-f(x) and f(-x), and then check the definitions above.

Ex. 17. Using algebra only, determine whether the function is even, odd, both, or neither.
1
• f(x) =
1 - x2 - x4
• g(x) = x7 + x5 - x4 + x
• h(x) = 2x2 + x
• j(x) = 3
p
• k(x) = x

Ex. 18. Suppose f is a function with domain R = (-1, 1), and f(3) = 20.
(i) If f is even, then f(-3) = .
(ii) If f is odd, then f(-3) = .
Solution:
(f even)
(i) f(-3) = f(3) = 20.
(f odd)
(ii) f(-3) = -f(3) = -20.

How can we tell whether a function is even or odd from its graph?
• A function is even if its graph has mirror symmetry in the y-axis.
• A function is odd if it is symmetric in the origin (has 180 rotational symmetry).
Ex. 19 (Challenge). How many functions with domain R are both even and odd?

Increasing and decreasing

Definition. Let I be an interval.
• A function f is increasing on I if
f(x1 ) < f(x2 )
for every x1 , x2 2 I such that x1 < x2 .
• A function f is decreasing on I if

f(x1 ) > f(x2 )

for every x1 , x2 2 I such that x1 < x2 .

Ex. 20. Consider the absolute value function A(x) = |x|,

x if x > 0,
A(x) = |x| =
-x if x < 0.

On what interval(s) is this function increasing? On what interval(s) is it decreasing?

 Repertoire of basic functions

Definition. A power function is a function defined by a relation of the form

f(x) = xa

where a is a constant.
• When a = 1, we call f the identity function.
• When a = -1, we call f the reciprocal function.
• When a = 1/n and n > 2 is a positive integer, we call f a root function.
p
(Recall that x1/n = n x for any natural number n > 2.)

Ex. 21. What is the domain of f(x) = xa when a = -1?

Solution.
1
f(x) = x-1 = .
x
This is defined when x 6= 0, undefined when x = 0.
Answer: {x | x 6= 0} .

Definition. A rational function is a ratio of two polynomial functions,

N(x)
R(x) = .
D(x)

Fact: The domain of a rational function is

{x | D(x) 6= 0}.

x-1
Ex. 22. Find the domains of g(x) = and h(x) = 3x.
x2 - 1
What’s wrong with the following reasoning?
x-1 1
Since g(x) = = , the domain of g is {x x 6= 1}.
(x + 1)(x - 1) x+1
Solution.
Set the denominator D(x) = 0:

x2 - 1 = 0 makes g(x) undefined

x = ±1

Answer: {x | x 6= ±1}
 Now we’ll look at h(x) = 3x. First we rewrite h(x) to spell out what the denominator is. . .
3x
h(x) = 3x =
1
. . . and then we set the denominator = 0:
1=0 makes h(x) undefined
No choice of x makes 1 = 0, obviously! So:
Answer: R

4x
Ex. 23. Find the domain of F(x) = p .
x - 4(x - 2)
Solution.
F is not a rational function.
But we can still ask the question, for what values of x is F(x) defined?

For F(x) to be defined, we need:

p
(1) the denominator x - 4(x - 2) 6= 0, and
p
(2) x - 4 must be a real number.

First we’ll check (1): when is the denominator equal to 0?

p
x - 4(x - 2) = 0
p
x - 4 = 0 or x = 2
x = 4 or x = 2
p
So 4 and 2 are not in the domain of F. Now we check (2): when is x - 4 a real number?

x-4 > 0
x>4
So only numbers x such that x > 4 can be in the domain of F. But note that we already
knew that 4 is not in the domain of F.
Answer: {x | x > 4}

Definition. A function f is called an algebraic function if it can be constructed using algebraic

operations (such as addition, subtraction, multiplication, and taking roots) starting with polynomi-
als.
Definition. A function f is called a transcendental function if it is not an algebraic func-
tion.

Ex. 24. Give three examples of algebraic functions. Then do the same for transcendental functions.
Is the absolute value function algebraic? Is exp algebraic? How about ln?
 Composition of functions

A function may be viewed as a machine, or a “black box,” that “eats” an input value x and “spits
out” the corresponding output value y = f(x):
f
x 7! y = f(x)
(the symbol 7! is read as “maps to”).

We can combine two functions f and g by doing them sequentially, one after another:
f g
x 7! y = f(x) 7! z = g f(x)
We call this the composite of f and g. It is a new, third function, which we give the symbol g f
(notice the order of f and g in this notation).
g f
x 7! z
You may find it helpful to read the symbol as “after.”

Definition. Given two functions f and g, the composite function g f (also called the composition
of g and f) is defined by the relation
(g f)(x) = g f(x) .

+ The domain of g f is the intersection of two sets:

\
Domain(f) {x f(x) 2 Domain(g)}.

Ex. 25. Find formulas for f g and g f if f and g are defined by

p
f(x) = x - 2, g(x) = x + 7.

Ex. 26. Find the domain of f g and the domain of g f, where f and g are as in the previous
example.
p
Ex. 27. Find a formula for G F, where F(x) = x + 3 and G(y) = y.
Domain of F = R.
Domain of G = [0, 1).
When is F(x) in the domain of G?
F(x) > 0
x > -3

Domain of G F:
\
Domain(f) {x f(x) 2 Domain(g)} = R \ [-3, 1) = [-3, 1) .
 Function arithmetic

Functions can also be combined using the arithmetic operations +, -, ·, and /.

For example, if f and g are two functions, we can define a new function h(x) = f(x) + g(x). The
name of this new function is f + g:

f + g (x) = f(x) + g(x).

(We write parentheses around the name of the function f + g because, if we didn’t, we’d end up
writing f + g(x), which, if you think about it, doesn’t make any sense—take the name of the
function f and add it to the output of g?)
The domain of the new function f + g is Domain(f) \ Domain(g).

Ex. 28.
• What’s the domain of f - g (x) = f(x) · g(x)? In other words, for what x-values is the
expression f(x) - g(x) defined?
• What’s the domain of fg (x) = f(x) · g(x)?
f(x)
• What’s the domain of f g (x) = ? (Careful!)
g(x)
 More preliminaries

This document is a refresher of some—but certainly not all—of the concepts you’ll need in
Calculus 1.

If it’s been a while since you’ve taken Precalculus, it is strongly recommended that you brush up on
all the material in Chapter 1 of our textbook with the possible exception of the hyperbolic functions
(see ? below).

In addition to the material covered in this document, be sure you are comfortable with:
• Linear functions (Section 1.2 in the OpenStax textbook)
• Point-slope and slope-intercept form of the equation of a line (Section 1.2)
• Polynomial and power functions (Section 1.2)
• “Behavior at infinity,” also known as “end behavior” (Section 1.2)
• Piecewise-defined functions (Section 1.2)
• Transformations of functions (Section 1.2)
• Radian measure (Section 1.3)
• The six basic trigonometric functions sin, cos, tan, csc, sec, and cot (Section 1.3)
• Trigonometric identities (Section 1.3)
• Inverse functions (Section 1.4)
• One-to-one functions, the Horizontal Line Test, and restrictions of a function’s domain
(Section 1.4)
• The inverse trigonometric functions sin-1 , cos-1 , tan-1 , csc-1 , sec-1 , cot-1 (Section 1.4)
• The laws of exponents (Section 1.5)
• Exponential functions f(x) = bx (Section 1.5)
• The number e (Section 1.5)
• Logarithmic functions f(x) = logb (x) (Section 1.5)
• Changing between logarithmic bases (Section 1.5)

?: Studying the hyperbolic trigonometric functions sinh, cosh, tanh, csch, sech, and coth
(Section 1.5) may be useful to students of electrical engineering, architecture, or physics, but
we won’t be discussing them in our course.
 Workbook Lesson 1
§2.1, Introduction to Calculus
Last revised: 2021-08-20 19:34

If the world and everything in it was flat—that is to say, linear —we wouldn’t need calculus.
First question: How fast is this line rising?

(4,1)

1 2 3 4 5 6 7

Algebra is all we need to answer this question.

y
The slope of the line, , tells us the line rises y units for every x units we move to the
x
right.

y 1
m= =
x 2
It doesn’t matter what point on the line we start from. The slope is constant—the line everywhere
rises at the same rate.
Second question: what’s the area between the line and the x-axis, let’s say between the two
values of x marked on the graph below?

1 2 3 4 5 6 7
a b

If we know the slope of the line is 12 , this is an easy geometry problem.

1 2 3 4 5 6 7
a b

The base of the triangle is

b a=4
units long. The slope (“rise over run”) tells us the height of the triangle must be 2.
Once we know the base and the height of the triangle, we can easily find its area:
1
A= 2
⇥ base ⇥ height = 4
 Average rate of change and the tangent problem

As soon as we start dealing with curves, it’s less obvious how to proceed.
How fast is this curve rising?
y = x2
9
8
7
6
5
4
3
2
1
x
-3 -2 -1 1 2 3

First of all, on one side of the y-axis, the curve isn’t rising at all—it’s falling.
Okay, so how fast is the curve rising for non-negative x?
Well, that depends on what point you’re starting from. If you’re starting at x = 0, the curve is
rising slowly. If you’re starting at, say, x = 2, the curve rises a lot faster.
We could use algebra to give a rough idea of how fast this curve rises. We could mark o↵ two
points on the curve and say something like, “the average rate of change from x = 0 to x = 3 is
the slope of the line through these two points.”
y = x2
9
8
7
6
5
4
3
2
1
x
-3 -2 -1 1 2 3
a b
Without calculus, that’s the best we can do—an approximation of how fast the curve rises,
depending on what part of the curve we’re looking at.
This is exactly the idea of the slope of the secant line, which you may have heard about in a previous
class.

Definition. Let y = f (x) be a function.

• The graph of the function f is the set of points x, f (x) for all x in the domain of f .
• A secant line of the graph is a line that meets the graph in two points, a, f (a) and
b, f (b) .
• The average rate of change in f (x) from x = a to x = b is the slope of the line through
the points a, f (a) and b, f (b) .
+ The symbol in an expression of the form means “the change in .” For a point on
the graph of y = f (x), the change in y, which we write as y, is the change in f (x). So the slope
of the secant line to the graph of y = f (x) through the points a, f (a) and b, f (b) is
y change in y f (b) f (a)
= = .
x change in x b a

Ex. 1. Find the average rate of change in f (x) = x2 from x = 0 to x = 3.

Solution to Ex. 1:
To find the two points a, f (a) and b, f (b) on the secant line, we take a = 0 and b = 3
and substitute the two values into the formula f (x) = x2 :

a= 0 f (0) = 02 = 0
b= 3 f (3) = 32 = 9

a, f (a) = (0, 0)
b, f (b) = (3, 9)
The average rate of change is the slope of the line through these two points:
y 9 0
= = 3
x 3 0

The problem with the average rate of change is, although it always approximates how fast the curve
rises or falls, sometimes it’s a bad approximation of how fast the height changes starting from a
given point (say, in the following example, starting from x = 0).

Ex. 2. Find the average rate of change in y = sin(x). . .

(a) . . . from x = 0 to x = ⇡.
y = sin(x)

1
3 /2
2 /2
1/2

π π π x
-π π
6 3 2

-1
We could get a better approximation if we take the two points of the secant line closer together.

Ex. 2 (continued). Find the average rate of change in y = sin(x).

(b) from x = 0 to x = ⇡/2. (Round to four decimal places for parts (b)–(d).)
(c) from x = 0 to x = ⇡/3.
(d) from x = 0 to x = ⇡/6.
Answers to Ex. 2:
The average rate of change in sin(x). . .
(a) . . . from x = 0 to x = ⇡ is 0 .
(b) . . . from x = 0 to x = ⇡/2 is 0.6366 .
(c) . . . from x = 0 to x = ⇡/3 is 0.8270 .
(d) . . . from x = 0 to x = ⇡/6 is 0.9549 .

What are these numbers telling us? It looks like as we take the rightmost point closer to x = 0,
the slope gets larger—which means, the curve rises more and more sharply.
y = sin(x)

1
3 /2
2 /2
1/2

π π π x
-π π
6 3 2

-1

But although the slopes are increasing as the rightmost point gets closer to x = 0, we know from
the graph that there’s a limit to how large the slope can get. The secant line is never going to look
anything like, say, a vertical line (which has infinite slope).
We now ask, will the slopes of the secant lines “home in” on some exact number as we take the
rightmost point closer and closer to x = 0? And if so, what is that exact number? Our textbook
calls this question the tangent problem.
To solve the tangent problem, we need more than algebra and geometry. We need calculus.
We’re not going to solve the tangent problem in this lesson. For now, I just want you to get a
sense of what algebra can’t do that calculus can.

Definition. Let y = f (x) be a function.

• The instantaneous rate of change of a function f (x) at x = a is the value to which
the average rate of change on intervals [a, b] approaches as b approaches a, provided that
there is such a value.
• The tangent line at x = a is the line through the point a, f (a) whose slope is the
instantaneous rate of change at x = a.

The method of exhaustion and the area problem

Let’s ask another question for which algebra and geometry can’t give us an exact answer. What’s
the area under the curve y = x2 from x = 0 to x = 1?
y = x2

x
1

Again, we can come up with an approximation.

The idea here is an ancient one. The Greeks figured out how to approximate the area of a circle
using what they called the method of exhaustion:

��

We can approximate the area under a curve by trying to “exhaust”—that is, cover all of—the area
by rectangles.
 y = x2 y = x2 y = x2 y = x2

1 1 1 1
��

x x x x
1 1 1 1

We can’t actually ever cover all the area with rectangles because one side of the region is curved.
But if you know what a pixel is, you understand that we can get pretty close to a curved shape using
blocks, provided that we make the blocks small enough to fit in a large number of them.
In the case of finding the area under a curve, we take more and more rectangles, thinner and thinner.
The total area of all the rectangles becomes a better approximation of the actual area.
As we take more and more, thinner and thinner rectangles, we know that the total area homes in
on some number—namely, the actual area under the curve. But to find that number exactly, we
need calculus. Geometry and algebra can’t do the job.

Limiting processes

What we’ve seen today are two examples of what we might call limiting processes.
• In the case of the area under a curve, we ask whether the total area of the rectangles “homes
in” on some number—which we’ll learn is called a limit—as the number of rectangles becomes
infinitely large.
• In the case of instantaneous velocity, we ask whether the slope of the secant line through two
points homes in on some number—a limit—as the distance between the two points becomes
infinitely small.
In calculus, we learn how to calculate with infinities. We find out that it’s not as hard as it sounds.
But you can understand why it took thousands of years for human beings to figure out how to do
it. For the ancient Greeks, infinity was a mysterious, mystical concept. Honestly, infinity seems
mysterious and mystical to most people—at least at first.
Infinity was tamed by Newton and Leibniz when they discovered the methods of calculus. You will
learn how to tame the infinite, too. Does that sound impressive? Taming the infinite sounds pretty
impressive to me.
But, you may wonder, what’s the point of it? What’s it useful for? How will calculus help me when
I’m working at a job in marketing, or in engineering, or in psychology, or in the social sciences?
Calculus is the mathematics of change. Whatever career you’re interested in, change plays some
role.
• When your company raises the price of a product they’re selling, how will that change the
number of units you sell? That’s the idea of marginal demand, which we calculate the same
way as instantaneous velocity.
 • When we ask someone to learn a more difficult task, how much longer will it take them to
learn it? That question (of psychology) can be answered by finding a tangent line.
• How much water can a curved tank hold? In Calculus II and III, future engineers learn how
to solve that problem, which is similar to finding the area under a curve.

Let’s do some practice problems, now that we have an idea where this is all leading.

Average velocity and instantaneous velocity

Definition. Suppose the position of a particle in motion is given by a function s(t). (For example,
if a stone is dropped from a great height, its position function is the particle’s height above ground
level.)
• The particle’s average velocity from time t = a to time t = b is the average rate of change
in its position function:
s(b) s(a)
b a
• The instantaneous velocity of a particle in motion is the instantaneous rate of change in its
position function—that is, the value to which the average velocity (from time a to time b)
approaches as b approaches a, provided that there is such a value.

Ex. 3 (§2.1—#1). Find the slope of the secant line to the graph of f (t) = t2 + 1 passing through
the point P = (1, 2) and the point Q = t, f (t) when. . .

(a) t = 1.1 (b) t = 1.01 (c) t = 1.001 (d) t = 1.0001

Then estimate the instantaneous velocity at t = 1 of a particle in motion whose position is given
by f (t) = t2 + 1.
Ex. 4. A stone is dropped from a window 300 feet above ground level. Its height in feet after t
seconds is h(t) = 300 16t2 . Find the average velocity of the stone for the time period beginning
when t = 3 and lasting 0.5 seconds.
Solution:
The average velocity is the slope of the secant line through two points.
• The first point is 3, h(3) . It corresponds to the beginning of the time period.
• The second point, which corresponds to the end of the time period 0.5 seconds later,
is 3.5, h(3.5) .
• Substituting t = 3 and t = 3.5 into the position function h(t) yields

h(3) = 156, h(3.5) = 104.

So the slope of the secant line through 3, h(3) and 3.5, h(3.5) is

h(3.5) h(3) 104 156

= = 104
3.5 3 0.5

Answer: Falling at 104 feet per second

Ex. 5 (§2.1—#18). A stone is tossed into the air from ground level with an initial velocity of
15 m/sec. Its height in meters after t seconds is h(t) = 15t 4.9t2 . Find the average velocity of
the stone for each of the given time intervals.
(a) [1, 1.05] (b) [1, 1.01] (c) [1, 1.005] (d) [1, 1.001]

Ex. 6 (§2.1—#19). Use the preceding exercise to estimate the instantaneous velocity of the stone
at t = 1 sec.
Ex. 7.
(a) Sketch the graph of f (x) = |x| over the interval [ 2, 4] and shade the region above the x-axis.
(b) Approximate the area between the graph of f and the x-axis from x = 2 to x = 4 by drawing
six rectangles and calculating their total area.
(c) Find the exact area between the graph of f and the x-axis from x = 2 to x = 4 using
geometry.

Solution:

First we shaded the region between the graph and the x-axis (yellow). Then we drew rectangles
contained in the shaded region.
�(�)=|�|
4

3 Applying the formula for the area of a

rectangle to each of the rectangles shown,
2 we get

1 Total area = 1 + 1 + 2 + 3
= 7.
-4 -3 -2 -1 1 2 3 4
�

�(�)=|�|
4

3 Applying the formula for the area of a

triangle to the two triangles shown,
2

Total area = 12 (2)(2) + 12 (4)(4)

1

= 10 .
-4 -3 -2 -1 1 2 3 4
�

Ex. 8.
p
(a) (§2.1—#26) Sketch the graph of y = 1 x2 over the interval [ 1, 1]. (Hint: Square both
sides of the equation and recall the standard form of the equation of a circle.)
p
(b) Approximate the area between the graph of y = 1 x2 and the x-axis from x = 1 to
x = 1 by drawing six rectangles and calculating their total area.
p
(c) Find the exact area between the graph of y = 1 x2 and the x-axis from x = 1 to x = 1
using geometry.
 Workbook Lesson 2
§2.2, Limit of a Function (Informal Definitions)
Last revised: 2021-05-28 10:35

Objectives
• Using correct notation, describe the limit of a function.
• Use a table of values to estimate the limit of a function or to identify when the limit does not exist. (Moved
to Lesson 5, §2.5)
• Define one-sided limits and provide examples.
• Explain the relationship between one-sided and two-sided limits.
• Using correct notation, describe an infinite limit.
• Define a vertical asymptote.

Ten simple rules (The Field Axioms)

The real numbers are the numbers on the number line. These include rational numbers and
irrational numbers, but not imaginary numbers, and certainly not 1, which isn’t a number at
all.
We denote the set of all real numbers either by the symbols ( 1, 1) or the symbol R (“dou-
blestruck” or “blackboard bold” R).
The real numbers, together with the operations of addition and multiplication, are an example of
what algebraists call a field.
The rules for doing arithmetic with numbers in a field are called the Field Axioms:
For any real numbers a, b, and c:
name addition multiplication
associativity (a + b) + c = a + (b + c) (ab)c = a(bc)
commutativity a+b=b+a ab = ba
distributivity a(b + c) = ab + ac (a + b)c = ac + bc
identity a+0=a=0+a a·1=a=1·a
inverses a + ( a) = 0 = ( a) + a aa 1 = 1 = a 1 a if a 6= 0
1
+ Recall that a 1
is another way of writing the reciprocal of a, that is,
a
.

These rules are as natural to us as walking, or throwing a ball. We don’t always think about the
rules of a field when we do arithmetic, but we should—all the alegbra we learn prior to calculus
follows from these ten simple rules.

Calculus begins with the definition of additional rules, defined in terms of what we call limits.
About 90% of undergraduate calculus involves a limit of one kind or another (e.g., derivatives,
integrals, limits of sequences, series), so it’s worth taking the time to understand this concept.
 The limit of a function: Informal definition
Let f be a function, and let a be a constant. The statement

lim f (x) = L
x!a

says that the limit of f (x), as x approaches a, is the number L.

Informal Definition 1: This means that the output f (x) can be made as near to L as we like, if
we take x sufficiently near a, but not equal to a.
+ It has nothing whatsoever to do with the value of f (x) when x equals a.

For example, we can say

lim x2 + 1 = 1
x!0

because the value of x2 + 1 is as close as we want to 1, whenever x is sufficiently near 0.

�(�) = �� +�
11
10
9
8
7
6
5
4
3
2
1
�
-3 -2 -1 1 2 3

�
(See applet on iCollege: “First example of a limit’’)
It might be that the value of the function x2 + 1 is 1 when x equals 0, but in general this has
nothing to do with the idea of a limit.

Consider a less obvious example:

x2 4
lim = 4.
x!2 x 2

What is the value of

x2 4
x 2
when x = 2?
0
Undefined: substituting gives , a meaningless expression.
0
 But for all x 6= 2 near 2, the value is near 4.
Indeed, since ⇠
x2 4 (x⇠⇠2)
(x + 2)⇠
= =x+2
x 2 x ⇠⇠
⇠ 2
for any x 6= 2, the graph of
x2 4
x 2
is identical to the graph of
x+2
everywhere except at x = 2.
�� - �
�(�) =
�-�

�
-3 -2 -1 1 2 3 4

-1 �

x2 4
From the graph, we see that the value of gets as close as we like to 4 if we take x
x 2
close enough to 2.

The takeaway from Informal Definition 1:

• The expression “lim f (x) ” represents the number that the output f (x) gets close to,
x!a
whenever the input x is close to a.
• If we want to abbreviate lim f (x) by a letter—say, L—we can write:
x!a

lim f (x) = L
x!a

• When we say that f (x) gets close to L, or x gets close to a, how “close” is close enough?
We’ll answer this question precisely in Section 2.5. For now, think of “close” as meaning “as
close as you like.”

Alternate notation for lim f (x) = L:

x!a

As x ! a, f (x) ! L.

• The symbol ! can be read aloud as “approaches” or “gets close to.”

 Determining a one-sided or two-sided limit of a function from its graph

It’s important to be able to “eyeball” the limit of a function from its graph.
Ex. 1. The Heaviside function is
⇢
0 if x < 0
H(x) =
1 if x 0

Find lim H(x) by inspecting the graph.

x!2

�(�)

Ex. 2. Does lim H(x) exist? That is, is there a number L such that lim H(x) = L? If so, say
x!0 x!0
what L is. If not, say why not.
Answer:
As x approaches 0 from the right, H(x) approaches 1.
As x approaches 0 from the left, H(x) approaches 0.
There is no single number that H(x) approaches as x approaches 0.
Therefore, lim H(x) does not exist.
x!0

Informal Definition 2: The statement

lim f (x) = L
x!a

says that the lefthand limit of f (x), as x approaches a, is the number L.

• It means that the output f (x) can be made as near to L as we like, for x < a sufficiently
near a.

Informal Definition 3: The statement

lim f (x) = L
x!a+

says that the righthand limit of f (x), as x approaches a, is the number L.

 • It means that the output f (x) can be made as near to L as we like, for x > a sufficiently
near a.
Theorem. lim f (x) = L if, and only if, lim f (x) = L = lim+ f (x).
x!a x!a x!a

If the lefthand limit lim f (x) and the righthand limit lim+ f (x) are not the same, then we say
x!a x!a
the “two-sided” limit lim f (x) does not exist.
x!a

y=f (x)
2.5

2.0
lim f (x) = 1
x!1
1.5
lim f (x) = 2
x!1+
1.0
lim f (x) does not exist
x!1
0.5

0.0 x
0.0 0.5 1.0 1.5 2.0
Ex. 3. The graph of a function y = f (x) is given below. Find the lefthand limit lim f (x) and the
x!2
righthand limit lim+ f (x). If lim f (x) exists, state its value; if it does not exist, give your answer
x!2 x!2
as “DNE.”
�-�
�(�) =
�-�

1.0

0.5

�
-3 -2 -1 1 2 3 4
�
-0.5

-1.0

(See applet on iCollege: “Visualizing one-sided and two-sided limits’)

Ex. 4 (compare with §2.2, #46–49) The graph of a function y = f (x) is given. State the
value of each quantity, if it exists. If it does not exist, explain why not.
156 Chapter 2 | Limits

(a) lim f (x)

calculator
x!6 to graph the function and determine the limit.
Was the conjecture correct? If not, why does the method of
tables fail?
(b) lim+ f (x)
44. x!6

⎛ ⎞ ⎛ ⎞
(c)θ lim f (x) ⎝ ⎠ θ ⎝ ⎠
x!6

−0.1 a. 0.1 e.
(d) f (6)
−0.01 b. 0.01 f.
(e) lim f (x)
x! 8
−0.001 c. 0.001 g.

(f)−0.0001
lim + f d.
(x) 0.0001 h.
x! 8

(g) lim f (x)

45. x! 8

⎛ ⎞
(h) f (8) a ⎝ ⎠ 46.

0.1 a. 47.

0.01 b. 48.

49.
 Infinite limits

1
Consider the graph of f (x) = . Does the limit of f (x) as x approaches 2 exist?
(x 2)2
�
�(�) =
(� - �)�
10

�
�
-3 -2 -1 1 2 3 4

The graph shows that the output f (x) can be made as large as we like by taking x near 2. Although
the outputs do not approach a number, we can say something about how f behaves close to 2.

Informal Definition 4:
• A (one-sided or two-sided) limit of f (x) as x approaches a is positive infinity if the values of
f (x) increase without bound as x approaches a.
• A (one-sided or two-sided) limit of f (x) as x approaches a is negative infinity if the values of
f (x) decrease without bound as x approaches a.
+ • The symbol 1 is not a number. In particular, we should not substitute x = ±1 in
an equation or an expression. We use the symbols ±1 only to indicate the behavior of
f (x) as x ! a.

Given the previous graph, we see that lim f (x) = 1 and lim+ f (x) = 1.
x!2 x!2

Since the one-sided limits as x ! 2 are the same, we see that lim f (x) = 1.
x!2

Definition: The graph of a function f (x) has a vertical asymptote at x = a if one or both of
the one-sided limits of f (x) as x ! a is positive or negative infinity.
 1 1 1
Ex. 5. Graph f (x) = . (Hint: is a transformation of .)
x 3 x 3 x
Write the equation of its vertical asymptote.
Then state the value of each quantity, if it exists. If it does not exist, explain why not.

(a) lim f (x)

x!3

(b) lim+ f (x)

x!3

(c) lim f (x)

x!3

(d) f (3)

Additional exercises

Ex. 6 (§2.2—#76). Sketch the graph of a function f with the given properties.

lim f (x) = 1,
x!2

lim f (x) = 3,
x!4

lim f (x) = 6,
x!4+

f (4) is not defined.

Ex. 7 (§2.2—#77). Sketch the graph of a function f with the given properties.

As x ! 1, f (x) ! 0. lim f (x) = 1. lim f (x) = 1.

x!1 x!1+
lim f (x) = f (0). f (0) = 1. As x ! 1, f (x) ! 1.
x!0
Ex. 8. Plot the graph of the function
x+2
f (x) =
x 3
using a graphing calculator, an online graphing tool
(like Geogebra), or software (like Mathematica or
Apple Grapher). (You may want to copy a sketch
of the graph in the space provided.) Use the graph to
state the value of each limit, if it exists. If it does not
exist, explain why.

(a) lim f (x) (b) lim+ f (x) (c) lim f (x)

x!3 x!3 x!3

Ex. 9. Plot the graph of the function

r
x2 + 1
f (x) = x
x2
using technology. Use the graph to state the value of
each limit, if it exists. If it does not exist, explain why.

(a) lim f (x) (b) lim+ f (x) (c) lim f (x)

x!0 x!0 x!0

Ex. 10. Sketch the graph of the piecewise-defined function below. Then use the graph to determine
the values of a for which lim f (x) exists.
x!a

8
< 2x + 1 if x  0
f (x) = 4x if 0 < x < 1
:
(x + 1)2 if x 1
 Workbook Lesson 3
§2.3, The Limit Laws
Last revised: 2021-06-17 13:29

Objectives
• Recognize the basic limit laws.
• Use the limit laws to evaluate the limit of a function.
• Evaluate the limit of a function by factoring.
• Use the limit laws to evaluate the limit of a polynomial or rational function.
• Evaluate the limit of a function by factoring or by using conjugates.
• Evaluate the limit of a function by using the Squeeze Theorem.
sin(x)
• Evaluate lim .
x!0 x

To find the exact value of a limit, we often use the following properties, called the Limit Laws.

In the equations below, a and c are constants, n is a positive integer, and f (x) and g(x)
are functions defined for all x 6= a in some open interval containing a.

Identity Law lim x = a

x!a

Constant Law lim c = c

x!a
⇥ ⇤
Sum Law lim f (x) + g(x) = lim f (x) + lim g(x)
x!a x!a x!a
⇥ ⇤
Di↵erence Law lim f (x) g(x) = lim f (x) lim g(x)
x!a x!a x!a

Constant Multiple Law lim cf (x) = c lim f (x)

x!a x!a
⇥ ⇤
Product Law lim f (x) g(x) = lim f (x) · lim g(x)
x!a x!a x!a

f (x) lim f (x)

Quotient Law lim = x!a if lim g(x) 6= 0
x!a g(x) lim g(x) x!a
x!a
⇥ ⇤n h in
Power Law lim f (x) = lim f (x)
x!a x!a
p p
Root Law lim n
x= n
a
x!a

(If n is even, we assume a > 0.)

Ex. 1. Derive the rule lim xn = an (from the Identity Law and Power Law).
x!a
 1
Ex. 2. Use the Limit Laws to find a formula for lim . When does this limit exist?
x!a f (x)

Ex. 3. Calculate lim (x4 3x)(x2 + 5x + 3). Justify each step by indicating the appropriate Limit
x! 1
Law.

lim (x4 3x)(x2 + 5x + 3) = lim (x4 3x) · lim (x2 + 5x + 3) (Product)

x! 1 x! 1 x! 1
 
= lim x4 lim 3x · lim x2 + lim 5x + lim 3 (Sum, Di↵.)
x! 1 x! 1 x! 1 x! 1 x! 1
 
4 2
= lim x 3 lim x · lim x + 5 lim x + lim 3 (C. Mult., Power)
x! 1 x! 1 x! 1 x! 1 x! 1
⇥ ⇤⇥ ⇤
= ( 1)4 3( 1) ( 1)2 + 5( 1) + 3 (Const., Id.)
= (1 + 3)(1 5 + 3)
= 4( 1)
= 4.

1 1
Ex. 4. Evaluate lim , if it exists.
x!0 x x2 +x
1 1 1 1
lim = lim Can’t use Quotient Law. . . Try simplifying.
x!0 x x2 + x x!0 x x(x + 1)
(x + 1) 1
= lim
x!0 x(x + 1)
x
= lim
x!0 x(x + 1)

1
= lim Why is this equality justified?
x!0 x + 1
lim 1
x!0
=
lim x + lim 1
x!0 x!0
1
=
0+1
= 1.
 (a + h)3 a3
Ex. 5. Evaluate lim , if it exists.
h!0 h
(a + h)3 a3 (a3 + 3a2 h + 3ah2 + h3 ) a3
lim = lim
h!0 h h!0 h
3a h + 3ah + h3
2 2
= lim
h!0 h
= lim 3a + 3ah + h2
2
h!0
2
= 3a . Are we done?

t2 + 6t + 9
Ex. 6. Find lim .
t! 3 t2 + 3t
t2 + 6t + 9 (t + 3)2
lim = lim
t! 3 t2 + 3t t! 3 t(t + 3)

1 (t + 3)2
= lim
t! 3 t t + 3
1
= lim · lim (t + 3)
t! 3 t t! 3
1
= ·0
3
= 0.

Direct Substitution: Special case (Polynomial and Rational Functions)

It is often the case that lim f (x) = f (a).

x!a

• This is not always true.

• But it is true for polynomial and rational functions.
• We will see in Section 2.4 it is true for a larger class of functions (called continuous functions).

Theorem (Direct Substitution for Polynomial and Rational Functions).

Let p(x) and q(x) be polynomial functions. Let a be a real number. Then
(Polynomials) lim p(x) = p(a) and
x!a

p(x) p(a)
(Rational Functions) lim = if q(a) 6= 0.
x!a q(x) q(a)

A proof of this Theorem can be found in the textbook.

Definition: A limit as x approaches a is called an indeterminate form if substituting x = a

results in a undefined expression.
 • The examples of indeterminate forms we encountered in Exercises 4, 5, and 6 demonstrate
that such limits can sometimes be evaluated by tinkering around with algebra until a Limit
Law can be applied.
0
• When substituting x = a yields the meaningless expression , we say the indeterminate form
0
is of type 0/0.
More examples of indeterminate forms of type 0/0:
2
lim x =0 lim x 5 =1 lim x3 =1
x!0 x x!0 x 5 x!0 x

1
• Other types of indeterminate forms include the meaningless expressions , 0 ⇥ 1, 1 1,
0 1 0
1
0 , 1 , and 1 .

Additional techniques for finding limits

We have already seen (in Exercises 4, 5, and 6) that, before the Limit Laws can be applied, a
certain amount of algebraic “tinkering” must be done first.
We now introduce additional techniques. The first involves the conjugate of a sum or di↵erence.

Recall:
• The conjugate of a sum (+) or di↵erence ( ) w ± A is w ⌥ A.

That is, the conjugate of w + A is w A, and the conjugate of w A is w + A.

• To rationalize the denominator of a fraction

N
w±A
means to multiply it by the following “special form of 1”:
w⌥A
w⌥A

• To rationalize the numerator of a fraction

w±A
D
means to multiply it by
w⌥A
.
w⌥A
 p
x+2 1
Ex. 7. Evaluate lim , if it exists.
x! 1 x+1

Solution:
p p p
x+2 1 x+2 1 x + 2 +1
lim = lim ·p
x! 1 x+1 x! 1 x+1 x + 2 +1
x+2 1
= lim p
x! 1 (x + 1) x + 2 + 1
x⇠
⇠ +⇠
1
= lim ⇠ p
x! 1 (x⇠+⇠1)
⇠ x+2+1
1
= lim p Can we apply the Limit Laws now?
x! 1 x+2+1
1
= .
2

The trick of “multiplying by a special form of 1” can also be useful for evaluating limits of complex
fractions (fractions that contain fractions).

1 1
x+1 2
Ex. 8. Evaluate lim , if it exists.
x!1 x 1
Hint: As written, the limit is an indeterminate form of type 0/0. Rewrite it by multiplying by the
complex fraction by the following special form of 1:
2(x + 1)
2(x + 1)
Do you see where the expression 2(x + 1) came from, and why the hint will result in simplifying
the numerator of the complex fraction?
 The Limit Laws for one-sided limits

In the fine print of the Limit Laws, we required that:

f (x) and g(x) are functions defined for all x 6= a over some open interval containing a.
The Limit Laws can be applied to one-sided limits by changing the condition to:
defined for all x in some open interval of the form (a, b)
(for righthand limits) or
defined for all x in some open interval of the form (b, a)
(for lefthand limits).

Ex. 9. Two limits are given below.

p p
lim x 3 lim+ x 3
x!3 x!3
p
(a) Sketch the graph of the function f (x) = x 3.
(b) Which of the one-sided limits exists?
(c) Evaluate the limit(s) for which the Limit Laws can be applied.
Before we move on to the next topic, we’ll do one more exercise that requires the use of one-sided
limits.

Ex. 10. Evaluate lim |x|, if it exists.

x!0

Solution:
Recall that ⇢
x if x 0
|x| =
x if x < 0

Sidebar : Is it always true that x < 0? No! Not if x is negative. . .

Resuming the problem,

lim |x| = lim+ x = 0 since |x| = x whenever x > 0.

x!0+ x!0

lim |x| = lim x=0 since |x| = x whenever x < 0.

x!0 x!0

Therefore, by the Theorem given in Section 2.2,

lim |x| = 0
x!0

since
lim |x| = 0 = lim+ |x|.
x!0 x!0

The meaning of “if, and only if”

In mathematics, what is the meaning of the phrase “if, and only if ”?

P : You did poorly in school.
Q: You end up serving in the military.

“P =) Q” means “If you did poorly in school, then you end up serving in the military.”
—Al Gore during the 2000 presidential campaign

“P (= Q” means “If you end up serving in the military, then you did poorly in school.”
—A misinterpretation of Gore’s statement by his critics

“P () Q” means P =) Q and P (= Q.
We pronounce the symbol “()” as “if, and only if.”
Two of these statements are indefensible insults to members of the Armed Services. One is a
warning to students, and not necessarily an insult.
All three of these statements have di↵erent meanings.
 Squeeze Theorem
Another useful property of limits is as follows.

Squeeze Theorem. Suppose 0.010

f (x)  g(x)  h(x) 0.005

f (x) = x 2
for all x 6= a in an open interval containing a. �� g(x) = cos(1/x)
-0.10 -0.05 0.05 0.10
If h(x) = -x 2
lim f (x) = L = lim h(x), -0.005
x!a x!a

then -0.010
lim g(x) = L.
x!a

We can use the Squeeze Theorem to evaluate limits when other methods fail.

In the next exercise, we’ll apply the Squeeze Theorem to find the important limit
sin(x)
lim ,
x!0 x
which cannot be found by applying the Limit Laws.

First, recall:
The region bounded by a circle and one of its central angles is called a sector of the circle.
A circle and its sector have areas proportional to their angle measures:
Area of sector Angle measure of sector
= . (?)
Area of circle Angle measure of circle

⇡r2 /4 1 ⇡/2
= =
⇡r2 4 2⇡

sin x
Ex. 11. Find lim .
x!0 x
sin x
First let’s take a guess. When x ⇡ 0, we see by graphing that sin x ⇡ x, so x
⇡ 1—at
least for x near 0.
Our guess is, the limit is 1. . . but this is just a guess. We need a proof to be convinced.
 sin(x)
We can’t use the Quotient Rule to evaluate the limit lim . (Why not?)
x!0 x
We’ll make a geometric argument.

Let x be the radian measure of a sector of the unit circle in the first
quadrant. (So 0 < x < ⇡2 .)

Label O = (0, 0), A = (cos x, sin x), and C = (1, 0).

B

1
A

For no apparent reason, we construct a right triangle OBC that

contains the sector.
O x C
D
We’ll also drop a perpendicular AD to the x-axis from A.

Claims to be proven:
x sin x tan x
(1) Area(sector OAC) = (2) Area( OAC) = (3) Area( OBC) =
2 2 2

Proofs of claims:
B

1
A

(1) x x
Area(sector OAC) = · ⇡(1)2 = by the above formula (?).
O x C
2⇡ 2
D

1
A

1 1
Area( OAC) = (1)(AD) = sin x
(2) 2 2
O x C because A = (cos x, sin x).
D

1
A

opp. BC
tan x = = = BC.
adj. 1
(3)
O x C 1 tan x
Area( OBC) = (1)(BC) = .
D
2 2
 The figures on the previous page show that

Area( OAC) < Area(sector OAC) < Area( OBC).

Let’s tinker with this until we get something we can use. . .

sin x x tan x sin x
< < =
2 2 2 2 cos x
2 cos x 2 2
< <
sin x x sin x
. . . yields:
sin x
cos x < < 1
x

sin x
By the Squeeze Theorem, lim+ = 1.
x!0 x

sin x
A similar argument shows that lim = 1, so
x!0 x
sin x
lim =1.
x!0 x

sin(23x)
Ex. 12. Find lim .
x!0 x
Solution:
✓ ◆
sin(23x) 1 sin(23x)
lim = lim 23 · ·
x!0 x x!0 23 x
sin(23x)
= 23 · lim
x!0 23x
= 23 · 1
= 23.

sin 4x sin 4x sin 4x x

Ex. 13. Find lim . Hint: = .
x!0 sin 6x sin 6x x sin 6x
Solution:
sin 4x sin 4x x
lim = lim Any ideas on what might help now?
x!0 sin 6x x!0 x sin 6x
4 sin 4x 6x
= lim
x!0 4x 6 sin 6x
4 sin 4x 6x
= lim · lim
x!0 4x x!0 6 sin 6x
✓ ◆ ✓ ◆
sin 4x 1 6x
= 4 lim · lim
x!0 4x 6 x!0 sin 6x
✓ ◆ ✓ ◆
sin 4x 1 1
= 4 lim · lim
x!0 4x 6 x!0 sin6x6x
1
= 4(1) · (1)
6
2
= .
3

Additional exercises

Ex. 14. Use Direct Substitution to find the value of the limit.

(a) (§2.3—#87) (b) (§2.3—#89) (c) (§2.3—#91) (d) (§2.3—#92)

lim x2 1 2 7x lim ln(e3x )
x!7 lim lim x!3
x!0 1 + sin(x) x!1 x 6
Ex. 15. Find the value of the limit. (Hint: See Example 2.19 in the textbook.)
1 1
x 3
lim
x!0 x 3

Ex. 16. Find the value of the limit. (Hint: See Example 2.20 in the textbook.)
1 3
lim
x!0 x x(x + 3)

Ex. 17. Find the value of the limit. (Hint: See Example 2.23 in the textbook.)

x2 + 5x + 14
lim
x! 2 x2 + x 2
Ex. 18 (§2.3—#107–114). Given that lim f (x) = 4, lim g(x) = 9, and lim h(x) = 6, find the
x!6 x!6 x!6
limits that exist. If the limit does not exist, explain why.
3 ⇥ ⇤
(a) lim 2f (x)g(x) h(x) (e) lim (x + 1) · f (x)
x!3 (c) lim x!3
x!3 2
⇥ ⇤ p
(b) lim f (x) + 13 g(x) (d) lim g(x) f (x) (f) lim f (x) · g(x) h(x)
x!3 x!3 x!3

Ex. 19. Use Direct Substitution to show that the limit leads to the indeterminate form 0/0. Then
evaluate the limit.
x2 16 sin(✓)
(a) (§2.3—#93) lim (e) (§2.3—#99) lim
t!4 x 4 ✓!⇡ tan(✓)

3x 18 x3 1
(b) (§2.3—#95) lim (f) (§2.3—#100) lim
t!4 2x 12 x!1 x2 1
(h + 1)2 1 2x2 + 3x 2
(c) (§2.3—#96) lim (g) (§2.3—#101) lim
h!0 h t!4 2x 1
p
t 9 x+4 1
(d) (§2.3—#97) lim p (h) (§2.3—#102) lim
t!9 t 3 x! 3 x+3
Ex. 20. Evaluate the limit.
x2 4x x2 + 2x + 1 cos ✓ 1
(a) lim (b) lim (c) lim
x!4 x2 3x 4 x! 1 x4 1 ✓!0 sin ✓

Ex. 21 (Example 2.24). Use the Squeeze Theorem to show that lim x cos(x) = 0.
x!0

Ex. 22. Show that the function

⇢
0 if x is rational,
f (x) =
x2 if x is irrational

satisfies lim f (x) = 0.

x!0
 Workbook Lesson 4
§2.4, Continuity
Last revised: 2021-08-20 19:53

Objectives
• Explain the three conditions for continuity at a number.
• Describe three kinds of discontinuities.
• Define continuity on an interval.
• State the Composite Function Theorem.
• Apply the Intermediate Value Theorem.

Introduction to continuity
Recall that, in general, the limit of f (x) as x approaches a has nothing to do with the value of
f (x) when x reaches a. For example, in the graph below on the left, the limit as x ! 2 of f (x)
is 4. The output value at x = 2—that is, f (2)—isn’t even defined.

Out[! ]=

The graph on the right shows a modified version of f . Let’s call the modified version g. The only
di↵erence between f and g is, there’s no hole at x = 2 in the graph of g. The limit of g(x) as
x ! 2 is the same as for f (x). But this time, the output at 2,

g(2) = 4,

is defined. Moreover, g(2) is the number to which the values of g(x) are approaching as x ap-
proaches 2.
Another word for “unbroken” is continuous. We can express the di↵erence between the behavior
of f and g at x = 2 by saying that g is continuous at x = 2, and f is not.
When Calculus was first being developed, a distinction was drawn between graphs you can draw
without lifting your pencil—which were called continuous—and ones you can’t. Later, in the 1800s,
that rough idea of continuity (that is, the property of being continuous) was refined and defined
more precisely. We’ll use the modern, more precise definition of “continuous.”
At first glance, our definition doesn’t seem to say anything about the graph of f (x) being unbroken:
Definition. A function f is continuous at a number a if lim f (x) = f (a).
x!a
In fact, this definition describes only what happens near a single point on the graph. It says that,
as x approaches a, the value of f (x) doesn’t just approach a limit—it actually reaches the value of
the limit when x is equal to a. That is, as the input x approaches a, the output f (x) approaches
f (a)—and gets there.
We’ll see in the next subsection (see ?) how this definition relates to the idea that a graph has no
breaks in it.

But first, we need to unpack the meaning of the equation in the definition of continuity:

lim f (x) = f (a).

x!a

This equation comes up so often, we’ll give it a special name: “the Direct Substitution Property.”

What is required for this equation to be a meaningful and true statement?

• a is in the domain of f (that is, f (a) is a real number)
• lim f (x) exists
x!a

• lim f (x) = f (a) (and since f (a) is a real number, lim f (x) 6= ±1)
x!a x!a

+ To decide whether a function f is continuous at a number a, verify that all three of the
bulleted statements are true.

Ex. 1. The graph of a function f is given.

y=f (x)
7

0 x
0 1 2 3 4 5 6 7

(a) At what numbers a does lim f (x) not exist?

x!a

(b) At what numbers is f (x) not continuous?

(c) At what numbers a does lim f (x) exist but f is not continuous at a?
x!a
 Continuity on an interval
We can extend the idea of continuity at a single number x = a to continuity over an entire interval
of x-values. (?): If (but not only if!) you can draw a function’s graph over an entire interval without
lifting your pencil, then the function is continuous “on” (or “over”) that interval:

Definition. A function f is continuous on an open interval (a, b) if f is continuous at every x

in (a, b).

f (x) = 3-x

x
1 2 3 4
-1
p
f (x) = 3 x is continuous at each x-value such that 1 < x < 3

Notice that we specified that the interval has to be open (that is, endpoints are not included). To
define continuity on a closed interval (endpoints included), we need to define what it means to be
continuous from one side, say, at an endpoint. The definition we just gave won’t work, because if
a function’s output isn’t defined on the far side of an endpoint, then one of the one-sided limits at
the endpoint won’t exist.

Definition. f is continuous from the left at a if lim f (x) = f (a), and continuous from the
x!a
right at a if lim+ f (x) = f (a).
x!a

Definition. A function f is continuous on an closed interval [a, b] if f is continuous at every

x in (a, b), continuous from the right at a, and continuous from the left at b.

That is, “continuous on [a, b]” means “continuous on both sides everywhere in the interior (a, b)
of the interval, and continuous from one side at each endpoint.”

Finally, you are invited to consider the problem of defining “continuous everywhere.” We need to
clarify what we mean by “everywhere.” Do we mean “at every number on the number line,” or do
we mean “at every number in the domain of f ”? If the former, then f (x) = 1/x is not continuous
“everywhere.” If the latter, then f (x) = 1/x is continuous “everywhere.”
Our textbook does not define “continuous everywhere.” But sometimes, speaking casually, we may
say that a function “is continuous,” or “is continuous everywhere.” This is imprecise—technically,
in this class, we should always specify on what interval (or at what number) the function is
continuous. Be mindful that, according to our definitions, “continuous” always means “continuous
on an interval” or “continuous at a number x = a”.
 Discontinuities
Our word for “not continuous” is discontinuous.

Definition. If a is in the domain of f , but f is not continuous at a, we say f is discontinuous

at a (or f has a discontinuity at a).

+ Our book has a slightly di↵erent definition. For f to be discontinuous at a, they do not
require f (a) to be defined. During our exams, you can safely ignore this subtle distinction. We’ll
only give you problems in which this issue doesn’t come up.

Ex. 2. The Heaviside function (see Lesson 2 for its graph) has a discontinuity at 0. This type of
discontinuity is called a jump discontinuity.

x2 3x 4
Ex. 3. The function f (x) = has one discontinuity. What is it? This is an example of
x 4
a removable discontinuity.
Sketch the graph:

⇢
1/x2 if x 6= 0
Ex. 4. The function f (x) = has one discontinuity. What is it? This is an
1 if x = 0
example of an infinite discontinuity.
Sketch the graph:

⇢
1 if x is rational
Ex. 5 (Challenge). The function I(x) = is discontinuous at every real
0 if x is irrational
number x. Can you explain why?
Definition: If f (x) is discontinuous at a, then
• f has a removable discontinuity at a if lim f (x) exists.
x!a
(Note: When we state that lim f (x) exists, we mean that lim f (x) = L for some real number L.)
x!a x!a

• f has a jump discontinuity at a if lim f (x) and lim+ f (x) both exist, but lim f (x) 6=
x!a x!a x!a
lim+ f (x).
x!a
(Note: When we state that lim f (x) and lim f (x) both exist, we mean that both limits are
x!a x!a+
equal to real numbers.)
• f has an infinite discontinuity at a if lim f (x) = ±1 or lim+ f (x) = ±1 (or both).
x!a x!a

+ A demonstration of the concept of discontinuity is provided on iCollege. Don’t sweat the

technical details referenced in the applet! Your instructor may cover them during lecture.
(See applet on iCollege: “Formal meaning of discontinuity”)

+ Recall: if you can draw the graph of a function over an interval without lifting your pencil,
then the function is continuous on that interval.

Ex. 6. The graph of a function is given below

(a) From the graph, state the intervals on which the function is continuous.
(b) Classify each discontinuity as jump, removable, or infinite.
 1
Ex. 7. Sketch the graph of the function f (x) = and explain why f is discontinuous at the
x+2
number a = 2.

Theorems for evaluating limits

Theorem. If f is a rational function, polynomial, or trigonometric function, then f is continuous

at every number in its domain.

Corollary: Direct Substitution for polynomials, rational functions, and trigonometric

functions.
If
• f is a rational function, polynomial, or trigonometric function, and
• a is in the domain of f ,
then
lim f (x) = f (a).
x!a

(Recall:) This equation is called the Direct Substitution Property.

Ex. 8. Find the limit, if it exists.

(a) lim cos(x) (b) lim tan(x) x2 1 x2 1

x!0 x!⇡/2 (c) lim (d) lim
x!3 2x 6 x!0 2x 6
Composite Function Theorem. If f is continuous at b and lim g(x) = b, then
x!a
⇣ ⌘
lim f (g(x)) = f lim g(x) = f (b).
x!a x!a

Corollary to the Composite Function Theorem. If g is continuous at a and f is continuous

at g(a), then f g is continuous at a.

Ex. 9. Use the Composite Function Theorem to prove its Corollary.

Theorem.
If f and g are continuous at x = a, then the following functions are continuous at x = a.

f g f ±g f ·g f /g (if g(a) 6= 0)

⇡
Ex. 10. Find lim h(x), where h(x) = x + sin x 2
.
x!⇡
Solution:
The function h is the combination of continuous functions using the operations in the previous
theorem, so h is continuous.
Since h is continuous, we can use the Direct Substution Property:
⇡
lim h(x) = h 2
=⇡ sin( ⇡2 ) = ⇡ 1
x!⇡
 Intermediate Value Theorem

The following is an example of an existence theorem. It asserts that there exists a number (call
it c) satisfying certain properties, but it does not tell us the value of c.

Intermediate Value Theorem (IVT). Suppose f is continuous on [a, b]. Suppose f (a) 6= f (b).
If z is any number between f (a) and f (b), then there exists a number c in the interval (a, b) such
that f (c) = z.

Ex. 11. Let f (x) = 1 x6 , and note that

f (0) = 1 and f (2) = 1 64 = 63.

Identify the numbers a, b, and z in the following statement.

By the Intermediate Value Theorem, f (c) = 0 for some number c between 0 and 2.
Recall: A zero (or root) of a function f is a value c such that f (c) = 0.

Ex. 12. Show there is a zero of the equation 4x3 6x2 + 3x 2 = 0 between 1 and 2.

• Since f is a polynomial, it is continuous, so the Intermediate Value Theorem can be used.

• Take a = 1, b = 2, and z = 0 for the Theorem.
• Check that f (a) 6= f (b), and that z is between f (a) and f (b):
f (1) = 1,
1 6= 12, and 0 is between 1 and 12 X
f (2) = 12.
• Now the Theorem guarantees there exists a number c in [1, 2] such that f (c) = z = 0.
y=f (x)

f (b)=12

z=0 x
1 c 2
f (a)=-1

Since f is continuous and z = 0 is between f (1) = 1 and f (2) = 12,

the IVT guarantees there is some number c between a = 1 and b = 2 such that f (c) = z = 0.
 Additional exercises: Limits and continuity

Ex. 13. Explain why each function is continuous or discontinuous.

(a) The temperature in your home as a function of time.
(b) The cost of an Uber ride as a function of the distance traveled.
(c) The electrical current supplied to a household appliance prior to, during, and after a blackout.

Ex. 14. Determine the value(s), if any, at which each function is discontinuous. Classify each
discontinuity as removable, jump, or infinite.
1 1 • k(x) = tan(2x)
• f (x) = p • h(t) = +1
x t
x 5 t+3
• g(x) = 2 • j(t) = t • `(t) =
x x e 2 t2 + 5t + 6
Ex. 15 (§2.4—#151). A particle moving along a line has a position function s(t), which is
continuous. Assume s(2) = 5 and s(5) = 2.
(a) Explain why there must be a value c such that 2 < c < 5 and s(c) = 4.
(b) Now suppose a second particle has a position function h(t) = s(t) t. Explain why there must
be a value d such that 2 < d < 5 and h(d) = 0.

Ex. 16. Let ⇢

3x2 4 if x  2,
h(x) =
x2 if x > 2.
Although h(0) < 10 and h(4) > 10, there is no value of x in the interval [0, 4] such that h(x) = 10.
Explain why this does not contradict the Intermediate Value Theorem.

Ex. 17. Define ⇢

x2 if x < 1,
f (x) =
x if x > 1.
Sketch the graph of f . Can you pick a value of k so that defining f (1) = k makes f (x) continuous
on the interval ( 1, 1)?
 Workbook Lesson 5
§2.5, The Epsilon-Delta Definition of a Limit
Last revised: 2021-06-15 07:33

Objectives
• Interpret an inequality of the form 0 < |x a| < c as a statement about the distance between x and a.
• Use a table of values to estimate the limit of a function or to identify when the limit does not exist. (Moved
from Lesson 2, §2.2)
• Describe the idea behind the epsilon-delta definition of a limit.
• Apply the epsilon-delta definition to find the limit of a function.
• Describe the epsilon-delta definitions of one-sided limits and infinite limits.
• Use the epsilon-delta definition to prove the limit laws.

Inequalities representing distance

The distance between two numbers w and a is |w a| 0.

Let c > 0. The inequality
|w a| < c
means that the distance |w a| between w and a is less than c. (We use the absolute value bars be-
cause the di↵erence w a might be negative, while distance is by definition never negative.)

Anchoring one end of a piece of string at the purple point on the number line below, pinch o↵ a
length of string—call the length c—and swing it around the purple point like a compass to see why
c is sometimes called the “radius” of the inequality.

Ex. 1. Where can x be on the number line if

0 < |x 7| < 2?

Answer in words, or by graphing on the number line.

Ex. 2. If |w a| = 0, what must be true about w and a?

Ex. 3. Graph the set of numbers w on the number line such |w 3|  1.

Guessing the limit of a function using a table of values

sin x
Ex. 4. Guess the value of lim using only a calculator.
x!0 x

Taking x closer and closer to 0, we find that:

sin(x)
x
x
±1. 0.841471
sin x
±0.5 0.958851 Guess: lim =1
±0.4 0.973546 x!0 x
±0.3 0.985067 sin x
(Notice that the function is undefined when x = 0.)
±0.2 0.993347 x
±0.1 0.998334 We showed in an earlier Lesson this guess is correct.
±0.05 0.999583
±0.01 0.999983
±0.005 0.999996

1
Ex. 5. Guess the value of lim sin .
x!0 x

Taking x closer and closer to 0, we find that:

1
x sin
x
± ⇡1 0
1
± 2⇡1
0 Guess: lim sin =0
x!0 x
1
± 3⇡ 0 This time our guess is wrong.
1
± 4⇡ 0 1
1
Can you explain why by looking at the graph of sin ?
± 5⇡ 0 x
1
± 10⇡ 0
1
± 100⇡ 0
 �(�) = sin(1/x)

1.0

0.5

�
-0.5 0.5

-0.5

-1.0

There’s something seriously wrong with our “informal” definition of a limit—it misleads us
1
into giving an incorrect answer. The limit of sin as x ! 0 does not exist.
x
In the next section of this Lesson, we will revise our informal definition of
lim f (x)
x!a
and give a precise definition that is reliable in all cases.

Formal definition of limit

Let’s get a more intimate understanding of the concept of a limit before we look at the true
definition.
(See applet on iCollege: “Epsilon-delta definition of limit”)
Here is the graph of a function. Let’s not worry about what the formula for this function is.
y = f(x)

2.5

2.0

1.5

f(a) 1.0

0.5

x
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 7 8 9 10
a

-0.5
A point a, f (a) on the graph is marked.
For what x-values is the output f (x) near f (a)?
You might say, well, how near do you want it? (Set a = 5 in the applet.) Let’s say I want the
output to be within four tenths of f (a) = 1. (Set E = .4 in the applet.)
�

��������� � > �

������ � > �

���� ��������� (�)

y = f(x)
���� ������ (�)

2.5

2.0

1.5
a = 5
e f (a) = 1.
L D = 3
f(a) 1.0 E = 0.4
e

0.5

x
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 7 8 9 10
a

-0.5
How close must the input x be to a
so that the corresponding output f (x) is within E of L ?

Imagine an old-fashioned radio with a knob you turn to change the station. You don’t have to tune
the knob to exactly the right frequency. Within a certain tolerance will be close enough to make
the radio station come in clearly.
So how close to a do our x-values have to be to give us output values that are all within the
tolerance shown? Is it enough to be within 3 units? (Set D = 3 in the applet.)
�

��������� � > �

������ � > �

���� ��������� (�)

y = f(x)
���� ������ (�)

2.5

2.0

1.5
a = 5
e f (a) = 1.
L D = 3
f(a) 1.0 E = 0.4
e

0.5

x
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 7 8 9 10
a

d d
-0.5
How close must the input x be to a
so that the corresponding output f (x) is within E of L ?

What about within 1 unit? (Set D = 1 in the applet.)

But 2 is no good—there are x values that give us outputs that aren’t within the tolerance. (Set
D = 2 in the applet to see, then go back to D = 1.)
Let’s give this margin of error along the x-axis a name—we’ll call D = 1 the margin. Clearly, any
smaller number will serve as a suitable margin, too.
The point is, all the inputs close enough to 5—that is, within the margin of 5—give us outputs
that are within the tolerance of four tenths.
Write:

The distance between the output f (x) and L is less than the tolerance E (?)
whenever the distance between the input x 6= a and a = 5 is less than the margin D.

That’s a lot of writing. Let’s rewrite this fact using symbols.

f (x) L <E whenever 0 < |x a| < D. (?)

��������� � > �

������ � > �

���� ��������� (�)

y = f(x)
���� ������ (�)

2.5

2.0

1.5
a = 5
e f (a) = 1.
L D = 1
f(a) 1.0 E = 0.4
e

0.5

x
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 7 8 9 10
a

d d
-0.5
How close must the input x be to a
so that the corresponding output f (x) is within E of L ?

But now suppose I want the output to be closer than 0.4 away. Suppose I change the tolerance to
some smaller number—say 0.1. (Set E = 0.1 in the applet.)
�

��������� � > �

���

������ � > �

���� ��������� (�)

y = f(x)
���� ������ (�)

2.5

2.0

1.5
a = 5
f (a) = 1.
L D = 1
e
f(a) 1.0 e E = 0.1

0.5

x
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 7 8 9 10
a

d d
-0.5
How close must the input x be to a
so that the corresponding output f (x) is within E of L ?

Now this fact (?) that I wrote (in both English and symbols) isn’t true anymore.
This statement guarantees that every x within D of a = 5 gives us an output that’s within the
tolerance of 1. But there are x values within the margin of 5 that yield outputs that are more than
0.1 away from the desired value, 1.
Well, is there some D I could choose to make this fact true again? Is there some D so that the
portion of the graph within the blue stripe, lies entirely within the red stripe?
Tinker with the slider and try to find a D that works. . .
Now let me ask you this. Is it the case that, no matter what E I pick, I can always pick a margin
D so small that the fact on the board holds true?
That is, can I always make the margin stripe so small that the portion of the graph it contains is
entirely contained in the yellow stripe—no matter how narrow I make the yellow stripe?
Yes. And this is the idea of a limit.

Formal definition. The statement

lim f (x) = L
x!a

(in words: “the limit of f (x) as x approaches a is L”) means that, given any tolerance E > 0,
there exists some margin D > 0 such that

f (x) L <E

whenever
0 < |x a| < D.
+ We don’t care what happens when x = a.

Note: Most authors use the Greek letters and " in the above definition rather than the Roman
letters D and E. (Of course, the names of variables don’t matter in mathematics!)

Ex. 7. Use the graph provided below to complete the statement:

|f (x) 2| < whenever <x< .

2.5

2.0
output f (x)

1.5

1.0

0.5

0.0
0 1 2 3 4 5 6 7
input x
Ex. 6. The graph of a function f is given. State whether or not each quantity exists. If it does
not exist, explain why not.

-4 -3 -2 -1 1 2 3 4 5 6

(a) lim f (x) (e) lim f (x) (i) lim f (x)

x! 3 x!0 x!2
(b) lim f (x) (f) lim+ f (x) (j) f (2)
x! 3+ x!0
(c) lim f (x) (g) lim f (x) (k) lim f (x)
x! 3 x!0 x!5

(d) f ( 3) (h) f (0) (l) lim+ f (x)

x!5
 Formal definitions of infinite limits

Let us revisit our definitions of infinite limits in order to make them more precise.

Formal definition (Infinite limits). The statement

lim f (x) = 1
x!a

means that for any number M , there is a D > 0 such that

f (x) > M

whenever x satisfies
0 < |x a| < D.

Similarly,
lim f (x) = 1
x!a
means that for any number M , there is a D > 0 such that

f (x) < M

whenever x satisfies
0 < |x a| < D.

The statements

lim f (x) = 1, lim f (x) = 1, lim f (x) = 1, lim f (x) = 1

x!a+ x!a+ x!a x!a

have similar definitions.

2x 2x 2x
Ex. 8. Find lim+ and lim . Then find lim , if it exists.
x!3 x 3 x!3 x 3 x!3 x 3
Solution:
As x ! 3 from the right, x 3 ! 0 through positive values, and 2x ! 6 > 0. So
2x 1
lim+ 2x ·
= lim+ |{z} = 1.
x!3 x 3 x!3 x 3
!6 | {z }
!1

(This is an intuitive explanation. To evaluate this limit without appealing to intuition, we

could write a formal proof—but you are not expected to do so.)
As x ! 3 from the left, x 3 ! 0 through negative values, and 2x ! 6 > 0. So
2x 1
lim 2x ·
= lim |{z} = 1.
x!3 x 3 x!3 x 3
!6 | {z }
! 1

2x 2x
Since lim+ 6= lim , the limit
x!3 x 3 x!3 x 3
2x
lim
x!3 x 3
does not exist.

1
Ex. 9. Find lim .
x!2 (x 2)4
Solution:
Solution:
1
As x ! 2 from the right, (x 2)4 ! 0 through positive values. That is, > 0 for
(x 2)4
x > 2 near 2. Thus
1
lim+ = 1.
x!2 (x 2)4

1
As x ! 2 from the left, (x 2)4 ! 0 through positive values. That is, > 0 for
(x 2)4
x < 2 near 2. Thus
1
lim = 1.
x!2 (x 2)4

Since the one-sided limits as x ! 2± are both 1, we have

1
lim = 1.
x!2 (x 2)4
 Epsilon-delta proofs (optional)

The formal definition of a limit can be thought of as a challenge. If for example

lim f (x) = 1,
x!0
this means:
Given any E > 0, you must be able to find a number D > 0
such that |f (x) 1| < E for any x such that 0 < |x 0| < D.

An epsilon-delta proof is an argument that proves it’s always possible to meet the challenge no
matter what E > 0 is.

Ex. 10. Let f (x) = 4x 5. Let E = 1. Find D > 0 such that |f (x) 7| < E whenever
0 < |x 3| < D.
Solution:
To find d, begin with the inequality |f (x) 7| < E = 1.

|f (x) 7| < 1
|(4x 5) 7| < 1
|4x 12| < 1
|4 · (x 3)| < 1
4|x 3| < 1 (|A · B| = |A| · |B|)
|x 3| < 14

1
Therefore, |f (x) 7| < 1 whenever 0 < |x 3| < 4
.
(Notice that, in this example, x = 3 makes |f (x) 7| < 1, so the ‘0 <’ isn’t used.)

Can we conclude from the work above that lim f (x) = 7?

x!3

No. . . this has to work for any E = 1, not only the single particular choice E = 1.
Ex. 11. Let f (x) = 1 2x. Show that lim f (x) = 7.
x!4

Solution:
The definition of a limit requires that we can find a D > 0 such that, for any given E > 0,
the following is true whenever 0 < |x 4| < D.

|f (x) ( 7)| < E

|8 2x| < E
2|x 4| < E
E
|x 4| <
2

Our work proves the following statement: for any E > 0, we have

|f (x) ( 7)| < E

whenever
E
0 < |x 4| < D = .
2

That is, we have proven that

lim f (x) = 7.
x!4
 Workbook Lesson 6
§3.1, Definition of the Derivative
Last revised: 2021-06-17 13:27

Objectives
• Recognize the meaning of the tangent to a curve at a point.
• Calculate the slope of a tangent line.
• Find an equation for the tangent line to the graph of a function f at the point a, f (a) .
• Identify the derivative as the limit of a di↵erence quotient.
• Calculate the derivative of a given function at a point.
• Describe the velocity as a rate of change.
• Explain the di↵erence between average velocity and instantaneous velocity.

Definition of the derivative

Recall:
A secant line to the graph of a function f is a line through two points a, f (a) and b, f (b)
on the graph.
Its slope
f (b) f (a)
(†)
b a
is the average rate of change in f from x = a to x = b.

Definition: The expression (†) is also known as a di↵erence quotient.

f(x)=100 - x 6 f(x)=100 - x 6 f(x)=100 - x 6

110 110 110

100 100 100

90 90 90

80 80 80
��
70 70 70

60 60 60

50 50 50

40 40 40
x x x
0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0

Ex. 1. Find the average rate of change in the function f (x) = 100 x6 from a = 0 to b.
(i) b = 2 (ii) b = 1 (iii) b = 0.1

Solution:
[100 26 ] [100 06 ]
(i) = 25 = 16.
2 0
[100 16 ] [100]
(ii) = 1.
1 0
[100 (0.1)6 ] [100]
(iii) = 0.000001.
1 0
The derivative at x = a is the limit of the slopes of secant lines to the graph through the points
P = a, f (a) and Q = b, f (b) , taking the limit as b approaches a.
f(x)=100 - x 6
110

100

90

80

70

60

50

40
x
0.0 0.5 1.0 1.5 2.0

Definition: Let f be a function. The derivative of f at x = a is

f (b) f (a)
def
f 0 (a) = lim . (*)
b!a b a
Since the names of variables don’t make any di↵erence, we can also write this as
deff (x) f (a)
f 0 (a) = lim . (*)
x!a x a
If this limit exists, we say f is di↵erentiable at a.

If we set x = a + h, we get an equivalent definition of the derivative of f at x = a:

Definition: Let f be a function. The derivative of f at x = a is

def f (a + h) f (a)
f 0 (a) = lim . (**)
h!0 h
dy
An alternate notation for the derivative, called Leibniz notation, is . Taking
dx
y = f (x) f (a),
x = x a = h,
we get a third way to write the definition of the derivative.
dy notation y
= lim (***)
dx x!0 x

Visually, the secant lines through P and nearby points Q = (x, f (x)) “limit” as Q ! P to a line
that “kisses” the graph of f at P , meeting the graph at P and at no other nearby points. We call
this line the tangent line at P to the graph of f :
Definition: If f 0 (a) exists, the tangent line to the curve y = f (x) at a point P = (a, f (a))
is defined to be the line with slope f 0 (a) through the point P .
+ A very old and very imprecise definition of a tangent line is:
a straight line which touches a curve, but does not cut it.
Here, “tangent” is opposed to a line which “cuts” the curve, as opposed to ‘kissing” it. (Compare
the definition of a continuous function as “a function whose graph can be drawn without lifting
one’s pencil”. . . )

Ex. 2. Find the derivative of f (x) = x2 at x = 5, using the definition of the derivative.
Solution:
We substitute a = 5 into definition (**) of the derivative of f at a, which we’ve copied here:
f (a + h) f (a)
f 0 (a) = lim (**)
h!0 h
We get:

f (5 + h) f (5) (5 + h)2 52
f 0 (5) = lim = lim
h!0 h h!0 h
25 + 10h + h2 25
= lim
h!0 h
h(10 + h)
= lim
h!0 h
= lim (10 + h)
h!0

= 10.

Ex. 3. Find an equation for the tangent line to the graph of f (x) = x2 at x = 5.
Solution:
The slope of the tangent line to f (x) at x = 5 is f 0 (a), which we found in the previous
exercise.
We now use the point-slope form of a linear equation,

y y0 = m(x x0 ),

substituting the point 5, f (5) = (5, 25) for (x0 , y0 ).

Both the equation
y 25 = 10(x 5)
and its simplified form
y = 10x 25
are acceptable answers.
Ex. 4.
(a) Find the derivative of f (x) = 3x 1 at x = a, using the definition of the derivative.
(b) Write an equation of a tangent line to the graph of f (x) at any point without doing any
additional scratchwork.

+ Notice that, for a linear function f (x) = mx+b, the slope of the tangent line and the slope of
all secant lines coincide. The derivative of a linear function is constant, f 0 (x) = m for all x.

Ex. 5. Find the tangent line to the curve y = 3/x at the point (3, 1).
Solution:
Check: The graph of f —that is, the set containing all points of the form x, f (x) —contains
the point a, f (a) = (3, 1). X
3 3 (3+h)
0 f (3 + h) f (3) 3+h
1 3+h
f (3) = lim = lim = lim
h!0 h h!0 h h!0 h
h 1 1
= lim = lim = .
h!0 h(3 + h) h!0 3 + h 3

1
We need the equation of a line with slope f 0 (3) = through the point (3, 1).
3
y y1 = f 0 (a) · (x a)
y 1 = f 0 (3) · (x 3)
1
y= 3
(x 3) + 1
3y = (x 3) + 3
x + 3y 6 = 0
 Determining di↵erentiability from a graph, and the formula for the tangent line

A function f is di↵erentiable at x = a (that is, f 0 (a) exists) if its graph looks like a non-vertical
straight line when we zoom in sufficiently around the point (a, f (a)).
That is, the tangent line at x = a is a close approximation of the graph of f at points x, f (x)
for x near a.
(See applet on iCollege: “The differentiation microscope”)

Counterexample. The function f (x) = |x| is not di↵erentiable at x = 0.

y = |x|
1.0
0.8
0.6
0.4
0.2
x
- 1.0 - 0.5 0.0 0.5 1.0

Intuitively : no matter how much we zoom in on the graph of f near (0, 0), the graph will
always have a “corner” (or “cusp”) in it.
Therefore, no line is tangent to the graph of f at the point (0, 0).
p
Counterexample. The function f (x) = 3
x is not di↵erentiable at x = 0.
y = x1/3
1.0

0.5

x
- 1.0 - 0.5 0.5 1.0

- 0.5

- 1.0

The graph ofpy = x1/3 can be obtained by reflecting the graph of y = x3 in the line y = x
because y = 3 x and y = x3 are inverse functions.
As x approaches 0, the tangent lines to the curve at x become steeper and steeper.
When x equals 0, the tangent line is vertical.
That is, the slope of the tangent line, f 0 (x), approaches infinity as x ! 0: in symbols,
lim f 0 (x) = 1.
x!0

Formula: The tangent line to the graph of y = f (x) at the point a, f (a) is

y = f 0 (a) · (x a) + f (a)
| {z }
slope

provided that
• a is in the domain of f (so that f (a) is defined),
• f is di↵erentiable at a (so that f 0 (a) is defined),
• lim f 0 (x) 6= ±1 (that is, the tangent line cannot be vertical).
x!a
 Finding a linear extension of a function that is di↵erentiable and continuous

Suppose we are given a function that is defined only for x  0. . . say, f (x) = ex for x  0.
�

-� -� � �

We wish to extend the function so that its graph is a line for x > 0 in such a way that the resulting
piecewise function is continuous (e.g., no gaps) and di↵erentiable (e.g., no sharp corners):
�

-� -� � �

How should we pick the y-intercept and the slope of the line? That is, what should a and b be in
order to make the piecewise function
⇢ x
e for x  0,
f (x) =
ax + b for x > 0
continuous and di↵erentiable?
• We can imagine the following real-world situation in which the above problem might need
to be solved: Let x = 0 represent the present moment, so that x < 0 represents the past.
Suppose a virus has been spreading faster and faster (exponentially) until now. We would
like to predict the number of infected, f (x), under the assumption that the rate of spread in
the future (time x > 0) stays the same as it is at the present moment.
• A mathematically similar problem is encountered in graphic and industrial design, economics,
and robotics: a curve is to be joined with a straight line at a point (called the break point)
in such a way that the result is perfectly smooth, with no corners or gaps. In this situation,
we might replace the exponential function ex by some other type of function—for example, a
polynomial. (Polynomials are often used in the design of fabricated objects and the modeling
of economic phenomena.)

+ Use the applet “ When is a piecewise function differentiable?” on iCollege to

experiment with di↵erent values of a and b and di↵erent functions defined for x  0.
 The derivative as a rate of change
The definitions of average velocity and instantaneous velocity were given in Section 1.1.
Here we restate those definitions and introduce some new shorthand. (New shorthand in blue.)
Definition. Let s(t) be a function that gives the position (or displacement) of a particle in
motion at time t.
• The average velocity of a particle in motion from time t = a to time t = b is the
average rate of change in its position function from t = a to t = b.
Let us denote the average velocity by vave for short, as follows:
s(b) s(a)
vave =
b a

• The instantaneous velocity of a particle in motion at time t = a is the instantaneous

rate of change in its position function at t = a.
Thus the instantaneous velocity at t = a is (by the definition of instantaneous rate of
change in Lesson 1, and by the definition (*) of the derivative on p. 2 of this document):
s(b) s(a)
s0 (a) = lim .
b!a b a
We write:
v(a) = s0 (a)
—that is, v(a) = s0 (a) is the (instantaneous) velocity at time a.
Taking b = a+h, so that h = b a ! 0 as b ! a, we get a more useful formula for v(a)
that should be used when working problems that ask for the instantaneous velocity:
s(b) s(a) s(a + h) s(a)
v(a) = s0 (a) = lim = lim . (?)
b!a b a h!0 h

Ex. 6. A stone is tossed into the air from ground level with an initial velocity of 15 m/sec. Its
height in meters after t seconds is s(t) = 15t 4.9t2 . Find the instantaneous velocity of the stone
at t = 1 sec.
Ex. 7. A co↵ee shop determines that the daily profit on scones, P (s), obtained by charging s dollars
per scone is modeled by the following equation:

P (s) = 20s2 + 140s 240

The co↵ee shop currently charges $3 per scone.

(a) Find P 0 (3), the rate of change in profit when the price is $3.
(b) Should the co↵ee shop consider raising or lowering its prices on scones?
Solution to part (a).

P (s) P (3)
P 0 (3.25) = lim
s!3 s 3
⇥ ⇤ ⇥ ⇤
20s2 + 140s 240 20(3)2 + 140(3) 240
= lim
s!3 s 3
2
20s + 140s 240
= lim
s!3 s 3
2
20(s 7s + 12)
= lim
s!3 s 3
20(s 4)(s 3)
= lim
s!3 s 3
= lim 20s + 80
s!3

= 20

Additional exercises

Ex. 8 (§3.1—#1, 5, 7, 8). Find the slope of the secant line between the values x1 and x2 for
each function.
p
• f (x) = 4x + 7, x1 = 2, x2 = 5 • h(x) = x 9, x1 = 10, x2 = 13
p 4
• g(x) = x, x1 = 1, x2 = 16 • j(x) = , x1 = 1, x2 = 3
3x 1
Ex. 9 (§3.1—#11).
(a) Find the slope of the tangent line to the parabola y = x2 + x at the point (1, 2).
(b) Find an equation for the tangent line to the parabola at the point (1, 2).

Ex. 10 (§3.1—#15).
7
(a) Find the slope f 0 (a) of the tangent line to the curve y = at the point a, f (a) for a = 3.
x
(b) Find an equation for the tangent line to the curve at x = 3.

Ex. 11. The graph of a function f is given. Arrange the following numbers from least to greatest.

0 f 0 ( 2) f 0 (0) f 0 (2) f 0 (4)

-1 1 2 3 4
Ex. 12 (§3.1—#21, 23, 25, 27). Find f 0 (a) using the definition of the derivative.
p
(a) f (x) = 5x + 4, a = 1 (c) f (x) = x, a = 4
1
(b) f (x) = x2 + 9x, a = 3 (d) f (x) = , a = 2
x
 Workbook Lesson 7
§3.2, The Derivative as a Function
Last revised: 2021-05-03 08:53

Objectives
• Define the derivative function of a given function.
• Graph a derivative function from the graph of a given function.
• State the connection between derivatives and continuity.
• Describe three conditions for when a function does not have a derivative.
• Explain the meaning of a higher-order derivative.

The derivative function

Let f (x) be a function. A second function, called the derivative of f , is defined by

def f (x + h) f (x)
f 0 (x) = lim .
h!0 h
This equation is just Equation (**) from Lesson 6 with the name of one variable changed.
+ The domain of the function f 0 —that is, the set of input values x for which f 0 (x) is defined—is
the set of x-values for which the above limit exists and is a real number.

Notations for the derivative:

• Here we list all the most common notations used to denote the derivative of y = f (x):
dy d⇥ ⇤
f 0 (x) f (x) Df (x) y0
dx dx

• In general, the notation

x=a
after an expression means “evaluate the expression by substituting x = a.” So, for instance,
the notation
dy
dx x=a
means the same thing as
f 0 (a).

Fail conditions for di↵erentiability:

Recall (from the previous lesson) that a function f fails to be di↵erentiable at x = a (that is,
f 0 (a) does not exist) if. . .
• the graph of f contains a corner or
• the tangent line at x = a is vertical.
 The variable of di↵erentiation (“di↵erentiating with respect to”)

Recall:
• When we discuss a function y = f (x), the input variable x is called the independent
variable.
• We call y the dependent variable because the value of y = f (x) depends on x.
• For another example with di↵erent variable names, if r is a function of t, and we write
r = g(t), then t is the independent variable and r is the dependent variable.

dy
We say that = f 0 (x) is the derivative of y = f (x) with respect to x, the independent variable.
dx
dy
• = f 0 (x) is the instantaneous rate of change in y = f (x) as x varies.
dx
dr
• = g 0 (t) is the instantaneous rate of change in r = g(t) as t varies.
dt
• The variable of di↵erentiation is the independent variable, with respect to which the
derivative is taken.

+ When unfamiliar variable names are used, you may find it helpful to identify which variable is
the dependent variable (what you take the derivative of ) and the independent variable (what you
take the derivative with respect to).

Graphing a derivative

Since f 0 is a function, we can graph it.

Ex. 1. Find the derivative of f (x) = x2 and sketch its graph.

 p
Ex. 2. Find the derivative of f (x) = x and sketch its graph.
Solution.

f (x + h) f (x)
f 0 (x) = lim
h!0 h
p p
x+h x
= lim
h!0 h
p p p p
x+h x x+h+ x
= lim ·p p
h!0 h x+h+ x
⇢x+h ⇢ x
= lim p p
h!0 h x + h + x
◆
h
= lim p p
h!0 ◆
h x+h+ x
1
= p .
2 x

1
f '(x) =
2 x
1.0

0.8

0.6

0.4

x
-1 1 2 3 4 5
 Di↵erentiability implies continuity

Theorem. If f is di↵erentiable at a, then f is continuous at a.

Proof.
We need to show that lim f (x) = f (a). We’ll do this by showing that lim f (x) f (a) = 0.
x!a x!a

We can make f (x) f (a) look like the slope of the secant line if we divide it by x a. We’ll
multiply by (x a) at the same time so that the value is not changed:
f (x) f (a)
f (x) f (a) = · (x a)
x a
Now we take the limit on both sides, and use the Limit Laws:

f (x) f (a)
lim [f (x) f (a)] = lim (x a)
x!a x!a x a

f (x) f (a) h i (†)
lim f (x) lim f (a) = lim [f (x) f (a)] = lim · lim x a = f 0 (a) · 0 = 0,
x!a x!a x!a x!a x a x!a
(*)
where the equality (†) is justified by the fact that, by definition of di↵erentiability of f at a,
f (a) f (x)
the limit f 0 (a) = lim exists.
x!a x a
Adding lim f (a) to both sides of (*) and then applying the Constant Law for Limits yields
x!a

lim f (x) = lim f (a) = f (a).

x!a x!a

Ex. 3.
The graph of the Heaviside function H(x) is shown below.
Is the Heaviside function di↵erentiable at x = 0? (That is, does H 0 (0) exist?)
Justify your answer.
�(�)

Answer: No.
If H 0 (0) did exist, then by the above Theorem, H would be continuous at x = 0.
Justification:
Since H is not continuous at x = 0, H 0 (0) must not exist.
 Higher derivatives

Recall:
If a function y = f (x) is di↵erentiable at every point of its domain, then f 0 (x) is a new
function with the same domain, called the (first) derivative of f .

This new function f 0 may (or may not) have a derivative of its own. If it does, the derivative of f 0
is denoted by

00 d dy d2 y d2 ⇥ ⇤
f = 2 f (x) D2 f or y 00 .
dx dx dx dx2

Definition: The derivative of the (first) derivative f 0 is called the second derivative of f .

Ex. 4. Find f 00 (x) if f (x) = x2 .

Solution:
We know f 0 (x) = 2x from Exercise 1.
f 0 (x + h) f (x)
f 00 (x) = lim
h!0 h
2(x + h) 2x
= lim
h!0 h
= 2.

The third derivative f 000 is the derivative of the second derivative f 00 .

Derivatives of higher order (for n 4) are written f (n) :
f 0, f 00 , f 000 , f (4) , f (5) , ...

Ex. 5. Find all higher-order derivatives f 00 , f 000 , f (4) , f (5) , . . . of f (x) = x2 .

Solution:
From Exercise 4, we know f 00 (x) = 2 for any value of the input x. Therefore,
f 00 (x + h) f 00 (x) 2 2
f 000 (x) = lim = lim = 0.
h!0 h h!0 h
Now
f 000 (x + h) f 000 (x)
f (4) (x) = lim =0
h!0 h
and we easily see that
f (n) (x) = 0 for n = 5, 6, 7, . . .
 Velocity and acceleration

The concepts of derivative and second derivative have natural interpretations in terms of mo-
tion.
Our first example will be the motion of a swinging pendulum. In later sections, we will see that
the same interpretation applies to a body undergoing one-dimensional “rectilinear” motion (that is,
motion in a straight line), such as a ball thrown straight up in the air.
Let x be time. Let f (x) be the displacement from an object’s initial position at time x = 0.
For the pendulum, define f (x) to be the positive or negative angle between the rest position
of the pendulum and the angle of the pendulum at time x.

(See applet on iCollege: “Pendulum”)

If f (x) = displacement at time x, then:

v(x) = f 0 (x) is velocity,
0 00
a(x) = v (x) = f (x) is acceleration, and
j(x) = a0 (x) = v 00 (x) = f 000 (x) is jerk (or lurch).

Let’s look at a displacement function that is simpler than that of the pendulum in the applet.
Ex. 6. Let x be time and suppose f (x) = sin(x) is the displacement function for a pendulum.
Identify the displacement, velocity and acceleration in the following graph.
1.0

0.5

2 4 6 8 10 12

-0.5

-1.0
Solution.
We recognize the solid blue graph as the graph of the function sin function.
At time t = ⇡/2, we have f (x) = displacement = 1 radian ⇡ 57 . At that instant, the
pendulum has stopped moving to the right, and is about to move to the left: its speed is 0.
Thus the dashed purple line is velocity f 0 (x).
By process of elimination, the dotted gold line must be acceleration. We verify that this
makes sense: at time t = ⇡/2, the pendulum is momentarily motionless, f 0 (⇡/2) = 0,
and accelerating in the leftward direction from its rightmost position, f (⇡/2) = 1. At that
instant, the acceleration is f 00 (⇡/2) = 1. The maximum acceleration in the rightward
direction occurs when the pendulum is motionless and in its leftmost position.

How should we interpret the acceleration at the instant t = ⇡, when the pendulum is in its rest
position f (⇡) = 0?

Hands-on demo of the relationship between the graphs of f , f 0 , and f 00

The applet “ Derivative sandbox,” provided on iCollege, illustrates the relationship between the
graph of a function f and the graph of its derivative f 0 . (The graph of f 00 can also be displayed.)
Interact with the applet by dragging the points indicated by crosshairs to change the shape of
the graph of f . Can you explain what the orange points on the graph of f represent, and how the
orange points relate to the graph of f 0 ?
 Additional exercises
Ex. 7 (§3.2—#54, 55, 58, 63). Find the derivative of each function using the definition of the
derivative. State the domain of the function and the domain of its derivative.

• f (x) = 2 3x • h(x) = 5x x2
1
• g(x) = 6 • j(x) = p
x
Ex. 8 (§3.2—#68, 73). The given limit is the derivative of a function y = f (x) at x = a. What
are f (x) and a?
(1 + h)2/3 1 eh 1
(a) lim (b) lim
h!0 h h!0 h

Ex. 10. The figure shows the graphs of f , f 0 , and f 00 . Identify each curve.
Ex. 11. The figure shows the graphs of f , f 0 , f 00 , and f 000 . Identify each curve.

Ex. 12. The graphs of a function f and its derivative f 0 are shown. Which is larger, f 0 ( 1) or
f 0 ( 1)?

-1 1 2 3 4
 Workbook Lesson 8
§3.3, Di↵erentiation Rules
Last revised: 2020-12-14 15:24

Objectives
• State the constant, constant multiple, and power rules.
• Apply the sum and di↵erence rules to combine derivatives.
• Use the product rule for finding the derivative of a product of functions.
• Use the quotient rule for finding the derivative of a quotient of functions.
• Extend the power rule to functions with negative exponents.
• Combine the di↵erentiation rules to find the derivative of a polynomial or rational function.

Basic di↵erentiation formulas

The definition of the derivative can be difficult and tedious to use.

It’s usually much faster and easier to use the Di↵erentiation Rules (last page of this document).
d
⇥ 2
• Once we’ve convinced ourselves that the derivative dx x ] of the function x2 is 2x (as we did
in Lesson 7), we can treat this fact as a rule.
d
⇥ n
• In general, the derivative dx x ] of any power function xn is nxn 1 .
(“The exponent moves out front and drops by 1.”)
• This “General Power Rule” is listed among several other basic di↵erentiation rules in the
Di↵erentiation Rules.
• All the Di↵erentiation Rules can be proven using the definition of the derivative.
• Some of the rules on the handout won’t be introduced until later in the course.

Let’s concentrate on Rules 1, 2, 4, 5, 6, and 7 first.

d
Ex. 1. Find the derivative [5] of the constant function f (x) = 5 using the definition of the
dx
derivative.
Since the average rate of change of a constant function is 0, we expect the instantaneous
rate of change to be 0 also.
d f (x + h) f (x)
[5] = lim
dx h!0 h
5 5
= lim
h!0 h
0
= lim
h!0 h

=0
where the last equality is justified by the fact that the limit as h ! 0 depends only on
nonzero values of h.
Ex. 1 is an example of the Constant Rule (Rule 1 on the Handout).
Intuitively: the instantaneous rate of change in a constant function is 0.

d e⇡ e ⇡
Ex. 2. Calculate: .
dt 2

d
Ex. 3. Find [s].
ds
Write f (s) = s.
d f (s + h) f (s) s+h s
[s] = lim = lim = 1.
ds h!0 h h!0 h

Proofs of the Power, Constant Multiple, Sum, and Di↵erence Rules (Rules 4–7 on the Handout)
can be found in the textbook.

Let’s look at some more problems similar to those you might see on an Exam.

1
Ex. 4. Find the derivative: t = 5s + 3 + 2 + 3s5 . Justify each step by stating which Rules were
s
used.
The dependent variable is t, and the independent variable is s,
so we are di↵erentiating with respect to t.
d⇥ ⇤ d ⇥ ⇤
t(s) = 5s + s 3 + 2 + 3s5
ds ds
d d ⇥ 3⇤ d d ⇥ 5⇤
= [5s] + s + [2] + 3s (Rule 6: Sum Rule)
ds ds ds ds
d d ⇥ 3⇤ d ⇥ 5⇤
= 5 [s] + s +0+3 s (Rule 5 and Rule 1)
ds ds ds
4
=5 3s + 15s4
3
=5 + 15s4 .
s4
 x3 + 2x + 1
Ex. 5. Di↵erentiate the function: y = p .
x
The dependent variable is y. The independent variable is x.
So we are di↵erentiating with respect to x.

dy d x3 + 2x + 1
=
dx dx x1/2
d ⇥ 5/2 ⇤
= x + 2x1/2 + x 1/2
dx
5 1 3/2
= x3/2 + x 1/2 x
2 2
3 p 2 3
= x x+ p p
2 x 2x x

2 p
3 t2
Ex. 6. Find the derivative: f (t) = p 27t + .
5
t 2
Independent variable: t

d d 1/5 1
f (t) = 2t 3t1/3 + t2
dt dt 2

1 6/5 1 2/3 1
= 2 5
t 3 3
t + (2)t
2
2 3 2/3 2
= t 6/5 t + t
5 3 2
2 1
= 6/5 +t
5t t2/3
2 1
= p5 6
p
3 2
+ t.
5 t t
 Product and Quotient Rules
We have seen (Sum Rule) that
d⇥ ⇤ d⇥ ⇤ d⇥ ⇤
f (x) + g(x) = f (x) + g(x) .
dx dx dx
That is,
The derivative of a sum of two functions, is the sum of the derivative of the two functions.
A similar rule holds for subtraction (Di↵erence Rule), but not for products:
d⇥ ⇤ d⇥ ⇤ d⇥ ⇤
f (x) · g(x) = f (x) · g(x) + f (x) · g(x).
dx dx dx
That is, the right-hand side of the Product Rule is
(first) · (derivative of the second) + (derivative of the first) · (second).
+ Since addition and multiplication are commutative, we can write this rule in a few di↵erent
ways, e.g.:
d⇥ ⇤
f (x) · g(x) = f (x) · g 0 (x) + f 0 (x)g(x)
dx
= f 0 (x) · g(x) + f (x)g 0 (x)
= g(x) · f 0 (x) + g 0 (x)f (x).

dy dy
Ex. 7. Find and .
dx dt
(a) y = tx2 + t3 x t x
(b) y = +
x2 t

Hint: To avoid having to use the Quotient Rule in part (b), rewrite the given formula without
fractions by using negative exponents:
t x
y = 2 + = tx 2 + xt 1
x t

Ex. 8. Di↵erentiate (5x12 + 2) (⇡ ⇡ 2 x4 ).

f (x) = 5x12 + 2 g(x) = ⇡ ⇡ 2 x4
For the Product Rule:
f 0 (x) = 60x11 g 0 (x) = 4⇡ 2 x3

d ⇥ ⇤ d ⇥ ⇤
5x12 + 2 ⇡ ⇡ 2 x4 = 5x12 + 2 ⇡ ⇡ 2 x4
dx dx
0 0
= 5x12 + 2 ⇡ ⇡ 2 x4 + 5x12 + 2 ⇡ ⇡ 2 x4
= 60x11 (⇡ ⇡ 2 x4 ) 4⇡ 2 x3 (5x12 + 2)
= 80⇡ 2 x15 + 60⇡x11 8⇡ 2 x3 .
 1 x2
Ex. 9. Prove that and have the same derivative.
1 x2 1 x2


d 1 (Recip. Rule) (1 x2 )0 ( 2x) 2x
= = = .
dx 1 x2 (1 x2 )2 (1 x2 )2 (1 x2 )2

d f (x) g(x)f 0 (x) f (x)g 0 (x)
For the Quotient Rule = , take
dx g(x) [g(x)]2
f (x) = x2 g(x) = 1 x2
f 0 (x) = 2x g 0 (x) = 2x


d x2 (Quot. Rule) (1 x2 )(x2 )0 (x2 )(1 x2 ) 0
=
dx 1 x2 (1 x2 )2
(1 x2 )(2x) (x2 )( 2x)
=
(1 x2 )2
2x 2x3 + 2x3
=
(1 x2 )2
2x
= .
(1 x2 )2

+ Silly but e↵ective trick for remembering the Quotient Rule:

‘‘Lo D hi over hi D lo, over the denominator squared must go.’’

+ Alternate strategy: Just avoid the Quotient Rule.

• In Section 3.6, we’ll learn the Chain Rule.
• After you know the Chain Rule, come back to this bulleted list.
f (x) ⇥ ⇤ 1
• Any quotient can be written as a product: f (x) · g(x) .
g(x)
• Then the Product Rule and Chain Rule can be used:
d h ⇥ ⇤ 1i ⇥ ⇤ 1 ⇥ ⇤ 2
f (x) · g(x) = f 0 (x) · g(x) f (x)g(x) · g(x)
dx
 Additional exercises

Ex. 10 (§3.3—#107, 113, 115).

✓ ◆
3 2 2 5 4x3 2x + 1
• f (x) = 5x x+1 • g(x) = x 2
+ 3 • h(x) =
x x x2

Ex. 11.

• I(x) = x5/2 x 2 1 p 5x3/2 + x5/2

• k(x) = 5
x • m(x) =
x x
2
1 1 2w w+4
• j(x) = + 2 • `(x) = x2.3 + ⇡ 2.3 • N (x) = p
x x w
 dy dy
Ex. 12. Find and .
dx dt
(a) y = t3 x + tx2 t x
(b) y =
x t2

Ex. 13 (§3.3—#119). Find the equation of the tangent line to the curve y = 2x3 + 4x2 5x 3
at the point ( 1, 4).

Ex. 14 (see also §3.3—#121).

The normal line to the graph of f at the point P is the line through the point P perpendicular
to the tangent line at point P . Find the equation of the normal line to the curve
2 3
y=
x x2
at the point (1, 1).
Ex. 17 (§3.3, Example 3.31). Find the points on the curve y = x3 7x2 + 8x + 1 where the
tangent is horizontal. (Hint: A horizontal line has slope = 0.)

Ex. 18 (§3.3—#146). The concentration of antibiotic in the bloodstream t hours after being
injected is given by the function
2t2 + t
C(t) = 3 ,
t + 50
where C(t) is measured in milligrams per liter of blood
(a) Find the rate of change of C(t).
(b) Determine the rate of change for t = 8, 12, 24, and 36.
(c) Briefly describe what seems to be occurring as the number of hours increases.
Ex. 19 (§3.3—#111, 116, 117). Di↵erentiate.
✓ ◆
4 13 x2 + 4 x+9
• f (x) = 3x 18x + • f (x) = 2 • f (x) =
x+1 x 4 x2 7x + 1
 6
Ex. 20 (§3.3—#141). Find an equation of the tangent line to the curve y = at the point
x 1
(3, 3).

Ex. 21. Find an equation of the normal line to the graph of f (x) = x + x2 at the point (0, 0).
(See above for the definition of the normal line.)
Ex. 22. The psychologist L. L. Thurstone suggested the following relationship between learning
time T = f (n) and the length n of a list:
p
T = f (n) = An n b,

where A and b are constants that depend on the person and the task.
dT
(a) Compute and interpret your results.
dn
(b) For a certain person, suppose A = 4 and b = 4. Compute f 0 (13) and f 0 (29) and interpret
your results.
 Di↵erentiation Rules
In the equations below, c is a (real) constant, and f (x) and g(x) are functions.
d⇥ ⇤
Recall: ⇤ means “the derivative of ⇤ with respect to x.’
dx

Basic formulas
d
1. Derivative of a constant [c] = 0
dx
d
2. Derivative of identity function [x] = 1
dx
d ⇥ ⇤
3. Chain Rule g f (x) = g 0 f (x) · f 0 (x)
dx

Arithmetic formulas
d c
4. Power Rule [x ] = cxc 1
dx
d⇥ ⇤
5. Constant Multiple Rule c · f (x) = c · f 0 (x)
dx
d⇥ ⇤
6. Sum Rule f (x) + g(x) = f 0 (x) + g 0 (x)
dx
d⇥ ⇤
7. Di↵erence Rule f (x) g(x) = f 0 (x) g 0 (x)
dx
d⇥ ⇤
8. Product Rule f (x) · g(x) = f (x) g 0 (x) + g(x) f 0 (x)
dx

d 1 g 0 (x)
9. Reciprocal Rule =⇥ ⇤2
dx g(x) g(x)

d f (x) g(x) f 0 (x) f (x) g 0 (x)
10. Quotient Rule = ⇥ ⇤2
dx g(x) g(x)

Transcendental functions
d⇥ ⇤ d⇥ ⇤
sin(x) = cos(x) cos(x) = sin x
dx dx
d⇥ ⇤ d⇥ ⇤
csc(x) = csc(x) cot(x) sec(x) = sec(x) tan(x)
dx dx
d⇥ ⇤ d⇥ ⇤
tan(x) = sec2 (x) cot(x) = csc2 (x)
dx dx
d x d⇥ ⇤ 1
[e ] = ex ln(x) =
dx dx x
d x d⇥ ⇤ 1
[b ] = bx ln(b) logb (x) =
dx dx x ln(b)
 Workbook Lesson 9
§3.5, Derivatives of Trigonometric Functions
Objectives
• Find the derivatives of the sine and cosine function.
• Find the derivatives of the standard trigonometric functions.
• Calculate the higher-order derivatives of the sine and cosine.

Let’s add two more Rules into the mix: the derivatives of sin and cos.

Ex. 1. Find the derivative: r = 4 cos(✓) 3 sin(✓).

Solution:
Independent variable: ✓
Dependent variable: r
dr d
= [4 cos(✓) 3 sin(✓)]
d✓ d✓
d d
=4 [cos(✓)] 3 [sin(✓)]
d✓ d✓
= 4[ sin(✓)] 3[cos(✓)]
= 4 sin(✓) 3 cos(✓).

d
⇥ ⇤ d
⇥ ⇤
Ex. 2. Show that (a) dx
csc(x) = csc(x) cot(x) and (b) dx
tan(x) = sec2 (x).
Solution:

(a): We know csc(x) = 1/ sin(x). For the Quotient Rule:

f (x) = 1 g(x) = sin(x)

f 0 (x) = 0 g 0 (x) = cos(x)

 ⇥ ⇤ ⇥ ⇤
d d 1 sin(x) · [0] [1] · cos(x)
[csc(x)] = =
dx dx sin(x) sin2 (x)
cos(x)
=
sin2 (x)
cos(x) 1
=
sin(x) sin(x)
= csc(x) cot(x).

(We could have used the Reciprocal Rule instead.)

 sin(x)
(b): We know tan(x) = . For the Quotient Rule:
cos(x)

f (x) = sin(x) g(x) = cos(x)

f 0 (x) = cos(x) g 0 (x) = sin(x)


d⇥ ⇤ d sin(x)
tan(x) =
dx dx cos x
⇥ ⇤
cos(x) · cos(x) sin(x) · sin(x)
=
cos2 (x)
cos2 (x) + sin2 (x)
= (⇤)
cos2 (x)
1
=
cos2 (x)
= sec2 (x).

Alternately, we could proceed (starting at (⇤)) as follows. . .

d⇥ ⇤
tan(x)
dx
⇥ ⇤
cos(x) · cos(x) sin(x) · sin(x)
=
cos2 (x)
cos2 (x) sin2 (x)
= +
cos2 (x) cos2 (x)
= 1 + tan2 (x).

All is well, because from the well-known identity

sin2 (x) + cos2 (x) = 1

we can deduce the (also well-known) identity

tan2 (x) + 1 = sec2 (x) :

sin2 (x) + cos2 (x) = 1 (Divide each term by cos2 (x)) (†)
1 + tan2 (x) = sec2 (x)

By the way, the identity

1 + cot2 (x) = csc2 (x) (Divide each term in (†) by sin2 (x))

can be similarly derived.

Ex. 3. Show that
d2 ⇥ ⇤
tan(x) = 2 sec2 (x) tan(x).
dx2
d
⇥ ⇤
(You may use the formula for dx tan(x) , which we derived in the previous exercise.)
Solution:

By formula,
d⇥ ⇤
tan(x) = sec2 (x).
dx

We set up the Product Rule

d
[f (x) · g(x)] = f 0 (x)g(x) + f (x)g 0 (x),
dx
for
sec2 (x) = sec(x) · sec(x)
as follows:
f (x) = sec(x) g(x) = sec(x)
f 0 (x) = sec(x) tan(x) g 0 (x) = sec(x) tan(x)

d2 ⇥ ⇤ d⇥ 2 ⇤ d⇥ ⇤
2
tan(x) = sec (x) = sec(x) · sec(x)
dx dx dx
= sec(x) · sec(x) tan(x) + sec(x) tan(x) · sec(x)
= 2 sec2 (x) tan(x).

Ex. 4. Show that the 27th derivative of sin(x) is cos(x).

 sin(x)
Ex. 5. Di↵erentiate without using the Quotient Rule.
x2
Solution:
2
Write f (x) = sin(x) and g(x) = x for the Product Rule.

2
f (x) = sin(x) g(x) = x
f 0 (x) = cos(x) g 0 (x) = 2x 3


d sin(x) d ⇥ ⇤
2 (Prod. Rule) 2
= sin(x) · x = cos(x) · x + sin(x) · ( 2x 3 )
dx x2 dx
cos(x) 2 sin(x)
= .
x2 x3

Additional exercises

Ex. 6 (§3.5—#175, 177, 179, 181). Di↵erentiate.

(a) y = x2 sec(x) + 1 sec(x)

(c) y =
x
(b) y = x2 cot(x) (d) y = (x + cos(x))(1 sin(x))
Ex. 7 (§3.5—#185). Find an equation of the tangent line to the curve y = sin(x) at the point
(0, 0).

Ex. 8 (§3.5—#197). Find all values of x at which the graph of f (x) = 3 sin(x) cos(x) has a
horizontal tangent line.
Ex. 9 (§3.5—#203). The number of hamburgers sold at a fast-food restaurant in Pasadena,
California, is approximately given by

y = 10 + 5 sin(x)

where y is the number of hamburgers sold and x represents the number of hours after the restaurant
opened at 11 a.m. The restaurant closes at 11 p.m.. Find y 0 and determine the intervals where
the number of burgers being sold is increasing.
 Workbook Lessons 10 and 11
§3.6, Chain Rule
§3.4, Derivatives as Rates of Change
Last revised: 2021-09-30 12:10

Objectives
• State the chain rule for the composition of two functions.
• Apply the chain rule together with the power rule.
• Apply the chain rule and the product/quotient rules correctly in combination when both are necessary.
• Recognize the chain rule for a composition of three or more functions.
• Find derivatives using all the Di↵erentiation Rules learned so far.
• Solve word problems that apply di↵erential calculus to problems in physics and business.

The Chain Rule

We know what the graph of (*) z = sin(x) looks like (z = vertical axis, x = horizontal axis). We
d(sin(x))
also know (**) = cos(x).
dx
Given what we know about the graphs of (*) and (**), sketch the graph of z = sin(x 1) and its
derivative.
(We don’t have any rules that tell us what the derivative of sin(x 1) is yet. At this point, just
guess what the graph of its derivative looks like.) (Hint: Use transformation of graphs.)
z

z = sin(x-1)

x
1 2 3 4 5 6
dz
dx

Now sketch the graph of sin(2x). Guess what the graph of its derivative must look like.
z
dz
dx
z = sin(2x)

x
1 2 3 4 5 6

d⇥ ⇤ d⇥ ⇤
The graphs we drew suggest that sin(x 1) = cos(x 1) and sin(2x) = 2 cos(2x).
dx dx
These guesses are correct.

There is a general rule we can use to find the derivative z = sin(f (x)) with respect to x.
Indeed, we can replace sin by any (di↵erentiable) function g.
Let f and g be two functions. The Chain Rule says that the composite function g f . . .
input to f y = output from f

x f y = f(x) g z = g(y) = g(f(x))

y = input to g z = output from g

. . . has derivative
d
g f (x) = g 0 f (x) · f 0 (x). (?)
dx

Mnemonic: Derivative of the outside, times derivative of the inside.

In Leibniz notation, equation (?) is
dz dz dy
= · .
dx dy dx
(See applet on iCollege: “Wheels and belts”)

d
Ex. 1. Find cos( 2x).
dx

Write y = f (x) = 2x and g(y) = cos(y) for the Chain Rule:

f g
x 7! y= 2x 7! cos( 2x)
inside outside
f (x) = 2x g(y) = cos(y)
f 0 (x) = 2 g 0 (y) = sin(y)

d (C.R.) (sin odd)

cos( 2x) = g 0 (f (x)) · f 0 (x) = sin( 2x) · ( 2) = 2 sin( 2x) = 2 sin(2x).
dx

Setup for the same problem, using Leibniz notation:

f g
x 7! y= 2x 7! z = cos( 2x)
inside outside
y = 2x z = cos(y)
dy dz
= 2 = sin(y)
dx dy

dz dz dy
= · = sin(y) · ( 2) = sin( 2x) · ( 2) = ...
dx dy dx
 p
4
Ex. 2. Di↵erentiate f (x) = 4x3 2.
f g
x 7! y = 4x3 2 7! (4x3 2)1/4
inside outside
f (x) = 4x3 2 g(y) = y 1/4
f 0 (x) = 12x2 g 0 (y) = 14 y 3/4

d 1 3x2
(4x3 2)1/4 = g 0 (f (x)) · f 0 (x) = g 0 (4x3 2) · (12x2 ) = (4x3 2) 3/4
· (12x2 ) = p .
dx 4 4
(4x3 2)3

✓ ◆
5 v
Ex. 3. Di↵erentiate q = sin .
v
✓ ◆
f 5 v g 5 v
v 7! y= 7! sin
v v


5 v 0 d 5 v d 1 2 5
f (v) = . f (v) = = [5v 1] = 5v = .
v dv v dv v2

✓ ◆
0 d d 5 v
g(y) = sin(y). g (y) = [ sin(y)] = [sin(y)] = cos (y) = cos .
dy dy v

 ✓ ◆ ◆ ✓ ◆ ✓
d 5 v 0 0 5 5 v
sin = g (f (v)) · f (v) = cos · 2
dv v v v
✓ ◆
5 5 v
= 2 cos .
v v

Ex. 4. Di↵erentiate f (x) = cos4 (x).

4
Note that cos4 (x) = cos(x) .
f g ⇥ ⇤4
x 7! y = cos(x) 7! cos(x)

d h 4
i d⇥ ⇤
cos(x) = 4 cos3 (x) · cos(x)
dx dx
= 4 cos3 (x) sin(x)
= 4 cos3 (x) sin(x).
 1
Ex. 5. Di↵erentiate .
(3x2 + 1)2


d 1 d ⇥ 2 ⇤
2 2
= (3x + 1) 2
dx (3x + 1) dx
d ⇥ 2 ⇤
= 2(3x2 + 1) 3
· 3x + 1
dx
= 2x(3x2 + 1) 3 (6x)
12x
= .
(3x2 + 1)3

Ex. 6. Di↵erentiate (7x 2)3 .

d⇥ ⇤ d⇥ ⇤
(7x 2)3 = 3(7x 2)2 · 7x 2
dx dx
= 21(7x 2)2 .

Ex. 7. Di↵erentiate (7x 2)3 (2x 1).

d⇥ ⇤ d⇥ ⇤ d⇥ ⇤
(7x 2)3 (2x 1) = (7x 2)3 · (2x 1) + (7x 2)3 · 2x 1 (Product Rule)
dx dx dx
d⇥ ⇤ d⇥ ⇤
= 3(7x 2)2 · 7x 2 · (2x 1) + (7x 2)3 · 2x 1 (Chain Rule)
dx dx
= 21(7x 2)2 · (2x 1) + 2(7x 2)3 .

Ex. 8. Di↵erentiate (7x 2)3 (2x 1)5 .

d⇥ ⇤ d⇥ ⇤ d⇥ ⇤
(7x 2)3 (2x 1)5 = (7x 2)3 · (2x 1)5 + (7x 2)3 · (2x 1)5 (Product Rule)
dx dx dx
d⇥ ⇤
= 21(7x 2)2 (2x 1)5 + (7x 2)3 · (2x 1)5 (Chain Rule)
dx
d⇥ ⇤
= 21(7x 2)2 (2x 1)5 + (7x 2)4 · 5(2x 1)4 · 2x 1
dx
= 21(7x 2)2 (2x 1)5 + 10(7x 2)4 .
Ex. 9. Prove that
(a) the derivative of an even function is odd, and
(b) the derivative of an odd function is even.
Solution:
f even =) f (x) = f ( x).
By Chain Rule, f 0 (x) = [f ( x)]0 = f 0 ( x) · ( 1) = f 0 ( x).
Thus f 0 is odd: f 0 (x) = f 0 ( x).

f odd =) f ( x) = f (x) =) f (x) = f ( x).

0 0 0
By Chain Rule, f (x) = [ f ( x)] = [f ( x)] = [f 0 ( x) · ( 1)] = f 0 ( x).
Thus f 0 is even: f 0 (x) = f 0 ( x).

Ex. 10.
• Suppose the motion of a particle is described by a displacement function s(t).
• As usual, let v(t) be the particle’s velocity, and let a(t) be the particle’s acceleration.
• Prove that
dv
a(t) = v(t)
ds
(note the use of Lebiniz notation).

Solution:
First, let’s clarify the meaning of each variable:
t time

s(t) displacement
ds
v(t) = velocity
dt
dv
a(t) = acceleration
dt
Comments:

By Chain Rule,
dv dv ds dv
a= = = v(t).
dt ds dt ds
dv
is acceleration, i.e. the rate of change in velocity with respect to time.
dt
dv
is the rate of velocity with respect to the displacement.
ds
Example:
v(s) = sin(s) and s(t) = 2t.
Then
dv d
= sin(s) = cos(s),
ds ds
while
ds
v(t) = =2
dt
and
d2 v dv
= v(t) = cos(s) · 2 = 2 cos(2t).
dt2 ds

Exercises 11–35 below are accompanied by fully worked solutions. You are not expected to practice
every single exercise.
+ But please be sure to study Exercises 24, 25, 26, 27, and 28!
+ Business students are encouraged to also study Exercises 32–35.
 Worked problems
Ex. 11. Di↵erentiate:
1
(a) p
(5x 1)3
1
(b) t sin
t
(c) sin(cos( 13 s3 ))
Solution:
" #
d 1 d ⇥ ⇤
p = (5x 1) 3/2 (a)
dx (5x 1)3 dx
3 d⇥ ⇤
= (5x 1) 5/2 · 5x 1
2 dx
15
= p .
2 (5x 1)5

d ⇥ ⇤ d d ⇥ ⇤
t sin(t 1 ) = [t] · sin(t 1 ) + t · sin(t 1 ) (b)
dt dt dt
d ⇥ ⇤
= sin(t 1 ) + t sin(t 1 )
dt
d
= sin(t ) + t cos(t 1 ) · [t 1 ]
1
dt
= sin(t 1 ) t cos(t 1 ) · t 2

1 cos 1t
= sin .
t t

d ⇥ ⇤ d ⇥ ⇤
sin cos( 13 s3 ) = cos cos( 13 s3 ) · cos( 13 s3 ) (c)
ds ds
d ⇥ 1 3⇤
= cos cos( 13 s3 ) · sin( 13 s3 ) · s
ds 3
= cos cos( 13 s3 ) · sin( 13 s3 ) · (s2 )
= s2 cos cos( 13 s3 ) sin( 13 s3 ).

dy
Ex. 12. If y(x) = 9x3 + 3x2 + 5 and x(t) = 7t2 + 10t + 2, find .
dt
Solution:
dy dy dx
= = (27x2 + 6x)(14t + 10)
dt dx dt
= 6x(9x + 2)(7t + 5)
= 6(7t2 + 10t + 2) 9(7t2 + 10t + 2) + 2 (7t + 5)
= 6(7t2 + 10t + 2)(63t2 + 90t + 20)(7t + 5).
 p
Ex. 13. Di↵erentiate y = x(x 1).
Solution:
0
y 0 = x1/2 (x 1)
0
= x3/2 x1/2
3p 1
= x p
2 x
.
2

0
y 0 = x1/2 (x 1)
1
= p (x 1) + x1/2 (1)
2 x
x 1
= p + 22x p
x
2 x
3x 1
= p .
2 x

p p
Ex. 14. Di↵erentiate g(u) = 2u+ 3u. Justify each step by giving the name of the di↵erentiation
rule you are using.
Solution:
p p 0
g 0 (u) = 2u + 3u
p 0 p 0
= 2u +3u Sum Rule
p 0 p p 0
= 2 u + 3 u Identity & Constant Multiple
p 0 p 0
= 2 u + 3 u1/2 Constant Multiple
p p
= 2 + 23 u 1/2 Identity & Power Rule
p p
= 2 + 2p3u .

Ex. 15.
(a) Di↵erentiate H(x) = (x + x 1 )3 without using the Chain Rule.
(b) Then di↵erentiate it using the Chain Rule.
Solution:
0
H 0 (x) = (x + x 1 )3
0
= x3 + 3x2 x 1
+ 3x(x 1 )2 + (x 1 )3
3 0
= x3 + 3x + 3x 1 + x
= 3x2 + 3 3x 2 3x 4

= 3x2 + 3 x32 x34 .

 0
H 0 (x) = (x + x 1 )3
0
= 3(x + x 1 )2 x + x 1
= 3(x + x 1 )2 (1 x 2 )
= 3(x + x1 )2 (1 x12 ).

Ex. 16.
p
x 1
(a) Di↵erentiate y = p using the Quotient Rule.
x+1
(b) Then di↵erentiate it using the Product Rule.
Solution:

(x1/2 + 1)(x1/2 1)0 (x1/2 1)(x1/2 + 1)0

y0 =
(x1/2 + 1)2
(x1/2 + 1)( 2 x 1/2 ) (x1/2 1)( 12 x 1/2 )
1
=
(x1/2 + 1)2
1
2
(1 + x 1/2 ) 12 (1 x 1/2 )
=
(x1/2 + 1)2
1 + x 1/2 1 + x 1/2
=
2(x1/2 + 1)2
1
= p p .
2 x( x + 1)2

1 0
y 0 = (x1/2 1) · (x1/2 + 1)
0
= (x1/2 1)0 (x1/2 + 1) 1 + (x1/2 1) (x1/2 + 1) 1
= ( 12 x 1/2 )(x1/2 + 1) 1 + (x1/2 1)( 1)(x1/2 + 1) 2 · 12 x 1/2
p
1 x 1
= p p p p
2 x( x + 1) 2 x( x + 1)2

Ex. 17.
t
(a) Di↵erentiate y = without using the Chain Rule.
(t 1)2
(b) Then di↵erentiate it using the Chain Rule.
Solution to part (a):
✓ ◆0
0 t
y =
(t1)2
✓ ◆0
t
=
t2 2t + 1
(t2 2t + 1) t(2t 2)
=
(t2 2t + 1)2
1 t2
=
(t2 2t + 1)2
(1 t)(1 + t)
=
(t 1)4
1 t
= .
(t 1)3

Solution to part (b):

z+1 t
Take f (t) = t 1 and g(z) = 2
, so that g f (t) = .
z (t 1)2
Then
f 0 (t) = 1
and

0 d z+1
g (z) =
dz z2
d ⇥ ⇤
= z 2 (z + 1)
dz
d ⇥ 1 2
⇤
= z +z
dz
2
= z 2z 3 ,
so

y 0 = g 0 f (t) · f 0 (t)
= g 0 (t 1) · 1
2 3
= (t 1) 2(t 1)
1 2
= +
(t 1)2 (t 1)3
1(t 1) 2
=
(t 1)3
t 1
= .
(t 1)3
 1
Ex. 18. Di↵erentiate y = without using the Quotient Rule or the Product Rule.
(1 + sec x)2
Solution:
2 0 2 sec x tan x
y 0 = (1 + sec x) = 2(1 + sec x) 3
sec x tan x = .
(1 + sec x)3

Ex. 19. Let g(x) = (x2 +1)3 (x3 +2)6 . Show that g 0 (x) = 6x(x2 +1)2 (x3 +2)5 (4x3 +3x+2).
Solution:
⇥ ⇤0
g 0 (x) = (x2 + 1)3 (x3 + 2)6
⇥ ⇤0 ⇥ ⇤0
= (x2 + 1)3 (x3 + 2)6 + (x2 + 1)3 (x3 + 2)6
= 18x2 (x2 + 1)3 (x3 + 2)5 + 6x(x2 + 1)2 (x3 + 2)6 Collect like terms
= 6x(x2 + 1)2 (x3 + 2)5 3x(x2 + 1) + (x3 + 2)
= 6x(x2 + 1)2 (x3 + 2)5 (4x3 + 3x + 2).

x 14 3x
Ex. 20. Let f (x) = p . Show that f 0 (x) = p .
7 3x 2 (7 3x)3
Solution:
1/2 0
f 0 (x) = x(7 3x)
0
= x (7 3x) 1/2 + (7 3x) 1/2
= x( 21 )(7 3x) 3/2 ( 3) + (7 3x) 1/2 Factor out (7 3x) 3/2

= (7 3x) 3/2 32 x + (7 3x)

3 1
2
x+7 2
( 3x + 14) 14 3x
= = = p .
(7 3x)3/2 (7 3x)3/2 2 (7 3x)3

Ex. 21. Di↵erentiate y = cos4 (sin3 x).

Solution:
⇥ ⇤4
y = cos (sin x)3 .
⇢
0
⇥ ⇤ 0
3 4
y = cos (sin x)
⇥ ⇤3 ⇥ ⇤0
= 4 cos (sin x)3 · cos (sin x)3
0
= 4 cos3 (sin3 x) · sin (sin x)3 · (sin x)3
= 4 cos3 (sin3 x) sin(sin3 x) · 3 (sin x)2 · (sin x)0
= 12 cos3 (sin3 x) sin(sin3 x) sin2 (x) cos(x).
 ⇢
x sin x1 if x 6= 0,
Ex. 22. Let f (x) = . Find a formula for f 0 (x). Does f 0 (0) exist?
0 if x = 0.
Solution:
For x 6= 0:
0
f 0 (x) = x sin(x 1 )
0
= sin(x 1 ) + x sin(x 1 )
= sin(x 1 ) + x cos(x 1 ) · 1
x2
= sin x1 1
x
cos x1 .

For x 6= 0:
f (x) f (0)
f 0 (0) = lim (definition of derivative)
x!0 x 0
x sin x1 0
= lim (f (x) = x sin x1 for x 6= 0)
x!0 x
= lim sin x1 does not exist. (See Section 2.5)
x!0

1 1 1
f 0 (0) does not exist. For x 6= 0, f 0 (x) = sin x x
cos x
.

Ex. 23.
p d x
(a) Write |x| = x2 , and use the Chain Rule to show that |x| = .
dx |x|
(b) Let f (x) = | sin x|. Find f 0 (x) and sketch the graphs of f and f 0 . Where is f not di↵erentiable?
Solution:
d d 1 x x
|x| = (x2 )1/2 = (x2 ) 1/2
· 2x = = .
dx dx 2 (x2 )1/2 |x|

d sin x d sin x
| sin x| = · (sin x) = cos x.
dx | sin x| dx | sin x|

d sin x
| sin x| = cos x is undefined () | sin x| = 0
dx | sin x|
() x = n⇡ where n is any integer.
 Worked problems with tangent lines

Ex. 24. Find an equation of the tangent line to the curve y = 1/(1 + x2 ), called the witch of
Maria Agnesi, at the point ( 1, 12 ).

Maria Gaetana Agnesi’s construction of the curve (1748)

Solution.
General form of the equation of a tangent line to the graph of y = y(x):

y y(a) = y 0 (a) · (x a)

Here a, f (a) = ( 1, 12 ), and since

2x
y0 = 1(1 + x2 ) 2 (2x) = ,
(1 + x2 )2
we have

1
y 0 ( 1) = .
2
1
y 2
= 12 (x + 1)

p
Ex. 25. Find an equation of the tangent line to the curve y = x x that is parallel to the line
y = 1 + 3x.
Solution:
p
y 0 = (x3/2 )0 = 3
2
a.

The slope of the line y = 1 + 3x is m = 3. p p p

The general form of the tangent line to y = x x at x = a is y a a = 32 a(x a).
3p
Solving 2 a = 3 yields a = 4.
y 8 = 3(x 4)
 x 1
Ex. 26. Find equations of the tangent lines to the curve y = that are parallel to the line
x+1
x 2y = 2.
(x + 1)(1) (x 1)(1) 2
y0 = 2
= .
(x + 1) (x + 1)2

Solution:
The slope of x 2y = 2 is m = 12 :

x 2 = 2y
y = 12 x 1
x 1 2
The slope of the tangent line to y = at x = a is y 0 (a) = . We solve the
x+1 (a + 1)2
equation y 0 (a) = m.
2
= 12
(a + 1)2
(a + 1)2 = 4
p
a= 1± 4 = 1 or 3

Since y( 3) = 2 and y(1) = 0, the desired tangent lines are

y 2 = 12 (x + 3) and y 0 = 12 (x 1) .

Ex. 27. Show that the curve y = 6x3 + 5x 3 has no tangent line with slope 4.
Solution:
y 0 = 18x2 + 5
We solve y 0 (a) = 4:

y 0 (a) = 18a2 + 5 = 4
18a2 = 1
a2 = 181 < 0 Impossible: the square of any real number is nonnegative.

We have shown that: if y 0 (a) = 4, then a2 < 0.

Since a2 < 0 is impossible, there is no tangent line with slope y 0 (a) = 4.
 Some physical applications of basic di↵erential calculus
We now turn our attention to a variety of applications (word problems) that can be solved using
the knowledge of derivatives we have developed thus far.

Ex. 28. The position of a particle in motion is given by the equation s = t3 6t2 + 9t, where s is
measured in meters and t is measured in seconds.
(a) Find the velocity at time t.
(b) What is the velocity after 2 s? After 4 s?
(c) When is the particle at rest?
(d) When is the particle moving forward (that is, in the positive direction)?
(Hint: When is v(t) > 0?)
(e) Draw a diagram to represent the motion of the particle.
(f) Find the total distance traveled by the particle during the first five seconds.
(g) Find the acceleration at time t.
(h) Find the acceleration after 4 s.

Solution.
(a) v(t) = s0 (t) = 3t2 12t + 9 .

(b) v(2) = 3 and v(4) = 9.

The velocity is 3m/s at time t = 2, and 9m/s at time t = 4.

(c) The equation v(t) = 0 has solutions t = 1 and t = 3.

The particle is at rest after t = 1 second and after t = 3 seconds.

(d) The particle moves forward when t < 1 or t > 3.

(f) The forward motion occurs when t < 1 or t > 3:

(forward motion) = |s(1) s(0)| + |s(5) s(3)| = |4 0| + |20 0| = 24m.

The backward motion occurs when 1 < t < 3:

(backward motion) = |s(3) s(1)| = |0 4| = 4m.

The total distance traveled is (forward motion) + (backward motion) = 24 + 4 = 28 m.

(g) a(t) = v 0 (t) = 6t 12.

(h) a(4) = 12 m/s2 .

Ex. 29. (Calculator) If a rock is thrown vertically upward from the surface of Mars with velocity
15 m/s, its height after t seconds is h = 15t 1.86t2 .
(a) What is the velocity of the rock after 2 s?
(b) What is the velocity of the rock when its height is 25 m on its way up?
(c) What is the velocity of the rock when its height is 25 m on its way down?
Solution.
(a): v(t) = h0 (t) = 15 3.72t, so v(2) = 7.56 m/s2 .
(b) and (c):
Suppose h(t) = 25. To solve the equation h(t) = 25, we use the quadratic formula:

1.86t2 15t + 25 = 0
p
15 ± 152 4(1.86)(25)
t=
2(1.86)
h(t) = 25 when t = t1 ⇡ 2.35 ot t = t2 ⇡ 5.71.
What is the approximate rate of change in the height at times t = t1 and t = t2 ?
v(t1 ) ⇡ 6.24 (upward)
v(t2 ) ⇡ 6.24 (downward)
Answer to (b): 6.24 m/s upward
Answer to (c): 6.24 m/s downward

kg · m
+ Recall: 1 newton (N) = 1
s2
is the force required to produce an acceleration of 1 m/s2
in a body of mass 1 kg.

Ex. 30. Newton’s Law of Gravitation says that the magnitude F of the force (in newtons N) exerted
GmM
by a body of mass on a body of mass M is F = , where G is the gravitational constant and
r2
r is the distance between the bodies.
dF
(a) Find , and explain its meaning. What does the minus sign indicate?
dr
(b) Suppose it is known that the earth attracts an object with a force that decreases at the rate
of 2 N/km when r = 20, 000 km. How fast does this force change when r = 10, 000 km?
Solution.
GmM 2
(a) F (r) = | {z } · r .
= GmM
r2
constants
dF
= 2 GmM r 3 is the rate of change in the gravitational force F between the two
dr
bodies with respect to the distance r between them. The minus sign indicates that, as the
distance (independent variable) r increases, (the dependent variable) F decreases.
 (When interpreting the meaning of the derivative in this and similar problems, regard the
independent variable as increasing.)
(b) Given that F 0 (20, 000) = 2, we want to find F 0 (10, 000). Using the fact that GmM =
20, 0003 . . .
2GmM
2 = F 0 (20, 000) =
20, 0003
GmM = 20, 0003

. . . we find that
2 · 20, 0003 16 ⇥ 1012
F 0 (10, 000) = = = 16.
10, 0003 1 ⇥ 1012
The force decreases at 16 N/km when r = 10, 000 km.

Ex. 31. If the equation of motion of a particle is given by s = A cos(!t + ), the particle is said
to undergo simple harmonic motion.
(a) Find the velocity of the particle at time t.
(b) When is the velocity 0?
(a) v(t) = s0 (t) = !A sin(!t + ).
(b)
v(t) = !A sin(!t + ) = 0
sin(!t + ) = 0

Since sin x = 0 only if x = n⇡, where n is any integer, we set x(t) = !t + = 0:

!t + = n⇡

n⇡
The velocity is 0 when t = where n is any integer.
!
+ A frictionless pendulum undergoes simple harmonic motion—provided that the pendulum
swings only through a small angle, and does not swing entirely around the pivot it hangs from! A
more realistic mathematical model would include a frictional force that causes the motion to slow
to an eventual halt. Such motion is called simple harmonic motion with damping. For details,
see any undergraduate Di↵erential Equations textbook.
 Additional applications of basic di↵erential calculus

Ex. 32. The cost of producing x ounces of gold from a new gold mine is C = f (x) dollars.
(a) What is the meaning of the derivative f 0 (x)? What are its units?
(b) What does the statement f 0 (800) = 17 mean?
(c) Do you think f 0 (x) will increase or decrease in the short term? What about in the long term?
Explain.
(a) Rate of change in production cost with respect to number of ounces of gold produced,
in dollars per ounce.
(b) After 800 oz of gold have been produced, the rate at which the production cost increases
is about $17 per ounce. So the cost of producing the 801st ounce is ⇡ $17.
(c) The production cost of the first ounce includes all startup costs. Initially, the rate of
increase in production costs will decrease due to increasingly efficient use of startup costs.
Eventually, the rate may increase due to costs specific to large-scale operations.

Ex. 33. The number N of locations of a popular co↵eehouse chain is given in the table. (The
numbers as of October 1 are given.)
Year 2004 2005 2006 2007 2008
N 8569 10, 241 12, 440 15, 011 16, 680
(a) Find the average rate of growth (i) from 2006 to 2008, (ii) from 2006 to 2007, (iii) from 2005
to 2006. Include the units.
(b) Estimate the instantaneous rate of growth in 2006 by taking the average of two average rates
of change. What are its units?
(a)

N
(Average rate of change from N = 2006 to N = 2008) = (i)
t
16, 680 12, 440
=
2
= 2120 locations/yr.
N
(Average rate of change from N = 2006 to N = 2007) = (ii)
t
= 2571 locations/yr.
N
(Average rate of change from N = 2006 to N = 2007) = (iii)
t
= 2199 locations/yr.

(b) We use the average rates of change from parts (ii) and (iii) because the instantaneous
2199 + 2571
rate of change, N 0 (t), is a two-sided limit: = 2385 locations/yr.
2
Ex. 34. (Calculator required) The cost (in dollars) of producing x units of a certain commodity is
C(x) = 5000 + 10x + 0.05x2 .
(a) Find the average rate of change of C with respect to x when the production level is changed
(i) from x = 100 to x = 105, (ii) from x = 100 to x = 101.
(b) Find the instantaneous rate of change of C with respect to x when x = 100. (This is called
the marginal cost.)
(a)
C
(i) = $20.25 per unit.
x
C
(ii) = $20.05 per unit.
x
(b) C 0 (100) = $20 per unit.

+ In the previous example,

x = (number of units)
only takes integer values. It does not make literal sense to speak about the limit

C(x + h) C(x)
C 0 (x) = lim , (1)
h!0 h
since C(x + h) is technically not defined when h ! 0 is so small that x + h is not an integer.
However, we can always replace C(x) by a smooth (i.e. di↵erentiable) approximating function.

Ex. 35. A manufacturer produces bolts of fabric with a fixed width. The quantity q of this fabric
(measured in yards) that is sold is a function of the selling price p (in dollars per yard), so we can
write q = f (p). Then the total revenue earned with selling price p is R(p) = pf (p).
(a) What does it mean to say that f (20) = 10, 000 and f 0 (20) = 350?
(b) Assuming the values in part (a), find R0 (20) and interpret your answer.

(a) q = f (p) = quantity of the fabric sold as a function of the selling price p.
f (20) = 10, 000 means that

10, 000 yards are sold when the price is $20/yd.

f 0 (20) = 350 means that:

As the selling price increases past $20/yd, the amount of fabric sold is decreasing at a
rate of 350 yards per $1/yd increase in the price.

(b) Di↵erentiating R(p) = pf (p) yields

R0 (p) = f (p) + pf 0 (p).

 Given that f (20) = 10, 000 and f 0 (20) = 350, we get

R0 (20) = f (20) + 20 · f 0 (20) = 10, 000 7, 000. (*)

Here, 7000 is the loss in dollars per $/yd due to selling less fabric (f 0 (20) < 0).
But the fact that R0 (20) = 3000 means that:

As the price of fabric per yard increases past $20, the total revenue is increasing at
$3000 per $/yd increase in price.

We conclude from equation (*) that the revenue 10, 000 makes up for the loss 7000 due
to increasing the price.

Additional exercises

Ex. 36 (§3.6—#245). If

h(x) = f g(x) , f (2) = 4, f 0 (2) = 4, f 0 (0) = 5, g(0) = 0, and g 0 (0) = 2,

what is h0 (0)?

Ex. 37 (§3.6—#215, 217, 219, 223, 233, 235). Di↵erentiate.

p 3
(a) y = 6(7x 4)3 (c) y = 4(x2 6x) + 3 (e) y = tan(x) + sin(x)
✓ ◆ ✓ ◆7
x x 7
(b) y = cos (d) y = + (f) y = sin cos(7x)
8 7 x
 x
Ex. 38 (§3.6—#241). Find an equation of the tangent line to the curve y = sin 2
at the
origin.

6 8
Ex. 39. (§3.6—#243). Find all points on the graph of the function f (x) = x x
at which
the tangent is horizontal.

Ex. 40 (§3.6—#254). A mass hanging from a vertical spring is in simple harmonic motion as
given by the following position function, where t is measured in seconds and s is in inches:
⇡
s(t) = 3 cos ⇡t + 4
.

Find the velocity of the mass at time t = 1.5 seconds.

Ex. 41 (§3.4—#155). The position function s(t) = t2 3t 4 represents the position of the
back of a car backing out of a driveway and then driving in a straight line, where s is in feet and t
is in seconds. When the position is 0, the back of the car is at the garage door. The car’s starting
position is s(0) = 4, that is, 4 feet inside the garage.
(a) Find the velocity of the car when s(t) = 0.
(b) Find the velocity of the car when s(t) = 14.

Ex. 42 (§3.4—#157). A potato is launched vertically upward with an initial velocity of 100 ft/s
from a potato gun at the top of an 85-foot-tall building. Its height above ground level after t
seconds is given by s(t) = 16t2 + 100t + 85.
(a) Find the velocity of the potato at 0.5 seconds.
(b) Find the velocity of the potato at 5.75 seconds.
(c) Find the speed of the potato at 0.5 seconds.
(d) Find the speed of the potato at 5.75 seconds.
(e) When does the potato reach its maximum height?
(f) What’s the acceleration of the potato at 0.5 s and 1.5 s?
(g) How long is the potato in the air?
Ex. 43 (§3.4, Example 3.36). A particle’s displacement is given by the function s = f (t),
where
f (t) = t3 9t2 + 24t + 4 (t 0).
Here time t is measured in seconds and displacement s is measured in feet.
(a) Find the velocity at time t.
(b) When is the particle at rest?
(c) When is the particle moving in the positive direction (from left to right)? When is the particle
moving in the negative direction (from right to left)?

Ex. 43. A spherical balloon is being inflated. Recall that the surface area of a sphere with radius
r is given by S = 4⇡r3 . Find the rate of increase in the surface area when r is 1 ft, 2 ft, and 3 ft.
What conclusion can you make?
 Workbook Lesson 12
§3.8, Implicit Di↵erentiation
Objectives
• Find the derivative of a complicated function by using implicit di↵erentiation.
• Use implicit di↵erentiation to determine the equation of a tangent line.

Implicit di↵erentiation
Consider the following equation of a circle:

x2 + y 2 = 25.

Is y a function of x?
We can, however, solve this equation for y. For each value of x, there are two values of y:
p
y = ± 25 x2 .

This equation determines two functions of x:

p p
f (x) = 25 x2 g(x) = 25 x2

The next equation isn’t nearly as easy to solve for y.

x3 + y 3 = 6xy

A computer algebra system gives

⇣p ⌘2/3 p
22/3 x3 (x3 32) x3 + 4 3 2x
y= qp .
2 3 x3 (x3 32) x3

Now suppose we’re asked to find a tangent line to this curve, say at the point (3, 3). The equation
of the tangent line would be
y 3 = f 0 (3)(x 3).
But we really don’t want to take the derivative of the last equation!
Fortunately, we don’t have to. We’ll assume that y can be solved as one or more (di↵erentiable)
functions of x. Then both sides of the equation can be di↵erentiated. We call this process
implicit di↵erentiation.

Ex. 1. Find the equation of the tangent line to the graph of x3 + y 3 = 6xy at (3, 3),
Solution:
d ⇥ 3 ⇤ d
x + y3 = [6xy]
dx dx
⇥ ⇤
3x2 + 3y 2 y 0 = 6 (1)(y) + (x)(y 0 )
3x2 + 3y 2 y 0 = 6y + 6xy 0

We can now solve for y 0 .

3x2 6y = 6xy 0 3y 2 y 0
x2 2y = (2x y 2 )y 0
0 x2 2y
y =
2x y2
Notice that y 0 is a function that has two inputs: we can indicate this fact by writing
y 0 = y 0 (x, y).
9 6
y 0 (3, 3) = = 1.
6 9
The equation of the tangent line at (3, 3), therefore, is
y 3= (x 3).

p
Ex. 2. Find the tangent line to x2/3 + y 2/3 = 4 at 3 3, 1 ⇡ ( 5.196, 1) and the tangent line
at (8, 0).
Solution:
p
At the point 3 3, 1 :
2 2
1/3
+ 1/3 y 0 = 0
3x 3y

2 3y 1/3
y0 =
3x1/3 2

y 1/3
y0 =
x1/3

p 1 1 1 1 1
y0 3 3, 1 = p = p = = =p .
( 3 3)1/3 ( 32 · 3)1/3 ( 33/2 )1/3 ( 1)1/3 (33/2 )1/3 3
Tangent line:
1 p
y 1 = p (x + 3 3)
3

1
y = p x+4
3
 At the point (8, 0):

01/3
y 0 (8, 0) =
81/3

y 0 = 0(x 8)

y=0

Ex. 3. Show that any tangent line at a point P to a circle with center O is perpendicular to the
radius OP .
Solution:
Let ⌃ be a circle with radius r. For simplicity assume its center is (0, 0). Its equation is
x2 + y 2 = r 2
Let P = (x0 , y0 ) 2 ⌃ (that is, P is a point on the circle). By implicit di↵erentiation,
x
2x + 2yy 0 = 0 =) y0 = ,
y
x0
so the tangent line to the circle at P is y 0 (x0 , y0 ) = . But the slope of the radius is
y0
y0 0
. Since these slopes are each other’s negative reciprocal, the tangent line at P is
x0 0
perpendicular to the radius OP .

Ex. 4. Show that the tangent to the ellipse

x2 y 2
+ 2 =1
a2 b
at the point (x0 , y0 ) is
x0 x y0 y
+ 2 = 1.
a2 b
Solution:
By implicit di↵erentiation,
2 2
2
x + 2 yy 0 = 0
a b
2b2 x
y0 =
2a2 y

b2 x 0
y 0 (x0 , y0 ) =
a2 y 0
 y0 y x0 x
The tangent line at (x0 , y0 ) is + 2 =1:
b2 a

b2 x 0
y y0 = (x x0 )
a2 y 0

y0 y y0 2 x0 x x0 2
= + 2
b2 b2 a2 a
y0 y x0 x x0 2 y0 2
+ = + 2
b2 a2 a2 b
y0 y x0 x x0 2 y0 2
+ = + 2 = 1.
b2 a2 a2 b

Additional exercises

dy
Ex. 5 (§3.8—#301, 307, 309). Find .
dx
(a) 6x2 + 3y 2 = 12 (b) y sin(xy) = y 2 + 2 (c) x3 y + xy 3 = 8
Ex. 6. Find y 00 and simplify fully.
sin(y) + cos(x) = 1

Ex. 7 (§3.8—#317). Find the equation of the normal line to the graph of x2 + 2xy 3y 2 = 0
at the point (1, 1).

Ex. 8 (§3.8—#318). Find all points on the graph of y 3 27y = x2 90 at which the tangent
line is vertical.
Ex. 9 (§3.8—#323). The number of cell phones produced when x dollars is spent on labor and
y dollars is spent on capital invested by a manufacturer can be modeled by the equation

60x3/4 y 1/4 = 3240.

dy
(a) Find .
dx
dy
(b) Evaluate at the point (81, 16).
dx
(c) Interpret the result of the previous part.
 Workbook Lesson 13
§3.7, Derivatives of Inverse Functions
Objectives
• Calculate the derivative of an inverse function.
• Recognize the derivatives of the standard inverse trigonometric functions.

Recall:
A function f is one-to-one (or invertible) if di↵erent input values yield di↵erent output
values. In symbols,
x1 6= x2 =) f (x1 ) 6= f (x2 )

The inverse of a function f —denoted f 1 and pronounced “f inverse”—is a function that

“undoes” f . That is, the following Cancellation Formulas hold:
1
f f (x) = x for all x in the domain of f ,
1
f f (y) = y for all y in the range of f ,

Theorem. If a function f is one-to-one, then its inverse function exists.

Counterexample. The function f (x) = x2 with domain ( 1, 1) is not one-to-one, because
there exist di↵erent input values (for example, x = 2 and x = 2) that yield the same output
value f (2) = 4 = f ( 2).
But, the function g(x) = x2 (same equation) with domain [0, 1) (di↵erent domain!) is
one-to-one.
The Horizontal Line Test can be used to check whether a function is one-to-one: if no
horizontal line meets the graph of the function in more than one point, then the function is
one-to-one, and therefore has an inverse.
�=�� �=��

� �

1 p
f (x) has no inverse The inverse of g(x) is g (y) = y
 1
To construct the graph of f , we reflect the graph of f in the line y = x.
• This reflection has the e↵ect of swapping the roles of x and y.
• That is, for each point a, f (a) on the graph of the original function f (shown in blue
below), we find the “shadow point” f (a), a on the graph of the inverse f 1 (shown in
orange below).
(See applet on iCollege: “Tangent and tangent to inverse”)
�

�
�=
� (�)

�
� � (�)

The Inverse Function Theorem

1
If a function f is both invertible and di↵erentiable, it seems reasonable that its inverse f is also
di↵erentiable.
After all, if the tangent line to the graph of f at the point a, f (a) has slope
y
f 0 (a) =
,
x
1
then shouldn’t the slope of the tangent line to the graph of f at the corresponding “shadow
point” f (a), a simply have slope
1 0 x 1
f f (a) = = 0 , (?)
y f (a)
since the roles of x and y are swapped in the graph of an inverse function?
The answer is, yes, provided that the following conditions are met:
• f is invertible and di↵erentiable.
• f 0 (a), the derivative of f at a, exists and is a real number.
• Furthermore, f 0 (a) cannot be equal to zero, otherwise equation (?) will be meaningless.
Now, whenever we deal with inverse functions, it’s easy to confuse the roles of x and y, since these
roles swap when we move from f to f 1 or vice versa.
If we stick with the convention that x-values are inputs to f , and y-values are outputs of f , then
y-values are inputs to f 1 and x-values are inputs to f , and equation (?) can be written either
as
0 1
f 1 f (x) = 0 .
f (x)
or
0 1
f 1 y = 0 .
f f 1 (y)

However, it is more common to use x to denote the input value no matter whether we are talking
about f or f 1 . If we adopt this convention, equation (?) becomes

1 0 1
f x = .
f0 f 1 (x)

We will follow the textbook and use the latter convention.

Inverse Function Theorem. Let f be a function. If f is invertible and di↵erentiable, then

1 0 1
f x =
f0 f 1 (x)

provided that
f0 f 1
(x) 6= 0.

1 0
Ex. 1. Find f (1) if f (x) = 2x + cos x.
Solution:
We know f is di↵erentiable. Is f one-to-one?
f 0 (x) = 2 sin x > 0, so f is increasing, so f is one-to-one.

1
Let’s find a = f (1).
1
f (1) = a
1 = f (a)
1 = 2a + cos a
a=0 (by inspection)

By the Inverse Function Theorem,

1 0 1 1 1 1
f (1) = = = = .
f0 f 1 (1) f 0 (a) 2 sin 0 2
 x+2
Ex. 2. Let f (x) = .
x
(a) Find the derivative of f .
(b) Find a formula for f 1 .
(c) Use the Inverse Function Theorem to find the derivative of f 1 .
(d) Find the derivative of f 1 without using the Inverse Function Theorem.
Solution:
x+2 (x)(1) (x + 2)(1) 2
(a) By the Quotient Rule, f 0 (x) = = 2
= 2.
x x x
1
(b) To find a formula for f , we write y = f (x) as follows:
x +2
y=
x
Then we solve for the input x:
xy = x + 2
xy x = 2
(y 1)x = 2
2
x=
y 1
1
Finally, we swap x and y to obtain a formula for f (x), bearing in mind that in this final
equation, x is the output of f :
1 2
f (x) = y = .
x 1
(c) First, let’s check the conditions for the Inverse Function Theorem:
• Since
x+2 1
f (x) = =1+2·
x x
1
is a transformation of the elementary function , we see that f passes the Horizontal
x
Line Test, so f is invertible.
2
• f is di↵erentiable except at x = 0, where f 0 (x) = is undefined.
x2
We can now apply the Inverse Function Theorem:
2 2 ✓ ◆2
1 0 1 1 1 x 1 1 2 2
f x = 0 = 0 2 = 2 = = =
1
f f (x) f x 1 2 x 12 2 2 x 1 (x 1)2

2
(d) We verify our answer to (c) by di↵erentiating f 1 (x) = :
x 1

d 2 (x 1)(0) 2(1)
=
dx x 1 (x 1)2
2
= .
(x 1)2
 p
Ex. 3. Let g(x) = 5
x.
1
(a) Find a formula for g and its derivative.
(b) Use the Inverse Function Theorem to find the derivative of g.
Solution:
1 0
(a) g 1 (x) = x5 and g (x) = 5x4 .
(b) Let us write f = g 1 , noting that, since f and g are inverse functions, g = f 1
.
Now
1 0 1
g 0 (x) = f (x) =
f0 f 1 (x)

1
=
g 1 0 g(x)
1
= p
g 1 0 5
x
1
= p 4
5 5x
1
= 4
5 x1/5
1 4/5
= x .
5

p
3
Ex. 4. Let g(x) = x 1.
1
(a) Find a formula for g and its derivative.
(b) Use the Inverse Function Theorem to find the derivative of g.
 Derivatives of inverse trigonometric functions

It is clear that the standard six standard trigonometric functions are not one-to-one, and thus do
not have inverses.
y=sinHxL y=cosHxL y=tanHxL
1.0 1.0 6
4
0.5 0.5
2

p p p p p p
-p - p -p - p -p - -2 p
2 -0.5 2 2 -0.5 2 2 2
-4
-1.0 -1.0 -6

y=cscHxL y=secHxL y=cotHxL

8 6
6 5 4
4
2
2
p 3p p p
-2 p
p
3p
2p p 2p -p - -2 p
2 2 2 2
-4 2 2
-5 -4
-6
-6

However, by restricting their domains, we can make them one-to-one functions:

y=sinHxL y=cosHxL y=tanHxL
1.0 1.0 6
4
0.5 0.5
2

p p p p p
- p - -2
2 -0.5 2 -0.5 2 2 2
-4
-1.0 -1.0 -6

y=cscHxL y=secHxL y=cotHxL

6
6
5 4
4
2 2
p p
p p p
2 -2 2 -2 p
2
-4 -4
-5
-6
-6

The inverses of these restricted functions are called the inverse trigonometric functions. They
are denoted by sin 1 , cos 1 , etc.
y=sin-1 HxL y=cos-1 HxL y=tan-1 HxL
p p
2 2

-1 1 -5 -4 -3 -2 -1 1 2 3 4 5

p
-
2 1

+ Alternate notation for the sin 1

function is arcsin, alternate notation for cos 1
is arccos,
and so on.

+ We will work only with the most common inverse trigonometric functions: sin 1 , cos 1 , and
1
tan .
The derivatives of inverse trigonometric functions are quite surprising in that their derivatives are
actually algebraic functions.

Ex. 5. We will find a formula for the derivative of g(x) = sin 1 (x).
(a) Write sin 1 (x) = ✓, so that sin(✓) = x, and draw a picture of a right triangle in which ✓ and
sin(✓) are labeled. Then label the remaining sides of the triangle.
0
(b) The derivative of g 1 (x) = sin(x) is g 1 (x) = cos(x). Use the diagram you drew in part (a)
0
to find an algebraic formula for g 1 g(x) .
(c) Use the Inverse Function Theorem to find the derivative of g(x) = sin 1 (x).

Solution:
(a)
opp.
Label one of the non-right angles as ✓. Since sin(✓) = , we will take the length of the
hyp.
hypotenuse to be 1 for simplicity, and label the side opposite ✓ as
x = sin(✓).

Then by the Pythagorean Theorem, the length of the adjacent side satisfies
adj.2 + x2 = 1,
so p
adj. = 1 x2 .

(b)
We have
g(x) = sin 1 (x) = ✓,
so
1 0
g g(x) = cos sin 1 (x) = cos(✓),
adj.
and from the fact that cos(✓) = it follows that
hyp.
0 p
g 1 g(x) = cos(✓) = 1 x2 .

(c)
1
g 0 (x) =
g 1 0 g(x)
1
=p .
1 x2
Ex. 6. Using the same technique as in the previous exercise, show that
d⇥ ⇤ 1
cos 1 (x) = p .
dx 1 x2

We can also find the derivatives of inverse functions by using implicit di↵erentiation.

Ex. 7. Find the derivative of tan 1 (x) without the Inverse Function Theorem.
Solution:
Set
1
✓ = tan x.
Then
tan(✓) = x,
and di↵erentiating the latter equation yields
d⇥ ⇤ d⇥ ⇤
tan(✓) = x
dx dx
d✓
sec2 (✓) · =1
dx
d✓ 1 1 1
= 2
= 2
= .
dx sec (✓) 1 + tan (✓) 1 + x2
3.7 EXERCISES
3.7 EXERCISES
EXERCISES
3.7
For
For the
the following
following exercises,
exercises, use
use the
the graph
graph of
of to
to 263.
263.
For the following exercises, use the graph of to 263.
a.
a. sketch
sketch the
the graph
graph of
of and
and
a. sketch the graph of and
Additional exercises
b. use part a. to estimate
b. use part a. to estimate
b.Ex.use8 part a. to estimate261, 262). Use the graph of y = f (x) to sketch the graph of y = f
(§3.7—#260, 1
(x).
260. 1 0
260. Then use the result to estimate f (1).
260.

(a)

For
For the
the following
following exercises,
exercises, use
use the
the functions
functions
For
find the following exercises, use the functions
find
find
a.
a. at
at and
and
a. at and
b.
b.
261.
261. b.
261.
c.
c. Then
Then use
use part
part b.
b. to
to find
find at
at
(b) c. Then use part b. to find at

264.
264.
264.
265.
265.
265.
266.
266.
266.
267.
267.
267.
For
For each
each of
of the
the following
following functions,
functions, find
find
262.
262. For each of the following functions, find
262.
(c) 268.
268.
268.

269.
269.
269.
270.
270.
270.
271.
271.
271.
272.
272.
272.
273.
273.
273.
For
For each
each of
of the
the given
given functions
functions
For each of the given functions
a. find the slope of the tangent line to its inver
 1 0
Ex. 9 (§3.7—#269). Let f (x) = x3 + 2x + 3. Find f (0).

2 1 0
Ex. 10 (§3.7—#271). Let f (x) = x , x < 0. Find f (1).
x

1 0
Ex. 11 (§3.7—#273). Let f (x) = tan(x) + 3x2 . Find f (0).
 Workbook Lesson 14
§3.9, Derivatives of Exponential and Logarithmic Functions
Last revised: 2021-02-18 12:45

Objectives
• Di↵erentiate exponential functions.
• Apply Logarithm Laws.
• Di↵erentiate logarithmic functions.
• Use logarithmic di↵erentiation to determine the derivative of a function.

Question. What does it mean to raise a number to an irrational exponent?

• When n = 1, 2, 3, . . . , we know 3n = (3)(3) · · · (3) by definition.
| {z }
n copies of 3
n n
• We also know that 3 = 1/3 for n = 1, 2, 3, . . . .
p
• Also by definition, 3p/q = q 3p when p and q are integers, provided that q 6= 0.
• At this point in your mathematical career, no one has ever defined for you the meaning of an
expression like 3⇡ (that is, an expression with an irrational exponent). We will do so in this
lesson.

Notation: Write
expb (x) = bx .

Assumption: (we’ll prove this assumption later) For any choice of positive real number b > 0, the
function expb (x) = bx is continuous.
• The function expb (x) = bx is called the exponential function with base b.
• Some authors require b 6= 1 in the definition of an exponential function, because the function
1x is a bit silly—its output is a constant, 1.
• For positive b > 1, the exponential function models exponential growth.
• For positive 0 < b < 1, the exponential function models exponential decay.
5
y = (1/10)x y = 10x
4

3
y = (1/2)x y = 2x
2

y = 1x
1

-1.0 -0.5 0.5 1.0

• The graphs suggest it is reasonable to assume that exponential functions are continuous.
(See applet on iCollege: “Graphs of exponential functions”)
+ Do not confuse the expression 2x with the expression x2 .

50 y = 2x

40

30

20 y = x2

10

-4 -2 2 4 6

Informal definition of the number e

The most common definition of the irrational number e ⇡ 2.7182 . . . is

1 n
lim 1 + n
.
n!1

However, our textbook doesn’t use this definition. We’ll follow the textbook’s approach and define
the number e informally, as follows.
Let’s consider the tangent line to the exponential function expb (x) = bx at the point (0, 1).
+ Notice that the point (0, 1) is on the graph of every exponential function, no matter what
the base b is, because for any positive number b we have b0 = 1.
3
m ≈ 1.0986

2 m ≈ 0.6931

y = 2x
y = 3x -1 1

-1

-2

-3

When the base is b = 2, the slope of the tangent line (dashed blue) is a little less than 1.
When the base is b = 3, the slope of the tangent line (dashed purple) is a little greater than 1.
It stands to reason that for some value of b between 2 and 3, the slope of the tangent at (0, 1) is
exactly 1.
3
m=1

y = ex
-1 1

-1

-2

-3

We’ll use this informal reasoning to define the number e. Toward the end of this lesson, we’ll revisit
the idea that
n
e = lim 1 + n1 .
n!1

Definition: The number e is the real number such that the tangent line to the exponential function
ex at the point (0, 1) has slope m = 1.
+ When the base is b = e, we omit the base and write

exp(x) = ex .

This definition of the number e only makes sense if we assume the following:
• There’s only one unique number that satisfies our definition of the number e.
• The tangent line at x = 0 exists—that is, the function ex is di↵erentiable at x = 0.

Has our reasoning so far seemed a little shaky? Well, it is. We’re making assumptions without
verifying that they’re true. But, as we said, we will give an alternative definition of e that doesn’t
rely on unproven assumptions toward the end of this lesson.
 Definition of the logarithmic functions

Recall:
1
• The inverse of a function f is a function, denoted by f , that “undoes” f .
1
• The fact that f “undoes” f is expressed by the Cancellation Formulas:

f 1 f (x) = x for all x in the domain of f ,

f f 1 (y) = y for all y in the range of f .

1
• The graph of the inverse function f is the mirror image of the graph of f reflected in the
line y = x.

Definition. The logarithmic function with base b, denoted by logb , is the inverse function of
the exponential function expb (x) = bx .
The Cancellation Formulas for an exponential function and its inverse are:
✓ ◆
logb expb (x) = logb bx = x for all real numbers x,

blogb (y) = y for all y > 0.

y = 2x y=x
4

(1,2)
2
y = log2 (x)
(0,1) (2,1)

(1,0)
-4 -2 2 4

-2

-4

Some special notation:

• The logarithmic function with base 10 is denoted by log and called the common logarithm.
• The logarithmic function with base e is denoted by ln and called the natural logarithm.
 y = expHxL y=x
4

H1,eL

He,1L
y = lnHxL

H0,1L

H1,0L
-4 -2 2 4

-2

Definition of bx for all real numbers x, and continuity of bx for any b > 0

-4
As mentioned above, you have probably never been taught the precise definition of an exponential
x
expression of the form b for x a real (possibly irrational) number. Let us now do so, using only
(1) the fact that exp(x) = ex and ln(x) = loge (x) are inverse functions, and (2) the rules of
exponents for rational exponents.
Theorem. For b > 0 and rational r,
br = er ln b .
Proof. By the second Cancellation Formula,
b = eln b (b > 0).
Thus
br = (eln b )r (r rational)
= er ln b .
We have just proven
br = er ln b for r rational. (*)

We now extend formula (*) by defining

bx = ex ln b for all x in R.
The exponential function bx has now been properly defined for any base b > 0. The fact that bx is
continuous for any b > 0 now follows by the Composite Function Theorem (Lesson 4) from the fact
(which we’ll assume without proof) that the natural exponential function ex and and the function
h(x) = ax (a = const) are both continuous: bx = exp (ln b)x .
 Derivatives of the exponential and logarithmic functions

So why do we care about e?

Here are three reasons:
• The exponential function ex is frequently used to model real-world situations in the sciences
(as you may have seen in earlier math classes).
• We can use the function ex to define bx for any b > 0 and any real number x (as seen on the
previous page).
• The formula for the derivative of ex is super easy.

The derivative of the exponential function with base e,

ex ,

is itself:
d ⇥ x⇤
e = ex .
dx

The derivative of the exponential function with any other base is a bit more complicated:
d ⇥ x⇤
b = bx ln(b).
dx

We can use implicit di↵erentiation to find the derivative of ln(x):

y = ln(x)
ey = x (Cancellation Formula)
d ⇥ y⇤ d⇥ ⇤
e = x
dx dx
dy
ey =1
dx
dy
eln(x) =1 (Substitution: y = ln(x))
dx
dy
x =1 (Cancellation Formula)
dx
dy 1
=
dx x
d ⇥ ⇤ 1
ln(x) =
dx x
We find the derivative of bx similarly:
y = logb (x)
by = x (Cancellation Formula)
d ⇥ y⇤ d⇥ ⇤
b = x
dx dx
dy
by ln(b) =1
dx
dy
blogb (x) ln(b) =1 (Substitution: y = logb (x))
dx
dy
x ln(b) =1 (Cancellation Formula)
dx
dy 1
=
dx x ln(b)
d⇥ ⇤ 1
logb (x) =
dx x ln(b)

To summarize:
d ⇥ x⇤ d⇥ ⇤ 1
e = ex ln(x) = for x > 0
dx dx x
d ⇥ x⇤ d⇥ ⇤ 1
b = bx ln(b) logb (x) = for x > 0
dx dx x ln(b)
(Note that the domain of the logarithmic functions is (0, 1).)

Exercises
x3 +1
Ex. 1. Di↵erentiate y = e .
Solution:
d h x3 +1 i (C.R.) x3 +1 d ⇥ 3 ⇤
e = e · x +1
dx dx
3 +1
= 3x2 ex

Ex. 2. Di↵erentiate y = ln(x3 + 1).

Solution:
d ⇥ ⇤ (C.R.) 1 d ⇥ ⇤
ln(x3 + 1) = 3 · ln(x3 + 1)
dx x + 1 dx
3x2
=
x3 + 1
Ex. 3. Di↵erentiate h(x) = xe2x .
Solution:

d ⇥ 2x ⇤ ⇥ ⇤0
xe = [x]0 e2x + x · e2x
dx
⇥
= e2x + x · e2x 2x]0

= e2x + 2xe2x

Ex. 4. Di↵erentiate f (x) = x ln(2x).

Solution:

d ⇥ ⇤0
[x ln(2x)] = [x]0 ln(2x) + x · ln(2x)
dx
1⇥ 0
= ln(2x) + x · 2x]
2x
1
= ln(2x) + x · · (2)
2x
= ln(2x) + 1

Ex. 5. Suppose u(x) is a di↵erentiable function. Prove the formula:

d⇥ ⇤ 1
ln u(x) = · u0 (x)
dx u(x)

Solution:

d⇥ ⇤ 1 d⇥ ⇤ 1
ln u(x) = · u(x) = · u0 (x).
dx u(x) dx u(x)
 
d x+1
Ex. 6. Compute ln p .
dx x 2
Solution:
We will apply the formula
d⇥ ⇤ 1
ln u(x) = · u0 (x),
dx u(x)
which was proven in the previous exercise.
x+1
u(x) = p .
x 2

d⇥ ⇤
u0 (x) = (x + 1)(x 2) 1/2
dx
d⇥ ⇤ d⇥ ⇤
= (x + 1) · (x 2) 1/2 + (x + 1) · (x 2) 1/2
dx dx
1 3/2 1/2
= (x + 1) · 2
(x 2) + 1 · (x 2)
1
= (x + 1)(x 2) 3/2 + 1 · (x 2) 1/2
.
2
x+1 1
= p +p .
2 (x 2)3 x 2

p
1 x 2
= .
u(x) x+1

Applying the formula now yields

d ⇥ ⇤ 1
ln u(x) = · u0 (x)
dx u(x)
✓ ◆
(x 2)1/2 1
= · (x + 1)(x 2) 3/2 + (x 2) 1/2
x+1 2
✓ ◆
1 1 1
= · (x + 1)(x 2) + 1
x+1 2
1 1
= (x 2) 1 +
2 x+1
1 1
= +
2(x 2) x + 1
(x + 1) 2(x 2)
= +
2(x 2)(x + 1) 2(x 2)(x + 1)
x 5
= .
2(x 2)(x + 1)
 
d
Ex. 7. Find ln |x| .
dx
Solution:

d 1
We will apply the formula ln u(x) = · u0 (x) .
dx u(x)

Note that |x| is a piecewise defined function that is not di↵erentiable at x = 0.

y = »x»
We will therefore have to calculate the derivative of |x| for x 6= 0 piecewise.
⇢
x if x 0,
u(x) = |x| =
x if x < 0.
8
< 1 if x 0, x
u0 (x) = undefined if x = 0,
:
1 if x < 0.
8
> 1
>
> if x 0,
>
< x
1
= undefined if x = 0,
u(x) >>
>
: 1
>
if x < 0.
x

Applying the formula, we get

8
> 1
>
> ·1 if x 0,
 >
< x
d 1
ln |x| = · u0 (x) = undefined if x = 0,
dx u(x) >
>
>
> 1
: · ( 1) if x < 0.
x
8
< 1 if x 6= 0,
= x
:
undefined if x = 0.
 Logarithm Laws
Recall the Logarithm Laws:
For x, y > 0 and r a real number,
(1) logb (xy) = logb (x) + logb (y)
(2) logb (x/y) = logb (x) logb (y)
(3) logb (xr ) = r logb (x)

The Logarithm Laws will be needed for our next technique, called logarithmic di↵erentiation.

(x2 + 5)4 sin x

Ex. 8. Write ln as a sum of logarithms. Cite the Logarithm Law you are using at
x3 + 1
each step.
Solution:

(x2 + 5)4 sin x (2)

ln = ln (x2 + 5)4 sin x ln(x3 + 1)
x3 + 1
(1)
= ln(x2 + 5)4 + ln(sin x) ln(x3 + 1)
(3)
= 4 ln(x2 + 5) + ln(sin x) ln(x3 + 1).

Ex. 9. Rewrite ln(a) + 12 ln(b) as a single logarithm. Cite the Logarithm Law you are using at each
step.
Solution:

(3) p
ln(a) + 12 ln(b) = ln(a) + ln b
(1) p
= ln a b
 e as a limit
(1) (2)
Theorem. e = lim (1 + x)1/x = lim (1 + n1 )n .
x!0 n!1

1
Proof. Let f (x) = ln(x). Then f 0 (1) = = 1.
1

f (1 + h) f (1)
1 = f 0 (1) = lim
h!0 h
ln(1 + h) ln(1)
= lim
h!0 h
1
= lim ln(1 + x)
x!0 x
⇥ ⇤1/x
= lim ln(1 + x)
x!0

Now
⇣ ⇥ ⇤1/x ⌘
e = exp(1) = exp lim ln(1 + x)
⇥ x!0 ⇤
= lim exp ln(1 + x)1/x
x!0
= lim (1 + x)1/x ,
x!0

which is (1).
For (2), take n = 1/x. As x ! 0+ , we have n ! 1, so
1 n
lim (1 + x)1/x = lim 1 + n
.
x!0 n!1

Many authors use the equation

1 n
e = lim 1 + n
n!1
as the definition of the number e. Then there is no need to prove the above Theorem, and:
• the natural logarithmic function
ln(x) = loge (x)
can be defined as the inverse of the function
exp(x) = ex ,

• exponential functions with other bases

expb (x) = bx
are defined by the formula
bx = eb ln(x) , and

• logarithmic functions with other bases can be defined as their inverses,

logb (x) = expb 1 (x).
 Logarithmic di↵erentiation
Consider the problem of finding the derivative of
p
x3/4 x2 + 1
y= .
(3x + 2)5

Although it would be tedious, we could certainly do this using only the Power, Product, Quotient,
and Chain Rules.

But there’s a better way.

+ The derivative of a function f (x) that only involves products, quotients, and powers (which
p
includes roots n x = x1/n ) can be found using the technique of logarithmic di↵erentiation.
p
dy x3/4 x2 + 1
Ex. 10. Find if y = .
dx (3x + 2)5

Solution (General steps for logarithmic di↵erentiation given in red).

(1) Take ln of both sides of the equation y = f (x), and apply the Logarithm Laws.

ln y = 34 ln x + 12 ln(x2 + 1) 5 ln(3x + 2).

dy
(2) Find using implicit di↵erentiation.
dx
 
d d 3 1
ln y = ln x + ln(x2 + 1) 5 ln(3x + 2)
dx dx 4 2
✓ ◆ ✓ ◆ ✓ ◆
1 dy 3 1 1 1 1
= + · 2x 5 ·3
y dx 4 x 2 x2 + 1 3x + 2
✓ ◆
dy 3 x 15
=y +
dx 4x x2 + 1 3x + 2

(3) Substitute y if a formula for y = y(x) was given.

p ✓ ◆
dy x3/4 x2 + 1 3 x 15
= 5
+ 2
dx (3x + 2) 4x x + 1 3x + 2
 Additional exercises

Ex. 11 (§3.9—#331, 333, 337, 339, 341, 343). Di↵erentiate.

p
• f (x) = x2 ex • f (x) = 24x + 4x2 • f (x) = ln 5x 7
x3 ln(x) ⇡ x
• f (x) = e • f (x) = x · ⇡ • f (x) = log(sec x)

Ex. 12. Find the equation of the tangent line to the curve

y = x4 + 2ex

at the point (0, 2).

Ex. 13. Find the equation of the tangent line to the curve

y = ln(x)

at the point (e, 1).

d ⇥ p ⇤ 1
Ex. 14. Show that ln x + x2 + 1 = p .
dx 2
x +1

Ex. 15. Use logarithmic di↵erentiation to find the derivative of y = xx .

 r
x 1
Ex. 16. Use logarithmic di↵erentiation to find the derivative of y = .
x4 + 1

Ex. 17 (§3.9—#357a). At which points on the graph of

y = x1/x (x > 0)

is the tangent line horizontal?

Ex. 18 (§3.9—#356). Find the equation of the tangent line to the curve

x3 x ln(y) + y 3 = 2x + 5
dy
at the point where x = 2. (Hint: Use implicit di↵erentiation to find .)
dx
2.0

1.5

1.0

0.5

-2 -1 1 2

-0.5

-1.0
 Workbook Lesson 15
§4.1, Related Rates
Last revised: 2021-06-03 14:09

Objectives
• Express changing quantities in terms of derivatives.
• Find relationships among the derivatives in a given problem.
• Use the Chain Rule to find the rate of change of one quantity that depends on the rate of change of other
quantities.

Strategy for Solving a Related Rates Problem

1. Write a Legend that assigns symbols to all variables involved in the problem. The Legend
should also state the meaning of each variable. Draw a picture, if applicable.
2. Identify the independent variable. Then state, in terms of the variables, the information that is
given and the rate to be determined.
3. Find an equation relating the variables in the Legend.
4. Di↵erentiate both sides of the equation found in Step 3 with respect to the independent variable.
5. Substitute all known values from Step 2 into the equation from Step 4, then solve for the
unknown rate of change.

Worked example
Ex. 1. Consider a 10 foot ladder that is leaning against a wall. If the bottom of the ladder slides
away from the wall at a rate of 1 ft/sec, how fast is the top of the ladder sliding down the wall
when the bottom of the ladder is 6 ft from the wall?

Recall that the velocity of a object that is moving over time t is given by the formula
d⇥ ⇤
velocity = position function .
dt

So, if y is the distance from the top of the ladder to the ground, then the speed at which the
dy
top of the ladder slides down the wall is .
dt
dx
Similarly, the speed at which the bottom of the ladder slides away from the wall is .
dt
Step 1: Write a Legend. Draw a picture, if applicable.
Legend:
x = distance from bottom of ladder to wall
y = distance from top of ladder to ground
t = time 10 y
dy
= velocity of top of ladder, sliding down wall
dt
dx
= velocity of bottom of ladder, sliding away from wall x
dt
Step 2: Identify the independent variable. Then state, in terms of the variables, the information
that is given and the rate to be determined.
• The independent variable is time t.
dy
• The desired rate of change is .
dt
• The given information is:

length of ladder = 10
dx
= velocity of bottom of ladder = 1
dt
x = distance of bottom of ladder from wall = 6

Step 3: Find an equation relating the variables in the Legend.

By Pythagorean Theorem,
x2 + y 2 = 102 . (?)

Step 4: Di↵erentiate both sides of the equation found in Step 3 with respect to the independent
variable.
d⇥ 2 ⇤ d ⇥ 2⇤
x + y2 = 10
dt dt
dx dy
2x + 2y =0 (Chain Rule)
dt dt

Step 5: Substitute all known values from Step 2 into the equation from Step 4, then solve for the
unknown rate of change.

dx dy
2x + 2y =0
dt x=6
dt x=6

dy
2(6)(1) + 2y =0
dt
dy
y = 6
dt
What is y when x = 6? We can use equation (?) to find out:

36 + y 2 = 100
y=8
dy
Now we can solve for the desired rate of change, :
dt
dy
8 = 6
dt
dy 6 3
= =
dt 8 4

3
The top of the ladder is sliding down the wall at a rate of 9 inches (= 4
feet) per second.
An activity is presented in the remainder of this document. It may be done in
groups of students, or on your own.
• On the next page, you’ll find eight exercises. (We’ve already done Ex. 1.)
• For each exercise, write a Legend (and draw a picture, if applicable).
• The subsequent page gives the instructor’s Legends for each problem.
– Do not look at the Legends until you’ve made a real e↵ort to come up
with your own.
– Compare your Legend with the instructor’s Legend.
– It’s okay if they’re not exactly the same—we may use di↵erent words and
variable names to express the same relationships.
• The remainder of this document presents the instructor’s solutions for each
of the exercises.
– Do not look at the instructor’s solution until you’ve made your best e↵ort
at figuring out your own solution.

Additional practice exercises can be found in the Final Exam Review on iCollege
(under Section 4.1).
 Related rates—Exercises
Ex. 1. Consider a 10 foot ladder that is leaning against a wall. If the bottom of the ladder slides away
from the wall at a rate of 1 ft/sec, how fast is the top of the ladder sliding down the wall when the bottom
of the ladder is 6 ft from the wall?
Ex. 2. Boyle’s Law states that when a sample of gas is compressed at a constant temperature, the pressure
P and volume V satisfy the equation P V = C, where C is a constant. Suppose that a certain instant
the volume is 600 cm3 , the pressure is 150 kilopascals (kPa), and the pressure is increasing at a rate of
20 kPa/min. At what rate is the volume decreasing at this instant?
Ex. 3. The length of a rectangle is increasing at a rate of 8 cm/s. Its width is increasing at a rate
of 3 cm/s. When the length is 20 cm and the width is 10 cm, how fast is the area of the rectangle
increasing?
Ex. 4. A baseball diamond is a square with side 90 ft. A batter hits the ball and runs toward first base
with a speed of 24 ft/s.

(a) At what rate is his distance to second base decreasing when he is halfway to first base?
(b) At what rate is his distance to third base increasing when he is halfway to first base?

Ex. 5. In an electrical circuit, two resistors with resistances R1 and R2 are connected in parallel as shown.

The total resistance R in ohms (⌦) satisfies the equation

1 1 1
= + .
R R1 R2
Suppose R1 and R2 are increasing at rates of 0.3 ⌦/sec and 0.2 ⌦/sec respectively. How fast is R
changing when R1 = 80 ⌦ and R2 = 100 ⌦?
Ex. 6. A plane flying at a constant speed of 300 km/h passes over a ground radar station at an altitude
of 1 km and climbs at an angle of 30 . At what rate is the distance from the plane to the radar station
increasing one minute later? (Hint: Draw a picture, then use the Law of Cosines.)
Ex. 7. A sprinter runs away from a tall stadium light. The light is 30 feet from the ground. If the sprinter
is 6 feet tall, and runs at 24 feet per second, at what rate does her shadow grow longer? (Hint: Use
similar triangles.)
Ex. 8. Water is leaking out of an inverted (i.e. upside-down) conical tank at a rate of 10, 000 cm3 /min.
At the same time, water is pumped into the tank at a constant rate. The tank has height 6 m and the
diameter at the top is 4 m. When the height of the water is 2 m, the water level is rising at a rate of 20
cm/min. Find the rate at which water is being pumped into the tank. (This exercise requires the formula
for the volume of a right circular cone, V = 13 ⇥ (area of base) ⇥ (height) = 13 ⇡r2 h.)
 Related rates—Legends

P = pressure dP
= change in pressure
Ex. 2. dt
V = volume dV
= change in volume
t = time (independent variable) dt

` = length
Ex. 3. w = width
A = area
t = time (independent variable)

x = runner’s distance to first base

Ex. 4. y = runner’s distance to second base
z = runner’s distance to third base
t = time (independent variable)

R1 = resistance in resistor #1
Ex. 5. R2 = resistance in resistor #2
R = total resistance
t = time (independent variable)

y = initial height of plane (constant)

Ex. 6. x = distance traveled by the plane (variable)
D = distance from plane to radar station (variable)
t = time (independent variable) 1 km
Ex. 7.

H = height of light (constant)

h = height of sprinter (constant)
x = distance from sprinter to base of light (increasing)
s = length of shadow (increasing)
H t = time (independent variable)

h
x s

Ex. 8.
C: rate at which water is pumped into tank (constant)
V: volume of water in the tank (increasing)
t: time (independent variable)
dV
: rate of change in volume of water in the tank
dt
h = AL : height of water (rising)
r = LM : radius of surface of water (increasing)
6 = AB : height of cone (constant)
2 = BC : radius of cone’s base (constant)
 Related rates—Solutions
Ex. 2. Boyle’s Law states that when a sample of gas is compressed at a constant temperature, the
pressure P and volume V satisfy the equation P V = C, where C is a constant. Suppose that a
certain instant the volume is 600 cm3 , the pressure is 150 kilopascals (kPa), and the pressure is
increasing at a rate of 20 kPa/min. At what rate is the volume decreasing at this instant?

P = pressure dP
= change in pressure
Step 1 (Legend): dt
V = volume dV
= change in volume
t = time (independent variable) dt

Step 2 (State what’s known and dP

= 20
what’s asked for, in terms of dt
the variables): dV
= ? when V = 600 and P = 150
dt
Step 3 (Equation relating the variables): PV = C (C = const)

Step 4 (Di↵erentiate with respect to dP dV

V +P = 0 (Product & Chain Rules)
the independent variable): dt dt

Step 5 (Substitute what’s known and dV 600

find the rate that was asked for): = · 20 = 80
dt V =600 150
P =150

The volume is decreasing at 80 cm3 /min.

Ex. 3. The length of a rectangle is increasing at a rate of 8 cm/s. Its width is increasing at a rate
of 3 cm/s. When the length is 20 cm and the width is 10 cm, how fast is the area of the rectangle
increasing?

` = length
Step 1 (Legend): w = width
A = area
t = time (independent variable)

d`
=8
Step 2 (State what’s known and dt
what’s asked for, in terms of the dw
=3
variables): dt
dA
= ? when ` = 20 and w = 10
dt
Step 3 (Equation relating the
A = `w
variables):

dA d
= [`w]
Step 4 (Di↵erentiate with respect to dt dt
the independent variable): dA d` dw
=w +`
dt dt dt

dA d` dw
= 10 + 20 ·
Step 5 (Substitute what’s known and dt `=20 dt `=20 dt `=20
w=10 w=10 w=10
find the rate that was asked for):
dA
= 20(3) + 3(8) = 140
dt `=20
w=10

The area of the rectangle is increasing at 140 cm per second.

Ex. 4. A baseball diamond is a square with side 90 ft. (See figure below.) A batter hits the ball
and, starting at home plate, runs toward first base with a speed of 24 ft/s.
(a) At what rate is his distance to second base decreasing when he is halfway to first base?
(b) At what rate is his distance to third base increasing when he is halfway to first base?

x = runner’s distance to first base

Legend: y = runner’s distance to second base
z = runner’s distance to third base
t = time (independent variable)
(a)
x2 + 902 = y 2
2xx0 = 2yy 0
xx0
y0 =
y
p p p
y x=45 = 452 + 902 = 452 + 4 · 452 = 45 5
45( 24) 24
y 0 x=45 = p = p
45 5 5

24
The runner’s distance from second base is decreasing at a rate of p
5
ft/s.

(b) For this part of the problem, we’ll let w = runner’s distance from home plate.
w2 + 902 = z 2
ww0
z0 =
z
p
z w=45 = 45 5
45(24) 24
y 0 w=45 = p =p
45 5 5
24
The runner’s distance from third base is increasing at a rate of p
5
ft/s.
Ex. 5. If two resistors with resistances R1 and R2 are connected in parallel as shown,

then the total resistance R in ohms (⌦) satisfies the equation

1 1 1
= + .
R R1 R2
Suppose R1 and R2 are increasing at rates of 0.3 ⌦/sec and 0.2 ⌦/sec respectively. How fast is R
changing when R1 = 80 ⌦ and R2 = 100 ⌦?

R1 = resistance in resistor #1
Legend: R2 = resistance in resistor #2
R = total resistance
t = time (independent variable)
dR1 dR2 dR
We find the relation between R1 , R2 , , , and .
dt dt dt
d ⇥ 1⇤ d ⇥ ⇤
R = R1 1 + R2 1
dt dt

2 dR 2 dR1 2 dR2
R = R1 R2
dt dt dt
dR
Solve for by multiplying both sides in the previous equation by R2 .
dt
✓ ◆
dR 2 1 dR1 1 dR2
=R + 2 (***)
dt R12 dt R2 dt
When R1 = 80 and R2 = 100, the value of R is
1 1 1 9
= + = .
R 80 100 400

Now substitute R1 = 80, R2 = 100, dR1 /dt = 0.3, and dR2 /dt = 0.2 into (***).
✓ ◆2 ✓ ◆
dR 400 1 3 1 2 107
= 2
+ 2
=
dt 9 80 10 100 10 810

107
R is changing at ⇡ 0.132099 ⌦/sec.
810
Ex. 6. A plane flying at a constant speed of 300 km/h passes over a ground radar station at an
altitude of 1 km and climbs at an angle of 30 . At what rate is the distance from the plane to the
radar station increasing one minute later?

Legend:
x = distance traveled by the plane (variable)
y = initial height of plane (constant)
D = distance from plane to radar station (variable)
t = time (independent variable)

D
1 km

By the Law of Cosines, writing ✓ = (angle opposite D) = 120 = 2⇡/3,

D 2 = x2 + y 2 2xy cos ✓
D 2 = x2 + 1 2x 1
2
(y = 1, ✓ = 2⇡/3)
D 2 = x2 + x + 1

By implicit di↵erentiation,
dD dx dx
2D = 2x +
dt dt dt
dD 2x + 1 dx
=
dt 2D dt
After 1 minute,
300 km
x= = 5 km,
60 min
so p p
D = D(5) = 52 + 5 + 1 = 31,
giving
dD 2(5) + 1 3300
= p · 300 = p ⇡ 296.349 km/h .
dt 2 31 2 31
Ex. 7. A sprinter runs away from a tall stadium light. The light is 30 feet from the ground. If
the sprinter is 6 feet tall, and runs at 24 feet per second, at what rate does her shadow grow
longer?

Legend:
H = height of light (constant)
h = height of sprinter (constant)
x = distance from sprinter to base of light (increasing)
H s = length of shadow (increasing)
t = time (independent variable)

h
x s

By similar triangles,
H h h
= .
x s
We solve for s:
h
s= x.
H h
Then
ds h
= x0
dt H h

and when h = 6, H = 30, and x0 = 24, we get

ds 6
= · 24 = 6 ft/sec .
dt 30 6
Ex. 8. Water is leaking out of an inverted (i.e. upside-down) conical tank at a rate of 10, 000
cm3 /min. At the same time, water is pumped into the tank at a constant rate. The tank has
height 6 m and the diameter at the top is 4 m. When the height of the water is 2 m, the water
level is rising at a rate of 20 cm/min. Find the rate at which water is being pumped into the
tank.
Given the points labeled as shown, we set up the legend.

Legend:
C: rate at which water is pumped into tank (constant)
V: volume of water in the tank (increasing) h = AL : height of water (rising)
t: time (independent variable) r = LM : radius of surface of water (increasing)
6 = AB : height of cone (constant)
dV
: rate of change in volume of water in the tank 2 = BC : radius of cone’s base (constant)
dt

We have
(rate of change in volume of water) = (speed of inflow) (speed of outflow)
that is,
dV
=C 10, 000.
dt
The formula for the volume of a cone tells us
1 1 1
V = (area of base)(height) = ⇡(LM )2 (AL) = ⇡r2 h.
3 3 3

Is r a function of h? Yes: by similar triangles,

LM BC
= ,
AL AB
so
BC 1
r = LM = AL = h.
AB 3
Thus
1 1 dV 1 dh
V = ⇡r2 h = 3 ⇡h3 , = ⇡h2 .
3 3 dt 9 dt
Noting that h = 2 m = 200 cm (must make length units of h and dh/dt match),
dV 1 800, 000⇡
h=200
= ⇡(200)2 20 = .
dt 9 9
dh
dt =20

800, 000⇡
Water is being pumped into the tank at a rate of C = 10, 000 + ⇡ 289, 253 cm3 /min.
9
 Additional exercises

Ex. 9 (§4.1—#17). The volume of a cube decreases at a rate of 10 m3/s. Find the rate at which
the side of the cube changes when the side of the cube is 2 m.

Ex. 10 (§4.1—#19). Recall that, in general, the surface area of a sphere with radius r is

A = 4⇡r2 .

The radius of a sphere decreases at a rate of 3 m/sec. Find the rate at which the surface area
decreases when the radius is 10 m.

Ex. 13 (§4.1—#11). A 6-foot-tall person walks away from a 10-ft. lamppost at a constant rate
of 3 ft./sec. What is the rate that the tip of the shadow moves away from the pole when the person
is 10 ft. away from the pole?

Ex. 11 (§4.1—#30). A trough has ends shaped like isosceles triangles with width 3 m and height
4 m. The trough is 10 m long. Water is being pumped into the trough at a rate of 5 m3 /min. At
what rate does the height of the water change when the water is 1 m deep?
Ex. 12 (§4.1—#7). Two airplanes are flying in the air at the same height: airplane A is flying
east at 250 mi/h and airplane B is flying north at 300 mi/h. If they are both heading to the same
airport, located 30 miles east of airplane A and 40 miles north of airplane B, at what rate is the
distance between the airplanes changing?

Ex. 14 (§4.1—#37). You are stationary on the ground and are watching a bird fly horizontally
at a rate of 10 m/sec. The bird is located 40 m above your head. How fast does the angle of
elevation change when the horizontal distance between you and the bird is 9 m?

Ex. 14 (§4.1—#39). A lighthouse (L) is on an island 4 mi away from the closest point, P , on the
beach (see image). If the lighthouse light rotates clockwise at a constant rate of 10 revolutions/min,
how fast does the beam of light move across the beach 2 mi away from the closest point on the
beach?
 Workbook Lesson 16
§4.2, Linear approximations and di↵erentials
Last revised: 2020-09-29 12:44

Objectives
• Describe the linear approximation to a function at a point.
• Write the linearization of a given function.
• Draw a graph that illustrates the use of di↵erentials to approximate the change in a quantity.
• Calculate the relative error and percentage error in using a di↵erential approximation.

Linear approximation to a di↵erentiable function

Linear functions are simpler and easier to work with than nonlinear functions. It therefore makes
sense to want to replace a nonlinear function by a linear approximation.
For a differentiable function f , the linear approximation is guaranteed to be a good approxi-
mation locally —that is, for x near some fixed input value a.
What exactly does this mean? �(�)= �
1.5

� (�) 1.0
• The linear approximation to a function f (x)
at the point x = a is given by the 0.5

tangent line to the graph of y = f (x) at

a, f (a) . �
-0.5 0.5 1.0 1.5 2.0
�
-0.5

• Suppose we need our approximation to be

within a certain tolerance—say, " = 0.05—
�(�)= �
of f (x). 1.5

� (�) 1.0

• For x “sufficiently near” a, the linear 0.5

approximation (height of the tangent line)

�
is within " of the value of f (height of the -0.5 0.5 1.0 1.5 2.0
�
graph of y = f (x)). -0.5

• How near to a must x be for the linear �(�)= �

1.5
approximation to be within the specified
tolerance "? That depends on the tangent 1.0

point. That is, if we change a, we may need

0.5
x to be closer to a in order for the linear ap- � (�)
proximation to be “close enough.” �
-0.5 0.5 1.0 1.5 2.0
�
-0.5

(See applet on iCollege: “Linear approximation”)

Recall:
The equation of the tangent line to the graph of f at the point a, f (a) is:

y = f 0 (a)(x a) + f (a)

Let’s rewrite this in function notation, and call it the linear approximation to (or linearization,
or tangent line approximation of) f near a:
L(x) = f 0 (a)(x a) + f (a)

When we say that “the linear approximation is close to f for x near a,” what we are saying is
that
|f (x) L(x)| < "
for all x sufficiently near a.

Ex. 1.
p
(a) Find the linear approximation L(x) to the function f (x) = x near 1.
(That is, find the equation of the tangent line to the graph of f at 1, f (1) and write it in
function notation L(x) = · · · .)
p
(b) Using a calculator to evaluate L(9.1), approximate the value of f (9.1) = 9.1.

You may say, why are we bothering p with this linear approximation stu↵, if in the end we’re just
going to use a calculator to evaluate 9.1?
That’s a fair question. The answer is that, when we are dealing with a complicated function
f , it is often the case that evaluating the linear approximation L is faster, easier, and yields an
approximation that is close enough for practical purposes to the true value.
Here are some examples of situations in which a linear approximation is used instead of an exact
calculation:
• In computer animation, when the position of many moving objects must be evaluated many
times per second (nonlinear functions require faster, more expensive graphics cards and pro-
cessors)
• In engineering, when an irregular shape can be approximated by a flat shape (after all, there
are no perfectly smooth surfaces in the real world)
• In physics, when a theoretical calculation by hand becomes much, much easier if a nonlinear
function is replaced by a linear function (e.g., sin(x) ⇡ x near x = 0)
• In statistics, when a simple description of a general trend is desired (a line is easier to
intuitively understand than a complicated curve)
• In economics, when an observed trend is “jittery” due to frequent fluctuations (think of the
stock market) and a short-term prediction is sought
 The di↵erential of a function
CT
Recall: In the following figure, the slope of the orange line is .
AC
5

4 T

2 A C

0
0 1 2 3 4 5

Consider the di↵erentiable function y = f (x) whose graph is shown below.

Starting at any point A = (x, y) on the graph, let x increase (or decrease) by adding a small number
x 6= 0 (positive or negative) to x. What are the coordinates of the point B whose horizontal
coordinate is x larger than x?
Let y = CB be the corresponding change in y. That is, y = f (x + x) f (x).
5 B

4 T
Dy=CB
3 dy = CT

2 A=Hx,yL C
Dx
1

0
0 1 2 3 4 5

!
The slope of the tangent line AT to f at A = (x, y) is
CT
= f 0 (x).
x
!
Since, near x, the graph of y = f (x) is close to the tangent line AT of f at x, we see that when
x is small, we have
⇣ ⌘ ⇣ ⌘
actual change in y = f (x) = y ⇡ CT = change in height of tangent line of f at x ,

so
y ⇡ f 0 (x) · x. (?)

def
Definition. The di↵erential of the function y = f (x) is: dy = f 0 (x) · x.
+ Since the di↵erential of the function y = x is
dx = 1 · x= x,
we often write the definition of the di↵erential as
dy = f 0 (x) dx.

+ It should be kept in mind that the variable dy depends on (the independent variables) x and
dx. So dy = dy(x, dx), that is, dy is really a function of two variables.

Ex. 2.
• d(sin x) = cos(x) x = cos(x) dx.
• If y = 6x2 + 3, then dy = 12x dx.
• If A(r) = ⇡r2 , then dA = 2⇡r dr.

dy
+ Until this point, we have not regarded the symbol
dx
as a fraction. Now that we have defined
dy
the di↵erential, it is clear that we can think of = f 0 (x) as an ordinary fraction.
dx

Ex. 3. Compute approximately the volume of metal in a hollow spherical shell of thickness 0.05
in., with inside radius 5 in.
Solution. The volume of the metal in the shell is the amount by which the volume of a sphere
increases when its radius changes from 5 to 5.05 in.
Using the formula for the volume of a sphere,
V = 43 ⇡r3 ,
we find that
dV = 4⇡r2 dr.

Taking r = 5 and dr = 0.05, we find

dV = 4⇡(25)(0.05) ⇡ 4 ⇥ 3.14 ⇥ 25 ⇥ 0.05 = 15.7 in3

The exact value of the volume of the shell is

Z 5.05
dV = V (5.05) V (5) = 15.865566418484217 . . .
5
Z
cubic inches. (The meaning of the symbol will be explained in Chapter 5—for now, let’s just
accept this calculation on faith.) The error in our estimate was < 0.17 in3 .
Ex. 4. Approximate (1.98)5 by hand.
Solution.
Take
y = x5 .
Then
dy = 5x4 dx.
Taking x = 2 and dx = 0.02, we find
80 8
dy = 5(16)( 0.02) = = = 1.6.
50 5
This means that y = x5 decreases by 1.6 when x decreases from 2 to 1.98. Hence
(1.98)5 ⇡ 32 1.6 = 30.4 .
The actual value is 30.431681596799997 . . . .

Calculating the amount of error

Consider a function f with an input that is a measured quantity.
Suppose the exact value of the measured quantity is a, but the measured value is a + dx. We say
the measurement error is dx (or x).
As a result of the error in measurement, an error occurs in the calculated quantity f (x). This type
of error is known as a propagated error and is given by
y = f (a + dx) f (a).

Since all measurements are prone to some degree of error, we do not know the exact value of a
measured quantity, so we cannot calculate the propagated error exactly.
However, given an estimate of the accuracy of a measurement, we can use di↵erentials to approx-
imate the propagated error y. Specifically, if f is di↵erentiable at a, then the propagated error
is
y ⇡ dy = f 0 (a) dx.

We do not know what a is—we only know the measured value a + dx. However, provided that
a + dx ⇡ a (that is, the measurement error dx is small), we have
y ⇡ dy ⇡ f 0 (a + dx) dx.

Ex. 5. Suppose the side length of a cube is measured to be 5 cm with an accuracy of 0.1 cm.
(a) Use di↵erentials to estimate the error in the computed volume of the cube.
(b) Compute the volume of the cube if the side length is 4.9 cm to compare the estimated error
with the actual potential error.
 Relative error and percentage error
The measurement error dx = x and the propagated error y are absolute errors. We are
typically interested in the size of an error relative to the size of the quantity being measured or
calculated.
q
In general, if a measured quantity q has an absolute error q, we define the relative error as ,
q
where q is the quantity’s true value.
The percentage error is the relative error expressed as a percentage.
• For example, if we measure the height of a ladder to be 63 in. when the actual height is 62
in., the absolute error is 63 62 = 1 in., but the relative error is
1
= 0.016,
62
or 1.6%.
• By comparison, if we measure the width of a piece of cardboard to be 8.25 in. when the
1
actual width is 8 in., our absolute error is in., whereas the relative error is
4
1
0.258 = ,
32
or 3.1%.
• Therefore, the percentage error in the measurement of the cardboard is larger, even though
0.25 in. is less than 1 in.

Ex. 6. An astronaut using a camera measures the radius of Earth as 4000 mi with an error of ±80
mi. Use di↵erentials to estimate the relative and percentage error of using this radius measurement
to calculate the volume of Earth, assuming the planet is a perfect sphere.
Solution:
If the measurement of the radius is accurate to within ±80, we have

80  dr  80.

We know from a previous exercise that the di↵erential of the volume V of a sphere is

dV = 4⇡r2 dr.

Using the measured radius of 4000 mi, we can estimate bounds on the propagated error dV :

4⇡(4000)2 (80)  dV  4⇡(4000)2 (80)

dV
To estimate the relative error , use the measured radius r = 4000 mi. to estimate V :
V
4
V ⇡ ⇡(4000)3 .
3
 Therefore,
4⇡(4000)2 (80) dV 4⇡(4000)2 (80)
.06 = 4   4 = .06.
3
⇡(4000)3 V 3
⇡(4000)3
The relative error is .06 and the percentage error is 6%.

Additional exercises

1
Ex. 7 (§4.2—#51). Find the linear approximation L(x) to f (x) = at a = 2.
x

Ex. 8 (§4.2—#53). Find the linear approximation L(x) of f (x) = sin(x) at a = ⇡2 .

Ex. 9 (§4.2—#55). Find the linear approximation L(x) of f (x) = sin2 (x) at a = 0.
Ex. 10 (§4.2—#69, 71). Find the di↵erential of the function.

(a) y = cos(x) x2 + 2
(b) y =
x 1

Ex. 11. Find the di↵erential of the function y = ln(cos(✓)).

1
Ex. 12 (§4.2—#73). Find the di↵erential of y = and evaluate at x = 1 and dx = 0.25.
x+1
Ex. 13 (§4.2—#79). Find the change in volume, dV , if the sides of a cube change from x to
x + dx.

Ex. 14 (§4.2—#81). Find the change in volume, dV , if the radius of a sphere changes from r
to r + dr.

Ex. 15 (§4.2—#84). A spherical ball is measured to have a radius of 5mm, with a possible
measurement error of 0.1mm. Use di↵erentials to estimate the maximum possible error, relative
error, and percentage error in computing the volume of the ball.
Ex. 16 (§4.2—#85). A pool has a rectangular base of 10 ft by 20 ft and a depth of 6 ft. What
is the change in volume if you only fill it up to 5.5 ft?
 Workbook Lesson 17
§4.3, Maxima and Minima
Last revised: 2021-03-02 11:34

Objectives
• Define absolute extrema and local extrema.
• Explain how to find the critical points of a function over a closed interval.
• Describe how to use critical points to locate absolute extrema over a closed interval.

Extreme values
Absolute max
Local max
Local min
Absolute min
3.0

2.5

2.0

1.5

1.0

0.5

-4 -2 2 4

-0.5

Let y = f (x) be a function. Let c be a number in the domain of f . The value f (c) is

• a global (or absolute) maximum value of f if

f (x)  f (c) for all x in the domain of f .

• a global (or absolute) minimum value of f if

f (x) f (c) for all x in the domain of f .

• a local (or relative) maximum value of f if for some d > 0,

f (x)  f (c) for all x in the domain of f such that |x c| < d.

• a local (or relative) minimum value of f if for some d > 0,

f (x) f (c) for all x in the domain of f such that |x c| < d.

The global and local maxima and minima are the extreme values (or extrema) of f .

Note: The conditions in (3) and (4) are sometimes stated “. . . for all x in the domain of f near c,’
or “for all x in an interval containing c.” (See Lesson 2.5, first Objective.)
Fact. Every global maximum of f is also a local maximum of f .

+ Note the di↵erence between an extreme value (output of a function—say, f (c)) and the input
value (say, c) that corresponds to that value.

Ex. 1. What are the extreme values (that is, the global and local maxima and minima) of these
functions?
• h(x) = 3(x 2)2 + 1
• f (x) = cos(x)
• g(x) = 4/x
• j(x) = 0
 The Extreme Value Theorem, Fermat’s Theorem, and critical numbers

Extreme Value Theorem (EVT). If f is a function that is continuous on [a, b] for some a < b,
then
• for some c 2 [a, b], f (c) is a global maximum of f on [a, b], and
• for some d 2 [a, b], f (d) is a global minimum of f on [a, b].

Ex. 2. The function y = g(x) pictured is continuous on its domain, but does not have a global
maximum. Why doesn’t this contradict the EVT?
Scratchwork:

-1 2

Answer: Left as exercise.

Commentary: If we restrict this function so that its domain is [ 1, 1], the restricted version of g
does have a global maximum, namely g(1) (since g is increasing).
Ex. 3. Define
1+x
f (x) =
, 5  x  5.
1 + x2
The notation “. . . , 5  x  5” means f is a function with domain [ 5, 5].
Prove that there exists a number c in the interval [ 5, 5] such that f has a global maximum f (c)
on [ 5, 5] by verifying the hypotheses of the EVT.
Solution:
The Extreme Value Theorem applies, because f is continuous on [ 5, 5]. Justification:
1+x
• is a rational function, so it is continuous at every point in its domain.
1 + x2
• Every number in [ 5, 5] is in the domain of f .

By EVT, there exists a number c in the interval [ 5, 5] such that f (c) is an global maximum of f
on [ 5, 5].
(There’s a d in the domain of f such that f (d) is a global minimum, too.)

Fermat’s Theorem. If f (c) is a local maximum or local minimum value of f , and f 0 (c) exists,
then f 0 (c) = 0.

Definition. A critical number (or critical point in the domain) of a function f is a number c
in the domain of f such that
(i) f 0 (c) does not exist, or
(ii) f 0 (c) = 0.

Ex. 4. Find the critical numbers of the function

f (x) = x3 (x2 16), 1 < x < 1.

Solution:
f is di↵erentiable everywhere, so we only need to check (ii) in the definition of a critical number.
f (x) = x5 16x3
f 0 (x) = 5x4 48x2 = x2 (5x2 48)
Set f 0 (x) = 0:
x2 (5x2 48) = 0
q q
48
has solutions x = 0 and x = ± 5 = ±4 35 .
q
3
Answer: The critical numbers of f are 0 and ±4 5
.
Question: Fermat’s Theorem says

f (c) extreme value of f =) c critical number of f

Does the converse implication,

f (c) extreme value of f (= c critical number of f

hold?
Answer: No. In the previous exercise (graph shown below), f (0) = 0 is not an extreme value, but
c = 0 is a critical number.
f (x) = x3 (x 4)(x + 4)

3 3
-4 4
5 5

+ Can you think of another function f such that f 0 (c) = 0 for some number c in its domain,
but f (c) is not an extreme value of f ?

Ex. 5. Find all critical numbers of f (x) = x3 + 6x2 15x.

Solution:
f 0 (x) = 3x2 + 12x 15
Solve f 0 (x) = 0:

3x2 + 12x 15 = 0
2
x + 4x 5=0
(x + 5)(x 1) = 0
x= 5 or x = 1

f 0 (x) does not exist:

Never.

5, 1
 ⇡
Ex. 6. Find all critical numbers of f (x) = x 2 cos x ( 2
 x  0).
Solution:
f 0 (x) = 1 + 2 sin x
Solve f 0 (x) = 0:

1 + 2 sin x = 0
1
sin x =
2
1.0

1.0

0.5
0.5

-1.0 -0.5 0.5 1.0

-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.5

-0.5
-0.5
-pê6

-1.0 -1.0

f 0 (x) does not exist:

Never.

⇡
6

p 1
Ex. 7. Find all critical numbers of h(p) = .
p2 + 4
Solution:
(p2 + 4)(1) (p 1)(2p) p2 + 2p + 4
h0 (p) = =
(p2 + 4)2 (p2 + 4)2
Solve h0 (p) = 0:

p2 + 2p + 4
=0
(p2 + 4)2
p2 + 2p + 4 = 0
p p p
2± 22 4( 1)(4)
p= 2( 1) = 1 ± 12 20 = 1 ± 5

h0 (p) does not exist: (p2 + 4)2 = 0

Never.

p
1± 5
Notice that in the previous three exercises, the instructions were the same—“find the critical
numbers”—but the methods used to actually do this were entirely di↵erent. (Factoring in Exercise
5, graphing in Exercise 6, and the Quadratic Formula in Exercise 7.)
• Get used to this. In calculus and later classes, the techniques we’ve learned are tools in our
toolbox: we are asked to solve problems using any tool we can, not to mechanically run
through the same step-by-step procedure over and over like robots.

Ex. 8. Find all critical numbers of F (x) = x4/5 (x 4)2 .

Solution:

4
F 0 (x) = x 1/5 (x 4)2 + 2x4/5 (x 4)
5
1
= x 1/5 (x 4) 4(x 4) + 10x
5
1
= x 1/5 (x 4)(14x 16)
5
2
= p (x 4)(7x 8)
55x

Solve F 0 (x) = 0:

2 8
p (x 4)(7x 8) = 0 x = 4 or x =
55x 7

F 0 (x) does not exist: x = 0

0, 87 , 4
 Finding global extreme values with the Closed Interval Theorem

Closed Interval Theorem. To find the global maximum and global minimum values of a contin-
uous function f on a closed interval [a, b]:
1. Find f (c) for all critical numbers c in the domain of [a, b].
2. Find f (a) and f (b).
3. The largest (smallest) of the numbers you found is the global maximum (minimum).

Ex. 9. Find the global maximum and minimum values of

f (x) = x3 3x2 + 1 ( 1
2
 x  1)
Solution:
Critical numbers:
f 0 (x) exists for every x in the domain [ 12 , 1] of f , so condition (i) in the definition of a
critical number is not true for any c in the interval [ 12 , 1].
f 0 (x) = 3x2 6x = 3x(x 2) = 0
x = 0: in domain X
x = 2: not in domain (reject)
Critical numbers: 0
Extreme values:
Endpoints:
1 1 3 1
f( 2) = 8 4 +1= 8

f (1) = 1 3+1= 1
Critical points:
f (0) = 1
Global minimum value: 1
Global maximum value: 1

-2 -1 1 2

-1

-2
 x
Ex. 10. Find all absolute maxima and minima of f (x) = on [0, 3].
x2 x+1
Solution:
⇥ ⇤
1 0
f 0 (x) = x(x2 x + 1)
f 0 (x) = (x2 x + 1) 1
x(x2 x + 1) 2
(2x 1)
1 2x2 x
f 0 (x) =
x2 x+1 (x2 x + 1)2

1 2x2 x
=0
x2 x+1 (x2 x + 1)2
x2 + 1
=0
(x2 x + 1)2
x = ±1

f (0) = 0
f (1) = 1
f (3) = 37

Absolute minimum value: f (0) = 0

Absolute maximum value: f (1) = 1

Ex. 11. Find all absolute maxima and minima of f (x) = (x2 1)3 on [ 1, 2].
Solution:
f 0 (x) = 3(x2 1)2 (2x) = 0
x = 0 or x = ±1

f ( 1) = 0
f (0) = 1
f (1) = 0
f (2) = 27

Absolute minimum value: f (0) = 1

Absolute maximum value: f (2) = 27
 Additional exercises

Ex. 12. The graph of a function is shown. State the absolute and local maximum and minimum
values of the function.
y
9
8
7
6
5
4
3
2
1
x
1 2 3 4 5 6 7 8 9

Ex. 13 (§4.3—#107). Sketch the graph of a function that is continuous on [ 4, 4] with absolute
maximum values at x = 2 and x = 3, a local minimum value at x = 1, and an absolute minimum
value at x = 4.
Ex. 16 (§4.3—#109, 111, 113, 115, 116, 117). Find the critical numbers of the function.
p p
(a) f (x) = 4 x x2 (c) f (x) = 4 x2 (e) f (x) = sin2 (x)
x2 1 1
(b) f (x) = ln(x 2) (d) f (x) = (f) f (x) = x +
x2 + 2x 3 x
Ex. 17 (§4.3—#90). Recall: the maximum or minimum value of a quadratic function f (x) =
B
Ax2 + Bx + C is given by the formula f 2A
. Prove this formula using calculus.

Ex. 18 (§4.3—#119, 121, 123, 127, 129, 133). Find the local and absolute maximum values
and the local and absolute minimum values of the function over the given interval.
2 (d) f (x) = sin(x) + cos(x) over [0, 2⇡]
(a) f (x) = x2 + over [1, 4]
x
1 (e) f (t) = x2 + 4x + 5 over ( 1, 1)
(b) f (x) = over (0, 1)
x x2
x2 + x + 6
(c) f (x) = x + sin(x) over [0, 2⇡] (f) f (t) = over ( 1, 1)
x 1
Ex. 19 (§4.3—#135, 139). Technology required. Use a graph to estimate the absolute maximum
and minimum values of the function. Then use calculus to find the exact maximum and minimum
values.
p
p 4 x2
(a) f (x) = 3x 1 x 2 (b) f (x) = p
4 + x2

Ex. 20 (§4.3—#141). A ball is thrown into the air and its height (in meters) is given by

h(t) = 4.9t2 + 60t + 5.

(a) Find the height at which the ball stops ascending.

(b) How long after it is thrown does this happen?
 Workbook Lesson 18
§4.4, The Mean Value Theorem
Last revised: 2021-09-30 12:08

Objectives
• Explain the conclusion of Rolle’s Theorem in plain English.
• Explain the conclusion of the Mean Value Theorem in plain English.
• Explain three consequences of the Mean Value Theorem (Corollaries 1–3 below).

Rolle’s Theorem
Rolle’s Theorem. Let f be a function. There is some number c in the interval (a, b) such that
f 0 (c) = 0 if:
• f is continuous on [a, b],
• f is di↵erentiable on (a, b), and
• f (a) = f (b).

To put it in other words, Rolle’s Theorem says that:

However complicated the graph of a function may be,
there must be some point at which the tangent is horizontal
if the conditions of Rolle’s Theorem are satisfied.

Consider, for example, the function whose graph is shown below.

- -

We check the conditions of Rolle’s Theorem:

• The function values are equal at the two marked points (dashed horizontal line).
• The function is continuous from one marked point to the other (the graph can be drawn
without lifting one’s pencil).
• The function is di↵erentiable in-between the two marked points (there are no sharp corners
or vertical tangents).
Rolle’s Theorem now guarantees that at some point between the two marked points, the graph has
a horizontal tangent. (Indeed, we can plainly see this is true at several points on the graph.)
+ Use the applet “ Rolle’s Theorem” on iCollege to explore the meaning of Rolle’s Theorem
with several di↵erent functions.

Proof of Rolle’s Theorem:

Let f be a function that is continuous on [a, b] and di↵erentiable on (a, b).
Suppose f (a) = f (b).
Case 1. f (x) = const.
Then f 0 (x) = 0 for all x, so any c in the interval (a, b) works.
Case 2. f (x) > f (a) for some x in the interval (a, b).

If there is some function value f (x) greater than f (a) = f (b),

where x is in-between a and b, then there must be some c
in-between a and b at which the tangent is horizontal.
The EVT can be applied: f is continuous on [a, b] X
⇥ ⇤
By EVT, there is some c in the interval [a, b] s.t. f (c) is a local max for f on a, b .
Since f (b) = f (a), the local max must occur at some c in the interval a, b .
(Why? Because we know f (a) = f (b) can’t be a local max by the hypothesis of this
case.)

f (c) is a local max X

Fermat’s Theorem can be applied: Thus f 0 (c) = 0.
f 0 (c) exists X

Case 3. f (x) < f (a) for some x in the interval (a, b).
(Similar to Case 2)
EVT =) there exists a number c in the interval [a, b] such that f (c) is a local min.
f (b) = f (a) =) a 6= c 6= b.
Fermat =) f 0 (c) = 0. ⇤
Ex. 1. Show that x3 + x 1 = 0 has exactly one (real) root.
Solution:
Claim: f (x) = x3 + x 1 has at least one root.
f (0) = 0 + 0 1= 1<0
f (1) = 1 + 1 1=1>0

So by , there is some number x0 in the interval (0, 1) such that f (x0 ) = 0.

| {z }
Which Theorem?

Claim: It is impossible for f (x) to have more than one root.

Assume for a contradiction that f has at least two roots, say f (a) = 0 = f (b).
f di↵erentiable on (a, b) X
Then Rolle’s Theorem applies: f continuous on [a, b] X
f (a) = f (b) X
By Rolle’s Theorem, there is some number c in the interval (a, b) s.t. f 0 (c) = 0.
However, f 0 (x) = 3x2 + 1 > 0, so it is impossible for f 0 (x) = 0.
We have a contradiction: f 0 (c) = 0 and f 0 (c) 6= 0.
Therefore, our assumption must have been false: f does not have at least two roots.

Mean Value Theorem

Rolle’s Theorem is used to prove a more general theorem: the Mean Value Theorem (see textbook
for proof).

Mean Value Theorem on a pedestrian bridge across East Zhushikou Avenue in Beijing

Mean Value Theorem. If f is di↵erentiable on (a, b) and f is continuous on [a, b], then there is
a number c in the interval (a, b) such that
f (b) f (a) = f 0 (c) · (b a). (*)
Note that (*) can be rewritten
f (b) f (a)
= f 0 (c).
b a

So the Theorem says that:

0 1 !
the average the instantaneous
@ rate of change A = rate of change for some c in the interval (a, b).
over [a, b] at c

The slope of the line through a, f (a) and b, f (b)

y
is = the average rate of change over [a, b].
x

The slope of the tangent line through c, f (c) is f 0 (c).

According to the Mean Value Theorem, these two slopes are equal
for some choice of c between a and b.

The Mean Value Theorem (and its special case, Rolle’s Theorem) just says there exists some c
between a and b.
Like some other theorems we have seen, the Mean Value Theorem does not tell you what the value
of c is. (Recall:) We call such a theorem an existence theorem.

Some other existence theorems:

• Intermediate Value Theorem
• Extreme Value Theorem
Definition. A constant M is an upper bound on a quantity y if y  M for all possible values
of y, and a lower bound if y M for all possible values of y.
• For example, in high school, the constant 100 was an upper bound on your final grade y in
math class, because 100 was the highest score allowed on your report card—no matter how
much extra credit you might have earned.

Ex. 2. Suppose f is a di↵erentiable function such that f (0) = 3 and f 0 (x)  5 for all values
of x. Use the Mean Value Theorem to find an upper bound on f (2).

f di↵erentiable on (0, 1) X
The MVT applies:
f continuous on [0, 1] X
By MVT, there is some c 2 (0, 2) s.t.
f (2) f (0) = f 0 (c)(2 0)
f (2) + 3 = 2f 0 (c)
f (2) + 3 = 2f 0 (c)  10 since f 0 (c)  5
f (2)  7

f (2)  7.

Ex. 3. Find a number c that satisfies the conclusion of the Mean Value Theorem for f (x) =
x3 3x + 2 on the interval [ 2, 2].

f di↵erentiable on ( 2, 2) X
The MVT applies:
f continuous on [ 2, 2] X
Use MVT:
f (2) f ( 2) = f 0 (c)(2 ( 2)) (by MVT)
4 = (8 2 + 2) ( 8 + 6 + 2) = 4f 0 (c)
f 0 (c) = 1

Now we know the value of the f 0 of c. But what’s the value of c? We can solve for it:
f 0 (x) = 3x2 3 (get a formula for f 0 . . . )
1 = f 0 (c) = 3c2 3 (. . . then plug in c, and use the fact that f 0 (c) = 1)
4 = 3c2
r
4
c=±
3
r
4
Both of ± are in [ 2, 2].
3
r
4
c=± .
3
Ex. 4. Before reading the proof to follow, explain why the following corollary to the Mean Value
Theorem is true.
Corollary 1. If f 0 (x) = 0 for all x in the interval (a, b), then f is constant on (a, b).

Proof:
For any x1 , x2 in the interval (a, b) such that x1 < x2 , by the Mean Value Theorem, there is
some c in the interval (x1 , x2 ) such that

f (x2 ) f (x1 ) = f 0 (c)(x2 x1 )

f (x2 ) f (x1 ) = 0
f (x2 ) = f (x1 )

(Why does this prove the theorem?)

⇤

Ex. 5. Complete the proof of the following corollary to the Mean Value Theorem.
Corollary 2. Suppose f and g are each di↵erentiable over an interval (a, b).
If f 0 (x) = g 0 (x) for all x in the interval (a, b), then there exists some constant
C such that f (x) = g(x) + C for every x in (a, b).
Proof:
Let h(x) = f (x) g(x). Then

h0 (x) =

for all x in the interval (a, b).

It follows that there exists some constant C such that h(x) = C for all x in (a, b)

.
Why?

We will present one more consequence of the Mean Value Theorem.

But first, let’s review two definitions you may have seen in earlier classes.

Definition: Let f be a function. Let I be an interval.

• f is increasing if, for every x1 and x2 in the interval I such that x1 < x2 ,
f (x1 ) < f (x2 ).
• f is decreasing if, for every x1 and x2 in the interval I such that x1 < x2 ,
f (x1 ) > f (x2 ).
Ex. 6. State what it would mean for a function to be not decreasing.
(Hint: See applet on iCollege: “Not increasing”)

1
Ex. 7. On what interval(s) is f (x) = increasing? On what interval(s) is it decreasing?
x

Corollary 3 (Increasing/Decreasing Test). Suppose f is continuous on the

closed interval [a, b] and di↵erentiable on the open interval (a, b).
(i) If f 0 (x) > 0 for all x in the interval (a, b), then f is increasing on [a, b].
(ii) If f 0 (x) < 0 for all x in the interval (a, b), then f is decreasing on [a, b].
Proof:
We will prove (i); the proof of (ii) is similar.
Assume for a contradiction that f 0 (x) > 0 for all x on the interval (a, b), but f is not an increasing
function on [a, b].
Since f is not increasing on [a, b], there exist numbers x1 and x2 in the interval [a, b] such that
f (x1 ) f (x2 ) and x 1 < x2 . (*)

Since f is di↵erentiable on the interval (a, b) and continuous on [a, b], by the Mean Value Theorem
there exists a number c in (a, b) such that
f (b) f (a)
f 0 (c) = .
b a

Subtracting f (x1 ) from both sides of (*) yields

f (x2 ) f (x1 )  0,
and we know that b a > 0 since a < b.
Therefore,
f (b) f (a)
f 0 (c) =  0.
b a
But contradicts that f 0 (x) > 0 for all x in (a, b).
⇤
 Exam questions for the theorems of Lessons 17 & 18 in previous semesters

Ex. 6.
(a) Complete the statement of the theorem.
Mean Value Theorem. There is a number c in the interval
(a, b) such that

if:
• f is on (a, b), and
• f is on [a, b].
(b) Suppose f is a di↵erentiable function such that f (0) = 3 and f 0 (x)  5 for all values of x.
Use the Mean Value Theorem to find the largest possible value for f (2).
(c) Find a number c that satisfies the conclusion of the Mean Value Theorem for g(x) = x3 3x+2
on the interval [ 2, 2].

Sample answer:
See Ex. 2 and Ex. 3 above.
Ex. 7.
Fermat’s Theorem. If f (c) is a local maximum or local mini-
mum value of f , and f 0 (c) exists, then f 0 (c) = 0.

A student argues that 0 is an extreme value of the function f (x) = x3 as follows:

(1) The function f (x) = x3 has derivative f 0 (x) = 3x2 .
(2) The domain of f 0 is ( 1, 1). In particular, f 0 (0) exists.
(3) f (0) = 0.
(4) Since f 0 (0) = 0, by Fermat’s Theorem f (0) = 0 is either a local maximum or a local minimum
value of f .
Which line contains an error? Explain what is wrong with the student’s reasoning.

Sample answer:
Line (4) is incorrect. The statement “if f 0 (c) = 0, then f (c) is a local min. or max.” is false.

Ex. 8.
Extreme Value Theorem (EVT). Suppose a < b. If f is a function that is
continuous on [a, b], then
• for some c in the interval [a, b], f (c) is a global maximum of f on [a, b]
• for some d in the interval [a, b], f (d) is a global minimum of f on [a, b]
Why can’t the Extreme Value Theorem be applied to find a global maximum of the Heaviside
function ⇢
1 if x 0
H(x) =
0 if x < 0
on the interval [ 1, 1]?

Sample answer:
The Extreme Value Theorem does not apply to the function H(x) on the interval [ 1, 1]
because H is not continuous on [ 1, 1].

+ Note that the Extreme Value Theorem will not be reprinted on exams or quizzes.
Ex. 9.
Use the Extreme Value Theorem to find the global maximum and minimum values of

f (x) = x3 3x2 + 1 ( 1
2
 x  1).

Ex. 10.
Prove that f (x) = x3 + x 1 has at least one root by following these steps.
(a) Find f (0) and f (1).
(b) What Theorem guarantees that there is some number x0 in the interval (0, 1) such that
f (x0 ) = 0?
(c) What facts must be known for the Theorem to apply?

Sample answer:
(a) f (0) = 1 and f (1) = 1
(b) Intermediate Value Theorem
(c) f is continuous, f (0) < 0, and f (1) > 0.
Ex. 11.
Rolle’s Theorem. Let f be a function. There is some number
c in the interval (a, b) such that f 0 (c) = 0 if:
• f is continuous on [a, b],
• f is di↵erentiable on (a, b), and
• f (a) = f (b).
Assume that the function
f (x) = x3 + x 1
has at least two roots a and b: that is,

f (a) = 0 = f (b).

Show that this assumption leads to a contradiction by following these steps.

(a) Check that Rolle’s Theorem applies to f on the interval [a, b] by verifying that
• f is di↵erentiable on (a, b)
• f is continuous on [a, b]
(b) Since Rolle’s Theorem applies, we must conclude there is some number c in the interval (a, b)
such that f 0 (c) = 0.
Prove this is impossible by solving the equation f 0 (x) = 0, and say why are there no (real)
solutions.

Sample answer:

(a) f is a polynomial, and any polynomial is continuous and di↵erentiable on ( 1, 1).

(b)
f 0 (x) = 0
3x2 + 1 = 0
1
x2 =
3
The equation has no solutions because x2 cannot be negative.
 Additional exercises

Ex. 12. The graph of a function f is shown. Verify that f satisfies the three hypotheses of Rolle’s
Theorem on the interval [0, 8]. Then estimate the value(s) of c that satisfy the conclusion of Rolle’s
Theorem on that interval.
y= f (x)
6
5
4
3
2
1
x
1 2 3 4 5 6 7 8 9

Ex. 13. The graph of a function g is shown.

y= g(x)
8
7
6
5
4
3
2
1
x
1 2 3 4 5 6 7 8 9

(a) Verify that g satisfies the hypotheses of the Mean Value Theorem on the interval [0, 9].
(b) Estimate the value(s) of c that satisfy the conclusion of the Mean Value Theorem on the
interval [0, 8].
(c) Estimate the value(s) of c that satisfy the conclusion of the Mean Value Theorem on the
interval [2, 7].
Ex. 14 (§4.4—#149).
• In order to be applied to a function f , the Mean Value Theorem requires that f must be
di↵erentiable on an interval (a, b).
• Prove that di↵erentiability is needed by drawing a counterexample of a function f with
domain [ 2, 2] on the blank coordinate system provided below.
• (That is, draw the graph of a function f which is not di↵erentiable on ( 2, 2) and which
does not satisfy the conclusion of the Mean Value Theorem for any number c such that
2 < c < 2.)
y= f (x)

x
-2 -1 1 2
 Workbook Lesson 19
§4.5, Derivatives and the Shape of a Graph
Last revised: 2020-09-29 12:44

Objectives
• Explain how the sign of the first derivative a↵ects the shape of a function’s graph.
• Use the Increasing/Decreasing Test to determine intervals of increase/decrease.
• State the First Derivative Test for critical points.
• Use concavity and inflection points to explain how the sign of the second derivative a↵ects the shape of a
function’s graph.
• Explain the concavity test for a function over an open interval.
• Explain the relationship between a function and its first and second derivatives.
• State the Second Derivative Test for local extrema.

Motivation
We can graph a function using calculators, computers, and even our phones. But we still teach
how to graph in calculus class. Why?
Consider the following graph of the function y = x6 + 4x5 3x 1.
7000

6000

5000

4000

3000

2000

1000

-4 -2 2 4

How many local maxima and minima are there? It looks like there’s just one, f ( 3.329 . . . ) =
265.354 . . . .
However, the tools of calculus tell us there must be two more.
Now look at this graph (top of next page):
 4

-1.5 -1.0 -0.5 0.5 1.0 1.5

-2

-4

-6

This is the same function, plotted on a di↵erent range of x-values. We see two more extreme
values—a local maximum f ( 0.657 . . . ) ⇡ 0.561 . . . and a local minimum f (0.597 . . . ) = 2.442 . . . .
We use calculus to ensure that we are not misled by technology. As we see in this example, a
graphing calculator can easily cause us to miss important features of the graph.

Finding intervals of increase/decrease

Recall from the previous lesson:
Increasing/Decreasing Test. Suppose f is continuous on the closed interval [a, b]
and di↵erentiable on the open interval (a, b).
(i) If f 0 (x) > 0 for all x in the interval (a, b), then f is increasing on [a, b].
(ii) If f 0 (x) < 0 for all x in the interval (a, b), then f is decreasing on [a, b].
Let us show how to find the intervals on which a function is increasing (or decreasing).

The rough outline of the process is:

Step 1. Find the critical numbers and mark them on the number line.

+ The critical numbers break up (or “partition”) the number line

into intervals.

Step 2. Determine the sign of f 0 (x) in each interval.

+ We’ll go into detail about how to do this in the next exercise.

Step 3. Apply the Increasing/Decreasing Test.

Ex. 1. Find the intervals on which f (x) = 3x4 4x3 12x2 + 5 is increasing, and the intervals on
which it is decreasing.
Solution.
Step 1. Find the critical numbers and mark them on the number line.
Critical numbers happen where f 0 (x) = 0 or f 0 (x) is undefined.
f 0 (x) = 0:

f 0 (x) = 12x3 12x2 24x = 12x(x2 x 2)

= 12x(x 2)(x + 1)

We see that f 0 (x) = 0 when x = 0, x = 2, or x = 1.

f 0 (x) is undefined:
This never happens, because f is a polynomial (and therefore its domain is all real
numbers).
Critical numbers on the number line:

-1 0 2

Step 2. Determine the sign of f 0 (x) in each interval.

We’ve already factored f 0 (x) = 12x(x 2)(x + 1).
This makes it easy to see where f 0 (x) is positive or negative—we can just find where
each factor is positive and negative, and then count the negative signs.
• An odd number of negative factors (e.g. positive ⇥ negative) yields a negative number.
• An even number of negative factors (e.g. positive ⇥ positive) yields a positive number.

interval sign of f 0 (x) 12x (x 2) (x + 1)

1<x< 1
1<x<0 + +
0<x<2 + +
2<x<1 + + + +

- + - +
-1 0 2

Step 3. Apply Increasing/Decreasing Test.

f is decreasing on ( 1, 1) and (0, 2), increasing on ( 1, 0) and (2, 1).
(We could use closed or open intervals here, because f 0 exists at each critical number.)
Ex. 2. Suppose the derivative of a function f is f 0 (x) = (x 3)2 (x + 1)4 (x 7)6 . On what
intervals is f increasing?

Ex. 3. Suppose the derivative of a function f is f 0 (x) = (x 3)2 (x + 1)4 (x 7)5 . On what
intervals is f increasing? Decreasing?

(x3)2 (x + 1)4
Ex. 4. Suppose the derivative of a function f is f 0 (x) = . On what intervals is f
(x 7)5
increasing? Decreasing? Can [3, 7] be one of the intervals of increase/decrease?
 The First Derivative Test
Recall: for di↵erentiable f ,
⇥ ⇤ (Fermat) ⇥ ⇤
f (c) is a local max or min value =)⇠
c is a critical number of f
⇠
(=

Given a critical number c, we want a test that tells us whether or not f (c) is a local max or
min.

First Derivative Test. Suppose c is a critical number of a continuous function f .

• If f 0 changes from positive to negative at c, then f (c) is a local max.
• If f 0 changes from negative to positive at c, then f (c) is a local min.
• If f 0 does not change sign at c, then f (c) is not a local max or a local min.

Ex. 5. Find the local maximum and local minimum values of f (x) = 3x4 4x3 12x2 + 5.
Solution:

Using the technical computing system

Mathematica (an app available on GSU com-
puters and free for download by GSU students),
we can quickly dispense with the gruntwork of
finding the critical numbers of f (x):

f 0 (x) = 0: When x = 0, x = 2, or x = 1
f 0 (x) is undefined: Never
Critical numbers on the number line:

- + - +
-1 0 2
local local local
min max min

f ( 1) = 0
Local max values: 5
f (0) = 5
Local min values: 0, 27
f (2) = 27
Ex. 6. Find the local maximum and local minimum values of

g(x) = x + 2 sin x (0  x  2⇡).

Solution:
The derivative
g 0 (x) = 1 + cos x
exists everywhere, so the only critical numbers c are those such that g 0 (c) = 0:

g 0 (x) = 0
1 + 2 cos x = 0
1
cos x =
2
2⇡ 4⇡
x= ,
3 3

2⇡ 4⇡
Critical numbers: ,
3 3

The fastest and simplest way to determine the sign chart is by looking at the graph of g 0 (x),
which we can quickly sketch without tech based on what we know about the cos function:
Graph of y = g 0 (x):
3

2p 4p
p 2p
3 3
-1

Sign chart for g 0 (x):

+ - +
2p 4p
3 3
local local
max min

Answer: (left as exercise)

Local max value:

Local min value:
 Concavity

Definition: If the graph of f lies above (below) all its tangents on an interval I, it is said to be
concave up (down) on I.
6 6

5 5

4 4

3 3

2 2

1 1

0 1 2 3 4 0 1 2 3 4

Definition: A point P on a curve y = f (x) is an inflection point if f is continuous at x and the

curve changes concavity from up to down or vice versa.

Concavity Test. Suppose f is twice di↵erentiable on an interval I. (That is, f 00 exists.)

(a) If f 00 (x) > 0 for all x in I, then the graph of f is concave up on I.
(b) If f 00 (x) < 0 for all x in I, then the graph of f is concave down on I.

Ex. 7. Sketch the graph of a function f satisfying all of the following conditions:
(i) f 0 (x) > 0 on ( 1, 1); f 0 (x) < 0 on (1, 1)
(ii) f 00 (x) > 0 on ( 1, 2) and (2, 1); f 00 (x) < 0 on ( 2, 2)
(iii) lim f (x) = 2; lim f (x) = 0
x! 1 x!1

Observations before graphing:

1. Horizontal asymptotes at y = 2 and y = 0.
2. f increases to its maximum at x = 1.
3. f has inflection points at x = ±2.
 Second Derivative Test. Suppose f 00 is continuous on an interval containing c.
(a) If f 0 (c) = 0 and f 00 (c) > 0, then f has a local maximum at c.
(b) If f 0 (c) = 0 and f 00 (c) < 0, then f has a local minimum at c.

For example, part (a) is true because the tangent at c is horizontal, and the concavity is up.

Ex. 8. Sketch the curve y = x4 4x3 by hand.

Solution:
f 0 (x) = 4x2 (x 3)
f 00 (x) = 12x(x 2)

Critical numbers:
f 0 is defined at all x in R.
f 0 (x) = 0 at x = 0 and x = 3.

2nd derivative test:

f 00 (3) = 36 > 0 so f (3) = 27 is a local max
f 00 (0) = 0 so . . . ?
(2nd derivative test yields no information where f 00 = 0.)

interval sign of f 00 (x) = 12(x 2) concavity

1<x<0 + up
0<x<2 down
2<x<1 + up

40

30

20

10

-2 -1 1 2 3 4

-10

-20
 Additional exercises

Ex. 9. Use the given graph of f to find the following.

y= f (x)

x
-2 -1 1 2

(a) The open intervals on which f is increasing.

(b) The open intervals on which f is decreasing.
(c) The open intervals on which the graph of f is concave upward.
(d) The open intervals on which the graph of f is concave downward.

Ex. 10. Use the given graph of g to find the following.

y= g(x)

x
-2 -1 1 2

(a) The open intervals on which f is increasing.

(b) The open intervals on which f is decreasing.
(c) The open intervals on which the graph of f is concave upward.
(d) The open intervals on which the graph of f is concave downward.
(e) The x-coordinate(s) of the point(s) of inflection.
Ex. 11 (§4.5—201). The graph of the derivative f 0 of a function f is given. List all intervals
on which f is increasing or decreasing.

Ex. 12 (§4.5—215). The graph of the derivative f 0 of a continuous function f is shown.

(a) On what intervals is f is increasing? Decreasing?
(b) At what values of x does f have a local maximum? Local minimum?
(c) On what intervals is the graph of f concave upward? Concave downward?
(d) What are the x-coordinates of the inflection points?
Ex. 13 (§4.5—221). List all intervals where the function

sin(x) + sin3 (x), ⇡ < x < ⇡,

is increasing or decreasing, and list all local minimum values and all local maximum values of f .

Ex. 14 (§4.5—225, 229). For each of the following functions. . .

(a) Find the intervals on which f is increasing or decreasing.
(b) Find the local maximum and minimum values of f .
(c) Find the intervals of concavity.
(d) Find the inflection points.

• f (x) = x3 6x2 • g(x) = x2 + x + 1

Ex. 15.
(a) Find the intervals on which f is increasing or decreasing.
(b) Find the local maximum and minimum values of f .
(c) Find the intervals of concavity.
(d) Find the inflection points.
(e) Sketch the graph of f .
p
• f (x) = 36x + 3x2 2x3 • h(x) = x 6 x
• g(x) = 12 x4 4x2 + 3 • j(✓) = 2 cos(✓) + cos2 (✓), 0  ✓  2⇡
 Some important theorems of Di↵erential Calculus
Last revised: 2021-03-09 12:35

§ Theorem Conditions Conclusion

(true if all conditions are satisfied)
(2.4) IVT f continuous on [a, b] There is some c in (a, b) such that f (c) = z.
f (a) 6= f (b)
z is between f (a) and f (b)
(4.3) EVT f continuous on [a, b] There is some c in [a, b] such that f (c) = min{f (x) a  x  b}.
There is some d in [a, b] such that f (d) = max{f (x) a  x  b}.
(4.3) Fermat f (c) local min or max value f 0 (c) = 0
f 0 (c) is a real number
(4.3) Closed Interval f continuous on [a, b] The min and max values of f on [a, b] are members of the following set:
f (a), f (b) [ {f (c) c is a critical number of f }.

(4.4) MVT f continuous on [a, b] There is some c in (a, b) such that

f di↵erentiable on (a, b) f (b) f (a) = f 0 (c) · (b a).

(4.4) Inc./Dec. Test f continuous on [a, b] If f 0 (x) > 0 for all x in (a, b), then f is increasing on [a, b].
f di↵erentiable on (a, b) If f 0 (x) < 0 for all x in (a, b), then f is decreasing on [a, b].
(4.5) 1st Deriv. Test c critical number of f If f 0 changes sign at c from + to , then f (c) is a local max value.
f continuous on [a, b] If f0 changes sign at c from to +, then f (c) is a local min value.
f 0 changes sign at c If f 0 does not change sign at c, then f (c) is not a local min or max value.

(4.5) Concavity Test f twice di↵erentiable on I, If f 00 (x) < 0 for all x in I, then the graph of f is concave up on I.
where I is an interval
If f 00 (x) < 0 for all x in I, then the graph of f is concave down on I.
(4.5) 2nd Deriv. Test f 00 continuous on an interval If f 0 (c) = 0 and f 00 (c) > 0, then f (c) is a local max value.
that contains c
If f 0 (c) = 0 and f 00 (c) < 0, then f (c) is a local min value.
 Workbook Lesson 20
§4.6, Limits at Infinity, Asymptotes, and Curve Sketching
Last revised: 2021-03-09 07:44

Objectives
• Calculate the limit of a function as x increases or decreases without bound.
• Recognize a horizontal asymptote on the graph of a function.
• Estimate the end behavior of a function as x increases or decreases without bound.
• Analyze a function and its derivatives to draw its graph.

12297_ch01_ptg01_hr_052-061.qk_12297_ch01_ptg01_hr_052-061 1/17/12 1:46 PM Page 59

Limits at infinity
k_12297_ch01_ptg01_hr_052-061 1/17/12 1:46
In the previous examples, wePM
sawPage
the 59
output of a function growing arbitrarily “large” (that is,
toward positive or negativeLicensed to:Now we look at limits where the input approaches ±1.
infinity).
x2 1
Ex. 1. What value does f (x) = approach as we take x ! 1? As we take x ! 1? SECTION
x2 + 1
Let’s begin by investigating the behavior of
x f !x"
SECTION 1.6 LIMITS INVOLVING INFINITY 59 x ! 2
0 !1
f !x" ! 2
&1 0 x "
Let’s begin by investigating the behavior
&2 0.600000
of the function f defined by
f !x" &3 0.800000 as x becomes large. The table at the left gives
&4 0.882353 decimal places, and the graph of f has been dr
1 x2 ! 1
!x" ! 20.923077
f&5
0 &10 "1
x 0.980198 y
0.600000 &50 0.999200 y
0.800000 as x becomes large. The table at &100
the left gives values of this function correct to six
0.999800
0.882353 decimal places, and the graph of&1000
f has been drawn
0.999998by a computer in Figure 7.
0.923077 0 1
0.980198 y
0.999200 y=1 FIGURE 7
0.999800
0.999998 As x grows larger and larger you can see tha
0 er to 1. In fact, it seems
x that we can make the
1 ≈-1 by taking x sufficiently large. This situation is
y=
≈+1
FIGURE 7 x2 ! 1
lim 2
x l# x " 1

As x grows larger and larger you can see that the values of f !x" get closer and clos-
er to 1. In fact, it seems that we can make the values of f !x" asInclose
general, we use
as we likethe
to notation
1
by taking x sufficiently large. This situation is expressed symbolically by writing lim f !x" !
x l#

x2 ! 1 to indicate that the values of f !x" approach L a

lim !1
x l# x 2 " 1

3 DEFINITION Let f be a function define

In general, we use the notation
lim f !x" ! L lim f !x" !
x l# x l#

means that the values of f !x" can be made a

to indicate that the values of f !x" approach L as x becomes largersufficiently
and larger.large.
Definition (Limits at infinity): Let f be a function defined on some interval (a, 1). Then

lim f (x) = L
x!1

means that for any E > 0, there is a number M such that |f (x) L| < E whenever x > M .
The statement
lim f (x) = L
x! 1

is defined similarly.

Definition: The line y = L is a horizontal asymptote of the curve y = f (x) if either

lim f (x) = L or lim f (x) = L.

x!1 x! 1

Again, the symbols ±1 do not represent numbers .

1
Ex. 2. Show that lim = 0, using the definition of a limit at infinity.
x!1 x

Solution: Let e > 0 be given. We need to find M such that

1
0 <E
x

for any x > M .

1 1 1
We may assume x > M > 0. (Why?) Then 0 = < E () < x.
x x E
Choose M = 1/E.

1
Ex. 3. By looking at the graph of f (x) = sin(x), we make the guess that lim f (x) = 0. Can
x x!1
you justify this guess without referring to the graph? (Justification given on next page.)
1
y= sinHxL
x

1.0

0.8

0.6

0.4

0.2

0.0 x
20 40 60 80
-0.2

-0.4
Justification:
Since sin(x) is always between 1 and 1, that is,

1  sin(x)  1,

we know
1 1 1
 sin(x)  (x > 0).
x x x
1 1
But we know that ! 0 as x ! 1. Similarly, ! 0 as x ! 1.
x x
1 1
As x ! 1, the value of sin(x) must approach 0 because it is “squeezed” between
x x
1
and , each of which approach 0.
x
(If we had to, we could make this argument into a careful formal proof.)

Ex. 4. Find lim sin(x), if it exists.

x!1

Solution:
As x ! 1, the value of sin(x) oscillates between 1 and 1 infinitely often. Therefore, the
limit does not exist: no number M is as required in the definition of a limit at infinity.
p p
Except for the laws lim xn = an and lim n x = n a, the Limit Laws are valid for limits as x ! ±1.
x!a x!a

1
Ex. 5. Let n be a positive integer. Taking it as known that ! 0 as x ! ±1, use the Limit
x
Laws to justify the following facts.
1 1
lim = 0, lim = 0.
x!1 xn x! 1 xn
 3x2 x 2
Ex. 6. Find lim . (Use the Limit Laws.)
x!1 5x2 + 4x + 1

To find the limit of any rational function as x ! ±1, start by dividing numerator and
denominator by the largest power of x appearing in the denominator.
Solution:
3x2 x 2
1 2
3x2 x 2 x2 3 x x2
lim 2 = lim = lim
x!1 5x + 4x + 1 x!1 5x2 + 4x + 1 x!1 5 + 4
x
+ 1
x2
x2 1 2
lim 3 x x2
x!1
= 4 1
lim 5 + x
+ x2
x!1

lim 3 lim 1 2 lim 1

2
x!1 x!1 x x!1 x
= 1
lim 5 + 4 lim + lim 12
x!1 x!1 x x!1 x

3 0 0 3
= = .
5+0+0 5

x5 1
Ex. 7. Find lim . (Use the Limit Laws.)
x!1 x3 + 1

x5 1 1 x!1
x5 1 x3
x2 x3
!1
lim 3 = lim x3 +1
= lim 1 x!1
x!1 x + 1 x!1 x!1 1+ !1
x3 x3

= 1.

p
t tt
Ex. 8. Find lim 3/2 .
t!1 2t + 3t 5

Not a rational function, but the same principle works:

p t1 t3/2
3/2
t t t t t t3/2
lim = lim 3/2 = lim 3/2
t!1 2t3/2 + 3t 5 t!1 2t + 3t 5 t!1 2t + 3t1 5t0
t3/2
1/2
t 1
= lim 1/2 3/2
t!1 2 + 3t 5t
1
p
t
1 1
= lim = .
t!1 2 + 3t
p 5
p
t t
2
 p
Ex. 9. Find lim x2 + 1 x.
x!1
p
Solution: The Subtraction Law doesn’t apply, because the limits lim x2 + 1 and lim x
x!1 x!1
don’t exist. (We would get the meaningless expression 1 1.)
Let’s try rationalizing (the numerator). . .
p
p p x2 + 1 + x
lim x2 + 1 x = lim x2 + 1 x p
x!1 x!1 x2 + 1 + x
2 2
(x + 1) x
= lim p
x!1 x2 + 1 + x
1
= lim p
x!1 x2 + 1 + x
p
It is incorrect to say lim x2 + 1 + x = 1 + 1 = 1. We can avoid this error that by
x!1
noticing that p
x2 + 1 + x > x ! 1 as x ! 1. (*)
p p
(Recall that means the nonnegative square root, so we always have x2 + 1 0.)
p
It follows that lim x2 + 1 + x = 1.
x!1

Ex. 10. Find lim x2 x, if it exists. Justify your answer.

x!1

Solution:
lim x2 x = lim x(x 1) = 1
x!1 x!1

because for all x > 1 we have x · (x 1) > x 1, and (x 1) ! 1 as x ! 1.

 Curve sketching
Consider the following functions:
2x
• f (x) =
x2 1
cos(x)
• g(x) =
2 + sin x
p
• h(x) = 3 x3 + 1
• j(x) = x5/3 5x2/3

Exercise. Sketch the graphs of the above functions.

It’s understandable if this exercise makes you a little bit nervous, at first glance. These don’t look
anything like the kinds of functions you were asked to graph in earlier classes.
But at this stage of your mathematical education, you already know everything you need to to
graph these functions. You know how to analyze the behavior of a function using calculus.
• You can describe the shape of the graph.
• You can say where the graph rises and falls, and where it levels o↵.
• You can find vertical asymptotes, which show where the function “blows up” (F (x) ! ±1).
• And you can find the end behavior of F (x) is (that is, the behavior of F (x) as x ! 1).

Before we tackle the above Exercise, let’s review some vocabulary and facts about functions’
behavior.

Review of preliminaries for curve sketching:

Let f be a function with domain D.
• f is even if f (x) = f ( x) for all x 2 D.
• f is odd if f (x) = f ( x) for all x 2 D.
– Are there any functions that are both odd and even?
– Are there any functions that are odd and periodic?
– Are there any functions that are even and periodic?
• If p > 0 is a constant and f (x + p) = f (x) for all x 2 D, we say f is periodic.
+ For example, cos(x + 4⇡) = cos x for all x in R, so cos is periodic.
• The smallest p as above is the period of f . We say f is p-periodic if p is the period of f .
Period of cos, sin: 2⇡
Period of tan: 2⇡
 Ex. 11. Prove that cot has period ⇡.
Solution:
For all x in the domain of cot,
1
cot(x) = .
tan(x)
Since 1 1
cot(x + ⇡) = = = cot(⇡),
tan(x + ⇡) tan(x + ⇡)
the function cot is ⇡-periodic.

• To sketch the graph of a p-periodic function, just sketch the graph an interval of length p,
then repeat it.
• If f is periodic with period p, then the period of the function
f (Ax)
for a constant A 6= 0 is p
.
|A|
2⇡
Ex. 12. Period of cos( x5 ): = 10⇡.
|1/5|
⇡
Ex. 13. Period of tan( 2x) + 1: = ⇡2 .
| 2|
Ex. 14. What is the period of cot(1 x)?

• The line y = L is a horizontal asymptote (HA) if limx!1 f (x) = L or limx! 1 f (x) = L.

Ex. 15. What are the horizontal asymptotes, if any, of the function y = 2x ?

• The line x = a is a vertical asymptote (VA) if limx!a± f (x) = ±1.

x 4
Ex. 16. What are the vertical asymptotes, if any, of f (x) = ?
x 3
There is a vertical asymptote at x = 3, because limx!a+ f (x) = ±1.

x2 6x + 9
Ex. 17. What are the vertical asymptotes, if any, of g(x) = ?
x 3
There are none, because limx!a g(x) = limx!a x 3 is a number for any a 2 R—even
for a = 3.

• To find the vertical asymptotes of a rational function, set the denominator equal to 0 after
canceling any common factors.
⇡
Ex. 18. Show that x = is an asymptote of y = tan x by using the definition.
2
 The 8-step process for sketching a curve
A. Find domain
B. Find x- and y-intercepts
C. Determine symmetry (even/odd, periodic)
D. Find any vertical or horizontal asymptotes
E. Find intervals where function is increasing, intervals where it is decreasing
F. Identify any local maxima and local minima
G. Determine intervals where function is concave up, intervals where it is concave down, and
identify any inflection points
H. Sketch the curve
 2x2
Ex. 19. Sketch the graph of f (x) = by hand.
x2 1

A. Domain
{x | x 6= ±1}

B. Intercepts
(0, 0)

C. Symmetry
Even:
We test whether or not f ( x) = f (x):

2( x)2 2x2 X
f ( x) = = = f (x)
( x)2 1 x2 1
X f is even.
Odd:
f is even and f (x) 6⌘ 0, so f is not odd.
Periodic:
f is not periodic.

D. Asymptotes
HA:

2
lim f (x) = lim 1 =2
x!1 x!1 1 x2

y=2
VA:
Noting that
• f is a rational function and
• f has no common factors in numerator and denominator,
we set the denominator equal to 0 to find two vertical asymptotes x = ±1 .
E. Increasing/decreasing
Start by finding a formula for f 0 (x):
4x
f 0 (x) =
(x2 1)2
By the Increasing/Decreasing Test (§3.3), f is increasing on an interval I if, and only
if, f 0 (x) > 0 for all x 2 I.
f 0 (x) > 0
4x
() >0
(x2 1)2
() 4x > 0 (x2 1)2 is always positive
() x < 0 Dividing by 4 reverses the inequality symbol
(Recall that the symbol () means “if, and only if,” or that two statements are “logically
equivalent,” i.e. both true or both false. We often don’t write this symbol, but when
we solve an inequality or an equation, it is silently implied between each step.)
Thus
f 0 (x) > 0 () x < 0.
This means
• f is increasing on the interval {x | x < 0}.
• f is decreasing on the interval {x | x > 0}.

F. Local max/min
f 0 (x) = 0 if, and only if, x = 0.
By the 1st Derivative Test (§3.3), f (0) = 0 is a local max.

G. Concavity and inflection points

12x2 + 4
f 00 (x) = >0
(x2 1)3
() (x2 1)3 > 0 12x2 + 4 is always positive
() |x| > 1
Thus
f 00 (x) > 0 () |x| > 1.
By the Concavity Test (§3.3), this means
• f is concave up on the intervals {x | x > 1} and {x | x < 1}.
• f is concave down on the interval {x | 1 < x < 1}.
 cos x
Ex. 20. Sketch the graph of f (x) = by hand.
2 + sin x

A. Domain
R

B. Intercepts
(2k+1)⇡
(0, 12 ) and 2
, 0 for any integer k

C. Symmetry
Even:
Want to test whether or not
cos x
f ( x) = f (x) = .
2 + sin x

cos( x) cos(x)
f ( x) = =
2 + sin( x) 2 sin(x)
f is not even.
Odd:
Want to test whether or not
cos x
f ( x) = f (x) = .
2 + sin x

cos( x) cos(x)
f ( x) = =
2 + sin( x) 2 sin(x)
f is not odd.
Periodic:
Want to test whether
f (x + p) = f (x)
for some p > 0 yet to be determined.
Guess: The period p of f is 2⇡. (We guess this because cos and sin are periodic
with period 2⇡.)
Check:
cos(x + 2⇡) cos(x)
f (x + 2⇡) = = = f (x)
2 + sin(x + 2⇡) 2 + sin(x)
X f is periodic (with period 2⇡)
At this point we should pick an interval I of length p = 2⇡.
We’ll pick
I = [0, 2⇡].

We can go back to the intercepts and note which ones have x 2 I:

⇡ 3⇡
x-int.: 2
,0 , 2
,0
y-int.: 0, 12

D. Asymptotes
HA: None
VA: None

E. Increasing/decreasing
Start by finding a formula for f 0 (x):
2 sin x + 1
f 0 (x) =
(2 + sin x)2

By the Increasing/Decreasing Test (§3.3), f is increasing on an interval I if, and only

if, f 0 (x) > 0 for all x 2 I.

f 0 (x) > 0
2 sin x + 1
() >0
(2 + sin x)2
() 2 sin x
1>0
1
() sin x <
2
7⇡ 11⇡
() <x<
6 6

Thus
7⇡ 11⇡
f 0 (x) > 0 () <x< .
6 6
This means
7⇡ 11⇡
• f is increasing on the interval {x | 6
<x< 6
}.
7⇡
• f is decreasing on the intervals {x | 0 < x < 6
} and {x | 11⇡
6
< x < 2⇡}.

F. Local max/min
7⇡ p1 11⇡ p1
Using the 1st Derivative Test, f 6
= 3
is a local min, and f 6
= 3
is a local max.
G. Concavity and inflection points
Want to know when
2 cos x(1 sin x)
f 00 (x) =
(2 + sin x)3
is positive.
• Since 1  sin x  1, we know (1 sin x) in the numerator is always nonnegative.
• Also since 1  sin x  1, we know 2 + sin x is always positive, so (2 + sin x)3 is
always positive.
• Since
1 sin x
f 00 (x) = 2 cos x fraction is always nonnegative
(2 + sin x)3

we conclude that
⇡ 3⇡
f 00 (x) > 0 () 2 cos x > 0 () cos x < 0 () <x<
2 2
for x 2 I = [0, 2⇡].
By the Concavity Test,
⇡ 3⇡
• f is concave up on the interval {x | 2
<x< 2
}.
• f is concave down on the intervals {x | 0 < x < ⇡2 } and {x | 3⇡
2
< x < 2⇡}.
 p
3
Ex. 21. Sketch the graph of f (x) = x3 + 1 by hand.

A. Domain
R

B. Intercepts
( 1, 0), (0, 1)

C. Symmetry
Even:
p p
f ( x) = 3
( x)3 + 1 6= 3 x3 + 1
p
3
(Take x = 1, then f ( 1) = 0 but f (1) = 2.)
f is not even.
Odd:
f is not odd.
Periodic:
f is not periodic because f (x) ! 1 as x ! 1

D. Asymptotes
HA:
limx!1 f (x) = 1 and limx! 1 f (x) = 1
No HA
VA:
No VA because f is continuous

E. Increasing/decreasing
f 0 (x) = 13 3x2 (x3 + 1) 2/3
= x2 (x3 + 1) 2/3

f 0 (x) > 0
x2
() p >0
3
(x3 + 1)2
p
() 3 (x3 + 1)2 > 0

Always true.
f is increasing on ( 1, 1)
F. Local max/min
Since f 0 (x) 6= 0 for all x 2 R, there are no local maxima or minima (hence no global
maxima or minima).

G. Concavity and inflection points

d ⇥ 2 3 ⇤
f 00 (x) = x (x + 1) 2/3
>0
dx
2 2 3 5/3
() x (x + 1) (3x2 ) + 2x(x3 + 1) 2/3
3
2x4 2x
() 3 5/3
+ 3 >0
(x + 1) (x + 1)2/3
2x4 + 2x(x3 + 1)
()
(x3 + 1)5/3
2x
() >0
(x3 + 1)5/3
() x > 0

Thus
f 00 (x) > 0 () x > 0.
By the Concavity Test (§3.3), this means
• f is concave up on the interval {x | x > 0}.
• f is concave down on the interval {x | x < 0}.
Ex. 22. Sketch the graph of f (x) = x5/3 5x2/3 by hand.

A. Domain
R

B. Intercepts
(0, 0) and (5, 0):
x5/3 5x2/3 = 0 =) x2/3 (x 5) = 0

C. Symmetry
No symmetry

D. Asymptotes
No asymptotes

E. Increasing/decreasing
5 10 5
f 0 (x) = x2/3 x 1/3
= x 1/3
(x 2)
3 3 3
f 0 (x) > 0
5 1/3
() x (x 2) > 0
3
() x < 0 or x > 2
• f is increasing on the intervals {x | x < 0} and {x | x > 2}.
• f is decreasing on the interval {x | 0 < x < 2}.

F. Local max/min
p
Using the 1st Derivative Test, f (0) = 0 is a local max, and f (2) = 3 3 4 is a local min.

G. Concavity and inflection points


00 d 5 2/3 10 1/3
f (x) = x x .
dx 3 3
10 10
f 00 (x) = x 1/3
+ x 4/3
>0
9 9
if and only if x > 1
• f is concave up on the interval {x | x > 1}.
• f is concave down on the interval {x | x < 1}.
 Additional exercises

Ex. 23. For the function f whose graph is given, state the following.

(a) lim f (x) (b) lim f (x) (c) lim f (x) (d) lim f (x)
x!1 x! 1 x!1 x!3

(e) The equations of all asymptotes

Ex. 24. Find the limit or show that it does not exist.
p
4x + 3 x + 3x2
(a) lim (e) lim
x!1 5x 1 x!1 4x 1
p
3t2 + t 1 + 4x6
(b) lim (f) lim
t! 1 t3 4t + 1 x! 1 2 x3
p
p (g) lim 25t2 + 2 5t
4 x t!1
(c) lim p p p
x!1 2 + x (h) lim x2 + ax x2 + bx
x!1
(u2 + 1)(2u2 1) 1 ex
(d) lim (i) lim
u! 1 (u2 + 2)2 x!1 1 + 2ex
 Workbook Lesson 21
§4.7, Applied Optimization Problems
Last revised: 2021-03-23 11:41

Objectives
• Set up and solve optimization problems in several applied fields.

Warmup

Problem:
• We have a pipe cleaner 8 inches long.
• We want to bend the pipe cleaner into the shape of a rectangle.
• How can we make the space inside the rectangle as large as possible?

We’ll need to bend the pipe cleaner into four segments (the four sides of the rectangle).
w w w w

Since the pipe cleaner is 8 inches long, the perimeter of the rectangle will also be 8 inches long.

perimeter = 8 A(w) = (4 - w)w

perimeter = 8 A(w) = (4 - w)w

An 8-inch long pipe cleaner is bent into the shape of a rectangle.

What are the dimensions of the rectangle with maximum area?
An 8-inch
We want to make the area of the rectangle long pipe
as large cleaner is bent
as possible. into theashape
What’s of a rectangle.
formula for the area?
What are the dimensions of the rectangle with maximum area?
Area = length ⇥ width
A = `w
In order to use the tools of single-variable calculus, we need to write the area A in terms of a single
variable.
Can we write the length ` in terms of the width w?
2` + 2w = 8 (perimeter = 8)
2` = 8 2w
`=4 w
Now we can write the area A as a function of a single variable, w:
A(w) = (4 w)w

Question: What value of w maximizes A(w)?

To answer the question and solve the problem, see applet on iCollege: “Pipe cleaner”
 Optimization problems
We will now look at a variety of word problems in which a certain quantity Q (for instance, cost,
profit, distance, angle, area, volume) is to be maximized or minimized.
We express the quantity as a function, Q = Q(x), and find its extreme values by solving the
equation Q0 (x) = 0.
Now, most real-world quantities depend on more than one variable. Even something as simple as
the area of a rectangle depends on two variables, length and width: A = A(`, w). So, in order to
pose a given question as an extreme value problem, we will often have to do some work to write Q
as a function of one variable.

Exercises
Ex. 1. Find the area of the largest rectangle that can be inscribed in a semicircle
of radius r. (Hint: Let the base of the rectangle be a line segment in the x-axis
with the origin as its midpoint.)
Ex. 2. The sum of two positive numbers is 16. What is the smallest possible
value of the sum of their squares?
Ex. 3. What is the minimal vertical distance between the parabolas y1 = x2 + 1
and y2 = x x2 ?
Ex. 4. A box with a square base and open top must have a volume of 32, 000
cm3 . Find the dimensions (` ⇥ w ⇥ h) of the box that minimize the amount of
material used.
Ex. 5. A right circular cylinder is inscribed in a sphere of radius r. Find the largest
possible volume of such a cylinder.
Ex. 6. A piece of wire 10 m long is cut into two pieces. One piece is bent into a
square and the other is bent into an equilateral triangle. How should the wire be
cut so that the total area enclosed is a maximum? (A calculator will be needed.)

+ The setups for Ex. 1 and Ex. 3 are outlined in the applets “ Inscribed rectangle” and
“ Vertical distance” on iCollege.
 Optimization problems—General Strategy
In an optimization problem, a quantity Q (e.g., cost, profit, distance, angle, area, volume, . . . ) is to be
maximized or minimized.
We express the quantity as a function, Q = Q(x), and find its extreme values by solving the equation
Q0 (x) = 0.
+ The following process is a suggestion, not a requirement. You are free to organize your work on
these problems however you like, as long as your work makes sense.

#4. A box with a square base and open top must have a volume of 32, 000 cm3 . Find the dimensions of
the box that minimize the amount of material used.
¨ Write a legend. Draw a picture, if applicable. DRAW PICTURE FIRST
It may be helpful to avoid using equations in this Key:
step.
Q: amount of material used
b: side length of square base
h: height of box
V: volume of box

≠ Identify what the question asks for. Want: Dimensions of box for min value of Q

Æ Write any constraints you are given. Constraints:

(A constraint is a relation between the variables V = b2 h = 32, 000
you listed in Step 1 other than Q.)

Ø Use the given constraints to write a formula Q depends on b and h:

for Q as a function of 1 variable.
Q = b2 + 4bh

But h is a function of b:
2
h = 32, 000b

So

Q = b2 +4b(32, 000b 2
) = b2 +128, 000b 1

∞ Solve the extreme value problem for Q.

Q0 (b) = 2b 128, 000b 2 = 0
2b3 128, 000 = 0
p
b3 = 3 64, 000 = 40. . . so b = 40 is an extreme value.

Is b = 40 a local min or a local max? Draw sign chart. Test points: Q0 (1) < 0 and
Q0 (1, 000, 000) > 0, so by the First Derivative Test, b = 40 is a local min.
The dimensions of the box are
40 ⇥ 40 ⇥ 20 , because if b = 40,
± Give the information asked for in Step 2. 32, 000
then h = 32,000
402
= = 20.
1, 600
 Optimization Problems—Legends
A: Area of rectangle
r: Radius of semicircle (constant)
Ex. 1.
2x: Width of rectangle
y: Height of rectangle

x: 1st number
Ex. 2. y: 2nd number
Q: sum of their squares

y 1 , y2 : heights of the two parabolas

Ex. 3.
Q = Q(x): vertical distance between the two parabolas at a given x-coordinate

Ex. 5.

r = AB: radius of sphere (constant)

b = BC: radius of base of cylinder
x = AC: half the height of the cylinder (height of triangle shown)
Q = Q(x): volume of cylinder

#6.

a : length of wire bent into the square

1
s= 4
: side length of square
a
1
t = 3 (10 a) : side length of triangle
h : height of triangle
Q : total area
Ex. 1.
Ex. 2.
Ex. 3.
Ex. 5.
Ex. 6. A piece of wire 10 m long is cut into two pieces. One piece is bent into a square and the
other is bent into an equilateral triangle. How should the wire be cut so that the total area enclosed
is a maximum?
a 10-a

10

a : length of wire bent into the square

1
s= 4
: side length of square
a
1
t = 3 (10 a) : side length of triangle
h : height of triangle
Q : total area

Want: How to cut wire into pieces of length a and 10 a so that Q is maximized.

Constraints:

a  10
p opp. h
3 1 ⇡
h= 2
t = p
2 3
(10 a) (= sin 3
= =
hyp. t p 3
t sin = t
t 3 2
(Since an equilateral triangle is required, the angle at each
vertex of the triangle is 60 = ⇡3 .)

Equation for Q: p
3

Q = area of square + area of triangle

1
= s2 + th
2
1 2 1
= a + p (10 a)2
16 12 3

Solve the extreme value problem for Q:

1
Q0 (a) = a + 1
p
6 3
(10 a) · ( 1) = 0
8
1
+ 6p1 3 a =
8
a 5
p
3 3
⇣p ⌘
1
a + 1
p p3 a = 5
p
8 6 3 3 3 3
p
1 9 3 4 5
8 9
a+ 18 4
a = p
3 3
p
9+4 3 5
72
a = p
3 3
a= p120 ⇡ 4.34965.
9 3+12
 Apply Closed Interval Theorem for s 2 [0, 10]:

⇣ ⇡ 4.81125
Q(0) ⌘
Q 9p120
3+12
⇡ 2.71853
Q(10) = 6.25

Q is maximized when a = 10.

Answer: Don’t cut the wire! The entire 10 m of the wire should be bent into a square.
 Additional exercises

Ex. 7 (§4.7—#318). Find two positive integers a and b such that a + b = 10 and such that
a2 + b2 is maximized.

Ex. 8 (§4.7—#320). You have 400 ft. of fencing to construct a rectangular pen for cattle. What
are the dimensions of the pen that maximize the area?

Ex. 9 (§4.7—#315). To carry a box on an airplane, the length + width + height of the box
must be less than or equal to 62 in. Assuming the height is fixed, show that the maximum volume
is
2
V = 31 12 h h.
What height allows you to have the largest volume?

Ex. 10 (§4.7—#347). Find the point on the line y = 5 2x that is closest to the origin.

Ex. 11 (§4.7—#349). Find the point on the parabola

y = x2

that is closest to the point (2, 0).

Ex. 12 (§4.7—#324, 325). A patient’s pulse measures 70 bpm, then 80 bpm, then 120 bpm.
To determine an accurate measurement of pulse, the doctor wants to know what value minimizes
the expression
(x 70)2 + (x 80)2 + (x 120)2 .
(a) What value minimizes the above expression?
(b) In the previous problem, assume the patient was nervous during the third measurement, so
we only weight that value half as much as the others. What is the value that minimizes the
following expression?
(x 70)2 + (x 80)2 + 12 (x 120)2

Ex. 13 (§4.7—#355). You are the manager of an apartment complex with 50 units. When you
set rent at $800/month, all apartments are rented. As you increase rent by $25/month, one fewer
apartment is rented. Maintenance costs run $50/month for each occupied unit. What is the rent
that maximizes the total amount of profit?
 connected to the ground. The first pole is tall and
False. For every continuous nonlinear
n find the value that maximizes the the second pole is tall. There is a distance of
between the two poles. Where should the wire be anchored
to the ground to minimize the amount of wire needed?
alse. For every continuous nonconstant
ed, finite domain, there exists at least one
Ex. 14 (§4.7—#354). You are building five identical pens adjacent to each other with a total
or maximizes the function.
area of 1000m2 , as shown in the following figure. What dimensions should you use to minimize the
amount
g exercises, set up of fencing?
and evaluate each
lem.

a suitcase on an airplane, the length

t of the box must be less than or equal
ing the height is fixed, show that the

e is What height

e the largest volume?

onstructing a cardboard box with the

You then Ex. 15 (§4.7—#323).
cut equal-size squares 323.
You[T]areYou are moving
moving into ainto a new
new apartmentand
apartment and notice
notice that there is a corner
where the hallway there
narrows is
from a
8 corner
ft to 6 where
ft. Whatthe hallway
is the narrows
length of from
the longest item that can be
so you may fold the edges. What are the What is the length of the longest item that can
carried
box with the largest horizontally around the corner?
volume? be carried horizontally around the corner?

sitive integer that minimizes the sum of

s reciprocal.

sitive integers such that their sum is

324. A patient’s pulse measures
maximize the sum of their squares. To determine an

exercises, consider the construction of a accurate measurement of pulse, the doctor wants to know
area. what value minimizes the expression
What value
of fencing to construct a minimizes it?
or cattle. What are the dimensions of the
the area?
 Workbook Lesson 22
§4.8, L’Hôpital’s Rule
Last revised: 2020-09-29 12:43

Objectives
• Recognize when to apply L’Hôpital’s rule.
• Identify indeterminate forms produced by quotients, products, subtractions, and powers, and apply L’Hôpital’s
rule in each case.
• Describe the relative growth rates of functions.

Indeterminate quotients
Recall: We used an ad hoc geometric argument to show that
sin x
lim = 1.
x!0 x

0
“Plugging in” x = 0 to evaluate this limit yields . In general, a limit of the form
0
f (x)
lim where f (x) ! 0 and g(x) ! 0 as x ! a
x!a g(x)

is called an indeterminate form of type 0/0.

Similarly, we define an indeterminate form of type 1/1 to be a limit of the form
f (x)
lim where both f (x) ! ±1 and g(x) ! ±1 as x ! a
x!a g(x)

Shortly, we will see several other types of indeterminate form, including 1 1, 0 · 1, 1/1,
10 , and 11 . In this section, we’ll look at a systematic method, called L’Hôpital’s rule, that can
be used to evaluate indeterminate forms.

L’Hôpital’s Rule. Let f and g be di↵erentiable functions, with g 0 (x) 6= 0 near a (except possibly
at a). Here, we allow a = ±1. Suppose that

lim f (x) = 0 and lim g(x) = 0

x!a x!a

or that
lim f (x) = ±1 and lim g(x) = ±1.
x!a x!a

(In other words, we have an indeterminate form of type 0/0 or 1/1 .) Then

f (x) f 0 (x)
lim = lim 0
x!a g(x) x!a g (x)

if the limit on the right-hand side exists, is 1, or is 1.

 ln x
Ex. 1. Show that lim = 1.
x!1 x 1

ex
Ex. 2. Show that lim = 1. (“There is a struggle between the numerator and the denomina-
x!1 x2
tor.” Which one outraces the other as x ! 1?)

ln x
Ex. 3. Show that lim p = 0.
x!1 3 x
 tan(x) x 1
Ex. 4. Show that lim = .
x!0 x3 3

sin(x)
Ex. 5. Show that lim = 0.
x!⇡ 1 cos x

1 sin ✓
Ex. 6. Find lim .
✓!⇡/2 csc ✓
Solution:
1 sin ✓ cos ✓
lim = lim
✓!⇡/2 csc ✓ ✓!⇡/2 cot ✓ csc ✓
cos ✓
= lim
✓!⇡/2 cos ✓ 1
sin ✓ sin ✓

sin ✓
= lim cos ✓ sin ✓
✓!⇡/2 cos ✓
= lim sin2 ✓ = 0.
✓!⇡/2
 x3
Ex. 7. Find lim 2.
x!1 ex
Solution:
x3 3x
lim x 2 = lim 2
x!1 e x!1 2ex

3
= lim 2 = 0.
x!1 4xex

Notice that the limit

3
2lim
x!1 4xex

is not an indeterminate form: although the denominator approaches 1, the numerator is fi-
nite.

+ Expressions involving 1 and division by 0 are not numbers.

For example, we defined

lim f (x) = 1 (†)

x!a

to mean that f (x) grows larger than any finite number as x ! a. But (†) is not a statement
that two numbers are equal, if by the word “number” you mean a point on the real number
line.1
In general, expressions that involve infinity symbols or division by zero should be understood
as statements about the behavior of a function f (x) as we take a limit.

Indeterminate products
What is the limit of 2x ln x as x ! 0+ ?
The factors f (x) = 2x and g(x) = ln x have competing behaviors as x ! 0+ .
• As x ! 0+ , the factor 2x gets closer and closer to 0.
• But ln x ! 1 as x ! 0+ .
• It’s not obvious which behavior will win out.
• If 2x wins, the limit is 0. If ln x wins, the limit is 1.
• It’s also possible in principle that the two competing behaviors reach a stalemate, in which
the case the limit is some finite nonzero number.
1
It is possible to “extend” the real numbers so that we can perform certain arithmetic operations involving 1,
but we will not do this in the calculus sequence.
When we encounter such a struggle f (x) · g(x), we can rewrite the product as
f (x) g(x)
h(x) = or h(x) = . (††)
1/g(x) 1/f (x)
Then l’Hôpital’s Rule can be applied to the limit of the quotient.

Definition.
• An indeterminate form of type 0 · 1 is a limit of the form
⇥ ⇤
lim f (x) · g(x) where f (x) ! 0 and g(x) ! ±1 as x ! a.
x!a

• An indeterminate form of type 1 · 1 is a limit of the form

⇥ ⇤
lim f (x) · g(x) where both f (x) ! ±1 and g(x) ! ±1 as x ! a.
x!a

Ex. 8. Find lim+ x ln x.

x!0

Solution:
ln x
lim+ x ln x = lim+
x!0 x!0 1/x
1/x
= lim+
x!0 1/x2
1
= lim+
x!0 1/x
= lim+ ( x) = 0.
x!0

What if we had started by writing

x
lim+ x ln x = lim+ ?
x!0 x!0 1/ ln x
We can still apply l’Hôpital’s Rule, since the righthand limit is an indeterminate form of type
0/0. But taking the derivative of the numerator and denominator yields a more complicated
expression
1
.
1/x(ln x)2
When deciding which of the two ways in (††) to rewrite the product, we should try to choose the
option that leads to a simpler limit.
x2
Ex. 9. Find lim x3 e .
x!1
Solution:

x2 x3
lim x3 e = lim 2 = 0 (see above).
x!1 x!1 ex
Ex. 10. Find lim cot 2x sin 6x.
x!0
Solution:
sin 6x
lim cot 2x sin 6x = lim
x!0 x!0 tan 2x

6 cos 6x
= lim = 3.
x!0 2 sec2 2x

Indeterminate di↵erences
I claim that ⇥p p ⇤
lim x x 1 = 0.
x!1 |{z} | {z }
# #
1 1

Can you prove this?2 This example illustrates

p the p
idea of two functions competing with each other
in a limit. Here, we have a stalemate, as x and x 1 turn out to be approximately equal when
x grows large.
On the other hand, 
1 1
lim = 1.
x!0 x | {zx2}
4
|{z}
# #
1 1

1 1
In this example, ! 1 wins, and ! 1 loses.
x4 x2
An indeterminate form of type 1 1 is a limit of the form
⇥ ⇤
lim f (x) g(x) where both f (x) ! 1 and g(x) ! 1 as x ! a.
x!a

In order to evaluate an indeterminate di↵erence, we should try to rewrite the di↵erence as a fraction,
because l’Hôpital’s Rule applies only to indeterminate quotients.

For example, we find that


1 1 1 x2 (l’Hôpital) 2x 2
lim 4 = lim = lim = lim 2 = 0.
x!0 x x2 x!0 x4 x!0 4x 3 x!0 4x

2
Hint: p p
p p p p x+ x 1 1
x x 1= x x 1 p p =p p .
x+ x 1 x+ x 1
Ex. 11. Find lim sec x tan x .
x!(⇡/2)

Solution:
✓ ◆ ✓ ◆
1 sin x cos x
lim sec x tan x = lim = lim = 0.
x!(⇡/2) x!(⇡/2) cos x x!(⇡/2) sin x

Indeterminate powers
⇥ ⇤g(x)
A limit of the form lim f (x) is an indeterminate form of. . .
x!a
0
. . . type 0 if f (x) ! 0 and g(x) ! 0 as x ! a,
. . . type 10 if f (x) ! 1 and g(x) ! 0 as x ! a,
. . . type 11 if f (x) ! 1 and g(x) ! 1 as x ! a.

To evaluate limits of these types, we can either take logarithms, writing

⇥ ⇤g(x)
y = f (x) =) ln y = g(x) · ln f (x).
|{z} | {z }
# #
1 1

⇥ ⇤g(x)
or rewrite f (x) using the identity bp = ep ln b (recall that this identity follows from the cancel-
lation law for exp and ln).
⇥ ⇤g(x)
f (x) = eg(x)·ln f (x) = exp g(x) · ln f (x) .
|{z} | {z }
# #
1 1

p
x
Ex. 12. Find limx!0+ x .
Solution:
p
x
p
Set y(x) = x . Then ln y = x ln x, and

ln x 1/x
lim+ ln y = lim+ = lim 1
x!0 x!0 x 1/2 x!0+ 2
x 3/2

= lim+ 2x 1 x3/2
x!0
p
= lim+ 2 x = 0.
x!0

Now use the fact that y = eln y :

lim y = lim+ eln y = e0 = 1 .

x!0+ x!0
Ex. 13. Find limx!0 (1 2x)1/x .
Solution:

y : = (1 2x)1/x .
1
ln y = ln(1 2x).
x
ln(1 2x)
lim ln y = lim
x!0 x!0 x
2/(1 2x)
= lim = 2.
x!0 1

lim y = lim eln y = e 2

= 1/e2 .
x!0 x!0

Some theorems proved by l’Hôpital’s Rule

⇥ ⇤g(x)
Theorem If lim f (x) = 0 and lim g(x) = 1, then lim f (x) = 0.
x!a x!a x!a

(We might restate this informally by saying, “The form 01 is not indeterminate.”)

f (x + h) f (x h)
Theorem. If f 0 is continuous, then lim = f 0 (x).
h!0 2h
f (x + h) 2f (x) + f (x h)
Theorem. If f 00 is continuous, then lim = f 00 (x).
h!0 h2

Ex. 14. Prove the previous two theorems.

 Additional exercises

Ex. 15. Suppose that

lim f (x) = 0, lim g(x) = 0, lim h(x) = 1,
x!a x!a x!a

lim p(x) = 1, lim q(x) = 1.

x!a x!a

• Which of the following are indeterminate forms?

• For any limit that is not an indeterminate form, evaluate it if possible.
⇥ ⇤
f (x) h(x) p(x) (g) lim p(x) q(x)
(a) lim (c) lim (e) lim x!a
x!a g(x) x!a p(x) x!a q(x)
⇥ ⇤
(h) lim p(x) + q(x)
f (x) p(x) ⇥ ⇤ x!a
(b) lim (d) lim (f) lim f (x) p(x)
x!a p(x) x!a f (x) x!a

Ex. 16 (§4.8—#357, 359, 363, 365, 369, 371, 375, 377, 383, 385, 387). Evaluate the limit,
if possible. If L’Hôpital’s Rule can’t be applied, explain why not.
ex x2 sin(x) tan(x) (j) lim x ln x4
(a) lim (d) lim (g) lim +
x!0
x!1 xk x!1 1/x x!0 x3
2
x a (1 + x) 1 ex x 1 (k) lim x2 e x
(b) lim (e) lim (h) lim x!1
x!a x2 a2 x!0 x x!0 x2
x ⇡
(f) lim 1 x
(c) lim x1/x x!⇡ sin(x)
(i) lim x sin x1 (l) lim 1 x
x!1 x!1 x!1
Ex. 17. If an initial amount A0 of money is invested in an interest rate r compounded n times a
year, the value of the investment after t years is
⇣ r ⌘nt
A = A0 1 + .
n
If we let n ! 1, we refer to the continuous compounding of interest. Use L’Hôpital’s Rule to
show that if interest is compounded continuously, then the amount after t years is

A = A0 ert .
 Workbook Lesson 24
§4.10, Antiderivatives
Last revised: 2021-04-22 12:35

Objectives
• Find the general antiderivative of a given function.
• Explain the terms and notation used for an indefinite integral.
• State the power rule for integrals.
• Use antidi↵erentiation to solve simple initial-value problems.

Ex. 1. If F 0 (x) = 3x2 , what can F be? Give two di↵erent answers.

Definition. A function F is an antiderivative of f on an open interval I if F 0 (x) = f (x) for all

x in I.

We haven’t given you any formulas for finding antiderivatives yet. This was intentional.
• At this stage, you are asked to simply guess what the antiderivative is.
• Check your answer by di↵erentiating.
• If your answer is a little bit o↵, modify your guess and try again.

Ex. 2. Using trial and error, find an antiderivative of f .

(a) f (x) = 0
(b) f (x) = sin x
(c) f (x) = ex
(d) f (x) = 1 5x3
1
(e) f (x) = p
3
x 1
Recall:
(Section 4.4) If the derivative of a function f is 0 on an open interval I = (a, b), then f is
constant on I.

We can use the above result to prove the following fact:

Corollary. Let F be an antiderivative of f on an open interval I. Every antiderivative G of f on
I is of the form
G(x) = F (x) + C
for some C = const.
Proof:
Let F and G be antiderivatives of f on I. Then

F 0 (x) = f (x) and G0 (x) = f (x)

for all x in I, so
F 0 (x) G0 (x) = f (x) f (x) = 0.
Since
F 0 (x) G0 (x)
is the derivative of the function
F (x) G(x),
the Theorem tells us that

F (x) G(x) = C for some C = const.

Therefore,
G(x) = F (x) + D
for some D = const.
⇤

Definition. If F is an antiderivative of f on an open interval I, we call the expression

F (x) + C (C = const)

the most general antiderivative of f on I.

Ex. 3. Find the most general antiderivative of
(a) f (x) = 1
(b) f (x) = sec2 x
(c) f (x) = xn , where n 0
3
(d) f (x) = x
p
2x5 x
(e) f (x) = 4 sin x +
x
Commentary on Ex. 3:
An antiderivative must be defined on an interval.
A formula for an antiderivative for (d) is
1 2
F (x) = x + C,
2
but what interval should we use?
The given function, f (x) = x 3 , is not defined on ( 1, 1).
Domain of f : ( 1, 0) [ (0, 1)
An antiderivative of f on ( 1, 0) is
1 2
F (x) = x + C1 .
2
An antiderivative of f on (0, 1) is
1 2
F (x) = x + C2 .
2

Di↵erential equations
Definition. An equation that involves an unknown function and its derivatives is called a di↵erential
equation. An initial value problem is a di↵erential equation for which the output of the function
is specified for a single input value.

y 0 = x2
Ex. 4. Find y given
y(2) = 4
Solution:
1
y(x) = x3 + C
3
1
4 = y(2) = · 23 + C
3
4
=C
3

1 4
y(x) = x3 +
3 3
 v(t) = s0 (t) = sin t cos t
Ex. 5. Find y given
s(0) = 0
Solution:

s0 (t) = sin t cos t

s(t) = cos t sin t + C
0 = cos 0 sin 0 + C
1=C

s(t) = cos t sin t + 1

Rectilinear motion
Recall:
d/dt d/dt
s(t) v(t) a(t)
position velocity acceleration

We can reverse the arrows by antidi↵erentiating.

• If acceleration and the initial position s(0) and speed v(0) are known, the position function
s(t) can be found by antidi↵erentiating twice.

During World War II, mechanical devices aboard rockets were used to carry out the process of
antidi↵erentiating twice in order to guide the missile to its target.
While the position of the rocket as it flew could not be determined by the technology of the day,
acceleration could be measured mechanically, as the following quote from a novelist (and former
engineering student) describes:

“a little pendulum was kept centered by a magnetic field. During launch, pulling gs,
the pendulum would swing aft, o↵ center. It had a coil attached to it. When the coil
moved through the magnetic field, electric current flowed in the coil. As the pendulum
was pushed o↵ center by the acceleration of the launch, current would flow—the more
acceleration, the more flow. So the Rocket . . . sensed acceleration first. . . . To get to
distance from acceleration, the Rocket had to [antidifferentiate] twice—needed a
moving coil, transformers, electrolytic cell, bridge of diodes, one tetrode. . . ”
—Thomas Pynchon, Gravity’s Rainbow
Ex. 6. A particle moving in a straight line has acceleration

a(t) = 12t 6.

Its initial velocity is v(0) = 6 cm/s and its initial displacement is s(0) = 9 cm.
(a) Find its position function.
(b) When is the first time after the initial time t = 0 that the displacement s(t) is 0?
(a)
v 0 (t) = 12t 6
2
v(t) = 6t 6t + C
6 = v(0) = C
s0 (t) = 6t2 6t 6
s(t) = 2t3 3t2 6t + C
9 = s(0) = C
s(t) = 2t3 3t2 6t + 9
 (b)
s(t) = 2t3 3t2 6t + 9
= t2 (2t 3) 3(2t 3) factor out (2t 3)
= (t2 3)(2t 3) = 0

p 3
t = ± 3 = ±1.732050 . . . or t= = 1.5
2

Ans.: 1.5 seconds after time t = 0.

Ex. 7. A stone is dropped o↵ a cli↵ and hits the ground at a speed of 120 ft/s. What is the height
of the cli↵?
Assume downward acceleration due to gravity is the constant

g = 32 ft/s2 .

Solution:
The acceleration is a constant function:

a(t) = 32

Assume time is t = 0 when the stone hits the ground.

v(t) = 32t + C
120 = v(0) = C
v(t) = 32t 120

Assume displacement is s(0) = 0 at time t = 0.

s(t) = 16t2 120t + D
0 = s(0) = D
s(t) = 16t2 120t

When was the stone at rest? That is, at what time t = T did we have v(T ) = 0?
0 = v(T ) = 32T 120
120 15
T = =
32 4
15
seconds before the stone hit the ground.
4
15
What was the displacement at time t = T = 4
?
15 225 15
s(T ) = s 4
= 16 · 16
120 · 4
= 225 + 450 = 225 ft .
Ex. 8. Show that the displacement at time t for rectilinear motion with constant acceleration A,
initial velocity v0 , and initial displacement s0 is
s = 12 At2 + v0 t + s0 .
Solution:

a(t) = v 0 (t) = A
v(t) = At + C1
v0 = v(0) = C1

v(t) = At + v0
v(t) = s0 (t) = At + v0
s(t) = 12 At2 + v0 t + C2
s0 = s(0) = C2

s(t) = 12 At2 + v0 t + s0 .

Ex. 9. A stone is thrown downward at a speed of 5m/s from a height of 450m above the ground
How long does it take to reach the ground? (Use g = 9.8m/s2 .)
Solution:
We could use the formula we got in the previous exercise, but we’ll do the work from scratch
for practice.
a(t) = g = 9.8 (Note sign.)
v(t) = 9.8t 5
s(t) = 4.9t2 5t + 450
We solve 0 = s(t) = 4.9t2 5t + 450 by the Quadratic Formula
p p
( 5) ± 25 4( 4.9)(450) 5 ± 8845
t= = .
2( 4.9) 9.8

We throw out the solution for negative time t < 0.

p p p
+ Without calculator: Note p > 25 = 5, so 5
8845 8845 > 0. Since the denominator is
5 8845
negative, the positive solution is .
9.8
p
5+ 8845
⇡ 9.0865 seconds.
9.8
 The indefinite integral

The act of finding an antiderivative of a function f is called integrating f .

For a function f and an antiderivative F , the family of functions

F (x) + C,

where C is any real number, is often called the family of antiderivatives of f .

It is also called the indefinite integral of f —in symbols,
Z
f (x) dx = F (x) + C

Methods for antidi↵erentiation

+ The definition of the derivative came with a formal process—the definition of the derivative—
that could be used to find the derivative of a function using algebra and limits. There is no such
process for finding the indefinite integral.
We find an antiderivative of a function f either
• by inspection, i.e. by thinking of a function whose derivative is f ,
• using a table, e.g. in Appendix A of your textbook,
• with special techniques such as “u-substitution” and “integration by parts” (these will be
discussed later in the calculus sequence) or
• “numerically,” i.e. by a process of approximation (often done by computer).

Rp Rp
Question. Can you evaluate the integrals 1 + 4x2 dx and sin x dx by inspection?
Answer: Almost certainly not!
We used an app (Mathematica) to find the first integral—you are certainly not expected to come
up with this! Z p
1 p 1 p
1 + 4x2 dx = x 1 + 4x2 + log( 1 + 4x2 + 2x) + C
2 4

The second integral simply cannot be expressed in terms of elementary functions.

Z p
sin x dx = ? No elementary solution.
A table that gives the antiderivatives of some common functions is provided at the end of this
document. Some of these “integration rules” should be memorized—for example, you should
memorize the formula Z
1
xn dx = xn+1 + C.
n+1

But if you hate memorization, it is recommended that when you are asked to find an antiderivative
of a function f , just ask yourself:

The derivative of what is f ?

You can always guess, then check by di↵erentiating.

• If your guess is a little o↵, adjust your guess and try again.
• If your guess is way o↵, and cannot be salvaged, it makes sense to consult the table of
integration formulas and memorize the appropriate rule for future use.

Ex. 10. Verify that each of the provided integration rules logically follows from the corresponding
di↵erentiation rule indicated in the table.

Additional exercises

Ex. 11 (§4.10—#465, 467, 469). Show that F (x) is an antiderivative of f (x).

(a) F (x) = 5x3 + 2x2 + 3x + 1, f (x) = 15x2 + 4x + 3
(b) F (x) = x2 ex , f (x) = ex (x2 + 2x)
(c) F (x) = ex , f (x) = ex
Ex. 12 (§4.10—#470–473). Find the most general antiderivative of the function.
1 • f (x) = ex + 3x x2
• f (x) = +x
x2
• f (x) = ex 3x2 + sin(x) • f (x) = x 1 + 4 sin(2x)

Ex. 13 (§4.10—#475, 477, 479, 481, 483, 489). Find an antiderivative of the function.

• f (x) = x + 12x2 • f (x) = sec2 (x) + 1

p 3
• f (x) = x • f (x) = sin2 (x) cos(x)
x1/3
• f (x) = • f (x) = 12 e 4x
+ sin(x)
x2/3
Ex. 14 (§4.10—#491, 493, 495). Evaluate the integral.
Z Z Z
3x + 2 + 2 p p
• sin(x) dx • dx • 4 x + 4 x dx
x2

Ex. 15 (§4.10—#499, 503). Solve the initial value problem.

2 x2
(a) f 0 (x) = x 3 , f (0) = 1 (b) f 0 (x) = , f (1) = 0
x2 2
Ex. 16 (§4.10—#505, 507). Find two di↵erent possible functions f .
(a) f 00 (x) = e x
(b) f 000 (x) = cos(x)

Ex. 17. The graph of a function f is shown. Which graph is an antiderivative of f and why?

g f h

Ex. 18. The graph of a function y = f (x) is shown. Make a rough sketch of its antiderivative F ,
given that F (0) = 1.
y = f (x )

x
1 2 3 4
 Integration rules

Integration rule Corresponding di↵erentiation rule

Z
1 d⇥ n
xn dx = xn+1 + C for n 6= 1 x ] = nxn 1
n+1 dx
Z
k dx = kx + C for k = const Special case of previous di↵erentiation rule (n = 1)

Z
1 d⇥ ⇤ 1
dx = ln |x| + C ln |x| =
x dx x
Z
d ⇥ x⇤
ex dx = ex + C e = ex
dx
Z
1 x d ⇥ x⇤
bx dx = b + C for 0 < b 6= 1 b = bx ln(b)
ln(b) dx
Z
d⇥ ⇤
cos(x) dx = sin(x) + C sin(x) = cos(x)
dx
Z
d⇥ ⇤
sin(x) dx = cos(x) + C cos(x) = sin(x)
dx
Z
d⇥ ⇤
sec2 (x) dx = tan(x) + C tan(x) = sec2 (x)
dx
Z
d⇥ ⇤
csc(x) cot(x) dx = csc(x) + C csc(x) = csc(x) cot(x)
dx
Z
d⇥ ⇤
sec(x) tan(x) dx = sec(x) + C sec(x) = sec(x) tan(x)
dx
Z
d⇥ ⇤
csc2 (x) dx = cot(x) + C cot(x) = csc2 (x)
dx
Z
1 d⇥ ⇤ 1
dx = tan 1 (x) + C tan 1 (x) =
1 + x2 dx 1 + x2
 Workbook Lesson 25
§5.1, Approximating Areas
Last revised: 2020-09-29 12:43

Objectives
• Use sigma (summation) notation to calculate sums and powers of integers.
• Use the sum of rectangular areas to approximate the area under a curve.
• Use Riemann sums to approximate area.

The method of exhaustion

In antiquity, the Greek mathematician Archimedes was fascinated with calculating the areas of
irregular shapes, such as the amount of space enclosed by a curve.
He used a process that has come to be known as the method of exhaustion (see Lesson 1), which
used smaller and smaller shapes, the areas of which could be calculated exactly, to fill an irregular
region and thereby obtain closer and closer approximations to the total area.
In this process, an area bounded by curves is filled with simple shapes like rectangles or triangles.
The areas of these shapes are individually easy to calculate. These areas are then summed to
approximate the area of the curved region.
In this section, we develop techniques to approximate the area between a curve defined by a function
f and the x-axis on a closed interval [a, b].
Like Archimedes, we first approximate the area under the curve using shapes of known area—namely,
rectangles.
By using smaller and smaller rectangles, we get closer and closer approximations to the area. Taking
a limit allows us to calculate the exact area under the curve.
In this section, we develop techniques to approximate the area between a curve y = f (x) and the
x-axis on a closed interval [a, b]. Like Archimedes, we first approximate the area under the curve
using shapes of known area (namely, rectangles). By using smaller and smaller rectangles, we get
closer and closer approximations to the area. Taking a limit allows us to calculate the exact area
under the curve.
We’ll begin by introducing some notation to make the calculations easier to follow. We then
consider the case when f (x) is continuous and nonnegative for all x. Later in the chapter, we’ll
consider more general cases.
 Sigma notation
The Greek capital letter ⌃, “sigma,” is used to express long sums of values in a compact form.
Say we want to add all the integers from 1 to 20. Without sigma notation, we have to write

1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 + 10 + 11 + 12 + 13 + 14 + 15 + 16 + 17 + 18 + 19 + 20.

We could probably skip writing a few terms and write

1 + 2 + 3 + 4 + · · · + 19 + 20,

which is better, but still tedious.

With sigma notation, we write this sum as
20
⌃ k (?)
k=1

which is both shorter and easier on the eyes.

In the above notation (?), called sigma notation (or sigma notation), the variable k is called the
index. Each term is evaluated, then we sum all the values, beginning with the value when k = 1
and ending with the value when k = 20.
We sometimes denote the terms to be added up in the form by ak (or bk , etc.), where k is the
index. Thus we might declare
k⇡
ak = sin
2
and write
6
⌃ ak .
k=1

6
k⇡
Ex. 1. Evaluate ⌃ ak , where ak = sin .
k=0 2
We can use any letter we like for the index. Mathematicians most often use i, j, k, m, and n for
indices.
The index is used only to keep track of the terms to be added. It does not factor into the calculation
of the sum itself. The index is therefore called a dummy variable. For example,
20 20
⌃ j and ⌃ k
j=1 k=1

mean exactly the same thing, 1 + 2 + 3 + · · · + 20.

+ The values of the index are always understood to be integers.

Ex. 2.
(a) Write the sum of terms 3k for k = 1, 2, 3, 4, 5 in sigma notation and evaluate.
1 1 1 1
(b) Write the sum + + + in sigma notation.
2 4 8 16
1 1 1 1 1
(c) Write the sum + + + + in sigma notation.
4 9 16 25 36
5
(d) Evaluate ⌃ (x + 1).
x=0

Rules for sigma notation

Ex. 3. Which of the following “rules” is true?
n
X n
X
cak = c ak
k=1 k=1

n
X n
X n
X
(ak + bk ) = ak + bk
k=1 k=1 k=1

n
X n
X n
X
(ak bk ) = ak bk
k=1 k=1 k=1

n n
! n
!
X X X
(ak · bk ) = ak · bk
k=1 k=1 k=1

Answer: The last formula is not true. (The remaining three are correct.)
Ex. 4. True or False: If the terms ak are all defined, then
10
X 9
X 11
X
ak = ak+1 = ak 1
k=1 k=0 k=2
no matter what the values of the ak are.

Some special sums

Ex. 5. What is the sum of the first n natural numbers? That is, find the value of
n
X
S= k.
k=1
Solution:
This solution is very creative. It has been claimed that the mathematician Carl Gauss discov-
ered this solution in elementary school. You are not expected to come up with such a novel
technique—we will simply use the formula which the following argument proves.
Double the sum, and rearrange terms:
2 ⇥ S = S + S = (1 + 2 + 3 + · · · + (n 2) + (n 1) + n) + (n + (n 1) + · · · + 3 + 2 + 1)
= (1 + n) + (2 + (n 1)) + (3 + (n 2)) + · · · + ((n 1) + 2) + (n + 1)
| {z }
n bracketed terms

= n(n + 1).

Out[136]=

Do you see it?

Since 2S = n(n + 1), we have
n(n + 1)
S= .
2
n
X n(n + 1)
k=
i=1
2
Two other useful formulas (which we will not prove, but may be required to complete some of the
exercises to come) are:
n
X n(n + 1)(2n + 1)
k2 =
i=1
6

n
X n2 (n + 1)2
k3 =
i=1
4

Ex. 6. Write using sigma notation and evaluate:

(a) The sum of the terms (k 3)2 for k = 1, 2, 3, 4, 5, 6.
(b) The sum of the terms (k 3 k 2 ) for k = 1, 2, 3, 4, 5, 6.
(c) The sum of the terms 4 + 3k for k = 1, 2, 3, . . . , 100.
 Approximating the area under a curve
Now that we have the necessary notation, we return to the problem at hand: approximating the
area under a curve.
Let f be a continuous, nonnegative function defined on the closed interval [a, b]. We want to
approximate the area A bounded by
• the curve y = f (x) above,
• the x-axis below,
• the line x = a on the left, and
• the line x = b on the right.
We call this area the area under the curve y = f (x) from a to b.

Definition. Let a = x0 < x1 < x2 < · · · < xn 1 < xn = b.

• The collection of n subintervals [x0 , x1 ], [x1 , x2 ], . . . , [xn 1 , xn ] is a partition P of [a, b].
• We write xk = xk xk 1 .
�� �� �� �� �� ��
� �

• The partition is regular if all the subintervals have the same width, in which we case we
write x for the shared width.
�� �� �� �� �� �� �� �� �� ��
� �

• For each subinterval [xk 1 , xk ], k = 1, 2, 3, . . . , n, choose any point x⇤k 2 [xk 1 , xk ].

The numbers x⇤k are called sample points for the partition P.
�� ��* �� ��* *
�� �� ��
� �

Ex. 7. What is the width of each subinterval in a regular partition of the interval [a, b] with n
subintervals?
Taking the sample points x⇤k as inputs to the function f (x) gives us the heights of rectangles, one
rectangle for each subinterval:
f (x )=x 2
400

�(�*� )300

200
�(�*� )
100
�(�*� ) �� �� �� ��
5 10 15 20
x
� �
-100
�*� �*� �*�

If we take the sample points to be the right endpoints of the subintervals [xk 1 , xk ], then the total
area of all the rectangles is a right-endpoint approximation of the area under the curve from a
to b:
f (x )=x 2
�(�*� )400

300

�(�*� )
200

�(�*� )100
�� �� �� ��
5 10 15 20
x
� �*
-100
�*� �*� ��

A left-endpoint approximation of the area under the curve from a to b is defined analo-
gously:
f (x )=x 2
�(�*� )400

300

�(�*� )
200

�(�*� )100
�� �� �� ��
5 10 15 20
x
� �
-100
�*� �*� �*�
The area of each rectangle is

base ⇥ height = (xk xk 1 ) · f (x⇤k )

= f (x⇤k ) xk .

The total area is n

X
f (x⇤k ) xk .
k=1

Ex. 8. Two of the three sums written below is a left-endpoint approximation under the curve
y = f (x) on [a, b]. The remaining sum is a right-endpoint approximation. Which is which?
Explain your reasoning.
n
X n
X n 1
X
f (xk 1 ) xk f (xk ) xk f (xk ) xk+1
k=1 k=1 k=0

Ex. 9.
(a) Find the right-endpoint approximation of the area under the curve y = x2 from x = 0 to
x = 20 with 5 rectangles.
(b) Sketch the rectangles in part (a) on the graph of y = x2 .
(c) Is this an overestimate or an underestimate of the actual area under the curve?
Ex. 10.
(a) Find the right-endpoint approximation of the area under the curve y = x2 from x = 4 to
x = 0 with 4 rectangles.
(b) Sketch the rectangles in part (a) on the graph of y = x2 .
(c) Is this an overestimate or an underestimate of the actual area under the curve?

Riemann sums

Let P be a partition of an interval [a, b]. In symbols, we’ll write:

P: a = x 0 < x 1 < x2 < · · · < x n 1 < xn = b

Suppose f is a function defined on [a, b]. A Riemann sum for the function f and the partition P
is a sum of the form n
X
f (x⇤k ) xk
k=1

where the x⇤k are sample points for the partition P (that is, each x⇤k is a number in the interval
[xk 1 , xk ].

Notice that a Riemann sum depends not only on the function f and the interval [a, b], but also on
the choice of partition P and the sample points x⇤k chosen for that partition.
That is, if I choose a certain partition and sample points, and you choose a di↵erent partition and
di↵erent sample points, then it’s unlikely that our Riemann sums will be equal, even when we’re
both approximating the area under the same curve y = f (x) over the same interval [a, b].

Our ultimate goal is not to approximate the area, but to find the exact area. To this end, we take
more and more, thinner and thinner rectangles. As the number of rectangles approaches 1, the
width of the rectangles approaches 0.

We may express this “limiting process” (see Lesson 1) by writing, “the limit of the Riemann sums
as the number of rectangles approaches 1”—in symbols,
n
X
lim f (x⇤k ) xk .
n!1
k=1

Or we might prefer to write, “the limit of the Riemann sums as the rectangles’ widths approach
0”—in symbols,
Xn
lim f (x⇤k ) xk .
max xk !0
k=1

However we write it, we are dealing here with a type of limit that is rather peculiar, because
if I take the limit of, say, left-endpoint approximations, and you take the limit of, say, right-
endpoint approximations, then there is no reason for us to believe that our limits will turn out to
be equal.
Fortunately, the following theorem—whose proof is beyond the scope of this class—guarantees that
whatever partitions we choose as the number of rectangles gets larger, and whatever sample points
we pick for each partition, the limit of the Riemann sums is “unique.” (Here, by “unique,” we
mean that if two di↵erent people compute the limit in di↵erent ways, then the values they get will
be equal.)
Theorem. For any function f that is continuous on the interval [a, b], if the limit of the Riemann
sums n
X
lim f (x⇤k ) xk
n!1
k=1

exists for some particular choice of sample points and partitions, then the limit exists for every
choice of sample points and partitions, and in every case, the value of the limit is the same.
If this limit exists and is a real number, we say f is integrable on the interval [a, b]. (We will
return to the idea of “integrability” in the next lesson.)
 Types of Riemann sums
We’ve already seen two ways to pick sample points—by choosing the left or right endpoints of each
subinterval of a partition. We now mention three other popular ways of picking sample points.

Definition. Let f be a function on [a, b] and let P : a = x0 < x1 < x2 < · · · < xn 1 < xn = b be
a partition of [a, b]. Consider the Riemann sum
n
X
f (x⇤k ) xk . (†)
k=1

• If we choose x⇤k so that f (x⇤k ) is the maximum value of f on [xk 1 , xk ]—in symbols,

f (x⇤k ) = max f (x)

xk 1 xxk

—for k = 1, 2, . . . , n, then the Riemann sum (†) is called an upper sum.

n
X
f (x⇤k ) xk .
k=1

• If we choose x⇤k so that f (x⇤k ) is the minimum value of f on [xk 1 , xk ]—in symbols,

f (x⇤k ) = min f (x)

xk 1 xxk

—for k = 1, 2, . . . , n, then the Riemann sum (?) is called a lower sum.

+ For an increasing function f , an upper sum is the same as a right-endpoint approxima-

tion.

Ex. 11. Explain why the previous sentence is true. Then make an analogous statement that’s true
for decreasing functions.
Ex. 12.
(a) Find and evaluate a lower sum with n = 4 rectangles that approximates the area under the
curve y = 10 x2 on [0, 1].
(b) Sketch the approximation.

Ex. 13.
(a) Find and evaluate an upper sum with n = 4 rectangles that approximates the area under the
curve y = 10 x2 on [0, 1].
(b) Sketch the approximation.
Ex. 14 (Midpoint approximation).
(a) Taking the sample points x⇤k to be the midpoints of the subintervals of the partition

a = x0 < x1 < x2 < · · · < xn = b,

find and evaluate a Riemann sum with n = 5 rectangles that approximates the area under the
curve y = 3x + 1 on [0, 10].
(b) Sketch the approximation.

Ex. 15 (The limit of the approximations).

(a) Write a right-endpoint approximation that approximates the area under the curve y = x2 from
a = 0 to b = 2 with n rectangles and a regular partition.
Hints:
b a 2
• x= = .
n n
• The right endpoint of the k th subinterval is
2 2k
xk = a + k · = .
n n

• The function value at the right endpoint is therefore

2k 4k 2
f (xk ) = f n
= .
n2

(b) Then take the limit as n ! 1 to find the exact area under the curve.
 Additional exercises

Ex. 16 (§5.1—#1). Are the given sums equal or unequal?

10 10 10 9
(a) ⌃ i and ⌃ k (c) ⌃ j(j 1) and ⌃ (k + 1)k
i=1 k=1 j=1 k=0

10 15 10 9
(b) ⌃ k and ⌃ (k 5) (d) ⌃ j(j 1) and ⌃ (k 2 k)
k=1 k=6 j=1 k=0

Ex. 17 (§5.1—#2, 3). Evaluate the sums

10 10
(a) ⌃ k (b) ⌃ k 2
k=5 k=5

Ex. 18 (§5.1—#19). Compute the left-endpoint Riemann sum L8 with 8 rectangles for

f (x) = x2 2x + 1

on [0, 2]. Illustrate with a graph.

Ex. 19 (§5.1—#23). Compute the left-endpoint Riemann sum and the right-endpoint Riemann
sum with 6 rectangles (L6 and R6 , respectively) for
p
f (x) = 9 (x 3)2

on [0, 6], and compare the values of L6 and R6 . Illustrate with a graph.

Ex. 20 (§5.1—#27). Express the left-endpoint sum R100 for f (x) = ln(x) on [1, e], but do not
evaluate the sum.

Ex. 21 (§5.1—#39). The following table gives the approximate increase in mean sea level in
inches over 20 years starting in the given year. Estimate the net change in mean sea level from
1870 to 2010.
starting year 1870 1890 1910 1930 1950 1970 1990
20-year change 0.3 1.5 0.2 2.8 0.7 1.1 1.5
Data source: Church & White, Sea-level rise from the late 19th to the early 21st century, Surv Geophys 32 (2011), pp. 585–602
 is 1 unit and its height is The base of the outer
triangle is and the height is

Use this information to argue that the area

Ex. 22 (§5.1—#59). A unit circle is made up of n sectors equivalent to the inner sector in the
figure.
of a unit circle is equal to π.

The base of the inner triangle is

b=1
unit and its height is
⇡
h = sin n
.
The base of the outer triangle is
⇡ ⇡ ⇡
B = cos n
+ sin n
tan n

and its height is

2⇡
H = B sin n
.
Use this information to argue that the area of a unit circle is equal to ⇡.
 Workbook Lesson 26
§5.2, The Definite Integral
Last revised: 2021-04-14 07:06

Objectives
• State the definition of the definite integral.
• Explain the terms integrand, limits of integration, and variable of integration.
• Explain when a function is integrable.
• Describe the relationship between the definite integral and net area.
• Use geometry and the properties of definite integrals to evaluate them.
• Calculate the average value of a function.

Definition of the definite integral

Recall: Last time, we saw that the area between a curve y = f (x) and the x-axis—which is usually
called the “area under the curve”—can be found as the the limit of Riemann sums:
Xn
lim f (x⇤i ) x.
n!1
i=1

However, the phrase “area under the curve” only makes sense if the values of f (x) are nonnegative—
that is, if the graph of f never dips below the x-axis.
If f has both positive values and negative values, the Riemann sum can include the area of rectangles
that lie below the x-axis.

We now lift the restriction that f (x) 0 for all x in [a, b], and define the definite integral to be
the limit of the Riemann sums.
Definition. For a function f defined on an interval [a, b], we define the definite integral of f
from a to b to be Z b n
X
f (x) dx = lim f (x⇤i ) x.
a n!1
i=1
Z
+ The integration symbol is an elongated S, meant to suggest summation (which is happening
behind the scenes in the Riemann sums whose limit is the definite integral).
 Z b
+ Think of the symbols and dx in the definite integral to be like parentheses—it is an error
Z a Z b
to write and omit the dx at the end. The expression f (x) between and dx is called the
a
integrand.

+ If we use a variable other than x as the input to the function f , the expression dx should be
changed accordingly. For example, the symbols
Z b Z b
f (t) dt and f (x) dx
a a
mean the same thing.
Z b
+ The variable ⇤ in the definite integral f (⇤) d⇤ is called the variable of integration.
a
The numbers a and b are called the limits of integration.

Net signed area and total area

Initially, we interpreted “area under the curve” only for continuous functions f that were everywhere
nonnegative.
Now that we are allowing the curve y = f (x) to dip below the x-axis, how can we interpret the
definite integral in terms of area?
Intuitively, the answer is simple: simply subtract the area under the x-axis (shaded below in purple)
from the area above the x-axis (shaded below in orange).

Definition. The net signed area “under” the curve y = f (x) over [a, b] is defined to be
A1 A2
where A1 is the area of the portion of the region between the curve y = f (x) (a  x  b) and the
x-axis that lies above the x-axis, and A2 is the area of the portion that lies below the x-axis.
Theorem. The definite integral of a continuous function f from a to b is equal to the net signed
area over [a, b]:
Z b
f (x) dx = A1 A2 .
a
Ex. 1. The graph of h consists of two straight lines and a semicircle. Evaulate the integrals by
interpreting them in terms of areas.
h(x)
6

x
2 4 6 8

-2

Z 3 Z 6 Z 8
(a) h(x) dx (b) h(x) dx (c) h(x) dx
0 3 0

Ex. 2. Recall that the formula for the area of a trapezoid is

y = 2x+3
12
a+b
A= h,
2 10

where h is the distance between the parallel sides of lengths 8

a and b.
6
Use the formula to find
4
Z 4
2x + 3 dx. 2
2

x
1 2 3 4 5
 Z 10
Ex. 3. Evaluate |x 5| dx.
0

Ex. 4.
(a) Write a formula for a Riemann sum using n rectangles and a right-endpoint approximation of
f (x) = 2x 1 on the interval [0, 3].
Z 3
(b) Use the definition of the integral to evaluate 2x 1 dx.
0
Solution (Ex. 4):
Rb
By definition of the definite integral a
f (⇤) d⇤,

Z b Z 3 n
f (x) dx = 2x 1 dx = lim ⌃ 2x⇤k 1 x,
a 0 n!1 k=1

where x⇤1 , . . . , x⇤n are sample points for a partition 0 = x0 < · · · < xn = 3 of [0, 3].
For a right-endpoint approximation, we take:
b a ✓ ◆
x= x⇤1 = right endpoint of the
= x1
n first subinterval [x0 , x1 ]
✓ ◆
x0 = a x⇤2 = right endpoint of the
= x2
next subinterval [x1 , x2 ]
x1 = a + x
..
x2 = a + 2 x .
✓ ◆
.. right endpoint of the
. x⇤k = k th subinterval [xk 1 , xk ] = xk
xk = a + k x
..
.. .
. ✓ ◆
right endpoint of the
xn = a + n x x⇤n = last subinterval [xn 1 , xn ] = xn

We now substitute and evaluate the limit, using a special sum formula along the way:
Z 3 n n 3 n 3
f (x) dx = lim ⌃ 2x⇤k 1 x = lim ⌃ 2xk 1 · = lim ⌃ 2k x 1 ·
0 n!1 k=1 n!1 k=1 n n!1 k=1 n
✓ ◆
n 6k 3
= lim ⌃ 1 ·
n!1 k=1 n n
✓ ◆
n 18k 3
= lim ⌃
n!1 k=1 n2 n
✓ ◆
n 18k n 3
= lim ⌃ ⌃
n!1 k=1 n2 k=1 n
✓ ◆
18 n
= lim 2 ⌃ k 3
n!1 n k=1

18 n(n + 1)
= lim 3
n!1 n2 2
18(n + 1)
= lim 3
n!1 2n
=9 3= 6
Ex. 5.
n
X 1 x2i
Express the limit lim x as a definite integral on the interval [2, 6].
n!1
i=1
4 + x2i
 Total area
Definition. Let A1 and A2 be as above. The total area “under” the curve y = f (x) over [a, b]
is Z b
f (x) dx = A1 + A2 .
a

Net signed area Total area

Ex. 5. Find the total area between f (x) = x 2 and the x-axis over the interval [0, 6]. (Hint: Sketch
the graph!)

Ex. 6. Find the signed area and the total area for g(x) = 1 |x| over the interval [0, 4].
(Hint: Sketch the graph!)
 Conditions for integrability
If the limit Z n
b X
f (x) dx = lim f (x⇤i ) x.
a n!1
i=1

exists, we say f is integrable on [a, b].

When can we be guaranteed that this limit exists? The following theorem gives a useful, but
somewhat restrictive condition:
Theorem. If f is continuous on [a, b], then f is integrable on [a, b].
Counterexample. The area under the curve y = x for x 0 is infinite. Why is this not a violation
of the Theorem?
We can relax the continuity requirement as follows:
Theorem. If f has a finite number of jump discontinuities, but is continuous at x for every other
x 2 [a, b], then f is integrable on [a, b].
This makes intuitive sense: the area under a graph with finitely many jump discontinuities can be
split up into pieces that don’t include a discontinuity, then the area of the pieces can be added.
7

0
0 1 2 3 4 5 6 7

Properties of the definite integral

The definition of a Riemann sum makes sense even if a > b. To see this, swap a and b. Then
b a a b
x= becomes = x instead. Then, when we take the limit on both sides of the
n n
equation !
Xn Xn
⇤ ⇤
f (xi )( x) = f (xi ) x
i=1 i=1
we get
Z a Z b
f (x) dx = f (x) dx. (1)
b a

Now, what happens if a = b?

 Z a
f (x) dx = 0. (2)
a

Some additional properties of the definite integral (assuming all integrals exist):
Z b Z b
cf (x) dx = c f (x) dx for c = const (3)
a a
Z b Z b Z b
⇥ ⇤
f (x) + g(x) dx = f (x) dx + g(x) dx (4)
a a a
Z b Z c Z c
f (x) dx + f (x) dx = f (x) dx (5)
a b a

Ex. 6. Prove that Z b

c dx = c(b a) for c = const.
a

(Use rule (3), and interpret the definite integral as area under a curve.)

Ex. 7. Prove that Z Z Z

b ⇥ ⇤ b b
f (x) g(x) dx = f (x) dx g(x) dx
a a a

Hint: Use rule (3), rule (4), and the definition of subtraction.
 Z 1
Ex. 8. Find (4 + 3x2 ) dx.
0
Z 1
1
Hint: x2 dx = .
0 3

Z 17 Z 12 Z 17
Ex. 9. Given that f (x) dx = 3 and f (x) dx = 2, find f (x) dx.
0 0 12
Hint: Use rules (1) and (5).

Comparison theorems for integrals

The following facts can be used to estimate lower or upper bounds for the value of a definite
integral.
Z b
If f (x) 0 for a  x  b, then f (x) dx 0. (6)
a
Z b Z b
If f (x) g(x) for a  x  b, then f (x) dx g(x) dx. (7)
a a
Z b
If m  f (x)  M for a  x  b, then m(b a)  f (x) dx  M (b a). (8)
a
 Z 4 p p
Ex. 10. Find lower and upper bounds for x dx. (Use the fact that f (x) = x is increasing
1
to find the minimum and maximum values of f on [1, 4].)

Average value of a function

Suppose the temperature inside an oven changes at a constant speed. Then the graph of the
temperature function would be a straight line:

0 1 2 3 4 5 6 7 8 9 10

What is the average temperature inside the oven over the ten-minute period we’ve graphed?
If we measure the temperature n times, at evenly spaced intervals, then the average temperature
is
y1 + y2 + · · · + yn
, (†)
n
no matter what n we use.
But what if the temperature changes “quadratically”? That is, what if the graph of the temperature
function follows a parabola?
5

0 1 2 3 4 5 6 7 8 9 10

-1
Now formula (†) can’t always give us the exact average temperature. After all, if Al takes the
average of n = 2 measurements of the temperature, at time x = 5 and x = 10, he gets
1.2 + 5
= 3.1,
2
but if Ana takes the average of n = 5 measurements, estimating the height of the graph at
x = 2, 4, 6, 8, 10, she gets
0.2 + 0.8 + 1.8 + 3.2 + 5 11
= = 5.5
5 5
—and these can’t both be right.

Al and Ana have approximated the average value. It’s reasonable to think that if we take more
and more measurements, our approximation will get more and more accurate.
Our formula for the average of finitely many numbers is
y1 + · · · + yn
.
n
If these represent the values of a function f over the interval [a, b], this becomes
y1 + · · · + yn f (x⇤1 ) + · · · + f (x⇤n )
=
n n
b a
for some values x⇤1 , . . . , x⇤n chosen from subintervals of equal width x= .
n
Solving for n, we have
b a
n= ,
x
so
y1 + · · · + yn f (x⇤1 ) + · · · + f (x⇤n ) f (x⇤1 ) + · · · + f (x⇤n )
= =
n n (b a)/ x
1 ⇥ ⇤
= f (x⇤1 ) x + · · · + f (x⇤n ) x
b a
n
X
1
= f (x⇤k ) x.
b a k=1

If we take more and more measurements, we are letting n ! 1. . .

n
X
1
lim f (x⇤k ) x
n!1 b a k=1

which, using the definition of a definite integral, is

n
X Z b
1 1
lim f (x⇤k ) x = f (x) dx.
n!1 b a k=1
b a a
We therefore define the average value of a function f on the interval [a, b] as
Z b
1
fave = f (x) dx.
b a a

Ex. 11. Find the average value of f (x) = x + 1 over the interval [0, 5].

Ex. 12. Find the average value of f (x) = 6 2x over the interval [0, 3].

⇥⇡ ⇤
Ex. 13. Find the average value of g(t) = csc(t) cot(t) on the interval 6
, ⇡4 .
Solution: p
Z ⇡/4
⇡/4 1 1 2 1
g(t) dt = [csc(t)]⇡/6 = = .
⇡/6 sin(⇡/4) sin(⇡/6) 2
 Additional exercises

Ex. 14 (§5.2—#60, 61, 62, 63). Express the limits as definite integrals over the indicated in-
terval.
n n
(a) lim ⌃ x⇤k x over [1, 3] (c) lim ⌃ sin2 (2⇡x⇤k ) x over [0, 1]
n!1 k=1 n!1 k=1

n ⇥ ⇤ n e xk
(b) lim ⌃ 5(x⇤k )3 4x⇤k x over [2, 7] (d) lim ⌃ x over [0, 1]
n!1 k=1 n!1 k=1 1 + xk

Ex. 15 (§5.2—#64, 67). Given the left Riemann sum Ln or the right Riemann sum Rn as
indicated, express as a definite integral the limit of the sum as n ! 1, identifying the correct
intervals.
✓ ◆
1 n k 1 3 n k
(a) Ln = ⌃ (b) Rn = ⌃ 3+3
n k=1 n n k=1 n
 61. over

Chapter 5 | Integration
62. over
Ex. 16 (§5.2—#71, 72). Evaluate the integrals of the functions graphed using the formulas for
areas of triangles and circles, and subtracting the areas below the x-axis.
63. over 72.
e limits as integrals. (a)
71. (b)

In the following exercises, given Ln or Rn as indicated,

express their limits as as definite integrals,
identifying the correct intervals.

over
64.

65.

r 72. 73.
66.

n or Rn as indicated,
as definite integrals,
67.

68.

69.

73.
In the following exercises, evaluate the integrals of the
Ex.functions
17 (§5.2—#76,
graphed using the79,formulas
81, 83). Evaluate
for areas the
of triangles integral using area formulas.
and
R circles, and subtracting the areas below the x-axis. R5p
3
(a) (3 x) dx (c) 1 4 (x 3)2 dx
70. 0
R6 R3
(b) 0
3 |x 3| dx (d) 2
3 |x| dx

the integrals of the

for areas of triangles
below the x-axis.
This OpenStax book is available for free at http://cnx.org/content/col11964/1.12

p://cnx.org/content/col11964/1.12
Ex. 18 (§5.2—#98, 101). Given that
Z 1 Z 1 Z 1
1 2 1 1
x dx = , x dx = , and x3 dx = ,
0 2 0 3 0 4
compute the integrals Z 1
(1 + x + x2 + x3 ) dx
0
and Z 1
(1 2x)3 dx.
0

Ex. 19 (§5.2—#104). Use the Comparison Theorems to show that

Z 3
(x2 6x + 9) dx 0.
0

Ex. 20 (§5.2—#107). Use the Comparison Theorems to show that

Z Z p
2 p 2
1 + x dx  1 + x2 dx.
1 1
 Rb
Ex. 21. Write as a single integral of the form a
f (x) dx:
Z 2 Z 5 Z 1
f (x) dx + f (x) dx f (x) dx
2 2 2

Ex. 22 (§5.2—#110). Find the average value fave of

f (x) = x2

between a = 1 and b = 1, and find a number c such that f (c) = fave .

Ex. 23 (§5.2—#112). Find the average value fave of

p
f (x) = 4 x2

between a = 0 and b = 2, and find a number c such that f (c) = fave .

 Workbook Lesson 27
§5.3, The Fundamental Theorem of Calculus
Last revised: 2020-10-01 10:37

Objectives
• State the meaning of the Fundamental Theorem of Calculus, Part 1.
• Use the Fundamental Theorem of Calculus, Part 1, to evaluate derivatives of integrals.
• State the meaning of the Fundamental Theorem of Calculus, Part 2.
• Use the Fundamental Theorem of Calculus, Part 2, to evaluate definite integrals.
• Explain the relationship between di↵erentiation and integration.
• Explain the meaning of the Mean Value Theorem for Integrals.

Evaluation Theorem
Not all integrable functions are continuous. But for continuous functions, the value of a definite
integral can be very easy to determine.
EvaluationR Theorem. Let f be a function that is continuous on [a, b]. If F is any antiderivative
b
of f , then a f (x) dx = F (b) F (a).

Considering how complicated the definition of the definite integral was, this formula seems to be a
minor miracle. The definition of the definite integral involved infinitely many values of f (x). But
this theorem says we can evaluate it knowing only two (!) values of an antiderivative F (x).
On the other hand, if we look at a certain physical application, this theorem becomes quite believ-
able. We know that the position function is an antiderivative of velocity:
d/dt
s(t) v(t).

The area under the velocity curve is equal to the change in distance:
Z b
v(t) dt = s(b) s(a).
a

This is exactly what the Evaluation Theorem says.

Ex. 1. Evaluate:
Z 1 Z ⇡/3
Z 4 p
(a) x2 dx (b) sec ✓ tan ✓ d✓ (c) t(1 + t) dt
2 ⇡/4 1
 ⇥ ⇤b
+ The notations . . . a means “evaluate what’s inside the brackets when b and a are substituted
b
for the variable, and then subtract.” The notation . . . a
is defined similarly. For example,
⇥ ⇤b b
F (x) a
= F (x) a
= F (b) F (a).

Solution to Ex. 1:
1
(a) An antiderivative of f (x) = x2 is F (x) = x3 .
3
Z 1 1
x3 13 ( 2)3 1 8
x2 dx = = = + = 3.
2 3 2 3 3 3 3

(b) An antiderivative of f (✓) = sec ✓ tan ✓ is F (✓) = sec ✓.

Z ⇡/3 p
⇥ ⇤⇡/3 1 1 1 1
sec ✓ tan ✓ d✓ = sec ✓ ⇡/4 = = p =2 2.
⇡/4 cos ⇡/3 cos ⇡/4 1/2 2/2
p
(c) An antiderivative of f (t) = t(1 + t) = t1/2 + t3/2 is F (t) = 23 t3/2 + 25 t5/2 .
Z  4
4 p 2 2
t(1 + t) dt = t3/2 + t5/2
1 3 5 1
 
2 3/2 2 5/2 2 3/2 2 5/2
= 4 + 4 1 + 1
3 5 3 5
 
16 64 2 2
= + +
3 5 3 5
80 192 10 6
= + +
15 15 15 15
256
= .
15

The relationship between the indefinite integral and the definite integral
The relationship between the indefinite and definite integral of a continuous function f (x) is
Z b Z b
f (x) dx = f (x) dx .
a a

Notice that Z b
f (x) dx = [F (b) + C] [F (a) + C] = F (b) F (a)
a
no matter what constant C we choose.
 Rectilinear motion problems
Ex. 2. A particle moves along a horizontal line with velocity
v(t) = t2 t 6
meters per second at time t.
(a) Find and interpret the displacement s(4) s(1) of the particle during the time period t 2 [1, 4].
R4
(b) Find the distance traveled 1 |v(t)| dt during this time period.
Solution:
(a) Since v(t) = s0 (t), by the Evaluation Theorem we have
Z 4
s(4) s(1) = v(t) dt
1
Z 4
= t2 t 6 dt
1
 3 4
t t2
= 6t
3 2 1
 
64 1 1
= 8 24 6
3 3 2
9
= .
2
R4
(b) How should we evaluate 1
|v(t)| dt?
We know that |v(t)| = v(t) when v(t) > 0, and |v(t)| = v(t) when v(t) < 0.
So where is v(t) > 0 and where is v(t) < 0?
Recall that the parabola
v(t) = t2 t 6 = (t 3)(t + 2)
opens upward and meets the horizontal axis at t = 2 and t = 3.
Therefore,
Z 4 Z 3 Z 4 Z 3 Z 4
|v(t)| dt = |v(t)| dt + |v(t)| dt = v(t) dt + v(t) dt
1 1 3 1 3
Z 3 Z 4
2
= t t 6 dt + t2 t 6 dt
1 3
 3 2 3  3 4
t t t t2
= 6t + 6t
3 2 1 3 2 2
22 17
= +
3 6
61
=
6
 Functions defined by an integral
Consider the following function:
✓ area under the curve ◆
f (x) = y = t2 .
from t = 0 to t = x
A formula for this function is. . . Z x
f (x) = t2 dt.
0
Functions defined by integrals appear often in physics, statistics, chemistry, electrical and civil
engineering. . .
Today we will see how to analyze a function like this by using di↵erential calculus. For example,
we’ll find the local maximum values of the sine integral function
Z x
sin t
Si(x) = t
dt.
0

+ We will also see that the natural logarithmic function ln can be defined by an integral.

Ex. 3. Define a function Z x

g(x) = f (t) dt,
a

where a = 1 and f (t) = 4t3 . What is the derivative of g(x)?

Solution:
First we substitute for f (t) and a. We get
Z x
g(x) = 4t3 dt. (*)
1

Now use the Evaluation Theorem to get something we can di↵erentiate.

Z x
⇥ ⇤x
g(x) = 4t3 dt = t4 1 = x4 1.
1

We know the derivative of x4 1. Our answer is

d ⇥ 4 ⇤
g 0 (x) = x 1 = 4x3 .
dx
 The Fundamental Theorem of Calculus, Part 1

Notice in the previous exercise that the expression we got for g 0 (x) was exactly the integrand in
the original equation (*) defining g.
That is,
g 0 (x) = 4x3 = f (x).
Why is this?

Let’s try to understand why this is true by looking at the situation geometrically.
• The definition of the derivative says
g(x+h) g(x)
g 0 (x) = lim h
.
h!0

• We know ✓ area under the curve ◆

g(x) = y = f (t)
from t = 0 to t = x
and ✓ area under the curve ◆
g(x + h) = y = f (t) .
from t = 0 to t = x + h
Therefore, the expression
g(x + h) g(x)
is the subtraction of two areas. This di↵erence is the area under the curve from x to x + h.
• How else can we express the area under the curve y = f (t) between x and x + h? Let’s try
to approximate it with a rectangle. The rectangle has to have base h. Let’s give it the height
f (x). Now we have:
g(x + h) g(x) ⇡ f (x) · h.

• Therefore,
g(x + h) g(x)
⇡ f (x). (**)
h
For continuous functions, taking the limit as h ! 0 in this last approximation (**) turns it into an
equality:
g(x + h) g(x)
g 0 (x) = lim = f (x).
h!0 h
Notice that this equation establishes a connection between integration (since g was defined by
an integral), and differentiation, the two branches of the science of calculus.
 Z x
0 d g(x + h) g(x)
g (x) = f (t) dt = lim = f (x).
dx a h!0 h

The above argument will serve as our proof for the first part of the Fundamental Theorem of
Calculus.

Fundamental Theorem of Calculus, Part 1.

Let f be a continuous function defined on [a, b]. Define a function g as
Z x
g(x) = f (t) dt (a  x  b),
a

Then g is di↵erentiable on the interval (a, b), and

g 0 (x) = f (x) (a < x < b).

Corollary.
Let f be a continuous function defined on [a, b]. If F is any antiderivative of f on [a, b], then
Z b
f (t) dt = F (b) F (a).
a

Ex. 4. Define a function Z x

1
`(x) = dt.
1 t
(Many authors define the natural logarithm function ln(x) by this integral.)
Use Part 1 of the Fundamental Theorem of Calculus to find the derivative of `(x).
Ex. 5. Find the derivative of the Fresnel function
Z x ✓ 2◆
⇡t
S(x) = sin dt.
0 2

Solution:
✓ ◆
d⇥ ⇤ ⇡t2
S(x) = sin .
dx 2

Ex. 6. Let
h(t) = 12t 8 sin(2t) + sin(4t).
Given the fact that
h0 (t) = 32 sin4 (x), (***)
find Z ⇡/2
sin4 (t) dt.
0
Solution:
Equation (***) tells us that the function

h(t) = 12t 8 sin(2t) + sin(4t)

is an antiderivative of
h0 (t) = 32 sin4 (x),
so
Z ⇡/2 Z ⇡/2 Z ⇡/2
1 0 1
4
sin (t) dt = h (t) dt = h0 (t) dt
0 0 32 32 0

1
= [12t 8 sin(2t) + sin(4t)]⇡/2
0
32
= [6⇡ 8 sin(⇡) + sin(2⇡)] 0
= 6⇡ 8.
Ex. 7. At what values of x does the following function have local maximum values?
Z x
sin t
Si(x) = dt (x > 0)
0 t
Solution:

d⇥ ⇤ sin x
Si(x) = = 0 (x > 0) () sin x = 0 (x > 0)
dx x
() x = k⇡ (k positive integer).

So the critical numbers of Si(x) are

⇡, 2⇡, 3⇡, . . .

Does Si(x) have local maxima or minima at these values of x?

Sign chart:

sin(x) x Si0 (x) = sin x

x
(0, ⇡) + + + Si(x) increasing
(⇡, 2⇡) + Si(x) decreasing
(2⇡, 3⇡) + + + Si(x) increasing
(3⇡, 4⇡) + Si(x) decreasing
(4⇡, 5⇡) + + Si(x) increasing
(5⇡, 6⇡) + Si(x) decreasing

Answer:
Si(x) has local maximum values at x = ⇡, 3⇡, 5⇡, . . . .
 The Fundamental Theorem of Calculus, Part 1

We actually already know the second part of the Fundamental Theorem of Calculus—we’ve been
calling it the Evaluation Theorem (see beginning of this document).
Fundamental Theorem of Calculus, Part 2.
Let f be an integrable function defined on [a, b]. If F is any antiderivative of f on [a, b], then
Z b
f (x) dx = F (b) F (a).
a

The observant student will notice that this looks very much like the Corollary to Part 1 of the FTC.
The only di↵erence is that, in the Corollary, we required f to be continuous—here, we only require
that f is integrable on the interval [a, b].

+ We’ve seen functions that are integrable on a closed interval [a, b], but not continuous. Can
you sketch the graph of such a function?

Mean Value Theorem for Integrals

Mean Value Theorem for Integrals. For a continuous function f defined on an interval [a, b],
there is a number c in [a, b] such that the rectangle with base [a, b] and height f (c) that has the
same area as the region under the graph of f from a to b.
+ One way to understand this fact is: “You can always chop o↵ the top of a (two-dimensional)
mountain at a certain height, and use it to fill in the valleys so that the mountaintop becomes
completely flat.” The height will be f (c).

Ex. 8. As written above, the Mean Value Theorem for Integrals says that, for a continuous
function f defined on the interval [a, b],
Z b
f (c) · (b a) = f (x) dx
a

for some number c in [a, b]. Show that this means exactly that:

A continuous function defined on [a, b] takes on

its average value for some number c in [a, b].

(See applet on iCollege: “Mean Value Theorem for Integrals”—image on next page)
 ( )

�� � (�)

����
��
'Integral Mean Value Theorem'. From the Wolfram Demonstrations Project
https://demonstrations.wolfram.com/IntegralMeanValueTheorem/

�� Author: Chris Boucher

��

�
� � �
“Shave the mountain”

Ex. 9. Find the average value of the function f (x) = 8 2x over the interval [0, 4] and find c in
[0, 4] such that f (c) equals the average value of f over [a, b].

R3
Ex. 10. Given the fact that 0 2x2 1 dx = 15, find c in the interval [0, 3] such that 2c2 1 equals
the average value of f (x) = 2x2 1 over [0, 3].
 Additional exercises
Ex. 11. Find a function f and a number a such that
Z x
f (t) p
6+ 2
dt = 2 x.
a t

(Hint: Use the Fundamental Theorem of Calculus, Part 1.)

Solution:

 Z x
d f (t) d⇥ p ⇤
6+ dt = 2 x
dx a t2 dx
✓ ◆
f (x) 1
2
=2 x 1/2 = x 1/2
x 2
f (x) = x3/2

Now we find a by substituting x = a.

Z a
f (t) p
6+ 2
dt = 2 a
a t
p
3+0= a

f (x) = x3/2 and a = 3

Ex. 12. Suppose h is a function such that h(1) = 2, h0 (1) = 2, h00 (1) = 3, h(2) = 6, h0 (2) = 5,
h00 (2) = 13, and h00 is continuous. Find
Z 2
h00 (u) du,
1

and justify each step.

Solution:

Z 2
h00 (u) du = h0 (2) h0 (1) (Corollary to FTC applies since h00 is continuous)
1
=5 2 = 3.

“The other information is unnecessary.”

Ex. 13 (§5.3—#161). Below is the graph of
Z x
g(x) = f (t) dt,
0

where f is a piecewise constant function.

(a) Over which interval(s) is f positive? Over which interval(s) is g negative?

(b) What are the maximum and minimum values of f ?
(c) What’s the average value of f ?

Ex. 14 (§5.3—#149, 151). Find the derivative.

Z x Z x
ds
(a) g(x) = ecos(t) dt (b) g(x) = p ds
1 4 16 s2
Ex. 15 (§5.3—#171, 173, 175, 177, 179, 181, 182, 183, 189). Use Part 2 of the Funda-
mental Theorem of Calculus to evaluate the integral.
Z 3 Z 4 ✓ ◆ Z 2⇡
2 2 1
(a) (x + 3x 5) dx (d) x dx (g) cos(✓) d✓
2 1/4 x2 0
Z 3 Z 4 Z ⇡/2
2 2 1
(b) (t 9)(4 t ) dt (e) p dx (h) sin(✓) d✓
2 1 2 x 0
Z 1 Z 16 Z 1✓ ◆
99 dt 1 1
(c) x dx (f) (i) dt
0 1 t1/4 2 t2 t3

Ex. 16 (§5.3—#191, 193). Use the evaluation theorem to express the integral as a function
of x.
Z x Z x
t
(a) e dt (b) sin(t) dt
1 x
 Workbook Lesson 28
§5.4, Integration Formulas and the Net Change Theorem
Last revised: 2020-09-29 12:42

Objectives
• Apply the basic integration formulas.
• Explain the significance of the net change theorem.
• Use the net change theorem to solve applied problems.
• Apply the integrals of odd and even functions.

Applying the basic integration formulas

Rules for integration were given in Lesson 24 (Section 4.10). Let’s warm up with a couple of
exercises that apply one of those formulas.

Ex. 1. Use the Power Rule for integration,

Z
1
xn dx = xn+1 + C for n 6= 1,
n+1
to find Z 4 p
t(1 + t) dt.
1

Ex. 2. Find the definite integral of f (x) = x2 3x over the interval [1, 3].
 The Net Change Theorem

The Net Change Theorem considers the integral of a rate of change F 0 (x).
It says that when a quantity changes, the new value equals the initial value plus the integral of the
rate of change of that quantity.
Net Change Theorem. The new value of a changing quantity equals the initial value plus the
integral of the rate of change:
Z b
F (b) = F (a) + F 0 (x) dx.
a

This is not news. If the function F is continuously di↵erentiable (that is, if the derivative of F is
continuous), then Z b
F 0 (x) dx = F (b) F (a)
a
by the Evaluation Theorem (also known as the Fundamental Theorem of Calculus, Part 2). Simply
adding F (a) to both sides yields the Net Change Theorem.
The significance of the net change theorem lies in the results. Net change can be applied to area,
distance, and volume, to name only a few applications. Net change accounts for negative quantities
automatically without having to write more than one integral.
For example, suppose we are given the velocity function of a particle in motion. The velocity
function accounts for both forward distance (v(t) > 0) and backward distance (v(t) < 0). To find
the change in position—that is, the net displacement—we integrate v(t).
Notice, however, that if we want the total distance traveled, we must count both forward distance
and backward distance as positive quantities. When the total distance traveled is asked for, we
must integrate |v(t)|, which is always nonnegative.

Ex. 3. Given a velocity function v(t) = 3t 5 (in meters per second) for a particle in motion from
time t = 0 to time t = 3, find the net displacement of the particle.

Ex. 4. Find the total di↵erence traveled by the particle in the previous exercise.
Ex. 5. Find the net displacement and total distance traveled in meters given the velocity function
f (t) = 12et 2 over the interval [0, 2].

We can apply the Net Change Theorem to rates of change other than the velocity of a moving
particle. For example, in the next exercise, we apply the Net Change Theorem to the rate of fuel
consumption.

Ex. 7. If the motor on a motorboat is started at t = 0 and the boat consumes gasoline at 5 t3
gal./hr for the first hour, how much gasoline is used in the first hour?
 Integrating even and odd functions

Recall:
A function f is even if
f ( x) = f (x)
for all x in the domain of f , odd if

f ( x) = f (x)

for all x in the domain of f .

When integrating an even or odd function, symmetry ensures that the following rule holds true.

Theorem. For a continuous even function f (x),

Z a Z a
f (x) dx = 2 f (x) dx.
a 0

For a continuous odd function g(x),

Z a
f (x) dx = 0.
a

5
3

2
-2 -1 1 2

1
-5

-2 -1 1 2

2
For an even function (e.g., x ), the For an odd function (e.g., x3 ), the
signed area from 0 to a is equal to the signed area from a to 0 cancels with
signed area from a to 0. the signed area from 0 to a.

Ex. 8. Evaluate the definite integral of the odd function 5 sin x over the interval [ ⇡, ⇡].
Ex. 9. Integrate the even function 3x8 2 from 2 to 2 and verify that the integration formula
for even functions holds.

Additional exercises

Ex. 10 (§5.4—#207, 209, 211). Evaluate.

Z ✓ ◆ Z Z
p 1 1
(a) x p dx (b) dx (c) (sin x cos x) dx
x 2x
Ex. 11 (§5.4—#213). Write an integral that expresses the increase in the perimeter P (s) of a
square when its side length s increases from 2 units to 4 units and evaluate the integral.

Ex. 12 (§5.4—#219). Write an integral that quantifies the change in the area of the surface of
a cube when its side length doubles from s units to 2s units and evaluate the integral.

Ex. 13 (§5.4—#221). Write an integral that quantifies the increase in the surface area of a
sphere as its radius doubles from R units to 2R units and evaluate the integral.
Ex. 14 (§5.4—#225). Suppose that a particle moves along a straight line with velocity defined
by v(t) = |2t 6|, where 0  t  6 (in meters per second). Find the displacement at time t and
the total distance traveled up to t = 6.

Ex. 15 (§5.4—#239). For a given motor vehicle, the maximum achievable deceleration from
braking is approximately 7 m/sec2 on dry concrete. On wet asphalt, it is approximately 2.5 m/sec2 .
Given that 1 mph corresponds to 0.447 m/sec, find the total distance that a car travels in meters
on dry concrete after the brakes are applied until it comes to a complete stop if the initial velocity is
67 mph (30 m/sec) or if the initial braking velocity is 56 mph (25 m/sec). Find the corresponding
distances if the surface is slippery wet asphalt.
 Workbook Lesson 29
§5.5, Substitution
Last revised: 2021-04-22 12:38

Objectives
• Use substitution to evaluate indefinite integrals.
• Use substitution to evaluate definite integrals.

Review: Change of variable

You may have encountered the idea of a change of variable in earlier math classes.
The idea is this: introducing a new variable sometimes makes an algebraic expression easier to work
with.
For example, solving the equation

(x 3)4 (x 3)2 = 0 (?)

becomes easier if we substitute as follows:

Let u = (x 3)2 .

Then equation (?), which is equivalent to

2
(x 3)2 (x 3)2 = 0,

becomes
u2 u = 0,
which can be solved quickly:

u(u 1) = 0
u=0 u=1
(x 3)2 = 0 (x 3)2 = 1
x=3 x 3 = ±1
x=2 x=4

Notice that we gave the solutions in terms of the original variable x.

The statement
Let u = (x 3)2 .
is an example of a change of variable.
+ Note that, when you make a change of variable in a math problem, you must tell the reader
by writing a statement like “Let u = . . . ”
 The Substitution Rule
As we’ve mentioned, the definition of the derivative comes with a formal process for di↵erentiating
a function, whereas antidi↵erentiation does not (see Lesson 24). So far, we have only been able to
evaluate indefinite integrals “by inspection.” That is, to evaluate
Z
f (x) dx,

we had to stare at f (x) until we thought of an antiderivative F (x), that is, a function F such
that
F 0 (x) = f (x)
on some interval. If we couldn’t think of such a function F (or find one in a table of integrals), we
couldn’t evaluate the integral.
Special techniques for evaluating complicated integrals exist, however. We’ll look at one such
technique today, often called u-substitution.
Formally, this technique amounts to reversing the Chain Rule.
In practice, we think of this technique as a “change of variable” that simplifies the integrand. That
is, we rewrite the integrand
y = y(x)
as a function of a new variable
y = y(u).

The u-substitution formula:

Z Z Z
du
y(x) dx = y u(x) · dx = y(u) du.
dx

Ex. 1.
du p
(a) Find a function u(x) such that both u and appear in the expression 2x 1 + x2 .
dx
R p
(b) Evaluate 2x 1 + x2 dx. Make sure to state the change of variables you used, e.g. by writing
“Let u = . . . ”
Solution:
du
Set u = 1 + x2 . Then = 2x.
dx
Z Z Z
2 1/2 1/2 du
(1 + x ) · |{z}
2x dx = u dx = u1/2 du
| {z } dx
du
y=u1/2
dx

= 23 u3/2 + C
= 23 (1 + x2 )3/2 + C
Notice that we give our answer in terms of the original variable x.

Why does the u-substitution formula work?

Justification: Let Y be an antiderivative of y, so that Y 0 = y. Then
Z Z
du (Chain Rule)
y u(x) · dx = Y 0 u(x) · u0 (x) dx = Y u(x) + C = Y u(x) + C
dx
Z Z
= Y 0 (u) du = y(u) du.

Theorem (Substitution Rule). If u = u(x) is a di↵erentiable function whose range is an interval

I, and y is a continuous function defined on I, then
Z Z
du
y u(x) · dx = y(u) du.
dx
R
Ex. 2. Evaluate sec2 x tan3 x dx.

Solution:
du
Set u = tan x. Then = sec2 x.
dx
Z Z Z
du
sec x tan x dx = (tan x) (sec x) dx = u3 dx
2 3 3 2
dx
1
= u4 + C
4
1
= tan4 x + C.
4

A popular trick for applying the Substitution Rule is as follows. It uses the notation for di↵eren-
tials.
Trick. If u = u(x), then du = u0 (x) dx, so
Z Z
0
y(u(x)) · u (x) dx = y(u) du.
 Group work: Substitution Rule, Day 1
Z
I. x2 cos x3 dx
Z p
sin t
II. p dt
t
Z
cos(⇡/x)
III. dx
x2
Z
dt
IV. p
cos2 t 1 + tan t
Z
V. sin ✓ sec2(cos ✓) d✓
Z p
⇡
VI. x cos x2 dx
0
Z 1
⇡t
VII. cos dt
1 4
Z 1
tan x dx
VIII. 2 4
11+x +x
Z
IX. csc ⇡t cos ⇡t dt

X. Average value of sin 4x on [ ⇡, ⇡]

 Substitution Rule, Day 1—Solutions
I.
Z Z
u=x 3 2 3 1
x cos(x ) dx = cos(u) du
du 3
= 3x2 1
dx = sin(u) + C
du = 3x2 dx 3
1 1
du = x2 dx = sin(x3 ) + C
3 3

II.
p Z p Z
u = t = t1/2 sin t 1 ⇣p ⌘
du 1 1/2 1 p dt = p sin t dt
= t = p t t
dt 2 2 t
= 2 cos(u) + C
1 p
2 du = p dt = 2 cos t + C
t

III. Z Z ⇣⇡ ⌘
cos(⇡/x) 2
dx = x cos dx
x2 x
Z
⇡ 1 1
u = = ⇡x = cos(u) du
x ⇡
du Z
= ⇡x 2 1
dx = cos(u) du
⇡
1 2 1
du = x dx = sin(u) + C
⇡ ⇡
1 ⇣⇡ ⌘
= sin +C
⇡ x
 Z
dt
IV. p
cos2 t 1 + tan t Z Z
dt sec2 (t)
p = p dt
cos2 (t) 1 + tan(t) 1 + tan(t)
Z
u = 1 + tan(t) = u 1/2
du
du = sec2 (t) dt
= 2u1/2 + C
p
= 2 1 + tan(t) + C

V. Z Z
2
u = cos(✓) sin(✓) sec (cos(✓)) d✓ = sec2 (u) du
du = sin(✓) d✓ = tan(u) + C
du = sin(✓) d✓
= tan (cos(✓)) + C

VI. Z p Z
x= ⇡
u = x2 2 1 u=⇡
x cos(x ) dx = cos(u) du
du = 2x dx x=0 2 u=0
 u=⇡
1 1
du = x dx = sin(u)
2 2 u=0
1
p = sin(⇡) sin(0)
x= ⇡ u=⇡ 2
x=0 u=0 = 0
VII. Z ✓ ◆ Z
1 ⇡/4
⇡t 4
cos dt = cos(u) du
⇡t ⇡ 1 4 ⇡ ⇡/4
u= = t  ⇡/4
4 4 4
⇡ = sin(u)
du = dt ⇡
4 ⇡/4
4
du = dt 4 h ⇣⇡ ⌘ ⇣ ⇡ ⌘i
⇡ = sin sin
⇡ 4 4
"p p !#
⇡ 4 2 2
x=1 u= =
4 ⇡ 2 2
⇡
x= 1 u= p
4 4 2
=
⇡

VIII.Z
1
tan x dx tan x
2 4
= 0 because f (x) = is odd:
1 1+x +x 1 + x2 + x4

?
f ( x) = f (x)
tan( x) ? tan x
2 4
=
1 + ( x) + ( x) 1 + x2 + x4
tan(x) X tan x
=
1 + x2 + x4 1 + x2 + x4

Note that the last equation follows from the prior equation because tan is
odd on [ 1, 1].

IX. Z Z
1
csc(⇡t) cos(⇡t) dt = cos(⇡t) dt
u = sin(⇡t) sin(⇡t)
Z
du = ⇡ cos(⇡t) dt = u 1
du
1
du = cos(⇡t) dt 1
⇡ = ln sin(⇡t) + C
⇡
X. Average value of sin 4x on [ ⇡, ⇡]
Z ⇡ Z ⇡
1 1
u = 4x sin(4x) dx = sin(4x) dx
⇡ ( ⇡) ⇡ 2⇡ ⇡
du = 4 dx Z 4⇡
1
1 = sin(u) du
du = dx 8⇡ 4⇡
4 
1
= cos(4⇡) cos( 4⇡)
8⇡
x=⇡ u = 4⇡ cos
is
x= 1 u = 4⇡ even
= 0
 Substitution Rule, Day 2—Solutions
I.
⇡
u= = ⇡y 1
y Z ✓ ◆ Z ⇣ ⌘
⇡ 1 1
du = ⇡y 2 dy cos dy = cos(u) du = sin ⇡y + C
y ⇡ ⇡
1
du = dy
⇡

II. Z Z
2 11 11
2 2 cos cos ( ) 1 d = cos sin2 ( ) d
sin ( ) + cos ( ) = 1 Z
22
sin2 ( ) = 1 cos2 ( ) = cos sin( ) d
Z
sin2 ( ) = cos2 ( ) 1
= u22 du
1 23
u = sin( ) = u +C
23
du = cos( ) d sin23
= +C
23

III. Z Z
1 1
dt = dt
(1 + 10t + 25t2 )(1 + 5t) (1 + 5t)3
Z
u = 1 + 5t 1
= u 3 du
du = 5 dt 5
1 1
1 =
10 (1 + 5t)2
+C
du = dt
5 1
= +C
10 + 100t + 250t2
IV. Z Z
5
(v + 3)(v 1) dv = (u + 4)u5 du
Z
= u6 + 4u5 du
u=v 1
u7 2u6
u+4=v+3 = + +C
7✓ 3 ◆
du = dv 6 3u + 14
=u +C
21
1
= (v 1)6 (3v + 11) + C
21

V. Z q Z
p p
p 4 ✓ d✓ = 2 u(4 u) du
u=4 ✓ Z
= 2 4u1/2u3/2 ) du
✓ ◆
Now solve for d✓: 8 3/2 2 5/2
p = 2 u u +C
3 5
✓=4 u ✓ ◆
3/2 20 3u
= 4u +C
✓ = (4 u)2 15
d✓ = 2(4 u) du 4 p
= (4 ✓)3/2 (8 + 3 ✓) + C
15

VI. Z Z
p 1
8x 1 + 4x dx = (2u 2)u1/2 du
4Z
1
= (u 1)u1/2 du
2Z
u = 1 + 4x 1
= u3/2 u1/2 du
2u 2 = 2x 2✓ ◆
1 2 5/2 2 3/2
1 = u u +C
du = dx 2 5 3
4 ✓ ◆
3u 5
= u3/2 +C
15
2
= (1 + 4x)3/2 (6x 1) + C
15
VII. Z Z ✓ ◆
p 1 1/2 1
x 2 3x dx = (u 2)u · du
3 3
Z
1
u=2 3x = u3/2 2u1/2 du
9
1 ✓ ◆
1 2 5/2 4 3/2
(u 2) = x = u u +C
3 9 5 3
✓ ◆
1 1 3/2 6u 20
du = dx = u +C
3 9 15
2
= (2 3x)3/2 (9x + 4) + C
135

VIII. Z p Z p
x3 x2 + 5 dx = x · x2 · x2 + 5 dx
Z
1
= (u 5)u1/2 du
2
Z
1
= u3/2 5u1/2 du
u = x2 + 5 2
✓ ◆
1 2 5/2 10 3/2
u 5 = x2 = u u +C
2 5 3
1 ✓ ◆
du = x dx 1 3/2 6u 50
2 = u +C
2 15
✓ ◆
1 2 3/2 6x 20
= (x + 5) +C
2 15
1 2
= (x + 5)3/2 (3x 10) + C
15
IX. Z p Z p
5
x 2x3 + 4 dx = x2 · x3 · 2x3 + 4 dx
Z
1
= (u 4)u1/2 dx
12
Z
1
= u3/2 4u1/2 du
12
✓ ◆
u = 2x3 + 4 1 2 5/2 8 3/2
= u u +C
1 12 5 3
du = x2 dx ✓
1 3/2 3u 20
◆
6 = u +C
1 3 6 15
✓ 3 ◆
(u 4) = x 1 6x 8
2 = 3
(2x + 4) 3/2
+C
6 15
✓ 3 ◆
1 3 3/2 3x 4
= 2(x + 2) +C
45 15
p
2 2 3
= (x + 2)3/2 (3x3 4) + C
45
 Back-substitution
Z
cos(⇡/y) 1 ⇡
[S] #29. Show that dy = sin + C.
y2 ⇡ y
u = ⇡/y

Z
sin21 ↵
+ Show that cos ↵(1 2 10
cos ↵) d↵ =
21
+ C.

u = sin x

Z
sin23 ↵
+ Show that cos (cos 2 11
1) d =
23
+ C.

u = sin x
 Z
du 1
+ Show that
(1 7u)2
=
7 49u
+ C.

u = 1 + 7u

Z
1 1
+ Show that
(1 + 10t + 25t2 )(1 + 5t)
dt =
10 + 100t + 250t2
+ C.

u = 1 + 5t

Z
1
+ Show that (x 3)(x + 2)7 dx =
72
(x + 2)8 (8x 29) + C.

u = 1 + 5t
 Z
1
[D] #13. Show that (v + 3)(v 1)5 dv = (v 1)6 (3v + 11) + C.
21
u=v 1

Z q p p p
4 3/2
[D] #18. Show that 4 ✓ d✓ = (4 ✓) (8 + 3 ✓) + C.
15
p
u=4 ✓

Z
p 2
[M] #1. Show that 8x 1 + 4x dx = (1 + 4x)3/2 (6x 1) + C.
15
p
u= 1 + 4x
 Z
p 2
[M] #2. Show that x 2 3x dx = (2 3x)3/2 (9x + 4) + C.
135
p
u= 2 + 3x

Z p 1
[M] #13. Show that x3 x2 + 5 dx = (x2 + 5)3/2 (3x2 10) + C.
15
p
u= x2 + 5

Z p
p 2 2 3
[M] #12. Show that x5 2x3 + 4 dx = (x + 2)3/2 (3x3 4) + C.
45
u = 2x3 + 4
 Group work: Substitution Rule, Day 2
Z
cos(⇡/y) 1 ⇡
I. Show that dy = sin + C.
y2 ⇡ y
Z
2 11 sin23
II. Show that cos (cos 1) d = + C.
23
Z
1 1
III. Show that dt = + C.
(1 + 10t + 25t2 )(1 + 5t) 10 + 100t + 250t2
Z
1
IV. Show that (v + 3)(v 1)5 dv =
(v 1)6 (3v + 11) + C.
21
Z q p p 3/2 p
4
V. Show that 4 ✓ d✓ = (4 ✓) (8 + 3 ✓) + C.
15
Z
p 2
VI. Show that 8x 1 + 4x dx = (1 + 4x)3/2 (6x 1) + C.
15
Z
p 2
VII. Show that x 2 3x dx = (2 3x)3/2 (9x + 4) + C.
135
Z p 1
VIII. Show that x x2 + 5 dx = (x2 + 5)3/2 (3x2 10) + C.
3
15
Z p
p 2 2 3
IX. Show that x5 2x3 + 4 dx = (x + 2)3/2 (3x3 4) + C.
45
 Additional exercises
Z
Ex. 3 (§5.5—#261). Evaluate the integral (x+1)4 dx by making the substitution u = x + 1.

R 7
Ex. 4 (§5.5—#261). Evaluate the integral (2x 3) dx by making the substitution u = 2x 3.

Z
x
Ex. 5 (§5.5—#265). Evaluate the integral p dx by making the substitution u = x2 + 1.
x2+1

Z
Ex. 6 (§5.5—#267). Evaluate the integral (x 1)(x2 2x)3 dx by making the substitution
u = x2 2x.
 Z
Ex. 7 (§5.5—#269). Evaluate the integral cos3 ✓ d✓ by making the substitution u = sin ✓.
(Hint: cos2 ✓ = 1 sin2 ✓.)

Ex. 8 (§5.5—#271, 273, 275, 279, 281, 283). Evaluate the indefinite integral.
Z Z Z
99 3 x2
(a) x(1 x) dx (c) cos (✓) sin(✓) d✓ (e) dx
(x3 3)2
Z Z Z
3 2 2
y5
(b) (11x 7) dx (d) t sin(t ) cos(t ) dt (f) dy
(1 y 3 )3/2
Ex. 8. Verify each equation using the Substitution Rule.
R p 2
(a) x 2 3x dx = (2 3x)3/2 (9x + 4) + C
135
R p 1
(b) x3 x2 + 5 dx = (x2 + 5)3/2 (3x2 10) + C
15
p
R 5p 2 2 3
(c) x 2x3 + 4 dx = (x + 2)3/2 (3x3 4) + C
45
 Workbook Lesson 30
§5.6, Integrals Involving Exponential and Logarithmic Functions
Last revised: 2020-09-29 12:42

Objectives
• Integrate functions involving exponential functions.
• Integrate functions involving logarithmic functions.

Recall (Sections 3.9 and 4.10):

Integration rule Corresponding di↵erentiation rule
Z
1 d⇥ ⇤ 1
dx = ln |x| + C ln |x| =
x dx x
Z
d ⇥ x⇤
ex dx = ex + C e = ex
dx
Z
1 x d ⇥ x⇤
bx dx = b + C for 0 < b 6= 1 b = bx ln(b)
ln(b) dx

We now add two integration rules:

Z
ln(x) dx = x ln(x) x+C

= x ln(x) 1 +C
Z
x
logb (x) dx = ln(x) 1 +C
ln(b)

Ex. 1. True/False: The derivative of ex is xex 1 .

p
Ex. 2. Find the most general antiderivative of ex 1 + ex .
Solution:
Take u = 1 + ex . Then du = ex dx, so
Z Z
x
p p
x
e 1 + e dx = u du

2
= u3/2 + C
3
2
= (1 + ex )3/2 + C
3
 R 3
Ex. 3. Integrate: 3x2 e2x dx.
Solution:
Take
u = 2x3 .
Then
du = 6x2 dx.
Since we want to match the expression 3x2 dx that actually appears in the integral, we rewrite
the previous equation as
1
du = 3x2 dx
2
Z Z
2 2x3 1
3x e dx = eu du
2
1
= eu + C
2
1 3
= e2x + C.
2

Z 2
Ex. 4. Find e2x dx.
0

Solution:
Take
u = 2x.
Then
du = 2 dx.
Since we want to match the expression dx that actually appears in the integral, we rewrite
the previous equation as
1
du = dx
2
Z 2 Z
2x 1
e dx = eu du
0 2
x=2
1
= eu
2 x=0
x=2
1 2x
= e
2 x=0

1 4
= e e0
2
e4 1
=
2
 Z 2
e1/x
Ex. 5. Find dx.
1 x2
Solution:
Let us rewrite the given integral as
Z 2
1
ex x 2
dx
1

Take
u = x 1.
Then
2
du = x dx.
2
Since we want to match the expression x dx that actually appears in the integral, we rewrite
the previous equation as
2
du = x dx
Now
Z 2 Z 2
e1/x 1
dx = ex x 2
dx
1 x2 1Z

= eu du

= eu |x=2
x=1

1 x=2
= ex
x=1

= e1/2 e1
p
=e e.
Ex. 6. Suppose a population of fruit flies increases at a rate of g(t) = 2e0.02t , in flies per day. If the
initial population of fruit flies is 100 flies, how many flies are in the population after 10 days?
Solution:
We apply the Net Change Theorem:
Z 10
G(10) = G(0) + 2e0.02t dt
0
1
Taking u = 0.02t, so that du = 0.02 dx and 0.02
dx = du, we see that the net change is
Z 10 Z 10
0.02t 2 u
2e dt = e du
0 0 0.02
t=10
2 u
= e
0.02 t=0
t=10
2 0.02t
= e
0.02 t=0
0.2
= 100e 100.
The initial population is
G(0) = 100,
so the population after 10 days is
Z 10
G(10) = G(0) + 2e0.02t dt = 100 + 100e0.2 100 ⇡ 122.
0

3
Ex. 7. Find the antiderivative of .
x 10
Solution:
1
First, let’s rewrite the integrand in the form :
⇤
Z Z
3 1
dx = 3 dx
x 10 x 10
Take u = x 10. Then du = dx:
Z Z Z
3 1 1
dx = 3 dx = 3 du = 3 ln |u| + C = 3 ln |x 10| + C
x 10 x 10 u
3
Is there any value of x that must be excluded for the integrand and its antiderivative
x 10
3 ln |x 10| to be defined? Yes—we must require x 6= 10.
Answer:
3 ln |x 10| + C, x 6= 10
 2x3 + 3x
Ex. 8. Find the antiderivative of .
x4 + 3x2
Solution:

u = x4 + 3x2
du = (4x3 + 6x) dx = 2(2x3 + 3x) dx
1
du = (2x3 + 3x) dx
2
Z Z
2x3 + 3x 1 du
4 2
dx =
x + 3x 2 u
1
= ln |u| + C
2
1
= ln |x4 + 3x2 | + C
2

Additional exercises

Ex. 9 (§5.6—#321, 329, 336, 337, 341). Evaluate the indefinite integral.
Z Z Z
3x dx 3
(a) e dx (c) 2
(e) x2 e x dx
x(ln x)
Z Z Z ln(1 t)
2 2 e
(b) dx (d) xe x dx (f) dt
x 1 t
Ex. 10 (§5.6—#355). Evaluate the indefinite integral.
Z 2 Z ⇡/3
1 + 2x + x2 sin x cos x
(a) 2 + x3
dx (b) dx
1 3x + 3x 0 sin x + cos x

Ex. 11 (§5.6—#383). Use the identity

Z x
1
ln(x) = dt
1 t
to derive the identity
1
ln = ln x.
x