Pre Calculus Textbook

Precalculus
Precalculus
David H. Collingwood
Department of Mathematics
University of Washington
K. David Prince
Minority Science and Engineering Program
College of Engineering
University of Washington
September 2, 2010
ii
Copyright c 2003 David H. Collingwood and K. David Prince. Permission is

granted to copy, distribute and/or modify this document under the terms of the
GNU Free Documentation License, Version 1.1 or any later version published by
the Free Software Foundation; with no Invariant Sections, with no Front-Cover,
and with no Back-Cover Texts. A copy of the license is included in the section
entitled “GNU Free Documentation License”.
Author Note
For most of you, this course will be unlike any mathematics course you
have previously encountered. Why is this?
Learning a new language

Colleges and universities have been designed to help us discover, share
and apply knowledge. As a student, the preparation required to carry out
this three part mission varies widely, depending upon the chosen field of
study. One fundamental prerequisite is fluency in a “basic language”;
this provides a common framework in which to exchange ideas, care-
fully formulate problems and actively work toward their solutions. In
modern science and engineering, college mathematics has become this
“basic language”, beginning with precalculus, moving into calculus and
progressing into more advanced courses. The difficulty is that college
mathematics will involve genuinely new ideas and the mystery of this
unknown can be sort of intimidating. However, everyone in this course
has the intelligence to succeed!
Is this course the same as high school Precalculus?

There are key differences between the way teaching and learning takes
place in high schools and universities. Our goal is much more than just
getting you to reproduce what was done in the classroom. Here are some
key points to keep in mind:
• The pace of this course will be faster than a high school class in
precalculus. Above that, we aim for greater command of the ma-
terial, especially the ability to extend what we have learned to new
situations.
• This course aims to help you build the stamina required to solve
challenging and lengthy multi-step problems.
• As a rule of thumb, this course should on average take 15 hours

of effort per week. That means that in addition to the 5 classroom
hours per week, you would spend 10 hours extra on the class. This
is only an average and my experience has shown that 12–15 hours
iii
iv
of study per week (outside class) is a more typical estimate. In other

words, for many students, this course is the equivalent of a half-
time job!
• Because the course material is developed in a highly cumulative

manner, we recommend that your study time be spread out evenly
over the week, rather than in huge isolated blocks. An analogy with
athletics is useful: If you are preparing to run a marathon, you must
train daily; if you want to improve your time, you must continually
push your comfort zone.
Prerequisites
This course assumes prior exposure to the “mathematics” in Chapters
1-12; these chapters cover functions, their graphs and some basic exam-
ples. This material is fully developed, in case you need to brush up on a
particular topic. If you have never encountered the concept of a function,
graphs of functions, linear functions or quadratic functions, this course
will probably seem too advanced. You are not assumed to have taken
a course which focuses on mathematical problem solving or multi-step
problem solving; that is the purpose of this course.
Internet
There is a great deal of archived information specific to this course that
can be accessed via the World Wide Web at the URL address
http://www.math.washington.edu/˜m120
Why are we using this text?

Prior to 1990, the performance of a student in precalculus at the Univer-
sity of Washington was not a predictor of success in calculus. For this
reason, the mathematics department set out to create a new course with
a specific set of goals in mind:
• A review of the essential mathematics needed to succeed in calculus.
• An emphasis on problem solving, the idea being to gain both experi-

ence and confidence in working with a particular set of mathemati-
cal tools.
This text was created to achieve these goals and the 2004-05 academic
year marks the eleventh year in which it has been used. Several thou-
sand students have successfully passed through the course.
v
Notation, Answers, etc.

This book is full of worked out examples. We use the the notation “Solu-
tion.” to indicate where the reasoning for a problem begins; the symbol
is used to indicate the end of the solution to a problem. There is a Table
of Contents that is useful in helping you find a topic treated earlier in
the course. It is also a good rough outline when it comes time to study
for the final examination. The book also includes an index at the end.
Finally, there is an appendix at the end of the text with ”answers” to most
of the problems in the text. It should be emphasized these are ”answers”
as opposed to ”solutions”. Any homework problems you may be asked to
turn in will require you include all your work; in other words, a detailed
solution. Simply writing down the answer from the back of the text would
never be sufficient; the answers are intended to be a guide to help insure
you are on the right track.
How to succeed in Math 120.

Most people learn mathematics by doing mathematics. That is, you learn
it by active participation; it is very unusual for someone to learn the ma-
terial by simply watching their instructor perform on Monday, Wednes-
day, and Friday. For this reason, the homework is THE heart of the
course and more than anything else, study time is the key to success
in Math 120. We advise 15 hours of study per week, outside class.
Also, during the first week, the number of study hours will probably be
even higher as you adjust to the viewpoint of the course and brush up
on algebra skills.
Here are some suggestions: Prior to a given class, make sure you have
looked over the reading assigned. If you can’t finish it, at least look it over
and get some idea of the topic to be discussed. Having looked over the
material ahead of time, you will get FAR MORE out of the lecture. Then,
after lecture, you will be ready to launch into the homework. If you follow
this model, it will minimize the number of times you leave class in a daze.
In addition, spread your study time out evenly over the week, rather than
waiting until the day before an assignment is due.
Acknowledgments
The efforts of several people have led to numerous changes, corrections
and improvements. We want to specifically thank Laura Acuña, Patrick
Averbeck, Jim Baxter, Sandi Bennett, Daniel Bjorkegren, Cindy Burton,
Michael D. Calac, Roll Jean Cheng, Jerry Folland, Dan Fox, Grant Gal-
braith, Peter Garfield, Richard J. Golob, Joel Grus, Fred Kuczmarski,
Julie Harris, Michael Harrison, Teri Hughes, Ron Irving, Ian Jannetty,
Mark Johnson, Michael Keynes, Don Marshall, Linda Martin, Patrick
vi
Perkins, Lisa Peterson, Ken Plochinski, Eric Rimbey, Tim Roberts, Aaron
Schlafly, David Schneider, Marilyn Stor, Lukas Svec, Sarah Swearinger,
Steve Tanner, Paul Tseng, and Rebecca Tyson. I am grateful to everyone
for their hard work and dedication to making this a better product for
our students.
The Minority Science and Engineering Program (MSEP) of the Col-
lege of Engineering supports the development of this textbook. It is also
authoring additional materials, namely, a student study guide and an
instructor guide. MSEP actively uses these all of these materials in its
summer mathematics program for freshman pre-engineers. We want to
thank MSEP for its contributions to this textbook.
We want to thank Intel Corporation for their grant giving us an ”Inno-
vation in Education” server donation. This computer hardware is used
to maintain and develop this textbook.
Comments
Send comments, corrections, and ideas to colling@math.washington.edu
or kdp@engr.washington.edu.
Preface
Have you ever noticed this peculiar feature of mathematics: When you
don’t know what is going on, it is really hard, difficult, and frustrating.
But, when you know what is going on, mathematics seems incredibly
easy, and you wonder why you had trouble with it in the first place!
Here is another feature of learning mathematics: When you are strug-
gling with a mathematical problem, there are times when the answer
seems to pop out at you. At first, nothing is there, then very suddenly,
in a flash, the answer is all there, and you sit wondering why you didn’t
“see” the solution sooner. We have a special name for this: It’s the “A-
Ha!” experience. Often the difficulty you have in studying mathematics
is that the rate at which you are having an A-Ha! experience might be so
low that you get discouraged or, even worse, you give up studying math-
ematics altogether. One purpose of this course is to introduce you to
some strategies that can help you increase the rate of your mathematical
A-Ha! experiences.
What is a story problem?

When we ask students if they like story problems, more often than not,
we hear statements like: “I hate story problems!” So, what is it about
these kinds of problems that causes such a negative reaction? Well,
the first thing you can say about story problems is that they are mostly
made up of words. This means you have to make a big effort to read
and understand the words of the problem. If you don’t like to read, story
problems will be troublesome.
The second thing that stands out with story problems is that they
force you to think about how things work. You have to give deep thought
to how things in the problem relate to each other. This in turn means that
story problems force you to connect many steps in the solution process.
You are no longer given a list of formulas to work using memorized steps.
So, in the end, the story problem is a multi-step process such that the
“A-Ha!” comes only after lots of intense effort.
All of this means you have to spend time working on story problems.
It is impossible to sit down and spend only a minute or two working
each problem. With story problems, you have to spend much more time
working toward a solution, and at the university, it is common to spend
vii
viii
an hour or more working each problem. So another aspect of working

these kinds of problems is that they demand a lot of work from you, the
problem solver.
We can conclude this: What works is work! Unfortunately, there is
no easy way to solve all story problems. There are, however, techniques
that you can use to help you work efficiently. In this course, you will
be presented with a wide range of mathematical tools, techniques, and
strategies that will prepare you for university level problem solving.
What are the BIG errors?

Before we look at how to make your problem solving more efficient, let’s
look at some typical situations that make problem solving inefficient. If
you want to be ready for university level mathematics, we are sure you
have heard somewhere: “You must be prepared!” This means you need
to have certain well-developed mathematical skills before you reach the
university. We would like to share with you the three major sources of
errors students make when working problems, especially when they are
working exam problems. Every time we sit down and review solutions
with a student who has just taken an exam, and who has lost a lot of
points in that exam, we find errors falling pretty much into three cate-
gories, and these errors are the major cause of inefficient mathematical
problem solving.
The first type of error that loses points is algebra. This is an error of
not knowing all of the algebraic rules. This type of error also includes
mistakes in the selection and use of mathematical symbols. Often, dur-
ing the problem solving process, you are required to introduce math-
ematical symbols. But, without these symbols, you cannot make any
further progress. Think of it this way: Without symbols, you cannot do
any mathematics involving equations!
The second error we see in problem solving has to do with visualiza-
tion. In this case, we’re talking about more than the graphics you can
get from a calculator. Graphing and curve sketching are very important
skills. But, in doing story problems, you might find it almost impossi-
ble to create a solution without first drawing a picture 1 of your problem.
Thus, by not drawing a good picture of the problem, students get stuck
in their exams, often missing the solution to a problem entirely.
Finally, the third big source of error is not knowing mathematical def-
initions. Actually, this is a huge topic, so we will only touch on some of
the main features of this kind of error. The key thing here is that by not
knowing mathematical definitions, it becomes very hard to know what to
do next in a multi-step solution to a story problem.
1
Whenever we talk about a picture of your problem, we mean not just the drawing
itself. In this case, the picture must include the drawing and the labels which clearly
signify the quantities related to your problem.
ix
Here is what it all boils down to: Mathematical definitions, for the
most part, provide little cookbook procedures for computing or measuring
something. For example, if you did not know the mathematical definition
of “speed,” you would not know that to measure speed, you first measure
your distance and you simultaneously measure the time it takes to cover
that distance. Notice this means you have two measuring instruments
working at the same time. The second thing you must do, according to
the definition of speed, is divide the distance you measured by the time
you measured. The result of your division is a number that you will call
speed. The definition is a step-by-step procedure that everyone agrees
to when talking about “speed.” So, it’s easy to understand that if you
are trying to solve a story problem requiring a speed computation and
you did not know the definition or you could not remember the definition
of speed, you are going to be “stuck” and no further progress will be
possible!
What does all of this mean for you? As you study your mathematics,
make sure you are the best you can be in these three areas: Algebra, Vi-
sualization, and Definitions. Do a little algebra every day. Always draw a
picture to go with all your problems. And, know your mathematical def-
initions without hesitation. Do this and you will see a very large portion
of your math errors disappear!
Problem Solving Strategies

This topic would require another book to fully develop. So, for now, we
would like to present some problem solving ideas you can start using
right away.
Let’s look now at a common scenario: A student reads a story problem
then exclaims, maybe with a little frustration: “If I only had the formula,
I could solve this problem!” Does this sound familiar? What is going on
here, and why is this student frustrated? Suppose you are this student.
What are you actually trying to do? Let’s break it down. First, you are
reading some descriptive information in words and you need to translate
this word information into symbols. If you had the symbolic information,
you would be in a position to mathematically solve your problem right
away.
Unfortunately, you cannot solve anything without first translating
your words into symbols. And, going directly from words to symbols is
usually very difficult! So, here we are looking for some alternative ap-
proach for translating words into symbols. Figure 1 is the answer to this
problem solving dilemma.
A lot is going on in Figure 1. Let’s consider some of the main features
of this diagram. First, it is suggesting that you are dealing with informa-
tion in three different forms: Words, Pictures, and Symbols. The arrows
in this diagram suggest that in any problem solving situation, you are
x
Words
Pictures Symbols
Figure 1: Problem solving as a transformation process.
actually translating information from one form to another. The arrows

also suggest that there are alternative paths you can take to get from one
form to another! This is a very, very important point: the idea that there
is more than one way to get from words to symbols.
Let’s rewind this discussion: You’re reading a story problem. But,
now, before giving any thought to what your formula is, that is, before
worrying about your symbolic information, you grab a blank sheet of pa-
per and start drawing a picture of your problem. And, to your picture
you add symbols denoting the quantities you need in your problem. At
this point in your problem solving, you are not trying to write any equa-
tions; you are only trying to see what your problem looks like. You are
also concentrating on another extremely important step: Deciding what
symbols to use in your problem!
Now you have a good picture of your problem. It shows not only what
the problem looks like, but symbolically shows all the problem’s variables
and constants. You can start using this information to mathematically
model your problem. The process of creating a mathematical model is ac-
tually nothing more than the arrow in the diagram going from pictures to
symbols. Mathematical modeling is the jump you make from the visual
information you have created to information contained in your formulas.
Let’s summarize the problem solving process. You start with a de-
scription of a problem that is presented to you mainly in the form of
words. Instead of trying to jump directly from words to symbols, you
jump from words to pictures. Once you have a good picture, you jump
from pictures to symbols. And, all the time, you are relying on mathe-
matical definitions as you interpret the words of your problem; on visu-
alization techniques as you draw pictures related to your problem; and,
on your algebra skills as you are formulating the equations you need to
xi
solve your problem.

There is one final thing to notice about the diagram in this section. All
of this discussion so far deals with the situation where your direction is
from
Words =⇒ Pictures =⇒ Symbols.

But when you study the diagram you see that the arrows go both ways!
So, we will leave you with this to think about: What does it mean, within
the context of problem solving, when you have
Symbols =⇒ Pictures =⇒ Words ?
An Example.
Here is a worked example that is taken from a typical homework assign-
ment for Section 1.1 of this book. See if you can recognize the multitude
of steps needed to arrive at the equations that allow us to compute a
solution. That is, try to identify the specific way in which information is
being transformed during the problem solving process.
This problem illustrates the principle used to make a good “squirt

gun”. A cylindrical tube has diameter 1 inch, then reduces to
diameter d. The tube is filled with oil and piston A moves to the
right 2 in/sec, as indicated. This will cause piston B to move to
the right m in/sec. Assume the oil does not compress; that means
the volume of the oil between the two pistons is always the same.
2 in/sec oil m in/sec

A B
1
1. If the diameter of the narrow part of the tube is 2
inch, what
is the speed of piston B?
in
2. If B moves 11 sec , what is the diameter of the narrow part of
the tube?
Solution.
The first thing to do with any story problem is to draw a picture of the
problem. In this case, you might re-sketch the picture so that it looks
3-dimensional: See Figure 2. As you draw, add in mathematical symbols
signifying quantities in the problem.
The next thing is to clearly define the variables in your problem:
xii
Volume Leaving Cylinder A.
Volume Entering Cylinder B.
d dB
A
x
B
Piston B.
xA
Piston A.
Figure 2: A re-sketch of the original given figure.
1. Let VA and VB stand for the change in volumes as piston A moves to

the right.
2. Let dA and dB represent the diameters of each cylinder.
3. Let rA and rB represent the radii of each cylinder.
4. Let sA and sB stand for the speeds of each piston.
5. Let xA and xB stand for the distance traveled by each piston.
Now that you have some symbols to work with, you can write the given
data down this way:
inches
1. sA = 2 second .
2. dA = 1.0 inch.
After you have studied this problem for a while, you would write down
some useful relationships:
1. The volume of any cylinder is
V = πr2h
where r is the radius of the cylinder, and h is its height or length.

From this, you can derive the volume of a cylinder in terms of its
diameter, d:
π 2
V= d h.
4
xiii
2. “Distance” = “Rate” × “Time”. In terms of this problem, you would

write
x = st,
where x is the distance your piston moves, and s is the speed of the
piston’s motion.
Now you are in a position to create a mathematical model that de-
scribes what is going on:
1. From the two relationships above, you can derive the volume equa-
tions for each cylinder so that the diameters and speed of the pistons
are included:
πd2AxA πd2A(sAt)
VA = =⇒ VA =
4 4
and
πd2BxB πd2B(sBt)
VB = =⇒ VB =
4 4
2. Since the oil does not compress, at each instant when piston A is
moving, you must have VA = VB, thus:
πd2A(sAt) πd2B(sBt)
= .
4 4
After canceling π, t, and 4, you end up with a mathematical model
describing this problem that you can use to answer all sorts of in-
teresting questions:
d2AsA = d2BsB.
Using your model, you can compute the following solutions:

in
1. Given: dB = 12 in, dA = 1.0 in, and sA = 2 sec , find sB. From your
model, you derive:
d2A · sA
d2AsA = d2BsB =⇒ sB =
d2B
from which you can compute
in
m = sB = 8 .
sec
2. Likewise, you can use your model to compute
r
2
dB = in,
11
exactly.
xiv
Contents
1 Warming Up 1
1.1 Units and Rates . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Total Change = Rate × Time . . . . . . . . . . . . . . . . . 4
1.3 The Modeling Process . . . . . . . . . . . . . . . . . . . . . 6
1.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2 Imposing Coordinates 11
2.1 The Coordinate System . . . . . . . . . . . . . . . . . . . . 11
2.1.1 Going from P to a Pair of Real Numbers. . . . . . 12
2.2 Three Features of a Coordinate System . . . . . . . . . . . 13
2.2.1 Scaling . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2.2 Axes Units . . . . . . . . . . . . . . . . . . . . . . . 14
2.3 A Key Step in all Modeling Problems . . . . . . . . . . . . 15
2.4 Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3 Three Simple Curves 25

3.1 The Simplest Lines . . . . . . . . . . . . . . . . . . . . . . . 25
3.2 Circles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.3 Intersecting Curves I . . . . . . . . . . . . . . . . . . . . . 28
3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4 Linear Modeling 35
4.1 The Earning Power Problem . . . . . . . . . . . . . . . . . 35
4.2 Relating Lines and Equations . . . . . . . . . . . . . . . . 37
4.3 Non-vertical Lines . . . . . . . . . . . . . . . . . . . . . . . 38
4.4 General Lines . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.5 Lines and Rate of Change . . . . . . . . . . . . . . . . . . . 41
4.6 Back to the Earning Power Problem . . . . . . . . . . . . . 44
4.7 What’s Needed to Build a Linear Model? . . . . . . . . . . 45
4.8 Linear Application Problems . . . . . . . . . . . . . . . . . 46
4.9 Perpendicular and Parallel Lines . . . . . . . . . . . . . . . 46
4.10 Intersecting Curves II . . . . . . . . . . . . . . . . . . . . . 47
4.11 Uniform Linear Motion . . . . . . . . . . . . . . . . . . . . 49
4.12 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
xv
xvi CONTENTS
4.13 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5 Functions and Graphs 57

5.1 Relating Data, Plots and Equations . . . . . . . . . . . . . 57
5.2 What is a Function? . . . . . . . . . . . . . . . . . . . . . . 59
5.2.1 The definition of a function
(equation viewpoint) . . . . . . . . . . . . . . . . . 60
(conceptual viewpoint) . . . . . . . . . . . . . . . . 63
5.3 The Graph of a Function . . . . . . . . . . . . . . . . . . . 64
5.4 The Vertical Line Test . . . . . . . . . . . . . . . . . . . . . 65
5.4.1 Imposed Constraints . . . . . . . . . . . . . . . . . 66
5.5 Linear Functions . . . . . . . . . . . . . . . . . . . . . . . . 66
5.6 Profit Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
6 Graphical Analysis 75
6.1 Visual Analysis of a Graph . . . . . . . . . . . . . . . . . . 75
6.1.1 Visualizing the domain and range . . . . . . . . . 75
6.1.2 Interpreting Points on the Graph . . . . . . . . . . 76
6.1.3 Interpreting Intercepts of a Graph . . . . . . . . . 78
6.1.4 Interpreting Increasing and Decreasing . . . . . . 78
6.2 Circles and Semicircles . . . . . . . . . . . . . . . . . . . . 79
6.3 Multipart Functions . . . . . . . . . . . . . . . . . . . . . . 81
6.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
7 Quadratic Modeling 87
7.1 Parabolas and Vertex Form . . . . . . . . . . . . . . . . . . 87
7.1.1 First Maneuver: Shifting . . . . . . . . . . . . . . . 89
7.1.2 Second Maneuver: Reflection . . . . . . . . . . . . 91
7.1.3 Third Maneuver: Vertical Dilation . . . . . . . . . 91
7.1.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . 91
7.2 Completing the Square . . . . . . . . . . . . . . . . . . . . 92
7.3 Interpreting the Vertex . . . . . . . . . . . . . . . . . . . . 95
7.4 Quadratic Modeling Problems . . . . . . . . . . . . . . . . 96
7.4.1 How many points determine a parabola? . . . . . 100
7.5 What’s Needed to Build a Quadratic Model? . . . . . . . . 103
7.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
8 Composition 107
8.1 The Formula for a Composition . . . . . . . . . . . . . . . 108
8.1.1 Some notational confusion . . . . . . . . . . . . . 111
8.2 Domain, Range, etc. for a Composition . . . . . . . . . . . 112
8.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
CONTENTS xvii
9 Inverse Functions 119

9.1 Concept of an Inverse Function . . . . . . . . . . . . . . . 119
9.1.1 An Example . . . . . . . . . . . . . . . . . . . . . . 120
9.1.2 A Second Example . . . . . . . . . . . . . . . . . . 121
9.1.3 A Third Example . . . . . . . . . . . . . . . . . . . 121
9.2 Graphical Idea of an Inverse . . . . . . . . . . . . . . . . . 122
9.2.1 One-to-one Functions . . . . . . . . . . . . . . . . 124
9.3 Inverse Functions . . . . . . . . . . . . . . . . . . . . . . . 125
9.3.1 Schematic Idea of an Inverse Function . . . . . . 126
9.3.2 Graphing Inverse Functions . . . . . . . . . . . . 127
9.4 Trying to Invert a Non one-to-one Function . . . . . . . . 127
9.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
9.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
10 Exponential Functions 133

10.1 Functions of Exponential Type . . . . . . . . . . . . . . . . 134
10.1.1 Reviewing the Rules of Exponents . . . . . . . . . 135
10.2 The Functions y = A0bx . . . . . . . . . . . . . . . . . . . . 137
10.2.1 The case b = 1 . . . . . . . . . . . . . . . . . . . . 138
10.2.2 The case b > 1 . . . . . . . . . . . . . . . . . . . . . 138
10.2.3 The case 0 < b < 1 . . . . . . . . . . . . . . . . . . 139
10.3 Piano Frequency Range . . . . . . . . . . . . . . . . . . . . 140
10.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
11 Exponential Modeling 145

11.1 The Method of Compound Interest . . . . . . . . . . . . . 146
11.1.1 Two Examples . . . . . . . . . . . . . . . . . . . . . 147
11.1.2 Discrete Compounding . . . . . . . . . . . . . . . 148
11.2 The Number e and
the Exponential Function . . . . . . . . . . . . . . . . . . . 149
11.2.1 Calculator drill . . . . . . . . . . . . . . . . . . . . 151
11.2.2 Back to the original problem... . . . . . . . . . . . 151
11.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
12 Logarithmic Functions 153

12.1 The Inverse Function of y = ex . . . . . . . . . . . . . . . . 153
12.2 Alternate form for
functions of exponential type . . . . . . . . . . . . . . . . . 156
12.3 The Inverse Function of y = bx . . . . . . . . . . . . . . . . 157
12.4 Measuring the Loudness of Sound . . . . . . . . . . . . . 159
12.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
13 Three Construction Tools 165

13.1 A Low-tech Exercise . . . . . . . . . . . . . . . . . . . . . . 165
13.2 Reflection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
13.3 Shifting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
xviii CONTENTS
13.4 Dilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

13.5 Vertex Form and Order of Operations . . . . . . . . . . . . 173
13.6 Summary of Rules . . . . . . . . . . . . . . . . . . . . . . . 174
13.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
14 Rational Functions 181

14.1 Modeling with Linear-to-linear Rational Functions . . . . 185
14.2 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
14.3 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
15 Measuring an Angle 191

15.1 Standard and Central Angles . . . . . . . . . . . . . . . . . 192
15.2 An Analogy . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
15.3 Degree Method . . . . . . . . . . . . . . . . . . . . . . . . . 193
15.4 Radian Method . . . . . . . . . . . . . . . . . . . . . . . . . 196
15.5 Areas of Wedges . . . . . . . . . . . . . . . . . . . . . . . . 199
15.5.1 Chord Approximation . . . . . . . . . . . . . . . . 201
15.6 Great Circle Navigation . . . . . . . . . . . . . . . . . . . . 202
15.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
15.8 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
16 Measuring Circular Motion 207

16.1 Different ways to measure
Cosmo’s speed . . . . . . . . . . . . . . . . . . . . . . . . . 207
16.2 Different Ways to Measure
Circular Motion . . . . . . . . . . . . . . . . . . . . . . . . 209
16.2.1 Three Key Formulas . . . . . . . . . . . . . . . . . 209
16.3 Music Listening Technology . . . . . . . . . . . . . . . . . 212
16.4 Belt and Wheel Problems . . . . . . . . . . . . . . . . . . . 215
16.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
17 The Circular Functions 221

17.1 Sides and Angles of a Right Triangle . . . . . . . . . . . . 221
17.2 The Trigonometric Ratios . . . . . . . . . . . . . . . . . . . 222
17.3 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
17.4 Circular Functions . . . . . . . . . . . . . . . . . . . . . . . 226
17.4.1 Are the trigonometric ratios functions? . . . . . . 227
17.4.2 Relating circular functions and right triangles . . 229
17.5 What About Other Circles? . . . . . . . . . . . . . . . . . . 230
17.6 Other Basic Circular Function . . . . . . . . . . . . . . . . 231
17.7 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
18 Trigonometric Functions 237

18.1 Easy Properties of Circular Functions . . . . . . . . . . . 237
18.2 Identities . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
18.3 Graphs of Circular Functions . . . . . . . . . . . . . . . . 242
CONTENTS xix
18.3.1 A matter of scaling . . . . . . . . . . . . . . . . . . 243

18.3.2 The sine and cosine graphs . . . . . . . . . . . . . 243
18.3.3 The tangent graph . . . . . . . . . . . . . . . . . . 246
18.4 Trigonometric Functions . . . . . . . . . . . . . . . . . . . 247
18.4.1 A Transition . . . . . . . . . . . . . . . . . . . . . . 247
18.4.2 Graphs of trigonometric functions . . . . . . . . . 249
18.4.3 Notation for trigonometric functions . . . . . . . . 249
18.5 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
19 Sinusoidal Functions 251

19.1 A special class of functions . . . . . . . . . . . . . . . . . . 251
19.1.1 How to roughly sketch a sinusoidal graph . . . . 254
19.1.2 Functions not in standard sinusoidal form . . . . 257
19.2 Examples of sinusoidal behavior . . . . . . . . . . . . . . . 259
19.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
19.4 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
20 Inverse Circular Functions 265

20.1 Solving Three Equations . . . . . . . . . . . . . . . . . . . 265
20.2 Inverse Circular Functions . . . . . . . . . . . . . . . . . . 268
20.3 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
20.4 How to solve trigonometric equations . . . . . . . . . . . . 274
20.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
20.6 Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Appendix 283
A Useful Formulas 285
B Answers 289
C GNU Free Documentation License 301

C.1 Applicability and Definitions . . . . . . . . . . . . . . . . . 302
C.2 Verbatim Copying . . . . . . . . . . . . . . . . . . . . . . . 303
C.3 Copying in Quantity . . . . . . . . . . . . . . . . . . . . . . 303
C.4 Modifications . . . . . . . . . . . . . . . . . . . . . . . . . . 304
C.5 Combining Documents . . . . . . . . . . . . . . . . . . . . 306
C.6 Collections of Documents . . . . . . . . . . . . . . . . . . . 306
C.7 Aggregation With Independent Works . . . . . . . . . . . . 306
C.8 Translation . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
C.9 Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
C.10 Future Revisions of This License . . . . . . . . . . . . . . . 307
References 309
Index 309
xx CONTENTS
Chapter 1
Warming Up
The basic theme of this book is to study precalculus within the context
of problem solving. This presents a challenge, since skill in problem
solving is as much an art or craft as it is a science. As a consequence,
the process of learning involves an active apprenticeship rather than a
passive reading of a text. We are going to start out by assembling a basic
toolkit of examples and techniques that are essential in everything that
follows. The main ideas discussed in the next couple of chapters will
surely be familiar; our perspective on their use and importance may be
new.
The process of going from equations to pictures involves the key con-
cept of a graph , while the reverse process of going from pictures (or raw
data) to equations is called modeling . Fortunately, the study of graphing
and modeling need not take place in a theoretical vacuum. For example,
imagine you have tossed a ball from the edge of a cliff. A number of nat-
ural questions arise: Where and when does the ball reach its maximum
height? Where and when does the ball hit the ground? Where is the ball
located after t seconds?
We can attack these questions from two directions. If
we knew some basic physics, then we would have equa- Path of tossed ball.
tions for the motion of the ball. Going from these equations
to the actual curved path of the ball becomes a graphing
problem; answering the questions requires that we really
Cliff.
understand the relationship between the symbolic equa-
tions and the curved path. Alternatively, we could ap- Ground level.
proach these questions without knowing any physics. The

idea would be to collect some data, keeping track of the
height and horizontal location of the ball at various times, Figure 1.1: Ball toss.
then find equations whose graphs will “best” reproduce the
collected data points; this would be a modeling approach to the prob-
lem. Modeling is typically harder than graphing, since it requires good
intuition and a lot of experience.
1
2 CHAPTER 1. WARMING UP
1.1 Units and Rates

A marathon runner passes the one-mile marker of the race with a clocked
speed of 18 feet/second. If a marathon is 26.2 miles in length and this
speed is maintained for the entire race, what will be the runner’s total
time?
This simple problem illustrates a key feature of modeling with mathe-
matics: Numbers don’t occur in isolation; a number typically comes with
some type of unit attached. To answer the question, we’ll need to recall
a formula which precisely relates “total distance traveled” to “speed” and
“elapsed time”. But, we must be VERY CAREFUL to use consistent
units. We are given speed units which involve distance in “feet” and the
length of the race involves distance units of “miles”. We need to make a
judgment call and decide on a single type of distance unit to use through-
out the problem; either choice is OK. Let’s use “feet”, then here is the fact
we need to recall:
(total distance traveled) = (constant speed) × (elapsed time) (1.1)
(ft) = (ft/sec) × (sec)
To apply the formula, let t represent the elapsed time in seconds and
first carry out a “conversion of units” using the conversion factor “5,280
ft/mile”. Recall, we can manipulate the units just like numbers, cancel-
ing common units on the top and bottom of a fraction:
////· ft
mile
26.2 mile × (5,280 ft/mile) = (26.2)(5,280)
mile
////
= (26.2)(5,280) ft.
Formula 1.1 can now be applied:
(26.2)(5,280) ft = 18 ft/sec × t
f/t

(26.2)(5,280)
= t
18 ft/sec
/
1
7,685.33 = 7,685.33 sec = t
1/sec
So, the runner would complete the race in t = 7,685.33 seconds. If we
wanted this answer in more sensible units, we would go through yet
another units conversion:
t = 7,685.33 sec × (1 min/60 sec) × (1 hr/60 min)

7,685.33
= hr
602
= 2.1348 hr.
1.1. UNITS AND RATES 3
The finish clock will display elapsed time in units of “hours : minutes :
seconds”. Two further conversions (see Exercise 1.4) lead to our runner
having a time of 2:08:05.33; this is a world class time!
Manipulation of units becomes especially important when we are work-
ing with the density of a substance, which is defined by
def mass
density = .
volume
For example, pure water has a density of 1 g/cm3. Notice, given any two
of the quantities “density, volume or mass,” we can solve for the remain-
ing unknown using the formula. For example, if 857 g of an unknown
substance has a volume of 2.1 liters, then the density would be
mass
d =
volume
857 g
=
2.1 L
g 1/L

857
= × ×
2.1 L
/ 1,000 cm3
= 0.408 g/cm3.
Example 1.1.1. A sphere of solid gold has a mass of 100 kg and the den-
sity of gold is 19.3 g/cm3. What is the radius of the sphere?
Solution. This problem is more involved. To answer this, let r be the

unknown radius of the sphere in units of cm. The volume of the sphere
is V = 34 πr3. Since the sphere is solid gold, the density of gold is the ratio
mass of sphere
density of gold =
volume of sphere
Plugging in what we know, we get the equation

! ! !
g 1000 g 105 g

100kg 100kg
19.3 = 4
= 4
= 4
cm3 3
πr3 3
πr3 1 kg 3
πr3
Solving for r3 we find
r3 = 1236.955516 cm3
from which we get

1/3
r = 1237 cm3 = 10.73457 cm
1.2 Total Change = Rate × Time

We live in a world where things are changing as time goes by: the tem-
perature during the day, the cost of tuition, the distance you will travel
after leaving this class, and so on. The ability to precisely describe how
a quantity is changing becomes especially important when making any
kind of experimental measurements. For this reason, let’s start with a
clear and careful definition. If a quantity is changing with respect to time
(like temperature, distance or cost), we can keep track of this using what
is called a rate (also sometimes called a rate of change ); this is defined
as follows:
def change in the quantity
rate =
change in time
This sort of thing comes up so frequently, there is special shorthand
notation commonly used: We let the Greek letter ∆ (pronounced “delta”)
be shorthand for the phrase “change in.” With this agreement, we can
rewrite our rate definition in this way:
def ∆ quantity
rate =
∆ time
But, now the question becomes: How do we calculate a rate? If you think
about it, to calculate “∆ quantity” in the rate definition requires that we
compare two quantities at two different times and see how they differ
(i.e., how they have changed). The two times of comparison are usually
called the final time and the initial time. We really need to be precise
about this, so here is what we mean:
∆ quantity = (value of quantity at final time) −
(value of quantity at initial time)
∆ time = (final time) − (initial time).
For example, suppose that on June 4 we measure that the tempera-
ture at 8:00 am is 65◦ F and at 10:00 am it is 71◦ F. So, the final time is
10:00 am, the initial time is 8:00 am and the temperature is changing
according to the
∆ quantity
rate =
∆ time
final value of quantity − initial value of quantity
=
final time − initial time
71 − 65 degrees
=
10:00 − 8:00 hours
= 3 deg/hr.
As a second example, suppose on June 5 the temperature at 8:00 am is
71◦ and at 10:00 am it is 65◦ . So, the final time is 10:00 am, the initial
1.2. TOTAL CHANGE = RATE × TIME 5
time is 8:00 am and the temperature is changing according to the

∆ quantity
rate =
∆ time
final value of quantity − initial value of quantity
=
final time − initial time
65 − 71 degrees
=
10:00 − 8:00 hours
= −3 deg/hr.
These two examples illustrate that a rate can be either a positive or a
negative number. More importantly, it highlights that we really need to
be careful when making a rate computation. In both examples, the initial
and final times are the same and the two temperatures involved are the
same, BUT whether they occur at the initial or final time is interchanged.
If we accidently mix this up, we will end up being off by a minus sign.
There are many situations where the rate is the same for all time
periods. In a case like this, we say we have a constant rate. For example,
imagine you are driving down the freeway at a constant speed of 60 mi/hr.
The fact that the speedometer needle indicates a steady speed of 60 mi/hr
means the rate your distance is changing is constant.
In cases when we have a constant rate, we often want to find the total
amount of change in the quantity over a specific time period. The key
principle in the background is this:
Total Change in some Quantity = Rate × Time (1.2)
It is important to mention that this formula only works when we have
a constant rate, but that will be the only situation we encounter in this
course. One of the main goals of calculus is to develop a version of (1.2)
that works for non-constant rates. Here is another example; others will
occur throughout the text.
Example 1.2.1. A water pipe mounted to the ceiling has a leak. It is
dripping onto the floor below and creates a circular puddle of water. The
surface area of this puddle is increasing at a constant rate of 4 cm2/hour.
Find the surface area and dimensions of the puddle after 84 minutes.
Solution. The quantity changing is “surface area” and we are given a
“rate” and “time.” Using (1.2) with time t = 84 minutes,
Total Surface Area = Rate × Time
cm2

84
= 4 × hr
hr 60
= 5.6 cm2.
The formula for the area of a circular region of radius rq
is given at the
back of this text. Using this, the puddle has radius r = 5.6
π
= 1.335 cm
at time t = 84 minutes.
1.3 The Modeling Process

Modeling is a method used in disciplines ranging from architecture to
zoology. This mathematical technique will crop up any time we are prob-
lem solving and consciously trying to both “describe” and “predict.” In-
evitably, mathematics is introduced to add structure to the model, but
the clean equations and formulas only arise after some (or typically a lot)
of preliminary work.
A model can be thought of as a caricature in that it will pick out cer-
tain features (like a nose or a face) and focus on those at the expense of
others. It takes a lot of experience to know which models are “good” and
“bad,” in the sense of isolating the right features. In the beginning, mod-
eling will lead to frustration and confusion, but by the end of this course
our comfort level will dramatically increase. Let’s look at an illustration
of the problem solving process.
Example 1.3.1. How much time do you anticipate studying precalculus
each week?
Solution. One possible response is simply to say “a little” or “way too
much!” You might not think these answers are the result of modeling,
but they are. They are a consequence of modeling the total amount of
study time in terms of categories such as “a little,” “some,” “lots,” “way
too much,” etc. By drawing on your past experiences with math classes
and using this crude model you arrived at a preliminary answer to the
question.
Let’s put a little more effort into the problem and try to come up with
a numerical estimate. If T is the number of hours spent on precalculus a
given week, it is certainly the case that:
T = (hours in class) + (hours reading text) + (hours doing homework)
Our time in class each week is known to be 5 hours. However, the other
two terms require a little more thought. For example, if we can comfort-
ably read and digest a page of text in (on average) 15 minutes and there
are r pages of text to read during the week, then
15
(hours reading text) = r hours.
60
As for homework, if a typical homework problem takes (on average) 25
minutes and there are h homework problems for the week, then
25
(hours doing homework) = h hours.
60
We now have a mathematical model for the weekly time commitment to
precalculus:
15 25
T =5+ r + h hours.
60 60
1.3. THE MODELING PROCESS 7
Is this a good model? Well, it is certainly more informative than our

original crude model in terms of categories like “a little” or “lots.” But, the
real plus of this model is that it clearly isolates the features being used
to make our estimated time commitment and it can be easily modified as
the amount of reading or homework changes. So, this is a pretty good
model. However, it isn’t perfect; some homework problems will take a lot
more than 25 minutes!
1.4 Exercises
Problem 1.1. (a) Verify that 7685.33 sec- “The cost of the book was more than $10 and
onds is 2 hours 8 minutes 5.33 seconds. the cost of the magazine was $4.” A first step
would be these “pseudo-equations”:
(b) Which is faster: 100 mph or 150 ft/s?
(Book cost) > $10 and (Magazine cost) = $4.
(c) Gina’s salary is 1 cent/second for a
(a) John’s salary is $56,000 a year and he
40 hour work week. Tiare’s salary is
pays no taxes.
$1400 for a 40 hour work week. Who
has a higher salary? (b) John’s salary is at most $56,000 a year
(d) Suppose it takes 180 credits to get a and he pays 15% of his salary in taxes.
baccalaureate degree. You accumu- (c) John’s salary is at least $56,000 a year
late credit at the rate of one credit per and he pays more than 28% of his salary
quarter for each hour that the class in taxes.
meets per week. For instance, a class
(d) The number of students taking Math 120
that meets three hours each week of
at the UW is somewhere between 1500
the quarter will count for three cred-
and 1800 each year.
its. In addition, suppose that you
spend 2.5 hours of study outside of class (e) The cost of a new red Porsche is more
for each hour in class. A quarter is than three times the cost of a new Ford
10 weeks long. How many total hours, F-150 pickup truck.
including time spent in class and time
(f) Each week, students spend at least two
spent studying out of class, must you in-
but no more than three hours studying
vest to get a degree?
for each credit hour.
Problem 1.2. Sarah can bicycle a loop around (g) Twice the number of happy math stu-
the north part of Lake Washington in 2 hours dents exceeds five times the number of
and 40 minutes. If she could increase her av- happy chemistry students. However, all
erage speed by 1 km/hr, it would reduce her of the happy math and chemistry stu-
time around the loop by 6 minutes. How many dents combined is less than half the to-
kilometers long is the loop? tal number of cheerful biology students.
(h) The difference between Cady’s high and
Problem 1.3. The density of lead is low midterm scores was 10%. Her final
11.34 g/cm3 and the density of aluminum is exam score was 97%.
2.69 g/cm3 . Find the radius of lead and alu-
minum spheres each having a mass of 50 kg. (i) The vote tally for Gov. Tush was within
one-hundredth of one percent of one-
Problem 1.4. Marathon runners keep track half the total number of votes cast.
of their speed using units of pace = min-
utes/mile. Problem 1.6. Which is a better deal: A 10 inch
(a) Lee has a speed of 16 ft/sec; what is his diameter pizza for $8 or a 15 inch diameter
pace? pizza for $16?
(b) Allyson has a pace of 6 min/mile; what
Problem 1.7. The famous theory of relativity
is her speed?
predicts that a lot of weird things will hap-
(c) Adrienne and Dave are both running pen when you approach the speed of light
a race. Adrienne has a pace of c = 3 × 108 m/sec. For example, here is a for-
5.7 min/mile and Dave is running mula that relates the mass mo (in kg) of an
10.3 mph. Who is running faster? object at rest and its mass when it is moving
at a speed v:
Problem 1.5. Convert each of the following m
m = q o .
sentences into “pseudo-equations.” For ex- 2
1 − cv2
ample, suppose you start with the sentence:
1.4. EXERCISES 9
(a) Suppose the object moving is Dave, who (a) Find the area and radius of the puddle
has a mass of mo = 66 kg at rest. What is after 1 minute, 92 minutes, 5 hours, 1
Dave’s mass at 90% of the speed of light? day.
At 99% of the speed of light? At 99.9% of
(b) Is the radius of the puddle increasing at
the speed of light?
a constant rate?
(b) How fast should Dave be moving to have
a mass of 500 kg?
Problem 1.11. During the 1950s, Seattle was
dumping an average of 20 million gallons of
Problem 1.8. During a typical evening in Seat- sewage into Lake Washington each day.
tle, Pagliacci receives phone orders for pizza (a) How much sewage went into Lake Wash-
delivery at a constant rate: 18 orders in a typ- ington in a week? In a year?
ical 4 minute period. How many pies are sold
in 4 hours? Assume Pagliacci starts taking or- (b) In order to illustrate the amounts in-
ders at 5 : 00 pm and the profit is a constant volved, imagine a rectangular prism
rate of $11 on 10 orders. When will phone order whose base is the size of a football
profit exceed $1,000? field (100 yards × 50 yards) with height
h yards. What are the dimensions of
such a rectangular prism containing the
Problem 1.9. A typical cell in the human body sewage dumped into Lake Washington in
contains molecules of deoxyribonucleic acid, a single day? (Note: There are 7.5 gal-
referred to as DNA for short. In the cell, this lons in one cubic foot. Dumping into
DNA is all twisted together in a tight little Lake Washington has stopped; now it
packet. But imagine unwinding (straightening goes into the Puget Sound.)
out) all of the DNA from a single typical cell
and laying it “end-to-end”; then the sum total
length will be approximately 2 meters. Problem 1.12. Dave has inherited an apple
orchard on which 60 trees are planted. Under
isolate DNA these conditions, each tree yields 12 bushels
from nucleus of apples. According to the local WSU exten-
cell sion agent, each time Dave removes a tree the
yield per tree will go up 0.45 bushels. Let x be
nucleus the number of trees in the orchard and N the
yield per tree.
(a) Find a formula for N in terms of the un-
known x. (Hint: Make a table of data
with one column representing various
lay out values of x and the other column the
end−to−end corresponding values of N. After you
complete the first few rows of the table,
you need to discover the pattern.)
2m (b) What possible reason(s) might explain
why the yield goes up when you remove
trees?
Assume the human body has 1014 cells con-
taining DNA. How many times would the sum
total length of DNA in your body wrap around Problem 1.13. Congress is debating a pro-
the equator of the earth? posed law to reduce tax rates. If the current
tax rate is r %, then the proposed rate after x
Problem 1.10. A water pipe mounted to the years is given by this formula:
ceiling has a leak and is dripping onto the floor r
below, creating a circular puddle of water. The .
1
area of the circular puddle is increasing at a 1+
1
constant rate of 11 cm2 /hour. 1+
x
Rewrite this formula as a simple fraction. Use (c) The temperature at 4:30 pm is 54◦ F and
your formula to calculate the new tax rate af- the temperature at 6:15 pm is 26◦ F.
ter 1, 2, 5 and 20 years. Would tax rates in- What are the initial time, the final time,
crease or decrease over time? Congress claims the initial temperature and the final
that this law would ultimately cut peoples’ tax temperature? What is the rate of change
rates by 75 %. Do you believe this claim? in the temperature between 4:30 pm and
6:15 pm?
Problem 1.14. (a) The temperature at 7:00
am is 44◦ F and the temperature at 10:00 Problem 1.15. (a) Solve for t: 3t −7 = 11+t.
am is 50◦ F. What are the initial time, the q
final time, the initial temperature and (b) Solve for a: 1 + a1 = 3.
the final temperature? What is the rate √
of change in the temperature between (c) Solve for x: x2 + a2 = 2a + x.
7:00 am and 10:00 am? (d) Solve for t: 1 − t > 4 − 2t.
(b) Assume it is 50 F at 10:00 am and
◦
the (e) Write as a single fraction:
rate of change in the temperature be-
tween 10:00 am and 2:00 pm is the 2 1
−
same as the rate in part (a). What is the x x+1
temperature at 2:00 pm?
Chapter 2
Imposing Coordinates
You find yourself visiting Spangle, WA and dinner time is approaching.

A friend has recommended Tiff’s Diner, an excellent restaurant; how will
you find it?
Of course, the solution to this simple problem amounts to locating a
“point” on a two-dimensional map. This idea will be important in many
problem solving situations, so we will quickly review the key ideas.
2.1 The Coordinate System

If we are careful, we can develop the flow of ideas under-
lying two-dimensional coordinate systems in such a way
Q
that it easily generalizes to three-dimensions. Suppose
we start with a blank piece of paper and mark two points;
let’s label these two points “P ” and “Q.” This presents the
P
basic problem of finding a foolproof method to reconstruct
the picture. Figure 2.1: Two points in a
The basic idea is to introduce a coordinate system for plane.
the plane (analogous to the city map grid of streets), al-
lowing us to catalog points in the plane using pairs of real numbers (anal-
ogous to the addresses of locations in the city).
Here are the details. Start by drawing two perpendicular lines, called
the horizontal axis and the vertical axis , each of which looks like a copy
of the real number line. We refer to the intersection point of these two
lines as the origin . Given P in the plane, the plan is to use these two axes
to obtain a pair of real numbers (x,y) that will give us the exact location of
P. With this in mind, the horizontal axis is often called the x-axis and the
vertical axis is often called the y-axis. Remember, a typical real number
line (like the x-axis or the y-axis) is divided into three parts: the positive
numbers, the negative numbers, and the number zero (see Figure 2.2(a)).
This allows us to specify positive and negative portions of the x-axis and
y-axis. Unless we say otherwise, we will always adopt the convention that
the positive x-axis consists of those numbers to the right of the origin on
11
12 CHAPTER 2. IMPOSING COORDINATES
the x-axis and the positive y-axis consists of those numbers above the
origin on the y-axis. We have just described the xy-coordinate system for
the plane:
Positive y-axis
Negative real Positive real Origin
numbers numbers
Negative x-axis Positive x-axis

Zero
Negative y-axis
(a) Number line. (b) xy-coordinate system.
Figure 2.2: Coordinates.
2.1.1 Going from P to a Pair of Real Numbers.

Imagine a coordinate system had been drawn on our piece
y-axis of paper in Figure 2.1. Let’s review the procedure of going
P ℓ
from a point P to a pair of real numbers:
y
1. First, draw two new lines passing through P, one
parallel to the x-axis and the other parallel to the
x-axis y-axis; call these ℓ and ℓ∗ , as pictured in Figure 2.3.
x
2. Notice that ℓ will cross the y-axis exactly once; the
ℓ∗
point on the y-axis where these two lines cross will
Figure 2.3: Coordinate be called “y.” Likewise, the line ℓ∗ will cross the
pairs. x-axis exactly once; the point on the x-axis where
these two lines cross will be called “x.”
3. If you begin with two different points, like P and Q in

Figure 2.1, you will see that the two pairs of points
you obtain will be different; i.e., if Q gives you the
pair (x∗ ,y∗ ), then either x 6= x∗ or y 6= y∗ . This shows
that two different points in the plane give two differ-
ent pairs of real numbers and describes the process
of assigning a pair of real numbers to the point P.
The great thing about the procedure we just described is that it is

reversible! In other words, suppose you start with a pair of real numbers,
say (x,y). Locate the number x on the x-axis and the number y on the
y-axis. Now draw two lines: a line ℓ parallel to the x-axis passing through
the number y on the y-axis and a line ℓ∗ parallel to the y-axis passing
through the number x on the x-axis. The two lines ℓ and ℓ∗ will intersect
2.2. THREE FEATURES OF A COORDINATE SYSTEM 13
in exactly one point in the plane, call it P. This procedure describes

how to go from a given pair of real numbers to a point in the plane. In
addition, if you start with two different pairs of real numbers, then the
corresponding two points in the plane are going to be different. In the
future, we will constantly be going back and forth between points in the
plane and pairs of real numbers using these ideas.
Definition 2.1.1. Coordinate System: Every point P in the xy-plane cor-

responds to a unique pair of real numbers (x, y), where x is a number on the
horizontal x-axis and y is a number on the vertical y-axis; for this reason,
we commonly use the notation “P = (x,y).”
Having specified positive and negative directions on y-axis

the horizontal and vertical axes, we can now divide our
two dimensional plane into four quadrants . The first
quadrant corresponds to all the points where both co- Second First
Quadrant Quadrant
ordinates are positive, the second quadrant consists of
points with the first coordinate negative and the second Third Fourth x-axis
Quadrant Quadrant
coordinate positive, etc. Every point in the plane will lie
in one of these four quadrants or on one of the two axes.
This quadrant terminology is useful to give a rough sense
of location, just as we use the terminology “Northeast,
Figure 2.4: Quadrants in the
Northwest, Southwest and Southeast” when discussing xy-plane.
locations on a map.
2.2 Three Features of a Coordinate System

A coordinate system involves scaling, labeling and units on each of the
axes.
2.2.1 Scaling
Sketch two xy coordinate systems. In the first, make the scale on each
axis the same. In the second, assume “one unit” on the x axis has the
same length as “two units” on the
y axis. Plot
2 the points (1,1), (−1,1),
− 45 , 16 3 9 2 4 1 1 1 1 4 3 9 4 16

25
, − ,
5 25
, − ,
5 25
, − ,
5 25
, (0,0), ,
5 25
, ,
5 25
, ,
5 25
, ,
5 25
, (1,1).
Both pictures illustrate how the points lie on a parabola in the xy-
coordinate system, but the aspect ratio has changed. The aspect ratio is
defined by this fraction:
def length of one unit on the vertical axis

aspect ratio = .
length of one unit on the horizontal axis
Figure 2.5(a) has aspect ratio 1, whereas Figure 2.5(b) has aspect ratio
1
2
. In problem solving, you will often need to make a rough assumption
about the relative axis scaling. This scaling will depend entirely on the
1.0 y-axis
0.8
0.6
1.0 y-axis
0.4 0.8
0.6
0.2 0.4
x-axis 0.2 x-axis
−1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0
(a) Aspect ratio = 1. (b) Aspect ratio = 12 .
Figure 2.5: Coordinates.
information given in the problem. Most graphing devices will allow you
to specify the aspect ratio.
2.2.2 Axes Units

Sometimes we are led to coordinate systems where each of the two axes
involve different types of units (labels). Here is a sample, that illustrates
the power of using pictures.
Example 2.2.1. As the marketing director of Turboweb software, you have

been asked to deliver a brief message at the annual stockholders meeting
on the performance of your product. Your staff has assembled this tabu-
lar collection of data; how can you convey the content of this table most
clearly?
TURBOWEB SALES (in $1000’s)

week sales week sales week sales week sales week sales
1 11.0517 11 30.0417 21 81.6617 31 221.98 41 603.403
2 12.214 12 33.2012 22 90.2501 32 245.325 42 666.863
3 13.4986 13 36.693 23 99.7418 33 271.126 43 736.998
4 14.9182 14 40.552 24 110.232 34 299.641 44 814.509
5 16.4872 15 44.8169 25 121.825 35 331.155 45 900.171
6 18.2212 16 49.5303 26 134.637 36 365.982 46 994.843
7 20.1375 17 54.7395 27 148.797 37 404.473 47 1099.47
8 22.2554 18 60.4965 28 164.446 38 447.012 48 1215.1
9 24.596 19 66.8589 29 181.741 39 494.024 49 1342.9
10 27.1828 20 73.8906 30 200.855 40 545.982 50 1,484.131
One idea is to simply flash an overhead slide of this data to the audi-
ence; this can be deadly! A better idea is to use a visual aid. Suppose
we let the variable x represent the week and the variable y represent the
gross sales (in thousands of dollars) in week x. We can then plot the
points (x,y) in the xy-coordinate system; see Figure 2.6.
Notice, the units on the two axes are very different: y-axis units are
“thousands of dollars” and x-axis units are “weeks.” In addition, the
aspect ratio of this coordinate system is not 1. The beauty of this picture
is the visual impact it gives your audience. From the coordinate plot we
can get a sense of how the sales figures are dramatically increasing. In
fact, this plot is good evidence you deserve a big raise!
2.3. A KEY STEP IN ALL MODELING PROBLEMS 15
y-axis (Thousands of Dollars)
Mathematical modeling is all about relating concrete 1,400
phenomena and symbolic equations, so we want to em- 1,200
1,000
brace the idea of visualization. Most typically, visualiza- 800
600
tion will involve plotting a collection of points in the plane. 400 x-axis
200 (Weeks)
This can be achieved by providing a “list” or a “prescrip-
tion” for plotting the points. The material we review in the 10 20 30 40 50
next couple of sections makes the transition from sym-

bolic mathematics to visual pictures go more smoothly. Figure 2.6: Turboweb sales.
2.3 A Key Step in all Modeling Problems

The initial problem solving or modeling step of deciding on a choice of
xy-coordinate system is called imposing a coordinate system : There will
often be many possible choices; it takes problem solving experience to
develop intuition for a “natural” choice. This is a key step in all modeling
problems.
Example 2.3.1. Return to the tossed ball scenario on page 1. How do we
decide where to draw a coordinate system in the picture?
Figure 2.7 on page 16 shows four natural choices of xy-coordinate
system. To choose a coordinate system we must specify the origin. The
four logical choices for the origin are either the top of the cliff, the bottom
of the cliff, the launch point of the ball or the landing point of the ball.
So, which choice do we make? The answer is that any of these choices
will work, but one choice may be more natural than another. For exam-
ple, Figure 2.7(b) is probably the most natural choice: in this coordinate
system, the motion of the ball takes place entirely in the first quadrant,
so the x and y coordinates of any point on the path of the ball will be
non-negative.
Example 2.3.2. Michael and Aaron are running toward each other, be-
ginning at opposite ends of a 10,000 ft. airport runway, as pictured in
Figure 2.8 on page 17. Where and when will these guys collide?
Solution. This problem requires that we find the “time” and “location” of
the collision. Our first step is to impose a coordinate system.
We choose the coordinate system so that Michael is initially located
at the point M = (0, 0) (the origin) and Aaron is initially located at the
point A = (10,000, 0). To find the coordinates of Michael after t seconds,
we need to think about how distance and time are related.
Since Michael is moving at the rate of 15 ft/second, then after one
second he is located 15 feet right of the origin; i.e., at the point (15, 0).
After 2 seconds, Michael has moved an additional 15 feet, for a total of
30 feet; so he is located at the point (30, 0), etc. Conclude Michael has
traveled 15t ft. to the right after t seconds; i.e., his location is the point
y-axis
y-axis
Path of tossed ball.
x-axis
Cliff.
Cliff.
x-axis
(a) Origin at the top of the ledge. (b) Origin at the bottom of the
ledge.
Path of tossed ball. y-axis

y-axis
x-axis
Cliff. Cliff.
x-axis
(c) Origin at the landing point. (d) Origin at the launch point.
Figure 2.7: Choices when imposing an xy-coordinate system.
M(t) = (15t, 0). Similarly, Aaron is located 8 ft. left of his starting location
after 1 second (at the point (9,992, 0)), etc. Conclude Aaron has traveled
8t ft. to the left after t seconds; i.e., his location is the point A(t) =
(10,000 − 8t, 0).
The key observation required to solve the problem is that the point
of collision occurs when the coordinates of Michael and Aaron are equal.
Because we are moving along the horizontal axis, this amounts to finding
where and when the x-coordinates of M(t) and A(t) agree. This is a
straight forward algebra problem:
15t = 10,000 − 8t (2.1)

23t = 10,000
t = 434.78
To the nearest tenth of a second, the runners collide after 434.8

seconds. Plugging t = 434.78 into either expression for the position:
M(434.8) = (15(434.8), 0) = (6,522, 0).
2.4. DISTANCE 17
ft ft
Michael: 15 sec Aaron: 8 sec y-axis
x-axis
M = (0,0) A = (10,000,0)
10,000ft
(a) The physical picture. (b) The xy-coordinate picture.
y-axis NOT TO SCALE!

Michael starts here. Aaron starts here.
M(t) = (15t,0) A(t) = (10,000 − 8t,0)

(0,0)
x-axis
M after t seconds. A after t seconds.
(c) Building a visual model
Michael: 10 mph. Aaron: 8 mph.
6,522 feet to collision point.
(d) Michael and Aaron’s collision point.
Figure 2.8: Michael and Aaron running head-on.
2.4 Distance
We end this Chapter with a discussion of direction and distance in the
plane. To set the stage, think about the following analogy:
Example 2.4.1. You are in an airplane flying from Denver to New York.
How far will you fly? To what extent will you travel north? To what extent
will you travel east?
Consider two points P = (x1,y1) and Q = (x2,y2) in the xy coordinate
system, where we assume that the units on each axis are the same ;
for example, both in units of “feet.” Imagine starting at the location P
(Denver) and flying to the location Q (New York) along a straight line
segment; see Figure 2.9(a). Now ask yourself this question: To what
overall extent have the x and y coordinates changed?
To answer this, we introduce visual and notational aides into this
figure. We have inserted an “arrow” pointing from the starting position P
to the ending position Q; see Figure 2.9(b). To simplify things, introduce
the notation ∆x to keep track of the change in the x-coordinate and ∆y
y-axis y-axis
Q Q = (x2 ,y2 )
y2 Ending
End (stop) point.
?? here. d
?? ∆y
Begin (start) here. y1

P P = (x1 ,y1 )
Beginning point.
x-axis x-axis
x1 x2
??
∆x
(a) Starting and stopping points. (b) Coordinates for P and Q.
Figure 2.9: The meaning of ∆x and ∆y.
to keep track of the change in the y-coordinate, as we move from P to Q.

Each of these quantities can now be computed:
∆x = change in x-coordinate going from P to Q (2.2)

= (x-coord of ending point) − (x-coord of beginning point)
= x2 − x1
∆y = change in y-coordinate going from P to Q
= (y-coord of ending point) − (y-coord of beginning point)
= y2 − y1.
We can interpret ∆x and ∆y using the right triangle in Figure 2.9(b).

This means we can use the Pythagorean Theorem to write:
d2 = (∆x)2 + (∆y)2;
that is,
p
d= (∆x)2 + (∆y)2,
which tells us the distance d from P to Q. In other words, d is the distance

we would fly if we had flown along that line segment connecting the two
points. As an example, if P = (1, 1) and Q = (5, 4), then ∆x = 5 − 1 = 4,
∆y = 4 − 1 = 3 and d = 5.
There is a subtle idea behind the way we defined ∆x and ∆y: You need
to specify the “beginning” and “ending” points used to do the calcula-
tion in Equations 2.2. What happens if we had reversed the choices in
Figure 2.9?
Then the quantities ∆x and ∆y will both be negative and the lengths of
the sides of the right triangle are computed by taking the absolute value
2.4. DISTANCE 19
of ∆x and ∆y. As far as a distance calculation is concerned, the previous

formula still works because of this algebra equality:
p
d = (∆x)2 + (∆y)2
p
= (|∆x|)2 + (|∆y|)2.
We will sometimes refer to ∆x and ∆y as directed distances in the x and

y directions. The notion of directed distance becomes important in our
discussion of lines in Chapter 4 and later when you learn about vectors ;
it is also very important in calculus.
For example, if P = (5, 4) and Q = (1, 1), then ∆x =
1 − 5 = −4, ∆y = 1 − 4 = −3 and d = 5. y-axis
P = (x1 , y1 )
Important Fact 2.4.2 (Distance formula). If P = (x1, y1) y1 Beginning
point.
and Q = (x2, y2) are two points in the plane, then the d
|∆y|
straight line distance between the points (in the same units
as the two axes) is given by the formula y2
Q = (x2 , y2 )
Ending point.
p
d = (∆x)2 + (∆y)2 x-axis
p x2 x1
= (x2 − x1)2 + (y2 − y1)2. (2.3)
|∆x|
Figure 2.10: A different di-

rection.
If your algebra is a little rusty, a very common mistake may crop up

when you are using the distance formula. For example,
!!!
p ? √ √
32 + 42 = 32 + 42 CAUTION
√ !!!
?
9 + 16 = 3+4
5 =
6 7.
Notice, you have an impossible situation: 5 is never equal to 7.

Example 2.4.3. Two cars depart from a four way inter-

section at the same time, one heading East and the other
heading North. Both cars are traveling at the constant
speed of 30 ft/sec. Find the distance (in miles) between the two cars after
1 hour 12 minutes. In addition, determine when the two cars would be
exactly 1 mile apart.
Solution. Begin with a picture of the situation. We have

y-axis
(North) indicated the locations of the two vehicles after t seconds
(0,b) and the distance d between them at time t. By the dis-
tance formula, the distance between them is
d
x-axis
p
(East)
d = (a − 0)2 + (0 − b)2
p
= a2 + b2.
(a,0)
This formula is a first step; the difficulty is that we have
traded the mystery distance d for two new unknown num-
Figure 2.11: Two departing bers a and b. To find the coordinate a for the Eastbound
cars. car, we know the car is moving at the rate of 30 ft/sec, so
it will travel 30t feet after t seconds; i.e., a = 30t. Similarly,
we find that b = 30t. Substituting into the formula for d
we arrive at
p
d = (30t)2 + (30t)2
p
= 2t2(30)2
√
= 30t 2.
First, we need to convert 1 hour and 12 minutes into seconds so that

our formula can be used:
1 hr 12 min = 1 + 12/60 hr
= 1.2 hr
60 min 60 sec

= (1.2hr)
hr min
= 4,320 sec.
Substituting t = 4,320 sec and recalling that 1 mile = 5,280 feet, we arrive
at
√
d = 129,600 2 feet
= 183,282 feet
= 34.71 miles.
For the second question, we specify the distance being 1 mile and want
to find when this occurs. The idea is to set d equal to 1 mile and solve for
2.4. DISTANCE 21
t. However, we need to be careful, since the units for d are feet:

√
30t 2 = d
= 5,280
Solving for t:
5,280
t = √
30 2
= 124.45 seconds
= 2 minutes 4 seconds.
The two cars will be 1 mile apart in 2 minutes, 4 seconds.
2.5 Exercises
Problem 2.1. In the following four cases, let P Problem 2.4. Erik’s disabled sailboat is float-
be the initial (starting) point and Q the ending ing at a stationary location 3 miles East and
point; recall Equation 2.2 and Figure 2.10 on 2 miles North of Kingston. A ferry leaves
Page 19. Compute d = the distance from P to Kingston heading due East toward Edmonds
√ and ∆y. Give your answer in exact form;
Q, ∆x at 12 mph. At the same time, Erik leaves the
eg. 2 is an exact√answer, whereas 1.41 is an sailboat in a dinghy heading due South at 10
approximation of 2. ft/sec (hoping to intercept the ferry). Edmonds
is 6 miles due East of Kingston.
(a) P = (0,0), Q = (1,1).
sailboat
(b) P = (2,1), Q = (1, − 1).
(c) P = (−1,2), Q = (4, − 1).
Kingston Edmonds
(d) P = (1,2), Q = (1 + 3t,3 + t), where t is a
constant.
Problem 2.2. Start with two points M = (a,b) North

and N = (s,t) in the xy-coordinate system. Let
d be the distance between these two points. Ballard
Answer these questions and make sure you
can justify your answers: UDub
p
(a) TRUE or FALSE: d = (a − s)2 + (b − t)2 .
(a) Compute Erik’s speed in mph and the
p
(b) TRUE or FALSE: d = (a − s)2 + (t − b)2 .
p Ferry speed in ft/sec.
(c) TRUE or FALSE: d = (s − a)2 + (t − b)2 .
(b) Impose a coordinate system and com-
(d) Suppose M is the beginning point and N plete this table of data concerning loca-
is the ending point; recall Equation 2.2 tions (i.e., coordinates) of Erik and the
and Figure 2.10 on Page 19. What is ∆x? ferry. Insert into the picture the loca-
What is ∆y? tions of the ferry and Erik after 7 min-
(e) Suppose N is the beginning point and M utes.
is the ending point; recall Equation 2.2 Distance
and Figure 2.10 on Page 19. What is ∆x? Time Ferry Erik
Between
What is ∆y? 0 sec
(f) If ∆x=0, what can you say about the re- 30 sec
lationship between the positions of the 7 min
two points M and N ? If ∆y=0, what can t hr
you say about the relationship between
the positions of the two points M and N? (c) Explain why Erik misses the ferry.
(Hint: Use some specific values for the (d) After 10 minutes, a Coast Guard boat
coordinates and draw some pictures to leaves Kingston heading due East at a
see what is going on.) speed of 25 ft/sec. Will the Coast Guard
boat catch the ferry before it reaches Ed-
Problem 2.3. Steve and Elsie are camping in monds? Explain.
the desert, but have decided to part ways.
Steve heads North, at 6 AM, and walks steadily Problem 2.5. Suppose two cars depart from
at 3 miles per hour. Elsie sleeps in, and starts a four way intersection at the same time, one
walking West at 3.5 miles per hour starting at heading north and the other heading west. The
8 AM. car heading north travels at the steady speed
When will the distance between them be 25
of 30 ft/sec and the car heading west travels
miles? at the steady speed of 58 ft/sec.
2.5. EXERCISES 23
(a) Find an expression for the distance be- Brooke

tween the two cars after t seconds. ocean
(b) Find the distance in miles between the
two cars after 3 hours 47 minutes. 5 mi
kayak reaches shore here
(c) When are the two cars 1 mile apart?
A Kono’s
shore
Problem 2.6. Allyson and Adrian have decided 6 mi

to connect their ankles with a bungee cord;
one end is tied to each person’s ankle. The (a) If she paddles along a straight line
cord is 30 feet long, but can stretch up to 90 course to the shore, find an expression
feet. They both start from the same location. that computes the total time to reach
Allyson moves 10 ft/sec and Adrian moves 8 lunch in terms of the location where
ft/sec in the directions indicated. Brooke beaches the boat.
(b) Determine the total time to reach Kono’s
if she paddles directly to the point “A”.
20 ft (c) Determine the total time to reach Kono’s
Building if she paddles directly to Kono’s.
(d) Do you think your answer to (b) or (c) is
the minimum time required for Brooke
to reach lunch?
(e) Determine the total time to reach Kono’s
if she paddles directly to a point on the
Allyson
30 ft shore half way between point “A” and
Kono’s. How does this time compare to
the times in parts (b) and (c)? Do you
need to modify your answer to part (d)?
start
Adrian
Problem 2.8. A spider is located at the po-
(a) Where are the two girls located after 2 sition (1,2) in a coordinate system, where the
seconds? units on each axis are feet. An ant is located at
the position (15,0) in the same coordinate sys-
(b) After 2 seconds, will the slack in the tem. Assume the location of the spider after t
bungee cord be used up? minutes is s(t) = (1 + 2t,2 + t) and the location
(c) Determine when the bungee cord first of the ant after t minutes is a(t) = (15 − 2t,2t).
becomes tight; i.e. there is no slack (a) Sketch a picture of the situation, indi-
in the line. Where are the girls located cating the locations of the spider and ant
when this occurs? at times t = 0,1,2,3,4,5 minutes. Label
the locations of the bugs in your picture,
(d) When will the bungee cord first touch using the notation s(0), s(1),...,s(5), a(0),
the corner of the building? (Hint: Use a(1), ..., a(5).
a fact about “similar triangles”.)
(b) When will the x-coordinate of the spider
equal 5? When will the y-coordinate of
the ant equal 5?
Problem 2.7. Brooke is located 5 miles out
(c) Where is the spider located when its
from the nearest point A along a straight
y-coordinate is 3?
shoreline in her seakayak. Hunger strikes and
she wants to make it to Kono’s for lunch; see (d) Where is each bug located when the
picture. Brooke can paddle 2 mph and walk 4 y-coordinate of the spider is twice as
mph. large as the y-coordinate of the ant?
(e) How far apart are the bugs when their (a) If x 6= 1,
x-coordinates coincide? Draw a pic-
ture, indicating the locations of each bug x2 − 1 x2 + (−1)1
=
when their x-coordinates coincide. x+1 x+1
x2 −1
(f) A sugar cube is located at the position = +
(9,6). Explain why each bug will pass x 1
through the position of the sugar cube. = x−1
Which bug reaches the sugar cube first?
(b)
(g) Find the speed of each bug along its line
of motion; which bug is moving faster? (x + y)2 − (x − y)2 = (x2 + y2 ) − x2 − y2
= 0
Problem 2.9. A Ferrari is heading south at a
constant speed on Broadway (a north/south
(c) If x 6= 4,
street) at the same time a Mercedes is heading
west on Aloha Avenue (an east/west street). 9(x − 4)2 32 (x − 4)2
The Ferrari is 624 feet north of the intersec- =
3x − 12 3x − 12
tion of Broadway and Aloha, at the same time
(3x − 12)2
that the Mercedes is 400 feet east of the inter- =
section. Assume the Mercedes is traveling at 3x − 12
the constant speed of 32 miles/hour. Find the = 3x − 12.
speed of the Ferrari so that a collision occurs
in the intersection of Broadway and Aloha.
Problem 2.12. Assume α, β are nonzero con-
Problem 2.10. Two planes flying opposite di- stants. Solve for x.
rections (North and South) pass each other 80 1
(a) αx + β =
miles apart at the same altitude. The North- αx−β
bound plane is flying 200 mph (miles per hour) (b) 1

+ 1
= 1
α β x
and the Southbound plane is flying 150 mph.
1 1
How far apart are the planes in 20 minutes? (c) α + β = x
When are the planes 300 miles apart?
Problem 2.11. Here is a list of some algebra Problem 2.13. Simplify as far as possible.
problems with ”solutions.” Some of the solu-
(a) (1 − t)2 + (2 + 2t)2
tions are correct and some are wrong. For
each problem, determine: (i) if the answer is (b) (t + 1)2 + (−t − 1)2 − 2
correct, (ii) if the steps are correct, (iii) identify 1 1
(c) (write as a single fraction)
−
any incorrect steps in the solution (noting that t−1 t+1
the answer may be correct but some steps may

p
(d) (2 + t)2 + 4t2
not be correct).
Chapter 3
Three Simple Curves
Before we discuss graphing, we first want to become ac- y-axis

quainted with the sorts of pictures that will arise. This is A typical curve.
surprisingly easy to accomplish: Impose an xy-coordinate
system on a blank sheet of paper. Take a sharp pencil and
begin moving it around on the paper. The resulting pic- x-axis
ture is what we will call a curve. For example, here is a
sample of the sort of “artwork” we are trying to visualize.
A number of examples in the text will involve basic curves
in the plane. When confronted with a curve in the plane,
the fundamental question we always try to answer is this:
Figure 3.1: A typical curve.
Can we give a condition (think of it as a “test”) that will
tell us precisely when a point in the plane lies on a curve?
Typically, the kind of condition we will give involves an equation in
two variables (like x and y). We consider the three simplest situations in
this chapter: horizontal lines, vertical lines and circles.
3.1 The Simplest Lines

Undoubtedly, the simplest curves in the plane are the y-axis
horizontal and vertical lines. For example, sketch a line (−1, 2) (0, 2) (2, 2) (x, 2), a typical point.
parallel to the x-axis passing through 2 on the y-axis; the ℓ

x-axis
result is a horizontal line ℓ, as pictured. This means the
line ℓ passes through the point (0, 2) in our coordinate sys- Figure 3.2: The points (x, y).
tem. A concise symbolic prescription for ALL of the points
on ℓ can be given using “set notation”:
ℓ = {(x, 2)|x is any real number}.
We read the right-hand side of this expression as “the set of all points
(x, 2) where x = any real number.” Notice, the points (x, y) on the line ℓ
are EXACTLY the ones that lead to solutions of the equation y = 2; i.e.,
take any point on this line, plug the coordinates into the equation y = 2
25
26 CHAPTER 3. THREE SIMPLE CURVES
and you get a true statement. Because the equation does not involve the
variable x and only constrains y to equal 2, we see that x can take on any
real value. In short, we see that plotting all of the solutions (x, y) to the
equation y = 2 gives the line ℓ. We usually refer to the set of all solutions
of the equation y = 2 as the graph of the equation y = 2.
As a second example, sketch the vertical line m pass-
y-axis
ing through 3 on the x-axis; this means the line m passes
m
through the point (3, 0) in our coordinate system. A con-
(3, y), a typical point. cise symbolic prescription for ALL of the points on m can
be given using “set notation”:
(3, 3)
m = {(3, y)|y is any real number}.
(3, 0)
x-axis Notice, the points (x, y) on the line m are EXACTLY
(3, −2)
the ones that lead to solutions of the equation x = 3; i.e.,
take any point on this line, plug the coordinates into the
equation x = 3 and you get a true statement. Because the
Figure 3.3: Stacked points. equation does not involve the variable y and only specifies
that x = 3, y can take on any real number value. Plotting all of the
solutions (x, y) to the equation x = 3 gives the line m. We usually refer to
the set of all solutions of the equation x = 3 as the graph of the equation
x = 3.
These two simple examples highlight our first clear connection be-
tween a geometric figure and an equation; the link is achieved by plot-
ting all of the solutions (x, y) of the equation in the xy-coordinate system.
These observations work for any horizontal or vertical line.
Definition 3.1.1. Horizontal and Vertical Lines: A horizontal line ℓ
passing through k on the y-axis is precisely a plot of all solutions (x, y) of
the equation y = k; i.e., ℓ is the graph of y = k. A vertical line m pass-
ing through h on the x-axis is precisely a plot of all solutions (x, y) of the
equation x = h; i.e., m is the graph of x = h.
3.2 Circles
Another common curve in the plane is a circle. Let’s see how to relate
a circle and an equation involving the variables x and y. As a special
case of the distance formula (2.3), suppose P = (0, 0) is the origin and
Q = (x, y) is any point in the plane; then
p
distance from P to Q = (x − 0)2 + (y − 0)2
p
= x2 + y2.
This calculation tells
p us that a point (x, y) is of distance r from the origin
if and only if r = x2 + y2 or, squaring each side, that x2 + y2 = r2. This
shows
{(x, y)|distance (x, y) to origin is r} = {(x, y)|x2 + y2 = r2 }. (3.1)
3.2. CIRCLES 27
What is the left-hand side of Equation 3.1? To pic- Pencil.
ture all points in the plane of distance r from the origin, r
fasten a pencil to one end of a non-elastic string (a string r

that will not stretch) of length r and tack the other end to
Start. Draw with a
the origin. Holding the string tight, the pencil point will tight string.
locate a point of distance r from the origin. We could visu-
alize all such points by simply moving the pencil around Figure 3.4: Drawing a circle.
the origin, all the while keeping the string tight.
What is the right-hand side of Equation 3.1? A point (x, y) in the
right-hand set is a solution to the equation x2 + y2 = r2; i.e., if we plug in
the coordinates we get a true statement. For example, in Figure
3.5 we
−r
r
plot eight solutions (r, 0), (−r, 0), (0, r), (0, −r), A = √2 , √2 , B = √r2 , √
r
2
,

−r √
C= √ 2
, −r2 , D = √−r √r
2
, 2 , of the equation. To see that the last point is
2 2
−r √r r2 r2
a solution, here is the sample calculation: √ 2
+ 2
= 2
+ 2
= r2.
Since the two sides of Equation 3.1 are equal, drawing (0, r)
D A
the circle of radius r is the same as plotting all of the solu-
tions of the equation x2 + y2 = r2. The same reasoning can (−r, 0) (r, 0)
be used to show that drawing a circle of radius r centered

at a point (h, k) is the same as plotting all of the solutions
C B
of the equation: (x − h)2 + (y − k)2 = r2. We usually refer to (0, −r)
the set of all solutions of the equation as the graph of the

Figure 3.5: Computing
equation. points.
Definition 3.2.1 (Circles). Let (h, k) be a given point in the

y-axis
xy-plane and r > 0 a given positive real number. The circle
(x,y)
of radius r centered at (h, k) is precisely all of the solutions
r
(x,y) of the equation
(h,k)
(x − h)2 + (y − k)2 = r2;
x-axis
i.e., the circle is the graph of this equation.
We refer to the equation in the box as the standard Figure 3.6: Defining a circle.
form of the equation of a circle. From this equation you
know both the center and radius of the circle described.
Be very careful with the minus signs “−” in the standard form for a circle
equation. For example, the equation
!!!
(x + 3)2 + (y − 1)2 = 7
CAUTION
!!!
is NOT in standard form. We can rewrite it in standard form:
√
(x − (−3))2 + (y − 1)2 = ( 7)2;
√
so, this equation describes a circle of radius 7 centered at (−3, 1).
Examples 3.2.2. Here are some of the ways we can discuss circles:
1. The circle of radius 1 centered at the origin is the graph of the equation
x2 + y2 = 1. This circle is called the unit circle and will be used
extensively.
2. A circle of radius 3 centered at the point (h, k) = (1, −1) is the graph of
the equation (x−1)2 +(y−(−1))2 = 32; or, equivalently (x−1)2 +(y+1)2 =
32; or, equivalently x2 + y2 − 2x + 2y = 7.
√
3. The circle of radius 5 centered at (2, −3) does not pass through the
origin; this is because (0, 0) is not a solution of the equation (x − 2)2 +
(y + 3)2 = 5.
3.3 Intersecting Curves I

In many problem solving situations, we will have two curves in the plane
and need to determine where the curves intersect one another. Before
we discuss a general procedure, let’s make sure we really understand
the meaning of the word “intersect.” From Latin, the word “inter” means
“within or in between” and the word “sectus” means “to cut.” So, the
intersection of two curves is the place where the curves “cut into” each
other; in other words, where the two curves cross one another.
If the pictures of two curves are given to us up front, we can often
visually decide whether or not they intersect. This is one good reason for
drawing a picture of any physical problem we are trying to solve. We will
need a small bag of tricks used for finding intersections of curves. We
begin with intersections involving the curves studied in this section.
Two different horizontal lines (or two different vertical lines) will never
intersect. However, a horizontal line always intersects a vertical line ex-
actly once; Figure 3.7(a). Given a circle and a horizontal or vertical line,
we may or may not have an intersection. Looking at Figure 3.7(b), you
can convince yourself a given horizontal or vertical line will intersect a
circle in either two points, one point or no points. This analysis is all pic-
torial; how do you find the explicit coordinates of an intersection point?
Let’s look at a sample problem to isolate the procedure used.
3.3. INTERSECTING CURVES I 29
y-axis Horizontal
line:
y = k.
Point:
(h,k). x-axis
Vertical
line:
x = h.
(a) Line equations. (b) Possible intersections.

Figure 3.7: Circles and lines.
Example 3.3.1. Glo-Tek Industries has designed a new halogen street

light fixture for the city of Seattle. According to the product literature, when
placed on a 50 foot light pole, the resulting useful illuminated area is a
circular disc 120 feet in diameter. Assume the light pole is located 20 feet
east and 40 feet north of the intersection of Parkside Ave. (a north/south
street) and Wilson St. (an east/west street). What portion of each street is
illuminated?
y-axis
(Parkside)
Solution. The illuminated area is a circular disc whose di- Illuminated
zone.
ameter and center are both known. Consequently, we
really need to study the intersection of this circle with the P
two streets. Begin by imposing the pictured coordinate
system; we will use units of feet for each axis. The illumi-
nated region will be a circular disc centered at the point x-axis
(20, 40) in the coordinate system; the radius of the disc (Wilson)
will be r = 60 feet.
R S
We need to find the points of intersection P, Q, R, and S
Q
of the circle with the x-axis and the y-axis. The equation
for the circle with r = 60 and center (h, k) = (20, 40) is
Figure 3.8: Illuminated
street.
(x − 20)2 + (y − 40)2 = 3600.
To find the circular disc intersection with the y-axis, we

have a system of two equations to work with:
(x − 20)2 + (y − 40)2 = 3,600;

x = 0.
To find the intersection points we simultaneously solve both equations.

To do this, we replace x = 0 in the first equation (i.e., we impose the
conditions of the second equation on the first equation) and arrive at

(0 − 20)2 + (y − 40)2 = 3,600;
400 + (y − 40)2 = 3,600;
(y − 40)2 = 3,200;
√
(y − 40) = ± 3,200
√
y = 40 ± 3,200
= −16.57 or 96.57.
Notice, we have two solutions. This means that the circle and y-axis
intersect at the points P = (0,96.57) and Q = (0, − 16.57). Similarly, to find
the circular disc intersection with the x-axis, we have a system of two
equations to work with:
(x − 20)2 + (y − 40)2 = 3,600;
y = 0.
Replace y = 0 in the first equation (i.e., we impose the conditions of the
second equation on the first equation) and arrive at
(x − 20)2 + (0 − 40)2 = 3,600;
(x − 20)2 = 2,000;
√
(x − 20) = ± 2,000;
√
x = 20 ± 2,000
= −24.72 or 64.72.
Conclude the circle and x-axis intersect at the points S = (64.72,0) and
R = (−24.72,0).
The procedure we used in the solution of Example 3.3.1 gives us a
general approach to finding the intersection points of circles with hori-
zontal and vertical lines; this will be important in the exercises.
3.4 Summary
• Every horizontal line has equation of the form y = c.
• Every vertical line has equation of the form x = c.
• Every circle has equation of the form
(x − h)2 + (y − k)2 = r2
where (h,k) is the center of the circle, and r is the circle’s radius.
3.5. EXERCISES 31
3.5 Exercises
Problem 3.1. This exercise emphasizes the rider
“mechanical aspects” of circles and their equa-
tions. 60 feet
(a) Find an equation whose graph is a circle 100 feet
of radius 3 centered at (−3,4).
operator
(b) Find an equation whose graph is a cir-
cle of diameter 21 centered at the point ground level
(3, − 11
3 ). 62 ft. tower
(c) Find four different equations whose 24 feet
graphs are circles of radius 2 through
(1,1). (a) Impose a coordinate system.
(d) Consider the equation (x − 1)2 + (y + 1)2 = (b) Suppose a rider is located at the point in
4. Which of the following points lie on the picture, 100 feet above the ground.
the graph of √this equation:√(1,1), (1, − 1), If the rider drops an ice cream cone
(1, − 3), (1 + 3,0), (0, − 1 − 3), (0,0). straight down, where will it land on the
ground?
Problem 3.2. Find the center and radius of (c) The ride operator is standing 24 feet to
each of the following circles. one side of the support tower on the level
(a) x2 − 6x + y2 + 2y − 2 = 0 ground at the location in the picture.
Determine the location(s) of a rider on
(b) x2 + 4x + y2 + 6y + 9 = 0 the Ferris Wheel so that a dropped ice
(c) x2 + 13 x + y2 − 10
= 127 cream cone lands on the operator. (Note:
3 y 9
There are two answers.)
(d) x2 + y2 = 23 x − y + 35
16
Problem 3.3. Water is flowing from a major Problem 3.5. Aleko’s Pizza has delivered a
broken water main at the intersection of two beautiful 16 inch diameter pie to Lee’s dorm
streets. The resulting puddle of water is circu- room. The pie is sliced into 8 equal sized
lar and the radius r of the puddle is given by pieces, but Lee is such a non-conformist he
the equation r = 5t feet, where t represents cuts off an edge as pictured. John then takes
time in seconds elapsed since the the main one of the remaining triangular slices. Who
broke. has more pizza and by how much?
(a) When the main broke, a runner was lo-
cated 6 miles from the intersection. The John’s part
runner continues toward the intersec-
tion at the constant speed of 17 feet per
second. When will the runner’s feet get
wet?
(b) Suppose, instead, that when the main
broke, the runner was 6 miles east, and
5000 feet north of the intersection. The
runner runs due west at 17 feet per sec-
ond. When will the runner’s feet get wet?
Problem 3.4. An amusement park Ferris

Wheel has a radius of 60 feet. The center of Lee’s part
the wheel is mounted on a tower 62 feet above
the ground (see picture). For these questions,
the wheel is not turning.
sailboat
Problem 3.6. A crawling tractor sprinkler is
located as pictured below, 100 feet South of a
sidewalk. Once the water is turned on, the Kingston Edmonds
sprinkler waters a circular disc of radius 20
feet and moves North along the hose at the rate
of 21 inch/second. The hose is perpendicular
to the 10 ft. wide sidewalk. Assume there is
grass on both sides of the sidewalk. North
Ballard
N
hose UDub
W E
S
(a) Find the equations for the lines along
which the ferry is moving and draw in
these lines.
sidewalk
(b) The sailboat has a radar scope that will

detect any object within 3 miles of the
tractor sprinkler sailboat. Looking down from above, as
in the picture, the radar region looks
like a circular disk. The boundary is
(a) Impose a coordinate system. Describe the ”edge” or circle around this disc, the
the initial coordinates of the sprinkler interior is the inside of the disk, and
and find equations of the lines forming the exterior is everything outside of the
the North and South boundaries of the disk (i.e. outside of the circle). Give
sidewalk. a mathematical (equation) description of
the boundary, interior and exterior of
(b) When will the water first strike the side- the radar zone. Sketch an accurate pic-
walk? ture of the radar zone by determining
where the line connecting Kingston and
(c) When will the water from the sprinkler Edmonds would cross the radar zone.
fall completely North of the sidewalk?
(c) When does the ferry enter the radar
(d) Find the total amount of time water from zone?
the sprinkler falls on the sidewalk.
(d) Where and when does the ferry exit the
(e) Sketch a picture of the situation after 33 radar zone?
minutes. Draw an accurate picture of
the watered portion of the sidewalk.
(e) How long does the ferry spend inside the
radar zone?
(f) Find the area of GRASS watered after
one hour.
Problem 3.8. Nora spends part of her sum-

Problem 3.7. Erik’s disabled sailboat is float- mer driving a combine during the wheat har-
ing stationary 3 miles East and 2 miles North vest. Assume she starts at the indicated posi-
of Kingston. A ferry leaves Kingston heading tion heading east at 10 ft/sec toward a circular
toward Edmonds at 12 mph. Edmonds is 6 wheat field of radius 200 ft. The combine cuts
miles due east of Kingston. After 20 minutes a swath 20 feet wide and begins when the cor-
the ferry turns heading due South. Ballard is ner of the machine labeled “a” is 60 feet north
8 miles South and 1 mile West of Edmonds. and 60 feet west of the western-most edge of
Impose coordinates with Ballard as the origin. the field.
3.5. EXERCISES 33
N
Problem 3.9. (a) Solve for x:
W E
x2 − 2x + 1
=x−2
S x+5
combine swath cut 20 ft
a (b) Solve for x:

center
wheat field x−3

=1
x+2
(c) If x = −2, find ALL solutions of the equa-

(a) When does Nora’s rig first start cutting tion
the wheat?
(x + 1)2 + (y − 1)2 = 10
(b) When does Nora’s rig first start cutting a
swath 20 feet wide?
(c) Find the total amount of time wheat is (d) If y = 3, find ALL solutions of the equa-
being cut during this pass across the tion
field.
(d) Estimate the area of the swath cut dur- 5(x + 1)2 + 2(y − 1)2 = 10
ing this pass across the field.
Chapter 4
Linear Modeling
Sometimes, we will begin a section by looking at a specific problem which

highlights the topic to be studied; this section offers the first such vista.
View these problems as illustrations of precalculus in action, rather than
confusing examples. Don’t panic, the essential algebraic skills will be
reviewed once the motivation is in place.
4.1 The Earning Power Problem

The government likes to gather all kinds of data. For example, Table 4.1
contains some data on the average annual income for full-time work-
ers; these data were taken from the 1990 Statistical Abstract of the U.S.
Given this information, a natural question would be: How can we predict
the future earning power of women and men? One way to answer this
(a) Women. (b) Men.
YEAR WOMEN (dollars) YEAR MEN (dollars)
1970 $5,616 1970 $9,521
1987 $18,531 1987 $28,313
Table 4.1: Earning power data.
question would be to use the data in the table to construct two different
mathematical models that predict the future (or past) earning power for
women or men. In order to do that, we would need to make some kind of
initial assumption about the type of mathematical model expected. Let’s
begin by drawing two identical xy-coordinate systems, where the x-axis
has units of “year” and the y-axis has units of “dollars;” see Figure 4.1. In
each coordinate system, the data in our table gives us two points to plot:
In the case of women, the data table gives us the points P = (1970, 5,616)
and Q = (1987, 18,531). Likewise, for the men, the data table gives us the
points R = (1970, 9,521) and S = (1987, 28,313).
To study the future earning power of men and women, we are going
to make an assumption: For women, if the earning power in year x is $y,
35
36 CHAPTER 4. LINEAR MODELING
PSfrag
y-axis (dollars) y-axis (dollars)
30000 30000
S = (1987,28313)
25000 25000
20000 Q = (1987,18531) 20000
15000 15000
10000 10000
R = (1970,5616)
5000 P = (1970,5616) 5000
x-axis (year) x-axis (year)
aca
1970
1975
1980
1985
1990
1995
1970
1975
1980
1985
1990
1995
(a) Data points for women. (b) Data points for men.
Figure 4.1: Visualizing the data.
then the point (x, y) lies on the line connecting P and Q. Likewise, for
men, if the earning power in year x is $y, then the point (x, y) lies on the
line connecting R and S.
In the real world, the validity of this kind of assumption would involve
a lot of statistical analysis. This kind of assumption leads us to what
is called a linear model, since we are demanding that the data points
predicted by the model (i.e., the points (x, y)) lie on a straight line in a
coordinate system. Now that we have made this assumption, our job is
to find a way to mathematically describe when a point (x, y) lies on one
of the two lines pictured in Figure 4.2.
Our goal in the next subsection is to review the mathematics nec-

essary to show that the lines in Figure 4.2 are the so-called graphs of
Equations 4.1 and 4.2.

30000 30000
S = (1987,28313)
25000 25000
20000 Q = (1987,18531) 20000

y (x,y) on the line:
15000 15000 This means men
y (x,y) on the line: This earn y dollars in
means women earn y year x.
10000 10000
dollars in year x. R = (1970,9521)
5000 P = (1970,5616) 5000
x-axis (year) x-axis (year)
x x
1970
1975
1980
1985
1990
1995
1970
1975
1980
1985
1990
1995
(a) Linear model for women. (b) Linear model for men.
Figure 4.2: Linear models of earning power.
4.2. RELATING LINES AND EQUATIONS 37

18,531 − 5,616
ywomen = (x − 1970) + 5,616 (4.1)
1987 − 1970
12,915
= (x − 1970) + 5,616
17
28,313 − 9,521
ymen = (x − 1970) + 9,521 (4.2)
1987 − 1970
18,792
= (x − 1970) + 9,521
17
4.2 Relating Lines and Equations

A systematic approach to studying equations and their graphs would
begin with the simple cases, gradually working toward the more com-
plicated. Thinking visually, the simplest curves in the plane would be
straight lines. As we discussed in Chapter 3, a point on the vertical line
in Figure 4.3(a) will always have the same x-coordinate; we refer to this
line as the graph of the equation x = h. Likewise, a point on the horizon-
tal line in Figure 4.3(b) will always have the same y-coordinate; we refer
to this line as the graph of the equation y = k. Figure 4.3(c) is different,
in the sense that neither the x nor the y coordinate is constant; i.e., as
you move a point along the line, both coordinates of the point are chang-
ing. It is reasonable to guess that this line is the graph of some equation
involving both x and y. The question is: What is the equation?
y-axis y-axis y-axis

graph of x = h (x,k) is a typical
point on this line (x,y) is a typical
point on this line
(h,y) is a typical
point on this line. graph of y = k graph of some
location k on y-axis equation involving
x-axis x and y
location h on x-axis x-axis x-axis
(a) Vertical line. (b) Horizontal line. (c) Sloped line.
Figure 4.3: Lines in a plane.

Here is the key geometric fact needed to model lines by mathematical

equations:
Important Fact 4.2.1. Two different points completely determine a straight

line.
This fact tells us that if you are given two different points on a line, you
can reconstruct the line in a coordinate system by simply lining a ruler
up with the two points. In our discussion, we will need to pay special
attention to the difference between vertical and non-vertical lines.
4.3 Non-vertical Lines

Assume in this section that ℓ is a non-vertical line in the plane; for ex-
ample, the line in Figure 4.3(c). If we are given two points P = (x1, y1)
and Q = (x2, y2) on a line ℓ, then Equation (2.2) on page 18 defined two
quantities we can calculate:
∆x = change in x going from P to Q = x2 − x1.
∆y = change in y going from P to Q = y2 − y1.

We define the slope of the line ℓ to be the ratio of ∆y by ∆x, which is
usually denoted by m:
∆y
def
m = (4.3)
∆x
y2 − y1
=
x2 − x1
change in y
slope of ℓ =
change in x
Notice, we are using the fact that the line is non-vertical to know that
this ratio is always defined; i.e., we will never have ∆x = 0 (which would
lead to illegal division by zero). There is some additional terminology that
goes along with the definition of the slope. The term ∆y is sometimes
called the rise of ℓ and ∆x is called the run of ℓ. For this reason, people
often refer to the slope of a line ℓ as “the rise over the run,” meaning
def rise of ℓ ∆y
slope of ℓ = = = m.
run of ℓ ∆x
In addition, notice that the calculation of ∆y involves taking the differ-
ence of two numbers; likewise, the calculation of ∆x involves taking the
difference of two numbers. For this reason, the slope of a line ℓ is some-
times called a difference quotient.
4.3. NON-VERTICAL LINES 39
For example, suppose P = (1, 1) and Q = (4, 5) lie on a

line ℓ. In this case, the rise = ∆y = 4 and the run = ∆x = 3, Q = (4,5)
5
so m = 34 is the slope of ℓ.
When computing ∆x, pay special attention that it is
P = (1,1)
the x-coordinate of the destination point Q minus the x- ∆y = 4
coordinate of the starting point P; likewise, when com-

1
puting ∆y, it is the y-coordinate of the destination point ∆x = 3
Q minus the y-coordinate of the starting point P. We can 1 4
reverse this order in both calculations and get the same
Figure 4.4: Computing the
slope:
slope of a line.
∆y y2 − y1 −(y2 − y1) y1 − y2 −∆y

m= = = = = .
∆x x2 − x1 −(x2 − x1) x1 − x2 −∆x
We CANNOT reverse the order in just one of the calculations and get !!!
the same slope:
CAUTION
y2 − y1 y2 − y1 y2 − y1 y1 − y2 !!!
m= 6= , and m= 6= .
x2 − x1 x1 − x2 x2 − x1 x2 − x1
It is very important to notice that the calculation of the ` ∗ ∗´

x 2 ,y 2 = Q ∗
slope of a line does not depend on the choice of the two
points P and Q. This is a real windfall, since we are then
always at liberty to pick our favorite two points on the (x 2 ,y 2 ) = Q
y2 − y1 y∗ ∗
2 − y1
line to determine the slope. The reason for this freedom (x 1 ,y 1 ) = P
x2 − x1
of choice is pretty easy to see by looking at a picture.
If we were to choose two other points P∗ = (x∗1, y∗1) and
x∗ − x∗
Q∗ = (x∗2, y∗2) on ℓ, then we would get two similar right ` 2 ∗´ 1
P ∗ = x∗
1 ,y 1
triangles: See Figure 4.5.
Basic properties of similar triangles tell us ratios of Figure 4.5: Using similar tri-
lengths of common sides are equal, so that angles.
y2 − y1 y∗2 − y∗1
m= = ∗ ;
x2 − x1 x2 − x∗1
but this just says the calculation of the slope is the same for any pair of
distinct points on ℓ. For example, lets redo the slope calculation when
P∗ = P = (x1, y1) and Q∗ = (x, y) represents an arbitrary point on the line.
Then the two ratios of lengths of common sides give us the equation
y − y1
m = ,
x − x1
y − y1 = m(x − x1).
This can be rewritten as
y = m(x − x1) + y1 (4.4)

or
y2 − y1
y = (x − x1) + y1. (4.5)
x2 − x1
Equation 4.4 is usually called the point slope formula for the line ℓ (since
the data required to write the equation amounts to a point (x1,y1) on
the line and the slope m), whereas Equation 4.5 is called the two point
formula for the line ℓ (since the data required amounts to the coordinates
of the points P and Q). In any event, we now see that
Important Fact 4.3.1. (x,y) lies on ℓ if and only if (x,y) is a solution to

y = m(x − x1) + y1.
We can plot the collection of ALL solutions to the equation in Fact 4.3.1,
which we refer to as the graph of the equation. As a subset of the xy-
coordinate system, the line
Important Fact 4.3.2. ℓ = {(x, m(x − x1) + y1) |x is any real number}.
4x − 1
graph of y = 3
Example 4.3.3. Consider the line ℓ, in Figure 4.6, through
3
(6, 23
3
) the two points P = (1,1) and Q = (4,5). Then the slope
(4,5) of ℓ is m = 4/3 and ℓ consists of all pairs of points (x,y)
(1,1)
such that the coordinates x and y satisfy the equation y =
4
5)
(−1, − 3
1)
(0, − 3 3
(x − 1) + 1. Letting x = 0, 1, 6 and −1, we conclude that
the following four points lie on the line ℓ: (0, 34 (0 − 1) + 1) =
(0, −1
3
), (1, 43 (1 − 1) + 1) = (1,1), (6, 34 (6 − 1) + 1) = (6, 23
3
) and
4 −5
Figure 4.6: Verifying points (−1, 3 (−1 − 1) + 1) = (−1, 3 ). By the same reasoning, the
on a line. point (0,0) does not lie on the line ℓ. As a set of points in the
plane, we have

4
ℓ= x, (x − 1) + 1 |x is any real number
3
Returning to the general situation, we can obtain a third general equa-

tion for a non-vertical line. To emphasize what is going on here, plug the
specific value x = 0 into Equation 4.4 and obtain the point R = (0,b) on
the line, where b = m(0 − x1) + y1 = −mx1 + y1. But, Equation (4.4) can be
written
y = m(x − x1) + y1
= mx − mx1 + y1
= mx + b.
4.4. GENERAL LINES 41
The point R is important; it is precisely the point where the line ℓ

crosses the y-axis, usually called the y-intercept. The slope intercept
equation of the line is the form
y = mx + b,
where the slope of the line is m and b is the y-intercept of the line.
Summary 4.3.4. Non-vertical Lines: Let ℓ be a non-vertical line in the
xy-plane. There are three ways to obtain an equation whose graph is ℓ,
depending on the data provided for ℓ:
1. If P = (x1,y1), Q = (x2,y2) are

two different
points on the line, then
y2 − y1
the two-point formula y = (x − x1) + y1 gives an equation
x2 − x1
whose graph is ℓ.
2. If P = (x1,y1) is a point on the line and m is the slope of ℓ, then
the point-slope formula y = m(x − x1) + y1 gives an equation whose
graph is ℓ.
3. If the line ℓ intersects the y-axis at the point (0,b) and m is the slope
of the line ℓ, then the slope-intercept formula y = mx + b gives an
equation whose graph is ℓ.
4.4 General Lines

Summarizing, any line in the plane is the graph of an equation involving
x and y and the equation always has the form
Ax + By + C = 0,
for some constants A, B, C. Equations like this are called linear equa-
tions. In general, non-vertical lines will be of the most interest to us,
since these are the lines that can be viewed as the graphs of functions ;
we will discuss this in Chapter 5.
4.5 Lines and Rate of Change

If we draw a non-vertical line in the xy coordinate system, then its slope
will be the rate of change of y with respect to x:
∆y
slope =
∆x
change in y
=
change in x
def
= rate of change of y with respect to x.
We should emphasize that this rate of change is a constant; in other

words, this rate is the same no matter where we compute the slope on
the line. The point-slope formula for a line can now be interpreted as
follows:
A line is determined by a point on the line and the rate of change of y
with respect to x.
An interesting thing to notice is how the units for x and y figure into
the rate of change calculation. For example, suppose that we have the
equation y = 10,000 x+200,000, which relates the value y of a house (in dol-
lars) to the number of years x you own it. For example, after 5 years, x = 5
and the value of the house would be y = 10,000 (5) + 200,000 = $250,000.
In this case, the equation y = 10,000 x + 200,000 is linear and already writ-
ten in slope-intercept form, so the slope can be read as m = 10,000. If
∆y
we carry along the units in the calculation of , then the numerator
∆x
involves “dollar” units and the denominator “years” units. That means
that carrying along units, the slope is actually m = 10,000 dollars/year.
In other words, the value of the house is changing at a rate of 10,000
dollars/year.
At the other extreme, if the units for both x and y are the same, then
the units cancel out in the rate of change calculation; in other words,
the slope is a unit-less quantity, simply a number. This sort of thing will
come up in the mathematics you see in chemistry and physics.
One important type of rate encountered is the speed
b
s
speed m
of a moving object. Suppose an object moves along a
straight line at a constant speed m: See Figure 4.7.
reference
point location of the ob-
If we specify a reference point, we can let b be the start-
initial location ject at time t
of the object ing location of the moving object, which is usually called
the initial location of the object. We can write down an
Figure 4.7: Motion along a equation relating the initial location b, the time t, the con-
line.
stant speed m and the location s at time t:
s = (location of object at time t)

s = (initial location of object) + (distance object travels in time t)
s = b + mt.
where t is in the same time units used to define the rate m. Notice, both
b and m would be constants given to us, so this is a linear equation
involving the variables s and t. We can graph the equation in the ts-
coordinate system: See Figure 4.8.
It is important to distinguish between this picture (the graph of s =
mt+b) and the path of the object in Figure 4.7. The graph of the equation
should be thought of as a visual aid attached to the equation s = mt + b.
The general idea is that using this visual aid can help answer various
questions involving the equation, which in turn will tell us things about
the motion of the object in Figure 4.7.
4.5. LINES AND RATE OF CHANGE 43
s-axis
Two other comments related to this discussion are im-
graph of s = mt + b
portant. First, concerning notation, the speed m is often
∆s
symbolized by v to denote constant velocity and b is writ- ∆t
∆s
ten as s◦ (the subscript “0” meaning “time zero”). With rate of change =
∆t
b = s-intercept
these changes, the equation becomes s = s◦ + vt, which is t-axis
the form in which it would be written in a typical physics
text. As a second note, if you return to Figure 4.8, you will Figure 4.8: The graph of
s = mt + b.
notice we only drew in the positive t axis. This was be-
cause t represented time, which is always a non-negative
quantity.
Example 4.5.1. Linda, Asia and Mookie are all playing frisbee. Mookie is
10 meters in front of Linda and always runs 5 m/sec. Asia is 34 meters
in front of Linda and always runs 4 m/sec. Linda yells “go!” and both
Mookie and Asia start running directly away from Linda to catch a tossed
frisbee. Find linear equations for the distances between Linda, Mookie and
Asia after t seconds.
Linda Mookie Asia
Solution. Let sM be the distance between Linda and Mookie and sA the
distance between Linda and Asia, after t seconds. An application of the
above formula tells us
sM = (initial distance between Linda and Mookie) + · · ·
· · · + (distance Mookie runs in t seconds)
sM = 10 + 5t.
Likewise,
sA = (initial distance between Linda and Asia) + · · ·
· · · + (distance Asia runs in t seconds)
sA = 34 + 4t.
If sMA is the distance between Mookie and Asia after t seconds, we com-
pute
sMA = sA − sM = (34 + 4t) − (10 + 5t) = (24 − t) meters.
In all cases, the distances we computed are given by linear equations of

the form s = b + mt, for appropriate b and rate m.
4.6 Back to the Earning Power Problem

y-axis (dollars)
We now return to the motivating problem at the start of
40000 this section. Recall the plot in Figure 4.1(b). We can
35000
U model the men’s earning power using the first and last
30000
T data points, using the ideas we have discussed about lin-
S ear equations. To do this, we should specify a “beginning
25000
point” and an “ending point” (recall Figure 2.9) and cal-
20000
y (x,y) on the line: culate the slope:
15000 means men earn y
dollars in year x
10000
R
Rbegin = (1970,9,521)
5000
x-axis (year) and
x
Send = (1987,28,313).
1980
1985
1995
1970
1975
1990
Figure 4.9: Linear model of

Men’s Earning Power.
We find that
∆y ∆x m
∆y
28,313 − 9,521 1987 − 1970
∆x
18,792 17 18,792
17
If we apply the “point-slope formula” for the equation of a line, we

arrive at the equation:
18,792
y= (x − 1970) + 9,521. (4.6)
17
The graph of this line will pass through the two points R and S in Fig-
ure 4.2. We sketch the graph in Figure 4.9, indicating two new points T
and U.
We can use the model in (4.6) to make predictions of two different
sorts: (i) predict earnings at some date, or (ii) predict when a desired
value for earnings will occur. For example, let’s graphically discuss the
earnings in 1995:
• Draw a vertical line x = 1995 up to the graph and label the intersec-
tion point U.
• Draw a horizontal line ℓ through U. The line ℓ crosses the y axis at
the point
18,792
(1995 − 1970) + 9521 = 37,156.
17
4.7. WHAT’S NEEDED TO BUILD A LINEAR MODEL? 45
• The coordinates of the point U = (1995, 37,156).
Conclude that $37,156 is the Men’s Earning Power in 1995. For another
example, suppose we wanted to know when men’s earning power will
equal $33,000? This means we seek a data point T on the men’s earning
curve whose y-coordinate is 33,000. By (4.6), T has the form

18,792
T = x, (x − 1970) + 9,521 .
17
We want this to be a data point of the form (x, 33,000). Setting these two
points equal and equating the second coordinates leads to an algebra
problem:
18,792
(x − 1970) + 9,521 = 33,000
17
x = 1991.24.
This means men’s earning power will be $33,000 at the end of the first
quarter of 1991. Graphically, we interpret this reasoning as follows:
• Draw a horizontal line y = 33,000 and label the intersection point T

on the model.
• Draw a vertical line ℓ through T . The line ℓ crosses the x axis at the
point 1991.24.
• The coordinates of the point T = (1991.24,33,000).
In the exercises, you will be asked to show that the women’s earning
power model is given by the equation
12,915
y= (x − 1970) + 5,616.
17
Using the two linear models for the earning power of men and women,
are women gaining on men? You will also be asked to think about this
question in the exercises.
4.7 What’s Needed to Build a Linear Model?

As we progress through this text, a number of different “types” of mathe-
matical models will be discussed. We will want to think about the infor-
mation needed to construct that particular kind of mathematical model.
Why would we care? For example, in a laboratory context, if we knew a
situation being studied was given by a linear model, this would effect the
amount of data collected. In the case of linear models, we can now make
this useful statement:
Important Fact 4.7.1. A linear model is completely determined by:
1. One data point and a slope (a rate of change), or
2. Two data points, or
3. An intercept and a slope (a rate of change).
4.8 Linear Application Problems

Example 4.8.1. The yearly resident tuition at the University of Washing-
ton was $1827 in 1989 and $2907 in 1995. Assume that the tuition growth at
the UW follows a linear model. What will be the tuition in the year 2000?
When will yearly tuition at the University of Washington be $10,000?
y-axis
10000 (dollars)
8000
Solution. If we consider a coordinate system where the
6000
4000
x-axis represents the year and the y-axis represents
2000
P Q x-axis dollars, we are given two data points: P = (1989, 1,827)
(year)
and Q = (1995, 2,907). Using the two-point formula for the
2020
2000
2010
1990
2030
2040
equation of line through P and Q, we obtain the equation

Figure 4.10: Linear tuition
model. y = 180(x − 1989) + 1,827.
The graph of this equation gives a line through the given

points as pictured in Figure 4.10.
If we let x = 2000, we get y = $3,807, which tells us the tuition in the
year 2000. On the other hand, if we set the equation equal to $10,000, we
can solve for x:
10,000 = 180(x − 1989) + 1,827
8,173 = 180(x − 1989)
2,034.4 = x.
Conclude the tuition is $10,000 in the year 2035.
4.9 Perpendicular and Parallel Lines

Here is a useful fact to keep in mind.
Important Facts 4.9.1. Two non-vertical lines in the plane are parallel
precisely when they both have the same slope. Two non-vertical lines are
perpendicular precisely when their slopes are negative reciprocals of one
another.
4.10. INTERSECTING CURVES II 47
Example 4.9.2. Let ℓ be a line in the plane passing through the points
(1, 1) and (6, −1). Find a linear equation whose graph is a line parallel to
ℓ passing through 5 on the y-axis. Find a linear equation whose graph is
perpendicular to ℓ and passes through (4, 6).
Solution. Letting P = (1, 1) and Q = (6, −1), apply the “two point formula”:
−2
y = (x − 1) + 1
5
2 7
= − x+ .
5 5
The graph of this equation will be ℓ. This equation is in slope intercept
form and we can read off that the slope is m = −2 5
. The desired line a is
2
parallel to ℓ; it must have slope m = − 5 and y-intercept 5. Plugging into
the “slope intercept form”:
−2
y= x + 5.
5
−1
The desired line b is a line perpendicular to ℓ (so its slope is m ′ = −2 =
5
5
2
)and passes through the point (4, 6), so we can use the “point slope
formula”:
5
y = (x − 4) + 6.
2
4.10 Intersecting Curves II

We have already encountered problems that require us to investigate the
intersection of two curves in the plane. Ultimately, this reduces to solving
a system of two (or more) equations in the variables x and y. A useful
tool when working with equations involving squared terms (i.e., x2 or y2),
is the quadratic formula .
Important Fact 4.10.1. Quadratic Formula: Consider the equation

az2 + bz + c = 0, where a,b,c are constants. The solutions for this equa-
tion are given by the formula
√
−b ± b2 − 4ac
z= .
2a
The solutions are real numbers if and only if b2 − 4ac ≥ 0.
The next example illustrates a typical application of the quadratic for-

mula. In addition, we describe a very useful technique for finding the
shortest distance between a “line” and a “point.”
North
flight path irrigated field Example 4.10.2. A crop dusting airplane flying a constant
speed of 120 mph is spotted 2 miles South and 1.5 miles
West East
Q East of the center of a circular irrigated field. The irrigated
field has a radius of 1 mile. Impose a coordinate system
2 miles
P crop
as pictured, with the center of the field the origin (0,0). The
1.5 miles
duster flight path of the duster is a straight line passing over the
South
labeled points P and Q. Assume that the point Q where the
Figure 4.11: The flight path plane exits the airspace above the field is the Western-most
of a crop duster. location of the field. Answer these questions:
1. Find a linear equation whose graph is the line along

which the crop duster travels.
2. Find the location P where the crop duster enters airspace

above the irrigated field.
3. How much time does the duster spend flying over the
irrigated field?
4. Find the shortest distance from the flight path to the

center of the irrigated field.
Solution.
1. Take Q = (−1, 0) and S = (1.5, −2) = duster spotting point. Construct

a line through Q and S. The slope is −0.8 = m and the line equation
becomes:
y = −0.8x − 0.8. (4.7)
2. The equation of the boundary of the irrigated region is x2 + y2 = 1.

We need to solve this equation AND the line equation y = −0.8x − 0.8
simultaneously. Plugging the line equation into the unit circle equa-
tion gives:
x2 + (−0.8x − 0.8)2 = 1
x2 + 0.64x2 + 1.28x + 0.64 = 1
1.64x2 + 1.28x − 0.36 = 0
Apply the quadratic formula and find x = −1, 0.2195. Conclude

that the x coordinate of P is 0.2195. To find the y coordinate, plug
into the line equation and get y = −0.9756. Conclude that P =
(0.2195, −0.9756)
4.11. UNIFORM LINEAR MOTION 49
3. Find the distance from P to Q by using the distance formula:

p
d = (−1 − 0.2195)2 + (0 − (−0.9756))2
= 1.562 miles
1.562miles
Now, 120mph
= 0.01302 hours = 47 seconds.
4. The idea is to construct a line perpendicular to the flight path pass-

ing through the origin of the coordinate system. This line will have
1
slope m = − −0.8 = 1.25. So this perpendicular line has equation
y = 1.25x. Intersecting this line with the flight path gives us the
point closest to the center of the field. The x-coordinate of this point
is found by setting the two line equations equal and solving:
−0.8x − 0.8 = 1.25x

x = −0.3902
This means that the closest point on the flight path is (−0.39, −0.49).
Apply the distance formula and the shortest distance to the flight
path is
p
d = (−0.39)2 + (−0.49)2
= 0.6263.
4.11 Uniform Linear Motion

When an object moves along a line in the xy-plane at a constant speed,
we say that the object exhibits uniform linear motion. Its location can
be described using a pair of linear equations involving a variable which
represents time. That is, we can find constants a, b, c, and d such that,
at any time t, the object’s location is given by (x,y), where
x = a + bt and y = c + dt.
Such equations are called parametric equations of motion. The motion

is defined in terms of the parameter t.
Since there are four constants to be determined, one needs four pieces
of information to determine these equations. Knowing the object’s loca-
tion at two points in time is sufficient.
Example 4.11.1. Bob is running in the xy-plane. He runs in a straight line
from the point (2,3) to the point (5, − 4), taking 6 seconds to do so. Find his
equations of motion.
Solution. We begin by setting a reference for our time parameter. Let’s let
t = 0 represent the instant when Bob is at the point (2,3). In this way, t
will represent the time since Bob left the point (2,3). When t = 6, we know
he will be at the point (5, − 4). This is enough information to determine
his equations of motion.
We seek constants a, b, c, and d so that at time t, Bob’s location is
given by
When t = 0, we know Bob’s location is (2,3). That is, x = 2 and y = 3.

Thus, with t = 0, we have the two equations
x = 2 = a + b(0) = a and y = 3 = c + d(0) = c.
and so a = 2 and c = 3. We’re half-way done.

When t = 6, we know Bob’s location is (5, − 4). Thus, with t = 6, we
have the two equations
x = 5 = a + b(6) = 2 + 6b and y = −4 = c + d(6) = 3 + 6d
which we can solve to find

1 7
b= and d = − .
2 6
So, we arrive at the equations of Bob’s motion
1 7
x = 2 + t and y = 3 − t.
2 6
Notice it is easy to check that these are correct. If we plug in t = 0, we
find x = 2, y = 3 as required. If we plug in t = 6, we find x = 5, y = −4, as
required. So we know we’ve done it right.
Now that we have these equations of motion, it is very easy to calculate
Bob’s location at any time. For instance, 30 seconds after leaving the
point (2,3), we can find that he is at the point (17, − 32) since
1 7
x = 2 + (30) = 17, y = 3 − (30) = −32.
2 6
Example 4.11.2. Olga is running in the xy-plane, and the coordinate are
given in meters (so, for example, the point (1,0) is one meter from the origin
(0,0)). She runs in a straight line, starting at the point (3,5) and running
along the line y = − 31 x + 6 at a speed of 7 meters per second, heading away
from the y-axis. What are her parametric equations of motion?
4.11. UNIFORM LINEAR MOTION 51
Solution. This example differs in some respects from the last example.
In particular, instead of knowing where the runner is at two points in
time, we only know one point, and have other information given to us
about the speed and path of the runner. One approach is to use this new
information to find where the runner is at some other point in time: this
will then give us exactly the same sort of information as we used in the
last example, and so we may solve it in an identical manner.
We know that Olga starts at the point (3,5). Letting t = 0 represent the
time when she starts, we then know that when t = 0, x = 3 and y = 5.
To get another point (and time), we can use the fact that we know what
line she travels along, and which direction she runs. We may consider
any point on the line in the correct direction: any will do. For instance,
the point (6,4) is on the line. We then need to find when Olga reaches
this point. To do this, we find the distance from her starting point to the
point (6,4), and divide this by her speed. The time she takes to get to (6,4)
is thus
p
(6 − 3)2 + (4 − 5)2
= 0.45175395 seconds .
7
At this point, we are now in the same situation as in the last example.
We know two facts: when t = 0, x = 3 and y = 5, and when t = 0.45175395,
x = 6 and y = 4. As we saw in the last example, this is enough information
to find the parametric equations of motion.
We seek a, b, c, and d such that Olga’s location t seconds after she
starts is (x,y) where
When t = 0, x = 3, and y = 5, so
x = 3 = a + b(0) = a and y = 5 = c + d(0) = c
and so a = 3 and c = 5. Also, when t = 0.45175395, x = 6 and y = 4, so
x = 6 = a+b(0.45175395) = 3+b(0.45175395) and y = 4 = c+d(0.45175395) = 5+d(0.45175395).
Solving these equations for b and d, we find
3 −1
b= = 6.64078311 and d = = −2.21359436.
0.45175395 0.45175395
Thus, Olga’s equations of motion are
x = 3 + 6.64078311t, y = 5 − 2.21359436t.
4.12 Summary
• The equation of every non-vertical line can be expressed in the form
y = m(x − h) + k (the point-slope form)
and
y = mx + b (the slope-intercept form)
• A vertical line has an equation of the form x = c.
• The shortest distance between a point, P, and a line, l, can be found

by determining a line l2 which passes through P and is perpendicu-
lar to l. Then the point at which l and l2 intersect is the point on l
which is closest to l. The distance from this point to P is the shortest
distance between P and l.
• The location of an object moving at constant speed along a line can

be described using a pair of equations (parametric equations)
x = a + bt, y = c + dt.
4.13. EXERCISES 53
4.13 Exercises
Problem 4.1. This exercise emphasizes the Problem 4.4. Complete Table 4.2 on page 54.
“mechanical aspects” of working with linear In many cases there may be several possible
equations. Find the equation of a line: correct answers.
(a) Passing through the points (1, − 1) and
Problem 4.5. The (average) sale price for
(−2,4).
single family property in Seattle and Port
(b) Passing through the point (−1, − 2) with Townsend is tabulated below:
slope m = 40.
YEAR SEATTLE PORT TOWNSEND
(c) With y-intercept b = −2 and slope m = 1970 $38,000 $8400
−2. 1990 $175,000 $168,400
(d) Passing through the point (4,11) and (a) Find a linear model relating the year x
having slope m = 0. and the sales price y for a single family
(e) Perpendicular to the line in (a) and pass- property in Seattle.
ing through (1,1). (b) Find a linear model relating the year x
(f) Parallel to the line in (b) and having y- and the sales price y for a single family
intercept b = −14. property in Port Townsend.
(c) Sketch the graph of both modeling equa-
(g) Having the equation 3x + 4y = 7.
tions in a common coordinate system;
(h) Crossing the x-axis at x = 1 and having restrict your attention to x ≥ 1970.
slope m = 1. (d) What is the sales price in Seattle and
Port Townsend in 1983 and 1998?
Problem 4.2. Sketch an accurate picture of (e) When will the average sales price in
the line having equation y = 2 − 12 x. Let α be Seattle and Port Townsend be equal and
an unknown constant. what is this price?
(a) Find the point of intersection between (f) When will the average sales price in
the line you have graphed and the line Port Townsend be $15,000 less than the
y = 1 + αx; your answer will be a point in Seattle sales price? What are the two
the xy plane whose coordinates involve sales prices at this time?
the unknown α. (g) When will the Port Townsend sales price
(b) Find α so that the intersection point in be $15,000 more than the Seattle sales
(a) has x-coordinate 10. price? What are the two sales prices at
this time?
(c) Find α so that the intersection point in
(h) When will the Seattle sales price be dou-
(a) lies on the x-axis.
ble the Port Townsend sales price?
(i) Is the Port Townsend sales price ever
Problem 4.3. (a) What is the area of the tri- double the Seattle sales price?
angle determined by the lines y = − 21 x +
5, y = 6x and the y-axis?
Problem 4.6. Consider the equation: αx2 +
(b) If b > 0 and m < 0, then the line y = 2α2 x + 1 = 0. Find the values of x that make
mx + b cuts off a triangle from the first this equation true (your answer will involve α).
quadrant. Express the area of that tri- Find values of α that make this equation true
angle in terms of m and b. (your answer will involve x).
(c) The lines y = mx + 5, y = x and the y-
axis form a triangle in the first quadrant. Problem 4.7. The cup on the 9th hole of a golf
Suppose this triangle has an area of 10 course is located dead center in the middle of a
square units. Find m. circular green that is 70 feet in diameter. Your
ball is located as in the picture below:
Point on Point on
Equation Slope y-intercept
the line the line
y = 2x + 1
(3, −4) (−1, 7)
−2 1
1
2
(0, 1)
1,000
(3, 3) (3, −2)
(5, −9)
Table 4.2: Linear equations table for Problem 4.4.
9 Problem 4.8. Allyson and Adrian have decided

cup to connect their ankles with a bungee cord;
one end is tied to each person’s ankle. The
green cord is 30 feet long, but can stretch up to 90
ball path
feet. They both start from the same location.
Allyson moves 10 ft/sec and Adrian moves 8
ft/sec in the directions indicated. Adrian stops
moving at time t = 5.5 sec, but Allyson keeps
50 feet on moving 10 ft/sec in the indicated direction.
rough
ball 40 feet
(a) Sketch an accurate picture of the situa-

The ball follows a straight line path and exits
tion at time t = 7 seconds. Make sure to
the green at the right-most edge. Assume the
label the locations of Allyson and Adrian;
ball travels a constant rate of 10 ft/sec.
also, compute the length of the bungee
(a) Where does the ball enter the green? cord at t = 7 seconds.
(b) When does the ball enter the green?
(c) How long does the ball spend inside the
green?
(d) Where is the ball located when it is
closest to the cup and when does this
occur. (b) Where is Allyson when the bungee
reaches its maximum length?
4.13. EXERCISES 55
Find the portion of the y-axis that Dave can-

20 ft not see. (Hint: Let a be the x-coordinate of the
point where line of sight #1 is tangent to the
silo; compute the slope of the line using two
Building
points (the tangent point and (12,0)). On the
other hand, compute the slope of line of sight
#1 by noting it is perpendicular to a radial line
through the tangency point. Set these two cal-
culations of the slope equal and solve for a.)
Allyson Problem 4.10. While speaking on the phone

30 ft
to a friend in Oslo, Norway, you learned that
the current temperature there was −23◦ Cel-
sius (−23◦ C). After the phone conversation,
you wanted to convert this temperature to
start Fahrenheit degrees ◦ F, but you could not find
Adrian
a reference with the correct formulas. You
then remembered that the relationship be-
tween ◦ F and ◦ C is linear.
Problem 4.9. Dave is going to leave academia
and go into business building grain silos. A (a) Using this and the knowledge that
grain silo is a cylinder with a hemispherical 32◦ F = 0◦ C and 212◦ F = 100◦ C, find an
top, used to store grain for farm animals. Here equation that computes Celsius temper-
is a 3D view, a cross-section, and the top view: ature in terms of Fahrenheit tempera-
ture; i.e., an equation of the form C=
“an expression involving only the vari-
able F .”
(b) Likewise, find an equation that com-

silo putes Fahrenheit temperature in terms
h
of Celsius temperature; i.e. an equation
of the form F= “an expression involving
only the variable C .”
3D-view
(c) How cold was it in Oslo in ◦ F?
r
cross-section
y-axis Problem 4.11. Pam is taking a train from the

town of Rome to the town of Florence. Rome is
located 30 miles due West of the town of Paris.
line of sight #1 Florence is 25 miles East, and 45 miles North
blind spot of Rome.
(12,0) On her trip, how close does Pam get to Paris?
x-axis
Dave
Problem 4.12. Angela, Mary and Tiff are all
standing near the intersection of University
line of sight #2 and 42nd streets. Mary and Tiff do not move,
but Angela runs toward Tiff at 12 ft/sec along
TOP VIEW a straight line, as pictured. Assume the roads
are 50 feet wide and Tiff is 60 feet north of the
If Dave is standing next to a silo of cross- nearest corner. Where is Angela located when
sectional radius r = 8 feet at the indicated po- she is closest to Mary and when does she reach
sition, his vision will be partially obstructed. this spot?
Problem 4.14. Margot is walking in a straight

tiff line from a point 30 feet due east of a statue in
a park toward a point 24 feet due north of the
statue. She walks at a constant speed of 4 feet
mary per second.
42 nd St. (a) Write parametric equations for Margot’s

position t seconds after she starts walk-
angela ing.
(b) Write an expression for the distance

from Margot’s position to the statue at
time t.
University Way
(c) Find the times when Margot is 28 feet
from the statue.
Problem 4.13. The infamous crawling tractor
sprinkler is located as pictured below, 100 feet
South of a 10 ft. wide sidewalk; notice the hose
and sidewalk are not perpendicular. Once the Problem 4.15. Juliet and Mercutio are mov-
water is turned on, the sprinkler waters a cir- ing at constant speeds in the xy-plane. They
cular disc of radius 20 feet and moves North start moving at the same time. Juliet starts at
along the hose at the rate of 12 inch/second. the point (0, − 6) and heads in a straight line
toward the point (10,5), reaching it in 10 sec-
(a) Impose a coordinate system. Describe onds. Mercutio starts at (9, − 14) and moves in
the initial coordinates of the sprinkler a straight line. Mercutio passes through the
and find the equation of the line forming same point on the x axis as Juliet, but 2 sec-
the southern boundary of the sidewalk. onds after she does.
(b) After 33 minutes, sketch a picture of How long does it take Mercutio to reach the
the wet portion of the sidewalk; find the y-axis?
length of the wet portion of the Southern
edge of the sidewalk.
Problem 4.16. (a) Solve for x:
(c) Find the equation of the line forming
the northern boundary of the sidewalk. 1 1
− = 3.
(Hint: You can use the properties of right x x+1
triangles.)
p
N (b) Solve for t: 2 = (1 + t)2 + (1 − 2t)2 .
hose p
(c) Solve for t: √3 = (1 + t)2 + (1 − 2t)2 .
5
W E
p
(d) Solve for t: 0 = (1 + t)2 + (1 − 2t)2 .
100 ft S
20 ft Problem 4.17. (a) Solve for x:
x4 − 4x2 + 2 = 0
100 ft. sidewalk
(b) Solve for y:
circular watered zone √
y−2 y=4
Chapter 5
Functions and Graphs
Pictures are certainly important in the work of an architect, but it is

perhaps less evident that visual aids can be powerful tools for solving
mathematical problems. If we start with an equation and attach a pic-
ture, then the mathematics can come to life. This adds a new dimension
to both interpreting and solving problems. One of the real triumphs of
modern mathematics is a theory connecting pictures and equations via
the concept of a graph. This transition from “equation” to “picture” (called
graphing) and its usefulness (called graphical analysis) are the theme of
the next two sections. The importance of these ideas is HUGE and can-
not be overstated. Every moment spent studying these ideas will pay
back dividends in this course and in any future mathematics, science or
engineering courses.
5.1 Relating Data, Plots and Equations

Imagine you are standing high atop an oceanside cliff and
gull line of motion
spot a seagull hovering in the air-current. Assuming the
gull moves up and down along a vertical line of motion,
how can we best describe its location at time t seconds? 1
0
0
1
There are three different (but closely linked) ways to cliff level
describe the location of the gull:
• a table of data of the gull’s height above cliff level at ocean
various times t;
Figure 5.1: Seagull’s height.
• a plot of the data in a “time” (seconds) vs. “height”
(feet) coordinate system;
• an equation relating time t (seconds) and height s
(feet).
To make sure we really understand how to pass back and forth be-
tween these three descriptive modes, imagine we have tabulated (Fig-
ure 5.2) the height of the gull above cliff level at one-second time intervals
57
58 CHAPTER 5. FUNCTIONS AND GRAPHS
for a 10 second time period. Here, a “negative height” means the gull is
below cliff level. We can try to visualize the meaning of this data by plot-
ting these 11 data points (t, s) in a time (sec.) vs. height (ft.) coordinate
system.
Feet
50
Gull Height (feet above cliff level) 40
30
t (sec) s (ft) t (sec) s (ft) t (sec) s (ft)
20
0 20 4 -10 8 20
10 2 4 6 8 10
1 6.88 5 -8.12 9 36.88
2 -2.5 6 -2.5 10 57.5 −10 Seconds
3 -8.12 7 6.88
(a) Symbolic data. (b) Visual data.
Figure 5.2: Symbolic versus visual view of data.
We can improve the quality of this description by increasing the num-

ber of data points. For example, if we tabulate the height of the gull above
cliff level at 1/2 second or 1/4 second time intervals (over the same 10
second time period), we might get these two plots:
Feet
Feet
50
50 40
40 30
30
20 20
10 10 4 Seconds
4 Seconds
−10 2 6 8 10 −10 2 6 8 10
1 1
(a) 2 second intervals. (b) 4 second intervals.
Figure 5.3: Shorter time intervals mean more data points.
We have focused on how to go from data to a plot, but the reverse pro-
cess is just as easy: A point (t, s) in any of these three plots is interpreted
to mean that the gull is s feet above cliff level at time t seconds.
Furthermore, increasing the amount of data, we see how the plotted
points are “filling in” a portion of a parabola. Of course, it is way too
tedious to create longer and longer tables of data. What we really want
is a “formula” (think of it as a prescription) that tells us how to produce
a data point for the gull’s height at any given time t. If we had such a
formula, then we could completely dispense with the tables of data and
just use the formula to crank out data points. For example, look at this
equation involving the variables t and s:
15
s= (t − 4)2 − 10.
8
5.2. WHAT IS A FUNCTION? 59
If we plug in t = 0, 1, 2, 9, 10, then we get s = 20, 6.88, −2.5, 36.88, 57.5, re-
spectively; this was some of our initial tabulated data. This same equa-
tion produces ALL of the data points for the other two plots, using 1/2
second and 1/4 second time intervals. (Granted, we have swept under the
rug the issue of “...where the heck the equation comes from...” ; that is
a consequence of mathematically modeling the motion of this gull. Right
now, we are focusing on how the equation relates to the data and the
plot, assuming the equation is in front of us to start with.) In addition, it
is very important to notice that having this equation produces an infinite
number of data points for our gull’s location, since we can plug in any
t value between 0 and 10 and get out a corresponding height s. In other
words, the equation is A LOT more powerful than a finite (usually called
discrete ) collection of tabulated data.
5.2 What is a Function?

Our lives are chock full of examples where two changing quantities are
related to one another:
• The cost of postage is related to the weight of the item.
• The value of an investment will depend upon the time elapsed.

• The population of cells in a growth medium will be related to the
amount of time elapsed.
• The speed of a chemical reaction will be related to the temperature
of the reaction vessel.
In all such cases, it would be beneficial to have a “procedure” whereby
we can assign a unique output value to any acceptable input value. For
example, given the time elapsed (an input value), we would like to predict
a unique future value of an investment (the output value). Informally,
this leads to the broadest (and hence most applicable) definition of what
we will call a function :
Definition 5.2.1. A function is a procedure for assigning a unique output
to any allowable input.
The key word here is “procedure.” Our discussion of the
y-axis
hovering seagull in 5.1 highlights three ways to produce P = (x,y)
y
such a “procedure” using data, plots of curves and equa-
x
tions. x-axis
• A table of data, by its very nature, will relate two Figure 5.4: Graph of a
columns of data: The output and input values are procedure.
listed as column entries of the table and reading across each row
is the “procedure” which relates an input with a unique output.
• Given a curve in Figure 5.4, consider the “procedure”

which associates to each x on the horizontal axis the y
coordinate of the pictured point P on the curve.
• Given an equation relating two quantities x and y, plug-

ging in a particular x value and going through the “pro-
cedure” of algebra often produces a unique output value
y.

(equation viewpoint)
Now we focus on giving a precise definition of a function, in the situa-
tion when the “procedure” relating two quantities is actually given by an
equation. Keep in mind, this is only one of three possible ways to de-
scribe a function; we could alternatively use tables of data or the plot of
a curve. We focus on the equation viewpoint first, since it is no doubt the
most familiar.
If we think of x and y as related physical quantities (e.g. time and dis-
tance ), then it is sometimes possible (and often desirable) to express one
of the variables in terms of the other. For example, by simple arithmetic,
the equations
1 p
3x + 2y = 4 x2 − x = y − 4 y x2 + 1 = 1,
2
can be rewritten as equivalent equations
1 1
y = (4 − 3x) 2x2 − 2x + 8 = y y= √ .
2 x2 +1
This leads to THE MOST IMPORTANT MATH DEFINITION IN THE WORLD:
Definition 5.2.2. A function is a package, consisting of three parts:
• An equation of the form
y = “a mathematical expression only involving the variable x,”
which we usually indicate via the shorthand notation y = f(x). This

equation has the very special property that each time we plug in
an x value, it produces exactly one (a unique) y value. We call the
mathematical expression f(x) ”the rule”.
• A set D of x-values we are allowed to plug into f(x), called the

”domain” of the function.
• The set R of output values f(x), where x varies over the domain,
called the ”range” of the function.
Any time we have a function y = f(x), we refer to x as the independent

variable (the “input data”) and y as the dependent variable (the “output
data”). This terminology emphasizes the fact that we have freedom in
the values of x we plug in, but once we specify an x value, the y value is
uniquely determined by the rule f(x).
Examples 5.2.3. (i) The equation y = −2x + 3 is in the y-axis

form y = f(x), where the rule is f(x) = −2x+3. Once we Graph of y = b
specify a domain of x values, we have a function. For
example, we could let the domain be all real numbers. b x-axis
(ii) Take the same rule f(x) = −2x + 3 from (i) and let the Figure 5.5: Constant func-
domain be all non-negative real numbers. This de- tion.
scribes a function. However, the functions f(x) = −2x+
3 on the domain of all non-negative real numbers and f(x) = −2x + 3
on the domain of all real numbers (from (i)) are different, even though
they share the same rule; this is because their domains differ! This
example illustrates the idea of what is called a restricted domain. In
other words, we started with the function in (i) on the domain of all
real numbers, then we “restricted” to the subset of non-negative real
numbers.
(iii) The equation y = b, where b is a constant, defines a

function on the domain of all real numbers, where the
rule is f(x) = b; we call these the constant functions.
Recall, in Chapter 3, we observed that the solutions
of the equation y = b, plotted in the xy coordinate sys-
tem, will give a horizontal line. For example, if b = 0,
you get the horizontal axis.
(iv) Consider the equation y = x1 , then the rule f(x) = x1

defines a function, as long as we do not plug in x = 0.
For example, take the domain to be the non-zero real
numbers.
√
(v) Consider the equation y = 1 − x2. Before we start
plugging in x values, we want to know the expression
under the radical symbol (square root symbol) is non-
negative; this insures the square root is a real num-
ber. This amounts to solving an inequality equation:
0 ≤ 1 − x2; i.e., −1
√ ≤ x ≤ 1. These remarks show that
the rule f(x) = 1 − x2 defines a function, where the
domain of x values is −1 ≤ x ≤ 1.
Typically, the domain of a function y = f(x) will either be the entire

number line, an interval on the number line, or a finite union of such
intervals. We summarize the notation used to represent intervals in Ta-
ble 5.1.
Common Intervals on the Number Line

Symbolic
Description Picture
Notation
All numbers x between a

and b, x possibly equal to a≤x≤b
either a or b a b

a<x<b
and b, x 6= a and x 6= b a b

and b, x 6= b and x possi- a≤x<b
bly equal to a a b

and b, x 6= a and x possi- a<x≤b
bly equal to b a b
Table 5.1: Interval Notations
We can interpret a function as a “prescription” that takes a given

x value (in the domain) and produces a single unique y value (in the
range). We need to be really careful and not fall into the trap of thinking
that every equation in the world is a function. For example, if we look at
this equation
x + y2 = 1
and plug in x = 0, the equation becomes
y2 = 1.
This equation has two solutions, y = ±1, so the conclusion is that plug-
ging in x = 0 does NOT produce a single output value. This violates one
of the conditions of our function definition, so the equation x + y2 = 1 is
NOT a function in the independent variable x. Notice, if you were to try
and solve this equation for y in terms of x, you’d first write y2 = 1 − x
and then take a square root (to isolate y); but the square root introduces
TWO roots, which is just another way of reflecting the fact there can be
two y values attached to a single x value. Alternatively, you can solve the
equation for x in terms of y, getting x = 1 − y2; this shows the equation
does define a function x = g(y) in the independent variable y.

(conceptual viewpoint)
Conceptually, you can think of a function as a “process”: An allowable
input goes into a “black box” and out pops a unique new value denoted
by the symbol f(x). Compare this with the machine making “hula-hoops”
in Figure 5.6. While you are problem solving, you will find this to be a
useful viewpoint when a function is described in words.
11111111111
00000000000
00000000000
11111111111
tube
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111 hoop
in
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
00000000000
11111111111
out in out
00000000000
11111111111
x y
tubes f(tube) domain f(x)
(a) A hoop machine as a (b) A function as a “process”

“process” taking “tubes” to which takes x to y.
“hoops.”
Figure 5.6: Viewing a function as a “process.”
Examples 5.2.4. Here are four examples of relationships that are func-
tions:
(i) The total amount of water used by a household since midnight

on a particular day. Let y be the total number of gallons of water
used by a household between 12:00am and a particular time t; we
will use time units of “hours.” Given a time t, the household will
have used a specific (unique) amount of water, call it S(t). Then y =
S(t) defines a function in the independent variable t with dependent
variable y. The domain would be 0 ≤ t ≤ 24 and the largest possible
value of S(t) on this domain is S(24). This tells us that the range would
be the set of values 0 ≤ y ≤ S(24).
(ii) The height of the center of a basketball as you dribble, de-

pending on time. Let s be the height of the basketball center at time
t seconds after you start dribbling. Given a time t, if we freeze the ac-
tion, the center of the ball has a single unique height above the floor,
call it h(t). So, the height of the basketball center is given by a func-
tion s = h(t). The domain would be a given interval of time you are
dribbling the ball; for example, maybe 0 ≤ t ≤ 2 (the first 2 seconds).
In this case, the range would be all of the possible heights attained
by the center of the basketball during this 2 seconds.
(iii) The state sales tax due on a taxable item. Let T be the state
tax (in dollars) due on a taxable item that sells for z dollars. Given a
taxable item that costs z dollars, the state tax due is a single unique
amount, call it W(z). So, T = W(z) is a function, where the indepen-

dent variable is z. The domain could be taken to be 0 ≤ z ≤ 1,000,000,
which would cover all items costing up to one-million dollars. The
range of the function would be the set of all values W(z), as z ranges
over the domain.
(iv) The speed of a chemical reaction depending on the temper-

ature. Let v be the speed of a particular chemical reaction and T
the temperature in Celsius ◦ C. Given a particular temperature T , one
could experimentally measure the speed of the reaction; there will be
a unique speed, call it r(T ). So, v = r(T ) is a function, where the inde-
pendent variable is T . The domain could be taken to be 0 ≤ T ≤ 100,
which would cover the range of temperatures between the freezing
and boiling points of water. The range of the function would be the
set of all speeds r(T ), as T ranges over the domain.
5.3 The Graph of a Function

Let’s start with a concrete example; the function f(x) = −2x + 3 on the
domain of all real numbers. We discussed this Example 5.2.3. Plug in
the specific x values, where x = −1, 0, 1, 2 and tabulate the resulting
y values of the function:
x y point (x,y) y-axis
-1 5 (-1,5)
0 3 (0,3) Graph of y = 2x + 3.
1 1 (1,1)
2 -1 (2,-1) x-axis
.. .. ..
. . .
x −2x + 3 (x, − 2x + 3)
(a) Tabulated data. (b) Visual data.
Figure 5.7: Symbolic versus visual view of data.
This tells us that the points (0, 3), (1, 1), (2, −1), (−1, 5) are solutions
of the equation y = −2x + 3. For example, if y = −2x + 3, x = 0, y = 3,
then 3 = −2 · 0 + 3 (which is true), or if y = −2x + 3, x = 2, y = −1, then
−1 = −2 · 2 + 3 (which is true), etc. In general, if we plug in x we get out
−2x + 3, so the point (x, − 2x + 3) is a solution to the function equation
y = f(x). We can plot all of these solutions in the xy-coordinate system.
The set of points we obtain, as we vary over all x in the domain, is called
the set of solutions of the equation y = −2x + 3:
Solutions = { (x, −2x + 3) | x any real number}.
Notice that plotting these points produces a line of slope m = −2 with

y-intercept 3. In other words, the graph of the function f(x) = −2x + 3 is
5.4. THE VERTICAL LINE TEST 65
the same as the graph of the equation y = −2x + 3, as we discussed in

Chapter 4.
In general, by definition, we say that a point (x,y) is a solution to the
function equation y = f(x) if plugging x and y into the equation gives a
true statement.
How can we find ALL the solutions of the equation y = f(x)? In general,
the definition of a function is “rigged” so it is easy to describe all solutions
of the equation y = f(x): Each time we specify an x value (in the domain),
there is only one y value, namely f(x). This means the point P = (x, f(x))
is the ONLY solution to the equation y = f(x) with first coordinate x. We
define the graph of the function y = f(x) to be the plot of all solutions of
this equation (in the xy coordinate system). It is common to refer to this
as either the “graph of f(x)” or the “graph of f.”
Graph = {(x,f(x)) | x in the domain} (5.1)
Important Procedure 5.3.1. Points on a graph. The description of the

graph of a function gives us a procedure to produce points on the graph
AND to test whether a given point is on the graph. On the one hand, if
you are given u in the domain of a function y = f(x), then you immediately
can plot the point (u, f(u)) on the graph. On the other hand, if someone
gives you a point (u, v), it will be on the graph only if v = f(u) is true. We
illustrate this in Example 5.3.2.
Example 5.3.2. The function s = h(t) = 15 8

(t − 4)2 − 10 de-
fines a function in the independent variable t. If we restrict 60
s-axis
to the domain 0 ≤ t ≤ 10, then the discussion in Chapter 7 50
40
tells us that the graph is a portion of a parabola: See Fig- 30
20
ure 5.8. Using the above procedure, you can verify that the 10 4 t-axis
data points discussed in the seagull example (in §5.1) all −10 2 6 8 10
lie on this parabola. On the other hand, the point (0,0) is Figure 5.8: s = h(t).
NOT on the graph, since h(0) = 20 6= 0.
5.4 The Vertical Line Test

There is a pictorial aspect of the graph of a function that is very revealing:
Since (x, f(x)) is the only point on the graph with first coordinate equal to
x, a vertical line passing through x on the x-axis (with x in the domain)
crosses the graph of y = f(x) once and only once. This gives us a decisive
way to test if a curve is the graph of a function.
Important Procedure 5.4.1. The vertical line test. Draw a curve in

the xy-plane and specify a set D of x-values. Suppose every vertical line
through a value in D intersects the curve exactly once. Then the curve is
the graph of some function on the domain D. If we can find a single vertical
line through some value in D that intersects the curve more than once, then
the curve is not the graph of a function on the domain D.
For example, draw any straight line m in the plane. By the vertical
line test, if the line m is not vertical, m is the graph of a function. On the
other hand, if the line m is vertical, then m is not the graph of a function.
These two situations are illustrated in Figure 5.9. As another example,
consider the equation x2 + y2 = 1, whose graph is the unit circle and
specify the domain D to be −1 ≤ x ≤1; recall Example 3.2.2. The vertical
line passing through the point 12 , 0 will intersect the unit circle twice;
by the vertical line test, the unit circle is not the graph of a function on
the domain −1 ≤ x ≤ 1.
y-axis y-axis y-axis

l crosses curve twice
l
m m
C
x-axis x-axis x-axis
Figure 5.9: Applying the vertical line test.
5.4.1 Imposed Constraints

In physical problems, it might be natural to constrain (meaning to “limit”
or “restrict”) the domain. As an example, suppose the height s (in feet) of
a ball above the ground after t seconds is given by the function
s = h(t) = −16t2 + 4.
We could look at the graph of the function in the ts-
s-axis
plane and we will review in Chapter 7 that the graph
Physically interesting looks like a parabola. The physical context of this prob-
portion of graph.
lem makes it natural to only consider the portion of the
graph in the first quadrant; why? One way of specifying
t-axis this quadrant would be to restrict the domain of possible
t values to lie between 0 and 12 ; notationally, we would
write this constraint as 0 ≤ t ≤ 12 .
Figure 5.10: Restricting the
domain.
5.5 Linear Functions

A major goal of this course is to discuss several different kinds of func-
tions. The work we did in Chapter 4 actually sets us up to describe one
5.6. PROFIT ANALYSIS 67
very useful type of function called a linear function. Back in Chapter 4,

we discussed how lines in the plane can be described using equations in
the variables x and y. One of the key conclusions was:
Important Fact 5.5.1. A non-vertical line in the plane will be the graph
of an equation y = mx + b, where m is the slope of the line and b is the
y-intercept.
Notice that any non-vertical line will satisfy the conditions of the ver-
tical line test, which means it must be the graph of a function. What is
the function? The answer is to use the equation in x and y we already ob-
tained in Chapter 4: The rule f(x) = mx + b on some specified domain will
have a line of slope m and y-intercept b as its graph. We call a function
of this form a linear function.
Example 5.5.2. You are driving 65 mph from the Kansas state line (mile
marker 0) to Salina (mile marker 130) along I-35. Describe a linear function
that calculates mile marker after t hours. Describe another linear function
that will calculate your distance from Salina after t hours.
Solution. Define a function d(t) to be the mile marker after

t hours. Using “distance=rate×time,” we conclude that y-axis
120
65t will be the distance traveled after t hours. Since we 100
started at mile marker 0, d(t) = 65t is the rule for the first 80
60
function. A reasonable domain would be to take 0 ≤ t ≤ 2, 40
since it takes 2 hours to reach Salina. 20
t-axis
For the second situation, we need to describe a dif- 0.5 1 1.5 2
ferent function, call it s(t), that calculates your distance Figure 5.11: Distance func-
tions.
from Salina after t hours. To describe the rule of s(t) we
can use the previous work:
s(t) = (mile marker Salina) −
(your mile marker at t hrs.)
= 130 − d(t)
= 130 − 65t.
For the rule s(t), the best domain would again be 0 ≤ t ≤ 2. We
have graphed these two functions in the same coordinate system: See
Figure 5.11 (Which function goes with which graph?).
5.6 Profit Analysis

Let’s give a first example of how to interpret the graph of a function in
the context of an application.
Example 5.6.1. A software company plans to bring a new product to mar-

ket. The sales price√per unit is $15 and the expense to produce and market
x units is $100(1 + x). What is the profit potential?
Two functions control the profit potential of the new software. The
first tells us the gross income, in dollars, on the sale of x units. All of
the costs involved in developing, supporting, distributing and marketing
x units are controlled by the expense equation (again in dollars):
g(x) = 15x (gross income function)
√
e(x) = 100(1 + x) (expense function)
A profit will be realized on the sale of x units whenever the gross
income exceeds expenses; i.e., this occurs when g(x) > e(x). A loss occurs
on the sale of x units when expenses exceed gross income; i.e., when
e(x) > g(x). Whenever the sale of x units yields zero profit (and zero loss),
we call x a break-even point ; i.e., when e(x) = g(x).
The above approach is “symbolic.” Let’s see how to study profit and
loss visually, by studying the graphs of the two functions g(x) and e(x).
To begin with, plot the graphs of the two individual functions in the xy-
coordinate system. We will focus on the situation when the sales figures
are between 0 to 100 units; so the domain of x values is the interval
0 ≤ x ≤ 100. Given any sales figure x, we can graphically relate three

1500 1500
gross income
expense
1000 1000
(x, 15x) (x, e(x))
500 500 Q
P x-axis (units sold) x-axis (units sold)
20 40 60 80 100 20 40 60 80 100
(a) Gross income graph. (b) Expenses graph.
Figure 5.12: Visualizing income and expenses.
things:
• x on the horizontal axis;
• a point on the graph of the gross income or expense function;
• y on the vertical axis.
If x = 20 units sold, there is a unique point P = (20, g(20)) = (20, 300) on
the gross income graph and a unique point Q = (20, e(20)) = (20, 547) on
the expenses graph. Since the y-coordinates of P and Q are the function
values at x = 20, the height of the point above the horizontal axis is
controlled by the function.
5.6. PROFIT ANALYSIS 69
dollars y
If we plot both graphs in the same coordinate (x, g(x))
system, we can visually study the distance between 1400
1200
points on each graph above x on the horizontal axis. 1000
(x, e(x))
In the first part of this plot, the expense graph is 800 (x, e(x))
B
600
above the income graph, showing a loss is realized; 400
200 (x, g(x))
the exact amount of the loss will be e(x) − g(x), which sold units x
is the length of the pictured line segment. Further to 20 40 60 80 100

the right, the two graphs cross at the point labeled Figure 5.13: Modelling profit and
“B”; this is the break-even point; i.e., expense and loss.
income agree, so there is zero profit (and zero loss).
Finally, to the right of B the income graph is above the expense graph, so
there is a profit; the exact amount of the profit will be g(x) − e(x), which is
the length of the right-most line segment. Our analysis will be complete
once we pin down the break-even point B. This amounts to solving the
equation g(x) = e(x).
√
15x = 100(1 + x)
√
15x − 100 = 100 x
225x2 − 3000x + 10000 = 10000x
225x2 − 13000x + 10000 = 0.
Applying the quadratic formula, we get two answers: x = 0.78 or 57.

Now, we face a problem: Which of these two solutions is the answer
to the original problem? We are going to argue that only the second
solution x = 57 gives us the break even point. What about the other
”solution” at x = 0.78?
√ Try plugging x = 0.78 into the original equation:
15(0.78) 6= 100(1 + 0.78). What has happened? Well, when going from
the second to the third line, both sides of the equation were squared.
Whenever we do this, we run the risk of adding extraneous solutions.
What should you do? After solving any equation, look back at your steps
and ask yourself whether or not you may have added (or lost) solutions.
In particular, be wary when squaring or taking the square root of both
sides of an equation. Always check your final answer in the original
equation.
We can now compute the coordinates of the break-even point using
either function:
B = (57, g(57)) = (57, 855) = (57, e(57)).

5.7 Exercises
gould
Problem 5.1. For each of the following func-
padelford
tions, find the expression for
f(x + h) − f(x)
.
h (a) Dave leaves Padelford Hall and walks at
Simplify each of your expressions far enough a constant speed until he reaches Gould
so that plugging in h = 0 would be allowed. Hall 10 minutes later.
(a) f(x) = x2 − 2x. (b) Dave leaves Padelford Hall and walks at
a constant speed. It takes him 6 min-
(b) f(x) = 2x + 3 utes to reach the half-way point. Then
(c) f(x) = x2 − 3 he gets confused and stops for 1 minute.
He then continues on to Gould Hall at
(d) f(x) = 4 − x2 the same constant speed he had when
(e) f(x) = −πx2 − π2 he originally left Padelford Hall.
√ (c) Dave leaves Padelford Hall and walks at
(f) f(x) = x − 1. (Hint: Rationalize the nu-
merator) a constant speed. It takes him 6 min-
utes to reach the half-way point. Then
he gets confused and stops for 1 minute
Problem 5.2. Here are the graphs of two lin- to figure out where he is. Dave then con-
ear functions on the domain 0 ≤ x ≤ 20. Find tinues on to Gould Hall at twice the con-
the formula for each of the rules y = f(x) and stant speed he had when he originally
y = g(x). Find the formula for a NEW func- left Padelford Hall.
tion v(x) that calculates the vertical distance
(d) Dave leaves Padelford Hall and walks at
between the two lines at x. Explain in terms
a constant speed. It takes him 6 min-
of the picture what v(x) is calculating. What is
utes to reach the half-way point. Dave
v(5)? What is v(20)? What are the smallest and
gets confused and stops for 1 minute to
largest values of v(x) on the domain 0 ≤ x ≤ 20?
figure out where he is. Dave is totally
y-axis lost, so he simply heads back to his of-
60 (20,60) fice, walking the same constant speed
g(x) he had when he originally left Padelford
Hall.
40 (e) Dave leaves Padelford heading for Gould
Hall at the same instant Angela leaves
(0,24) Gould Hall heading for Padelford Hall.
20 f(x) (20,20) Both walk at a constant speed, but An-
gela walks twice as fast as Dave. Indi-
(0,4) cate a plot of “distance from Padelford”
x-axis
vs. “time” for both Angela and Dave.
10 20
(f) Suppose you want to sketch the graph of
a new function s = g(t) that keeps track
Problem 5.3. Dave leaves his office in of Dave’s distance s from Gould Hall at
Padelford Hall on his way to teach in Gould time t. How would your graphs change
Hall. Below are several different scenarios. in (a)-(e)?
In each case, sketch a plausible (reasonable)
graph of the function s = d(t) which keeps
track of Dave’s distance s from Padelford Hall Problem 5.4. At 5 AM one day, a monk be-
at time t. Take distance units to be “feet” and gan a trek from his monastery by the sea to
time units to be “minutes.” Assume Dave’s the monastery at the top of a mountain. He
path to Gould Hall is along a straight line reached the mountain-top monastery at 11
which is 2400 feet long. AM, spent the rest of the day in meditation,
5.7. EXERCISES 71
and then slept the night there. In the morning, y-axis

at 5 AM, he began walking back to the seaside
monastery. Though walking downhill should
have been faster, he dawdled in the beautiful
sunshine, and ending up getting to the seaside
monastery at exactly 11 AM.
(a) Was there necessarily a time during each

trip when the monk was in exactly the
same place on both days? Why or why
not?
x-axis
(b) Suppose the monk walked faster on the

second day, and got back at 9 AM. What
is your answer to part (a) in this case? Recall the procedure 5.3.1 on page 65.
(a) Find the x and y intercepts of the graph.
(c) Suppose the monk started later, at 10 (b) Find the exact coordinates of all points
AM, and reached the seaside monastery (x,y) on the graph which have y-
at 3 PM. What is your answer to part (a) coordinate equal to 5.
in this case?
(c) Find the coordinates of all points (x,y) on
the graph which have y-coordinate equal
to -3.
(d) Which of these points √ is on the
√ graph:
Problem 5.5. Sketch a reasonable graph for (1, − 2), (−1,3), (2.4,8), ( 3,7 − 3 3).
each of the following functions. Specify a rea-
sonable domain and range and state any as- (e) Find the exact coordinatesp of the point
√
sumptions you are making. Finally, describe (x,y) on the graph with x = 1 + 2.
the largest and smallest values of your func-
tion.
Problem 5.7. After winning the lottery, you
(a) Height of a person depending on age. decide to buy your own island. The island is
located 1 km offshore from a straight portion of
the mainland. There is currently no source of
(b) Height of the top of your head as you electricity on the island, so you want to run a
jump on a pogo stick for 5 seconds. cable from the mainland to the island. An elec-
trical power sub-station is located 4 km from
(c) The amount of postage you must put your island’s nearest location to the shore. It
on a first class letter, depending on the costs $50,000 per km to lay a cable in the wa-
weight of the letter. ter and $30,000 per km to lay a cable over the
land.
(d) Distance of your big toe from the ground
ocean your island
as you ride your bike for 10 seconds.
cable path
1 km
(e) Your height above the water level in a power
swimming pool after you dive off the high
board.
x
4 km
Problem 5.6. Here is a picture of the graph of (a) Explain why we can assume the cable
the function f(x) = 3x2 − 3x − 2. follows the path indicated in the picture;
i.e. explain why the path consists of Problem 5.9. Which of the curves in Fig-
two line segments, rather than a weird ure 5.14 represent the graph of a function? If
curved path AND why it is OK to assume the curve is not the graph of a function, de-
the cable reaches shore to the right of scribe what goes wrong and how you might
the power station and the left of the is- “fix it.” When you describe how to “fix” the
land. graph, you are allowed to cut the curve into
pieces and such that each piece is the graph
(b) Let x be the distance downshore from
of a function. Many of these problems have
the power sub-station to where the cable
more than one correct answer.
reaches the land. Find a function f(x) in
the variable x that computes the cost to
lay a cable out to your island. Problem 5.10. Find an EXACT answer for
each problem.
(c) Make a table of values of f(x), where
x = 0, 12 ,1, 32 ,2,. . . , 72 ,4. Use these calcula- (a) Solve for x
tions to estimate the installation of min- x 5 30
imal cost. + = 2
x+3 x−7 x − 4x − 21
(b) Solve for x

Problem 5.8. This problem deals with the
√ x
“mechanical aspects” of working with the rule 5x − 4 = + 2
of a function. For each of the functions listed 2
in (a)-(c), calculate: f(0), f(−2), f(x + 3), f(♥),
f(♥ + △). (c) Solve for x
√ √
(a) The function f(x) = 21 (x − 3) on the do- x + x − 20 = 10
main of all real numbers.
(d) Solve for t
(b) The function f(x) = 2x2 − 6x on the do-
main of all real numbers. √ √
2t − 1 + 3t + 3 = 5
(c) The function f(x) = 4π2 .
5.7. EXERCISES 73
(a) (b) (c) (d)
(e) (f) (g) (h)
(i) (j) (k) (l)
(m) (n) (o) (p)
Figure 5.14: Curves to consider for Problem 5.9

Chapter 6
Graphical Analysis
We ended the previous section with an in-depth look at a “Profit Analysis

Problem.” In that discussion, we looked at the graphs of the relevant
functions and used these as visual aids to help us answer the questions
posed. This was a concrete illustration of what is typically called “graph-
ical analysis of a function.” This is a fundamental technique we want
to carry forward throughout the course. Let’s highlight the key ideas for
future reference.
6.1 Visual Analysis of a Graph

A variety of information can be visually read off of a function graph.
To see this, we ask ourselves the following question: What is the most
basic qualitative feature of a graph? To answer this, we need to return
to the definition of the graph (see Equation (5.1) on page 65) and the
surrounding discussion. The key thing about the graph of a function
f(x) is that it keeps track of a particular set of points in the plane whose
coordinates are related by the function rule. To be precise, a point P =
(x,y) will be on the graph of the function f(x) exactly when y = f(x).
6.1.1 Visualizing the domain and range

A function is a package that consists of a rule y = f(x), a domain of
allowed x-values and a range of output y-values. The domain can be
visualized as a subset of the x-axis and the range as a subset of the
y-axis. If you are handed the domain, it is graphically easy to describe
the range values obtained; here is the procedure:
Important Procedure 6.1.1. Look at all points on the graph correspond-

ing to domain values on the x-axis, then project these points to the y-axis.
The collection of all values you obtain on the y-axis will be the range of
the function. This idea of “projection” is illustrated in the two graphs be-
75
76 CHAPTER 6. GRAPHICAL ANALYSIS
low. We use arrows “→” to indicate going from a domain x-value, up to the
graph, then over to the y-axis:
range = [1,9]
graph of
f (x) = −2x + 3
domain = [3,5]
range = domain
[−7, − 3] = [−3,1] graph of
f (x) = −2x + 3
(a) Domain: [3, 5]. (b) Domain: [−3, 1].
Figure 6.1: Example projections.
6.1.2 Interpreting Points on the Graph

We can visually detect where a function has positive or negative values:
Important Fact 6.1.2. The function values f(x) control the height of the
point P(x) = (x,f(x)) on the graph above the x-axis; if the function value f(x)
is negative, the point P(x) is below the x-axis.
f(x) units above x-axis
f(x3 ) units above the x-axis

P (x) = (x,f(x))
P (x3 ) = (x3 ,f(x3 ))

x2
x x3 x-axis
|f(x)| units below the x-axis P(x2 ) = (x2 ,f(x2 ))
y-axis
Figure 6.2: Interpreting points on a graph.
In Figure 6.2 we can now divide the domain (in this case the whole num-
ber line) into segments where the function is above, below or crossing the
axis. Keeping track of this information on a number line is called a sign
plot for the function. We include a “shadow” of the graph in Figure 6.3
6.1. VISUAL ANALYSIS OF A GRAPH 77
positive positive
negative negative x-axis
Figure 6.3: Sign plot.
to emphasize how we arrived at our “positive” and “negative” labeling of

the sign plot; in practice we would only provide a labeled number line.
By moving through a sequence of x values we can investigate how the
corresponding points on the graph move “up and down”; this then gives
us a dynamic visual sense of how the function values are changing. For
example, in Figure 6.4, suppose we let x move from 1 to 5, left to right;
we have indicated how the corresponding points on the curve will move
and how the function values will change.
y-axis
f(1) P
#2: points on graph move from P to Q
f(2)
#3: f(x) values move
like this
graph of y = f(x)
f(3)
f(5) Q
f(4)
1 5
x-axis
#1: x values move from 1 to 5
Figure 6.4: Dynamic interpretation of a graph.

6.1.3 Interpreting Intercepts of a Graph

y-axis
The places where a graph crosses the axes are often sig-
y-intercept = (0,f(0)) nificant. We isolate each as an important feature to look
(x 3 ,0) (x 2 ,0) x-axis
for when doing graphical analysis. The graph of the func-
(x 1 ,0) tion y = f(x) crosses the y-axis at the point (0,f(0)); so,
the y-intercept of the graph is just f(0). The graph of the
x-intercepts have the form “(x,0)”
function y = f(x) crosses the x-axis at points of the form
Figure 6.5: Intercepts of a
(x0,f(x0)), where f(x0) = 0. The values x0 are called roots
graph.
or zeros of the function f(x). There can be at most one
y-intercept, but there can be several x-intercepts or no
x-intercept: See Figure 6.5
The graph of a function y = f(x) crosses the vertical line x = h at the
point (h,f(h)). To find where the graph of a function y = f(x) crosses the
horizontal line y = k, first solve the equation k = f(x) for x. If the equation
k = f(x) has solutions x1, x2, x3, x4, then the points of intersection would
have the coordinates given.
y-axis upper semicircle radius

y-axis r = 2 centered at (2,1)
(h,(f(h))
y=k
graph of f(x)
(x 1 ,k) (x 2 ,k) (x 3 ,k) (x 4 ,k)

y=k y= 1
2
x1 x2 x3 x4 x-axis
x=h x=h x-axis
(a) General curve. (b) Semicircle.
Figure 6.6: Crossing horizontal and vertical lines.
As another example, the graph in Figure 6.6(b) above will cross the
horizontal line y = k twice if and only if 1 ≤ k < 3; the graph will cross the
horizontal line y = k once if and only if k = 3. The graph will not intersect
the line y = 21 and the graph will cross the vertical line x = h if and only if
0 ≤ h ≤ 4.
6.1.4 Interpreting Increasing and Decreasing

y-axis We use certain terms to describe how the function values
x-axis
are changing over some domain of x values. Typically, we
want to study what is happening to the values f(x) as x
ll
hi
up
dow
moves from “left to right” in some interval. This can be
nh = local extrema
ill
linked graphically with the study of “uphill” and “down-
Figure 6.7: Graphically in- hill” portions of the function graph: If you were “walking
terpreting increasing and de- to the right” along the graph, the function values are in-
creasing.
creasing if you are walking uphill. Likewise, if you were
6.2. CIRCLES AND SEMICIRCLES 79
“walking to the right” along the graph, the function values are decreasing
if you are walking downhill.
Once we understand where the graph is moving uphill and downhill,
we can isolate the places where we change from moving uphill to down-
hill, or vice versa; these “peaks” and “valleys” are called local maxima
and local minima. Some folks refer to either case as a local extrema.
People have invested a lot of time (centuries!) and energy (lifetimes!) into
the study of how to find local extrema for particular function graphs. We
will see some basic examples in this course and others will surface in
future courses once you have the tools of calculus at your disposal. Ex-
amples range from business applications that involve optimizing profit to
understanding the three-dimensional shape a of biological molecule.
Example 6.1.3. A hang glider launches from a gliderport ft above gliderport

in La Jolla. The launch point is located at the edge of a 600
500 ft. high cliff over the Pacific Ocean. The elevation of 400
200
the pilot above the gliderport after t minutes is given by the minutes
graph in Figure 6.8: 5 10

−200
1. When is the pilot climbing and descending? −400
−600
2. When is the pilot at the glider port elevation?

Figure 6.8: Hanglider eleva-
3. How much time does the pilot spend flying level? tions.
Solution.
1. Graphically, we need to determine the portions of the graph that
are increasing or decreasing. In this example, it is increasing when
0 ≤ t ≤ 2 and 7 ≤ t ≤ 9. And, it is decreasing when 3 ≤ t ≤ 5 and
9 ≤ t ≤ 10.
2. Graphically, this question amounts to asking when the elevation
is 0, which is the same as finding when the graph crosses the hori-
zontal axis. We can read off there are four such times: t = 0, 4, 8, 10.
3. Graphically, we need to determine the portions of the graph that are
made up of horizontal line segments. This happens when 2 ≤ t ≤ 3
and 5 ≤ t ≤ 7. So, our pilot flies level for a total of 3 minutes.
6.2 Circles and Semicircles

Back in Chapter 3, we discussed equations whose graphs were circles:
We found that the graph of the equation
(x − h)2 + (y − k)2 = r2 (6.1)
is a circle of radius r centered at the point (h, k). It is possible to ma-

nipulate this equation and become confused. We could rewrite this as
(y − k)2 = r2 − (x − h)2, then take the square root of each side. However,
the resulting equivalent equation would be
p
y = k ± r2 − (x − h)2
and the presence of that ± sign is tricky; it means we have two equations:
p
y = k + r2 − (x − h)2 or
p
y = k − r2 − (x − h)2.
Each of these two equations defines a function:
p
f(x) = k + r2 − (x − h)2 or (6.2)
p
g(x) = k − r2 − (x − h)2. (6.3)
So, even though the Equation 6.1 is not a function, we were able to ob-
tain two different functions f(x) and g(x) from the original equation. The
relationship between the graph of the original equation and the graphs
of the two functions in (6.2) and (6.3) is as follows: The upper semicircle
is the graph of the function f(x) and the lower semicircle is the graph of
the function g(x).
y-axis upper semicircle y-axis
(h,k)
(h,k)
lower semicircle
x-axis x-axis
Graph of y = f(x) Graph of y = g(x)
Figure 6.9: Upper and lower semicircles.
Example 6.2.1. A tunnel connecting two portions of a space station has

a circular cross-section of radius 15 feet. Two walkway decks are con-
structed in the tunnel. Deck A is along a horizontal diameter and another
parallel Deck B is 2 feet below Deck A. Because the space station is in
a weightless environment, you can walk vertically upright along Deck A,
or vertically upside down along Deck B. You have been assigned to paint
“safety stripes” on each deck level, so that a 6 foot person can safely walk
upright along either deck. Determine the width of the “safe walk zone” on
each deck.
6.3. MULTIPART FUNCTIONS 81
Solution. Impose a coordinate system so that the origin is

at the center of the circular cross section of the tunnel; by Deck A
symmetry the walkway is centered about the origin. With
this coordinate system, the graph of the equation x2 +y2 = x-axis
152 = 225 will be the circular cross-section of the tunnel.

Deck B
In the case of Deck A, we basically need to determine r = 15ft
how close to each edge of the tunnel a 6 foot high person
can stand without hitting his or her head on the tunnel; (a) Cross-section of tunnel.
a similar remark applies to Deck B. This means we are
really trying to fit two six-foot-high rectangular safe walk Deck A
zones into the picture:
Our job is to find the coordinates of the four points Deck B
P, Q, R, and S. Let’s denote by x1, x2, x3,and x4 the x- = Safe walk zone
coordinates of these four points, then P = (x1, 6), Q = (b) Walk zones.
(x2, 6), R = (x3, −8), and S = (x4, −8). To find x1, x2, x3,
and x4, we need to find the intersection of the circle in Figure 6.10: Space station
Figure 6.10(b) with two horizontal lines: tunnels.
• Intersecting the the upper semicircle with the hori-

zontal line having equation y = 6 will determine x1
and
√ x2; the upper semicircle is the graph of f(x) =
225 − x2.
• Intersecting the lower semicircle with the horizon-

tal line having equation y = −8 will determine x3
and
√ x4; the lower semicircle is the graph of g(x) =
− 225 − x2.
For Deck A, we simultaneously solve the system of equations
√
y = 225 − x2
.
y =6
2
Plugging in y = 6 into√ the first equation of the system gives x = 225 −
2
6 = 189; i.e., x = ± 189 = ±13.75. This tells us that P = (−13.75, 6) and
Q = (13.75, 6). In a similar way, for Deck B, we find R = (−12.69, −8) and
S = (12.69, −8).
In the case of Deck A, we would paint a safety stripe 13.75 feet to the
right and left of the centerline. In the case of Deck B, we would paint a
safety stripe 12.69 feet to the right and left of the centerline.
6.3 Multipart Functions

So far, in all of our examples we have been able to write f(x) as a nice
compact expression in the variable x. Sometimes we have to work harder.
As an example of what we have in mind, consider the graph in Fig-

ure 6.11(a):
y-axis 

 −1 if 0≤x<1
1 

 1 if 1≤x<2
1 2 3 4 x-axis f(x) = −1 if 2≤x<3


−1

 1 if 3≤x<4

−1 if x=4
(a) Graphing a multipart (b) Writing a multipart function.
function.
Figure 6.11: A multipart function.
The curve we are trying to describe in this picture is made up of five

pieces; four little line segments and a single point. The first thing to
notice is that on the domain 0 ≤ x ≤ 4, this curve will define the graph of
some function f(x). To see why this is true, imagine a vertical line moving
from left to right within the domain 0 ≤ x ≤ 4 on the x-axis; any one of
these vertical lines will intersect the curve exactly once, so by the vertical
line test, the curve must be the graph of a function. Mathematicians use
the shorthand notation above to describe this function. Notice how the
rule for f(x) involves five cases; each of these cases corresponds to one
of the five pieces that make up the curve. Finally, notice the care with
the “open” and “closed” circles is really needed if we want to make sure
the curve defines a function; in terms of the rule, these open and closed
circles translate into strict inequalities like < or weak inequalities like ≤.
This is an example of what we call a multipart function.
The symbolic appearance of multipart functions can be somewhat
frightening. The key point is that the graph (and rule) of the function
will be broken up into a number of separate cases. To study the graph
or rule, we simply “home in” on the appropriate case. For example, in
the above illustration, suppose we wanted to compute f(3.56). First, we
would find which of the five cases covers x = 3.56, then apply that part of
the rule to compute f(3.56) = 1.
Our first multipart function example illustrated how to go from a
graph in the plane to a rule for f(x); we can reverse this process and
go from the rule to the graph.
Example 6.3.1. Sketch the graph of the multipart function


 1 √ if x ≤ −1
g(x) = 2
1 + 1 − x if −1 ≤ x ≤ 1

1 if x ≥ 1
6.3. MULTIPART FUNCTIONS 83
y-axis second part

Solution. The graph of g(x) will consist of three pieces. of graph
The first case consists of the graph of the function y =

g(x) = 1 on the domain x ≤ −1, this consists of all points
x-axis
on the horizontal line y = 1 to the left of and including the
first part of graph third part of graph
point (−1,1). We have “lassoed” this portion of the graph
in Figure 6.12. Likewise, the third case in the definition = graph of g(x)
yields the graph of the function y = g(x) = 1 on the domain

Figure 6.12: Multipart func-
x ≥ 1; this is just all points on the horizontal line y = 1 tion g(x).
to the right of and including the point (1, 1). Finally, we
need to analyze the middle√ case, which means we need to
look at the graph of 1 + 1 − x2 on the domain −1 ≤ x ≤ 1. This is just
the upper semicircle of the circle of radius 1 centered at (0,1). If we paste
these three pieces together, we arrive at the graph of g(x).
Example 6.3.2. You are dribbling a basketball and the

function s = h(t) keeps track of the height of the ball’s cen- 5
4
ter above the floor after t seconds. Sketch a reasonable
feet
3
graph of s = h(t). 2
1 2
seconds
Solution. If we take the domain to be 0 ≤ t ≤ 2 (the first 2 Figure 6.13: Dribbling.

seconds), a reasonable graph might look like Figure 6.13.
This is a multipart function. Three portions of the graph are decreasing
and two portions are increasing. Why doesn’t the graph touch the t axis?
6.4 Exercises
Problem 6.1. The absolute value function is Problem 6.4. (a) Let f(x) = x + |2x − 1|. Find
defined by the multipart rule: all solutions to the equation

x if 0 ≤ x
|x| = f(x) = 8.
−x if x < 0
The graph of the absolute value function is pic-

(b) Let g(x) = 3x − 3 + |x + 5|. Find all values
tured below:
of a which satisfy the equation
y-axis
g(a) = 2a + 8.
y = |x|
(c) Let h(x) = |x| − 3x + 4. Find all solutions
to the equation
x-axis h(x − 1) = x − 2.
Problem 6.5. Express the area of the shaded

region below as a function of x. The dimen-
(a) Calculate: |0|, |2|, | − 3|.
sions in the figure are centimeters.
(b) Solve for x: |x| = 4; |x| = 0, |x| = −1.
(c) Sketch the graph of y = 12 x + 2 and y = |x|
in the same coordinate system. Find
where the two graphs intersect, label the
coordinates of these point(s), then find
the area of the region bounded by the
two graphs.
5
tions, graph f(x) and g(x) = |f(x)|, and give the 3
multipart rule for g(x).
(a) f(x) = −0.5x − 1
(b) f(x) = 2x − 5
x
(c) f(x) = x + 3
6
Problem 6.3. Solve each of the following equa-
tions for x.
(a) g(x) = 17, where g(x) = |3x + 5|
(b) f(x) = 1.5 where Problem 6.6. Pizzeria Buonapetito makes a
triangular-shaped pizza with base width of 30
2x if x < 3, inches and height 20 inches as shown. Alice
f(x) =
4−x if x ≥ 3. wants only a portion of the pizza and does so
by making a vertical cut through the pizza and
(c) h(x) = −1 where taking the shaded portion. Letting x be the
bottom length of Alice’s portion and y be the
−8 − 4x if x ≤ −2, length of the cut as shown, answer the follow-
h(x) =
1 + 13 x if x > −2. ing questions:
6.4. EXERCISES 85
Problem 6.8. Arthur is going for a run. From

his starting point, he runs due east at 10 feet
per second for 250 feet. He then turns, and
20 runs north at 12 feet per second for 400 feet.
y He then turns, and runs west at 9 feet per sec-
ond for 90 feet.
Express the (straight-line) distance from
x Arthur to his starting point as a function of
10 20 t, the number of seconds since he started.
Problem 6.9. A baseball diamond is a square

with sides of length 90 ft. Assume Edgar hits a
20 home run and races around the bases (coun-
y terclockwise) at a speed of 18 ft/sec. Express
the distance between Edgar and home plate as
a function of time t. (Hint: This will be a mul-
tipart function.) Try to sketch a graph of this
x
function.
(a) Find a formula for y as a multipart func-

tion of x, for 0 ≤ x ≤ 30. Sketch the graph
of this function and calculate the range.
(b) Find a formula for the area of Alice’s 90 ft
portion as a multipart function of x, for Edgar
0 ≤ x ≤ 30.
(c) If Alice wants her portion to have half
the area of the pizza, where should she
make the cut? d(t)
Problem 6.7. This problem deals with cars

traveling between Bellevue and Spokane,
which are 280 miles apart. Let t be the time
in hours, measured from 12:00 noon; for ex- home plate
ample, t = −1 is 11:00 am.
(a) Joan drives from Bellevue to Spokane at
a constant speed, departing from Belle-
vue at 11:00 am and arriving in Spokane Problem 6.10. Pagliacci Pizza has designed a
at 3:30 pm. Find a function j(t) that cardboard delivery box from a single piece of
computes her distance from Bellevue at cardboard, as pictured.
time t. Sketch the graph, specify the do-
main and determine the range.
(a) Find a polynomial function v(x) that
(b) Steve drives from Spokane to Bellevue computes the volume of the box in terms
at 70 mph, departing from Spokane at of x. What is the degree of v?
12:00 noon. Find a function s(t) for his
distance from Bellevue at time t. Sketch (b) Find a polynomial function a(x) that
the graph, specify the domain and deter- computes the exposed surface area of
mine the range. the closed box in terms of x. What is
(c) Find a function d(t) that computes the the degree of a? What are the explicit di-
distance between Joan and Steve at mensions if the exposed surface area of
time t. the closed box is 600 sq. inches?
x (a) When will the ditch be completely full?
x (b) Find a multipart function that models

the vertical cross-section of the ditch.
20 in (c) What is the width of the filled portion of
the ditch after 1 hour and 18 minutes?
(d) When will the filled portion of the ditch
x
be 42 feet wide? 50 feet wide? 73 feet
x wide?
50 in Problem 6.12. The graph of a function y =

g(x) on the domain −6 ≤ x ≤ 6 consistes of line
remove shaded squares and fold to get: segments and semicircles of radius 2 connect-
ing the points (−6,0),(−4,4), (0,4), (4,4), (6,0).
y-axis
x-axis
Problem 6.11. The vertical cross-section of a

drainage ditch is pictured below: (a) What is the range of g?
(b) Where is the function increasing? Where
is the function decreasing?
(c) Find the multipart formula for y = g(x).
(d) If we restrict the function to the smaller
domain −5 ≤ x ≤ 0, what is the range?
3D−view of ditch (e) If we restrict the function to the smaller

domain 0 ≤ x ≤ 4, what is the range?
20 ft 20 ft
Problem 6.13. (a) Simply as far as possible

R R
1 a
R R − .
1 a+1
1+ a
vertical cross-section
(b) Find a, b, c that simultaneously satisfy
Here, R indicates a circle of radius 10 feet and these three equations:
all of the indicated circle centers lie along the
common horizontal line 10 feet above and par- a+b−c = 5
allel to the ditch bottom. Assume that water is 2a − 3b + c = 4
flowing into the ditch so that the level above a+b+c = −1
the bottom is rising 2 inches per minute.
Chapter 7
Quadratic Modeling
If you kick a ball through the air enough times, you will find its path
tends to be parabolic. Before we can answer any detailed questions about
this situation, we need to get our hands on a precise mathematical model
for a parabolic shaped curve. This means we seek a function y = f(x)
whose graph reproduces the path of the ball.
ground level
Figure 7.1: Possible paths for a kicked ball are parabolic.
7.1 Parabolas and Vertex Form

OK, suppose we sit down with an xy-coordinate system and draw four
random parabolas; let’s label them I, II, III, and IV: See Figure 7.2. The
relationship between these parabolas and the fixed coordinate system
can vary quite a bit: The key distinction between these four curves is
that only I and IV are the graphs of functions; this follows from the ver-
tical line test. A parabola that is the graph of a function is called a
standard parabola. We can see that any standard parabola has three
basic features:
87
88 CHAPTER 7. QUADRATIC MODELING
y-axis
II
III
I IV
x-axis
not graphs of functions
Figure 7.2: Relationship between a fixed coordinate system and various parabolas.
• the parabola will either open “upward” or “downward”;
• the graph will have either a “highest point” or “lowest point,” called
the vertex ;
• the parabola will be symmetric about some vertical line called the
axis of symmetry.
Our first task is to describe the mathematical model for any standard
parabola. In other words, what kind of function equations y = f(x) give
us standard parabolas as their graphs? Our approach is geometric and
visual:
• Begin with one specific example, then show every other standard
parabola can be obtained from it via some specific geometric ma-
neuvers.
• As we perform these geometric maneuvers, we keep track of how the

function equation for the curve is changing.
This discussion will amount to a concrete application of a more general

set of tools developed in the following section of this chapter.
Using a graphing device, it is an easy matter to plot
y-axis
35
30
the graph of y = x2 and see we are getting the parabola
25 pictured in Figure 7.3. The basic idea is to describe how
20
15
we can manipulate this graph and obtain any standard
10 parabola. In the end, we will see that standard parabolas
5
x-axis are obtained as the graphs of functions having the form
−6 −4 −2 2 4 6
Figure 7.3: Graph of y = x2 . y = ax2 + bx + c,

7.1. PARABOLAS AND VERTEX FORM 89
for various constants a, b, and c, with a 6= 0. A function of this type is

called a quadratic function and these play a central role throughout the
course. We will divide our task into two steps.
First we show every standard parabola arises as the graph of a func-
tion having the form
y = a(x − h)2 + k,
for some constants a, h, and k, with a 6= 0. This is called the vertex

form of a quadratic function. Notice, if we were to algebraically expand
out this equation, we could rewrite it in the y = ax2 + bx + c form. For
example, suppose we start with the vertex form y = 2(x − 1)2 + 3, so
that a = 2, h = 1, k = 3. Then we can rewrite the equation in the form
y = ax2 + bx + c as follows:
2(x − 1)2 + 3 = 2(x2 − 2x + 1) + 3 = 2x2 − 4x + 5,
so a = 2, b = −4, c = 5. The second step is to show any quadratic function

can be written in vertex form; the underlying algebraic technique used
here is called completing the square. This is a bit more involved. For
example, if you are simply handed the quadratic function y = −3x2+6x−1,
it not at all obvious why the vertex form is obtained by this equality:
−3x2 + 6x − 1 = −3(x − 1)2 + 2.
The reason behind this equality is the technique of completing the square.
In the end, we will almost always be interested in the vertex form of a
quadratic. This is because a great deal of qualitative information about
the parabolic graph can simply be “read off” from this form.
7.1.1 First Maneuver: Shifting

Suppose we start with the graph in Figure 7.3 and hori-
40 y-axis
zontally shift it h units to the right. To be specific, con-
sider the two cases h = 2 and h = 4. To visualize this, 30
imagine making a wire model of the graph, set in on top 20
of the curve, then slide the wire model h units to the right. 10
What you will obtain are the two “dashed curves” in Fig- x-axis
ure 7.4. We will call the process just described a hori- −6 −4 −2 2 4 6
zontal shift. Since the “dashed curves” are no longer the Figure 7.4: Shift to the right.
original parabola in Figure 7.3, the corresponding func-
tion equations must have changed.
Using a graphing device, you can check that the corresponding equa-
tions for the dashed graphs would be
y = (x − 2)2
= x2 − 4x + 4,
which is the plot with lowest point (2, 0) and

y = (x − 4)2
= x2 − 8x + 16,
which is the plot with lowest point (4, 0). In general, if h is positive, the
graph of the function y = (x − h)2 is the parabola obtained by shifting the
graph of y = x2 by h units to the right.
Next, if h is negative, shifting h units to the right is the
y-axis
40
same as shifting |h| units left! On the domain −6 ≤ x ≤ 6,
30 Figure 7.5 indicates this for the cases h = −2, − 4, using
20 “dashed curves” for the shifted graphs and a solid line
10
for the graph of y = x2. Using a graphing device, we can
x-axis check that the corresponding equations for the dashed
−6 −4 −2 2 4 6
graphs would be
Figure 7.5: Shift to the left.
y = (x − (−2))2
= (x + 2)2
= x2 + 4x + 4,
which is the plot with lowest point (−2, 0) and
y = (x − (−4))2
= (x + 4)2
= x2 + 8x + 16,
which is the plot with lowest point (−4, 0). In general, if h is negative, the
graph of the function y = (x − h)2 gives the parabola obtained by shifting
the graph of y = x2 by |h| units to the LEFT.
The conclusion thus far is this: Begin with the graph of y = x2 in
Figure 7.3. Horizontally shifting this graph h units to the right gives a
new (standard) parabola whose equation is y = (x − h)2.
y-axis
We can also imagine vertically shifting the graph in
40 Figure 7.3. This amounts to moving the graph k units
30 vertically upward. It turns out that this vertically shifted
20
graph corresponds to the graph of the function y = x2 + k.
x-axis
10
We can work out a few special cases and use a graphing
−6 −4 −2 2 4 6 device to illustrate what all this really means.
−10
Figure 7.6 illustrates the graphs of y = x2 + k in the
Figure 7.6: Vertical shifts. cases when k = 4, 10 and k = −4, −10, leading to vertically
shifted graphs. Positive values of k lead to the upper two
“dashed curves” and negative values of k lead to the lower two “dashed
curves”; the plot of y = x2 is again the solid line. The equations giving
these graphs would be y = x2 − 10, y = x2 − 4, y = x2 + 4 and y = x2 + 10,
from bottom to top dashed plot.
If we combine horizontal and vertical shifting, we end up with the
graphs of functions of the form y = (x − h)2 + k. Figure 7.7(a) illustrates
7.1. PARABOLAS AND VERTEX FORM 91
the four cases with corresponding equations y = (x±2)2±4; as an exercise,

identify which equation goes with each curve.
7.1.2 Second Maneuver: Reflection

Next, we can reflect any of the curves y = p(x) obtained
y-axis
by horizontal or vertical shifting across the x-axis. This 40
procedure will produce a new curve which is the graph of 30
the new function y = −p(x). For example, begin with the 20
four dashed curves in the previous figure. Here are the 10
reflected parabolas and their equations are y = −(x ± 2)2 ± x-axis

4: See Figure 7.7(b). −6 −4 −2 2 4 6
(a) Combined shifts.

7.1.3 Third Maneuver: Vertical Dilation y-axis
30
2
If a is a positive number, the graph of y = ax is usually 20
called a vertical dilation of the graph of y = x2. There are x-axis

10
two cases to distinguish here: −6 −4 −2

−10
2 4 6
−20
• If a > 1, we have a vertically expanded graph. −30
• If 0 < a < 1, we have a vertically compressed graph.

(b) Reflections.
This is illustrated for a = 2 (upper dashed plot) and a = y-axis
1/2 (lower dashed plot): See Figure 7.7(c). 30
25
20
7.1.4 Conclusion 15
10
Starting with y = x2 in Figure 7.3, we can combine to- x-axis

5
gether all three of the operations: shifting, reflection and −6 −4 −2 2 4 6
dilation. This will lead to the graphs of functions that

(c) Vertical dilations.
have the form:
Figure 7.7: Shifts, reflec-
y = a(x − h)2 + k,
tions, and dilations.
for some a, h and k, a 6= 0. If you think about it for awhile,
it seems pretty easy to believe that any standard parabola
arises from the one in Figure 7.3 using our three geomet-
ric maneuvers. In other words, what we have shown is
that any standard parabola is the graph of a quadratic equation in vertex
form. Let’s summarize.
Important Fact 7.1.1. A standard parabola is the graph of a function
y = f(x) = a(x − h)2 + k, for some constants a, h, and k and a 6= 0. The
vertex of the parabola is (h, k) and the axis of symmetry is the line x = h. If
a > 0, then the parabola opens upward; if a < 0, then the parabola opens
downward.
Example 7.1.2. Describe a sequence of geometric operations leading from

the graph of y = x2 to the graph of y = f(x) = −3(x − 1)2 + 2.
reflect across x-axis Solution. To begin with, we can make some initial conclu-
horizontal shift by h = 1
sions about the specific shifts, reflections and dilations
−3(x − 1)2 + 2 involved, based on looking at the vertex form of the equa-
vertical shift by k = 2
tion. In addition, by Fact 7.1.1, we know that the vertex
vertical dilate by 3 of the graph of y = f(x) is (1, 2), the line x = 1 is a vertical
axis of symmetry and the parabola opens downward.
(a) What do the symbols of an We need to be a little careful about the order in which
equation mean?
we apply the four operations highlighted. We will illus-
20 y-axis trate a procedure that works. The full explanation for the
15 success of our procedure involves function compositions
10
and we will return to that at the end of Chapter 8. The
5
x-axis order in which we will apply our geometric maneuvers is
−6 −4 −2
−5
2 4 6
as follows:
−10
−15
horizontal shift ⇒ vertical dilate
⇒ reflect
(b) What does the equation
look like? ⇒ vertical shift
Figure 7.8(b) illustrates the four curves obtained by ap-
Figure 7.8: Interpreting an
equation. plying these successive steps, in this order. As a refer-
ence, we include the graph of y = x2 as a “dashed curve”:
• A horizontal shift by h = 1 yields the graph of
y = (x − 1)2; this is the fat parabola opening upward
with vertex (1, 0).
• A dilation by a = 3 yields the graph of y = 3(x −
1)2; this is the skinny parabola opening upward with
vertex (1, 0).
• A reflection yields the graph of y = −3(x − 1)2; this is
the downward opening parabola with vertex (1, 0).
• A vertical shift by k = 2 yields the graph of y = −3(x −
1)2 + 2; this is the downward opening parabola with
vertex (1, 2).
7.2 Completing the Square

By now it is pretty clear we can say a lot about the graph of a quadratic
function which is in vertex form. We need a procedure for rewriting a
given quadratic function in vertex form. Let’s first look at an example.
7.2. COMPLETING THE SQUARE 93
Example 7.2.1. Find the vertex form of the quadratic function y = −3x2 +
6x − 1.
Solution. Since our goal is to put the function in vertex form, we can write
down what this means, then try to solve for the unknown constants. Our
first step would be to write
−3x2 + 6x − 1 = a(x − h)2 + k,
for some constants a, h, k. Now, expand the right hand side of this
equation and factor out coefficients of x and x2:
−3x2 + 6x − 1 = a(x − h)2 + k
−3x2 + 6x − 1 = a(x2 − 2xh + h2) + k
−3x2 + 6x − 1 = ax2 − 2xah + ah2 + k
(−3)x2 + (6)x + (−1) = (a)x2 + (−2ah)x + (ah2 + k).
If this is an equation, then it must be the case that the coefficients of
like powers of x match up on the two sides of the equation in Figure 7.9.
Now we have three equations and three unknowns (the a, h, k) and we
Equal
z}|{ z }| {
(−3) x2 + (6) x + (−1) = (a) x2 + (−2ah) x + (ah2 + k)
| {z } | {z } |{z} | {z }
Equal
Equal
Figure 7.9: Balancing the coefficients.
can proceed to solve for these:

−3 = a
6 = −2ah
−1 = ah2 + k
The first equation just hands us the value of a = −3. Next, we can plug
this value of a into the second equation, giving us
6 = −2ah
= −2(−3)h
= 6h,
so h = 1. Finally, plug the now known values of a and h into the third
equation:
−1 = ah2 + k
= −3(12) + k
= −3 + k,
so k = 2. Our conclusion is then
−3x2 + 6x − 1 = −3(x − 1)2 + 2.
Notice, this is the quadratic we studied in Example 7.1.2 on page 92.
The procedure used in the preceding example will always work to
rewrite a quadratic function in vertex form. We refer to this as completing
the square.
Example 7.2.2. Describe the relationship between the graphs of y = x2
and y = f(x) = −4x2 + 5x + 2.
Solution. We will go through the algebra to complete the

20 y-axis
square, then interpret what this all means in terms of
10
graphical maneuvers. We have
x-axis
−6 −4 −2 2 4 6 −4x2 + 5x + 2 = a(x − h)2 + k
(−4)x2 + 5x + 2 = ax2 + (−2ah)x + (ah2 + k).
−10
−20
This gives us three equations:
Figure 7.10: Maneuvering
−4 = a
y = x2 .
5 = −2ah
2 = ah2 + k.
We conclude that a = −4, h = 85 = 0.625 and k = 16
57
= 3.562. So, this tells
us that we can obtain the graph of y = f(x) from that of y = x2 by these
steps:
• Horizontally shifting by h = 0.625 units gives y = (x − 0.625)2.
• Vertically dilate by the factor a = 4 gives y = 4(x − 0.625)2.
• Reflecting across the x-axis gives y = −4(x − 0.625)2.
• Vertically shifting by k = 3.562 gives y = f(x) = −4(x − 0.625)2 + 3.562.
Example 7.2.3. A drainage canal has a cross-section in the shape of a

parabola. Suppose that the canal is 10 feet deep and 20 feet wide at the
top. If the water depth in the ditch is 5 feet, how wide is the surface of the
water in the ditch?
7.3. INTERPRETING THE VERTEX 95
Solution. Impose an xy-coordinate system so that the 00000000000000

11111111111111
00000000000000
11111111111111
20 feet
centerline
parabolic cross-section of the canal is symmetric about 00000000000000

11111111111111
the y-axis and its vertex is the origin. The vertex form of 00000000000000
11111111111111
00000000000000
11111111111111
10 feet
any such parabola is y = f(x) = ax2, for some a > 0; this

is because (h, k) = (0, 0) is the vertex and the parabola Figure 7.11: A drainage
opens upward! The dimension information given tells us canal.
that the points (10, 10) and (−10, 10) are on the graph of
f(x). Plugging into the expression for f, we conclude that 10 = 100a, so
a = 0.1 and f(x) = (0.1)x2. Finally, if the water√ is 5 feet deep, we must
2
solve the equation: 5 = (0.1)x , leading to x = ± 50 = ±7.07. Conclude the
surface of the water is 14.14 feet wide when the water is 5 feet deep.
7.3 Interpreting the Vertex
minimum „ « maximum
−b „
−b
«
value f value f
2a 2a
vertex
vertex
−b −b
2a 2a
Figure 7.12: The vertex as the extremum of the quadratic function.
If we begin with a quadratic function y = f(x) = ax2 + bx + c, we know

the graph will be a parabola. Graphically, the vertex will correspond to
either the “highest point” or “lowest point” on the graph. If a > 0, the
vertex is the lowest point on the graph; if a < 0, the vertex is the highest
point on the graph. The maximum or minimum value of the function is
the second coordinate of the vertex and the value of the variable x for
which this extreme value is achieved is the first coordinate of the vertex.
As we know, it is easy to read off the vertex coordinates when a quadratic
function is written in vertex form. If instead we are given a quadratic
function y = ax2 + bx + c, we can use the technique of completing the
square and arrive at a formula for the coordinates of the vertex in terms
of a, b, and c. We summarize this below and label the two situations
(upward or downward opening parabola) in the Figure 7.12. Keep in
mind, it is always possible to obtain this formula by simply completing

the square.
Important Fact 7.3.1. In applications involving a quadratic function
f(x) = ax2 + bx + c,
the vertex has coordinates P = −b −b

2a
, f 2a
. The second coordinate of the
vertex will detect the maximum or minimum value of f(x); this is often a
key step in problem solving.
Example 7.3.2. Discuss the graph of the quadratic function y = f(x) =
−2x2 + 11x − 4.
Solution. We need to place the equation y = f(x) in vertex

2
form. We can simply compute a = −2, h = −b 2a
= 11
4
and
−4 −2 4 6 8
11 89
−20
k = f( 4 ) = 8 , using Fact 7.3.1:
−40
f(x) = −2x2 + 11x − 4
−60 2
11 89
−80 = −2 x − + .
4 8
Figure 7.13: Sketching y =
f(x). This means that the graph of f(x) is a parabola opening
downward with vertex 11 , 89 and axis x = 11
4 8 4
; see Fig-
ure 7.13.
7.4 Quadratic Modeling Problems

The real importance of quadratic functions stems from the connection
with motion problems. Imagine one of the three kicked ball scenarios in
Figure 7.1 and impose a coordinate system with the kicker located at the
origin. We can study the motion of the ball in two ways:
• Regard time t as the important variable and try to find a function
y(t) which describes the height of the ball t seconds after the ball
is kicked; this would just be the y-coordinate of the ball at time t.
The function y(t) is a quadratic function. If we had this function in
hand, we could determine when the ball hits the ground by solving
the equation 0 = y(t), but we would not be able to determine where
the ball hits the ground.
• A second approach is to forget about the time variable and simply
try to find a function y = f(x) whose graph models the exact path of
the ball. In particular, we could find where the ball hits the ground
by solving 0 = f(x), but we would not be able to determine when the
ball hits the ground.
7.4. QUADRATIC MODELING PROBLEMS 97
Example 7.4.1. Figure 7.14(a) shows a ball is located on path of kicked ball
the edge of a cliff. The ball is kicked and its height (in feet)
cliff
above the level ground is given by the function s = y(t) = ground level
−16t2 + 48t + 50, where t represents seconds elapsed after
kicking the ball. What is the maximum height of the ball
and when is this height achieved? When does the ball hit (a) What it looks like
the ground? How high is the cliff? physically.
Solution. The function y(t) is a quadratic function with

a negative leading coefficient, so its graph in the ts- 50
coordinate system will be a downward opening parabola.

−2 −1 1 2 3 4 5
We use a graphing device to get the picture in Fig-
ure 7.14(b).
−50
The vertex is the highest point on the graph, which −100
can be found by writing y(t) in standard form using

Fact 7.3.1: (b) What it looks like
graphically.
y(t) = −16t2 + 48t + 50
2 Figure 7.14: Different views
3 of the ball’s trajectory.
= −16 t − + 86.
2
The vertex of the graph of y(t) is 32 , 86 , so the maximum

height of the ball above the level ground is 86 feet, occur-

ing at time t = 32 .
The ball hits the ground when its height above the ground is zero;
using the quadratic formula:
y(t) = −16t2 + 48t + 50
= 0 p
−48 ± (48)2 − 4 (−16) (50)
t =
2 · 16
= 3.818 sec or − 0.818 sec
Conclude the ball hits the ground after 3.818 seconds. Finally, the height
of the cliff is the height of the ball zero seconds after release; i.e., y(0) = 50
feet is the height of the cliff.
Here are two items to consider carefully:

!!!
1. The graph of y(t) is NOT the path followed by the ball! Finding the
CAUTION
actual path of the ball is not possible unless additional information
!!!
is given. Can you see why?
2. The function y(t) is defined for all t; however, in the context of the
problem, there is no physical meaning when t < 0.
The next example illustrates how we must be very careful to link the
question being asked with an appropriate function.
Example 7.4.2. A hot air balloon takes off from the edge of a mountain
lake. Impose a coordinate system as pictured in Figure 7.15 and assume
2
that the path of the balloon follows the graph of y = f(x) = − 2500 x2 + 54 x.
The land rises at a constant incline from the lake at the rate of 2 vertical
feet for each 20 horizontal feet. What is the maximum height of the balloon
above lake level? What is the maximum height of the balloon above ground
level? Where does the balloon land on the ground? Where is the balloon
50 feet above the ground?
height above lake (ft) Solution. In the coordinate system indicated, the origin is
balloon
200 balloon path

the takeoff point and the graph of y = f(x) is the path of
the balloon. Since f(x) is a quadratic function with a neg-
100 ative leading coefficient, its graph will be a parabola which
opens downward. The difficulty with this problem is that
lake
500 1000
(ft)
at any instant during the balloon’s flight, the “height of
Figure 7.15: Visualizing. the balloon above the ground” and the “height of the bal-
loon above the lake level” are different! The picture in
Figure 7.16 highlights this difference; consequently, two different func-
tions will be needed to study these two different quantities.
height of balloon above lake level

200
A A
100
height of balloon above ground level
B B height of ground above lake level

lake
500 1000
Figure 7.16: The height of the balloon y as a function of x.
The function y = f(x) keeps track of the height of the balloon above
lake level at a given x location on the horizontal axis. The line ℓ with
2 1
slope m = 20 = 10 passing through the origin models the ground level.
This says that the function
1
y = x
10
keeps track of the height of the ground above lake level at a given x
location on the horizontal axis.
We can determine the maximum height of the balloon above lake level
by analyzing the parabolic graph of y = f(x). Putting f(x) in vertex form,
via Fact 7.3.1,

2
f(x) = − (x − 500)2 + 200.
2500
The vertex of the graph of y = f(x) is (500, 200). This just tells us that the
maximum height of the balloon above lake level is 200 feet. To find the
landing point, we need to solve the system of equations
2

y = − 2500 x2 + 45 x
1 .
y = 10 x
As usual, plugging the second equation into the first and solving for x,
we get
1 2 2 4
x = − x + x
10 2500 5
x2 = 875x
x2 − 875x = 0
x(x − 875) = 0
height above lake level (feet)
vertex (high point above lake)
200
100
landing point
horizontal distance
from launch (feet)
takeoff point 500 1000
Figure 7.17: Locating the takeoff and landing points.
From the algebra, we see there are two solutions: x = 0 or x = 875;

these correspond to the takeoff and landing points of the balloon, which
are the two places the flight path and ground coincide. (Notice, if we had
divided out x from the last equation, we would only get one solution; the
tricky point is that we can’t divide by zero!) The balloon lands at the
position where x = 875 and to find the y coordinate of this landing point
we plug x = 875 into our function for the balloon height above lake level:
y = f(875) = 87.5 feet. So, the landing point has coordinates (875,87.5).
Next, we want to study the height of the balloon above the ground. Let
y = g(x) be the function which represents the height of the balloon above
the ground when the horizontal coordinate is x. We find

height of the balloon elevation of ground
g(x) = above lake level with − above lake level with
horizontal coordinate x horizontal coordinate x
= f(x) − g(x)

2 2 4 1
= − x + x −
|{z} x
2500 5 10
| {z } minus | {z }
(balloon above lake level) (ground above lake level)
2
= − (x − 437.5)2 + 153.12,
2500
Notice that g(x) itself is a NEW quadratic function with a negative leading
coefficient, so the graph of y = g(x) will be a downward opening parabola.
The vertex of this parabola will be (437.5, 153.12), so the highest elevation
of the balloon above the ground is 153.12 feet.
We can now sketch the graph of g(x) and the horizontal line deter-
mined by y = 50 in a common coordinate system, as below. Finding
where the balloon is 50 feet above the ground amounts to finding where
these two graphs intersect. We need to now solve the system of equations
2 7

y = − 2500 x2 + 10
x
.
y = 50
Plug the second equation into the first and apply the quadratic for-
mula to get x = 796.54 or 78.46. This tells us the two possible x coor-
dinates when the balloon is 50 feet above the ground. In terms of the
original coordinate system imposed, the two places where the balloon is
50 feet above the ground are (78.46, 57.85) and (796.54, 129.6).
7.4.1 How many points determine a parabola?

We all recall from elementary geometry that two distinct points in the
plane will uniquely determine a line; in fact, we used this to derive equa-
tions for lines in the plane. We could then ask if there is a similar char-
acterization of parabolas.
Important Fact 7.4.3. Let P = (x1, y1), Q = (x2, y2) and R = (x3, y3) be
three distinct non-collinear points in the plane such that the x-coordinates
are all different. Then there exists a unique standard parabola passing
vertex (high point above ground)
graph of height above ground function
100
line y = 50
500 1000
places where 50 feet above ground
Figure 7.18: Finding heights above the ground.
through these three points. This parabola is the graph of a quadratic func-
tion y = f(x) = ax2 + bx + c and we can find these coefficients by simultane-
ously solving the system of three equations and three unknowns obtained
by assuming P, Q and R are points on the graph of y = f(x):
 2


 ax 1 + bx1 + c = y 1 


 

 
2
ax2 + bx2 + c = y2 .

 


 

 
ax23 + bx3 + c = y3.
Example 7.4.4. Assume the value of a particular house in Seattle has

increased in value according to a quadratic function y = v(x), where the
units of y are in dollars and x represents the number of years the property
has been owned. Suppose the house was purchased on January 1, 1970
and valued at $50,000. In 1980, the value of the house on January 1
was $80,000. Finally, on January 1, 1990 the value was $200,000. Find
the value function v(x), determine the value on January 1, 1996 and find
when the house will be valued at $1,000,000.
Solution. The goal is to explicitly find the value of the function y = v(x).
We are going to work in a xy-coordinate system in which the first co-
ordinate of any point represents time and the second coordinate repre-
sents value. We need to decide what kind of units will be used. The
x-variable, which represents time, will denote the number of years the
house is owned. For the y-variable, which represents value, we could
use dollars. But, instead, we will follow a typical practice in real estate
and use the units of K, where K = $1,000. For example, a house valued
at $235,600 would be worth 235.6 K. These will be the units we use, which
essentially saves us from drowning in a sea of zeros!
We are given three pieces of information about the value of a particular
house. This leads to three points in our coordinate system: P = (0, 50),
Q = (10, 80) and R = (20, 200). If we plot these points, they do not lie
on a common line, so we know there is a unique quadratic function
v(x) = ax2 + bx + c whose graph (which will be a parabola) passes through
these three points. In order to find the coefficients a, b, and c, we need
to solve the system of equations:
 
 a02 + b0 + c = 50 
a(10)2 + b(10) + c = 80 ,
 2 
a(20) + b(20) + c = 200
which is equivalent to the system

 
 c = 50 
100a + 10b + c = 80 .
 
400a + 20b + c = 200
Plugging c = 50 into the second two equations gives the system

100a + 10b = 30
(7.1)
400a + 20b = 150
30−10b
Solve the first equation for a, obtaining a = 100
, then plug this into
the second equation to get:

30 − 10b
400 + 20b = 150
100
120 − 40b + 20b = 150
3
b=− .
2
Now, plug b = − 23 into the first equation of Equation 7.1 to get 100a +
3 9

10 − 2 = 30; i.e., a = 20 . We conclude that
9 2 3
y = v(x) = x − x + 50,
20 2
keeping in mind the units here are K.
To find the value of the house on January 1, 1996, we simply note this
9
is after x = 26 years of ownership. Plugging in, we get y = v(26) = 20 (26)2 −
3
2
26 + 50 = 315.2; i.e., the value of the house is $315,200. To find when
7.5. WHAT’S NEEDED TO BUILD A QUADRATIC MODEL? 103
the house will be worth $1,000,000, we note that $1,000,000 = 1,000 K and
need to solve the equation
9 2 3
1000 = v(x) = x − x + 50
20 2
9 2 3
0 = x − x − 950.
20 2
By the quadratic formula,
q
3 3 2 9

2
± − 2
− 4 (−950)
x = 9
20
2 20
√
1.5 ± 1712.25
=
0.9
= 47.64 or − 44.31.
Because x represents time, we can ignore the negative solution and so

the value of the house will be $1,000,000 after approximately 47.64 years
of ownership.
7.5 What’s Needed to Build a Quadratic Model?

Back in Fact 4.7.1 on page 46, we highlighted the information required to
determine a linear model. We now describe the quadratic model analog.
Important Facts 7.5.1. A quadratic model is completely determined by:
1. Three distinct non-collinear points, or
2. The vertex and one other point on the graph.
The first approach is just Fact 7.4.3. The second approach is based
on the vertex form of a quadratic function. The idea is that we know any
quadratic function f(x) has the form
f(x) = a(x − h)2 + k,
where (h, k) is the vertex. If we are given h and k, together with another
point (x0, y0) on the graph, then plugging in gives this equation:
y0 = a(x0 − h)2 + k.
The only unknown in this equation is a, which we can solve for using
algebra. A couple of the exercises will depend upon these observations.
7.6 Summary
• A quadratic function is one of the form
f(x) = ax2 + bx + c.
where a 6= 0.
• The graph of a quadratic function is a parabola which is symmetric

about the vertical line through the highest (or lowest) point on the
graph. This highest (or lowest) point is known as the vertex of the
graph; its location is given by (h,k) where
b
h=− and k = f(h).
2a
• If a > 0, then the vertex is the lowest (or minimum) point on the
graph, and the parabola ”opens upward”. If a < 0, then the vertex
is the highest (or maximum) point on the graph, and the parabola
”opens downward”.
• Every quadratic function can be expressed in the form
f(x) = a(x − h)2 + k
where (h,k) is the vertex of the function’s graph.

7.7. EXERCISES 105
7.7 Exercises
Problem 7.1. Write the following quadratic Problem 7.5. A hot air balloon takes off from
functions in vertex form, find the vertex, the the edge of a plateau. Impose a coordinate sys-
axis of symmetry and sketch a rough graph. tem as pictured below and assume that the
path the balloon follows is the graph of the
(a) f(x) = 2x2 − 16x + 41. 4
quadratic function y = f(x) = − 2500 x2 + 54 x.
(b) f(x) = 3x2 − 15x − 77. The land drops at a constant incline from the
plateau at the rate of 1 vertical foot for each
(c) f(x) = x2 − 73 x + 13.
5 horizontal feet. Answer the following ques-
(d) f(x) = 2x2 . tions:
1 2 height above plateau (feet)
(e) f(x) = 100 x .
balloon
Problem 7.2. (a) Sketch the graph of the

function f(x) = x2 − 3x + 4 on the interval
−3 ≤ x ≤ 5. What is the maximum value
takeoff
of f(x) on that interval? What is the min- horizontal distance
imum value of f(x) on that interval? from launch (feet)
ground incline
(b) Sketch the graph of the function f(x) =
x2 − 3x + 4 on the interval 2 ≤ x ≤ 7. What
is the maximum value of f(x) on that in-
terval? What is the minimum value of
f(x) on that interval? (a) What is the maximum height of the bal-
loon above plateau level?
(c) Sketch the graph of the function g(x) =
−(x + 3)2 + 3 on the interval 0 ≤ x ≤ 4. (b) What is the maximum height of the bal-
What is the maximum value of g(x) on loon above ground level?
that interval? What is the minimum (c) Where does the balloon land on the
value of g(x) on that interval? ground?
(d) Where is the balloon 50 feet above the
Problem 7.3. The initial price of buzz.com
ground?
stock is $10 per share. After 20 days the stock
price is $20 per share and after 40 days the
price is $25 per share. Assume the price of Problem 7.6. (a) Suppose f(x) = 3x2 − 2.
the stock is modeled by a quadratic function. Does the point (1,2) lie on the graph of
(a) Find the quadratic function s(t) giving y = f(x)? Why or why not?
the stock price after t days. If you buy (b) If b is a constant, where does the line
1000 shares after 30 days, what is the y = 1 + 2b intersect the graph of y =
cost? x2 + bx + b?
(b) To maximize profit, when should you sell (c) If a is a constant, where does the
shares? How much will the profit be on line y = 1 − a2 intersect the graph of
your 1000 shares purchased in (a)? y = x2 − 2ax + 1?
(c) When will the stock become worthless? (d) Where does the graph of y = −2x2 + 3x + 10
intersect the graph of y = x2 + x − 10?
Problem 7.4. Sketch the graph of
y = x2 − 2x − 3. Label the coordinates of the Problem 7.7. Sylvia has an apple orchard.
x and y intercepts of the graph. In the One season, her 100 trees yielded 140 apples
same coordinate system, sketch the graph of per tree. She wants to increase her production
y = |x2 − 2x − 3|, give the multipart rule and by adding more trees to the orchard. However,
label the x and y intercepts of the graph. she knows that for every 10 additional trees
she plants, she will lose 4 apples per tree (i.e.,
the yield per tree will decrease by 4 apples). (a) Write an expression for the distance be-
How many trees should she have in the or- tween Sven and Rudyard t seconds after
chard to maximize her production of apples? they start walking.
Problem 7.8. Rosalie is organizing a circus (b) When are Sven and Rudyard closest?
performance to raise money for a charity. She What is the minimum distance between
is trying to decide how much to charge for tick- them?
ets. From past experience, she knows that the
number of people who will attend is a linear
function of the price per ticket. If she charges Problem 7.14. After a vigorous soccer match,
5 dollars, 1200 people will attend. If she Tina and Michael decide to have a glass of
charges 7 dollars, 970 people will attend. How their favorite refreshment. They each run in
much should she charge per ticket to make the a straight line along the indicated paths at a
most money? speed of 10 ft/sec. Parametrize the motion of
Tina and Michael individually. Find when and
Problem 7.9. A Norman window is a rectan- where Tina and Michael are closest to one an-
gle with a semicircle on top. Suppose that the other; also compute this minimum distance.
perimeter of a particular Norman window is to
be 24 feet. What should its dimensions be in (200,300)
(−50,275)
order to maximize the area of the window and, soy milk
therefore, allow in as much light as possible?
beet juice
Problem 7.10. Jun has 300 meters of fenc-
ing to make a rectangular enclosure. She also
wants to use some fencing to split the enclo- (400,50)
sure into two parts with a fence parallel to two
Tina
of the sides. What dimensions should the en-
closure have to have the maximum possible Michael
area?
Problem 7.11. Steve likes to entertain friends

at parties with “wire tricks.” Suppose he takes Problem 7.15. If the graph of the quadratic
a piece of wire 60 inches long and cuts it into function f(x) = x2 + dx + 3d has its vertex on
two pieces. Steve takes the first piece of wire the x-axis, what are the possible values of d?
and bends it into the shape of a perfect circle. What if f(x) = x2 + 3dx − d2 + 1 ?
He then proceeds to bend the second piece of
wire into the shape of a perfect square. Where Problem 7.16. For each of the following equa-
should Steve cut the wire so that the total area tions, find the value(s) of the constant α so that
of the circle and square combined is as small the equation has exactly one solution, and de-
as possible? What is this minimal area? What termine the solution for each value.
should Steve do if he wants the combined area
to be as large as possible? (a) αx2 + x + 1 = 0
Problem 7.12. In each case, find a quadratic (b) x2 + αx + 1 = 0

function whose graph passes through the (c) x2 + x + α = 0
given points:
(d) x2 + αx + 4α + 1 = 0
(a) (0,0), (1,1) and (3, − 1).
(b) (−1,1), (1, − 2) and (3,4).
Problem 7.17. (a) Solve for t
(c) (2,1), (3,2) and (5,1).
(d) (0,1), (1,1) and (1,3). s = 2(t − 1)2 + 1
Problem 7.13. Sven starts walking due south (b) Solve for x
at 5 feet per second from a point 120 feet north
of an intersection. At the same time Rudyard y = x2 + 2x + 3
starts walking due east at 4 feet per second
from a point 150 feet west of the intersection.
Chapter 8
Composition
A new home takes its shape from basic building materials and the skill-
ful use of construction tools. Likewise, we can build new functions
from known functions through the application of analogous mathemati-
cal tools. There are five tools we want to develop: composition, reflection,
shifting, dilation, arithmetic. We will handle composition in this section,
then discuss the others in the following two sections.
To set the stage, let’s look at a simple botany experi- oxygen rate hr 1
ment. Imagine a plant growing under a particular steady 1

light source. Plants continually give off oxygen gas to 0.8
0.6
the environment at some rate; common units would be
0.4
liters/hour. If we leave this plant unbothered, we mea- 0.2
sure that the plant puts out 1 liter/hour; so, the oxygen hours
output is a steady constant rate. However, if we apply a 2 4 6 8 10
flash of high intensity green light at the time t = 1 and (a) Flash at t = 1.
measure the oxygen output of the plant, we obtain the
plot in Figure 8.1(a). oxygen rate hr 1
Using what we know from the previous section on 1

quadratic functions, we can check that a reasonable 0.8
0.6
model for the graph is this multipart function f(t) (on the 0.4
domain 0 ≤ t ≤ 10): 0.2 hours
 2 4 6 8 10
 1 if t ≤ 1 (b) Flash at t = 5.
2 2 8
f(t) = t − 3 t + 3 if 1 ≤ t ≤ 3
 3
1 if 3 ≤ t Figure 8.1: Light flashes.
Suppose we want to model the oxygen consumption

when a green light pulse occurs at time t = 5 (instead of
time t = 1), what is the mathematical model? For starters,
it is pretty easy to believe that the graph for this new sit-
uation will look like the new graph in Figure 8.1(b).
But, can we somehow use the model f(t) in hand (the known function)
to build the model we want (the new function)? We will return in Exam-
107
108 CHAPTER 8. COMPOSITION
ple 8.2.4 to see the answer is yes; first, we need to develop the tool of
function composition.
8.1 The Formula for a Composition

The basic idea is to start with two functions f and g and produce a new
function called their composition. There are two basic steps in this pro-
cess and we are going to focus on each separately. The first step is fairly
mechanical, though perhaps somewhat unnatural. It involves combining
the formulas for the functions f and g together to get a new formula;
we will focus on that step in this subsection. The next step is of varying
complexity and involves analyzing how the domains and ranges of f and g
affect those of the composition; we defer that to the next subsection once
we have the mechanics down.
in out in out
u x y
the “g” function x = g(u) the “f” function y = f(x)
in out
“composed” function y = f(g(u))
Figure 8.2: Visualizing a composite function y = h(u) = f(g(u)).
Here is a very common occurrence: We are handed a function y = f(x),

which means given an x value, the rule f(x) produces a new y value. In
addition, it may happen that the variable x is itself related to a third
variable u by some different function equation x = g(u). Given u, the
rule g(u) will produce a value of x; from this x we can use the rule f(x)
to produce a y value. In other words, we can regard y as a function
depending on the new independent variable u. It is important to know
the mechanics of working with this kind of setup. Abstractly, we have
just described a situation where we take two functions and build a new
8.1. THE FORMULA FOR A COMPOSITION 109
function which “composes” the original ones together; schematically the

situation looks like this:
Example 8.1.1. A pebble is tossed into a pond. The radius of the first cir-
cular ripple is measured to increase at the constant rate of 2.3 ft/sec. What
is the area enclosed by the leading ripple after 6 seconds have elapsed?
How much time must elapse so that the area enclosed by the leading ripple
is 300 square feet?
Solution. We know that an object tossed into a pond leading ripple after t seconds
will generate a series of concentric ripples, which grow leading ripple after 2 seconds
steadily larger. We are asked questions that relate the
area of the circular region bounded by the leading ripple
and time elapsed. r = r(t)
Let r denote the radius of the leading ripple after t sec- leading ripple after 1 second
onds; units of feet. The area A of a disc bounded by a Figure 8.3: Concentric rip-
2
leading ripple will be A = A(r) = πr . This exhibits A as a ples.
function in the variable r. However, the radius is changing with respect
to time:
feet

r = r(t) = radius after t seconds = 2.3 t seconds = 2.3t feet.
sec
So, r = r(t) is a function of t. In the expression A = A(r), replace “r” by
“r(t),” then
A = π(2.3t)2 = 5.29πt2.
The new function a(t) = 5.29πt2 gives a precise relationship between area
and time.
To answer our first question, a(6) = 598.3 feet2 is the area of the region
bounded by the leading ripple afterp6 seconds. On the other hand, if
a(t) = 300 ft2;, 300 = 5.29πt2, so t = ± 300/(5.29π) = ±4.25. Since t repre-
sents time, only the positive solution t = 4.25 seconds makes sense.
We can formalize the key idea used in solving this problem, which is
familiar from previous courses. Suppose that
y = f(x)
and that additionally the independent variable x is itself a function of a
different independent variable t; i.e.,
x = g(t).
Then we can replace every occurrence of “x” in f(x) by the expression
“g(t),” thereby obtaining y as a function in the independent variable t.
We usually denote this new function of t:
y = f(g(t)).
We refer to f(g(t)) as the composition of f and g or the composite function.

The process of forming the composition of two functions is a mechan-
ical procedure. If you are handed the actual formulas for y = f(x) and
x = g(t), then Procedure 8.1.2 is what you need.
Important Procedure 8.1.2. To obtain the formula for f(g(t)), replace ev-
ery occurrence of “x” in f(x) by the expression “g(t).”
Here are some examples of how to do this:
Examples 8.1.3. Use the composition procedure in each of these cases.
(i) If y = f(x) = 2 and x = g(t) = 2t, then f(g(t)) = f(2t) = 2.
(ii) If y = 3x − 7 and x = g(t) = 4, then f(g(t)) = f(4) = 3 · 4 − 7 = 5.
(iii) If y = f(x) = x2 + 1 and x = g(t) = 2t − 1, then
f(g(t)) = f(2t − 1)
= (2t − 1)2 + 1
= 4t2 − 4t + 2.
p
(iv) If y = f(x) = 2 + 1 + (x − 3)2 and x = g(t) = 2t2 − 1, then
f(g(t)) = f(2t2 − 1)
p
= 2 + 1 + (2t2 − 1 − 3)2
p
= 2 + 4t4 − 16t2 + 17.
(v) If y = f(x) = x2 and x = g(t) = t + ♥, then
f(g(t)) = f(t + ♥)
= (t + ♥)2
= t2 + 2t♥ + ♥2.
It is natural to ask: What good is this whole business about composi-

tions? One way to think of it is that we can use composite functions to
break complicated functions into simpler parts. For example,
p
y = h(x) = x2 + 1
√
can be written as the composition f(g(x)), where y = f(z) = z and z =
g(x) = x2+1. Each of the functions f and g is “simpler” than the original h,
which can help when studying h.
Examples 8.1.4. Here we use composite functions to “simplify” a given
function.
8.1. THE FORMULA FOR A COMPOSITION 111
(i) The function y = (x−3)1 2 +2 can be written as a composition y = f(g(x)),

where y = f(z) = z21+2 and z = g(x) = x − 3.
(ii) The upper semicircle

p of radius 2 centered at (1,2) is the graph of the
function y = 2 + 4 − (x − 1)2. This function
√ can be written as a com-
position y = f(g(x)), where y = f(z) = 2 + 4 − z2 and z = g(x) = x − 1.
8.1.1 Some notational confusion

In our discussion above, we have used different letters to represent the
domain variables of two functions we are composing. Typically, we have
been writing: If y = f(x) and x = g(t), then y = f(g(t)) is the composition.
This illustrates that the three variables t, x, and y can all be of different
types. For example, t might represent time, x could be speed and y could
be distance.
If we are given two functions that involve the same independent vari-
able, like f(x) = x2 and g(x) = 2x + 1, then we can still form a new function
f(g(x)) by following the same prescription as in Procedure 8.1.2:
Important Procedure 8.1.5. To obtain the formula for f(g(x)), replace

every occurrence of “x” in f(x) by the expression “g(x).”
For our example, this gives us:
f(g(x)) = f(2x + 1) = (2x + 1)2.
Here are three other examples:

√ √
• If f(x) = x, g(x) = 2x2 + 1, then f(g(x)) = 2x2 + 1.
• If f(x) = x1 , g(x) = 2x + 1, then f(g(x)) = 1

2x+1
.
• If f(x) = x2, g(x) = △ − x, then f(g(x)) = △2 − 2x△ + x2.

y-axis
10 Example 8.1.6. Let f(x) = x2, g(x) = x + 1 and h(x) = x − 1.
8 Find the formulas for f(g(x)), g(f(x)), f(h(x)) and h(f(x)).
6 Discuss the relationship between the graphs of these four
4
functions.
2
x-axis
Solution. If we apply Procedure 8.1.5, we obtain the com-
−3 −2 −1 1 2 3
position formulas. The four graphs are given on the do-
Figure 8.4: Sketching com- main −3 ≤ x ≤ 3, together with the graph of f(x) = x2.
posite functions.
f(g(x)) = f(x + 1) = (x + 1)2
g(f(x)) = g(x2) = x2 + 1
f(h(x)) = f(x − 1) = (x − 1)2
h(f(x)) = h(x2) = x2 − 1.
We can identify each graph by looking at its vertex:
• f(x) has vertex (0,0)
• f(g(x)) has vertex (-1,0)
• g(f(x)) has vertex (0,1)
• f(h(x)) has vertex (1,0)
• h(f(x)) has vertex (0,-1)
Horizontal or vertical shifting of the graph of f(x) = x2 gives the other four
graphs: See Figure 8.4.
8.2 Domain, Range, etc. for a Composition

A function is a “package” consisting of a rule, a domain of allowed input
values, and a range of output values. When we start to compose func-
tions, we sometimes need to worry about how the domains and ranges
of the composing functions affect the composed function. First off, when
you form the composition f(g(x)) of f(x) and g(x), the range values for
g(x) must lie within the domain values for f(x). This may require that
you modify the range values of g(x) by changing its domain. The domain
values for f(g(x)) will be the domain values for g(x).
8.2. DOMAIN, RANGE, ETC. FOR A COMPOSITION 113
-axis
in out in out
the “g” function the “f” function

domain g range of f(g(x))
range of g
domain of f
Figure 8.5: What is the domain and range of a composite function?
In practical terms, here is how one deals with the domain issues for a
composition. This is a refinement of Procedure 8.1.5 on page 111.
Important Procedure 8.2.1. To obtain the formula for f(g(x)), replace

every occurrence of “x” in f(x) by the expression “g(x).” In addition, if
there is a condition on the domain of f that involves x, then replace every
occurrence of “x” in that condition by the expression “g(x).”
The next example illustrates how to use this principle.
Example 8.2.2. Start with the function y = f(x) = x2 on the domain

−1 ≤ x ≤ 1. Find the rule and domain of y = f(g(x)), where g(x) = x − 1.
Solution. We can apply the first statement in Procedure 8.2.1 to find the
rule for y = f(g(x)):
y = f(g(x))
= f(x − 1) = (x − 1)2
= x2 − 2x + 1.
To find the domain of y = f(g(x)), we apply the second statement in

Procedure 8.2.1; this will require that we solve an inequality equation:
−1 ≤ g(x) ≤ 1
−1 ≤ x − 1 ≤ 1
0≤ x ≤2
The conclusion is that y = f(g(x)) = x2 −2x+1 on the domain 0 ≤ x ≤ 2.

√
Example 8.2.3. Let y = f(z) = z, z = g(x) = x + 1. What is the largest
possible domain so that the composition f(g(x)) makes sense?
y
z z = g(x) = x + 1
y = f(z)
range
desired range
z
−1
domain x
required domain
√
(a) y = z. (b) z = x = 1.
Figure 8.6: Finding the largest domain for f(g(x)).
Solution. The largest possible domain for y = f(z) will consist of all non-
negative real numbers; this is also the range of the function f(z): See
Figure 8.6(a).
To find the largest domain for the composition, we try to find a domain
of x-values so that the range of z = g(x) is the domain of y = f(z). So, in
this case, we want the range of g(x) to be all non-negative real numbers,
denoted 0 ≤ z. We graph z = g(x) in the xz-plane, mark the desired range
0 ≤ z on the vertical z-axis, then determine which x-values would lead to
points on the graph with second coordinates in this zone. We find that
the domain of all x-values greater or equal to −1 (denoted −1 ≤ x)√leads
to the desired range. In summary, the composition y = f(g(x)) = x + 1
is defined on the domain of x-values −1 ≤ x.
Let’s return to the botany experiment that opened this section and see
how composition of functions can be applied to the situation. Recall that
plants continually give off oxygen gas to the environment at some rate;
common units would be liters/hour.
Example 8.2.4. A plant is growing under a particular steady light source.

If we apply a flash of high intensity green light at the time t = 1 and
measure the oxygen output of the plant, we obtain the plot below and the
mathematical model f(t).

 1 if t ≤ 1
2 2 8
f(t) = t − 3 t + 3 if 1 ≤ t ≤ 3
 3
1 if 3 ≤ t
Now, suppose instead we apply the flash of high intensity green light at
the time t = 5. Verify that the mathematical model for this experiment is
given by f(g(t)), where g(t) = t − 4.
8.2. DOMAIN, RANGE, ETC. FOR A COMPOSITION 115
Solution. Our expectation is that the plot for this new ex- oxygen rate 1
hr
periment will have the “parabolic dip” shifted over to oc- 1
cur starting at time t = 5 instead of at time t = 1. In other 0.8
0.6
words, we expect the graph in Figure 8.7(b).
0.4
Our job is to verify that this graph is obtained from the
0.2
function f(g(t)), where g(t) = t − 4. This is a new terrain hours
for us, since we need to look at a composition involving a 2 4 6 8 10

multipart function. Here is how to proceed: When we are
(a) Flash at t = 1.
calculating a composition involving a multipart function,
we need to look at each of the parts separately, so there oxygen rate 1
hr
will be three cases to consider: 1
First part: f(t) = 1 when t ≤ 1. To get the formula for 0.8
0.6
f(g(t)), we now appeal to Procedure 8.2.1 and just replace
0.4
every occurrence of t in f(t) by g(t). That gives us this
0.2
NEW domain condition and function equation: hours
2 4 6 8 10
f(g(t)) = f(t − 4) = 1 when t − 4 ≤ 1 (b) Flash at t = 5.
= 1 when t ≤ 5. Figure 8.7: Applying light at

2 2 8 time t.
Second part: f(t) = 3
t− 3
+ 3 when 1 ≤ t ≤ 3. We
t
now appeal to Procedure 8.2.1 and just replace every oc-
currence of t in this function by g(t). That gives us this NEW domain
condition and function equation:
2 8
f(g(t)) = f(t − 4) = (t − 4)2 − (t − 4) + 3 when 1 ≤ t − 4 ≤ 3
3 3
2 2 73
= t − 8t + when 5 ≤ t ≤ 7.
3 3
Third part: f(t) = 1 when 3 ≤ t. We now appeal to Procedure 8.2.1 and
just replace every occurrence of t in this function by g(t). That gives us
this NEW domain condition and function equation:
f(g(t)) = f(t − 4) = 1 when 3 ≤ t − 4

= 1 when 7 ≤ t.
The multipart rule for this composition can now be written down and
using a graphing device you can verify its graph is the model for our
experiment.

 1 if t ≤ 5
2 2 73
f(g(t)) = t − 8t + if 5 ≤ t ≤ 7
 3 3
1 if 7 ≤ t
8.3 Exercises
Problem 8.1. For this problem, f(t) = t − 1, Problem 8.5. A car leaves Seattle heading
g(t) = −t − 1 and h(t) = |t|. east. The speed of the car in mph after m min-
utes is given by the function
(a) Compute the multipart rules for h(f(t))
and h(g(t)) and sketch their graphs. 70m2
C(m) = .
(b) Compute the multipart rules for f(h(t)) 10 + m2
and g(h(t)) and sketch their graphs. (a) Find a function m = f(s) that converts
(c) Compute the multipart rule for h(h(t)−1) seconds s into minutes m. Write out
and sketch the graph. the formula for the new function C(f(s));
what does this function calculate?
(b) Find a function m = g(h) that converts
Problem 8.2. Write each of the following func-
hours h into minutes m. Write out the
tions as a composition of two simpler func-
formula for the new function C(g(h));
tions: (There is more than one correct answer.)
what does this function calculate?
(a) y = (x − 11)5 . (c) Find a function z = v(s) that converts
√
(b) y = 3 1 + x2 . mph s into ft/sec z. Write out the for-
mula for the new function v(C(m); what
(c) y = 2(x − 3)5 − 5(x − 3)2 + 12 (x − 3) + 11. does this function calculate?
1
(d) y = x2 +3
.
p√ Problem 8.6. Compute the compositions
(e) y = x + 1.
p f(g(x)), f(f(x)) and g(f(x)) in each case:
(f) y = 2 − 5 − (3x − 1)2 . (a) f(x) = x2 , g(x) = x + 3.
√
(b) f(x) = 1/x, g(x) = x.
Problem 8.3. (a) Let f(x) be a linear func-
(c) f(x) = 9x + 2, g(x) = 91 (x − 2).
tion, f(x) = ax + b for constants a and b.
Show that f(f(x)) is a linear function. (d) f(x) = 6x2 + 5, g(x) = x − 4.
√
(b) Find a function g(x) such that g(g(x)) = (e) f(x) = 4x3 − 3, g(x) = 3 2x + 6
6x − 8. (f) f(x) = 2x + 1, g(x) = x3 .
(g) f(x) = 3, g(x) = 4x2 + 2x + 1.
Problem 8.4. Let f(x) = 21 x + 3.
(h) f(x) = −4, g(x) = 0.
(a) Sketch the graphs of f(x),f(f(x)),f(f(f(x)))
on the interval −2 ≤ x ≤ 10. √
Problem 8.7. Let y = f(z) = 4 − z2 and
(b) Your graphs should all intersect at the z = g(x) = 2x + 3. Compute the composition
point (6,6). The value x = 6 is called y = f(g(x)). Find the largest possible domain
a fixed point of the function f(x) since of x-values so that the composition y = f(g(x))
f(6) = 6; that is, 6 is fixed - it doesn’t is defined.
move when f is applied to it. Give an ex-
planation for why 6 is a fixed point for Problem 8.8. Suppose you have a function
any function f(f(f(...f(x)...))). y = f(x) such that the domain of f(x) is
(c) Linear functions (with the exception of 1 ≤ x ≤ 6 and the range of f(x) is −3 ≤ y ≤ 5.
f(x) = x) can have at most one fixed (a) What is the domain of f(2(x − 3)) ?
point. Quadratic functions can have at
(b) What is the range of f(2(x − 3)) ?
most two. Find the fixed points of the
function g(x) = x2 − 2. (c) What is the domain of 2f(x) − 3 ?
(d) Give a quadratic function whose fixed (d) What is the range of 2f(x) − 3 ?
points are x = −2 and x = 3. (e) Can you find constants B and C so that
the domain of f(B(x − C)) is 8 ≤ x ≤ 9?
8.3. EXERCISES 117
(f) Can you find constants A and D so that this simplified expression:
the range of Af(x) + D is 0 ≤ y ≤ 1?
f(x + h) − f(x)
.
h
1
(a) f(x) = x−1 .
Problem 8.9. For each of the given functions
y = f(x), simplify the following expression so (b) f(x) = (2x + 1)2 .
√
that h is no longer a factor in the denomina- (c) f(x) = 25 − x2 .
tor, then calculate the result of setting h = 0 in
Chapter 9
Inverse Functions
The experimental sciences are loaded with examples of functions relat-

ing time and some measured quantity. In this case, time represents our
“input” and the quantity we are measuring is the “output.” For example,
maybe you have just mixed together some chemical reactants in a ves-
sel. As time goes by, you measure the fraction of reactants remaining,
tabulate your results, then sketch a graph as indicated in Figure 9.1.
Viewing the input value as “time” and the output value fraction
as “fraction of product,” we could find a function y = f(t) 1
modeling this data. Using this function, you can easily
compute the fraction of reactants remaining at any time
in the future. However, it is probably just as interesting
to know how to predict the time when a given fraction 0 time
of reactants exists. In other words, we would like a new
Figure 9.1: Fraction of reac-
function that allows us to input a “fraction of reactants” tants as a function of time.
and get out the “time” when this occurs. This “reverses”
the input/output roles in the original function. Is there a systematic way
to find the new function if we know y = f(t)? The answer is yes and
depends upon the general theory of inverse functions.
9.1 Concept of an Inverse Function

Suppose you are asked to solve the following three equations for x. How
do you proceed?
(x + 2) = 64
(x + 2)2 = 64
(x + 2)3 = 64.
In the first equation, you add “−2” to each side, then obtain x = 62. In
the third equation, you take the cube root of both sides of the equation,
giving you x + 2 = 4, then subtract 2 getting x = 2. In second equation,
you take a square root of both sides, BUT you need to remember both the
119
120 CHAPTER 9. INVERSE FUNCTIONS
positive and negative results when doing this. So, you are reduced down
to x + 2 = ±8 or that x = −10 or 6. Why is it that in two of these cases you
obtain a single solution, while in the remaining case there are two dif-
ferent answers? We need to sort this out, since the underlying ideas will
surface when we address the inverse circular functions in Chapter 20.
Let’s recall the conceptual idea of a function: A func-
In Out tion is a process which takes a number x and outputs a
x f(x)
new number f(x). So far, we’ve only worked with this pro-
cess from “left to right;” i.e., given x, we simply put it into
The function
a symbolic rule and out pops a new number f(x). This is
Figure 9.2: A function as a all pretty mechanical and straightforward.
process.
9.1.1 An Example
Let’s schematically interpret what happens for the specific concrete ex-
ample y = f(x) = 3x − 1, when x = −1, − 21 , 0, 21 , 1, 2: See Figure 9.3.
in out in out
1 1
−1 3x − 1 −4 2
3x − 1 2
in out in out
− 21 3x − 1 − 25 1 3x − 1 2
in out in out
0 3x − 1 −1 2 3x − 1 5
Figure 9.3: Function process y = 3x − 1.
We could try to understand the function process in this example in

“reverse order,” going “right to left;” namely, you might ask what x value
can be run through the process so you end up with the number 11? This
is somewhat like the “Jeopardy” game show: You know what the answer
is, you want to find the question. For our example, if we start out with
some given y values, then we can define a “reverse process” x = 31 (y + 1),
which returns the x value required so that f(x) = y: See Figure 9.4.
9.1. CONCEPT OF AN INVERSE FUNCTION 121
eplacements
(y+1) 1 (y+1) 1
−1 −4
3 2 3 2
(y+1) (y+1)
− 21
3
− 52 1 3
2
(y+1) (y+1)
0 −1 2 5
3 3
Figure 9.4: The reverse process x = 13 (y + 1).
9.1.2 A Second Example

If we begin with a linear function y = f(x) = mx + b, where m 6= 0, then we
can always find a “reverse process” for the function. To find it, you must
solve the equation y = f(x) for x in terms of y:
y = mx + b
y − b = mx
1
(y − b) = x
m
So, if m = 3 and b = −1, we just have the first example above.
For another example, suppose y = −0.8x + 2; then m = −0.8 and b = 2.
In this case, the reverse process is −1.25(y − 2) = x. If we are given the
value y = 11, we simply compute that x = −11.25; i.e., f(−11.25) = 11.
9.1.3 A Third Example

The previous examples hide a subtle point that can arise when we try to
understand the “reverse process” for a given function. Suppose we begin
with the function y = f(x) = (x − 1)2 + 1. Figure 9.5 is a schematic of how
the function works when we plug in x = −1, − 21 , 0, 21 , 1, 2; what is being
illustrated is a “forward process”, in that each input generates a unique
output.
For this example, if we start out with some given y values, then we can
try to define a “reverse process” x = ??? which returns an x value required
so that f(x) = y. Unfortunately, there is no way to obtain a single formula
for this reverse process; Figure 9.6(a) shows what happens if you are
given y = 3 and you try to solve for x.
in out in out
−1
2
(x − 1) + 1 5 1 (x − 1)2 + 1 5
2 4
in out in out
2
− 21 (x − 1) + 1 13
4
1 (x − 1)2 + 1 1
in out in out
0 (x − 1)2 + 1 2 2 (x − 1)2 + 1 2
Figure 9.5: Function process y = f(x) = (x − 1)2 + 1.
The conclusion is that the “reverse process” has two outputs. This
violates the rules required for a function, so this is NOT a function. The
solution is to create two new “reverse processes.”
√
Each of these “reverse processes” has a unique output;
1+ 2 reverse
process
in other words, each of these “reverse processes” defines
x = −
√ a function.
+ y− 1 3
So, given y √= 3, there are √TWO possible x √values,
√
1− 2 namely x = 1 ± 2, so that f(1 + 2) = 3 and f(1 − 2) = 3.
In other words, the reverse process is not given by a single
(a) Reverse process but not a
function. equation; there are TWO POSSIBLE reverse processes.
reverse
process
1+
√
2 x =+
√
y−1 3 9.2 Graphical Idea of an Inverse
We have seen that finding inverses is related to solving
reverse
process equations. However so far, the discussion has been sym-
1−
√
2 x =−
√
y−1 3
bolic; we have pushed around a few equations and in
the end generated some confusion. Let’s use the tools of
Chapter 6 to visualize what is going on here. Suppose we
(b) Two new reverse processes
are given the graph of a function f(x) as in Figure 9.7(a).
that are functions. What input x values result in an output value of 3? This
involves finding all x such that f(x) = 3. Graphically, this
Figure 9.6: What to do if a means we are trying to find points on the graph of f(x)
reverse process is not a func- so that their y-coordinates are 3. The easiest way to to do
tion.
this is to draw the line y = 3 and find where it intersects
the graph.
In Figure 9.7(b) we can see the points of intersection
are (−5, 3), (−1, 3), and (9, 3). That means that x = −5, −1,
9 produce the output value 3; i.e., f(−5) = f(−1) = f(9) = 3.
9.2. GRAPHICAL IDEA OF AN INVERSE 123
y = f(x) y = f(x)
5 5
y=3
−5 5 10 −5 5 10
−5 −5
(a) Given the function y = f(x). (b) What values of x give

f(x) = 3?
Figure 9.7: Using the horizontal line y = 3 to find values on the x-axis.
This leads to our first important fact about the “reverse process” for a
function:
Important Fact 9.2.1. Given a number c, the x values such that f(x) = c
can be found by finding the x-coordinates of the intersection points of the
graphs of y = f(x) and y = c.
Example 9.2.2. Graph y = f(x) = x2 and discuss the meaning of Fact 9.2.1
when c = 3, 1, 6.
Solution. We graph y = x2 and the lines y = 1, y = 3 and y = x2

y = 6. Let’s use c = 6 as an example. We need to simul-
y=6
taneously solve the equations y = x2 and √ y = 6. Putting
2
these together, we get x = 6 or x = ± 6 ≈ ±2.449; i.e., y=3
f(±2.449) = 6. If c = 3, we get x = ±1.732; i.e., f(±1.732) = 3. y=1
Finally, if c = 1, we get x = ±1; i.e., f(±1) = 1.

Figure 9.8: Graph of y =
The pictures so far indicate another very important
f(x) = x2 .
piece of information. For any number c, we can tell ex-
actly “how many” input x values lead to the same output value c, just by
counting the number of times the graphs of y = f(x) and y = c intersect.
Important Fact 9.2.3. For any function f(x) and any number c, the num-
ber of x values so that f(x) = c is the number of times the graphs of y = c
and y = f(x) intersect.
Examples 9.2.4.
(i) If f(x) is a linear function f(x) = mx + b, m 6= 0, then the graph of

f(x) intersects a given horizontal line y = c EXACTLY once; i.e., the
equation c = f(x) always has a unique solution.
(ii) If f(x) = d is a constant function and c = d, then every input x value

in the domain leads to the output value c. On the other hand, if c 6= d,
then no input x value will lead to the output value c. For example, if
f(x) = 1 and c = 1, then every real number can be input to produce
an output of 1; if c = 2, then no input value of x will lead to an output
of 2.
f(x) = mx + b
y-axis y-axis
y=c
f(x) = 1
x-axis x-axis
any of these inputs

the only input which leads to an output of 1
leads to an output of c
linear functions constant functions
Figure 9.9: Does a horizontal line y = c intersect a curve once or more than once?
9.2.1 One-to-one Functions

For a specified domain, one-to-one functions are functions with the prop-
erty: Given any number c, there is at most one input x value in the do-
main so that f(x) = c. Among our examples thus far, linear functions
(degree 1 polynomials) are always one-to-one. However, f(x) = x2 is not
one-to-one; we’ve already seen that it can have two values for some of its
inverses. By Fact 9.2.3, we can quickly come up with what’s called the
horizontal line test.
Important Fact 9.2.5 (Horizontal Line Test). On a given domain of x-values,

if the graph of some function f(x) has the property that every horizontal line
crosses the graph at most only once, then the function is one-to-one on this
domain.
9.3. INVERSE FUNCTIONS 125
Example 9.2.6. By the horizontal line test, it is easy to y-axis

see that f(x) = x3 is one-to-one on the domain of all real
numbers.
Although it isn’t common, it’s quite nice when a func- x-axis horizontal
tion is one-to-one because we don’t need to worry as lines
much about the number of input x values producing the

same output y value. In effect, this is saying that we can
define a “reverse process” for the function y = f(x) which
will also be a function; this is the key theme of the next Figure 9.10: A one-to-one
section. function f(x) = x3 .
9.3 Inverse Functions

Let’s now come face to face with the problem of finding the “reverse pro-
cess” for a given function y = f(x). It is important to keep in mind that
the domain and range of the function will both play an important role
in this whole development. For example, Figure 9.11 shows the function
f(x) = x2 with three different domains specified and the corresponding
range values.
range range
range
domain domain domain
Figure 9.11: Possible domains for a given range.
These comments set the stage for a third important fact. Since the
domain and range of the function and its inverse rule are going to be
intimately related, we want to use notation that will highlight this fact.
We have been using the letters x and y for the domain (input) and range
(output) variables of f(x) and the “reverse process” is going to reverse
these roles. It then seems natural to simply write y (instead of c) for the
input values of the “reverse process” and x for its output values.
Important Fact 9.3.1. Suppose a function f(x) is one-to-one on a domain

of x values. Then define a NEW FUNCTION by the rule
f−1(y) = the x value so that f(x) = y.
The domain of y values for the function f−1(y) is equal to the range of the
function f(x).
The rule defined here is the “reverse process” for the given function. It
is referred to as the inverse function and we read f−1(y) as “...eff inverse
of y...”.
!!!
CAUTION Both the “domain” of f(x) and the “rule” f(x) have equal influence on
!!! whether the inverse rule is a function. Keep in mind, you do NOT get an
inverse function automatically from functions that are not one-to-one!
9.3.1 Schematic Idea of an Inverse Function

Suppose that f(x) is one-to-one, so that f−1(y) is a function. As a result,
we can model f−1(y) as a black box. What does it do? If we put in y in
the input side, we should get out the x such that f(x) = y.
Now, let’s try to unravel something very special that is
in out happening on a symbolic level. What would happen if we
f−1 (y) plugged f(a) into the inverse function for some number a?
y x such that Then the inverse rule f−1(f(a)) tells us that we want to
f(x) = y
find some x so that f(x) = f(a). But, we already know
Figure 9.12: A new function x = a works and since f−1(y) is a function (hence gives us
x = f−1 (y). unique answers), the output of f−1(f(a)) is just a. Sym-
bolically, this means we have Fact 9.3.2.
Important Fact 9.3.2. For every a value in the domain of f(x), we have
f−1(f(a)) = a. (9.1)
This is better shown in the black box picture of Figure 9.13.
in out in out
a f(x) f(a) f−1 (x) a
Figure 9.13: Visualizing f−1 (f(a)) = a.
A good way to get an idea of what an inverse function is doing is to

remember that f−1(y) reverses the process of f(x). We can think of f−1(y)
as a “black box” running f(x) backwards.
9.4. TRYING TO INVERT A NON ONE-TO-ONE FUNCTION 127
9.3.2 Graphing Inverse Functions

How can we get the graph of an inverse function? The idea is to manip-
ulate the graph of our original one-to-one function in some prescribed
way, ending up with the graph of f−1(y). This isn’t as hard as it sounds,
but some confusion with the variables enters into play. Remember that
a typical point on the graph of a function y = f(x) looks like (x, f(x)). Now
let’s take a look at the inverse function x = f−1(y). Given a number y in
the domain of f−1(y), y = f(x) for some x in the domain of f(x); i.e., we are
using the fact that the domain of f−1 equals the range of f. The function
f−1(y) takes the number f(x) and sends it to x, by Fact 9.3.2. So when
f(x) is the input value, x becomes the output value. Conclude a point on
the graph of f−1(y) looks like (f(x), x). It’s similar to the graph of y = f(x),
only the x and y coordinates have reversed! What does that do to the
graph? Essentially, you reorient the picture so that the positive x-axis
and positive y-axis are interchanged. Figure 9.14 shows the process for
√
the function y = f(x) = x3 and its inverse function x = f−1(y) = 3 y. We
place some ∗ symbols on the graph to help keep track of what is happen-
ing.
+y−axis
* +x−axis
*
* +x−axis * +y−axis * +y−axis
rotate 900 *
clockwise
+x−axis flip across *
horiz axis
* *
√
Figure 9.14: Graphically finding x = f−1 (y) = 3 y.
9.4 Trying to Invert a Non one-to-one Func-

tion
√
Suppose we blindly try to show that y is the inverse function for y = x2,
without worrying about all of this one-to-one stuff. We’ll start out with
√
the number −7. If f−1(y) = y, then we know that
f−1(f(−7)) = f−1(49) = 7.
On the other hand, the formula in Fact 9.3.2 tells us that we must have
f−1(f(−7)) = −7,
√
so we have just shown 7 = −7! So clearly f−1(y) 6= y. Even if we try
√
f−1(y) = − y, we produce a contradiction. It seems that if you didn’t have
to worry about negative numbers, things would be all right. Then you
√
could say that f−1(y) = y. Let’s try to see what this means graphically.
Let’s set f(x) = x2, but only for non-negative x-values. That means
that we want to erase the graph to the left of the y-axis (so remember - no
negative x-values allowed). The graph would then look like Figure 9.15.
+y
y = x2
+x
√
inverse function y
+x +y
domain non-negative x domain non-negative y
Figure 9.15: Restricting the domain: No negative x-values.
This is a now a one-to-one function! And now, one can see that its
√
inverse function is y. Similarly, we could have taken f(x) = x2 but only
√
for the non-positive x-values. In that case, f−1(y) = − y. In effect, we
have split the graph of y = x2 into two parts, each of which is the graph
of a one-to-one function; Figure 9.16.
+y
y = x2
+x
domain non-negative y
+x +y
√
domain non-positive x inverse function − y
Figure 9.16: Restricting the domain: No positive x-values.
It is precisely this splitting into two cases that √

leads us to multiple
√
2
solutions of an equation like x = 5. We obtain x = 5 and x = − 5; one
9.5. SUMMARY 129
solution comes from the side of the graph to the left of the y-axis, and
the other from the right of the y-axis. This is because we have separate
inverse functions for the left and right side of the graph of y = x2.
9.5 Summary
• Two functions f and g are inverses if
f(g(x)) = x and g(f(x)) = x
for all x in the domain of f and the domain of g.
• A function f is one-to-one if every equation
f(x) = k
has at most one solution. If there is a value of k such that the

equation f(x) = k has more than one solution, then f is not one-to-
one.
• A function is one-to-one if every horizontal line intersects the func-

tion’s graph at most once.
• A function has an inverse if the function is one-to-one, and every

one-to-one function has an inverse.
• The domain of a function is the range of its inverse, and the range
of a function is the domain of its inverse.
• The graph of a function and its inverse are mirror images of each
other across the line y = x.
9.6 Exercises
2 A B
Problem 9.1. Let f(x) = 3x−4 on the largest
domain for which the formula makes sense.
(a) Find the domain and range of f(x), then

sketch the graph.
(b) Find the domain, range and rule for

the inverse function f−1 , then sketch its C
D
graph.
Problem 9.2. Find the inverse function of

each of the following functions. Specify the do-
mains of the inverse functions.
1 Problem 9.5. Show that, for every value of a,

(a) f(x) = x+8 the function
5
1
5 f(x) = a +
(b) h(x) = x−a
x+3
√ is its own inverse.
(c) g(x) = 4 3 − x − 7
√ √ Problem 9.6. Clovis is standing at the edge
(d) j(x) = x + x − 1
of a cliff, which slopes 4 feet downward from
him for every 1 horizontal foot. He launches
p
(e) k(x) = 16 − x2 , 0 ≤ x ≤ 4
a small model rocket from where he is stand-
ing. With the origin of the coordinate system
located where he is standing, and the x-axis
Problem 9.3. For this problem, y = f(x) = extending horizontally, the path of the rocket
2x2 − 3x − 1 on the domain of all real numbers. is described by the formula y = −2x2 + 120x.
(a) Give a function h = f(x) relating the
(a) Sketch the function graph and find the
height h of the rocket above the sloping
coordinates of the vertex P = (a,b).
ground to its x-coordinate.
(b) Explain why y = f(x) does not have an (b) Find the maximum height of the rocket
inverse function on the domain of all real above the sloping ground. What is its
numbers. x-coordinate when it is at its maximum
height?
(c) Restrict y = f(x) to the domain {a ≤
x} and find the formula for the inverse (c) Clovis measures its height h of the
function f−1 (y). What are the domain rocket above the sloping ground while it
and range of the inverse function? is going up. Give a function x = g(h) re-
lating the x-coordinate of the rocket to
(d) Restrict y = f(x) to the domain {x ≤ h.
a} and find the formula for the inverse
(d) Does this function still work when the
function f−1 (y). What are the domain
rocket is going down? Explain.
and range of the inverse function?

Problem 9.4. Which of the following graphs tions: (1) sketch the function, (2) find the in-
are one-to-one? If they are not one-to-one, sec- verse function, and (3) sketch the inverse func-
tion the graph up into parts that are one-to- tion. In each case, indicate the correct do-
one. mains and ranges. (4) Finally, make sure you
9.6. EXERCISES 131
test each of the functions you propose as an (b) After how many hours will the surface of
inverse with the following compositions: the water have width of 6 feet?
?
f(f−1 (x)) = x (c) Give a function t = f−1 (w) relating the
time to the width of the surface of the
and water. Make sure to specify the domain
?
f−1 (f(x)) = x. and compute the range too.
(a) f(x) = 3x − 2
(b) f(x) = 21 x + 5
Problem 9.9. A biochemical experiment in-
(c) f(x) = −x2 + 3, x ≥ 0 volves combining together two protein ex-
(d) f(x) = x2 + 2x + 5, x ≤ −1 tracts. Suppose a function φ(t) monitors the
√ amount (nanograms) of extract A remaining at
(e) f(x) = 4 − x2 , 0 ≤ x ≤ 2
time t (nanoseconds). Assume you know these
facts:
Problem 9.8. A trough has a semicircular
cross section with a radius of 5 feet. Water 1. The function φ is invertible; i.e., it has
starts flowing into the trough in such a way an inverse function.
that the depth of the water is increasing at a 2. φ(0) = 6, φ(1) = 5, φ(2) = 3, φ(3) = 1,
rate of 2 inches per hour. φ(4) = 0.5, φ(10) = 0.
water (a) At what time do you know there will be

3 nanograms of extract A remaining?
(b) What is φ−1 (0.5) and what does it tell
you?
5 ft
(c) (True or False) There is exactly one time
cross-section of
when the amount of extract A remaining
trough
is 4 nanograms.
(a) Give a function w = f(t) relating the (d) Calculate φ(φ−1 (1)) =
width w of the surface of the water to (e) Calculate φ−1 (φ(6)) =
the time t, in hours. Make sure to spec-
ify the domain and compute the range (f) What is the domain and range of φ?
too.
Chapter 10
Exponential Functions
If we start with a single yeast cell under favorable growth conditions, then
it will divide in one hour to form two identical “daughter cells”. In turn,
after another hour, each of these daughter cells will divide to produce
two identical cells; we now have four identical “granddaughter cells” of
the original parent cell. Under ideal conditions, we can imagine how this
“doubling effect” will continue:
cells TIME
t=0 hours
t=1 hours
t=2 hours
t=3 hours
Figure 10.1: Observing cell growth.
The question is this: Can we find a function of t that

Total Number of
will predict (i.e. model) the number of yeast cells after t yeast cells
hours
hours? If we tabulate some data (as at right), the conclu-
0 1=20
sion is that the formula
1 2=21
2 4=22
N(t) = 2t
3 8=23
4 16=24
predicts the number of yeast cells after t hours. Now, 5 32=25
let’s make a very slight change. Suppose that instead 6 64=26
of starting with a single cell, we begin with a population
of 3 × 106 cells; a more realistic situation. If we assume Table 10.1: Cell growth data.
133
134 CHAPTER 10. EXPONENTIAL FUNCTIONS
that the population of cells will double every hour, then

reasoning as above will lead us to conclude that the formula
N(t) = (3 × 106)2t
gives the population of cells after t hours. Now, as long as t represents
a non-negative integer, we know how to calculate N(t). For example, if
t = 6, then
N(t) = (3 × 106)26
= (3 × 106)(2 · 2 · 2 · 2 · 2 · 2)
= (3 × 106)64
= 192 × 106.
The key point is that computing N(t) only involves simple arithmetic.
But what happens if we want to know the population of cells after 6.37
hours? That would require that we work with the formula
N(t) = (3 × 106)26.37
and the rules of arithmetic do not suffice to calculate N(t). We are stuck,
since we must understand the meaning of an expression like 26.37. In
order to proceed, we will need to review the algebra required to make
sense of raising a number (such as 2) to a non-integer power. √
We need to
understand the precise meaning of expressions like: 2 , 2 , 2−π, etc.
6.37 5
10.1 Functions of Exponential Type
this is a variable this is a fixed positive integer

x b
y=b y=x
this is a fixed number this is a variable
Exponential Picture Monomial Picture
Figure 10.2: Viewing the difference between exponential and monomial functions.
On a symbolic level, the class of functions we are trying to motivate is

easily introduced. We have already studied the monomials y = xb, where
x was our input variable and b was a fixed positive integer exponent.
What happens if we turn this around, interchanging x and b, defining a
new rule:
y = f(x) = bx. (10.1)
10.1. FUNCTIONS OF EXPONENTIAL TYPE 135
We refer to x as the power and b the base. An expression of this sort

is called a function of exponential type. Actually, if your algebra is a
bit rusty, it is easy to initially confuse functions of exponential type and
monomials (see Figure 10.2).
10.1.1 Reviewing the Rules of Exponents

To be completely honest, making sense of the expression
y = bx for all numbers x requires the tools of Calculus,
but it is possible to establish a reasonable comfort level
3
by handling the case when x is a rational number. If b ≥ 0
and n is a positive integer (i.e. n = 1, 2, 3, 4, . . . ), then we 2 n=2
n=4
can try to solve the equation 1
n=6
tn = b. (10.2) -1.5 -1 -0.5

0
0 0.5 1 1.5
-1
-2
A solution t to this equation is called an nth root of
-3
b. This leads to complications, depending on whether n
is even or odd. In the odd case, for any real number b,
notice that the graph of y = b will always cross the graph 3
of y = tn exactly once, leading to one solution of (10.2). 2 n=3

On the other hand, if n is even and b < 0, then the n=5
n=7
1
graph of y = tn will miss the graph of y = b, implying there
are no solutions to the equation in (10.2). (There will be -1.5 -1 -0.5
0
0 0.5 1 1.5
complex solutions to equations such as t2 = √ −1, involv- -1
ing the imaginary complex numbers ±i = ± −1, but we -2

are only working with real numbers in this course.) Also,
-3
again in the case when n is even, it can happen that there
are two solutions to (10.2). We do not want to constantly
worry about this even/odd distinction, so we will hence-
forth assume b > 0. To eliminate possible ambiguity, we Figure 10.3: Even and odd
monomials.
will single out a particular nth-root; we define the sym-
bols:
√
n 1
b = b n = the largest real nth root of b. (10.3)
√
Thus, whereas ±1 are both 4th-roots of 1, we have defined 4 1 = 1.
In order to manipulate y = bx for rational x, we need to recall some
basic facts from algebra.
Important Facts 10.1.1 (Working with rational exponents). For all posi-
tive integers p and q, and any real number base b > 0, we have
p
√ p √
q q
b =
q b = bp.
For any rational numbers r and s, and for all positive bases a and b:
n n even n
y-axis y=t y-axis y=t n odd
= solution
y=b
y=b
t-axis
y=b*
t-axis
no solution or two solutions

exactly one solution
Figure 10.4: How many solutions to tn = b?.
1. Product of power rule: brbs = br+s
2. Power of power rule: (br)s = brs
3. Power of product rule: (ab)r = arbr
4. Zero exponent rule: b0 = 1

1
5. Negative power rule: b−r = br
These rules have two important consequences, one theoretical and the
other more practical. On the first count, recall that any rational number
r can be written in the form r = qp , where p and q are integers. Conse-
quently, using these rules, we see that the expression y = bx defines a
function of x, whenever x is a rational number. On the more practical
side of things, using the rules we can calculate and manipulate certain
expressions. For example,
2
√ 2
3
27 3 = 27 = 32 = 9;
5
√ −5 1 1
3
8− 3 = 8 = 2−5 = 5 = .
2 32
The sticky point which remains is knowing that f(x) = bx actually
defines a function for all real values of x. This is not easy to verify and
we are simply going to accept it as a fact. The difficulty is that we need
the fundamentally new concept of a limit, which is the starting point of
a Calculus course. Once we know the expression does define a function,
we can also verify that the rules of Fact 10.1.1 carry through for all real
10.2. THE FUNCTIONS Y = A0BX 137
exponent powers. Your calculator should

√
have a “y to the x key”, allowing
2
you to calculate expressions such as π involving non-rational powers.
Here are the key modeling functions we will work with in this Chapter.
Definition 10.1.2. A function of exponential type has the form
A(x) = A0bx,
for some b > 0, b 6= 1, and A0 6= 0.
We will refer to the formula in Definition 10.1.2 as the standard ex-
ponential form. Just as with standard forms for quadratic functions, we
sometimes need to do a little calculation to put an equation in standard
form. The constant A0 is called the initial value of the exponential func-
tion; this is because if x represents time, then A(0) = A0b0 = A0 is the
value of the function at time x = 0; i.e. the initial value of the function.
2x−1
Example 10.1.3. Write the equations y = 83x and y = 7 12 in standard
exponential form.
Solution. In both cases, we just use the rules of exponents to maneuver
the given equation into standard form:
y = 83x
= (83)x
= 512x
and
2x−1
1
y=7
2
2x −1
1 1
=7
2 2
2!x
1
=7 2
2
x
1
= 14
4
10.2 The Functions y = A0bx

We know f(x) = 2x defines a function of x, so we can study basic qual-
itative features of its graph. The data assembled in the solution of the
“Doubling Effect” beginning this Chapter, plus the rules of exponents,
produce a number of points on the graph. This graph exhibits four key
qualitative features that deserve mention:
Point on the
x 2x x
graph of y = 2x y=2
.. .. .. (3,8)
. . .
-2 1/4 (-2, 1/4)
-1 1/2 (-1, 1/2)
(2,4)
0 1 (0, 1) (−1,1/2)
1 2 (1, 2) (−2,1/4) (1,2)
(0,1)
2 4 (2, 4)
3 8 (3, 8) −1 1
.. .. ..
. . .
(a) Data points from y = 2x . (b) Graph of y = 2x .
Figure 10.5: Visualizing y = 2x .
• The graph is always above the horizontal axis; i.e. the function
values are always positive.
• The graph has y-intercept 1 and is increasing.
• The graph becomes closer and closer to the horizontal axis as we

move left; i.e. the x-axis is a horizontal asymptote for the left-
hand portion of the graph.
• The graph becomes higher and higher above the horizontal axis as
we move to the right; i.e., the graph is unbounded as we move to
the right.
The special case of y = 2x is representative of the function y = bx, but
there are a few subtle points that need to be addressed. First, recall we
are always assuming that our base b > 0. We will consider three separate
cases: b = 1, b > 1, and 0 < b < 1.
10.2.1 The case b = 1

In the case b = 1, we are working with the function y = 1x = 1; this is not
too exciting, since the graph is just a horizontal line. We will ignore this
case.
10.2.2 The case b > 1

If b > 1, the graph of the function y = bx is qualitatively similar to the
situation for b = 2, which we just considered. The only difference is the
exact amount of “concavity” in the graph, but the four features high-
lighted above are still valid. Figure 10.6(a) indicates how these graphs
10.2. THE FUNCTIONS Y = A0BX 139
compare for three different values of b. Functions of this type exhibit

what is typically referred to as exponential growth ; this codifies the fact
that the function values grow rapidly as we move to the right along the
x-axis.
y-axis y-axis
all graphs pass through (0,1)

all graphs pass through (0,1)
x-axis x-axis
(a) Graph of y = bx , b > 1. (b) Graph of y = bx , 0 < b < 1.

Figure 10.6: Visualizing cases for b.
10.2.3 The case 0 < b < 1

We can understand the remaining case 0 < b < 1, by using the remarks
above and our work in Chapter 13. x First, with this condition on b, notice
that b1 > 1, so the graph of y = b1 is of the type in Figure 10.6(a). Now,
using the rules of exponents:
−x −1!x
1 1
y= = = bx.
b b
−x
By the reflection principle, the graph of y = b1 is obtained by re-
1 x
flecting the graph of y = ( b ) about the y-axis. Putting these remarks
together, if 0 < b < 1, we conclude that the graph of y = bx will look
like Figure 10.6(b). Notice, the graphs in Figure 10.6(b) share qualitative
features, mirroring the features outlined previously, with the “asymptote”
and “unbounded” portions of the graph interchanged. Graphs of this sort
are often said to exhibit exponential decay, in the sense that the function
values rapidly approach zero as we move to the right along the x-axis.
Important Facts 10.2.1 (Features of Exponential Type Functions). Let

b be a positive real number, not equal to 1. The graph of y = bx has these
four properties:
1. The graph is always above the horizontal axis.
2. The graph has y-intercept 1.
3. If b > 1 (resp. 0 < b < 1), the graph becomes closer and closer to the
horizontal axis as we move to the left (resp. move to the right); this
says the x-axis is a horizontal asymptote for the left-hand portion of
the graph (resp. right-hand portion of the graph).
4. If b > 1 (resp. 0 < b < 1), the graph becomes higher and higher above
the horizontal axis as we move to the right (resp. move to the left);
this says that the graph is unbounded as we move to the right (resp.
move to the left).
If A0 > 0, the graph of the function y = A0bx is a vertically expanded or

compressed version of the graph of y = bx. If A0 < 0, we additionally
reflect about the x-axis.
10.3 Piano Frequency Range

A sound wave will cause your eardrum to move back and forth. In the
case of a so-called pure tone, this motion is modeled by a function of the
form
d(t) = A sin(2πft),
where f is called the frequency, in units of “periods/unit time”, called

“Hertz” and abbreviated “Hz”. The coefficient A is related to the actual
displacement of the eardrum, which is, in turn, related to the loudness
of the sound. A person can typically perceive sounds ranging from 20 Hz
to 20,000 Hz.
A# C# D# F# G# A# C# D# F# G#A# C#D# F# G# A# C# D# F# G#A# C# D# F# G# A# C# D# F# G# A# C# D# F# G#A#
A B C D E F G A B C D E F G A B C D E F G A B C D E F G A B C D E F G A B C D E F G A B C D E F G A B C
220 Hz
middle C
Figure 10.7: A piano keyboard.
A piano keyboard layout is shown in Figure 10.7. The white keys are
labelled A, B, C, D, E, F, and G, with the sequence running from left to
right and repeating for the length of the keyboard. The black keys fit into
this sequence as “sharps”, so that the black key between A and B is “A
sharp”, denoted A# . Thus, starting at any A key, the 12 keys to the right
are A, A# , B, C, C# , D, D# , E, F, F# , G, and G# . The sequence then repeats.
Notice that between some adjacent pairs of white keys there is no black
key.
A piano keyboard is commonly tuned according to a rule requiring
that each key (white and black) has a frequency 21/12 times the frequency
of the key to its immediate left. This makes the ratio of adjacent keys
always the same (21/12), and it means that keys 12 keys apart have a ratio
of frequencies exactly equal to 2 (since (21/12)12 = 2). Two such keys are
10.3. PIANO FREQUENCY RANGE 141
said to be an octave apart. Assuming that the key A below middle C has
a frequency of 220 Hz, we can determine the frequency of every key on
the keyboard. For instance, the A# to the right of this key has frequency
220 × 21/12 = 220 × 1.059463094... ≈ 233.08188Hz. The B to the right of this
key has frequency 233.08188 × 21/12 ≈ 246.94165Hz.
10.4 Exercises
Problem 10.1. Let’s brush up on the required (d) Anja, a third member of your lab work-
calculator skills. Use a calculator to approxi- ing with the same yeast cells, took these
mate: two measurements: 7.246 × 106 cells after
4 hours; 16.504 × 106 cells after 6 hours.
(a) 3π Should you be worried by Anja’s results?
√
(b) 42+ 5 If Anja’s measurements are correct, does
your model over estimate or under esti-
(c) ππ mate the number of yeast cells at time
√ t?
(d) 5− 3
2
(e) 3π
√ Problem 10.4. (a) Find the frequency of
(f) 11π−7 middle C.
(b) Find the frequency of A above middle C.

Problem 10.2. Put each equation in standard
exponential form: (c) What is the frequency of the lowest note
on the keyboard? Is there a way to solve
(a) y = 3(2−x ) this without simply computing the fre-
quency of every key below A220?
(b) y = 4−x/2
(c) y = ππx (d) The Bosendorfer piano is famous, due in

part, to the fact it includes additional
x
1 3+ 2
(d) y = 1 keys at the left hand end of the key-
3
board, extending to the C below the bot-
5
(e) y = 0.3452x−7 tom A on a standard keyboard. What
is the lowest frequency produced by a
(f) y = 4(0.0003467)−0.4x+2 Bosendorfer?
Problem 10.3. A colony of yeast cells is esti- Problem 10.5. You have a chess board as pic-
mated to contain 106 cells at time t = 0. After tured, with squares numbered 1 through 64.
collecting experimental data in the lab, you de- You also have a huge change jar with an un-
cide that the total population of cells at time t limited number of dimes. On the first square
hours is given by the function you place one dime. On the second square you
stack 2 dimes. Then you continue, always dou-
y = 106 e0.495105t . bling the number from the previous square.
(a) How many cells are present after one (a) How many dimes will you have stacked
hour? on the 10th square?
(b) (True or False) The population of yeast (b) How many dimes will you have stacked
cells will double every 1.4 hours. on the nth square?
(c) Cherie, another member of your lab, (c) How many dimes will you have stacked
looks at your notebook and says : ...that on the 64th square?
formula is wrong, my calculations pre-
dict the formula for the number of yeast (d) Assuming a dime is 1 mm thick, how
cells is given by the function high will this last pile be?
y = 106 (2.042727)0.693147t . (e) The distance from the earth to the sun
is approximately 150 million km. Relate
Should you be worried by Cherie’s re- the height of the last pile of dimes to this
mark? distance.
10.4. EXERCISES 143
calculates the fraction of hemoglobin satu-

63 64
rated with oxygen at a given pressure p.
(a) The graphs of M(p) and H(p) are given

below on the domain 0 ≤ p ≤ 100; which
is which?
fraction
1
0.8
0.6
10 9
0.4
1 2 3 8
0.2
p
20 40 60 80 100
Problem 10.6. Myoglobin and hemoglobin are
oxygen carrying molecules in the human body.
Hemoglobin is found inside red blood cells, (b) If the pressure in the lungs is 100 torrs,
which flow from the lungs to the muscles what is the level of oxygen saturation of
through the bloodstream. Myoglobin is found the hemoglobin in the lungs?
in muscle cells. The function (c) The pressure in an active muscle is 20
p torrs. What is the level of oxygen satu-
Y = M(p) =
1+p ration of myoglobin in an active muscle?
calculates the fraction of myoglobin saturated What is the level of hemoglobin in an ac-
with oxygen at a given pressure p torrs. For tive muscle?
example, at a pressure of 1 torr, M(1) = 0.5,
(d) Define the efficiency of oxygen transport
which means half of the myoglobin (i.e. 50%)
at a given pressure p to be M(p) − H(p).
is oxygen saturated. (Note: More precisely, you
What is the oxygen transport efficiency
need to use something called the “partial pres-
at 20 torrs? At 40 torrs? At 60 torrs?
sure”, but the distinction is not important for
Sketch the graph of M(p) − H(p); are
this problem.) Likewise, the function
there conditions under which transport
p2.8 efficiency is maximized (explain)?
Y = H(p) =
262.8 + p2.8
Chapter 11
Exponential Modeling
Example 11.0.1. A computer industry spokesperson has predicted that

the number of subscribers to geton.com, an internet provider, will grow ex-
ponentially for the first 5 years. Assume this person is correct. If geton.com
has 100,000 subscribers after 6 months and 750,000 subscribers after 12
months, how many subscribers will there be after 5 years?
Solution. The solution to this problem offers a template for many expo-
nential modeling applications. Since, we are assuming that the number
of subscribers N(x), where x represents years, is a function of exponential
type,
N(x) = N◦ bx,
for some N◦ and b > 1. We are given two pieces of information about the
values of N(x):
N(0.5) = 100,000; i.e., N0b0.5 = 100,000, and
N(1) = 750,000; i.e., N0b = 750,000.
We can use these two equations to solve for the two unknowns N◦ and b
as follows: If we divide the second equation by the first, we get
b1
= 7.5
b0.5 √
b0.5 = b1/2 = b = 7.5
∴ b = 56.25.
Plugging this value of b into either equation (say the first one), we can
100,000
solve for N◦ : N◦ = (56.25)0.5 = 13,333. We conclude that the number of
geton.com subscribers will be predicted by
N(x) = 13,333(56.25)x.
In five years, we obtain N(5) = 7,508,300,000,000 subscribers, which ex-
ceeds the population of the Earth (which is between 5 and 6 billion)!
145
146 CHAPTER 11. EXPONENTIAL MODELING
800000 Q
There are two important conclusions we can draw from
600000
this problem. First, the given information provides us
400000
with two points on the graph of the function N(x):
200000
P
P = (0.5, 100,000)
0.2 0.4 0.6 0.8 1 Q = (1, 750,000).
Figure 11.1: Finding the More importantly, this example illustrates a very impor-
equation for N(x) = N0 bx . tant principal we can use when modeling with functions
of exponential type.
Important Fact 11.0.2. A function of exponential type can be determined
if we are given two data points on its graph.
!!!
When you use the above strategy to find the base b of the exponen-
CAUTION
tial model, make sure to write down a lengthy decimal approximation.
!!!
As a rule of thumb, go for twice as many significant digits as you are
otherwise using in the problem.
11.1 The Method of Compound Interest

You walk into a Bank with P0 dollars (usually called principal), wishing
to invest the money in a savings account. You expect to be rewarded by
the Bank and paid interest, so how do you compute the total value of the
account after t years?
The future value of the account is really a function of the number of
years t elapsed, so we can write this as a function P(t). Our goal is to see
that P(t) is a function of exponential type. In order to compute the future
value of the account, the Bank provides any savings account investor
with two important pieces of information:
r = annual (decimal) interest rate
n = the number of compounding periods per year
The number n tells us how many times each year the Bank will compute
the total value P(t) of the account. For example, if n = 1, the calculation
is done at one-year intervals; if n = 12, the calculation is done each
month, etc. The bank will compute the value of your account after a
typical compounding period by using the periodic rate of return nr . For
example, if the interest rate percentage is 12% and the compounding
period is monthly (i.e., n = 12), then the annual (decimal) interest rate is
0.12 and the periodic rate is 0.12
12
= 0.01.
!!!
The number r always represents the decimal interest rate, which is a
CAUTION
decimal between 0 and 1. If you are given the interest rate percentage
!!!
(which is a positive number between 0 and 100), you need to convert
to a decimal by dividing by 100.
11.1. THE METHOD OF COMPOUND INTEREST 147
11.1.1 Two Examples

Let’s consider an example: P0 = $1,000 invested at the annual interest
percentage of 8% compounded yearly, so n = 1 and r = 0.08. To compute
the value P(1) after one year, we will have
P(1) = P0 + (periodic rate)P0
= P0 + rP0 = P0(1 + r)
= $1,000(1 + 0.08) = $1,080.
To compute the value after two years, we need to apply the periodic
rate to the value of the account after one year:
P(2) = P(1) + (periodic rate)P(1)
= P0(1 + r) + rP0(1 + r) = P0(1 + r)2
= $1,000(1 + 0.08)2 = $1,166.40.
Notice, the amount the Bank has paid after two years is $166.40, which
is slightly bigger than twice the $80 paid after one year. To compute the
value after three years, we need to apply the periodic rate to the value of
the account after two years:
P(3) = P(2) + (periodic rate)P(2)
= P0(1 + r)2 + rP0(1 + r)2 = P0(1 + r)3
= $1,000(1 + 0.08)3 = $1,259.71.
Again, notice the amount the Bank is paying after three years is $259.71,
which is slightly larger than three times the $80 paid after one year.
Continuing on in this way, to find the value after t years, we arrive at the
formula
P(t) = P0(1 + r)t
= $1,000(1.08)t.
In particular, after 5 and 10 years, the value of the account (to the nearest
dollar) will be $1,469 and $2,159, respectively.
As a second example, suppose we begin with the same $1,000 and
the same annual interest percentage 8%, but now compound monthly,
so n = 12 and r = 0.08. The value of the account after one compounding
period is P(1/12), since a month is one-twelfth of a year. Arguing as
before, paying special attention that the periodic rate is now nr = 0.08
12
, we
have
P(1/12) = P0 + (periodic rate)P0

.08
= P0 1 +
12
= $1,000(1 + 0.006667) = $1,006.67.
After two compounding periods, the value is P(2/12),
P(2/12) = P(1/12) + (periodic rate)P(1/12)

.08 0.08 .08
= P0 1 + + P0 1 +
12 12 12
2
.08
= P0 1 +
12
= $1,000(1 + 0.006667)2 = $1,013.38.
Continuing on in this way, after k compounding periods have elapsed,

k
the value will be P 12 , which is computed as
k
.08
P(k/12) = P0 1+ .
12
It is possible to rewrite this formula to give us the value after t years,

noting that t years will lead to 12t compounding periods; i.e., set k = 12t
in the previous formula:
12t
.08
P(t) = P0 1 +
12
For example, after 1, 5 and 10 years, the value of the account, to the
nearest dollar, would be $1,083, $1,490, and $2,220.
11.1.2 Discrete Compounding

The two examples above highlight a general formula for computing the
future value of an account.
Important Fact 11.1.1 (Discrete compounding). Suppose an account is

opened with P0 principal. If the decimal interest rate is r and the number of
compounding periods per year is n, then the value P(t) of the account after
t years will be
r nt
P(t) = P0 1 + .
n
Notice, the future value P(t) is a function of exponential type; the base
is the number 1 + nr , which will be greater than one. Since P0 > 0, the
graph will be qualitatively similar to the ones pictured in Figure 10.6(a).
Example 11.1.2. At birth, your Uncle Hans secretly purchased a $5,000

U.S. Savings Bond for $2,500. The conditions of the bond state that the
U.S. Government will pay a minimum annual interest rate of r = 8.75%,
compounded quarterly. Your Uncle has given you the bond as a gift, subject
to the condition that you cash the bond at age 35 and buy a red Porsche.
11.2. THE NUMBER E AND THE EXPONENTIAL FUNCTION 149
On your way to the Dealer, you receive a call from your tax accountant
informing you of a 28% tax on the capital gain you realize through cashing
in the bond; the capital gain is the selling price of the bond minus the
purchase price. Before stepping onto the showroom floor, compute how
much cash will you have on hand, after the U.S. Government shares in
your profits.
Solution. The value of your bond after 35 years is computed by the for-
mula in Fact 11.1.1, using P0 = $2,500, r = 0.0875, n = 4, and t = 35.
Plugging this all in, we find that the selling price of the bond is
4(35)
0.0875
P(35) = $2,500 1 + = $51,716.42.
4
The capital gain will be $51,716.42 - $2,500 = $49,216.42 and the tax
due is $(49,216.42)(0.28) = $13,780.60. You are left with $51,716.42 -
$13,780.60 = $37,935.82. Better make that a used Porsche!
11.2 The Number e and

the Exponential Function
What happens to the future value of an investment of P0 dollars as the
number of compounding periods is increased? For example, return to
our earlier example: P0 = $1000 and an annual interest percentage of 8%.
After 1 year, the table below indicates the value of the investment for
various compounding periods: yearly, quarterly, monthly, weekly, daily,
and hourly.
Value after 1 year

n Compounding (to nearest dol-
Period
lar)
1 yearly $1,000(1 + 0.08)1 = $1,080.00
0.08 4

4 quarterly $1,000 1 + 4
= $1,082.43
0.08 12

12 monthly $1,000 1 + 12
= $1,083.00
0.08 52

52 weekly $1,000 1 + 52
= $1,083.22
0.08 365

365 daily $1,000 1 + 365
= $1,083.28
0.08 8,760

8,760 hourly $1,000 1 + 8,760
=$1,083.29
We could continue on, considering “minute” and “second” compounding

and what we will find is that the value will be at most $1,083.29. This
illustrates a general principal:
Important Fact 11.2.1. Initially increasing the number of compounding
periods makes a significant difference in the future value; however, even-
tually there appears to be a limiting value.
Let’s see if we can understand mathematically why this is happening.
The first step is to recall the discrete compounding formula:
r nt
P(t) = P0 1 + .
n
If our desire is to study the effect of increasing the number of compound-
ing periods, this means we want to see what happens to this formula as
n gets BIG. To analyze this, it is best to rewrite the expression using a
substitution trick: Set z = nr , so that n = rz and nr = 1z . Plugging in, we
have
r nt
P(t) = P0 1 +
n
rzt
1
= P0 1 + (11.1)
z
z rt
1
= P0 1+ .
z
So, since r is a fixed number and z = nr , letting n get BIG is the same
as letting z become BIG in (11.1). This all means we need to answer
1 z

this new question: What happens to the expression 1 + z as z becomes
large? On the one hand, the power in the expression is getting large; at
the same time, the base is getting close to 1. This makes it very tricky
to make quick predictions about the outcome. It is best to first tabulate
z
some numerical data for the values of y = g(z) = 1 + 1z and look at a
plot of this function graph on the domain 0.01 ≤ z ≤ 100:
See Figure 11.2.
1 z
You can see from this plot, the graph of y = 1 + z approaches the
“dashed” horizontal asymptote, as z becomes BIG. We will let the letter “e”
represent the spot where this horizontal line crosses the vertical axis and
e ≈ 2.7182818. This number is only an approximation, since e is known to
be an irrational number. What sets√ this irrational number apart from the
ones you are familiar with (e.g. 2, π, etc.) is that defining the number
e requires a “limiting” process. This will be studied a lot more in your
Calculus course. The new number e is a positive number greater than 1,
so we can study the function:
y = ex. (11.2)
Since e > 1, the graph will share the properties in Figure 10.6(a). This
function is usually referred to as THE exponential function. Scientific
calculators will have a key of the form “ exp(x) ” or “ ex ”.
11.2. THE NUMBER E AND THE EXPONENTIAL FUNCTION 151
z
1
z 1+ z
1 2 2.5
2 2.25
2
3 2.37037
1.5
4 2.4414
20 2.65329 1
100 2.70481 0.5
1000 2.71692
20 40 60 80 100
109 2.71828
1 z

(a) Data points
z for (b) The graph of 1 + z .
1 + 1z .
Figure 11.2: What happens when z get very large?
11.2.1 Calculator drill

Plugging in x = 1, you can compute an approximation to e on your cal-
culator; you should get e = 2.7183 to four decimal
√ places. Make sure
you can compute expressions like e3, eπ, and e = e1/2; to four decimal
places, you should get 20.0855, 23.1407, and 1.6487.
11.2.2 Back to the original problem...

We can now return to our future value formula (11.1) and conclude that
as the number of compounding periods increases, the future value is
approaching a limiting value:
zrt
1
P(t) = P0 1+ =⇒ P0ert.
z
The right hand limiting formula Q(t) = P0ert computes the future value
using what is usually referred to as continuous compounding. From the
investors viewpoint, this is the best possible scheme for computing future
value.
Important Fact 11.2.2 (Continuous compounding). The future value of

P0 dollars principal invested at an annual decimal interest rate of r un-
der continuous compounding after t years is Q(t) = P0ert; this value is al-
nt
ways greater than the value of P0 1 + nr , for any discrete compounding
scheme. In fact, P0ert is the limiting value.
11.3 Exercises
Problem 11.1. In 1968, the U.S. minimum (a) Sketch rough graphs of these two func-
wage was $1.60 per hour. In 1976, the min- tions.
imum wage was $2.30 per hour. Assume the
minimum wage grows according to an expo- (b) The graph of the equation x2 − y2 = 1
nential model w(t), where t represents the time is shown below; this is called the unit
in years after 1960. hyperbola. For any value a, show that
the point (x,y) = (cosh(a), sinh(a)) is on
(a) Find a formula for w(t). the unit hyperbola. (Hint: Verify that
(b) What does the model predict for the min- [cosh(x)]2 − [sinh(x)]2 = 1, for all x.)
imum wage in 1960?
y
(c) If the minimum wage was $5.15 in 1996,
is this above, below or equal to what the
model predicts.
Problem 11.2. The town of Pinedale,

Wyoming, is experiencing a population boom.
In 1990, the population was 860 and five years
later it was 1210. (−1,0) (1,0) x
(a) Find a linear model l(x) and an expo-
nential model p(x) for the population of
Pinedale in the year 1990+x.
(b) What do these models estimate the pop-
ulation of Pinedale to be in the year
2000?
(c) A hanging cable is modeled by a portion
Problem 11.3. In 1989, research scientists of the graph of the function
published a model for predicting the cumu-
x−h
lative number of AIDS cases reported in the y = a cosh( ) + C,
United States: a

t − 1980
3 for appropriate constants a, h and C.
a(t) = 155 , (thousands) The constant h depends on how the co-
10
ordinate system is imposed. A cable for
where t is the year. This paper was considered a suspension bridge hangs from two 100
a “relief”, since there was a fear the correct ft. high towers located 400 ft. apart. Im-
model would be of exponential type. Pick two pose a coordinate system so that the pic-
data points predicted by the research model ture is symmetric about the y-axis and
a(t) to construct a new exponential model b(t) the roadway coincides with the x-axis.
for the number of cumulative AIDS cases. Dis- The hanging cable constant is a = 500
cuss how the two models differ and explain the and h = 0. Find the minimum distance
use of the word “relief”. from the cable to the road.
Problem 11.4. Define two new functions: towers

x −x
e +e
y = cosh(x) = cable
2
100 ft
and d
x −x
e −e
y = sinh(x) = . roadway
2
These are called the basic hyperbolic trigono- 400 ft
metric functions.
Chapter 12
Logarithmic Functions
If we invest P0 = $1,000 at an annual rate of r = 8% compounded contin-

uously, how long will it take for the account to have a value of $5000?
The formula P(t) = 1,000e0.08t gives the value after t years, so we need
to solve the equation:
5,000 = 1,000e0.08t
5 = e0.08t.
Unfortunately, algebraic manipulation will not lead to a further simplifi-

cation of this equation; we are stuck! The required technique involves the
theory of inverse functions. Assuming we can find the inverse function of
f(t) = et, we can apply f−1(t) to each side of the equation and solve for t:
f−1(5) = f−1(e0.08t) = 0.08t

(12.5)f−1(5) = t
The goal in this section is to describe the function f−1, which is usually
denoted by the symbol f−1(t) = ln(t) and called the natural logarithm
function. On your calculator, you will find a button dedicated to this
function and we can now compute ln(5) = 1.60944. Conclude that the
solution is t = 20.12 years.
12.1 The Inverse Function of y = ex

If we sketch a picture of the exponential function on the domain of all real
numbers and keep in mind the properties in Fact 10.2.1, then every hor-
izontal line above the x-axis intersects the graph of y = ex exactly once:
See Figure 12.1(a). The range of the exponential function will consist
of all possible y-coordinates of points on the graph. Using the graphical
techniques of Chapter 6, we can see that the range of will be all POSITIVE
real numbers: See Figure 12.1(b).
153
154 CHAPTER 12. LOGARITHMIC FUNCTIONS
PSfrag y=e
x
y = ex
range = all
positive numbers
graphs of y=c,
c>1 cross exponential
graph exactly once.
these horizontal
−1 1 lines miss graph −1 1
of exponential function.
domain = all real numbers
(a) Horizontal line test for (b) The domain and range for
y = ex . y = ex .
Figure 12.1: Properties needed to find the inverse of f(x) = ex .
By the horizontal line test, this means the exponential function is one-
to-one and the inverse rule f−1(c) will define a function

the unique solu-
!



 tion of the equation , if c > 0

−1
f (c) = c = ex (12.1)


 if c ≤ 0.
 (undefined),

y=e
x reflecting line y = x
This inverse function is called the natural logarithm
function, denoted ln(c). We can sketch the graph of the
y=ln (x)
the natural logarithm as follows: First, by Fact 9.2.1, the
domain of the function ln(y) = x is just the range of the
−1 1 exponential function, which we noted is all positive num-
bers. Likewise, the range of the function ln(y) = x is the
domain of the exponential function, which we noted is all
Figure 12.2: Visualizing the
real numbers. Interchanging x and y, the graph of the
y = ln(x). natural logarithm function y = ln(x) can be obtained by
flipping the graph of y = ex across the line y = x:
Important Facts 12.1.1 (Graphical features of natural log). The function
y = ln(x) has these features:
• The largest domain is the set of positive numbers; e.g. ln(−1) makes
no sense.
• The graph has x-intercept 1 and is increasing.
• The graph becomes closer and closer to the vertical axis as we ap-
proach x = 0; i.e. the y-axis is a vertical asymptote for the graph.
• The graph is unbounded as we move to the right.

Any time we are working with an inverse function, symbolic properties
are useful. Here are the important ones related to the natural logarithm.
Important Facts 12.1.2 (Natural log properties). We have the following
properties:
12.1. THE INVERSE FUNCTION OF Y = EX 155
(a) For any real number x, ln(ex) = x.
(b) For any positive number x, eln(x) = x.
(c) ln(bt) = t ln(b), for b > 0 and t any real number;
(d) ln(ba) = ln(a) + ln(b), for all a, b > 0;
(e) ln ab = ln(b) − ln(a), for all a, b > 0.

The properties (c)-(e) are related to three of the rules of exponents in

Facts 10.1.1. Here are the kinds of basic symbolic maneuvers you can
pull off using these properties:
Examples 12.1.3.
(i) ln(83) = 3 ln(8) = 6.2383; ln(6π) = ln(6) + ln(π) = 2.9365; ln 53 =

ln(3) − ln(5) = −0.5108.

√
(ii) ln x = lnx1/2 = 12 ln(x); ln x2 − 1 = ln((x − 1)(x + 1)) = ln(x − 1) +

5
ln(x + 1); ln x2x+1 = ln x5 − ln x2 + 1 = 5 ln(x) − ln x2 + 1 .

Examples 12.1.4.
(i) Given the equation 3x+1 = 12, we can solve for x:
3x+1 = 12
ln 3x+1 = ln(12)

(x + 1) ln(3) = ln(12)
ln(12)
x= − 1 = 1.2619.
ln(3)
(ii) Given the equation etan(x) = 2, we can solve for x:
etan(x) = 2
ln etan(x) = ln(2)

tan(x) = ln(2) = 0.6931

x = 0.6061 + kπ,
where k is any integer 0, ± 1, ±2, . . . .
Example 12.1.5. If $2,000 is invested in a continuously compounding

savings account and we want the value after 12 years to be $130,000,
what is the required annual interest rate? If, instead, the same $2,000
is invested in a continuously compounding savings account with r = 6.4%
annual interest, when will the exact account value be be $130,000?
Solution. In the first scenario,
130,000 = 2,000e12r
65 = e12r
ln(65) = ln e12r

ln(65) = 12r
ln(65)
r= = 0.3479.
12
This gives an annual interest rate of 34.79%. In the second scenario, we
study the equation
130,000 = 2,000e(0.064)t
65 = e(0.064)t
ln(65)
t= = 65.22.
0.064
So, it takes over 65 years to accumulate $130,000 under the second
scheme.
12.2 Alternate form for

functions of exponential type
The standard model for an exponential function is A(t) = A0bt, for some
b > 0, b 6= 1, and A0 6= 0. Using the properties of the natural logarithm
function,
t
bt = eln(b) = et ln(b).
This means that every function as in Definition 10.1.2 can be re-written

using the exponential function et. Another way of saying this is that you
really only need the function keys “ et ” and “ ln(t) ” on your calculator.
Important Fact 12.2.1 (Observation). A function of exponential type can

be written in the form
A(t) = A0eat,
for some constants A0 6= 0 and a 6= 0.
By studying the sign of the constant a, we can determine whether the

function exhibits exponential growth or decay. For example, given the
function A(t) = eat, if a > 0 (resp. a < 0), then the function exhibits
exponential growth (resp. decay).
Examples 12.2.2.
12.3. THE INVERSE FUNCTION OF Y = BX 157
(a) The function A(t) = 200 (2t) exhibits exponential growth and can be
re-written as:
A(t) = 200 et ln(2) = 200e0.69315t

(b) The function A(t) = 4e−0.2t exhibits exponential decay and can be re-
written as:
t
A(t) = 4e−0.2t = 4 e−0.2 = 4 0.81873t .

12.3 The Inverse Function of y = bx

For some topics in Chemistry and Physics (e.g. acid base equilibria and
acoustics) it is useful to have on hand an inverse function for y = bx,
where b > 0 and b 6= 1. Just as above, we would show that f(x) = bx is
one-to-one, the range is all positive numbers and obtain the graph using
ideas in Figure 12.2. We will refer to the inverse rule as the logarithm
function base b, denoted logb(x), defined by the rule:

the unique solu-
!



 tion of the equation , if c > 0

logb(c) = c = bx


 if c ≤ 0.
 (undefined),

We will need to consider two cases, depending on the magnitude of b:

The important qualitative features of the logarithm function y = logb(x)
mirror Fact 12.1.1:
x reflecting line y = x x reflecting line y = x

y=b y=b
y=logb(x)
−1 1 −1 1
y=log (x)
b
The case b > 1 The case 0 < b < 1
Figure 12.3: Cases to consider for b.

Important Facts 12.3.1 (Graphical features of general logs). The func-

tion y = logb(x) has these features:
• The largest domain is the set of positive numbers; e.g. logb(−1) is not
defined.
• The graph has x-intercept 1 and is increasing if b > 1 (resp. decreas-

ing if 0 < b < 1).
• The graph becomes closer and closer to the vertical axis as we ap-
proach x = 0; this says the y-axis is a vertical asymptote for the
graph.
• The graph is unbounded as we move to the right.
Important Facts 12.3.2 (Log properties). Fix a positive base b, b 6= 1.
(a) For any real number x, logb(bx) = x.
(b) For any positive number x, blogb (x) = x.
(c) logb(rt) = t logb(r), for r > 0 and t any real number;
(d) logb(rs) = logb(r) + logb(s), for all r, s > 0;
(e) logb( sr ) = logb(r) − logb(s), for all r, s > 0.
It is common to simplify terminology and refer to the function logb(x)

as the log base b function, dropping the longer phrase “logarithm”. Some
scientific calculators will have a key devoted to this function. Other cal-
culators may have a key labeled “log(x)”, which is usually understood
to mean the log base 10. However, many calculators only have the key
“ln(x)”. This is not cause for alarm, since it is always possible to express
logb(x) in terms of the natural log function. Let’s see how to do this, since
it is a great application of the Log Properties listed in Fact 12.3.2.
Suppose we start with y = logb(x). We will rewrite this in terms of the
natural log by carrying out a sequence of algebraic steps below; make
sure you see why each step is justified.
y = logb(x)
by = x
ln(by) = ln(x)
y ln(b) = ln(x)
ln(x)
y= .
ln(b)
We have just verified a useful conversion formula:

12.4. MEASURING THE LOUDNESS OF SOUND 159
Important Fact 12.3.3 (Log conversion formula). For x a positive number

and b > 0, b 6= 1 a base,
ln(x)
logb(x) = .
ln(b)
For example,
ln(5)
log10(5) = = 0.699
ln(10)
ln(11)
log0.02(11) = = −0.613
ln(0.02)
ln( 12 )

1
log20 = = −0.2314
2 ln(20)
The conversion formula allows one to proceed slightly differently when

solving equations involving functions of exponential type. This is illus-
trated in the next example.
Example 12.3.4. Ten years ago, you purchased a house valued at $80,000.
Your plan is to sell the house at some point in the future, when the value
is at least $1,000,000. Assume that the future value of the house can
be computed using quarterly compounding and an annual interest rate of
4.8%. How soon can you sell the house?
Solution. We can use the future value formula to obtain the equation
4t
0.048
1,000,000 = 80,000 1 +
4
4t
12.5 = (1.012)
Using the log base b = 1.012,
log1.012(12.5) = log1.012 (1.012)4t

log1.012(12.5) = 4t
ln(12.5)
t= = 52.934.
4 ln(1.012)
Since you have already owned the house for 10 years, you would need to
wait nearly 43 years to sell at the desired price.
12.4 Measuring the Loudness of Sound

As we noted earlier, the reception of a sound wave by the ear gives rise to
a vibration of the eardrum with a definite frequency and a definite am-
plitude. This vibration may also be described in terms of the variation of
air pressure at the same point, which causes the eardrum to move. The
perception that rustling leaves and a jet aircraft sound different involves
two concepts: (1) the fact that the frequencies involved may differ; (2)
the intuitive notion of “loudness”. This loudness is directly related to the
force being exerted on the eardrum, which we refer to as the intensity
of the sound. We can try to measure the intensity using some sort of
scale. This becomes challenging, since the human ear is an amazing in-
strument, capable of hearing a large range of sound intensities. For that
reason, a logarithmic scale becomes most useful. The sound pressure
level β of a sound is defined by the equation

I
β = 10 log10 , (12.2)
I0
where I0 is an arbitrary reference intensity which is taken to correspond
with the average faintest sound which can be heard and I is the intensity
of the sound being measured. The units used for β are called decibels,
abbreviated “db”. (Historically, the units of loudness were called bels,
in honor of Alexander Graham Bell, referring to the quantity log10 II0 .)
Notice, in the case of sound of intensity I = I0, we have a sound pressure
level of

I0
β = 10 log10 = 10 log10(1) = 10(0) = 0.
I0
We refer to any sound of intensity I0 as having a sound pressure level
at the threshold of hearing. At the other end of the scale, a sound of
intensity the maximum the eardrum can tolerate has an average sound
pressure level of about 120 db. The Table 12.4(a) gives a hint of the
sound pressure levels associated to some common sounds.
Source of Noise Sound Pressure pain threshold

Level in db
Threshold of pain 120 120
Riveter 95 100
Zone of Hearing
Busy Street Traffic 70 80
60
Ordinary Conversation 65 db
40
Quiet Auto 50 20
Background Radio 40 0
Whisper 20 20 100 1000 10,000 20,000

Rustle of Leaves 10 Hz
hearing threshold
Threshold of Hearing 0
(a) Sources of noise levels. (b) Graphing noise levels.
Figure 12.4: Considering noise levels.
It turns out that the above comments on the threshold of hearing and
pain are really only averages and depend upon the frequency of the given
12.4. MEASURING THE LOUDNESS OF SOUND 161
sound. In fact, while the threshold of pain is on average close to 120

db across all frequencies between 20 Hz and 20,000 Hz, the threshold
of hearing is much more sensitive to frequency. For example, for a tone
of 20 Hz (something like the ground-shaking rumble of a passing freight
train), the sound pressure level needs to be relatively high to be heard;
100 db on average. As the frequency increases, the required sound pres-
sure level for hearing tends to drop down to 0 db around 2000 Hz. An
examination by a hearing specialist can determine the precise sensitiv-
ities of your ear across the frequency range, leading to a plot of your
“envelope of hearing”; a sample plot is given in Figure 12.4(b). Such a
plot would differ from person to person and is helpful in isolating hearing
problems.
Example 12.4.1. A loudspeaker manufacturer advertises that their model
no. 801 speaker produces a sound pressure level of 87 db when a refer-
ence test tone is applied. A competing speaker company advertises that
their model X-1 speaker produces a sound pressure level of 93 db when
fed the same test signal. What is the ratio of the two sound intensities pro-
duced by these speakers? If you wanted to find a speaker which produces
a sound of intensity twice that of the no. 801 when fed the test signal,
what is its sound pressure level?
Solution. If we let I1 and I2 refer to the sound intensities of the two speak-
ers reproducing the test signal, then we have two equations:

I1
87 = 10 log10
I
0
I2
93 = 10 log10
I0
Using log properties, we can solve the first equation for I1:

I1
87 = 10 log10 = 10 log10(I1) − 10 log10(I0)
I0
log10(I1) = 8.7 + log10(I0)
10log10 (I1 ) = 108.7+log10 (I0 )
I1 = 108.710log10 (I0 ) = 108.7I0.
Similarly, we find that I2 = 109.3I0. This means that the ratio of the inten-
sities will be
I2 109.3I0
= 8.7 = 100.6 = 3.98.
I1 10 I0
This means that the test signal on the X − 1 speaker produces a sound
pressure level nearly 4 times that of the same test signal on the no. 801
speaker.
To finish the problem, imagine a third speaker which produces a

sound pressure level β, which is twice that of the first speaker. If I3
is the corresponding intensity of the sound, then as above, I3 = 10(β/10)I0.
We are assuming that I3 = 2I1, so this gives us the equation
1
I1 = I3
2
1
108.7I0 = 10(β/10)I0
2
8.7
1 (β/10)
log10 10 = log10 10
2

1
+ log10 10(β/10)

8.7 = log10
2

β
8.7 = −0.30103 +
10
90 = β
So, the test signal on the third speaker must produce a sound pressure
level of 90 db.
12.5. EXERCISES 163
12.5 Exercises
Problem 12.1. These problems will help you (c) A rule of thumb used by many people to
develop your skills with logarithms. determine the length of time to double
an investment is the rule of 70. The rule
(a) Compute: log5 3, loge 11, log√2 π, log2 10, says it takes about t = 70r years to dou-
log10 2. ble the investment. Graphically compare
(b) Solve for x: 35 = ex , log3 x = e, log3 5 = this rule to the one isolated in part b. of
xe3 . this problem.
(c) Solve each of these equations for x in

Problem 12.5. The length of some fish are
terms of y: y = 10x , 3y = 10x , y = 103x .
modeled by a von Bertalanffy growth function.
For Pacific halibut, this function has the form
Problem 12.2. As light from the surface pene-
L(t) = 200 (1 − 0.956 e−0.18t )
trates water, its intensity is diminished. In the
clear waters of the Caribbean, the intensity is where L(t) is the length (in centimeters) of a
decreased by 15 percent for every 3 meters of fish t years old.
depth. Thus, the intensity will have the form
of a general exponential function. (a) What is the length of a new-born halibut
at birth?
(a) If the intensity of light at the water’s sur-
face is I◦ , find a formula for I(d), the in- (b) Use the formula to estimate the length
tensity of light at a depth of d meters. of a 6–year–old halibut.
Your formula should depend on I◦ and
(c) At what age would you expect the hal-
d.
ibut to be 120 cm long?
(b) At what depth will the light intensity be
(d) What is the practical (physical) signifi-
decreased to 1% of its surface intensity?
cance of the number 200 in the formula
for L(t)?
Problem 12.3. Rewrite each function in the
form y = A◦ eat , for appropriate constants A◦
Problem 12.6. A cancerous cell lacks nor-
and a.
mal biological growth regulation and can di-
(a) y = 13(3t ) vide continuously. Suppose a single mouse
skin cell is cancerous and its mitotic cell cy-
(b) y = 2( 81 )t cle (the time for the cell to divide once) is 20
(c) y = −7(1.567)t−3 hours. The number of cells at time t grows
according to an exponential model.
(d) y = −17(2.005)−t
(a) Find a formula C(t) for the number of
(e) y = 3(14.24)4t cancerous skin cells after t hours.
(b) Assume a typical mouse skin cell is
Problem 12.4. (a) If you invest Po dollars spherical of radius 50×10−4 cm. Find the
at 7% annual interest and the future combined volume of all cancerous skin
value is computed by continuous com- cells after t hours. When will the volume
pounding, how long will it take for your of cancerous cells be 1 cm3 ?
money to double?
(b) Suppose you invest Po dollars at r% an- Problem 12.7. Your Grandfather purchased
nual interest and the future value is a house for $55,000 in 1952 and it has in-
computed by continuous compounding. creased in value according to a function y =
If you want the value of the account to v(x), where x is the number of years owned.
double in 2 years, what is the required These questions probe the future value of the
interest rate? house under various mathematical models.
(a) Suppose the value of the house is (d) There also was an exponentially-growing
$75,000 in 1962. Assume v(x) is a linear population of anteaters on board. At
function. Find a formula for v(x). What the start of the voyage there were
is the value of the house in 1995? When 17 anteaters, and the population of
will the house be valued at $200,000? anteaters doubled every 2.8 weeks. How
long into the voyage were there 200 ants
(b) Suppose the value of the house is per anteater?
$75,000 in 1962 and $120,000 in 1967.
Assume v(x) is a quadratic function.
Find a formula for v(x). What is the Problem 12.10. The populations of termites
value of the house in 1995? When will and spiders in a certain house are growing ex-
the house be valued at $200,000? ponentially. The house contains 100 termites
the day you move in. After 4 days, the house
(c) Suppose the value of the house is
contains 200 termites. Three days after mov-
$75,000 in 1962. Assume v(x) is a func-
ing in, there are two times as many termites
tion of exponential type. Find a formula
as spiders. Eight days after moving in, there
for v(x). What is the value of the house
were four times as many termites as spiders.
in 1995? When will the house be valued
How long (in days) does it take the popula-
at $200,000?
tion of spiders to triple?
Problem 12.8. Solve the following equations Problem 12.11. In 1987, the population of
for x: Mexico was estimated at 82 million people,
with an annual growth rate of 2.5%. The 1987
(a) log3 (5) = log2 (x) population of the United States was estimated
at 244 million with an annual growth rate
(b) 10log2 (x) = 3
of 0.7 %. Assume that both populations are
x
(c) 35 = 7 growing exponentially.
(d) log2 (ln(x)) = 3 (a) When will Mexico double its 1987 popu-
lation?
(e) ex = 105
(b) When will the United States and Mexico
(f) 23x+5 = 32 have the same population?
Problem 12.9. A ship embarked on a long voy- Problem 12.12. The cities of Abnarca and
age. At the start of the voyage, there were 500 Bonipto have populations that are growing ex-
ants in the cargo hold of the ship. One week ponentially. In 1980, Abnarca had a popula-
into the voyage, there were 800 ants. Suppose tion of 25,000 people. In 1990, its population
the population of ants is an exponential func- was 29,000.
tion of time. Bonipto had a population of 34,000 in 1980.
The population of Bonipto doubles every 55
(a) How long did it take the population to
years.
double?
(a) How long does it take the population of
(b) How long did it take the population to
Abnarca to double?
triple?
(b) When will Abnarca’s population equal
(c) When were there be 10,000 ants on that of Bonipto?
board?
Chapter 13
Three Construction Tools
Sometimes the composition of two functions can be understood by graph-

ical manipulation. When we discussed quadratic functions and parabo-
las in the previous section, certain key graphical maneuvers were laid
out. In this section, we extend those graphical techniques to general
function graphs.
13.1 A Low-tech Exercise

This section is all about building new functions from ones
we already have in hand. This can be approached symbol- y = f(x)
ically or graphically. Let’s begin with a simple hands-on
exercise involving the curve in Figure 13.1.
By the vertical line test, we know this represents the
graph of a function y = f(x). With this picture and a piece Figure 13.1: Start with some
of bendable wire we can build an INFINITE number of new curve.
functions from the original function. Begin by making a
“model” of this graph by bending a piece of wire to the exact shape of
the graph and place it right on top of the curve. The wire model can be
manipulated in a variety of ways: slide the model back and forth horizon-
tally, up and down vertically, expand or compress the model horizontally
or vertically.
Another way to build new curves from old ones is to exploit the built in
symmetry of the xy-coordinate system. For example, imagine reflecting
the graph of y = f(x) across the x-axis or the y-axis.
In all of the above cases, we moved from the original
wire model of our function graph to a new curve that (by rotate
the vertical line test) is the graph of a new function. The
big caution in all this is that we are NOT ALLOWED to
NOT a function graph
rotate or twist the curve; this kind of maneuver does lead
Figure 13.2: Rotating a
to a new curve, but it may not be the graph of a function: curve.
See Figure 13.2.
The pictures in Figure 13.3 highlight most of what we have to say in
165
166 CHAPTER 13. THREE CONSTRUCTION TOOLS
PSfrag this section; the hard work remaining is a symbolic reinterpretation of

these graphical operations.
left right
up
down
Horizontal shift. Vertical shift.
pull push push

pull
expand horizontally compress horizontally
Horizontal expansion. Horizontal compression.
pull pull
push push
expand compress
vertically vertically
Vertical expansion. Vertical compression.
reflect across y-axis

reflect across
x-axis
Vertical reflection. Horizontal reflection.
Figure 13.3: Shifting, dilating, and reflecting y = f(x).
13.2 Reflection
In order to illustrate the technique of reflection, we will use a concrete
example:
Function: y = p(x) = 2x + 2
Domain: −2 ≤ x ≤ 2
Range: −2 ≤ y ≤ 6
As we know, the graph of y = p(x) on the domain −2 ≤ x ≤ 2 is a line of

slope 2 with y-intercept 2, as pictured in Figure 13.4(a). Now, start with
13.2. REFLECTION 167
the function equation y = p(x) = 2x + 2 and replace every occurrence of

“y” by “−y.” This produces the new equation −y = 2x + 2; or, equivalently,
y = q(x) = −2x − 2.
The domain is still −2 ≤ x ≤ 2, but the range will 8 y-axis

6
change; we obtain the new range by replacing “y” by “−y” 4
in the original range: −2 ≤ −y ≤ 6; so 2 ≥ y ≥ −6. The 2
x-axis
graph of this function is a DIFFERENT line; this one has −2 −1 −2 1 2
slope −2 and y-intercept −2. We contrast these two curves −4
−6
in Figure 13.4(b), where q(x) is graphed as the “dashed −8
line” in the same picture with the original p(x). Once we
(a) Graph of p(x) = 2x + 2.
do this, it is easy to see how the graph of q(x) is really just
the original line reflected across the x-axis. 8 y-axis
6
Next, take the original function equation y = p(x) = 4
2
2x + 2 and replace every occurrence of “x” by “−x.” This x-axis
produces a new equation y = 2(−x) + 2; or, equivalently, −2 −1 −2 1 2
−4
−6
y = r(x) = −2x + 2. −8
The domain must also be checked by replacing “x” by (b) Graph of q(x) = −2x − 2.
“−x” in the original domain condition: −2 ≤ −x ≤ 2, so 8 y-axis
2 ≥ x ≥ −2. It just so happens in this case, the domain 6
4
is unchanged. This is yet another DIFFERENT line; this 2 x-axis
one has slope −2 and y-intercept 2. We contrast these −2 −1 −2 1 2
two curves in Figure 13.4(c), where r(x) is graphed as the −4
“dashed line” in the same picture with the original p(x). −6
−8
Once we do this, it is easy to see how the graph of r(x) is
really just the original curve reflected across the y-axis. (c) Graph of r(x) = −2x + 2.
This example illustrates a general principle referred to Figure 13.4: Reflecting y =
as the reflection principle. p(x).
Important Facts 13.2.1 (Reflection). Let y = f(x) be a

function equation.
(i) We can reflect the graph across the x-axis and the re-
sulting curve is the graph of the new function obtained
by replacing “y” by “−y” in the original equation. The
domain is the same as the domain for y = f(x). If the range for y = f(x)
is c ≤ y ≤ d, then the range of −y = f(x) is c ≤ −y ≤ d. In other words,
the reflection across the x-axis is the graph of y = −f(x).
(ii) We can reflect the graph across the y-axis and the re-
sulting curve is the graph of the new function obtained
by replacing “x” by “−x” in the original equation. The
range is the same as the range for y = f(x). If the
domain for y = f(x) is a ≤ x ≤ b, then the domain
of y = f(−x) is a ≤ −x ≤ b. Using composition no-

tation, the reflection across the y-axis is the graph of
y = f(−x).
Example 13.2.2. Consider the parallelogram-shaped region R with ver-

tices (0, 2), (0, −2), (1, 0), and (−1, 0). Use the reflection principle to find
functions whose graphs bound R.
Solution. Here is a picture of the region R: First off, us-

P = (0,2) ing the two point formula for the equation of a line, we
l3 l
2 find that the line ℓ1 passing through the points P = (0, 2)
Q = (1,0) and Q = (1, 0) is the graph of the function y = f1(x) =
x-axis −2x + 2. By Fact 13.2.1 (i), ℓ2 is the graph of the equa-
l
4
l
1
tion −y = −2x + 2, which we can write as the function
Region R
y = f2(x) = 2x − 2. By Fact 13.2.1 (ii) applied to ℓ2, the line
ℓ3 is the graph of the function y = f3(x) = −2x − 2. Finally,
y-axis
by Fact 13.2.1 (i) applied to ℓ3, the line ℓ4 is the graph of
Figure 13.5: The region R. the equation −y = −2x − 2, which we can write as the
function y = f4(x) = 2x + 2.
Figure 13.6 illustrates the fact that we need to be
y-axis careful about the domain of the original function when
y = f(−x) y = f(x) using the reflection
p principle. For example, consider
y = f(x) = 1 + 1 − (x − 3)2. The largest possible domain of
−4 −2 2 4
x-axis x-values is 2 ≤ x ≤ 4 and the graph is an upper semicircle
of radius 1 centered at the point (3, 1).
−y = f(x)
Reflection
p across the x-axis gives the graph of
Figure 13.6: Reflecting the y = −1 − 1 − (x − 3)2 on the same domain; p reflection
semicircle. across the y-axis gives the graph of y = 1 + 1 − (x + 3)2
on the new domain −4 ≤ x ≤ −2.
13.3 Shifting
√
2.5 y-axis Let’s start out with the function y = f(x) = 4 − x2, which
2 has a largest possible domain −2 ≤ x ≤ 2. From Chap-
1.5 ter 6, the graph of this equation is an upper semicircle
1 of radius 2 centered at the origin (0, 0). Sliding the graph
0.5 back and forth horizontally or vertically (or both), never
x-axis
rotating or twisting, we are led to the “dashed curves” be-
−2 −1 0 1 2
low (contrasted with the original graph which is plotted
Figure 13.7: Graph of y = with a solid curve). This describes some shifted curves on
√
4 − x2 .
a pictorial level, but what are the underlying equations?
For this example, we can use the fact that all of the shifted curves are
still semicircles and Chapter 6 tells us how to find their equations.
13.3. SHIFTING 169
The lower right-hand dashed semicircle is of radiusp 2 and is centered

at (3, 0), so the corresponding equation must be y = 4 − (x − 3)2. The
upper left-hand dashed semicircle is of radius
√ 2 and centered at (0, 3), so
the corresponding equation must be y = 3+ 4 − x2. The upper right-hand
dashed semicircle is of radiusp2 and centered at (3, 3), so the correspond-
ing equation must be y = 3 + 4 − (x − 3)2.
Keeping this same example, we can continue this kind
5
of shifting more generally by thinking about the effect of
4
making
√ the following three replacements in the equation
y = 4 − x2: 3
1
x x−h
−2 −1 1 2 3 4 5
y y−k
Figure 13.8: Shifting the up-
(x and y) (x − h and y − k). per semicircle.
These substitutions lead to three new equations, each

the equation of a semicircle:

p Upper semicircle with radius
y = 4 − (x − h)2; ⇐
2 and center (h, 0).

y − k = 4 − x2; and, ⇐
2 and center (0, k).

y − k = 4 − (x − h)2. ⇐
2 and center (h, k).
There are three potentially confusing points with this

y-axis original
example: curve
• Be careful with the sign (i.e., ±) of h and k. In Fig- x-axis
ure 13.9, if h = 1, we horizontally shift the semicircle h negative h positive

1 unit to the right; whereas, if h = −1, we horizon- curve shifted curve shifted
|h| units left h units right
tally shift the semicircle −1 units to the right. But, if h is negative if h is positive
shifting −1 unit to the right is the same as shifting 1 Figure 13.9: Potentially con-
unit to the left! In other words, if h is positive, then fusing points.
a horizontal shift by h will move the graph |h| units
to the right; if h is negative, then a horizontal shift
by h will move the graph |h| units to the left.
• If k is positive, then a vertical shift by k will move

the graph |k| units up; if k is negative, then a vertical
shift by k will move the graph |k| units down. These
conventions insure that the “positivity” of h and k
match up with “rightward” and “upward” movement
of the graph.
• When shifting, the domain of allowed x-values may

change.
This example illustrates an important general principle referred to as
the shifting principle.
Important Facts 13.3.1 (Shifting). Let y = f(x) be a function equation.
(i) If we replace “x” by “x − h” in the original function equation, then
the graph of the resulting new function y = f(x − h) is obtained by
horizontally shifting the graph of f(x) by h. If h is positive, the picture
shifts to the right h units; if h is negative, the picture shifts to the left
h units. If the domain of f(x) is an interval a ≤ x ≤ b, then the domain
of f(x − h) is a ≤ x − h ≤ b. The range remains unchanged under
horizontal shifting.
(ii) If we replace “y” by “y − k” in the original function equation, then
the graph of the resulting new function y = f(x) + k is obtained by
vertically shifting the graph of f(x) by k. If k is positive, the picture
shifts upward k units; if k is negative, the picture shifts downward
k units. If the range of f(x) is an interval c ≤ y ≤ d, then the range
of f(x) + k is c ≤ y − k ≤ d. The domain remains unchanged under
vertical shifting.
13.4 Dilation
y-axis
To introduce the next graphical principle we will look at
1
0.75 the function
0.5
x
0.25
2 y = f(x) = 2 .
x +1
−6 −4 −2 −0.25 4 6
x-axis
−0.5 Using a graphing device, we have produced a plot of the
−0.75
−1
graph on the domain −6 ≤ x ≤ 6. Figure 13.10 shows the
Figure 13.10: Graph of y = curve has a high point H (like a “mountain peak”) and a
x
f(x) = x2 +1 . low point L (like a “valley”). Using a graphing device, we
can determine that the high point is H = (1, 12 ) (it lies on
the line with equation y = 21 ) and the low point is L = −1, − 12 (it lies on
the line with equation line y = − 21 ), so the range is − 12 ≤ y ≤ 12 . Draw two
new horizontal lines with equations y = 2 · ± 21 = ±1. Grab the high point
H on the curve and uniformly pull straight up, so that the high point now
lies on the horizontal line y = 1 at (1, 1). Repeat this process by pulling
L straight downward, so that the low point is now on the line y = −1
at (−1, −1). We end up with the ”stretched dashed curve” illustrated in
Figure 13.11(a). In terms of the original function equation y = x2x+1 , we
are simply describing the graphical effect of multiplying the y-coordinate
of every point on the curve by the positive number 2. In other words, the
2x x

dashed curve is the graph of y = x2 +1 = 2 x2 +1 .
13.4. DILATION 171
Next, draw the two horizontal lines with equations y = 12 · ± 12 = ± 41 .

Grab the high point H on the curve (in Figure 13.10) and uniformly
push straight down, so that the high point now lies on the horizontal
line y = 14 at (1, 41 ). Repeat this process at the low point L by pushing the
curve straight upward, so that the low point is now on the line y = − 14
at (−1, − 41 ). We end up with the new ”dashed curve” illustrated in Fig-
ure 13.11(b). In terms of the original function equation y = x2x+1 , we are
simply describing the graphical effect of multiplying the y-coordinate of
every point on the curve by the positive number 12 . In other words, the
dashed curve is the graph of y = 2(x2x+1) .
We could repeat this process systematically: 1
y-axis
0.75
0.5
• Pick a positive number c. 0.25
• Draw the two horizontal lines y = c · ± 21 . If c > 1, then −6 −4 −2 2

−0.25
4 6
x-axis
the graph of y = 2c is parallel and above the graph of −0.5
−0.75
y = 12 . On the other hand, if 0 < c < 1, then this new −1
line is parallel and below y = 21 .

(a) Vertical expansion.
• Uniformly deform the original graph (in Fig-
1 y-axis
ure 13.10) so that the new curve has it’s high and 0.75
low points just touching y = ± c2 . This will involve 111

000
000
111
0.5
000
111
0.25
vertically stretching or compressing, depending on 11 11
00 00
00 11
11
−6 00
−4 −2 111
000 2 4 6
whether 1 < c or 0 < c < 1, respectively. A number of 00 11
11 00 000
111
−0.25 x-axis
000
111
−0.5
possibilities are pictured in Figure 13.11(c). 11
00 −0.75
00
11
00
11 −1
We refer to each new dashed curve as a vertical dilation
(b) Vertical compression.
of the original (solid) curve. This example illustrates an
important principle. 1 1111
0000
0000
1111
y-axis
0.75 0000
1111
Important Facts 13.4.1 (Vertical dilation). Let c > 0 be a 0.5
0.25
positive number and y = f(x) a function equation. 00 11
11 00111
000
11
00
−6 11
00
−4 111
000
−2 2 4 6
00
11 00
11 000
111 −0.25 x-axis
y
(i) If we replace “ y ” by “ ” in the original equation, then
c
−0.5
−0.75
the graph of the resulting new equation is obtained by −1
vertical dilation of the graph of y = f(x). The domain

(c) Many possibilities.
of x-values is not affected.
Figure 13.11: Dilating y =
(ii) If c > 1, then the graph of yc = f(x) (i.e., y = cf(x)) is a f(x).
vertically stretched version of the original graph.
(iii) If 0 < c < 1, then the graph of yc = f(x) (i.e., y = cf(x)) is

a vertically compressed version of the original graph.
If we combine dilation with reflection across the x-axis,

we can determine the graphical relationship between y =
f(x) and y = cf(x), for any constant c. The key observation
is that reflection across the x-axis corresponds to the case c = −1.
Example 13.4.2. Describe the relationship between the graphs of

p
y = f(x) 1 − (x + 1)2,
=
p
y = −f(x) = − 1 − (x + 1)2, and
p
y = −4f(x) = −4 1 − (x + 1)2.
step 1: start with upper Solution. The graph of y = f(x) is an upper semicircle of
semicircle
radius 1 centered at the point (−1,0). To obtain the pic-
ture of the graph of y = −4f(x), we first reflect y = f(x)
x-axis across the x-axis; this gives us the graph of y = −f(x).
step 2: reflect across Then, we vertically dilate this picture by a factor of c = 4
step 3: stretch curve from
to get the graph of y4 = −f(x), which is the same as the
step 2: to get
y = −4f(x) graph of the equation y = −4f(x). See Figure 13.12.
Figure 13.12: Reflecting and Let’s return to the original example y = x2x+1 and in-
dilating a lower semicircle.
vestigate a different type of dilation where the action is
taking place in the horizontal direction (whereas it was in
the vertical direction before). Grab the right-hand end of the graph (in
Figure 13.10) and pull to the right, while at the same time pulling the
left-hand end to the left. We can quantify this by stipulating that the
1 1

high point H = 1, 2 of the original curve moves to the new location 2, 2
and the low point L = −1, − 12 moves to the new location −2, − 21 .

1 y-axis 1 y-axis
0.75 0.75
0.5 0.5
0.25 0.25
11
00
00
11 11
00
00
11 00
11
−6 −4 −2 2 4 6 −6 11
00
−4 −2 2 4 6
00
11 −0.25 x-axis 00
11 00
11 −0.25 x-axis
−0.5 −0.5
−0.75 −0.75
−1 −1
(a) Stretching. (b) Compressing.

Figure 13.13: Horizontally dilating y = x2x+1 .
The result will be somewhat analogous to stretching a spring. By the

same token, we could push the right-hand end to the left and push the
left-hand end to the right, like compressing a spring. We can quantify this
by stipulating that the high point H = 1, 21 of the original curve moves to
1 1 1

the new location 2 , 2 and the low point L = −1, − 2 moves to the new
location − 21 , − 21 . These two situations are indicated in Figure 13.13. We
refer to each of the dashed curves as a horizontal dilation of the original
(solid) curve.
13.5. VERTEX FORM AND ORDER OF OPERATIONS 173
The tricky point is to understand what happens on the level of the

original equation. In the case of the stretched graph in Figure 13.13(a),
you can use a graphing device to verify that this looks like the graph
x/2
of y = (x/2)2 +1 ; in other words, we replaced “x” by “x/2” in the original
equation. In the case of the compressed graph in Figure 13.13(b), you can
use a graphing device to verify that this looks like the graph of y = (2x)2x2 +1 ;
x
in other words, we replaced “x” by “ 1/2 = 2x” in the original equation.
The process just described leads to a general principle.
Important Facts 13.4.3 (Horizontal dilation). Let c > 0 be a positive
number and y = f(x) a function equation.
(i) If we replace “ x” by “ xc ” in the original function equation, then the
x

graph of the resulting new function y = f c is obtained by a horizon-
= f(x). If the domain of f(x) is a ≤ x ≤ b,
tal dilation of the graph of y
then the domain of y = f xc is a ≤ xc ≤ b.
(ii) If c > 1, then the graph of y = f xc is a horizontal stretch.

(iii) If 0 < c < 1, then the graph of y = f xc is a horizontal compression.

13.5 Vertex Form and Order of Operations

Using the language of function compositions we can clarify our discus-
sion in Example 7.1.2. Let’s revisit that example:
Example 13.5.1. The problem is to describe a sequence of geometric ma-
neuvers that transform the graph of y = x2 into the graph of y = −3(x − 1)2 + 2.
Solution. The idea is to rewrite y = −3(x−1)2 +2 as a composition of y = x2
with four other functions, each of which corresponds to a horizontal shift,
vertical shift, reflection or dilation. Once we have done this, we can
read off the order of geometric operations using the order of composition.
Along the way, pay special attention to the exact order in which we will
be composing our functions; this will make a big difference.
To begin with, we can isolate four key numbers in the equation:
Reflect Horizontal shift by h = 1

z}|{ z}|{
− 3
|{z} (x − 1 )2 + 2
|{z}
Dilate by 3 Vertical shift by k = 2
We want to use each number to define a new function, then compose

these in the correct order. We will also give our starting function y = x2 a
specific name to make things definite:
f(x) = x2 h(x) = x − 1
v(x) = x + 2 r(x) = −x
d(x) = 3x.
Now, verify that

    
h i

v r d f h(x)  = v r d f[x − 1]
  

2
= v r d{(x − 1) }
h i
2
= v r(3(x − 1) )
= v[−3(x − 1)2]
= −3(x − 1)2 + 2
= −3x2 + 6x − 1.
13.6 Summary of Rules

For quick reference, we summarize the consequence of
shifting and expanding symbolically and pictorially. The
2 running example for Tables 13.1, 13.2, and 13.3 will be
a multipart function y = f(x) whose graph, seen in Fig-
ure 13.14, consists of a line segment and a quarter circle
on the domain −2 ≤ x ≤ 2:
−2 2
Figure 13.14: A multipart

function. x+2
√ if −2 ≤ x ≤ 0
f(x) =
4 − x if 0 ≤ x ≤ 2
2
13.6. SUMMARY OF RULES 175
Reflection
Symbolic New Equa- Graphical
Picture
Change tion Consequence
A reflection 2
Replace x with
y = f(−x) across the
−x.
y-axis. −2 2
−2 2
Replace f(x) A reflection
y = −f(x) across the
with −f(x).
x-axis. −2
Table 13.1: Reflecting y = f(x).

Shifting (Assume c > 0 )

Picture
Change tion Consequence
2
Replace x with A shift to the
y = f(x − c)
(x − c). right c units.
2 4
2
Replace x with A shift to the
y = f(x + c)
(x + c). left c units.
−4 −2
Replace f(x) A shift up c

y = f(x) + c 2
with (f(x) + c). units.
−2 2
Replace f(x) A shift down c −2 2

y = f(x) − c
with (f(x) − c). units.
2
Table 13.2: Shifting y = f(x).

13.6. SUMMARY OF RULES 177
Dilation
Change tion Consequence Picture
4
c=2
x 3
If c > 1, replace A horizontal 2
y=f
x with xc .

c expansion. 1
−4−3−2−1 0 1 2 3 4
1 4
c= 2
x 3
If 0 < c < 1, re- A horizontal
y=f 2
place x with xc .

c compression. 1
−4 −3−2−1 1
0
0
1
0 1 2 3 4
0
1
11
00
00
11
4
00
11
c=2 00
11
00
11
3
If c > 1, replace A vertical 2
y = cf(x)
f(x) with (cf(x)). expansion. 1
−4−3−2−1 11
00
00
11
0 1 2 3 4
00
11
1 4
c=
If 0 < c < 1, re- 2
3
A vertical
place f(x) with y = cf(x) 2
compression. 0
1
0
1
1
0
1
(cf(x)). 0
1
−4−3−2−1 11
00
00
11
0 1 2 3 4
00
11
Table 13.3: Dilating y = f(x).

13.7 Exercises
Problem 13.1. On a single set of axes, sketch y
a picture of the graphs
√ of the √
following four
√
equations: y = −x+ 2, y = −x− 2, y = x+ 2,
π
√
and y = x − 2. These equations determine
lines, which in turn bound a diamond shaped
region in the plane.
f(x)
π/2
(a) Show that the unit circle sits inside this
diamond tangentially; i.e. show that the
unit circle intersects each of the four
lines exactly once.
(b) Find the intersection points between the
unit circle and each of the four lines.
x
(c) Construct a diamond shaped region in −1 1
which the circle of radius 1 centered at
(−2, − 1) sits tangentially. Use the tech-
niques of this section to help.
Problem 13.2. The graph of a function y = f(x)

is pictured with domain −2.5 ≤ x ≤ 3.5. Problem 13.4. (a) Each of the six functions
y = f(x) below can be written in the
“standard form”
y = A|B(x − C)| + D,
for some constants A,B,C,D. Find these

constants, describe the precise order of
graphical operations involved in going
from the graph of y = |x| to the graph of
y = f(x) (paying close attention to the or-
der), write out the multipart rule, sketch
the graph and calculate the coordinates
of the “vertex” of the graph.
(a1) f(x) = |x − 2|
(a2) f(x) = 2|x + 3|
Sketch the graph of each of the new func-
(a3) f(x) = |2x − 1|
tions listed below.
(a) g(x) = 2f(x + 1) (a4) f(x) = |2(x − 1)|
(b) h(x) = 12 f(2x − 1) (a5) f(x) = 3|2x − 1| + 5
(c) j(x) = 5f( 13 x + 2) − 2 (a6) f(x) = −2|x + 3| − 1
(b) Solve the following inequalities using

Problem 13.3. The graph of a function y = f(x)
your work in the previous part of this
is pictured with domain −1 ≤ x ≤ 1. Sketch the
problem:
graph of the new function
1 (b1) |x − 2| ≤ 3
y = g(x) = f(3x) − 0.5.
π
(b2) 1 ≤ 2|x + 3| ≤ 5
Find theplargest possible domain of the func-
tion y = g(x). (b3) y = 3|2x − 1| + 5 ≥ 10
13.7. EXERCISES 179
10 y-axis
(c) The graphs of y = 3|2x − 1| + 5
and y = −|x − 3| + 10 intersect to form a
8
bounded region of the plane. Find the
vertices of this region and sketch a pic- 6
ture.
4
Problem 13.5. Consider the function y = f(x)
2
with multipart definition
x-axis


 0 if x ≤ −1 2 4 6 8 10

2x + 2 if −1 ≤ x ≤ 0
f(x) =

 −x + 2 if 0 ≤ x ≤ 2 (c) Assume that the maximum value of the

0 if x ≥ 2 function a(x) in (a) occurs when x = 4.
Find the maximum value of z = a(x) and
(a) Sketch the graph of y = f(x).
z = 2a(3x + 3) + 1.
(b) Is y = f(x) an even function? Is y = f(x)
an odd function? (A function y = f(x) is
called even if f(x) = f(−x) for all x in the Problem 13.7. Describe how each graph dif-
domain. A function y = f(x) is called odd fers from that of y = x2 .
if f(−x) = −f(x) for all x in the domain.)
(a) y = 2x2
(c) Sketch the reflection of the graph across
the x-axis and y-axis. Obtain the re- (b) y = x2 − 5
sulting multipart equations for these re- (c) y = (x − 4)2
flected curves.
(d) y = (3x − 12)2
(d) Sketch the vertical dilations y = 2f(x)
and y = 12 f(x). (e) y = 2(3x − 12)2 − 5
(e) Sketch the horizontal dilations y = f(2x)
and y = f( 12 x). Problem 13.8. In each case, start with the
(f) Find a number c > 0 so that the highest function y = |x| and perform the operations
point on the graph of the vertical dilation described to the graph, in the order specified.
y = cf(x) has y-coordinate 11. Write out the resulting rule for the function
and sketch the final graph you obtain.
(g) Using horizontal dilation, find a number
c > 0 so that the function values f( xc ) are (a) (1) horizontally compress by a factor of
non-zero for all − 25 < x < 5. 2; (2) horizontally shift to the left by 2;
(3) vertically stretch by a factor of 7; (4)
(h) Using horizontal dilation, find positive
vertically shift up 2 units.
numbers c,d > 0 so that the function
values f( 1c (x − d)) are non-zero precisely (b) (1) horizontally stretch by a factor of 2;
when 0 < x < 1. (2) horizontally shift to the right by 2; (3)
vertically compress by a factor of 7; (4)
vertically shift down 2 units.
Problem 13.6. An isosceles triangle has sides
of length x, x and y. In addition, assume the (c) (1) horizontally shift to the right 2 units;
triangle has perimeter 12. (2) horizontally compress by a factor of
3.
(a) Find the rule for a function that com-
putes the area of the triangle as a func-
tion of x. Describe the largest possible
Problem 13.9. (a) Begin with the function
domain of this function.
y = f(x) = 2x .
(b) The graph of z = a(x) from part (a) is
given below. Sketch the graph and find (a1) Rewrite each of the following func-
the rule for the function z = 2a(3x+3)+1; tions in standard exponential form:
make sure to specify the domain and f(2x), f(x − 1), f(2x − 1), f(2(x − 1)),
range of this new function. 3f(x), 3f(2(x − 1)).
(a2) Is the function 3f(2(x−1))+1 a func- cal dilation, horizontal shifting, horizon-
tion of exponential type? tal dilation) to the standard exponential
(a3) Sketch the graphs of f(x), f(2x), f(2(x− model y = Ao bx . For which of the four
1)), 3f(2(x − 1)) and 3f(2(x − 1)) + 1 operations is the resulting function still
in the same coordinate system a standard exponential model?
and explain which graphical op-
eration(s) (vertical shifting, vertical
Problem 13.10. Begin with a sketch of the
dilation, horizontal shifting, hori-
graph of the function y = 2x on the domain
zontal dilation) have been carried
of all real numbers. Describe how to use the
out.
“four tools” of Chapter 13 to obtain the graphs
(b) In general, explain what happens when of these functions: y = −2x , y = 2−x , y = 3(2x ),
you apply the four construction tools y = 31 (2x ), y = 3 + 2x , y = 2x − 2, y = 2x−2 ,
of Chapter 13 (vertical shifting, verti- y = 2x+2 , y = 23x , y = 2x/3 .
Chapter 14
Rational Functions
A rational function is a function of the form f(x) = p(x)

q(x)
where p(x) and q(x)
are polynomials. For example, the following are all rational functions.
x x2 + 1 4x5 − 4x2 − 8 x6
f(x) = g(x) = h(x) = 3 j(x) = 1
3x + 4 3x − 5 x + x2 − x + 1 x8 + 5x − 2
There is a huge variety of rational functions. In this course, we will

concentrate our efforts exclusively toward understanding the simplest
type of rational functions: linear-to-linear rational functions. Linear-
to-linear rational functions are rational functions in which the numera-
tor and the denominator are both linear polynomials. The following are
linear-to-linear-rational functions.
x 5x − 6 0.34x − 0.113 4x + 43
k(x) = m(x) = n(x) = p(x) = 5
3x + 4 2x + 1 x−1 8
x − 1.117
The simplest example of a linear-to-linear rational

function is f(x) = x1 whose graph is shown in Figure 14.1. 10
This is an important example for the study of this class of

functions, as we shall see. 5
Let’s consider the graph of this function, f(x) = x1 . We

first begin by considering the domain of f. Since the nu-
merator of x1 is a constant, and the denominator is just x, -10 -5
0
0 5 10
the only way we can run into difficulty when evaluating

this function is if we try to divide by zero; that is, the only -5
value of x not in the domain of this function is zero. The

function is defined for all x except x = 0. At x = 0, there -10
must be a gap, or hole, in the graph. Figure 14.1: The graph of

To get the graph started we might simply give ourselves f(x) = x1 .
a point on the graph. For instance, we see that f(1) =
1/1 = 1, so the point (1,1) is on the graph. Then, if we use larger values of
x, we see that 1/x becomes smaller as x grows. For instance, f(2) = 1/2,
f(10) = 1/10, and f(1000) = 1/1000. In addition, we see that there is no
limit to how small we can make 1/x simply by taking x big enough. Want
181
182 CHAPTER 14. RATIONAL FUNCTIONS
1/x to be less than 0.001? Just pick x bigger than 1000. Want 1/x to be
less than 0.000001? Just use x bigger than 1,000,000.
What this means graphically is that as x gets bigger (starting from
x = 1), the curve y = 1/x gets closer and closer to the x-axis. As a result,
we say that the x-axis is a horizontal asymptote for this function.
We see the same behavior for negative values of x. If x is large, and
negative (think -1000, or -1000000), then 1/x is very small (i.e., close to
zero), and it gets smaller the larger x becomes. Graphically, this means
that as x gets large in the negative direction, the curve y = 1/x gets closer
and closer to the x-axis. We say that, in both the positive and negative
directions, y = 1/x is asymptotic to the x-axis.
A similar thing happens when we consider x near zero. If x is a small
positive number (think 1/2, or 1/10, or 1/10000), then 1/x is a large
positive number. What’s more, if we think of x as getting closer to zero,
1/x gets bigger and bigger. Plus, there is no bound on how big we can
make 1/x simply by taking x as close to zero as we need to.
For instance, can 1/x be as big as 10000? All you need to do is make
x smaller than 1/10000.
Graphically, what this means is that as x approaches zero from the
positive side, the y value gets larger and larger. As a result, the curve
approaches the y-axis as y gets larger. We say that the y − axis is a
vertical asymptote for the curve y = 1/x.
We see the same phenomenon as x approaches zero from the negative
side: y = 1/x gets larger in the negative direction (i.e., it gets more and
more negative). The curve gets closer and closer to the negative y-axis as
y becomes more and more negative. Again, we say that the y − axis is a
vertical asymptote for the curve y = 1/x.
It turns out that every linear-to-linear rational function has a graph
that looks essentially the same as the graph of y = 1/x. Let’s see why.
Consider the linaer-to-linear rational function f(x) = ax+b
cx+d
. If we divide
cx + d into ax + b, the result is
ax + b a b − ad
c
f(x) = = +
cx + d c cx + d
which we can rewrite as

a ad 1 a ad 1 a bc − ad 1
f(x) = + b − · = + b− · d
= + · .
c c cx + d c c c(x + c ) c c 2
x + dc
If we now let

a bc − ad d
A= ,B= , and C = ,
c c2 c
then we can write
1
f(x) = A + B .
x+C
183
If we let g(x) = 1/x then we have shown that f(x) = A + Bg(x + C), and so
the graph of the function f is just a horizontally shifted, vertically shifted
and vertically dilated version of the graph of g. Also, if B turned out to be
negative the graph would be vertically flipped, too.
Why is that useful? It means that the graph of a linear-to-linear ra-
tional function can only take one of two forms. Either it looks like this:
Or like this:
We can sketch an accurate graph of a linear-to-linear rational function

by sketching the asymptotes and then sketching in just one point on the
graph. That will be enough information to nail down a decent sketch. We
can always plot more points to give us more precision, but one point is
enough to capture the essence of the graph.
Given a linear-to-linear function that we wish to graph, we must first
determine the asymptotes. There will be one horizontal asymptote and
one vertical asymptote.
The vertical asymptote will be a vertical line with equation x = k where
k is the one x value which is not in the domain of the function. That is,
find the value of x which makes the denominator zero and that will tell
you the vertical asymptote.
The horizontal asymptote is a little more involved. However, we can
quickly get to a shortcut. The essence of a horizontal asymptote is that it
describes what value the function is approximately equal to for very large
values of x. To study what a linear-to-linear function is like when x is
very large, we can perform the following algebraic manipulation:
ax + b ax + b 1/x a + bx
f(x) = = · =
cx + d cx + d 1/x c + dx
While dividing by x is troublesome if x equals zero, here we are assuming

x is very large, so it is certainly not zero.
Now, consider this last expression in the above equation. If x is very
large, then
b
≈0
x
(where ≈ means ”is approximately”) and likewise
d
≈ 0.
x
Hence, when x is very large,
a+0 a
f(x) ≈ = .
c+0 c
We can interpret this by saying that when x is very large, the function
f(x) is is close to a constant, and that constant is ac . Thus, the horizontal
ax + b
asymptote of f(x) = is the horizontal line y = ac .
cx + d
3x−1
Example 14.0.1. Sketch the graph of the function f(x) = 2x+7
.
Solution. We begin by finding the asymptotes of f.

The denominator is equal to zero when 2x + 7 = 0, i.e., when x = −7/2.
As a result, the vertical asymptote for this function is the vertical line
x = −7/2.
By taking the ratio of the coefficients of x in the numerator and de-
nominator, we can find that the horizontal asymptote is the horizontal
line
3
y= .
2
10
0
-10 -5 0 5 10
-5
-10
3x−1
Figure 14.2: The graph of f(x) = 2x+7 .
We then sketch these two asymptotes. The last thing we need is a

single point. For instance, we may evaluate f(0):
−1
f(0) =
7
14.1. MODELING WITH LINEAR-TO-LINEAR RATIONAL FUNCTIONS185
and so the point (0, −1/7) is on the graph. With this information, we know
that the curve lies below the horizontal asymptote to the right of the
vertical asymptote, and consequently the curve lies above the horizontal
asymptote to the left of the vertical asymptote.
We graph the result in Figure 14.2.
14.1 Modeling with Linear-to-linear Rational

Functions
As we have done with other sorts of functions, such as linear and quadratic,
we can also model using linear-to-linear rational functions. One reason
for using this type of function is their asymptotic nature. Many chang-
ing quantities in the world continually increase or decrease, but with
bounds on how large or small they can get. For instance, a population
may steadily decrease, but a population can never be negative. Con-
versely, a population may steadily increase, but due to environmental
and other factors we may hypothesize that the population will always
stay below some upper bound. As a result, the population may ”level
off”. This leveling off behavior is exemplified by the asymptotic nature of
the linear-to-linear rational functions, and so this type of function pro-
vides a way to model such behavior.
Given any linear-to-linear rational function, we can always divide the
numerator and the denominator by the coefficient of x in the denomina-
tor. In this way, we can always assume that the coefficient of x in the
denominator of a function we seek is 1. This is illustrated in the next
example.
2x+3
Example 14.1.1. Let f(x) = 5x−7
. Then
1
2x + 3 5
f(x) = · 1
5x − 7 5
2
5
+ 53
x
= .
x − 75
Thus, in general, when we seek a linear-to-linear rational function,

we will be looking for a function of the form
ax + b
f(x) =
x+c
and thus there are three parameters we need to determine.
Note that for a function of this form, the horizontal asymptote is y = a
and the vertical asymptote is x = −c.
Since these functions have three parameters (i.e., a, b and c), we will
need three pieces of information to nail down the function.
There are essentially three types of modeling problems that require the
determination of a linear-to-linear function. The three types are based
on the kind of information given about the function. The three types are:
1. You know three points the the graph of the function passes through;
2. You know one of the function’s asymptotes and two points the graph
passes through;
3. You know both asymptotes and one point the graph passes through.
Notice that in all cases you know three pieces of information. Since
a linear-to-linear function is determined by three parameters, this is ex-
actly the amount of information needed to determine the function.
The worst case, in terms of the amount of algebra you need to do, is
the first case. Let’s look at an example of the algebra involved with this
sort.
Example 14.1.2. Find the linear-to-linear rational function f(x) such that
f(10) = 20, f(20) = 32 and f(25) = 36.
Solution. Since f(x) is a linear-to-linear rational function, we know
ax + b
f(x) =
x+c
for constants a, b, and c. We need to find a, b and c.
We know three things.
First, f(10) = 20. So
10a + b
f(10) = = 20,
10 + c
10a + b = 200 + 20c. (14.1)
Second, f(20) = 32. So
20a + b
f(20) = = 32,
20 + c
20a + b = 640 + 32c. (14.2)
Third, f(25) = 36. So
25a + b
f(25) = = 36,
25 + c
14.1. MODELING WITH LINEAR-TO-LINEAR RATIONAL FUNCTIONS187
25a + b = 900 + 36c. (14.3)
These three numbered equations are enough algebraic material to

solve for a, b, and c. Here is one way to do that.
Subtract equation 14.1 from equation 14.2 to get
10a = 440 + 12c (14.4)
and subtract equation 14.2 from equation 14.3 to get
5a = 260 + 4c (14.5)
Note that we’ve eliminated b. Now multiply this last equation by 2 to

get
10a = 520 + 8c
Subtract equation 14.4 from this to get
0 = 80 − 4c
which easily give us c = 20.

Plugging this value into equation 14.4, we can find a = 68, and then
we can find b = −80.
Thus,
68x − 80
f(x) = .
x + 20
We can check that we have done the algebra correctly by evaluating
f(x) at x = 10, x = 20 and x = 25. If we get f(10) = 20, f(20) = 32 and
f(25) = 36, then we’ll know our work is correct.
Algebraically, this was the worst situation of the three, since it re-
quired the most algebra. If, instead of knowing three points, we know
one or both of the asymptotes, then we can easily find a and/or c, and
so cut down on the amount of algebra needed. However, the method is
essentially identical.
Let’s now apply these ideas to a real world problem.
Example 14.1.3. Clyde makes extra money selling tickets in front of the
Safeco Field. The amount he charges for a ticket depends on how many he
has. If he only has one ticket, he charges $100 for it. If he has 10 tickets,
he charges $80 a piece. But if he has a large number of tickets, he will
sell them for $50 each. How much will he charge for a ticket if he holds 20
tickets?
Solution. We want to give a linear-to-linear rational function relating the

price of a ticket y to the number of tickets x that Clyde is holding. As we
saw above, we can assume the function is of the form
ax + b
y=
x+c
where a, b and c are numbers. Note that y = a is the horizontal asymp-
tote. When x is very large, y is close to 50. This means the line y = 50 is
a horizontal asymptote. Thus a = 50 and
50x + b
y= .
x+c
Next we plug in the point (1,100) to get a linear equation in b and c.
50 · 1 + b
100 =
1+c
100 · (1 + c) = 50 + b
50 = b − 100c
Similarly, plugging in (10,80) and doing a little algebra (do it now!)
gives another linear equation 300 = b − 80c. Solving these two linear
equations simultaneously gives c = 12.5 and b = 1300. Thus our function
is
50x + 1300
y=
x + 12.5
and, if Clyde holds 20 tickets, he will charge
50 · 20 + 1300
y= = $70.77
20 + 12.5
per ticket.
14.2 Summary
• Every linear-to-linear rational function has a graph which is a shifted,
scaled version of the curve y = 1/x. As a result, they have one verti-
cal asymptote, and one horizontal asymptote.
• Every linear-to-linear rational function f can be expressed in the

form
ax + b
f(x) = .
x+c
This function has horizontal asymptote y = a and vertical asymp-
tote x = −c.
14.3. EXERCISES 189
14.3 Exercises
Problem 14.1. Give the domain of each of (e) In the long run, what will be the ratio of
the following functions. Find the x- and y- the prices of the ukuleles?
intercepts of each function. Sketch a graph
and indicate any vertical or horizontal asymp- Problem 14.4. Isobel is producing and selling
totes. Give equations for the asymptotes. casette tapes of her rock band. When she had
(a) f(x) = 2x
(b) g(x) = 3x+2 sold 10 tapes, her net profit was $6. When she
x−1 2x−5
x+1 had sold 20 tapes, however, her net profit had
(c) h(x) = x−2 (d) j(x) = 4x−12
x+8 shrunk to $4 due to increased production ex-
8x+16 9x+24 penses. But when she had sold 30 tapes, her
(e) k(x) = 5x− 1
(f) m(x) = 35x−100
2
net profit had rebounded to $8.
Problem 14.2. Oscar is hunting magnetic (a) Give a quadratic model relating Isobel’s
fields with his gauss meter, a device for mea- net profit y to the number of tapes sold
suring the strength and polarity of magnetic x.
fields. The reading on the meter will increase (b) Divide the profit function in part (a) by
as Oscar gets closer to a magnet. Oscar is in the number of tapes sold x to get a model
a long hallway at the end of which is a room relating average profit w per tape to the
containing an extremely strong magnet. When number of tapes sold.
he is far down the hallway from the room, the
(c) How many tapes must she sell in order
meter reads a level of 0.2. He then walks down
to make $1.20 per tape in net profit?
the hallway and enters the room. When he has
gone 6 feet into the room, the meter reads 2.3.
Problem 14.5. Find the linear-to-linear func-
Eight feet into the room, the meter reads 4.4.
tion whose graph passes through the points
(a) Give a linear-to-linear rational model re- (1,1), (5,2) and (20,3). What is its horizontal
lating the meter reading y to how many asymptote?
feet x Oscar has gone into the room.
Problem 14.6. Find the linear-to-linear func-
(b) How far must he go for the meter to tion whose graph has y = 6 as a horizontal
reach 10? 100? asymptote and passes through (0,10) and (3,7).
(c) Considering your function from part (a)
and the results of part (b), how far into Problem 14.7. The more you study for a cer-
the room do you think the magnet is?
tain exam, the better your performance on it.
If you study for 10 hours, your score will be
65%. If you study for 20 hours, your score will
Problem 14.3. In 1975 I bought an old Mar-
be 95%. You can get as close as you want to
tin ukulele for $200. In 1995 a similar uke
a perfect score just by studying long enough.
was selling for $900. In 1980 I bought a new
Assume your percentage score is a linear-to-
Kamaka uke for $100. In 1990 I sold it for
linear function of the number of hours that
$400.
you study.
(a) Give a linear model relating the price p of If you want a score of 80%, how long do
the Martin uke to the year t. Take t = 0 you need to study?
in 1975.
Problem 14.8. A street light is 10 feet North of
(b) Give a linear model relating the price q a straight bike path that runs East-West. Olav
of the Kamaka uke to the year t. Again is bicycling down the path at a rate of 15 MPH.
take t = 0 in 1975. At noon, Olav is 33 feet West of the point on
(c) When is the value of the Martin twice the the bike path closest to the street light. (See
value of the Kamaka? the picture). The relationship between the in-
tensity C of light (in candlepower) and the dis-
(d) Give a function f(t) which gives the ratio tance d (in feet) from the light source is given
of the price of the Martin to the price of by C = dk2 , where k is a constant depending on
the Kamaka. the light source.
(a) From 20 feet away, the street light has Problem 14.10. The number of customers
an intensity of 1 candle. What is k? in a local dive shop depends on the amount
(b) Find a function which gives the intensity of money spent on advertising. If the shop
of the light shining on Olav as a function spends nothing on advertising, there will be
of time, in seconds. 100 customers/day. If the shop spends
$100, there will be 200 customers/day. As
(c) When will the light on Olav have maxi- the amount spent on advertising increases,
mum intensity? the number of customers/day increases and
(d) When will the intensity of the light be 2 approaches (but never exceeds) 400 cus-
candles? tomers/day.
(Top View−looking down)
(a) Find a linear to linear rational function
y = f(x) that calculates the number y of
10ft
olav customers/day if $x is spent on adver-
path tising.
33ft
(b) How much must the shop spend on ad-
vertising to have 300 customers/day.
Problem 14.9. For each of the following find
the linear to linear function f(x) satisfying the (c) Sketch the graph of the function y = f(x)
given requirements: on the domain 0 ≤ x ≤ 5000.
(a) f(0) = 0, f(10) = 10, f(20) = 15
(d) Find the rule, domain and range for the
(b) f(0) = 10, f(5) = 4, f(20) = 3 inverse function from part (c). Explain
(c) f(10) = 20, f(30) = 25, and the graph of in words what the inverse function cal-
f(x) has y = 30 as its horizontal asymp- culates.
tote
Chapter 15
Measuring an Angle
So far, the equations we have studied have an algebraic Cosmo

S
character, involving the variables x and y, arithmetic op- motors
around
erations and maybe extraction of roots. Restricting our Q the
attention to such equations would limit our ability to de- circle
P R
scribe certain natural phenomena. An important example 20 feet
involves understanding motion around a circle, and it can

be motivated by analyzing a very simple scenario: Cosmo
the dog, tied by a 20 foot long tether to a post, begins Figure 15.1: Cosmo the dog
walking around a circle. A number of very natural ques- walking a circular path.
tions arise:
Natural Questions 15.0.1. How can we measure the angles ∠SPR, ∠QPR, and
∠QPS? How can we measure the arc lengths arc(RS), arc(SQ) and arc(RQ)? How
can we measure the rate Cosmo is moving around the circle? If we know how to
measure angles, can we compute the coordinates of R, S, and Q? Turning this
around, if we know how to compute the coordinates of R, S, and Q, can we then
measure the angles ∠SPR, ∠QPR, and ∠QPS ? Finally, how can we specify the
direction Cosmo is traveling?
We will answer all of these questions and see how the theory which
evolves can be applied to a variety of problems. The definition and basic
properties of the circular functions will emerge as a central theme in this
Chapter. The full problem-solving power of these functions will become
apparent in our discussion of sinusoidal functions in Chapter 19.
The xy-coordinate system is well equipped to study straight line mo-
tion between two locations. For problems of this sort, the important
tool is the distance formula. However, as Cosmo has illustrated, not all
two-dimensional motion is along a straight line. In this section, we will
describe how to calculate length along a circular arc, which requires a
quick review of angle measurement.
191
192 CHAPTER 15. MEASURING AN ANGLE
15.1 Standard and Central Angles
An angle is the union of two rays emanating from a common point called
the vertex of the angle. A typical angle can be dynamically generated
by rotating a single ray from one position to another, sweeping counter-
clockwise or clockwise : See Figure 15.2. We often insert a curved arrow
to indicate the direction in which we are sweeping out the angle. The ray
ℓ1 is called the initial side and ℓ2 the terminal side of the angle ∠AOB.
(terminal side)
l2
B
SWEEP CLOCKWISE
vertex vertex A (initial side)

l1 l1 l1
O A O A (initial side) O
(terminal side)
START SWEEP COUNTERCLOCKWISE B
l2
Figure 15.2: Angle ∠AOB.
Working with angles, we need to agree on a standard frame of refer-

ence for viewing them. Within the usual xy-coordinate system, imagine a
model of ∠AOB in Figure 15.2 constructed from two pieces of rigid wire,
welded at the vertex. Sliding this model around inside the xy-plane will
not distort its shape, only its position relative to the coordinate axis. So,
we can slide the angle into position so that the initial side is coincident
with the positive x-axis and the vertex is the origin. Whenever we do
this, we say the angle is in standard position. Once an angle is in stan-
dard position, we can construct a circle centered at the origin and view
our standard angle as cutting out a particular “pie shaped wedge” of the
corresponding disc.
Notice, the shaded regions in Figure 15.3 depend on whether we sweep
the angle counterclockwise or clockwise from the initial side. The portion
of the “pie wedge” along the circle edge, which is an arc, is called the arc
subtended by the angle. We can keep track of this arc using the notation
arc(AB). A central angle is any angle with vertex at the center of a circle,
but its initial side may or may not be the positive x-axis. For example,
∠QPS in the Figure beginning this Chapter is a central angle which is not
in standard position.
15.2. AN ANALOGY 193
y-axis l2 l2
y-axis
B B
arc subtended
arc(AB)
arc(AB)
vertex vertex
x-axis x-axis
O l1 O l1
A A
COUNTERCLOCKWISE CLOCKWISE
Figure 15.3: Standard angles and arcs.
15.2 An Analogy
To measure the dimensions of a box you would use a ruler. In other
words, you use an instrument (the ruler) as a standard against which
you measure the box. The ruler would most likely be divided up into
either English units (inches) or metric units (centimeters), so we could
express the dimensions in a couple of different ways, depending on the
units desired.
By analogy, to measure the size of an angle, we need a standard
against which any angle can be compared. In this section, we will de-
scribe two standards commonly used: the degree method and the radian
method of angle measurement. The key idea is this: Beginning with a
circular region, describe how to construct a “basic” pie shaped wedge
whose interior angle becomes the standard unit of angle measurement.
15.3 Degree Method

Begin by drawing a circle of radius r, call it Cr, centered at the origin.
Divide this circle into 360 equal sized pie shaped wedges, beginning with
the the point (r,0) on the circle; i.e. the place where the circle crosses the
x-axis.
We will refer to the arcs along the outside edges of these wedges as
one-degree arcs. Why 360 equal sized arcs? The reason for doing so
is historically tied to the fact that the ancient Babylonians did so as
they developed their study of astronomy. (There is actually an alternate
system based on dividing the circular region into 400 equal sized wedges.)
Any central angle which subtends one of these 360 equal sized arcs is
circle Cr
etc.
a total of 360 this angle is
equal sized r DEFINED to
pie shaped typical have measure
wedges inside (r,0) wedge r 1 degree
this disk
etc.
***NOT TO SCALE***
Figure 15.4: Wedges as 1◦ arcs.
defined to have a measure of one degree, denoted 1◦ .

We can now use one-degree arcs to measure any angle: Begin by slid-
ing the angle ∠AOB into standard central position, as in Figure 15.3.
Piece together consecutive one-degree arcs in a counterclockwise or clock-
wise direction, beginning from the initial side and working toward the
terminal side, approximating the angle ∠AOB to the nearest degree. If
we are allowed to divide a one-degree arc into a fractional portion, then
we could precisely determine the number m of one-degree arcs which
consecutively fit together into the given arc. If we are sweeping counter-
clockwise from the initial side of the angle, m is defined to be the degree
measure of the angle. If we sweep in a clockwise direction, then −m is
defined to be the degree measure of the angle. So, in Figure 15.3, the
left-hand angle has positive degree measure while the right-hand angle
has negative degree measure. Simple examples would be angles like the
ones in Figure 15.5.
Notice, with our conventions, the rays determining an angle with mea-
sure −135◦ sit inside the circle in the same position as those for an angle
of measure 225◦; the minus sign keeps track of sweeping the positive
x-axis clockwise (rather than counterclockwise).
We can further divide a one-degree arc into 60 equal arcs, each called
a one minute arc. Each one-minute arc can be further divided into 60
equal arcs, each called a one second arc. This then leads to angle mea-
sures of one minute, denoted 1 ′ and one second, denoted 1 ′′ :
1◦ = 60 minutes
= 3600 seconds.
15.3. DEGREE METHOD 195
180◦
90◦ 270◦
45◦ 315◦
−135◦
Figure 15.5: Examples of common angles.
For example, an angle of measure θ = 5 degrees 23 minutes 18 sec-

onds is usually denoted 5◦ 23 ′ 18 ′′ . We could express this as a decimal of
degrees:
◦ ←In degrees!
◦ ′ ′′ 23 18
5 23 18 = 5+ +
60 3600
= 5.3883◦.
As another example, suppose we have an angle with measure 75.456◦ and
we wish to convert this into degree/minute/second units. First, since
75.456◦ = 75◦ +0.456◦, we need to write 0.456◦ in minutes by the calculation:
minutes
0.456 degree × 60 = 27.36 ′.
degree
This tells us that 75.456◦ = 75◦ 27.36 ′ = 75◦ 27 ′ + 0.36 ′. Now we need to write
0.36 ′ in seconds via the calculation:
0.36 minutes × 60 seconds/minute = 21.6 ′′ .
In other words, 75.456◦ = 75◦ 27 ′ 21.6 ′′.

Degree measurement of an angle is very closely tied to direction in the
plane, explaining its use in map navigation. With some additional work,
it is also possible to relate degree measure and lengths of circular arcs.
To do this carefully, first go back to Figure 15.3 and recall the situation
where an arc arc(AB) is subtended by the central angle ∠AOB. In this

situation, the arc length of arc(AB), commonly denoted by the letter s, is
the distance from A to B computed along the circular arc; keep in mind,
this is NOT the same as the straight line distance between the points A
and B.
For example, consider the six angles pictured above, of measures 90◦ ,
180◦ , 270◦, 45◦ , −135◦ , and 315◦ . If the circle is of radius r and we want to
compute the lengths of the arcs subtended by these six angles, then this
can be done using the formula for the circumference of a circle (on the
back of this text) and the following general principle:
Important Fact 15.3.1.
(length of a part) = (fraction of the part) × (length of the whole)
For example, the circumference of the entire circle of radius r is 2πr;

this is the “length of the whole” in the general principle. The arc sub-
90
tended by a 90◦ angle is 360 = 14 of the entire circumference; this is the
“fraction of the part” in the general principle. The boxed formula implies:

◦ 1 πr
s = arc length of the 90 arc = 2πr = .
4 2
s = distance
along the arc Similarly, a 180◦ angle subtends an arc of length
s = 7πr
πr,
r a 315◦ angle subtends an arc of length s = 315
360
2πr = 4
,
θ degrees
etc. In general, we arrive at this formula:
r Important Fact 15.3.2 (Arc length in degrees). Start with

a central angle of measure θ degrees inside a circle of ra-
Cr
dius r. Then this angle will subtend an arc of length
Figure 15.6: The definition

2π
of arc length. s= rθ
360
15.4 Radian Method

The key to understanding degree measurement was the description of a
“basic wedge” which contained an interior angle of measure 1◦ ; this was
straightforward and familiar to all of us. Once this was done, we could
proceed to measure any angle in degrees and compute arc lengths as in
Fact 15.3.2. However, the formula for the length of an arc subtended by
an angle measured in degrees is sort of cumbersome, involving the curi-
2π
ous factor 360 . Our next goal is to introduce an alternate angle measure-
ment scheme called radian measure that begins with a different “basic
wedge”. As will become apparent, a big selling point of radian measure is
that arc length calculations become easy.
15.4. RADIAN METHOD 197
r
r
r r
equilateral
wedge
r (r,0)
r
this angle is
DEFINED to
circle Cr have
measure 1
radian
Figure 15.7: Constructing an equilateral wedge.
As before, begin with a circle Cr of radius r. Construct an equilateral

wedge with all three sides of equal length r; see Figure 15.7. We define
the measure of the interior angle of this wedge to be 1 radian.
Once we have defined an angle of measure 1 radian, we can define an
angle of measure 2 radians by putting together two equilateral wedges.
Likewise, an angle of measure 12 radian is obtained by symmetrically di-
viding an equilateral wedge in half, etc.
Reasoning in this way, we can piece together equilateral wedges or
fractions of such to compute the radian measure of any angle. It is im-
portant to notice an important relationship between the radian measure
of an angle and arc length calculations. In the five angles pictured above,
1 radian, 2 radian, 3 radian, 12 radian and 41 radian, the length of the arcs
subtended by these angles θ are r, 2r, 3r, 12 r, and 14 r. In other words, a
pattern emerges that gives a very simple relationship between the length
s of an arc and the radian measure of the subtended angle:
Important Fact 15.4.1 (Arc length in radians). s = distance

along the arc
Start with a central angle of measure θ radians inside a
r
circle of radius r. Then this angle will subtend an arc of
length s = θr. θ radians
These remarks allow us to summarize the definition of r
the radian measure θ of ∠AOB inside a circle of radius r

by the formula: Cr
s
r
if angle is swept counterclockwise Figure 15.9: Defining arc
θ=
− sr if angle is swept clockwise length when angles are mea-
sured in radians.
3r
2r
r 3 radians = θ
2 radians = θ
r r
r
r
r
1 radian = θ
1
r 4
radian = θ
1
2
radian = θ
r
r
2 r r
4
r r
Figure 15.8: Measuring angles in radians.
y-axis length of
The units of θ are sometimes abbreviated as rad. It is
arc(AB) = s
important to appreciate the role of the radius of the circle
B Cr when using radian measure of an angle: An angle of
radian measure θ will subtend an arc of length |θ| on the
θ radian
O A
unit circle. In other words, radian measure of angles is
x-axis
exactly the same as arc length on the unit circle; we
couldn’t hope for a better connection!
circle radius r
The difficulty with radian measure versus degree mea-
Figure 15.10: Arc length sure is really one of familiarity. Let’s view a few common
after imposing a coordinate angles in radian measure. It is easiest to start with the
system. case of angles in central standard position within the unit
circle. Examples of basic angles would be fractional parts
1
of one complete revolution around the unit circle; for example, 12 revo-
1 1 1 1 3
lution, 8 revolution, 6 revolution, 4 revolution, 2 revolution and 4 revo-
lution. One revolution around the unit circle describes an arc of length
2π and so the subtended angle (1 revolution) is 2π radians. We can now
1
easily find the radian measure of these six angles. For example, 12 rev-
1
olution would describe an angle of measure ( 12 )2π rad= π6 rad. Similarly,
the other five angles pictured below have measures π4 rad, π3 rad, π2 rad, π
rad and 3π 2
rad.
All of these examples have positive radian measure. For an angle with
negative radian measure, such as θ = − π2 radians, we would locate B
by rotating 14 revolution clockwise, etc. From these calculations and our
15.5. AREAS OF WEDGES 199
eplacements
π π
π
4 3
6
1 1 1
12
revolution 8
revolution 6
revolution
π π 3π
2 2
1 1 3
4
revolution 2
revolution 4
revolution
Figure 15.11: Common angles measured in radians.
previous examples of degree measure we find that
180 degrees = π radians. (15.1)
Solving this equation for degrees or radians will provide conversion

formulas relating the two types of angle measurement. The formula
π 2π
also helps explain the origin of the curious conversion factor 180 = 360
in Fact 15.3.2.
15.5 Areas of Wedges

The beauty of radian measure is that it is rigged so that we can easily
compute lengths of arcs and areas of circular sectors (i.e. “pie-shaped
regions”). This is a key reason why we will almost always prefer to work
with radian measure.
Example 15.5.1. If a 16 inch pizza is cut into 12 equal slices, what is the
area of a single slice?
This can be solved using a general principle:
(Area of a part) = (area of the whole) × (fraction of the part)

So, for our pizza:

(area one slice) = (area whole pie) × (fraction of pie)

2 1
= (8 π)
12
16π
= .
3
Let’s apply the same reasoning to find the area of a
B circular sector. We know the area of the circular disc
bounded by a circle of radius r is πr2. Let Rθ be the “pie
shaped wedge” cut out by an angle ∠AOB with positive
Rθ
O
θ measure θ radians. Using the above principle
area(Rθ) = (area of disc bounded by Cr)
A
× (portion of disc accounted for by Rθ)

Cr 2 θ 1
= (πr ) = r2θ.
2π 2
Figure 15.12: Finding the
area of a “pie shaped wedge”. For example, if r = 3 in. and θ = π4 rad, then the area of
the pie shaped wedge is 98 π sq. in.
Important Fact 15.5.2 (Wedge area). Start with a central angle with pos-
itive measure θ radians inside a circle of radius r. The area of the “pie
shaped region” bounded by the angle is 21 r2θ.
Example 15.5.3. A water drip irrigation arm 1200 feet long rotates around
a pivot P once every 12 hours. How much area is covered by the arm in
one hour? in 37 minutes? How much time is required to drip irrigate 1000
square feet?
Solution. The irrigation arm will complete one revolution in 12 hours.
The angle swept out by one complete revolution is 2π radians, so after t
hours the arm sweeps out an angle θ(t) given by
2π radians π
θ(t) = × t hours = t radians.
12 hours 6
Consequently, by Important Fact 15.5.2 the area A(t) of the irrigated
region after t hours is
1 1 π
A(t) = (1200)2θ(t) = (1200)2 t = 120,000πt square feet.
2 2 6
After 1 hour, the irrigated area is A(1) = 120,000π = 376,991 sq. ft. Like-
37
wise, after 37 minutes, which is 60 hours, the area of the irrigated region
37 37
is A( 60 ) = 120,000π( 60) = 232,500 square feet. To answer the final ques-
tion, we need tosolve the equation A(t) = 1000; i.e., 120,000πt = 1000, so
1 3600 seconds
t= hours × = 9.55 seconds.
120π hour
15.5. AREAS OF WEDGES 201
15.5.1 Chord Approximation

Our ability to compute arc lengths can be used as an estimating tool for
distances between two points. Let’s return to the situation posed at the
beginning of this section: Cosmo the dog, tied by a 20 foot long tether
to a post in the ground, begins at location R and walks counterclockwise
to location S. Furthermore, let’s suppose you are standing at the center
of the circle determined by the tether and you measure the angle from R
to S to be 5◦ ; see the left-hand figure. Because the angle is small, notice
that the straight line distance d from R to S is approximately the same as
the arc length s subtended by the angle ∠RPS; the right-hand picture in
Figure 15.13 is a blow-up:
S
5◦ s
R d
P 20 feet
5◦
P
20 feet R
Figure 15.13: Using the arc length s to approximate the chord d.
Example 15.5.4. Estimate the distance from R to S.
Solution. We first convert the angle into radian measure via (15.1): 5◦ =
0.0873 radians. By Fact 15.3.2, the arc s has length 1.745 feet = 20.94 inches.
This is approximately equal to the distance from R to S, since the angle
is small.
S We call a line segment connecting two points on a cir-

cle a chord of the circle. The above example illustrates
s
a general principal for approximating the length of any
chord
R
chord. A smaller angle will improve the accuracy of the
O arc length approximation.
Important Fact 15.5.5 (Chord Approximation). In Fig-
ure 15.14, if the central angle is small, then s ≈ |RS|.
Figure 15.14: Chord ap-

proximation.
15.6 Great Circle Navigation
A basic problem is to find the shortest route between any two locations
on the earth. We will review how to coordinatize the surface of the earth
and recall the fact that the shortest path between two points is measured
along a great circle.
View the earth as a sphere of radius r = 3,960 miles. We could slice
the earth with a two-dimensional plane P0 which is both perpendicular
to a line connecting the North and South poles and passes through the
center of the earth. Of course, the resulting intersection will trace out a
circle of radius r = 3,960 miles on the surface of the earth, which we call
the equator. We call the plane P0 the equatorial plane. Slicing the earth
with any other plane P parallel to P0, we can consider the right triangle
pictured below and the angle θ:
North Pole line of latitude North Pole

b
b
θ
r r
θ
equatorial plane P0 r
r
equator 90◦ − θ◦
equator
center of earth center of earth South Pole

South Pole
Figure 15.15: Measuring latitude.
Essentially two cases arise, depending on whether or not the plane P

is above or below the equatorial plane. The plane P slices the surface of
the earth in a circle, which we call a line of latitude. This terminology
is somewhat incorrect, since these lines of latitude are actually circles
on the surface of the earth, but the terminology is by now standard. De-
pending on whether this line of latitude lies above or below the equatorial
plane, we refer to it as the θ◦ North line of latitude (denoted θ◦ N) or the
θ◦ South line of latitude (denoted θ◦ S). Notice, the radius b of a line of
latitude can vary from a maximum of 3,960 miles (in the case of θ = 0◦ ),
15.6. GREAT CIRCLE NAVIGATION 203
to a minimum of 0 miles, (when θ = 90◦ ). When b = 0, we are at the North

or South poles on the earth.
In a similar spirit, we could imagine slicing the earth with a plane Q
which is perpendicular to the equatorial plane and passes through the
center of the earth. The resulting intersection will trace out a circle of
radius 3,960 miles on the surface of the earth, which is called a line of
longitude. Half of a line of longitude from the North Pole to the South
Pole is called a meridian. We distinguish one such meridian; the one
which passes through Greenwich, England as the Greenwich meridian.
Longitudes are measured using angles East or West of Greenwich. Pic-
tured below, the longitude of A is θ. Because θ is east of Greenwich, θ
measures longitude East, typically written θ◦ E; west longitudes would
be denoted as θ◦ W. All longitudes are between 0◦ and 180◦ . The meridian
which is 180◦ West (and 180◦ E) is called the International Date Line.
Introducing the grid of latitude and longitude lines on
Greenwich, England
the earth amounts to imposing a coordinate system. In North Pole
line of longitude
other words, any position on the earth can be determined Greenwich A
International
by providing the longitude and latitude of the point. The meridian
r Date Line
θ
usual convention is to list longitude first. For example, r
Equator
Seattle has coordinates 122.0333◦ W, 47.6◦ N. Since the la-
center of earth South Pole
bels “N and S” are attached to latitudes and the labels “E
and W” are attached to longitudes, there is no ambiguity Figure 15.16: The Interna-
here. This means that Seattle is on the line of longitude tional Date Line.
122.0333◦ West of the Greenwich meridian and on the line
of latitude 47.6◦ North of the equator. In the figure below, we indicate the
key angles ψ = 47.6◦ and θ = 122.0333◦ by inserting the three indicated
radial line segments.
not great circle
N
Seattle, WA great circles
Ψ r
θ
Greenwich
Meridian
S
not great circle
Figure 15.17: Distances along great circles.

Now that we have imposed a coordinate system on the earth, it is

natural to study the distance between two locations. A great circle of a
sphere is defined to be a circle lying on the sphere with the same center
as the sphere. For example, the equator and any line of longitude are
great circles. However, lines of latitude are not great circles (except the
special case of the equator). Great circles are very important because
they are used to find the shortest distance between two points on the
earth. The important fact from geometry is summarized below.
Important Fact 15.6.1 (Great Circles). The shortest distance between

two points on the earth is measured along a great circle connecting them.
Example 15.6.2. What is the shortest distance from the North Pole to Seat-
tle, WA ?
N
Solution. The line of longitude 122.0333◦ W is a great circle
connecting the North Pole and Seattle. So, the shortest
W distance will be the arc length s subtended by the angle
∠NOW pictured in Figure 15.18. Since the latitude of
O Seattle is 47.6◦, the angle ∠EOW has measure 47.6◦ . Since
E
equator ∠EON is a right angle (i.e., 90◦ ), ∠NOW has measure 42.4◦ .
By Fact 15.4.1 and Equation 15.1,
S
Greenwich
Meridian s = (3960 miles)(42.4◦)(0.01745 radians/degree)
Figure 15.18: Distance be- = 2943.7 miles,
tween the North Pole and which is the shortest distance from the pole to Seattle.
Seattle, Washington.
15.7 Summary
• 360◦ = 2π radians
• A circular arc with radius r and angle θ has length s, with
s = rθ
when θ is measured in radians.
• A circular wedge with radius r and angle θ has area A, with
1
A = r2θ
2
when θ is measured in radians.
15.8. EXERCISES 205
15.8 Exercises
Problem 15.1. Let ∠AOB be an angle of mea- (c) Suppose bug A lands on the end of the
sure θ. blade farthest from the pivot. Assume
the wiper turns through an angle of 110◦ .
(a) Convert θ = 13.4o into degrees/ min-
In one cycle (back and forth) of the wiper
utes/ seconds and into radians.
blade, how far has the bug traveled?
(b) Convert θ = 1o 4 ′ 44 ′′ into degrees and ra-
dians.
(d) Suppose bug B lands on the end of the
(c) Convert θ = 0.1 radian into degrees and wiper blade closest to the pivot. Assume
degrees/ minutes/ seconds. the wiper turns through an angle of 110◦ .
In one cycle of the wiper blade, how far
has the bug traveled?
Problem 15.2. A nautical mile is a unit of dis-
tance frequently used in ocean navigation. It is
defined as the length of an arc s along a great (e) Suppose bug C lands on an intermediate
circle on the earth when the subtending angle location of the wiper blade. Assume the
has measure 1 ′ = “one minute” = 1/60 of wiper turns through an angle of 110◦ . If
one degree. Assume the radius of the earth is bug C travels 28 inches after one cycle
3,960 miles. of the wiper blade, determine the loca-
tion of bug C on the wiper blade.
(a) Find the length of one nautical mile to
the nearest 10 feet.
(b) A vessel which travels one nautical mile
in one hours time is said to have the
speed of one knot ; this is the usual nav- Problem 15.4. A water treatment facility oper-
igational measure of speed. If a vessel is ates by dripping water from a 60 foot long arm
traveling 26 knots, what is the speed in whose end is mounted to a central pivot. The
mph (miles per hour)? water then filters through a layer of charcoal.
The arm rotates once every 8 minutes.
(c) If a vessel is traveling 18 mph, what is
the speed in knots?
(a) Find the area of charcoal covered with
water after 1 minute.
Problem 15.3. The rear window wiper blade
on a station wagon has a length of 16 inches. (b) Find the area of charcoal covered with
The wiper blade is mounted on a 22 inch arm, water after 1 second.
6 inches from the pivot point.
(c) How long would it take to cover 100
square feet of charcoal with water?
6"
16" (d) How long would it take to cover 3245

square feet of charcoal with water?
Problem 15.5. Astronomical measurements

are often made by computing the small angle
formed by the extremities of a distant object
(a) If the wiper turns through an angle of
and using the estimating technique in 15.5.1.
110◦ , how much area is swept clean?
In the picture below, the full moon is shown
o
(b) Through how much of an angle would to form an angle of 21 when the distance indi-
the wiper sweep if the area cleaned was cated is 248,000 miles. Estimate the diameter
10 square inches? of the moon.
moon (a) θ = 45◦

(b) θ = 80o
248,000 miles
(c) θ = 3 radians
(d) θ = 2.46 radians
(e) θ = 97o 23 ′ 3 ′′
o
1/2
(f) θ = 35o 24 ′ 2 ′′
earth Problem 15.8. Matilda is planning a walk

around the perimeter of Wedge Park, which is
shaped like a circular wedge, as shown below.
Problem 15.6. An aircraft is flying at the The walk around the park is 2.1 miles, and the
speed of 500 mph at an elevation of 10 park has an area of 0.25 square miles.
miles above the earth, beginning at the North If θ is less than 90 degrees, what is the
pole and heading South along the Greenwich value of the radius, r?
meridian. A spy satellite is orbiting the earth
at an elevation of 4800 miles above the earth r
in a circular orbit in the same plane as the
Greenwich meridian. Miraculously, the plane
and satellite always lie on the same radial line θ
from the center of the earth. Assume the ra-
dius of the earth is 3960 miles.
r
satellite
Problem 15.9. Let C6 be the circle of radius

plane
6 inches centered at the origin in the xy-
coordinate system. Compute the areas of the
shaded regions in the picture below; the inner
earth circle in the rightmost picture is the unit cir-
cle:
y=x
y=−x
(a) When is the plane directly over a loca-

tion with latitude 74◦ 30 ′ 18 ′′ N for the first C C
6
time? 6
(b) How fast is the satellite moving?

y=x
(c) When is the plane directly over the equa-
tor and how far has it traveled?
(d) How far has the satellite traveled when
the plane is directly over the equator?
y=−(1/4)x + 2
Problem 15.7. Find the area of the sector of a C

circle of radius 11 inches if the measure θ of a 6
central angle of this sector is:
Chapter 16
Measuring Circular Motion

Cosmo moves
counterclockwise,
If Cosmo begins at location R and walks counterclockwise, S
maintaining
a tight tether.
always maintaining a tight tether, how can we measure
Cosmo’s speed? R
This is a “dynamic question” and requires that we dis- P 20 feet
cuss ways of measuring circular motion. In contrast, if

we take a snapshot and ask to measure the specific angle
∠RPS, this is a “static question”, which we answered in
Figure 16.1: How fast is
the previous section. Cosmo moving?
16.1 Different ways to measure

Cosmo’s speed
If Cosmo starts at location R and arrives at location S after some amount
of time, we could study
measure ∠RPS
ω= .
time required to go from R to S
The funny Greek letter “ω” on the left of side of the equation is pro-
nounced “oh-meg-a”.
We will refer to this as an angular speed. Typical
degrees degrees
units are “ minute ”, “ second ”, “ radians

minute
”, etc. For example, if the angle
◦
swept out by Cosmo after 8 seconds is 40 , then Cosmo’s angular speed
40◦ 5◦
is 8 seconds = sec . Using (15.1), we can convert to radian units and get
π rad rad
ω = 36 sec
= 53 π min . This is a new example of a rate and we can ask to
find the total change,
in the spirit of (1.2). If we are given ω in units of
rad
deg
“ time ” or “ time ”, we have
θ = ωt,
which computes the measure of the angle θ swept out after time t (i.e.
the total change in the angle). Angular speed places emphasis upon the
207
208 CHAPTER 16. MEASURING CIRCULAR MOTION
“size of the angle being swept out per unit time” by the moving object,
starting from some initial position. We need to somehow indicate the
direction in which the angle is being swept out. This can be done by in-
dicating “clockwise” our “counterclockwise”. Alternatively, we can adopt
the convention that the positive rotational direction is counterclockwise,
then insert a minus sign to indicate rotation clockwise. For example,
saying that Cosmo is moving at an angular speed of ω = − π2 rad
sec
means he
π rad
is moving clockwise 2 sec .
Another way to study the rate of a circular motion is to count the
number of complete circuits of the circle per unit time. This sort of rate
has the form
Number of Revolutions
;
Unit of Time
we will also view this as an angular speed. If we take “minutes” to be
the preferred unit of time, we arrive at the common measurement called
revolutions per minute, usually denoted RPM or rev/min. For example, if
Cosmo completes one trip around the circle every 2 minutes, then Cosmo
is moving at a rate of 12 RPM. If instead, Cosmo completes one trip around
the circle every 12 seconds, then we could first express Cosmo’s speed in
1
units of revolutions/second as 12 rev/second, then convert to RPM units:
1 rev sec

60 = 5 RPM.
12 sec min
As a variation, if we measure that Cosmo completed 37 of a revolution in
2 minutes, then Cosmo’s angular speed is computed by
3
7
rev 3
= RPM.
2 min 14
The only possible ambiguity involves the direction of revolution: the ob-
ject can move clockwise or counterclockwise.
The one shortcoming of using angular speed is that we are not directly
keeping track of the distance the object is traveling. This is fairly easy
to remedy. Returning to Figure 16.1, the circumference of the circle of
motion is 2π(20) = 40π feet. This is the distance traveled per revolution,
so we can now make conversions of angular speed into “distance traveled
per unit time”; this is called the linear speed. If Cosmo is moving 12 RPM,
then he has a linear speed of
1 rev 40π ft ft

v= = 20π .
2 min rev min
Likewise, if Cosmo is moving π7 rad

sec
, then
π rad 1 rev 40π ft 20π ft

v= = .
7 sec 2π rad rev 7 sec
16.2. DIFFERENT WAYS TO MEASURE CIRCULAR MOTION 209
Important Fact 16.1.1. This discussion is an example of what is usually

called “units analysis”. The key idea we have illustrated is how to convert
between two different types of units:
rev
converts to ft
−→
min min
16.2 Different Ways to Measure

Circular Motion
The discussion of Cosmo applies to circular motion of any object. As a
matter of convention, we usually use the Greek letter ω to denote angular
speed and v for linear speed. If an object is moving around a circle of
radius r at a constant rate, then we can measure it’s speed in two ways:
• The angular speed
“revolutions” “degrees swept” “radians swept”

ω= or or .
“unit time” “unit time” “unit time”
• The linear speed
“distance traveled”
v= .
“per unit time”
Important Facts 16.2.1 (Measuring and converting). We can convert be-

tween angular and linear speeds using these facts:
• 1 revolution = 360◦ = 2π radians;
• The circumference of a circle of radius r units is 2πr units.
16.2.1 Three Key Formulas

If an object begins moving around a circle, there are a number of quan-
tities we can try to relate. Some of these are “static quantities”: Take
a visual “snapshot” of the situation after a certain amount of time has
elapsed, then we can measure the radius, angle swept, arc length and
time elapsed. Other quantities of interest are “dynamic quantities”: This
means something is CHANGING with respect to time; in our case, the lin-
ear speed (which measures distance traveled per unit time) and angular
speed (which measures angle swept per unit time) fall into this category.
STATIC QUANTITIES DYNAMIC QUANTITIES
S S
1. arc length s 5. angular speed ω
2. angle swept in time r 6. linear speed v

θ
R P R
P
3. radius r
4. elapsed time t
...take a “snapshot” after time t... ...see what happens per unit time...
Figure 16.2: Measuring linear and angular speed.
We now know two general relationships for circular motion:

(i) s = rθ, where s=arclength (a linear distance), r=radius of the circular
path and θ=angle swept in RADIAN measure; this was the content
of Fact 15.4.1 on page 197.
(ii) θ = ωt, where θ is the measure of an angle swept, ω= angular speed
and t represents time elapsed. This is really just a consequence of
units manipulation.
Notice how the units work in these formulas. If r=20 feet and θ = 1.3
radians, then the arc length s = 20(1.3) feet= 26 feet; this is the length
of the arc of radius 20 feet that is subtending the angle θ. If ω = 3
rad
rad/second and t = 5 seconds, then θ = 3 seconds × 5 seconds = 15 radians.
If we replace “θ” in s = rθ of (i) with θ = ωt in (ii), then we get
s = rωt.
This gives us a relationship between arclength s (a distance) and time t.
Plug in the fact that the linear speed is defined to be v = “distance”
t
and we
get
s rωt
v= = = rω.
t t
All of these observations are summarized below.
Important Facts 16.2.2 (Three really useful formulas). If we measure
angles θ in RADIANS and ω in units of radians per unit time, we have
these three formulas:
s = rθ (16.1)
θ = ωt (16.2)
v = rω (16.3)
16.2. DIFFERENT WAYS TO MEASURE CIRCULAR MOTION 211
Example 16.2.3. You are riding a stationary exercise bike

and the speedometer reads a steady speed of 40 MPH
(miles per hour). If the rear wheel is 28 inches in diame-
ter, determine the angular speed of a location on the rear
tire. A pebble becomes stuck to the tread of the rear tire. *
pebble sticks to tread here
Describe the location of the pebble after 1 second and 0.1
second. Figure 16.3: Where is the
pebble after t seconds?
Solution. The tires will be rotating in a counterclockwise
direction and the radius r = 12 28 = 14 inches. The other given quantity,
“40 MPH”, involves miles, so we need to decide which common units to
work with. Either will work, but since the problem is focused on the
wheel, we will utilize inches.
If the speedometer reads 40 MPH, this is the linear speed of a specified
location on the rear tire. We need to convert this into an angular speed,
using unit conversion formulas. First, the linear speed of the wheel is
miles ft in 1 hr min

v = 40 5280 12 1
hr mile ft 60 min 60 sec
in
= 704 .
sec
Now, the angular speed ω of the wheel will be
inches
704 second
ω = inches
2(14)π revolution
revolution
= 8
second
= 480 RPM
It is then an easy matter to convert this to
revolution degrees

ω = 8 360
second revolution
degrees
= 2,880 .
second
If the pebble begins at the “6 o’clock” position (the place the tire touches
the ground on the wheel), then after 1 second the pebble will go through 8
revolutions, so will be in the “6 o’clock” position again. After 0.1 seconds,
the pebble will go through
rev
8 (0.1 sec) = 0.8 rev
sec
deg

= (0.8 rev) 360
rev
◦
= 288 .
Keeping in mind that the rotation is counterclockwise, we can view the

location of the pebble after 0.1 seconds as pictured below:
288◦ counterclock-
wise rotation
after 0.1 second
*
located here at starts
time = 0.1 sec here
*
pebble sticks to tread in 6 o’clock position
Figure 16.4: Computing the pebble’s position after t = 0.1 sec.
We solved the previous problem using the “unit conversion method”.

There is an alternate approach available, which uses one of the formulas
in Fact 16.2.2. Here is how you could proceed: First, as above, we know
the linear speed is v = 704 in/sec. Using the “v = ωr” formula, we have
in
704 = ω(14 in)
sec
rad
ω = 50.28 .
sec
Notice how the units worked out in the calculation: the “time” unit comes
from v and the “angular” unit will always be radians. As a comparison
with the solution above, we can convert ω into RPM units:
rad 1 rev

ω = 50.28
sec 2π rad
rev
= 8 .
sec
All of the problems in this section can be worked using either the “unit
conversion method” or the “v = ωr method”.
16.3 Music Listening Technology

The technology of reproducing music has gone through a revolution since
the early 1980’s. The “old” stereo long playing record (the LP ) and the
16.3. MUSIC LISTENING TECHNOLOGY 213
“new” digital compact disc (the CD ) are two methods of storing musical
data for later reproduction in a home stereo system. These two technolo-
gies adopt different perspectives as to which notion of circular speed is
best to work with.
Long playing stereo records are thin vinyl plastic discs of radius 6
inches onto which small spiral grooves are etched into the surface; we
can approximately view this groove as a circle. The LP is placed on a
flat 12 inch diameter platter which turns at a constant angular speed of
33 13 RPM. An arm on a pivot (called the tone arm) has a needle mounted
on the end (called the cartridge), which is placed in the groove on the
outside edge of the record. Because the grooves wobble microscopically
from side-to-side, the needle will mimic this motion. In turn, this sets
a magnet (mounted on the opposite end of the needle) into motion. This
moving magnet sits inside a coil of wire, causing a small varying voltage;
the electric signal is then fed to your stereo, amplified and passed onto
your speakers, reproducing music!
amp
LP turning at 33 13 RPM
speakers
tonearm
needle
Figure 16.5: Reproducing music using analogue technology.
This is known as analogue technology and is based upon the idea of

maintaining a constant angular speed of 33 31 RPM for the storage medium
(our LP). (Older analogue technologies used 45 RPM and 78 RPM records.
However, 33 13 RPM became the consumer standard for stereo music.)
With an LP, the beginning of the record (the lead-in groove ) would be
on the outermost edge of the record and the end of the record (the exit
groove ) would be close to the center. Placing the needle in the lead-in
groove, the needle gradually works its way to the exit groove. However,
whereas the angular speed of the LP is a constant 33 13 RPM, the linear
speed at the needle can vary quite a bit, depending on the needle loca-
tion.
Example 16.3.1 (Analogue LP’s). The “lead-in groove” is

6 inches from the center of an LP, while the “exit groove” is
1
1 inch from the center. What is the linear speed (MPH) of
6
the needle in the “lead-in groove”? What is the linear speed
(MPH) of the needle in the “exit groove”? Find the location
Figure 16.6: Lead-in and
exit grooves. of the needle if the linear speed is 1 MPH.
Solution. This is a straightforward application of Fact 16.2.1.
Let v6 (resp. v1) be the linear speed at the lead in groove (resp. exit
groove); the subscript keeps track of the needle radial location. Since the
groove is approximately a circle,
1 rev inches

v6 = 33 2(6)π
3 min rev
in
= 1257
min
in min

1257 min 60 hour
= ft
12 in

5280 mile ft
= 1.19 MPH
Similarly, v1 = 0.2 MPH. To answer the remaining question, let r be the
radial distance from the center of the LP to the needle location on the
record. If vr = 1 MPH:
mile
1 = vr
hour
1 rev in min 1 ft 1 mile

= 33 2rπ 60
3 min rev hr 12 in 5280 ft
So, when the needle is r = 5.04 inches from the center, the linear speed is
1 MPH.
In the early 1980’s, a new method of storing and repro-
laser support arm
laser moves back and forth ducing music was introduced; this medium is called the
digital compact disc, referred to as a CD for short. This is
a thin plastic disc of diameter 4.5 inches, which appears
to the naked eye to have a shiny silver coating on one side.
Upon microscopic examination one would find concentric
circles of pits in the silver coating. This disc is placed in
spinning CD
a CD player, which spins the disc. A laser located above
the spinning disc will project onto the spinning disc. The
pits in the silver coating will cause the reflected laser light
Figure 16.7: Reproducing to vary in intensity. A sensor detects this variation, con-
music using digital technol-
ogy.
verting it to a digital signal (the analogue to digital or AD
conversion ). This is fed into a digital to analogue or DA
conversion device, which sends a signal to your stereo,
again producing music.
16.4. BELT AND WHEEL PROBLEMS 215
The technology of CD ′ s differs from that of LP ′ s in two crucial ways.

First, the circular motion of the spinning CD is controlled so that the
target on the disc below the laser is always moving at a constant linear
speed of 1.2 meters
sec
inches
= 2,835 minute . Secondly, the beginning location of the
laser will be on the inside portion of the disc, working its way outward to
the end. In this context, it makes sense to study how the angular speed
of the CD is changing, as the laser position changes.
Example 16.3.2 (Digital CD’s). What is the angular speed

(in RPM) of a CD if the laser is at the beginning, located 34
2 ′′
inches from the center of the disc ? What is the angular
speed (in RPM) of a CD if the laser is at the end, located 2
inches from the center of the disc ? Find the location of the 3 ′′
4
Start of CD End of CD
laser if the angular speed is 350 RPM.
Figure 16.8: Computing the
Solution. This is an application of Fact 16.2.1. Let ω3/4 be
angular speed of a CD.
the angular speed at the start and ω2 the angular speed
at the end of the CD ; the subscript is keeping track of the
laser distance from the CD center.
(2835 inches/min)
ω2 = = 225.6 RPM
(2(2)π inches/rev)
2835 inches/min)
ω3/4 = = 601.6 RPM
(2(0.75)π inches/rev)
To answer the remaining question, let r be the radial distance from
the center of the CD to the laser location on the CD. If the angular speed
ωr at this location is 350 RPM, we have the equation
350 RPM = ωr
inches

2,835 minute
= inches

2rπ revolution
1.289 inches = r.
So, when the laser is 1.289 inches from the center, the CD is moving
350 RPM.
16.4 Belt and Wheel Problems

The industrial revolution spawned a number of elaborate machines in-
volving systems of belts and wheels. Computing the speed of various
belts and wheels in such a system may seem complicated at first glance.
The situation can range from a simple system of two wheels with a belt
connecting them, to more elaborate designs. We call problems of this sort
belt and wheel problems, or more generally, connected wheel problems.
Solving problems of this type always uses the same strategy, which we
will first highlight by way of an example.
J
E G
C D F A B
H
I
Figure 16.9: Two typical connected wheel scenarios.
front sprocket radius = 5 inches
Example 16.4.1. You are riding a stationary exercise bike.

Assume the rear wheel is 28 inches in diameter, the rear
sprocket has radius 2 inches and the front sprocket has
rear sprocket radius = 2 inches
radius 5 inches. How many revolutions per minute of the
front sprocket produces a forward speed of 40 MPH on the
(a) A stationary exercise bike. bike (miles per hour)?
radius A = 14 inches Solution. There are 3 wheels involved with a belt (the bicy-
A
cle chain) connecting two of the wheels. In this problem,
we are provided with the linear speed of wheel A (which is
C B
40 MPH) and we need to find the angular speed of wheel
radius B =
radius C = 5 inches
2 inches C=front sprocket.
Denote by vA, vB, and vC the linear speeds of each of
(b) A model of the bike’s the wheels A, B, and C, respectively. Likewise, let ωA, ωB,
connected wheels. and ωC denote the angular speeds of each of the wheels A,
B, and C, respectively. In addition, the chain connecting
Figure 16.10: Visualizing the wheels B and C will have a linear speed, which we will
the connected wheels of an denote by vchain. The strategy is broken into a sequence
exercise bike.
of steps which leads us from the known linear speed vA to
the angular speed ωC of wheel C:
vA
• Step 1: Given vA, find ωA. Use the fact ωA = rA
.
• Step 2: Observe ωA = ωB; this is because the wheel

and rear sprocket are both rigidly mounted on a
common axis of rotation.
• Step 3: Given
ωB, find vB. Use the fact vB = rBωB =
rB
rBωA = rA vA.
• Step 4: Observe vB = vchain = vC; this is because the

chain is directly connecting the two sprockets and
assumed not to slip.
16.4. BELT AND WHEEL PROBLEMS 217
vC
• Step 5: Given vC, find ωC. Use the fact ωC = rC
=

vB rB
rC
= rA rC
vA.
Saying that the speedometer reads 40 MPH is the same as saying

that the linear speed of a location on the rear wheel is vA = 40 MPH.
Converting this into angular speed was carried out in our solution to
Example 16.2.3 above; we found that ωA = 480 RPM. This completes
Step 1 and so by Step 2, ωA = ωB = 480 RPM. For Step 3, we convert
ωB = 480 RPM into linear speed following Fact 16.2.1:
revolution inches

vB = 480 (2(2)π)
minute revolution
inches
= 6,032 .
minute
By Step 4, conclude that the linear speed of wheel C is vC = 6,032 inches/min.
Finally, to carry out Step 5, we convert the linear speed into angular
speed:
6,032 inches

min
ωC =
2(5)π inches

rev
= 192 RPM
rev
= 3.2 .
sec
In conclusion, the bike rider must pedal the front sprocket at the rate of
rev
3.2 sec .
This example indicates the basic strategy used in all belt/wheel prob-
lems.
Important Facts 16.4.2 (Belt and Wheel Strategy). Three basic facts are
used in all such problems:
• Using “unit conversion” or Fact 16.2.2 allows us to go from linear

speed v to angular speed ω, and vice versa.
• If two wheels are fastened rigidly to a common axle, then they have
the same angular speed. (Caution: two wheels fastened to a common
axle typically do not have the same linear speed!)
• If two wheels are connected by a belt (or chain), the linear speed of
the belt coincides with the linear speed of each wheel.
16.5 Exercises
Problem 16.1. The restaurant in the Space Problem 16.4. Lee is running around the
Needle in Seattle rotates at the rate of one rev- perimeter of a circular track at a rate of 10
olution per hour. ft/sec. The track has a radius of 100 yards.
(a) Through how many radians does it turn After 10 seconds, Lee turns and runs along a
in 100 minutes? radial line to the center of the circle. Once he
reaches the center, he turns and runs along a
(b) How long does it take the restaurant to radial line to his starting point on the perime-
rotate through 4 radians? ter. Assume Lee does not slow down when he
(c) How far does a person sitting by the win- makes these two turns.
dow move in 100 minutes if the radius of (a) Sketch a picture of the situation.
the restaurant is 21 meters?
(b) How far has Lee traveled once he returns
to his starting position?
Problem 16.2. You are riding a bicycle along
a level road. Assume each wheel is 26 inches (c) How much time will elapse during Lee’s
in diameter, the rear sprocket has a radius of circuit?
3 inches and the front sprocket has a radius (d) Find the area of the pie shaped sector
of 7 inches. How fast do you need to pedal (in enclosed by Lee’s path.
revolutions per minute) to achieve a speed of
35 mph?
Problem 16.5. John has been hired to de-
front wheel sign an exciting carnival ride. Tiff, the car-
nival owner, has decided to create the worlds
rear wheel greatest ferris wheel. Tiff isn’t into math; she
simply has a vision and has told John these
constraints on her dream: (i) the wheel should
rotate counterclockwise with an angular speed
of 12 RPM; (ii) the linear speed of a rider should
be 200 mph; (iii) the lowest point on the ride
should be 4 feet above the level ground.
rear sprocket front sprocket

12 RPM
Problem 16.3. Answer the following angular
θ
speed questions.
P
(a) A wheel of radius 22 ft. is rotating 11
RPM counterclockwise. Considering a
point on the rim of the rotating wheel, 4 feet
what is the angular speed ω in rad/sec
and the linear speed v in ft/sec? (a) Find the radius of the ferris wheel.
(b) A wheel of radius 8 in. is rotating (b) Once the wheel is built, John suggests
15o /sec. What is the linear speed v, the that Tiff should take the first ride. The
angular speed in RPM and the angular wheel starts turning when Tiff is at the
speed in rad/sec? location P, which makes an angle θ with
(c) You are standing on the equator of the the horizontal, as pictured. It takes her
earth (radius 3960 miles). What is your 1.3 seconds to reach the top of the ride.
linear and angular speed? Find the angle θ.
(d) An auto tire has radius 12 inches. If you (c) Poor engineering causes Tiff’s seat to fly
are driving 65 mph, what is the angular off in 6 seconds. Describe where Tiff is
speed in rad/sec and the angular speed located (an angle description) the instant
in RPM? she becomes a human missile.
16.5. EXERCISES 219
Problem 16.6. Michael and Aaron are on the (e) Assume Michael has traveled 88 feet
“UL-Tossum” ride at Funworld. This is a from the position P to a new position Q.
merry-go-round of radius 20 feet which spins How many seconds will this take? What
counterclockwise 60 RPM. The ride is driven will be the angle swept out by Michael?
by a belt connecting the outer edge of the ride
to a drive wheel of radius 3 feet:
Aaron length of arc (PQ) is 88 ft
Drive wheel
radius 3 ft
Michael
O P
O P
main ride
drive belt radius 20 ft
Q
(a) Assume Michael is seated on the edge of

the ride, as pictured. What is Michael’s
linear speed in mph and ft/sec? Problem 16.7. You are riding a bicycle along a
level road. Assume each wheel is 28 inches in
(b) What is the angular speed of the drive diameter, the rear sprocket has radius 3 inches
wheel in RPM? and the front sprocket has radius r inches.
(c) Suppose Aaron is seated 16 feet from the Suppose you are pedaling the front sprocket
rev
center of the ride. What is the angular at the rate of 1.5 sec and your forward speed is
speed of Aaron in RPM? What is the lin- 11 mph on the bike. What is the radius of the
ear speed of Aaron in ft/sec? front sprocket?
(d) After 0.23 seconds Michael will be lo-
cated at S as pictured. What is the an- Problem 16.8. You are designing a system of
gle ∠POS in degrees? What is the angle wheels and belts as pictured below. You want
∠POS in radians? How many feet has wheel A to rotate 20 RPM while wheel B rotates
Michael traveled? 42 RPM. Wheel A has a radius of 6 inches,
wheel B has a radius of 7 inches and wheel C
has a radius of 1 inch. Assume wheels C and
S D are rigidly fastened to the same axle. What
is the radius r of wheel D?
θ
O P D
A C B
Chapter 17
The Circular Functions
Suppose Cosmo begins at location R and walks in a coun-

terclockwise direction, always maintaining a tight 20 ft S1
S2
long tether. As Cosmo moves around the circle, how can
we describe his location at any given instant?
In one sense, we have already answered this question:
P 20 feet
The measure of ∠RPS1 exactly pins down a location on R
S3
the circle of radius 20 feet. But, we really might prefer
a description of the horizontal and vertical coordinates of
S4
Cosmo; this would tie in better with the coordinate system
we typically use. Solving this problem will require NEW Figure 17.1: Cosmo moves
functions, called the circular functions. counterclockwise maintain-
ing a tight tether. Where’s
Cosmo?
17.1 Sides and Angles of a Right Triangle
Example 17.1.1. You are preparing to make your final The billiard table layout.
shot at the British Pocket Billiard World Championships.
4 ft
The position of your ball is as in Figure 17.2, and you must this pocket
for the
6 ft
play the ball off the left cushion into the lower-right corner find
5 ft big money
this
pocket, as indicated by the dotted path. For the big money, location
where should you aim to hit the cushion?

12 ft
Solution. This problem depends on two basic facts. First, 4

5−x θ
the angles of entry and exit between the path the cushion θ
will be equal. Secondly, the two obvious right triangles x
in this picture are similar triangles. Let x represent the

distance from the bottom left corner to the impact point Mathmatically modeling the bank shot.
of the ball’s path: Figure 17.2: A pocket bil-

liard banking problem.
Properties of similar triangles tell us that the ratios of
4
common sides are equal: 5−x = 12
x
. If we solve this equation
15
for x, we obtain x = 4 = 3.75 feet.
221
222 CHAPTER 17. THE CIRCULAR FUNCTIONS
This discussion is enough to win the tourney. But, of course, there

are still other questions we can ask about this simple example: What is
the angle θ? That is going to require substantially more work; indeed the
bulk of this Chapter! It turns out, there is a lot of mathematical mileage
in the idea of studying ratios of sides of right triangles. The first step,
which will get the ball rolling, is to introduce new functions whose very
definition involves relating sides and angles of right triangles.
17.2 The Trigonometric Ratios

B
From elementary geometry, the sum of the angles of any
hypotenuse triangle will equal 180◦ . Given a right triangle △ABC, since
side opposite θ one of the angles is 90◦ , the remaining two angles must be
θ acute angles ; i.e., angles of measure between 0◦ and 90◦ .
A C
side adjacent θ If we specify one of the acute angles in a right triangle
Figure 17.3: Labeling the △ABC, say angle θ, we can label the three sides using
sides of a right triangle. this terminology. We then consider the following three
ratios of side lengths, referred to as trigonometric ratios :
def length of side opposite θ

sin(θ) = (17.1)
length of hypotenuse
def length of side adjacent θ

cos(θ) = (17.2)
length of hypotenuse
def length of side opposite θ

tan(θ) = . (17.3)
length of side adjacent to θ
For example, we have three right triangles in Figure 17.4; you can
verify that the Pythagorean Theorem holds in each of the cases. In the
5
left-hand triangle, sin(θ) = 13 , cos(θ) = 12
13
5
, tan(θ) = 12 . In the middle
1 1
triangle, sin(θ) = √2 , cos(θ) = √2 , tan(θ) = 1. In the right-hand triangle,
√
sin(θ) = 12 , cos(θ) = 23 , tan(θ) = √13 . The symbols “sin”, “cos”, and “tan”
are abbreviations for the words sine, cosine and tangent, respectively.
As we have defined them, the trigonometric ratios depend on the dimen-
sions of the triangle. However, the same ratios are obtained for any right
triangle with acute angle θ. This follows from the properties of similar
triangles. Consider Figure 17.5. Notice △ABC and △ADE are similar. If
we use △ABC to compute cos(θ), then we find cos(θ) = |AC| |AB|
. On the other
|AE|
hand, if we use △ADE, we obtain cos(θ) = |AD|
. Since the ratios of com-
|AC| |AE|
mon sides of similar triangles must agree, we have cos(θ) = |AB|
= |AD|
,
17.2. THE TRIGONOMETRIC RATIOS 223
√
2 2
13
5 1 1
θ θ θ
√
12 1 3
Figure 17.4: Computing trigonometric ratios for selected right triangles.
which is what we wanted to be true. The same argument can be used

to show that sin(θ) and tan(θ) can be computed using any right triangle
with acute angle θ.
Except for some “rigged” right triangles, it is not easy D
to calculate the trigonometric ratios. Before the 1970’s, B
approximate values of sin(θ), cos(θ), tan(θ) were listed
in long tables or calculated using a slide rule. Today, a
scientific calculator saves the day on these computations.
θ
Most scientific calculators will give an approximation for A
C E
the values of the trigonometric ratios. However, it is good
to keep in mind we can compute the EXACT values of the Figure 17.5: Applying
π π π π
trigonometric ratios when θ = 0, 6 , 4 , 3 , 2 radians or, trigonometric ratios to any
◦ ◦ ◦ ◦ ◦ right triangle.
equivalently, when θ = 0 , 30 , 45 , 60 , 90 .
Angle θ Trigonometric Ratio

Deg Rad sin(θ) cos(θ) tan(θ)
0◦ 0 0 1 0
√
π 1 3 √1
30◦ 6 2 2 3
√ √
π 2 2
45◦ 4 2 2
1
π
√
3 1
√
60◦ 3 2 2
3
π
90◦ 2
1 0 Undefined
Table 17.1: Exact Trigonometric Ratios

Some people make a big deal of “approximate” vs. “exact” answers; we

won’t worry about it here, unless we are specifically asked for an exact
answer. However, here is something we will make a big deal about:
!!!
CAUTION When computing values of cos(θ), sin(θ), and tan(θ) on your calcu-
!!! lator, make sure you are using the correct “angle mode” when entering
θ; i.e. “degrees” or “radians”.
For example, if θ = 1◦ , then cos(1◦ ) = 0.9998, sin(1◦ ) = 0.0175, and
tan(1◦ ) = 0.0175. In contrast, if θ = 1 radians, then cos(1) = 0.5403,
sin(1) = 0.8415, and tan(1) = 1.5574.
17.3 Applications
When confronted with a situation involving a right trian-
gle where the measure of one acute angle θ and one side
h
are known, we can solve for the remaining sides using the
θ
h sin(θ) a tan(θ) appropriate trigonometric ratios. Here is the key picture
θ
to keep in mind:
h cos(θ) a
Figure 17.6: What do these Important Facts 17.3.1 (Trigonometric ratios).

ratios mean? Given a right triangle, the trigonometric ratios relate the
lengths of the sides as shown in Figure 17.6.
Example 17.3.2. To measure the distance across a river for a new bridge,
surveyors placed poles at locations A, B and C. The length |AB| = 100 feet
and the measure of the angle ∠ABC is 31◦ 18 ′ . Find the distance to span the
river. If the measurement of the angle ∠ABC is only accurate within ±2 ′ ,
find the possible error in |AC|.
Solution. The trigonometric ratio relating these two sides

would be the tangent and we can convert θ into decimal
C
form, arriving at:
d
B 100 A |AC| d
tan(31◦18 ′ ) = tan(31.3◦) = =
|BA| 100
310 18′
Figure 17.7: The distance therefore d = 60.8 feet.

spanning a river.
This tells us that the bridge needs to span a gap of
60.8 feet. If the measurement of the angle was in error by +2 ′ , then
tan(31◦20 ′ ) = tan(31.3333◦) = 0.6088 and the span is 60.88 ft. On the
other hand, if the measurement of the angle was in error by −2 ′ , then
tan(31◦16 ′ ) = tan(31.2667◦) = 0.6072 and the span is 60.72 ft.
17.3. APPLICATIONS 225
Example 17.3.3. A plane is flying 2000 feet above sea T

level toward a mountain. The pilot observes the top of the E
P
mountain to be 18◦ above the horizontal, then immediately
2000 ft L
flies the plane at an angle of 20◦ above horizontal. The
airspeed of the plane is 100 mph. After 5 minutes, the 000
111
S sealevel
000
111
000
111
000
111
plane is directly above the top of the mountain. How high

is the plane above the top of the mountain (when it passes Figure 17.8: Flying toward a
over)? What is the height of the mountain? mountain.
Solution. We can compute the hypotenuse of △LPT by us-

ing the speed and time information about the plane:
25
|PT | = (100 mph)(5 minutes)(1 hour/60 minutes) = miles.
3
The definitions of the trigonometric ratios show:
25
|TL| = sin(20◦ ) = 2.850 miles, and
3
25
|PL| = cos(20◦ ) = 7.831 miles.
3
With this data, we can now find |EL|:
|EL| = |PL| tan(18◦ ) = 2.544 miles.
The height of the plane above the peak is |TE| = |TL| − |EL| = 2.850 − 2.544 =
0.306 miles = 1,616 feet. The elevation of the peak above sea level is
given by: Peak elevation = plane altitude + |EL| = |SP| + |EL| = 2,000 +
(2.544)(5,280) = 15,432 feet.
Example 17.3.4. A Forest Service helicopter needs to de- B

termine the width of a deep canyon. While hovering, they 400 ft γ
A
measure the angle γ = 48◦ at position B (see picture), then
α
descend 400 feet to position A and make two measure- β
C E 00
11
canyon
00
11 D
ments of α = 13◦ (the measure of ∠EAD), β = 53◦ (the mea- 00
11
sure of ∠CAD). Determine the width of the canyon to the

Figure 17.9: Finding the
nearest foot. width of a canyon.
Solution. We will need to exploit three right triangles in

the picture: △BCD, △ACD, and △ACE. Our goal is to compute |ED| =
|CD| − |CE|, which suggests more than one right triangle will come into
play.
The first step is to use △BCD and △ACD to obtain a system of two
equations and two unknowns involving some of the side lengths; we will
then solve the system. From the definitions of the trigonometric ratios,
|CD| = (400 + |AC|) tan(48◦)
|CD| = |AC| tan(53◦ ).
Plugging the second equation into the first and rearranging we get
400 tan(48◦ )
|AC| = = 2,053 feet.
tan(53◦) − tan(48◦ )
Plugging this back into the second equation of the system gives
|CD| = (2053) tan(53◦) = 2724 feet.
The next step is to relate △ACD and △ACE, which can now be done
in an effective way using the calculations above. Notice that the measure
of ∠CAE is β − α = 40◦ . We have
|CE| = |AC| tan(40◦ ) = (2053) tan(40◦ ) = 1,723 feet.
As noted above, |ED| = |CD| − |CE| = 2,724 − 1,723 = 1,001 feet is the width
of the canyon.
17.4 Circular Functions

If Cosmo is located somewhere in the first quadrant of
S = (x,y)
Figure 17.1, represented by the location S, we can use the
trigonometric ratios to describe his coordinates. Impose
20
y the indicated xy-coordinate system with origin at P and
θ extract the pictured right triangle with vertices at P and
P x R S. The radius is 20 ft. and applying Fact 17.3.1 gives
S = (x, y) = (20 cos(θ), 20 sin(θ)).
Unfortunately, we run into a snag if we allow Cosmo to

Figure 17.10: Cosmo on a wander into the second, third or fourth quadrant, since
circular path.
then the angle θ is no longer acute.
17.4. CIRCULAR FUNCTIONS 227
17.4.1 Are the trigonometric ratios functions?

Recall that sin(θ), cos(θ), and tan(θ) are defined for acute
angles θ inside a right triangle. We would like to say that P A unit circle with
radius = 1.
these three equations actually define functions where the
variable is an angle θ. Having said this, it is natural to ask 1
θ
if these three equations can be extended to be defined for
ANY angle θ. For example, we need to explain how sin 2π 3
O 1 R
is defined.
To start, we begin with the unit circle pictured in the
xy-coordinate system. Let θ = ∠ROP be the angle in stan-
dard central position shown in Figure 17.11. If θ is pos- Figure 17.11: Coordinates
itive (resp. negative), we adopt the convention that θ is of points on the unit circle.
swept out by counterclockwise (resp. clockwise) rotation
of the initial side OR. The objective is to find the coordinates of the point
P in this figure. Notice that each coordinate of P (the x-coordinate and
the y-coordinate) will depend on the given angle θ. For this reason, we
need to introduce two new functions involving the variable θ.
Definition 17.4.1. Let θ be an angle in standard central

position inside the unit circle, as in Figure 17.11. This angle
111 11
000 00
00
11
determines a point P on the unit circle. Define two new r =111
000
111
000
111
00
11
1 kilometer
000 00
11
000 11
111 00
0.025 rad
sec
functions, cos(θ) and sin(θ), on the domain of all θ values
as follows: Michael
starts here
def
cos(θ) = horizontal x-coordinate of P on unit circle
def
sin(θ) = vertical y-coordinate of P on unit circle.
Figure 17.12: A circular
We refer to sin(θ) and cos(θ) as the basic circular func- driving track.
tions. Keep in mind that these functions have variables
which are angles (either in degree or radian measure). These functions
will be on your calculator. Again, BE CAREFUL to check the angle mode
setting on your calculator (“degrees” or “radians”) before doing a calcula-
tion.
Example 17.4.2. Michael is test driving a vehicle counter-

y-axis
clockwise around a desert test track which is circular of M(t) = (x(t),y(t))
radius 1 kilometer. He starts at the location pictured, trav-
0.025 rad
eling 0.025 rad
sec
. Impose coordinates as pictured. Where is sec
θ(t)
Michael located (in xy-coordinates) after 18 seconds? x-axis
Michael
starts here
Solution. Let M(t) be the point on the circle of motion rep-
resenting Michael’s location after t seconds and θ(t) the
angle swept out the by Michael after t seconds. Since we
Figure 17.13: Modeling
are given the angular speed, we get Michael’s location.
θ(t) = 0.025t radians.
Since the angle θ(t) is in central standard position, we get
M(t) = (cos(θ(t)), sin(θ(t))) = (cos(0.025t), sin(0.025t)).
So, after 18 seconds Michael’s location will be M(18) = (0.9004, 0.4350).

!!!
Interpreting the coordinates of the point P = (cos(θ), sin(θ)) in Fig-
CAUTION
ure 17.11 only works if the angle θ is viewed in central standard po-
!!!
sition. You must do some additional work if the angle is placed in a
different position; see the next Example.
Example 17.4.3. Both Angela and Michael are test driving

y-axis vehicles counterclockwise around a desert test track which
is circular of radius 1 kilometer. They start at the loca-
0.025 rad
Angela
r = 1 kilometer sec tions shown in Figure 17.14(a). Michael is traveling 0.025
starts here x-axis rad/sec and Angela is traveling 0.03 rad/sec. Impose co-
Michael
0.03 rad
sec
starts here ordinates as pictured. Where are the drivers located (in
xy-coordinates) after 18 seconds?
(a) Angela and Michael on the Solution. Let M(t) be the point on the circle of motion rep-
same test track. resenting Michael’s location after t seconds. Likewise, let
A(t) be the point on the circle of motion representing An-
y-axis
gela’s location after t seconds. Let θ(t) be the angle swept
M(t)
0.025 rad
out the by Michael and α(t) the angle swept out by Angela
sec
Angela
starts here
β(t) θ(t)
x-axis
after t seconds.
α(t) Michael
starts here
Since we are given the angular speeds, we get
0.03 rad
sec
A(t) θ(t) = 0.025t radians, and
α(t) = 0.03t radians.
(b) Modeling the motion of
Angela and Michael. From the previous Example 17.4.2,
Figure 17.14: Visualizing

M(t) = (cos(0.025t), sin(0.025t)), and
motion on a circular track. M(18) = (0.9004, 0.4350).
Angela’s angle α(t) is NOT in central standard position, so
we must observe that α(t) + π = β(t), where β(t) is in cen-
tral standard position: See Figure 17.14(b). We conclude
that
A(t) = (cos(β(t)), sin(β(t)))
= (cos(π + 0.03t), sin(π + 0.03t)).
So, after 18 seconds Angela’s location will be A(18) =
(−0.8577, −0.5141).
17.4. CIRCULAR FUNCTIONS 229
17.4.2 Relating circular functions and right triangles

If the point P on the unit circle is located in the first
quadrant, then we can compute cos(θ) and sin(θ) using unit circle (radius = 1)
trigonometric ratios. In general, it’s useful to relate right P
triangles, the unit circle and the circular functions. To de- sin(θ)
scribe this connection, given θ we place it in central stan- O

θ
R
dard position in the unit circle, where ∠ROP = θ. Draw
a line through P perpendicular to the x-axis, obtaining cos(θ)
an inscribed right triangle. Such a right triangle has hy-
potenuse of length 1, vertical side of length labeled b and Figure 17.15: The point P in
the first quadrant.
horizontal side of length labeled a. There are four cases:
See Figure 17.16.
P P
b b 1
0 θ θ
θ
a 0
1
a a 0
1
0
1 11
00 1
0 0
1
O R O R 00
11
O R θ O 0a
1 0
1
b R
b 00
11
P
P
CASE I CASE II CASE III CASE IV
Figure 17.16: Possible positions of θ on the unit circle.
Case I has already been discussed, arriving at cos(θ) = a and sin(θ) =

b. In Case II , we can interpret cos(θ) = −a, sin(θ) = b. We can reason
similarly in the other Cases III and IV, using Figure 17.16, and we arrive
at this conclusion:
Important Facts 17.4.4 (Circular functions and triangles). View θ as in

Figure 17.16 and form the pictured inscribed right triangles. Then we can
interpret cos(θ) and sin(θ) in terms of these right triangles as follows:
Case I : cos(θ) = a, sin(θ) = b

Case II : cos(θ) = −a, sin(θ) = b
Case III : cos(θ) = −a, sin(θ) = −b
Case IV : cos(θ) = a, sin(θ) = −b
17.5 What About Other Circles?

What happens if we begin with a circle Cr with radius r
T
P
(possibly different than 1) and want to compute the coor-
θ dinates of points on this circle?
O R S The circular functions can be used to answer this more
general question. Picture our circle Cr centered at the
Cr origin in the same picture with unit circle C1 and the angle
unit circle
θ in standard central position for each circle. As pictured,
Figure 17.17: Points on
we can view θ = ∠ROP = ∠SOT . If P = (x,y) is our point
other circles.
on the unit circle corresponding to the angle θ, then the
calculation below shows how to compute coordinates on
general circles:
P = (x,y)
= (cos(θ), sin(θ)) ∈ C1 ⇔ x2 + y2 = 1
⇔ r2x2 + r2y2 = r2
⇔ (rx)2 + (ry)2 = r2
⇔ T = (rx, ry)
= (r cos(θ), r sin(θ)) ∈ Cr.
Important Fact 17.5.1. Let Cr be a circle of radius r centered at the origin

and θ = ∠SOT an angle in standard central position for this circle, as in
Figure 17.17. Then the coordinates of T = (r cos(θ), r sin(θ)).
y-axis
Examples 17.5.2. Consider the picture below, with θ = 0.8
β = π − α = 2.9416
radians and α = 0.2 radians. What are the coordinates of
R
U Q the labeled points?
T θ
α S P
B O A x-axis
Solution. The angle θ is in standard central position; α is
circle radius = 1 a central angle, but it is not in standard position. Notice,
circle radius = 2
circle radius = 3 β = π − α = 2.9416 is an angle in standard central posi-
Figure 17.18: Coordinates
tion which locates the same points U, T, S as the angle α.
of points on circles. Applying Definition 17.4.1 on page 227:
P = (cos(0.8), sin(0.8)) = (0.6967, 0.7174)

Q = (2 cos(0.8), 2 sin(0.8)) = (1.3934, 1.4347)
R = (3 cos(0.8), 3 sin(0.8)) = (2.0901, 2.1521)
S = (cos(2.9416), sin(2.9416)) = (−0.9801, 0.1987)
T = (2 cos(2.9416), 2 sin(2.9416)) = (−1.9602, 0.3973)
U = (3 cos(2.9416), 3 sin(2.9416)) = (−2.9403, 0.5961).
17.6. OTHER BASIC CIRCULAR FUNCTION 231
Example 17.5.3. Suppose Cosmo begins at the position R

in the figure, walking around the circle of radius 20 feet
with an angular speed of 54 RPM counterclockwise. After 3 S
minutes have elapsed, describe Cosmo’s precise location.
P 20 feet
Solution. Cosmo has traveled 3 54 = 125
revolutions. If θ is R
12 radians

the angle traveled after 3 minutes, θ = 5 rev 2π rev =
24π
5
radians = 15.08 radians. By (15.5.1), we have x =
20 cos 24π
5
rad = −16.18 feet and y = 20 sin 24π 5
rad =
11.76 feet. Conclude that Cosmo is located at the point Figure 17.19: Where is
Cosmo after 3 minutes?
S = (−16.18, 11.76). Using (15.1), θ = 864◦ = 2(360◦) + 144◦ ;
this means that Cosmo walks counterclockwise around
the circle two complete revolutions, plus 144◦.
17.6 Other Basic Circular Function

Given any angle θ, our constructions offer a concrete link between the
cosine and sine functions and right triangles inscribed inside the unit
circle: See Figure 17.20.
P P
θ θ
θ R
O R O R O R θO
P
P
CASE I CASE II CASE III CASE IV
Figure 17.20: Computing the slope of a line using the function tan(θ).
The slope of the hypotenuse of these inscribed triangles is just the

slope of the line through OP. Since P = (cos(θ), sin(θ)) and O = (0, 0):
∆y sin(θ)
Slope = = ;
∆x cos(θ)
this would be valid as long as cos(θ) 6= 0. This calculation motivates a

new circular function called the tangent of θ by the rule
sin(θ)
tan(θ) = , provided cos(θ) 6= 0.
cos(θ)
The only time cos(θ) = 0 is when the corresponding point P on the

unit circle has x-coordinate 0. But, this only happens at the positions
(0, 1) and (0, −1) on the unit circle, corresponding to angles of the form
θ = ± π2 , ± 3π
2
, ± 5π
2
, · · · . These are the cases when the inscribed right tri-
angle would “degenerate” to having zero width and the line segment OP
becomes vertical. In summary, we then have this general idea to keep in
mind:
Important Fact 17.6.1. The slope of a line = tan(θ), where θ is the
angle the line makes with the x-axis (or any other horizontal line)
Three other commonly used circular functions come up from time to
time. The cotangent function y = cot(θ), the secant function y = sec(θ)
and the cosecant function y = csc(θ) are defined by the formulas:
def 1 def 1 def 1
sec(θ) = , csc(θ) = , cot(θ) = .
cos(θ) sin(θ) tan(θ)
Just as with the tangent function, one needs to worry about the values
of θ for which these functions are undefined (due to division by zero). We
will not need these functions in this text.
North Example 17.6.2. Three airplanes depart SeaTac Airport.

Alaska Northwest A NorthWest flight is heading in a direction 50◦ counter-
clockwise from East, an Alaska flight is heading 115◦ coun-
50 0
terclockwise from East and a Delta flight is heading 20◦
115 0
West East clockwise from East. Find the location of the Northwest
SeaTac 20 0 flight when it is 20 miles North of SeaTac. Find the loca-
South Delta
tion of the Alaska flight when it is 50 miles West of SeaTac.
Find the location of the Delta flight when it is 30 miles East
(a) The flight paths of three of SeaTac.
airplanes.
Solution. We impose a coordinate system in Fig-
Alaska x = 30 ure 17.21(a), where “East” (resp. “North”) points along the
Q
Northwest positive x-axis (resp. positive y-axis). To solve the prob-
N
lem, we will find the equation of the three lines represent-
ing the flight paths, then determine where they intersect
P
y = 20
the appropriate horizontal or vertical line. The Northwest
W E and Alaska directions of flight are angles in standard cen-
R
x = −50
Delta
tral position; the Delta flight direction will be −20◦ . We can
S imagine right triangles with their hypotenuses along the
(b) Modeling the paths of each directions of flight, then using the tangent function, we
flight. have these three immediate conclusions:
Figure 17.21: Visualizing
and modeling departing air-
slope NW line = tan(50◦ ) = 1.19,
planes. slope Alaska line = tan(115◦ ) = −2.14, and
slope Delta line = tan(−20◦ ) = −0.364.
17.6. OTHER BASIC CIRCULAR FUNCTION 233
All three flight paths pass through the origin (0,0) of our coordinate
system, so the equations of the lines through the flight paths will be:
NW flight : y = 1.19x,
Alaska flight : y = −2.14x,
Delta flight : y = −0.364x.
The Northwest flight is 20 miles North of SeaTac when y = 20; plugging

into the equation of the line of flight gives 20 = 1.19x, so x = 16.81 and
the plane location will be P = (16.81, 20). Similarly, the Alaska flight is
50 miles West of SeaTac when x = −50; plugging into the equation of the
line of flight gives y = (−2.14)(−50) = 107 and the plane location will be
Q = (−50, 107). Finally, check that the Delta flight is at R = (30, −10.92)
when it is 30 miles East of SeaTac.
17.7 Exercises
Problem 17.1. John has been hired to de- forest) as a potential landing site, but are un-
sign an exciting carnival ride. Tiff, the car- certain whether it is wide enough. They make
nival owner, has decided to create the world’s two measurements from A (see picture) finding
greatest ferris wheel. Tiff isn’t into math; she α = 25o and β = 54o . They rise vertically 100
simply has a vision and has told John these feet to B and measure γ = 47o . Determine the
constraints on her dream: (i) the wheel should width of the clearing to the nearest foot.
rotate counterclockwise with an angular speed
B
of 12 RPM; (ii) the linear speed of a rider should
be 200 mph; (iii) the lowest point on the ride 100 feet γ
should be 4 feet above the level ground. Recall, A
we worked on this in Exercise 16.5.
α
β
C E clearing D
12 RPM
θ
Problem 17.4. Marla is running clockwise
P
around a circular track. She runs at a con-
stant speed of 3 meters per second. She takes
4 feet
46 seconds to complete one lap of the track.
From her starting point, it takes her 12 sec-
onds to reach the northermost point of the
(a) Impose a coordinate system and find the track.
coordinates T (t) = (x(t),y(t)) of Tiff at Impose a coordinate system with the cen-
time t seconds after she starts the ride. ter of the track at the origin, and the northern-
most point on the positive y-axis.
(b) Tiff becomes a human missile after 6
seconds on the ride. Find Tiff’s coordi- (a) Give Marla’s coordinates at her starting
nates the instant she becomes a human point.
missile. (b) Give Marla’s coordinates when she has
been running for 10 seconds.
(c) Find the equation of the tangential line
along which Tiff travels the instant she (c) Give Marla’s coordinates when she has
becomes a human missile. Sketch a pic- been running for 901.3 seconds.
ture indicating this line and her initial
direction of motion along it when the Problem 17.5. A merry-go-round is rotating
seat detaches. at the constant angular speed of 3 RPM coun-
terclockwise. The platform of this ride is a cir-
cular disc of radius 24 feet. You jump onto the
Problem 17.2. (a) Find the equation of a ride at the location pictured below.
line passing through the point (-1,2) and
making an angle of 13o with the x-axis.
(Note: There are two answers; find them rotating 3 RPM
both.)
(b) Find the equation of a line making an jump on here
θ
angle of 8o with the y-axis and passing
through the point (1,1). (Note: There are
two answers; find them both.)
Problem 17.3. The crew of a helicopter needs

to land temporarily in a forest and spot a flat
horizontal piece of ground (a clearing in the
17.7. EXERCISES 235
(a) If θ = 34o , then what are your xy- relatively flat area nearby the tower (not nec-
coordinates after 4 minutes? essarily the same altitude as the bottom of the
tower), and standing some unknown distance
(b) If θ = 20o , then what are your xy-
away from the tower, you make three mea-
coordinates after 45 minutes?
surements all at the same height above sea
(c) If θ = −14o , then what are your xy- level. You observe that the top of the old tower
coordinates after 6 seconds? Draw an makes an angle of 39◦ above level. You move
accurate picture of the situation. 110 feet away from the original measurement
and observe that the old top of the tower now
(d) If θ = −2.1 rad, then what are your
makes an angle of 34◦ above level. Finally, af-
xy-coordinates after 2 hours and 7 sec-
ter the new construction is complete, you ob-
onds? Draw an accurate picture of the
serve that the new top of the tower, from the
situation.
same point as the second measurement was
(e) If θ = 2.1 rad, then what are your xy- made, makes an angle of 40◦ above the hori-
coordinates after 5 seconds? Draw an zontal. All three measurements are made at
accurate picture of the situation. the same height above sea level and are in line
with the tower. Find the height of the addition
to the tower, to the nearest foot.
Problem 17.6. Shirley is on a ferris wheel
which spins at the rate of 3.2 revolutions per
minute. The wheel has a radius of 45 feet, and Problem 17.9. Charlie and Alexandra are
the center of the wheel is 59 feet above the running around a circular track with radius
ground. After the wheel starts moving, Shirley 60 meters. Charlie started at the western-
takes 16 seconds to reach the top of the wheel. most point of the track, and, at the same time,
How high above the ground is she when Alexandra started at the northernmost point.
the wheel has been moving for 9 minutes? They both run counterclockwise. Alexandra
runs at 4 meters per second, and will take ex-
actly 2 minutes to catch up to Charlie.
Problem 17.7. The top of the Boulder Dam
Impose a coordinate system, and give the
has an angle of elevation of 1.2 radians from
x- and y-coordinates of Charlie after one
a point on the Colorado River. Measuring the
minute of running.
angle of elevation to the top of the dam from
a point 155 feet farther down river is 0.9 radi-
ans; assume the two angle measurements are Problem 17.10. George and Paula are run-
taken at the same elevation above sea level. ning around a circular track. George starts at
How high is the dam? the westernmost point of the track, and Paula
starts at the easternmost point. The illustra-
downriver tion below shows their starting positions and
running directions. They start running toward
each other at constant speeds. George runs at
dam 9 feet per second. Paula takes 50 seconds to
run a lap of the track. George and Paula pass
0.9 1.2 each other after 11 seconds.
N
155 ft a
George Paula
Problem 17.8. A radio station obtains a per-
mit to increase the height of their radio tower
on Queen Anne Hill by no more than 100 feet.
You are the head of the Queen Anne Commu-
nity Group and one of your members asks you After running for 3 minutes, how far east
to make sure that the radio station does not of his starting point is George?
exceed the limits of the permit. After finding a
Problem 17.11. A kite is attached to 300 feet (d) What are the coordinates of the bug af-
of string, which makes a 42 degree angle with ter 1 second? After 0 seconds? After 3
the level ground. The kite pilot is holding the seconds? After 22 seconds?
string 4 feet above the ground.
ω=4π/9rad/sec
bug lands here
kite
1.2 rad
o 2 in
42
4 feet
ground level
(a) How high above the ground is the kite?

(b) Suppose that power lines are located
250 feet in front of the kite flyer. Is
any portion of the kite or string over the
power lines?
bug lands here
Problem 17.12. In the pictures below, a bug

has landed on the rim of a jelly jar and is mov-
2 in
ing around the rim. The location where the
bug initially lands is described and its angu-
lar speed is given. Impose a coordinate sys- ω=4π/9rad/sec
tem with the origin at the center of the circle
of motion. In each of the cases, answer these
questions:
(a) Find an angle θ0 in standard central po-
sition that gives the bugs initial location. ω= 4π/9rad/sec
(In some cases, this is the angle given in
the picture; in other cases, you will need bug lands
to do something.) here
(b) The location angle of the bug at time t is 0.5 rad

given by the formula θ(t) = θ0 + ωt. Plug
in the values for θ0 and ω to explicitly
2 in
obtain a formula for θ(t).
(c) Find the coordinates of the bug at time
t.
Chapter 18
Trigonometric Functions
Our definitions of the circular functions are based upon the unit circle.
This makes it easy to visualize many of their properties.
18.1 Easy Properties of Circular Functions

How can we determine the range of function values for y-axis
(0,1)
cos(θ) and sin(θ)? To begin with, recall the abstract sin(θ)
P(θ)
ball moves
counterclockwise
definition for the range of a function f(θ): 1
θ (1,0)
(−1,0)
x-axis
cos(θ)
Range of f = {f(θ) : θ is in the domain}.
UNIT CIRCLE
(0, − 1)
Using the unit circle constructions of the basic circular Figure 18.1: Visualizing the
functions, it is easy to visualize the range of cos(θ) and range of sin(θ) and cos(θ).
sin(θ). Beginning at the position (1, 0), imagine a ball
moving counterclockwise around the unit circle. If we
“freeze” the motion at any point in time, we will have swept out an angle
θ and the corresponding position P(θ) on the circle will have coordinates
P(θ) = (cos(θ), sin(θ)).
y-axis light source

ball moves from 0 to
π radians around ball moves from 0
2
unit to π
2
radians
circle around unit circle
light
x-axis source
x-axis
y-axis
(a) What do you see on the (b) What do you see on the
y-axis? x-axis?
Figure 18.2: Projecting the coordinates of points onto the y-axis and the x-axis.
By studying the coordinates of the ball as it moves in the first quad-

rant, we will be studying cos(θ) and sin(θ), for 0 ≤ θ ≤ π/2 radians.
237
238 CHAPTER 18. TRIGONOMETRIC FUNCTIONS
We can visualize this very concretely. Imagine a light source as in Fig-

ure 18.2(a); then a shadow projects onto the vertical y-axis. The shadow
locations you would see on the y-axis are precisely the values sin(θ), for
0 ≤ θ ≤ π/2 radians. Similarly, imagine a light source as in Figure 18.2(b);
then a shadow projects onto the horizontal x-axis. The shadow locations
you would see on the x-axis are precisely the values cos(θ), for 0 ≤ θ ≤ π/2
radians.
There are two visual conclusions: First, the function values of sin(θ)
vary from 0 to 1 as θ varies from 0 to π/2. Secondly, the function values
of cos(θ) vary from 1 to 0 as θ varies from 0 to π/2. Of course, we can go
ahead and continue analyzing the motion as the ball moves into the sec-
ond, third and fourth quadrant, ending up back at the starting position
(1, 0). See Figure 18.3.
light source
y-axis
ball moves from
0 to 2π radians ball moves from 0
#2 around unit to 2π radians
#1 #2 #1
circle around nit circle
x-axis x-axis
#3 #4
#3
#4
light
source
(a) What do you see on the (b) What do you see on the
y-axis? x-axis?
Figure 18.3: Analyzing the values of the sine and cosine functions.
The conclusion is that after one complete counterclockwise rotation,

the values of sin(θ) and cos(θ) range over the interval [−1, 1]. As the
ball moves through the four quadrants, we have indicated the “order” in
which these function values are assumed by labeling arrows #1 — #4:
For example, for the sine function, look at Figure 18.3(a). The values of
the sine function vary from 0 up to 1 while the ball moves through the
first quadrant (arrow labeled #1), then from 1 down to 0 (arrow labeled
#2), then from 0 down to −1 (arrow labeled #3), then from −1 up to 0
(arrow labeled #4).
What about the tangent function? We have seen that the tangent
function computes the slope of the hypotenuse of an inscribed triangle.
This means we can determine the range of values of tan(θ) by investigat-
ing the possible slopes for these inscribed triangles. We will maintain the
above model of a ball moving around the unit circle.
We look at two cases, each starting at (1, 0). In the first quadrant, the
ball moves counterclockwise and in the fourth quadrant it moves clock-
wise: In the first quadrant, we notice that these hypotenuse slopes are
always non-negative, beginning with slope 0 (the degenerate right triangle
when θ = 0) then increasing. In fact, as the angle θ approaches π/2 ra-
dians, the ball is getting closer to the position (0, 1) and the hypotenuse
18.1. EASY PROPERTIES OF CIRCULAR FUNCTIONS 239
y-axis y-axis
8 7
6
5
4
1
x-axis 1
x-axis
2
6
7
(a) What happens to the (b) What happens to the

slopes of these triangles? slopes of these triangles?
Figure 18.4: Analyzing the values of the tangent function.
is approaching a vertical line. This tells us that as θ varies from 0 to

π/2 (but not equal to π/2), these slopes attain all possible non-negative
values. In other words, the range of values for tan(θ) on the domain
0 ≤ θ < π/2 will be 0 ≤ z < ∞. Similar reasoning shows that as the ball
moves in the fourth quadrant, the slopes of the hypotenuses of the tri-
angles are always non-positive, varying from 0 to ANY negative value. In
other words, the range of values for tan(θ) on the domain −π/2 < θ ≤ 0
will be −∞ < z ≤ 0.
On your calculator, you can verify the visual conclusions we just es-
tablished by studying the values of tan(θ) for θ close (but not equal) to π2
radians = 90◦ :
tan(89◦ ) = 57.29 tan(−89◦ ) = −57.29

tan(89.9◦) = 572.96 tan(−89.9◦) = −572.96
tan(89.99◦) = 5729.58 tan(−89.99◦) = −5729.58
.. ..
. .
The fact that the values of the tangent function become arbitrarily large
as we get close to ±π/2 radians means the function output values are
unbounded.
Important Fact 18.1.1 (Circular function values). For any angle θ, we

always have −1 ≤ cos(θ) ≤ 1 and −1 ≤ sin(θ) ≤ 1. On domain 0 ≤ θ ≤ 2π,
the range of both cos(θ) and sin(θ) is −1 ≤ z ≤ 1. In contrast, on the
domain of all θ values for which tangent is defined, the range of tan(θ) is
all real numbers.
For the sine and cosine functions, if the domain is not 0 ≤ θ ≤ 2π,
then we need to consider the “periodic qualities” of the circular functions
to determine the range. This is discussed below.
18.2 Identities
There are dozens of formulas that relate the values of two or more cir-
cular functions; these are usually lumped under the heading of Trigono-
metric Identities. In this course, we only need a couple frequently used
identities.
If we take the point P = (cos(θ), sin(θ)) on the unit circle, correspond-
ing to the standard central position angle θ, then recall the equation of
the unit circle tells us x2 +y2 = 1. But, since the x coordinate is cos(θ) and
the y coordinate is sin(θ), we have (cos(θ))2 + (sin(θ))2 = 1. It is common
notational practice to write (cos(θ))2 = cos2(θ) and (sin(θ))2 = sin2(θ).
This leads to the most important of all trigonometric identities:
Important Fact 18.2.1 (Trigonometric identity). For any angle θ, we

have the identity cos2(θ) + sin2(θ) = 1.
Adding any multiple of 2π radians (or 360◦ ) to an angle will not change
the values of the circular functions. If we focus on radians for a moment,
this says that knowing the values of cos(θ) and sin(θ) on the domain
0 ≤ θ ≤ 2π determines the values for any other possible angle.
There is something very general going on here, so let’s pause a mo-
ment to make a definition and then an observation.
Definition 18.2.2 (Periodic function). For c > 0, a function f(θ) is called

c-periodic if two things are true:
(i) f(θ + c) = f(θ) holds for all θ;
(ii) There is no smaller d, 0 < d < c, such that f(θ + d) = f(θ) holds for all
θ.
We usually call c the period of the function.
Using this new terminology, we conclude that the sine and cosine
circular functions are 2π-periodic. In the case of the tangent circular
function, it is also true that tan(θ) = tan(θ + 2πn), for every integer n.
However, referring back to the unit circle definitions of the circular func-
tions, we have tan(θ) = tan(θ + nπ), for all integers n. If you take n = 1,
then this tells us that the tangent circular function is π-periodic. We
summarize this information below.
Important Fact 18.2.3 (Periodicity identity). For any angle θ and any
integer n = 0, ±1, ±2, ±3, . . . , we have cos(θ) = cos(θ+2πn), sin(θ) = sin(θ+
2πn), and tan(θ) = tan(θ + nπ).
18.2. IDENTITIES 241
Next, we draw an angle θ and its negative in the same unit

(cos(θ), sin(θ)) = P θ circle
unit circle picture in standard central position. We have θ
indicated the points Pθ and P−θ used to define the circu- O
R
lar functions. It is clear from the picture in Figure 18.5 −θ
that Pθ and P−θ have the same x-coordinate, but the y- (cos(−θ), sin(−θ)) = P −θ
coordinates are negatives of one another. This gives the Figure 18.5: Visualizing a
next identity: trigonometric identity.
Important Fact 18.2.4 (Even/Odd identity). For any angle θ, sin(−θ) =

− sin(θ), and cos(−θ) = cos(θ).
We can use the terminology of even and odd functions here. In this
language, this result says that the cosine function is an even function
and the sine function is an odd function.
Next, draw the angles θ and θ + π in the same unit
circle picture in standard central position. We have in- P θ = (cos(θ), sin(θ))
θ
dicated the corresponding points Pθ and Pθ+π on the unit
R
circle and their coordinates in terms of the circular func- θ+π
unit P θ+π =
tions: From the picture in Figure 18.6, the x-coordinate of circle (cos(θ + π), (sin(θ + π))
Pθ must be the “negative” of the x-coordinate of Pθ+π and Figure 18.6: Visualizing
similarly, the y-coordinate of Pθ must be the “negative” of Fact 18.2.5.
the y-coordinate of Pθ+π. This gives us the next identity:
Important Fact 18.2.5 (Plus π identity). For any angle θ, we have sin(θ+
π) = − sin(θ), and cos(θ + π) = − cos(θ).
Important Fact 18.2.6. For any angle θ, we have sin(π − θ) = sin(θ) and
cos(π − θ) = − cos(θ).
For example, we have sin 5π π

1
6
= sin 6
= 2 . This calculation leads to a
computational observation: Combining Table 17.1 with the previous two
identities we can compute the EXACT value of cos(θ), sin(θ), and tan(θ)
at an angle θ which is a multiple of 30◦ = π6 radians or 45◦ = π4 radians.
Here are some sample calculations together with a reference as to “why”
each equality is valid:
Example 18.2.7.
(i) cos(−45◦ ) = cos(45◦ ) Fact 18.2.4 on page 241

√
2
= Table 17.1 on page 223
2
(ii) sin(225◦ ) = sin(45◦ + 180◦ ) Fact 18.2.3 on page 240

= − sin(45◦ ) Fact 18.2.5 on page 241
√
2
=− Table 17.1 on page 223
2

2π π
(iii) cos = cos − + π Fact 18.2.3 on page 240
3 3
π
= − cos − Fact 18.2.5 on page 241
π 3
= − cos Fact 18.2.4 on page 241
3
1
=− Table 17.1 on page 223
2
18.3 Graphs of Circular Functions

We have introduced three new functions of the variable θ and it is impor-
tant to understand and interpret the pictures of their graphs. To do this,
we need to settle on a coordinate system in which to work. The horizontal
axis will correspond to the independent variable, so this should be the θ-
axis. We will label the vertical axis, which corresponds to the dependent
variable, the z-axis. With these conventions, beginning with any of the
circular functions z = sin(θ), z = cos(θ), or z = tan(θ), the graph will be
a subset of the θz-coordinate system. Precisely, given a circular function
z = f(θ), the graph consists of all pairs (θ, f(θ)), where θ varies over a do-
main of allowed values. We will record and discuss these graphs below;
a graphing device will painlessly produce these for us!
There is a point of possible confusion that needs attention. We pur-
posely did not use the letter “y” for the dependent variable of the circular
functions. This is to avoid possible confusion with our construction of
the sine and cosine functions using the unit circle. Since we viewed the
unit circle inside the xy-coordinate system, the x-coordinates (resp. y-
coordinates) of points on the unit circle are computed by cos(θ) (resp.
sin(θ)).
18.3. GRAPHS OF CIRCULAR FUNCTIONS 243
z y
P = (cos(θ),sin(θ))
θ
θ x
Coordinate system used to Coordinate system used to DE-

GRAPH the circular functions. FINE the circular functions.
Figure 18.7: The zθ versus xy coordinate systems.
18.3.1 A matter of scaling

The first issue concerns scaling of the axes used in graphing the circular
functions. As we know, the definition of radian measure is directly tied
to the lengths of arcs subtended by angles in the unit circle:
Important Fact 18.3.1. An angle of measure 1 radian inside the unit

circle will subtend an arc of length 1.
Since length is a good intuitive scaling quantity, it is natural to scale

the θ-axis so that the length of 1 radian on the θ-axis (horizontal axis)
is the same length as 1 unit on the vertical axis. For this reason, we
will work primarily with radian measure when sketching the graphs of
circular functions. If we need to work explicitly with degree measure for
angles, then we can always convert radians to degrees using the fact:
360◦ = 2π radians.
18.3.2 The sine and cosine graphs

Using Fact 18.1.1, we know that −1 ≤ sin(θ) ≤ 1 and −1 ≤ cos(θ) ≤ 1.
Pictorially, this tells us that the graphs of z = sin(θ) and z = cos(θ) lie
between the horizontal lines z = 1 and z = −1; i.e. the graphs lie inside
the darkened band pictured in Figure 18.8.
By Fact 18.2.3, we know that the values of the sine and cosine repeat
themselves every 2π radians. Consequently, if we know the graphs of the
sine and cosine on the domain 0 ≤ θ ≤ 2π, then the picture will repeat for
the interval 2π ≤ θ ≤ 4π, −2π ≤ θ ≤ 0, etc.
z-axis
z=1
θ-axis
z = −1
Figure 18.8: Visualizing the range of sin(θ) and cos(θ).
z-axis
y=1
Repeat Repeat Repeat Repeat

θ-axis
−4π −2π 0 2π 4π 6π
y = −1
picture in here repeats each 2π units
Figure 18.9: On what intervals will the graph repeat?
Sketching the graph of z = sin (θ) for 0 ≤ θ ≤ 2π can be roughly

π 1
π √2 π √3
achieved by plotting points. For example, 6 , 2 , 4 , 2 , 3 , 2 , and
π
3π
5π √3 7π √2 11π 1
2
, 1 lie on the graph, as do 2
, −1 , 3 , − 2 , 4 , − 2 , 6 , − 2 , and
(2π, 0), etc. If we return to our analysis of the range of values for the
sine function in Figure 18.2, it is easy to see where sin(θ) is positive or
negative; combined with Chapter 4, this tells us where the graph is above
and below the horizontal axis (see Figure 18.10).
z-axis
θ-axis
−2π −π π 2π 3π
Figure 18.10: Where is the graph positive or negative?
We now include a software plot of the graph of sine function, observing

the three qualitative features just isolated: bounding, periodicity and
sign properties (see Figure 18.11).
18.3. GRAPHS OF CIRCULAR FUNCTIONS 245
z-axis
1
θ-axis
−2π −π π 2π 3π
−1
one period
Figure 18.11: The graph of z = sin(θ).
We could repeat this analysis to arrive at the graph π

−θ
of the cosine. Instead, we will utilize an identity. Given 2
an angle θ, place it in central standard position in the

unit circle, as one of the four cases of Figure 17.16. For θ
example, we have pictured Case I in this figure. Since π
the sum of the angles in a triangle is 180◦ = π radians, unit
circle
2
we know that θ, π/2, and π2 − θ are the three angles of the

inscribed right triangle. From the picture in Figure 18.12, Figure 18.12: Visualizing
it then follows that the conversion identity.
side adjacent to θ
cos(θ) =
hypotenuse
side opposite to π2 − θ

=
hypotenuse
π
= sin −θ .
2
Using the same reasoning this identity is valid for all θ. This gives us
another useful identity:
Important Fact 18.3.2 (Conversion identity). For any angle θ, cos(θ) =
sin π2 − θ , and sin(θ) = cos π2 − θ .
This identity can be used to sketch the graph of the cosine function.
First, we do a calculation using our new identity:
cos(θ) = cos(−θ) Fact 18.2.4 on page 241

π
= sin − (−θ) Fact 18.3.2 (above)
2 π
= sin θ − − Since: (a + b) = (a − (−b)) = (b − (−a))
2
By the horizontal shifting principle in Fact 13.3.1 on page 170, the graph
of z = cos(θ) is obtained by horizontally shifting the graph of z = sin(θ)
by π2 units to the left. Here is a plot of the graph of the cosine function:
See Figure 18.13.
z-axis
1
θ-axis
−2π −π π 2π 3π
−1
one period
Figure 18.13: The graph of z = cos(θ).
18.3.3 The tangent graph

θ = −π
2 graph heads this
As we have already seen, unlike the sine and cosine circu-
way, getting close
to vertical line as θ
lar functions, the tangent function is NOT defined for all
gets close to π sin(θ)
2 values of θ. Since tan(θ) = cos(θ) , here are some properties
−π
2
π
2 θ-axis we can immediately deduce:
graph heads this
way, getting close
to vertical line as θ • The function z = tan(θ) is undefined if and only if
gets close to − π
z-axis θ= π
2
2
θ = π2 + kπ, where k = 0, ±1, ±2, ±3, · · · .
Figure 18.14: The behav-
ior of tan(θ) as θ approaches • The function z = tan(θ) = 0 if and only if θ = kπ,
asymptotes. where k = 0, ±1, ±2, ±3, · · · .
• By Fact 18.2.3, the tangent function is π-periodic,

so the picture of the graph will repeat itself every π-
units and it is enough to understand the graph when
− π2 < θ < π2 .
• On the domain 0 ≤ θ < π2 , tan(θ) ≥ 0; on the domain
− π2 < θ ≤ 0, tan(θ) ≤ 0.
In the θz-coordinate system, the vertical lines θ = π2 + kπ, where
k = 0, ±1, ±2, · · · will be vertical asymptotes for the graph of the tangent
function. Using our slope interpretation in Figure 18.4, what becomes
clear is this: As the values of θ get close to π2 , the graph is getting close
to the vertical line θ = π2 AND becoming farther and farther away from
the horizontal axis: To understand this numerically, first suppose
θ is
π π π π

slightly smaller than 2 , say θ = 2 − 0.1 , 2 − 0.01 , and 2 − 0.001 . Then
the calculation of tan(θ) involves dividing a number very close to 1 by a
very small positive number:
π
tan − 0.1 = 9.9666,
π2
tan − 0.01 = 99.9967, and
π2
tan − 0.001 = 1000.
2
18.4. TRIGONOMETRIC FUNCTIONS 247
θ = −π
2
z-axis one period
Conclude that as θ “approaches π2 from below”, the val- θ = 5π
2
ues of tan(θ) are becoming larger and larger. This says etc. etc.
that the function values become “unbounded”. Likewise, θ-axis
imagine the case when θ is slightly bigger than − π2 , say −2π −π π 2π 3π
π π π

θ = − 2 + 0.1 , − 2 + 0.01 , and − 2 + 0.001 . Then the
calculation of tan(θ) involves dividing a number very close θ = − 3π θ= π θ = 3π
2 2 2
to −1 by a very small positive number: Figure 18.15: The graph of
π z = tan(θ).
tan − + 0.1 = −9.9666,
π2
tan − + 0.01 = −99.9967, and
π2
tan − + 0.001 = −1000.
2
Conclude that as θ “approaches − π2 from above”, the values of tan(θ) are
becoming negative numbers of increasingly larger magnitude:
Again, this tells us the function values are becoming “unbounded”.
The graph of z = tan(θ) for − π2 < θ < π2 can be roughly achieved by com-
bining the calculations as in Example 18.2.7 and the qualitative features
highlighted. Figure 18.15 shows a software plot.
18.4 Trigonometric Functions

To become successful mathematical modelers, we must have wide variety
of functions in our toolkit. As an illustration, the graph below might
represent the height of the tide above some reference level over the course
of several days. The curve drawn is clearly illustrating that the height of
the tide is “periodic” as a function of time t; in other words, the behavior
of the tide repeats itself as time goes by. However, if we try to model this
periodic behavior, the only weapon at our disposal would be the circular
functions and these require an angle variable, not a time variable such
as t; we are stuck!
Modeling the tide graph requires the trigonometric functions, which lie
at the heart of studying all kinds of periodic behavior. We have no desire
to “reinvent the wheel”, so let’s use our previous work on the circular
functions to define the trigonometric functions.
18.4.1 A Transition
Given a real number t, is there a sensible way to define cos(t) and sin(t)?
The answer is yes and depends on the ideas surrounding radian measure
of angles. Given the positive real number t, we can certainly imagine an
angle of measure t radians inside the unit circle (in standard position)
and we know the arc subtended by this angle has length t (this is why
we use the unit circle).
feet
+20
+15
+10
+5
0
−5 t (time)
−10
−15
−20
Figure 18.16: A periodic function with input variable t.
P(t) = (x,y)
arc length = t
We already know that cos(t radians) and sin(t radians)
compute the x and y coordinates of the point P(t). In
t rads
effect, we are just using the measure of the angle t to
(1,0)
help us locate the point P(t). An alternate way to locate
P(t) is to move along the circumference counterclockwise,
unit circle beginning at (1, 0), until we have an arc of length t; that
Figure 18.17: The circular again puts us at the point P(t). In the case of an angle
functions with input variable of measure −t radians, the point P(−t) can be located by
t.
moving along the circumference clockwise, beginning at
(1, 0), until we have an arc of length t.
Definition 18.4.1 (Trigonometric functions). Let t be a real number. We

DEFINE the sine function y = sin(t), the cosine function y = cos(t) and the
tangent function y = tan(t) by the rules
def
sin(t) = y-coordinate of P(t) = sin(t radians)
def
cos(t) = x-coordinate of P(t) = cos(t radians)
def sin(t)
tan(t) = = tan(t radians)
cos(t)
We refer to these as the basic trigonometric functions. If we are work-

ing with radian measure and t is a real number, then there is no differ-
ence between evaluating a trigonometric function at the real number t
and evaluating the corresponding circular function at the angle of mea-
sure t radians.
Example 18.4.2. Assume that the number of hours of daylight in Seattle

2π
during 1994 is given by the function d(t) = 3.7sin 366 (t − 80.5) + 12, where
18.4. TRIGONOMETRIC FUNCTIONS 249
t represents the day of the year and t = 0 corresponds to January 1. How

many hours of daylight will there be on May 11?
Solution. To solve the problem, you need to consult a calendar, finding

every month has 31 days, except: February has 28 days and April, June,
September and November have 30 days. May 11 is the 31+28+31+30+11 =
131st day of the year. So, there will be d(131) = 3.7 sin(2(50.5)π/366) + 12 =
14.82 hours of daylight on May 11.
18.4.2 Graphs of trigonometric functions

The graphs of the trigonometric functions y = sin(t), y = cos(t), and
y = tan(t) will look just like Figures 18.11, 18.13, and 18.15, except that
the horizontal axis becomes the t-axis and the vertical axis becomes the
y-axis.
18.4.3 Notation for trigonometric functions

In many texts, you will find the sine function written as y = sin t; i.e. the
parenthesis around the “t” are omitted. A similar comment applies to all
of the trigonometric functions. We will never do this and the reasoning
is simply this: Maintaining the parenthesis, as in y = sin(t), emphasizes
the fact that we are dealing with a function and the “input values” are
located between the parenthesis. For example, if we write the function
y = sin(t2 +2t+1), it is crystal clear that the sine function is applied to the
expression “t2 + 2t + 1”; using the alternate notation yields the expression
y = sin t2 + 2t + 1, which is interpreted to mean y = (sin t2) + (2t + 1).
!!!
As a rule, whenever you see an expression involving sin(· · · ), cos(· · · ),
CAUTION
or tan(· · · ), we assume “ · · · ” is in units of RADIANS, unless otherwise
!!!
noted. When computing values on your calculator, MAKE SURE YOU
ARE USING RADIAN MODE!
18.5 Exercises
Problem 18.1. Work the following problems Problem 18.5. Start with the equation
without using ANY calculators. sin(θ) = cos(θ). Use the unit circle interpre-
tation of the circular functions to find the so-
(a) Sketch y = sin(x).
lutions of this equation; make sure to describe
(b) Sketch y = sin2 (x). your reasoning.
1
(c) Sketch y = .
1 + sin2 (x) Problem 18.6. A six foot long rod is attached
at one end A to a point on a wheel of radius 2
feet, centered at the origin. The other end B is
Problem 18.2. Sketch the graphs of these free to move back and forth along the x-axis.
functions: The point A is at (2,0) at time t = 0, and the
(a) f(t) = | sin(t)|. wheel rotates counterclockwise at 3 rev/sec.
(b) f(t) = | cos(t)|. 2
A
(c) f(t) = | tan(t)|. A B B
-2 2 4 6 8 2 4 6 8
-2
Problem 18.3. Solve the following: time t=0
time t > 0
24
(a) If cos(θ) = 25 , what are the two possible
values of sin(θ)? (a) As the point A makes one complete rev-
olution, indicate in the picture the direc-
(b) If sin(θ) = −0.8 and θ is in the third tion and range of motion of the point B.
quadrant of the xy plane, what is cos(θ)?
(b) Find the coordinates of the point A as a
(c) If sin(θ) = 73 , what is sin( π2 − θ)?
function of time t.
(c) Find the coordinates of the point B as a
Problem 18.4. These graphs represent peri-
function of time t.
odic functions. Describe the period in each
case. (d) What is the x-coordinate of the point B
when t = 1? You should be able to find
this two ways: with your function from
part (c), and using some common sense
(where is point A after one second?).
(e) Is the function you found in (c) a sinu-
soidal function? Explain.
Chapter 19
Sinusoidal Functions
A migrating salmon is heading up a portion of the 5
Columbia River. It’s depth d(t) (in feet) below the water 4
depth
3
surface is measured and plotted over a 30 minute period,
2
as a function of time t (minutes). What is the formula for 1
d(t)?
5 10 15 20 25 30
In order to answer the question, we need to introduce time
an important new family of functions called the sinusoidal Figure 19.1: The depth of a
functions. These functions will play a central role in mod- salmon as a function of time.
eling any kind of periodic phenomena. The amazing fact
is that almost any function you will encounter can be approximated by a
sum of sinusoidal functions; a result that has far-reaching implications
in all of our lives.
19.1 A special class of functions

Beginning with the trigonometric function y = sin(x), what is the most
general function we can build using the graphical techniques of shifting
and stretching?
y-axis
1
y = sin(x)
horizontally shift
vertically vertically
x-axis
π
dilate shift
−π −π π
horizontally dilate 2 2
−1
Figure 19.2: Visualizing the geometric operations available for curve sketching.
The graph of y = sin(x) can be manipulated in four basic ways: hori-

zontally shift, vertically shift, horizontally dilate or vertically dilate. Each
of these “geometric operations” corresponds to a simple change in the
“symbolic formula” for the function, as discussed in Chapter 13.
251
252 CHAPTER 19. SINUSOIDAL FUNCTIONS
If we vertically shift the graph by D units upward, the resulting curve

would be the graph of the function y = sin(x)+D; see Facts 13.3.1. Recall,
the effect of the sign of D: If D is negative, the effect of shifting D units
upward is the same as shifting |D| units downward. Notice, the function
y = sin(x) + D is still a periodic function, having the same period 2π as
y = sin(x). Notice, whereas the graph of the function y = sin(x) oscillates
between the horizontal lines y = ±1, the graph of y = sin(x) + D oscillates
between y = D ± 1. For this reason, we sometimes refer to the constant
D as the mean of the function y = sin(x) + D. In Figure 19.3, notice that
the graph of y = sin(x) +D is symmetrically split by the horizontal “mean”
line y = D.
y-axis y-axis y = sin(x) + D
y = sin(x)
y = D
D
x-axis x-axis
shift D units
Figure 19.3: Interpreting the mean.
Next, consider the effect of horizontally shifting the graph of y = sin(x)

by C units to the right. By Facts 13.3.1, the new curve is the graph of
the function y = sin(x − C). Also, recall the effect of the sign of C: If
C is negative, the effect of shifting C units right is the same as shifting
|C| units left. If the domain of sin(x) is 0 ≤ x ≤ 2π, then the domain of
sin(x − C) is 0 ≤ x − C ≤ 2π, again by Facts 13.3.1. Rewriting this, the
domain of sin(x − C) is C ≤ x ≤ 2π + C and the graph will go through
precisely one period on this domain. In other words, the new function
sin(x − C) is still 2π-periodic. The constant C is usually called the phase
shift of y = sin(x − C). Looking at Figure 19.4, it is possible to interpret C
graphically: C will be a point where the graph crosses the horizontal axis
on its way up from a minimum to a maximum.
y-axis y-axis C
y = sin(x) y = sin(x − C)
x-axis x-axis
shift C units
Figure 19.4: Interpreting the phase shift.
Vertically dilating the graph, either by vertical expansion or compres-

sion, leads to a new curve. The graph of this vertically dilated curve
is y = A sin(x), for some positive constant A. Furthermore, if A > 1,
19.1. A SPECIAL CLASS OF FUNCTIONS 253
the graph of y = A sin(x) is a vertically expanded version of y = sin(x),

whereas, if 0 < A < 1, then the graph of y = A sin(x) is a vertically com-
pressed version of y = sin(x). Notice, the function y = A sin(x) is still 2π-
periodic. What has changed is the band of oscillation: whereas the graph
of the function y = sin(x) stays between the horizontal lines y = ±1, the
graph of y = A sin(x) oscillates between the horizontal lines y = ±A. We
usually refer to A as the amplitude of the function y = A sin(x).
y-axis y = A sin(x)
y-axis
y = sin(x)
A
x-axis
x-axis
stretch A units
Figure 19.5: Interpreting the amplitude.
Finally, horizontally dilating the graph, either by horizontal expansion

or compression, leads to a new curve. The equation of this horizontally
dilated curve is y = sin(cx), for some constant c > 0. We know that y =
sin(x) is a 2π-periodic function and observe that horizontally dilation still
results in a periodic function, but the period will typically NOT be 2π. For
future purposes, it is useful to rewrite the equation for the horizontally
stretched curve in a way more directly highlighting the period. To begin
with, once the horizontal stretching factor c is known, we could rewrite
2π
c= , for some B 6= 0.
B
y-axis y-axis y = sin(( 2π

B
)x)
y = sin(x)
x-axis
x-axis
stretch
Figure 19.6: Interpreting the period.
Here is the point of this yoga with the horizontal dilating constant: If
we let the values of x range over the interval [0, B], then 2π
B
x will
range over
2π
the interval [0,2π]. In other words, the function y = sin B x is B-periodic
and we can read off the period of y = sin 2π B
x by viewing the constant
in this mysterious way. The four constructions outlined lead to a new
family of functions.
Definition 19.1.1 (The Sinusoidal Function). Let A, B, C and D be fixed

constants, where A and B are both positive. Then we can form the new
function

2π
y = A sin (x − C) + D,
B
which is called a sinusoidal function. The four constants can be interpreted

graphically as indicated:
y-axis y-axis
x-axis x-axis
C
all four
operations
y = sin(x)
y = A sin(( 2π
B
)(x − c)) + D
Figure 19.7: Putting it all together for the sinusoidal function.
19.1.1 How to roughly sketch a sinusoidal graph

Important Procedure 19.1.2. Given a sinusoidal function in the standard
form

2π
y = A sin (x − C) + D,
B
once the constants A, B, C, and D are specified, any graphing device can
produce an accurate graph. However, it is pretty straightforward to sketch
a rough graph by hand and the process will help reinforce the graphical
meaning of the constants A, B, C, and D. Here is a “five step procedure”
one can follow, assuming we are given A, B, C, and D. It is a good idea to
follow Example 19.1.3 as you read this procedure; that way it will seem a
lot less abstract.
1. Draw the horizontal line given by the equation y = D; this line will
2π

“split” the graph of y = A sin B (x − C) + D into symmetrical upper
and lower halves.
2. Draw the two horizontal lines given by the equations y = D±A. These
two lines determine a horizontal strip inside which the graph of the
sinusoidal function will oscillate. Notice, the points where the sinu-
soidal function has a maximum value lie on the line y = D + A. Like-
wise, the points where the sinusoidal function has a minimum value
lie on the line y = D − A. Of course, we do not yet have a prescription
that tells us where these maxima (peaks) and minima (valleys) are
located; that will come out of the next steps.
3. Since we are given the period B, we know these important facts: (1)
The period B is the horizontal distance between two successive max-
ima (peaks) in the graph. Likewise, the period B is the horizontal
distance between two successive minima (valleys) in the graph. (2)
The horizontal distance between a maxima (peak) and the successive
minima (valley) is 21 B.
4. Plot the point (C, D). This will be a place where the graph of the
sinusoidal function will cross the mean line y = D on its way up
from a minima to a maxima. This is not the only place where the
graph crosses the mean line; it will also cross at the points obtained
from (C, D) by horizontally shifting by any integer multiple of 12 B.
For example, here are three places the graph crosses the mean line:
(C, D), (C + 21 B, D), (C + B, D)
5. Finally, midway between (C, D) and (C+ 21 B, D) there will be a maxima

(peak); i.e. at the point (C + 14 B, D + A). Likewise, midway between
(C + 21 B, D) and (C + B, D) there will be a minima (valley); i.e. at the
point (C + 34 B,D − A). It is now possible to roughly sketch the graph on
the domain C ≤ x ≤ C + B by connecting the points described. Once
this portion of the graph is known, the fact that the function is periodic
tells us to simply repeat the picture in the intervals C + B ≤ x ≤ C + 2B,
C − B ≤ x ≤ C, etc.
To make sense of this procedure, let’s do an explicit example to see

how these five steps produce a rough sketch.
eplacements
Example 19.1.3. The temperature (in ◦ C) of Adri-N’s dorm room varies
π

during the day according to the sinusoidal function d(t) = 6 sin 12 (t − 11) +
19, where t represents hours after midnight. Roughly sketch the graph of
d(t) over a 24 hour period.. What is the temperature of the room at 2:00
pm? What is the maximum and minimum temperature of the room?
Solution. We begin with the rough sketch. Start by taking an inventory

of the constants in this sinusoidal function:
π
2π
d(t) = 6 sin (t − 11) + 19 = A sin (t − C) + D.
12 B
Conclude that A = 6, B = 24, C = 11, D = 19. Following the first four

steps of the procedure outlined, we can sketch the lines y = D = 19,
y = D ± A = 19 ± 6 and three points where the graph crosses the mean
line (see Figure19.8).
30
25 y = 25
d(t) graph
will oscillate 20
inside this y = 19
strip (11, 19) (23, 19) (35, 19)
15
y = 13
10
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38
Figure 19.8: Sketching the mean D and amplitude A.
According to the fifth step in the sketching procedure, we can plot the
maxima (C + 41 B, D + A) = (17, 25) and the minima (C + 43 B, D − A) = (29, 13).
We then “connect the dots” to get a rough sketch on the domain 11 ≤ t ≤
35.
30 maxima
(17, 25)
25 y = 25
d(t) graph
will oscillate (23, 19)
20
inside this y = 19
strip (11, 19) (35, 19)
15
y = 13
(29, 13)
10 minima
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38
Figure 19.9: Visualizing the maximum and minimum over one period.
Finally, we can use the fact the function has period 24 to sketch the
graph to the right and left by simply repeating the picture every 24 hori-
zontal units.
y-axis
30 maxima
maxima maxima
(−7,25) (17,25) (41,25)
25 y = 25
20 (11,19) (23,19) (35,19)

y = 19
(−13,19) (−1,19) (47,19) (59,19)
15
y = 13
10 (5,13) (29,13) (53,13)
minima minima minima
t-axis
−12 −8 −4 2 6 10 14 18 22 26 30 34 38
−14 −10 −6 −2 4 8 12 16 20 24 28 32 36
Figure 19.10: Repeat sketch for every full period.
We restrict the picture to the domain 0 ≤ t ≤ 24 and obtain the com-

puter generated graph pictured in Figure 19.11; as you can see, our
rough graph is very accurate. The temperature at 2:00 p.m. is just
d(14) = 23.24◦ C. From the graph, the maximum value of the function will
be D + A = 25◦ C and the minimum value will be D − A = 13◦ C.
25
temp (C)
20
15
10
5 10 15 20
t (hours)
Figure 19.11: The computer generated solution.
19.1.2 Functions not in standard sinusoidal form

Any time we are given a trigonometric function written in the standard
form

2π
y = A sin (x − C) + D,
B
for constants A, B, C, and D (with A and B positive), the summary in
Definition 19.1.1 tells us everything we could possibly want to know
about the graph. But, there are two ways in which we might encounter a
trigonometric type function that is not in this standard form:
• The constants A or B might benegative. For example, y = −2 sin(2x−
7) − 3 and y = 3 sin − 21 x + 1 + 4 are examples that fail to be in
standard form.
• We might use the cosine function in place of the sine function. For
example, something like y = 2 cos(3x + 1) − 2 fails to be in standard
sinusoidal form.
Now what do we do? Does this mean we need to repeat the analysis that
led to Definition 19.1.1? It turns out that if we use our trig identities
just right, then we can move any such equation into standard form and
read off the amplitude, period, phase shift and mean. In other words,
equations that fail to be in standard sinusoidal form for either of these
two reasons will still define sinusoidal functions. We illustrate how this
is done by way of some examples:
Examples 19.1.4.
(i) Start with y = −2 sin(2x−7) −3, then here are the steps with reference
to the required identities to put the equation in standard form:
y = −2 sin(2x − 7) − 3
= 2 (− sin(2x − 7)) − 3
= 2 sin(2x − 7 + π) + (−3) Fact 18.2.5 on page 241

2π 7−π
= 2 sin x− + (−3).
π 2
This function is now in the standard form of Definition 19.1.1, so it is

a sinusoidal function with phase shift C = 7−π
2
= 1.93, mean D = −3,
amplitude A = 2 and period B = π.
(ii) Start with y = 3 sin(− 21 x+1) +4, then here are the steps with reference
to the required identities to put the equation in standard form:

1
y = 3 sin − x + 1 + 4
2

1
= 3 sin − x−1 +4
2

1
= 3 − sin x−1 +4 Fact 18.2.4 on page 241
2

1
= 3 sin x−1+π +4 Fact 18.2.5 on page 241
2

2π
= 3 sin (x − [2 − 2π]) + 4
4π
19.2. EXAMPLES OF SINUSOIDAL BEHAVIOR 259
This function is now in the standard form of Definition 19.1.1, so it

is a sinusoidal function with phase shift C = 2 − 2π, mean D = 4,
amplitude A = 3 and period B = 4π.
(iii) Start with y = 2 cos(3x+1) −2, then here are the steps to put the equa-
tion in standard form. A key simplifying step is to use the identity:
cos(t) = sin( π2 + t).
y = 2 cos(3x + 1) − 2
π
= 2 sin + 3x + 1 − 2
2 h π i
= 2 sin 3x − −1 − + (−2)
2
!
2π 1h πi
= 2 sin 2π
x− −1 − + (−2)
3
3 2
This function is now in the standard form of Definition 19.1.1, so it is

a sinusoidal function with phase shift C = 31 [−1 − π2 ], mean D = −2,
amplitude A = 2 and period B = 2π3
.
19.2 Examples of sinusoidal behavior

Problems involving sinusoidal behavior come in two basic flavors. On the
one hand, we could be handed an explicit sinusoidal function

2π
y = A sin (x − C) + D
B
and asked various questions. The answers typically require either direct
calculation or interpretation of the constants. Example 19.1.3 is typical
of this kind of problem. On the other hand, we might be told a particular
situation is described by a sinusoidal function and provided some data
or a graph. In order to further analyze the problem, we need a “formula”,
which means finding the constants A, B, C, and D. This is a typical
scenario in a “mathematical modeling problem”: the process of observing
data, THEN obtaining a mathematical formula. To find A, take half the
difference between the largest and smallest values of f(x). The period B
is most easily found by measuring the distance between two successive
maxima (peaks) or minima (valleys) in the graph. The mean D is the
average of the largest and smallest values of f(x). The phase shift C
(which is usually the most tricky quantity to get your hands on) is found
by locating a “reference point”. This “reference point” is a location where
the graph crosses the mean line y = D on its way up from a minimum
to a maximum. The funny thing is that the phase shift C is NOT unique;
there are an infinite number of correct choices. One choice that will work
is C = (x-coordinate of a maximum) − B4 . Any other choice of C will differ

from this one by a multiple of the period B.
max value − min value

A =
2
B = distance between two successive peaks (or valleys)
B
C = x-coordinate of a maximum −
4
max value + min value
D = .
2
Example 19.2.1. Assume that the number of hours of daylight in Seattle
is given by a sinusoidal function d(t) of time. During 1994, assume the
longest day of the year is June 21 with 15.7 hours of daylight and the
shortest day is December 21 with 8.3 hours of daylight. Find a formula
d(t) for the number of hours of daylight on the tth day of the year.
Solution. Because the function d(t) is assumed to be si-

d-axis (hours of daylight)
15 2π

12.5
nusoidal, it has the form y = A sin B (t − C) + D, for con-
10 stants A, B, C, and D. We simply need to use the given
7.5 information to find these constants. The largest value of
5
the function is 15.7 and the smallest value is 8.3. Know-
50 100 150 200 250 300 350
t-axis (days) ing this, from the above discussion we can read off :
Figure 19.12: Hours of day-
light in Seattle in 1994.
15.7 + 8.3 15.7 − 8.3
D= = 12 A= = 3.7.
2 2
To find the period, we need to compute the time between two successive
maximum values of d(t). To find this, we can simply double the time
length of one-half period, which would be the length of time between
successive maximum and minimum values of d(t). This gives us the
equation
B = 2(days between June 21 and December 21) = 2(183) = 366.
Locating the final constant C requires the most thought. Recall, the
longest day of the year is June 21, which is day 172 of the year, so
B 366
C = (day with max daylight) − = 172 − = 80.5.
4 4
In summary, this shows that

2π
d(t) = 3.7 sin (t − 80.5) + 12.
366
A rough sketch, following the procedure outlined above, gives this graph
on the domain 0 ≤ t ≤ 366; we have included the mean line y = 12 for
reference.
19.2. EXAMPLES OF SINUSOIDAL BEHAVIOR 261
We close with the example that started this section.

Example 19.2.2. The depth of a migrating salmon below the water sur-
face changes according to a sinusoidal function of time. The fish varies
between 1 and 5 feet below the surface of the water. It takes the fish
1.571 minutes to move from its minimum depth to its successive maximum
depth. It is located at a maximum depth when t = 4.285 minutes. What
is the formula for the function d(t) that predicts the depth of the fish after
t minutes? What was the depth of the salmon when it was first spotted?
During the first 10 minutes, how many times will the salmon be exactly 4
feet below the surface of the water?
Solution. We know that d(t) = A sin( 2π

B
(x − C)) + D, for ap- 5
propriate constants A, B, C, and D. We need to use the 4
depth
given information to extract these four constants. The 3
2
amplitude and mean are easily found using the above for-
1
mulas:
2 4 6 8 10
max depth − min depth 5−1 time
A= = =2 Figure 19.13: Depth of a mi-
2 2
max depth + min depth 5+1 grating salmon.
D= = = 3.
2 2
The period can be found by noting that the information about the time
between a successive minimum and maximum depth will be half of a
period (look at the picture in Figure 19.13):
B = 2(1.571) = 3.142
Finally, to find C we
B 3.142
C = (time of maximum depth) − = 4.285 − = 3.50.
4 4
The formula is now

2π
d(t) = 2 sin (t − 3.5) + 3 = 2 sin(2t − 7) + 3
3.142
The depth of the salmon when it was first spotted is just
d(0) = 2 sin(−7) + 3 = 1.686 feet.
Finally, graphically, the last question amounts to determining how many

times the graph of d(t) crosses the line y = 4 on the domain [0,10].
This can be done using Figure 19.13. A simultaneous picture of the
two graphs is given, from which we can see the salmon is exactly 4 feet
below the surface of the water six times during the first 10 minutes.
19.3 Summary
• A sinusoidal function is one of the form

2π
f(t) = A sin (t − C) + D
B
where A, B, C, and D are constants.
– A is the amplitude of the function; this is half the vertical dis-

tance between a high point and a low point on its graph.
– B is the period of the function; this is the horizontal distance be-
tween two consecutive high points (or low points) on its graph.
– C is the phase shift of the function; it is multi-valued, but one
choice for C is a value of t at which the function is increasing
and equal to D.
– D is the mean value of the function; it is the y-value of the hor-
izontal line about which the graph of the function is balanced.
• The graph of a sinusoidal function is a shifted, scaled version of the

graph of y = sin t.
19.4. EXERCISES 263
19.4 Exercises
Problem 19.1. Find the amplitude, period, a machine begins recording a plot of volume per
phase shift and the mean of the following si- breath versus time (in seconds). Let b(t) be a
nusoidal functions. function of time t that tells us the volume (in
liters) of a breath that starts at time t. Dur-
(a) y = sin(2x − π) + 1 ing the test, the smallest volume per breath
(b) y = 6 sin(πx) − 1 is 0.6 liters and this first occurs for a breath
that starts 5 seconds into the test. The largest
(c) y = 3 sin(x + 2.7) + 5.2 volume per breath is 1.8 liters and this first
(d) y = 5.6 sin 32 x − 7 − 12.1
occurs for a breath beginning 55 seconds into
the test.
(e) y = 2.1 sin πx + 44.3 − 9.8

(a) Find a formula for the function b(t)
(f) y = 3.9 (sin(22.34(x + 18)) − 11) whose graph will model the test data for
this patient.
(g) y = 11.2 sin π5 (x − 9.2) + 8.3

(b) If the patient begins a breath every 5

seconds, what are the breath volumes
Problem 19.2. A weight is attached to a spring during the first minute of the test?
suspended from a beam. At time t = 0, it
is pulled down to a point 10 cm above the
ground and released. After that, it bounces up Problem 19.4. Suppose the high tide in Seat-
and down between its minimum height of 10 tle occurs at 1:00 a.m. and 1:00 p.m. at which
cm and a maximum height of 26 cm, and its time the water is 10 feet above the height of
height h(t) is a sinusoidal function of time t. low tide. Low tides occur 6 hours after high
It first reaches a maximum height 0.6 seconds tides. Suppose there are two high tides and
after starting. two low tides every day and the height of the
tide varies sinusoidally.
(a) Follow the procedure outlined in this
section to sketch a rough graph of h(t). (a) Find a formula for the function y = h(t)
Draw at least two complete cycles of the that computes the height of the tide
oscillation, indicating where the maxima above low tide at time t. (In other words,
and minima occur. y = 0 corresponds to low tide.)
(b) What are the mean, amplitude, phase (b) What is the tide height at 11:00 a.m.?
shift and period for this function?
(c) Give four different possible values for Problem 19.5. Your seat on a Ferris Wheel is
the phase shift. at the indicated position at time t = 0.
(d) Write down a formula for the function

h(t) in standard sinusoidal form; i.e. as
in 19.1.1 on Page 254.
Start
(e) What is the height of the weight after 53 feet
0.18 seconds?
(f) During the first 10 seconds, how many
times will the weight be exactly 22 cm
above the floor? (Note: This problem
does not require inverse trigonometry.)
Let t be the number of seconds elapsed after
the wheel begins rotating counterclockwise.
Problem 19.3. A respiratory ailment called You find it takes 3 seconds to reach the top,
“Cheyne-Stokes Respiration” causes the vol- which is 53 feet above the ground. The wheel
ume per breath to increase and decrease in a is rotating 12 RPM and the diameter of the
sinusoidal manner, as a function of time. For wheel is 50 feet. Let d(t) be your height above
one particular patient with this condition, a the ground at time t.
(a) Argue that d(t) is a sinusoidal function, Problem 19.7. The voltage output(in volts) of
describing the amplitude, phase shift, an electrical circuit at time t seconds is given
period and mean. by the function
(b) When are the first and second times you
V(t) = 23 sin(5πt−3π)+1 .
are exactly 28 feet above the ground?
(c) After 29 seconds, how many times will (a) What is the initial voltage output of the
you have been exactly 28 feet above the circuit?
ground?
(b) Is the voltage output of the circuit ever
equal to zero? Explain.
Problem 19.6. In Exercise 17.12, we studied
the situation below: A bug has landed on the (c) The function V(t) = 2p(t) , where p(t) =
rim of a jelly jar and is moving around the rim. 3 sin(5πt − 3π) + 1. Put the sinusoidal
The location where the bug initially lands is function p(t) in standard form and
described and its angular speed is given. Im- sketch the graph for 0 ≤ t ≤ 1. Label the
pose a coordinate system with the origin at the coordinates of the extrema on the graph.
center of the circle of motion. In each of the
(d) Calculate the maximum and minimum
cases, the earlier exercise found the coordi-
voltage output of the circuit.
nates P(t) of the bug at time t. For each of the
scenarios below, answer these two questions: (e) During the first second, determine when
(a) Both coordinates of P(t) = (x(t),y(t)) are the voltage output of the circuit is 10
sinusoidal functions in the variable t. volts.
Sketch a rough graph of the functions (f) A picture of the graph of y = V(t) on the
x(t) and y(t) on the domain 0 ≤ t ≤ 9. domain 0 ≤ t ≤ 1 is given; label the coor-
(b) Use the graph sketches to help you find dinates of the extrema on the graph.
the the amplitude, mean, period and
phase shift for each function. Write x(t)
volts
and y(t) in standard sinusoidal form.
15
ω=4π/9rad/sec
12.5
bug lands here
10
7.5
1.2 rad
5
2.5
2 in
t axis
0.2 0.4 0.6 0.8 1
(g) Restrict the function V(t) to the domain

0.1 ≤ t ≤ 0.3; explain why this function
has an inverse and find the formula for
the inverse rule. Restrict the function
V(t) to the domain 0.3 ≤ t ≤ 0.5; ex-
bug lands here plain why this function has an inverse
and find the formula for the inverse rule.
2 in
ω=4π/9rad/sec
ω= 4π/9rad/sec
bug lands
here
0.5 rad
2 in
Chapter 20
Inverse Circular Functions
An aircraft is flying at an altitude 10 miles above the el-
10 miles
evation of an airport. If the airplane begins a steady de- θ
scent 100 miles from the airport, what is the angle θ of
100 miles
descent? Figure 20.1: An aircraft de-
The only natural circular function we can use is z = cending toward an airport.
tan(θ), leading to the equation:
10 1
tan(θ) = = .
100 10
The problem is that this equation does not tell us the value of θ. More-
over, none of the equation solving techniques at our disposal (which all
amount to algebraic manipulations) will help us solve the equation for θ.
What we need is an inverse function θ = f−1(z); then we could use the
fact that tan−1(tan(θ)) = θ and obtain:

−1 −1 1
θ = tan (tan(θ)) = tan .
10
Computationally, without even thinking about what is going on, any sci-
entific calculator will allow us to compute values of an inverse circular
function and leads to a solution of our problem. In this example, you will
−1 1 ◦
find θ = tan 10
= 5.71 . Punch this into your calculator and verify it!
20.1 Solving Three Equations

Example 20.1.1. Find all values of θ (an angle) that make this equation
true: sin(θ) = 21 .
Solution. We begin with a graphical reinterpretation: the solutions corre-

spond to the places where the graphs of z = sin(θ) and z = 21 intersect in
the θz-coordinate system. Recalling Figure 18.11, we can picture these
two graphs simultaneously as below:
265
266 CHAPTER 20. INVERSE CIRCULAR FUNCTIONS
cross cross z-axis cross cross

1
1
Graph of z = 2
A B
θ-axis
−2π −π π 2π 3π
−1
Graph of z = sin(θ) one period
1
Figure 20.2: Where does sin(θ) cross z = 2 ?
The first thing to notice is that these two graphs will cross an infi-
nite number of times, so there are infinitely many solutions to Exam-
ple 20.1.1! However, notice there is a predictable spacing of the crossing
points, which is just a manifestation of the periodicity of the sine func-
tion. In fact, if we can find the two crossing points labeled “A” and “B”,
then all other crossing points are obtained by adding multiples of 2π to
either “A” or “B”.By Table 17.1, θ = π6 radians is a special angle where we
computed sin π6 = 21 , which tells us that the crossing point labeled “A”
is the point ( π6 , 12 ). Using the identities in Facts 18.2.4 and 18.2.5, notice
that

5π π
sin = sin − + π
6 6
π
= − sin −
6
π
= − − sin
π 6
= sin
6
1
= .
2
So, θ = 5π6
is the only other angle θ between 0 and 2π such that Ex-
ample 20.1.1 holds. This corresponds to the crossing point labeled “B”,
which has coordinates 5π 1

,
6 2
. In view of the remarks above, the crossing
points come in two flavors:

π 1
+ 2kπ, , k = 0, ±1, ±2, ±3, · · · , and
6 2

5π 1
+ 2kπ, , k = 0, ±1, ±2, ±3, · · · .
6 2
Taking this example as a model, we can tackle the more general prob-
lem: For a fixed real number c, describe the solution(s) of the equation
20.1. SOLVING THREE EQUATIONS 267
c = f(θ) for each of the circular functions z = f(θ). Studying solutions

of these equations will force us to come to grips with three important
issues:
• For what values of c does f(θ) = c have a solution?
• For a given value of c, how many solutions does f(θ) = c have?
• Can we restrict the domain so that the resulting function is one-to-

one?
All of these questions must be answered before we can come to grips with
any understanding of the inverse functions. Using the graphs of the cir-
cular functions, it is an easy matter to arrive at the following qualitative
conclusions.
Important Fact 20.1.2. None of the circular functions is one-to-one on

the domain of all θ values. The equations c = sin(θ) and c = cos(θ) have
a solution if and only if −1 ≤ c ≤ 1; if c is in this range, there are infinitely
many solutions. The equation c = tan(θ) has a solution for any value of c
and there are infinitely many solutions.
Example 20.1.3.
√ If two sides of a right triangle have
lengths 1 and 3 as pictured below, what are the acute
angles α and β?
β
Solution. By the Pythagorean Theorem the remaining side √
3
has length
r √ 2 α
1+ 3 = 2.
√ 1
Since tan(α) = 3, we need to solve this equation √ for α. Figure 20.3: What are the
Graphically, we need to determine where z = 3 crosses values for α and β?
the graph of the tangent function:
From Fact 20.1.2, there will be infinitely many solutions to our equa-
tion, but notice that there is exactly one solution in the interval − π2 , π2
and we can find it using Table 17.1:
√
3
π
◦ sin(60◦) 2 √
tan = tan(60 ) = = 1
= 3.
3 cos(60◦) 2
So, α = π3 radians = 60◦ is the only acute angle solution and β = 180◦ −
60◦ − 90◦ = 30◦ .
θ = −π z-axis one period

2
5π
θ= 2
only place graphs

cross √ in this period
is ( π
3
, 3)
etc. etc.
θ-axis
−2π −π π 2π 3π
3π
θ = − 3π
2 π θ= 2
θ= 2
√
Figure 20.4: Where does the line z = 3 cross z = tan θ?
20.2 Inverse Circular Functions

Except for specially chosen angles, we have not addressed the serious
problem of FINDING values of the inverse rules attached to the circular
function equations. (Our previous examples were “rigged”, so that we
could use Table 17.1.) To proceed computationally, we need to obtain
the inverse circular functions. If we were to proceed in a sloppy manner,
then a first attempt at defining the inverse circular functions would be to
write
✘ sin−1(z) = solutions θ of the equation z = sin(θ).

✘ cos−1(z) = solutions θ of the equation z = cos(θ). (20.1)
✘ tan−1(z) = solutions θ of the equation z = tan(θ).
There are two main problems with these rules as they stand. First, to
have a solution θ in the case of sin−1(z) and cos−1(z), we need to restrict z
so that −1 ≤ z ≤ 1. Secondly, having made this restriction on z in the first
two cases, there is no unique solution; rather, there are an infinite num-
ber of solutions. This means that the rules sin−1, cos−1, and tan−1 as they
now stand do not define functions. Given what we have reviewed about
inverse functions, the only way to proceed is to restrict each circular
function to a domain of θ values on which it becomes one-to-one, then
we can appeal to Fact 9.3.1 and conclude the inverse function makes
sense.
20.2. INVERSE CIRCULAR FUNCTIONS 269
At this stage a lot of choice (flexibility) enters into determining the

domain on which we should try to invert each circular function. In effect,
there are an infinite number of possible choices. If z = f(θ) denotes one
of the three circular functions, there are three natural criteria we use
to guide the choice of a restricted domain, which we will call a principal
domain :
• The domain of f(θ) should include the angles between 0 and π2 , since
these are the possible acute angles in a right triangle.
• On the restricted domain, the function f(θ) should take on all pos-
sible values in the range of f(θ). In addition, the function should be
one-to-one on this restricted domain.
• The function f(θ) should be “continuous” on this restricted domain;
i.e. the graph on this domain could be traced with a pencil, without
lifting it off the paper.
In the case of z = sin(θ), the principal domain − π2 ≤ θ ≤ π2 satisfies our
criteria and the picture is given below. Notice, we would not want to take
the interval 0 ≤ θ ≤ π, since z = sin(θ) doesn’t achieve negative values on
this domain; in addition, it’s not one-to-one there.
z-axis y-axis principal domain
1
on the unit circle
θ
θ-axis
x-axis
−π −π
2
π
2
π
sine restricted to
−1 principal domain
Figure 20.5: Principal domain for sin(θ).
In the case of z = cos(θ), the principal domain 0 ≤ θ ≤ π satisfies our

criteria and the picture is given below. Notice, we would not want to take
the interval − π2 ≤ θ ≤ π2 , since z = cos(θ) doesn’t achieve negative values
on this domain; in addition, it’s not one-to-one there.
In the case of z = tan(θ), the principal domain − π2 < θ < π2 satisfies
our criteria and the picture is given below. Notice, we would not want to
take the interval 0 ≤ θ ≤ π, since z = tan(θ) does not have a continuous
graph on this interval; in other words, we do not include the endpoints
since tan(θ) is undefined there.
Important Facts 20.2.1 (Inverse circular functions). Restricting each cir-
cular function to its principal domain, its inverse rule f−1(z) = θ will define
a function.
z-axis y-axis principal domain

1
on the unit circle
θ
θ-axis
3π x-axis
−π −π
2
π
2
π 2
cosine restricted to
−1 principal domain
Figure 20.6: Principal domain for cos(θ).
(i) If −1 ≤ z ≤ 1, then sin−1(z) is the unique angle θ in the principal

domain − π2 ≤ θ ≤ π2 with the property that sin(θ) = z.
(ii) If −1 ≤ z ≤ 1, then cos−1(z) is the unique angle θ in the principal

domain 0 ≤ θ ≤ π with the property that cos(θ) = z.
(iii) For any real number z, tan−1(z) is the unique angle θ in the principal
domain − π2 < θ < π2 with the property that tan(θ) = z.
We refer to the functions defined above as the inverse circular func-

tions. These are sometimes referred to as the “arcsine”, “arccosine” and
“arctangent” functions, though we will not use that terminology. The in-
verse circular functions give us one solution for each of these equations:
c = cos(θ)
c = sin(θ)
c = tan(θ);
these are called the principal solutions. We also can refer to these as the
principal values of the inverse circular function rules θ = f−1(z).
As usual, be careful with “radian mode” and “degree mode” when
!!! making calculations. For example, if your calculator is in “degree”
CAUTION mode and you type in “tan−1(18)”, the answer given is “86.82”. This
!!! means that an angle of measure θ = 86.82◦ has tan(86.82) = 18. If your
calculator is in “radian” mode and you type in “sin−1(0.9)”, the answer
given is “1.12”. This means that an angle of measure θ = 1.12 radians
has sin(1.12) = 0.9.
There is a key property of the inverse circular functions which is use-
ful in equation solving; it is just a direct translation of Fact 9.3.2 into our
current context:
Important Facts 20.2.2 (Composition identities). We have the following

equalities involving compositions of circular functions and their inverses:
z-axis
y-axis principal domain
on the unit circle
1 θ
θ-axis x-axis
π
−π −π
2 2
π
−1
tangent function
restricted to prin-
cipal domain
Figure 20.7: Principal domain for tan(θ).
(a) If − π2 ≤ θ ≤ π2 , then sin−1(sin(θ)) = θ.

(b) If 0 ≤ θ ≤ π, then cos−1(cos(θ)) = θ.
(c) If − π2 < θ < π2 , then tan−1(tan(θ)) = θ.
We have been very explicit about the allowed θ values for the equations
in Fact 20.2.2. This is important and an Exercise will touch on this issue.
20.3 Applications
As a simple application of Fact 20.2.2, we can return to the beginning of
this section and justify the reasoning used to find the angle of descent of
the aircraft:

−1 −1 1
θ = tan ((tan(θ)) = tan = 0.09967 rad = 5.71◦.
10
Let’s look at some other applications.
Example 20.3.1. Find two acute angles θ so that the following equation
is satisfied:
9 252 2

= 1 − cos (θ) .
4 cos2(θ) 16
Solution. Begin by multiplying each side of the equation by cos2(θ) and
rearranging terms:
9 252
1 − cos2(θ) cos2(θ)

=
4 16
252 252 9
0 = cos4(θ) − cos2(θ) + .
16 16 4
To solve this equation for θ, we use what is called the technique of substi-
tution. The central idea is to bring the quadratic formula into the picture
by making the substitution z = cos2(θ):
252 2 252 9
0=
z − z+ .
16 16 4
Applying the quadratic formula, we obtain
r
2
252 252 252 9
16
± 16
− 4 16 4
z= 2(252 )
= 0.9386 or 0.06136.
16
We now use the fact that z = cos2(θ) and note the cosine of an acute angle
is non-negative to conclude that
cos2(θ) = 0.9386 or cos2(θ) = 0.06136
cos(θ) = 0.9688 or cos(θ) = 0.2477.
Finally, we use the inverse cosine function to arrive at our two acute
angle solutions:
cos(θ) = 0.9688 ⇒ θ = cos−1(0.9688) = 14.35◦
cos(θ) = 0.2477 ⇒ θ = cos−1(0.2477) = 75.66◦
11
00 Example 20.3.2. A 32 ft ladder leans against a building
00
11
00000000000000000000000000
11111111111111111111111111
11111111111111111111111111
00000000000000000000000000
00
11
00000000000000000000000000
11111111111111111111111111
00000000000000000000000000
11111111111111111111111111
00000000000000000000000000
11111111111111111111111111
00
11
00000000000000000000000000
11111111111111111111111111
00000000000000000000000000
11111111111111111111111111
00000000000000000000000000
11111111111111111111111111
00000000000000000000000000
11111111111111111111111111
(as shown below) making an angle α with the wall. OSHA
00000000000000000000000000
11111111111111111111111111
00000000000000000000000000
11111111111111111111111111
00000000000000000000000000
11111111111111111111111111
00000000000000000000000000
11111111111111111111111111 (Occupational Safety and Health Administration) specifies
a “safety range” for the angle α to be 15◦ ≤ α ≤ 44◦ . If the
α
h 32 base of the ladder is d = 10 feet from the house, is this a
safe placement? Find the highest and lowest points safely
d accessible.
Figure 20.8: A ladder prob- 10
Solution. If d = 10, then
sin(α) = 32 , so the principal so-
lem. lution is α = sin −1 10
= 18.21 ◦
; this lies within the safety
32
zone. From the picture, it is clear that the highest point safely reached
will occur precisely when α = 15◦ and as this angle increases, the height
h decreases until we reach the lowest safe height when α = 44◦ . We need
to solve two right triangles. If α = 15◦ , then h = 32 cos(15◦ ) = 30.9 ft. If
α = 44◦ , then h = 32 cos(44◦) = 23.02 ft.
Example 20.3.3. A Coast Guard jet pilot makes contact with a small
unidentified propeller plane 15 miles away at the same altitude in a di-
rection 0.5 radians counterclockwise from East. The prop plane flies in
the direction 1.0 radians counterclockwise from East. The jet has been in-
structed to allow the prop plane to fly 10 miles before intercepting. In what
direction should the jet fly to intercept the prop plane? If the prop plane is
flying 200 mph, how fast should the jet be flying to intercept?
North Intercept point
Solution. A picture of the situation is shown in Fig-

find speed to
ure 20.9(a). After a look at the picture, three right trian- intercept
10 miles
1 rad
East
gles pop out and beg to be exploited. We highlight these in prop plane
spotted here
the Figure 20.9(b), by imposing a coordinate system and Jet 15 miles
labeling the various sides of our triangles. We will work West 0.5 rad East
θ = intercept heading
South
in radian units and label θ to be the required intercept
heading. (a) The physical layout.
We will first determine the sides x + y and u + w of the
Intercept point
large right triangle. To do this, we have y-axis
1 rad
u
10
d
x = 15 cos(0.5) = 13.164 miles, u+w
prop plane
w spotted
y = 10 cos(1.0) = 5.403 miles,
15
0.5 rad θ
here
x y x-axis
w = 15 sin(0.5) = 7.191 miles, and Jet
x+y
u = 10 sin(1.0) = 8.415 miles.
(b) Modeling the problem.
w+u
We now have tan(θ) = x+y
= 0.8405, so the principal solu-
Figure 20.9: Visualizing the
tion is θ = 0.699 radians, which is about 40.05◦. This is the Coast Guard problem.
only acute angle solution, so we have found the required
intercept heading.
To find the intercept speed, first compute your
distance to the intercept point, which is the length
of the hypotenuse of the big right triangle: d =
p
(18.567)2 + (15.606)2 = 24.254 miles. You need to travel
this distance in the same amount of time T it takes the
10
prop plane to travel 10 miles at 200 mph; i.e. T = 200 = 0.05 hours. Thus,
the intercept speed s is
distance traveled 24.254

s= = = 485 mph.
time T elapsed 0.05
Later you will use an alternative approach to this problem using ve-
locity vectors.
In certain applications, knowledge of the principal solutions for the
equations c = cos(θ), c = sin(θ), and c = tan(θ) is not sufficient. Here is
a typical example of this, illustrating the reasoning required.
Example 20.3.4. A rigid 14 ft pole is used to vault. The vaulter leaves

and returns to the ground when the tip is 6 feet high, as indicated. What
are the angles of the pole with the ground on takeoff and landing?
14 ft β
α 6 ft 6 ft
leaving ground airborne returning to ground
Figure 20.10: Various angles of a vaulter’s pole.
y-axis
1 Solution. From the obvious right triangles in the picture,
6
y= 3 we are interested in finding angles θ where sin(θ) = 14 = 73 .
Q 7 P
β The idea is to proceed in three steps:
α
1 x-axis • Find the principal solution of the equation sin(θ) = 73 ;
• Find all solutions of the equation sin(θ) = 37 ;
Figure 20.11: Modeling the

• Use the constraints of the problem to find α and β
problem with a unit circle. among the set of all solutions.
Solving the equation sin(θ) = 73 involves finding the points
on the unit circle with y-coordinate equal to 37 . From the
picture, we see there are two such points, labeled P and Q.
The coordinates of these points will be P = cos(α), 73 and Q = cos(β), 73 .

Notice, α is the principal solution of our equation sin(θ) = 37 , since

0 ≤ α ≤ 90◦ ; so α = sin−1 37 = 25.38◦ . In general, the solutions come

in two basic flavors:

θ = α + 2k(180◦ )
= 25.38◦ + 2k(180◦ ),
or
θ = β + 2k(180◦),
where k = 0, ± 1, ± 2, ± 3, . . . . To find the angle β, we can use basic
properties of the circular functions:
sin(β) = sin(α) = − sin(−α) = sin(180◦ − α) = sin(154.62◦).
This tells us β = 154.62◦.
20.4 How to solve trigonometric equations

So far, our serious use of the inverse trigonometric functions has focused
on situations that ultimately involve triangles. However, many trigono-
metric modeling problems have nothing to do with triangles and so we
20.4. HOW TO SOLVE TRIGONOMETRIC EQUATIONS 275
need to free ourselves from the necessity of relying on such a geomet-

ric picture. There are two general strategies for finding solutions to the
equations c = sin(θ), c = cos(θ), and c = tan(θ):
• The first strategy is summarized in Procedure 20.4.1. This method

has the advantage of offering a “prescription” for solving the equa-
tions; the disadvantage is you can lose intuition toward interpreting
your answers.
• The second strategy is graphical in nature and is illustrated in Ex-

ample 20.4.2 below. This method usually clarifies interpretation of
the answers, but it does require more work since an essential step is
to roughly sketch the graph of the trigonometric function (following
the procedure of Chapter 19 or using a graphing device).
Each approach has its merits as you will see in the exercises.
Important Procedure 20.4.1. To find ALL solutions to the equations c =

sin(θ), c = cos(θ), and c = tan(θ), we can lay out a foolproof strategy.
Step Sine case Cosine case Tangent case
1. Find principal solu-

θ = sin−1 (c) θ = cos−1 (c) θ = tan−1 (c)
tion
2. Find symmetry solu-

θ = −sin−1 (c) + π θ = −cos−1 (c) not applicable
tion
3. Write out multiples of

2kπ 2kπ kπ
period k = 0,± 1,± 2,· · ·
4. Obtain general princi-

θ = sin−1 (c) + 2kπ θ = cos−1 (c) + 2kπ θ = tan−1 (c) + kπ
pal solutions
5. Obtain general sym-

θ = −sin−1 (c) + π + 2kπ θ = −cos−1 (c) + 2kπ not applicable
metry solutions
Example 20.4.2. Assume that the number of hours of daylight in your

2π

hometown during 1994 is given by the function d(t) = 3.7 sin 366 (t − 80.5) +
12, where t represents the day of the year. Find the days of the year during
which there will be approximately 14 hours of daylight?1
Solution. To begin, we want to roughly sketch the graph of y = d(t) on

the domain 0 ≤ t ≤ 366. If you apply the graphing procedure discussed
in Chapter 19, you obtain the sinusoidal graph below on the larger do-
main −366 ≤ t ≤ 732. (The reason we use a larger domain is so that
the “pattern” that will arise in the strategy described below is more ev-
ident. Ultimately, we will restrict our attention to the smaller domain
0 ≤ t ≤ 366.) To determine when there will be 14 hours of daylight, we
need to solve the equation 14 = d(t). Graphically, this amounts to finding
y-axis
15
12.5
10
7.5
t-axis
−200 200 400 600
Figure 20.12: Where does the sinusoidal function d(t) cross the line y = 14.
the places where the line y = 14 intersects the graph of d(t). As can be
seen in Figure 20.12, there are several such intersection points.
We now outline a “three step strategy” to find all of these intersection
points (which amounts to solving the equation 14 = d(t)):
1. Principal Solution. We will find one solution by using the inverse

sine function. If we start with the function y = sin(t) on its principal
domain −π 2
≤ t ≤ π2 , then we can compute the domain of d(t) =
2π
3.7 sin 366 (t − 80.5) + 12:
π 2π π
− ≤ (t − 80.5) ≤
2 366 2
−11 ≤ t ≤ 172.
Now, using the inverse sine function we can find the principal solu-
tion to the equation 14 = d(t):

2π
14 = 3.7 sin (t − 80.5) + 12 (20.2)
366

2π
0.54054 = sin (t − 80.5)
366
π
0.57108 = sin−1(0.54054) = (t − 80.5)
183
t = 113.8
Notice, this answer is in the domain −11 ≤ t ≤ 172. In effect, we

have found THE ONLY SOLUTION on this domain. Conclude that
there will be about 14 hours of daylight on the 114th day of the year.
1
You can get the actual data from the naval observatory at this world wide web
address: <http://tycho.usno.navy.mil/time.html>
20.4. HOW TO SOLVE TRIGONOMETRIC EQUATIONS 277
2. Symmetry Solution. To find another solution to the equation 14 =

d(t), we will use symmetry properties of the graph of y = d(t). This
is where having the graph of y = d(t) is most useful. We know the
a maxima on the graph occurs at the point M = (172, 15.7); review
Example 19.2.1 for a discussion of why this is the case. From the
graph, we can see there are two symmetrically located intersection
points on either side of M. The principal solution gives the intersec-
tion point (113.8, 14). This point is 58.2 horizontal units to the left of
M; see the picture below. So a symmetrically positioned intersection
point will be (172 + 58.2, 14) = (230.2, 14).
y-axis M
15
12.5
10
7.5
t-axis
−200 200 400 600
Figure 20.13: Finding the symetry solution.
In other words, t = 230.2 is a second solution to the equation 14 =

d(t). We call this the symmetry solution.
3. Other Solutions. To find all other solutions of 14 = d(t), we add

integer multiples of the period B = 366 to the t-coordinates of the
principal and symmetric intersection points. On the domain −366 ≤
t ≤ 732 we get these six intersection points; refer to the picture of
the graph:
(−252.2, 14), (113.8, 14), (479.8, 14),

(−135.8, 14), (230.2, 14), (596.2, 14).
So, on the domain −366 ≤ t ≤ 732 we have these six solutions to the
equation 14 = d(t):
t = −252.2, −135.8, 113.8, 230.2, 479.8, 596.2.
To conclude the problem, we only are interested in solutions in the

domain 0 ≤ t ≤ 366, so the answers are t = 113.8, 230.2; i.e. on days 114
and 230 there will be about 14 hours of daylight.
20.5 Summary
• The inverse sine function, sin−1 x, is defined as the inverse of the sine
function, sin x, restricted to the domain −π/2 ≤ x ≤ π/2.
• Inverse cosine and inverse tangent are defined similarly.
• To find the solutions to an equation of the form

2π
f(t) = A sin (t − C) + D = k,
B
proceed as follows:
1. Sketch the graph of the function f(t), including a few periods

and t = C.
2. Use algebra and the inverse sine function (sin−1) to find one
solution. This is the principal solution; it is the solution nearest
to t = C. Call this solution P.
3. Use the principal solution and your knowledge of the graph of
the function to find the symmetry solution. Call this solution S.
4. All solutions are then of the form
P, P ± B, P ± 2B, P ± 3B, . . .
or
S, S ± B, S ± 2B, S ± 3B, . . . .
20.6. EXERCISES 279
20.6 Exercises
Problem 20.1. Let’s make sure we can handle Problem 20.4. Hugo bakes world famous
the symbolic and mechanical aspects of work- scones. The key to his success is a special
ing with the inverse trigonometric functions: oven whose temperature varies according to a
sinusoidal function; assume the temperature
(a) Set your calculator to “radian mode” and
(in degrees Fahrenheit) of the oven t minutes
compute to four decimal places:
after inserting the scones is given by
√
(a1) sin−1 (x), for x = 0,1,−1, 3
2 ,0.657,

π 3π

−3 y = s(t) = 15 sin t− + 415
11 ,2. 5 2
√
3
(a2) cos−1 (x), for x = 0,1,−1, 2 ,0.657, (a) Find the amplitude, phase shift, period
−3
11 ,2. and mean for s(t), then sketch the graph
√
(a3) tan−1 (x), for x = 0,1,−1, 3 on the domain 0 ≤ t ≤ 20 minutes.
2 ,0.657,
−3
11 ,2. (b) What is the maximum temperature of
the oven? Give all times when the
(b) Redo part (a) with your calculator set in oven achieves this maximum tempera-
“degree mode”. ture during the first 20 minutes.
(c) Find four values of x that satisfy the (c) What is the minimum temperature of
equation 5 sin(2x2 + x − 1) = 2. the oven? Give all times when the
(d) Find four solutions to the equation oven achieves this minimum tempera-
5 tan(2x2 + x − 1) = 2 ture during the first 20 minutes.
(d) During the first 20 minutes of baking,
Problem 20.2. For each part of the problem calculate the total amount of time the
below: oven temperature is at least 410o F.
• Sketch the graphs of f(x) and g(x) on the (e) During the first 20 minutes of baking,
same set of axes. calculate the total amount of time the
oven temperature is at most 425o F.
• Set f(x) = g(x) and find the principal and
symmetry solutions. (f) During the first 20 minutes of baking,
calculate the total amount of time the
• Indicate the other solutions of the equa- oven temperature is between 410o F and
tion f(x) = g(x) on your graph and de- 425o F.
scribe their relationship to the principal
and symmetry solutions.
Problem 20.5. The temperature in Gavin’s
(a) f(x) = sin x − π2 , g(x) = 31 .

oven is a sinusoidal function of time. Gavin
sets his oven so that it has a maximum tem-
(b) f(x) = sin x + π6 , g(x) = −1.

perature of 300◦ F and a minimum tempera-
(c) f(x) = sin (2x − 1), g(x) = 14 . ture of 240◦ . Once the temperature hits 300◦ ,
it takes 20 minutes before it is 300◦ again.
(d) f(x) = 10 cos (2x + 1) − 5, g(x) = −1.
Gavin’s cake needs to be in the oven for 30
minutes at temperatures at or above 280 ◦ . He
Problem 20.3. Assume that the number of puts the cake into the oven when it is at 270◦
hours of daylight in New Orleans in 1994 is and rising. How long will Gavin need to leave
given by the function D(x) = 37 sin 365
2π
x + 35 the cake in the oven?

3 ,
where x represents the number of days after
March 21. Problem 20.6. Maria started observing Elas-
ticman’s height at midnight. At 3 AM, he was
(a) Find the number of hours of daylight on at his shortest: only 5 feet tall. At 9 AM, he
January 1, May 18 and October 5. was at his tallest: 11 feet tall.
(b) On what days of the year will there be Elasticman’s height is a sinusoidal function of
approximately 10 hours of daylight? time.
In the 24 hours after Maria began observing (i) Find where Tiffany passes Michael the
Elasticman, how much of the time will Elastic- first time.
man be less than 6 feet tall?
(j) Find when Tiffany passes Michael the
second time.
Problem 20.7. Suppose
(k) Find where Tiffany passes Michael the
2π second time.
T (t) = 23 sin (t − 7) + 66
24
is the temperature (in degrees Fahrenheit) at Problem 20.9. A communications satellite or-
time t, where t is measured in hours after mid- bits the earth t miles above the surface. As-
night on Sunday. You paint the exterior door sume the radius of the earth is 3,960 miles.
to your house at 5 p.m. on Monday. The paint The satellite can only “see” a portion of the
information states that 48 hours of 75◦ F dry- earth’s surface, bounded by what is called a
ing time is required; i.e., you can only count horizon circle. This leads to a two-dimensional
time periods when the temperature is at least cross-sectional picture we can use to study the
75◦ F. When will the door be dry? size of the horizon slice:
Problem 20.8. Tiffany and Michael begin run- satellite

ning around a circular track of radius 100 Earth
yards. They start at the locations pictured.
Michael is running 0.025 rad/sec counter-
clockwise and Tiffany is running 0.03 rad/sec horizon circle
counterclockwise. Impose coordinates as pic-
tured. center of Earth
satellite
α
Earth
r= 100 yards 0.025 rad/sec α
Tiff starts here
Michael starts here t

CROSS−SECTION
0.03 rad/sec
(a) Find a formula for α in terms of t.
(b) If t = 30,000 miles, what is alpha? What
(a) Where is each runner located (in percentage of the circumference of the
xy-coordinates) after 8 seconds? earth is covered by the satellite? What
would be the minimum number of such
(b) How far has each runner traveled after satellites required to cover the circum-
8 seconds? ference?
(c) Find the angle swept out by Michael (c) If t = 1,000 miles, what is alpha? What
after t seconds. percentage of the circumference of the
(d) Find the angle swept out by Tiffany earth is covered by the satellite? What
after t seconds. would be the minimum number of such
satellites required to cover the circum-
(e) Find the xy-coordinates of Michael and ference?
Tiffany after t seconds.
(d) Suppose you wish to place a satellite
(f) Find the first time when Michael’s into orbit so that 20% of the circumfer-
x-coordinate is -50. ence is covered by the satellite. What is
(g) Find the first time when Tiffany’s the required distance t?
x-coordinate is -50.
(h) Find when Tiffany passes Michael the Problem 20.10. Answer the following ques-
first time. tions:
20.6. EXERCISES 281
(a) If y = sin(x) on the domain − π2 ≤ x ≤ π2 , cording to her design, if the atmospheric pres-
what is the domain D and range R of sure exerted on the rocket is less than 10
y = 2 sin(3x − 1) + 3? How many solutions pounds/sq.in., the laughing gas chamber in-
does the equation 4 = 2 sin(3x−1)+3 have side the rocket will explode. Tiff worked from
on the domain D and what are they? a formula p = (14.7)e−h/10 pounds/sq.in. for
the atmospheric pressure h miles above sea
(b) If y = sin(t) on the domain − π2 ≤ t ≤ π2 , level. Assume that the rocket is launched at
what is the domain D and range R of an angle of α above level ground at sea level
2π
y = 8 sin( 1.2 (t−0.3))+18. How many solu- with an initial speed of 1400 feet/sec. Also,
2π
tions does the equation 22 = 8 sin( 1.2 (t − assume the height (in feet) of the rocket at
0.3))+18 have on the domain D and what time t seconds is given by the equation y(t) =
are they? −16t2 + 1400 sin(α)t.
(c) If y = sin(t) on the domain − π2 ≤ t ≤ (a) At what altitude will the rocket explode?
π
2 , what is the domain D and range R
of y = 27 sin( 366 2π
(t − 80.5)) + 45. How (b) If the angle of launch is α = 12o , de-
many solutions does the equation 40 = termine the minimum atmospheric pres-
2π
27 sin( 366 (t − 80.5)) + 45 have on the do- sure exerted on the rocket during its
main D and what are they? flight. Will the rocket explode in mid-
air?
(d) If y = cos(x) on the domain 0 ≤ x ≤ π,
(c) If the angle of launch is α = 82o , de-
what is the domain D and range R of
termine the minimum atmospheric pres-
y = 4 cos(2x+1)−3? How many solutions
sure exerted on the rocket during its
does the equation −1 = 4 cos(2x + 1) − 3
flight. Will the rocket explode in mid-
have on the domain D and what are
air?
they?
(d) Find the largest launch angle α so that
(e) If y = tan(x) on the domain − π2 < x < π2 ,
the rocket will not explode.
what is the domain D and range R of y =
2 tan(−x + 5) + 13? How many solutions
does the equation 100 = 2 tan(−x + 5) + 13 Problem 20.12. Let’s make sure we can han-
have on the domain D and what are dle the symbolic and mechanical aspects of
they? working with the inverse trigonometric func-
tions:
Problem 20.11. Tiffany is a model rocket en- (a) Find four solutions of tan(2x2 +x −1) = 5.
thusiast. She has been working on a pres- (b) Solve for x: tan−1 (2x2 + x − 1) = 0.5
surized rocket filled with laughing gas. Ac-
Appendix
283
Appendix A
Useful Formulas
Abbreviations
inch = in liter = L
feet = ft Joule = J
yard = yd calorie = cal
mile = mi atmosphere = atm
millimeter = mm Coulomb = C
centimeter = cm radian = rad
meter = m
degree = deg
kilometer = km
= ◦
second = sec = s
minute = min miles per hour = mi/hr
hour = hr = mph
year = yr
feet per second = ft/sec
ounce = oz
= ft/s
pound = lb
gram = g meters per second = m/sec
kilogram = kg = m/s
quart = qt
revolutions
gallon = gal = rev/min
per minute
milliliter = ml = RPM
285
286 APPENDIX A. USEFUL FORMULAS
Conversion Factors
Length Energy
1 in = 2.54 cm 1 J = 1 kg m2/s2
1 ft = 0.3048 m
1 cal = 4.184 J
1 mi = 1.609344 km
Mass
Volume
1 gal = 3.7854 L 1 oz = 28.3495 g
1 qt = 0.946353 L 1 lb = 0.453592 kg
Formulas from Plane and Solid Geometry

Rectangle
r
• Perimeter = 2ℓ + 2w
• Area = ℓw
w
Rectanglular prism
• Surface Area = 2(ℓw + ℓh + wh)
ℓ
• Volume = ℓhw
Triangle
h
• Perimeter = a + b + c
w
1
• Area = 2
bh ℓ
a
h
c
Right circular cylinder
• Surface Area = 2πr2 + 2πrh
b
• Volume = πr2h
Circle
r
• Perimeter = 2πr h
2
• Area = πr
287
Sphere Right circular cone

• Surface Area = 4πr2 • Surface Area = πr2 + πrs
• Volume = 43 πr3 • Volume = 13 πr2h
Constants
Avogadro’s number = N = 6.022142 × 1023
speed of light = c = 2.99792 × 108 m/s2
density of water = 1 g/cm3
mass of earth = 5.9736 × 1024 kg
earth’s equatorial radius = 3,960 mi = 6.38 × 106 m
acceleration of gravity at earth’s surface = 32 ft/sec2 = 9.8 m/s2
Algebra
• axay = ax+y
• (ax)y = axy
• a0 = 1
a x ax
• = x
b b
ax
• = ax−y
ay
√
• n a = a1/n
• (a + b)(c + d) = ac + ad + bc + bd
• (a + b)2 = a2 + 2ab + b2
• (a + b)3 = a3 + 3a2b + 3ab2 + b3

√
2 −b ± b2 − 4ac
• Quadratic Formula: If ax + bx + c = 0, then x =
2a
288 APPENDIX A. USEFUL FORMULAS
2
b2

2 b
• Completing the Square: ax + bx + c = a x + − +c
2a 4a
Trigonometry
• sin(−x) = − sin x • cos(π − x) = − cos x
• cos(−x) = cos x • sin(π + x) = − sin x
π
• sin − x = cos x • cos(π + x) = − cos x
2
π • sin(x + y) = sin x cos y + cos x sin y
• cos − x = sin x
2
π • cos(x + y) = cos x cos y − sin x sin y
• sin + x = cos x
2 • sin2 x + cos2 x = 1
π
• cos + x = − sin x • sin 2x = 2 sin x cos x
2
• sin(π − x) = sin x • cos 2x = cos2 x − sin2 x
Appendix B
Answers
Answer 1.1 (b) 150 ft/sec. (c) Gina. (d) 6300 hours. Answer 1.14 (a) Initial time = 7am, Initial temperature =
44◦ F, Final time = 10am, Final temperature = 50◦ F, rate of
Answer 1.2 68.4444 km. change = 2◦ F/hr. (b) 58◦ F. (c) Initial time = 4.5pm, Initial
temperature = 54◦ F, Final time = 6.25pm, Final tempera-
ture = 26◦ F, rate of change = -16◦ F/hr.
Answer 1.3 r = 10.172 cm for lead; r = 16.433 cm for alu-
minum.
Answer 1.15 (a) t = 9. (b) a = 1/8. (c) x = −3a/4. (d)
x+2
t > 3. (e) x(x+1) .
Answer 1.4 (a) 5.5 min/mi = 5:30 pace. (b) 14 32 ft/sec.
(c) Adrienne.
√ √
Answer 2.1 (a) d = 2, ∆x = √ 1 = ∆y = 1. (b) d = 5,
Answer 1.5 (a) (John’s Salary) = $56000 and (taxes) = $0. √ −1, ∆y = −2. (c) d = 34, ∆x = 5, ∆y = −3. (d)
∆x =
(b) (John’s Salary) ≤$56000 and (taxes) = 0.15× (John’s d = 10t2 + 2t + 1, ∆x = 3t, ∆y = 1 + t.
Salary). (c) (John’s Salary) ≥ $56000 and (John’s taxes)
> 0.28× (John’s Salary). (d) 1500 ≤(number of 120 stu- Answer 2.2 (a-c) True. (d) ∆x = s − a,∆y = t − b. (e)
dents each year) ≤ 1800. (e) (cost red Porsche) > 3× (cost ∆x = a − s,∆y = b − t. (f) ∆x = 0 means the points line
F-150 pickup). (f) 2 hours ≤ (weekly study time per credit on the same vertical line; ∆y = 0 means the points line on
hour) ≤ 3 hours (g) 2× (number of happy math students) the same horizontal line.
> 5× (number of happy chemistry students). (happy math
students) + (happy chemistry students) < 12 × (number of
cheerful biology students). (h) (Cady’s high score) - (Cady’s Answer 2.3 Just after 12:29 PM that afternoon.
low score) = 10%. (Cady’s final exam score) = 97%. (i)
(0.9999)× 21 (total votes cast)≤ (Tush votes) ≤ (1.0001)× 21 Answer 2.4 (a) Erik= 6.818 mph, Ferry = 17.6 ft/sec. (b)
(total votes cast). Impose coordinates with Kingston the origin and units of
miles on each axis; then Edmonds is located at (6,0) and
Answer 1.6 Go for the 15 inch pie. Erik’s sailboat is at (3,2). The table rows have these en-
tries:
(0,0), (0.1,0), (1.4,0), (12t,0).
Answer 1.7 (a) m90 = 151kg,m99 = 468kg,m99.9 =
m (3,2), (3,1.9432), (3,1.2045), (3,2 − 6.818t)
1476kg. (b) v = 2.974 × 108 sec . p
3.606, 3.491, 2.003, (12t − 3)2 + (2 − 6.818t)2 (c) Use
coordinates as in (b), then when the ferry reaches (3,0),
Answer 1.8 1080 pizzas sold in 4 hours. Profit reaches Erik is at (3,0.296). (d) CG vessel does not catch the ferry
$1000 at about 8 : 22 pm. before Edmonds.
Answer 1.9 About 5 × 106 times around the equator.

Answer 2.5 (a) d(t) = (65.3)t (b) 227 minutes, 168.4
miles (c) t = 80.86 seconds.
Answer 1.10 (a) Radii are about r = 0.2416, 2.317, 4.184
and 9.167 cm at the indicated times. (b) No.
Answer 2.6 (a) Allyson’s coordinate position: (0ft,20ft).
Adrienne’s coordinate position: (−16ft,0ft). (b) After 2
Answer 1.11 (a) 140 million gallons per week; about 7300 seconds there will still be slack in the bungee cord. (c)
million gallons per year. (b) 100 yds × 50 yds ×20 yds. t ≈ 2.34. Use this time to find where Allyson and Adrienne
are located. (d) t = 5.5 seconds.
Answer 1.12 (a) N = 12 + 0.45(60 − x). (b) Reduced com-
petition for resources (eg. water, nutrients, etc.). Answer 2.7 Impose a coordinate system with origin at
A and the shore the horizontal x axis; let x be the lo-
Answer 1.13 Formula simplifies to r(1+x)
1+2x
. Rates are cation
√ on the shore where Brooke beaches. Equation
1
about 0.67r, 0.6r, 0.54r and 0.51r at the indicated times. 2
25 + x2 + 14 (6 − x) gives time to reach Kono’s. T(0) = 4
Rates decrease over time, but will never be less than 0.5r. hr, T(6) = 3.9 hr. Neither time will be the minimum time.
289
290 APPENDIX B. ANSWERS
Answer 2.8 (b) t = 2; t = 52 . (c) (3,3) (d) spider=( 37 , 38 ), Answer 3.8 (a) 6.92 seconds. (b) 7.67 seconds. (c) 38.16
ant=( 41 , 4 ). (e) 1.5 feet. (f) Spider reaches (9,6) when seconds. (d) 7332 ft2 < area < 7632 ft2 .
3 3
t
√ = 4; ant reaches (9,6)
√ when t = 3. (g) spider speed is
5 ft/sec; ant speed is 8 ft/sec. Answer 3.9 (a) x = 11/5.
p (b) no solutions. (c) (−2, − 2)
and (−2,4). (d) (−1 ± 2/5,3).
Answer 2.9 49.92 mph (exactly).
Answer 4.1 (a) y = − 35 (x − 1) − 1 (b) y = 40(x + 1) − 2
Answer 2.10 141.46 miles. They are 300 miles apart at (c) y = −2x − 2 (d) y = 11 (e) m = 53 , y = 53 (x − 1) + 1 (f)
time 0.826 hr = 49.6 minutes. y = 40x − 14 (g) y = − 43 x + 47 (h) y = x − 1
2
Answer 2.11 (a) Final answer is correct, but second Answer 4.2 (a)( 1+2α , 1+4α
1+2α
). (b) α = −2/5. (c) α = −1/4.
equality is wrong. (b) Final answer should be 4xy; key
fact is that (x + y)2 = x2 + 2xy + y2 , etc. (c) Answer and Answer 4.3 (a) area = 25 b
(b) area = − 2m
2
(c) m = − 14
13
steps correct.
q Answer 4.4 In some cases, answers in this problem are

Answer 2.12 (a) x = ± 1+β 2
(b) x = αβ
(c) x = β unique. In other cases, the answers are not unique. Here
α2 α+β αβ+1 is a possible solution set:
2
Answer 2.13 (a) 5t2 + 6t + 5 (b) 2t2 + 4t (c) t 2 −1
(d) Eqn Slope y-int Pt on line Pt on line
√
5t2 + 4t + 4 y = 2x + 1 2 1 (0, 1) (− 1 , 0)
2
Answer 3.1 (a) (x + 3)2 + (y − 4)2 = 9 (b) (x − 3)2 +

“ ”2 y = − 11
4
x + 17
4
− 11
4
17
4
(3, − 4) (−1,7)
y + 11
3
1
= 16 (c) Draw a vertical and horizontal line
through (1,1). On the vertical line, a circle of radius 2 y = −2x + 1 −2 1 (0, 1) (1 , 0)
2
will have a center at either (1,3) or (1,−1). Likewise, the
horizontal line will have circles at either (−1, 1) or (3, 1).
y= 1
2
x+1 1
2
1 (0, 1) (−2, 0)
Eqn C(h,k) (x − h)2 + (y − k)2 = r2

y = 1,000 0 1,000 (0, 1,000) (0, 1,000)
1. (3, 1) (x − 3)2 + (y − 1)2 =4
2. (1, 3) (x − 1)2 + (y − 3)2 =4
y= 0 0 0 (0, 0) (1, 0)
3. (−1, 1) (x + 1)2 + (y − 1)2 =4
4. (1, −1) (x − 1)2 + (y + 1)2 =4
x= 3 Undef None (3, 3) (3, −2)
√ √
(d) (1,1), (1, − 3), (1 + 3,0), (0, − 1 − 3)
y = x − 14 1 −14 (5, − 9) (0, −14)
√
Answer 3.2 (a) center (3,-1), radius 2 p3 (b) center (-2,-3),
radius 2 (c) center (-1/6,5/3), radius 203/12 (d) center Answer 4.5 (a) y = 6,850(x − 1970) + 38,000 (b) y =
√
(3/4, -1/2), radius 3 8,000(x − 1970) + 8,400 (d) Tabularize your answer:
Answer 3.3 (a) Wet in 24.0 minutes. (b) Wet in Port

25.4154041 minutes. Year Seattle Townsend
1983 $127,050 $112,400
Answer 3.4 (a) Imposing with x-axis along ground and y- 1998 $229,800 $232,400
axis along tower, wheel modeled by x2 + (y − 62)2 = 602 .
(b) 46.43 ft. to right of tower. (c) (−24,7) and (−24,117). (e) 1995.74, $214,313 (f) 1982.70, Sea = $124,965 (g) 2008.78,
Sea = $303,661 (h) 1972.32, Sea = $53,892 (i) No.
Answer 3.5 Lee has 2.265 in2 more pie.
Answer 4.6
√
Answer 3.6 (a) Impose a coordinate system so that the −2α2 ±
4α4 − 4α
tractor is at the origin at t = 0 seconds. With this coor- x=
2α
dinate system, the south edge of the sidewalk is modeled
√
by ys = 100; the north edge is modeled by, yn = 110. (b) −x2 ± x4 − 8x
t = 32 minutes. (c) t = 52 minutes. (d) 20 minutes. α=
4x
Answer 3.7 (a)] The equation for eastward travel from Answer 4.7 Line of travel is y = (0.667)(x − 35). (a)
Kingston is y = 8. Southward travel along x = −1. (b) (−13.46,−32.31) (b) 3.19 seconds (c) 5.82 seconds (d)
Boundary: (x + 2)2 + (y − 10)2 = 9; Interior:(x + 2)2 + (y − (10.78,−16.17) at about t = 6.1 sec.
10)2 < 9; Exterior:(x+2)2 +(y−10)2 > 9. (c) 3.82 minutes.
(d) Plug x = −1 into circle equation to find exit point. Use Answer 4.8 (a) Allyson at (0,70); Adrian at (−44,0).
this to find when the ferry exits the radar zone. Be careful Bungee is 83.1401 ft. long. (b) Occurs at time t = 7.7546
to reference all times relative to when the ferry departed seconds and Allyson is at (0,77.546). Allyson’s final loca-
Kingston. (e) 20.32 minutes. tion is 77.546 ft. from her starting point.
291
Answer 4.9 The lines of tangency have equations y = Answer 5.8 (a) − 23 ,− 52 , x2 , 21 (♥ − 3). (b) 0, 20, 6x + 2x2 ,
± √2 (x − 12). The non-visible portion of the y-axis is 2♥2 − 6♥, 2(♥ + △ − 3)(♥ + △). (c) Always 4π2 .
5√ √
− 245 5
≤ y≤ 24 5
5
.
Answer 5.9 Use the vertical line test. For example, (a) is
not a function, by the vertical line test; you can split it
Answer 4.10 C = 95 (F − 32) and F = 9
5
C + 32. In Oslo, the into two function graphs by slicing the ellipse symmetri-
temperature is −9.4◦ F. cally into upper and lower halves. On the other hand, (o)
is a function, by the vertical line test; etc.
Answer 4.11 At the closest point, she will be
26.22471828 miles from Paris. √ 5.10 (a) x = 5.
Answer (b) x = 4, 8. (c) x = 36. (d)
242− 56400
2
Answer 4.12 Impose coordinates with Angela’s initial lo-
cation as the origin. Angela is closest to Mary at Answer 6.1 (a) 0, 2, 3. (b) x = ±4, x = 0, no solution. (c)
(18.8356,41.4383); this takes approximately 3.8 seconds. Intersect at (4,4) and (− 43 , 43 ). Area is 16
3
.
Answer 4.13 (a) Impose coordinates with sprinkler ini- Answer 6.2 (a)
tial location as the origin (0,0). Line equation becomes
y = − 51 x + 100. (b) Sprinkler is located at (0,82.5) at time −0.5x − 1 if x ≤ −2
| − 0.5x − 1| =
t = 33 minutes. Circular boundary of watered zone hits 0.5x + 1 if x > −2
southern edge of sidewalk at the points (−6.708,101.3416)
and (13.4386,97.3123). (c) y = − 51 x + 110.198. y
3
Answer 4.14 (a) With the origin set at the statue, x = 2.5
30 − 3.123475t,
√ y = 2.49878t. (b) The distance to the
statue is 16t2 − 187.4085t + 900 feet. (c) Margot will be 2
28 feet from the statue after 0.655672 seconds, and af-
1.5
ter 11.057360 seconds (assuming she continues past the
point due north of the statue). 1
0.5
Answer 4.15 18.923076923 seconds
x
p √ -4 -2 2 4
Answer 4.16 (a) (−1 ± 7/3)/2 = (−3 ± 21)/6. (b)
t =√−0.463325,0.863325; or the exact answer would be
Answer 6.3 (a) x = 4 and x = − 22 (b) x = 0.75 (c) the
(1 ± 11)/5. (c) t = 0.2. (d) No real number solutions. 3
equation has no solutions.
p √
p (a)√There are four√answers: x = ±
Answer 4.17 2+ 2 Answer 6.4 (a) x = −7 and x = 3 (b) a = 3 (c) x = 8/3
and x = ± 2 − 2 (b) y = 6 + 2 5.
1
Answer 6.5 For 0 ≤ x ≤ 6, the area is 6
x(x + 18).
Answer 5.1 (a) −2 + h + 2x. (b) 2. (c) h + 2x. (d) −h − 2x.
1 √
(e) −π(h + 2x) (f) √h+x−1+ . Answer 6.6 (a) The rule is
x−1

2x if 0 ≤ x ≤ 10
Answer 5.2 g(x) = 95 x + 24, f(x) = 45 x + 4, v(x) = x + 20, y=
−x + 30 if 10 ≤ x ≤ 30
v(5) = 25, minimum v(x) = 20, maximum v(x) = 40.
and the range is 0 ≤ y ≤ 20.
Answer 5.3 For example, in (a), suppose Dave has con- (b) The rule for the area function a(x) is
stant speed v ft/min. Then the function s = d(t) = vt
will compute the distance Dave travels in t minutes. The x2 if 0 ≤ x ≤ 10
a(x) =
graph would be a line with s-intercept 0 and slope v; the − 12 x2 + 30x − 150 if 10 ≤ x ≤ 30
domain would be 0 ≤ t ≤ 2400
v
; etc.
(c) x = 12.6795 inches.
Answer 5.4
Answer 6.7 (a)
Answer 5.5 Several possible answers for each one.
0 if t < −1
√ j(t) =
62.22(t + 1) if −1 ≤ t ≤ 3.5
Answer 5.6 (a) x-intercepts= 16 (3 ± 33). y-intercept =
1
√ 1
√
−2. (b) ( 6 (3 − 93),5) and ( 6 (3 + 93),5). (c) None. (d) (b)
√ √ √
if t < 0
p p
Yes, No, No, Yes. (e) ( 1 + 2,−2−3 1 + 2+3(1+ 2)). 280
s(t) =
280 − 70t if 0 ≤ t ≤ 4
p
Answer 5.7 (b) f(x) = 30000x + 50000 (4 − x)2 + 1. (c)
(c) The table of values (x,f(x)) will be: (0., 206155), 
(0.5,197003), (1, 188114), (1.5,179629), (2,171803),  280 if t < −1


(2.5,165139), (3.,160711), (3.5,160902), (4,170000). Min-  280 − 62.22(t + 1) if −1 ≤ t < 0
imal cost occurs for some 2.5 < x < 3.5; the exact answer d(t) = 217.78 − 132.22t if 0 ≤ t < 1.6471

is x = 13
4
, but we cannot solve this in our class since it 
 132.22t − 217.78
 if 1.6471 ≤ t < 3.5
requires the tools of Calculus. 70t if 3.5 ≤ t < 4
Answer 6.8 The distance to his starting point t seconds Answer 7.2 (a) maximum value is 22; minimum value is
after he starts is 1.75 (b) maximum value is 32; minimum value is 2 (c)
 maximum value is -6; minimum value is -46
 10t
 p if 0 ≤ t ≤ 25,
d(t) =
2
q250 + (12(t − 25))
2 if 25 ≤ t ≤ 175
3
, Answer 7.3 (a) Multipart function: s(t) = − 1601
t2 + 85 t+10

 for 0 ≤ t ≤ 114.031, s(t) = 0 for t ≥ 114.031.; $23,125. (b)
400 + (250 − 9(t − 175
2
3
))2 if 175
3
≤ t≤ 205
3
.
Sell at time t = 50 days; $2500. (c) On day t = 115 it is
worthless.
Answer 6.9
 Answer 7.4 The parabola has x intercepts at −1,3 and y

 18t if 0≤ x≤ 5
 p
2 + 182 (t − 5)2 intercept at -3. The vertex of the parabola is (1, − 4).
90 if 5 ≤ x ≤ 10
y = d(t) = p

 902 + (90 − 18(t − 10))2 if 10 ≤ x ≤ 15 

90 − 18(t − 15) if 15 ≤ x ≤ 20  x2 − 2x − 3 if x ≤ −1
2
|x − 2x − 3| = −x2 + 2x + 3 if −1 < x < 3

x2 − 2x − 3 if x ≥ 3
Here is the graph of d(t)
Answer 7.5 (a) 100 ft. (b) 156.25 ft. (c) (625, − 125). (d)
ft
When x = 54.81 ft. or x = 570.19 ft.; i.e. at (54.81,39.04)
140 and (570.19, − 64.04).
120
100 Answer 7.6 (a) No, since f(1) = 1 6= 2. (b) The points
80 (1,1+2b) and (−(1+b),1+2b). (c) Only the point (a,1−a2 ).
60 (d) The points (−2.2701, − 7.1168) and (2.9368,1.5612).
40
20 Answer 7.7 She should have 225 trees in the orchard.
t sec
5 10 15 20
Answer 7.8 She should charge $7.72 to make the most
money.
Answer 6.10 (a) v(x) = 2x3 − 70x2 + 500x; degree 3. (b)
a(x) = 1000 − 50x − 2x2 , degree 2. To get 600 sq. in. Answer 7.9 The radius of the circular part should be
dimensions are: 6.3746 × 7.2508× 18.6254.
24
Answer 6.11 (a) 2 hours. (b) Impose coordinates with x ≈ 3.36059492 feet.
4+π
axis the bottom of the ditch and the y axis the pictured
centerline. The short side of the rectangular part should also be equal
 to this.
 20 if x ≤ −40

 p
2


 10 + 100 − (x + 40) if −40 ≤ x ≤ −30


p
2 Answer 7.10 The enclosure should be 50 meters by 75
 10 − 100 − (x + 20) if −30 ≤ x ≤ −20
meters.
y= 0 if −20 ≤ x ≤ 20


p

 10 − p100 − (x − 20)2 if 20 ≤ x ≤ 30

 2 Answer 7.11 Cut so the pieces have lengths 26.394 in.

 10 + 100 − (x − 40)
 if 30 ≤ x ≤ 40
and 33.606 in.; bend the 26.394 inch piece into a circle.
20 if 40 ≤ x
√ Answer 7.12 (a) y = − 23 x2 + 53 x. (b) y = 1.125x2 − 1.5x −
(c) 2(40 − 91) feet. (d) 42 ft. wide: 0.3008 minutes. 50 ft.
wide: 8.038 minutes. 73 ft. wide: 116.205 minutes. 1.625. (c) y = −0.5x2 + 3.5x − 4. (d) No solution.
Answer 6.12 (a) 0 ≤ y ≤ 6. (b) Increasing: −6 ≤ x ≤ −2 Answer

p 7.13 (a) The distance between Sven and Rudyard
and 2 ≤ x ≤ 4. Decreasing: −2 ≤ x ≤ 2 and 4 ≤ x ≤ 6. is 41t2 − 2400t + 36900 feet, where t is the number of
(c) seconds after they start moving. (b) They will be closest
together after they have been moving for 29.268292 sec-
 onds, and will be 42.166916 feet apart at that time.

 2x + 12 if −6 ≤ x ≤ −4
 p
4 + p4 − (x + 2)2 if −4 ≤ x ≤ 0
y=
 4 − 4 − (x − 2)2
 if 0≤ x≤ 4 Answer 7.14 Michael M(t) = (5.547t,8.32t), Tina T(t) =
 (400 − 8.944t,50 + 4.472t). Study the distance SQUARED
−2x + 12 if 4≤ x≤ 6
from M(t) to T(t). Michael and Tina will be closest when
(d) 2 ≤ y ≤ 6. (e) 2 ≤ y ≤ 4. t = 26.641 sec; they are 54.3 ft at that instant.
Answer 6.13 (a) 0. (b) a = 13

, b = − 53 , c = −3. Answer 7.15 In the first case, d = 0 or d = 12. In the
5
second case, d = ± √2 .
13
Answer 7.1 (a) 2(x−4)2 +9, Vertex: (4, 9), Axis: x = 4. (b)
3(x − 5/2)2 − 383/4, Vertex: (5/2, -383/4), Axis: x = 5/2. √ (d) There are two possible values for α: If
Answer 7.16
(c) (x− 3/14)2 + 2539/196, Vertex: (3/14, 2539/196), Axis: α = 8 + 2 17, then√ the unique solution
√ of the equation
x = 3/14. (d)2(x − 0)2 − 0, Vertex: (0, 0), Axis: x = 0. (e) will be x = −4 − 17. If α = 8 − 2 17, √ then the unique
(1/100)(x − 0)2 − 0, Vertex: (0, 0), Axis: x = 0. solution of the equation will be x = −4 + 17.
293
1
√ √ √
Answer 7.17 (a) t = 2
(2± 2 s − 1). (b) x = −1± y − 2. Answer 8.6 (a) f(g(x)) = (x + 3)2 , f(f(x)) = x4 , g(f(x)) =
x2 + 3. (b) f(g(x)) = √1x , f(f(x)) = x, g(f(x)) = √1x . (c)
Answer 8.1 (a) f(g(x)) = x, f(f(x)) = 81x + 20, g(f(x)) = x. (d) f(g(x)) =
6(x − 4)2 + 5, f(f(x)) = 6(6x2 + 5)2 + 5, g(f(x)) = 6x2 + 1.
−t + 1 if t ≤ 1 (e) f(g(x)) = 8x + 21, f(f(x)) = 4(4x3 − 3)3 − 3, g(f(x)) = 2x.
h(f(t)) =
t−1 if 1 ≤ t (f) f(g(x)) = 2x3 + 1, f(f(x)) = 4x + 3, g(f(x)) = (2x + 1)3 .
(g) f(g(x)) = 3, f(f(x)) = 3, g(f(x)) = 43. (h) f(g(x)) = −4,
−t − 1 if t ≤ −1 f(f(x)) = −4, g(f(x)) = 0.
h(g(t)) =
t+1 if −1 ≤ t
(b) Answer 8.7 −5/2 ≤ x ≤ −1/2

−t − 1 if t ≤ 0 Answer 8.8 (a) 27 ≤ x ≤ 6. (b) −3 ≤ y ≤ 5. (c) 1 ≤ x ≤ 6.
f(h(t)) =
t−1 if 0 ≤ t (d) −9 ≤ y ≤ 7. (e) B = 5,C = 39 . (f) A = 18 , D = 38 .
5

t−1 if t ≤ 0 −1 −1
g(h(t)) = Answer 8.9 (a) , set h = 0 to get (b)
−t − 1 if 0 ≤ t (x−1)(x+h−1) (x−1) 2
√ −h−2x
4+4h+8x, set h = 0 to get 4+8x. (c) √ ,
(c) 25−x 2 + 25−(x+h) 2
−x
 set h = 0 to get √ .
25−x 2

 −t − 1 if t ≤ −1

t+1 if −1 ≤ t ≤ 0
h(h(t) − 1) = Answer 9.1 (a) domain of f ={x|x 6= 4
}; range=f= {y|y 6= 0}.

 −t + 1 if 0≤ t≤ 1 3

t−1 if 1≤ t (b) f−1 (y) = 2+4y .
3y
y
Answer 9.2
3
√
2.5 3+ 17+8y
Answer 9.3 (c) f−1 (y) = 4
on the domain {y|y ≥
2 − 17
8
}.
1.5
Answer 9.4 Only (B) is one-to-one on the entire domain.
1
0.5 Answer 9.5
x 2
-3 -2 -1 1 2 3 Answer √9.6 (a) h = f(x) = −2x + 124x. (c) x = g(h) =
-0.5 31 − 21 3844 − 2h.
-1
Answer 9.7 (a) (1)
Answer 8.2 (a) y = f(g(x)),

√ if f(x) = x5 ,
g(x) = x − 11. (b) y
y = f(g(x)), if f(x) = 3 x, g(x) = 1 + x2 . (c) y = f(g(x)), if
f(x) = 2x5 − 5x2 + (1/2)x + 11, g(x) = x − 3. (d) y = f(g(x)),
√
if f(x) = (1/x), g(x) = x2 + 3. (e) y = f(g(x)), if f(x)
√ = x,
√
g(x) = x + 1. (f) y = f(g(x)), if f(x) = 2 − 5 − x2 ,
g(x) = 3x − 1. x
Answer 8.4 (c) x = −1 and x = 2
s
Answer 8.5 (a) f(s) = 60
;
y+2
70s2 (2) x = f−1 (y) = 3
has domain and range all real num-
C(f(s)) = bers. (3)
10(602 ) + s2
x
C(f(s)) computes mph when you input seconds s. (b)
g(h) = 60h;
252000h2
C(g(h)) =
10 + 3600h2 y
C(g(h)) computes mph when you input hours h. (c)

22
v(s) = 15 s;
308m2
v(C(m)) =
30 + 3m2 (4) f(f−1 (y)) = f( y+2 ) = 3( y+2 ) − 2 = y + 2 − 2 = y;
3 3
v(C(m)) computes ft/sec when you input minutes m. f (f(x)) = f (3x − 2) = 3x−2+2
−1 −1
3
= x.
q
Answer 9.8 (a) w = f(t) = 2 25 − ( 6t − 5)2 . Domain: Answer 12.5 (a) 8.8 cm. (b) 135 cm. (c) 4.84 yrs. (d)
0 ≤ t ≤ 30 hours; Range: Maximum possible length of the halibut.
q 0 ≤ w ≤ 10 ft. (b) t = 6 hours.
(c) t = f−1 (w) = 30 − 6 25 − ( w
2
)2 . Domain: 0 ≤ w ≤ 10
ft.; Range: 0 ≤ t ≤ 30 hours. Answer 12.6 (a) C(t) = 1.03526t .
Answer 9.9 (a) 2 nanoseconds. (f) domain φ={0 ≤ t ≤ 10}; Answer 12.7 (c) v(x) = 55000(1.0315)x . Valued at
range φ={0 ≤ y ≤ 6}. $200,000 during 1993.
Answer 10.1 (a) 31.5443; (b) 355.1134; (c) 36.4622; (d) Answer 12.8 (a) x = 2.7606. (b) 1.392. (c) 0.3552. (d)
0.0616; (e) 51,168; (f) 0.009794; x = e8 . (e) x = 11.513. (f)-0.61. (g) sin(x) = −0.6931;
x = −0.7658 + 2kπ or x = 3.9074 + 2kπ.
Answer 10.2 (a) y = 3( 21 )x (b) y = ( 21 )x . (d) y = 1 √1 x
27
( ) .
3
Answer 12.9 (d) 8.617 weeks.
Answer 10.3 (a) 1.64 × 106 cells. (b) True. (c) The two
formulas are identical. Answer 12.10 31.699250014 days
Answer 10.4 (a) 261.31 Hz. (b) 440 Hz. (c) 27.5 Hz. (d) Answer 12.11 (a) 2015 (b) 2048.
16.35 Hz.
Answer 12.12 (a) 46.701735 years (b) 137.3113631016
Answer 10.5 (a) 29 = 512. (b) 2n−1 . (c) 263 = 9.2 × 1018 . years after 1980
(d) 9.2 × 1015 meters.
Answer 13.1 The graph is given below:
Answer 10.6 (a) M(p) is the higher curve. (d)
y
fraction 2
0.8
1.5
0.6
1
0.4
0.5
0.2
x
-2 -1.5 -1 -0.5 0.5 1 1.5 2
p
20 40 60 80 100 -0.5
Answer 11.1 (a) w(t) = $1.113(1.046408)t . (b) $1.11. (c) -1

It is below; should be $5.70 by the model.
-1.5
Answer 11.2 (a) p(x) = 860(1.070674)x , l(x) = 70x + 860.
(b) p(10) = 1702,l(10) = 1560.
-2
Answer 11.3 If we use 1989 and 2000 in a(t), we get
two data points and a corresponding exponential model: Answer 13.2 (a)
E(t) = 15.918(1.243301)( t − 1980). The exponential model
grows faster than the cubic model and eventually exceeds
6
a(t).
4
Answer 11.4 (c) The curve is modeled by y =
x
500cosh( 500 ) − 440.536 and the minimum height is 59.46
feet. 2
Answer 12.1 (a) 0.6826; 2.3979; 3.3030; 3.3219;0.3010. 0

-3 -2 -1 0 1 2
(b) 3.555; 19.8; 0.0729. (c) x = log10 y; x = log10 (3y);
x = (1/3)log10 (y). -2
Answer 12.2 (a) I(d) = I◦ (0.94727)d . (b) 85 meters. -4
Answer 12.3 (a) y = 13e1.09861t . (b) y = 2e−2.0794t . -6
Answer 12.4 (a) 9.9 years. (b) 34.66%. (b)

295
(b) No. (c)

1.5


 0 if x ≤ −2
1 
x+2 if −2 ≤ x ≤ 0
y = f(−x) =
 −2x + 2
 if 0≤ x≤ 1

0.5
0 if x≥ 1
0
-0.5 0 0.5 1 1.5 2 y
4
-0.5
3
-1
2
-1.5
1
(c)
x
-4 -3 -2 -1 1 2 3 4
-1
10 

 0 if x ≤ −1

−2x − 2 if −1 ≤ x ≤ 0
y = −f(x)
5 
 x−2 if 0≤ x≤ 2

0 if x≥ 2
0
-12 -10 -8 -6 -4 -2 0 2 4 y
1
-5
x
-4 -3 -2 -1 1 2 3 4
-10 -1
Answer 13.3 − 13 ≤ x ≤ 0. -2
Answer 13.4 (a5) (1) horizontally dilate (compress) by a -3

factor of 2; (2) horizontal shift right by 12 ; (3) vertical dilate
(expand) by a factor of 3; vertical shift up by 5. -4

−6x + 8 if x < 21
f(x) = (d) Here are the graphs of y = 2f(x) and y = 1
f(x), respec-
6x + 2 if x ≥ 21 2
tively:
y
30
y
25 4
20 3
15 2
10 1
5 x
-4 -3 -2 -1 1 2 3 4
x
-2 -1 1 2 3 4 -1
Answer 13.5 (a)

y y
4 4
3 3
2 2
1 1
x x
-4 -3 -2 -1 1 2 3 4 -4 -3 -2 -1 1 2 3 4
-1 -1
(e) Here are the graphs of y = f(2x) and y = f( 12 x), respec- Answer 14.3 (a) m(t) = 35t + 200. (b) k(t) = 30t − 50. (c)
tively: 1987. (d) r(t) = 35t+200
30t−50
. (e) 76 .
y
4 Answer 14.4 (a) y = 0.03x2 −1.1x+14 (b) w = 0.03x−1.1+
14
x
(c) 70 or 6 23
3
41 35
11
x + 11 41x + 35
2 Answer 14.5 f(x) = 65
= . The horizon-
x + 11 11x + 65
1 41
tal asymptote is y = .
11
x
-4 -3 -2 -1 1 2 3 4
6x + 10
-1 Answer 14.6 f(x) =
x+1
y Answer 14.7 You should study for 11.25 hours.

4
400
3 Answer 14.8 (a) k = 400 (b) I(t) = 484t 2 −1452t+1189
(c)
t = 1.5 (d) t = 1.05 and 1.95.
2
Answer 14.9
1
400x+20000
x Answer 14.10 (a)y = f(x) = x+200
. (b) x = $400.
-4 -2 2 4
(d)x = f−1 (y) = 200y−20000
400−y
. Domain f−1 = {100 ≤ y ≤
-1 388.46}, Range f−1 = {0 ≤ x ≤5000}. The inverse function
takes the number of customers per day as an input value
(f) c = 11
. (g) c = 5
. (h) c = d = 1
. and gives the amount the shop spent on advertising as an
2 2 3
output value.
√
Answer 13.6 (a) a(x) =√ 2(6 − x) 3x − 9 with domain 3 ≤ Answer 15.1 (a) 13o 24′ or 0.233874 rads. (b) 1.0788
x ≤ 6. (b) 1 + 36(1 − x) x on√the domain 0 ≤ x ≤ 1. 1 + degs or .01882 rads. (c) 5.7296 degs or 5o 43′ 46.5".
√
36(1−x)
√ x (c) Max of a(x) is 4 3 and max of 2a(3x+3)+1
is 8 3 + 1 (when x = 13 .).
Answer 15.2 (a) 6080 ft. (b) 29.95 mph. (c) 15.63 knots.
Answer 13.7 (c) Horizontally shift the graph of y = x2 to Answer 15.3 (a) 430 sq. in. (b) 2.56◦ . (c) 84.47 in. (d) 23.04
the right 4 units. in. (e) 7.29 in.
Answer 13.8 (a) y = 7|2x + 4| + 2. Answer 15.4 (a) 1413.7 sq. ft. (c) 4.244 sec.
Answer 13.9 (a1) f(2x) = 4x . f(2x − 1) = 1 x

4 . Answer 15.5 2164.208272472 miles.
2
Answer 15.6 (a) 2.147 hrs. (b) 1103 mph. (c) 12.47 hrs.,
Answer 13.10 y = 13 (2x ) is obtained by vertically dilating
6236 miles. (d) 13760 miles.
y = 2x ; it is vertically compressed. y = 2x/3 is a horizontal
dilation of y = 2x ; it is horizontally stretched.
Answer 15.7 (b) 84.47 sq. in. (c) 181.5 sq. in.
Answer 14.1 (a) domain={x|x 6= 1}; range={y|y 6= 2}; zero Answer 15.8 0.685078 miles.
at x = 0; horizontal asymptote y = 2; vertical asymptote
x = 1; graph below:
Answer 15.9 Middle picture: shaded area= 12.537 sq. in.
y
15 Answer 16.1 (a) 10π/3 rad = 10.47 radians. (b) 4/2π rev *
1 hour/rev = .64 hours = 38.2 minutes. (c) Using (2.2.2),
10
(21)(10π/3) = 219.91 meters.
5
Answer 16.2 194 RPM.
x
-4 -2 2 4
-5 Answer 16.3 (a) ω = 11π 30
rad/sec, v = 121π
15
ft/sec. (b)
-10
v = 2.094 in/sec, ω = 2.5 RPM.
-15 Answer 16.4 (b) 700 ft. (c) 70 sec. (d) 15000 sq. ft.
0.2x−10.4
Answer 14.2 (a) y = x−10
. (b) x = 9.143 ft.; x = 9.916 Answer 16.5 (a) r = 233.427 ft. (b) θ = 0.06283 rad. (c)
ft. (c) 10 ft. 2π/5 rad counterclockwise from P.
297
2
Answer 16.6 (a) 40π ft/sec; 85.68 mph. (b) 400 RPM. (c)
60 RPM; 32π ft/sec. (d) 1.445 rad = 82.8◦ ; 28.9 ft. (e) 0.7 1.5
sec; 1.4π rad. 1
0.5
Answer 16.7 4.4 inches.
-6 -4 -2 2 4 6
-0.5
Answer 16.8 r = 2.45 inches. -1
-1.5
Answer 17.1 (a) If you impose coordinates with the cen- -2
ter of the wheel at (0,237.427), then ground level coin-
cides with the x-axis. (a) T(t) = (x(t),y(t)), where x(t) =
233.427cos( 2π5
t − 0.06283) and y(t) = 233.427sin( 2π 5
t− Answer 18.2 (a)
0.06283) + 237.427. (b) T(6) = (85.93,454.45). (c) First find
the slope of a radial line from the wheel center out to Tiff’s 2
launch point. 1.5
1
Answer 17.2 (a) y = ±0.2309(x + 1) + 2
0.5
Answer 17.3 290 ft. -6 -4 -2 2 4 6

-0.5
-1
Answer 17.4 (a) (-21.91218, -1.498834) (b) (-5.92564,
21.14892) (c) (19.07064, -10.89497) -1.5
-2
Answer 17.5 (a) (19.9,13.42). (b) (22.55,8.21). (c)
(−1.674,23.942). (d) (23.882,2.375). (b)
2
Answer 17.6 101.496936 feet above the ground.
1.5
1
Answer 17.7 The dam is 383 feet high.
0.5
Answer 17.8 108 ft. -6 -4 -2 2 4 6

-0.5
Answer 17.9 With the center of the track at the ori- -1

gin, and the northernmost point on the positive y- -1.5
axis, Charlie’s location after one minue of running is -2
(59.84016,4.37666).
(c)
Answer 17.10 105.2718216 feet
Answer 17.11 (a) 204.74 ft. (b) no.

15
Answer 17.12 Top right scenario: (a) θ◦ = 1.2 rad.

(b)θ(t) = 1.2 + 4πt
9
(c) b(t) = (2cos(1.2 + 4πt
9
),2sin(1.2 +
4πt
9
)). (d) b(1) = (−1.710,1.037). b(0) = (0.725,1.864).
b(3) = (1.252, − 1.560). b(22) = (1.753,0.962). 10
1
Answer 18.1 (b) sin2 (x) = 2
(1 − cos(2x)).
5
2
1.5
1
0.5 0
-6 -4 -2 0 2 4 6
-6 -4 -2 2 4 6
-0.5
-1
-1.5 √
40
-2 Answer 18.3 (a) 7/25 or -7/25. (b) -0.6. (c) ± 7
.
(c) Answer 18.4 In the first case,

Answer 19.7 (a)2 volts. (b) Never zero since 2x is positive

2π
for all x. (c)p(t) = 3sin( 2/5 (t− 35 )+1, so A = 3, D = 1, B =
2/5, C = 3/5. (d) 0.25 ≤ V(t) ≤ 16. (e)t = 0.65635 + k(0.4)
and t = 0.74365 + k(0.4), k = 0, ± 1,±2, . .. are ALL solu-
one period
tions. Four of these lie in the domain 0 ≤ t ≤ 1. (f) Maxima
have coordinates (0.3 + k(0.4),16) and minima have coor-
dinates (0.1 + k(0.4),0.25), where k = 0, ± 1, ±2,. . .. (g) If
we restrict V(t) to 0.5 ≤ t ≤ 0.7, the inverse function has
rule:
ln(y)−ln(2)
arcsin( 3 ln(2) ) + 3π
t= .
5π
π 5π
Answer 18.5 4
+2kπ and 4
+2kπ, k = 0,±1, ±2,±3, .. . . If we restrict V(t) to 0.5 + k(0.4) ≤ t ≤ 0.7 + k(0.4), the
inverse function has rule:
Answer 18.6 (c) On domain t ≥ 0, arcsin( ln(y)−ln(2) ) + 3π
3 ln(2)
t = k(0.4) + .
q 5π
B = B(t) = 2cos(6πt) + 36 − 4(sin(6πt))2 .
In particular, if we restrict V(t) to 0.1 ≤ t ≤ 0.3, the inverse
function has rule:
(e) 0.1023, 0.2310.
arcsin( ln(y)−ln(2)
3 ln(2)
) + 3π
t = −0.4 + .
Answer 19.1 (a) 1, π, π
, 1. (b) 6, 2, 0, -1. 5π
2
If we restrict V(t) to 0.7 ≤ t ≤ 0.9, the inverse function has
2π rule:
Answer 19.2 (d) h(t) = 8sin( 1.2 (t − 0.3)) + 18. ln(y)−ln(2)
−arcsin( 3 ln(2) ) + 4π
t= .
5π
2π
Answer 19.3 (a) b(t) = 0.6sin( 100 (t − 30)) + 1.2. If we restrict V(t) to 0.7 + k(0.4) ≤ t ≤ 0.9 + k(0.4), the
inverse function has rule:
Answer 19.4 (a) h(t) = 5sin( π 6
(t − 10) + 5, where t indi- −arcsin(
ln(y)−ln(2)
) + 4π
cates hours after midnight. (b) 7.5 ft. above low tide. 3 ln(2)
t = k(0.4) + .
5π
Answer 19.5 (a) A = 25, B = 5 seconds, C = 1.75, D = 28. In particular, if we restrict V(t) to 0.3 ≤ t ≤ 0.5, the inverse
(b) t=1.75 and 4.25 seconds function has rule:
−arcsin( ln(y)−ln(2)
3 ln(2)
) + 4π
Answer 19.6 First scenario: x(t) = 2π
2sin( 4.5 (t − Cx )), t = −0.4 + .
5π
where Cx = −(1.2 + π 2
9
)( 4π ); y(t) = 2π
2sin( 4.5 (t − Cy )),
9 Answer 20.1 (a1) 0, 1.5708, −1.5708, 1.0472, 0.7168,
where Cy = −(1.2)( 4π ). Plots are below:
−0.2762, not defined.
Answer 20.2 (a) Principal solution: x = 1.9106, symmetry

x
solution: x = 4.3726; graph below with these two solutions
2 graphically indicated:
1 y
2
1.5
t
2 4 6 8 1
0.5
-1
x
-10 -5 5 10
-0.5
-2
-1
-1.5
y
2 -2
Answer 20.3 (a) 9.39, 13.63, 11.09. (b) Feb. 3, Nov. 4.

1
Answer 20.4 (a) A = 15, D = 415, B = 10,C = 15/2. Note
that C = 10k + 15/2, k = 0,±1, ±2,. . . are also all valid
t choices for the phase shift. (b) maximum temperature=
2 4 6 8
430o F. (c) minimum temperature= 400o F. (d) 12.1635
-1 minutes. (e) 14.6456 minutes. (f) 6.80907 minutes.
Answer 20.5 The cake should be in the oven for

-2 67.572141357 minutes.
299
Answer 20.6 6.42529 hours 8
Answer 20.7 8.9286 hours of dry time each day. 6
Answer 20.8 The key fact to use over and over 4

is this: M(t)=M’s location after t seconds =
(100cos(0.025t),100sin(0.025t)); T(t)= T’s location after 2
t seconds = (100cos(0.03t + π),100sin(0.03t + π)).
x
Answer 20.9 (a)α = arcsin[3960/(3960 + t)]. (b) α = -2 -1.5 -1 -0.5 0.5 1 1.5 2
6.696degs. The interior angle is 166.608 degs and so one
satellite covers 46% of the circumference. Thus you need -2
3 (not two-point-something) satellites to cover the earth’s
circumference. (c)α = 52.976deg. The interior angle is
74.047 degs and so one satellite covers 20.57% of the cir-
cumference. Thus you need 5 satellites to cover the earth’s Answer 20.11 (a) 3.852624 miles (a) 14.34lb/in2 ; no ex-
circumference. (d) Get an equation for the interior angle plosion (b) 8.32lb/in2 ; explosion (c) 54.6 degrees
in terms of t. Solve 2(90 − arcsin[3960/(3960 + t)]) = 20%
of 360 degs = 72 degs. You’ll get t = 934.83 miles.
Answer 20.10 (a) Domain: 13 − π 6

≤ x ≤ 31 + π 6
; Range: Answer 20.12 (a) Many possible answers; for example:
1 π
1 ≤ y ≤ 5. One solution: x = 3 + 18 ; graph is below: −1.9293, −1.3677, 0.8677, 1.4293.
Appendix C
GNU Free Documentation

License
Version 1.1, March 2000
c 2000 Free Software Foundation, Inc.

Copyright
59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
Everyone is permitted to copy and distribute verbatim copies of this li-
cense document, but changing it is not allowed.
Preamble
The purpose of this License is to make a manual, textbook, or other

written document “free” in the sense of freedom: to assure everyone the
effective freedom to copy and redistribute it, with or without modifying
it, either commercially or noncommercially. Secondarily, this License
preserves for the author and publisher a way to get credit for their work,
while not being considered responsible for modifications made by others.
This License is a kind of “copyleft”, which means that derivative works
of the document must themselves be free in the same sense. It com-
plements the GNU General Public License, which is a copyleft license
designed for free software.
We have designed this License in order to use it for manuals for free
software, because free software needs free documentation: a free pro-
gram should come with manuals providing the same freedoms that the
software does. But this License is not limited to software manuals; it can
be used for any textual work, regardless of subject matter or whether it
is published as a printed book. We recommend this License principally
for works whose purpose is instruction or reference.
301
302 APPENDIX C. GNU FREE DOCUMENTATION LICENSE
C.1 Applicability and Definitions

This License applies to any manual or other work that contains a notice
placed by the copyright holder saying it can be distributed under the
terms of this License. The “Document”, below, refers to any such manual
or work. Any member of the public is a licensee, and is addressed as
“you”.
A “Modified Version” of the Document means any work containing the
Document or a portion of it, either copied verbatim, or with modifications
and/or translated into another language.
A “Secondary Section” is a named appendix or a front-matter section
of the Document that deals exclusively with the relationship of the pub-
lishers or authors of the Document to the Document’s overall subject (or
to related matters) and contains nothing that could fall directly within
that overall subject. (For example, if the Document is in part a textbook
of mathematics, a Secondary Section may not explain any mathematics.)
The relationship could be a matter of historical connection with the sub-
ject or with related matters, or of legal, commercial, philosophical, ethical
or political position regarding them.
The “Invariant Sections” are certain Secondary Sections whose titles
are designated, as being those of Invariant Sections, in the notice that
says that the Document is released under this License.
The “Cover Texts” are certain short passages of text that are listed, as
Front-Cover Texts or Back-Cover Texts, in the notice that says that the
Document is released under this License.
A “Transparent” copy of the Document means a machine-readable
copy, represented in a format whose specification is available to the
general public, whose contents can be viewed and edited directly and
straightforwardly with generic text editors or (for images composed of
pixels) generic paint programs or (for drawings) some widely available
drawing editor, and that is suitable for input to text formatters or for
automatic translation to a variety of formats suitable for input to text
formatters. A copy made in an otherwise Transparent file format whose
markup has been designed to thwart or discourage subsequent modifi-
cation by readers is not Transparent. A copy that is not “Transparent” is
called “Opaque”.
Examples of suitable formats for Transparent copies include plain
ASCII without markup, Texinfo input format, LATEX input format, SGML
or XML using a publicly available DTD, and standard-conforming sim-
ple HTML designed for human modification. Opaque formats include
PostScript, PDF, proprietary formats that can be read and edited only by
proprietary word processors, SGML or XML for which the DTD and/or
processing tools are not generally available, and the machine-generated
HTML produced by some word processors for output purposes only.
The “Title Page” means, for a printed book, the title page itself, plus
C.2. VERBATIM COPYING 303
such following pages as are needed to hold, legibly, the material this
License requires to appear in the title page. For works in formats which
do not have any title page as such, “Title Page” means the text near the
most prominent appearance of the work’s title, preceding the beginning
of the body of the text.
C.2 Verbatim Copying

You may copy and distribute the Document in any medium, either com-
mercially or noncommercially, provided that this License, the copyright
notices, and the license notice saying this License applies to the Docu-
ment are reproduced in all copies, and that you add no other conditions
whatsoever to those of this License. You may not use technical measures
to obstruct or control the reading or further copying of the copies you
make or distribute. However, you may accept compensation in exchange
for copies. If you distribute a large enough number of copies you must
also follow the conditions in section 3.
You may also lend copies, under the same conditions stated above,
and you may publicly display copies.
C.3 Copying in Quantity

If you publish printed copies of the Document numbering more than
100, and the Document’s license notice requires Cover Texts, you must
enclose the copies in covers that carry, clearly and legibly, all these Cover
Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on the
back cover. Both covers must also clearly and legibly identify you as the
publisher of these copies. The front cover must present the full title with
all words of the title equally prominent and visible. You may add other
material on the covers in addition. Copying with changes limited to the
covers, as long as they preserve the title of the Document and satisfy
these conditions, can be treated as verbatim copying in other respects.
If the required texts for either cover are too voluminous to fit legibly,
you should put the first ones listed (as many as fit reasonably) on the
actual cover, and continue the rest onto adjacent pages.
If you publish or distribute Opaque copies of the Document number-
ing more than 100, you must either include a machine-readable Trans-
parent copy along with each Opaque copy, or state in or with each Opaque
copy a publicly-accessible computer-network location containing a com-
plete Transparent copy of the Document, free of added material, which
the general network-using public has access to download anonymously
at no charge using public-standard network protocols. If you use the
latter option, you must take reasonably prudent steps, when you begin
distribution of Opaque copies in quantity, to ensure that this Transpar-

ent copy will remain thus accessible at the stated location until at least
one year after the last time you distribute an Opaque copy (directly or
through your agents or retailers) of that edition to the public.
It is requested, but not required, that you contact the authors of the
Document well before redistributing any large number of copies, to give
them a chance to provide you with an updated version of the Document.
C.4 Modifications
You may copy and distribute a Modified Version of the Document under
the conditions of sections 2 and 3 above, provided that you release the
Modified Version under precisely this License, with the Modified Version
filling the role of the Document, thus licensing distribution and modifica-
tion of the Modified Version to whoever possesses a copy of it. In addition,
you must do these things in the Modified Version:
• Use in the Title Page (and on the covers, if any) a title distinct from
that of the Document, and from those of previous versions (which
should, if there were any, be listed in the History section of the
Document). You may use the same title as a previous version if the
original publisher of that version gives permission.
• List on the Title Page, as authors, one or more persons or enti-

ties responsible for authorship of the modifications in the Modified
Version, together with at least five of the principal authors of the
Document (all of its principal authors, if it has less than five).
• State on the Title page the name of the publisher of the Modified
Version, as the publisher.
• Preserve all the copyright notices of the Document.
• Add an appropriate copyright notice for your modifications adjacent

to the other copyright notices.
• Include, immediately after the copyright notices, a license notice

giving the public permission to use the Modified Version under the
terms of this License, in the form shown in the Addendum below.
• Preserve in that license notice the full lists of Invariant Sections and
required Cover Texts given in the Document’s license notice.
• Include an unaltered copy of this License.
• Preserve the section entitled “History”, and its title, and add to it an
item stating at least the title, year, new authors, and publisher of
C.4. MODIFICATIONS 305
the Modified Version as given on the Title Page. If there is no section

entitled “History” in the Document, create one stating the title, year,
authors, and publisher of the Document as given on its Title Page,
then add an item describing the Modified Version as stated in the
previous sentence.
• Preserve the network location, if any, given in the Document for
public access to a Transparent copy of the Document, and likewise
the network locations given in the Document for previous versions
it was based on. These may be placed in the “History” section. You
may omit a network location for a work that was published at least
four years before the Document itself, or if the original publisher of
the version it refers to gives permission.
• In any section entitled “Acknowledgements” or “Dedications”, pre-
serve the section’s title, and preserve in the section all the substance
and tone of each of the contributor acknowledgements and/or ded-
ications given therein.
• Preserve all the Invariant Sections of the Document, unaltered in
their text and in their titles. Section numbers or the equivalent are
not considered part of the section titles.
• Delete any section entitled “Endorsements”. Such a section may not
be included in the Modified Version.
• Do not retitle any existing section as “Endorsements” or to conflict
in title with any Invariant Section.
If the Modified Version includes new front-matter sections or appen-
dices that qualify as Secondary Sections and contain no material copied
from the Document, you may at your option designate some or all of these
sections as invariant. To do this, add their titles to the list of Invariant
Sections in the Modified Version’s license notice. These titles must be
distinct from any other section titles.
You may add a section entitled “Endorsements”, provided it contains
nothing but endorsements of your Modified Version by various parties –
for example, statements of peer review or that the text has been approved
by an organization as the authoritative definition of a standard.
You may add a passage of up to five words as a Front-Cover Text, and
a passage of up to 25 words as a Back-Cover Text, to the end of the list
of Cover Texts in the Modified Version. Only one passage of Front-Cover
Text and one of Back-Cover Text may be added by (or through arrange-
ments made by) any one entity. If the Document already includes a cover
text for the same cover, previously added by you or by arrangement made
by the same entity you are acting on behalf of, you may not add another;
but you may replace the old one, on explicit permission from the previous
publisher that added the old one.
The author(s) and publisher(s) of the Document do not by this License

give permission to use their names for publicity for or to assert or imply
endorsement of any Modified Version.
C.5 Combining Documents

You may combine the Document with other documents released under
this License, under the terms defined in section 4 above for modified
versions, provided that you include in the combination all of the Invariant
Sections of all of the original documents, unmodified, and list them all
as Invariant Sections of your combined work in its license notice.
The combined work need only contain one copy of this License, and
multiple identical Invariant Sections may be replaced with a single copy.
If there are multiple Invariant Sections with the same name but different
contents, make the title of each such section unique by adding at the end
of it, in parentheses, the name of the original author or publisher of that
section if known, or else a unique number. Make the same adjustment
to the section titles in the list of Invariant Sections in the license notice
of the combined work.
In the combination, you must combine any sections entitled “History”
in the various original documents, forming one section entitled “His-
tory”; likewise combine any sections entitled “Acknowledgements”, and
any sections entitled “Dedications”. You must delete all sections entitled
“Endorsements.”
C.6 Collections of Documents

You may make a collection consisting of the Document and other docu-
ments released under this License, and replace the individual copies of
this License in the various documents with a single copy that is included
in the collection, provided that you follow the rules of this License for
verbatim copying of each of the documents in all other respects.
You may extract a single document from such a collection, and dis-
tribute it individually under this License, provided you insert a copy of
this License into the extracted document, and follow this License in all
other respects regarding verbatim copying of that document.
C.7 Aggregation With Independent Works

A compilation of the Document or its derivatives with other separate and
independent documents or works, in or on a volume of a storage or dis-
tribution medium, does not as a whole count as a Modified Version of the
C.8. TRANSLATION 307
Document, provided no compilation copyright is claimed for the compila-

tion. Such a compilation is called an “aggregate”, and this License does
not apply to the other self-contained works thus compiled with the Docu-
ment, on account of their being thus compiled, if they are not themselves
derivative works of the Document.
If the Cover Text requirement of section 3 is applicable to these copies
of the Document, then if the Document is less than one quarter of the
entire aggregate, the Document’s Cover Texts may be placed on covers
that surround only the Document within the aggregate. Otherwise they
must appear on covers around the whole aggregate.
C.8 Translation
Translation is considered a kind of modification, so you may distribute
translations of the Document under the terms of section 4. Replacing In-
variant Sections with translations requires special permission from their
copyright holders, but you may include translations of some or all In-
variant Sections in addition to the original versions of these Invariant
Sections. You may include a translation of this License provided that
you also include the original English version of this License. In case of a
disagreement between the translation and the original English version of
this License, the original English version will prevail.
C.9 Termination
You may not copy, modify, sublicense, or distribute the Document ex-
cept as expressly provided for under this License. Any other attempt
to copy, modify, sublicense or distribute the Document is void, and will
automatically terminate your rights under this License. However, par-
ties who have received copies, or rights, from you under this License will
not have their licenses terminated so long as such parties remain in full
compliance.
C.10 Future Revisions of This License

The Free Software Foundation may publish new, revised versions of the
GNU Free Documentation License from time to time. Such new versions
will be similar in spirit to the present version, but may differ in detail to
address new problems or concerns. See http://www.gnu.org/copyleft/.
Each version of the License is given a distinguishing version number.
If the Document specifies that a particular numbered version of this Li-
cense ”or any later version” applies to it, you have the option of following
the terms and conditions either of that specified version or of any later
version that has been published (not as a draft) by the Free Software
Foundation. If the Document does not specify a version number of this
License, you may choose any version ever published (not as a draft) by
the Free Software Foundation.
ADDENDUM: How to use this License for your

documents
To use this License in a document you have written, include a copy of
the License in the document and put the following copyright and license
notices just after the title page:
Copyright c YEAR YOUR NAME. Permission is granted to copy,

distribute and/or modify this document under the terms of
the GNU Free Documentation License, Version 1.1 or any later
version published by the Free Software Foundation; with the
Invariant Sections being LIST THEIR TITLES, with the Front-
Cover Texts being LIST, and with the Back-Cover Texts being
LIST. A copy of the license is included in the section entitled
“GNU Free Documentation License”.
If you have no Invariant Sections, write “with no Invariant Sections”

instead of saying which ones are invariant. If you have no Front-Cover
Texts, write “no Front-Cover Texts” instead of “Front-Cover Texts being
LIST”; likewise for Back-Cover Texts.
If your document contains nontrivial examples of program code, we
recommend releasing these examples in parallel under your choice of
free software license, such as the GNU General Public License, to permit
their use in free software.
Index
x-axis, 11 circles,great, 206

x-axis,positive, 11 circles,point coordinates on, 232
xy-coordinate system, 11, 12 circles,unit, 28
y-axis, 11 circular function, 231
y-axis,positive, 12 circular function,triangles, 231
y-intercept, 41 circular function, 193, 223, 228,
229
nth root, 135 circular function,circles, 232
circular function,inverse, 267
adjacent, 224
circular function,special values, 225
amplitude, 255
circular motion, 211
analogue LP’s, 215
compound interest, 146, 148
angle, 194
compounding periods, 146
angle,central, 194
continuous compounding, 151
angle,initial side, 194
converting units, 1
angle,standard position, 194
coordinates,imposing, 11
angle,terminal side, 194
cosecant function, 234
angle,vertex, 194
cosine function, 224, 250
angular speed, 209
cotangent function, 234
arc,length, 201
curves,intersecting, 28
arc,subtended, 194
arccosine function, 272
db, 160
arcsine function, 272
decibel, 160
arctangent function, 272
decreasing function, 78
area,sector, 201
degree, 196
aspect ratio, 13
degree method, 195
axis scaling, 13
degree,minute, 196
axis units, 14
degree,second, 196
axis,horizontal, 11
density, 3
axis,vertical, 11
dependent variable, 61
belt/wheel problems, 217 difference quotient, 38
digital compact disc, 217
CD’s, 217 dilation, 172
central angle, 194 dilation,horizontal, 175
chord, 204 dilation,vertical, 173
circle, 25 directed distance, 19
circles, 26, 79 distance, directed, 19
circles,circular function, 232 distance,between two points, 17, 19
309
310 INDEX
domain, 60 graphing, 1
great circle, 204, 206
e, 149, 150
envelope of hearing, 161 horizon circle, 282
equation,quadratic, 47 horizontal line, 26
equatorial plane, 204 horizontal axis, 11
even function, 243
exponential decay, 139 identity,composition, 272
exponential function, 149 identity,even/odd, 243
exponential growth, 139 identity,key, 242
exponential modeling, 145 identity,periodicity, 242
exponential type, 139 imposing coordinates, 11, 15
independent variable, 61
function, 60 interest, 146
function,circular, 193, 223, 228, 229, intersecting curves, 28
231 intervals, 62
function,cosine, 224, 250 inverse circular function, 267, 270,
function,decreasing, 78 271
function,even, 243 inverse function, 267
function,exponential, 149
knot, 207
function,exponential type, 139
function,logarithm base b, 157 latitude, 204
function,logarithmic, 153 line,horizontal, 26
function,multipart, 81 line,vertical, 26
function,natural logarithm, 154 linear speed, 210
function,odd, 243 linear functions, 66
function,periodic, 242 linear modeling, 35
function,picturing, 57, 59 lines, 35, 41
function,rational, 183 lines,horizontal, 25
function,sine, 224, 250 lines,parallel, 46
function,sinusoidal, 193, 249, 253 lines,perpendicular, 46
function,tangent, 224, 250 lines,point slope formula, 40
function,trigonometric, 249 lines,slope intercept formula, 41
function,cos(θ), 224 lines,two point formula, 40
function,cos(x), 250 lines,vertical, 25
function,cos−1(z), 271 logarithm conversion formula, 158
function,sin(θ), 224 logarithm function base b, 157
function,sin(x), 250 logarithmic function, 153
function,sin−1(z), 271 longitude, 205
function,tan(theta), 224 loudness of sound, 159
function,tan(x), 250 LP’s, 215
function,tan−1(z), 271
mean, 254
graph, 1, 26, 27, 40 meridian, 205
graph,circular function, 244 meridian,Greenwich, 205
graph,sin(θ), 246 modeling, 1
INDEX 311
modeling,exponential, 145 RPM, 210

modeling,linear, 35 rules of exponents, 135
modeling,sinusoidal, 253
motion,circular, 211 scaling, 13
mulitpart function, 82 secant function, 234
multipart functions, 81 sector,area, 201
semicircles, 79
natural logarithm, 153 shifting, 170
natural logarithm function, 154 shifting,principle, 172
natural logarithm function, prop- sign plot, 77
erties, 154 sine function, 224, 250
nautical mile, 207 sinusoidal function, 249
navigation, 204 sinusoidal function, 193, 253
sinusoidal modeling, 253
odd function, 243 slope, 38
origin, 11 solve the triangle, 267
sound pressure level, 160
parabola,three points determine, 101
speed,angular, 209
parametric equations, 49
speed,circular, 211
period, 255
speed,linear, 210
periodic, 242, 249
standard position, 194
periodic rate, 146
standard angle, 194
phase shift, 254
standard form, 27
piano frequency range, 140
picturing a function, 57, 59 tangent function, 224, 250
positive x-axis, 11 triangle,sides, 223
positive y-axis, 12 trigonometric function, 249
principal, 146 trigonometric ratios, 225
principal domain, 271
principal domain, cosine, 271 uniform linear motion, 49
principal domain, sine, 271 unit circle, 28
principal domain, tangent, 271 units, 1
principal solution, 270
vertical axis, 11
Pythagorean Theorem, 18
vertical line test, 65
quadrants, 13 vertical lines, 26
quadratic formula, 47
radian, 200
radian method, 198
range, 61
rate, 4, 41
rate of change, 4
rational function, 183
reflection, 168
restricted domain, 61
right triangles, 231

Pre Calculus Textbook

Uploaded by

Copyright:

Available Formats

Pre Calculus Textbook

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pre Calculus Textbook

Uploaded by

Copyright:

Available Formats

Precalculus

Copyright c 2003 David H. Collingwood and K. David Prince. Permission is

Learning a new language

Is this course the same as high school Precalculus?

• As a rule of thumb, this course should on average take 15 hours

of study per week (outside class) is a more typical estimate. In other

• Because the course material is developed in a highly cumulative

Why are we using this text?

• A review of the essential mathematics needed to succeed in calculus.

• An emphasis on problem solving, the idea being to gain both experi-

Notation, Answers, etc.

How to succeed in Math 120.

What is a story problem?

an hour or more working each problem. So another aspect of working

What are the BIG errors?

Problem Solving Strategies

Figure 1: Problem solving as a transformation process.

actually translating information from one form to another. The arrows

solve your problem.

Words =⇒ Pictures =⇒ Symbols.

This problem illustrates the principle used to make a good “squirt

2 in/sec oil m in/sec

Volume Leaving Cylinder A.

Volume Entering Cylinder B.

Figure 2: A re-sketch of the original given figure.

1. Let VA and VB stand for the change in volumes as piston A moves to

2. Let dA and dB represent the diameters of each cylinder.

3. Let rA and rB represent the radii of each cylinder.

4. Let sA and sB stand for the speeds of each piston.

5. Let xA and xB stand for the distance traveled by each piston.

1. The volume of any cylinder is

where r is the radius of the cylinder, and h is its height or length.

2. “Distance” = “Rate” × “Time”. In terms of this problem, you would

Using your model, you can compute the following solutions:

3 Three Simple Curves 25

5 Functions and Graphs 57

9 Inverse Functions 119

10 Exponential Functions 133

11 Exponential Modeling 145

12 Logarithmic Functions 153

13 Three Construction Tools 165

13.4 Dilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

14 Rational Functions 181

15 Measuring an Angle 191

16 Measuring Circular Motion 207

17 The Circular Functions 221

18 Trigonometric Functions 237

18.3.1 A matter of scaling . . . . . . . . . . . . . . . . . . 243

19 Sinusoidal Functions 251

20 Inverse Circular Functions 265

A Useful Formulas 285

C GNU Free Documentation License 301

proach these questions without knowing any physics. The

1.1 Units and Rates

t = 7,685.33 sec × (1 min/60 sec) × (1 hr/60 min)

Solution. This problem is more involved. To answer this, let r be the

Plugging in what we know, we get the equation

Solving for r3 we find

from which we get

1.2 Total Change = Rate × Time